From mboxrd@z Thu Jan 1 00:00:00 1970 From: Leonard Crestez Subject: Re: [PATCH V3 3/9] cpufreq: Cap the default transition delay value to 10 ms Date: Wed, 16 Aug 2017 12:42:45 +0300 Message-ID: <1502876565.23210.15.camel@nxp.com> References: <1b93c94cb8b4914314e4f50304c3cb11c53d8b14.1500373914.git.viresh.kumar@linaro.org> <1500983686.30745.28.camel@nxp.com> <20170726060634.GY352@vireshk-i7> <1501174453.7957.30.camel@nxp.com> <20170728052843.GT352@vireshk-i7> <20170816063431.GB24299@vireshk-i7> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Return-path: Received: from mail-bn3nam01on0057.outbound.protection.outlook.com ([104.47.33.57]:18560 "EHLO NAM01-BN3-obe.outbound.protection.outlook.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751666AbdHPJmx (ORCPT ); Wed, 16 Aug 2017 05:42:53 -0400 In-Reply-To: <20170816063431.GB24299@vireshk-i7> Sender: linux-pm-owner@vger.kernel.org List-Id: linux-pm@vger.kernel.org To: Viresh Kumar , Rafael Wysocki , Shawn Guo , Sascha Hauer , Fabio Estevam Cc: linux-pm@vger.kernel.org, Vincent Guittot , linux@dominikbrodowski.net, linux-kernel@vger.kernel.org On Wed, 2017-08-16 at 12:04 +0530, Viresh Kumar wrote: > On 28-07-17, 10:58, Viresh Kumar wrote: > > > > At this point I really feel that this is a hardware specific problem > > and it was working by chance until now. And I am not sure if we > > shouldn't be stopping this patch from getting merged just because of > > that. > > > > At least you can teach your distribution to go increase the sampling > > rate from userspace to make it all work. > Its been 3 weeks since my last email on this thread and no reply yet > from any of the IMX maintainers. Can someone please help here ? > > @Shawn: Can you help debugging a bit here, to see what's get screwed > up due to this commit ? Its just that your platform isn't able to > change freq at 10 ms rate. > > @Rafael: I am not sure, but should we be stopping this patch because > some hardware isn't able to change freq at 10ms interval and is just > faking the transition delay to start with ? > > Maybe we get this merged again and the IMX guys can figure out what's > wrong on their platform and how to fix it ? I reported the initial issue but did not have the time to do a more thorough investigation, this is more complicated than it seems. I said this before but maybe it got lost: I don't think the odd behavior I noticed justifies keeping the patch from merging. -- Regards, Leonrd