diff options
Diffstat (limited to 'drivers/infiniband')
96 files changed, 1250 insertions, 919 deletions
diff --git a/drivers/infiniband/core/cm.c b/drivers/infiniband/core/cm.c index 142170473e75..8670e58675c6 100644 --- a/drivers/infiniband/core/cm.c +++ b/drivers/infiniband/core/cm.c @@ -36,6 +36,7 @@ MODULE_LICENSE("Dual BSD/GPL"); #define CM_DESTROY_ID_WAIT_TIMEOUT 10000 /* msecs */ #define CM_DIRECT_RETRY_CTX ((void *) 1UL) +#define CM_MRA_SETTING 24 /* 4.096us * 2^24 = ~68.7 seconds */ static const char * const ibcm_rej_reason_strs[] = { [IB_CM_REJ_NO_QP] = "no QP", @@ -167,7 +168,7 @@ struct cm_port { struct cm_device { struct kref kref; struct list_head list; - spinlock_t mad_agent_lock; + rwlock_t mad_agent_lock; struct ib_device *ib_device; u8 ack_delay; int going_down; @@ -241,7 +242,6 @@ struct cm_id_private { u8 initiator_depth; u8 retry_count; u8 rnr_retry_count; - u8 service_timeout; u8 target_ack_delay; struct list_head work_list; @@ -285,7 +285,7 @@ static struct ib_mad_send_buf *cm_alloc_msg(struct cm_id_private *cm_id_priv) if (!cm_id_priv->av.port) return ERR_PTR(-EINVAL); - spin_lock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); + read_lock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); mad_agent = cm_id_priv->av.port->mad_agent; if (!mad_agent) { m = ERR_PTR(-EINVAL); @@ -311,7 +311,7 @@ static struct ib_mad_send_buf *cm_alloc_msg(struct cm_id_private *cm_id_priv) m->ah = ah; out: - spin_unlock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); + read_unlock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); return m; } @@ -1297,10 +1297,10 @@ static __be64 cm_form_tid(struct cm_id_private *cm_id_priv) if (!cm_id_priv->av.port) return cpu_to_be64(low_tid); - spin_lock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); + read_lock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); if (cm_id_priv->av.port->mad_agent) hi_tid = ((u64)cm_id_priv->av.port->mad_agent->hi_tid) << 32; - spin_unlock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); + read_unlock(&cm_id_priv->av.port->cm_dev->mad_agent_lock); return cpu_to_be64(hi_tid | low_tid); } @@ -1872,7 +1872,7 @@ static void cm_process_work(struct cm_id_private *cm_id_priv, static void cm_format_mra(struct cm_mra_msg *mra_msg, struct cm_id_private *cm_id_priv, - enum cm_msg_response msg_mraed, u8 service_timeout, + enum cm_msg_response msg_mraed, const void *private_data, u8 private_data_len) { cm_format_mad_hdr(&mra_msg->hdr, CM_MRA_ATTR_ID, cm_id_priv->tid); @@ -1881,7 +1881,7 @@ static void cm_format_mra(struct cm_mra_msg *mra_msg, be32_to_cpu(cm_id_priv->id.local_id)); IBA_SET(CM_MRA_REMOTE_COMM_ID, mra_msg, be32_to_cpu(cm_id_priv->id.remote_id)); - IBA_SET(CM_MRA_SERVICE_TIMEOUT, mra_msg, service_timeout); + IBA_SET(CM_MRA_SERVICE_TIMEOUT, mra_msg, CM_MRA_SETTING); if (private_data && private_data_len) IBA_SET_MEM(CM_MRA_PRIVATE_DATA, mra_msg, private_data, @@ -1960,7 +1960,7 @@ static void cm_dup_req_handler(struct cm_work *work, switch (cm_id_priv->id.state) { case IB_CM_MRA_REQ_SENT: cm_format_mra((struct cm_mra_msg *) msg->mad, cm_id_priv, - CM_MSG_RESPONSE_REQ, cm_id_priv->service_timeout, + CM_MSG_RESPONSE_REQ, cm_id_priv->private_data, cm_id_priv->private_data_len); break; @@ -2454,7 +2454,7 @@ static void cm_dup_rep_handler(struct cm_work *work) cm_id_priv->private_data_len); else if (cm_id_priv->id.state == IB_CM_MRA_REP_SENT) cm_format_mra((struct cm_mra_msg *) msg->mad, cm_id_priv, - CM_MSG_RESPONSE_REP, cm_id_priv->service_timeout, + CM_MSG_RESPONSE_REP, cm_id_priv->private_data, cm_id_priv->private_data_len); else @@ -3094,26 +3094,13 @@ out: return -EINVAL; } -int ib_send_cm_mra(struct ib_cm_id *cm_id, - u8 service_timeout, - const void *private_data, - u8 private_data_len) +int ib_prepare_cm_mra(struct ib_cm_id *cm_id) { struct cm_id_private *cm_id_priv; - struct ib_mad_send_buf *msg; enum ib_cm_state cm_state; enum ib_cm_lap_state lap_state; - enum cm_msg_response msg_response; - void *data; unsigned long flags; - int ret; - - if (private_data && private_data_len > IB_CM_MRA_PRIVATE_DATA_SIZE) - return -EINVAL; - - data = cm_copy_private_data(private_data, private_data_len); - if (IS_ERR(data)) - return PTR_ERR(data); + int ret = 0; cm_id_priv = container_of(cm_id, struct cm_id_private, id); @@ -3122,58 +3109,33 @@ int ib_send_cm_mra(struct ib_cm_id *cm_id, case IB_CM_REQ_RCVD: cm_state = IB_CM_MRA_REQ_SENT; lap_state = cm_id->lap_state; - msg_response = CM_MSG_RESPONSE_REQ; break; case IB_CM_REP_RCVD: cm_state = IB_CM_MRA_REP_SENT; lap_state = cm_id->lap_state; - msg_response = CM_MSG_RESPONSE_REP; break; case IB_CM_ESTABLISHED: if (cm_id->lap_state == IB_CM_LAP_RCVD) { cm_state = cm_id->state; lap_state = IB_CM_MRA_LAP_SENT; - msg_response = CM_MSG_RESPONSE_OTHER; break; } fallthrough; default: - trace_icm_send_mra_unknown_err(&cm_id_priv->id); + trace_icm_prepare_mra_unknown_err(&cm_id_priv->id); ret = -EINVAL; goto error_unlock; } - if (!(service_timeout & IB_CM_MRA_FLAG_DELAY)) { - msg = cm_alloc_msg(cm_id_priv); - if (IS_ERR(msg)) { - ret = PTR_ERR(msg); - goto error_unlock; - } - - cm_format_mra((struct cm_mra_msg *) msg->mad, cm_id_priv, - msg_response, service_timeout, - private_data, private_data_len); - trace_icm_send_mra(cm_id); - ret = ib_post_send_mad(msg, NULL); - if (ret) - goto error_free_msg; - } - cm_id->state = cm_state; cm_id->lap_state = lap_state; - cm_id_priv->service_timeout = service_timeout; - cm_set_private_data(cm_id_priv, data, private_data_len); - spin_unlock_irqrestore(&cm_id_priv->lock, flags); - return 0; + cm_set_private_data(cm_id_priv, NULL, 0); -error_free_msg: - cm_free_msg(msg); error_unlock: spin_unlock_irqrestore(&cm_id_priv->lock, flags); - kfree(data); return ret; } -EXPORT_SYMBOL(ib_send_cm_mra); +EXPORT_SYMBOL(ib_prepare_cm_mra); static struct cm_id_private *cm_acquire_mraed_id(struct cm_mra_msg *mra_msg) { @@ -3377,7 +3339,6 @@ static int cm_lap_handler(struct cm_work *work) cm_format_mra((struct cm_mra_msg *) msg->mad, cm_id_priv, CM_MSG_RESPONSE_OTHER, - cm_id_priv->service_timeout, cm_id_priv->private_data, cm_id_priv->private_data_len); spin_unlock_irq(&cm_id_priv->lock); @@ -3786,7 +3747,8 @@ static void cm_process_send_error(struct cm_id_private *cm_id_priv, spin_lock_irq(&cm_id_priv->lock); if (msg != cm_id_priv->msg) { spin_unlock_irq(&cm_id_priv->lock); - cm_free_priv_msg(msg); + cm_free_msg(msg); + cm_deref_id(cm_id_priv); return; } cm_free_priv_msg(msg); @@ -4378,7 +4340,7 @@ static int cm_add_one(struct ib_device *ib_device) return -ENOMEM; kref_init(&cm_dev->kref); - spin_lock_init(&cm_dev->mad_agent_lock); + rwlock_init(&cm_dev->mad_agent_lock); cm_dev->ib_device = ib_device; cm_dev->ack_delay = ib_device->attrs.local_ca_ack_delay; cm_dev->going_down = 0; @@ -4494,9 +4456,9 @@ static void cm_remove_one(struct ib_device *ib_device, void *client_data) * The above ensures no call paths from the work are running, * the remaining paths all take the mad_agent_lock. */ - spin_lock(&cm_dev->mad_agent_lock); + write_lock(&cm_dev->mad_agent_lock); port->mad_agent = NULL; - spin_unlock(&cm_dev->mad_agent_lock); + write_unlock(&cm_dev->mad_agent_lock); ib_unregister_mad_agent(mad_agent); ib_port_unregister_client_groups(ib_device, i, cm_counter_groups); diff --git a/drivers/infiniband/core/cm_trace.h b/drivers/infiniband/core/cm_trace.h index 944d9071245d..4a4987da69d4 100644 --- a/drivers/infiniband/core/cm_trace.h +++ b/drivers/infiniband/core/cm_trace.h @@ -229,7 +229,7 @@ DEFINE_CM_ERR_EVENT(send_drep); DEFINE_CM_ERR_EVENT(dreq_unknown); DEFINE_CM_ERR_EVENT(send_unknown_rej); DEFINE_CM_ERR_EVENT(rej_unknown); -DEFINE_CM_ERR_EVENT(send_mra_unknown); +DEFINE_CM_ERR_EVENT(prepare_mra_unknown); DEFINE_CM_ERR_EVENT(mra_unknown); DEFINE_CM_ERR_EVENT(qp_init); DEFINE_CM_ERR_EVENT(qp_rtr); diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index fedcdb56fb6b..9b471548e7ae 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -46,7 +46,6 @@ MODULE_LICENSE("Dual BSD/GPL"); #define CMA_CM_RESPONSE_TIMEOUT 20 #define CMA_MAX_CM_RETRIES 15 -#define CMA_CM_MRA_SETTING (IB_CM_MRA_FLAG_DELAY | 24) #define CMA_IBOE_PACKET_LIFETIME 16 #define CMA_PREFERRED_ROCE_GID_TYPE IB_GID_TYPE_ROCE_UDP_ENCAP @@ -72,6 +71,8 @@ static const char * const cma_events[] = { static void cma_iboe_set_mgid(struct sockaddr *addr, union ib_gid *mgid, enum ib_gid_type gid_type); +static void cma_netevent_work_handler(struct work_struct *_work); + const char *__attribute_const__ rdma_event_msg(enum rdma_cm_event_type event) { size_t index = event; @@ -144,19 +145,6 @@ struct iw_cm_id *rdma_iw_cm_id(struct rdma_cm_id *id) } EXPORT_SYMBOL(rdma_iw_cm_id); -/** - * rdma_res_to_id() - return the rdma_cm_id pointer for this restrack. - * @res: rdma resource tracking entry pointer - */ -struct rdma_cm_id *rdma_res_to_id(struct rdma_restrack_entry *res) -{ - struct rdma_id_private *id_priv = - container_of(res, struct rdma_id_private, res); - - return &id_priv->id; -} -EXPORT_SYMBOL(rdma_res_to_id); - static int cma_add_one(struct ib_device *device); static void cma_remove_one(struct ib_device *device, void *client_data); @@ -1047,6 +1035,7 @@ __rdma_create_id(struct net *net, rdma_cm_event_handler event_handler, get_random_bytes(&id_priv->seq_num, sizeof id_priv->seq_num); id_priv->id.route.addr.dev_addr.net = get_net(net); id_priv->seq_num &= 0x00ffffff; + INIT_WORK(&id_priv->id.net_work, cma_netevent_work_handler); rdma_restrack_new(&id_priv->res, RDMA_RESTRACK_CM_ID); if (parent) @@ -2211,8 +2200,8 @@ static int cma_ib_handler(struct ib_cm_id *cm_id, case IB_CM_REP_RECEIVED: if (state == RDMA_CM_CONNECT && (id_priv->id.qp_type != IB_QPT_UD)) { - trace_cm_send_mra(id_priv); - ib_send_cm_mra(cm_id, CMA_CM_MRA_SETTING, NULL, 0); + trace_cm_prepare_mra(id_priv); + ib_prepare_cm_mra(cm_id); } if (id_priv->id.qp) { event.status = cma_rep_recv(id_priv); @@ -2473,8 +2462,8 @@ static int cma_ib_req_handler(struct ib_cm_id *cm_id, if (READ_ONCE(conn_id->state) == RDMA_CM_CONNECT && conn_id->id.qp_type != IB_QPT_UD) { - trace_cm_send_mra(cm_id->context); - ib_send_cm_mra(cm_id, CMA_CM_MRA_SETTING, NULL, 0); + trace_cm_prepare_mra(cm_id->context); + ib_prepare_cm_mra(cm_id); } mutex_unlock(&conn_id->handler_mutex); @@ -5241,9 +5230,9 @@ static int cma_netevent_callback(struct notifier_block *self, if (!memcmp(current_id->id.route.addr.dev_addr.dst_dev_addr, neigh->ha, ETH_ALEN)) continue; - INIT_WORK(¤t_id->id.net_work, cma_netevent_work_handler); cma_id_get(current_id); - queue_work(cma_wq, ¤t_id->id.net_work); + if (!queue_work(cma_wq, ¤t_id->id.net_work)) + cma_id_put(current_id); } out: spin_unlock_irqrestore(&id_table_lock, flags); diff --git a/drivers/infiniband/core/cma_trace.h b/drivers/infiniband/core/cma_trace.h index dc622f3778be..3456d5f3aa47 100644 --- a/drivers/infiniband/core/cma_trace.h +++ b/drivers/infiniband/core/cma_trace.h @@ -55,7 +55,7 @@ DECLARE_EVENT_CLASS(cma_fsm_class, DEFINE_CMA_FSM_EVENT(send_rtu); DEFINE_CMA_FSM_EVENT(send_rej); -DEFINE_CMA_FSM_EVENT(send_mra); +DEFINE_CMA_FSM_EVENT(prepare_mra); DEFINE_CMA_FSM_EVENT(send_sidr_req); DEFINE_CMA_FSM_EVENT(send_sidr_rep); DEFINE_CMA_FSM_EVENT(disconnect); diff --git a/drivers/infiniband/core/device.c b/drivers/infiniband/core/device.c index b4e3e4beb7f4..d4263385850a 100644 --- a/drivers/infiniband/core/device.c +++ b/drivers/infiniband/core/device.c @@ -1352,6 +1352,9 @@ static void ib_device_notify_register(struct ib_device *device) down_read(&devices_rwsem); + /* Mark for userspace that device is ready */ + kobject_uevent(&device->dev.kobj, KOBJ_ADD); + ret = rdma_nl_notify_event(device, 0, RDMA_REGISTER_EVENT); if (ret) goto out; @@ -1468,10 +1471,9 @@ int ib_register_device(struct ib_device *device, const char *name, return ret; } dev_set_uevent_suppress(&device->dev, false); - /* Mark for userspace that device is ready */ - kobject_uevent(&device->dev.kobj, KOBJ_ADD); ib_device_notify_register(device); + ib_device_put(device); return 0; diff --git a/drivers/infiniband/core/iwcm.c b/drivers/infiniband/core/iwcm.c index f4486cbd8f45..62410578dec3 100644 --- a/drivers/infiniband/core/iwcm.c +++ b/drivers/infiniband/core/iwcm.c @@ -368,12 +368,9 @@ EXPORT_SYMBOL(iw_cm_disconnect); /* * CM_ID <-- DESTROYING * - * Clean up all resources associated with the connection and release - * the initial reference taken by iw_create_cm_id. - * - * Returns true if and only if the last cm_id_priv reference has been dropped. + * Clean up all resources associated with the connection. */ -static bool destroy_cm_id(struct iw_cm_id *cm_id) +static void destroy_cm_id(struct iw_cm_id *cm_id) { struct iwcm_id_private *cm_id_priv; struct ib_qp *qp; @@ -442,20 +439,22 @@ static bool destroy_cm_id(struct iw_cm_id *cm_id) iwpm_remove_mapinfo(&cm_id->local_addr, &cm_id->m_local_addr); iwpm_remove_mapping(&cm_id->local_addr, RDMA_NL_IWCM); } - - return iwcm_deref_id(cm_id_priv); } /* - * This function is only called by the application thread and cannot - * be called by the event thread. The function will wait for all - * references to be released on the cm_id and then kfree the cm_id - * object. + * Destroy cm_id. If the cm_id still has other references, wait for all + * references to be released on the cm_id and then release the initial + * reference taken by iw_create_cm_id. */ void iw_destroy_cm_id(struct iw_cm_id *cm_id) { - if (!destroy_cm_id(cm_id)) + struct iwcm_id_private *cm_id_priv; + + cm_id_priv = container_of(cm_id, struct iwcm_id_private, id); + destroy_cm_id(cm_id); + if (refcount_read(&cm_id_priv->refcount) > 1) flush_workqueue(iwcm_wq); + iwcm_deref_id(cm_id_priv); } EXPORT_SYMBOL(iw_destroy_cm_id); @@ -1035,8 +1034,10 @@ static void cm_work_handler(struct work_struct *_work) if (!test_bit(IWCM_F_DROP_EVENTS, &cm_id_priv->flags)) { ret = process_event(cm_id_priv, &levent); - if (ret) - WARN_ON_ONCE(destroy_cm_id(&cm_id_priv->id)); + if (ret) { + destroy_cm_id(&cm_id_priv->id); + WARN_ON_ONCE(iwcm_deref_id(cm_id_priv)); + } } else pr_debug("dropping event %d\n", levent.event); if (iwcm_deref_id(cm_id_priv)) diff --git a/drivers/infiniband/core/mad_rmpp.c b/drivers/infiniband/core/mad_rmpp.c index 8af0619a39cd..b4b10e8a6495 100644 --- a/drivers/infiniband/core/mad_rmpp.c +++ b/drivers/infiniband/core/mad_rmpp.c @@ -158,7 +158,7 @@ static struct ib_mad_send_buf *alloc_response_msg(struct ib_mad_agent *agent, ah = ib_create_ah_from_wc(agent->qp->pd, recv_wc->wc, recv_wc->recv_buf.grh, agent->port_num); if (IS_ERR(ah)) - return (void *) ah; + return ERR_CAST(ah); hdr_len = ib_get_mad_data_offset(recv_wc->recv_buf.mad->mad_hdr.mgmt_class); msg = ib_create_send_mad(agent, recv_wc->wc->src_qp, diff --git a/drivers/infiniband/core/ucaps.c b/drivers/infiniband/core/ucaps.c index 6853c6d078f9..de5cb8bf0a61 100644 --- a/drivers/infiniband/core/ucaps.c +++ b/drivers/infiniband/core/ucaps.c @@ -170,7 +170,7 @@ int ib_create_ucap(enum rdma_user_cap type) ucap->dev.class = &ucaps_class; ucap->dev.devt = MKDEV(MAJOR(ucaps_base_dev), type); ucap->dev.release = ucap_dev_release; - ret = dev_set_name(&ucap->dev, ucap_names[type]); + ret = dev_set_name(&ucap->dev, "%s", ucap_names[type]); if (ret) goto err_device; diff --git a/drivers/infiniband/core/umem_odp.c b/drivers/infiniband/core/umem_odp.c index e9fa22d31c23..c752ae9fad6c 100644 --- a/drivers/infiniband/core/umem_odp.c +++ b/drivers/infiniband/core/umem_odp.c @@ -41,65 +41,72 @@ #include <linux/hugetlb.h> #include <linux/interval_tree.h> #include <linux/hmm.h> +#include <linux/hmm-dma.h> #include <linux/pagemap.h> #include <rdma/ib_umem_odp.h> #include "uverbs.h" -static inline int ib_init_umem_odp(struct ib_umem_odp *umem_odp, - const struct mmu_interval_notifier_ops *ops) +static void ib_init_umem_implicit_odp(struct ib_umem_odp *umem_odp) { - int ret; + umem_odp->is_implicit_odp = 1; + umem_odp->umem.is_odp = 1; + mutex_init(&umem_odp->umem_mutex); +} + +static int ib_init_umem_odp(struct ib_umem_odp *umem_odp, + const struct mmu_interval_notifier_ops *ops) +{ + struct ib_device *dev = umem_odp->umem.ibdev; + size_t page_size = 1UL << umem_odp->page_shift; + struct hmm_dma_map *map; + unsigned long start; + unsigned long end; + size_t nr_entries; + int ret = 0; umem_odp->umem.is_odp = 1; mutex_init(&umem_odp->umem_mutex); - if (!umem_odp->is_implicit_odp) { - size_t page_size = 1UL << umem_odp->page_shift; - unsigned long start; - unsigned long end; - size_t ndmas, npfns; - - start = ALIGN_DOWN(umem_odp->umem.address, page_size); - if (check_add_overflow(umem_odp->umem.address, - (unsigned long)umem_odp->umem.length, - &end)) - return -EOVERFLOW; - end = ALIGN(end, page_size); - if (unlikely(end < page_size)) - return -EOVERFLOW; - - ndmas = (end - start) >> umem_odp->page_shift; - if (!ndmas) - return -EINVAL; - - npfns = (end - start) >> PAGE_SHIFT; - umem_odp->pfn_list = kvcalloc( - npfns, sizeof(*umem_odp->pfn_list), GFP_KERNEL); - if (!umem_odp->pfn_list) - return -ENOMEM; - - umem_odp->dma_list = kvcalloc( - ndmas, sizeof(*umem_odp->dma_list), GFP_KERNEL); - if (!umem_odp->dma_list) { + start = ALIGN_DOWN(umem_odp->umem.address, page_size); + if (check_add_overflow(umem_odp->umem.address, + (unsigned long)umem_odp->umem.length, &end)) + return -EOVERFLOW; + end = ALIGN(end, page_size); + if (unlikely(end < page_size)) + return -EOVERFLOW; + + nr_entries = (end - start) >> PAGE_SHIFT; + if (!(nr_entries * PAGE_SIZE / page_size)) + return -EINVAL; + + map = &umem_odp->map; + if (ib_uses_virt_dma(dev)) { + map->pfn_list = kvcalloc(nr_entries, sizeof(*map->pfn_list), + GFP_KERNEL | __GFP_NOWARN); + if (!map->pfn_list) ret = -ENOMEM; - goto out_pfn_list; - } + } else + ret = hmm_dma_map_alloc(dev->dma_device, map, + (end - start) >> PAGE_SHIFT, + 1 << umem_odp->page_shift); + if (ret) + return ret; - ret = mmu_interval_notifier_insert(&umem_odp->notifier, - umem_odp->umem.owning_mm, - start, end - start, ops); - if (ret) - goto out_dma_list; - } + ret = mmu_interval_notifier_insert(&umem_odp->notifier, + umem_odp->umem.owning_mm, start, + end - start, ops); + if (ret) + goto out_free_map; return 0; -out_dma_list: - kvfree(umem_odp->dma_list); -out_pfn_list: - kvfree(umem_odp->pfn_list); +out_free_map: + if (ib_uses_virt_dma(dev)) + kfree(map->pfn_list); + else + hmm_dma_map_free(dev->dma_device, map); return ret; } @@ -118,7 +125,6 @@ struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_device *device, { struct ib_umem *umem; struct ib_umem_odp *umem_odp; - int ret; if (access & IB_ACCESS_HUGETLB) return ERR_PTR(-EINVAL); @@ -130,16 +136,10 @@ struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_device *device, umem->ibdev = device; umem->writable = ib_access_writable(access); umem->owning_mm = current->mm; - umem_odp->is_implicit_odp = 1; umem_odp->page_shift = PAGE_SHIFT; umem_odp->tgid = get_task_pid(current->group_leader, PIDTYPE_PID); - ret = ib_init_umem_odp(umem_odp, NULL); - if (ret) { - put_pid(umem_odp->tgid); - kfree(umem_odp); - return ERR_PTR(ret); - } + ib_init_umem_implicit_odp(umem_odp); return umem_odp; } EXPORT_SYMBOL(ib_umem_odp_alloc_implicit); @@ -260,74 +260,41 @@ err_put_pid: } EXPORT_SYMBOL(ib_umem_odp_get); -void ib_umem_odp_release(struct ib_umem_odp *umem_odp) +static void ib_umem_odp_free(struct ib_umem_odp *umem_odp) { + struct ib_device *dev = umem_odp->umem.ibdev; + /* * Ensure that no more pages are mapped in the umem. * * It is the driver's responsibility to ensure, before calling us, * that the hardware will not attempt to access the MR any more. */ - if (!umem_odp->is_implicit_odp) { - mutex_lock(&umem_odp->umem_mutex); - ib_umem_odp_unmap_dma_pages(umem_odp, ib_umem_start(umem_odp), - ib_umem_end(umem_odp)); - mutex_unlock(&umem_odp->umem_mutex); - mmu_interval_notifier_remove(&umem_odp->notifier); - kvfree(umem_odp->dma_list); - kvfree(umem_odp->pfn_list); - } - put_pid(umem_odp->tgid); - kfree(umem_odp); + mutex_lock(&umem_odp->umem_mutex); + ib_umem_odp_unmap_dma_pages(umem_odp, ib_umem_start(umem_odp), + ib_umem_end(umem_odp)); + mutex_unlock(&umem_odp->umem_mutex); + mmu_interval_notifier_remove(&umem_odp->notifier); + if (ib_uses_virt_dma(dev)) + kfree(umem_odp->map.pfn_list); + else + hmm_dma_map_free(dev->dma_device, &umem_odp->map); } -EXPORT_SYMBOL(ib_umem_odp_release); -/* - * Map for DMA and insert a single page into the on-demand paging page tables. - * - * @umem: the umem to insert the page to. - * @dma_index: index in the umem to add the dma to. - * @page: the page struct to map and add. - * @access_mask: access permissions needed for this page. - * - * The function returns -EFAULT if the DMA mapping operation fails. - * - */ -static int ib_umem_odp_map_dma_single_page( - struct ib_umem_odp *umem_odp, - unsigned int dma_index, - struct page *page, - u64 access_mask) +void ib_umem_odp_release(struct ib_umem_odp *umem_odp) { - struct ib_device *dev = umem_odp->umem.ibdev; - dma_addr_t *dma_addr = &umem_odp->dma_list[dma_index]; - - if (*dma_addr) { - /* - * If the page is already dma mapped it means it went through - * a non-invalidating trasition, like read-only to writable. - * Resync the flags. - */ - *dma_addr = (*dma_addr & ODP_DMA_ADDR_MASK) | access_mask; - return 0; - } + if (!umem_odp->is_implicit_odp) + ib_umem_odp_free(umem_odp); - *dma_addr = ib_dma_map_page(dev, page, 0, 1 << umem_odp->page_shift, - DMA_BIDIRECTIONAL); - if (ib_dma_mapping_error(dev, *dma_addr)) { - *dma_addr = 0; - return -EFAULT; - } - umem_odp->npages++; - *dma_addr |= access_mask; - return 0; + put_pid(umem_odp->tgid); + kfree(umem_odp); } +EXPORT_SYMBOL(ib_umem_odp_release); /** * ib_umem_odp_map_dma_and_lock - DMA map userspace memory in an ODP MR and lock it. * * Maps the range passed in the argument to DMA addresses. - * The DMA addresses of the mapped pages is updated in umem_odp->dma_list. * Upon success the ODP MR will be locked to let caller complete its device * page table update. * @@ -355,9 +322,6 @@ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt, struct hmm_range range = {}; unsigned long timeout; - if (access_mask == 0) - return -EINVAL; - if (user_virt < ib_umem_start(umem_odp) || user_virt + bcnt > ib_umem_end(umem_odp)) return -EFAULT; @@ -383,11 +347,11 @@ int ib_umem_odp_map_dma_and_lock(struct ib_umem_odp *umem_odp, u64 user_virt, if (fault) { range.default_flags = HMM_PFN_REQ_FAULT; - if (access_mask & ODP_WRITE_ALLOWED_BIT) + if (access_mask & HMM_PFN_WRITE) range.default_flags |= HMM_PFN_REQ_WRITE; } - range.hmm_pfns = &(umem_odp->pfn_list[pfn_start_idx]); + range.hmm_pfns = &(umem_odp->map.pfn_list[pfn_start_idx]); timeout = jiffies + msecs_to_jiffies(HMM_RANGE_DEFAULT_TIMEOUT); retry: @@ -415,22 +379,17 @@ retry: for (pfn_index = 0; pfn_index < num_pfns; pfn_index += 1 << (page_shift - PAGE_SHIFT), dma_index++) { - if (fault) { - /* - * Since we asked for hmm_range_fault() to populate - * pages it shouldn't return an error entry on success. - */ - WARN_ON(range.hmm_pfns[pfn_index] & HMM_PFN_ERROR); - WARN_ON(!(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)); - } else { - if (!(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)) { - WARN_ON(umem_odp->dma_list[dma_index]); - continue; - } - access_mask = ODP_READ_ALLOWED_BIT; - if (range.hmm_pfns[pfn_index] & HMM_PFN_WRITE) - access_mask |= ODP_WRITE_ALLOWED_BIT; - } + /* + * Since we asked for hmm_range_fault() to populate + * pages it shouldn't return an error entry on success. + */ + WARN_ON(fault && range.hmm_pfns[pfn_index] & HMM_PFN_ERROR); + WARN_ON(fault && !(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)); + if (!(range.hmm_pfns[pfn_index] & HMM_PFN_VALID)) + continue; + + if (range.hmm_pfns[pfn_index] & HMM_PFN_DMA_MAPPED) + continue; hmm_order = hmm_pfn_to_map_order(range.hmm_pfns[pfn_index]); /* If a hugepage was detected and ODP wasn't set for, the umem @@ -443,15 +402,6 @@ retry: __func__, hmm_order, page_shift); break; } - - ret = ib_umem_odp_map_dma_single_page( - umem_odp, dma_index, hmm_pfn_to_page(range.hmm_pfns[pfn_index]), - access_mask); - if (ret < 0) { - ibdev_dbg(umem_odp->umem.ibdev, - "ib_umem_odp_map_dma_single_page failed with error %d\n", ret); - break; - } } /* upon success lock should stay on hold for the callee */ if (!ret) @@ -471,45 +421,38 @@ EXPORT_SYMBOL(ib_umem_odp_map_dma_and_lock); void ib_umem_odp_unmap_dma_pages(struct ib_umem_odp *umem_odp, u64 virt, u64 bound) { - dma_addr_t dma_addr; - dma_addr_t dma; - int idx; - u64 addr; struct ib_device *dev = umem_odp->umem.ibdev; + u64 addr; lockdep_assert_held(&umem_odp->umem_mutex); virt = max_t(u64, virt, ib_umem_start(umem_odp)); bound = min_t(u64, bound, ib_umem_end(umem_odp)); for (addr = virt; addr < bound; addr += BIT(umem_odp->page_shift)) { - idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift; - dma = umem_odp->dma_list[idx]; - - /* The access flags guaranteed a valid DMA address in case was NULL */ - if (dma) { - unsigned long pfn_idx = (addr - ib_umem_start(umem_odp)) >> PAGE_SHIFT; - struct page *page = hmm_pfn_to_page(umem_odp->pfn_list[pfn_idx]); - - dma_addr = dma & ODP_DMA_ADDR_MASK; - ib_dma_unmap_page(dev, dma_addr, - BIT(umem_odp->page_shift), - DMA_BIDIRECTIONAL); - if (dma & ODP_WRITE_ALLOWED_BIT) { - struct page *head_page = compound_head(page); - /* - * set_page_dirty prefers being called with - * the page lock. However, MMU notifiers are - * called sometimes with and sometimes without - * the lock. We rely on the umem_mutex instead - * to prevent other mmu notifiers from - * continuing and allowing the page mapping to - * be removed. - */ - set_page_dirty(head_page); - } - umem_odp->dma_list[idx] = 0; - umem_odp->npages--; + u64 offset = addr - ib_umem_start(umem_odp); + size_t idx = offset >> umem_odp->page_shift; + unsigned long pfn = umem_odp->map.pfn_list[idx]; + + if (!hmm_dma_unmap_pfn(dev->dma_device, &umem_odp->map, idx)) + goto clear; + + if (pfn & HMM_PFN_WRITE) { + struct page *page = hmm_pfn_to_page(pfn); + struct page *head_page = compound_head(page); + /* + * set_page_dirty prefers being called with + * the page lock. However, MMU notifiers are + * called sometimes with and sometimes without + * the lock. We rely on the umem_mutex instead + * to prevent other mmu notifiers from + * continuing and allowing the page mapping to + * be removed. + */ + set_page_dirty(head_page); } + umem_odp->npages--; +clear: + umem_odp->map.pfn_list[idx] &= ~HMM_PFN_FLAGS; } } EXPORT_SYMBOL(ib_umem_odp_unmap_dma_pages); diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 3c3bb670c805..bc9fe3ceca4d 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -193,7 +193,7 @@ _ib_uverbs_lookup_comp_file(s32 fd, struct uverbs_attr_bundle *attrs) fd, attrs); if (IS_ERR(uobj)) - return (void *)uobj; + return ERR_CAST(uobj); uverbs_uobject_get(uobj); uobj_put_read(uobj); diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index c5e78bbefbd0..75fde0fe9989 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -572,7 +572,7 @@ struct ib_ah *rdma_create_ah(struct ib_pd *pd, struct rdma_ah_attr *ah_attr, GFP_KERNEL : GFP_ATOMIC); if (IS_ERR(slave)) { rdma_unfill_sgid_attr(ah_attr, old_sgid_attr); - return (void *)slave; + return ERR_CAST(slave); } ah = _rdma_create_ah(pd, ah_attr, flags, NULL, slave); rdma_lag_put_ah_roce_slave(slave); diff --git a/drivers/infiniband/hw/bnxt_re/debugfs.c b/drivers/infiniband/hw/bnxt_re/debugfs.c index af91d16c3c77..e632f1661b92 100644 --- a/drivers/infiniband/hw/bnxt_re/debugfs.c +++ b/drivers/infiniband/hw/bnxt_re/debugfs.c @@ -170,6 +170,9 @@ static int map_cc_config_offset_gen0_ext0(u32 offset, struct bnxt_qplib_cc_param case CMDQ_MODIFY_ROCE_CC_MODIFY_MASK_TCP_CP: *val = ccparam->tcp_cp; break; + case CMDQ_MODIFY_ROCE_CC_MODIFY_MASK_INACTIVITY_CP: + *val = ccparam->inact_th; + break; default: return -EINVAL; } @@ -203,7 +206,7 @@ static ssize_t bnxt_re_cc_config_get(struct file *filp, char __user *buffer, return simple_read_from_buffer(buffer, usr_buf_len, ppos, (u8 *)(buf), rc); } -static void bnxt_re_fill_gen0_ext0(struct bnxt_qplib_cc_param *ccparam, u32 offset, u32 val) +static int bnxt_re_fill_gen0_ext0(struct bnxt_qplib_cc_param *ccparam, u32 offset, u32 val) { u32 modify_mask; @@ -247,7 +250,9 @@ static void bnxt_re_fill_gen0_ext0(struct bnxt_qplib_cc_param *ccparam, u32 offs ccparam->tcp_cp = val; break; case CMDQ_MODIFY_ROCE_CC_MODIFY_MASK_TX_QUEUE: + return -EOPNOTSUPP; case CMDQ_MODIFY_ROCE_CC_MODIFY_MASK_INACTIVITY_CP: + ccparam->inact_th = val; break; case CMDQ_MODIFY_ROCE_CC_MODIFY_MASK_TIME_PER_PHASE: ccparam->time_pph = val; @@ -258,17 +263,20 @@ static void bnxt_re_fill_gen0_ext0(struct bnxt_qplib_cc_param *ccparam, u32 offs } ccparam->mask = modify_mask; + return 0; } static int bnxt_re_configure_cc(struct bnxt_re_dev *rdev, u32 gen_ext, u32 offset, u32 val) { struct bnxt_qplib_cc_param ccparam = { }; + int rc; - /* Supporting only Gen 0 now */ - if (gen_ext == CC_CONFIG_GEN0_EXT0) - bnxt_re_fill_gen0_ext0(&ccparam, offset, val); - else - return -EINVAL; + if (gen_ext != CC_CONFIG_GEN0_EXT0) + return -EOPNOTSUPP; + + rc = bnxt_re_fill_gen0_ext0(&ccparam, offset, val); + if (rc) + return rc; bnxt_qplib_modify_cc(&rdev->qplib_res, &ccparam); return 0; diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index 9082b3fd2b47..063801384b2b 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -1774,10 +1774,7 @@ int bnxt_re_destroy_srq(struct ib_srq *ib_srq, struct ib_udata *udata) ib_srq); struct bnxt_re_dev *rdev = srq->rdev; struct bnxt_qplib_srq *qplib_srq = &srq->qplib_srq; - struct bnxt_qplib_nq *nq = NULL; - if (qplib_srq->cq) - nq = qplib_srq->cq->nq; if (rdev->chip_ctx->modes.toggle_bits & BNXT_QPLIB_SRQ_TOGGLE_BIT) { free_page((unsigned long)srq->uctx_srq_page); hash_del(&srq->hash_entry); @@ -1785,8 +1782,6 @@ int bnxt_re_destroy_srq(struct ib_srq *ib_srq, struct ib_udata *udata) bnxt_qplib_destroy_srq(&rdev->qplib_res, qplib_srq); ib_umem_release(srq->umem); atomic_dec(&rdev->stats.res.srq_count); - if (nq) - nq->budget--; return 0; } @@ -1827,7 +1822,6 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq, struct ib_udata *udata) { struct bnxt_qplib_dev_attr *dev_attr; - struct bnxt_qplib_nq *nq = NULL; struct bnxt_re_ucontext *uctx; struct bnxt_re_dev *rdev; struct bnxt_re_srq *srq; @@ -1873,7 +1867,6 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq, srq->qplib_srq.eventq_hw_ring_id = rdev->nqr->nq[0].ring_id; srq->qplib_srq.sg_info.pgsize = PAGE_SIZE; srq->qplib_srq.sg_info.pgshft = PAGE_SHIFT; - nq = &rdev->nqr->nq[0]; if (udata) { rc = bnxt_re_init_user_srq(rdev, pd, srq, udata); @@ -1908,8 +1901,6 @@ int bnxt_re_create_srq(struct ib_srq *ib_srq, goto fail; } } - if (nq) - nq->budget++; active_srqs = atomic_inc_return(&rdev->stats.res.srq_count); if (active_srqs > rdev->stats.res.srq_watermark) rdev->stats.res.srq_watermark = active_srqs; @@ -3079,7 +3070,6 @@ int bnxt_re_destroy_cq(struct ib_cq *ib_cq, struct ib_udata *udata) ib_umem_release(cq->umem); atomic_dec(&rdev->stats.res.cq_count); - nq->budget--; kfree(cq->cql); return 0; } diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c index 457eecb99f96..be34c605d516 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c @@ -1113,7 +1113,7 @@ int bnxt_qplib_create_qp(struct bnxt_qplib_res *res, struct bnxt_qplib_qp *qp) qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_FORCE_COMPLETION; if (qp->wqe_mode == BNXT_QPLIB_WQE_MODE_VARIABLE) qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_VARIABLE_SIZED_WQE_ENABLED; - if (_is_ext_stats_supported(res->dattr->dev_cap_flags) && !res->is_vf) + if (bnxt_ext_stats_supported(res->cctx, res->dattr->dev_cap_flags, res->is_vf)) qp_flags |= CMDQ_CREATE_QP_QP_FLAGS_EXT_STATS_ENABLED; req.qp_flags = cpu_to_le32(qp_flags); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c index f231e886ad9d..9efd32a3dc55 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c @@ -846,7 +846,12 @@ int bnxt_qplib_qext_stat(struct bnxt_qplib_rcfw *rcfw, u32 fid, req.resp_size = sbuf.size / BNXT_QPLIB_CMDQE_UNITS; req.resp_addr = cpu_to_le64(sbuf.dma_addr); - req.function_id = cpu_to_le32(fid); + if (bnxt_qplib_is_chip_gen_p7(rcfw->res->cctx) && rcfw->res->is_vf) + req.function_id = + cpu_to_le32(CMDQ_QUERY_ROCE_STATS_EXT_VF_VALID | + (fid << CMDQ_QUERY_ROCE_STATS_EXT_VF_NUM_SFT)); + else + req.function_id = cpu_to_le32(fid); req.flags = cpu_to_le16(CMDQ_QUERY_ROCE_STATS_EXT_FLAGS_FUNCTION_ID); bnxt_qplib_fill_cmdqmsg(&msg, &req, &resp, &sbuf, sizeof(req), diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index 8d753e6e0c71..e02721a9e288 100644 --- a/drivers/infiniband/hw/cxgb4/cm.c +++ b/drivers/infiniband/hw/cxgb4/cm.c @@ -191,7 +191,7 @@ static void start_ep_timer(struct c4iw_ep *ep) static int stop_ep_timer(struct c4iw_ep *ep) { pr_debug("ep %p stopping\n", ep); - del_timer_sync(&ep->timer); + timer_delete_sync(&ep->timer); if (!test_and_set_bit(TIMEOUT, &ep->com.flags)) { c4iw_put_ep(&ep->com); return 0; diff --git a/drivers/infiniband/hw/hfi1/aspm.c b/drivers/infiniband/hw/hfi1/aspm.c index a3c53be4072c..9b508eaf441d 100644 --- a/drivers/infiniband/hw/hfi1/aspm.c +++ b/drivers/infiniband/hw/hfi1/aspm.c @@ -191,7 +191,7 @@ void aspm_disable_all(struct hfi1_devdata *dd) for (i = 0; i < dd->first_dyn_alloc_ctxt; i++) { rcd = hfi1_rcd_get_by_index(dd, i); if (rcd) { - del_timer_sync(&rcd->aspm_timer); + timer_delete_sync(&rcd->aspm_timer); spin_lock_irqsave(&rcd->aspm_lock, flags); rcd->aspm_intr_enable = false; spin_unlock_irqrestore(&rcd->aspm_lock, flags); diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index 368b6be3226f..e908f529335d 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -5576,7 +5576,7 @@ static int init_rcverr(struct hfi1_devdata *dd) static void free_rcverr(struct hfi1_devdata *dd) { if (dd->rcverr_timer.function) - del_timer_sync(&dd->rcverr_timer); + timer_delete_sync(&dd->rcverr_timer); } static void handle_rxe_err(struct hfi1_devdata *dd, u32 unused, u64 reg) @@ -12308,7 +12308,7 @@ static void free_cntrs(struct hfi1_devdata *dd) int i; if (dd->synth_stats_timer.function) - del_timer_sync(&dd->synth_stats_timer); + timer_delete_sync(&dd->synth_stats_timer); cancel_work_sync(&dd->update_cntr_work); ppd = (struct hfi1_pportdata *)(dd + 1); for (i = 0; i < dd->num_pports; i++, ppd++) { diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c index 50826e7cdb7e..3da90f2eb8e7 100644 --- a/drivers/infiniband/hw/hfi1/driver.c +++ b/drivers/infiniband/hw/hfi1/driver.c @@ -1303,7 +1303,7 @@ void shutdown_led_override(struct hfi1_pportdata *ppd) */ smp_rmb(); if (atomic_read(&ppd->led_override_timer_active)) { - del_timer_sync(&ppd->led_override_timer); + timer_delete_sync(&ppd->led_override_timer); atomic_set(&ppd->led_override_timer_active, 0); /* Ensure the atomic_set is visible to all CPUs */ smp_wmb(); diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index d6fbd9c2b8b4..b35f92e7d865 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -985,7 +985,7 @@ static void stop_timers(struct hfi1_devdata *dd) for (pidx = 0; pidx < dd->num_pports; ++pidx) { ppd = dd->pport + pidx; if (ppd->led_override_timer.function) { - del_timer_sync(&ppd->led_override_timer); + timer_delete_sync(&ppd->led_override_timer); atomic_set(&ppd->led_override_timer_active, 0); } } diff --git a/drivers/infiniband/hw/hfi1/mad.h b/drivers/infiniband/hw/hfi1/mad.h index b6e3141253c4..d6dde762921a 100644 --- a/drivers/infiniband/hw/hfi1/mad.h +++ b/drivers/infiniband/hw/hfi1/mad.h @@ -124,7 +124,6 @@ struct opa_mad_notice_attr { } __packed ntc_2048; }; - u8 class_data[]; }; #define IB_VLARB_LOWPRI_0_31 1 diff --git a/drivers/infiniband/hw/hfi1/pio.c b/drivers/infiniband/hw/hfi1/pio.c index 5a91cbda4aee..764286da2ce8 100644 --- a/drivers/infiniband/hw/hfi1/pio.c +++ b/drivers/infiniband/hw/hfi1/pio.c @@ -1361,16 +1361,6 @@ void sc_flush(struct send_context *sc) sc_wait_for_packet_egress(sc, 1); } -/* drop all packets on the context, no waiting until they are sent */ -void sc_drop(struct send_context *sc) -{ - if (!sc) - return; - - dd_dev_info(sc->dd, "%s: context %u(%u) - not implemented\n", - __func__, sc->sw_index, sc->hw_context); -} - /* * Start the software reaction to a context halt or SPC freeze: * - mark the context as halted or frozen diff --git a/drivers/infiniband/hw/hfi1/pio.h b/drivers/infiniband/hw/hfi1/pio.h index d07cc6ea7c63..ab0f9a3a8d12 100644 --- a/drivers/infiniband/hw/hfi1/pio.h +++ b/drivers/infiniband/hw/hfi1/pio.h @@ -246,7 +246,6 @@ void sc_disable(struct send_context *sc); int sc_restart(struct send_context *sc); void sc_return_credits(struct send_context *sc); void sc_flush(struct send_context *sc); -void sc_drop(struct send_context *sc); void sc_stop(struct send_context *sc, int bit); struct pio_buf *sc_buffer_alloc(struct send_context *sc, u32 dw_len, pio_release_cb cb, void *arg); diff --git a/drivers/infiniband/hw/hfi1/sdma.c b/drivers/infiniband/hw/hfi1/sdma.c index b67d23b1f286..16a749d16ee9 100644 --- a/drivers/infiniband/hw/hfi1/sdma.c +++ b/drivers/infiniband/hw/hfi1/sdma.c @@ -1521,24 +1521,6 @@ void sdma_all_running(struct hfi1_devdata *dd) } /** - * sdma_all_idle() - called when the link goes down - * @dd: hfi1_devdata - * - * This routine moves all engines to the idle state. - */ -void sdma_all_idle(struct hfi1_devdata *dd) -{ - struct sdma_engine *sde; - unsigned int i; - - /* idle all engines */ - for (i = 0; i < dd->num_sdma; ++i) { - sde = &dd->per_sdma[i]; - sdma_process_event(sde, sdma_event_e70_go_idle); - } -} - -/** * sdma_start() - called to kick off state processing for all engines * @dd: hfi1_devdata * @@ -1575,7 +1557,7 @@ void sdma_exit(struct hfi1_devdata *dd) sde->this_idx); sdma_process_event(sde, sdma_event_e00_go_hw_down); - del_timer_sync(&sde->err_progress_check_timer); + timer_delete_sync(&sde->err_progress_check_timer); /* * This waits for the state machine to exit so it is not diff --git a/drivers/infiniband/hw/hfi1/sdma.h b/drivers/infiniband/hw/hfi1/sdma.h index d77246b48434..91dfd5d0c419 100644 --- a/drivers/infiniband/hw/hfi1/sdma.h +++ b/drivers/infiniband/hw/hfi1/sdma.h @@ -373,7 +373,6 @@ void sdma_start(struct hfi1_devdata *dd); void sdma_exit(struct hfi1_devdata *dd); void sdma_clean(struct hfi1_devdata *dd, size_t num_engines); void sdma_all_running(struct hfi1_devdata *dd); -void sdma_all_idle(struct hfi1_devdata *dd); void sdma_freeze_notify(struct hfi1_devdata *dd, int go_idle); void sdma_freeze(struct hfi1_devdata *dd); void sdma_unfreeze(struct hfi1_devdata *dd); diff --git a/drivers/infiniband/hw/hfi1/tid_rdma.c b/drivers/infiniband/hw/hfi1/tid_rdma.c index c465966a1d9c..78bf4a48c035 100644 --- a/drivers/infiniband/hw/hfi1/tid_rdma.c +++ b/drivers/infiniband/hw/hfi1/tid_rdma.c @@ -3965,7 +3965,7 @@ static int hfi1_stop_tid_reap_timer(struct rvt_qp *qp) lockdep_assert_held(&qp->s_lock); if (qpriv->s_flags & HFI1_R_TID_RSC_TIMER) { - rval = del_timer(&qpriv->s_tid_timer); + rval = timer_delete(&qpriv->s_tid_timer); qpriv->s_flags &= ~HFI1_R_TID_RSC_TIMER; } return rval; @@ -3975,7 +3975,7 @@ void hfi1_del_tid_reap_timer(struct rvt_qp *qp) { struct hfi1_qp_priv *qpriv = qp->priv; - del_timer_sync(&qpriv->s_tid_timer); + timer_delete_sync(&qpriv->s_tid_timer); qpriv->s_flags &= ~HFI1_R_TID_RSC_TIMER; } @@ -4781,7 +4781,7 @@ static int hfi1_stop_tid_retry_timer(struct rvt_qp *qp) lockdep_assert_held(&qp->s_lock); if (priv->s_flags & HFI1_S_TID_RETRY_TIMER) { - rval = del_timer(&priv->s_tid_retry_timer); + rval = timer_delete(&priv->s_tid_retry_timer); priv->s_flags &= ~HFI1_S_TID_RETRY_TIMER; } return rval; @@ -4791,7 +4791,7 @@ void hfi1_del_tid_retry_timer(struct rvt_qp *qp) { struct hfi1_qp_priv *priv = qp->priv; - del_timer_sync(&priv->s_tid_retry_timer); + timer_delete_sync(&priv->s_tid_retry_timer); priv->s_flags &= ~HFI1_S_TID_RETRY_TIMER; } diff --git a/drivers/infiniband/hw/hfi1/user_exp_rcv.c b/drivers/infiniband/hw/hfi1/user_exp_rcv.c index cf2d29098406..62b4f16dab27 100644 --- a/drivers/infiniband/hw/hfi1/user_exp_rcv.c +++ b/drivers/infiniband/hw/hfi1/user_exp_rcv.c @@ -53,7 +53,7 @@ int hfi1_user_exp_rcv_init(struct hfi1_filedata *fd, int ret = 0; fd->entry_to_rb = kcalloc(uctxt->expected_count, - sizeof(struct rb_node *), + sizeof(*fd->entry_to_rb), GFP_KERNEL); if (!fd->entry_to_rb) return -ENOMEM; diff --git a/drivers/infiniband/hw/hfi1/verbs.c b/drivers/infiniband/hw/hfi1/verbs.c index 33af2196ef31..49e0f79b950c 100644 --- a/drivers/infiniband/hw/hfi1/verbs.c +++ b/drivers/infiniband/hw/hfi1/verbs.c @@ -1900,7 +1900,7 @@ void hfi1_unregister_ib_device(struct hfi1_devdata *dd) if (!list_empty(&dev->memwait)) dd_dev_err(dd, "memwait list not empty!\n"); - del_timer_sync(&dev->mem_timer); + timer_delete_sync(&dev->mem_timer); verbs_txreq_exit(dev); kfree(dev_cntr_descs); diff --git a/drivers/infiniband/hw/hns/Makefile b/drivers/infiniband/hw/hns/Makefile index 7917af8e6380..baf592e6f21b 100644 --- a/drivers/infiniband/hw/hns/Makefile +++ b/drivers/infiniband/hw/hns/Makefile @@ -4,6 +4,7 @@ # ccflags-y := -I $(srctree)/drivers/net/ethernet/hisilicon/hns3 +ccflags-y += -I $(src) hns-roce-hw-v2-objs := hns_roce_main.o hns_roce_cmd.o hns_roce_pd.o \ hns_roce_ah.o hns_roce_hem.o hns_roce_mr.o hns_roce_qp.o \ diff --git a/drivers/infiniband/hw/hns/hns_roce_ah.c b/drivers/infiniband/hw/hns/hns_roce_ah.c index 4fc5b9d5fea8..307c35888b30 100644 --- a/drivers/infiniband/hw/hns/hns_roce_ah.c +++ b/drivers/infiniband/hw/hns/hns_roce_ah.c @@ -33,7 +33,6 @@ #include <linux/pci.h> #include <rdma/ib_addr.h> #include <rdma/ib_cache.h> -#include "hnae3.h" #include "hns_roce_device.h" #include "hns_roce_hw_v2.h" diff --git a/drivers/infiniband/hw/hns/hns_roce_device.h b/drivers/infiniband/hw/hns/hns_roce_device.h index 560a1d9de408..1dcc9cbb4678 100644 --- a/drivers/infiniband/hw/hns/hns_roce_device.h +++ b/drivers/infiniband/hw/hns/hns_roce_device.h @@ -1027,6 +1027,26 @@ struct hns_roce_dev { atomic64_t *dfx_cnt; }; +enum hns_roce_trace_type { + TRACE_SQ, + TRACE_RQ, + TRACE_SRQ, +}; + +static inline const char *trace_type_to_str(enum hns_roce_trace_type type) +{ + switch (type) { + case TRACE_SQ: + return "SQ"; + case TRACE_RQ: + return "RQ"; + case TRACE_SRQ: + return "SRQ"; + default: + return "UNKNOWN"; + } +} + static inline struct hns_roce_dev *to_hr_dev(struct ib_device *ib_dev) { return container_of(ib_dev, struct hns_roce_dev, ib_dev); diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 160e8927d364..fa8747656f25 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -43,13 +43,15 @@ #include <rdma/ib_umem.h> #include <rdma/uverbs_ioctl.h> -#include "hnae3.h" #include "hns_roce_common.h" #include "hns_roce_device.h" #include "hns_roce_cmd.h" #include "hns_roce_hem.h" #include "hns_roce_hw_v2.h" +#define CREATE_TRACE_POINTS +#include "hns_roce_trace.h" + enum { CMD_RST_PRC_OTHERS, CMD_RST_PRC_SUCCESS, @@ -738,6 +740,8 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, else ret = set_ud_wqe(qp, wr, wqe, &sge_idx, owner_bit); + trace_hns_sq_wqe(qp->qpn, wqe_idx, wqe, 1 << qp->sq.wqe_shift, + wr->wr_id, TRACE_SQ); if (unlikely(ret)) { *bad_wr = wr; goto out; @@ -807,6 +811,9 @@ static void fill_rq_wqe(struct hns_roce_qp *hr_qp, const struct ib_recv_wr *wr, wqe = hns_roce_get_recv_wqe(hr_qp, wqe_idx); fill_recv_sge_to_wqe(wr, wqe, max_sge, hr_qp->rq.rsv_sge); + + trace_hns_rq_wqe(hr_qp->qpn, wqe_idx, wqe, 1 << hr_qp->rq.wqe_shift, + wr->wr_id, TRACE_RQ); } static int hns_roce_v2_post_recv(struct ib_qp *ibqp, @@ -943,7 +950,7 @@ static void fill_wqe_idx(struct hns_roce_srq *srq, unsigned int wqe_idx) static void update_srq_db(struct hns_roce_srq *srq) { struct hns_roce_dev *hr_dev = to_hr_dev(srq->ibsrq.device); - struct hns_roce_v2_db db; + struct hns_roce_v2_db db = {}; hr_reg_write(&db, DB_TAG, srq->srqn); hr_reg_write(&db, DB_CMD, HNS_ROCE_V2_SRQ_DB); @@ -984,6 +991,9 @@ static int hns_roce_v2_post_srq_recv(struct ib_srq *ibsrq, fill_recv_sge_to_wqe(wr, wqe, max_sge, srq->rsv_sge); fill_wqe_idx(srq, wqe_idx); srq->wrid[wqe_idx] = wr->wr_id; + + trace_hns_srq_wqe(srq->srqn, wqe_idx, wqe, 1 << srq->wqe_shift, + wr->wr_id, TRACE_SRQ); } if (likely(nreq)) { @@ -1311,6 +1321,8 @@ static int __hns_roce_cmq_send_one(struct hns_roce_dev *hr_dev, tail = csq->head; for (i = 0; i < num; i++) { + trace_hns_cmdq_req(hr_dev, &desc[i]); + csq->desc[csq->head++] = desc[i]; if (csq->head == csq->desc_num) csq->head = 0; @@ -1325,6 +1337,8 @@ static int __hns_roce_cmq_send_one(struct hns_roce_dev *hr_dev, if (hns_roce_cmq_csq_done(hr_dev)) { ret = 0; for (i = 0; i < num; i++) { + trace_hns_cmdq_resp(hr_dev, &csq->desc[tail]); + /* check the result of hardware write back */ desc_ret = le16_to_cpu(csq->desc[tail++].retval); if (tail == csq->desc_num) @@ -4302,8 +4316,7 @@ static inline int get_pdn(struct ib_pd *ib_pd) } static void modify_qp_reset_to_init(struct ib_qp *ibqp, - struct hns_roce_v2_qp_context *context, - struct hns_roce_v2_qp_context *qpc_mask) + struct hns_roce_v2_qp_context *context) { struct hns_roce_dev *hr_dev = to_hr_dev(ibqp->device); struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); @@ -5122,7 +5135,7 @@ static int hns_roce_v2_set_abs_fields(struct ib_qp *ibqp, if (cur_state == IB_QPS_RESET && new_state == IB_QPS_INIT) { memset(qpc_mask, 0, hr_dev->caps.qpc_sz); - modify_qp_reset_to_init(ibqp, context, qpc_mask); + modify_qp_reset_to_init(ibqp, context); } else if (cur_state == IB_QPS_INIT && new_state == IB_QPS_INIT) { modify_qp_init_to_init(ibqp, context, qpc_mask); } else if (cur_state == IB_QPS_INIT && new_state == IB_QPS_RTR) { @@ -5313,6 +5326,7 @@ static void v2_set_flushed_fields(struct ib_qp *ibqp, return; spin_lock_irqsave(&hr_qp->sq.lock, sq_flag); + trace_hns_sq_flush_cqe(hr_qp->qpn, hr_qp->sq.head, TRACE_SQ); hr_reg_write(context, QPC_SQ_PRODUCER_IDX, hr_qp->sq.head); hr_reg_clear(qpc_mask, QPC_SQ_PRODUCER_IDX); hr_qp->state = IB_QPS_ERR; @@ -5322,6 +5336,7 @@ static void v2_set_flushed_fields(struct ib_qp *ibqp, return; spin_lock_irqsave(&hr_qp->rq.lock, rq_flag); + trace_hns_rq_flush_cqe(hr_qp->qpn, hr_qp->rq.head, TRACE_RQ); hr_reg_write(context, QPC_RQ_PRODUCER_IDX, hr_qp->rq.head); hr_reg_clear(qpc_mask, QPC_RQ_PRODUCER_IDX); spin_unlock_irqrestore(&hr_qp->rq.lock, rq_flag); @@ -6248,6 +6263,7 @@ static irqreturn_t hns_roce_v2_aeq_int(struct hns_roce_dev *hr_dev, eq->sub_type = sub_type; ++eq->cons_index; aeqe_found = IRQ_HANDLED; + trace_hns_ae_info(event_type, aeqe, eq->eqe_size); atomic64_inc(&hr_dev->dfx_cnt[HNS_ROCE_DFX_AEQE_CNT]); diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h index 91a5665465ff..bc7466830eaf 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.h +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.h @@ -34,6 +34,7 @@ #define _HNS_ROCE_HW_V2_H #include <linux/bitops.h> +#include "hnae3.h" #define HNS_ROCE_V2_MAX_RC_INL_INN_SZ 32 #define HNS_ROCE_V2_MTT_ENTRY_SZ 64 diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index cf89a8db4f64..e7a497cc125c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -37,7 +37,6 @@ #include <rdma/ib_smi.h> #include <rdma/ib_user_verbs.h> #include <rdma/ib_cache.h> -#include "hnae3.h" #include "hns_roce_common.h" #include "hns_roce_device.h" #include "hns_roce_hem.h" @@ -763,7 +762,7 @@ static int hns_roce_register_device(struct hns_roce_dev *hr_dev) if (ret) return ret; } - dma_set_max_seg_size(dev, UINT_MAX); + dma_set_max_seg_size(dev, SZ_2G); ret = ib_register_device(ib_dev, "hns_%d", dev); if (ret) { dev_err(dev, "ib_register_device failed!\n"); diff --git a/drivers/infiniband/hw/hns/hns_roce_mr.c b/drivers/infiniband/hw/hns/hns_roce_mr.c index 09da3496843b..93a48b41955b 100644 --- a/drivers/infiniband/hw/hns/hns_roce_mr.c +++ b/drivers/infiniband/hw/hns/hns_roce_mr.c @@ -38,6 +38,7 @@ #include "hns_roce_device.h" #include "hns_roce_cmd.h" #include "hns_roce_hem.h" +#include "hns_roce_trace.h" static u32 hw_index_to_key(int ind) { @@ -159,6 +160,7 @@ static int hns_roce_mr_enable(struct hns_roce_dev *hr_dev, if (IS_ERR(mailbox)) return PTR_ERR(mailbox); + trace_hns_mr(mr); if (mr->type != MR_TYPE_FRMR) ret = hr_dev->hw->write_mtpt(hr_dev, mailbox->buf, mr); else @@ -1146,6 +1148,7 @@ int hns_roce_mtr_create(struct hns_roce_dev *hr_dev, struct hns_roce_mtr *mtr, struct ib_device *ibdev = &hr_dev->ib_dev; int ret; + trace_hns_buf_attr(buf_attr); /* The caller has its own buffer list and invokes the hns_roce_mtr_map() * to finish the MTT configuration. */ diff --git a/drivers/infiniband/hw/hns/hns_roce_restrack.c b/drivers/infiniband/hw/hns/hns_roce_restrack.c index 356d98816949..f637b73b946e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_restrack.c +++ b/drivers/infiniband/hw/hns/hns_roce_restrack.c @@ -4,7 +4,6 @@ #include <rdma/rdma_cm.h> #include <rdma/restrack.h> #include <uapi/rdma/rdma_netlink.h> -#include "hnae3.h" #include "hns_roce_common.h" #include "hns_roce_device.h" #include "hns_roce_hw_v2.h" diff --git a/drivers/infiniband/hw/hns/hns_roce_trace.h b/drivers/infiniband/hw/hns/hns_roce_trace.h new file mode 100644 index 000000000000..59ceb591b3a1 --- /dev/null +++ b/drivers/infiniband/hw/hns/hns_roce_trace.h @@ -0,0 +1,216 @@ +/* SPDX-License-Identifier: GPL-2.0+ */ +/* + * Copyright (c) 2025 Hisilicon Limited. + */ + +#undef TRACE_SYSTEM +#define TRACE_SYSTEM hns_roce + +#if !defined(__HNS_ROCE_TRACE_H) || defined(TRACE_HEADER_MULTI_READ) +#define __HNS_ROCE_TRACE_H + +#include <linux/tracepoint.h> +#include <linux/string_choices.h> +#include "hns_roce_device.h" +#include "hns_roce_hw_v2.h" + +DECLARE_EVENT_CLASS(flush_head_template, + TP_PROTO(unsigned long qpn, u32 pi, + enum hns_roce_trace_type type), + TP_ARGS(qpn, pi, type), + + TP_STRUCT__entry(__field(unsigned long, qpn) + __field(u32, pi) + __field(enum hns_roce_trace_type, type) + ), + + TP_fast_assign(__entry->qpn = qpn; + __entry->pi = pi; + __entry->type = type; + ), + + TP_printk("%s 0x%lx flush head 0x%x.", + trace_type_to_str(__entry->type), + __entry->qpn, __entry->pi) +); + +DEFINE_EVENT(flush_head_template, hns_sq_flush_cqe, + TP_PROTO(unsigned long qpn, u32 pi, + enum hns_roce_trace_type type), + TP_ARGS(qpn, pi, type)); +DEFINE_EVENT(flush_head_template, hns_rq_flush_cqe, + TP_PROTO(unsigned long qpn, u32 pi, + enum hns_roce_trace_type type), + TP_ARGS(qpn, pi, type)); + +#define MAX_SGE_PER_WQE 64 +#define MAX_WQE_SIZE (MAX_SGE_PER_WQE * HNS_ROCE_SGE_SIZE) +DECLARE_EVENT_CLASS(wqe_template, + TP_PROTO(unsigned long qpn, u32 idx, void *wqe, u32 len, + u64 id, enum hns_roce_trace_type type), + TP_ARGS(qpn, idx, wqe, len, id, type), + + TP_STRUCT__entry(__field(unsigned long, qpn) + __field(u32, idx) + __array(u32, wqe, + MAX_WQE_SIZE / sizeof(__le32)) + __field(u32, len) + __field(u64, id) + __field(enum hns_roce_trace_type, type) + ), + + TP_fast_assign(__entry->qpn = qpn; + __entry->idx = idx; + __entry->id = id; + __entry->len = len / sizeof(__le32); + __entry->type = type; + for (int i = 0; i < __entry->len; i++) + __entry->wqe[i] = le32_to_cpu(((__le32 *)wqe)[i]); + ), + + TP_printk("%s 0x%lx wqe(0x%x/0x%llx): %s", + trace_type_to_str(__entry->type), + __entry->qpn, __entry->idx, __entry->id, + __print_array(__entry->wqe, __entry->len, + sizeof(__le32))) +); + +DEFINE_EVENT(wqe_template, hns_sq_wqe, + TP_PROTO(unsigned long qpn, u32 idx, void *wqe, u32 len, u64 id, + enum hns_roce_trace_type type), + TP_ARGS(qpn, idx, wqe, len, id, type)); +DEFINE_EVENT(wqe_template, hns_rq_wqe, + TP_PROTO(unsigned long qpn, u32 idx, void *wqe, u32 len, u64 id, + enum hns_roce_trace_type type), + TP_ARGS(qpn, idx, wqe, len, id, type)); +DEFINE_EVENT(wqe_template, hns_srq_wqe, + TP_PROTO(unsigned long qpn, u32 idx, void *wqe, u32 len, u64 id, + enum hns_roce_trace_type type), + TP_ARGS(qpn, idx, wqe, len, id, type)); + +TRACE_EVENT(hns_ae_info, + TP_PROTO(int event_type, void *aeqe, unsigned int len), + TP_ARGS(event_type, aeqe, len), + + TP_STRUCT__entry(__field(int, event_type) + __array(u32, aeqe, + HNS_ROCE_V3_EQE_SIZE / sizeof(__le32)) + __field(u32, len) + ), + + TP_fast_assign(__entry->event_type = event_type; + __entry->len = len / sizeof(__le32); + for (int i = 0; i < __entry->len; i++) + __entry->aeqe[i] = le32_to_cpu(((__le32 *)aeqe)[i]); + ), + + TP_printk("event %2d aeqe: %s", __entry->event_type, + __print_array(__entry->aeqe, __entry->len, sizeof(__le32))) +); + +TRACE_EVENT(hns_mr, + TP_PROTO(struct hns_roce_mr *mr), + TP_ARGS(mr), + + TP_STRUCT__entry(__field(u64, iova) + __field(u64, size) + __field(u32, key) + __field(u32, pd) + __field(u32, pbl_hop_num) + __field(u32, npages) + __field(int, type) + __field(int, enabled) + ), + + TP_fast_assign(__entry->iova = mr->iova; + __entry->size = mr->size; + __entry->key = mr->key; + __entry->pd = mr->pd; + __entry->pbl_hop_num = mr->pbl_hop_num; + __entry->npages = mr->npages; + __entry->type = mr->type; + __entry->enabled = mr->enabled; + ), + + TP_printk("iova:0x%llx, size:%llu, key:%u, pd:%u, pbl_hop:%u, npages:%u, type:%d, status:%d", + __entry->iova, __entry->size, __entry->key, + __entry->pd, __entry->pbl_hop_num, __entry->npages, + __entry->type, __entry->enabled) +); + +TRACE_EVENT(hns_buf_attr, + TP_PROTO(struct hns_roce_buf_attr *attr), + TP_ARGS(attr), + + TP_STRUCT__entry(__field(unsigned int, region_count) + __field(unsigned int, region0_size) + __field(int, region0_hopnum) + __field(unsigned int, region1_size) + __field(int, region1_hopnum) + __field(unsigned int, region2_size) + __field(int, region2_hopnum) + __field(unsigned int, page_shift) + __field(bool, mtt_only) + ), + + TP_fast_assign(__entry->region_count = attr->region_count; + __entry->region0_size = attr->region[0].size; + __entry->region0_hopnum = attr->region[0].hopnum; + __entry->region1_size = attr->region[1].size; + __entry->region1_hopnum = attr->region[1].hopnum; + __entry->region2_size = attr->region[2].size; + __entry->region2_hopnum = attr->region[2].hopnum; + __entry->page_shift = attr->page_shift; + __entry->mtt_only = attr->mtt_only; + ), + + TP_printk("rg cnt:%u, pg_sft:0x%x, mtt_only:%s, rg 0 (sz:%u, hop:%u), rg 1 (sz:%u, hop:%u), rg 2 (sz:%u, hop:%u)\n", + __entry->region_count, __entry->page_shift, + str_yes_no(__entry->mtt_only), + __entry->region0_size, __entry->region0_hopnum, + __entry->region1_size, __entry->region1_hopnum, + __entry->region2_size, __entry->region2_hopnum) +); + +DECLARE_EVENT_CLASS(cmdq, + TP_PROTO(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc), + TP_ARGS(hr_dev, desc), + + TP_STRUCT__entry(__string(dev_name, dev_name(hr_dev->dev)) + __field(u16, opcode) + __field(u16, flag) + __field(u16, retval) + __array(u32, data, 6) + ), + + TP_fast_assign(__assign_str(dev_name); + __entry->opcode = le16_to_cpu(desc->opcode); + __entry->flag = le16_to_cpu(desc->flag); + __entry->retval = le16_to_cpu(desc->retval); + for (int i = 0; i < 6; i++) + __entry->data[i] = le32_to_cpu(desc->data[i]); + ), + + TP_printk("%s cmdq opcode:0x%x, flag:0x%x, retval:0x%x, data:%s\n", + __get_str(dev_name), __entry->opcode, + __entry->flag, __entry->retval, + __print_array(__entry->data, 6, sizeof(__le32))) +); + +DEFINE_EVENT(cmdq, hns_cmdq_req, + TP_PROTO(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc), + TP_ARGS(hr_dev, desc)); +DEFINE_EVENT(cmdq, hns_cmdq_resp, + TP_PROTO(struct hns_roce_dev *hr_dev, + struct hns_roce_cmq_desc *desc), + TP_ARGS(hr_dev, desc)); + +#endif /* __HNS_ROCE_TRACE_H */ + +#undef TRACE_INCLUDE_FILE +#define TRACE_INCLUDE_FILE hns_roce_trace +#undef TRACE_INCLUDE_PATH +#define TRACE_INCLUDE_PATH . +#include <trace/define_trace.h> diff --git a/drivers/infiniband/hw/irdma/cm.c b/drivers/infiniband/hw/irdma/cm.c index ce8d821bdad8..23207f13ac1b 100644 --- a/drivers/infiniband/hw/irdma/cm.c +++ b/drivers/infiniband/hw/irdma/cm.c @@ -3303,7 +3303,7 @@ void irdma_cleanup_cm_core(struct irdma_cm_core *cm_core) if (!cm_core) return; - del_timer_sync(&cm_core->tcp_timer); + timer_delete_sync(&cm_core->tcp_timer); destroy_workqueue(cm_core->event_wq); cm_core->dev->ws_reset(&cm_core->iwdev->vsi); diff --git a/drivers/infiniband/hw/irdma/ctrl.c b/drivers/infiniband/hw/irdma/ctrl.c index 6aed6169c07d..99a7f1a6c0b5 100644 --- a/drivers/infiniband/hw/irdma/ctrl.c +++ b/drivers/infiniband/hw/irdma/ctrl.c @@ -3131,7 +3131,7 @@ int irdma_sc_cqp_init(struct irdma_sc_cqp *cqp, writel(0, cqp->dev->hw_regs[IRDMA_CCQPSTATUS]); ibdev_dbg(to_ibdev(cqp->dev), - "WQE: sq_size[%04d] hw_sq_size[%04d] sq_base[%p] sq_pa[%pK] cqp[%p] polarity[x%04x]\n", + "WQE: sq_size[%04d] hw_sq_size[%04d] sq_base[%p] sq_pa[%p] cqp[%p] polarity[x%04x]\n", cqp->sq_size, cqp->hw_sq_size, cqp->sq_base, (u64 *)(uintptr_t)cqp->sq_pa, cqp, cqp->polarity); return 0; diff --git a/drivers/infiniband/hw/irdma/main.c b/drivers/infiniband/hw/irdma/main.c index 1ee8969595d3..1e840bbd619d 100644 --- a/drivers/infiniband/hw/irdma/main.c +++ b/drivers/infiniband/hw/irdma/main.c @@ -1,10 +1,8 @@ // SPDX-License-Identifier: GPL-2.0 OR Linux-OpenIB /* Copyright (c) 2015 - 2021 Intel Corporation */ #include "main.h" -#include "../../../net/ethernet/intel/ice/ice.h" MODULE_ALIAS("i40iw"); -MODULE_AUTHOR("Intel Corporation, <e1000-rdma@lists.sourceforge.net>"); MODULE_DESCRIPTION("Intel(R) Ethernet Protocol Driver for RDMA"); MODULE_LICENSE("Dual BSD/GPL"); @@ -61,7 +59,7 @@ static void irdma_log_invalid_mtu(u16 mtu, struct irdma_sc_dev *dev) } static void irdma_fill_qos_info(struct irdma_l2params *l2params, - struct iidc_qos_params *qos_info) + struct iidc_rdma_qos_params *qos_info) { int i; @@ -85,12 +83,13 @@ static void irdma_fill_qos_info(struct irdma_l2params *l2params, } } -static void irdma_iidc_event_handler(struct ice_pf *pf, struct iidc_event *event) +static void irdma_iidc_event_handler(struct iidc_rdma_core_dev_info *cdev_info, + struct iidc_rdma_event *event) { - struct irdma_device *iwdev = dev_get_drvdata(&pf->adev->dev); + struct irdma_device *iwdev = dev_get_drvdata(&cdev_info->adev->dev); struct irdma_l2params l2params = {}; - if (*event->type & BIT(IIDC_EVENT_AFTER_MTU_CHANGE)) { + if (*event->type & BIT(IIDC_RDMA_EVENT_AFTER_MTU_CHANGE)) { ibdev_dbg(&iwdev->ibdev, "CLNT: new MTU = %d\n", iwdev->netdev->mtu); if (iwdev->vsi.mtu != iwdev->netdev->mtu) { l2params.mtu = iwdev->netdev->mtu; @@ -98,25 +97,26 @@ static void irdma_iidc_event_handler(struct ice_pf *pf, struct iidc_event *event irdma_log_invalid_mtu(l2params.mtu, &iwdev->rf->sc_dev); irdma_change_l2params(&iwdev->vsi, &l2params); } - } else if (*event->type & BIT(IIDC_EVENT_BEFORE_TC_CHANGE)) { + } else if (*event->type & BIT(IIDC_RDMA_EVENT_BEFORE_TC_CHANGE)) { if (iwdev->vsi.tc_change_pending) return; irdma_prep_tc_change(iwdev); - } else if (*event->type & BIT(IIDC_EVENT_AFTER_TC_CHANGE)) { - struct iidc_qos_params qos_info = {}; + } else if (*event->type & BIT(IIDC_RDMA_EVENT_AFTER_TC_CHANGE)) { + struct iidc_rdma_priv_dev_info *iidc_priv = cdev_info->iidc_priv; if (!iwdev->vsi.tc_change_pending) return; l2params.tc_changed = true; ibdev_dbg(&iwdev->ibdev, "CLNT: TC Change\n"); - ice_get_qos_params(pf, &qos_info); - irdma_fill_qos_info(&l2params, &qos_info); + + irdma_fill_qos_info(&l2params, &iidc_priv->qos_info); if (iwdev->rf->protocol_used != IRDMA_IWARP_PROTOCOL_ONLY) - iwdev->dcb_vlan_mode = qos_info.num_tc > 1 && !l2params.dscp_mode; + iwdev->dcb_vlan_mode = + l2params.num_tc > 1 && !l2params.dscp_mode; irdma_change_l2params(&iwdev->vsi, &l2params); - } else if (*event->type & BIT(IIDC_EVENT_CRIT_ERR)) { + } else if (*event->type & BIT(IIDC_RDMA_EVENT_CRIT_ERR)) { ibdev_warn(&iwdev->ibdev, "ICE OICR event notification: oicr = 0x%08x\n", event->reg); if (event->reg & IRDMAPFINT_OICR_PE_CRITERR_M) { @@ -151,10 +151,8 @@ static void irdma_iidc_event_handler(struct ice_pf *pf, struct iidc_event *event */ static void irdma_request_reset(struct irdma_pci_f *rf) { - struct ice_pf *pf = rf->cdev; - ibdev_warn(&rf->iwdev->ibdev, "Requesting a reset\n"); - ice_rdma_request_reset(pf, IIDC_PFR); + ice_rdma_request_reset(rf->cdev, IIDC_FUNC_RESET); } /** @@ -166,14 +164,15 @@ static int irdma_lan_register_qset(struct irdma_sc_vsi *vsi, struct irdma_ws_node *tc_node) { struct irdma_device *iwdev = vsi->back_vsi; - struct ice_pf *pf = iwdev->rf->cdev; + struct iidc_rdma_core_dev_info *cdev_info; struct iidc_rdma_qset_params qset = {}; int ret; + cdev_info = iwdev->rf->cdev; qset.qs_handle = tc_node->qs_handle; qset.tc = tc_node->traffic_class; qset.vport_id = vsi->vsi_idx; - ret = ice_add_rdma_qset(pf, &qset); + ret = ice_add_rdma_qset(cdev_info, &qset); if (ret) { ibdev_dbg(&iwdev->ibdev, "WS: LAN alloc_res for rdma qset failed.\n"); return ret; @@ -194,19 +193,20 @@ static void irdma_lan_unregister_qset(struct irdma_sc_vsi *vsi, struct irdma_ws_node *tc_node) { struct irdma_device *iwdev = vsi->back_vsi; - struct ice_pf *pf = iwdev->rf->cdev; + struct iidc_rdma_core_dev_info *cdev_info; struct iidc_rdma_qset_params qset = {}; + cdev_info = iwdev->rf->cdev; qset.qs_handle = tc_node->qs_handle; qset.tc = tc_node->traffic_class; qset.vport_id = vsi->vsi_idx; qset.teid = tc_node->l2_sched_node_id; - if (ice_del_rdma_qset(pf, &qset)) + if (ice_del_rdma_qset(cdev_info, &qset)) ibdev_dbg(&iwdev->ibdev, "WS: LAN free_res for rdma qset failed.\n"); } -static int irdma_init_interrupts(struct irdma_pci_f *rf, struct ice_pf *pf) +static int irdma_init_interrupts(struct irdma_pci_f *rf, struct iidc_rdma_core_dev_info *cdev) { int i; @@ -217,12 +217,12 @@ static int irdma_init_interrupts(struct irdma_pci_f *rf, struct ice_pf *pf) return -ENOMEM; for (i = 0; i < rf->msix_count; i++) - if (ice_alloc_rdma_qvector(pf, &rf->msix_entries[i])) + if (ice_alloc_rdma_qvector(cdev, &rf->msix_entries[i])) break; if (i < IRDMA_MIN_MSIX) { - for (; i > 0; i--) - ice_free_rdma_qvector(pf, &rf->msix_entries[i]); + while (--i >= 0) + ice_free_rdma_qvector(cdev, &rf->msix_entries[i]); kfree(rf->msix_entries); return -ENOMEM; @@ -233,54 +233,65 @@ static int irdma_init_interrupts(struct irdma_pci_f *rf, struct ice_pf *pf) return 0; } -static void irdma_deinit_interrupts(struct irdma_pci_f *rf, struct ice_pf *pf) +static void irdma_deinit_interrupts(struct irdma_pci_f *rf, struct iidc_rdma_core_dev_info *cdev) { int i; for (i = 0; i < rf->msix_count; i++) - ice_free_rdma_qvector(pf, &rf->msix_entries[i]); + ice_free_rdma_qvector(cdev, &rf->msix_entries[i]); kfree(rf->msix_entries); } static void irdma_remove(struct auxiliary_device *aux_dev) { - struct iidc_auxiliary_dev *iidc_adev = container_of(aux_dev, - struct iidc_auxiliary_dev, - adev); - struct ice_pf *pf = iidc_adev->pf; struct irdma_device *iwdev = auxiliary_get_drvdata(aux_dev); + struct iidc_rdma_core_auxiliary_dev *iidc_adev; + struct iidc_rdma_core_dev_info *cdev_info; + + iidc_adev = container_of(aux_dev, struct iidc_rdma_core_auxiliary_dev, adev); + cdev_info = iidc_adev->cdev_info; + ice_rdma_update_vsi_filter(cdev_info, iwdev->vsi_num, false); irdma_ib_unregister_device(iwdev); - ice_rdma_update_vsi_filter(pf, iwdev->vsi_num, false); - irdma_deinit_interrupts(iwdev->rf, pf); + irdma_deinit_interrupts(iwdev->rf, cdev_info); - pr_debug("INIT: Gen2 PF[%d] device remove success\n", PCI_FUNC(pf->pdev->devfn)); + kfree(iwdev->rf); + + pr_debug("INIT: Gen2 PF[%d] device remove success\n", PCI_FUNC(cdev_info->pdev->devfn)); } -static void irdma_fill_device_info(struct irdma_device *iwdev, struct ice_pf *pf, - struct ice_vsi *vsi) +static void irdma_fill_device_info(struct irdma_device *iwdev, + struct iidc_rdma_core_dev_info *cdev_info) { + struct iidc_rdma_priv_dev_info *iidc_priv = cdev_info->iidc_priv; struct irdma_pci_f *rf = iwdev->rf; - rf->cdev = pf; + rf->sc_dev.hw = &rf->hw; + rf->iwdev = iwdev; + rf->cdev = cdev_info; + rf->hw.hw_addr = iidc_priv->hw_addr; + rf->pcidev = cdev_info->pdev; + rf->hw.device = &rf->pcidev->dev; + rf->pf_id = iidc_priv->pf_id; rf->gen_ops.register_qset = irdma_lan_register_qset; rf->gen_ops.unregister_qset = irdma_lan_unregister_qset; - rf->hw.hw_addr = pf->hw.hw_addr; - rf->pcidev = pf->pdev; - rf->pf_id = pf->hw.pf_id; - rf->default_vsi.vsi_idx = vsi->vsi_num; - rf->protocol_used = pf->rdma_mode & IIDC_RDMA_PROTOCOL_ROCEV2 ? - IRDMA_ROCE_PROTOCOL_ONLY : IRDMA_IWARP_PROTOCOL_ONLY; + + rf->default_vsi.vsi_idx = iidc_priv->vport_id; + rf->protocol_used = + cdev_info->rdma_protocol == IIDC_RDMA_PROTOCOL_ROCEV2 ? + IRDMA_ROCE_PROTOCOL_ONLY : IRDMA_IWARP_PROTOCOL_ONLY; rf->rdma_ver = IRDMA_GEN_2; rf->rsrc_profile = IRDMA_HMC_PROFILE_DEFAULT; rf->rst_to = IRDMA_RST_TIMEOUT_HZ; rf->gen_ops.request_reset = irdma_request_reset; rf->limits_sel = 7; rf->iwdev = iwdev; + mutex_init(&iwdev->ah_tbl_lock); - iwdev->netdev = vsi->netdev; - iwdev->vsi_num = vsi->vsi_num; + + iwdev->netdev = iidc_priv->netdev; + iwdev->vsi_num = iidc_priv->vport_id; iwdev->init_state = INITIAL_STATE; iwdev->roce_cwnd = IRDMA_ROCE_CWND_DEFAULT; iwdev->roce_ackcreds = IRDMA_ROCE_ACKCREDS_DEFAULT; @@ -292,19 +303,18 @@ static void irdma_fill_device_info(struct irdma_device *iwdev, struct ice_pf *pf static int irdma_probe(struct auxiliary_device *aux_dev, const struct auxiliary_device_id *id) { - struct iidc_auxiliary_dev *iidc_adev = container_of(aux_dev, - struct iidc_auxiliary_dev, - adev); - struct ice_pf *pf = iidc_adev->pf; - struct ice_vsi *vsi = ice_get_main_vsi(pf); - struct iidc_qos_params qos_info = {}; + struct iidc_rdma_core_auxiliary_dev *iidc_adev; + struct iidc_rdma_core_dev_info *cdev_info; + struct iidc_rdma_priv_dev_info *iidc_priv; + struct irdma_l2params l2params = {}; struct irdma_device *iwdev; struct irdma_pci_f *rf; - struct irdma_l2params l2params = {}; int err; - if (!vsi) - return -EIO; + iidc_adev = container_of(aux_dev, struct iidc_rdma_core_auxiliary_dev, adev); + cdev_info = iidc_adev->cdev_info; + iidc_priv = cdev_info->iidc_priv; + iwdev = ib_alloc_device(irdma_device, ibdev); if (!iwdev) return -ENOMEM; @@ -314,10 +324,10 @@ static int irdma_probe(struct auxiliary_device *aux_dev, const struct auxiliary_ return -ENOMEM; } - irdma_fill_device_info(iwdev, pf, vsi); + irdma_fill_device_info(iwdev, cdev_info); rf = iwdev->rf; - err = irdma_init_interrupts(rf, pf); + err = irdma_init_interrupts(rf, cdev_info); if (err) goto err_init_interrupts; @@ -326,8 +336,7 @@ static int irdma_probe(struct auxiliary_device *aux_dev, const struct auxiliary_ goto err_ctrl_init; l2params.mtu = iwdev->netdev->mtu; - ice_get_qos_params(pf, &qos_info); - irdma_fill_qos_info(&l2params, &qos_info); + irdma_fill_qos_info(&l2params, &iidc_priv->qos_info); if (iwdev->rf->protocol_used != IRDMA_IWARP_PROTOCOL_ONLY) iwdev->dcb_vlan_mode = l2params.num_tc > 1 && !l2params.dscp_mode; @@ -339,7 +348,7 @@ static int irdma_probe(struct auxiliary_device *aux_dev, const struct auxiliary_ if (err) goto err_ibreg; - ice_rdma_update_vsi_filter(pf, iwdev->vsi_num, true); + ice_rdma_update_vsi_filter(cdev_info, iwdev->vsi_num, true); ibdev_dbg(&iwdev->ibdev, "INIT: Gen2 PF[%d] device probe success\n", PCI_FUNC(rf->pcidev->devfn)); auxiliary_set_drvdata(aux_dev, iwdev); @@ -351,7 +360,7 @@ err_ibreg: err_rt_init: irdma_ctrl_deinit_hw(rf); err_ctrl_init: - irdma_deinit_interrupts(rf, pf); + irdma_deinit_interrupts(rf, cdev_info); err_init_interrupts: kfree(iwdev->rf); ib_dealloc_device(&iwdev->ibdev); @@ -367,7 +376,7 @@ static const struct auxiliary_device_id irdma_auxiliary_id_table[] = { MODULE_DEVICE_TABLE(auxiliary, irdma_auxiliary_id_table); -static struct iidc_auxiliary_drv irdma_auxiliary_drv = { +static struct iidc_rdma_core_auxiliary_drv irdma_auxiliary_drv = { .adrv = { .id_table = irdma_auxiliary_id_table, .probe = irdma_probe, diff --git a/drivers/infiniband/hw/irdma/main.h b/drivers/infiniband/hw/irdma/main.h index bb0b6494ccb2..674acc952168 100644 --- a/drivers/infiniband/hw/irdma/main.h +++ b/drivers/infiniband/hw/irdma/main.h @@ -29,7 +29,8 @@ #include <linux/io-64-nonatomic-lo-hi.h> #endif #include <linux/auxiliary_bus.h> -#include <linux/net/intel/iidc.h> +#include <linux/net/intel/iidc_rdma.h> +#include <linux/net/intel/iidc_rdma_ice.h> #include <rdma/ib_smi.h> #include <rdma/ib_verbs.h> #include <rdma/ib_pack.h> diff --git a/drivers/infiniband/hw/irdma/osdep.h b/drivers/infiniband/hw/irdma/osdep.h index 4b4f78288d12..3f73ceacccb6 100644 --- a/drivers/infiniband/hw/irdma/osdep.h +++ b/drivers/infiniband/hw/irdma/osdep.h @@ -5,8 +5,8 @@ #include <linux/pci.h> #include <linux/bitfield.h> -#include <linux/net/intel/iidc.h> #include <rdma/ib_verbs.h> +#include <net/dscp.h> #define STATS_TIMER_DELAY 60000 diff --git a/drivers/infiniband/hw/irdma/pble.c b/drivers/infiniband/hw/irdma/pble.c index e7ce6840755f..37ce35cb10e7 100644 --- a/drivers/infiniband/hw/irdma/pble.c +++ b/drivers/infiniband/hw/irdma/pble.c @@ -108,7 +108,7 @@ static int add_sd_direct(struct irdma_hmc_pble_rsrc *pble_rsrc, chunk->vaddr = sd_entry->u.bp.addr.va + offset; chunk->fpm_addr = pble_rsrc->next_fpm_addr; ibdev_dbg(to_ibdev(dev), - "PBLE: chunk_size[%lld] = 0x%llx vaddr=0x%pK fpm_addr = %llx\n", + "PBLE: chunk_size[%lld] = 0x%llx vaddr=0x%p fpm_addr = %llx\n", chunk->size, chunk->size, chunk->vaddr, chunk->fpm_addr); return 0; diff --git a/drivers/infiniband/hw/irdma/type.h b/drivers/infiniband/hw/irdma/type.h index 59b34afa867b..527c6da2c1ac 100644 --- a/drivers/infiniband/hw/irdma/type.h +++ b/drivers/infiniband/hw/irdma/type.h @@ -567,7 +567,7 @@ struct irdma_sc_vsi { u8 qos_rel_bw; u8 qos_prio_type; u8 stats_idx; - u8 dscp_map[IIDC_MAX_DSCP_MAPPING]; + u8 dscp_map[DSCP_MAX]; struct irdma_qos qos[IRDMA_MAX_USER_PRIORITY]; u64 hw_stats_regs[IRDMA_HW_STAT_INDEX_MAX_GEN_1]; bool dscp_mode:1; @@ -695,7 +695,7 @@ struct irdma_l2params { u16 qs_handle_list[IRDMA_MAX_USER_PRIORITY]; u16 mtu; u8 up2tc[IRDMA_MAX_USER_PRIORITY]; - u8 dscp_map[IIDC_MAX_DSCP_MAPPING]; + u8 dscp_map[DSCP_MAX]; u8 num_tc; u8 vsi_rel_bw; u8 vsi_prio_type; diff --git a/drivers/infiniband/hw/irdma/utils.c b/drivers/infiniband/hw/irdma/utils.c index e73b14fd95ef..d66b4f7a84ec 100644 --- a/drivers/infiniband/hw/irdma/utils.c +++ b/drivers/infiniband/hw/irdma/utils.c @@ -963,7 +963,7 @@ void irdma_terminate_del_timer(struct irdma_sc_qp *qp) int ret; iwqp = qp->qp_uk.back_qp; - ret = del_timer(&iwqp->terminate_timer); + ret = timer_delete(&iwqp->terminate_timer); if (ret) irdma_qp_rem_ref(&iwqp->ibqp); } @@ -1570,7 +1570,7 @@ void irdma_hw_stats_stop_timer(struct irdma_sc_vsi *vsi) { struct irdma_vsi_pestat *devstat = vsi->pestat; - del_timer_sync(&devstat->stats_timer); + timer_delete_sync(&devstat->stats_timer); } /** diff --git a/drivers/infiniband/hw/irdma/verbs.c b/drivers/infiniband/hw/irdma/verbs.c index eeb932e58730..1e8c92826de2 100644 --- a/drivers/infiniband/hw/irdma/verbs.c +++ b/drivers/infiniband/hw/irdma/verbs.c @@ -4871,5 +4871,4 @@ void irdma_ib_dealloc_device(struct ib_device *ibdev) irdma_rt_deinit_hw(iwdev); irdma_ctrl_deinit_hw(iwdev->rf); - kfree(iwdev->rf); } diff --git a/drivers/infiniband/hw/mana/cq.c b/drivers/infiniband/hw/mana/cq.c index 0fc4e2679218..28e154bbb50f 100644 --- a/drivers/infiniband/hw/mana/cq.c +++ b/drivers/infiniband/hw/mana/cq.c @@ -15,14 +15,12 @@ int mana_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr, struct ib_device *ibdev = ibcq->device; struct mana_ib_create_cq ucmd = {}; struct mana_ib_dev *mdev; - struct gdma_context *gc; bool is_rnic_cq; u32 doorbell; u32 buf_size; int err; mdev = container_of(ibdev, struct mana_ib_dev, ib_dev); - gc = mdev_to_gc(mdev); cq->comp_vector = attr->comp_vector % ibdev->num_comp_vectors; cq->cq_handle = INVALID_MANA_HANDLE; @@ -65,7 +63,7 @@ int mana_ib_create_cq(struct ib_cq *ibcq, const struct ib_cq_init_attr *attr, ibdev_dbg(ibdev, "Failed to create kernel queue for create cq, %d\n", err); return err; } - doorbell = gc->mana_ib.doorbell; + doorbell = mdev->gdma_dev->doorbell; } if (is_rnic_cq) { diff --git a/drivers/infiniband/hw/mana/device.c b/drivers/infiniband/hw/mana/device.c index b31089320aa5..165c0a1e67d1 100644 --- a/drivers/infiniband/hw/mana/device.c +++ b/drivers/infiniband/hw/mana/device.c @@ -101,103 +101,95 @@ static int mana_ib_probe(struct auxiliary_device *adev, const struct auxiliary_device_id *id) { struct mana_adev *madev = container_of(adev, struct mana_adev, adev); + struct gdma_context *gc = madev->mdev->gdma_context; + struct mana_context *mc = gc->mana.driver_data; struct gdma_dev *mdev = madev->mdev; struct net_device *ndev; - struct mana_context *mc; struct mana_ib_dev *dev; u8 mac_addr[ETH_ALEN]; int ret; - mc = mdev->driver_data; - dev = ib_alloc_device(mana_ib_dev, ib_dev); if (!dev) return -ENOMEM; ib_set_device_ops(&dev->ib_dev, &mana_ib_dev_ops); - - dev->ib_dev.phys_port_cnt = mc->num_ports; - - ibdev_dbg(&dev->ib_dev, "mdev=%p id=%d num_ports=%d\n", mdev, - mdev->dev_id.as_uint32, dev->ib_dev.phys_port_cnt); - dev->ib_dev.node_type = RDMA_NODE_IB_CA; - - /* - * num_comp_vectors needs to set to the max MSIX index - * when interrupts and event queues are implemented - */ - dev->ib_dev.num_comp_vectors = mdev->gdma_context->max_num_queues; - dev->ib_dev.dev.parent = mdev->gdma_context->dev; - - ndev = mana_get_primary_netdev(mc, 0, &dev->dev_tracker); - if (!ndev) { - ret = -ENODEV; - ibdev_err(&dev->ib_dev, "Failed to get netdev for IB port 1"); - goto free_ib_device; - } - ether_addr_copy(mac_addr, ndev->dev_addr); - addrconf_addr_eui48((u8 *)&dev->ib_dev.node_guid, ndev->dev_addr); - ret = ib_device_set_netdev(&dev->ib_dev, ndev, 1); - /* mana_get_primary_netdev() returns ndev with refcount held */ - netdev_put(ndev, &dev->dev_tracker); - if (ret) { - ibdev_err(&dev->ib_dev, "Failed to set ib netdev, ret %d", ret); - goto free_ib_device; - } - - ret = mana_gd_register_device(&mdev->gdma_context->mana_ib); - if (ret) { - ibdev_err(&dev->ib_dev, "Failed to register device, ret %d", - ret); - goto free_ib_device; - } - dev->gdma_dev = &mdev->gdma_context->mana_ib; - - dev->nb.notifier_call = mana_ib_netdev_event; - ret = register_netdevice_notifier(&dev->nb); - if (ret) { - ibdev_err(&dev->ib_dev, "Failed to register net notifier, %d", - ret); - goto deregister_device; - } - - ret = mana_ib_gd_query_adapter_caps(dev); - if (ret) { - ibdev_err(&dev->ib_dev, "Failed to query device caps, ret %d", - ret); - goto deregister_net_notifier; - } - - ib_set_device_ops(&dev->ib_dev, &mana_ib_stats_ops); - - ret = mana_ib_create_eqs(dev); - if (ret) { - ibdev_err(&dev->ib_dev, "Failed to create EQs, ret %d", ret); - goto deregister_net_notifier; - } - - ret = mana_ib_gd_create_rnic_adapter(dev); - if (ret) - goto destroy_eqs; - + dev->ib_dev.num_comp_vectors = gc->max_num_queues; + dev->ib_dev.dev.parent = gc->dev; + dev->gdma_dev = mdev; xa_init_flags(&dev->qp_table_wq, XA_FLAGS_LOCK_IRQ); - ret = mana_ib_gd_config_mac(dev, ADDR_OP_ADD, mac_addr); - if (ret) { - ibdev_err(&dev->ib_dev, "Failed to add Mac address, ret %d", - ret); - goto destroy_rnic; + + if (mana_ib_is_rnic(dev)) { + dev->ib_dev.phys_port_cnt = 1; + ndev = mana_get_primary_netdev(mc, 0, &dev->dev_tracker); + if (!ndev) { + ret = -ENODEV; + ibdev_err(&dev->ib_dev, "Failed to get netdev for IB port 1"); + goto free_ib_device; + } + ether_addr_copy(mac_addr, ndev->dev_addr); + addrconf_addr_eui48((u8 *)&dev->ib_dev.node_guid, ndev->dev_addr); + ret = ib_device_set_netdev(&dev->ib_dev, ndev, 1); + /* mana_get_primary_netdev() returns ndev with refcount held */ + netdev_put(ndev, &dev->dev_tracker); + if (ret) { + ibdev_err(&dev->ib_dev, "Failed to set ib netdev, ret %d", ret); + goto free_ib_device; + } + + dev->nb.notifier_call = mana_ib_netdev_event; + ret = register_netdevice_notifier(&dev->nb); + if (ret) { + ibdev_err(&dev->ib_dev, "Failed to register net notifier, %d", + ret); + goto free_ib_device; + } + + ret = mana_ib_gd_query_adapter_caps(dev); + if (ret) { + ibdev_err(&dev->ib_dev, "Failed to query device caps, ret %d", ret); + goto deregister_net_notifier; + } + + ib_set_device_ops(&dev->ib_dev, &mana_ib_stats_ops); + + ret = mana_ib_create_eqs(dev); + if (ret) { + ibdev_err(&dev->ib_dev, "Failed to create EQs, ret %d", ret); + goto deregister_net_notifier; + } + + ret = mana_ib_gd_create_rnic_adapter(dev); + if (ret) + goto destroy_eqs; + + ret = mana_ib_gd_config_mac(dev, ADDR_OP_ADD, mac_addr); + if (ret) { + ibdev_err(&dev->ib_dev, "Failed to add Mac address, ret %d", ret); + goto destroy_rnic; + } + } else { + dev->ib_dev.phys_port_cnt = mc->num_ports; + ret = mana_eth_query_adapter_caps(dev); + if (ret) { + ibdev_err(&dev->ib_dev, "Failed to query ETH device caps, ret %d", ret); + goto free_ib_device; + } } - dev->av_pool = dma_pool_create("mana_ib_av", mdev->gdma_context->dev, - MANA_AV_BUFFER_SIZE, MANA_AV_BUFFER_SIZE, 0); + dev->av_pool = dma_pool_create("mana_ib_av", gc->dev, MANA_AV_BUFFER_SIZE, + MANA_AV_BUFFER_SIZE, 0); if (!dev->av_pool) { ret = -ENOMEM; goto destroy_rnic; } - ret = ib_register_device(&dev->ib_dev, "mana_%d", - mdev->gdma_context->dev); + ibdev_dbg(&dev->ib_dev, "mdev=%p id=%d num_ports=%d\n", mdev, + mdev->dev_id.as_uint32, dev->ib_dev.phys_port_cnt); + + ret = ib_register_device(&dev->ib_dev, mana_ib_is_rnic(dev) ? "mana_%d" : "manae_%d", + gc->dev); if (ret) goto deallocate_pool; @@ -208,15 +200,16 @@ static int mana_ib_probe(struct auxiliary_device *adev, deallocate_pool: dma_pool_destroy(dev->av_pool); destroy_rnic: - xa_destroy(&dev->qp_table_wq); - mana_ib_gd_destroy_rnic_adapter(dev); + if (mana_ib_is_rnic(dev)) + mana_ib_gd_destroy_rnic_adapter(dev); destroy_eqs: - mana_ib_destroy_eqs(dev); + if (mana_ib_is_rnic(dev)) + mana_ib_destroy_eqs(dev); deregister_net_notifier: - unregister_netdevice_notifier(&dev->nb); -deregister_device: - mana_gd_deregister_device(dev->gdma_dev); + if (mana_ib_is_rnic(dev)) + unregister_netdevice_notifier(&dev->nb); free_ib_device: + xa_destroy(&dev->qp_table_wq); ib_dealloc_device(&dev->ib_dev); return ret; } @@ -227,25 +220,24 @@ static void mana_ib_remove(struct auxiliary_device *adev) ib_unregister_device(&dev->ib_dev); dma_pool_destroy(dev->av_pool); + if (mana_ib_is_rnic(dev)) { + mana_ib_gd_destroy_rnic_adapter(dev); + mana_ib_destroy_eqs(dev); + unregister_netdevice_notifier(&dev->nb); + } xa_destroy(&dev->qp_table_wq); - mana_ib_gd_destroy_rnic_adapter(dev); - mana_ib_destroy_eqs(dev); - unregister_netdevice_notifier(&dev->nb); - mana_gd_deregister_device(dev->gdma_dev); ib_dealloc_device(&dev->ib_dev); } static const struct auxiliary_device_id mana_id_table[] = { - { - .name = "mana.rdma", - }, + { .name = "mana.rdma", }, + { .name = "mana.eth", }, {}, }; MODULE_DEVICE_TABLE(auxiliary, mana_id_table); static struct auxiliary_driver mana_driver = { - .name = "rdma", .probe = mana_ib_probe, .remove = mana_ib_remove, .id_table = mana_id_table, diff --git a/drivers/infiniband/hw/mana/main.c b/drivers/infiniband/hw/mana/main.c index eda9c5b971de..41a24a186f9d 100644 --- a/drivers/infiniband/hw/mana/main.c +++ b/drivers/infiniband/hw/mana/main.c @@ -4,6 +4,7 @@ */ #include "mana_ib.h" +#include "linux/pci.h" void mana_ib_uncfg_vport(struct mana_ib_dev *dev, struct mana_ib_pd *pd, u32 port) @@ -243,7 +244,6 @@ void mana_ib_dealloc_ucontext(struct ib_ucontext *ibcontext) int mana_ib_create_kernel_queue(struct mana_ib_dev *mdev, u32 size, enum gdma_queue_type type, struct mana_ib_queue *queue) { - struct gdma_context *gc = mdev_to_gc(mdev); struct gdma_queue_spec spec = {}; int err; @@ -252,7 +252,7 @@ int mana_ib_create_kernel_queue(struct mana_ib_dev *mdev, u32 size, enum gdma_qu spec.type = type; spec.monitor_avl_buf = false; spec.queue_size = size; - err = mana_gd_create_mana_wq_cq(&gc->mana_ib, &spec, &queue->kmem); + err = mana_gd_create_mana_wq_cq(mdev->gdma_dev, &spec, &queue->kmem); if (err) return err; /* take ownership into mana_ib from mana */ @@ -479,7 +479,7 @@ int mana_ib_create_dma_region(struct mana_ib_dev *dev, struct ib_umem *umem, { unsigned long page_sz; - page_sz = ib_umem_find_best_pgsz(umem, PAGE_SZ_BM, virt); + page_sz = ib_umem_find_best_pgsz(umem, dev->adapter_caps.page_size_cap, virt); if (!page_sz) { ibdev_dbg(&dev->ib_dev, "Failed to find page size.\n"); return -EINVAL; @@ -494,7 +494,7 @@ int mana_ib_create_zero_offset_dma_region(struct mana_ib_dev *dev, struct ib_ume unsigned long page_sz; /* Hardware requires dma region to align to chosen page size */ - page_sz = ib_umem_find_best_pgoff(umem, PAGE_SZ_BM, 0); + page_sz = ib_umem_find_best_pgoff(umem, dev->adapter_caps.page_size_cap, 0); if (!page_sz) { ibdev_dbg(&dev->ib_dev, "Failed to find page size.\n"); return -EINVAL; @@ -551,6 +551,7 @@ int mana_ib_mmap(struct ib_ucontext *ibcontext, struct vm_area_struct *vma) int mana_ib_get_port_immutable(struct ib_device *ibdev, u32 port_num, struct ib_port_immutable *immutable) { + struct mana_ib_dev *dev = container_of(ibdev, struct mana_ib_dev, ib_dev); struct ib_port_attr attr; int err; @@ -560,10 +561,12 @@ int mana_ib_get_port_immutable(struct ib_device *ibdev, u32 port_num, immutable->pkey_tbl_len = attr.pkey_tbl_len; immutable->gid_tbl_len = attr.gid_tbl_len; - immutable->core_cap_flags = RDMA_CORE_PORT_RAW_PACKET; - if (port_num == 1) { - immutable->core_cap_flags |= RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; + + if (mana_ib_is_rnic(dev)) { + immutable->core_cap_flags = RDMA_CORE_PORT_IBA_ROCE_UDP_ENCAP; immutable->max_mad_size = IB_MGMT_MAD_SIZE; + } else { + immutable->core_cap_flags = RDMA_CORE_PORT_RAW_PACKET; } return 0; @@ -572,12 +575,14 @@ int mana_ib_get_port_immutable(struct ib_device *ibdev, u32 port_num, int mana_ib_query_device(struct ib_device *ibdev, struct ib_device_attr *props, struct ib_udata *uhw) { - struct mana_ib_dev *dev = container_of(ibdev, - struct mana_ib_dev, ib_dev); + struct mana_ib_dev *dev = container_of(ibdev, struct mana_ib_dev, ib_dev); + struct pci_dev *pdev = to_pci_dev(mdev_to_gc(dev)->dev); memset(props, 0, sizeof(*props)); + props->vendor_id = pdev->vendor; + props->vendor_part_id = dev->gdma_dev->dev_id.type; props->max_mr_size = MANA_IB_MAX_MR_SIZE; - props->page_size_cap = PAGE_SZ_BM; + props->page_size_cap = dev->adapter_caps.page_size_cap; props->max_qp = dev->adapter_caps.max_qp_count; props->max_qp_wr = dev->adapter_caps.max_qp_wr; props->device_cap_flags = IB_DEVICE_RC_RNR_NAK_GEN; @@ -596,6 +601,8 @@ int mana_ib_query_device(struct ib_device *ibdev, struct ib_device_attr *props, props->max_ah = INT_MAX; props->max_pkeys = 1; props->local_ca_ack_delay = MANA_CA_ACK_DELAY; + if (!mana_ib_is_rnic(dev)) + props->raw_packet_caps = IB_RAW_PACKET_CAP_IP_CSUM; return 0; } @@ -603,6 +610,7 @@ int mana_ib_query_device(struct ib_device *ibdev, struct ib_device_attr *props, int mana_ib_query_port(struct ib_device *ibdev, u32 port, struct ib_port_attr *props) { + struct mana_ib_dev *dev = container_of(ibdev, struct mana_ib_dev, ib_dev); struct net_device *ndev = mana_ib_get_netdev(ibdev, port); if (!ndev) @@ -623,7 +631,7 @@ int mana_ib_query_port(struct ib_device *ibdev, u32 port, props->active_width = IB_WIDTH_4X; props->active_speed = IB_SPEED_EDR; props->pkey_tbl_len = 1; - if (port == 1) { + if (mana_ib_is_rnic(dev)) { props->gid_tbl_len = 16; props->port_cap_flags = IB_PORT_CM_SUP; props->ip_gids = true; @@ -696,6 +704,41 @@ int mana_ib_gd_query_adapter_caps(struct mana_ib_dev *dev) caps->max_recv_sge_count = resp.max_recv_sge_count; caps->feature_flags = resp.feature_flags; + caps->page_size_cap = PAGE_SZ_BM; + if (mdev_to_gc(dev)->pf_cap_flags1 & GDMA_DRV_CAP_FLAG_1_GDMA_PAGES_4MB_1GB_2GB) + caps->page_size_cap |= (SZ_4M | SZ_1G | SZ_2G); + + return 0; +} + +int mana_eth_query_adapter_caps(struct mana_ib_dev *dev) +{ + struct mana_ib_adapter_caps *caps = &dev->adapter_caps; + struct gdma_query_max_resources_resp resp = {}; + struct gdma_general_req req = {}; + int err; + + mana_gd_init_req_hdr(&req.hdr, GDMA_QUERY_MAX_RESOURCES, + sizeof(req), sizeof(resp)); + + err = mana_gd_send_request(mdev_to_gc(dev), sizeof(req), &req, sizeof(resp), &resp); + if (err) { + ibdev_err(&dev->ib_dev, + "Failed to query adapter caps err %d", err); + return err; + } + + caps->max_qp_count = min_t(u32, resp.max_sq, resp.max_rq); + caps->max_cq_count = resp.max_cq; + caps->max_mr_count = resp.max_mst; + caps->max_pd_count = 0x6000; + caps->max_qp_wr = min_t(u32, + 0x100000 / GDMA_MAX_SQE_SIZE, + 0x100000 / GDMA_MAX_RQE_SIZE); + caps->max_send_sge_count = 30; + caps->max_recv_sge_count = 15; + caps->page_size_cap = PAGE_SZ_BM; + return 0; } @@ -740,7 +783,7 @@ int mana_ib_create_eqs(struct mana_ib_dev *mdev) spec.eq.log2_throttle_limit = LOG2_EQ_THROTTLE; spec.eq.msix_index = 0; - err = mana_gd_create_mana_eq(&gc->mana_ib, &spec, &mdev->fatal_err_eq); + err = mana_gd_create_mana_eq(mdev->gdma_dev, &spec, &mdev->fatal_err_eq); if (err) return err; @@ -791,7 +834,7 @@ int mana_ib_gd_create_rnic_adapter(struct mana_ib_dev *mdev) mana_gd_init_req_hdr(&req.hdr, MANA_IB_CREATE_ADAPTER, sizeof(req), sizeof(resp)); req.hdr.req.msg_version = GDMA_MESSAGE_V2; - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.notify_eq_id = mdev->fatal_err_eq->id; if (mdev->adapter_caps.feature_flags & MANA_IB_FEATURE_CLIENT_ERROR_CQE_SUPPORT) @@ -816,7 +859,7 @@ int mana_ib_gd_destroy_rnic_adapter(struct mana_ib_dev *mdev) gc = mdev_to_gc(mdev); mana_gd_init_req_hdr(&req.hdr, MANA_IB_DESTROY_ADAPTER, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp); @@ -843,7 +886,7 @@ int mana_ib_gd_add_gid(const struct ib_gid_attr *attr, void **context) } mana_gd_init_req_hdr(&req.hdr, MANA_IB_CONFIG_IP_ADDR, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.op = ADDR_OP_ADD; req.sgid_type = (ntype == RDMA_NETWORK_IPV6) ? SGID_TYPE_IPV6 : SGID_TYPE_IPV4; @@ -873,7 +916,7 @@ int mana_ib_gd_del_gid(const struct ib_gid_attr *attr, void **context) } mana_gd_init_req_hdr(&req.hdr, MANA_IB_CONFIG_IP_ADDR, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.op = ADDR_OP_REMOVE; req.sgid_type = (ntype == RDMA_NETWORK_IPV6) ? SGID_TYPE_IPV6 : SGID_TYPE_IPV4; @@ -896,7 +939,7 @@ int mana_ib_gd_config_mac(struct mana_ib_dev *mdev, enum mana_ib_addr_op op, u8 int err; mana_gd_init_req_hdr(&req.hdr, MANA_IB_CONFIG_MAC_ADDR, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.op = op; copy_in_reverse(req.mac_addr, mac, ETH_ALEN); @@ -917,8 +960,11 @@ int mana_ib_gd_create_cq(struct mana_ib_dev *mdev, struct mana_ib_cq *cq, u32 do struct mana_rnic_create_cq_req req = {}; int err; + if (!mdev->eqs) + return -EINVAL; + mana_gd_init_req_hdr(&req.hdr, MANA_IB_CREATE_CQ, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.gdma_region = cq->queue.gdma_region; req.eq_id = mdev->eqs[cq->comp_vector]->id; @@ -950,7 +996,7 @@ int mana_ib_gd_destroy_cq(struct mana_ib_dev *mdev, struct mana_ib_cq *cq) return 0; mana_gd_init_req_hdr(&req.hdr, MANA_IB_DESTROY_CQ, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.cq_handle = cq->cq_handle; @@ -976,7 +1022,7 @@ int mana_ib_gd_create_rc_qp(struct mana_ib_dev *mdev, struct mana_ib_qp *qp, int err, i; mana_gd_init_req_hdr(&req.hdr, MANA_IB_CREATE_RC_QP, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.pd_handle = pd->pd_handle; req.send_cq_handle = send_cq->cq_handle; @@ -1012,7 +1058,7 @@ int mana_ib_gd_destroy_rc_qp(struct mana_ib_dev *mdev, struct mana_ib_qp *qp) int err; mana_gd_init_req_hdr(&req.hdr, MANA_IB_DESTROY_RC_QP, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.rc_qp_handle = qp->qp_handle; err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp); @@ -1035,7 +1081,7 @@ int mana_ib_gd_create_ud_qp(struct mana_ib_dev *mdev, struct mana_ib_qp *qp, int err, i; mana_gd_init_req_hdr(&req.hdr, MANA_IB_CREATE_UD_QP, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.pd_handle = pd->pd_handle; req.send_cq_handle = send_cq->cq_handle; @@ -1070,7 +1116,7 @@ int mana_ib_gd_destroy_ud_qp(struct mana_ib_dev *mdev, struct mana_ib_qp *qp) int err; mana_gd_init_req_hdr(&req.hdr, MANA_IB_DESTROY_UD_QP, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.qp_handle = qp->qp_handle; err = mana_gd_send_request(gc, sizeof(req), &req, sizeof(resp), &resp); diff --git a/drivers/infiniband/hw/mana/mana_ib.h b/drivers/infiniband/hw/mana/mana_ib.h index 6903946677e5..42bebd6cd4f7 100644 --- a/drivers/infiniband/hw/mana/mana_ib.h +++ b/drivers/infiniband/hw/mana/mana_ib.h @@ -60,6 +60,7 @@ struct mana_ib_adapter_caps { u32 max_recv_sge_count; u32 max_inline_data_size; u64 feature_flags; + u64 page_size_cap; }; struct mana_ib_queue { @@ -543,6 +544,11 @@ static inline void mana_put_qp_ref(struct mana_ib_qp *qp) complete(&qp->free); } +static inline bool mana_ib_is_rnic(struct mana_ib_dev *mdev) +{ + return mdev->gdma_dev->dev_id.type == GDMA_DEVICE_MANA_IB; +} + static inline struct net_device *mana_ib_get_netdev(struct ib_device *ibdev, u32 port) { struct mana_ib_dev *mdev = container_of(ibdev, struct mana_ib_dev, ib_dev); @@ -642,6 +648,7 @@ int mana_ib_query_gid(struct ib_device *ibdev, u32 port, int index, void mana_ib_disassociate_ucontext(struct ib_ucontext *ibcontext); int mana_ib_gd_query_adapter_caps(struct mana_ib_dev *mdev); +int mana_eth_query_adapter_caps(struct mana_ib_dev *mdev); int mana_ib_create_eqs(struct mana_ib_dev *mdev); diff --git a/drivers/infiniband/hw/mana/mr.c b/drivers/infiniband/hw/mana/mr.c index f99557ec7767..6d974d0a8400 100644 --- a/drivers/infiniband/hw/mana/mr.c +++ b/drivers/infiniband/hw/mana/mr.c @@ -5,8 +5,8 @@ #include "mana_ib.h" -#define VALID_MR_FLAGS \ - (IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ) +#define VALID_MR_FLAGS (IB_ACCESS_LOCAL_WRITE | IB_ACCESS_REMOTE_WRITE | IB_ACCESS_REMOTE_READ |\ + IB_ACCESS_REMOTE_ATOMIC | IB_ZERO_BASED) #define VALID_DMA_MR_FLAGS (IB_ACCESS_LOCAL_WRITE) @@ -24,6 +24,9 @@ mana_ib_verbs_to_gdma_access_flags(int access_flags) if (access_flags & IB_ACCESS_REMOTE_READ) flags |= GDMA_ACCESS_FLAG_REMOTE_READ; + if (access_flags & IB_ACCESS_REMOTE_ATOMIC) + flags |= GDMA_ACCESS_FLAG_REMOTE_ATOMIC; + return flags; } @@ -48,7 +51,10 @@ static int mana_ib_gd_create_mr(struct mana_ib_dev *dev, struct mana_ib_mr *mr, req.gva.virtual_address = mr_params->gva.virtual_address; req.gva.access_flags = mr_params->gva.access_flags; break; - + case GDMA_MR_TYPE_ZBVA: + req.zbva.dma_region_handle = mr_params->zbva.dma_region_handle; + req.zbva.access_flags = mr_params->zbva.access_flags; + break; default: ibdev_dbg(&dev->ib_dev, "invalid param (GDMA_MR_TYPE) passed, type %d\n", @@ -144,11 +150,18 @@ struct ib_mr *mana_ib_reg_user_mr(struct ib_pd *ibpd, u64 start, u64 length, dma_region_handle); mr_params.pd_handle = pd->pd_handle; - mr_params.mr_type = GDMA_MR_TYPE_GVA; - mr_params.gva.dma_region_handle = dma_region_handle; - mr_params.gva.virtual_address = iova; - mr_params.gva.access_flags = - mana_ib_verbs_to_gdma_access_flags(access_flags); + if (access_flags & IB_ZERO_BASED) { + mr_params.mr_type = GDMA_MR_TYPE_ZBVA; + mr_params.zbva.dma_region_handle = dma_region_handle; + mr_params.zbva.access_flags = + mana_ib_verbs_to_gdma_access_flags(access_flags); + } else { + mr_params.mr_type = GDMA_MR_TYPE_GVA; + mr_params.gva.dma_region_handle = dma_region_handle; + mr_params.gva.virtual_address = iova; + mr_params.gva.access_flags = + mana_ib_verbs_to_gdma_access_flags(access_flags); + } err = mana_ib_gd_create_mr(dev, mr, &mr_params); if (err) diff --git a/drivers/infiniband/hw/mana/qp.c b/drivers/infiniband/hw/mana/qp.c index c928af58f38b..14fd7d6c54a2 100644 --- a/drivers/infiniband/hw/mana/qp.c +++ b/drivers/infiniband/hw/mana/qp.c @@ -635,7 +635,6 @@ static int mana_ib_create_ud_qp(struct ib_qp *ibqp, struct ib_pd *ibpd, { struct mana_ib_dev *mdev = container_of(ibpd->device, struct mana_ib_dev, ib_dev); struct mana_ib_qp *qp = container_of(ibqp, struct mana_ib_qp, ibqp); - struct gdma_context *gc = mdev_to_gc(mdev); u32 doorbell, queue_size; int i, err; @@ -654,7 +653,7 @@ static int mana_ib_create_ud_qp(struct ib_qp *ibqp, struct ib_pd *ibpd, goto destroy_queues; } } - doorbell = gc->mana_ib.doorbell; + doorbell = mdev->gdma_dev->doorbell; err = create_shadow_queue(&qp->shadow_rq, attr->cap.max_recv_wr, sizeof(struct ud_rq_shadow_wqe)); @@ -736,7 +735,7 @@ static int mana_ib_gd_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int err; mana_gd_init_req_hdr(&req.hdr, MANA_IB_SET_QP_STATE, sizeof(req), sizeof(resp)); - req.hdr.dev_id = gc->mana_ib.dev_id; + req.hdr.dev_id = mdev->gdma_dev->dev_id; req.adapter = mdev->adapter_handle; req.qp_handle = qp->qp_handle; req.qp_state = attr->qp_state; diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index 33f525b744f2..e279e69b9a51 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -43,7 +43,7 @@ #define MAX_VFS 80 #define MAX_PEND_REQS_PER_FUNC 4 -#define MAD_TIMEOUT_MS 2000 +#define MAD_TIMEOUT_SEC 2 #define mcg_warn(fmt, arg...) pr_warn("MCG WARNING: " fmt, ##arg) #define mcg_error(fmt, arg...) pr_err(fmt, ##arg) @@ -270,7 +270,7 @@ static int send_join_to_wire(struct mcast_group *group, struct ib_sa_mad *sa_mad if (!ret) { /* calls mlx4_ib_mcg_timeout_handler */ queue_delayed_work(group->demux->mcg_wq, &group->timeout_work, - msecs_to_jiffies(MAD_TIMEOUT_MS)); + secs_to_jiffies(MAD_TIMEOUT_SEC)); } return ret; @@ -309,7 +309,7 @@ static int send_leave_to_wire(struct mcast_group *group, u8 join_state) if (!ret) { /* calls mlx4_ib_mcg_timeout_handler */ queue_delayed_work(group->demux->mcg_wq, &group->timeout_work, - msecs_to_jiffies(MAD_TIMEOUT_MS)); + secs_to_jiffies(MAD_TIMEOUT_SEC)); } return ret; @@ -1091,7 +1091,7 @@ static void _mlx4_ib_mcg_port_cleanup(struct mlx4_ib_demux_ctx *ctx, int destroy for (i = 0; i < MAX_VFS; ++i) clean_vf_mcast(ctx, i); - end = jiffies + msecs_to_jiffies(MAD_TIMEOUT_MS + 3000); + end = jiffies + secs_to_jiffies(MAD_TIMEOUT_SEC + 3); do { count = 0; mutex_lock(&ctx->mcg_table_lock); diff --git a/drivers/infiniband/hw/mlx5/fs.c b/drivers/infiniband/hw/mlx5/fs.c index 251246c73b33..680627f1de33 100644 --- a/drivers/infiniband/hw/mlx5/fs.c +++ b/drivers/infiniband/hw/mlx5/fs.c @@ -1645,11 +1645,6 @@ static struct mlx5_ib_flow_handler *create_flow_rule(struct mlx5_ib_dev *dev, return _create_flow_rule(dev, ft_prio, flow_attr, dst, 0, NULL); } -enum { - LEFTOVERS_MC, - LEFTOVERS_UC, -}; - static struct mlx5_ib_flow_handler *create_leftovers_rule(struct mlx5_ib_dev *dev, struct mlx5_ib_flow_prio *ft_prio, struct ib_flow_attr *flow_attr, @@ -1659,43 +1654,32 @@ static struct mlx5_ib_flow_handler *create_leftovers_rule(struct mlx5_ib_dev *de struct mlx5_ib_flow_handler *handler = NULL; static struct { - struct ib_flow_attr flow_attr; struct ib_flow_spec_eth eth_flow; - } leftovers_specs[] = { - [LEFTOVERS_MC] = { - .flow_attr = { - .num_of_specs = 1, - .size = sizeof(leftovers_specs[0]) - }, - .eth_flow = { - .type = IB_FLOW_SPEC_ETH, - .size = sizeof(struct ib_flow_spec_eth), - .mask = {.dst_mac = {0x1} }, - .val = {.dst_mac = {0x1} } - } - }, - [LEFTOVERS_UC] = { - .flow_attr = { - .num_of_specs = 1, - .size = sizeof(leftovers_specs[0]) - }, - .eth_flow = { - .type = IB_FLOW_SPEC_ETH, - .size = sizeof(struct ib_flow_spec_eth), - .mask = {.dst_mac = {0x1} }, - .val = {.dst_mac = {} } - } - } - }; + struct ib_flow_attr flow_attr; + } leftovers_wc = { .flow_attr = { .num_of_specs = 1, + .size = sizeof(leftovers_wc) }, + .eth_flow = { + .type = IB_FLOW_SPEC_ETH, + .size = sizeof(struct ib_flow_spec_eth), + .mask = { .dst_mac = { 0x1 } }, + .val = { .dst_mac = { 0x1 } } } }; - handler = create_flow_rule(dev, ft_prio, - &leftovers_specs[LEFTOVERS_MC].flow_attr, - dst); + static struct { + struct ib_flow_spec_eth eth_flow; + struct ib_flow_attr flow_attr; + } leftovers_uc = { .flow_attr = { .num_of_specs = 1, + .size = sizeof(leftovers_uc) }, + .eth_flow = { + .type = IB_FLOW_SPEC_ETH, + .size = sizeof(struct ib_flow_spec_eth), + .mask = { .dst_mac = { 0x1 } }, + .val = { .dst_mac = {} } } }; + + handler = create_flow_rule(dev, ft_prio, &leftovers_wc.flow_attr, dst); if (!IS_ERR(handler) && flow_attr->type == IB_FLOW_ATTR_ALL_DEFAULT) { handler_ucast = create_flow_rule(dev, ft_prio, - &leftovers_specs[LEFTOVERS_UC].flow_attr, - dst); + &leftovers_uc.flow_attr, dst); if (IS_ERR(handler_ucast)) { mlx5_del_flow_rules(handler->rule); ft_prio->refcount--; @@ -3461,7 +3445,6 @@ DECLARE_UVERBS_NAMED_OBJECT( &UVERBS_METHOD(MLX5_IB_METHOD_STEERING_ANCHOR_DESTROY)); const struct uapi_definition mlx5_ib_flow_defs[] = { -#if IS_ENABLED(CONFIG_INFINIBAND_USER_ACCESS) UAPI_DEF_CHAIN_OBJ_TREE_NAMED( MLX5_IB_OBJECT_FLOW_MATCHER), UAPI_DEF_CHAIN_OBJ_TREE( @@ -3472,7 +3455,6 @@ const struct uapi_definition mlx5_ib_flow_defs[] = { UAPI_DEF_CHAIN_OBJ_TREE_NAMED( MLX5_IB_OBJECT_STEERING_ANCHOR, UAPI_DEF_IS_OBJ_SUPPORTED(mlx5_ib_shared_ft_allowed)), -#endif {}, }; diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index d07cacaa0abd..ce7610740412 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -485,6 +485,10 @@ static int translate_eth_ext_proto_oper(u32 eth_proto_oper, u16 *active_speed, *active_width = IB_WIDTH_2X; *active_speed = IB_SPEED_NDR; break; + case MLX5E_PROT_MASK(MLX5E_200GAUI_1_200GBASE_CR1_KR1): + *active_width = IB_WIDTH_1X; + *active_speed = IB_SPEED_XDR; + break; case MLX5E_PROT_MASK(MLX5E_400GAUI_8_400GBASE_CR8): *active_width = IB_WIDTH_8X; *active_speed = IB_SPEED_HDR; @@ -493,10 +497,18 @@ static int translate_eth_ext_proto_oper(u32 eth_proto_oper, u16 *active_speed, *active_width = IB_WIDTH_4X; *active_speed = IB_SPEED_NDR; break; + case MLX5E_PROT_MASK(MLX5E_400GAUI_2_400GBASE_CR2_KR2): + *active_width = IB_WIDTH_2X; + *active_speed = IB_SPEED_XDR; + break; case MLX5E_PROT_MASK(MLX5E_800GAUI_8_800GBASE_CR8_KR8): *active_width = IB_WIDTH_8X; *active_speed = IB_SPEED_NDR; break; + case MLX5E_PROT_MASK(MLX5E_800GAUI_4_800GBASE_CR4_KR4): + *active_width = IB_WIDTH_4X; + *active_speed = IB_SPEED_XDR; + break; default: return -EINVAL; } @@ -4422,17 +4434,6 @@ static void mlx5_ib_stage_cong_debugfs_cleanup(struct mlx5_ib_dev *dev) mlx5_core_native_port_num(dev->mdev) - 1); } -static int mlx5_ib_stage_uar_init(struct mlx5_ib_dev *dev) -{ - dev->mdev->priv.uar = mlx5_get_uars_page(dev->mdev); - return PTR_ERR_OR_ZERO(dev->mdev->priv.uar); -} - -static void mlx5_ib_stage_uar_cleanup(struct mlx5_ib_dev *dev) -{ - mlx5_put_uars_page(dev->mdev, dev->mdev->priv.uar); -} - static int mlx5_ib_stage_bfrag_init(struct mlx5_ib_dev *dev) { int err; @@ -4662,9 +4663,6 @@ static const struct mlx5_ib_profile pf_profile = { STAGE_CREATE(MLX5_IB_STAGE_CONG_DEBUGFS, mlx5_ib_stage_cong_debugfs_init, mlx5_ib_stage_cong_debugfs_cleanup), - STAGE_CREATE(MLX5_IB_STAGE_UAR, - mlx5_ib_stage_uar_init, - mlx5_ib_stage_uar_cleanup), STAGE_CREATE(MLX5_IB_STAGE_BFREG, mlx5_ib_stage_bfrag_init, mlx5_ib_stage_bfrag_cleanup), @@ -4722,9 +4720,6 @@ const struct mlx5_ib_profile raw_eth_profile = { STAGE_CREATE(MLX5_IB_STAGE_CONG_DEBUGFS, mlx5_ib_stage_cong_debugfs_init, mlx5_ib_stage_cong_debugfs_cleanup), - STAGE_CREATE(MLX5_IB_STAGE_UAR, - mlx5_ib_stage_uar_init, - mlx5_ib_stage_uar_cleanup), STAGE_CREATE(MLX5_IB_STAGE_BFREG, mlx5_ib_stage_bfrag_init, mlx5_ib_stage_bfrag_cleanup), diff --git a/drivers/infiniband/hw/mlx5/mlx5_ib.h b/drivers/infiniband/hw/mlx5/mlx5_ib.h index ace2df3e1d9f..fde859d207ae 100644 --- a/drivers/infiniband/hw/mlx5/mlx5_ib.h +++ b/drivers/infiniband/hw/mlx5/mlx5_ib.h @@ -351,6 +351,7 @@ struct mlx5_ib_flow_db { #define MLX5_IB_UPD_XLT_PD BIT(4) #define MLX5_IB_UPD_XLT_ACCESS BIT(5) #define MLX5_IB_UPD_XLT_INDIRECT BIT(6) +#define MLX5_IB_UPD_XLT_DOWNGRADE BIT(7) /* Private QP creation flags to be passed in ib_qp_init_attr.create_flags. * @@ -1005,7 +1006,6 @@ enum mlx5_ib_stages { MLX5_IB_STAGE_ODP, MLX5_IB_STAGE_COUNTERS, MLX5_IB_STAGE_CONG_DEBUGFS, - MLX5_IB_STAGE_UAR, MLX5_IB_STAGE_BFREG, MLX5_IB_STAGE_PRE_IB_REG_UMR, MLX5_IB_STAGE_WHITELIST_UID, @@ -1473,8 +1473,8 @@ void mlx5_ib_odp_cleanup_one(struct mlx5_ib_dev *ibdev); int __init mlx5_ib_odp_init(void); void mlx5_ib_odp_cleanup(void); int mlx5_odp_init_mkey_cache(struct mlx5_ib_dev *dev); -void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries, - struct mlx5_ib_mr *mr, int flags); +int mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries, + struct mlx5_ib_mr *mr, int flags); int mlx5_ib_advise_mr_prefetch(struct ib_pd *pd, enum ib_uverbs_advise_mr_advice advice, @@ -1495,8 +1495,11 @@ static inline int mlx5_odp_init_mkey_cache(struct mlx5_ib_dev *dev) { return 0; } -static inline void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries, - struct mlx5_ib_mr *mr, int flags) {} +static inline int mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries, + struct mlx5_ib_mr *mr, int flags) +{ + return -EOPNOTSUPP; +} static inline int mlx5_ib_advise_mr_prefetch(struct ib_pd *pd, diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index b7c8c926c578..6dd813bac5b2 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -525,7 +525,7 @@ static void queue_adjust_cache_locked(struct mlx5_cache_ent *ent) ent->fill_to_high_water = false; if (ent->pending) queue_delayed_work(ent->dev->cache.wq, &ent->dwork, - msecs_to_jiffies(1000)); + secs_to_jiffies(1)); else mod_delayed_work(ent->dev->cache.wq, &ent->dwork, 0); } @@ -576,7 +576,7 @@ static void __cache_work_func(struct mlx5_cache_ent *ent) "add keys command failed, err %d\n", err); queue_delayed_work(cache->wq, &ent->dwork, - msecs_to_jiffies(1000)); + secs_to_jiffies(1)); } } } else if (ent->mkeys_queue.ci > 2 * ent->limit) { @@ -1026,7 +1026,7 @@ void mlx5_mkey_cache_cleanup(struct mlx5_ib_dev *dev) mlx5r_destroy_cache_entries(dev); destroy_workqueue(dev->cache.wq); - del_timer_sync(&dev->delay_timer); + timer_delete_sync(&dev->delay_timer); } struct ib_mr *mlx5_ib_get_dma_mr(struct ib_pd *pd, int acc) @@ -2051,7 +2051,7 @@ static int mlx5_revoke_mr(struct mlx5_ib_mr *mr) ent->in_use--; if (ent->is_tmp && !ent->tmp_cleanup_scheduled) { mod_delayed_work(ent->dev->cache.wq, &ent->dwork, - msecs_to_jiffies(30 * 1000)); + secs_to_jiffies(30)); ent->tmp_cleanup_scheduled = true; } spin_unlock_irq(&ent->mkeys_queue.lock); diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index 86d8fa63bf69..eaa2f9f5f3a9 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -34,6 +34,9 @@ #include <linux/kernel.h> #include <linux/dma-buf.h> #include <linux/dma-resv.h> +#include <linux/hmm.h> +#include <linux/hmm-dma.h> +#include <linux/pci-p2pdma.h> #include "mlx5_ib.h" #include "cmd.h" @@ -158,41 +161,50 @@ static void populate_klm(struct mlx5_klm *pklm, size_t idx, size_t nentries, } } -static u64 umem_dma_to_mtt(dma_addr_t umem_dma) -{ - u64 mtt_entry = umem_dma & ODP_DMA_ADDR_MASK; - - if (umem_dma & ODP_READ_ALLOWED_BIT) - mtt_entry |= MLX5_IB_MTT_READ; - if (umem_dma & ODP_WRITE_ALLOWED_BIT) - mtt_entry |= MLX5_IB_MTT_WRITE; - - return mtt_entry; -} - -static void populate_mtt(__be64 *pas, size_t idx, size_t nentries, - struct mlx5_ib_mr *mr, int flags) +static int populate_mtt(__be64 *pas, size_t start, size_t nentries, + struct mlx5_ib_mr *mr, int flags) { struct ib_umem_odp *odp = to_ib_umem_odp(mr->umem); - dma_addr_t pa; + bool downgrade = flags & MLX5_IB_UPD_XLT_DOWNGRADE; + struct pci_p2pdma_map_state p2pdma_state = {}; + struct ib_device *dev = odp->umem.ibdev; size_t i; if (flags & MLX5_IB_UPD_XLT_ZAP) - return; + return 0; for (i = 0; i < nentries; i++) { - pa = odp->dma_list[idx + i]; - pas[i] = cpu_to_be64(umem_dma_to_mtt(pa)); + unsigned long pfn = odp->map.pfn_list[start + i]; + dma_addr_t dma_addr; + + pfn = odp->map.pfn_list[start + i]; + if (!(pfn & HMM_PFN_VALID)) + /* ODP initialization */ + continue; + + dma_addr = hmm_dma_map_pfn(dev->dma_device, &odp->map, + start + i, &p2pdma_state); + if (ib_dma_mapping_error(dev, dma_addr)) + return -EFAULT; + + dma_addr |= MLX5_IB_MTT_READ; + if ((pfn & HMM_PFN_WRITE) && !downgrade) + dma_addr |= MLX5_IB_MTT_WRITE; + + pas[i] = cpu_to_be64(dma_addr); + odp->npages++; } + return 0; } -void mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries, - struct mlx5_ib_mr *mr, int flags) +int mlx5_odp_populate_xlt(void *xlt, size_t idx, size_t nentries, + struct mlx5_ib_mr *mr, int flags) { if (flags & MLX5_IB_UPD_XLT_INDIRECT) { populate_klm(xlt, idx, nentries, mr, flags); + return 0; } else { - populate_mtt(xlt, idx, nentries, mr, flags); + return populate_mtt(xlt, idx, nentries, mr, flags); } } @@ -303,8 +315,7 @@ static bool mlx5_ib_invalidate_range(struct mmu_interval_notifier *mni, * estimate the cost of another UMR vs. the cost of bigger * UMR. */ - if (umem_odp->dma_list[idx] & - (ODP_READ_ALLOWED_BIT | ODP_WRITE_ALLOWED_BIT)) { + if (umem_odp->map.pfn_list[idx] & HMM_PFN_VALID) { if (!in_block) { blk_start_idx = idx; in_block = 1; @@ -687,7 +698,7 @@ static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp, { int page_shift, ret, np; bool downgrade = flags & MLX5_PF_FLAGS_DOWNGRADE; - u64 access_mask; + u64 access_mask = 0; u64 start_idx; bool fault = !(flags & MLX5_PF_FLAGS_SNAPSHOT); u32 xlt_flags = MLX5_IB_UPD_XLT_ATOMIC; @@ -695,12 +706,14 @@ static int pagefault_real_mr(struct mlx5_ib_mr *mr, struct ib_umem_odp *odp, if (flags & MLX5_PF_FLAGS_ENABLE) xlt_flags |= MLX5_IB_UPD_XLT_ENABLE; + if (flags & MLX5_PF_FLAGS_DOWNGRADE) + xlt_flags |= MLX5_IB_UPD_XLT_DOWNGRADE; + page_shift = odp->page_shift; start_idx = (user_va - ib_umem_start(odp)) >> page_shift; - access_mask = ODP_READ_ALLOWED_BIT; if (odp->umem.writable && !downgrade) - access_mask |= ODP_WRITE_ALLOWED_BIT; + access_mask |= HMM_PFN_WRITE; np = ib_umem_odp_map_dma_and_lock(odp, user_va, bcnt, access_mask, fault); if (np < 0) diff --git a/drivers/infiniband/hw/mlx5/qpc.c b/drivers/infiniband/hw/mlx5/qpc.c index d3dcc272200a..146d03ae40bd 100644 --- a/drivers/infiniband/hw/mlx5/qpc.c +++ b/drivers/infiniband/hw/mlx5/qpc.c @@ -21,8 +21,10 @@ mlx5_get_rsc(struct mlx5_qp_table *table, u32 rsn) spin_lock_irqsave(&table->lock, flags); common = radix_tree_lookup(&table->tree, rsn); - if (common) + if (common && !common->invalid) refcount_inc(&common->refcount); + else + common = NULL; spin_unlock_irqrestore(&table->lock, flags); @@ -178,6 +180,18 @@ static int create_resource_common(struct mlx5_ib_dev *dev, return 0; } +static void modify_resource_common_state(struct mlx5_ib_dev *dev, + struct mlx5_core_qp *qp, + bool invalid) +{ + struct mlx5_qp_table *table = &dev->qp_table; + unsigned long flags; + + spin_lock_irqsave(&table->lock, flags); + qp->common.invalid = invalid; + spin_unlock_irqrestore(&table->lock, flags); +} + static void destroy_resource_common(struct mlx5_ib_dev *dev, struct mlx5_core_qp *qp) { @@ -609,8 +623,20 @@ err_destroy_rq: int mlx5_core_destroy_rq_tracked(struct mlx5_ib_dev *dev, struct mlx5_core_qp *rq) { + int ret; + + /* The rq destruction can be called again in case it fails, hence we + * mark the common resource as invalid and only once FW destruction + * is completed successfully we actually destroy the resources. + */ + modify_resource_common_state(dev, rq, true); + ret = destroy_rq_tracked(dev, rq->qpn, rq->uid); + if (ret) { + modify_resource_common_state(dev, rq, false); + return ret; + } destroy_resource_common(dev, rq); - return destroy_rq_tracked(dev, rq->qpn, rq->uid); + return 0; } static void destroy_sq_tracked(struct mlx5_ib_dev *dev, u32 sqn, u16 uid) diff --git a/drivers/infiniband/hw/mlx5/umr.c b/drivers/infiniband/hw/mlx5/umr.c index 793f3c5c4d01..5be4426a2884 100644 --- a/drivers/infiniband/hw/mlx5/umr.c +++ b/drivers/infiniband/hw/mlx5/umr.c @@ -840,7 +840,17 @@ int mlx5r_umr_update_xlt(struct mlx5_ib_mr *mr, u64 idx, int npages, size_to_map = npages * desc_size; dma_sync_single_for_cpu(ddev, sg.addr, sg.length, DMA_TO_DEVICE); - mlx5_odp_populate_xlt(xlt, idx, npages, mr, flags); + /* + * npages is the maximum number of pages to map, but we + * can't guarantee that all pages are actually mapped. + * + * For example, if page is p2p of type which is not supported + * for mapping, the number of pages mapped will be less than + * requested. + */ + err = mlx5_odp_populate_xlt(xlt, idx, npages, mr, flags); + if (err) + return err; dma_sync_single_for_device(ddev, sg.addr, sg.length, DMA_TO_DEVICE); sg.length = ALIGN(size_to_map, MLX5_UMR_FLEX_ALIGNMENT); diff --git a/drivers/infiniband/hw/mthca/mthca_catas.c b/drivers/infiniband/hw/mthca/mthca_catas.c index ffb98eaaf1c2..6eabef9aa211 100644 --- a/drivers/infiniband/hw/mthca/mthca_catas.c +++ b/drivers/infiniband/hw/mthca/mthca_catas.c @@ -171,7 +171,7 @@ void mthca_start_catas_poll(struct mthca_dev *dev) void mthca_stop_catas_poll(struct mthca_dev *dev) { - del_timer_sync(&dev->catas_err.timer); + timer_delete_sync(&dev->catas_err.timer); if (dev->catas_err.map) iounmap(dev->catas_err.map); diff --git a/drivers/infiniband/hw/mthca/mthca_mr.c b/drivers/infiniband/hw/mthca/mthca_mr.c index 192f83fd7c8a..dacb8ceeebe0 100644 --- a/drivers/infiniband/hw/mthca/mthca_mr.c +++ b/drivers/infiniband/hw/mthca/mthca_mr.c @@ -144,7 +144,7 @@ static int mthca_buddy_init(struct mthca_buddy *buddy, int max_order) buddy->max_order = max_order; spin_lock_init(&buddy->lock); - buddy->bits = kcalloc(buddy->max_order + 1, sizeof(long *), + buddy->bits = kcalloc(buddy->max_order + 1, sizeof(*buddy->bits), GFP_KERNEL); buddy->num_free = kcalloc((buddy->max_order + 1), sizeof *buddy->num_free, GFP_KERNEL); diff --git a/drivers/infiniband/hw/qib/qib_driver.c b/drivers/infiniband/hw/qib/qib_driver.c index 4fcbef99e400..bdd724add147 100644 --- a/drivers/infiniband/hw/qib/qib_driver.c +++ b/drivers/infiniband/hw/qib/qib_driver.c @@ -768,7 +768,7 @@ int qib_reset_device(int unit) ppd = dd->pport + pidx; if (atomic_read(&ppd->led_override_timer_active)) { /* Need to stop LED timer, _then_ shut off LEDs */ - del_timer_sync(&ppd->led_override_timer); + timer_delete_sync(&ppd->led_override_timer); atomic_set(&ppd->led_override_timer_active, 0); } diff --git a/drivers/infiniband/hw/qib/qib_fs.c b/drivers/infiniband/hw/qib/qib_fs.c index b9f4a2937c3a..2098de762bf5 100644 --- a/drivers/infiniband/hw/qib/qib_fs.c +++ b/drivers/infiniband/hw/qib/qib_fs.c @@ -90,7 +90,7 @@ static int create_file(const char *name, umode_t mode, int error; inode_lock(d_inode(parent)); - *dentry = lookup_one_len(name, parent, strlen(name)); + *dentry = lookup_noperm(&QSTR(name), parent); if (!IS_ERR(*dentry)) error = qibfs_mknod(d_inode(parent), *dentry, mode, fops, data); @@ -433,7 +433,7 @@ static int remove_device_files(struct super_block *sb, char unit[10]; snprintf(unit, sizeof(unit), "%u", dd->unit); - dir = lookup_one_len_unlocked(unit, sb->s_root, strlen(unit)); + dir = lookup_noperm_unlocked(&QSTR(unit), sb->s_root); if (IS_ERR(dir)) { pr_err("Lookup of %s failed\n", unit); diff --git a/drivers/infiniband/hw/qib/qib_iba7220.c b/drivers/infiniband/hw/qib/qib_iba7220.c index 78dfe98ebcf7..302c0d19f57d 100644 --- a/drivers/infiniband/hw/qib/qib_iba7220.c +++ b/drivers/infiniband/hw/qib/qib_iba7220.c @@ -1656,7 +1656,7 @@ static void qib_7220_quiet_serdes(struct qib_pportdata *ppd) ppd->cpspec->chase_end = 0; if (ppd->cpspec->chase_timer.function) /* if initted */ - del_timer_sync(&ppd->cpspec->chase_timer); + timer_delete_sync(&ppd->cpspec->chase_timer); if (ppd->cpspec->ibsymdelta || ppd->cpspec->iblnkerrdelta || ppd->cpspec->ibdeltainprog) { @@ -2605,7 +2605,7 @@ static int qib_7220_set_ib_cfg(struct qib_pportdata *ppd, int which, u32 val) * wait forpending timer, but don't clear .data (ppd)! */ if (ppd->cpspec->chase_timer.expires) { - del_timer_sync(&ppd->cpspec->chase_timer); + timer_delete_sync(&ppd->cpspec->chase_timer); ppd->cpspec->chase_timer.expires = 0; } break; diff --git a/drivers/infiniband/hw/qib/qib_iba7322.c b/drivers/infiniband/hw/qib/qib_iba7322.c index 9db29916e35a..7b4bf06c3b38 100644 --- a/drivers/infiniband/hw/qib/qib_iba7322.c +++ b/drivers/infiniband/hw/qib/qib_iba7322.c @@ -2512,7 +2512,7 @@ static void qib_7322_mini_quiet_serdes(struct qib_pportdata *ppd) ppd->cpspec->chase_end = 0; if (ppd->cpspec->chase_timer.function) /* if initted */ - del_timer_sync(&ppd->cpspec->chase_timer); + timer_delete_sync(&ppd->cpspec->chase_timer); /* * Despite the name, actually disables IBC as well. Do it when @@ -4239,7 +4239,7 @@ static int qib_7322_set_ib_cfg(struct qib_pportdata *ppd, int which, u32 val) * wait forpending timer, but don't clear .data (ppd)! */ if (ppd->cpspec->chase_timer.expires) { - del_timer_sync(&ppd->cpspec->chase_timer); + timer_delete_sync(&ppd->cpspec->chase_timer); ppd->cpspec->chase_timer.expires = 0; } break; diff --git a/drivers/infiniband/hw/qib/qib_init.c b/drivers/infiniband/hw/qib/qib_init.c index 4100656fe9a3..33c23adec101 100644 --- a/drivers/infiniband/hw/qib/qib_init.c +++ b/drivers/infiniband/hw/qib/qib_init.c @@ -796,19 +796,19 @@ static void qib_stop_timers(struct qib_devdata *dd) int pidx; if (dd->stats_timer.function) - del_timer_sync(&dd->stats_timer); + timer_delete_sync(&dd->stats_timer); if (dd->intrchk_timer.function) - del_timer_sync(&dd->intrchk_timer); + timer_delete_sync(&dd->intrchk_timer); for (pidx = 0; pidx < dd->num_pports; ++pidx) { ppd = dd->pport + pidx; if (ppd->hol_timer.function) - del_timer_sync(&ppd->hol_timer); + timer_delete_sync(&ppd->hol_timer); if (ppd->led_override_timer.function) { - del_timer_sync(&ppd->led_override_timer); + timer_delete_sync(&ppd->led_override_timer); atomic_set(&ppd->led_override_timer_active, 0); } if (ppd->symerr_clear_timer.function) - del_timer_sync(&ppd->symerr_clear_timer); + timer_delete_sync(&ppd->symerr_clear_timer); } } diff --git a/drivers/infiniband/hw/qib/qib_mad.c b/drivers/infiniband/hw/qib/qib_mad.c index ef02f2bfddb2..568deb77ab4d 100644 --- a/drivers/infiniband/hw/qib/qib_mad.c +++ b/drivers/infiniband/hw/qib/qib_mad.c @@ -2441,7 +2441,7 @@ void qib_notify_free_mad_agent(struct rvt_dev_info *rdi, int port_idx) struct qib_devdata, verbs_dev); if (dd->pport[port_idx].cong_stats.timer.function) - del_timer_sync(&dd->pport[port_idx].cong_stats.timer); + timer_delete_sync(&dd->pport[port_idx].cong_stats.timer); if (dd->pport[port_idx].ibport_data.smi_ah) rdma_destroy_ah(&dd->pport[port_idx].ibport_data.smi_ah->ibah, diff --git a/drivers/infiniband/hw/qib/qib_sd7220.c b/drivers/infiniband/hw/qib/qib_sd7220.c index 1dc3ccf0cf1f..c4ee120ac7fb 100644 --- a/drivers/infiniband/hw/qib/qib_sd7220.c +++ b/drivers/infiniband/hw/qib/qib_sd7220.c @@ -1375,7 +1375,7 @@ void toggle_7220_rclkrls(struct qib_devdata *dd) void shutdown_7220_relock_poll(struct qib_devdata *dd) { if (dd->cspec->relock_timer_active) - del_timer_sync(&dd->cspec->relock_timer); + timer_delete_sync(&dd->cspec->relock_timer); } static unsigned qib_relock_by_timer = 1; diff --git a/drivers/infiniband/hw/qib/qib_verbs.c b/drivers/infiniband/hw/qib/qib_verbs.c index 5fcb41970ad9..9832567a8bb8 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.c +++ b/drivers/infiniband/hw/qib/qib_verbs.c @@ -1655,7 +1655,7 @@ void qib_unregister_ib_device(struct qib_devdata *dd) if (!list_empty(&dev->memwait)) qib_dev_err(dd, "memwait list not empty!\n"); - del_timer_sync(&dev->mem_timer); + timer_delete_sync(&dev->mem_timer); while (!list_empty(&dev->txreq_free)) { struct list_head *l = dev->txreq_free.next; struct qib_verbs_txreq *tx; diff --git a/drivers/infiniband/hw/usnic/usnic_ib_main.c b/drivers/infiniband/hw/usnic/usnic_ib_main.c index 4ddcd5860e0f..11eca39b73a9 100644 --- a/drivers/infiniband/hw/usnic/usnic_ib_main.c +++ b/drivers/infiniband/hw/usnic/usnic_ib_main.c @@ -397,7 +397,7 @@ static void *usnic_ib_device_add(struct pci_dev *dev) if (!us_ibdev) { usnic_err("Device %s context alloc failed\n", netdev_name(pci_get_drvdata(dev))); - return ERR_PTR(-EFAULT); + return NULL; } us_ibdev->ufdev = usnic_fwd_dev_alloc(dev); @@ -517,8 +517,8 @@ static struct usnic_ib_dev *usnic_ib_discover_pf(struct usnic_vnic *vnic) } us_ibdev = usnic_ib_device_add(parent_pci); - if (IS_ERR_OR_NULL(us_ibdev)) { - us_ibdev = us_ibdev ? us_ibdev : ERR_PTR(-EFAULT); + if (!us_ibdev) { + us_ibdev = ERR_PTR(-EFAULT); goto out; } @@ -586,10 +586,10 @@ static int usnic_ib_pci_probe(struct pci_dev *pdev, } pf = usnic_ib_discover_pf(vf->vnic); - if (IS_ERR_OR_NULL(pf)) { - usnic_err("Failed to discover pf of vnic %s with err%ld\n", - pci_name(pdev), PTR_ERR(pf)); - err = pf ? PTR_ERR(pf) : -EFAULT; + if (IS_ERR(pf)) { + err = PTR_ERR(pf); + usnic_err("Failed to discover pf of vnic %s with err%d\n", + pci_name(pdev), err); goto out_clean_vnic; } diff --git a/drivers/infiniband/hw/usnic/usnic_uiom.c b/drivers/infiniband/hw/usnic/usnic_uiom.c index f948b76f984d..3fbf99757b11 100644 --- a/drivers/infiniband/hw/usnic/usnic_uiom.c +++ b/drivers/infiniband/hw/usnic/usnic_uiom.c @@ -56,7 +56,7 @@ static int usnic_uiom_dma_fault(struct iommu_domain *domain, unsigned long iova, int flags, void *token) { - usnic_err("Device %s iommu fault domain 0x%pK va 0x%lx flags 0x%x\n", + usnic_err("Device %s iommu fault domain 0x%p va 0x%lx flags 0x%x\n", dev_name(dev), domain, iova, flags); return -ENOSYS; diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c index 614009fb9632..583debe4b9a2 100644 --- a/drivers/infiniband/sw/rdmavt/qp.c +++ b/drivers/infiniband/sw/rdmavt/qp.c @@ -1297,7 +1297,7 @@ int rvt_error_qp(struct rvt_qp *qp, enum ib_wc_status err) if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) { qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR); - del_timer(&qp->s_timer); + timer_delete(&qp->s_timer); } if (qp->s_flags & RVT_S_ANY_WAIT_SEND) @@ -2546,7 +2546,7 @@ void rvt_stop_rc_timers(struct rvt_qp *qp) /* Remove QP from all timers */ if (qp->s_flags & (RVT_S_TIMER | RVT_S_WAIT_RNR)) { qp->s_flags &= ~(RVT_S_TIMER | RVT_S_WAIT_RNR); - del_timer(&qp->s_timer); + timer_delete(&qp->s_timer); hrtimer_try_to_cancel(&qp->s_rnr_timer); } } @@ -2575,7 +2575,7 @@ static void rvt_stop_rnr_timer(struct rvt_qp *qp) */ void rvt_del_timers_sync(struct rvt_qp *qp) { - del_timer_sync(&qp->s_timer); + timer_delete_sync(&qp->s_timer); hrtimer_cancel(&qp->s_rnr_timer); } EXPORT_SYMBOL(rvt_del_timers_sync); @@ -2596,7 +2596,7 @@ static void rvt_rc_timeout(struct timer_list *t) qp->s_flags &= ~RVT_S_TIMER; rvp->n_rc_timeouts++; - del_timer(&qp->s_timer); + timer_delete(&qp->s_timer); trace_rvt_rc_timeout(qp, qp->s_last_psn + 1); if (rdi->driver_f.notify_restart_rc) rdi->driver_f.notify_restart_rc(qp, diff --git a/drivers/infiniband/sw/rxe/Kconfig b/drivers/infiniband/sw/rxe/Kconfig index c180e7ebcfc5..1ed5b63f8afc 100644 --- a/drivers/infiniband/sw/rxe/Kconfig +++ b/drivers/infiniband/sw/rxe/Kconfig @@ -1,7 +1,7 @@ # SPDX-License-Identifier: GPL-2.0-only config RDMA_RXE tristate "Software RDMA over Ethernet (RoCE) driver" - depends on INET && PCI && INFINIBAND + depends on INET && PCI && INFINIBAND && 64BIT depends on INFINIBAND_VIRT_DMA select NET_UDP_TUNNEL select CRC32 diff --git a/drivers/infiniband/sw/rxe/rxe.c b/drivers/infiniband/sw/rxe/rxe.c index b248c68bf9b1..3a77d6db1720 100644 --- a/drivers/infiniband/sw/rxe/rxe.c +++ b/drivers/infiniband/sw/rxe/rxe.c @@ -101,6 +101,8 @@ static void rxe_init_device_param(struct rxe_dev *rxe, struct net_device *ndev) rxe->attr.odp_caps.per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_READ; rxe->attr.odp_caps.per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_ATOMIC; rxe->attr.odp_caps.per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_SRQ_RECV; + rxe->attr.odp_caps.per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_FLUSH; + rxe->attr.odp_caps.per_transport_caps.rc_odp_caps |= IB_ODP_SUPPORT_ATOMIC_WRITE; } } diff --git a/drivers/infiniband/sw/rxe/rxe_cq.c b/drivers/infiniband/sw/rxe/rxe_cq.c index fec87c9030ab..fffd144d509e 100644 --- a/drivers/infiniband/sw/rxe/rxe_cq.c +++ b/drivers/infiniband/sw/rxe/rxe_cq.c @@ -56,11 +56,8 @@ int rxe_cq_from_init(struct rxe_dev *rxe, struct rxe_cq *cq, int cqe, err = do_mmap_info(rxe, uresp ? &uresp->mi : NULL, udata, cq->queue->buf, cq->queue->buf_size, &cq->queue->ip); - if (err) { - vfree(cq->queue->buf); - kfree(cq->queue); + if (err) return err; - } cq->is_user = uresp; diff --git a/drivers/infiniband/sw/rxe/rxe_loc.h b/drivers/infiniband/sw/rxe/rxe_loc.h index feb386d98d1d..876702058c84 100644 --- a/drivers/infiniband/sw/rxe/rxe_loc.h +++ b/drivers/infiniband/sw/rxe/rxe_loc.h @@ -70,9 +70,9 @@ int copy_data(struct rxe_pd *pd, int access, struct rxe_dma_info *dma, void *addr, int length, enum rxe_mr_copy_dir dir); int rxe_map_mr_sg(struct ib_mr *ibmr, struct scatterlist *sg, int sg_nents, unsigned int *sg_offset); -int rxe_mr_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, - u64 compare, u64 swap_add, u64 *orig_val); -int rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value); +enum resp_states rxe_mr_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, + u64 compare, u64 swap_add, u64 *orig_val); +enum resp_states rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value); struct rxe_mr *lookup_mr(struct rxe_pd *pd, int access, u32 key, enum rxe_mr_lookup_type type); int mr_check_range(struct rxe_mr *mr, u64 iova, size_t length); @@ -140,6 +140,12 @@ static inline int qp_mtu(struct rxe_qp *qp) return IB_MTU_4096; } +static inline bool is_odp_mr(struct rxe_mr *mr) +{ + return IS_ENABLED(CONFIG_INFINIBAND_ON_DEMAND_PAGING) && mr->umem && + mr->umem->is_odp; +} + void free_rd_atomic_resource(struct resp_res *res); static inline void rxe_advance_resp_resource(struct rxe_qp *qp) @@ -187,13 +193,16 @@ static inline unsigned int wr_opcode_mask(int opcode, struct rxe_qp *qp) /* rxe_odp.c */ extern const struct mmu_interval_notifier_ops rxe_mn_ops; -#ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING +#if defined CONFIG_INFINIBAND_ON_DEMAND_PAGING int rxe_odp_mr_init_user(struct rxe_dev *rxe, u64 start, u64 length, u64 iova, int access_flags, struct rxe_mr *mr); int rxe_odp_mr_copy(struct rxe_mr *mr, u64 iova, void *addr, int length, enum rxe_mr_copy_dir dir); -int rxe_odp_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, - u64 compare, u64 swap_add, u64 *orig_val); +enum resp_states rxe_odp_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, + u64 compare, u64 swap_add, u64 *orig_val); +int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, u64 iova, + unsigned int length); +enum resp_states rxe_odp_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value); #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */ static inline int rxe_odp_mr_init_user(struct rxe_dev *rxe, u64 start, u64 length, u64 iova, @@ -206,9 +215,19 @@ static inline int rxe_odp_mr_copy(struct rxe_mr *mr, u64 iova, void *addr, { return -EOPNOTSUPP; } -static inline int +static inline enum resp_states rxe_odp_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, - u64 compare, u64 swap_add, u64 *orig_val) + u64 compare, u64 swap_add, u64 *orig_val) +{ + return RESPST_ERR_UNSUPPORTED_OPCODE; +} +static inline int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, u64 iova, + unsigned int length) +{ + return -EOPNOTSUPP; +} +static inline enum resp_states rxe_odp_do_atomic_write(struct rxe_mr *mr, + u64 iova, u64 value) { return RESPST_ERR_UNSUPPORTED_OPCODE; } diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c index 868d2f0b74e9..bcb97b3ea58a 100644 --- a/drivers/infiniband/sw/rxe/rxe_mr.c +++ b/drivers/infiniband/sw/rxe/rxe_mr.c @@ -323,7 +323,7 @@ int rxe_mr_copy(struct rxe_mr *mr, u64 iova, void *addr, return err; } - if (mr->umem->is_odp) + if (is_odp_mr(mr)) return rxe_odp_mr_copy(mr, iova, addr, length, dir); else return rxe_mr_copy_xarray(mr, iova, addr, length, dir); @@ -424,7 +424,7 @@ err1: return err; } -int rxe_flush_pmem_iova(struct rxe_mr *mr, u64 iova, unsigned int length) +static int rxe_mr_flush_pmem_iova(struct rxe_mr *mr, u64 iova, unsigned int length) { unsigned int page_offset; unsigned long index; @@ -433,16 +433,6 @@ int rxe_flush_pmem_iova(struct rxe_mr *mr, u64 iova, unsigned int length) int err; u8 *va; - /* mr must be valid even if length is zero */ - if (WARN_ON(!mr)) - return -EINVAL; - - if (length == 0) - return 0; - - if (mr->ibmr.type == IB_MR_TYPE_DMA) - return -EFAULT; - err = mr_check_range(mr, iova, length); if (err) return err; @@ -454,7 +444,7 @@ int rxe_flush_pmem_iova(struct rxe_mr *mr, u64 iova, unsigned int length) if (!page) return -EFAULT; bytes = min_t(unsigned int, length, - mr_page_size(mr) - page_offset); + mr_page_size(mr) - page_offset); va = kmap_local_page(page); arch_wb_cache_pmem(va + page_offset, bytes); @@ -468,11 +458,33 @@ int rxe_flush_pmem_iova(struct rxe_mr *mr, u64 iova, unsigned int length) return 0; } +int rxe_flush_pmem_iova(struct rxe_mr *mr, u64 start, unsigned int length) +{ + int err; + + /* mr must be valid even if length is zero */ + if (WARN_ON(!mr)) + return -EINVAL; + + if (length == 0) + return 0; + + if (mr->ibmr.type == IB_MR_TYPE_DMA) + return -EFAULT; + + if (is_odp_mr(mr)) + err = rxe_odp_flush_pmem_iova(mr, start, length); + else + err = rxe_mr_flush_pmem_iova(mr, start, length); + + return err; +} + /* Guarantee atomicity of atomic operations at the machine level. */ DEFINE_SPINLOCK(atomic_ops_lock); -int rxe_mr_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, - u64 compare, u64 swap_add, u64 *orig_val) +enum resp_states rxe_mr_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, + u64 compare, u64 swap_add, u64 *orig_val) { unsigned int page_offset; struct page *page; @@ -524,27 +536,15 @@ int rxe_mr_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, kunmap_local(va); - return 0; + return RESPST_NONE; } -#if defined CONFIG_64BIT -/* only implemented or called for 64 bit architectures */ -int rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value) +enum resp_states rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value) { unsigned int page_offset; struct page *page; u64 *va; - /* ODP is not supported right now. WIP. */ - if (mr->umem->is_odp) - return RESPST_ERR_UNSUPPORTED_OPCODE; - - /* See IBA oA19-28 */ - if (unlikely(mr->state != RXE_MR_STATE_VALID)) { - rxe_dbg_mr(mr, "mr not in valid state\n"); - return RESPST_ERR_RKEY_VIOLATION; - } - if (mr->ibmr.type == IB_MR_TYPE_DMA) { page_offset = iova & (PAGE_SIZE - 1); page = ib_virt_dma_to_page(iova); @@ -572,20 +572,12 @@ int rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value) } va = kmap_local_page(page); - /* Do atomic write after all prior operations have completed */ smp_store_release(&va[page_offset >> 3], value); - kunmap_local(va); - return 0; -} -#else -int rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value) -{ - return RESPST_ERR_UNSUPPORTED_OPCODE; + return RESPST_NONE; } -#endif int advance_dma_data(struct rxe_dma_info *dma, unsigned int length) { diff --git a/drivers/infiniband/sw/rxe/rxe_odp.c b/drivers/infiniband/sw/rxe/rxe_odp.c index 9f6e2bb2a269..dbc5a5600eb7 100644 --- a/drivers/infiniband/sw/rxe/rxe_odp.c +++ b/drivers/infiniband/sw/rxe/rxe_odp.c @@ -4,6 +4,7 @@ */ #include <linux/hmm.h> +#include <linux/libnvdimm.h> #include <rdma/ib_umem_odp.h> @@ -26,7 +27,7 @@ static bool rxe_ib_invalidate_range(struct mmu_interval_notifier *mni, start = max_t(u64, ib_umem_start(umem_odp), range->start); end = min_t(u64, ib_umem_end(umem_odp), range->end); - /* update umem_odp->dma_list */ + /* update umem_odp->map.pfn_list */ ib_umem_odp_unmap_dma_pages(umem_odp, start, end); mutex_unlock(&umem_odp->umem_mutex); @@ -44,12 +45,11 @@ static int rxe_odp_do_pagefault_and_lock(struct rxe_mr *mr, u64 user_va, int bcn { struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); bool fault = !(flags & RXE_PAGEFAULT_SNAPSHOT); - u64 access_mask; + u64 access_mask = 0; int np; - access_mask = ODP_READ_ALLOWED_BIT; if (umem_odp->umem.writable && !(flags & RXE_PAGEFAULT_RDONLY)) - access_mask |= ODP_WRITE_ALLOWED_BIT; + access_mask |= HMM_PFN_WRITE; /* * ib_umem_odp_map_dma_and_lock() locks umem_mutex on success. @@ -124,8 +124,8 @@ int rxe_odp_mr_init_user(struct rxe_dev *rxe, u64 start, u64 length, return err; } -static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, - u64 iova, int length, u32 perm) +static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, u64 iova, + int length) { bool need_fault = false; u64 addr; @@ -137,7 +137,7 @@ static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, while (addr < iova + length) { idx = (addr - ib_umem_start(umem_odp)) >> umem_odp->page_shift; - if (!(umem_odp->dma_list[idx] & perm)) { + if (!(umem_odp->map.pfn_list[idx] & HMM_PFN_VALID)) { need_fault = true; break; } @@ -147,23 +147,28 @@ static inline bool rxe_check_pagefault(struct ib_umem_odp *umem_odp, return need_fault; } +static unsigned long rxe_odp_iova_to_index(struct ib_umem_odp *umem_odp, u64 iova) +{ + return (iova - ib_umem_start(umem_odp)) >> umem_odp->page_shift; +} + +static unsigned long rxe_odp_iova_to_page_offset(struct ib_umem_odp *umem_odp, u64 iova) +{ + return iova & (BIT(umem_odp->page_shift) - 1); +} + static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u32 flags) { struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); bool need_fault; - u64 perm; int err; if (unlikely(length < 1)) return -EINVAL; - perm = ODP_READ_ALLOWED_BIT; - if (!(flags & RXE_PAGEFAULT_RDONLY)) - perm |= ODP_WRITE_ALLOWED_BIT; - mutex_lock(&umem_odp->umem_mutex); - need_fault = rxe_check_pagefault(umem_odp, iova, length, perm); + need_fault = rxe_check_pagefault(umem_odp, iova, length); if (need_fault) { mutex_unlock(&umem_odp->umem_mutex); @@ -173,7 +178,7 @@ static int rxe_odp_map_range_and_lock(struct rxe_mr *mr, u64 iova, int length, u if (err < 0) return err; - need_fault = rxe_check_pagefault(umem_odp, iova, length, perm); + need_fault = rxe_check_pagefault(umem_odp, iova, length); if (need_fault) return -EFAULT; } @@ -190,13 +195,13 @@ static int __rxe_odp_mr_copy(struct rxe_mr *mr, u64 iova, void *addr, size_t offset; u8 *user_va; - idx = (iova - ib_umem_start(umem_odp)) >> umem_odp->page_shift; - offset = iova & (BIT(umem_odp->page_shift) - 1); + idx = rxe_odp_iova_to_index(umem_odp, iova); + offset = rxe_odp_iova_to_page_offset(umem_odp, iova); while (length > 0) { u8 *src, *dest; - page = hmm_pfn_to_page(umem_odp->pfn_list[idx]); + page = hmm_pfn_to_page(umem_odp->map.pfn_list[idx]); user_va = kmap_local_page(page); if (!user_va) return -EFAULT; @@ -255,8 +260,9 @@ int rxe_odp_mr_copy(struct rxe_mr *mr, u64 iova, void *addr, int length, return err; } -static int rxe_odp_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, - u64 compare, u64 swap_add, u64 *orig_val) +static enum resp_states rxe_odp_do_atomic_op(struct rxe_mr *mr, u64 iova, + int opcode, u64 compare, + u64 swap_add, u64 *orig_val) { struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); unsigned int page_offset; @@ -277,9 +283,9 @@ static int rxe_odp_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, return RESPST_ERR_RKEY_VIOLATION; } - idx = (iova - ib_umem_start(umem_odp)) >> umem_odp->page_shift; - page_offset = iova & (BIT(umem_odp->page_shift) - 1); - page = hmm_pfn_to_page(umem_odp->pfn_list[idx]); + idx = rxe_odp_iova_to_index(umem_odp, iova); + page_offset = rxe_odp_iova_to_page_offset(umem_odp, iova); + page = hmm_pfn_to_page(umem_odp->map.pfn_list[idx]); if (!page) return RESPST_ERR_RKEY_VIOLATION; @@ -304,11 +310,11 @@ static int rxe_odp_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, kunmap_local(va); - return 0; + return RESPST_NONE; } -int rxe_odp_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, - u64 compare, u64 swap_add, u64 *orig_val) +enum resp_states rxe_odp_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, + u64 compare, u64 swap_add, u64 *orig_val) { struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); int err; @@ -324,3 +330,91 @@ int rxe_odp_atomic_op(struct rxe_mr *mr, u64 iova, int opcode, return err; } + +int rxe_odp_flush_pmem_iova(struct rxe_mr *mr, u64 iova, + unsigned int length) +{ + struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); + unsigned int page_offset; + unsigned long index; + struct page *page; + unsigned int bytes; + int err; + u8 *va; + + err = rxe_odp_map_range_and_lock(mr, iova, length, + RXE_PAGEFAULT_DEFAULT); + if (err) + return err; + + while (length > 0) { + index = rxe_odp_iova_to_index(umem_odp, iova); + page_offset = rxe_odp_iova_to_page_offset(umem_odp, iova); + + page = hmm_pfn_to_page(umem_odp->map.pfn_list[index]); + if (!page) { + mutex_unlock(&umem_odp->umem_mutex); + return -EFAULT; + } + + bytes = min_t(unsigned int, length, + mr_page_size(mr) - page_offset); + + va = kmap_local_page(page); + arch_wb_cache_pmem(va + page_offset, bytes); + kunmap_local(va); + + length -= bytes; + iova += bytes; + page_offset = 0; + } + + mutex_unlock(&umem_odp->umem_mutex); + + return 0; +} + +enum resp_states rxe_odp_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value) +{ + struct ib_umem_odp *umem_odp = to_ib_umem_odp(mr->umem); + unsigned int page_offset; + unsigned long index; + struct page *page; + int err; + u64 *va; + + /* See IBA oA19-28 */ + err = mr_check_range(mr, iova, sizeof(value)); + if (unlikely(err)) { + rxe_dbg_mr(mr, "iova out of range\n"); + return RESPST_ERR_RKEY_VIOLATION; + } + + err = rxe_odp_map_range_and_lock(mr, iova, sizeof(value), + RXE_PAGEFAULT_DEFAULT); + if (err) + return RESPST_ERR_RKEY_VIOLATION; + + page_offset = rxe_odp_iova_to_page_offset(umem_odp, iova); + index = rxe_odp_iova_to_index(umem_odp, iova); + page = hmm_pfn_to_page(umem_odp->map.pfn_list[index]); + if (!page) { + mutex_unlock(&umem_odp->umem_mutex); + return RESPST_ERR_RKEY_VIOLATION; + } + /* See IBA A19.4.2 */ + if (unlikely(page_offset & 0x7)) { + mutex_unlock(&umem_odp->umem_mutex); + rxe_dbg_mr(mr, "misaligned address\n"); + return RESPST_ERR_MISALIGNED_ATOMIC; + } + + va = kmap_local_page(page); + /* Do atomic write after all prior operations have completed */ + smp_store_release(&va[page_offset >> 3], value); + kunmap_local(va); + + mutex_unlock(&umem_odp->umem_mutex); + + return RESPST_NONE; +} diff --git a/drivers/infiniband/sw/rxe/rxe_param.h b/drivers/infiniband/sw/rxe/rxe_param.h index 003f681e5dc0..767870568372 100644 --- a/drivers/infiniband/sw/rxe/rxe_param.h +++ b/drivers/infiniband/sw/rxe/rxe_param.h @@ -53,12 +53,9 @@ enum rxe_device_param { | IB_DEVICE_MEM_WINDOW | IB_DEVICE_FLUSH_GLOBAL | IB_DEVICE_FLUSH_PERSISTENT -#ifdef CONFIG_64BIT | IB_DEVICE_MEM_WINDOW_TYPE_2B | IB_DEVICE_ATOMIC_WRITE, -#else - | IB_DEVICE_MEM_WINDOW_TYPE_2B, -#endif /* CONFIG_64BIT */ + RXE_MAX_SGE = 32, RXE_MAX_WQE_SIZE = sizeof(struct rxe_send_wqe) + sizeof(struct ib_sge) * RXE_MAX_SGE, diff --git a/drivers/infiniband/sw/rxe/rxe_qp.c b/drivers/infiniband/sw/rxe/rxe_qp.c index 91d329e90308..f2af3e0aef35 100644 --- a/drivers/infiniband/sw/rxe/rxe_qp.c +++ b/drivers/infiniband/sw/rxe/rxe_qp.c @@ -811,9 +811,14 @@ static void rxe_qp_do_cleanup(struct work_struct *work) spin_unlock_irqrestore(&qp->state_lock, flags); qp->qp_timeout_jiffies = 0; - if (qp_type(qp) == IB_QPT_RC) { - del_timer_sync(&qp->retrans_timer); - del_timer_sync(&qp->rnr_nak_timer); + /* In the function timer_setup, .function is initialized. If .function + * is NULL, it indicates the function timer_setup is not called, the + * timer is not initialized. Or else, the timer is initialized. + */ + if (qp_type(qp) == IB_QPT_RC && qp->retrans_timer.function && + qp->rnr_nak_timer.function) { + timer_delete_sync(&qp->retrans_timer); + timer_delete_sync(&qp->rnr_nak_timer); } if (qp->recv_task.func) diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index 54ba9ee1acc5..711f73e0bbb1 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -649,10 +649,6 @@ static enum resp_states process_flush(struct rxe_qp *qp, struct rxe_mr *mr = qp->resp.mr; struct resp_res *res = qp->resp.res; - /* ODP is not supported right now. WIP. */ - if (mr->umem->is_odp) - return RESPST_ERR_UNSUPPORTED_OPCODE; - /* oA19-14, oA19-15 */ if (res && res->replay) return RESPST_ACKNOWLEDGE; @@ -706,7 +702,7 @@ static enum resp_states atomic_reply(struct rxe_qp *qp, if (!res->replay) { u64 iova = qp->resp.va + qp->resp.offset; - if (mr->umem->is_odp) + if (is_odp_mr(mr)) err = rxe_odp_atomic_op(mr, iova, pkt->opcode, atmeth_comp(pkt), atmeth_swap_add(pkt), @@ -753,7 +749,16 @@ static enum resp_states atomic_write_reply(struct rxe_qp *qp, value = *(u64 *)payload_addr(pkt); iova = qp->resp.va + qp->resp.offset; - err = rxe_mr_do_atomic_write(mr, iova, value); + /* See IBA oA19-28 */ + if (unlikely(mr->state != RXE_MR_STATE_VALID)) { + rxe_dbg_mr(mr, "mr not in valid state\n"); + return RESPST_ERR_RKEY_VIOLATION; + } + + if (is_odp_mr(mr)) + err = rxe_odp_do_atomic_write(mr, iova, value); + else + err = rxe_mr_do_atomic_write(mr, iova, value); if (err) return err; diff --git a/drivers/infiniband/sw/rxe/rxe_task.c b/drivers/infiniband/sw/rxe/rxe_task.c index 80332638d9e3..6f8f353e9583 100644 --- a/drivers/infiniband/sw/rxe/rxe_task.c +++ b/drivers/infiniband/sw/rxe/rxe_task.c @@ -85,17 +85,17 @@ static bool is_done(struct rxe_task *task) /* do_task is a wrapper for the three tasks (requester, * completer, responder) and calls them in a loop until - * they return a non-zero value. It is called either - * directly by rxe_run_task or indirectly if rxe_sched_task - * schedules the task. They must call __reserve_if_idle to - * move the task to busy before calling or scheduling. - * The task can also be moved to drained or invalid - * by calls to rxe_cleanup_task or rxe_disable_task. - * In that case tasks which get here are not executed but - * just flushed. The tasks are designed to look to see if - * there is work to do and then do part of it before returning - * here with a return value of zero until all the work - * has been consumed then it returns a non-zero value. + * they return a non-zero value. It is called indirectly + * when rxe_sched_task schedules the task. They must + * call __reserve_if_idle to move the task to busy before + * calling or scheduling. The task can also be moved to + * drained or invalid by calls to rxe_cleanup_task or + * rxe_disable_task. In that case tasks which get here + * are not executed but just flushed. The tasks are + * designed to look to see if there is work to do and + * then do part of it before returning here with a return + * value of zero until all the work has been consumed then + * it returns a non-zero value. * The number of times the task can be run is limited by * max iterations so one task cannot hold the cpu forever. * If the limit is hit and work remains the task is rescheduled. @@ -234,24 +234,6 @@ void rxe_cleanup_task(struct rxe_task *task) spin_unlock_irqrestore(&task->lock, flags); } -/* run the task inline if it is currently idle - * cannot call do_task holding the lock - */ -void rxe_run_task(struct rxe_task *task) -{ - unsigned long flags; - bool run; - - WARN_ON(rxe_read(task->qp) <= 0); - - spin_lock_irqsave(&task->lock, flags); - run = __reserve_if_idle(task); - spin_unlock_irqrestore(&task->lock, flags); - - if (run) - do_task(task); -} - /* schedule the task to run later as a work queue entry. * the queue_work call can be called holding * the lock. diff --git a/drivers/infiniband/sw/rxe/rxe_task.h b/drivers/infiniband/sw/rxe/rxe_task.h index a63e258b3d66..a8c9a77b6027 100644 --- a/drivers/infiniband/sw/rxe/rxe_task.h +++ b/drivers/infiniband/sw/rxe/rxe_task.h @@ -47,8 +47,6 @@ int rxe_init_task(struct rxe_task *task, struct rxe_qp *qp, /* cleanup task */ void rxe_cleanup_task(struct rxe_task *task); -void rxe_run_task(struct rxe_task *task); - void rxe_sched_task(struct rxe_task *task); /* keep a task from scheduling */ diff --git a/drivers/infiniband/sw/siw/Kconfig b/drivers/infiniband/sw/siw/Kconfig index ae4a953e2a03..186f182b80e7 100644 --- a/drivers/infiniband/sw/siw/Kconfig +++ b/drivers/infiniband/sw/siw/Kconfig @@ -3,6 +3,7 @@ config RDMA_SIW depends on INET && INFINIBAND depends on INFINIBAND_VIRT_DMA select CRC32 + select NET_CRC32C help This driver implements the iWARP RDMA transport over the Linux TCP/IP network stack. It enables a system with a diff --git a/drivers/infiniband/sw/siw/siw.h b/drivers/infiniband/sw/siw/siw.h index 385067e07faf..f5fd71717b80 100644 --- a/drivers/infiniband/sw/siw/siw.h +++ b/drivers/infiniband/sw/siw/siw.h @@ -693,29 +693,9 @@ static inline void siw_crc_oneshot(const void *data, size_t len, u8 out[4]) return siw_crc_final(&crc, out); } -static inline __wsum siw_csum_update(const void *buff, int len, __wsum sum) -{ - return (__force __wsum)crc32c((__force __u32)sum, buff, len); -} - -static inline __wsum siw_csum_combine(__wsum csum, __wsum csum2, int offset, - int len) -{ - return (__force __wsum)crc32c_combine((__force __u32)csum, - (__force __u32)csum2, len); -} - static inline void siw_crc_skb(struct siw_rx_stream *srx, unsigned int len) { - const struct skb_checksum_ops siw_cs_ops = { - .update = siw_csum_update, - .combine = siw_csum_combine, - }; - __wsum crc = (__force __wsum)srx->mpa_crc; - - crc = __skb_checksum(srx->skb, srx->skb_offset, len, crc, - &siw_cs_ops); - srx->mpa_crc = (__force u32)crc; + srx->mpa_crc = skb_crc32c(srx->skb, srx->skb_offset, len, srx->mpa_crc); } #define siw_dbg(ibdev, fmt, ...) \ @@ -738,7 +718,7 @@ static inline void siw_crc_skb(struct siw_rx_stream *srx, unsigned int len) "MEM[0x%08x] %s: " fmt, mem->stag, __func__, ##__VA_ARGS__) #define siw_dbg_cep(cep, fmt, ...) \ - ibdev_dbg(&cep->sdev->base_dev, "CEP[0x%pK] %s: " fmt, \ + ibdev_dbg(&cep->sdev->base_dev, "CEP[0x%p] %s: " fmt, \ cep, __func__, ##__VA_ARGS__) void siw_cq_flush(struct siw_cq *cq); diff --git a/drivers/infiniband/sw/siw/siw_cq.c b/drivers/infiniband/sw/siw/siw_cq.c index f3c2226aff94..25b3c741b66b 100644 --- a/drivers/infiniband/sw/siw/siw_cq.c +++ b/drivers/infiniband/sw/siw/siw_cq.c @@ -72,7 +72,7 @@ int siw_reap_cqe(struct siw_cq *cq, struct ib_wc *wc) wc->opcode = map_wc_opcode[cqe->opcode]; wc->status = map_cqe_status[cqe->status].ib; siw_dbg_cq(cq, - "idx %u, type %d, flags %2x, id 0x%pK\n", + "idx %u, type %d, flags %2x, id 0x%p\n", cq->cq_get % cq->num_cqe, cqe->opcode, cqe->flags, (void *)(uintptr_t)cqe->id); } else { diff --git a/drivers/infiniband/sw/siw/siw_mem.c b/drivers/infiniband/sw/siw/siw_mem.c index dcb963607c8b..d5ddeb17bd22 100644 --- a/drivers/infiniband/sw/siw/siw_mem.c +++ b/drivers/infiniband/sw/siw/siw_mem.c @@ -18,30 +18,6 @@ #define SIW_STAG_MAX_INDEX 0x00ffffff /* - * The code avoids special Stag of zero and tries to randomize - * STag values between 1 and SIW_STAG_MAX_INDEX. - */ -int siw_mem_add(struct siw_device *sdev, struct siw_mem *m) -{ - struct xa_limit limit = XA_LIMIT(1, SIW_STAG_MAX_INDEX); - u32 id, next; - - get_random_bytes(&next, 4); - next &= SIW_STAG_MAX_INDEX; - - if (xa_alloc_cyclic(&sdev->mem_xa, &id, m, limit, &next, - GFP_KERNEL) < 0) - return -ENOMEM; - - /* Set the STag index part */ - m->stag = id << 8; - - siw_dbg_mem(m, "new MEM object\n"); - - return 0; -} - -/* * siw_mem_id2obj() * * resolves memory from stag given by id. might be called from: @@ -181,10 +157,10 @@ int siw_check_mem(struct ib_pd *pd, struct siw_mem *mem, u64 addr, */ if (addr < mem->va || addr + len > mem->va + mem->len) { siw_dbg_pd(pd, "MEM interval len %d\n", len); - siw_dbg_pd(pd, "[0x%pK, 0x%pK] out of bounds\n", + siw_dbg_pd(pd, "[0x%p, 0x%p] out of bounds\n", (void *)(uintptr_t)addr, (void *)(uintptr_t)(addr + len)); - siw_dbg_pd(pd, "[0x%pK, 0x%pK] STag=0x%08x\n", + siw_dbg_pd(pd, "[0x%p, 0x%p] STag=0x%08x\n", (void *)(uintptr_t)mem->va, (void *)(uintptr_t)(mem->va + mem->len), mem->stag); diff --git a/drivers/infiniband/sw/siw/siw_mem.h b/drivers/infiniband/sw/siw/siw_mem.h index e74cfcd6dbc1..8e769d30e2ac 100644 --- a/drivers/infiniband/sw/siw/siw_mem.h +++ b/drivers/infiniband/sw/siw/siw_mem.h @@ -12,7 +12,6 @@ void siw_umem_release(struct siw_umem *umem); struct siw_pbl *siw_pbl_alloc(u32 num_buf); dma_addr_t siw_pbl_get_buffer(struct siw_pbl *pbl, u64 off, int *len, int *idx); struct siw_mem *siw_mem_id2obj(struct siw_device *sdev, int stag_index); -int siw_mem_add(struct siw_device *sdev, struct siw_mem *m); int siw_invalidate_stag(struct ib_pd *pd, u32 stag); int siw_check_mem(struct ib_pd *pd, struct siw_mem *mem, u64 addr, enum ib_access_flags perms, int len); diff --git a/drivers/infiniband/sw/siw/siw_qp_rx.c b/drivers/infiniband/sw/siw/siw_qp_rx.c index 32554eba1eac..a10820e33887 100644 --- a/drivers/infiniband/sw/siw/siw_qp_rx.c +++ b/drivers/infiniband/sw/siw/siw_qp_rx.c @@ -38,7 +38,7 @@ static int siw_rx_umem(struct siw_rx_stream *srx, struct siw_umem *umem, p = siw_get_upage(umem, dest_addr); if (unlikely(!p)) { - pr_warn("siw: %s: [QP %u]: bogus addr: %pK, %pK\n", + pr_warn("siw: %s: [QP %u]: bogus addr: %p, %p\n", __func__, qp_id(rx_qp(srx)), (void *)(uintptr_t)dest_addr, (void *)(uintptr_t)umem->fp_addr); @@ -51,7 +51,7 @@ static int siw_rx_umem(struct siw_rx_stream *srx, struct siw_umem *umem, pg_off = dest_addr & ~PAGE_MASK; bytes = min(len, (int)PAGE_SIZE - pg_off); - siw_dbg_qp(rx_qp(srx), "page %pK, bytes=%u\n", p, bytes); + siw_dbg_qp(rx_qp(srx), "page %p, bytes=%u\n", p, bytes); dest = kmap_atomic(p); rv = skb_copy_bits(srx->skb, srx->skb_offset, dest + pg_off, @@ -105,11 +105,11 @@ static int siw_rx_kva(struct siw_rx_stream *srx, void *kva, int len) { int rv; - siw_dbg_qp(rx_qp(srx), "kva: 0x%pK, len: %u\n", kva, len); + siw_dbg_qp(rx_qp(srx), "kva: 0x%p, len: %u\n", kva, len); rv = skb_copy_bits(srx->skb, srx->skb_offset, kva, len); if (unlikely(rv)) { - pr_warn("siw: [QP %u]: %s, len %d, kva 0x%pK, rv %d\n", + pr_warn("siw: [QP %u]: %s, len %d, kva 0x%p, rv %d\n", qp_id(rx_qp(srx)), __func__, len, kva, rv); return rv; diff --git a/drivers/infiniband/sw/siw/siw_verbs.c b/drivers/infiniband/sw/siw/siw_verbs.c index fd7b266a221b..2b2a7b8e93b0 100644 --- a/drivers/infiniband/sw/siw/siw_verbs.c +++ b/drivers/infiniband/sw/siw/siw_verbs.c @@ -936,7 +936,7 @@ int siw_post_send(struct ib_qp *base_qp, const struct ib_send_wr *wr, rv = -EINVAL; break; } - siw_dbg_qp(qp, "opcode %d, flags 0x%x, wr_id 0x%pK\n", + siw_dbg_qp(qp, "opcode %d, flags 0x%x, wr_id 0x%p\n", sqe->opcode, sqe->flags, (void *)(uintptr_t)sqe->id); @@ -1102,7 +1102,7 @@ int siw_post_receive(struct ib_qp *base_qp, const struct ib_recv_wr *wr, siw_dbg_qp(qp, "error %d\n", rv); *bad_wr = wr; } - return rv > 0 ? 0 : rv; + return rv; } int siw_destroy_cq(struct ib_cq *base_cq, struct ib_udata *udata) @@ -1332,7 +1332,7 @@ struct ib_mr *siw_reg_user_mr(struct ib_pd *pd, u64 start, u64 len, struct siw_device *sdev = to_siw_dev(pd->device); int rv; - siw_dbg_pd(pd, "start: 0x%pK, va: 0x%pK, len: %llu\n", + siw_dbg_pd(pd, "start: 0x%p, va: 0x%p, len: %llu\n", (void *)(uintptr_t)start, (void *)(uintptr_t)rnic_va, (unsigned long long)len); @@ -1525,7 +1525,7 @@ int siw_map_mr_sg(struct ib_mr *base_mr, struct scatterlist *sl, int num_sle, mem->len = base_mr->length; mem->va = base_mr->iova; siw_dbg_mem(mem, - "%llu bytes, start 0x%pK, %u SLE to %u entries\n", + "%llu bytes, start 0x%p, %u SLE to %u entries\n", mem->len, (void *)(uintptr_t)mem->va, num_sle, pbl->num_buf); } diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index abe0522b7df4..91f866e3fb8b 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -329,14 +329,6 @@ struct ipoib_dev_priv { unsigned long flags; - /* - * This protects access to the child_intfs list. - * To READ from child_intfs the RTNL or vlan_rwsem read side must be - * held. To WRITE RTNL and the vlan_rwsem write side must be held (in - * that order) This lock exists because we have a few contexts where - * we need the child_intfs, but do not want to grab the RTNL. - */ - struct rw_semaphore vlan_rwsem; struct mutex mcast_mutex; struct rb_root path_tree; @@ -399,6 +391,9 @@ struct ipoib_dev_priv { struct ib_event_handler event_handler; struct net_device *parent; + /* 'child_intfs' and 'list' membership of all child devices are + * protected by the netdev instance lock of 'dev'. + */ struct list_head child_intfs; struct list_head list; int child_type; @@ -512,6 +507,8 @@ int ipoib_intf_init(struct ib_device *hca, u32 port, const char *format, void ipoib_ib_dev_flush_light(struct work_struct *work); void ipoib_ib_dev_flush_normal(struct work_struct *work); void ipoib_ib_dev_flush_heavy(struct work_struct *work); +void ipoib_queue_work(struct ipoib_dev_priv *priv, + enum ipoib_flush_level level); void ipoib_ib_tx_timeout_work(struct work_struct *work); void ipoib_ib_dev_cleanup(struct net_device *dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index 5cde275daa94..10b0dbda6cd5 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -40,6 +40,7 @@ #include <linux/ip.h> #include <linux/tcp.h> +#include <net/netdev_lock.h> #include <rdma/ib_cache.h> #include "ipoib.h" @@ -781,16 +782,20 @@ static void ipoib_napi_enable(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); - napi_enable(&priv->recv_napi); - napi_enable(&priv->send_napi); + netdev_lock_ops_to_full(dev); + napi_enable_locked(&priv->recv_napi); + napi_enable_locked(&priv->send_napi); + netdev_unlock_full_to_ops(dev); } static void ipoib_napi_disable(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); - napi_disable(&priv->recv_napi); - napi_disable(&priv->send_napi); + netdev_lock_ops_to_full(dev); + napi_disable_locked(&priv->recv_napi); + napi_disable_locked(&priv->send_napi); + netdev_unlock_full_to_ops(dev); } int ipoib_ib_dev_stop_default(struct net_device *dev) @@ -1172,24 +1177,11 @@ out: } static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, - enum ipoib_flush_level level, - int nesting) + enum ipoib_flush_level level) { - struct ipoib_dev_priv *cpriv; struct net_device *dev = priv->dev; int result; - down_read_nested(&priv->vlan_rwsem, nesting); - - /* - * Flush any child interfaces too -- they might be up even if - * the parent is down. - */ - list_for_each_entry(cpriv, &priv->child_intfs, list) - __ipoib_ib_dev_flush(cpriv, level, nesting + 1); - - up_read(&priv->vlan_rwsem); - if (!test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags) && level != IPOIB_FLUSH_HEAVY) { /* Make sure the dev_addr is set even if not flushing */ @@ -1253,10 +1245,14 @@ static void __ipoib_ib_dev_flush(struct ipoib_dev_priv *priv, ipoib_ib_dev_down(dev); if (level == IPOIB_FLUSH_HEAVY) { + netdev_lock_ops(dev); if (test_bit(IPOIB_FLAG_INITIALIZED, &priv->flags)) ipoib_ib_dev_stop(dev); - if (ipoib_ib_dev_open(dev)) + result = ipoib_ib_dev_open(dev); + netdev_unlock_ops(dev); + + if (result) return; if (netif_queue_stopped(dev)) @@ -1280,7 +1276,7 @@ void ipoib_ib_dev_flush_light(struct work_struct *work) struct ipoib_dev_priv *priv = container_of(work, struct ipoib_dev_priv, flush_light); - __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_LIGHT, 0); + __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_LIGHT); } void ipoib_ib_dev_flush_normal(struct work_struct *work) @@ -1288,7 +1284,7 @@ void ipoib_ib_dev_flush_normal(struct work_struct *work) struct ipoib_dev_priv *priv = container_of(work, struct ipoib_dev_priv, flush_normal); - __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_NORMAL, 0); + __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_NORMAL); } void ipoib_ib_dev_flush_heavy(struct work_struct *work) @@ -1297,10 +1293,35 @@ void ipoib_ib_dev_flush_heavy(struct work_struct *work) container_of(work, struct ipoib_dev_priv, flush_heavy); rtnl_lock(); - __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_HEAVY, 0); + __ipoib_ib_dev_flush(priv, IPOIB_FLUSH_HEAVY); rtnl_unlock(); } +void ipoib_queue_work(struct ipoib_dev_priv *priv, + enum ipoib_flush_level level) +{ + if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) { + struct ipoib_dev_priv *cpriv; + + netdev_lock(priv->dev); + list_for_each_entry(cpriv, &priv->child_intfs, list) + ipoib_queue_work(cpriv, level); + netdev_unlock(priv->dev); + } + + switch (level) { + case IPOIB_FLUSH_LIGHT: + queue_work(ipoib_workqueue, &priv->flush_light); + break; + case IPOIB_FLUSH_NORMAL: + queue_work(ipoib_workqueue, &priv->flush_normal); + break; + case IPOIB_FLUSH_HEAVY: + queue_work(ipoib_workqueue, &priv->flush_heavy); + break; + } +} + void ipoib_ib_dev_cleanup(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 3b463db8ce39..f2f5465f2a90 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -49,6 +49,7 @@ #include <linux/jhash.h> #include <net/arp.h> #include <net/addrconf.h> +#include <net/netdev_lock.h> #include <net/pkt_sched.h> #include <linux/inetdevice.h> #include <rdma/ib_cache.h> @@ -132,6 +133,52 @@ static int ipoib_netdev_event(struct notifier_block *this, } #endif +struct ipoib_ifupdown_work { + struct work_struct work; + struct net_device *dev; + netdevice_tracker dev_tracker; + bool up; +}; + +static void ipoib_ifupdown_task(struct work_struct *work) +{ + struct ipoib_ifupdown_work *pwork = + container_of(work, struct ipoib_ifupdown_work, work); + struct net_device *dev = pwork->dev; + unsigned int flags; + + rtnl_lock(); + flags = dev->flags; + if (pwork->up) + flags |= IFF_UP; + else + flags &= ~IFF_UP; + + if (dev->flags != flags) + dev_change_flags(dev, flags, NULL); + rtnl_unlock(); + netdev_put(dev, &pwork->dev_tracker); + kfree(pwork); +} + +static void ipoib_schedule_ifupdown_task(struct net_device *dev, bool up) +{ + struct ipoib_ifupdown_work *work; + + if ((up && (dev->flags & IFF_UP)) || + (!up && !(dev->flags & IFF_UP))) + return; + + work = kmalloc(sizeof(*work), GFP_KERNEL); + if (!work) + return; + work->dev = dev; + netdev_hold(dev, &work->dev_tracker, GFP_KERNEL); + work->up = up; + INIT_WORK(&work->work, ipoib_ifupdown_task); + queue_work(ipoib_workqueue, &work->work); +} + int ipoib_open(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); @@ -154,17 +201,10 @@ int ipoib_open(struct net_device *dev) struct ipoib_dev_priv *cpriv; /* Bring up any child interfaces too */ - down_read(&priv->vlan_rwsem); - list_for_each_entry(cpriv, &priv->child_intfs, list) { - int flags; - - flags = cpriv->dev->flags; - if (flags & IFF_UP) - continue; - - dev_change_flags(cpriv->dev, flags | IFF_UP, NULL); - } - up_read(&priv->vlan_rwsem); + netdev_lock_ops_to_full(dev); + list_for_each_entry(cpriv, &priv->child_intfs, list) + ipoib_schedule_ifupdown_task(cpriv->dev, true); + netdev_unlock_full_to_ops(dev); } else if (priv->parent) { struct ipoib_dev_priv *ppriv = ipoib_priv(priv->parent); @@ -199,17 +239,10 @@ static int ipoib_stop(struct net_device *dev) struct ipoib_dev_priv *cpriv; /* Bring down any child interfaces too */ - down_read(&priv->vlan_rwsem); - list_for_each_entry(cpriv, &priv->child_intfs, list) { - int flags; - - flags = cpriv->dev->flags; - if (!(flags & IFF_UP)) - continue; - - dev_change_flags(cpriv->dev, flags & ~IFF_UP, NULL); - } - up_read(&priv->vlan_rwsem); + netdev_lock_ops_to_full(dev); + list_for_each_entry(cpriv, &priv->child_intfs, list) + ipoib_schedule_ifupdown_task(cpriv->dev, false); + netdev_unlock_full_to_ops(dev); } return 0; @@ -426,17 +459,20 @@ static int ipoib_match_gid_pkey_addr(struct ipoib_dev_priv *priv, } } + if (test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) + return matches; + /* Check child interfaces */ - down_read_nested(&priv->vlan_rwsem, nesting); + netdev_lock(priv->dev); list_for_each_entry(child_priv, &priv->child_intfs, list) { matches += ipoib_match_gid_pkey_addr(child_priv, gid, - pkey_index, addr, - nesting + 1, - found_net_dev); + pkey_index, addr, + nesting + 1, + found_net_dev); if (matches > 1) break; } - up_read(&priv->vlan_rwsem); + netdev_unlock(priv->dev); return matches; } @@ -531,9 +567,11 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) set_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); ipoib_warn(priv, "enabling connected mode " "will cause multicast packet drops\n"); + netdev_lock_ops(dev); netdev_update_features(dev); - dev_set_mtu(dev, ipoib_cm_max_mtu(dev)); + netif_set_mtu(dev, ipoib_cm_max_mtu(dev)); netif_set_real_num_tx_queues(dev, 1); + netdev_unlock_ops(dev); rtnl_unlock(); priv->tx_wr.wr.send_flags &= ~IB_SEND_IP_CSUM; @@ -543,9 +581,11 @@ int ipoib_set_mode(struct net_device *dev, const char *buf) if (!strcmp(buf, "datagram\n")) { clear_bit(IPOIB_FLAG_ADMIN_CM, &priv->flags); + netdev_lock_ops(dev); netdev_update_features(dev); - dev_set_mtu(dev, min(priv->mcast_mtu, dev->mtu)); + netif_set_mtu(dev, min(priv->mcast_mtu, dev->mtu)); netif_set_real_num_tx_queues(dev, dev->num_tx_queues); + netdev_unlock_ops(dev); rtnl_unlock(); ipoib_flush_paths(dev); return (!rtnl_trylock()) ? -EBUSY : 0; @@ -1212,6 +1252,7 @@ void ipoib_ib_tx_timeout_work(struct work_struct *work) int err; rtnl_lock(); + netdev_lock_ops(priv->dev); if (!test_bit(IPOIB_FLAG_ADMIN_UP, &priv->flags)) goto unlock; @@ -1226,6 +1267,7 @@ void ipoib_ib_tx_timeout_work(struct work_struct *work) netif_tx_wake_all_queues(priv->dev); unlock: + netdev_unlock_ops(priv->dev); rtnl_unlock(); } @@ -1992,9 +2034,9 @@ static int ipoib_ndo_init(struct net_device *ndev) dev_hold(priv->parent); - down_write(&ppriv->vlan_rwsem); + netdev_lock(priv->parent); list_add_tail(&priv->list, &ppriv->child_intfs); - up_write(&ppriv->vlan_rwsem); + netdev_unlock(priv->parent); } return 0; @@ -2004,8 +2046,6 @@ static void ipoib_ndo_uninit(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); - ASSERT_RTNL(); - /* * ipoib_remove_one guarantees the children are removed before the * parent, and that is the only place where a parent can be removed. @@ -2015,9 +2055,9 @@ static void ipoib_ndo_uninit(struct net_device *dev) if (priv->parent) { struct ipoib_dev_priv *ppriv = ipoib_priv(priv->parent); - down_write(&ppriv->vlan_rwsem); + netdev_lock(ppriv->dev); list_del(&priv->list); - up_write(&ppriv->vlan_rwsem); + netdev_unlock(ppriv->dev); } ipoib_neigh_hash_uninit(dev); @@ -2167,7 +2207,6 @@ static void ipoib_build_priv(struct net_device *dev) priv->dev = dev; spin_lock_init(&priv->lock); - init_rwsem(&priv->vlan_rwsem); mutex_init(&priv->mcast_mutex); INIT_LIST_HEAD(&priv->path_list); @@ -2372,10 +2411,10 @@ static void set_base_guid(struct ipoib_dev_priv *priv, union ib_gid *gid) netif_addr_unlock_bh(netdev); if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) { - down_read(&priv->vlan_rwsem); + netdev_lock_ops_to_full(priv->dev); list_for_each_entry(child_priv, &priv->child_intfs, list) set_base_guid(child_priv, gid); - up_read(&priv->vlan_rwsem); + netdev_unlock_full_to_ops(priv->dev); } } @@ -2415,6 +2454,14 @@ static int ipoib_set_mac(struct net_device *dev, void *addr) set_base_guid(priv, (union ib_gid *)(ss->__data + 4)); + if (!test_bit(IPOIB_FLAG_SUBINTERFACE, &priv->flags)) { + struct ipoib_dev_priv *cpriv; + + netdev_lock_ops_to_full(dev); + list_for_each_entry(cpriv, &priv->child_intfs, list) + queue_work(ipoib_workqueue, &cpriv->flush_light); + netdev_unlock_full_to_ops(dev); + } queue_work(ipoib_workqueue, &priv->flush_light); return 0; @@ -2526,7 +2573,7 @@ static struct net_device *ipoib_add_port(const char *format, ib_register_event_handler(&priv->event_handler); /* call event handler to ensure pkey in sync */ - queue_work(ipoib_workqueue, &priv->flush_heavy); + ipoib_queue_work(priv, IPOIB_FLUSH_HEAVY); ndev->rtnl_link_ops = ipoib_get_link_ops(); @@ -2624,9 +2671,11 @@ static void ipoib_remove_one(struct ib_device *device, void *client_data) rtnl_lock(); + netdev_lock(priv->dev); list_for_each_entry_safe(cpriv, tcpriv, &priv->child_intfs, list) unregister_netdevice_queue(cpriv->dev, &head); + netdev_unlock(priv->dev); unregister_netdevice_queue(priv->dev, &head); unregister_netdevice_many(&head); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c index 368e5d77416d..86983080d28b 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_verbs.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_verbs.c @@ -280,15 +280,15 @@ void ipoib_event(struct ib_event_handler *handler, dev_name(&record->device->dev), record->element.port_num); if (record->event == IB_EVENT_CLIENT_REREGISTER) { - queue_work(ipoib_workqueue, &priv->flush_light); + ipoib_queue_work(priv, IPOIB_FLUSH_LIGHT); } else if (record->event == IB_EVENT_PORT_ERR || record->event == IB_EVENT_PORT_ACTIVE || record->event == IB_EVENT_LID_CHANGE) { - queue_work(ipoib_workqueue, &priv->flush_normal); + ipoib_queue_work(priv, IPOIB_FLUSH_NORMAL); } else if (record->event == IB_EVENT_PKEY_CHANGE) { - queue_work(ipoib_workqueue, &priv->flush_heavy); + ipoib_queue_work(priv, IPOIB_FLUSH_HEAVY); } else if (record->event == IB_EVENT_GID_CHANGE && !test_bit(IPOIB_FLAG_DEV_ADDR_SET, &priv->flags)) { - queue_work(ipoib_workqueue, &priv->flush_light); + ipoib_queue_work(priv, IPOIB_FLUSH_LIGHT); } } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c index 562df2b3ef18..243e8f555eca 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_vlan.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_vlan.c @@ -53,8 +53,7 @@ static bool is_child_unique(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv) { struct ipoib_dev_priv *tpriv; - - ASSERT_RTNL(); + bool result = true; /* * Since the legacy sysfs interface uses pkey for deletion it cannot @@ -73,13 +72,17 @@ static bool is_child_unique(struct ipoib_dev_priv *ppriv, if (ppriv->pkey == priv->pkey) return false; + netdev_lock(ppriv->dev); list_for_each_entry(tpriv, &ppriv->child_intfs, list) { if (tpriv->pkey == priv->pkey && - tpriv->child_type == IPOIB_LEGACY_CHILD) - return false; + tpriv->child_type == IPOIB_LEGACY_CHILD) { + result = false; + break; + } } + netdev_unlock(ppriv->dev); - return true; + return result; } /* @@ -98,8 +101,6 @@ int __ipoib_vlan_add(struct ipoib_dev_priv *ppriv, struct ipoib_dev_priv *priv, int result; struct rdma_netdev *rn = netdev_priv(ndev); - ASSERT_RTNL(); - /* * We do not need to touch priv if register_netdevice fails, so just * always use this flow. @@ -267,6 +268,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) ppriv = ipoib_priv(pdev); rc = -ENODEV; + netdev_lock(ppriv->dev); list_for_each_entry_safe(priv, tpriv, &ppriv->child_intfs, list) { if (priv->pkey == pkey && priv->child_type == IPOIB_LEGACY_CHILD) { @@ -278,9 +280,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) goto out; } - down_write(&ppriv->vlan_rwsem); list_del_init(&priv->list); - up_write(&ppriv->vlan_rwsem); work->dev = priv->dev; INIT_WORK(&work->work, ipoib_vlan_delete_task); queue_work(ipoib_workqueue, &work->work); @@ -291,6 +291,7 @@ int ipoib_vlan_delete(struct net_device *pdev, unsigned short pkey) } out: + netdev_unlock(ppriv->dev); rtnl_unlock(); return rc; |