linux-pm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Doug Smythies" <dsmythies@telus.net>
To: "'Rafael J. Wysocki'" <rjw@rjwysocki.net>
Cc: 'Srinivas Pandruvada' <srinivas.pandruvada@linux.intel.com>,
	'LKML' <linux-kernel@vger.kernel.org>,
	'Linux PM' <linux-pm@vger.kernel.org>
Subject: RE: [Update][PATCH v3 2/3] cpufreq: intel_pstate: Do not reinit performance limits in ->setpolicy
Date: Wed, 29 Mar 2017 15:01:17 -0700	[thread overview]
Message-ID: <004101d2a8d7$ff31bb50$fd9531f0$@net> (raw)
In-Reply-To: jZL3cbFe8sVy3jZL4c6fog

Hi Rafael,

Sorry for the delay, but I didn't notice until today that this
commit causes a regression, at least in my computer.

I have not figured out exactly what is wrong, as I must
admit I am finding these policy interactions difficult
to follow.

On 2017.03.02 14:29 Rafael J. Wysocki wrote:
>From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
>
> If the current P-state selection algorithm is set to "performance"
> in intel_pstate_set_policy(), the limits may be initialized from

... [cut] ...

Going back to kernel 4.11-rc1 I get this after boot**:

$ uname -a
Linux s15 4.11.0-rc1-stock #217 SMP Sun Mar 5 15:34:38 PST 2017 x86_64 x86_64 x86_64 GNU/Linux

$ grep . /sys/devices/system/cpu/intel_pstate/*
/sys/devices/system/cpu/intel_pstate/max_perf_pct:100  <<< Correct
/sys/devices/system/cpu/intel_pstate/min_perf_pct:43   <<< Correct
/sys/devices/system/cpu/intel_pstate/no_turbo:0
/sys/devices/system/cpu/intel_pstate/num_pstates:23
/sys/devices/system/cpu/intel_pstate/status:active
/sys/devices/system/cpu/intel_pstate/turbo_pct:18

$ grep . /sys/devices/system/cpu/cpu0/cpufreq/*
/sys/devices/system/cpu/cpu0/cpufreq/affected_cpus:0
grep: /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq: Permission denied
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_max_freq:3800000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_min_freq:1600000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_transition_latency:4294967295
/sys/devices/system/cpu/cpu0/cpufreq/related_cpus:0
/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors:performance powersave
/sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq:1600805
/sys/devices/system/cpu/cpu0/cpufreq/scaling_driver:intel_pstate
/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor:powersave  <<< Notice
/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq:3800000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq:1600000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed:<unsupported>

$ sudo rdmsr --bitfield 15:8 -d -a 0x199 <<< Requested P-States
16
16
17
17
16
16
16
16

Going back to kernel 4.11-rc2 I get this after boot**:

~$ uname -a
Linux s15 4.11.0-rc2-stock #218 SMP Sun Mar 12 23:57:44 PDT 2017 x86_64 x86_64 x86_64 GNU/Linux

$ grep . /sys/devices/system/cpu/intel_pstate/*
/sys/devices/system/cpu/intel_pstate/max_perf_pct:100  <<< Correct
/sys/devices/system/cpu/intel_pstate/min_perf_pct:100  <<< Incorrect
/sys/devices/system/cpu/intel_pstate/no_turbo:0
/sys/devices/system/cpu/intel_pstate/num_pstates:23
/sys/devices/system/cpu/intel_pstate/status:active
/sys/devices/system/cpu/intel_pstate/turbo_pct:18

$ grep . /sys/devices/system/cpu/cpu0/cpufreq/*
/sys/devices/system/cpu/cpu0/cpufreq/affected_cpus:0
grep: /sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_cur_freq: Permission denied
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_max_freq:3800000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_min_freq:1600000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/cpuinfo_transition_latency:4294967295
/sys/devices/system/cpu/cpu0/cpufreq/related_cpus:0
/sys/devices/system/cpu/cpu0/cpufreq/scaling_available_governors:performance powersave
/sys/devices/system/cpu/cpu0/cpufreq/scaling_cur_freq:1600805
/sys/devices/system/cpu/cpu0/cpufreq/scaling_driver:intel_pstate
/sys/devices/system/cpu/cpu0/cpufreq/scaling_governor:powersave  <<< Notice
/sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq:3800000  <<< Correct
/sys/devices/system/cpu/cpu0/cpufreq/scaling_min_freq:3800000  <<< Incorrect
/sys/devices/system/cpu/cpu0/cpufreq/scaling_setspeed:<unsupported>

$ sudo rdmsr --bitfield 15:8 -d -a 0x199  <<< Requested P-States
38   <<<< All are pinned, system is idle
38
38
38
38
38
38
38

**: After boot means > 1 minute after boot, because my distro (Ubtunu)
starts up using the performance governor and then changes to powersave
after 1 minute.

... Doug

  parent reply	other threads:[~2017-03-29 22:01 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-02-28 13:13 [PATCH 0/2] cpufreq: intel_pstate: Two fixes related to limis Rafael J. Wysocki
2017-02-28 13:14 ` [PATCH 1/2] cpufreq: intel_pstate: Fix global settings in active mode Rafael J. Wysocki
2017-02-28 22:21   ` Rafael J. Wysocki
2017-02-28 13:16 ` [PATCH 2/2] cpufreq: intel_pstate: Do not reinit performance limits in ->setpolicy Rafael J. Wysocki
2017-02-28 23:00 ` [PATCH v2 0/3] cpufreq: intel_pstate: Fixes related to limis Rafael J. Wysocki
2017-02-28 23:07   ` [PATCH v2 1/3] cpufreq: intel_pstate: Fix global settings in active mode Rafael J. Wysocki
2017-02-28 23:09   ` [PATCH v2 2/3] cpufreq: intel_pstate: Do not reinit performance limits in ->setpolicy Rafael J. Wysocki
2017-03-02 17:18     ` Rafael J. Wysocki
2017-03-02 17:22       ` Rafael J. Wysocki
2017-03-02 22:29     ` [Update][PATCH v3 " Rafael J. Wysocki
2017-03-29 22:01     ` Doug Smythies [this message]
2017-03-29 22:16     ` Doug Smythies
2017-03-29 22:17       ` Rafael J. Wysocki
2017-02-28 23:11   ` [PATCH v2 3/3] cpufreq: intel_pstate: Fix intel_pstate_verify_policy() Rafael J. Wysocki

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to='004101d2a8d7$ff31bb50$fd9531f0$@net' \
    --to=dsmythies@telus.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=rjw@rjwysocki.net \
    --cc=srinivas.pandruvada@linux.intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).