Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1762399Ab3DBUxB (ORCPT ); Tue, 2 Apr 2013 16:53:01 -0400 Received: from mail.skyhub.de ([78.46.96.112]:55806 "EHLO mail.skyhub.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1760449Ab3DBUw7 (ORCPT ); Tue, 2 Apr 2013 16:52:59 -0400 Date: Tue, 2 Apr 2013 22:40:14 +0200 From: Borislav Petkov To: Jacob Shin Cc: "Rafael J. Wysocki" , cpufreq@vger.kernel.org, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, Viresh Kumar , Thomas Renninger Subject: Re: [PATCH V3 2/2] cpufreq: AMD "frequency sensitivity feedback" powersave bias for ondemand governor Message-ID: <20130402204014.GH17675@pd.tnic> Mail-Followup-To: Borislav Petkov , Jacob Shin , "Rafael J. Wysocki" , cpufreq@vger.kernel.org, linux-pm@vger.kernel.org, linux-kernel@vger.kernel.org, Viresh Kumar , Thomas Renninger References: <1364926304-1799-1-git-send-email-jacob.shin@amd.com> <1364926304-1799-3-git-send-email-jacob.shin@amd.com> <20130402192352.GC17675@pd.tnic> <20130402200337.GA17919@jshin-Toonie> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20130402200337.GA17919@jshin-Toonie> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1787 Lines: 57 On Tue, Apr 02, 2013 at 03:03:37PM -0500, Jacob Shin wrote: > Well, so this powersave_bias also works as a tunable knob. > > From ondemand side, if /sys/../ondemand/powersave_bias is 0, then we > (AMD sensitivity) don't get called and you get the default ondemand > behavior. > > Like existing powersave_bias, users can tune the value to whatever > they want, to get a specturum of less to more aggressive power savings > vs performance. > > I thought tunable would be more flexible .. out in the field or what > not .. no? Ok, yes, that is the default on current systems which don't have hw feedback. But, on hw with such counters, I think the default should be to use the hw feedback feature so that hardware can already do more informed decisions for users. As Thomas said, I hardly doubt users even know about that knob. So if we can make the freq sensitivity thing work out of the box and without user intervention, then we should strive to do that, no? IOW: if (!powersave_bias) { /* user hasn't touched knob */ if (HAS_FEEDBACK_INTERFACE) od_ops.powersave_bias_target(...); __cpufreq_driver_target(..) else od_ops.powersave_bias_target(..) __cpufreq_driver_target(..) } The only change is that on hw feedback systems, you don't get the old behavior with powersave_bias == 0. Question is, do you even want it all that much but would rather leave the hw do much more informed decisions than the ondemand governor. Hmmm. -- Regards/Gruss, Boris. Sent from a fat crate under my desk. Formatting is fine. -- -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/