Re: [PATCH v2 2/2] cpufreq: intel_pstate: Conditional frequency invariant accounting

[Rafael J. Wysocki]

On Fri, Oct 4, 2019 at 10:28 AM Vincent Guittot
<vincent.guittot@xxxxxxxxxx> wrote:
>
> On Fri, 4 Oct 2019 at 10:24, Giovanni Gherdovich <ggherdovich@xxxxxxx> wrote:
> >
> > On Thu, 2019-10-03 at 20:31 -0700, Srinivas Pandruvada wrote:
> > > On Thu, 2019-10-03 at 20:05 +0200, Rafael J. Wysocki wrote:
> > > > On Wednesday, October 2, 2019 2:29:26 PM CEST Giovanni Gherdovich
> > > > wrote:
> > > > > From: Srinivas Pandruvada <srinivas.pandruvada@xxxxxxxxxxxxxxx>
> > > > >
> > > > > intel_pstate has two operating modes: active and passive. In "active"
> > > > > mode, the in-built scaling governor is used and in "passive" mode, the
> > > > > driver can be used with any governor like "schedutil". In "active" mode
> > > > > the utilization values from schedutil is not used and there is a
> > > > > requirement from high performance computing use cases, not to readas
> > > > > well any APERF/MPERF MSRs.
> > > >
> > > > Well, this isn't quite convincing.
> > > >
> > > > In particular, I don't see why the "don't read APERF/MPERF MSRs" argument
> > > > applies *only* to intel_pstate in the "active" mode.  What about
> > > > intel_pstate in the "passive" mode combined with the "performance"
> > > > governor?  Or any other governor different from "schedutil" for that
> > > > matter?
> > > >
> > > > And what about acpi_cpufreq combined with any governor different from
> > > > "schedutil"?
> > > >
> > > > Scale invariance is not really needed in all of those cases right now
> > > > AFAICS, or is it?
> > >
> > > Correct. This is just part of the patch to disable in active mode
> > > (particularly in HWP and performance mode).
> > >
> > > But this patch is 2 years old. The folks who wanted this, disable
> > > intel-pstate and use userspace governor with acpi-cpufreq. So may be
> > > better to address those cases too.
> >
> > I disagree with "scale invariance is needed only by the schedutil governor";
> > the two other users are the CPU's estimated utilization in the wakeup path,
> > via cpu_util_without(), as well as the load-balance path, via cpu_util() which
> > is used by update_sg_lb_stats().
> >
> > Also remember that scale invariance is applied to both PELT signals util_avg
> > and load_avg; schedutil uses the former but not the latter.
>
> You have been quicker than me to reply. I was about to say the exact
> same things.
> scale invariance also helps the scheduler in task placement by
> stabilizing the metrics whatever the running frequency so a task will
> not be seen as a big task just because of a CPU running at lower
> frequency

So avoiding it just in one specific driver/governor configuration
would be inconsistent at best.

I guess that leaves us with the impact reduction option, realistically.