Merge branch 'nvme-5.8' of git://git.infradead.org/nvme into block-5.8

Pull NVMe fixes from Christoph.

* 'nvme-5.8' of git://git.infradead.org/nvme:
  nvme-multipath: fix bogus request queue reference put
  nvme-multipath: fix deadlock due to head->lock
  nvme: don't protect ns mutation with ns->head->lock
  nvme-multipath: fix deadlock between ana_work and scan_work
  nvme: fix possible deadlock when I/O is blocked
  nvme-rdma: assign completion vector correctly
  nvme-loop: initialize tagset numa value to the value of the ctrl
  nvme-tcp: initialize tagset numa value to the value of the ctrl
  nvme-pci: initialize tagset numa value to the value of the ctrl
  nvme-pci: override the value of the controller's numa node
  nvme: set initial value for controller's numa node

Author: axboe

drivers/nvme/host/core.c

@@ -1974,7 +1974,6 @@ static int __nvme_revalidate_disk(struct gendisk *disk, struct nvme_id_ns *id)
 	if (ns->head->disk) {
 		nvme_update_disk_info(ns->head->disk, ns, id);
 		blk_queue_stack_limits(ns->head->disk->queue, ns->queue);
-		revalidate_disk(ns->head->disk);
 	}
 #endif
 	return 0;
@@ -4174,6 +4173,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 	ctrl->dev = dev;
 	ctrl->ops = ops;
 	ctrl->quirks = quirks;
+	ctrl->numa_node = NUMA_NO_NODE;
 	INIT_WORK(&ctrl->scan_work, nvme_scan_work);
 	INIT_WORK(&ctrl->async_event_work, nvme_async_event_work);
 	INIT_WORK(&ctrl->fw_act_work, nvme_fw_act_work);

drivers/nvme/host/multipath.c

@@ -409,15 +409,14 @@ static void nvme_mpath_set_live(struct nvme_ns *ns)
 {
 	struct nvme_ns_head *head = ns->head;
 
-	lockdep_assert_held(&ns->head->lock);
-
 	if (!head->disk)
 		return;
 
-	if (!(head->disk->flags & GENHD_FL_UP))
+	if (!test_and_set_bit(NVME_NSHEAD_DISK_LIVE, &head->flags))
 		device_add_disk(&head->subsys->dev, head->disk,
 				nvme_ns_id_attr_groups);
 
+	mutex_lock(&head->lock);
 	if (nvme_path_is_optimized(ns)) {
 		int node, srcu_idx;
 
@@ -426,9 +425,10 @@ static void nvme_mpath_set_live(struct nvme_ns *ns)
 			__nvme_find_path(head, node);
 		srcu_read_unlock(&head->srcu, srcu_idx);
 	}
+	mutex_unlock(&head->lock);
 
-	synchronize_srcu(&ns->head->srcu);
-	kblockd_schedule_work(&ns->head->requeue_work);
+	synchronize_srcu(&head->srcu);
+	kblockd_schedule_work(&head->requeue_work);
 }
 
 static int nvme_parse_ana_log(struct nvme_ctrl *ctrl, void *data,
@@ -483,14 +483,12 @@ static inline bool nvme_state_is_live(enum nvme_ana_state state)
 static void nvme_update_ns_ana_state(struct nvme_ana_group_desc *desc,
 		struct nvme_ns *ns)
 {
-	mutex_lock(&ns->head->lock);
 	ns->ana_grpid = le32_to_cpu(desc->grpid);
 	ns->ana_state = desc->state;
 	clear_bit(NVME_NS_ANA_PENDING, &ns->flags);
 
 	if (nvme_state_is_live(ns->ana_state))
 		nvme_mpath_set_live(ns);
-	mutex_unlock(&ns->head->lock);
 }
 
 static int nvme_update_ana_state(struct nvme_ctrl *ctrl,
@@ -640,31 +638,37 @@ static ssize_t ana_state_show(struct device *dev, struct device_attribute *attr,
 }
 DEVICE_ATTR_RO(ana_state);
 
-static int nvme_set_ns_ana_state(struct nvme_ctrl *ctrl,
+static int nvme_lookup_ana_group_desc(struct nvme_ctrl *ctrl,
 		struct nvme_ana_group_desc *desc, void *data)
 {
-	struct nvme_ns *ns = data;
+	struct nvme_ana_group_desc *dst = data;
 
-	if (ns->ana_grpid == le32_to_cpu(desc->grpid)) {
-		nvme_update_ns_ana_state(desc, ns);
-		return -ENXIO; /* just break out of the loop */
-	}
+	if (desc->grpid != dst->grpid)
+		return 0;
 
-	return 0;
+	*dst = *desc;
+	return -ENXIO; /* just break out of the loop */
 }
 
 void nvme_mpath_add_disk(struct nvme_ns *ns, struct nvme_id_ns *id)
 {
 	if (nvme_ctrl_use_ana(ns->ctrl)) {
+		struct nvme_ana_group_desc desc = {
+			.grpid = id->anagrpid,
+			.state = 0,
+		};
+
 		mutex_lock(&ns->ctrl->ana_lock);
 		ns->ana_grpid = le32_to_cpu(id->anagrpid);
-		nvme_parse_ana_log(ns->ctrl, ns, nvme_set_ns_ana_state);
+		nvme_parse_ana_log(ns->ctrl, &desc, nvme_lookup_ana_group_desc);
 		mutex_unlock(&ns->ctrl->ana_lock);
+		if (desc.state) {
+			/* found the group desc: update */
+			nvme_update_ns_ana_state(&desc, ns);
+		}
 	} else {
-		mutex_lock(&ns->head->lock);
 		ns->ana_state = NVME_ANA_OPTIMIZED; 
 		nvme_mpath_set_live(ns);
-		mutex_unlock(&ns->head->lock);
 	}
 
 	if (bdi_cap_stable_pages_required(ns->queue->backing_dev_info)) {
@@ -686,6 +690,14 @@ void nvme_mpath_remove_disk(struct nvme_ns_head *head)
 	kblockd_schedule_work(&head->requeue_work);
 	flush_work(&head->requeue_work);
 	blk_cleanup_queue(head->disk->queue);
+	if (!test_bit(NVME_NSHEAD_DISK_LIVE, &head->flags)) {
+		/*
+		 * if device_add_disk wasn't called, prevent
+		 * disk release to put a bogus reference on the
+		 * request queue
+		 */
+		head->disk->queue = NULL;
+	}
 	put_disk(head->disk);
 }
 

drivers/nvme/host/nvme.h

@@ -364,6 +364,8 @@ struct nvme_ns_head {
 	spinlock_t		requeue_lock;
 	struct work_struct	requeue_work;
 	struct mutex		lock;
+	unsigned long		flags;
+#define NVME_NSHEAD_DISK_LIVE	0
 	struct nvme_ns __rcu	*current_path[];
 #endif
 };

drivers/nvme/host/pci.c

@@ -1593,7 +1593,7 @@ static int nvme_alloc_admin_tags(struct nvme_dev *dev)
 
 		dev->admin_tagset.queue_depth = NVME_AQ_MQ_TAG_DEPTH;
 		dev->admin_tagset.timeout = ADMIN_TIMEOUT;
-		dev->admin_tagset.numa_node = dev_to_node(dev->dev);
+		dev->admin_tagset.numa_node = dev->ctrl.numa_node;
 		dev->admin_tagset.cmd_size = sizeof(struct nvme_iod);
 		dev->admin_tagset.flags = BLK_MQ_F_NO_SCHED;
 		dev->admin_tagset.driver_data = dev;
@@ -1669,6 +1669,8 @@ static int nvme_pci_configure_admin_queue(struct nvme_dev *dev)
 	if (result)
 		return result;
 
+	dev->ctrl.numa_node = dev_to_node(dev->dev);
+
 	nvmeq = &dev->queues[0];
 	aqa = nvmeq->q_depth - 1;
 	aqa |= aqa << 16;
@@ -2257,7 +2259,7 @@ static void nvme_dev_add(struct nvme_dev *dev)
 		if (dev->io_queues[HCTX_TYPE_POLL])
 			dev->tagset.nr_maps++;
 		dev->tagset.timeout = NVME_IO_TIMEOUT;
-		dev->tagset.numa_node = dev_to_node(dev->dev);
+		dev->tagset.numa_node = dev->ctrl.numa_node;
 		dev->tagset.queue_depth =
 				min_t(int, dev->q_depth, BLK_MQ_MAX_DEPTH) - 1;
 		dev->tagset.cmd_size = sizeof(struct nvme_iod);

drivers/nvme/host/rdma.c

@@ -470,7 +470,7 @@ static int nvme_rdma_create_queue_ib(struct nvme_rdma_queue *queue)
 	 * Spread I/O queues completion vectors according their queue index.
 	 * Admin queues can always go on completion vector 0.
 	 */
-	comp_vector = idx == 0 ? idx : idx - 1;
+	comp_vector = (idx == 0 ? idx : idx - 1) % ibdev->num_comp_vectors;
 
 	/* Polling queues need direct cq polling context */
 	if (nvme_rdma_poll_queue(queue))

drivers/nvme/host/tcp.c

@@ -1532,7 +1532,7 @@ static struct blk_mq_tag_set *nvme_tcp_alloc_tagset(struct nvme_ctrl *nctrl,
 		set->ops = &nvme_tcp_admin_mq_ops;
 		set->queue_depth = NVME_AQ_MQ_TAG_DEPTH;
 		set->reserved_tags = 2; /* connect + keep-alive */
-		set->numa_node = NUMA_NO_NODE;
+		set->numa_node = nctrl->numa_node;
 		set->flags = BLK_MQ_F_BLOCKING;
 		set->cmd_size = sizeof(struct nvme_tcp_request);
 		set->driver_data = ctrl;
@@ -1544,7 +1544,7 @@ static struct blk_mq_tag_set *nvme_tcp_alloc_tagset(struct nvme_ctrl *nctrl,
 		set->ops = &nvme_tcp_mq_ops;
 		set->queue_depth = nctrl->sqsize + 1;
 		set->reserved_tags = 1; /* fabric connect */
-		set->numa_node = NUMA_NO_NODE;
+		set->numa_node = nctrl->numa_node;
 		set->flags = BLK_MQ_F_SHOULD_MERGE | BLK_MQ_F_BLOCKING;
 		set->cmd_size = sizeof(struct nvme_tcp_request);
 		set->driver_data = ctrl;

drivers/nvme/target/loop.c

@@ -340,7 +340,7 @@ static int nvme_loop_configure_admin_queue(struct nvme_loop_ctrl *ctrl)
 	ctrl->admin_tag_set.ops = &nvme_loop_admin_mq_ops;
 	ctrl->admin_tag_set.queue_depth = NVME_AQ_MQ_TAG_DEPTH;
 	ctrl->admin_tag_set.reserved_tags = 2; /* connect + keep-alive */
-	ctrl->admin_tag_set.numa_node = NUMA_NO_NODE;
+	ctrl->admin_tag_set.numa_node = ctrl->ctrl.numa_node;
 	ctrl->admin_tag_set.cmd_size = sizeof(struct nvme_loop_iod) +
 		NVME_INLINE_SG_CNT * sizeof(struct scatterlist);
 	ctrl->admin_tag_set.driver_data = ctrl;
@@ -512,7 +512,7 @@ static int nvme_loop_create_io_queues(struct nvme_loop_ctrl *ctrl)
 	ctrl->tag_set.ops = &nvme_loop_mq_ops;
 	ctrl->tag_set.queue_depth = ctrl->ctrl.opts->queue_size;
 	ctrl->tag_set.reserved_tags = 1; /* fabric connect */
-	ctrl->tag_set.numa_node = NUMA_NO_NODE;
+	ctrl->tag_set.numa_node = ctrl->ctrl.numa_node;
 	ctrl->tag_set.flags = BLK_MQ_F_SHOULD_MERGE;
 	ctrl->tag_set.cmd_size = sizeof(struct nvme_loop_iod) +
 		NVME_INLINE_SG_CNT * sizeof(struct scatterlist);