linux-kernel - Re: sched/cpufreq: Rework schedutil governor performance estimation

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAKfTPtC7pOtb-srrgQLFbTueLLDqHay+GQBm9=sNsnZDg_UYSQ@mail.gmail.com>
Date: Sun, 11 Feb 2024 14:36:02 +0100
From: Vincent Guittot <vincent.guittot@...aro.org>
To: Doug Smythies <dsmythies@...us.net>
Cc: Ingo Molnar <mingo@...nel.org>, "Rafael J. Wysocki" <rafael@...nel.org>, linux-pm@...r.kernel.org, 
	linux-kernel@...r.kernel.org
Subject: Re: sched/cpufreq: Rework schedutil governor performance estimation -
 Regression bisected

On Sat, 10 Feb 2024 at 00:16, Doug Smythies <dsmythies@...us.net> wrote:
>
> Hi Vincent,
> Thank you for your quick reply.
>
> On 2024.02.09.14:11 Vincent wrote:
> On Fri, 9 Feb 2024 at 22:38, Doug Smythies <dsmythies@...us.net> wrote:
> >>
> >> Hi,
> >>
> >> I noticed a regression in the 6.8rc series kernels. Bisecting the kernel pointed to:
> >>
> >> # first bad commit: [9c0b4bb7f6303c9c4e2e34984c46f5a86478f84d]
> >> sched/cpufreq: Rework schedutil governor performance estimation
> >>
> >> There was previous bisection and suggestion of reversion,
> >> but I guess it wasn't done in the end. [1]
> >
> > This has been fixed with
> > https://lore.kernel.org/all/170539970061.398.16662091173685476681.tip-bot2@tip-bot2/
>
> Okay, thanks. I didn't find that one.
>
> >> The regression: reduced maximum CPU frequency is ignored.
>
> > This seems to be something new.
> > schedutil doesn't impact the max_freq and it's up to cpufreq driver
> > select the final freq which should stay within the limits
>
> Okay. All I know is this is the commit that caused the regression.

Could you check if the fix solved your problem ?

> I do not know why, but I do wonder if there could any relationship with
> the old, never fixed, problem of incorrect stale frequencies reported
> under the same operating conditions. See the V2 note:
> https://lore.kernel.org/all/001d01d9d3a7$71736f50$545a4df0$@telus.net/

IIUC the problem is that policy->cur is not used by intel_cpufreq and
stays set to the last old/init value.
Do I get it right that this is only informative ?

Normally cpufreq governor checks the new limits and updates current
freq if necessary except when fast switch is enabled.

>
> where I haven't been able to figure out a solution.
>
> >> Conditions:
> >> CPU frequency scaling driver: intel_cpufreq (a.k.a intel_pstate in passive mode)
> >> CPU frequency scaling governor: schedutil
> >> HWP (HardWare Pstate) control (a.k.a. Intel_speedshift): Enabled
> >> Processor: Intel(R) Core(TM) i5-10600K CPU @ 4.10GHz
> >>
> >> I did not check any other conditions, i.e. HWP disabled or the acpi-cpufreq driver.
> >>
> >> Example: A 100% load on CPU 5.
> >>
> >> sudo turbostat --quiet --Summary --show Busy%,Bzy_MHz,IRQ,PkgWatt,PkgTmp,RAMWatt,GFXWatt,CorWatt --interval 15
> >> Busy%   Bzy_MHz IRQ     PkgTmp  PkgWatt CorWatt GFXWatt RAMWatt
> >> 8.42    4636    21823   67      28.40   27.56   0.00    2.59
> >> 8.40    4577    17724   66      27.57   26.73   0.00    2.59
> >> 8.35    4637    19535   66      28.65   27.81   0.00    2.60
> >> 8.41    4578    20723   66      27.73   26.89   0.00    2.59
> >> 8.40    4558    19156   67      27.39   26.55   0.00    2.58
> >> 8.34    4502    18127   67      26.79   25.96   0.00    2.57
> >>
> >> grep . /sys/devices/system/cpu/cpu*/cpufreq/scaling_max_freq
> >> /sys/devices/system/cpu/cpu0/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu10/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu11/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu1/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu2/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu3/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu4/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu6/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu7/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu8/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu9/cpufreq/scaling_max_freq:2400000
> >>
> >> grep . /sys/devices/system/cpu/cpu5/cpufreq/*
> >> /sys/devices/system/cpu/cpu5/cpufreq/affected_cpus:5
> >> /sys/devices/system/cpu/cpu5/cpufreq/base_frequency:4100000
> >> /sys/devices/system/cpu/cpu5/cpufreq/cpuinfo_max_freq:4800000
> >> /sys/devices/system/cpu/cpu5/cpufreq/cpuinfo_min_freq:800000
> >> /sys/devices/system/cpu/cpu5/cpufreq/cpuinfo_transition_latency:20000
> >> /sys/devices/system/cpu/cpu5/cpufreq/energy_performance_available_preferences:default performance balance_performance balance_power
> >> power
> >> /sys/devices/system/cpu/cpu5/cpufreq/energy_performance_preference:balance_performance
> >> /sys/devices/system/cpu/cpu5/cpufreq/related_cpus:5
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_available_governors:conservative ondemand userspace powersave performance schedutil
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_cur_freq:4799998
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_driver:intel_cpufreq
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_governor:schedutil
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_max_freq:2400000
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_min_freq:800000
> >> /sys/devices/system/cpu/cpu5/cpufreq/scaling_setspeed:<unsupported>
> >>
> >> [1] https://lore.kernel.org/all/CAKfTPtDCQuJjpi6=zjeWPcLeP+ZY5Dw7XDrZ-LpXqEAAUbXLhA@mail.gmail.com/
>
>