summaryrefslogtreecommitdiff
path: root/block/blk-mq-sched.c
diff options
context:
space:
mode:
authorMing Lei <ming.lei@redhat.com>2017-05-26 19:53:19 +0800
committerJens Axboe <axboe@fb.com>2017-05-26 14:12:03 -0600
commitab42f35d9cb5ac49b5a2a11f940e74f58f207280 (patch)
treed3fedd45516727f64f0efa1f494755cbabd8a6cc /block/blk-mq-sched.c
parentcdbe0206783b4e86abb0150303f137e448b5e531 (diff)
downloadlwn-ab42f35d9cb5ac49b5a2a11f940e74f58f207280.tar.gz
lwn-ab42f35d9cb5ac49b5a2a11f940e74f58f207280.zip
blk-mq: merge bio into sw queue before plugging
Before blk-mq is introduced, I/O is merged to elevator before being putted into plug queue, but blk-mq changed the order and makes merging to sw queue basically impossible. Then it is observed that throughput of sequential I/O is degraded about 10%~20% on virtio-blk in the test[1] if mq-deadline isn't used. This patch moves the bio merging per sw queue before plugging, like what blk_queue_bio() does, and the performance regression is fixed under this situation. [1]. test script: sudo fio --direct=1 --size=128G --bsrange=4k-4k --runtime=40 --numjobs=16 --ioengine=libaio --iodepth=64 --group_reporting=1 --filename=/dev/vdb --name=virtio_blk-test-$RW --rw=$RW --output-format=json RW=read or write Reviewed-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Ming Lei <ming.lei@redhat.com> Signed-off-by: Jens Axboe <axboe@fb.com>
Diffstat (limited to 'block/blk-mq-sched.c')
0 files changed, 0 insertions, 0 deletions