Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 14 Dec 2020 16:37:17 +0530
From:   Viresh Kumar <viresh.kumar@linaro.org>
To:     "Rafael J. Wysocki" <rafael@kernel.org>
Cc:     "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
Message-ID: <20201214110717.boz634eih2ymjtju@vireshk-i7>
References: <20360841.iInq7taT2Z@kreacher>
 <1916732.tSaCp9PeQq@kreacher>
 <20201208085146.pzem6t3mt44xwxkm@vireshk-i7>
 <CAJZ5v0idoNOPU5-toOw+uTRvjJz60Ddc2xV7rMQeufY_EW58uQ@mail.gmail.com>
 <20201209051642.ddwgds4gznxt3lfn@vireshk-i7>
 <CAJZ5v0iOvG0PNQDXN00oKCzyZmaF71UB+DJ+zHL5P3xRCAk1tQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAJZ5v0iOvG0PNQDXN00oKCzyZmaF71UB+DJ+zHL5P3xRCAk1tQ@mail.gmail.com>
User-Agent: NeoMutt/20180716-391-311a52
Precedence: bulk

On 09-12-20, 16:32, Rafael J. Wysocki wrote:
> 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.

Right and we end up increasing the frequency here..

> 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.

Sure, I am not against sharing the code path, but all we need is
something like this here:

     if (sugov_cpu_is_busy(sg_cpu) && sg_cpu->util < prev_util)
             sg_cpu->util = prev_util;
     else
             next_f = get_next_freq(sg_policy, sg_cpu->util, sg_cpu->max);

i.e. we don't need to call get_next_freq() in this case at all.

-- 
viresh