diff options
| author | Robert Elliott <elliott@hp.com> | 2014-05-20 16:46:26 -0500 |
|---|---|---|
| committer | Jens Axboe <axboe@fb.com> | 2014-05-20 15:49:03 -0600 |
| commit | da41a589f52464e24ddefe76814ee35bfb07950c (patch) | |
| tree | e3270f36a1a5766e2433b70b0eef4e2205096075 /block/blk-core.c | |
| parent | eba7176826ddab1d04c51bb2d5f2bbf22865444c (diff) | |
| download | op-kernel-dev-da41a589f52464e24ddefe76814ee35bfb07950c.zip op-kernel-dev-da41a589f52464e24ddefe76814ee35bfb07950c.tar.gz | |
blk-mq: Micro-optimize blk_queue_nomerges() check
In blk_mq_make_request(), do the blk_queue_nomerges() check
outside the call to blk_attempt_plug_merge() to eliminate
function call overhead when nomerges=2 (disabled)
Signed-off-by: Robert Elliott <elliott@hp.com>
Signed-off-by: Jens Axboe <axboe@fb.com>
Diffstat (limited to 'block/blk-core.c')
| -rw-r--r-- | block/blk-core.c | 8 |
1 files changed, 4 insertions, 4 deletions
diff --git a/block/blk-core.c b/block/blk-core.c index fe81e19..5b6f768 100644 --- a/block/blk-core.c +++ b/block/blk-core.c @@ -1471,6 +1471,8 @@ bool bio_attempt_front_merge(struct request_queue *q, struct request *req, * added on the elevator at this point. In addition, we don't have * reliable access to the elevator outside queue lock. Only check basic * merging parameters without querying the elevator. + * + * Caller must ensure !blk_queue_nomerges(q) beforehand. */ bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio, unsigned int *request_count) @@ -1480,9 +1482,6 @@ bool blk_attempt_plug_merge(struct request_queue *q, struct bio *bio, bool ret = false; struct list_head *plug_list; - if (blk_queue_nomerges(q)) - goto out; - plug = current->plug; if (!plug) goto out; @@ -1561,7 +1560,8 @@ void blk_queue_bio(struct request_queue *q, struct bio *bio) * Check if we can merge with the plugged list before grabbing * any locks. */ - if (blk_attempt_plug_merge(q, bio, &request_count)) + if (!blk_queue_nomerges(q) && + blk_attempt_plug_merge(q, bio, &request_count)) return; spin_lock_irq(q->queue_lock); |
