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

Groups > linux.kernel > #1468256 > unrolled thread

Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection algorithm for Core

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: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection  algorithm for Core Wanpeng Li <kernellwp@gmail.com> - 2016-08-23 05:50 +0200
    Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state  selection algorithm for Core Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> - 2016-08-23 06:10 +0200
      Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection  algorithm for Core Wanpeng Li <kernellwp@gmail.com> - 2016-08-23 07:00 +0200
        Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection algorithm for Core "Rafael J. Wysocki" <rjw@rjwysocki.net> - 2016-08-23 19:30 +0200

#1468256 — Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection algorithm for Core

Hi Rafael,
2016-08-01 7:38 GMT+08:00 Rafael J. Wysocki <rjw@rjwysocki.net>:
> From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>
> The PID-base P-state selection algorithm used by intel_pstate for
> Core processors is based on very weak foundations.  Namely, its
> decisions are mostly based on the values of the APERF and MPERF
> feedback registers and it only estimates the actual utilization to
> check if it is not extremely low (in order to avoid getting stuck
> in the highest P-state in that case).
>
> Since it generally causes the CPU P-state to ramp up quickly, it
> leads to satisfactory performance, but the metric used by it is only
> really valid when the CPU changes P-states by itself (ie. in the turbo
> range) and if the P-state value set by the driver is treated by the
> CPU as the upper limit on turbo P-states selected by it.

Do you mean CPU will ignore the request value which is submitted by
intel_pstate driver in the turbo range, but respect the upper limit on
the turbo P-states which is submitted by intel_pstate driver?

Regards,
Wanpeng Li

[toc] | [next] | [standalone]

#1468263 — Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection algorithm for Core

On Tue, 2016-08-23 at 11:48 +0800, Wanpeng Li wrote:
> Hi Rafael,
> 2016-08-01 7:38 GMT+08:00 Rafael J. Wysocki <rjw@rjwysocki.net>:
> > 
> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > 
> > The PID-base P-state selection algorithm used by intel_pstate for
> > Core processors is based on very weak foundations.  Namely, its
> > decisions are mostly based on the values of the APERF and MPERF
> > feedback registers and it only estimates the actual utilization to
> > check if it is not extremely low (in order to avoid getting stuck
> > in the highest P-state in that case).
> > 
> > Since it generally causes the CPU P-state to ramp up quickly, it
> > leads to satisfactory performance, but the metric used by it is
> > only
> > really valid when the CPU changes P-states by itself (ie. in the
> > turbo
> > range) and if the P-state value set by the driver is treated by the
> > CPU as the upper limit on turbo P-states selected by it.
> 
> Do you mean CPU will ignore the request value which is submitted by
> intel_pstate driver in the turbo range, but respect the upper limit
> on
> the turbo P-states which is submitted by intel_pstate driver?
Any request to request a P-state even upper limit can be totally
ignored in turbo range. 

Thanks,
Srinivas

> 
> Regards,
> Wanpeng Li
> --
> To unsubscribe from this list: send the line "unsubscribe linux-pm"
> in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

[toc] | [prev] | [next] | [standalone]

#1468282

2016-08-23 12:08 GMT+08:00 Srinivas Pandruvada
<srinivas.pandruvada@linux.intel.com>:
> On Tue, 2016-08-23 at 11:48 +0800, Wanpeng Li wrote:
>> Hi Rafael,
>> 2016-08-01 7:38 GMT+08:00 Rafael J. Wysocki <rjw@rjwysocki.net>:
>> >
>> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>> >
>> > The PID-base P-state selection algorithm used by intel_pstate for
>> > Core processors is based on very weak foundations.  Namely, its
>> > decisions are mostly based on the values of the APERF and MPERF
>> > feedback registers and it only estimates the actual utilization to
>> > check if it is not extremely low (in order to avoid getting stuck
>> > in the highest P-state in that case).
>> >
>> > Since it generally causes the CPU P-state to ramp up quickly, it
>> > leads to satisfactory performance, but the metric used by it is
>> > only
>> > really valid when the CPU changes P-states by itself (ie. in the
>> > turbo
>> > range) and if the P-state value set by the driver is treated by the
>> > CPU as the upper limit on turbo P-states selected by it.
>>
>> Do you mean CPU will ignore the request value which is submitted by
>> intel_pstate driver in the turbo range, but respect the upper limit
>> on
>> the turbo P-states which is submitted by intel_pstate driver?
> Any request to request a P-state even upper limit can be totally
> ignored in turbo range.

How to understand Rafael's description "if the P-state value set by
the driver is treated by the CPU as the upper limit on the turbo
P-states selected by it"? The upper limit will be totally ignored in
turbo range.

Regards,
Wanpeng Li

[toc] | [prev] | [next] | [standalone]

#1468730 — Re: [RFC][PATCH 7/7] cpufreq: intel_pstate: Change P-state selection algorithm for Core

On Tuesday, August 23, 2016 12:50:14 PM Wanpeng Li wrote:
> 2016-08-23 12:08 GMT+08:00 Srinivas Pandruvada
> <srinivas.pandruvada@linux.intel.com>:
> > On Tue, 2016-08-23 at 11:48 +0800, Wanpeng Li wrote:
> >> Hi Rafael,
> >> 2016-08-01 7:38 GMT+08:00 Rafael J. Wysocki <rjw@rjwysocki.net>:
> >> >
> >> > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> >> >
> >> > The PID-base P-state selection algorithm used by intel_pstate for
> >> > Core processors is based on very weak foundations.  Namely, its
> >> > decisions are mostly based on the values of the APERF and MPERF
> >> > feedback registers and it only estimates the actual utilization to
> >> > check if it is not extremely low (in order to avoid getting stuck
> >> > in the highest P-state in that case).
> >> >
> >> > Since it generally causes the CPU P-state to ramp up quickly, it
> >> > leads to satisfactory performance, but the metric used by it is
> >> > only
> >> > really valid when the CPU changes P-states by itself (ie. in the
> >> > turbo
> >> > range) and if the P-state value set by the driver is treated by the
> >> > CPU as the upper limit on turbo P-states selected by it.
> >>
> >> Do you mean CPU will ignore the request value which is submitted by
> >> intel_pstate driver in the turbo range, but respect the upper limit
> >> on
> >> the turbo P-states which is submitted by intel_pstate driver?
> > Any request to request a P-state even upper limit can be totally
> > ignored in turbo range.
> 
> How to understand Rafael's description "if the P-state value set by
> the driver is treated by the CPU as the upper limit on the turbo
> P-states selected by it"? The upper limit will be totally ignored in
> turbo range.

It wasn't ignored, though, by some older CPU models.

Thanks,
Rafael

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web