diff options
Diffstat (limited to 'drivers/infiniband')
60 files changed, 653 insertions, 319 deletions
diff --git a/drivers/infiniband/Kconfig b/drivers/infiniband/Kconfig index ee270e065ba9..2a972ed6851b 100644 --- a/drivers/infiniband/Kconfig +++ b/drivers/infiniband/Kconfig @@ -61,9 +61,12 @@ config INFINIBAND_ON_DEMAND_PAGING pages on demand instead. config INFINIBAND_ADDR_TRANS - bool + bool "RDMA/CM" depends on INFINIBAND default y + ---help--- + Support for RDMA communication manager (CM). + This allows for a generic connection abstraction over RDMA. config INFINIBAND_ADDR_TRANS_CONFIGFS bool diff --git a/drivers/infiniband/core/cache.c b/drivers/infiniband/core/cache.c index e337b08de2ff..ecc55e98ddd3 100644 --- a/drivers/infiniband/core/cache.c +++ b/drivers/infiniband/core/cache.c @@ -291,14 +291,18 @@ static int find_gid(struct ib_gid_table *table, const union ib_gid *gid, * so lookup free slot only if requested. */ if (pempty && empty < 0) { - if (data->props & GID_TABLE_ENTRY_INVALID) { - /* Found an invalid (free) entry; allocate it */ - if (data->props & GID_TABLE_ENTRY_DEFAULT) { - if (default_gid) - empty = curr_index; - } else { - empty = curr_index; - } + if (data->props & GID_TABLE_ENTRY_INVALID && + (default_gid == + !!(data->props & GID_TABLE_ENTRY_DEFAULT))) { + /* + * Found an invalid (free) entry; allocate it. + * If default GID is requested, then our + * found slot must be one of the DEFAULT + * reserved slots or we fail. + * This ensures that only DEFAULT reserved + * slots are used for default property GIDs. + */ + empty = curr_index; } } @@ -420,8 +424,10 @@ int ib_cache_gid_add(struct ib_device *ib_dev, u8 port, return ret; } -int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, - union ib_gid *gid, struct ib_gid_attr *attr) +static int +_ib_cache_gid_del(struct ib_device *ib_dev, u8 port, + union ib_gid *gid, struct ib_gid_attr *attr, + unsigned long mask, bool default_gid) { struct ib_gid_table *table; int ret = 0; @@ -431,11 +437,7 @@ int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, mutex_lock(&table->lock); - ix = find_gid(table, gid, attr, false, - GID_ATTR_FIND_MASK_GID | - GID_ATTR_FIND_MASK_GID_TYPE | - GID_ATTR_FIND_MASK_NETDEV, - NULL); + ix = find_gid(table, gid, attr, default_gid, mask, NULL); if (ix < 0) { ret = -EINVAL; goto out_unlock; @@ -452,6 +454,17 @@ out_unlock: return ret; } +int ib_cache_gid_del(struct ib_device *ib_dev, u8 port, + union ib_gid *gid, struct ib_gid_attr *attr) +{ + unsigned long mask = GID_ATTR_FIND_MASK_GID | + GID_ATTR_FIND_MASK_GID_TYPE | + GID_ATTR_FIND_MASK_DEFAULT | + GID_ATTR_FIND_MASK_NETDEV; + + return _ib_cache_gid_del(ib_dev, port, gid, attr, mask, false); +} + int ib_cache_gid_del_all_netdev_gids(struct ib_device *ib_dev, u8 port, struct net_device *ndev) { @@ -489,7 +502,7 @@ static int __ib_cache_gid_get(struct ib_device *ib_dev, u8 port, int index, return -EINVAL; if (table->data_vec[index].props & GID_TABLE_ENTRY_INVALID) - return -EAGAIN; + return -EINVAL; memcpy(gid, &table->data_vec[index].gid, sizeof(*gid)); if (attr) { @@ -728,7 +741,7 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, unsigned long gid_type_mask, enum ib_cache_gid_default_mode mode) { - union ib_gid gid; + union ib_gid gid = { }; struct ib_gid_attr gid_attr; struct ib_gid_table *table; unsigned int gid_type; @@ -736,7 +749,9 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, table = ib_dev->cache.ports[port - rdma_start_port(ib_dev)].gid; - make_default_gid(ndev, &gid); + mask = GID_ATTR_FIND_MASK_GID_TYPE | + GID_ATTR_FIND_MASK_DEFAULT | + GID_ATTR_FIND_MASK_NETDEV; memset(&gid_attr, 0, sizeof(gid_attr)); gid_attr.ndev = ndev; @@ -747,12 +762,12 @@ void ib_cache_gid_set_default_gid(struct ib_device *ib_dev, u8 port, gid_attr.gid_type = gid_type; if (mode == IB_CACHE_GID_DEFAULT_MODE_SET) { - mask = GID_ATTR_FIND_MASK_GID_TYPE | - GID_ATTR_FIND_MASK_DEFAULT; + make_default_gid(ndev, &gid); __ib_cache_gid_add(ib_dev, port, &gid, &gid_attr, mask, true); } else if (mode == IB_CACHE_GID_DEFAULT_MODE_DELETE) { - ib_cache_gid_del(ib_dev, port, &gid, &gid_attr); + _ib_cache_gid_del(ib_dev, port, &gid, + &gid_attr, mask, true); } } } diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 51a641002e10..a693fcd4c513 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -382,6 +382,8 @@ struct cma_hdr { #define CMA_VERSION 0x00 struct cma_req_info { + struct sockaddr_storage listen_addr_storage; + struct sockaddr_storage src_addr_storage; struct ib_device *device; int port; union ib_gid local_gid; @@ -866,7 +868,6 @@ static int cma_modify_qp_rtr(struct rdma_id_private *id_priv, { struct ib_qp_attr qp_attr; int qp_attr_mask, ret; - union ib_gid sgid; mutex_lock(&id_priv->qp_mutex); if (!id_priv->id.qp) { @@ -889,12 +890,6 @@ static int cma_modify_qp_rtr(struct rdma_id_private *id_priv, if (ret) goto out; - ret = ib_query_gid(id_priv->id.device, id_priv->id.port_num, - rdma_ah_read_grh(&qp_attr.ah_attr)->sgid_index, - &sgid, NULL); - if (ret) - goto out; - BUG_ON(id_priv->cma_dev->device != id_priv->id.device); if (conn_param) @@ -1340,11 +1335,11 @@ static bool validate_net_dev(struct net_device *net_dev, } static struct net_device *cma_get_net_dev(struct ib_cm_event *ib_event, - const struct cma_req_info *req) + struct cma_req_info *req) { - struct sockaddr_storage listen_addr_storage, src_addr_storage; - struct sockaddr *listen_addr = (struct sockaddr *)&listen_addr_storage, - *src_addr = (struct sockaddr *)&src_addr_storage; + struct sockaddr *listen_addr = + (struct sockaddr *)&req->listen_addr_storage; + struct sockaddr *src_addr = (struct sockaddr *)&req->src_addr_storage; struct net_device *net_dev; const union ib_gid *gid = req->has_gid ? &req->local_gid : NULL; int err; @@ -1359,11 +1354,6 @@ static struct net_device *cma_get_net_dev(struct ib_cm_event *ib_event, if (!net_dev) return ERR_PTR(-ENODEV); - if (!validate_net_dev(net_dev, listen_addr, src_addr)) { - dev_put(net_dev); - return ERR_PTR(-EHOSTUNREACH); - } - return net_dev; } @@ -1490,15 +1480,51 @@ static struct rdma_id_private *cma_id_from_event(struct ib_cm_id *cm_id, } } + /* + * Net namespace might be getting deleted while route lookup, + * cm_id lookup is in progress. Therefore, perform netdevice + * validation, cm_id lookup under rcu lock. + * RCU lock along with netdevice state check, synchronizes with + * netdevice migrating to different net namespace and also avoids + * case where net namespace doesn't get deleted while lookup is in + * progress. + * If the device state is not IFF_UP, its properties such as ifindex + * and nd_net cannot be trusted to remain valid without rcu lock. + * net/core/dev.c change_net_namespace() ensures to synchronize with + * ongoing operations on net device after device is closed using + * synchronize_net(). + */ + rcu_read_lock(); + if (*net_dev) { + /* + * If netdevice is down, it is likely that it is administratively + * down or it might be migrating to different namespace. + * In that case avoid further processing, as the net namespace + * or ifindex may change. + */ + if (((*net_dev)->flags & IFF_UP) == 0) { + id_priv = ERR_PTR(-EHOSTUNREACH); + goto err; + } + + if (!validate_net_dev(*net_dev, + (struct sockaddr *)&req.listen_addr_storage, + (struct sockaddr *)&req.src_addr_storage)) { + id_priv = ERR_PTR(-EHOSTUNREACH); + goto err; + } + } + bind_list = cma_ps_find(*net_dev ? dev_net(*net_dev) : &init_net, rdma_ps_from_service_id(req.service_id), cma_port_from_service_id(req.service_id)); id_priv = cma_find_listener(bind_list, cm_id, ib_event, &req, *net_dev); +err: + rcu_read_unlock(); if (IS_ERR(id_priv) && *net_dev) { dev_put(*net_dev); *net_dev = NULL; } - return id_priv; } diff --git a/drivers/infiniband/core/iwpm_util.c b/drivers/infiniband/core/iwpm_util.c index 9821ae900f6d..da12da1c36f6 100644 --- a/drivers/infiniband/core/iwpm_util.c +++ b/drivers/infiniband/core/iwpm_util.c @@ -114,7 +114,7 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, struct sockaddr_storage *mapped_sockaddr, u8 nl_client) { - struct hlist_head *hash_bucket_head; + struct hlist_head *hash_bucket_head = NULL; struct iwpm_mapping_info *map_info; unsigned long flags; int ret = -EINVAL; @@ -142,6 +142,9 @@ int iwpm_create_mapinfo(struct sockaddr_storage *local_sockaddr, } } spin_unlock_irqrestore(&iwpm_mapinfo_lock, flags); + + if (!hash_bucket_head) + kfree(map_info); return ret; } diff --git a/drivers/infiniband/core/mad.c b/drivers/infiniband/core/mad.c index c50596f7f98a..b28452a55a08 100644 --- a/drivers/infiniband/core/mad.c +++ b/drivers/infiniband/core/mad.c @@ -59,7 +59,7 @@ module_param_named(recv_queue_size, mad_recvq_size, int, 0444); MODULE_PARM_DESC(recv_queue_size, "Size of receive queue in number of work requests"); static struct list_head ib_mad_port_list; -static u32 ib_mad_client_id = 0; +static atomic_t ib_mad_client_id = ATOMIC_INIT(0); /* Port list lock */ static DEFINE_SPINLOCK(ib_mad_port_list_lock); @@ -377,7 +377,7 @@ struct ib_mad_agent *ib_register_mad_agent(struct ib_device *device, } spin_lock_irqsave(&port_priv->reg_lock, flags); - mad_agent_priv->agent.hi_tid = ++ib_mad_client_id; + mad_agent_priv->agent.hi_tid = atomic_inc_return(&ib_mad_client_id); /* * Make sure MAD registration (if supplied) diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c index cc2966380c0c..c0e4fd55e2cc 100644 --- a/drivers/infiniband/core/roce_gid_mgmt.c +++ b/drivers/infiniband/core/roce_gid_mgmt.c @@ -255,6 +255,7 @@ static void bond_delete_netdev_default_gids(struct ib_device *ib_dev, struct net_device *rdma_ndev) { struct net_device *real_dev = rdma_vlan_dev_real_dev(event_ndev); + unsigned long gid_type_mask; if (!rdma_ndev) return; @@ -264,21 +265,22 @@ static void bond_delete_netdev_default_gids(struct ib_device *ib_dev, rcu_read_lock(); - if (rdma_is_upper_dev_rcu(rdma_ndev, event_ndev) && - is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) == - BONDING_SLAVE_STATE_INACTIVE) { - unsigned long gid_type_mask; - + if (((rdma_ndev != event_ndev && + !rdma_is_upper_dev_rcu(rdma_ndev, event_ndev)) || + is_eth_active_slave_of_bonding_rcu(rdma_ndev, real_dev) + == + BONDING_SLAVE_STATE_INACTIVE)) { rcu_read_unlock(); + return; + } - gid_type_mask = roce_gid_type_mask_support(ib_dev, port); + rcu_read_unlock(); - ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev, - gid_type_mask, - IB_CACHE_GID_DEFAULT_MODE_DELETE); - } else { - rcu_read_unlock(); - } + gid_type_mask = roce_gid_type_mask_support(ib_dev, port); + + ib_cache_gid_set_default_gid(ib_dev, port, rdma_ndev, + gid_type_mask, + IB_CACHE_GID_DEFAULT_MODE_DELETE); } static void enum_netdev_ipv4_ips(struct ib_device *ib_dev, diff --git a/drivers/infiniband/core/ucma.c b/drivers/infiniband/core/ucma.c index 74329483af6d..eab43b17e9cf 100644 --- a/drivers/infiniband/core/ucma.c +++ b/drivers/infiniband/core/ucma.c @@ -159,6 +159,23 @@ static void ucma_put_ctx(struct ucma_context *ctx) complete(&ctx->comp); } +/* + * Same as ucm_get_ctx but requires that ->cm_id->device is valid, eg that the + * CM_ID is bound. + */ +static struct ucma_context *ucma_get_ctx_dev(struct ucma_file *file, int id) +{ + struct ucma_context *ctx = ucma_get_ctx(file, id); + + if (IS_ERR(ctx)) + return ctx; + if (!ctx->cm_id->device) { + ucma_put_ctx(ctx); + return ERR_PTR(-EINVAL); + } + return ctx; +} + static void ucma_close_event_id(struct work_struct *work) { struct ucma_event *uevent_close = container_of(work, struct ucma_event, close_work); @@ -683,7 +700,7 @@ static ssize_t ucma_resolve_ip(struct ucma_file *file, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - if (!rdma_addr_size_in6(&cmd.src_addr) || + if ((cmd.src_addr.sin6_family && !rdma_addr_size_in6(&cmd.src_addr)) || !rdma_addr_size_in6(&cmd.dst_addr)) return -EINVAL; @@ -734,7 +751,7 @@ static ssize_t ucma_resolve_route(struct ucma_file *file, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1050,7 +1067,7 @@ static ssize_t ucma_connect(struct ucma_file *file, const char __user *inbuf, if (!cmd.conn_param.valid) return -EINVAL; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1092,7 +1109,7 @@ static ssize_t ucma_accept(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1120,7 +1137,7 @@ static ssize_t ucma_reject(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1139,7 +1156,7 @@ static ssize_t ucma_disconnect(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); @@ -1167,15 +1184,10 @@ static ssize_t ucma_init_qp_attr(struct ucma_file *file, if (cmd.qp_state > IB_QPS_ERR) return -EINVAL; - ctx = ucma_get_ctx(file, cmd.id); + ctx = ucma_get_ctx_dev(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); - if (!ctx->cm_id->device) { - ret = -EINVAL; - goto out; - } - resp.qp_attr_mask = 0; memset(&qp_attr, 0, sizeof qp_attr); qp_attr.qp_state = cmd.qp_state; @@ -1316,13 +1328,13 @@ static ssize_t ucma_set_option(struct ucma_file *file, const char __user *inbuf, if (copy_from_user(&cmd, inbuf, sizeof(cmd))) return -EFAULT; + if (unlikely(cmd.optlen > KMALLOC_MAX_SIZE)) + return -EINVAL; + ctx = ucma_get_ctx(file, cmd.id); if (IS_ERR(ctx)) return PTR_ERR(ctx); - if (unlikely(cmd.optlen > KMALLOC_MAX_SIZE)) - return -EINVAL; - optval = memdup_user(u64_to_user_ptr(cmd.optval), cmd.optlen); if (IS_ERR(optval)) { @@ -1384,7 +1396,7 @@ static ssize_t ucma_process_join(struct ucma_file *file, else return -EINVAL; - ctx = ucma_get_ctx(file, cmd->id); + ctx = ucma_get_ctx_dev(file, cmd->id); if (IS_ERR(ctx)) return PTR_ERR(ctx); diff --git a/drivers/infiniband/core/umem.c b/drivers/infiniband/core/umem.c index 9a4e899d94b3..2b6c9b516070 100644 --- a/drivers/infiniband/core/umem.c +++ b/drivers/infiniband/core/umem.c @@ -119,7 +119,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, umem->length = size; umem->address = addr; umem->page_shift = PAGE_SHIFT; - umem->pid = get_task_pid(current, PIDTYPE_PID); /* * We ask for writable memory if any of the following * access flags are set. "Local write" and "remote write" @@ -132,7 +131,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, IB_ACCESS_REMOTE_ATOMIC | IB_ACCESS_MW_BIND)); if (access & IB_ACCESS_ON_DEMAND) { - put_pid(umem->pid); ret = ib_umem_odp_get(context, umem, access); if (ret) { kfree(umem); @@ -148,7 +146,6 @@ struct ib_umem *ib_umem_get(struct ib_ucontext *context, unsigned long addr, page_list = (struct page **) __get_free_page(GFP_KERNEL); if (!page_list) { - put_pid(umem->pid); kfree(umem); return ERR_PTR(-ENOMEM); } @@ -231,7 +228,6 @@ out: if (ret < 0) { if (need_release) __ib_umem_release(context->device, umem, 0); - put_pid(umem->pid); kfree(umem); } else current->mm->pinned_vm = locked; @@ -274,8 +270,7 @@ void ib_umem_release(struct ib_umem *umem) __ib_umem_release(umem->context->device, umem, 1); - task = get_pid_task(umem->pid, PIDTYPE_PID); - put_pid(umem->pid); + task = get_pid_task(umem->context->tgid, PIDTYPE_PID); if (!task) goto out; mm = get_task_mm(task); diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 13cb5e4deb86..21a887c9523b 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -691,6 +691,7 @@ ssize_t ib_uverbs_reg_mr(struct ib_uverbs_file *file, mr->device = pd->device; mr->pd = pd; + mr->dm = NULL; mr->uobject = uobj; atomic_inc(&pd->usecnt); mr->res.type = RDMA_RESTRACK_MR; @@ -765,6 +766,11 @@ ssize_t ib_uverbs_rereg_mr(struct ib_uverbs_file *file, mr = uobj->object; + if (mr->dm) { + ret = -EINVAL; + goto put_uobjs; + } + if (cmd.flags & IB_MR_REREG_ACCESS) { ret = ib_check_mr_access(cmd.access_flags); if (ret) diff --git a/drivers/infiniband/core/uverbs_ioctl.c b/drivers/infiniband/core/uverbs_ioctl.c index 8c93970dc8f1..8d32c4ae368c 100644 --- a/drivers/infiniband/core/uverbs_ioctl.c +++ b/drivers/infiniband/core/uverbs_ioctl.c @@ -234,6 +234,15 @@ static int uverbs_validate_kernel_mandatory(const struct uverbs_method_spec *met return -EINVAL; } + for (; i < method_spec->num_buckets; i++) { + struct uverbs_attr_spec_hash *attr_spec_bucket = + method_spec->attr_buckets[i]; + + if (!bitmap_empty(attr_spec_bucket->mandatory_attrs_bitmask, + attr_spec_bucket->num_attrs)) + return -EINVAL; + } + return 0; } diff --git a/drivers/infiniband/core/uverbs_std_types_flow_action.c b/drivers/infiniband/core/uverbs_std_types_flow_action.c index cbcec3da12f6..b4f016dfa23d 100644 --- a/drivers/infiniband/core/uverbs_std_types_flow_action.c +++ b/drivers/infiniband/core/uverbs_std_types_flow_action.c @@ -363,28 +363,28 @@ static int UVERBS_HANDLER(UVERBS_METHOD_FLOW_ACTION_ESP_MODIFY)(struct ib_device static const struct uverbs_attr_spec uverbs_flow_action_esp_keymat[] = { [IB_UVERBS_FLOW_ACTION_ESP_KEYMAT_AES_GCM] = { - .ptr = { + { .ptr = { .type = UVERBS_ATTR_TYPE_PTR_IN, UVERBS_ATTR_TYPE(struct ib_uverbs_flow_action_esp_keymat_aes_gcm), .flags = UVERBS_ATTR_SPEC_F_MIN_SZ_OR_ZERO, - }, + } }, }, }; static const struct uverbs_attr_spec uverbs_flow_action_esp_replay[] = { [IB_UVERBS_FLOW_ACTION_ESP_REPLAY_NONE] = { - .ptr = { + { .ptr = { .type = UVERBS_ATTR_TYPE_PTR_IN, /* No need to specify any data */ .len = 0, - } + } } }, [IB_UVERBS_FLOW_ACTION_ESP_REPLAY_BMP] = { - .ptr = { + { .ptr = { .type = UVERBS_ATTR_TYPE_PTR_IN, UVERBS_ATTR_STRUCT(struct ib_uverbs_flow_action_esp_replay_bmp, size), .flags = UVERBS_ATTR_SPEC_F_MIN_SZ_OR_ZERO, - } + } } }, }; diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index 7eff3aeffe01..6ddfb1fade79 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -1656,6 +1656,7 @@ struct ib_mr *ib_alloc_mr(struct ib_pd *pd, if (!IS_ERR(mr)) { mr->device = pd->device; mr->pd = pd; + mr->dm = NULL; mr->uobject = NULL; atomic_inc(&pd->usecnt); mr->need_inval = false; diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c index f6c739ec8b62..20b9f31052bf 100644 --- a/drivers/infiniband/hw/bnxt_re/main.c +++ b/drivers/infiniband/hw/bnxt_re/main.c @@ -185,12 +185,65 @@ static void bnxt_re_shutdown(void *p) bnxt_re_ib_unreg(rdev, false); } +static void bnxt_re_stop_irq(void *handle) +{ + struct bnxt_re_dev *rdev = (struct bnxt_re_dev *)handle; + struct bnxt_qplib_rcfw *rcfw = &rdev->rcfw; + struct bnxt_qplib_nq *nq; + int indx; + + for (indx = BNXT_RE_NQ_IDX; indx < rdev->num_msix; indx++) { + nq = &rdev->nq[indx - 1]; + bnxt_qplib_nq_stop_irq(nq, false); + } + + bnxt_qplib_rcfw_stop_irq(rcfw, false); +} + +static void bnxt_re_start_irq(void *handle, struct bnxt_msix_entry *ent) +{ + struct bnxt_re_dev *rdev = (struct bnxt_re_dev *)handle; + struct bnxt_msix_entry *msix_ent = rdev->msix_entries; + struct bnxt_qplib_rcfw *rcfw = &rdev->rcfw; + struct bnxt_qplib_nq *nq; + int indx, rc; + + if (!ent) { + /* Not setting the f/w timeout bit in rcfw. + * During the driver unload the first command + * to f/w will timeout and that will set the + * timeout bit. + */ + dev_err(rdev_to_dev(rdev), "Failed to re-start IRQs\n"); + return; + } + + /* Vectors may change after restart, so update with new vectors + * in device sctructure. + */ + for (indx = 0; indx < rdev->num_msix; indx++) + rdev->msix_entries[indx].vector = ent[indx].vector; + + bnxt_qplib_rcfw_start_irq(rcfw, msix_ent[BNXT_RE_AEQ_IDX].vector, + false); + for (indx = BNXT_RE_NQ_IDX ; indx < rdev->num_msix; indx++) { + nq = &rdev->nq[indx - 1]; + rc = bnxt_qplib_nq_start_irq(nq, indx - 1, + msix_ent[indx].vector, false); + if (rc) + dev_warn(rdev_to_dev(rdev), + "Failed to reinit NQ index %d\n", indx - 1); + } +} + static struct bnxt_ulp_ops bnxt_re_ulp_ops = { .ulp_async_notifier = NULL, .ulp_stop = bnxt_re_stop, .ulp_start = bnxt_re_start, .ulp_sriov_config = bnxt_re_sriov_config, - .ulp_shutdown = bnxt_re_shutdown + .ulp_shutdown = bnxt_re_shutdown, + .ulp_irq_stop = bnxt_re_stop_irq, + .ulp_irq_restart = bnxt_re_start_irq }; /* RoCE -> Net driver */ diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c index 3a78faba8d91..50d8f1fc98d5 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c @@ -336,22 +336,32 @@ static irqreturn_t bnxt_qplib_nq_irq(int irq, void *dev_instance) return IRQ_HANDLED; } +void bnxt_qplib_nq_stop_irq(struct bnxt_qplib_nq *nq, bool kill) +{ + tasklet_disable(&nq->worker); + /* Mask h/w interrupt */ + NQ_DB(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements); + /* Sync with last running IRQ handler */ + synchronize_irq(nq->vector); + if (kill) + tasklet_kill(&nq->worker); + if (nq->requested) { + irq_set_affinity_hint(nq->vector, NULL); + free_irq(nq->vector, nq); + nq->requested = false; + } +} + void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq) { if (nq->cqn_wq) { destroy_workqueue(nq->cqn_wq); nq->cqn_wq = NULL; } + /* Make sure the HW is stopped! */ - synchronize_irq(nq->vector); - tasklet_disable(&nq->worker); - tasklet_kill(&nq->worker); + bnxt_qplib_nq_stop_irq(nq, true); - if (nq->requested) { - irq_set_affinity_hint(nq->vector, NULL); - free_irq(nq->vector, nq); - nq->requested = false; - } if (nq->bar_reg_iomem) iounmap(nq->bar_reg_iomem); nq->bar_reg_iomem = NULL; @@ -361,6 +371,40 @@ void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq) nq->vector = 0; } +int bnxt_qplib_nq_start_irq(struct bnxt_qplib_nq *nq, int nq_indx, + int msix_vector, bool need_init) +{ + int rc; + + if (nq->requested) + return -EFAULT; + + nq->vector = msix_vector; + if (need_init) + tasklet_init(&nq->worker, bnxt_qplib_service_nq, + (unsigned long)nq); + else + tasklet_enable(&nq->worker); + + snprintf(nq->name, sizeof(nq->name), "bnxt_qplib_nq-%d", nq_indx); + rc = request_irq(nq->vector, bnxt_qplib_nq_irq, 0, nq->name, nq); + if (rc) + return rc; + + cpumask_clear(&nq->mask); + cpumask_set_cpu(nq_indx, &nq->mask); + rc = irq_set_affinity_hint(nq->vector, &nq->mask); + if (rc) { + dev_warn(&nq->pdev->dev, + "QPLIB: set affinity failed; vector: %d nq_idx: %d\n", + nq->vector, nq_indx); + } + nq->requested = true; + NQ_DB_REARM(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements); + + return rc; +} + int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq, int nq_idx, int msix_vector, int bar_reg_offset, int (*cqn_handler)(struct bnxt_qplib_nq *nq, @@ -372,41 +416,17 @@ int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq, resource_size_t nq_base; int rc = -1; - nq->pdev = pdev; - nq->vector = msix_vector; if (cqn_handler) nq->cqn_handler = cqn_handler; if (srqn_handler) nq->srqn_handler = srqn_handler; - tasklet_init(&nq->worker, bnxt_qplib_service_nq, (unsigned long)nq); - /* Have a task to schedule CQ notifiers in post send case */ nq->cqn_wq = create_singlethread_workqueue("bnxt_qplib_nq"); if (!nq->cqn_wq) - goto fail; - - nq->requested = false; - memset(nq->name, 0, 32); - sprintf(nq->name, "bnxt_qplib_nq-%d", nq_idx); - rc = request_irq(nq->vector, bnxt_qplib_nq_irq, 0, nq->name, nq); - if (rc) { - dev_err(&nq->pdev->dev, - "Failed to request IRQ for NQ: %#x", rc); - goto fail; - } - - cpumask_clear(&nq->mask); - cpumask_set_cpu(nq_idx, &nq->mask); - rc = irq_set_affinity_hint(nq->vector, &nq->mask); - if (rc) { - dev_warn(&nq->pdev->dev, - "QPLIB: set affinity failed; vector: %d nq_idx: %d\n", - nq->vector, nq_idx); - } + return -ENOMEM; - nq->requested = true; nq->bar_reg = NQ_CONS_PCI_BAR_REGION; nq->bar_reg_off = bar_reg_offset; nq_base = pci_resource_start(pdev, nq->bar_reg); @@ -419,7 +439,13 @@ int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq, rc = -ENOMEM; goto fail; } - NQ_DB_REARM(nq->bar_reg_iomem, nq->hwq.cons, nq->hwq.max_elements); + + rc = bnxt_qplib_nq_start_irq(nq, nq_idx, msix_vector, true); + if (rc) { + dev_err(&nq->pdev->dev, + "QPLIB: Failed to request irq for nq-idx %d", nq_idx); + goto fail; + } return 0; fail: diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h index ade9f13c0fd1..72352ca80ace 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h @@ -467,7 +467,10 @@ struct bnxt_qplib_nq_work { struct bnxt_qplib_cq *cq; }; +void bnxt_qplib_nq_stop_irq(struct bnxt_qplib_nq *nq, bool kill); void bnxt_qplib_disable_nq(struct bnxt_qplib_nq *nq); +int bnxt_qplib_nq_start_irq(struct bnxt_qplib_nq *nq, int nq_indx, + int msix_vector, bool need_init); int bnxt_qplib_enable_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq, int nq_idx, int msix_vector, int bar_reg_offset, int (*cqn_handler)(struct bnxt_qplib_nq *nq, diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c index 80027a494730..2852d350ada1 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.c @@ -582,19 +582,29 @@ fail: return -ENOMEM; } -void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw) +void bnxt_qplib_rcfw_stop_irq(struct bnxt_qplib_rcfw *rcfw, bool kill) { - unsigned long indx; - - /* Make sure the HW channel is stopped! */ - synchronize_irq(rcfw->vector); tasklet_disable(&rcfw->worker); - tasklet_kill(&rcfw->worker); + /* Mask h/w interrupts */ + CREQ_DB(rcfw->creq_bar_reg_iomem, rcfw->creq.cons, + rcfw->creq.max_elements); + /* Sync with last running IRQ-handler */ + synchronize_irq(rcfw->vector); + if (kill) + tasklet_kill(&rcfw->worker); if (rcfw->requested) { free_irq(rcfw->vector, rcfw); rcfw->requested = false; } +} + +void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw) +{ + unsigned long indx; + + bnxt_qplib_rcfw_stop_irq(rcfw, true); + if (rcfw->cmdq_bar_reg_iomem) iounmap(rcfw->cmdq_bar_reg_iomem); rcfw->cmdq_bar_reg_iomem = NULL; @@ -614,6 +624,31 @@ void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw) rcfw->vector = 0; } +int bnxt_qplib_rcfw_start_irq(struct bnxt_qplib_rcfw *rcfw, int msix_vector, + bool need_init) +{ + int rc; + + if (rcfw->requested) + return -EFAULT; + + rcfw->vector = msix_vector; + if (need_init) + tasklet_init(&rcfw->worker, + bnxt_qplib_service_creq, (unsigned long)rcfw); + else + tasklet_enable(&rcfw->worker); + rc = request_irq(rcfw->vector, bnxt_qplib_creq_irq, 0, + "bnxt_qplib_creq", rcfw); + if (rc) + return rc; + rcfw->requested = true; + CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, rcfw->creq.cons, + rcfw->creq.max_elements); + + return 0; +} + int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev, struct bnxt_qplib_rcfw *rcfw, int msix_vector, @@ -675,27 +710,17 @@ int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev, rcfw->creq_qp_event_processed = 0; rcfw->creq_func_event_processed = 0; - rcfw->vector = msix_vector; if (aeq_handler) rcfw->aeq_handler = aeq_handler; + init_waitqueue_head(&rcfw->waitq); - tasklet_init(&rcfw->worker, bnxt_qplib_service_creq, - (unsigned long)rcfw); - - rcfw->requested = false; - rc = request_irq(rcfw->vector, bnxt_qplib_creq_irq, 0, - "bnxt_qplib_creq", rcfw); + rc = bnxt_qplib_rcfw_start_irq(rcfw, msix_vector, true); if (rc) { dev_err(&rcfw->pdev->dev, "QPLIB: Failed to request IRQ for CREQ rc = 0x%x", rc); bnxt_qplib_disable_rcfw_channel(rcfw); return rc; } - rcfw->requested = true; - - init_waitqueue_head(&rcfw->waitq); - - CREQ_DB_REARM(rcfw->creq_bar_reg_iomem, 0, rcfw->creq.max_elements); init.cmdq_pbl = cpu_to_le64(rcfw->cmdq.pbl[PBL_LVL_0].pg_map_arr[0]); init.cmdq_size_cmdq_lvl = cpu_to_le16( diff --git a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h index c7cce2e4185e..46416dfe8830 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_rcfw.h @@ -195,7 +195,10 @@ struct bnxt_qplib_rcfw { void bnxt_qplib_free_rcfw_channel(struct bnxt_qplib_rcfw *rcfw); int bnxt_qplib_alloc_rcfw_channel(struct pci_dev *pdev, struct bnxt_qplib_rcfw *rcfw, int qp_tbl_sz); +void bnxt_qplib_rcfw_stop_irq(struct bnxt_qplib_rcfw *rcfw, bool kill); void bnxt_qplib_disable_rcfw_channel(struct bnxt_qplib_rcfw *rcfw); +int bnxt_qplib_rcfw_start_irq(struct bnxt_qplib_rcfw *rcfw, int msix_vector, + bool need_init); int bnxt_qplib_enable_rcfw_channel(struct pci_dev *pdev, struct bnxt_qplib_rcfw *rcfw, int msix_vector, diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 6f2b26126c64..2be2e1ac1b5f 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -315,7 +315,7 @@ static void advance_oldest_read(struct t4_wq *wq) * Deal with out-of-order and/or completions that complete * prior unsignalled WRs. */ -void c4iw_flush_hw_cq(struct c4iw_cq *chp) +void c4iw_flush_hw_cq(struct c4iw_cq *chp, struct c4iw_qp *flush_qhp) { struct t4_cqe *hw_cqe, *swcqe, read_cqe; struct c4iw_qp *qhp; @@ -339,6 +339,13 @@ void c4iw_flush_hw_cq(struct c4iw_cq *chp) if (qhp == NULL) goto next_cqe; + if (flush_qhp != qhp) { + spin_lock(&qhp->lock); + + if (qhp->wq.flushed == 1) + goto next_cqe; + } + if (CQE_OPCODE(hw_cqe) == FW_RI_TERMINATE) goto next_cqe; @@ -390,6 +397,8 @@ void c4iw_flush_hw_cq(struct c4iw_cq *chp) next_cqe: t4_hwcq_consume(&chp->cq); ret = t4_next_hw_cqe(&chp->cq, &hw_cqe); + if (qhp && flush_qhp != qhp) + spin_unlock(&qhp->lock); } } diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index feeb8ee6f4a2..44161ca4d2a8 100644 --- a/drivers/infiniband/hw/cxgb4/device.c +++ b/drivers/infiniband/hw/cxgb4/device.c @@ -875,6 +875,11 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev) rdev->status_page->db_off = 0; + init_completion(&rdev->rqt_compl); + init_completion(&rdev->pbl_compl); + kref_init(&rdev->rqt_kref); + kref_init(&rdev->pbl_kref); + return 0; err_free_status_page_and_wr_log: if (c4iw_wr_log && rdev->wr_log) @@ -893,13 +898,15 @@ destroy_resource: static void c4iw_rdev_close(struct c4iw_rdev *rdev) { - destroy_workqueue(rdev->free_workq); kfree(rdev->wr_log); c4iw_release_dev_ucontext(rdev, &rdev->uctx); free_page((unsigned long)rdev->status_page); c4iw_pblpool_destroy(rdev); c4iw_rqtpool_destroy(rdev); + wait_for_completion(&rdev->pbl_compl); + wait_for_completion(&rdev->rqt_compl); c4iw_ocqp_pool_destroy(rdev); + destroy_workqueue(rdev->free_workq); c4iw_destroy_resource(&rdev->resource); } diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h index cc929002c05e..831027717121 100644 --- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h +++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h @@ -185,6 +185,10 @@ struct c4iw_rdev { struct wr_log_entry *wr_log; int wr_log_size; struct workqueue_struct *free_workq; + struct completion rqt_compl; + struct completion pbl_compl; + struct kref rqt_kref; + struct kref pbl_kref; }; static inline int c4iw_fatal_error(struct c4iw_rdev *rdev) @@ -1049,7 +1053,7 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size); void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size); u32 c4iw_ocqp_pool_alloc(struct c4iw_rdev *rdev, int size); void c4iw_ocqp_pool_free(struct c4iw_rdev *rdev, u32 addr, int size); -void c4iw_flush_hw_cq(struct c4iw_cq *chp); +void c4iw_flush_hw_cq(struct c4iw_cq *chp, struct c4iw_qp *flush_qhp); void c4iw_count_rcqes(struct t4_cq *cq, struct t4_wq *wq, int *count); int c4iw_ep_disconnect(struct c4iw_ep *ep, int abrupt, gfp_t gfp); int c4iw_flush_rq(struct t4_wq *wq, struct t4_cq *cq, int count); diff --git a/drivers/infiniband/hw/cxgb4/mem.c b/drivers/infiniband/hw/cxgb4/mem.c index e90f2fd8dc16..1445918e3239 100644 --- a/drivers/infiniband/hw/cxgb4/mem.c +++ b/drivers/infiniband/hw/cxgb4/mem.c @@ -489,10 +489,10 @@ struct ib_mr *c4iw_get_dma_mr(struct ib_pd *pd, int acc) err_dereg_mem: dereg_mem(&rhp->rdev, mhp->attr.stag, mhp->attr.pbl_size, mhp->attr.pbl_addr, mhp->dereg_skb, mhp->wr_waitp); -err_free_wr_wait: - c4iw_put_wr_wait(mhp->wr_waitp); err_free_skb: kfree_skb(mhp->dereg_skb); +err_free_wr_wait: + c4iw_put_wr_wait(mhp->wr_waitp); err_free_mhp: kfree(mhp); return ERR_PTR(ret); diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index de77b6027d69..ae167b686608 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -1343,12 +1343,12 @@ static void __flush_qp(struct c4iw_qp *qhp, struct c4iw_cq *rchp, qhp->wq.flushed = 1; t4_set_wq_in_error(&qhp->wq); - c4iw_flush_hw_cq(rchp); + c4iw_flush_hw_cq(rchp, qhp); c4iw_count_rcqes(&rchp->cq, &qhp->wq, &count); rq_flushed = c4iw_flush_rq(&qhp->wq, &rchp->cq, count); if (schp != rchp) - c4iw_flush_hw_cq(schp); + c4iw_flush_hw_cq(schp, qhp); sq_flushed = c4iw_flush_sq(qhp); spin_unlock(&qhp->lock); diff --git a/drivers/infiniband/hw/cxgb4/resource.c b/drivers/infiniband/hw/cxgb4/resource.c index 3cf25997ed2b..0ef25ae05e6f 100644 --- a/drivers/infiniband/hw/cxgb4/resource.c +++ b/drivers/infiniband/hw/cxgb4/resource.c @@ -260,12 +260,22 @@ u32 c4iw_pblpool_alloc(struct c4iw_rdev *rdev, int size) rdev->stats.pbl.cur += roundup(size, 1 << MIN_PBL_SHIFT); if (rdev->stats.pbl.cur > rdev->stats.pbl.max) rdev->stats.pbl.max = rdev->stats.pbl.cur; + kref_get(&rdev->pbl_kref); } else rdev->stats.pbl.fail++; mutex_unlock(&rdev->stats.lock); return (u32)addr; } +static void destroy_pblpool(struct kref *kref) +{ + struct c4iw_rdev *rdev; + + rdev = container_of(kref, struct c4iw_rdev, pbl_kref); + gen_pool_destroy(rdev->pbl_pool); + complete(&rdev->pbl_compl); +} + void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { pr_debug("addr 0x%x size %d\n", addr, size); @@ -273,6 +283,7 @@ void c4iw_pblpool_free(struct c4iw_rdev *rdev, u32 addr, int size) rdev->stats.pbl.cur -= roundup(size, 1 << MIN_PBL_SHIFT); mutex_unlock(&rdev->stats.lock); gen_pool_free(rdev->pbl_pool, (unsigned long)addr, size); + kref_put(&rdev->pbl_kref, destroy_pblpool); } int c4iw_pblpool_create(struct c4iw_rdev *rdev) @@ -310,7 +321,7 @@ int c4iw_pblpool_create(struct c4iw_rdev *rdev) void c4iw_pblpool_destroy(struct c4iw_rdev *rdev) { - gen_pool_destroy(rdev->pbl_pool); + kref_put(&rdev->pbl_kref, destroy_pblpool); } /* @@ -331,12 +342,22 @@ u32 c4iw_rqtpool_alloc(struct c4iw_rdev *rdev, int size) rdev->stats.rqt.cur += roundup(size << 6, 1 << MIN_RQT_SHIFT); if (rdev->stats.rqt.cur > rdev->stats.rqt.max) rdev->stats.rqt.max = rdev->stats.rqt.cur; + kref_get(&rdev->rqt_kref); } else rdev->stats.rqt.fail++; mutex_unlock(&rdev->stats.lock); return (u32)addr; } +static void destroy_rqtpool(struct kref *kref) +{ + struct c4iw_rdev *rdev; + + rdev = container_of(kref, struct c4iw_rdev, rqt_kref); + gen_pool_destroy(rdev->rqt_pool); + complete(&rdev->rqt_compl); +} + void c4iw_rqtpool_free(struct c4iw_rdev *rdev, u32 addr, int size) { pr_debug("addr 0x%x size %d\n", addr, size << 6); @@ -344,6 +365,7 @@ void c4iw_rqtpool_free(struct c4iw_rdev *rdev, u32 addr, int size) rdev->stats.rqt.cur -= roundup(size << 6, 1 << MIN_RQT_SHIFT); mutex_unlock(&rdev->stats.lock); gen_pool_free(rdev->rqt_pool, (unsigned long)addr, size << 6); + kref_put(&rdev->rqt_kref, destroy_rqtpool); } int c4iw_rqtpool_create(struct c4iw_rdev *rdev) @@ -380,7 +402,7 @@ int c4iw_rqtpool_create(struct c4iw_rdev *rdev) void c4iw_rqtpool_destroy(struct c4iw_rdev *rdev) { - gen_pool_destroy(rdev->rqt_pool); + kref_put(&rdev->rqt_kref, destroy_rqtpool); } /* diff --git a/drivers/infiniband/hw/hfi1/affinity.c b/drivers/infiniband/hw/hfi1/affinity.c index a97055dd4fbd..b5fab55cc275 100644 --- a/drivers/infiniband/hw/hfi1/affinity.c +++ b/drivers/infiniband/hw/hfi1/affinity.c @@ -412,7 +412,6 @@ static void hfi1_cleanup_sdma_notifier(struct hfi1_msix_entry *msix) static int get_irq_affinity(struct hfi1_devdata *dd, struct hfi1_msix_entry *msix) { - int ret; cpumask_var_t diff; struct hfi1_affinity_node *entry; struct cpu_mask_set *set = NULL; @@ -424,10 +423,6 @@ static int get_irq_affinity(struct hfi1_devdata *dd, extra[0] = '\0'; cpumask_clear(&msix->mask); - ret = zalloc_cpumask_var(&diff, GFP_KERNEL); - if (!ret) - return -ENOMEM; - entry = node_affinity_lookup(dd->node); switch (msix->type) { @@ -458,6 +453,9 @@ static int get_irq_affinity(struct hfi1_devdata *dd, * finds its CPU here. */ if (cpu == -1 && set) { + if (!zalloc_cpumask_var(&diff, GFP_KERNEL)) + return -ENOMEM; + if (cpumask_equal(&set->mask, &set->used)) { /* * We've used up all the CPUs, bump up the generation @@ -469,6 +467,8 @@ static int get_irq_affinity(struct hfi1_devdata *dd, cpumask_andnot(diff, &set->mask, &set->used); cpu = cpumask_first(diff); cpumask_set_cpu(cpu, &set->used); + + free_cpumask_var(diff); } cpumask_set_cpu(cpu, &msix->mask); @@ -482,7 +482,6 @@ static int get_irq_affinity(struct hfi1_devdata *dd, hfi1_setup_sdma_notifier(msix); } - free_cpumask_var(diff); return 0; } diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index e6a60fa59f2b..e6bdd0c1e80a 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -5944,6 +5944,7 @@ static void is_sendctxt_err_int(struct hfi1_devdata *dd, u64 status; u32 sw_index; int i = 0; + unsigned long irq_flags; sw_index = dd->hw_to_sw[hw_context]; if (sw_index >= dd->num_send_contexts) { @@ -5953,10 +5954,12 @@ static void is_sendctxt_err_int(struct hfi1_devdata *dd, return; } sci = &dd->send_contexts[sw_index]; + spin_lock_irqsave(&dd->sc_lock, irq_flags); sc = sci->sc; if (!sc) { dd_dev_err(dd, "%s: context %u(%u): no sc?\n", __func__, sw_index, hw_context); + spin_unlock_irqrestore(&dd->sc_lock, irq_flags); return; } @@ -5978,6 +5981,7 @@ static void is_sendctxt_err_int(struct hfi1_devdata *dd, */ if (sc->type != SC_USER) queue_work(dd->pport->hfi1_wq, &sc->halt_work); + spin_unlock_irqrestore(&dd->sc_lock, irq_flags); /* * Update the counters for the corresponding status bits. diff --git a/drivers/infiniband/hw/hfi1/driver.c b/drivers/infiniband/hw/hfi1/driver.c index 46d1475b2154..bd837a048bf4 100644 --- a/drivers/infiniband/hw/hfi1/driver.c +++ b/drivers/infiniband/hw/hfi1/driver.c @@ -433,31 +433,43 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, bool do_cnp) { struct hfi1_ibport *ibp = to_iport(qp->ibqp.device, qp->port_num); + struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); struct ib_other_headers *ohdr = pkt->ohdr; struct ib_grh *grh = pkt->grh; u32 rqpn = 0, bth1; - u16 pkey, rlid, dlid = ib_get_dlid(pkt->hdr); + u16 pkey; + u32 rlid, slid, dlid = 0; u8 hdr_type, sc, svc_type; bool is_mcast = false; + /* can be called from prescan */ if (pkt->etype == RHF_RCV_TYPE_BYPASS) { is_mcast = hfi1_is_16B_mcast(dlid); pkey = hfi1_16B_get_pkey(pkt->hdr); sc = hfi1_16B_get_sc(pkt->hdr); + dlid = hfi1_16B_get_dlid(pkt->hdr); + slid = hfi1_16B_get_slid(pkt->hdr); hdr_type = HFI1_PKT_TYPE_16B; } else { is_mcast = (dlid > be16_to_cpu(IB_MULTICAST_LID_BASE)) && (dlid != be16_to_cpu(IB_LID_PERMISSIVE)); pkey = ib_bth_get_pkey(ohdr); sc = hfi1_9B_get_sc5(pkt->hdr, pkt->rhf); + dlid = ib_get_dlid(pkt->hdr); + slid = ib_get_slid(pkt->hdr); hdr_type = HFI1_PKT_TYPE_9B; } switch (qp->ibqp.qp_type) { + case IB_QPT_UD: + dlid = ppd->lid; + rlid = slid; + rqpn = ib_get_sqpn(pkt->ohdr); + svc_type = IB_CC_SVCTYPE_UD; + break; case IB_QPT_SMI: case IB_QPT_GSI: - case IB_QPT_UD: - rlid = ib_get_slid(pkt->hdr); + rlid = slid; rqpn = ib_get_sqpn(pkt->ohdr); svc_type = IB_CC_SVCTYPE_UD; break; @@ -482,7 +494,6 @@ void hfi1_process_ecn_slowpath(struct rvt_qp *qp, struct hfi1_packet *pkt, dlid, rlid, sc, grh); if (!is_mcast && (bth1 & IB_BECN_SMASK)) { - struct hfi1_pportdata *ppd = ppd_from_ibp(ibp); u32 lqpn = bth1 & RVT_QPN_MASK; u8 sl = ibp->sc_to_sl[sc]; diff --git a/drivers/infiniband/hw/hfi1/hfi.h b/drivers/infiniband/hw/hfi1/hfi.h index 32c48265405e..cac2c62bc42d 100644 --- a/drivers/infiniband/hw/hfi1/hfi.h +++ b/drivers/infiniband/hw/hfi1/hfi.h @@ -1537,13 +1537,13 @@ void set_link_ipg(struct hfi1_pportdata *ppd); void process_becn(struct hfi1_pportdata *ppd, u8 sl, u32 rlid, u32 lqpn, u32 rqpn, u8 svc_type); void return_cnp(struct hfi1_ibport *ibp, struct rvt_qp *qp, u32 remote_qpn, - u32 pkey, u32 slid, u32 dlid, u8 sc5, + u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh); void return_cnp_16B(struct hfi1_ibport *ibp, struct rvt_qp *qp, - u32 remote_qpn, u32 pkey, u32 slid, u32 dlid, + u32 remote_qpn, u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh); typedef void (*hfi1_handle_cnp)(struct hfi1_ibport *ibp, struct rvt_qp *qp, - u32 remote_qpn, u32 pkey, u32 slid, u32 dlid, + u32 remote_qpn, u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh); #define PKEY_CHECK_INVALID -1 @@ -2437,7 +2437,7 @@ static inline void hfi1_make_16b_hdr(struct hfi1_16b_header *hdr, ((slid >> OPA_16B_SLID_SHIFT) << OPA_16B_SLID_HIGH_SHIFT); lrh2 = (lrh2 & ~OPA_16B_DLID_MASK) | ((dlid >> OPA_16B_DLID_SHIFT) << OPA_16B_DLID_HIGH_SHIFT); - lrh2 = (lrh2 & ~OPA_16B_PKEY_MASK) | (pkey << OPA_16B_PKEY_SHIFT); + lrh2 = (lrh2 & ~OPA_16B_PKEY_MASK) | ((u32)pkey << OPA_16B_PKEY_SHIFT); lrh2 = (lrh2 & ~OPA_16B_L4_MASK) | l4; hdr->lrh[0] = lrh0; diff --git a/drivers/infiniband/hw/hfi1/init.c b/drivers/infiniband/hw/hfi1/init.c index 33eba2356742..6309edf811df 100644 --- a/drivers/infiniband/hw/hfi1/init.c +++ b/drivers/infiniband/hw/hfi1/init.c @@ -88,9 +88,9 @@ * pio buffers per ctxt, etc.) Zero means use one user context per CPU. */ int num_user_contexts = -1; -module_param_named(num_user_contexts, num_user_contexts, uint, S_IRUGO); +module_param_named(num_user_contexts, num_user_contexts, int, 0444); MODULE_PARM_DESC( - num_user_contexts, "Set max number of user contexts to use"); + num_user_contexts, "Set max number of user contexts to use (default: -1 will use the real (non-HT) CPU count)"); uint krcvqs[RXE_NUM_DATA_VL]; int krcvqsset; @@ -1209,30 +1209,49 @@ static void finalize_asic_data(struct hfi1_devdata *dd, kfree(ad); } -static void __hfi1_free_devdata(struct kobject *kobj) +/** + * hfi1_clean_devdata - cleans up per-unit data structure + * @dd: pointer to a valid devdata structure + * + * It cleans up all data structures set up by + * by hfi1_alloc_devdata(). + */ +static void hfi1_clean_devdata(struct hfi1_devdata *dd) { - struct hfi1_devdata *dd = - container_of(kobj, struct hfi1_devdata, kobj); struct hfi1_asic_data *ad; unsigned long flags; spin_lock_irqsave(&hfi1_devs_lock, flags); - idr_remove(&hfi1_unit_table, dd->unit); - list_del(&dd->list); + if (!list_empty(&dd->list)) { + idr_remove(&hfi1_unit_table, dd->unit); + list_del_init(&dd->list); + } ad = release_asic_data(dd); spin_unlock_irqrestore(&hfi1_devs_lock, flags); - if (ad) - finalize_asic_data(dd, ad); + + finalize_asic_data(dd, ad); free_platform_config(dd); rcu_barrier(); /* wait for rcu callbacks to complete */ free_percpu(dd->int_counter); free_percpu(dd->rcv_limit); free_percpu(dd->send_schedule); free_percpu(dd->tx_opstats); + dd->int_counter = NULL; + dd->rcv_limit = NULL; + dd->send_schedule = NULL; + dd->tx_opstats = NULL; sdma_clean(dd, dd->num_sdma); rvt_dealloc_device(&dd->verbs_dev.rdi); } +static void __hfi1_free_devdata(struct kobject *kobj) +{ + struct hfi1_devdata *dd = + container_of(kobj, struct hfi1_devdata, kobj); + + hfi1_clean_devdata(dd); +} + static struct kobj_type hfi1_devdata_type = { .release = __hfi1_free_devdata, }; @@ -1265,6 +1284,8 @@ struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev, size_t extra) return ERR_PTR(-ENOMEM); dd->num_pports = nports; dd->pport = (struct hfi1_pportdata *)(dd + 1); + dd->pcidev = pdev; + pci_set_drvdata(pdev, dd); INIT_LIST_HEAD(&dd->list); idr_preload(GFP_KERNEL); @@ -1331,9 +1352,7 @@ struct hfi1_devdata *hfi1_alloc_devdata(struct pci_dev *pdev, size_t extra) return dd; bail: - if (!list_empty(&dd->list)) - list_del_init(&dd->list); - rvt_dealloc_device(&dd->verbs_dev.rdi); + hfi1_clean_devdata(dd); return ERR_PTR(ret); } diff --git a/drivers/infiniband/hw/hfi1/pcie.c b/drivers/infiniband/hw/hfi1/pcie.c index 83d66e862207..c1c982908b4b 100644 --- a/drivers/infiniband/hw/hfi1/pcie.c +++ b/drivers/infiniband/hw/hfi1/pcie.c @@ -163,9 +163,6 @@ int hfi1_pcie_ddinit(struct hfi1_devdata *dd, struct pci_dev *pdev) resource_size_t addr; int ret = 0; - dd->pcidev = pdev; - pci_set_drvdata(pdev, dd); - addr = pci_resource_start(pdev, 0); len = pci_resource_len(pdev, 0); diff --git a/drivers/infiniband/hw/hfi1/platform.c b/drivers/infiniband/hw/hfi1/platform.c index d486355880cb..cbf7faa5038c 100644 --- a/drivers/infiniband/hw/hfi1/platform.c +++ b/drivers/infiniband/hw/hfi1/platform.c @@ -199,6 +199,7 @@ void free_platform_config(struct hfi1_devdata *dd) { /* Release memory allocated for eprom or fallback file read. */ kfree(dd->platform_config.data); + dd->platform_config.data = NULL; } void get_port_type(struct hfi1_pportdata *ppd) diff --git a/drivers/infiniband/hw/hfi1/qsfp.c b/drivers/infiniband/hw/hfi1/qsfp.c index 1869f639c3ae..b5966991d647 100644 --- a/drivers/infiniband/hw/hfi1/qsfp.c +++ b/drivers/infiniband/hw/hfi1/qsfp.c @@ -204,6 +204,8 @@ static void clean_i2c_bus(struct hfi1_i2c_bus *bus) void clean_up_i2c(struct hfi1_devdata *dd, struct hfi1_asic_data *ad) { + if (!ad) + return; clean_i2c_bus(ad->i2c_bus0); ad->i2c_bus0 = NULL; clean_i2c_bus(ad->i2c_bus1); diff --git a/drivers/infiniband/hw/hfi1/ruc.c b/drivers/infiniband/hw/hfi1/ruc.c index 3daa94bdae3a..c0071ca4147a 100644 --- a/drivers/infiniband/hw/hfi1/ruc.c +++ b/drivers/infiniband/hw/hfi1/ruc.c @@ -733,6 +733,20 @@ static inline void hfi1_make_ruc_bth(struct rvt_qp *qp, ohdr->bth[2] = cpu_to_be32(bth2); } +/** + * hfi1_make_ruc_header_16B - build a 16B header + * @qp: the queue pair + * @ohdr: a pointer to the destination header memory + * @bth0: bth0 passed in from the RC/UC builder + * @bth2: bth2 passed in from the RC/UC builder + * @middle: non zero implies indicates ahg "could" be used + * @ps: the current packet state + * + * This routine may disarm ahg under these situations: + * - packet needs a GRH + * - BECN needed + * - migration state not IB_MIG_MIGRATED + */ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth0, u32 bth2, int middle, @@ -777,6 +791,12 @@ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, else middle = 0; + if (qp->s_flags & RVT_S_ECN) { + qp->s_flags &= ~RVT_S_ECN; + /* we recently received a FECN, so return a BECN */ + becn = true; + middle = 0; + } if (middle) build_ahg(qp, bth2); else @@ -784,11 +804,6 @@ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, bth0 |= pkey; bth0 |= extra_bytes << 20; - if (qp->s_flags & RVT_S_ECN) { - qp->s_flags &= ~RVT_S_ECN; - /* we recently received a FECN, so return a BECN */ - becn = true; - } hfi1_make_ruc_bth(qp, ohdr, bth0, bth1, bth2); if (!ppd->lid) @@ -806,6 +821,20 @@ static inline void hfi1_make_ruc_header_16B(struct rvt_qp *qp, pkey, becn, 0, l4, priv->s_sc); } +/** + * hfi1_make_ruc_header_9B - build a 9B header + * @qp: the queue pair + * @ohdr: a pointer to the destination header memory + * @bth0: bth0 passed in from the RC/UC builder + * @bth2: bth2 passed in from the RC/UC builder + * @middle: non zero implies indicates ahg "could" be used + * @ps: the current packet state + * + * This routine may disarm ahg under these situations: + * - packet needs a GRH + * - BECN needed + * - migration state not IB_MIG_MIGRATED + */ static inline void hfi1_make_ruc_header_9B(struct rvt_qp *qp, struct ib_other_headers *ohdr, u32 bth0, u32 bth2, int middle, @@ -839,6 +868,12 @@ static inline void hfi1_make_ruc_header_9B(struct rvt_qp *qp, else middle = 0; + if (qp->s_flags & RVT_S_ECN) { + qp->s_flags &= ~RVT_S_ECN; + /* we recently received a FECN, so return a BECN */ + bth1 |= (IB_BECN_MASK << IB_BECN_SHIFT); + middle = 0; + } if (middle) build_ahg(qp, bth2); else @@ -846,11 +881,6 @@ static inline void hfi1_make_ruc_header_9B(struct rvt_qp *qp, bth0 |= pkey; bth0 |= extra_bytes << 20; - if (qp->s_flags & RVT_S_ECN) { - qp->s_flags &= ~RVT_S_ECN; - /* we recently received a FECN, so return a BECN */ - bth1 |= (IB_BECN_MASK << IB_BECN_SHIFT); - } hfi1_make_ruc_bth(qp, ohdr, bth0, bth1, bth2); hfi1_make_ib_hdr(&ps->s_txreq->phdr.hdr.ibh, lrh0, diff --git a/drivers/infiniband/hw/hfi1/ud.c b/drivers/infiniband/hw/hfi1/ud.c index bcf3b0bebac8..69c17a5ef038 100644 --- a/drivers/infiniband/hw/hfi1/ud.c +++ b/drivers/infiniband/hw/hfi1/ud.c @@ -628,7 +628,7 @@ int hfi1_lookup_pkey_idx(struct hfi1_ibport *ibp, u16 pkey) } void return_cnp_16B(struct hfi1_ibport *ibp, struct rvt_qp *qp, - u32 remote_qpn, u32 pkey, u32 slid, u32 dlid, + u32 remote_qpn, u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh) { u64 pbc, pbc_flags = 0; @@ -687,7 +687,7 @@ void return_cnp_16B(struct hfi1_ibport *ibp, struct rvt_qp *qp, } void return_cnp(struct hfi1_ibport *ibp, struct rvt_qp *qp, u32 remote_qpn, - u32 pkey, u32 slid, u32 dlid, u8 sc5, + u16 pkey, u32 slid, u32 dlid, u8 sc5, const struct ib_grh *old_grh) { u64 pbc, pbc_flags = 0; diff --git a/drivers/infiniband/hw/hns/hns_roce_cq.c b/drivers/infiniband/hw/hns/hns_roce_cq.c index 14734d0d0b76..3a485f50fede 100644 --- a/drivers/infiniband/hw/hns/hns_roce_cq.c +++ b/drivers/infiniband/hw/hns/hns_roce_cq.c @@ -377,6 +377,7 @@ struct ib_cq *hns_roce_ib_create_cq(struct ib_device *ib_dev, hr_cq->set_ci_db = hr_cq->db.db_record; *hr_cq->set_ci_db = 0; + hr_cq->db_en = 1; } /* Init mmt table and write buff address to mtt table */ diff --git a/drivers/infiniband/hw/hns/hns_roce_hem.c b/drivers/infiniband/hw/hns/hns_roce_hem.c index 0eeabfbee192..63b5b3edabcb 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hem.c +++ b/drivers/infiniband/hw/hns/hns_roce_hem.c @@ -912,7 +912,7 @@ int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev, obj_per_chunk = buf_chunk_size / obj_size; num_hem = (nobj + obj_per_chunk - 1) / obj_per_chunk; bt_chunk_num = bt_chunk_size / 8; - if (table->type >= HEM_TYPE_MTT) + if (type >= HEM_TYPE_MTT) num_bt_l0 = bt_chunk_num; table->hem = kcalloc(num_hem, sizeof(*table->hem), @@ -920,7 +920,7 @@ int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev, if (!table->hem) goto err_kcalloc_hem_buf; - if (check_whether_bt_num_3(table->type, hop_num)) { + if (check_whether_bt_num_3(type, hop_num)) { unsigned long num_bt_l1; num_bt_l1 = (num_hem + bt_chunk_num - 1) / @@ -939,8 +939,8 @@ int hns_roce_init_hem_table(struct hns_roce_dev *hr_dev, goto err_kcalloc_l1_dma; } - if (check_whether_bt_num_2(table->type, hop_num) || - check_whether_bt_num_3(table->type, hop_num)) { + if (check_whether_bt_num_2(type, hop_num) || + check_whether_bt_num_3(type, hop_num)) { table->bt_l0 = kcalloc(num_bt_l0, sizeof(*table->bt_l0), GFP_KERNEL); if (!table->bt_l0) @@ -1039,14 +1039,14 @@ void hns_roce_cleanup_hem_table(struct hns_roce_dev *hr_dev, void hns_roce_cleanup_hem(struct hns_roce_dev *hr_dev) { hns_roce_cleanup_hem_table(hr_dev, &hr_dev->cq_table.table); - hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.irrl_table); if (hr_dev->caps.trrl_entry_sz) hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.trrl_table); + hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.irrl_table); hns_roce_cleanup_hem_table(hr_dev, &hr_dev->qp_table.qp_table); hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtpt_table); - hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_table); if (hns_roce_check_whether_mhop(hr_dev, HEM_TYPE_CQE)) hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_cqe_table); + hns_roce_cleanup_hem_table(hr_dev, &hr_dev->mr_table.mtt_table); } diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index 47e1b6ac1e1a..8013d69c5ac4 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -722,6 +722,7 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) free_mr->mr_free_pd = to_hr_pd(pd); free_mr->mr_free_pd->ibpd.device = &hr_dev->ib_dev; free_mr->mr_free_pd->ibpd.uobject = NULL; + free_mr->mr_free_pd->ibpd.__internal_mr = NULL; atomic_set(&free_mr->mr_free_pd->ibpd.usecnt, 0); attr.qp_access_flags = IB_ACCESS_REMOTE_WRITE; @@ -1036,7 +1037,7 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) do { ret = hns_roce_v1_poll_cq(&mr_free_cq->ib_cq, ne, wc); - if (ret < 0) { + if (ret < 0 && hr_qp) { dev_err(dev, "(qp:0x%lx) starts, Poll cqe failed(%d) for mr 0x%x free! Remain %d cqe\n", hr_qp->qpn, ret, hr_mr->key, ne); diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c index 8b84ab7800d8..1f0965bb64ee 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v2.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v2.c @@ -71,6 +71,11 @@ static int set_rwqe_data_seg(struct ib_qp *ibqp, struct ib_send_wr *wr, return -EINVAL; } + if (wr->opcode == IB_WR_RDMA_READ) { + dev_err(hr_dev->dev, "Not support inline data!\n"); + return -EINVAL; + } + for (i = 0; i < wr->num_sge; i++) { memcpy(wqe, ((void *)wr->sg_list[i].addr), wr->sg_list[i].length); @@ -137,8 +142,8 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, unsigned long flags; unsigned int ind; void *wqe = NULL; - u32 tmp_len = 0; bool loopback; + u32 tmp_len; int ret = 0; u8 *smac; int nreq; @@ -148,7 +153,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, ibqp->qp_type != IB_QPT_GSI && ibqp->qp_type != IB_QPT_UD)) { dev_err(dev, "Not supported QP(0x%x)type!\n", ibqp->qp_type); - *bad_wr = NULL; + *bad_wr = wr; return -EOPNOTSUPP; } @@ -182,7 +187,9 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, qp->sq.wrid[(qp->sq.head + nreq) & (qp->sq.wqe_cnt - 1)] = wr->wr_id; - owner_bit = ~(qp->sq.head >> ilog2(qp->sq.wqe_cnt)) & 0x1; + owner_bit = + ~(((qp->sq.head + nreq) >> ilog2(qp->sq.wqe_cnt)) & 0x1); + tmp_len = 0; /* Corresponding to the QP type, wqe process separately */ if (ibqp->qp_type == IB_QPT_GSI) { @@ -456,6 +463,7 @@ static int hns_roce_v2_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, } else { dev_err(dev, "Illegal qp_type(0x%x)\n", ibqp->qp_type); spin_unlock_irqrestore(&qp->sq.lock, flags); + *bad_wr = wr; return -EOPNOTSUPP; } } @@ -540,16 +548,20 @@ static int hns_roce_v2_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, } if (i < hr_qp->rq.max_gs) { - dseg[i].lkey = cpu_to_le32(HNS_ROCE_INVALID_LKEY); - dseg[i].addr = 0; + dseg->lkey = cpu_to_le32(HNS_ROCE_INVALID_LKEY); + dseg->addr = 0; } /* rq support inline data */ - sge_list = hr_qp->rq_inl_buf.wqe_list[ind].sg_list; - hr_qp->rq_inl_buf.wqe_list[ind].sge_cnt = (u32)wr->num_sge; - for (i = 0; i < wr->num_sge; i++) { - sge_list[i].addr = (void *)(u64)wr->sg_list[i].addr; - sge_list[i].len = wr->sg_list[i].length; + if (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) { + sge_list = hr_qp->rq_inl_buf.wqe_list[ind].sg_list; + hr_qp->rq_inl_buf.wqe_list[ind].sge_cnt = + (u32)wr->num_sge; + for (i = 0; i < wr->num_sge; i++) { + sge_list[i].addr = + (void *)(u64)wr->sg_list[i].addr; + sge_list[i].len = wr->sg_list[i].length; + } } hr_qp->rq.wrid[ind] = wr->wr_id; @@ -606,6 +618,8 @@ static void hns_roce_free_cmq_desc(struct hns_roce_dev *hr_dev, dma_unmap_single(hr_dev->dev, ring->desc_dma_addr, ring->desc_num * sizeof(struct hns_roce_cmq_desc), DMA_BIDIRECTIONAL); + + ring->desc_dma_addr = 0; kfree(ring->desc); } @@ -1074,6 +1088,7 @@ static int hns_roce_v2_profile(struct hns_roce_dev *hr_dev) if (ret) { dev_err(hr_dev->dev, "Configure global param fail, ret = %d.\n", ret); + return ret; } /* Get pf resource owned by every pf */ @@ -1365,6 +1380,8 @@ static int hns_roce_v2_write_mtpt(void *mb_buf, struct hns_roce_mr *mr, roce_set_bit(mpt_entry->byte_12_mw_pa, V2_MPT_BYTE_12_PA_S, mr->type == MR_TYPE_MR ? 0 : 1); + roce_set_bit(mpt_entry->byte_12_mw_pa, V2_MPT_BYTE_12_INNER_PA_VLD_S, + 1); mpt_entry->byte_12_mw_pa = cpu_to_le32(mpt_entry->byte_12_mw_pa); mpt_entry->len_l = cpu_to_le32(lower_32_bits(mr->size)); @@ -2162,6 +2179,7 @@ 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_dev *hr_dev = to_hr_dev(ibqp->device); struct hns_roce_qp *hr_qp = to_hr_qp(ibqp); /* @@ -2274,7 +2292,8 @@ static void modify_qp_reset_to_init(struct ib_qp *ibqp, context->rq_db_record_addr = hr_qp->rdb.dma >> 32; qpc_mask->rq_db_record_addr = 0; - roce_set_bit(context->byte_76_srqn_op_en, V2_QPC_BYTE_76_RQIE_S, 1); + roce_set_bit(context->byte_76_srqn_op_en, V2_QPC_BYTE_76_RQIE_S, + (hr_dev->caps.flags & HNS_ROCE_CAP_FLAG_RQ_INLINE) ? 1 : 0); roce_set_bit(qpc_mask->byte_76_srqn_op_en, V2_QPC_BYTE_76_RQIE_S, 0); roce_set_field(context->byte_80_rnr_rx_cqn, V2_QPC_BYTE_80_RX_CQN_M, @@ -2592,10 +2611,12 @@ static void modify_qp_init_to_init(struct ib_qp *ibqp, roce_set_field(qpc_mask->byte_4_sqpn_tst, V2_QPC_BYTE_4_SQPN_M, V2_QPC_BYTE_4_SQPN_S, 0); - roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, hr_qp->qpn); - roce_set_field(qpc_mask->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, 0); + if (attr_mask & IB_QP_DEST_QPN) { + roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, + V2_QPC_BYTE_56_DQPN_S, hr_qp->qpn); + roce_set_field(qpc_mask->byte_56_dqpn_err, + V2_QPC_BYTE_56_DQPN_M, V2_QPC_BYTE_56_DQPN_S, 0); + } roce_set_field(context->byte_168_irrl_idx, V2_QPC_BYTE_168_SQ_SHIFT_BAK_M, V2_QPC_BYTE_168_SQ_SHIFT_BAK_S, @@ -2650,8 +2671,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, return -EINVAL; } - if ((attr_mask & IB_QP_ALT_PATH) || (attr_mask & IB_QP_ACCESS_FLAGS) || - (attr_mask & IB_QP_PKEY_INDEX) || (attr_mask & IB_QP_QKEY)) { + if (attr_mask & IB_QP_ALT_PATH) { dev_err(dev, "INIT2RTR attr_mask (0x%x) error\n", attr_mask); return -EINVAL; } @@ -2800,10 +2820,12 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, V2_QPC_BYTE_140_RR_MAX_S, 0); } - roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, attr->dest_qp_num); - roce_set_field(qpc_mask->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, - V2_QPC_BYTE_56_DQPN_S, 0); + if (attr_mask & IB_QP_DEST_QPN) { + roce_set_field(context->byte_56_dqpn_err, V2_QPC_BYTE_56_DQPN_M, + V2_QPC_BYTE_56_DQPN_S, attr->dest_qp_num); + roce_set_field(qpc_mask->byte_56_dqpn_err, + V2_QPC_BYTE_56_DQPN_M, V2_QPC_BYTE_56_DQPN_S, 0); + } /* Configure GID index */ port_num = rdma_ah_get_port_num(&attr->ah_attr); @@ -2845,7 +2867,7 @@ static int modify_qp_init_to_rtr(struct ib_qp *ibqp, if (ibqp->qp_type == IB_QPT_GSI || ibqp->qp_type == IB_QPT_UD) roce_set_field(context->byte_24_mtu_tc, V2_QPC_BYTE_24_MTU_M, V2_QPC_BYTE_24_MTU_S, IB_MTU_4096); - else + else if (attr_mask & IB_QP_PATH_MTU) roce_set_field(context->byte_24_mtu_tc, V2_QPC_BYTE_24_MTU_M, V2_QPC_BYTE_24_MTU_S, attr->path_mtu); @@ -2922,11 +2944,9 @@ static int modify_qp_rtr_to_rts(struct ib_qp *ibqp, return -EINVAL; } - /* If exist optional param, return error */ - if ((attr_mask & IB_QP_ALT_PATH) || (attr_mask & IB_QP_ACCESS_FLAGS) || - (attr_mask & IB_QP_QKEY) || (attr_mask & IB_QP_PATH_MIG_STATE) || - (attr_mask & IB_QP_CUR_STATE) || - (attr_mask & IB_QP_MIN_RNR_TIMER)) { + /* Not support alternate path and path migration */ + if ((attr_mask & IB_QP_ALT_PATH) || + (attr_mask & IB_QP_PATH_MIG_STATE)) { dev_err(dev, "RTR2RTS attr_mask (0x%x)error\n", attr_mask); return -EINVAL; } @@ -3161,7 +3181,8 @@ static int hns_roce_v2_modify_qp(struct ib_qp *ibqp, (cur_state == IB_QPS_RTR && new_state == IB_QPS_ERR) || (cur_state == IB_QPS_RTS && new_state == IB_QPS_ERR) || (cur_state == IB_QPS_SQD && new_state == IB_QPS_ERR) || - (cur_state == IB_QPS_SQE && new_state == IB_QPS_ERR)) { + (cur_state == IB_QPS_SQE && new_state == IB_QPS_ERR) || + (cur_state == IB_QPS_ERR && new_state == IB_QPS_ERR)) { /* Nothing */ ; } else { @@ -4478,7 +4499,7 @@ static int hns_roce_v2_create_eq(struct hns_roce_dev *hr_dev, ret = hns_roce_cmd_mbox(hr_dev, mailbox->dma, 0, eq->eqn, 0, eq_cmd, HNS_ROCE_CMD_TIMEOUT_MSECS); if (ret) { - dev_err(dev, "[mailbox cmd] creat eqc failed.\n"); + dev_err(dev, "[mailbox cmd] create eqc failed.\n"); goto err_cmd_mbox; } @@ -4694,6 +4715,8 @@ static const struct pci_device_id hns_roce_hw_v2_pci_tbl[] = { {0, } }; +MODULE_DEVICE_TABLE(pci, hns_roce_hw_v2_pci_tbl); + static int hns_roce_hw_v2_get_cfg(struct hns_roce_dev *hr_dev, struct hnae3_handle *handle) { diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index 9d48bc07a9e6..96fb6a9ed93c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -199,7 +199,7 @@ static int hns_roce_query_device(struct ib_device *ib_dev, memset(props, 0, sizeof(*props)); - props->sys_image_guid = cpu_to_be32(hr_dev->sys_image_guid); + props->sys_image_guid = cpu_to_be64(hr_dev->sys_image_guid); props->max_mr_size = (u64)(~(0ULL)); props->page_size_cap = hr_dev->caps.page_size_cap; props->vendor_id = hr_dev->vendor_id; diff --git a/drivers/infiniband/hw/hns/hns_roce_qp.c b/drivers/infiniband/hw/hns/hns_roce_qp.c index e289a924e789..baaf906f7c2e 100644 --- a/drivers/infiniband/hw/hns/hns_roce_qp.c +++ b/drivers/infiniband/hw/hns/hns_roce_qp.c @@ -620,7 +620,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, to_hr_ucontext(ib_pd->uobject->context), ucmd.db_addr, &hr_qp->rdb); if (ret) { - dev_err(dev, "rp record doorbell map failed!\n"); + dev_err(dev, "rq record doorbell map failed!\n"); goto err_mtt; } } @@ -660,6 +660,7 @@ static int hns_roce_create_qp_common(struct hns_roce_dev *hr_dev, goto err_rq_sge_list; } *hr_qp->rdb.db_record = 0; + hr_qp->rdb_en = 1; } /* Allocate QP buf */ @@ -955,7 +956,14 @@ int hns_roce_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, } if (cur_state == new_state && cur_state == IB_QPS_RESET) { - ret = 0; + if (hr_dev->caps.min_wqes) { + ret = -EPERM; + dev_err(dev, "cur_state=%d new_state=%d\n", cur_state, + new_state); + } else { + ret = 0; + } + goto out; } diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h index d5d8c1be345a..2f2b4426ded7 100644 --- a/drivers/infiniband/hw/i40iw/i40iw.h +++ b/drivers/infiniband/hw/i40iw/i40iw.h @@ -207,6 +207,7 @@ struct i40iw_msix_vector { u32 irq; u32 cpu_affinity; u32 ceq_id; + cpumask_t mask; }; struct l2params_work { diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c index 4cfa8f4647e2..f7c6fd9ff6e2 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.c +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c @@ -2093,7 +2093,7 @@ static int i40iw_addr_resolve_neigh_ipv6(struct i40iw_device *iwdev, if (netif_is_bond_slave(netdev)) netdev = netdev_master_upper_dev_get(netdev); - neigh = dst_neigh_lookup(dst, &dst_addr); + neigh = dst_neigh_lookup(dst, dst_addr.sin6_addr.in6_u.u6_addr32); rcu_read_lock(); if (neigh) { diff --git a/drivers/infiniband/hw/i40iw/i40iw_hw.c b/drivers/infiniband/hw/i40iw/i40iw_hw.c index 6139836fb533..c9f62ca7643c 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_hw.c +++ b/drivers/infiniband/hw/i40iw/i40iw_hw.c @@ -331,7 +331,7 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) switch (info->ae_id) { case I40IW_AE_LLP_FIN_RECEIVED: if (qp->term_flags) - continue; + break; if (atomic_inc_return(&iwqp->close_timer_started) == 1) { iwqp->hw_tcp_state = I40IW_TCP_STATE_CLOSE_WAIT; if ((iwqp->hw_tcp_state == I40IW_TCP_STATE_CLOSE_WAIT) && @@ -360,7 +360,7 @@ void i40iw_process_aeq(struct i40iw_device *iwdev) break; case I40IW_AE_LLP_CONNECTION_RESET: if (atomic_read(&iwqp->close_timer_started)) - continue; + break; i40iw_cm_disconn(iwqp); break; case I40IW_AE_QP_SUSPEND_COMPLETE: diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c index 9cd0d3ef9057..05001e6da1f8 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_main.c +++ b/drivers/infiniband/hw/i40iw/i40iw_main.c @@ -687,7 +687,6 @@ static enum i40iw_status_code i40iw_configure_ceq_vector(struct i40iw_device *iw struct i40iw_msix_vector *msix_vec) { enum i40iw_status_code status; - cpumask_t mask; if (iwdev->msix_shared && !ceq_id) { tasklet_init(&iwdev->dpc_tasklet, i40iw_dpc, (unsigned long)iwdev); @@ -697,9 +696,9 @@ static enum i40iw_status_code i40iw_configure_ceq_vector(struct i40iw_device *iw status = request_irq(msix_vec->irq, i40iw_ceq_handler, 0, "CEQ", iwceq); } - cpumask_clear(&mask); - cpumask_set_cpu(msix_vec->cpu_affinity, &mask); - irq_set_affinity_hint(msix_vec->irq, &mask); + cpumask_clear(&msix_vec->mask); + cpumask_set_cpu(msix_vec->cpu_affinity, &msix_vec->mask); + irq_set_affinity_hint(msix_vec->irq, &msix_vec->mask); if (status) { i40iw_pr_err("ceq irq config fail\n"); diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 40e4f5ab2b46..68679ad4c6da 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -394,6 +394,7 @@ static struct i40iw_pbl *i40iw_get_pbl(unsigned long va, list_for_each_entry(iwpbl, pbl_list, list) { if (iwpbl->user_base == va) { + iwpbl->on_list = false; list_del(&iwpbl->list); return iwpbl; } @@ -614,6 +615,7 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, return ERR_PTR(-ENOMEM); iwqp = (struct i40iw_qp *)mem; + iwqp->allocated_buffer = mem; qp = &iwqp->sc_qp; qp->back_qp = (void *)iwqp; qp->push_idx = I40IW_INVALID_PUSH_PAGE_INDEX; @@ -642,7 +644,6 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, goto error; } - iwqp->allocated_buffer = mem; iwqp->iwdev = iwdev; iwqp->iwpd = iwpd; iwqp->ibqp.qp_num = qp_num; @@ -1898,6 +1899,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, goto error; spin_lock_irqsave(&ucontext->qp_reg_mem_list_lock, flags); list_add_tail(&iwpbl->list, &ucontext->qp_reg_mem_list); + iwpbl->on_list = true; spin_unlock_irqrestore(&ucontext->qp_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_CQ: @@ -1908,6 +1910,7 @@ static struct ib_mr *i40iw_reg_user_mr(struct ib_pd *pd, spin_lock_irqsave(&ucontext->cq_reg_mem_list_lock, flags); list_add_tail(&iwpbl->list, &ucontext->cq_reg_mem_list); + iwpbl->on_list = true; spin_unlock_irqrestore(&ucontext->cq_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_MEM: @@ -2045,14 +2048,18 @@ static void i40iw_del_memlist(struct i40iw_mr *iwmr, switch (iwmr->type) { case IW_MEMREG_TYPE_CQ: spin_lock_irqsave(&ucontext->cq_reg_mem_list_lock, flags); - if (!list_empty(&ucontext->cq_reg_mem_list)) + if (iwpbl->on_list) { + iwpbl->on_list = false; list_del(&iwpbl->list); + } spin_unlock_irqrestore(&ucontext->cq_reg_mem_list_lock, flags); break; case IW_MEMREG_TYPE_QP: spin_lock_irqsave(&ucontext->qp_reg_mem_list_lock, flags); - if (!list_empty(&ucontext->qp_reg_mem_list)) + if (iwpbl->on_list) { + iwpbl->on_list = false; list_del(&iwpbl->list); + } spin_unlock_irqrestore(&ucontext->qp_reg_mem_list_lock, flags); break; default: diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.h b/drivers/infiniband/hw/i40iw/i40iw_verbs.h index 9067443cd311..76cf173377ab 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.h +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.h @@ -78,6 +78,7 @@ struct i40iw_pbl { }; bool pbl_allocated; + bool on_list; u64 user_base; struct i40iw_pble_alloc pble_alloc; struct i40iw_mr *iwmr; diff --git a/drivers/infiniband/hw/mlx4/mr.c b/drivers/infiniband/hw/mlx4/mr.c index 17f4f151a97f..61d8b06375bb 100644 --- a/drivers/infiniband/hw/mlx4/mr.c +++ b/drivers/infiniband/hw/mlx4/mr.c @@ -346,7 +346,7 @@ int mlx4_ib_umem_calc_optimal_mtt_size(struct ib_umem *umem, u64 start_va, /* Add to the first block the misalignment that it suffers from. */ total_len += (first_block_start & ((1ULL << block_shift) - 1ULL)); last_block_end = current_block_start + current_block_len; - last_block_aligned_end = round_up(last_block_end, 1 << block_shift); + last_block_aligned_end = round_up(last_block_end, 1ULL << block_shift); total_len += (last_block_aligned_end - last_block_end); if (total_len & ((1ULL << block_shift) - 1ULL)) diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 50af8915e7ec..199648adac74 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -673,7 +673,8 @@ static int set_qp_rss(struct mlx4_ib_dev *dev, struct mlx4_ib_rss *rss_ctx, MLX4_IB_RX_HASH_SRC_PORT_TCP | MLX4_IB_RX_HASH_DST_PORT_TCP | MLX4_IB_RX_HASH_SRC_PORT_UDP | - MLX4_IB_RX_HASH_DST_PORT_UDP)) { + MLX4_IB_RX_HASH_DST_PORT_UDP | + MLX4_IB_RX_HASH_INNER)) { pr_debug("RX Hash fields_mask has unsupported mask (0x%llx)\n", ucmd->rx_hash_fields_mask); return (-EOPNOTSUPP); diff --git a/drivers/infiniband/hw/mlx5/Kconfig b/drivers/infiniband/hw/mlx5/Kconfig index bce263b92821..fb4d77be019b 100644 --- a/drivers/infiniband/hw/mlx5/Kconfig +++ b/drivers/infiniband/hw/mlx5/Kconfig @@ -1,6 +1,7 @@ config MLX5_INFINIBAND tristate "Mellanox Connect-IB HCA support" depends on NETDEVICES && ETHERNET && PCI && MLX5_CORE + depends on INFINIBAND_USER_ACCESS || INFINIBAND_USER_ACCESS=n ---help--- This driver provides low-level InfiniBand support for Mellanox Connect-IB PCI Express host channel adapters (HCAs). diff --git a/drivers/infiniband/hw/mlx5/main.c b/drivers/infiniband/hw/mlx5/main.c index daa919e5a442..69716a7ea993 100644 --- a/drivers/infiniband/hw/mlx5/main.c +++ b/drivers/infiniband/hw/mlx5/main.c @@ -52,7 +52,6 @@ #include <linux/mlx5/port.h> #include <linux/mlx5/vport.h> #include <linux/mlx5/fs.h> -#include <linux/mlx5/fs_helpers.h> #include <linux/list.h> #include <rdma/ib_smi.h> #include <rdma/ib_umem.h> @@ -180,7 +179,7 @@ static int mlx5_netdev_event(struct notifier_block *this, if (rep_ndev == ndev) roce->netdev = (event == NETDEV_UNREGISTER) ? NULL : ndev; - } else if (ndev->dev.parent == &ibdev->mdev->pdev->dev) { + } else if (ndev->dev.parent == &mdev->pdev->dev) { roce->netdev = (event == NETDEV_UNREGISTER) ? NULL : ndev; } @@ -2417,7 +2416,7 @@ static void set_proto(void *outer_c, void *outer_v, u8 mask, u8 val) MLX5_SET(fte_match_set_lyr_2_4, outer_v, ip_protocol, val); } -static void set_flow_label(void *misc_c, void *misc_v, u8 mask, u8 val, +static void set_flow_label(void *misc_c, void *misc_v, u32 mask, u32 val, bool inner) { if (inner) { @@ -4757,7 +4756,7 @@ mlx5_ib_get_vector_affinity(struct ib_device *ibdev, int comp_vector) { struct mlx5_ib_dev *dev = to_mdev(ibdev); - return mlx5_get_vector_affinity(dev->mdev, comp_vector); + return mlx5_get_vector_affinity_hint(dev->mdev, comp_vector); } /* The mlx5_ib_multiport_mutex should be held when calling this function */ @@ -5427,9 +5426,7 @@ static void mlx5_ib_stage_cong_debugfs_cleanup(struct mlx5_ib_dev *dev) static int mlx5_ib_stage_uar_init(struct mlx5_ib_dev *dev) { dev->mdev->priv.uar = mlx5_get_uars_page(dev->mdev); - if (!dev->mdev->priv.uar) - return -ENOMEM; - return 0; + return PTR_ERR_OR_ZERO(dev->mdev->priv.uar); } static void mlx5_ib_stage_uar_cleanup(struct mlx5_ib_dev *dev) diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 1520a2f20f98..90a9c461cedc 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -866,25 +866,28 @@ static int mr_umem_get(struct ib_pd *pd, u64 start, u64 length, int *order) { struct mlx5_ib_dev *dev = to_mdev(pd->device); + struct ib_umem *u; int err; - *umem = ib_umem_get(pd->uobject->context, start, length, - access_flags, 0); - err = PTR_ERR_OR_ZERO(*umem); + *umem = NULL; + + u = ib_umem_get(pd->uobject->context, start, length, access_flags, 0); + err = PTR_ERR_OR_ZERO(u); if (err) { - *umem = NULL; - mlx5_ib_err(dev, "umem get failed (%d)\n", err); + mlx5_ib_dbg(dev, "umem get failed (%d)\n", err); return err; } - mlx5_ib_cont_pages(*umem, start, MLX5_MKEY_PAGE_SHIFT_MASK, npages, + mlx5_ib_cont_pages(u, start, MLX5_MKEY_PAGE_SHIFT_MASK, npages, page_shift, ncont, order); if (!*npages) { mlx5_ib_warn(dev, "avoid zero region\n"); - ib_umem_release(*umem); + ib_umem_release(u); return -EINVAL; } + *umem = u; + mlx5_ib_dbg(dev, "npages %d, ncont %d, order %d, page_shift %d\n", *npages, *ncont, *order, *page_shift); @@ -1458,13 +1461,12 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, int access_flags = flags & IB_MR_REREG_ACCESS ? new_access_flags : mr->access_flags; - u64 addr = (flags & IB_MR_REREG_TRANS) ? virt_addr : mr->umem->address; - u64 len = (flags & IB_MR_REREG_TRANS) ? length : mr->umem->length; int page_shift = 0; int upd_flags = 0; int npages = 0; int ncont = 0; int order = 0; + u64 addr, len; int err; mlx5_ib_dbg(dev, "start 0x%llx, virt_addr 0x%llx, length 0x%llx, access_flags 0x%x\n", @@ -1472,6 +1474,17 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, atomic_sub(mr->npages, &dev->mdev->priv.reg_pages); + if (!mr->umem) + return -EINVAL; + + if (flags & IB_MR_REREG_TRANS) { + addr = virt_addr; + len = length; + } else { + addr = mr->umem->address; + len = mr->umem->length; + } + if (flags != IB_MR_REREG_PD) { /* * Replace umem. This needs to be done whether or not UMR is @@ -1479,6 +1492,7 @@ int mlx5_ib_rereg_user_mr(struct ib_mr *ib_mr, int flags, u64 start, */ flags |= IB_MR_REREG_TRANS; ib_umem_release(mr->umem); + mr->umem = NULL; err = mr_umem_get(pd, addr, len, access_flags, &mr->umem, &npages, &page_shift, &ncont, &order); if (err) diff --git a/drivers/infiniband/hw/mlx5/qp.c b/drivers/infiniband/hw/mlx5/qp.c index 7ed4b70f6447..2193dc1765fb 100644 --- a/drivers/infiniband/hw/mlx5/qp.c +++ b/drivers/infiniband/hw/mlx5/qp.c @@ -259,7 +259,11 @@ static int set_rq_size(struct mlx5_ib_dev *dev, struct ib_qp_cap *cap, } else { if (ucmd) { qp->rq.wqe_cnt = ucmd->rq_wqe_count; + if (ucmd->rq_wqe_shift > BITS_PER_BYTE * sizeof(ucmd->rq_wqe_shift)) + return -EINVAL; qp->rq.wqe_shift = ucmd->rq_wqe_shift; + if ((1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) < qp->wq_sig) + return -EINVAL; qp->rq.max_gs = (1 << qp->rq.wqe_shift) / sizeof(struct mlx5_wqe_data_seg) - qp->wq_sig; qp->rq.max_post = qp->rq.wqe_cnt; } else { @@ -480,11 +484,6 @@ static int qp_has_rq(struct ib_qp_init_attr *attr) return 1; } -static int first_med_bfreg(void) -{ - return 1; -} - enum { /* this is the first blue flame register in the array of bfregs assigned * to a processes. Since we do not use it for blue flame but rather @@ -510,6 +509,12 @@ static int num_med_bfreg(struct mlx5_ib_dev *dev, return n >= 0 ? n : 0; } +static int first_med_bfreg(struct mlx5_ib_dev *dev, + struct mlx5_bfreg_info *bfregi) +{ + return num_med_bfreg(dev, bfregi) ? 1 : -ENOMEM; +} + static int first_hi_bfreg(struct mlx5_ib_dev *dev, struct mlx5_bfreg_info *bfregi) { @@ -537,10 +542,13 @@ static int alloc_high_class_bfreg(struct mlx5_ib_dev *dev, static int alloc_med_class_bfreg(struct mlx5_ib_dev *dev, struct mlx5_bfreg_info *bfregi) { - int minidx = first_med_bfreg(); + int minidx = first_med_bfreg(dev, bfregi); int i; - for (i = first_med_bfreg(); i < first_hi_bfreg(dev, bfregi); i++) { + if (minidx < 0) + return minidx; + + for (i = minidx; i < first_hi_bfreg(dev, bfregi); i++) { if (bfregi->count[i] < bfregi->count[minidx]) minidx = i; if (!bfregi->count[minidx]) @@ -2451,18 +2459,18 @@ enum { static int ib_rate_to_mlx5(struct mlx5_ib_dev *dev, u8 rate) { - if (rate == IB_RATE_PORT_CURRENT) { + if (rate == IB_RATE_PORT_CURRENT) return 0; - } else if (rate < IB_RATE_2_5_GBPS || rate > IB_RATE_300_GBPS) { + + if (rate < IB_RATE_2_5_GBPS || rate > IB_RATE_300_GBPS) return -EINVAL; - } else { - while (rate != IB_RATE_2_5_GBPS && - !(1 << (rate + MLX5_STAT_RATE_OFFSET) & - MLX5_CAP_GEN(dev->mdev, stat_rate_support))) - --rate; - } - return rate + MLX5_STAT_RATE_OFFSET; + while (rate != IB_RATE_PORT_CURRENT && + !(1 << (rate + MLX5_STAT_RATE_OFFSET) & + MLX5_CAP_GEN(dev->mdev, stat_rate_support))) + --rate; + + return rate ? rate + MLX5_STAT_RATE_OFFSET : rate; } static int modify_raw_packet_eth_prio(struct mlx5_core_dev *dev, diff --git a/drivers/infiniband/hw/nes/nes_nic.c b/drivers/infiniband/hw/nes/nes_nic.c index 0a75164cedea..007d5e8a0121 100644 --- a/drivers/infiniband/hw/nes/nes_nic.c +++ b/drivers/infiniband/hw/nes/nes_nic.c @@ -461,7 +461,7 @@ static bool nes_nic_send(struct sk_buff *skb, struct net_device *netdev) /** * nes_netdev_start_xmit */ -static int nes_netdev_start_xmit(struct sk_buff *skb, struct net_device *netdev) +static netdev_tx_t nes_netdev_start_xmit(struct sk_buff *skb, struct net_device *netdev) { struct nes_vnic *nesvnic = netdev_priv(netdev); struct nes_device *nesdev = nesvnic->nesdev; diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 7d3763b2e01c..3f9afc02d166 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -401,49 +401,47 @@ int qedr_mmap(struct ib_ucontext *context, struct vm_area_struct *vma) { struct qedr_ucontext *ucontext = get_qedr_ucontext(context); struct qedr_dev *dev = get_qedr_dev(context->device); - unsigned long vm_page = vma->vm_pgoff << PAGE_SHIFT; - u64 unmapped_db = dev->db_phys_addr; + unsigned long phys_addr = vma->vm_pgoff << PAGE_SHIFT; unsigned long len = (vma->vm_end - vma->vm_start); - int rc = 0; - bool found; + unsigned long dpi_start; + + dpi_start = dev->db_phys_addr + (ucontext->dpi * ucontext->dpi_size); DP_DEBUG(dev, QEDR_MSG_INIT, - "qedr_mmap called vm_page=0x%lx vm_pgoff=0x%lx unmapped_db=0x%llx db_size=%x, len=%lx\n", - vm_page, vma->vm_pgoff, unmapped_db, dev->db_size, len); - if (vma->vm_start & (PAGE_SIZE - 1)) { - DP_ERR(dev, "Vma_start not page aligned = %ld\n", - vma->vm_start); + "mmap invoked with vm_start=0x%pK, vm_end=0x%pK,vm_pgoff=0x%pK; dpi_start=0x%pK dpi_size=0x%x\n", + (void *)vma->vm_start, (void *)vma->vm_end, + (void *)vma->vm_pgoff, (void *)dpi_start, ucontext->dpi_size); + + if ((vma->vm_start & (PAGE_SIZE - 1)) || (len & (PAGE_SIZE - 1))) { + DP_ERR(dev, + "failed mmap, adrresses must be page aligned: start=0x%pK, end=0x%pK\n", + (void *)vma->vm_start, (void *)vma->vm_end); return -EINVAL; } - found = qedr_search_mmap(ucontext, vm_page, len); - if (!found) { - DP_ERR(dev, "Vma_pgoff not found in mapped array = %ld\n", + if (!qedr_search_mmap(ucontext, phys_addr, len)) { + DP_ERR(dev, "failed mmap, vm_pgoff=0x%lx is not authorized\n", vma->vm_pgoff); return -EINVAL; } - DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping doorbell bar\n"); - - if ((vm_page >= unmapped_db) && (vm_page <= (unmapped_db + - dev->db_size))) { - DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping doorbell bar\n"); - if (vma->vm_flags & VM_READ) { - DP_ERR(dev, "Trying to map doorbell bar for read\n"); - return -EPERM; - } - - vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); + if (phys_addr < dpi_start || + ((phys_addr + len) > (dpi_start + ucontext->dpi_size))) { + DP_ERR(dev, + "failed mmap, pages are outside of dpi; page address=0x%pK, dpi_start=0x%pK, dpi_size=0x%x\n", + (void *)phys_addr, (void *)dpi_start, + ucontext->dpi_size); + return -EINVAL; + } - rc = io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, - PAGE_SIZE, vma->vm_page_prot); - } else { - DP_DEBUG(dev, QEDR_MSG_INIT, "Mapping chains\n"); - rc = remap_pfn_range(vma, vma->vm_start, - vma->vm_pgoff, len, vma->vm_page_prot); + if (vma->vm_flags & VM_READ) { + DP_ERR(dev, "failed mmap, cannot map doorbell bar for read\n"); + return -EINVAL; } - DP_DEBUG(dev, QEDR_MSG_INIT, "qedr_mmap return code: %d\n", rc); - return rc; + + vma->vm_page_prot = pgprot_writecombine(vma->vm_page_prot); + return io_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff, len, + vma->vm_page_prot); } struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, diff --git a/drivers/infiniband/sw/rxe/rxe_opcode.c b/drivers/infiniband/sw/rxe/rxe_opcode.c index 61927c165b59..4cf11063e0b5 100644 --- a/drivers/infiniband/sw/rxe/rxe_opcode.c +++ b/drivers/infiniband/sw/rxe/rxe_opcode.c @@ -390,7 +390,7 @@ struct rxe_opcode_info rxe_opcode[RXE_NUM_OPCODE] = { .name = "IB_OPCODE_RC_SEND_ONLY_INV", .mask = RXE_IETH_MASK | RXE_PAYLOAD_MASK | RXE_REQ_MASK | RXE_COMP_MASK | RXE_RWR_MASK | RXE_SEND_MASK - | RXE_END_MASK, + | RXE_END_MASK | RXE_START_MASK, .length = RXE_BTH_BYTES + RXE_IETH_BYTES, .offset = { [RXE_BTH] = 0, diff --git a/drivers/infiniband/sw/rxe/rxe_req.c b/drivers/infiniband/sw/rxe/rxe_req.c index 7bdaf71b8221..785199990457 100644 --- a/drivers/infiniband/sw/rxe/rxe_req.c +++ b/drivers/infiniband/sw/rxe/rxe_req.c @@ -728,7 +728,6 @@ next_wqe: rollback_state(wqe, qp, &rollback_wqe, rollback_psn); if (ret == -EAGAIN) { - kfree_skb(skb); rxe_run_task(&qp->req.task, 1); goto exit; } diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index a65c9969f7fc..955ff3b6da9c 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -742,7 +742,6 @@ static enum resp_states read_reply(struct rxe_qp *qp, err = rxe_xmit_packet(rxe, qp, &ack_pkt, skb); if (err) { pr_err("Failed sending RDMA reply.\n"); - kfree_skb(skb); return RESPST_ERR_RNR; } @@ -954,10 +953,8 @@ static int send_ack(struct rxe_qp *qp, struct rxe_pkt_info *pkt, } err = rxe_xmit_packet(rxe, qp, &ack_pkt, skb); - if (err) { + if (err) pr_err_ratelimited("Failed sending ack\n"); - kfree_skb(skb); - } err1: return err; @@ -1141,7 +1138,6 @@ static enum resp_states duplicate_request(struct rxe_qp *qp, if (rc) { pr_err("Failed resending result. This flow is not handled - skb ignored\n"); rxe_drop_ref(qp); - kfree_skb(skb_copy); rc = RESPST_CLEANUP; goto out; } diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 2cb52fd48cf1..73a00a1c06f6 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -761,7 +761,6 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, struct ib_send_wr *wr, unsigned int mask; unsigned int length = 0; int i; - int must_sched; while (wr) { mask = wr_opcode_mask(wr->opcode, qp); @@ -791,14 +790,7 @@ static int rxe_post_send_kernel(struct rxe_qp *qp, struct ib_send_wr *wr, wr = wr->next; } - /* - * Must sched in case of GSI QP because ib_send_mad() hold irq lock, - * and the requester call ip_local_out_sk() that takes spin_lock_bh. - */ - must_sched = (qp_type(qp) == IB_QPT_GSI) || - (queue_count(qp->sq.queue) > 1); - - rxe_run_task(&qp->req.task, must_sched); + rxe_run_task(&qp->req.task, 1); if (unlikely(qp->req.state == QP_STATE_ERROR)) rxe_run_task(&qp->comp.task, 1); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 161ba8c76285..cf291f90b58f 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -1094,7 +1094,7 @@ drop_and_unlock: spin_unlock_irqrestore(&priv->lock, flags); } -static int ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) +static netdev_tx_t ipoib_start_xmit(struct sk_buff *skb, struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); struct rdma_netdev *rn = netdev_priv(dev); diff --git a/drivers/infiniband/ulp/srp/Kconfig b/drivers/infiniband/ulp/srp/Kconfig index c74ee9633041..99db8fe5173a 100644 --- a/drivers/infiniband/ulp/srp/Kconfig +++ b/drivers/infiniband/ulp/srp/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_SRP tristate "InfiniBand SCSI RDMA Protocol" - depends on SCSI + depends on SCSI && INFINIBAND_ADDR_TRANS select SCSI_SRP_ATTRS ---help--- Support for the SCSI RDMA Protocol over InfiniBand. This diff --git a/drivers/infiniband/ulp/srpt/Kconfig b/drivers/infiniband/ulp/srpt/Kconfig index 31ee83d528d9..fb8b7182f05e 100644 --- a/drivers/infiniband/ulp/srpt/Kconfig +++ b/drivers/infiniband/ulp/srpt/Kconfig @@ -1,6 +1,6 @@ config INFINIBAND_SRPT tristate "InfiniBand SCSI RDMA Protocol target support" - depends on INFINIBAND && TARGET_CORE + depends on INFINIBAND && INFINIBAND_ADDR_TRANS && TARGET_CORE ---help--- Support for the SCSI RDMA Protocol (SRP) Target driver. The |