Re: [PATCH v3 3/3] scsi: target: Move LUN stats to per CPU

[Hannes Reinecke <hare@suse.com>]

On 9/18/25 00:12, Mike Christie wrote:
> The atomic use in the main I/O path is causing perf issues when using
> higher performance backend devices and multiple queues (more than
> 10 when using vhost-scsi) like with this fio workload:
> 
> [global]
> bs=4K
> iodepth=128
> direct=1
> ioengine=libaio
> group_reporting
> time_based
> runtime=120
> name=standard-iops
> rw=randread
> numjobs=16
> cpus_allowed=0-15
> 
> To fix this issue, this moves the LUN stats to per CPU.
> 
> Note: I forgot to include this patch with the delayed/ordered per CPU
> tracking and per device/device entry per CPU stats. With this patch you
> get the full 33% improvements when using fast backends, multiple queues
> and multiple IO submiters.
> 
> Signed-off-by: Mike Christie <michael.christie@oracle.com>
> Reviewed-by: Dmitry Bogdanov <d.bogdanov@yadro.com>
> ---
>   drivers/target/target_core_device.c          |  1 +
>   drivers/target/target_core_fabric_configfs.c |  2 +-
>   drivers/target/target_core_internal.h        |  1 +
>   drivers/target/target_core_stat.c            | 67 +++++++-------------
>   drivers/target/target_core_tpg.c             | 23 ++++++-
>   drivers/target/target_core_transport.c       | 22 +++++--
>   include/target/target_core_base.h            |  8 +--
>   7 files changed, 65 insertions(+), 59 deletions(-)
> 
Ho-hum.

That only works if both submission and completion paths do run on the
_same_ cpu. Are we sure that they do?

Cheers,

Hannes
-- 
Dr. Hannes Reinecke                  Kernel Storage Architect
hare@suse.com                               +49 911 74053 688
SUSE Software Solutions GmbH, Frankenstr. 146, 90461 Nürnberg
HRB 36809 (AG Nürnberg), GF: I. Totev, A. McDonald, W. Knoblich