Hi, scsi uses one global atomic variable to track queue depth for each LUN/request queue. This way can't scale well when there is lots of CPU cores and the disk is very fast. Broadcom guys has complained that their high end HBA can't reach top performance because .device_busy is operated in IO path. Replace the atomic variable sdev->device_busy with sbitmap for tracking scsi device queue depth. Test on scsi_debug shows this way improve IOPS > 20%. Meantime the IOPS difference is just ~1% compared with bypassing .device_busy on scsi_debug via patches[1] The 1st 6 patches moves percpu allocation hint into sbitmap, since the improvement by doing percpu allocation hint on sbitmap is observable. Meantime export helpers for SCSI. Patch 7 and 8 prepares for the conversion by returning budget token from .get_budget callback, meantime passes the budget token to driver via 'struct blk_mq_queue_data' in .queue_rq(). The last two patches changes SCSI for switching to track device queue depth via sbitmap. Broadcom Guys, please test this patchset and see if expected performance can be reached. Please comment and review! thanks, Ming [1] https://lore.kernel.org/linux-block/20200119071432.18558-6-ming.lei@xxxxxxxxxx/ Ming Lei (10): sbitmap: maintain allocation round_robin in sbitmap sbitmap: add helpers for updating allocation hint sbitmap: remove sbitmap_clear_bit_unlock sbitmap: move allocation hint into sbitmap sbitmap: export sbitmap_weight sbitmap: add helper of sbitmap_calculate_shift blk-mq: return budget token from .get_budget callback blk-mq: pass budget token to dirver via blk_mq_queue_data scsi: add scsi_device_busy() to read sdev->device_busy scsi: replace sdev->device_busy with sbitmap block/blk-mq-sched.c | 20 ++- block/blk-mq.c | 37 +++-- block/blk-mq.h | 11 +- block/kyber-iosched.c | 3 +- drivers/dma/idxd/device.c | 2 +- drivers/dma/idxd/submit.c | 2 +- drivers/scsi/mpt3sas/mpt3sas_scsih.c | 2 +- drivers/scsi/scsi.c | 2 + drivers/scsi/scsi_lib.c | 47 +++--- drivers/scsi/scsi_priv.h | 1 + drivers/scsi/scsi_scan.c | 21 ++- drivers/scsi/scsi_sysfs.c | 4 +- drivers/scsi/sg.c | 2 +- include/linux/blk-mq.h | 5 +- include/linux/sbitmap.h | 84 +++++++---- include/scsi/scsi_cmnd.h | 2 + include/scsi/scsi_device.h | 8 +- lib/sbitmap.c | 213 +++++++++++++++------------ 18 files changed, 285 insertions(+), 181 deletions(-) Cc: Omar Sandoval <osandov@xxxxxx> Cc: Sathya Prakash <sathya.prakash@xxxxxxxxxxxx> Cc: Chaitra P B <chaitra.basappa@xxxxxxxxxxxx> Cc: Suganath Prabu Subramani <suganath-prabu.subramani@xxxxxxxxxxxx> Cc: Kashyap Desai <kashyap.desai@xxxxxxxxxxxx> Cc: Sumit Saxena <sumit.saxena@xxxxxxxxxxxx> Cc: Shivasharan S <shivasharan.srikanteshwara@xxxxxxxxxxxx> Cc: Ewan D. Milne <emilne@xxxxxxxxxx> Cc: Hannes Reinecke <hare@xxxxxxx> Cc: Bart Van Assche <bart.vanassche@xxxxxxx> -- 2.20.1
