public inbox for cpufreq@vger.kernel.org
 help / color / mirror / Atom feed
From: Borislav Petkov <bp@suse.de>
To: "Rafael J. Wysocki" <rjw@sisk.pl>,
	Stratos Karafotis <stratosk@semaphore.gr>
Cc: Viresh Kumar <viresh.kumar@linaro.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Ingo Molnar <mingo@redhat.com>, "H. Peter Anvin" <hpa@zytor.com>,
	linux-pm@vger.kernel.org, cpufreq@vger.kernel.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v3 1/3] cpufreq: ondemand: Change the calculation of target frequency
Date: Thu, 6 Jun 2013 12:01:38 +0200	[thread overview]
Message-ID: <20130606100138.GC21181@pd.tnic> (raw)
In-Reply-To: <105446113.ZumbZWCbSi@vostro.rjw.lan>

On Wed, Jun 05, 2013 at 10:35:05PM +0200, Rafael J. Wysocki wrote:
> On Wednesday, June 05, 2013 08:13:26 PM Stratos Karafotis wrote:
> > Hi Borislav,
> > 
> > On 06/05/2013 07:17 PM, Borislav Petkov wrote:
> > > On Wed, Jun 05, 2013 at 07:01:25PM +0300, Stratos Karafotis wrote:
> > >> Ondemand calculates load in terms of frequency and increases it only
> > >> if the load_freq is greater than up_threshold multiplied by current
> > >> or average frequency. This seems to produce oscillations of frequency
> > >> between min and max because, for example, a relatively small load can
> > >> easily saturate minimum frequency and lead the CPU to max. Then, the
> > >> CPU will decrease back to min due to a small load_freq.
> > >
> > > Right, and I think this is how we want it, no?
> > >
> > > The thing is, the faster you finish your work, the faster you can become
> > > idle and save power.
> > 
> > This is exactly the goal of this patch. To use more efficiently middle
> > frequencies to finish faster the work.

Hold on, you say above "easily saturate minimum frequency and lead the
CPU to max". I read this as we jump straight to max P-state where we
even boost.

"CPU to max" finishes the work faster than middle frequencies, if you're
CPU-bound.

> > > If you switch frequencies in a staircase-like manner, you're going to
> > > take longer to finish, in certain cases, and burn more power while doing
> > > so.
> > 
> > This is not true with this patch. It switches to middle frequencies
> > when the load < up_threshold.

This is worth investigating wrt hightened power consumption, as Rafael
suggested.

-- 
Regards/Gruss,
    Boris.

Sent from a fat crate under my desk. Formatting is fine.
--

  reply	other threads:[~2013-06-06 10:01 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-06-05 16:01 [PATCH v3 1/3] cpufreq: ondemand: Change the calculation of target frequency Stratos Karafotis
2013-06-05 16:17 ` Borislav Petkov
2013-06-05 16:58   ` David C Niemi
2013-06-06  9:55     ` Borislav Petkov
2013-06-06  9:57       ` Viresh Kumar
2013-06-06 13:50       ` David C Niemi
2013-06-05 17:13   ` Stratos Karafotis
2013-06-05 20:35     ` Rafael J. Wysocki
2013-06-06 10:01       ` Borislav Petkov [this message]
2013-06-06 10:10         ` Viresh Kumar
2013-06-06 12:10           ` Borislav Petkov
2013-06-06 16:46             ` Stratos Karafotis
2013-06-06 17:11               ` Borislav Petkov
2013-06-06 17:32                 ` Stratos Karafotis
2013-06-07 19:14       ` Stratos Karafotis
2013-06-07 20:57         ` Rafael J. Wysocki
2013-06-08  9:56           ` Stratos Karafotis
2013-06-08 11:18             ` Rafael J. Wysocki
  -- strict thread matches above, loose matches on Subject: below --
2013-06-06 12:54 Stratos Karafotis
2013-06-06 13:15 ` Borislav Petkov
2013-06-06 12:56 Stratos Karafotis
2013-06-08 12:34 Stratos Karafotis
2013-06-08 14:05 ` Rafael J. Wysocki
2013-06-08 20:31   ` Stratos Karafotis
2013-06-08 22:18     ` Rafael J. Wysocki
2013-06-09 16:26       ` Borislav Petkov
2013-06-09 18:08         ` Stratos Karafotis
2013-06-09 20:58           ` Rafael J. Wysocki
2013-06-09 21:14             ` Borislav Petkov
2013-06-09 22:11               ` Rafael J. Wysocki
2013-06-10 21:57             ` Stratos Karafotis
2013-06-10 23:24               ` Rafael J. Wysocki
2013-06-13 21:22                 ` Stratos Karafotis
2013-06-13 21:40                   ` Borislav Petkov
2013-06-13 22:04                     ` Stratos Karafotis
2013-06-13 22:38                       ` Borislav Petkov
2013-06-13 22:15                     ` Rafael J. Wysocki
2013-06-13 22:37                       ` Borislav Petkov
2013-06-14 12:46                         ` Rafael J. Wysocki
2013-06-14 12:44                           ` Borislav Petkov
2013-06-14 12:55                             ` Rafael J. Wysocki
2013-06-14 15:53                               ` Stratos Karafotis

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=20130606100138.GC21181@pd.tnic \
    --to=bp@suse.de \
    --cc=cpufreq@vger.kernel.org \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pm@vger.kernel.org \
    --cc=mingo@redhat.com \
    --cc=rjw@sisk.pl \
    --cc=stratosk@semaphore.gr \
    --cc=tglx@linutronix.de \
    --cc=viresh.kumar@linaro.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