> Il giorno 02 set 2017, alle ore 17:17, Ming Lei <ming.lei@xxxxxxxxxx> ha scritto: > > Hi, > > In Red Hat internal storage test wrt. blk-mq scheduler, we > found that I/O performance is much bad with mq-deadline, especially > about sequential I/O on some multi-queue SCSI devcies(lpfc, qla2xxx, > SRP...) > > Turns out one big issue causes the performance regression: requests > are still dequeued from sw queue/scheduler queue even when ldd's > queue is busy, so I/O merge becomes quite difficult to make, then > sequential IO degrades a lot. > > The 1st five patches improve this situation, and brings back > some performance loss. > > Patch 6 ~ 7 uses q->queue_depth as hint for setting up > scheduler queue depth. > > Patch 8 ~ 15 improve bio merge via hash table in sw queue, > which makes bio merge more efficient than current approch > in which only the last 8 requests are checked. Since patch > 6~14 converts to the scheduler way of dequeuing one request > from sw queue one time for SCSI device, and the times of > acquring ctx->lock is increased, and merging bio via hash > table decreases holding time of ctx->lock and should eliminate > effect from patch 14. > > With this changes, SCSI-MQ sequential I/O performance is > improved much, Paolo reported that mq-deadline performance > improved much[2] in his dbench test wrt V2. Also performanc > improvement on lpfc/qla2xx was observed with V1.[1] > > Also Bart worried that this patchset may affect SRP, so provide > test data on SCSI SRP this time: > > - fio(libaio, bs:4k, dio, queue_depth:64, 64 jobs) > - system(16 cores, dual sockets, mem: 96G) > > |v4.13-rc6+* |v4.13-rc6+ | patched v4.13-rc6+ > ----------------------------------------------------- > IOPS(K) | DEADLINE | NONE | NONE > ----------------------------------------------------- > read | 587.81 | 511.96 | 518.51 > ----------------------------------------------------- > randread | 116.44 | 142.99 | 142.46 > ----------------------------------------------------- > write | 580.87 | 536.4 | 582.15 > ----------------------------------------------------- > randwrite | 104.95 | 124.89 | 123.99 > ----------------------------------------------------- > > > |v4.13-rc6+ |v4.13-rc6+ | patched v4.13-rc6+ > ----------------------------------------------------- > IOPS(K) | DEADLINE |MQ-DEADLINE |MQ-DEADLINE > ----------------------------------------------------- > read | 587.81 | 158.7 | 450.41 > ----------------------------------------------------- > randread | 116.44 | 142.04 | 142.72 > ----------------------------------------------------- > write | 580.87 | 136.61 | 569.37 > ----------------------------------------------------- > randwrite | 104.95 | 123.14 | 124.36 > ----------------------------------------------------- > > *: v4.13-rc6+ means v4.13-rc6 with block for-next > > > Please consider to merge to V4.4. > > [1] http://marc.info/?l=linux-block&m=150151989915776&w=2 > [2] https://marc.info/?l=linux-block&m=150217980602843&w=2 > > V4: > - add Reviewed-by tag > - some trival change: typo fix in commit log or comment, > variable name, no actual functional change > > V3: > - totally round robin for picking req from ctx, as suggested > by Bart > - remove one local variable in __sbitmap_for_each_set() > - drop patches of single dispatch list, which can improve > performance on mq-deadline, but cause a bit degrade on > none because all hctxs need to be checked after ->dispatch > is flushed. Will post it again once it is mature. > - rebase on v4.13-rc6 with block for-next > > V2: > - dequeue request from sw queues in round roubin's style > as suggested by Bart, and introduces one helper in sbitmap > for this purpose > - improve bio merge via hash table from sw queue > - add comments about using DISPATCH_BUSY state in lockless way, > simplifying handling on busy state, > - hold ctx->lock when clearing ctx busy bit as suggested > by Bart > > Tested-by: Paolo Valente <paolo.valente@xxxxxxxxxx> > Ming Lei (14): > blk-mq-sched: fix scheduler bad performance > sbitmap: introduce __sbitmap_for_each_set() > blk-mq: introduce blk_mq_dispatch_rq_from_ctx() > blk-mq-sched: move actual dispatching into one helper > blk-mq-sched: improve dispatching from sw queue > blk-mq-sched: don't dequeue request until all in ->dispatch are > flushed > blk-mq-sched: introduce blk_mq_sched_queue_depth() > blk-mq-sched: use q->queue_depth as hint for q->nr_requests > block: introduce rqhash helpers > block: move actual bio merge code into __elv_merge > block: add check on elevator for supporting bio merge via hashtable > from blk-mq sw queue > block: introduce .last_merge and .hash to blk_mq_ctx > blk-mq-sched: refactor blk_mq_sched_try_merge() > blk-mq: improve bio merge from blk-mq sw queue > > block/blk-mq-debugfs.c | 1 + > block/blk-mq-sched.c | 186 +++++++++++++++++++++++++++++++----------------- > block/blk-mq-sched.h | 23 ++++++ > block/blk-mq.c | 93 +++++++++++++++++++++++- > block/blk-mq.h | 7 ++ > block/blk-settings.c | 2 + > block/blk.h | 55 ++++++++++++++ > block/elevator.c | 93 ++++++++++++++---------- > include/linux/blk-mq.h | 3 + > include/linux/sbitmap.h | 54 ++++++++++---- > 10 files changed, 399 insertions(+), 118 deletions(-) > > -- > 2.9.5 >
