From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Renninger Subject: Re: [ANNOUNCE] cpupowerutils - cpufrequtils extended with quite some features Date: Sat, 12 Mar 2011 00:53:53 +0100 Message-ID: <201103120053.53565.trenn@suse.de> References: <201103111247.00212.trenn@suse.de> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: Sender: cpufreq-owner@vger.kernel.org List-ID: Content-Type: Text/Plain; charset="us-ascii" To: Corentin Chary Cc: linux-pm@lists.linux-foundation.org, discuss@lesswatts.org, cpufreq@vger.kernel.org Hi, I excluded some people/lists... On Friday, March 11, 2011 04:56:01 PM Corentin Chary wrote: ... > > The tool should compile and work on as much architectures as > > possible. > > Hi Thomas, > Do you think handling really vendor specific "boosts" like EeePC's SHE > is in cpupower scope ? Not really. > Still, it would not solve the issue of setting SHE mode in a standard > way, or associating it with a governor. I am not familiar with this stuff. This is an oem/vendor specific way to modify the front-side-bus throughput on (Intel atom only?) eeepc like machines (acer, asus and others?)? You could try to experiment with registering your own delayed work queue in eeepc-laptopc.c and try to find best tunings and fittings for such requirements. If it works out and/or other xy-laptop.c or whatever drivers provide similar functionality it could get moved up into a more generic framework. A separate one next to cpuidle and cpufreq, for example fsb_mod? -> then cpupower should get enhanced to show its statistics, provide easy knobs to tune things and provide a manpage what this is all about... Thomas PS: Wouldn't the cpufreq subsystem suffice to register for this SHE thing? Could it be that SHE and real CPU freq (speedstep) is mutual exclusive? Is the only problem that p4_clockmod cannot be loaded then? I know it's needed to "unlock" certain eeepcs, but this not typical? Are there any statistics how much p4_clockmod saves power on such machines? Thomas