Linux Power Management development
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@kernel.org>
To: Doug Smythies <dsmythies@telus.net>
Cc: "'Rafael J. Wysocki'" <rafael@kernel.org>,
	'srinivas pandruvada' <srinivas.pandruvada@linux.intel.com>,
	"'Zhang, Rui'" <rui.zhang@intel.com>,
	'Thomas Gleixner' <tglx@linutronix.de>,
	stable@vger.kernel.org, x86@kernel.org, linux-pm@vger.kernel.org,
	'Feng Tang' <feng.tang@intel.com>
Subject: Re: CPU excessively long times between frequency scaling driver calls - bisected
Date: Thu, 24 Feb 2022 08:58:26 -0800	[thread overview]
Message-ID: <20220224165826.GW4285@paulmck-ThinkPad-P17-Gen-1> (raw)
In-Reply-To: <002901d8299b$af50d6d0$0df28470$@telus.net>

On Thu, Feb 24, 2022 at 08:29:22AM -0800, Doug Smythies wrote:
> 
> On 2022.02.24 04:08:30 Paul E. McKenney wrote:
> > On Thu, Feb 24, 2022 at 04:08:30PM +0800, Feng Tang wrote:
> >> On Wed, Feb 23, 2022 at 03:23:20PM +0100, Rafael J. Wysocki wrote:
> >>> On Wed, Feb 23, 2022 at 1:40 AM Feng Tang <feng.tang@intel.com> wrote:
> >>> 
> >>> But this is not related to idle as such, but to the fact that idle
> >>> sometimes stops the scheduler tick which otherwise would run the
> >>> cpufreq governor callback on a regular basis.
> >>> 
> >>> It is stopping the tick that gets us into trouble, so I would avoid
> >>> doing it if the current performance state is too aggressive.
> >> 
> >> I've tried to simulate Doug's environment by using his kconfig, and
> >> offline my 36 CPUs Desktop to leave 12 CPUs online, and on it I can
> >> still see Local timer interrupts when there is no active load, with
> >> the longest interval between 2 timer interrupts is 4 seconds, while
> >> idle class's task_tick_idle() will do nothing, and CFS'
> >> task_tick_fair() will in turn call cfs_rq_util_change()
> >
> > Every four seconds?  Could you please post your .config?
> >
> >							Thanx, Paul
> 
> I steal the kernel config from the Ubuntu mainline PPA.
> See also earlier on this thread:
> 
> https://lore.kernel.org/linux-pm/CAAYoRsXkyWf0vmEE2HvjF6pzCC4utxTF=7AFx1PJv4Evh=C+Ow@mail.gmail.com/
> 
> but relevant part copied here:
> 
> > I steal the kernel configuration file from the Ubuntu mainline PPA
> > [1], what they call "lowlatency", or 1000Hz tick. I make these
> > changes before compile:
> >
> > scripts/config --disable DEBUG_INFO
> > scripts/config --disable SYSTEM_TRUSTED_KEYS
> > scripts/config --disable SYSTEM_REVOCATION_KEYS
> >
> > I [will] also send you the config and dmesg files in an off-list email.
> >
> > [1] https://kernel.ubuntu.com/~kernel-ppa/mainline/v5.17-rc3/
> 
> I put the same one I sent Feng on my web site where I was
> sharing stuff with Srinivas (coded to avoid the barrage of bots):
> 
> double u double u double u dot smythies dot com/~doug/linux/s18/hwp/srinivas/long_dur/

Thank you!

I don't see CONFIG_FAST_NO_HZ=y in your .config, so that is not the
reason for your every-four-second timers.  ;-)

(CONFIG_FAST_NO_HZ is being removed, FYI.)

							Thanx, Paul

  reply	other threads:[~2022-02-24 16:59 UTC|newest]

Thread overview: 46+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <003f01d81c8c$d20ee3e0$762caba0$@telus.net>
2022-02-08  2:39 ` CPU excessively long times between frequency scaling driver calls - bisected Feng Tang
2022-02-08  7:13   ` Doug Smythies
2022-02-08  9:15     ` Feng Tang
2022-02-09  6:23       ` Doug Smythies
2022-02-10  7:45         ` Zhang, Rui
2022-02-13 18:54           ` Doug Smythies
2022-02-14 15:17             ` srinivas pandruvada
2022-02-15 21:35               ` Doug Smythies
2022-02-22  7:34               ` Feng Tang
2022-02-22 18:04                 ` Rafael J. Wysocki
2022-02-23  0:07                   ` Doug Smythies
2022-02-23  0:32                     ` srinivas pandruvada
2022-02-23  0:40                       ` Feng Tang
2022-02-23 14:23                         ` Rafael J. Wysocki
2022-02-24  8:08                           ` Feng Tang
2022-02-24 14:44                             ` Paul E. McKenney
2022-02-24 16:29                               ` Doug Smythies
2022-02-24 16:58                                 ` Paul E. McKenney [this message]
2022-02-25  0:29                               ` Feng Tang
2022-02-25  1:06                                 ` Paul E. McKenney
2022-02-25 17:45                             ` Rafael J. Wysocki
2022-02-26  0:36                               ` Doug Smythies
2022-02-28  4:12                                 ` Feng Tang
2022-02-28 19:36                                   ` Rafael J. Wysocki
2022-03-01  5:52                                     ` Feng Tang
2022-03-01 11:58                                       ` Rafael J. Wysocki
2022-03-01 17:18                                         ` Doug Smythies
2022-03-01 17:34                                           ` Rafael J. Wysocki
2022-03-02  4:06                                             ` Doug Smythies
2022-03-02 19:00                                               ` Rafael J. Wysocki
2022-03-03 23:00                                                 ` Doug Smythies
2022-03-04  6:59                                                   ` Doug Smythies
2022-03-16 15:54                                                     ` Doug Smythies
2022-03-17 12:30                                                       ` Rafael J. Wysocki
2022-03-17 13:58                                                         ` Doug Smythies
2022-03-24 14:04                                                           ` Doug Smythies
2022-03-24 18:17                                                             ` Rafael J. Wysocki
2022-03-25  0:03                                                               ` Doug Smythies
2022-03-03  5:27                                               ` Feng Tang
2022-03-03 12:02                                                 ` Rafael J. Wysocki
2022-03-04  5:13                                                   ` Feng Tang
2022-03-04 16:23                                                     ` Paul E. McKenney
2022-02-23  2:49                   ` Feng Tang
2022-02-23 14:11                     ` Rafael J. Wysocki
2022-02-23  9:40                   ` Thomas Gleixner
2022-02-23 14:23                     ` 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=20220224165826.GW4285@paulmck-ThinkPad-P17-Gen-1 \
    --to=paulmck@kernel.org \
    --cc=dsmythies@telus.net \
    --cc=feng.tang@intel.com \
    --cc=linux-pm@vger.kernel.org \
    --cc=rafael@kernel.org \
    --cc=rui.zhang@intel.com \
    --cc=srinivas.pandruvada@linux.intel.com \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=x86@kernel.org \
    /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