On 8/19/20 5:20 PM, John Garry wrote: > Hi all, > > Here is v8 of the patchset. > > In this version of the series, we keep the shared sbitmap for driver tags, > and introduce changes to fix up the tag budgeting across request queues. > We also have a change to count requests per-hctx for when an elevator is > enabled, as an optimisation. I also dropped the debugfs changes - more on > that below. > > Some performance figures: > > Using 12x SAS SSDs on hisi_sas v3 hw. mq-deadline results are included, > but it is not always an appropriate scheduler to use. > > Tag depth 4000 (default) 260** > > Baseline (v5.9-rc1): > none sched: 2094K IOPS 513K > mq-deadline sched: 2145K IOPS 1336K > > Final, host_tagset=0 in LLDD *, ***: > none sched: 2120K IOPS 550K > mq-deadline sched: 2121K IOPS 1309K > > Final ***: > none sched: 2132K IOPS 1185 > mq-deadline sched: 2145K IOPS 2097 > > * this is relevant as this is the performance in supporting but not > enabling the feature > ** depth=260 is relevant as some point where we are regularly waiting for > tags to be available. Figures were are a bit unstable here. > *** Included "[PATCH V4] scsi: core: only re-run queue in > scsi_end_request() if device queue is busy" > > A copy of the patches can be found here: > https://github.com/hisilicon/kernel-dev/tree/private-topic-blk-mq-shared-tags-v8 > > The hpsa patch depends on: > https://lore.kernel.org/linux-scsi/20200430131904.5847-1-hare@xxxxxxx/ > > And the smartpqi patch is not to be accepted. > > Comments (and testing) welcome, thanks! > > Differences to v7: > - Add null_blk and scsi_debug support > - Drop debugfs tags patch - it's too difficult to be the same between > hostwide and non-hostwide, as discussed: > https://lore.kernel.org/linux-scsi/1591810159-240929-1-git-send-email-john.garry@xxxxxxxxxx/T/#mb3eb462d8be40273718505989abd12f8228c15fd > And from commit 6bf0eb550452 ("sbitmap: Consider cleared bits in > sbitmap_bitmap_show()"), I guess not many used this anyway... > - Add elevator per-hctx request count for optimisation > - Break up "blk-mq: rename blk_mq_update_tag_set_depth()" into 2x patches > - Pass flags for avoid per-hq queue tags init/free for hostwide tags > - Add Don's reviewed-tag and tested-by tags to appropiate patches > - (@Don, please let me know if issue with how I did this) > - Add "scsi: core: Show nr_hw_queues in sysfs" > - Rework megaraid SAS patch to have module param (Kashyap) > - rebase > > V7 is here for more info: > https://lore.kernel.org/linux-scsi/1591810159-240929-1-git-send-email-john.garry@xxxxxxxxxx/T/#t > > Hannes Reinecke (5): > blk-mq: Rename blk_mq_update_tag_set_depth() > blk-mq: Free tags in blk_mq_init_tags() upon error > scsi: Add host and host template flag 'host_tagset' > hpsa: enable host_tagset and switch to MQ > smartpqi: enable host tagset > > John Garry (10): > blk-mq: Pass flags for tag init/free > blk-mq: Use pointers for blk_mq_tags bitmap tags > blk-mq: Facilitate a shared sbitmap per tagset > blk-mq: Relocate hctx_may_queue() > blk-mq: Record nr_active_requests per queue for when using shared > sbitmap > blk-mq: Record active_queues_shared_sbitmap per tag_set for when using > shared sbitmap > null_blk: Support shared tag bitmap > scsi: core: Show nr_hw_queues in sysfs > scsi: hisi_sas: Switch v3 hw to MQ > scsi: scsi_debug: Support host tagset > > Kashyap Desai (2): > blk-mq, elevator: Count requests per hctx to improve performance > scsi: megaraid_sas: Added support for shared host tagset for > cpuhotplug > > Ming Lei (1): > blk-mq: Rename BLK_MQ_F_TAG_SHARED as BLK_MQ_F_TAG_QUEUE_SHARED > Now that Jens merged the block bits in his tree, wouldn't it be better to re-send the SCSI bits only, thereby avoiding a potential merge error later on? Cheers, Hannes -- Dr. Hannes Reinecke Kernel Storage Architect hare@xxxxxxx +49 911 74053 688 SUSE Software Solutions Germany GmbH, Maxfeldstr. 5, 90409 Nürnberg HRB 36809 (AG Nürnberg), GF: Felix Imendörffer
