block: Protect less code with sysfs_lock in blk_{un,}register_queue()

The __blk_mq_register_dev(), blk_mq_unregister_dev(),
elv_register_queue() and elv_unregister_queue() calls need to be
protected with sysfs_lock but other code in these functions not.
Hence protect only this code with sysfs_lock. This patch fixes a
locking inversion issue in blk_unregister_queue() and also in an
error path of blk_register_queue(): it is not allowed to hold
sysfs_lock around the kobject_del(&q->kobj) call.

Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Jens Axboe <axboe@kernel.dk>

1 files changed, 28 insertions, 9 deletions

diff --git a/block/blk-sysfs.c b/block/blk-sysfs.c
index 4a6a40ffd78e..cbea895a5547 100644
--- a/block/blk-sysfs.c
+++ b/block/blk-sysfs.c
@@ -853,6 +853,10 @@ struct kobj_type blk_queue_ktype = {
         .release        = blk_release_queue,
 };
 
+/**
+ * blk_register_queue - register a block layer queue with sysfs
+ * @disk: Disk of which the request queue should be registered with sysfs.
+ */
 int blk_register_queue(struct gendisk *disk)
 {
         int ret;
@@ -909,11 +913,12 @@ int blk_register_queue(struct gendisk *disk)
         if (q->request_fn || (q->mq_ops && q->elevator)) {
                 ret = elv_register_queue(q);
                 if (ret) {
+                        mutex_unlock(&q->sysfs_lock);
                         kobject_uevent(&q->kobj, KOBJ_REMOVE);
                         kobject_del(&q->kobj);
                         blk_trace_remove_sysfs(dev);
                         kobject_put(&dev->kobj);
-                        goto unlock;
+                        return ret;
                 }
         }
         ret = 0;
@@ -923,6 +928,13 @@ unlock:
 }
 EXPORT_SYMBOL_GPL(blk_register_queue);
 
+/**
+ * blk_unregister_queue - counterpart of blk_register_queue()
+ * @disk: Disk of which the request queue should be unregistered from sysfs.
+ *
+ * Note: the caller is responsible for guaranteeing that this function is called
+ * after blk_register_queue() has finished.
+ */
 void blk_unregister_queue(struct gendisk *disk)
 {
         struct request_queue *q = disk->queue;
@@ -935,8 +947,9 @@ void blk_unregister_queue(struct gendisk *disk)
                 return;
 
         /*
-         * Protect against the 'queue' kobj being accessed
-         * while/after it is removed.
+         * Since sysfs_remove_dir() prevents adding new directory entries
+         * before removal of existing entries starts, protect against
+         * concurrent elv_iosched_store() calls.
          */
         mutex_lock(&q->sysfs_lock);
 
@@ -944,18 +957,24 @@ void blk_unregister_queue(struct gendisk *disk)
         queue_flag_clear(QUEUE_FLAG_REGISTERED, q);
         spin_unlock_irq(q->queue_lock);
 
-        wbt_exit(q);
-
+        /*
+         * Remove the sysfs attributes before unregistering the queue data
+         * structures that can be modified through sysfs.
+         */
         if (q->mq_ops)
                 blk_mq_unregister_dev(disk_to_dev(disk), q);
-
-        if (q->request_fn || (q->mq_ops && q->elevator))
-                elv_unregister_queue(q);
+        mutex_unlock(&q->sysfs_lock);
 
         kobject_uevent(&q->kobj, KOBJ_REMOVE);
         kobject_del(&q->kobj);
         blk_trace_remove_sysfs(disk_to_dev(disk));
-        kobject_put(&disk_to_dev(disk)->kobj);
 
+        wbt_exit(q);
+
+        mutex_lock(&q->sysfs_lock);
+        if (q->request_fn || (q->mq_ops && q->elevator))
+                elv_unregister_queue(q);
         mutex_unlock(&q->sysfs_lock);
+
+        kobject_put(&disk_to_dev(disk)->kobj);
 }