On 2023-03-27 03:43, John Garry wrote:
It's easy to get scsi_debug to error on throughput testing when we have multiple shosts: $ lsscsi [7:0:0:0] disk Linux scsi_debug 0191 [0:0:0:0] disk Linux scsi_debug 0191 $ fio --filename=/dev/sda --filename=/dev/sdb --direct=1 --rw=read --bs=4k --iodepth=256 --runtime=60 --numjobs=40 --time_based --name=jpg --eta-newline=1 --readonly --ioengine=io_uring --hipri --exitall_on_error jpg: (g=0): rw=read, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=io_uring, iodepth=256 ... fio-3.28 Starting 40 processes [ 27.521809] hrtimer: interrupt took 33067 ns [ 27.904660] sd 7:0:0:0: [sdb] tag#171 FAILED Result: hostbyte=DID_ABORT driverbyte=DRIVER_OK cmd_age=0s [ 27.904660] sd 0:0:0:0: [sda] tag#58 FAILED Result: hostbyte=DID_ABORT driverbyte=DRIVER_OK cmd_age=0s fio: io_u error [ 27.904667] sd 0:0:0:0: [sda] tag#58 CDB: Read(10) 28 00 00 00 27 00 00 01 18 00 on file /dev/sda[ 27.904670] sd 0:0:0:0: [sda] tag#62 FAILED Result: hostbyte=DID_ABORT driverbyte=DRIVER_OK cmd_age=0s The issue is related to how the driver manages submit queues and tags. A single array of submit queues - sdebug_q_arr - with its own set of tags is shared among all shosts. As such, for occasions when we have more than one shost it is possible to overload the submit queues and run out of tags. The struct sdebug_queue is to manage tags and hold the associated queued command entry pointer (for that tag). Since the tagset iters are now used for functions like sdebug_blk_mq_poll(), there is no need to manage these queues. Indeed, blk-mq already provides what we need for managing tags and queues. Drop sdebug_queue and all its usage in the driver. Signed-off-by: John Garry <john.g.garry@xxxxxxxxxx>
Acked-by: Douglas Gilbert <dgilbert@xxxxxxxxxxxx> Thanks.
