dm: run queue on re-queue
authorJunichi Nomura <j-nomura@ce.jp.nec.com>
Tue, 26 May 2015 08:25:54 +0000 (08:25 +0000)
committerMike Snitzer <snitzer@redhat.com>
Tue, 26 May 2015 13:57:36 +0000 (09:57 -0400)
Without kicking queue, requeued request may stay forever in
the queue if there are no other I/O activities to the device.

The original error had been in v2.6.39 with commit 7eaceaccab5f
("block: remove per-queue plugging"), which replaced conditional
plugging by periodic runqueue.

Commit 9d1deb83d489 in v4.1-rc1 removed the periodic runqueue
and the problem started to manifest.

Fixes: 9d1deb83d489 ("dm: don't schedule delayed run of the queue if nothing to do")
Signed-off-by: Jun'ichi Nomura <j-nomura@ce.jp.nec.com>
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
drivers/md/dm.c

index a930b72314ac985da702f8b47a8054a75b2e2ba8..0bf79a0bad374291db6a22f45c1cad70ce7188a8 100644 (file)
@@ -1164,6 +1164,7 @@ static void old_requeue_request(struct request *rq)
 
        spin_lock_irqsave(q->queue_lock, flags);
        blk_requeue_request(q, rq);
+       blk_run_queue_async(q);
        spin_unlock_irqrestore(q->queue_lock, flags);
 }