From: "Rafael J. Wysocki" <rafael@kernel.org>
To: Viresh Kumar <viresh.kumar@linaro.org>
Cc: "Rafael J. Wysocki" <rafael@kernel.org>,
"Rafael J. Wysocki" <rjw@rjwysocki.net>,
Linux PM <linux-pm@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Doug Smythies <dsmythies@telus.net>,
Giovanni Gherdovich <ggherdovich@suse.com>
Subject: Re: [PATCH v1 2/4] cpufreq: schedutil: Adjust utilization instead of frequency
Date: Wed, 9 Dec 2020 16:32:06 +0100 [thread overview]
Message-ID: <CAJZ5v0iOvG0PNQDXN00oKCzyZmaF71UB+DJ+zHL5P3xRCAk1tQ@mail.gmail.com> (raw)
In-Reply-To: <20201209051642.ddwgds4gznxt3lfn@vireshk-i7>
On Wed, Dec 9, 2020 at 6:16 AM Viresh Kumar <viresh.kumar@linaro.org> wrote:
>
> On 08-12-20, 18:01, Rafael J. Wysocki wrote:
> > On Tue, Dec 8, 2020 at 9:52 AM Viresh Kumar <viresh.kumar@linaro.org> wrote:
> > >
> > > On 07-12-20, 17:29, Rafael J. Wysocki wrote:
> > > > From: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > > >
> > > > When avoiding reduction of the frequency after the target CPU has
> > > > been busy since the previous frequency update, adjust the utilization
> > > > instead of adjusting the frequency, because doing so is more prudent
> > > > (it is done to counter a possible utilization deficit after all) and
> > > > it will allow some code to be shared after a subsequent change.
> > > >
> > > > Signed-off-by: Rafael J. Wysocki <rafael.j.wysocki@intel.com>
> > > > ---
> > > > kernel/sched/cpufreq_schedutil.c | 11 ++++-------
> > > > 1 file changed, 4 insertions(+), 7 deletions(-)
> > > >
> > > > Index: linux-pm/kernel/sched/cpufreq_schedutil.c
> > > > ===================================================================
> > > > --- linux-pm.orig/kernel/sched/cpufreq_schedutil.c
> > > > +++ linux-pm/kernel/sched/cpufreq_schedutil.c
> > > > @@ -437,7 +437,7 @@ static void sugov_update_single(struct u
> > > > {
> > > > struct sugov_cpu *sg_cpu = container_of(hook, struct sugov_cpu, update_util);
> > > > struct sugov_policy *sg_policy = sg_cpu->sg_policy;
> > > > - unsigned int cached_freq = sg_policy->cached_raw_freq;
> > > > + unsigned long prev_util = sg_cpu->util;
> > > > unsigned int next_f;
> > > >
> > > > sugov_iowait_boost(sg_cpu, time, flags);
> > > > @@ -451,17 +451,14 @@ static void sugov_update_single(struct u
> > > > sugov_get_util(sg_cpu);
> > > > sugov_iowait_apply(sg_cpu, time);
> > > >
> > > > - next_f = get_next_freq(sg_policy, sg_cpu->util, sg_cpu->max);
> > > > /*
> > > > * Do not reduce the frequency if the CPU has not been idle
> > > > * recently, as the reduction is likely to be premature then.
> > > > */
> > > > - if (sugov_cpu_is_busy(sg_cpu) && next_f < sg_policy->next_freq) {
> > > > - next_f = sg_policy->next_freq;
> > > > + if (sugov_cpu_is_busy(sg_cpu) && sg_cpu->util < prev_util)
> > > > + sg_cpu->util = prev_util;
> > > >
> > > > - /* Restore cached freq as next_freq has changed */
> > > > - sg_policy->cached_raw_freq = cached_freq;
> > > > - }
> > > > + next_f = get_next_freq(sg_policy, sg_cpu->util, sg_cpu->max);
> > >
> > > I don't think we can replace freq comparison by util, or at least it will give
> > > us a different final frequency and the behavior is changed.
> > >
> > > Lets take an example, lets say current freq is 1 GHz and max is 1024.
Ah, so that's in the freq-dependent case.
In the freq-invariant case next_f doesn't depend on the current frequency.
> > > Round 1: Lets say util is 1000
> > >
> > > next_f = 1GHz * 1.25 * 1000/1024 = 1.2 GHz
> > >
> > > Round 2: Lets say util has come down to 900 here,
> > >
> > > before the patch:
> > >
> > > next_f = 1.2 GHz * 1.25 * 900/1024 = 1.31 GHz
> > >
> > > after the patch:
> > >
> > > next_f = 1.2 GHz * 1.25 * 1000/1024 = 1.45 GHz
> > >
> > > Or did I make a mistake here ?
> >
> > I think so, if my understanding is correct.
> >
> > Without the patch, next_f will be reset to the previous value
> > (sq_policy->next_freq) if the CPU has been busy and the (new) next_f
> > is less than that value.
> >
> > So the "new" next_f before the patch is 1.31 GHz, but because it is
> > less than the previous value (1.45 GHz), it will be reset to that
> > value, unless I'm missing something.
>
> The prev frequency here was 1.2 GHz (after Round 1). 1.45 GHz is the
> value we get after this patch, as we take the earlier utilization
> (1000) into account instead of 900.
So I have misunderstood your example.
In the non-invariant case (which is or shortly will be relevant for
everybody interested) cpuinfo.max_freq goes into the calculation
instead of the current frequency and the mapping between util and freq
is linear. In the freq-dependent case it is not linear, of course.
So I guess the concern is that this changes the behavior in the
freq-dependent case which may not be desirable.
Fair enough, but I'm not sure if that is enough of a reason to avoid
sharing the code between the "perf" and "freq" paths.
next prev parent reply other threads:[~2020-12-09 15:32 UTC|newest]
Thread overview: 34+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-12-07 16:25 [PATCH v1 0/4] cpufreq: Allow drivers to receive more information from the governor Rafael J. Wysocki
2020-12-07 16:28 ` [PATCH v1 1/4] cpufreq: schedutil: Add util to struct sg_cpu Rafael J. Wysocki
2020-12-08 8:33 ` Viresh Kumar
2020-12-09 17:17 ` Rafael J. Wysocki
2020-12-07 16:29 ` [PATCH v1 2/4] cpufreq: schedutil: Adjust utilization instead of frequency Rafael J. Wysocki
2020-12-08 8:51 ` Viresh Kumar
2020-12-08 17:01 ` Rafael J. Wysocki
2020-12-09 5:16 ` Viresh Kumar
2020-12-09 15:32 ` Rafael J. Wysocki [this message]
2020-12-14 11:07 ` Viresh Kumar
2020-12-07 16:35 ` [PATCH v1 3/4] cpufreq: Add special-purpose fast-switching callback for drivers Rafael J. Wysocki
2020-12-08 9:02 ` Viresh Kumar
2020-12-15 4:16 ` Viresh Kumar
2020-12-15 15:38 ` Rafael J. Wysocki
2020-12-07 16:38 ` [PATCH v1 4/4] cpufreq: intel_pstate: Implement the ->adjust_perf() callback Rafael J. Wysocki
2020-12-08 12:43 ` Peter Zijlstra
2020-12-08 17:10 ` Rafael J. Wysocki
2020-12-08 16:30 ` [PATCH v1 0/4] cpufreq: Allow drivers to receive more information from the governor Giovanni Gherdovich
2020-12-08 17:13 ` Rafael J. Wysocki
2020-12-08 19:14 ` Doug Smythies
2020-12-13 19:12 ` Doug Smythies
2020-12-18 15:32 ` Peter Zijlstra
2020-12-14 20:01 ` [PATCH v2 0/3] " Rafael J. Wysocki
2020-12-14 20:04 ` [PATCH v2 1/3] cpufreq: schedutil: Add util to struct sg_cpu Rafael J. Wysocki
2020-12-14 20:08 ` [PATCH v2 2/3] cpufreq: Add special-purpose fast-switching callback for drivers Rafael J. Wysocki
2020-12-14 20:09 ` [PATCH v2 3/3] cpufreq: intel_pstate: Implement the ->adjust_perf() callback Rafael J. Wysocki
2020-12-15 3:29 ` Srinivas Pandruvada
2020-12-15 4:16 ` [PATCH v2 0/3] cpufreq: Allow drivers to receive more information from the governor Viresh Kumar
2020-12-17 15:26 ` Doug Smythies
2020-12-21 10:41 ` Rafael J. Wysocki
2020-12-18 16:11 ` Giovanni Gherdovich
2020-12-21 16:11 ` Rafael J. Wysocki
2020-12-23 13:06 ` Giovanni Gherdovich
2020-12-28 19:11 ` 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=CAJZ5v0iOvG0PNQDXN00oKCzyZmaF71UB+DJ+zHL5P3xRCAk1tQ@mail.gmail.com \
--to=rafael@kernel.org \
--cc=dsmythies@telus.net \
--cc=ggherdovich@suse.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-pm@vger.kernel.org \
--cc=peterz@infradead.org \
--cc=rjw@rjwysocki.net \
--cc=srinivas.pandruvada@linux.intel.com \
--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;
as well as URLs for NNTP newsgroup(s).