MIME-Version: 1.0
References: <alpine.DEB.2.22.394.2112132215060.215073@hadrien>
 <CAJZ5v0iBU8gw8+-5nxj2cKzf7tyN=p3Adcm4Z5bn=oVYhU28bQ@mail.gmail.com>
 <alpine.DEB.2.22.394.2112172022100.2968@hadrien> <87r1abt1d2.fsf@riseup.net>
 <alpine.DEB.2.22.394.2112172258480.2968@hadrien> <87fsqqu6by.fsf@riseup.net>
 <alpine.DEB.2.22.394.2112180654470.3139@hadrien> <878rwitdu3.fsf@riseup.net>
 <alpine.DEB.2.22.394.2112181138210.3130@hadrien> <871r29tvdj.fsf@riseup.net>
 <alpine.DEB.2.22.394.2112190734070.3181@hadrien> <87wnk0s0tf.fsf@riseup.net>
In-Reply-To: <87wnk0s0tf.fsf@riseup.net>
From:   "Rafael J. Wysocki" <rafael@kernel.org>
Date:   Tue, 21 Dec 2021 18:04:36 +0100
Message-ID: <CAJZ5v0i7gBtm6x+zUUzhxXjmYhPwr=JxvOuMZ0aD9qxnjE9YKw@mail.gmail.com>
Subject: Re: cpufreq: intel_pstate: map utilization into the pstate range
To:     Francisco Jerez <currojerez@riseup.net>
Cc:     Julia Lawall <julia.lawall@inria.fr>,
        "Rafael J. Wysocki" <rafael@kernel.org>,
        Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>,
        Len Brown <lenb@kernel.org>,
        Viresh Kumar <viresh.kumar@linaro.org>,
        Linux PM <linux-pm@vger.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Juri Lelli <juri.lelli@redhat.com>,
        Vincent Guittot <vincent.guittot@linaro.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sun, Dec 19, 2021 at 11:10 PM Francisco Jerez <currojerez@riseup.net> wrote:
>
> Julia Lawall <julia.lawall@inria.fr> writes:
>
> > On Sat, 18 Dec 2021, Francisco Jerez wrote:

[cut]

> > I did some experiements with forcing different frequencies.  I haven't
> > finished processing the results, but I notice that as the frequency goes
> > up, the utilization (specifically the value of
> > map_util_perf(sg_cpu->util) at the point of the call to
> > cpufreq_driver_adjust_perf in sugov_update_single_perf) goes up as well.
> > Is this expected?
> >
>
> Actually, it *is* expected based on our previous hypothesis that these
> workloads are largely latency-bound: In cases where a given burst of CPU
> work is not parallelizable with any other tasks the thread needs to
> complete subsequently, its overall runtime will decrease monotonically
> with increasing frequency, therefore the number of instructions executed
> per unit of time will increase monotonically with increasing frequency,
> and with it its frequency-invariant utilization.

But shouldn't these two effects cancel each other if the
frequency-invariance mechanism works well?