Re: [PATCH 1/1] cpufreq: pcc-cpufreq: Re-introduce deadband effect to reduce number of frequency changes

From	Viresh Kumar <viresh.kumar@linaro.org>
Newsgroups	linux.kernel
Subject	Re: [PATCH 1/1] cpufreq: pcc-cpufreq: Re-introduce deadband effect to reduce number of frequency changes
Date	2016-10-05 07:30 +0200
Message-ID	<soMWd-4MN-1@gated-at.bofh.it> (permalink)
References	(5 earlier) <si6wG-8E-11@gated-at.bofh.it> <sj9st-7HE-35@gated-at.bofh.it> <sj9st-7HE-33@gated-at.bofh.it> <sjcA2-17g-31@gated-at.bofh.it> <skgrV-10r-67@gated-at.bofh.it>
Organization	linux.* mail to news gateway

Show all headers | View raw

On 22-09-16, 19:54, Andreas Herrmann wrote:
> On Mon, Sep 19, 2016 at 10:39:18PM +0300, Stratos Karafotis wrote:
> > Just an idea. I would split the frequency range (max_freq - min_freq)
> > into ~10 steps. But I'm not familiar with the affected systems and
> > of course I can't prove this is an ideal approach.
> 
> I've modified the pcc-cpufreq specific map_load_to_freq function to do
> just that (map load values to 10 discrete frequency values) instead of
> falling back to the deadband (pre-commit-6393d6-version).

And in that you can remove the deadband frequencies from the table to
get almost the same results, as I already stated in an earlier email.

-- 
viresh

Back to linux.kernel | Previous | Next | Find similar | Unroll thread

Thread

Re: [PATCH 1/1] cpufreq: pcc-cpufreq: Re-introduce deadband effect  to reduce number of frequency changes Viresh Kumar <viresh.kumar@linaro.org> - 2016-10-05 07:30 +0200

csiph-web