[RFC PATCH] blk-mq: fix potential I/O hang caused by batch wakeup

[Yang Yang]

The depth is 62, and the wake_batch is 8. In the following situation,
the task would hang forever.

  t1:                 t2:                          t3:
  blk_mq_get_tag      .                            .
  io_schedule         .                            .
                      elevator_switch              .
                      blk_mq_freeze_queue          .
                      blk_freeze_queue_start       .
                      blk_mq_freeze_queue_wait     .
                                                   blk_mq_submit_bio
                                                   __bio_queue_enter

Fix this issue by waking up all the waiters sleeping on tags after
freezing the queue.

Signed-off-by: Yang Yang <yang.yang@xxxxxxxx>
---
 block/blk-core.c | 2 --
 block/blk-mq.c   | 4 +++-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/block/blk-core.c b/block/blk-core.c
index a16b5abdbbf5..e1eacfad6e5b 100644
--- a/block/blk-core.c
+++ b/block/blk-core.c
@@ -298,8 +298,6 @@ void blk_queue_start_drain(struct request_queue *q)
 	 * prevent I/O from crossing blk_queue_enter().
 	 */
 	blk_freeze_queue_start(q);
-	if (queue_is_mq(q))
-		blk_mq_wake_waiters(q);
 	/* Make blk_queue_enter() reexamine the DYING flag. */
 	wake_up_all(&q->mq_freeze_wq);
 }
diff --git a/block/blk-mq.c b/block/blk-mq.c
index 4ecb9db62337..9eb3139e713a 100644
--- a/block/blk-mq.c
+++ b/block/blk-mq.c
@@ -125,8 +125,10 @@ void blk_freeze_queue_start(struct request_queue *q)
 	if (++q->mq_freeze_depth == 1) {
 		percpu_ref_kill(&q->q_usage_counter);
 		mutex_unlock(&q->mq_freeze_lock);
-		if (queue_is_mq(q))
+		if (queue_is_mq(q)) {
+			blk_mq_wake_waiters(q);
 			blk_mq_run_hw_queues(q, false);
+		}
 	} else {
 		mutex_unlock(&q->mq_freeze_lock);
 	}
-- 
2.34.1