From: Ming Lei ming.lei@redhat.com
commit 0a67b5a926e63ff5492c3c675eab5900580d056d upstream.
cecf5d87ff20 ("block: split .sysfs_lock into two locks") starts to release & actuire sysfs_lock again during switching elevator. So it isn't enough to prevent switching elevator from happening by simply clearing QUEUE_FLAG_REGISTERED with holding sysfs_lock, because in-progress switch still can move on after re-acquiring the lock, meantime the flag of QUEUE_FLAG_REGISTERED won't get checked.
Fixes this issue by checking 'q->elevator' directly & locklessly after q->kobj is removed in blk_unregister_queue(), this way is safe because q->elevator can't be changed at that time.
Fixes: cecf5d87ff20 ("block: split .sysfs_lock into two locks") Cc: Christoph Hellwig hch@infradead.org Cc: Hannes Reinecke hare@suse.com Cc: Greg KH gregkh@linuxfoundation.org Cc: Mike Snitzer snitzer@redhat.com Cc: Bart Van Assche bvanassche@acm.org Signed-off-by: Ming Lei ming.lei@redhat.com Signed-off-by: Jens Axboe axboe@kernel.dk Signed-off-by: Jack Wang jinpu.wang@cloud.ionos.com Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- block/blk-sysfs.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
--- a/block/blk-sysfs.c +++ b/block/blk-sysfs.c @@ -977,7 +977,6 @@ EXPORT_SYMBOL_GPL(blk_register_queue); void blk_unregister_queue(struct gendisk *disk) { struct request_queue *q = disk->queue; - bool has_elevator;
if (WARN_ON(!q)) return; @@ -993,7 +992,6 @@ void blk_unregister_queue(struct gendisk */ mutex_lock(&q->sysfs_lock); blk_queue_flag_clear(QUEUE_FLAG_REGISTERED, q); - has_elevator = !!q->elevator; mutex_unlock(&q->sysfs_lock);
mutex_lock(&q->sysfs_dir_lock); @@ -1009,7 +1007,11 @@ void blk_unregister_queue(struct gendisk blk_trace_remove_sysfs(disk_to_dev(disk));
mutex_lock(&q->sysfs_lock); - if (q->request_fn || has_elevator) + /* + * q->kobj has been removed, so it is safe to check if elevator + * exists without holding q->sysfs_lock. + */ + if (q->request_fn || q->elevator) elv_unregister_queue(q); mutex_unlock(&q->sysfs_lock); mutex_unlock(&q->sysfs_dir_lock);