[PATCH] Correct unplugs on nr_queued

From: Jens Axboe <axboe@suse.de>

There's a small discrepancy in when we decide to unplug a queue based on
q->unplug_thresh.  Basically it doesn't work for tagged queues, since
q->rq.count[READ] + q->rq.count[WRITE] is just the number of allocated
requests, not the number of requests stuck in the io scheduler.  We could
just change the nr_queued == to a nr_queued >=, however that is still
suboptimal.

This patch adds accounting for requests that have been dequeued from the io
scheduler, but not freed yet.  These are q->in_flight.  allocated_requests
- q->in_flight == requests_in_scheduler.  So the condition correctly
becomes

	if (requests_in_scheduler == q->unplug_thresh)

instead.  I did a quick round of testing, and for dbench on a SCSI disk the
number of timer induced unplugs was reduced from 13 to 5 :-).  Not a huge
number, but there might be cases where it's more significant.  Either way,
it gets ->unplug_thresh always right, which the old logic didn't.

1 files changed, 5 insertions, 0 deletions

diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
index 572f96e6940a..44c722d4b67b 100644
--- a/include/linux/blkdev.h
+++ b/include/linux/blkdev.h
@@ -348,6 +348,8 @@ struct request_queue
 
 	atomic_t		refcnt;
 
+	unsigned int		in_flight;
+
 	/*
 	 * sg stuff
 	 */
@@ -377,6 +379,9 @@ struct request_queue
 #define blk_fs_request(rq)	((rq)->flags & REQ_CMD)
 #define blk_pc_request(rq)	((rq)->flags & REQ_BLOCK_PC)
 #define blk_noretry_request(rq)	((rq)->flags & REQ_FAILFAST)
+#define blk_rq_started(rq)	((rq)->flags & REQ_STARTED)
+
+#define blk_account_rq(rq)	(blk_rq_started(rq) && blk_fs_request(rq))
 
 #define blk_pm_suspend_request(rq)	((rq)->flags & REQ_PM_SUSPEND)
 #define blk_pm_resume_request(rq)	((rq)->flags & REQ_PM_RESUME)