Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required RT CPU capacity

From	Juri Lelli <Juri.Lelli@arm.com>
Newsgroups	linux.kernel
Subject	Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required RT CPU capacity
Date	2016-09-02 11:40 +0200
Message-ID	<scT73-67d-1@gated-at.bofh.it> (permalink)
References	(3 earlier) <scfjj-40K-19@gated-at.bofh.it> <scgS6-4O7-21@gated-at.bofh.it> <scgS6-4O7-19@gated-at.bofh.it> <sch1M-4Rl-19@gated-at.bofh.it> <scI1X-7ip-13@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 01/09/16 14:48, Steve Muckle wrote:
> On Wed, Aug 31, 2016 at 06:00:02PM +0100, Juri Lelli wrote:
> > > Another problem is that we have many semi related knobs; we have the
> > > global RT runtime limit knob, but that doesn't affect cpufreq (maybe it
> > > should)
> > 
> > Maybe we could create this sort of link when using the cgroup RT
> > throttling interface as well? It should still then fit well once we
> > replace the underlying mechanism with DL reservations. And, AFAIK, the
> > interface is used by Android folks already.
> 
> I'm not sure how the upper bounds can be used to infer CPU frequency...
> On my Nexus 6p (an Android device), the global RT runtime limit
> seems to be set at 950ms/1sec, the root cgroup is set to 800ms/1sec, and
> bg_non_interactive is set at 700ms/1sec.
> 

Right, unfortunately. Still too coarse grained (as Thomas is also saying
in his last reply, if I read it correctly). Doesn't pay off the added
complexity I'm afraid.

Thread

[PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required RT CPU capacity Steve Muckle <steve.muckle@linaro.org> - 2016-08-26 20:50 +0200
  Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required RT CPU capacity "Rafael J. Wysocki" <rjw@rjwysocki.net> - 2016-08-31 03:30 +0200
    Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Steve Muckle <steve.muckle@linaro.org> - 2016-08-31 16:50 +0200
  Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Peter Zijlstra <peterz@infradead.org> - 2016-08-31 16:40 +0200
    Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Steve Muckle <steve.muckle@linaro.org> - 2016-08-31 17:10 +0200
      Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Peter Zijlstra <peterz@infradead.org> - 2016-08-31 18:50 +0200
        Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Juri Lelli <Juri.Lelli@arm.com> - 2016-08-31 19:00 +0200
          Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Peter Zijlstra <peterz@infradead.org> - 2016-09-01 09:20 +0200
          Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Steve Muckle <steve.muckle@linaro.org> - 2016-09-01 23:50 +0200
            Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Juri Lelli <Juri.Lelli@arm.com> - 2016-09-02 11:40 +0200
              Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Thomas Gleixner <tglx@linutronix.de> - 2016-09-02 14:20 +0200
        Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required RT CPU capacity "Rafael J. Wysocki" <rjw@rjwysocki.net> - 2016-09-01 00:50 +0200
        Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Thomas Gleixner <tglx@linutronix.de> - 2016-09-02 10:20 +0200
      Re: [PATCH 2/2] sched: cpufreq: use rt_avg as estimate of required  RT CPU capacity Thomas Gleixner <tglx@linutronix.de> - 2016-08-31 19:10 +0200

csiph-web