On Tue, May 18, 2021 at 05:01:16PM +0100, John Garry wrote: > On 18/05/2021 13:51, John Garry wrote: > > n 18/05/2021 13:00, Ming Lei wrote: > > > > > 'Before 620K' could be caused by busy queue when batching > > > > > submission isn't > > > > > applied, so merge chance is increased. This patch applies batching > > > > > submission, so queue becomes not busy enough. > > > > > > > > > > BTW, what is the queue depth of sdev and can_queue of shost > > > > > for your hisilision SAS? > > > > sdev queue depth is 64 (see hisi_sas_slave_configure()) and host > > > > depth is > > > > 4096 - 96 (for reserved tags) = 4000 > > > OK, so queue depth should get IO merged if there are too many requests > > > queued. > > > > > > What is the same read test result without shared tags? still 430K? > > > > I never left a driver switch in place to disable it. > > > > I can forward-port "reply-map" support, which is not too difficult and I > > will let you know the result. > > The 'after' results are similar to without shared sbitmap, i.e using > reply-map: > > reply-map: > 450K (read), 430K IOPs (randread) OK, that is expected result. After shared sbitmap, IO merge gets improved when batching submission is bypassed, meantime IOPS of random IO drops because cpu utilization is increased. So that isn't a regression, let's live with this awkward situation, :-( Thanks, Ming
