On 11/17/21 7:17 PM, Karan Tilak Kumar (kartilak) wrote: > Hello btrace community, > > I came across blktrace and blkparse tools as a part of my learning on blk-mq. > I'm trying to test some code in my driver and I want to test a few things by turning multiqueue on and off, and compare results. > I'm running IOs using medusa on my multipath device /dev/dm-2 > > When I turn off mq, I still see IOs queued from each CPU: > > CPU0 (dm-2): > Reads Queued: 1,980, 199,438KiB Writes Queued: 1,344, 172,032KiB > Read Dispatches: 3,429, 199,894KiB Write Dispatches: 2,688, 172,032KiB > Reads Requeued: 0 Writes Requeued: 0 > Reads Completed: 6,826, 199,130KiB Writes Completed: 5,376, 172,032KiB > Read Merges: 0, 0KiB Write Merges: 0, 0KiB > Read depth: 70 Write depth: 64 > IO unplugs: 3,177 Timer unplugs: 0 > CPU1 (dm-2): > Reads Queued: 2,101, 247,313KiB Writes Queued: 1,856, 237,568KiB > Read Dispatches: 3,986, 247,177KiB Write Dispatches: 3,709, 237,376KiB > Reads Requeued: 0 Writes Requeued: 0 > Reads Completed: 7,975, 247,177KiB Writes Completed: 7,545, 241,344KiB > Read Merges: 0, 0KiB Write Merges: 0, 0KiB > Read depth: 70 Write depth: 64 > IO unplugs: 3,899 Timer unplugs: 0 > CPU2 (dm-2): > Reads Queued: 1,907, 224,925KiB Writes Queued: 1,760, 225,280KiB > Read Dispatches: 3,635, 224,793KiB Write Dispatches: 3,524, 225,536KiB > Reads Requeued: 0 Writes Requeued: 0 > Reads Completed: 7,273, 224,793KiB Writes Completed: 7,044, 225,536KiB > Read Merges: 0, 0KiB Write Merges: 0, 0KiB > Read depth: 70 Write depth: 64 > IO unplugs: 3,625 Timer unplugs: 1 > CPU3 (dm-2): > Reads Queued: 1,572, 173,940KiB Writes Queued: 1,408, 180,224KiB > Read Dispatches: 2,885, 173,753KiB Write Dispatches: 2,815, 180,160KiB > Reads Requeued: 0 Writes Requeued: 0 > Reads Completed: 5,766, 173,541KiB Writes Completed: 5,631, 180,160KiB > Read Merges: 0, 0KiB Write Merges: 0, 0KiB > Read depth: 70 Write depth: 64 > IO unplugs: 2,922 Timer unplugs: 0 > > > #cat /sys/module/scsi_mod/parameters/use_blk_mq > N > #cat /sys/module/dm_mod/parameters/use_blk_mq > N > > > When multiqueue is off, we don't expect IOs from all CPUs. Is this correct? > I've tried running medusa with one thread, and with eight threads on separate occasions, and I see IOs from all CPUs each time. > Is this expected? Please advise. Yes this is expected, non blk-mq will still submit from any CPU that happens to be issuing the IO. -- Jens Axboe
