From mboxrd@z Thu Jan 1 00:00:00 1970 From: Len Brown Subject: Re: RFC: /sys/power/policy_preference Date: Thu, 17 Jun 2010 14:44:41 -0400 (EDT) Message-ID: References: Mime-Version: 1.0 Content-Type: MULTIPART/MIXED; BOUNDARY="8323328-1729434706-1276800281=:4214" Return-path: In-reply-to: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linux-pm-bounces@lists.linux-foundation.org Errors-To: linux-pm-bounces@lists.linux-foundation.org To: Jason Chagas Cc: Linux Power Management List List-Id: linux-pm@vger.kernel.org This message is in MIME format. The first part should be readable text, while the remaining parts are likely unreadable without MIME-aware tools. --8323328-1729434706-1276800281=:4214 Content-Type: TEXT/PLAIN; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable X-MIME-Autoconverted: from 8bit to quoted-printable by smtp1.linux-foundation.org id o5HJpspH018258 On Wed, 16 Jun 2010, Jason Chagas wrote: > > powersave (can sacrifice measurable performance) > > =A0 =A0 =A0 =A0scheduler: enable sched_smt_power_savings > > =A0 =A0 =A0 =A0cpuidle: enable all C-states, subject to QOS > > =A0 =A0 =A0 =A0ondemand: disable turbo mode, powersave_bias=3D10 > > =A0 =A0 =A0 =A0msr_ia32_energy_perf_bias: 11 of 15 > > >=20 > Do you have any pointers to info (or code) on how the different > C&P-states are selected within each level? I'm not proposing any changes to how the cpuidle menu govenor selects C-states, or how it obeys PM_QOS, except for 1 thing. I am proposing that in "max_performance" mode that C-states be disabled entirely, to avoid any state latency whatsoever. thanks, Len Brown, Intel Open Source Technology Center --8323328-1729434706-1276800281=:4214 Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Disposition: inline Content-Transfer-Encoding: 7bit --8323328-1729434706-1276800281=:4214--