nvme: change namespaces_mutext to namespaces_rwsem

namespaces_mutext is used to synchronize the operations on ctrl
namespaces list. Most of the time, it is a read operation.

On the other hand, there are many interfaces in nvme core that
need this lock, such as nvme_wait_freeze, and even more interfaces
will be added. If we use mutex here, circular dependency could be
introduced easily. For example:
context A                  context B
nvme_xxx                   nvme_xxx
hold namespaces_mutext     require namespaces_mutext
sync context B

So it is better to change it from mutex to rwsem.

Reviewed-by: Keith Busch <keith.busch@intel.com>
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

drivers/nvme/host/core.c

@@ -1125,13 +1125,13 @@ static void nvme_update_formats(struct nvme_ctrl *ctrl)
 	struct nvme_ns *ns, *next;
 	LIST_HEAD(rm_list);
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_write(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list) {
 		if (ns->disk && nvme_revalidate_disk(ns->disk)) {
 			list_move_tail(&ns->list, &rm_list);
 		}
 	}
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_write(&ctrl->namespaces_rwsem);
 
 	list_for_each_entry_safe(ns, next, &rm_list, list)
 		nvme_ns_remove(ns);
@@ -2441,7 +2441,7 @@ static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
 	struct nvme_ns *ns;
 	int ret;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	if (list_empty(&ctrl->namespaces)) {
 		ret = -ENOTTY;
 		goto out_unlock;
@@ -2458,14 +2458,14 @@ static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
 	dev_warn(ctrl->device,
 		"using deprecated NVME_IOCTL_IO_CMD ioctl on the char device!\n");
 	kref_get(&ns->kref);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 
 	ret = nvme_user_cmd(ctrl, ns, argp);
 	nvme_put_ns(ns);
 	return ret;
 
 out_unlock:
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 	return ret;
 }
 
@@ -2894,7 +2894,7 @@ static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 {
 	struct nvme_ns *ns, *ret = NULL;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list) {
 		if (ns->head->ns_id == nsid) {
 			if (!kref_get_unless_zero(&ns->kref))
@@ -2905,7 +2905,7 @@ static struct nvme_ns *nvme_find_get_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 		if (ns->head->ns_id > nsid)
 			break;
 	}
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 	return ret;
 }
 
@@ -3016,9 +3016,9 @@ static void nvme_alloc_ns(struct nvme_ctrl *ctrl, unsigned nsid)
 
 	__nvme_revalidate_disk(disk, id);
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_write(&ctrl->namespaces_rwsem);
 	list_add_tail(&ns->list, &ctrl->namespaces);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_write(&ctrl->namespaces_rwsem);
 
 	nvme_get_ctrl(ctrl);
 
@@ -3072,9 +3072,9 @@ static void nvme_ns_remove(struct nvme_ns *ns)
 	list_del_rcu(&ns->siblings);
 	mutex_unlock(&ns->ctrl->subsys->lock);
 
-	mutex_lock(&ns->ctrl->namespaces_mutex);
+	down_write(&ns->ctrl->namespaces_rwsem);
 	list_del_init(&ns->list);
-	mutex_unlock(&ns->ctrl->namespaces_mutex);
+	up_write(&ns->ctrl->namespaces_rwsem);
 
 	synchronize_srcu(&ns->head->srcu);
 	nvme_mpath_check_last_path(ns);
@@ -3100,12 +3100,12 @@ static void nvme_remove_invalid_namespaces(struct nvme_ctrl *ctrl,
 	struct nvme_ns *ns, *next;
 	LIST_HEAD(rm_list);
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_write(&ctrl->namespaces_rwsem);
 	list_for_each_entry_safe(ns, next, &ctrl->namespaces, list) {
 		if (ns->head->ns_id > nsid)
 			list_move_tail(&ns->list, &rm_list);
 	}
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_write(&ctrl->namespaces_rwsem);
 
 	list_for_each_entry_safe(ns, next, &rm_list, list)
 		nvme_ns_remove(ns);
@@ -3185,9 +3185,9 @@ static void nvme_scan_work(struct work_struct *work)
 	}
 	nvme_scan_ns_sequential(ctrl, nn);
  done:
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_write(&ctrl->namespaces_rwsem);
 	list_sort(NULL, &ctrl->namespaces, ns_cmp);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_write(&ctrl->namespaces_rwsem);
 	kfree(id);
 }
 
@@ -3220,9 +3220,9 @@ void nvme_remove_namespaces(struct nvme_ctrl *ctrl)
 	if (ctrl->state == NVME_CTRL_DEAD)
 		nvme_kill_queues(ctrl);
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_write(&ctrl->namespaces_rwsem);
 	list_splice_init(&ctrl->namespaces, &ns_list);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_write(&ctrl->namespaces_rwsem);
 
 	list_for_each_entry_safe(ns, next, &ns_list, list)
 		nvme_ns_remove(ns);
@@ -3411,7 +3411,7 @@ int nvme_init_ctrl(struct nvme_ctrl *ctrl, struct device *dev,
 	ctrl->state = NVME_CTRL_NEW;
 	spin_lock_init(&ctrl->lock);
 	INIT_LIST_HEAD(&ctrl->namespaces);
-	mutex_init(&ctrl->namespaces_mutex);
+	init_rwsem(&ctrl->namespaces_rwsem);
 	ctrl->dev = dev;
 	ctrl->ops = ops;
 	ctrl->quirks = quirks;
@@ -3472,7 +3472,7 @@ void nvme_kill_queues(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 
 	/* Forcibly unquiesce queues to avoid blocking dispatch */
 	if (ctrl->admin_q)
@@ -3491,7 +3491,7 @@ void nvme_kill_queues(struct nvme_ctrl *ctrl)
 		/* Forcibly unquiesce queues to avoid blocking dispatch */
 		blk_mq_unquiesce_queue(ns->queue);
 	}
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_kill_queues);
 
@@ -3499,10 +3499,10 @@ void nvme_unfreeze(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list)
 		blk_mq_unfreeze_queue(ns->queue);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_unfreeze);
 
@@ -3510,13 +3510,13 @@ void nvme_wait_freeze_timeout(struct nvme_ctrl *ctrl, long timeout)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list) {
 		timeout = blk_mq_freeze_queue_wait_timeout(ns->queue, timeout);
 		if (timeout <= 0)
 			break;
 	}
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_wait_freeze_timeout);
 
@@ -3524,10 +3524,10 @@ void nvme_wait_freeze(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list)
 		blk_mq_freeze_queue_wait(ns->queue);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_wait_freeze);
 
@@ -3535,10 +3535,10 @@ void nvme_start_freeze(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list)
 		blk_freeze_queue_start(ns->queue);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_start_freeze);
 
@@ -3546,10 +3546,10 @@ void nvme_stop_queues(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list)
 		blk_mq_quiesce_queue(ns->queue);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_stop_queues);
 
@@ -3557,10 +3557,10 @@ void nvme_start_queues(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list)
 		blk_mq_unquiesce_queue(ns->queue);
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 EXPORT_SYMBOL_GPL(nvme_start_queues);
 

drivers/nvme/host/multipath.c

@@ -44,12 +44,12 @@ void nvme_kick_requeue_lists(struct nvme_ctrl *ctrl)
 {
 	struct nvme_ns *ns;
 
-	mutex_lock(&ctrl->namespaces_mutex);
+	down_read(&ctrl->namespaces_rwsem);
 	list_for_each_entry(ns, &ctrl->namespaces, list) {
 		if (ns->head->disk)
 			kblockd_schedule_work(&ns->head->requeue_work);
 	}
-	mutex_unlock(&ctrl->namespaces_mutex);
+	up_read(&ctrl->namespaces_rwsem);
 }
 
 static struct nvme_ns *__nvme_find_path(struct nvme_ns_head *head)

drivers/nvme/host/nvme.h

@@ -141,7 +141,7 @@ struct nvme_ctrl {
 	struct blk_mq_tag_set *tagset;
 	struct blk_mq_tag_set *admin_tagset;
 	struct list_head namespaces;
-	struct mutex namespaces_mutex;
+	struct rw_semaphore namespaces_rwsem;
 	struct device ctrl_device;
 	struct device *device;	/* char device */
 	struct cdev cdev;