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


Groups > linux.kernel > #1657587 > unrolled thread

Re: hackbench vs select_idle_sibling; was: [tip:sched/core] sched/fair, cpumask: Export for_each_cpu_wrap()

Started byMatt Fleming <matt@codeblueprint.co.uk>
First post2017-06-05 15:10 +0200
Last post2017-06-06 11:30 +0200
Articles 2 — 2 participants

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.


Contents

  Re: hackbench vs select_idle_sibling; was: [tip:sched/core]  sched/fair, cpumask: Export for_each_cpu_wrap() Matt Fleming <matt@codeblueprint.co.uk> - 2017-06-05 15:10 +0200
    Re: hackbench vs select_idle_sibling; was: [tip:sched/core]  sched/fair, cpumask: Export for_each_cpu_wrap() Peter Zijlstra <peterz@infradead.org> - 2017-06-06 11:30 +0200

#1657587 — Re: hackbench vs select_idle_sibling; was: [tip:sched/core] sched/fair, cpumask: Export for_each_cpu_wrap()

FromMatt Fleming <matt@codeblueprint.co.uk>
Date2017-06-05 15:10 +0200
SubjectRe: hackbench vs select_idle_sibling; was: [tip:sched/core] sched/fair, cpumask: Export for_each_cpu_wrap()
Message-ID<tOZVE-3jm-33@gated-at.bofh.it>
On Fri, 19 May, at 04:00:35PM, Matt Fleming wrote:
> On Wed, 17 May, at 12:53:50PM, Peter Zijlstra wrote:
> > 
> > Please test..
> 
> Results are still coming in but things do look better with your patch
> applied.
> 
> It does look like there's a regression when running hackbench in
> process mode and when the CPUs are not fully utilised, e.g. check this
> out:

This turned out to be a false positive; your patch improves things as
far as I can see.

[toc] | [next] | [standalone]


#1658528

FromPeter Zijlstra <peterz@infradead.org>
Date2017-06-06 11:30 +0200
Message-ID<tPiYi-6PD-21@gated-at.bofh.it>
In reply to#1657587
On Mon, Jun 05, 2017 at 02:00:21PM +0100, Matt Fleming wrote:
> On Fri, 19 May, at 04:00:35PM, Matt Fleming wrote:
> > On Wed, 17 May, at 12:53:50PM, Peter Zijlstra wrote:
> > > 
> > > Please test..
> > 
> > Results are still coming in but things do look better with your patch
> > applied.
> > 
> > It does look like there's a regression when running hackbench in
> > process mode and when the CPUs are not fully utilised, e.g. check this
> > out:
> 
> This turned out to be a false positive; your patch improves things as
> far as I can see.

Hooray, I'll move it to a part of the queue intended for merging.

Thanks!

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web