Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1636379

Re: [RFC PATCHSET v2] sched/fair: fix load balancer behavior when cgroup is in use

From Tejun Heo <tj@kernel.org>
Newsgroups linux.kernel
Subject Re: [RFC PATCHSET v2] sched/fair: fix load balancer behavior when cgroup is in use
Date 2017-05-05 15:30 +0200
Message-ID <tDLt0-2gd-21@gated-at.bofh.it> (permalink)
References <tDvxT-8pV-5@gated-at.bofh.it> <tDH61-7L0-9@gated-at.bofh.it>
Organization linux.* mail to news gateway

Show all headers | View raw


Hello, Vincent.

On Fri, May 05, 2017 at 10:46:53AM +0200, Vincent Guittot wrote:
> schbench results looks better with this version
> Latency percentiles (usec)
> 50.0000th: 212
> 75.0000th: 292
> 90.0000th: 385
> 95.0000th: 439
> *99.0000th: 671
> 99.5000th: 7992
> 99.9000th: 12176
> min=0, max=14855
> 
> p99 is back to a normal value but p99.5 stays higher than mainline

By how much and is that with the weight adjustment on the cgroup?  I
can't get reliable numbers on p99.5 and beyond on my test setup.  I
ordered the hikey board and will try to replicate your setup once it
arrives.

> I have also checked load_avg and runnable_load_avg value and there is
> something incorrect. I will provide details on the related patch

Sure, will respond there.

Thanks.

-- 
tejun

Back to linux.kernel | Previous | NextPrevious in thread | Next in thread | Find similar | Unroll thread


Thread

[RFC PATCHSET v2] sched/fair: fix load balancer behavior when cgroup  is in use Tejun Heo <tj@kernel.org> - 2017-05-04 22:30 +0200
  [PATCH 2/3] sched/fair: Add load_weight->runnable_load_{sum|avg} Tejun Heo <tj@kernel.org> - 2017-05-04 22:40 +0200
    Re: [PATCH 2/3] sched/fair: Add load_weight->runnable_load_{sum|avg} Dietmar Eggemann <dietmar.eggemann@arm.com> - 2017-05-05 15:30 +0200
      Re: [PATCH 2/3] sched/fair: Add load_weight->runnable_load_{sum|avg} Tejun Heo <tj@kernel.org> - 2017-05-05 15:30 +0200
        Re: [PATCH 2/3] sched/fair: Add load_weight->runnable_load_{sum|avg} Dietmar Eggemann <dietmar.eggemann@arm.com> - 2017-05-05 15:50 +0200
  [PATCH 3/3] sched/fair: Propagate runnable_load_avg independently  from load_avg Tejun Heo <tj@kernel.org> - 2017-05-04 22:40 +0200
    Re: [PATCH 3/3] sched/fair: Propagate runnable_load_avg independently  from load_avg Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-05 12:50 +0200
      Re: [PATCH 3/3] sched/fair: Propagate runnable_load_avg independently  from load_avg Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-05 14:20 +0200
        Re: [PATCH 3/3] sched/fair: Propagate runnable_load_avg  independently from load_avg Tejun Heo <tj@kernel.org> - 2017-05-05 15:30 +0200
    Re: [PATCH 3/3] sched/fair: Propagate runnable_load_avg independently  from load_avg Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-05 19:00 +0200
  [PATCH 1/3] sched/fair: Peter's shares_type patch Tejun Heo <tj@kernel.org> - 2017-05-04 22:40 +0200
    Re: [PATCH 1/3] sched/fair: Peter's shares_type patch Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-05 12:50 +0200
      Re: [PATCH 1/3] sched/fair: Peter's shares_type patch Tejun Heo <tj@kernel.org> - 2017-05-05 17:40 +0200
        Re: [PATCH 1/3] sched/fair: Peter's shares_type patch Tejun Heo <tj@kernel.org> - 2017-05-10 17:20 +0200
          Re: [PATCH 1/3] sched/fair: Peter's shares_type patch Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-10 18:10 +0200
            Re: [PATCH 1/3] sched/fair: Peter's shares_type patch Peter Zijlstra <peterz@infradead.org> - 2017-05-11 09:00 +0200
      Re: [PATCH 1/3] sched/fair: Peter's shares_type patch Peter Zijlstra <peterz@infradead.org> - 2017-05-05 17:50 +0200
  Re: [RFC PATCHSET v2] sched/fair: fix load balancer behavior when  cgroup is in use Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-05 10:50 +0200
    Re: [RFC PATCHSET v2] sched/fair: fix load balancer behavior when  cgroup is in use Tejun Heo <tj@kernel.org> - 2017-05-05 15:30 +0200
      Re: [RFC PATCHSET v2] sched/fair: fix load balancer behavior when  cgroup is in use Vincent Guittot <vincent.guittot@linaro.org> - 2017-05-05 15:40 +0200

csiph-web