From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751760AbdGBHYa (ORCPT ); Sun, 2 Jul 2017 03:24:30 -0400 Received: from isilmar-4.linta.de ([136.243.71.142]:39924 "EHLO isilmar-4.linta.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750771AbdGBHY2 (ORCPT ); Sun, 2 Jul 2017 03:24:28 -0400 Date: Sun, 2 Jul 2017 09:23:50 +0200 From: Dominik Brodowski To: Viresh Kumar Cc: Rafael Wysocki , linux-pm@vger.kernel.org, Vincent Guittot , linux-doc@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH 3/6] cpufreq: governor: Drop min_sampling_rate Message-ID: <20170702072350.GB24454@light.dominikbrodowski.net> References: <713af1a417a9a77f0c41976b25874687ac235e8e.1498733506.git.viresh.kumar@linaro.org> <20170629180123.GA2443@light.dominikbrodowski.net> <20170630033425.GU29665@vireshk-i7> <20170630045306.GA8069@light.dominikbrodowski.net> <20170630054033.GZ29665@vireshk-i7> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="bg08WKrSYDhXBjb5" Content-Disposition: inline In-Reply-To: <20170630054033.GZ29665@vireshk-i7> User-Agent: Mutt/1.8.3 (2017-05-23) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org --bg08WKrSYDhXBjb5 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Fri, Jun 30, 2017 at 11:10:33AM +0530, Viresh Kumar wrote: > On 30-06-17, 06:53, Dominik Brodowski wrote: > > On Fri, Jun 30, 2017 at 09:04:25AM +0530, Viresh Kumar wrote: > > > On 29-06-17, 20:01, Dominik Brodowski wrote: > > > > On Thu, Jun 29, 2017 at 04:29:06PM +0530, Viresh Kumar wrote: > > > > > The cpufreq core and governors aren't supposed to set a limit on = how > > > > > fast we want to try changing the frequency. This is currently don= e for > > > > > the legacy governors with help of min_sampling_rate. > > > > >=20 > > > > > At worst, we may end up setting the sampling rate to a value lowe= r than > > > > > the rate at which frequency can be changed and then one of the CP= Us in > > > > > the policy will be only changing frequency for ever. > > > >=20 > > > > Is it safe to issue requests to change the CPU frequency so frequen= tly, > > >=20 > > > Well, I assumed so. I am not sure the hardware would break though. > > > Overheating ? > > >=20 > > > > even > > > > on historic hardware such as speedstep-{ich,smi,centrino}? In the p= ast, >=20 > speedstep-smi is the only one which sets transition_latency to > CPUFREQ_ETERNAL and the others are putting some meaningful values. So > yes, they should be doing DVFS dynamically. >=20 > > > > these checks more or less disallowed the running of dynamic frequen= cy > > > > scaling at least on speedstep-smi[*], > > >=20 > > > We must by doing dynamic freq scaling even without this patch. I don't > > > see why you say the above then. > > >=20 > > > All we do here is that we get rid of the limit on how soon we can > > > change the freq again. > >=20 > > Well, as I understand it, first generation "speedstep" was designed mor= e or > > less to switch frequencies only when AC power was lost or restored. > >=20 > > The Linux implementation merely said: "no on-the-fly changes", but swit= ch > > frequencies whenever a user explicitly requested such a change (presuma= bly > > only every once in an unspecified while). > >=20 > > This same reasoning may be present in other drivers using CPUFREQ_ETERN= AL. >=20 > Thanks for the explanation here and I am convinced that this series > has at least done one thing wrong. And that is removal of > max_transition_latency from governors and allowing ondemand to run on > such platforms (which may end up breaking them). >=20 > So I will actually modify that patch and set max_transition_latency to > CPUFREQ_ETERNAL for ondemand/conservative instead of 10ms. Also we > should do the same for schedutil as well, so that will also use the > max_transition_latency field. >=20 > But I hope, this patch will still be fine. Right ? Indeed, I have no comments otherwise. Thanks! Best Dominik --bg08WKrSYDhXBjb5 Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEmgXaWKgmjrvkPhLCmpdgiUyNow0FAllYn3sACgkQmpdgiUyN ow0etRAAtQHgq1kN5qmw22w+zDaphoUDb4EgGDGZQg5+43u6znamZAG1bbiy8aKn dzvvnfv0StRqk91ch8cdKeYxeouLyrpu27Wnfkt3AClugsp5lM5UplG+pFrTQFYR 4oPbvBMVh1jjApkkwGlqabQXqIWO9eQ+Oc5dBbokUGtVUgktOMX/iQH0fJ5n0UIf QrEP146aldKXuBXQZObIF2kC3/gqSnAqEOTZvKGfmmAGPG2SRN/lnnjuCRDj3xy2 FOIZFCouH0LTTTYxpVuKHdPo40HtezZ9BuKiV1Ej1l8OAgTBwiMZsh0flfKCxP1v fa4jP495SUsXNnZ58HRIaLrpokGU1De1wGQurAD6ViNVNPlZjfjueWSloO2m9xzo DsLuQ2LXx2av6cICUUPN/GWz856DuMLmi3oI8CCJL2voSSmxbSpveC06E5sjOBEc nUKsyYlc36+2n8PdBSluzxXqXdVqLnGK6glcgCLjUPneLldE4Kq/B1HujL683yUC aopepE8+iUmrAYtmjxNozZhczqLxY7ME6ViCTPrudvGn5VHdxwzN6ohJK/V7F0Fc k/nOGvwxVNbbBz1JzQ6RuNFZXllutOROgWwOo4hE352F8PpNfpQxYyfZVsZS33J6 kIUMGMlCdPYPBQkBciZjptw36DG2OkmYsuljULKSQ9AcoDLZ/bE= =CMCa -----END PGP SIGNATURE----- --bg08WKrSYDhXBjb5--