On 10/5/21 7:34 AM, John Garry wrote:

On 05/10/2021 13:35, Jens Axboe wrote:

quoted

quoted

Baseline is 1b2d1439fc25 (block/for-next) Merge branch 'for-5.16/io_uring'
into for-next

Let's get this queued up for testing, thanks John.

Cheers, appreciated

@Kashyap, You mentioned that when testing you saw a performance 
regression from v5.11 -> v5.12 - any idea on that yet? Can you describe 
the scenario, like IO scheduler and how many disks and the type? Does 
disabling host_tagset_enable restore performance?

FWIW, I ran my usual peak testing on this and didn't observe any regressions.
Caveat being that a) no scheduler is involved, and b) no shared tags. But
at least that fast case/path is fine.

-- 
Jens Axboe