Commit 84da7acc authored by Ming Lei's avatar Ming Lei Committed by Jens Axboe

block: avoid double io accounting for flush request

For flush request, rq->end_io() may be called two times, one is from
timeout handling(blk_mq_check_expired()), another is from normal
completion(__blk_mq_end_request()).

Move blk_account_io_flush() after flush_rq->ref drops to zero, so
io accounting can be done just once for flush request.

Fixes: b6866318 ("block: add iostat counters for flush requests")
Reviewed-by: default avatarBart Van Assche <bvanassche@acm.org>
Reviewed-by: default avatarChristoph Hellwig <hch@lst.de>
Tested-by: default avatarJohn Garry <john.garry@huawei.com>
Signed-off-by: default avatarMing Lei <ming.lei@redhat.com>
Link: https://lore.kernel.org/r/20210511152236.763464-2-ming.lei@redhat.comSigned-off-by: default avatarJens Axboe <axboe@kernel.dk>
parent 8c390ff9
...@@ -219,8 +219,6 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error) ...@@ -219,8 +219,6 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error)
unsigned long flags = 0; unsigned long flags = 0;
struct blk_flush_queue *fq = blk_get_flush_queue(q, flush_rq->mq_ctx); struct blk_flush_queue *fq = blk_get_flush_queue(q, flush_rq->mq_ctx);
blk_account_io_flush(flush_rq);
/* release the tag's ownership to the req cloned from */ /* release the tag's ownership to the req cloned from */
spin_lock_irqsave(&fq->mq_flush_lock, flags); spin_lock_irqsave(&fq->mq_flush_lock, flags);
...@@ -230,6 +228,7 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error) ...@@ -230,6 +228,7 @@ static void flush_end_io(struct request *flush_rq, blk_status_t error)
return; return;
} }
blk_account_io_flush(flush_rq);
/* /*
* Flush request has to be marked as IDLE when it is really ended * Flush request has to be marked as IDLE when it is really ended
* because its .end_io() is called from timeout code path too for * because its .end_io() is called from timeout code path too for
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment