Merge branch 'act_tc-offload-originating-device'

Paul Blakey says:

====================
net/sched: Pass originating device to drivers offloading ct connection

Currently, drivers register to a ct zone that can be shared by multiple
devices. This can be inefficient for the driver to offload, as it
needs to handle all the cases where the tuple can come from,
instead of where it's most likely will arive from.

For example, consider the following tc rules:
tc filter add dev dev1 ... flower action ct commit zone 5 \
   action mirred egress redirect dev dev2

tc filter add dev dev2 ... flower action ct zone 5 \
   action goto chain chain 2
tc filter add dev dev2 ... flower ct_state +trk+est ... \
   action mirred egress redirect dev dev1

Both dev2 and dev1 register to the zone 5 flow table (created
by act_ct). A tuple originating on dev1, going to dev2, will
be offloaded to both devices, and both will need to offload
both directions, resulting in 4 total rules. The traffic
will only hit originiating tuple on dev1, and reply tuple
on dev2.

By passing the originating device that created the connection
with the tuple, dev1 can choose to offload only the originating
tuple, and dev2 only the reply tuple. Resulting in a more
efficient offload.

The first patch adds an act_ct nf conntrack extension, to
temporarily store the originiating device from the skb before
offloading the connection once the connection is established.
Once sent to offload, it fills the tuple originating device.

The second patch get this information from tuples
which pass in openvswitch.

The third patch is Mellanox driver ct offload implementation using
this information to provide a hint to firmware of where this
offloaded tuple packets will arrive from (LOCAL or UPLINK port),
and thus increase insertion rate.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 49 insertions, 5 deletions

diff --git a/drivers/net/ethernet/mellanox/mlx5/core/dev.c b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
index a8b84d53dfb0..ba6dad97e308 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/dev.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/dev.c
@@ -538,7 +538,7 @@ int mlx5_rescan_drivers_locked(struct mlx5_core_dev *dev)
 	return add_drivers(dev);
 }
 
-static bool mlx5_same_hw_devs(struct mlx5_core_dev *dev, struct mlx5_core_dev *peer_dev)
+bool mlx5_same_hw_devs(struct mlx5_core_dev *dev, struct mlx5_core_dev *peer_dev)
 {
 	u64 fsystem_guid, psystem_guid;
 
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
index 9f33729e7fc4..4a0d38d219ed 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en/tc_ct.c
@@ -14,6 +14,7 @@
 #include <linux/workqueue.h>
 #include <linux/refcount.h>
 #include <linux/xarray.h>
+#include <linux/if_macvlan.h>
 
 #include "lib/fs_chains.h"
 #include "en/tc_ct.h"
@@ -326,7 +327,33 @@ mlx5_tc_ct_rule_to_tuple_nat(struct mlx5_ct_tuple *tuple,
 }
 
 static int
-mlx5_tc_ct_set_tuple_match(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec,
+mlx5_tc_ct_get_flow_source_match(struct mlx5_tc_ct_priv *ct_priv,
+				 struct net_device *ndev)
+{
+	struct mlx5e_priv *other_priv = netdev_priv(ndev);
+	struct mlx5_core_dev *mdev = ct_priv->dev;
+	bool vf_rep, uplink_rep;
+
+	vf_rep = mlx5e_eswitch_vf_rep(ndev) && mlx5_same_hw_devs(mdev, other_priv->mdev);
+	uplink_rep = mlx5e_eswitch_uplink_rep(ndev) && mlx5_same_hw_devs(mdev, other_priv->mdev);
+
+	if (vf_rep)
+		return MLX5_FLOW_CONTEXT_FLOW_SOURCE_LOCAL_VPORT;
+	if (uplink_rep)
+		return MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+	if (is_vlan_dev(ndev))
+		return mlx5_tc_ct_get_flow_source_match(ct_priv, vlan_dev_real_dev(ndev));
+	if (netif_is_macvlan(ndev))
+		return mlx5_tc_ct_get_flow_source_match(ct_priv, macvlan_dev_real_dev(ndev));
+	if (mlx5e_get_tc_tun(ndev) || netif_is_lag_master(ndev))
+		return MLX5_FLOW_CONTEXT_FLOW_SOURCE_UPLINK;
+
+	return MLX5_FLOW_CONTEXT_FLOW_SOURCE_ANY_VPORT;
+}
+
+static int
+mlx5_tc_ct_set_tuple_match(struct mlx5_tc_ct_priv *ct_priv,
+			   struct mlx5_flow_spec *spec,
 			   struct flow_rule *rule)
 {
 	void *headers_c = MLX5_ADDR_OF(fte_match_param, spec->match_criteria,
@@ -341,8 +368,7 @@ mlx5_tc_ct_set_tuple_match(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec,
 
 		flow_rule_match_basic(rule, &match);
 
-		mlx5e_tc_set_ethertype(priv->mdev, &match, true, headers_c,
-				       headers_v);
+		mlx5e_tc_set_ethertype(ct_priv->dev, &match, true, headers_c, headers_v);
 		MLX5_SET(fte_match_set_lyr_2_4, headers_c, ip_protocol,
 			 match.mask->ip_proto);
 		MLX5_SET(fte_match_set_lyr_2_4, headers_v, ip_protocol,
@@ -438,6 +464,23 @@ mlx5_tc_ct_set_tuple_match(struct mlx5e_priv *priv, struct mlx5_flow_spec *spec,
 			 ntohs(match.key->flags));
 	}
 
+	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_META)) {
+		struct flow_match_meta match;
+
+		flow_rule_match_meta(rule, &match);
+
+		if (match.key->ingress_ifindex & match.mask->ingress_ifindex) {
+			struct net_device *dev;
+
+			dev = dev_get_by_index(&init_net, match.key->ingress_ifindex);
+			if (dev && MLX5_CAP_ESW_FLOWTABLE(ct_priv->dev, flow_source))
+				spec->flow_context.flow_source =
+					mlx5_tc_ct_get_flow_source_match(ct_priv, dev);
+
+			dev_put(dev);
+		}
+	}
+
 	return 0;
 }
 
@@ -770,7 +813,7 @@ mlx5_tc_ct_entry_add_rule(struct mlx5_tc_ct_priv *ct_priv,
 	if (ct_priv->ns_type == MLX5_FLOW_NAMESPACE_FDB)
 		attr->esw_attr->in_mdev = priv->mdev;
 
-	mlx5_tc_ct_set_tuple_match(netdev_priv(ct_priv->netdev), spec, flow_rule);
+	mlx5_tc_ct_set_tuple_match(ct_priv, spec, flow_rule);
 	mlx5e_tc_match_to_reg_match(spec, ZONE_TO_REG, entry->tuple.zone, MLX5_CT_ZONE_MASK);
 
 	zone_rule->rule = mlx5_tc_rule_insert(priv, spec, attr);
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
index bb677329ea08..6f8baa0f2a73 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
+++ b/drivers/net/ethernet/mellanox/mlx5/core/mlx5_core.h
@@ -305,5 +305,6 @@ static inline u32 mlx5_sriov_get_vf_total_msix(struct pci_dev *pdev)
 bool mlx5_eth_supported(struct mlx5_core_dev *dev);
 bool mlx5_rdma_supported(struct mlx5_core_dev *dev);
 bool mlx5_vnet_supported(struct mlx5_core_dev *dev);
+bool mlx5_same_hw_devs(struct mlx5_core_dev *dev, struct mlx5_core_dev *peer_dev);
 
 #endif /* __MLX5_CORE_H__ */