diff options
Diffstat (limited to 'drivers/infiniband')
57 files changed, 676 insertions, 395 deletions
diff --git a/drivers/infiniband/core/addr.c b/drivers/infiniband/core/addr.c index a6cb379a4ebc..437522ca97b4 100644 --- a/drivers/infiniband/core/addr.c +++ b/drivers/infiniband/core/addr.c @@ -61,6 +61,7 @@ struct addr_req { void (*callback)(int status, struct sockaddr *src_addr, struct rdma_dev_addr *addr, void *context); unsigned long timeout; + struct delayed_work work; int status; u32 seq; }; @@ -268,6 +269,7 @@ int rdma_translate_ip(const struct sockaddr *addr, return ret; ret = rdma_copy_addr(dev_addr, dev, NULL); + dev_addr->bound_dev_if = dev->ifindex; if (vlan_id) *vlan_id = rdma_vlan_dev_vlan_id(dev); dev_put(dev); @@ -280,6 +282,7 @@ int rdma_translate_ip(const struct sockaddr *addr, &((const struct sockaddr_in6 *)addr)->sin6_addr, dev, 1)) { ret = rdma_copy_addr(dev_addr, dev, NULL); + dev_addr->bound_dev_if = dev->ifindex; if (vlan_id) *vlan_id = rdma_vlan_dev_vlan_id(dev); break; @@ -293,7 +296,7 @@ int rdma_translate_ip(const struct sockaddr *addr, } EXPORT_SYMBOL(rdma_translate_ip); -static void set_timeout(unsigned long time) +static void set_timeout(struct delayed_work *delayed_work, unsigned long time) { unsigned long delay; @@ -301,7 +304,7 @@ static void set_timeout(unsigned long time) if ((long)delay < 0) delay = 0; - mod_delayed_work(addr_wq, &work, delay); + mod_delayed_work(addr_wq, delayed_work, delay); } static void queue_req(struct addr_req *req) @@ -316,8 +319,7 @@ static void queue_req(struct addr_req *req) list_add(&req->list, &temp_req->list); - if (req_list.next == &req->list) - set_timeout(req->timeout); + set_timeout(&req->work, req->timeout); mutex_unlock(&lock); } @@ -405,10 +407,10 @@ static int addr4_resolve(struct sockaddr_in *src_in, fl4.saddr = src_ip; fl4.flowi4_oif = addr->bound_dev_if; rt = ip_route_output_key(addr->net, &fl4); - if (IS_ERR(rt)) { - ret = PTR_ERR(rt); - goto out; - } + ret = PTR_ERR_OR_ZERO(rt); + if (ret) + return ret; + src_in->sin_family = AF_INET; src_in->sin_addr.s_addr = fl4.saddr; @@ -423,8 +425,6 @@ static int addr4_resolve(struct sockaddr_in *src_in, *prt = rt; return 0; -out: - return ret; } #if IS_ENABLED(CONFIG_IPV6) @@ -509,6 +509,11 @@ static int addr_resolve(struct sockaddr *src_in, struct dst_entry *dst; int ret; + if (!addr->net) { + pr_warn_ratelimited("%s: missing namespace\n", __func__); + return -EINVAL; + } + if (src_in->sa_family == AF_INET) { struct rtable *rt = NULL; const struct sockaddr_in *dst_in4 = @@ -522,8 +527,12 @@ static int addr_resolve(struct sockaddr *src_in, if (resolve_neigh) ret = addr_resolve_neigh(&rt->dst, dst_in, addr, seq); - ndev = rt->dst.dev; - dev_hold(ndev); + if (addr->bound_dev_if) { + ndev = dev_get_by_index(addr->net, addr->bound_dev_if); + } else { + ndev = rt->dst.dev; + dev_hold(ndev); + } ip_rt_put(rt); } else { @@ -539,19 +548,63 @@ static int addr_resolve(struct sockaddr *src_in, if (resolve_neigh) ret = addr_resolve_neigh(dst, dst_in, addr, seq); - ndev = dst->dev; - dev_hold(ndev); + if (addr->bound_dev_if) { + ndev = dev_get_by_index(addr->net, addr->bound_dev_if); + } else { + ndev = dst->dev; + dev_hold(ndev); + } dst_release(dst); } - addr->bound_dev_if = ndev->ifindex; - addr->net = dev_net(ndev); + if (ndev->flags & IFF_LOOPBACK) { + ret = rdma_translate_ip(dst_in, addr, NULL); + /* + * Put the loopback device and get the translated + * device instead. + */ + dev_put(ndev); + ndev = dev_get_by_index(addr->net, addr->bound_dev_if); + } else { + addr->bound_dev_if = ndev->ifindex; + } dev_put(ndev); return ret; } +static void process_one_req(struct work_struct *_work) +{ + struct addr_req *req; + struct sockaddr *src_in, *dst_in; + + mutex_lock(&lock); + req = container_of(_work, struct addr_req, work.work); + + if (req->status == -ENODATA) { + src_in = (struct sockaddr *)&req->src_addr; + dst_in = (struct sockaddr *)&req->dst_addr; + req->status = addr_resolve(src_in, dst_in, req->addr, + true, req->seq); + if (req->status && time_after_eq(jiffies, req->timeout)) { + req->status = -ETIMEDOUT; + } else if (req->status == -ENODATA) { + /* requeue the work for retrying again */ + set_timeout(&req->work, req->timeout); + mutex_unlock(&lock); + return; + } + } + list_del(&req->list); + mutex_unlock(&lock); + + req->callback(req->status, (struct sockaddr *)&req->src_addr, + req->addr, req->context); + put_client(req->client); + kfree(req); +} + static void process_req(struct work_struct *work) { struct addr_req *req, *temp_req; @@ -569,20 +622,23 @@ static void process_req(struct work_struct *work) true, req->seq); if (req->status && time_after_eq(jiffies, req->timeout)) req->status = -ETIMEDOUT; - else if (req->status == -ENODATA) + else if (req->status == -ENODATA) { + set_timeout(&req->work, req->timeout); continue; + } } list_move_tail(&req->list, &done_list); } - if (!list_empty(&req_list)) { - req = list_entry(req_list.next, struct addr_req, list); - set_timeout(req->timeout); - } mutex_unlock(&lock); list_for_each_entry_safe(req, temp_req, &done_list, list) { list_del(&req->list); + /* It is safe to cancel other work items from this work item + * because at a time there can be only one work item running + * with this single threaded work queue. + */ + cancel_delayed_work(&req->work); req->callback(req->status, (struct sockaddr *) &req->src_addr, req->addr, req->context); put_client(req->client); @@ -625,6 +681,7 @@ int rdma_resolve_ip(struct rdma_addr_client *client, req->context = context; req->client = client; atomic_inc(&client->refcount); + INIT_DELAYED_WORK(&req->work, process_one_req); req->seq = (u32)atomic_inc_return(&ib_nl_addr_request_seq); req->status = addr_resolve(src_in, dst_in, addr, true, req->seq); @@ -679,7 +736,7 @@ void rdma_addr_cancel(struct rdma_dev_addr *addr) req->status = -ECANCELED; req->timeout = jiffies; list_move(&req->list, &req_list); - set_timeout(req->timeout); + set_timeout(&req->work, req->timeout); break; } } @@ -785,9 +842,8 @@ static int netevent_callback(struct notifier_block *self, unsigned long event, if (event == NETEVENT_NEIGH_UPDATE) { struct neighbour *neigh = ctx; - if (neigh->nud_state & NUD_VALID) { - set_timeout(jiffies); - } + if (neigh->nud_state & NUD_VALID) + set_timeout(&work, jiffies); } return 0; } @@ -798,7 +854,7 @@ static struct notifier_block nb = { int addr_init(void) { - addr_wq = alloc_workqueue("ib_addr", WQ_MEM_RECLAIM, 0); + addr_wq = alloc_ordered_workqueue("ib_addr", WQ_MEM_RECLAIM); if (!addr_wq) return -ENOMEM; diff --git a/drivers/infiniband/core/cma.c b/drivers/infiniband/core/cma.c index 31bb82d8ecd7..0eb393237ba2 100644 --- a/drivers/infiniband/core/cma.c +++ b/drivers/infiniband/core/cma.c @@ -623,22 +623,11 @@ static inline int cma_validate_port(struct ib_device *device, u8 port, if ((dev_type != ARPHRD_INFINIBAND) && rdma_protocol_ib(device, port)) return ret; - if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) { + if (dev_type == ARPHRD_ETHER && rdma_protocol_roce(device, port)) ndev = dev_get_by_index(&init_net, bound_if_index); - if (ndev && ndev->flags & IFF_LOOPBACK) { - pr_info("detected loopback device\n"); - dev_put(ndev); - - if (!device->get_netdev) - return -EOPNOTSUPP; - - ndev = device->get_netdev(device, port); - if (!ndev) - return -ENODEV; - } - } else { + else gid_type = IB_GID_TYPE_IB; - } + ret = ib_find_cached_gid_by_port(device, gid, gid_type, port, ndev, NULL); @@ -1044,6 +1033,8 @@ int rdma_init_qp_attr(struct rdma_cm_id *id, struct ib_qp_attr *qp_attr, } else ret = iw_cm_init_qp_attr(id_priv->cm_id.iw, qp_attr, qp_attr_mask); + qp_attr->port_num = id_priv->id.port_num; + *qp_attr_mask |= IB_QP_PORT; } else ret = -ENOSYS; @@ -2569,21 +2560,6 @@ static int cma_resolve_iboe_route(struct rdma_id_private *id_priv) goto err2; } - if (ndev->flags & IFF_LOOPBACK) { - dev_put(ndev); - if (!id_priv->id.device->get_netdev) { - ret = -EOPNOTSUPP; - goto err2; - } - - ndev = id_priv->id.device->get_netdev(id_priv->id.device, - id_priv->id.port_num); - if (!ndev) { - ret = -ENODEV; - goto err2; - } - } - supported_gids = roce_gid_type_mask_support(id_priv->id.device, id_priv->id.port_num); gid_type = cma_route_gid_type(addr->dev_addr.network, diff --git a/drivers/infiniband/core/roce_gid_mgmt.c b/drivers/infiniband/core/roce_gid_mgmt.c index db958d3207ef..94a9eefb3cfc 100644 --- a/drivers/infiniband/core/roce_gid_mgmt.c +++ b/drivers/infiniband/core/roce_gid_mgmt.c @@ -42,6 +42,8 @@ #include <rdma/ib_cache.h> #include <rdma/ib_addr.h> +static struct workqueue_struct *gid_cache_wq; + enum gid_op_type { GID_DEL = 0, GID_ADD @@ -560,7 +562,7 @@ static int netdevice_queue_work(struct netdev_event_work_cmd *cmds, } INIT_WORK(&ndev_work->work, netdevice_event_work_handler); - queue_work(ib_wq, &ndev_work->work); + queue_work(gid_cache_wq, &ndev_work->work); return NOTIFY_DONE; } @@ -693,7 +695,7 @@ static int addr_event(struct notifier_block *this, unsigned long event, dev_hold(ndev); work->gid_attr.ndev = ndev; - queue_work(ib_wq, &work->work); + queue_work(gid_cache_wq, &work->work); return NOTIFY_DONE; } @@ -740,6 +742,10 @@ static struct notifier_block nb_inet6addr = { int __init roce_gid_mgmt_init(void) { + gid_cache_wq = alloc_ordered_workqueue("gid-cache-wq", 0); + if (!gid_cache_wq) + return -ENOMEM; + register_inetaddr_notifier(&nb_inetaddr); if (IS_ENABLED(CONFIG_IPV6)) register_inet6addr_notifier(&nb_inet6addr); @@ -764,4 +770,5 @@ void __exit roce_gid_mgmt_cleanup(void) * ib-core is removed, all physical devices have been removed, * so no issue with remaining hardware contexts. */ + destroy_workqueue(gid_cache_wq); } diff --git a/drivers/infiniband/core/uverbs_cmd.c b/drivers/infiniband/core/uverbs_cmd.c index 8ba9bfb073d1..c551d2b275fd 100644 --- a/drivers/infiniband/core/uverbs_cmd.c +++ b/drivers/infiniband/core/uverbs_cmd.c @@ -1153,7 +1153,7 @@ ssize_t ib_uverbs_resize_cq(struct ib_uverbs_file *file, int out_len) { struct ib_uverbs_resize_cq cmd; - struct ib_uverbs_resize_cq_resp resp; + struct ib_uverbs_resize_cq_resp resp = {}; struct ib_udata udata; struct ib_cq *cq; int ret = -EINVAL; @@ -1296,7 +1296,6 @@ ssize_t ib_uverbs_destroy_cq(struct ib_uverbs_file *file, struct ib_uobject *uobj; struct ib_cq *cq; struct ib_ucq_object *obj; - struct ib_uverbs_event_queue *ev_queue; int ret = -EINVAL; if (copy_from_user(&cmd, buf, sizeof cmd)) @@ -1313,7 +1312,6 @@ ssize_t ib_uverbs_destroy_cq(struct ib_uverbs_file *file, */ uverbs_uobject_get(uobj); cq = uobj->object; - ev_queue = cq->cq_context; obj = container_of(cq->uobject, struct ib_ucq_object, uobject); memset(&resp, 0, sizeof(resp)); @@ -1935,7 +1933,8 @@ static int modify_qp(struct ib_uverbs_file *file, goto out; } - if (!rdma_is_port_valid(qp->device, cmd->base.port_num)) { + if ((cmd->base.attr_mask & IB_QP_PORT) && + !rdma_is_port_valid(qp->device, cmd->base.port_num)) { ret = -EINVAL; goto release_qp; } @@ -2005,28 +2004,13 @@ static int modify_qp(struct ib_uverbs_file *file, rdma_ah_set_port_num(&attr->alt_ah_attr, cmd->base.alt_dest.port_num); - if (qp->real_qp == qp) { - if (cmd->base.attr_mask & IB_QP_AV) { - ret = ib_resolve_eth_dmac(qp->device, &attr->ah_attr); - if (ret) - goto release_qp; - } - ret = ib_security_modify_qp(qp, - attr, - modify_qp_mask(qp->qp_type, - cmd->base.attr_mask), - udata); - } else { - ret = ib_security_modify_qp(qp, - attr, - modify_qp_mask(qp->qp_type, - cmd->base.attr_mask), - NULL); - } + ret = ib_modify_qp_with_udata(qp, attr, + modify_qp_mask(qp->qp_type, + cmd->base.attr_mask), + udata); release_qp: uobj_put_obj_read(qp); - out: kfree(attr); @@ -2103,7 +2087,6 @@ ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file, struct ib_uverbs_destroy_qp cmd; struct ib_uverbs_destroy_qp_resp resp; struct ib_uobject *uobj; - struct ib_qp *qp; struct ib_uqp_object *obj; int ret = -EINVAL; @@ -2117,7 +2100,6 @@ ssize_t ib_uverbs_destroy_qp(struct ib_uverbs_file *file, if (IS_ERR(uobj)) return PTR_ERR(uobj); - qp = uobj->object; obj = container_of(uobj, struct ib_uqp_object, uevent.uobject); /* * Make sure we don't free the memory in remove_commit as we still @@ -3019,7 +3001,6 @@ int ib_uverbs_ex_destroy_wq(struct ib_uverbs_file *file, { struct ib_uverbs_ex_destroy_wq cmd = {}; struct ib_uverbs_ex_destroy_wq_resp resp = {}; - struct ib_wq *wq; struct ib_uobject *uobj; struct ib_uwq_object *obj; size_t required_cmd_sz; @@ -3053,7 +3034,6 @@ int ib_uverbs_ex_destroy_wq(struct ib_uverbs_file *file, if (IS_ERR(uobj)) return PTR_ERR(uobj); - wq = uobj->object; obj = container_of(uobj, struct ib_uwq_object, uevent.uobject); /* * Make sure we don't free the memory in remove_commit as we still @@ -3743,10 +3723,8 @@ ssize_t ib_uverbs_destroy_srq(struct ib_uverbs_file *file, struct ib_uverbs_destroy_srq cmd; struct ib_uverbs_destroy_srq_resp resp; struct ib_uobject *uobj; - struct ib_srq *srq; struct ib_uevent_object *obj; int ret = -EINVAL; - enum ib_srq_type srq_type; if (copy_from_user(&cmd, buf, sizeof cmd)) return -EFAULT; @@ -3756,9 +3734,7 @@ ssize_t ib_uverbs_destroy_srq(struct ib_uverbs_file *file, if (IS_ERR(uobj)) return PTR_ERR(uobj); - srq = uobj->object; obj = container_of(uobj, struct ib_uevent_object, uobject); - srq_type = srq->srq_type; /* * Make sure we don't free the memory in remove_commit as we still * needs the uobject memory to create the response. diff --git a/drivers/infiniband/core/uverbs_main.c b/drivers/infiniband/core/uverbs_main.c index 3d2609608f58..c023e2c81b8f 100644 --- a/drivers/infiniband/core/uverbs_main.c +++ b/drivers/infiniband/core/uverbs_main.c @@ -250,6 +250,7 @@ void ib_uverbs_release_file(struct kref *ref) if (atomic_dec_and_test(&file->device->refcount)) ib_uverbs_comp_dev(file->device); + kobject_put(&file->device->kobj); kfree(file); } @@ -917,7 +918,6 @@ err: static int ib_uverbs_close(struct inode *inode, struct file *filp) { struct ib_uverbs_file *file = filp->private_data; - struct ib_uverbs_device *dev = file->device; mutex_lock(&file->cleanup_mutex); if (file->ucontext) { @@ -939,7 +939,6 @@ static int ib_uverbs_close(struct inode *inode, struct file *filp) ib_uverbs_release_async_event_file); kref_put(&file->ref, ib_uverbs_release_file); - kobject_put(&dev->kobj); return 0; } diff --git a/drivers/infiniband/core/verbs.c b/drivers/infiniband/core/verbs.c index c973a83c898b..7f8fe443df46 100644 --- a/drivers/infiniband/core/verbs.c +++ b/drivers/infiniband/core/verbs.c @@ -452,6 +452,19 @@ int ib_get_gids_from_rdma_hdr(const union rdma_network_hdr *hdr, } EXPORT_SYMBOL(ib_get_gids_from_rdma_hdr); +/* + * This function creates ah from the incoming packet. + * Incoming packet has dgid of the receiver node on which this code is + * getting executed and, sgid contains the GID of the sender. + * + * When resolving mac address of destination, the arrived dgid is used + * as sgid and, sgid is used as dgid because sgid contains destinations + * GID whom to respond to. + * + * This is why when calling rdma_addr_find_l2_eth_by_grh() function, the + * position of arguments dgid and sgid do not match the order of the + * parameters. + */ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, const struct ib_wc *wc, const struct ib_grh *grh, struct rdma_ah_attr *ah_attr) @@ -507,11 +520,6 @@ int ib_init_ah_from_wc(struct ib_device *device, u8 port_num, } resolved_dev = dev_get_by_index(&init_net, if_index); - if (resolved_dev->flags & IFF_LOOPBACK) { - dev_put(resolved_dev); - resolved_dev = idev; - dev_hold(resolved_dev); - } rcu_read_lock(); if (resolved_dev != idev && !rdma_is_upper_dev_rcu(idev, resolved_dev)) @@ -1268,20 +1276,36 @@ out: } EXPORT_SYMBOL(ib_resolve_eth_dmac); -int ib_modify_qp(struct ib_qp *qp, - struct ib_qp_attr *qp_attr, - int qp_attr_mask) +/** + * ib_modify_qp_with_udata - Modifies the attributes for the specified QP. + * @qp: The QP to modify. + * @attr: On input, specifies the QP attributes to modify. On output, + * the current values of selected QP attributes are returned. + * @attr_mask: A bit-mask used to specify which attributes of the QP + * are being modified. + * @udata: pointer to user's input output buffer information + * are being modified. + * It returns 0 on success and returns appropriate error code on error. + */ +int ib_modify_qp_with_udata(struct ib_qp *qp, struct ib_qp_attr *attr, + int attr_mask, struct ib_udata *udata) { + int ret; - if (qp_attr_mask & IB_QP_AV) { - int ret; - - ret = ib_resolve_eth_dmac(qp->device, &qp_attr->ah_attr); + if (attr_mask & IB_QP_AV) { + ret = ib_resolve_eth_dmac(qp->device, &attr->ah_attr); if (ret) return ret; } + return ib_security_modify_qp(qp, attr, attr_mask, udata); +} +EXPORT_SYMBOL(ib_modify_qp_with_udata); - return ib_security_modify_qp(qp->real_qp, qp_attr, qp_attr_mask, NULL); +int ib_modify_qp(struct ib_qp *qp, + struct ib_qp_attr *qp_attr, + int qp_attr_mask) +{ + return ib_modify_qp_with_udata(qp, qp_attr, qp_attr_mask, NULL); } EXPORT_SYMBOL(ib_modify_qp); diff --git a/drivers/infiniband/hw/bnxt_re/bnxt_re.h b/drivers/infiniband/hw/bnxt_re/bnxt_re.h index 08772836fded..85527532c49d 100644 --- a/drivers/infiniband/hw/bnxt_re/bnxt_re.h +++ b/drivers/infiniband/hw/bnxt_re/bnxt_re.h @@ -51,6 +51,8 @@ #define BNXT_RE_PAGE_SIZE_8M BIT(23) #define BNXT_RE_PAGE_SIZE_1G BIT(30) +#define BNXT_RE_MAX_MR_SIZE BIT(30) + #define BNXT_RE_MAX_QPC_COUNT (64 * 1024) #define BNXT_RE_MAX_MRW_COUNT (64 * 1024) #define BNXT_RE_MAX_SRQC_COUNT (64 * 1024) @@ -60,6 +62,13 @@ #define BNXT_RE_RQ_WQE_THRESHOLD 32 +/* + * Setting the default ack delay value to 16, which means + * the default timeout is approx. 260ms(4 usec * 2 ^(timeout)) + */ + +#define BNXT_RE_DEFAULT_ACK_DELAY 16 + struct bnxt_re_work { struct work_struct work; unsigned long event; diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.c b/drivers/infiniband/hw/bnxt_re/ib_verbs.c index c7bd68311d0c..f0e01b3ac711 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.c +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.c @@ -145,10 +145,8 @@ int bnxt_re_query_device(struct ib_device *ibdev, ib_attr->fw_ver = (u64)(unsigned long)(dev_attr->fw_ver); bnxt_qplib_get_guid(rdev->netdev->dev_addr, (u8 *)&ib_attr->sys_image_guid); - ib_attr->max_mr_size = ~0ull; - ib_attr->page_size_cap = BNXT_RE_PAGE_SIZE_4K | BNXT_RE_PAGE_SIZE_8K | - BNXT_RE_PAGE_SIZE_64K | BNXT_RE_PAGE_SIZE_2M | - BNXT_RE_PAGE_SIZE_8M | BNXT_RE_PAGE_SIZE_1G; + ib_attr->max_mr_size = BNXT_RE_MAX_MR_SIZE; + ib_attr->page_size_cap = BNXT_RE_PAGE_SIZE_4K; ib_attr->vendor_id = rdev->en_dev->pdev->vendor; ib_attr->vendor_part_id = rdev->en_dev->pdev->device; @@ -174,9 +172,11 @@ int bnxt_re_query_device(struct ib_device *ibdev, ib_attr->max_mr = dev_attr->max_mr; ib_attr->max_pd = dev_attr->max_pd; ib_attr->max_qp_rd_atom = dev_attr->max_qp_rd_atom; - ib_attr->max_qp_init_rd_atom = dev_attr->max_qp_rd_atom; - ib_attr->atomic_cap = IB_ATOMIC_HCA; - ib_attr->masked_atomic_cap = IB_ATOMIC_HCA; + ib_attr->max_qp_init_rd_atom = dev_attr->max_qp_init_rd_atom; + if (dev_attr->is_atomic) { + ib_attr->atomic_cap = IB_ATOMIC_HCA; + ib_attr->masked_atomic_cap = IB_ATOMIC_HCA; + } ib_attr->max_ee_rd_atom = 0; ib_attr->max_res_rd_atom = 0; @@ -201,7 +201,7 @@ int bnxt_re_query_device(struct ib_device *ibdev, ib_attr->max_fast_reg_page_list_len = MAX_PBL_LVL_1_PGS; ib_attr->max_pkeys = 1; - ib_attr->local_ca_ack_delay = 0; + ib_attr->local_ca_ack_delay = BNXT_RE_DEFAULT_ACK_DELAY; return 0; } @@ -390,15 +390,17 @@ int bnxt_re_del_gid(struct ib_device *ibdev, u8 port_num, return -EINVAL; ctx->refcnt--; if (!ctx->refcnt) { - rc = bnxt_qplib_del_sgid - (sgid_tbl, - &sgid_tbl->tbl[ctx->idx], true); - if (rc) + rc = bnxt_qplib_del_sgid(sgid_tbl, + &sgid_tbl->tbl[ctx->idx], + true); + if (rc) { dev_err(rdev_to_dev(rdev), "Failed to remove GID: %#x", rc); - ctx_tbl = sgid_tbl->ctx; - ctx_tbl[ctx->idx] = NULL; - kfree(ctx); + } else { + ctx_tbl = sgid_tbl->ctx; + ctx_tbl[ctx->idx] = NULL; + kfree(ctx); + } } } else { return -EINVAL; @@ -588,10 +590,10 @@ static int bnxt_re_create_fence_mr(struct bnxt_re_pd *pd) /* Create a fence MW only for kernel consumers */ mw = bnxt_re_alloc_mw(&pd->ib_pd, IB_MW_TYPE_1, NULL); - if (!mw) { + if (IS_ERR(mw)) { dev_err(rdev_to_dev(rdev), "Failed to create fence-MW for PD: %p\n", pd); - rc = -EINVAL; + rc = PTR_ERR(mw); goto fail; } fence->mw = mw; @@ -612,30 +614,13 @@ int bnxt_re_dealloc_pd(struct ib_pd *ib_pd) int rc; bnxt_re_destroy_fence_mr(pd); - if (ib_pd->uobject && pd->dpi.dbr) { - struct ib_ucontext *ib_uctx = ib_pd->uobject->context; - struct bnxt_re_ucontext *ucntx; - /* Free DPI only if this is the first PD allocated by the - * application and mark the context dpi as NULL - */ - ucntx = container_of(ib_uctx, struct bnxt_re_ucontext, ib_uctx); - - rc = bnxt_qplib_dealloc_dpi(&rdev->qplib_res, - &rdev->qplib_res.dpi_tbl, - &pd->dpi); + if (pd->qplib_pd.id) { + rc = bnxt_qplib_dealloc_pd(&rdev->qplib_res, + &rdev->qplib_res.pd_tbl, + &pd->qplib_pd); if (rc) - dev_err(rdev_to_dev(rdev), "Failed to deallocate HW DPI"); - /* Don't fail, continue*/ - ucntx->dpi = NULL; - } - - rc = bnxt_qplib_dealloc_pd(&rdev->qplib_res, - &rdev->qplib_res.pd_tbl, - &pd->qplib_pd); - if (rc) { - dev_err(rdev_to_dev(rdev), "Failed to deallocate HW PD"); - return rc; + dev_err(rdev_to_dev(rdev), "Failed to deallocate HW PD"); } kfree(pd); @@ -667,23 +652,22 @@ struct ib_pd *bnxt_re_alloc_pd(struct ib_device *ibdev, if (udata) { struct bnxt_re_pd_resp resp; - if (!ucntx->dpi) { + if (!ucntx->dpi.dbr) { /* Allocate DPI in alloc_pd to avoid failing of * ibv_devinfo and family of application when DPIs * are depleted. */ if (bnxt_qplib_alloc_dpi(&rdev->qplib_res.dpi_tbl, - &pd->dpi, ucntx)) { + &ucntx->dpi, ucntx)) { rc = -ENOMEM; goto dbfail; } - ucntx->dpi = &pd->dpi; } resp.pdid = pd->qplib_pd.id; /* Still allow mapping this DBR to the new user PD. */ - resp.dpi = ucntx->dpi->dpi; - resp.dbr = (u64)ucntx->dpi->umdbr; + resp.dpi = ucntx->dpi.dpi; + resp.dbr = (u64)ucntx->dpi.umdbr; rc = ib_copy_to_udata(udata, &resp, sizeof(resp)); if (rc) { @@ -960,7 +944,7 @@ static int bnxt_re_init_user_qp(struct bnxt_re_dev *rdev, struct bnxt_re_pd *pd, qplib_qp->rq.nmap = umem->nmap; } - qplib_qp->dpi = cntx->dpi; + qplib_qp->dpi = &cntx->dpi; return 0; rqfail: ib_umem_release(qp->sumem); @@ -1530,13 +1514,24 @@ int bnxt_re_modify_qp(struct ib_qp *ib_qp, struct ib_qp_attr *qp_attr, if (qp_attr_mask & IB_QP_MAX_QP_RD_ATOMIC) { qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_MAX_RD_ATOMIC; - qp->qplib_qp.max_rd_atomic = qp_attr->max_rd_atomic; + /* Cap the max_rd_atomic to device max */ + qp->qplib_qp.max_rd_atomic = min_t(u32, qp_attr->max_rd_atomic, + dev_attr->max_qp_rd_atom); } if (qp_attr_mask & IB_QP_SQ_PSN) { qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_SQ_PSN; qp->qplib_qp.sq.psn = qp_attr->sq_psn; } if (qp_attr_mask & IB_QP_MAX_DEST_RD_ATOMIC) { + if (qp_attr->max_dest_rd_atomic > + dev_attr->max_qp_init_rd_atom) { + dev_err(rdev_to_dev(rdev), + "max_dest_rd_atomic requested%d is > dev_max%d", + qp_attr->max_dest_rd_atomic, + dev_attr->max_qp_init_rd_atom); + return -EINVAL; + } + qp->qplib_qp.modify_flags |= CMDQ_MODIFY_QP_MODIFY_MASK_MAX_DEST_RD_ATOMIC; qp->qplib_qp.max_dest_rd_atomic = qp_attr->max_dest_rd_atomic; @@ -2403,7 +2398,7 @@ struct ib_cq *bnxt_re_create_cq(struct ib_device *ibdev, } cq->qplib_cq.sghead = cq->umem->sg_head.sgl; cq->qplib_cq.nmap = cq->umem->nmap; - cq->qplib_cq.dpi = uctx->dpi; + cq->qplib_cq.dpi = &uctx->dpi; } else { cq->max_cql = min_t(u32, entries, MAX_CQL_PER_POLL); cq->cql = kcalloc(cq->max_cql, sizeof(struct bnxt_qplib_cqe), @@ -2905,6 +2900,7 @@ int bnxt_re_poll_cq(struct ib_cq *ib_cq, int num_entries, struct ib_wc *wc) spin_lock_irqsave(&cq->cq_lock, flags); budget = min_t(u32, num_entries, cq->max_cql); + num_entries = budget; if (!cq->cql) { dev_err(rdev_to_dev(cq->rdev), "POLL CQ : no CQL to use"); goto exit; @@ -3031,6 +3027,11 @@ int bnxt_re_req_notify_cq(struct ib_cq *ib_cq, else if (ib_cqn_flags & IB_CQ_SOLICITED) type = DBR_DBR_TYPE_CQ_ARMSE; + /* Poll to see if there are missed events */ + if ((ib_cqn_flags & IB_CQ_REPORT_MISSED_EVENTS) && + !(bnxt_qplib_is_cq_empty(&cq->qplib_cq))) + return 1; + bnxt_qplib_req_notify_cq(&cq->qplib_cq, type); return 0; @@ -3245,6 +3246,12 @@ struct ib_mr *bnxt_re_reg_user_mr(struct ib_pd *ib_pd, u64 start, u64 length, struct scatterlist *sg; int entry; + if (length > BNXT_RE_MAX_MR_SIZE) { + dev_err(rdev_to_dev(rdev), "MR Size: %lld > Max supported:%ld\n", + length, BNXT_RE_MAX_MR_SIZE); + return ERR_PTR(-ENOMEM); + } + mr = kzalloc(sizeof(*mr), GFP_KERNEL); if (!mr) return ERR_PTR(-ENOMEM); @@ -3388,8 +3395,26 @@ int bnxt_re_dealloc_ucontext(struct ib_ucontext *ib_uctx) struct bnxt_re_ucontext *uctx = container_of(ib_uctx, struct bnxt_re_ucontext, ib_uctx); + + struct bnxt_re_dev *rdev = uctx->rdev; + int rc = 0; + if (uctx->shpg) free_page((unsigned long)uctx->shpg); + + if (uctx->dpi.dbr) { + /* Free DPI only if this is the first PD allocated by the + * application and mark the context dpi as NULL + */ + rc = bnxt_qplib_dealloc_dpi(&rdev->qplib_res, + &rdev->qplib_res.dpi_tbl, + &uctx->dpi); + if (rc) + dev_err(rdev_to_dev(rdev), "Deallocte HW DPI failed!"); + /* Don't fail, continue*/ + uctx->dpi.dbr = NULL; + } + kfree(uctx); return 0; } diff --git a/drivers/infiniband/hw/bnxt_re/ib_verbs.h b/drivers/infiniband/hw/bnxt_re/ib_verbs.h index 6c160f6a5398..a0bb7e33d7ca 100644 --- a/drivers/infiniband/hw/bnxt_re/ib_verbs.h +++ b/drivers/infiniband/hw/bnxt_re/ib_verbs.h @@ -59,7 +59,6 @@ struct bnxt_re_pd { struct bnxt_re_dev *rdev; struct ib_pd ib_pd; struct bnxt_qplib_pd qplib_pd; - struct bnxt_qplib_dpi dpi; struct bnxt_re_fence_data fence; }; @@ -127,7 +126,7 @@ struct bnxt_re_mw { struct bnxt_re_ucontext { struct bnxt_re_dev *rdev; struct ib_ucontext ib_uctx; - struct bnxt_qplib_dpi *dpi; + struct bnxt_qplib_dpi dpi; void *shpg; spinlock_t sh_lock; /* protect shpg */ }; diff --git a/drivers/infiniband/hw/bnxt_re/main.c b/drivers/infiniband/hw/bnxt_re/main.c index 1fce5e73216b..ceae2d92fb08 100644 --- a/drivers/infiniband/hw/bnxt_re/main.c +++ b/drivers/infiniband/hw/bnxt_re/main.c @@ -333,6 +333,7 @@ static int bnxt_re_net_stats_ctx_alloc(struct bnxt_re_dev *rdev, bnxt_re_init_hwrm_hdr(rdev, (void *)&req, HWRM_STAT_CTX_ALLOC, -1, -1); req.update_period_ms = cpu_to_le32(1000); req.stats_dma_addr = cpu_to_le64(dma_map); + req.stat_ctx_flags = STAT_CTX_ALLOC_REQ_STAT_CTX_FLAGS_ROCE; bnxt_re_fill_fw_msg(&fw_msg, (void *)&req, sizeof(req), (void *)&resp, sizeof(resp), DFLT_HWRM_CMD_TIMEOUT); rc = en_dev->en_ops->bnxt_send_fw_msg(en_dev, BNXT_ROCE_ULP, &fw_msg); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.c b/drivers/infiniband/hw/bnxt_re/qplib_fp.c index f05500bcdcf1..9af1514e5944 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.c @@ -1128,6 +1128,11 @@ int bnxt_qplib_post_send(struct bnxt_qplib_qp *qp, } /* Each SGE entry = 1 WQE size16 */ wqe_size16 = wqe->num_sge; + /* HW requires wqe size has room for atleast one SGE even if + * none was supplied by ULP + */ + if (!wqe->num_sge) + wqe_size16++; } /* Specifics */ @@ -1364,6 +1369,11 @@ int bnxt_qplib_post_recv(struct bnxt_qplib_qp *qp, rqe->flags = wqe->flags; rqe->wqe_size = wqe->num_sge + ((offsetof(typeof(*rqe), data) + 15) >> 4); + /* HW requires wqe size has room for atleast one SGE even if none + * was supplied by ULP + */ + if (!wqe->num_sge) + rqe->wqe_size++; /* Supply the rqe->wr_id index to the wr_id_tbl for now */ rqe->wr_id[0] = cpu_to_le32(sw_prod); @@ -1885,6 +1895,25 @@ flush_rq: return rc; } +bool bnxt_qplib_is_cq_empty(struct bnxt_qplib_cq *cq) +{ + struct cq_base *hw_cqe, **hw_cqe_ptr; + unsigned long flags; + u32 sw_cons, raw_cons; + bool rc = true; + + spin_lock_irqsave(&cq->hwq.lock, flags); + raw_cons = cq->hwq.cons; + sw_cons = HWQ_CMP(raw_cons, &cq->hwq); + hw_cqe_ptr = (struct cq_base **)cq->hwq.pbl_ptr; + hw_cqe = &hw_cqe_ptr[CQE_PG(sw_cons)][CQE_IDX(sw_cons)]; + + /* Check for Valid bit. If the CQE is valid, return false */ + rc = !CQE_CMP_VALID(hw_cqe, raw_cons, cq->hwq.max_elements); + spin_unlock_irqrestore(&cq->hwq.lock, flags); + return rc; +} + static int bnxt_qplib_cq_process_res_raweth_qp1(struct bnxt_qplib_cq *cq, struct cq_res_raweth_qp1 *hwcqe, struct bnxt_qplib_cqe **pcqe, diff --git a/drivers/infiniband/hw/bnxt_re/qplib_fp.h b/drivers/infiniband/hw/bnxt_re/qplib_fp.h index 36b7b7db0e3f..19176e06c98a 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_fp.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_fp.h @@ -449,6 +449,7 @@ int bnxt_qplib_create_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq); int bnxt_qplib_destroy_cq(struct bnxt_qplib_res *res, struct bnxt_qplib_cq *cq); int bnxt_qplib_poll_cq(struct bnxt_qplib_cq *cq, struct bnxt_qplib_cqe *cqe, int num, struct bnxt_qplib_qp **qp); +bool bnxt_qplib_is_cq_empty(struct bnxt_qplib_cq *cq); void bnxt_qplib_req_notify_cq(struct bnxt_qplib_cq *cq, u32 arm_type); void bnxt_qplib_free_nq(struct bnxt_qplib_nq *nq); int bnxt_qplib_alloc_nq(struct pci_dev *pdev, struct bnxt_qplib_nq *nq); diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.c b/drivers/infiniband/hw/bnxt_re/qplib_sp.c index fde18cf0e406..ef91ab786dd4 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.c +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.c @@ -51,6 +51,19 @@ const struct bnxt_qplib_gid bnxt_qplib_gid_zero = {{ 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 } }; /* Device */ + +static bool bnxt_qplib_is_atomic_cap(struct bnxt_qplib_rcfw *rcfw) +{ + int rc; + u16 pcie_ctl2; + + rc = pcie_capability_read_word(rcfw->pdev, PCI_EXP_DEVCTL2, + &pcie_ctl2); + if (rc) + return false; + return !!(pcie_ctl2 & PCI_EXP_DEVCTL2_ATOMIC_REQ); +} + int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw, struct bnxt_qplib_dev_attr *attr) { @@ -81,6 +94,8 @@ int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw, /* Extract the context from the side buffer */ attr->max_qp = le32_to_cpu(sb->max_qp); + /* max_qp value reported by FW for PF doesn't include the QP1 for PF */ + attr->max_qp += 1; attr->max_qp_rd_atom = sb->max_qp_rd_atom > BNXT_QPLIB_MAX_OUT_RD_ATOM ? BNXT_QPLIB_MAX_OUT_RD_ATOM : sb->max_qp_rd_atom; @@ -129,6 +144,7 @@ int bnxt_qplib_get_dev_attr(struct bnxt_qplib_rcfw *rcfw, attr->tqm_alloc_reqs[i * 4 + 3] = *(++tqm_alloc); } + attr->is_atomic = bnxt_qplib_is_atomic_cap(rcfw); bail: bnxt_qplib_rcfw_free_sbuf(rcfw, sbuf); return rc; diff --git a/drivers/infiniband/hw/bnxt_re/qplib_sp.h b/drivers/infiniband/hw/bnxt_re/qplib_sp.h index a543f959098b..2ce7e2a32cf0 100644 --- a/drivers/infiniband/hw/bnxt_re/qplib_sp.h +++ b/drivers/infiniband/hw/bnxt_re/qplib_sp.h @@ -42,6 +42,8 @@ #define BNXT_QPLIB_RESERVED_QP_WRS 128 +#define PCI_EXP_DEVCTL2_ATOMIC_REQ 0x0040 + struct bnxt_qplib_dev_attr { char fw_ver[32]; u16 max_sgid; @@ -70,6 +72,7 @@ struct bnxt_qplib_dev_attr { u32 max_inline_data; u32 l2_db_size; u8 tqm_alloc_reqs[MAX_TQM_ALLOC_REQ]; + bool is_atomic; }; struct bnxt_qplib_pd { diff --git a/drivers/infiniband/hw/cxgb3/iwch_provider.c b/drivers/infiniband/hw/cxgb3/iwch_provider.c index 29d30744d6c9..0cd0c1fa27d4 100644 --- a/drivers/infiniband/hw/cxgb3/iwch_provider.c +++ b/drivers/infiniband/hw/cxgb3/iwch_provider.c @@ -718,7 +718,7 @@ static struct ib_mr *iwch_alloc_mr(struct ib_pd *pd, struct iwch_mr *mhp; u32 mmid; u32 stag = 0; - int ret = 0; + int ret = -ENOMEM; if (mr_type != IB_MR_TYPE_MEM_REG || max_num_sg > T3_MAX_FASTREG_DEPTH) @@ -731,10 +731,8 @@ static struct ib_mr *iwch_alloc_mr(struct ib_pd *pd, goto err; mhp->pages = kcalloc(max_num_sg, sizeof(u64), GFP_KERNEL); - if (!mhp->pages) { - ret = -ENOMEM; + if (!mhp->pages) goto pl_err; - } mhp->rhp = rhp; ret = iwch_alloc_pbl(mhp, max_num_sg); @@ -751,7 +749,8 @@ static struct ib_mr *iwch_alloc_mr(struct ib_pd *pd, mhp->attr.state = 1; mmid = (stag) >> 8; mhp->ibmr.rkey = mhp->ibmr.lkey = stag; - if (insert_handle(rhp, &rhp->mmidr, mhp, mmid)) + ret = insert_handle(rhp, &rhp->mmidr, mhp, mmid); + if (ret) goto err3; pr_debug("%s mmid 0x%x mhp %p stag 0x%x\n", __func__, mmid, mhp, stag); diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index e16fcaf6b5a3..be07da1997e6 100644 --- a/drivers/infiniband/hw/cxgb4/cq.c +++ b/drivers/infiniband/hw/cxgb4/cq.c @@ -963,6 +963,7 @@ struct ib_cq *c4iw_create_cq(struct ib_device *ibdev, goto err3; if (ucontext) { + ret = -ENOMEM; mm = kmalloc(sizeof *mm, GFP_KERNEL); if (!mm) goto err4; diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c index bfc77596acbe..cb7fc0d35d1d 100644 --- a/drivers/infiniband/hw/cxgb4/qp.c +++ b/drivers/infiniband/hw/cxgb4/qp.c @@ -569,7 +569,7 @@ static int build_rdma_read(union t4_wr *wqe, struct ib_send_wr *wr, u8 *len16) { if (wr->num_sge > 1) return -EINVAL; - if (wr->num_sge) { + if (wr->num_sge && wr->sg_list[0].length) { wqe->read.stag_src = cpu_to_be32(rdma_wr(wr)->rkey); wqe->read.to_src_hi = cpu_to_be32((u32)(rdma_wr(wr)->remote_addr >> 32)); diff --git a/drivers/infiniband/hw/hfi1/chip.c b/drivers/infiniband/hw/hfi1/chip.c index 2ba00b89df6a..94b54850ec75 100644 --- a/drivers/infiniband/hw/hfi1/chip.c +++ b/drivers/infiniband/hw/hfi1/chip.c @@ -12847,7 +12847,12 @@ static void remap_intr(struct hfi1_devdata *dd, int isrc, int msix_intr) /* clear from the handled mask of the general interrupt */ m = isrc / 64; n = isrc % 64; - dd->gi_mask[m] &= ~((u64)1 << n); + if (likely(m < CCE_NUM_INT_CSRS)) { + dd->gi_mask[m] &= ~((u64)1 << n); + } else { + dd_dev_err(dd, "remap interrupt err\n"); + return; + } /* direct the chip source to the given MSI-X interrupt */ m = isrc / 8; diff --git a/drivers/infiniband/hw/hfi1/qp.c b/drivers/infiniband/hw/hfi1/qp.c index 650305cc0373..1a7af9f60c13 100644 --- a/drivers/infiniband/hw/hfi1/qp.c +++ b/drivers/infiniband/hw/hfi1/qp.c @@ -647,18 +647,17 @@ void qp_iter_print(struct seq_file *s, struct qp_iter *iter) qp->pid); } -void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, - gfp_t gfp) +void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp) { struct hfi1_qp_priv *priv; - priv = kzalloc_node(sizeof(*priv), gfp, rdi->dparms.node); + priv = kzalloc_node(sizeof(*priv), GFP_KERNEL, rdi->dparms.node); if (!priv) return ERR_PTR(-ENOMEM); priv->owner = qp; - priv->s_ahg = kzalloc_node(sizeof(*priv->s_ahg), gfp, + priv->s_ahg = kzalloc_node(sizeof(*priv->s_ahg), GFP_KERNEL, rdi->dparms.node); if (!priv->s_ahg) { kfree(priv); diff --git a/drivers/infiniband/hw/hfi1/qp.h b/drivers/infiniband/hw/hfi1/qp.h index 1eb9cd7b8c19..6fe542b6a927 100644 --- a/drivers/infiniband/hw/hfi1/qp.h +++ b/drivers/infiniband/hw/hfi1/qp.h @@ -123,8 +123,7 @@ void hfi1_migrate_qp(struct rvt_qp *qp); /* * Functions provided by hfi1 driver for rdmavt to use */ -void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, - gfp_t gfp); +void *qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp); void qp_priv_free(struct rvt_dev_info *rdi, struct rvt_qp *qp); unsigned free_all_qps(struct rvt_dev_info *rdi); void notify_qp_reset(struct rvt_qp *qp); diff --git a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c index 37d5d29597a4..2540b65e242c 100644 --- a/drivers/infiniband/hw/hns/hns_roce_hw_v1.c +++ b/drivers/infiniband/hw/hns/hns_roce_hw_v1.c @@ -228,14 +228,14 @@ int hns_roce_v1_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr, switch (wr->opcode) { case IB_WR_RDMA_READ: ps_opcode = HNS_ROCE_WQE_OPCODE_RDMA_READ; - set_raddr_seg(wqe, atomic_wr(wr)->remote_addr, - atomic_wr(wr)->rkey); + set_raddr_seg(wqe, rdma_wr(wr)->remote_addr, + rdma_wr(wr)->rkey); break; case IB_WR_RDMA_WRITE: case IB_WR_RDMA_WRITE_WITH_IMM: ps_opcode = HNS_ROCE_WQE_OPCODE_RDMA_WRITE; - set_raddr_seg(wqe, atomic_wr(wr)->remote_addr, - atomic_wr(wr)->rkey); + set_raddr_seg(wqe, rdma_wr(wr)->remote_addr, + rdma_wr(wr)->rkey); break; case IB_WR_SEND: case IB_WR_SEND_WITH_INV: @@ -661,9 +661,11 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) union ib_gid dgid; u64 subnet_prefix; int attr_mask = 0; - int i; + int i, j; int ret; + u8 queue_en[HNS_ROCE_V1_RESV_QP] = { 0 }; u8 phy_port; + u8 port = 0; u8 sl; priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; @@ -709,27 +711,35 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) attr.rnr_retry = 7; attr.timeout = 0x12; attr.path_mtu = IB_MTU_256; + attr.ah_attr.type = RDMA_AH_ATTR_TYPE_ROCE; rdma_ah_set_grh(&attr.ah_attr, NULL, 0, 0, 1, 0); rdma_ah_set_static_rate(&attr.ah_attr, 3); subnet_prefix = cpu_to_be64(0xfe80000000000000LL); for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { + phy_port = (i >= HNS_ROCE_MAX_PORTS) ? (i - 2) : + (i % HNS_ROCE_MAX_PORTS); + sl = i / HNS_ROCE_MAX_PORTS; + + for (j = 0; j < caps->num_ports; j++) { + if (hr_dev->iboe.phy_port[j] == phy_port) { + queue_en[i] = 1; + port = j; + break; + } + } + + if (!queue_en[i]) + continue; + free_mr->mr_free_qp[i] = hns_roce_v1_create_lp_qp(hr_dev, pd); - if (IS_ERR(free_mr->mr_free_qp[i])) { + if (!free_mr->mr_free_qp[i]) { dev_err(dev, "Create loop qp failed!\n"); goto create_lp_qp_failed; } hr_qp = free_mr->mr_free_qp[i]; - sl = i / caps->num_ports; - - if (caps->num_ports == HNS_ROCE_MAX_PORTS) - phy_port = (i >= HNS_ROCE_MAX_PORTS) ? (i - 2) : - (i % caps->num_ports); - else - phy_port = i % caps->num_ports; - - hr_qp->port = phy_port + 1; + hr_qp->port = port; hr_qp->phy_port = phy_port; hr_qp->ibqp.qp_type = IB_QPT_RC; hr_qp->ibqp.device = &hr_dev->ib_dev; @@ -739,23 +749,22 @@ static int hns_roce_v1_rsv_lp_qp(struct hns_roce_dev *hr_dev) hr_qp->ibqp.recv_cq = cq; hr_qp->ibqp.send_cq = cq; - rdma_ah_set_port_num(&attr.ah_attr, phy_port + 1); - rdma_ah_set_sl(&attr.ah_attr, phy_port + 1); - attr.port_num = phy_port + 1; + rdma_ah_set_port_num(&attr.ah_attr, port + 1); + rdma_ah_set_sl(&attr.ah_attr, sl); + attr.port_num = port + 1; attr.dest_qp_num = hr_qp->qpn; memcpy(rdma_ah_retrieve_dmac(&attr.ah_attr), - hr_dev->dev_addr[phy_port], + hr_dev->dev_addr[port], MAC_ADDR_OCTET_NUM); memcpy(&dgid.raw, &subnet_prefix, sizeof(u64)); - memcpy(&dgid.raw[8], hr_dev->dev_addr[phy_port], 3); - memcpy(&dgid.raw[13], hr_dev->dev_addr[phy_port] + 3, 3); + memcpy(&dgid.raw[8], hr_dev->dev_addr[port], 3); + memcpy(&dgid.raw[13], hr_dev->dev_addr[port] + 3, 3); dgid.raw[11] = 0xff; dgid.raw[12] = 0xfe; dgid.raw[8] ^= 2; rdma_ah_set_dgid_raw(&attr.ah_attr, dgid.raw); - attr_mask |= IB_QP_PORT; ret = hr_dev->hw->modify_qp(&hr_qp->ibqp, &attr, attr_mask, IB_QPS_RESET, IB_QPS_INIT); @@ -812,6 +821,9 @@ static void hns_roce_v1_release_lp_qp(struct hns_roce_dev *hr_dev) for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { hr_qp = free_mr->mr_free_qp[i]; + if (!hr_qp) + continue; + ret = hns_roce_v1_destroy_qp(&hr_qp->ibqp); if (ret) dev_err(dev, "Destroy qp %d for mr free failed(%d)!\n", @@ -963,7 +975,7 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) msecs_to_jiffies(HNS_ROCE_V1_FREE_MR_TIMEOUT_MSECS) + jiffies; int i; int ret; - int ne; + int ne = 0; mr_work = container_of(work, struct hns_roce_mr_free_work, work); hr_mr = (struct hns_roce_mr *)mr_work->mr; @@ -976,6 +988,10 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) for (i = 0; i < HNS_ROCE_V1_RESV_QP; i++) { hr_qp = free_mr->mr_free_qp[i]; + if (!hr_qp) + continue; + ne++; + ret = hns_roce_v1_send_lp_wqe(hr_qp); if (ret) { dev_err(dev, @@ -985,7 +1001,6 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) } } - ne = HNS_ROCE_V1_RESV_QP; do { ret = hns_roce_v1_poll_cq(&mr_free_cq->ib_cq, ne, wc); if (ret < 0) { @@ -995,7 +1010,8 @@ static void hns_roce_v1_mr_free_work_fn(struct work_struct *work) goto free_work; } ne -= ret; - msleep(HNS_ROCE_V1_FREE_MR_WAIT_VALUE); + usleep_range(HNS_ROCE_V1_FREE_MR_WAIT_VALUE * 1000, + (1 + HNS_ROCE_V1_FREE_MR_WAIT_VALUE) * 1000); } while (ne && time_before_eq(jiffies, end)); if (ne != 0) @@ -2181,7 +2197,7 @@ static int hns_roce_v1_poll_one(struct hns_roce_cq *hr_cq, } wc->wr_id = wq->wrid[wq->tail & (wq->wqe_cnt - 1)]; ++wq->tail; - } else { + } else { /* RQ conrespond to CQE */ wc->byte_len = le32_to_cpu(cqe->byte_cnt); opcode = roce_get_field(cqe->cqe_byte_4, @@ -3533,10 +3549,12 @@ static int check_qp_db_process_status(struct hns_roce_dev *hr_dev, old_cnt = roce_get_field(old_send, ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S); - if (cur_cnt - old_cnt > SDB_ST_CMP_VAL) + if (cur_cnt - old_cnt > + SDB_ST_CMP_VAL) { success_flags = 1; - else { - send_ptr = roce_get_field(old_send, + } else { + send_ptr = + roce_get_field(old_send, ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_M, ROCEE_SDB_SEND_PTR_SDB_SEND_PTR_S) + roce_get_field(sdb_retry_cnt, @@ -3641,6 +3659,7 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) struct hns_roce_dev *hr_dev; struct hns_roce_qp *hr_qp; struct device *dev; + unsigned long qpn; int ret; qp_work_entry = container_of(work, struct hns_roce_qp_work, work); @@ -3648,8 +3667,9 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) dev = &hr_dev->pdev->dev; priv = (struct hns_roce_v1_priv *)hr_dev->hw->priv; hr_qp = qp_work_entry->qp; + qpn = hr_qp->qpn; - dev_dbg(dev, "Schedule destroy QP(0x%lx) work.\n", hr_qp->qpn); + dev_dbg(dev, "Schedule destroy QP(0x%lx) work.\n", qpn); qp_work_entry->sche_cnt++; @@ -3660,7 +3680,7 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) &qp_work_entry->db_wait_stage); if (ret) { dev_err(dev, "Check QP(0x%lx) db process status failed!\n", - hr_qp->qpn); + qpn); return; } @@ -3674,7 +3694,7 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) ret = hns_roce_v1_modify_qp(&hr_qp->ibqp, NULL, 0, hr_qp->state, IB_QPS_RESET); if (ret) { - dev_err(dev, "Modify QP(0x%lx) to RST failed!\n", hr_qp->qpn); + dev_err(dev, "Modify QP(0x%lx) to RST failed!\n", qpn); return; } @@ -3683,14 +3703,14 @@ static void hns_roce_v1_destroy_qp_work_fn(struct work_struct *work) if (hr_qp->ibqp.qp_type == IB_QPT_RC) { /* RC QP, release QPN */ - hns_roce_release_range_qp(hr_dev, hr_qp->qpn, 1); + hns_roce_release_range_qp(hr_dev, qpn, 1); kfree(hr_qp); } else kfree(hr_to_hr_sqp(hr_qp)); kfree(qp_work_entry); - dev_dbg(dev, "Accomplished destroy QP(0x%lx) work.\n", hr_qp->qpn); + dev_dbg(dev, "Accomplished destroy QP(0x%lx) work.\n", qpn); } int hns_roce_v1_destroy_qp(struct ib_qp *ibqp) diff --git a/drivers/infiniband/hw/hns/hns_roce_main.c b/drivers/infiniband/hw/hns/hns_roce_main.c index c3b41f95e70a..d9777b662eba 100644 --- a/drivers/infiniband/hw/hns/hns_roce_main.c +++ b/drivers/infiniband/hw/hns/hns_roce_main.c @@ -125,8 +125,6 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, return -ENODEV; } - spin_lock_bh(&hr_dev->iboe.lock); - switch (event) { case NETDEV_UP: case NETDEV_CHANGE: @@ -144,7 +142,6 @@ static int handle_en_event(struct hns_roce_dev *hr_dev, u8 port, break; } - spin_unlock_bh(&hr_dev->iboe.lock); return 0; } diff --git a/drivers/infiniband/hw/i40iw/i40iw.h b/drivers/infiniband/hw/i40iw/i40iw.h index da2eb5a281fa..9b1566468744 100644 --- a/drivers/infiniband/hw/i40iw/i40iw.h +++ b/drivers/infiniband/hw/i40iw/i40iw.h @@ -527,6 +527,7 @@ enum i40iw_status_code i40iw_add_mac_addr(struct i40iw_device *iwdev, int i40iw_modify_qp(struct ib_qp *, struct ib_qp_attr *, int, struct ib_udata *); void i40iw_cq_wq_destroy(struct i40iw_device *iwdev, struct i40iw_sc_cq *cq); +void i40iw_cleanup_pending_cqp_op(struct i40iw_device *iwdev); void i40iw_rem_pdusecount(struct i40iw_pd *iwpd, struct i40iw_device *iwdev); void i40iw_add_pdusecount(struct i40iw_pd *iwpd); void i40iw_rem_devusecount(struct i40iw_device *iwdev); diff --git a/drivers/infiniband/hw/i40iw/i40iw_cm.c b/drivers/infiniband/hw/i40iw/i40iw_cm.c index 6ae98aa7f74e..5a2fa743676c 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_cm.c +++ b/drivers/infiniband/hw/i40iw/i40iw_cm.c @@ -3487,7 +3487,8 @@ static void i40iw_cm_disconn_true(struct i40iw_qp *iwqp) if (((original_hw_tcp_state == I40IW_TCP_STATE_CLOSED) || (original_hw_tcp_state == I40IW_TCP_STATE_TIME_WAIT) || (last_ae == I40IW_AE_RDMAP_ROE_BAD_LLP_CLOSE) || - (last_ae == I40IW_AE_LLP_CONNECTION_RESET))) { + (last_ae == I40IW_AE_LLP_CONNECTION_RESET) || + iwdev->reset)) { issue_close = 1; iwqp->cm_id = NULL; if (!iwqp->flush_issued) { @@ -4265,6 +4266,8 @@ void i40iw_cm_disconnect_all(struct i40iw_device *iwdev) cm_node = container_of(list_node, struct i40iw_cm_node, connected_entry); attr.qp_state = IB_QPS_ERR; i40iw_modify_qp(&cm_node->iwqp->ibqp, &attr, IB_QP_STATE, NULL); + if (iwdev->reset) + i40iw_cm_disconn(cm_node->iwqp); i40iw_rem_ref_cm_node(cm_node); } } diff --git a/drivers/infiniband/hw/i40iw/i40iw_ctrl.c b/drivers/infiniband/hw/i40iw/i40iw_ctrl.c index a027e2072477..9ec1ae9a82c9 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_ctrl.c +++ b/drivers/infiniband/hw/i40iw/i40iw_ctrl.c @@ -1970,6 +1970,8 @@ static enum i40iw_status_code i40iw_sc_ccq_destroy(struct i40iw_sc_cq *ccq, ret_code = i40iw_cqp_poll_registers(cqp, tail, 1000); } + cqp->process_cqp_sds = i40iw_update_sds_noccq; + return ret_code; } diff --git a/drivers/infiniband/hw/i40iw/i40iw_main.c b/drivers/infiniband/hw/i40iw/i40iw_main.c index e0f47cc2effc..ae8463ff59a7 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_main.c +++ b/drivers/infiniband/hw/i40iw/i40iw_main.c @@ -243,6 +243,8 @@ static void i40iw_destroy_cqp(struct i40iw_device *iwdev, bool free_hwcqp) if (free_hwcqp) dev->cqp_ops->cqp_destroy(dev->cqp); + i40iw_cleanup_pending_cqp_op(iwdev); + i40iw_free_dma_mem(dev->hw, &cqp->sq); kfree(cqp->scratch_array); iwdev->cqp.scratch_array = NULL; @@ -274,13 +276,12 @@ static void i40iw_disable_irq(struct i40iw_sc_dev *dev, /** * i40iw_destroy_aeq - destroy aeq * @iwdev: iwarp device - * @reset: true if called before reset * * Issue a destroy aeq request and * free the resources associated with the aeq * The function is called during driver unload */ -static void i40iw_destroy_aeq(struct i40iw_device *iwdev, bool reset) +static void i40iw_destroy_aeq(struct i40iw_device *iwdev) { enum i40iw_status_code status = I40IW_ERR_NOT_READY; struct i40iw_sc_dev *dev = &iwdev->sc_dev; @@ -288,7 +289,7 @@ static void i40iw_destroy_aeq(struct i40iw_device *iwdev, bool reset) if (!iwdev->msix_shared) i40iw_disable_irq(dev, iwdev->iw_msixtbl, (void *)iwdev); - if (reset) + if (iwdev->reset) goto exit; if (!dev->aeq_ops->aeq_destroy(&aeq->sc_aeq, 0, 1)) @@ -304,19 +305,17 @@ exit: * i40iw_destroy_ceq - destroy ceq * @iwdev: iwarp device * @iwceq: ceq to be destroyed - * @reset: true if called before reset * * Issue a destroy ceq request and * free the resources associated with the ceq */ static void i40iw_destroy_ceq(struct i40iw_device *iwdev, - struct i40iw_ceq *iwceq, - bool reset) + struct i40iw_ceq *iwceq) { enum i40iw_status_code status; struct i40iw_sc_dev *dev = &iwdev->sc_dev; - if (reset) + if (iwdev->reset) goto exit; status = dev->ceq_ops->ceq_destroy(&iwceq->sc_ceq, 0, 1); @@ -335,12 +334,11 @@ exit: /** * i40iw_dele_ceqs - destroy all ceq's * @iwdev: iwarp device - * @reset: true if called before reset * * Go through all of the device ceq's and for each ceq * disable the ceq interrupt and destroy the ceq */ -static void i40iw_dele_ceqs(struct i40iw_device *iwdev, bool reset) +static void i40iw_dele_ceqs(struct i40iw_device *iwdev) { u32 i = 0; struct i40iw_sc_dev *dev = &iwdev->sc_dev; @@ -349,32 +347,31 @@ static void i40iw_dele_ceqs(struct i40iw_device *iwdev, bool reset) if (iwdev->msix_shared) { i40iw_disable_irq(dev, msix_vec, (void *)iwdev); - i40iw_destroy_ceq(iwdev, iwceq, reset); + i40iw_destroy_ceq(iwdev, iwceq); iwceq++; i++; } for (msix_vec++; i < iwdev->ceqs_count; i++, msix_vec++, iwceq++) { i40iw_disable_irq(dev, msix_vec, (void *)iwceq); - i40iw_destroy_ceq(iwdev, iwceq, reset); + i40iw_destroy_ceq(iwdev, iwceq); } } /** * i40iw_destroy_ccq - destroy control cq * @iwdev: iwarp device - * @reset: true if called before reset * * Issue destroy ccq request and * free the resources associated with the ccq */ -static void i40iw_destroy_ccq(struct i40iw_device *iwdev, bool reset) +static void i40iw_destroy_ccq(struct i40iw_device *iwdev) { struct i40iw_sc_dev *dev = &iwdev->sc_dev; struct i40iw_ccq *ccq = &iwdev->ccq; enum i40iw_status_code status = 0; - if (!reset) + if (!iwdev->reset) status = dev->ccq_ops->ccq_destroy(dev->ccq, 0, true); if (status) i40iw_pr_err("ccq destroy failed %d\n", status); @@ -810,7 +807,7 @@ static enum i40iw_status_code i40iw_setup_ceqs(struct i40iw_device *iwdev, iwceq->msix_idx = msix_vec->idx; status = i40iw_configure_ceq_vector(iwdev, iwceq, ceq_id, msix_vec); if (status) { - i40iw_destroy_ceq(iwdev, iwceq, false); + i40iw_destroy_ceq(iwdev, iwceq); break; } i40iw_enable_intr(&iwdev->sc_dev, msix_vec->idx); @@ -912,7 +909,7 @@ static enum i40iw_status_code i40iw_setup_aeq(struct i40iw_device *iwdev) status = i40iw_configure_aeq_vector(iwdev); if (status) { - i40iw_destroy_aeq(iwdev, false); + i40iw_destroy_aeq(iwdev); return status; } @@ -1442,12 +1439,11 @@ static enum i40iw_status_code i40iw_save_msix_info(struct i40iw_device *iwdev, /** * i40iw_deinit_device - clean up the device resources * @iwdev: iwarp device - * @reset: true if called before reset * * Destroy the ib device interface, remove the mac ip entry and ipv4/ipv6 addresses, * destroy the device queues and free the pble and the hmc objects */ -static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset) +static void i40iw_deinit_device(struct i40iw_device *iwdev) { struct i40e_info *ldev = iwdev->ldev; @@ -1464,7 +1460,7 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset) i40iw_destroy_rdma_device(iwdev->iwibdev); /* fallthrough */ case IP_ADDR_REGISTERED: - if (!reset) + if (!iwdev->reset) i40iw_del_macip_entry(iwdev, (u8)iwdev->mac_ip_table_idx); /* fallthrough */ case INET_NOTIFIER: @@ -1474,26 +1470,26 @@ static void i40iw_deinit_device(struct i40iw_device *iwdev, bool reset) unregister_inet6addr_notifier(&i40iw_inetaddr6_notifier); } /* fallthrough */ + case PBLE_CHUNK_MEM: + i40iw_destroy_pble_pool(dev, iwdev->pble_rsrc); + /* fallthrough */ case CEQ_CREATED: - i40iw_dele_ceqs(iwdev, reset); + i40iw_dele_ceqs(iwdev); /* fallthrough */ case AEQ_CREATED: - i40iw_destroy_aeq(iwdev, reset); + i40iw_destroy_aeq(iwdev); /* fallthrough */ case IEQ_CREATED: - i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_IEQ, reset); + i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_IEQ, iwdev->reset); /* fallthrough */ case ILQ_CREATED: - i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_ILQ, reset); + i40iw_puda_dele_resources(&iwdev->vsi, I40IW_PUDA_RSRC_TYPE_ILQ, iwdev->reset); /* fallthrough */ case CCQ_CREATED: - i40iw_destroy_ccq(iwdev, reset); - /* fallthrough */ - case PBLE_CHUNK_MEM: - i40iw_destroy_pble_pool(dev, iwdev->pble_rsrc); + i40iw_destroy_ccq(iwdev); /* fallthrough */ case HMC_OBJS_CREATED: - i40iw_del_hmc_objects(dev, dev->hmc_info, true, reset); + i40iw_del_hmc_objects(dev, dev->hmc_info, true, iwdev->reset); /* fallthrough */ case CQP_CREATED: i40iw_destroy_cqp(iwdev, true); @@ -1670,6 +1666,7 @@ static int i40iw_open(struct i40e_info *ldev, struct i40e_client *client) status = i40iw_hmc_init_pble(&iwdev->sc_dev, iwdev->pble_rsrc); if (status) break; + iwdev->init_state = PBLE_CHUNK_MEM; iwdev->virtchnl_wq = alloc_ordered_workqueue("iwvch", WQ_MEM_RECLAIM); i40iw_register_notifiers(); iwdev->init_state = INET_NOTIFIER; @@ -1693,7 +1690,7 @@ static int i40iw_open(struct i40e_info *ldev, struct i40e_client *client) } while (0); i40iw_pr_err("status = %d last completion = %d\n", status, iwdev->init_state); - i40iw_deinit_device(iwdev, false); + i40iw_deinit_device(iwdev); return -ERESTART; } @@ -1774,9 +1771,12 @@ static void i40iw_close(struct i40e_info *ldev, struct i40e_client *client, bool iwdev = &hdl->device; iwdev->closing = true; + if (reset) + iwdev->reset = true; + i40iw_cm_disconnect_all(iwdev); destroy_workqueue(iwdev->virtchnl_wq); - i40iw_deinit_device(iwdev, reset); + i40iw_deinit_device(iwdev); } /** diff --git a/drivers/infiniband/hw/i40iw/i40iw_puda.c b/drivers/infiniband/hw/i40iw/i40iw_puda.c index db41ab40da9c..71050c5d29a0 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_puda.c +++ b/drivers/infiniband/hw/i40iw/i40iw_puda.c @@ -408,6 +408,9 @@ enum i40iw_status_code i40iw_puda_send(struct i40iw_sc_qp *qp, set_64bit_val(wqe, 0, info->paddr); set_64bit_val(wqe, 8, LS_64(info->len, I40IWQPSQ_FRAG_LEN)); set_64bit_val(wqe, 16, header[0]); + + /* Ensure all data is written before writing valid bit */ + wmb(); set_64bit_val(wqe, 24, header[1]); i40iw_debug_buf(qp->dev, I40IW_DEBUG_PUDA, "PUDA SEND WQE", wqe, 32); @@ -1411,10 +1414,10 @@ static void i40iw_ieq_handle_exception(struct i40iw_puda_rsrc *ieq, if (!list_empty(rxlist)) { tmpbuf = (struct i40iw_puda_buf *)rxlist->next; - plist = &tmpbuf->list; while ((struct list_head *)tmpbuf != rxlist) { if ((int)(buf->seqnum - tmpbuf->seqnum) < 0) break; + plist = &tmpbuf->list; tmpbuf = (struct i40iw_puda_buf *)plist->next; } /* Insert buf before tmpbuf */ diff --git a/drivers/infiniband/hw/i40iw/i40iw_utils.c b/drivers/infiniband/hw/i40iw/i40iw_utils.c index 56d986924a4c..e311ec559f4e 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_utils.c +++ b/drivers/infiniband/hw/i40iw/i40iw_utils.c @@ -337,6 +337,7 @@ struct i40iw_cqp_request *i40iw_get_cqp_request(struct i40iw_cqp *cqp, bool wait */ void i40iw_free_cqp_request(struct i40iw_cqp *cqp, struct i40iw_cqp_request *cqp_request) { + struct i40iw_device *iwdev = container_of(cqp, struct i40iw_device, cqp); unsigned long flags; if (cqp_request->dynamic) { @@ -350,6 +351,7 @@ void i40iw_free_cqp_request(struct i40iw_cqp *cqp, struct i40iw_cqp_request *cqp list_add_tail(&cqp_request->list, &cqp->cqp_avail_reqs); spin_unlock_irqrestore(&cqp->req_lock, flags); } + wake_up(&iwdev->close_wq); } /** @@ -365,6 +367,56 @@ void i40iw_put_cqp_request(struct i40iw_cqp *cqp, } /** + * i40iw_free_pending_cqp_request -free pending cqp request objs + * @cqp: cqp ptr + * @cqp_request: to be put back in cqp list + */ +static void i40iw_free_pending_cqp_request(struct i40iw_cqp *cqp, + struct i40iw_cqp_request *cqp_request) +{ + struct i40iw_device *iwdev = container_of(cqp, struct i40iw_device, cqp); + + if (cqp_request->waiting) { + cqp_request->compl_info.error = true; + cqp_request->request_done = true; + wake_up(&cqp_request->waitq); + } + i40iw_put_cqp_request(cqp, cqp_request); + wait_event_timeout(iwdev->close_wq, + !atomic_read(&cqp_request->refcount), + 1000); +} + +/** + * i40iw_cleanup_pending_cqp_op - clean-up cqp with no completions + * @iwdev: iwarp device + */ +void i40iw_cleanup_pending_cqp_op(struct i40iw_device *iwdev) +{ + struct i40iw_sc_dev *dev = &iwdev->sc_dev; + struct i40iw_cqp *cqp = &iwdev->cqp; + struct i40iw_cqp_request *cqp_request = NULL; + struct cqp_commands_info *pcmdinfo = NULL; + u32 i, pending_work, wqe_idx; + + pending_work = I40IW_RING_WORK_AVAILABLE(cqp->sc_cqp.sq_ring); + wqe_idx = I40IW_RING_GETCURRENT_TAIL(cqp->sc_cqp.sq_ring); + for (i = 0; i < pending_work; i++) { + cqp_request = (struct i40iw_cqp_request *)(unsigned long)cqp->scratch_array[wqe_idx]; + if (cqp_request) + i40iw_free_pending_cqp_request(cqp, cqp_request); + wqe_idx = (wqe_idx + 1) % I40IW_RING_GETSIZE(cqp->sc_cqp.sq_ring); + } + + while (!list_empty(&dev->cqp_cmd_head)) { + pcmdinfo = (struct cqp_commands_info *)i40iw_remove_head(&dev->cqp_cmd_head); + cqp_request = container_of(pcmdinfo, struct i40iw_cqp_request, info); + if (cqp_request) + i40iw_free_pending_cqp_request(cqp, cqp_request); + } +} + +/** * i40iw_free_qp - callback after destroy cqp completes * @cqp_request: cqp request for destroy qp * @num: not used @@ -546,8 +598,12 @@ void i40iw_rem_ref(struct ib_qp *ibqp) cqp_info->in.u.qp_destroy.scratch = (uintptr_t)cqp_request; cqp_info->in.u.qp_destroy.remove_hash_idx = true; status = i40iw_handle_cqp_op(iwdev, cqp_request); - if (status) - i40iw_pr_err("CQP-OP Destroy QP fail"); + if (!status) + return; + + i40iw_rem_pdusecount(iwqp->iwpd, iwdev); + i40iw_free_qp_resources(iwdev, iwqp, qp_num); + i40iw_rem_devusecount(iwdev); } /** diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.c b/drivers/infiniband/hw/i40iw/i40iw_verbs.c index 4dbe61ec7a77..02d871db7ca5 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.c +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.c @@ -426,9 +426,13 @@ void i40iw_free_qp_resources(struct i40iw_device *iwdev, struct i40iw_qp *iwqp, u32 qp_num) { + struct i40iw_pbl *iwpbl = &iwqp->iwpbl; + i40iw_dealloc_push_page(iwdev, &iwqp->sc_qp); if (qp_num) i40iw_free_resource(iwdev, iwdev->allocated_qps, qp_num); + if (iwpbl->pbl_allocated) + i40iw_free_pble(iwdev->pble_rsrc, &iwpbl->pble_alloc); i40iw_free_dma_mem(iwdev->sc_dev.hw, &iwqp->q2_ctx_mem); i40iw_free_dma_mem(iwdev->sc_dev.hw, &iwqp->kqp.dma_mem); kfree(iwqp->kqp.wrid_mem); @@ -483,7 +487,7 @@ static int i40iw_setup_virt_qp(struct i40iw_device *iwdev, struct i40iw_qp *iwqp, struct i40iw_qp_init_info *init_info) { - struct i40iw_pbl *iwpbl = iwqp->iwpbl; + struct i40iw_pbl *iwpbl = &iwqp->iwpbl; struct i40iw_qp_mr *qpmr = &iwpbl->qp_mr; iwqp->page = qpmr->sq_page; @@ -688,19 +692,22 @@ static struct ib_qp *i40iw_create_qp(struct ib_pd *ibpd, ucontext = to_ucontext(ibpd->uobject->context); if (req.user_wqe_buffers) { + struct i40iw_pbl *iwpbl; + spin_lock_irqsave( &ucontext->qp_reg_mem_list_lock, flags); - iwqp->iwpbl = i40iw_get_pbl( + iwpbl = i40iw_get_pbl( (unsigned long)req.user_wqe_buffers, &ucontext->qp_reg_mem_list); spin_unlock_irqrestore( &ucontext->qp_reg_mem_list_lock, flags); - if (!iwqp->iwpbl) { + if (!iwpbl) { err_code = -ENODATA; i40iw_pr_err("no pbl info\n"); goto error; } + memcpy(&iwqp->iwpbl, iwpbl, sizeof(iwqp->iwpbl)); } } err_code = i40iw_setup_virt_qp(iwdev, iwqp, &init_info); @@ -1161,8 +1168,10 @@ static struct ib_cq *i40iw_create_cq(struct ib_device *ibdev, memset(&req, 0, sizeof(req)); iwcq->user_mode = true; ucontext = to_ucontext(context); - if (ib_copy_from_udata(&req, udata, sizeof(struct i40iw_create_cq_req))) + if (ib_copy_from_udata(&req, udata, sizeof(struct i40iw_create_cq_req))) { + err_code = -EFAULT; goto cq_free_resources; + } spin_lock_irqsave(&ucontext->cq_reg_mem_list_lock, flags); iwpbl = i40iw_get_pbl((unsigned long)req.user_cq_buffer, @@ -2063,7 +2072,7 @@ static int i40iw_dereg_mr(struct ib_mr *ib_mr) ucontext = to_ucontext(ibpd->uobject->context); i40iw_del_memlist(iwmr, ucontext); } - if (iwpbl->pbl_allocated) + if (iwpbl->pbl_allocated && iwmr->type != IW_MEMREG_TYPE_QP) i40iw_free_pble(iwdev->pble_rsrc, palloc); kfree(iwmr); return 0; diff --git a/drivers/infiniband/hw/i40iw/i40iw_verbs.h b/drivers/infiniband/hw/i40iw/i40iw_verbs.h index 07c3fec77de6..9067443cd311 100644 --- a/drivers/infiniband/hw/i40iw/i40iw_verbs.h +++ b/drivers/infiniband/hw/i40iw/i40iw_verbs.h @@ -170,7 +170,7 @@ struct i40iw_qp { struct i40iw_qp_kmode kqp; struct i40iw_dma_mem host_ctx; struct timer_list terminate_timer; - struct i40iw_pbl *iwpbl; + struct i40iw_pbl iwpbl; struct i40iw_dma_mem q2_ctx_mem; struct i40iw_dma_mem ietf_mem; struct completion sq_drained; diff --git a/drivers/infiniband/hw/mlx4/cm.c b/drivers/infiniband/hw/mlx4/cm.c index 1e6c526450d9..fedaf8260105 100644 --- a/drivers/infiniband/hw/mlx4/cm.c +++ b/drivers/infiniband/hw/mlx4/cm.c @@ -323,6 +323,9 @@ int mlx4_ib_multiplex_cm_handler(struct ib_device *ibdev, int port, int slave_id mad->mad_hdr.attr_id == CM_REP_ATTR_ID || mad->mad_hdr.attr_id == CM_SIDR_REQ_ATTR_ID) { sl_cm_id = get_local_comm_id(mad); + id = id_map_get(ibdev, &pv_cm_id, slave_id, sl_cm_id); + if (id) + goto cont; id = id_map_alloc(ibdev, slave_id, sl_cm_id); if (IS_ERR(id)) { mlx4_ib_warn(ibdev, "%s: id{slave: %d, sl_cm_id: 0x%x} Failed to id_map_alloc\n", @@ -343,6 +346,7 @@ int mlx4_ib_multiplex_cm_handler(struct ib_device *ibdev, int port, int slave_id return -EINVAL; } +cont: set_local_comm_id(mad, id->pv_cm_id); if (mad->mad_hdr.attr_id == CM_DREQ_ATTR_ID) diff --git a/drivers/infiniband/hw/mlx4/cq.c b/drivers/infiniband/hw/mlx4/cq.c index 4f5a143fc0a7..ff931c580557 100644 --- a/drivers/infiniband/hw/mlx4/cq.c +++ b/drivers/infiniband/hw/mlx4/cq.c @@ -102,7 +102,7 @@ static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf * int err; err = mlx4_buf_alloc(dev->dev, nent * dev->dev->caps.cqe_size, - PAGE_SIZE * 2, &buf->buf, GFP_KERNEL); + PAGE_SIZE * 2, &buf->buf); if (err) goto out; @@ -113,7 +113,7 @@ static int mlx4_ib_alloc_cq_buf(struct mlx4_ib_dev *dev, struct mlx4_ib_cq_buf * if (err) goto err_buf; - err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf, GFP_KERNEL); + err = mlx4_buf_write_mtt(dev->dev, &buf->mtt, &buf->buf); if (err) goto err_mtt; @@ -219,7 +219,7 @@ struct ib_cq *mlx4_ib_create_cq(struct ib_device *ibdev, uar = &to_mucontext(context)->uar; } else { - err = mlx4_db_alloc(dev->dev, &cq->db, 1, GFP_KERNEL); + err = mlx4_db_alloc(dev->dev, &cq->db, 1); if (err) goto err_cq; diff --git a/drivers/infiniband/hw/mlx4/main.c b/drivers/infiniband/hw/mlx4/main.c index 75b2f7d4cd95..d1b43cbbfea7 100644 --- a/drivers/infiniband/hw/mlx4/main.c +++ b/drivers/infiniband/hw/mlx4/main.c @@ -1155,7 +1155,7 @@ static void mlx4_ib_disassociate_ucontext(struct ib_ucontext *ibcontext) * call to mlx4_ib_vma_close. */ put_task_struct(owning_process); - msleep(1); + usleep_range(1000, 2000); owning_process = get_pid_task(ibcontext->tgid, PIDTYPE_PID); if (!owning_process || diff --git a/drivers/infiniband/hw/mlx4/mcg.c b/drivers/infiniband/hw/mlx4/mcg.c index 3405e947dc1e..b73f89700ef9 100644 --- a/drivers/infiniband/hw/mlx4/mcg.c +++ b/drivers/infiniband/hw/mlx4/mcg.c @@ -1091,7 +1091,7 @@ static void _mlx4_ib_mcg_port_cleanup(struct mlx4_ib_demux_ctx *ctx, int destroy if (!count) break; - msleep(1); + usleep_range(1000, 2000); } while (time_after(end, jiffies)); flush_workqueue(ctx->mcg_wq); diff --git a/drivers/infiniband/hw/mlx4/mlx4_ib.h b/drivers/infiniband/hw/mlx4/mlx4_ib.h index c2b9cbf4da05..9db82e67e959 100644 --- a/drivers/infiniband/hw/mlx4/mlx4_ib.h +++ b/drivers/infiniband/hw/mlx4/mlx4_ib.h @@ -185,7 +185,6 @@ enum mlx4_ib_qp_flags { MLX4_IB_QP_LSO = IB_QP_CREATE_IPOIB_UD_LSO, MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK = IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK, MLX4_IB_QP_NETIF = IB_QP_CREATE_NETIF_QP, - MLX4_IB_QP_CREATE_USE_GFP_NOIO = IB_QP_CREATE_USE_GFP_NOIO, /* Mellanox specific flags start from IB_QP_CREATE_RESERVED_START */ MLX4_IB_ROCE_V2_GSI_QP = MLX4_IB_QP_CREATE_ROCE_V2_GSI, diff --git a/drivers/infiniband/hw/mlx4/qp.c b/drivers/infiniband/hw/mlx4/qp.c index 996e9058e515..75c0e6c5dd56 100644 --- a/drivers/infiniband/hw/mlx4/qp.c +++ b/drivers/infiniband/hw/mlx4/qp.c @@ -634,8 +634,8 @@ static void mlx4_ib_free_qp_counter(struct mlx4_ib_dev *dev, static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, struct ib_qp_init_attr *init_attr, - struct ib_udata *udata, int sqpn, struct mlx4_ib_qp **caller_qp, - gfp_t gfp) + struct ib_udata *udata, int sqpn, + struct mlx4_ib_qp **caller_qp) { int qpn; int err; @@ -691,14 +691,14 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (qp_type == MLX4_IB_QPT_SMI || qp_type == MLX4_IB_QPT_GSI || (qp_type & (MLX4_IB_QPT_PROXY_SMI | MLX4_IB_QPT_PROXY_SMI_OWNER | MLX4_IB_QPT_PROXY_GSI | MLX4_IB_QPT_TUN_SMI_OWNER))) { - sqp = kzalloc(sizeof (struct mlx4_ib_sqp), gfp); + sqp = kzalloc(sizeof(struct mlx4_ib_sqp), GFP_KERNEL); if (!sqp) return -ENOMEM; qp = &sqp->qp; qp->pri.vid = 0xFFFF; qp->alt.vid = 0xFFFF; } else { - qp = kzalloc(sizeof (struct mlx4_ib_qp), gfp); + qp = kzalloc(sizeof(struct mlx4_ib_qp), GFP_KERNEL); if (!qp) return -ENOMEM; qp->pri.vid = 0xFFFF; @@ -780,7 +780,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, goto err; if (qp_has_rq(init_attr)) { - err = mlx4_db_alloc(dev->dev, &qp->db, 0, gfp); + err = mlx4_db_alloc(dev->dev, &qp->db, 0); if (err) goto err; @@ -788,7 +788,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, } if (mlx4_buf_alloc(dev->dev, qp->buf_size, qp->buf_size, - &qp->buf, gfp)) { + &qp->buf)) { memcpy(&init_attr->cap, &backup_cap, sizeof(backup_cap)); err = set_kernel_sq_size(dev, &init_attr->cap, qp_type, @@ -797,7 +797,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, goto err_db; if (mlx4_buf_alloc(dev->dev, qp->buf_size, - PAGE_SIZE * 2, &qp->buf, gfp)) { + PAGE_SIZE * 2, &qp->buf)) { err = -ENOMEM; goto err_db; } @@ -808,20 +808,20 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (err) goto err_buf; - err = mlx4_buf_write_mtt(dev->dev, &qp->mtt, &qp->buf, gfp); + err = mlx4_buf_write_mtt(dev->dev, &qp->mtt, &qp->buf); if (err) goto err_mtt; qp->sq.wrid = kmalloc_array(qp->sq.wqe_cnt, sizeof(u64), - gfp | __GFP_NOWARN); + GFP_KERNEL | __GFP_NOWARN); if (!qp->sq.wrid) qp->sq.wrid = __vmalloc(qp->sq.wqe_cnt * sizeof(u64), - gfp, PAGE_KERNEL); + GFP_KERNEL, PAGE_KERNEL); qp->rq.wrid = kmalloc_array(qp->rq.wqe_cnt, sizeof(u64), - gfp | __GFP_NOWARN); + GFP_KERNEL | __GFP_NOWARN); if (!qp->rq.wrid) qp->rq.wrid = __vmalloc(qp->rq.wqe_cnt * sizeof(u64), - gfp, PAGE_KERNEL); + GFP_KERNEL, PAGE_KERNEL); if (!qp->sq.wrid || !qp->rq.wrid) { err = -ENOMEM; goto err_wrid; @@ -859,7 +859,7 @@ static int create_qp_common(struct mlx4_ib_dev *dev, struct ib_pd *pd, if (init_attr->create_flags & IB_QP_CREATE_BLOCK_MULTICAST_LOOPBACK) qp->flags |= MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK; - err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp, gfp); + err = mlx4_qp_alloc(dev->dev, qpn, &qp->mqp); if (err) goto err_qpn; @@ -1127,10 +1127,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, int err; int sup_u_create_flags = MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK; u16 xrcdn = 0; - gfp_t gfp; - gfp = (init_attr->create_flags & MLX4_IB_QP_CREATE_USE_GFP_NOIO) ? - GFP_NOIO : GFP_KERNEL; /* * We only support LSO, vendor flag1, and multicast loopback blocking, * and only for kernel UD QPs. @@ -1140,8 +1137,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, MLX4_IB_SRIOV_TUNNEL_QP | MLX4_IB_SRIOV_SQP | MLX4_IB_QP_NETIF | - MLX4_IB_QP_CREATE_ROCE_V2_GSI | - MLX4_IB_QP_CREATE_USE_GFP_NOIO)) + MLX4_IB_QP_CREATE_ROCE_V2_GSI)) return ERR_PTR(-EINVAL); if (init_attr->create_flags & IB_QP_CREATE_NETIF_QP) { @@ -1154,7 +1150,6 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, return ERR_PTR(-EINVAL); if ((init_attr->create_flags & ~(MLX4_IB_SRIOV_SQP | - MLX4_IB_QP_CREATE_USE_GFP_NOIO | MLX4_IB_QP_CREATE_ROCE_V2_GSI | MLX4_IB_QP_BLOCK_MULTICAST_LOOPBACK) && init_attr->qp_type != IB_QPT_UD) || @@ -1179,7 +1174,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, case IB_QPT_RC: case IB_QPT_UC: case IB_QPT_RAW_PACKET: - qp = kzalloc(sizeof *qp, gfp); + qp = kzalloc(sizeof(*qp), GFP_KERNEL); if (!qp) return ERR_PTR(-ENOMEM); qp->pri.vid = 0xFFFF; @@ -1188,7 +1183,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, case IB_QPT_UD: { err = create_qp_common(to_mdev(pd->device), pd, init_attr, - udata, 0, &qp, gfp); + udata, 0, &qp); if (err) { kfree(qp); return ERR_PTR(err); @@ -1217,8 +1212,7 @@ static struct ib_qp *_mlx4_ib_create_qp(struct ib_pd *pd, } err = create_qp_common(to_mdev(pd->device), pd, init_attr, udata, - sqpn, - &qp, gfp); + sqpn, &qp); if (err) return ERR_PTR(err); diff --git a/drivers/infiniband/hw/mlx4/srq.c b/drivers/infiniband/hw/mlx4/srq.c index e32dd58937a8..0facaf5f6d23 100644 --- a/drivers/infiniband/hw/mlx4/srq.c +++ b/drivers/infiniband/hw/mlx4/srq.c @@ -135,14 +135,14 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, if (err) goto err_mtt; } else { - err = mlx4_db_alloc(dev->dev, &srq->db, 0, GFP_KERNEL); + err = mlx4_db_alloc(dev->dev, &srq->db, 0); if (err) goto err_srq; *srq->db.db = 0; - if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, &srq->buf, - GFP_KERNEL)) { + if (mlx4_buf_alloc(dev->dev, buf_size, PAGE_SIZE * 2, + &srq->buf)) { err = -ENOMEM; goto err_db; } @@ -167,7 +167,7 @@ struct ib_srq *mlx4_ib_create_srq(struct ib_pd *pd, if (err) goto err_buf; - err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf, GFP_KERNEL); + err = mlx4_buf_write_mtt(dev->dev, &srq->mtt, &srq->buf); if (err) goto err_mtt; diff --git a/drivers/infiniband/hw/mlx5/mr.c b/drivers/infiniband/hw/mlx5/mr.c index 763bb5b36144..2c40a2e989d2 100644 --- a/drivers/infiniband/hw/mlx5/mr.c +++ b/drivers/infiniband/hw/mlx5/mr.c @@ -582,6 +582,15 @@ static void clean_keys(struct mlx5_ib_dev *dev, int c) } } +static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev) +{ + if (!mlx5_debugfs_root) + return; + + debugfs_remove_recursive(dev->cache.root); + dev->cache.root = NULL; +} + static int mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev) { struct mlx5_mr_cache *cache = &dev->cache; @@ -600,38 +609,34 @@ static int mlx5_mr_cache_debugfs_init(struct mlx5_ib_dev *dev) sprintf(ent->name, "%d", ent->order); ent->dir = debugfs_create_dir(ent->name, cache->root); if (!ent->dir) - return -ENOMEM; + goto err; ent->fsize = debugfs_create_file("size", 0600, ent->dir, ent, &size_fops); if (!ent->fsize) - return -ENOMEM; + goto err; ent->flimit = debugfs_create_file("limit", 0600, ent->dir, ent, &limit_fops); if (!ent->flimit) - return -ENOMEM; + goto err; ent->fcur = debugfs_create_u32("cur", 0400, ent->dir, &ent->cur); if (!ent->fcur) - return -ENOMEM; + goto err; ent->fmiss = debugfs_create_u32("miss", 0600, ent->dir, &ent->miss); if (!ent->fmiss) - return -ENOMEM; + goto err; } return 0; -} - -static void mlx5_mr_cache_debugfs_cleanup(struct mlx5_ib_dev *dev) -{ - if (!mlx5_debugfs_root) - return; +err: + mlx5_mr_cache_debugfs_cleanup(dev); - debugfs_remove_recursive(dev->cache.root); + return -ENOMEM; } static void delay_time_func(unsigned long ctx) @@ -692,6 +697,11 @@ int mlx5_mr_cache_init(struct mlx5_ib_dev *dev) if (err) mlx5_ib_warn(dev, "cache debugfs failure\n"); + /* + * We don't want to fail driver if debugfs failed to initialize, + * so we are not forwarding error to the user. + */ + return 0; } @@ -825,7 +835,7 @@ static int mr_umem_get(struct ib_pd *pd, u64 start, u64 length, access_flags, 0); err = PTR_ERR_OR_ZERO(*umem); if (err < 0) { - mlx5_ib_err(dev, "umem get failed (%ld)\n", PTR_ERR(umem)); + mlx5_ib_err(dev, "umem get failed (%d)\n", err); return err; } @@ -1779,7 +1789,7 @@ mlx5_ib_sg_to_klms(struct mlx5_ib_mr *mr, mr->ndescs = sg_nents; for_each_sg(sgl, sg, sg_nents, i) { - if (unlikely(i > mr->max_descs)) + if (unlikely(i >= mr->max_descs)) break; klms[i].va = cpu_to_be64(sg_dma_address(sg) + sg_offset); klms[i].bcount = cpu_to_be32(sg_dma_len(sg) - sg_offset); diff --git a/drivers/infiniband/hw/mlx5/odp.c b/drivers/infiniband/hw/mlx5/odp.c index ae0746754008..3d701c7a4c91 100644 --- a/drivers/infiniband/hw/mlx5/odp.c +++ b/drivers/infiniband/hw/mlx5/odp.c @@ -939,7 +939,7 @@ static int mlx5_ib_mr_initiator_pfault_handler( if (qp->ibqp.qp_type != IB_QPT_RC) { av = *wqe; - if (av->dqp_dct & be32_to_cpu(MLX5_WQE_AV_EXT)) + if (av->dqp_dct & cpu_to_be32(MLX5_EXTENDED_UD_AV)) *wqe += sizeof(struct mlx5_av); else *wqe += sizeof(struct mlx5_base_av); diff --git a/drivers/infiniband/hw/nes/nes_hw.c b/drivers/infiniband/hw/nes/nes_hw.c index 8f9d8b4ad583..b0adf65e4bdb 100644 --- a/drivers/infiniband/hw/nes/nes_hw.c +++ b/drivers/infiniband/hw/nes/nes_hw.c @@ -551,7 +551,7 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { if ((0x0F000100 == (pcs_control_status0 & 0x0F000100)) || (0x0F000100 == (pcs_control_status1 & 0x0F000100))) int_cnt++; - msleep(1); + usleep_range(1000, 2000); } if (int_cnt > 1) { spin_lock_irqsave(&nesadapter->phy_lock, flags); @@ -592,7 +592,7 @@ struct nes_adapter *nes_init_adapter(struct nes_device *nesdev, u8 hw_rev) { break; } } - msleep(1); + usleep_range(1000, 2000); } } } diff --git a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c index 2f30bda8457a..27d5e8d9f08d 100644 --- a/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c +++ b/drivers/infiniband/hw/ocrdma/ocrdma_verbs.c @@ -744,7 +744,8 @@ err: if (is_uctx_pd) { ocrdma_release_ucontext_pd(uctx); } else { - status = _ocrdma_dealloc_pd(dev, pd); + if (_ocrdma_dealloc_pd(dev, pd)) + pr_err("%s: _ocrdma_dealloc_pd() failed\n", __func__); } exit: return ERR_PTR(status); @@ -1901,6 +1902,7 @@ struct ib_srq *ocrdma_create_srq(struct ib_pd *ibpd, goto err; if (udata == NULL) { + status = -ENOMEM; srq->rqe_wr_id_tbl = kzalloc(sizeof(u64) * srq->rq.max_cnt, GFP_KERNEL); if (srq->rqe_wr_id_tbl == NULL) diff --git a/drivers/infiniband/hw/qedr/verbs.c b/drivers/infiniband/hw/qedr/verbs.c index 548e4d1e998f..2ae71b8f1ba8 100644 --- a/drivers/infiniband/hw/qedr/verbs.c +++ b/drivers/infiniband/hw/qedr/verbs.c @@ -53,6 +53,14 @@ #define DB_ADDR_SHIFT(addr) ((addr) << DB_PWM_ADDR_OFFSET_SHIFT) +static inline int qedr_ib_copy_to_udata(struct ib_udata *udata, void *src, + size_t len) +{ + size_t min_len = min_t(size_t, len, udata->outlen); + + return ib_copy_to_udata(udata, src, min_len); +} + int qedr_query_pkey(struct ib_device *ibdev, u8 port, u16 index, u16 *pkey) { if (index > QEDR_ROCE_PKEY_TABLE_LEN) @@ -378,7 +386,7 @@ struct ib_ucontext *qedr_alloc_ucontext(struct ib_device *ibdev, uresp.sges_per_srq_wr = dev->attr.max_srq_sge; uresp.max_cqes = QEDR_MAX_CQES; - rc = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); + rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp)); if (rc) goto err; @@ -499,7 +507,7 @@ struct ib_pd *qedr_alloc_pd(struct ib_device *ibdev, uresp.pd_id = pd_id; - rc = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); + rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp)); if (rc) { DP_ERR(dev, "copy error pd_id=0x%x.\n", pd_id); dev->ops->rdma_dealloc_pd(dev->rdma_ctx, pd_id); @@ -729,7 +737,7 @@ static int qedr_copy_cq_uresp(struct qedr_dev *dev, uresp.db_offset = DB_ADDR_SHIFT(DQ_PWM_OFFSET_UCM_RDMA_CQ_CONS_32BIT); uresp.icid = cq->icid; - rc = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); + rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp)); if (rc) DP_ERR(dev, "copy error cqid=0x%x.\n", cq->icid); @@ -1238,7 +1246,7 @@ static int qedr_copy_qp_uresp(struct qedr_dev *dev, uresp.atomic_supported = dev->atomic_cap != IB_ATOMIC_NONE; uresp.qp_id = qp->qp_id; - rc = ib_copy_to_udata(udata, &uresp, sizeof(uresp)); + rc = qedr_ib_copy_to_udata(udata, &uresp, sizeof(uresp)); if (rc) DP_ERR(dev, "create qp: failed a copy to user space with qp icid=0x%x.\n", diff --git a/drivers/infiniband/hw/qib/qib_qp.c b/drivers/infiniband/hw/qib/qib_qp.c index 5984981e7dd4..a343e3b5d4cb 100644 --- a/drivers/infiniband/hw/qib/qib_qp.c +++ b/drivers/infiniband/hw/qib/qib_qp.c @@ -104,10 +104,9 @@ const struct rvt_operation_params qib_post_parms[RVT_OPERATION_MAX] = { }; -static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map, - gfp_t gfp) +static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map) { - unsigned long page = get_zeroed_page(gfp); + unsigned long page = get_zeroed_page(GFP_KERNEL); /* * Free the page if someone raced with us installing it. @@ -126,7 +125,7 @@ static void get_map_page(struct rvt_qpn_table *qpt, struct rvt_qpn_map *map, * zero/one for QP type IB_QPT_SMI/IB_QPT_GSI. */ int qib_alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, - enum ib_qp_type type, u8 port, gfp_t gfp) + enum ib_qp_type type, u8 port) { u32 i, offset, max_scan, qpn; struct rvt_qpn_map *map; @@ -160,7 +159,7 @@ int qib_alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, max_scan = qpt->nmaps - !offset; for (i = 0;;) { if (unlikely(!map->page)) { - get_map_page(qpt, map, gfp); + get_map_page(qpt, map); if (unlikely(!map->page)) break; } @@ -317,16 +316,16 @@ u32 qib_mtu_from_qp(struct rvt_dev_info *rdi, struct rvt_qp *qp, u32 pmtu) return ib_mtu_enum_to_int(pmtu); } -void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, gfp_t gfp) +void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp) { struct qib_qp_priv *priv; - priv = kzalloc(sizeof(*priv), gfp); + priv = kzalloc(sizeof(*priv), GFP_KERNEL); if (!priv) return ERR_PTR(-ENOMEM); priv->owner = qp; - priv->s_hdr = kzalloc(sizeof(*priv->s_hdr), gfp); + priv->s_hdr = kzalloc(sizeof(*priv->s_hdr), GFP_KERNEL); if (!priv->s_hdr) { kfree(priv); return ERR_PTR(-ENOMEM); diff --git a/drivers/infiniband/hw/qib/qib_verbs.h b/drivers/infiniband/hw/qib/qib_verbs.h index da0db5485ddc..a52fc67b40d7 100644 --- a/drivers/infiniband/hw/qib/qib_verbs.h +++ b/drivers/infiniband/hw/qib/qib_verbs.h @@ -274,11 +274,11 @@ int qib_get_counters(struct qib_pportdata *ppd, * Functions provided by qib driver for rdmavt to use */ unsigned qib_free_all_qps(struct rvt_dev_info *rdi); -void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp, gfp_t gfp); +void *qib_qp_priv_alloc(struct rvt_dev_info *rdi, struct rvt_qp *qp); void qib_qp_priv_free(struct rvt_dev_info *rdi, struct rvt_qp *qp); void qib_notify_qp_reset(struct rvt_qp *qp); int qib_alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, - enum ib_qp_type type, u8 port, gfp_t gfp); + enum ib_qp_type type, u8 port); void qib_restart_rc(struct rvt_qp *qp, u32 psn, int wait); #ifdef CONFIG_DEBUG_FS diff --git a/drivers/infiniband/sw/rdmavt/qp.c b/drivers/infiniband/sw/rdmavt/qp.c index 727e81cc2c8f..8876ee7bc326 100644 --- a/drivers/infiniband/sw/rdmavt/qp.c +++ b/drivers/infiniband/sw/rdmavt/qp.c @@ -118,10 +118,9 @@ const int ib_rvt_state_ops[IB_QPS_ERR + 1] = { EXPORT_SYMBOL(ib_rvt_state_ops); static void get_map_page(struct rvt_qpn_table *qpt, - struct rvt_qpn_map *map, - gfp_t gfp) + struct rvt_qpn_map *map) { - unsigned long page = get_zeroed_page(gfp); + unsigned long page = get_zeroed_page(GFP_KERNEL); /* * Free the page if someone raced with us installing it. @@ -173,7 +172,7 @@ static int init_qpn_table(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt) rdi->dparms.qpn_res_start, rdi->dparms.qpn_res_end); for (i = rdi->dparms.qpn_res_start; i <= rdi->dparms.qpn_res_end; i++) { if (!map->page) { - get_map_page(qpt, map, GFP_KERNEL); + get_map_page(qpt, map); if (!map->page) { ret = -ENOMEM; break; @@ -342,14 +341,14 @@ static inline unsigned mk_qpn(struct rvt_qpn_table *qpt, * Return: The queue pair number */ static int alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, - enum ib_qp_type type, u8 port_num, gfp_t gfp) + enum ib_qp_type type, u8 port_num) { u32 i, offset, max_scan, qpn; struct rvt_qpn_map *map; u32 ret; if (rdi->driver_f.alloc_qpn) - return rdi->driver_f.alloc_qpn(rdi, qpt, type, port_num, gfp); + return rdi->driver_f.alloc_qpn(rdi, qpt, type, port_num); if (type == IB_QPT_SMI || type == IB_QPT_GSI) { unsigned n; @@ -374,7 +373,7 @@ static int alloc_qpn(struct rvt_dev_info *rdi, struct rvt_qpn_table *qpt, max_scan = qpt->nmaps - !offset; for (i = 0;;) { if (unlikely(!map->page)) { - get_map_page(qpt, map, gfp); + get_map_page(qpt, map); if (unlikely(!map->page)) break; } @@ -672,7 +671,6 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, struct ib_qp *ret = ERR_PTR(-ENOMEM); struct rvt_dev_info *rdi = ib_to_rvt(ibpd->device); void *priv = NULL; - gfp_t gfp; size_t sqsize; if (!rdi) @@ -680,18 +678,9 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, if (init_attr->cap.max_send_sge > rdi->dparms.props.max_sge || init_attr->cap.max_send_wr > rdi->dparms.props.max_qp_wr || - init_attr->create_flags & ~(IB_QP_CREATE_USE_GFP_NOIO)) + init_attr->create_flags) return ERR_PTR(-EINVAL); - /* GFP_NOIO is applicable to RC QP's only */ - - if (init_attr->create_flags & IB_QP_CREATE_USE_GFP_NOIO && - init_attr->qp_type != IB_QPT_RC) - return ERR_PTR(-EINVAL); - - gfp = init_attr->create_flags & IB_QP_CREATE_USE_GFP_NOIO ? - GFP_NOIO : GFP_KERNEL; - /* Check receive queue parameters if no SRQ is specified. */ if (!init_attr->srq) { if (init_attr->cap.max_recv_sge > rdi->dparms.props.max_sge || @@ -719,14 +708,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, sz = sizeof(struct rvt_sge) * init_attr->cap.max_send_sge + sizeof(struct rvt_swqe); - if (gfp == GFP_NOIO) - swq = __vmalloc( - sqsize * sz, - gfp | __GFP_ZERO, PAGE_KERNEL); - else - swq = vzalloc_node( - sqsize * sz, - rdi->dparms.node); + swq = vzalloc_node(sqsize * sz, rdi->dparms.node); if (!swq) return ERR_PTR(-ENOMEM); @@ -741,7 +723,8 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, } else if (init_attr->cap.max_recv_sge > 1) sg_list_sz = sizeof(*qp->r_sg_list) * (init_attr->cap.max_recv_sge - 1); - qp = kzalloc_node(sz + sg_list_sz, gfp, rdi->dparms.node); + qp = kzalloc_node(sz + sg_list_sz, GFP_KERNEL, + rdi->dparms.node); if (!qp) goto bail_swq; @@ -751,7 +734,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, kzalloc_node( sizeof(*qp->s_ack_queue) * rvt_max_atomic(rdi), - gfp, + GFP_KERNEL, rdi->dparms.node); if (!qp->s_ack_queue) goto bail_qp; @@ -766,7 +749,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, * Driver needs to set up it's private QP structure and do any * initialization that is needed. */ - priv = rdi->driver_f.qp_priv_alloc(rdi, qp, gfp); + priv = rdi->driver_f.qp_priv_alloc(rdi, qp); if (IS_ERR(priv)) { ret = priv; goto bail_qp; @@ -786,11 +769,6 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, qp->r_rq.wq = vmalloc_user( sizeof(struct rvt_rwq) + qp->r_rq.size * sz); - else if (gfp == GFP_NOIO) - qp->r_rq.wq = __vmalloc( - sizeof(struct rvt_rwq) + - qp->r_rq.size * sz, - gfp | __GFP_ZERO, PAGE_KERNEL); else qp->r_rq.wq = vzalloc_node( sizeof(struct rvt_rwq) + @@ -824,7 +802,7 @@ struct ib_qp *rvt_create_qp(struct ib_pd *ibpd, err = alloc_qpn(rdi, &rdi->qp_dev->qpn_table, init_attr->qp_type, - init_attr->port_num, gfp); + init_attr->port_num); if (err < 0) { ret = ERR_PTR(err); goto bail_rq_wq; @@ -1280,9 +1258,7 @@ int rvt_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, if (attr_mask & IB_QP_TIMEOUT) { qp->timeout = attr->timeout; - qp->timeout_jiffies = - usecs_to_jiffies((4096UL * (1UL << qp->timeout)) / - 1000UL); + qp->timeout_jiffies = rvt_timeout_to_jiffies(qp->timeout); } if (attr_mask & IB_QP_QKEY) diff --git a/drivers/infiniband/sw/rxe/rxe_net.c b/drivers/infiniband/sw/rxe/rxe_net.c index c3a140ed4df2..08f3f90d2912 100644 --- a/drivers/infiniband/sw/rxe/rxe_net.c +++ b/drivers/infiniband/sw/rxe/rxe_net.c @@ -441,6 +441,8 @@ static void rxe_skb_tx_dtor(struct sk_buff *skb) if (unlikely(qp->need_req_skb && skb_out < RXE_INFLIGHT_SKBS_PER_QP_LOW)) rxe_run_task(&qp->req.task, 1); + + rxe_drop_ref(qp); } int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb) @@ -473,6 +475,7 @@ int rxe_send(struct rxe_dev *rxe, struct rxe_pkt_info *pkt, struct sk_buff *skb) return -EAGAIN; } + rxe_add_ref(pkt->qp); atomic_inc(&pkt->qp->skb_out); kfree_skb(skb); diff --git a/drivers/infiniband/sw/rxe/rxe_resp.c b/drivers/infiniband/sw/rxe/rxe_resp.c index be944d5aa9af..a958ee918a49 100644 --- a/drivers/infiniband/sw/rxe/rxe_resp.c +++ b/drivers/infiniband/sw/rxe/rxe_resp.c @@ -1219,6 +1219,9 @@ void rxe_drain_req_pkts(struct rxe_qp *qp, bool notify) kfree_skb(skb); } + if (notify) + return; + while (!qp->srq && qp->rq.queue && queue_head(qp->rq.queue)) advance_consumer(qp->rq.queue); } diff --git a/drivers/infiniband/sw/rxe/rxe_verbs.c b/drivers/infiniband/sw/rxe/rxe_verbs.c index 073e66783f1d..af90a7d42b96 100644 --- a/drivers/infiniband/sw/rxe/rxe_verbs.c +++ b/drivers/infiniband/sw/rxe/rxe_verbs.c @@ -914,6 +914,9 @@ static int rxe_post_recv(struct ib_qp *ibqp, struct ib_recv_wr *wr, spin_unlock_irqrestore(&rq->producer_lock, flags); + if (qp->resp.state == QP_STATE_ERROR) + rxe_run_task(&qp->resp.task, 1); + err1: return err; } @@ -1240,6 +1243,8 @@ int rxe_register_device(struct rxe_dev *rxe) addrconf_addr_eui48((unsigned char *)&dev->node_guid, rxe->ndev->dev_addr); dev->dev.dma_ops = &dma_virt_ops; + dma_coerce_mask_and_coherent(&dev->dev, + dma_get_required_mask(dev->dev.parent)); dev->uverbs_abi_ver = RXE_UVERBS_ABI_VERSION; dev->uverbs_cmd_mask = BIT_ULL(IB_USER_VERBS_CMD_GET_CONTEXT) diff --git a/drivers/infiniband/ulp/ipoib/ipoib.h b/drivers/infiniband/ulp/ipoib/ipoib.h index ff50a7bd66d8..7ac25059c40f 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib.h +++ b/drivers/infiniband/ulp/ipoib/ipoib.h @@ -336,6 +336,7 @@ struct ipoib_dev_priv { unsigned long flags; struct rw_semaphore vlan_rwsem; + struct mutex mcast_mutex; struct rb_root path_tree; struct list_head path_list; diff --git a/drivers/infiniband/ulp/ipoib/ipoib_cm.c b/drivers/infiniband/ulp/ipoib/ipoib_cm.c index 7cbcfdac6529..d69410c2ed97 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_cm.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_cm.c @@ -39,6 +39,7 @@ #include <linux/vmalloc.h> #include <linux/moduleparam.h> #include <linux/sched/signal.h> +#include <linux/sched/mm.h> #include "ipoib.h" @@ -510,7 +511,6 @@ static int ipoib_cm_rx_handler(struct ib_cm_id *cm_id, case IB_CM_REQ_RECEIVED: return ipoib_cm_req_handler(cm_id, event); case IB_CM_DREQ_RECEIVED: - p = cm_id->context; ib_send_cm_drep(cm_id, NULL, 0); /* Fall through */ case IB_CM_REJ_RECEIVED: @@ -954,7 +954,7 @@ void ipoib_cm_dev_stop(struct net_device *dev) break; } spin_unlock_irq(&priv->lock); - msleep(1); + usleep_range(1000, 2000); ipoib_drain_cq(dev); spin_lock_irq(&priv->lock); } @@ -1047,9 +1047,8 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_ .sq_sig_type = IB_SIGNAL_ALL_WR, .qp_type = IB_QPT_RC, .qp_context = tx, - .create_flags = IB_QP_CREATE_USE_GFP_NOIO + .create_flags = 0 }; - struct ib_qp *tx_qp; if (dev->features & NETIF_F_SG) @@ -1057,10 +1056,6 @@ static struct ib_qp *ipoib_cm_create_tx_qp(struct net_device *dev, struct ipoib_ min_t(u32, priv->ca->attrs.max_sge, MAX_SKB_FRAGS + 1); tx_qp = ib_create_qp(priv->pd, &attr); - if (PTR_ERR(tx_qp) == -EINVAL) { - attr.create_flags &= ~IB_QP_CREATE_USE_GFP_NOIO; - tx_qp = ib_create_qp(priv->pd, &attr); - } tx->max_send_sge = attr.cap.max_send_sge; return tx_qp; } @@ -1131,10 +1126,11 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, struct sa_path_rec *pathrec) { struct ipoib_dev_priv *priv = ipoib_priv(p->dev); + unsigned int noio_flag; int ret; - p->tx_ring = __vmalloc(ipoib_sendq_size * sizeof *p->tx_ring, - GFP_NOIO, PAGE_KERNEL); + noio_flag = memalloc_noio_save(); + p->tx_ring = vzalloc(ipoib_sendq_size * sizeof(*p->tx_ring)); if (!p->tx_ring) { ret = -ENOMEM; goto err_tx; @@ -1142,9 +1138,10 @@ static int ipoib_cm_tx_init(struct ipoib_cm_tx *p, u32 qpn, memset(p->tx_ring, 0, ipoib_sendq_size * sizeof *p->tx_ring); p->qp = ipoib_cm_create_tx_qp(p->dev, p); + memalloc_noio_restore(noio_flag); if (IS_ERR(p->qp)) { ret = PTR_ERR(p->qp); - ipoib_warn(priv, "failed to allocate tx qp: %d\n", ret); + ipoib_warn(priv, "failed to create tx qp: %d\n", ret); goto err_qp; } @@ -1206,7 +1203,7 @@ static void ipoib_cm_tx_destroy(struct ipoib_cm_tx *p) goto timeout; } - msleep(1); + usleep_range(1000, 2000); } } diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c index 7871379342f4..184a22f48027 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ethtool.c @@ -52,7 +52,8 @@ static const struct ipoib_stats ipoib_gstrings_stats[] = { IPOIB_NETDEV_STAT(tx_bytes), IPOIB_NETDEV_STAT(tx_errors), IPOIB_NETDEV_STAT(rx_dropped), - IPOIB_NETDEV_STAT(tx_dropped) + IPOIB_NETDEV_STAT(tx_dropped), + IPOIB_NETDEV_STAT(multicast), }; #define IPOIB_GLOBAL_STATS_LEN ARRAY_SIZE(ipoib_gstrings_stats) diff --git a/drivers/infiniband/ulp/ipoib/ipoib_ib.c b/drivers/infiniband/ulp/ipoib/ipoib_ib.c index efe7402f4885..2e075377242e 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_ib.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_ib.c @@ -256,6 +256,8 @@ static void ipoib_ib_handle_rx_wc(struct net_device *dev, struct ib_wc *wc) ++dev->stats.rx_packets; dev->stats.rx_bytes += skb->len; + if (skb->pkt_type == PACKET_MULTICAST) + dev->stats.multicast++; skb->dev = dev; if ((dev->features & NETIF_F_RXCSUM) && @@ -709,6 +711,27 @@ static int recvs_pending(struct net_device *dev) return pending; } +static void check_qp_movement_and_print(struct ipoib_dev_priv *priv, + struct ib_qp *qp, + enum ib_qp_state new_state) +{ + struct ib_qp_attr qp_attr; + struct ib_qp_init_attr query_init_attr; + int ret; + + ret = ib_query_qp(qp, &qp_attr, IB_QP_STATE, &query_init_attr); + if (ret) { + ipoib_warn(priv, "%s: Failed to query QP\n", __func__); + return; + } + /* print according to the new-state and the previous state.*/ + if (new_state == IB_QPS_ERR && qp_attr.qp_state == IB_QPS_RESET) + ipoib_dbg(priv, "Failed modify QP, IB_QPS_RESET to IB_QPS_ERR, acceptable\n"); + else + ipoib_warn(priv, "Failed to modify QP to state: %d from state: %d\n", + new_state, qp_attr.qp_state); +} + int ipoib_ib_dev_stop_default(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); @@ -728,7 +751,7 @@ int ipoib_ib_dev_stop_default(struct net_device *dev) */ qp_attr.qp_state = IB_QPS_ERR; if (ib_modify_qp(priv->qp, &qp_attr, IB_QP_STATE)) - ipoib_warn(priv, "Failed to modify QP to ERROR state\n"); + check_qp_movement_and_print(priv, priv->qp, IB_QPS_ERR); /* Wait for all sends and receives to complete */ begin = jiffies; @@ -770,7 +793,7 @@ int ipoib_ib_dev_stop_default(struct net_device *dev) ipoib_drain_cq(dev); - msleep(1); + usleep_range(1000, 2000); } ipoib_dbg(priv, "All sends and receives done.\n"); diff --git a/drivers/infiniband/ulp/ipoib/ipoib_main.c b/drivers/infiniband/ulp/ipoib/ipoib_main.c index 6e86eeee370e..6c77df34869d 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_main.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_main.c @@ -233,6 +233,7 @@ static netdev_features_t ipoib_fix_features(struct net_device *dev, netdev_featu static int ipoib_change_mtu(struct net_device *dev, int new_mtu) { struct ipoib_dev_priv *priv = ipoib_priv(dev); + int ret = 0; /* dev->mtu > 2K ==> connected mode */ if (ipoib_cm_admin_enabled(dev)) { @@ -256,9 +257,34 @@ static int ipoib_change_mtu(struct net_device *dev, int new_mtu) ipoib_dbg(priv, "MTU must be smaller than the underlying " "link layer MTU - 4 (%u)\n", priv->mcast_mtu); - dev->mtu = min(priv->mcast_mtu, priv->admin_mtu); + new_mtu = min(priv->mcast_mtu, priv->admin_mtu); - return 0; + if (priv->rn_ops->ndo_change_mtu) { + bool carrier_status = netif_carrier_ok(dev); + + netif_carrier_off(dev); + + /* notify lower level on the real mtu */ + ret = priv->rn_ops->ndo_change_mtu(dev, new_mtu); + + if (carrier_status) + netif_carrier_on(dev); + } else { + dev->mtu = new_mtu; + } + + return ret; +} + +static void ipoib_get_stats(struct net_device *dev, + struct rtnl_link_stats64 *stats) +{ + struct ipoib_dev_priv *priv = ipoib_priv(dev); + + if (priv->rn_ops->ndo_get_stats64) + priv->rn_ops->ndo_get_stats64(dev, stats); + else + netdev_stats_to_stats64(stats, &dev->stats); } /* Called with an RCU read lock taken */ @@ -1534,6 +1560,7 @@ static void ipoib_flush_neighs(struct ipoib_dev_priv *priv) int i, wait_flushed = 0; init_completion(&priv->ntbl.flushed); + set_bit(IPOIB_NEIGH_TBL_FLUSH, &priv->flags); spin_lock_irqsave(&priv->lock, flags); @@ -1578,7 +1605,6 @@ static void ipoib_neigh_hash_uninit(struct net_device *dev) ipoib_dbg(priv, "ipoib_neigh_hash_uninit\n"); init_completion(&priv->ntbl.deleted); - set_bit(IPOIB_NEIGH_TBL_FLUSH, &priv->flags); /* Stop GC if called at init fail need to cancel work */ stopped = test_and_set_bit(IPOIB_STOP_NEIGH_GC, &priv->flags); @@ -1808,6 +1834,7 @@ static const struct net_device_ops ipoib_netdev_ops_pf = { .ndo_get_vf_stats = ipoib_get_vf_stats, .ndo_set_vf_guid = ipoib_set_vf_guid, .ndo_set_mac_address = ipoib_set_mac, + .ndo_get_stats64 = ipoib_get_stats, }; static const struct net_device_ops ipoib_netdev_ops_vf = { @@ -1820,6 +1847,7 @@ static const struct net_device_ops ipoib_netdev_ops_vf = { .ndo_tx_timeout = ipoib_timeout, .ndo_set_rx_mode = ipoib_set_mcast_list, .ndo_get_iflink = ipoib_get_iflink, + .ndo_get_stats64 = ipoib_get_stats, }; void ipoib_setup_common(struct net_device *dev) @@ -1850,6 +1878,7 @@ 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); INIT_LIST_HEAD(&priv->child_intfs); @@ -2146,14 +2175,14 @@ static struct net_device *ipoib_add_port(const char *format, priv->dev->dev_id = port - 1; result = ib_query_port(hca, port, &attr); - if (!result) - priv->max_ib_mtu = ib_mtu_enum_to_int(attr.max_mtu); - else { + if (result) { printk(KERN_WARNING "%s: ib_query_port %d failed\n", hca->name, port); goto device_init_failed; } + priv->max_ib_mtu = ib_mtu_enum_to_int(attr.max_mtu); + /* MTU will be reset when mcast join happens */ priv->dev->mtu = IPOIB_UD_MTU(priv->max_ib_mtu); priv->mcast_mtu = priv->admin_mtu = priv->dev->mtu; @@ -2184,12 +2213,14 @@ static struct net_device *ipoib_add_port(const char *format, printk(KERN_WARNING "%s: ib_query_gid port %d failed (ret = %d)\n", hca->name, port, result); goto device_init_failed; - } else - memcpy(priv->dev->dev_addr + 4, priv->local_gid.raw, sizeof (union ib_gid)); + } + + memcpy(priv->dev->dev_addr + 4, priv->local_gid.raw, + sizeof(union ib_gid)); set_bit(IPOIB_FLAG_DEV_ADDR_SET, &priv->flags); result = ipoib_dev_init(priv->dev, hca, port); - if (result < 0) { + if (result) { printk(KERN_WARNING "%s: failed to initialize port %d (ret = %d)\n", hca->name, port, result); goto device_init_failed; @@ -2212,6 +2243,7 @@ static struct net_device *ipoib_add_port(const char *format, goto register_failed; } + result = -ENOMEM; if (ipoib_cm_add_mode_attr(priv->dev)) goto sysfs_failed; if (ipoib_add_pkey_attr(priv->dev)) @@ -2337,6 +2369,7 @@ static int __init ipoib_init_module(void) ipoib_sendq_size = max3(ipoib_sendq_size, 2 * MAX_SEND_CQE, IPOIB_MIN_QUEUE_SIZE); #ifdef CONFIG_INFINIBAND_IPOIB_CM ipoib_max_conn_qp = min(ipoib_max_conn_qp, IPOIB_CM_MAX_CONN_QP); + ipoib_max_conn_qp = max(ipoib_max_conn_qp, 0); #endif /* diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c index 057f58e6afca..93e149efc1f5 100644 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c @@ -684,15 +684,10 @@ void ipoib_mcast_start_thread(struct net_device *dev) int ipoib_mcast_stop_thread(struct net_device *dev) { struct ipoib_dev_priv *priv = ipoib_priv(dev); - unsigned long flags; ipoib_dbg_mcast(priv, "stopping multicast thread\n"); - spin_lock_irqsave(&priv->lock, flags); - cancel_delayed_work(&priv->mcast_task); - spin_unlock_irqrestore(&priv->lock, flags); - - flush_workqueue(priv->wq); + cancel_delayed_work_sync(&priv->mcast_task); return 0; } @@ -748,6 +743,14 @@ void ipoib_mcast_remove_list(struct list_head *remove_list) { struct ipoib_mcast *mcast, *tmcast; + /* + * make sure the in-flight joins have finished before we attempt + * to leave + */ + list_for_each_entry_safe(mcast, tmcast, remove_list, list) + if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)) + wait_for_completion(&mcast->done); + list_for_each_entry_safe(mcast, tmcast, remove_list, list) { ipoib_mcast_leave(mcast->dev, mcast); ipoib_mcast_free(mcast); @@ -838,6 +841,7 @@ void ipoib_mcast_dev_flush(struct net_device *dev) struct ipoib_mcast *mcast, *tmcast; unsigned long flags; + mutex_lock(&priv->mcast_mutex); ipoib_dbg_mcast(priv, "flushing multicast list\n"); spin_lock_irqsave(&priv->lock, flags); @@ -856,15 +860,8 @@ void ipoib_mcast_dev_flush(struct net_device *dev) spin_unlock_irqrestore(&priv->lock, flags); - /* - * make sure the in-flight joins have finished before we attempt - * to leave - */ - list_for_each_entry_safe(mcast, tmcast, &remove_list, list) - if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)) - wait_for_completion(&mcast->done); - ipoib_mcast_remove_list(&remove_list); + mutex_unlock(&priv->mcast_mutex); } static int ipoib_mcast_addr_is_valid(const u8 *addr, const u8 *broadcast) @@ -982,14 +979,6 @@ void ipoib_mcast_restart_task(struct work_struct *work) netif_addr_unlock(dev); local_irq_restore(flags); - /* - * make sure the in-flight joins have finished before we attempt - * to leave - */ - list_for_each_entry_safe(mcast, tmcast, &remove_list, list) - if (test_bit(IPOIB_MCAST_FLAG_BUSY, &mcast->flags)) - wait_for_completion(&mcast->done); - ipoib_mcast_remove_list(&remove_list); /* diff --git a/drivers/infiniband/ulp/iser/iscsi_iser.c b/drivers/infiniband/ulp/iser/iscsi_iser.c index 5a887efb4bdf..37b33d708c2d 100644 --- a/drivers/infiniband/ulp/iser/iscsi_iser.c +++ b/drivers/infiniband/ulp/iser/iscsi_iser.c @@ -83,6 +83,7 @@ static struct scsi_host_template iscsi_iser_sht; static struct iscsi_transport iscsi_iser_transport; static struct scsi_transport_template *iscsi_iser_scsi_transport; static struct workqueue_struct *release_wq; +static DEFINE_MUTEX(unbind_iser_conn_mutex); struct iser_global ig; int iser_debug_level = 0; @@ -550,12 +551,14 @@ iscsi_iser_conn_stop(struct iscsi_cls_conn *cls_conn, int flag) */ if (iser_conn) { mutex_lock(&iser_conn->state_mutex); + mutex_lock(&unbind_iser_conn_mutex); iser_conn_terminate(iser_conn); iscsi_conn_stop(cls_conn, flag); /* unbind */ iser_conn->iscsi_conn = NULL; conn->dd_data = NULL; + mutex_unlock(&unbind_iser_conn_mutex); complete(&iser_conn->stop_completion); mutex_unlock(&iser_conn->state_mutex); @@ -977,13 +980,21 @@ static int iscsi_iser_slave_alloc(struct scsi_device *sdev) struct iser_conn *iser_conn; struct ib_device *ib_dev; + mutex_lock(&unbind_iser_conn_mutex); + session = starget_to_session(scsi_target(sdev))->dd_data; iser_conn = session->leadconn->dd_data; + if (!iser_conn) { + mutex_unlock(&unbind_iser_conn_mutex); + return -ENOTCONN; + } ib_dev = iser_conn->ib_conn.device->ib_device; if (!(ib_dev->attrs.device_cap_flags & IB_DEVICE_SG_GAPS_REG)) blk_queue_virt_boundary(sdev->request_queue, ~MASK_4K); + mutex_unlock(&unbind_iser_conn_mutex); + return 0; } diff --git a/drivers/infiniband/ulp/iser/iser_initiator.c b/drivers/infiniband/ulp/iser/iser_initiator.c index 12ed62ce9ff7..2a07692007bd 100644 --- a/drivers/infiniband/ulp/iser/iser_initiator.c +++ b/drivers/infiniband/ulp/iser/iser_initiator.c @@ -137,8 +137,10 @@ iser_prepare_write_cmd(struct iscsi_task *task, if (unsol_sz < edtl) { hdr->flags |= ISER_WSV; - hdr->write_stag = cpu_to_be32(mem_reg->rkey); - hdr->write_va = cpu_to_be64(mem_reg->sge.addr + unsol_sz); + if (buf_out->data_len > imm_sz) { + hdr->write_stag = cpu_to_be32(mem_reg->rkey); + hdr->write_va = cpu_to_be64(mem_reg->sge.addr + unsol_sz); + } iser_dbg("Cmd itt:%d, WRITE tags, RKEY:%#.4X " "VA:%#llX + unsol:%d\n", diff --git a/drivers/infiniband/ulp/iser/iser_verbs.c b/drivers/infiniband/ulp/iser/iser_verbs.c index c538a38c91ce..26a004e97ae0 100644 --- a/drivers/infiniband/ulp/iser/iser_verbs.c +++ b/drivers/infiniband/ulp/iser/iser_verbs.c @@ -708,8 +708,14 @@ iser_calc_scsi_params(struct iser_conn *iser_conn, unsigned short sg_tablesize, sup_sg_tablesize; sg_tablesize = DIV_ROUND_UP(max_sectors * 512, SIZE_4K); - sup_sg_tablesize = min_t(unsigned, ISCSI_ISER_MAX_SG_TABLESIZE, - device->ib_device->attrs.max_fast_reg_page_list_len); + if (device->ib_device->attrs.device_cap_flags & + IB_DEVICE_MEM_MGT_EXTENSIONS) + sup_sg_tablesize = + min_t( + uint, ISCSI_ISER_MAX_SG_TABLESIZE, + device->ib_device->attrs.max_fast_reg_page_list_len); + else + sup_sg_tablesize = ISCSI_ISER_MAX_SG_TABLESIZE; iser_conn->scsi_sg_tablesize = min(sg_tablesize, sup_sg_tablesize); } |