From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756646AbYCMRA7 (ORCPT ); Thu, 13 Mar 2008 13:00:59 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754245AbYCMRAt (ORCPT ); Thu, 13 Mar 2008 13:00:49 -0400 Received: from mga03.intel.com ([143.182.124.21]:61091 "EHLO mga03.intel.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753022AbYCMRAs convert rfc822-to-8bit (ORCPT ); Thu, 13 Mar 2008 13:00:48 -0400 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.25,495,1199692800"; d="scan'208";a="218229135" X-MimeOLE: Produced By Microsoft Exchange V6.5 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 8BIT Subject: RE: [linux-pm] [PATCH] cpuidle: avoid singing capacitors Date: Thu, 13 Mar 2008 09:47:30 -0700 Message-ID: <924EFEDD5F540B4284297C4DC59F3DEEB33CDD@orsmsx423.amr.corp.intel.com> In-Reply-To: <20080313173437.5f00f70e@mjolnir.drzeus.cx> X-MS-Has-Attach: X-MS-TNEF-Correlator: Thread-Topic: [linux-pm] [PATCH] cpuidle: avoid singing capacitors Thread-Index: AciFKHfZy/8woBidSAyYrG8Kdr6/IwAAQ1DA References: <20080310100008.GA9520@elf.ucw.cz><20080310134915.45ae7446@mjolnir.drzeus.cx><200803121511.17389.lenb@kernel.org> <20080313173437.5f00f70e@mjolnir.drzeus.cx> From: "Pallipadi, Venkatesh" To: "Pierre Ossman" , "Len Brown" Cc: , "Pavel Machek" , "LKML" , "Adam Belay" , "Andi Kleen" , "Lee Revell" X-OriginalArrivalTime: 13 Mar 2008 16:45:42.0686 (UTC) FILETIME=[AD3FEBE0:01C88529] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org >-----Original Message----- >From: linux-kernel-owner@vger.kernel.org >[mailto:linux-kernel-owner@vger.kernel.org] On Behalf Of Pierre Ossman >Sent: Thursday, March 13, 2008 9:35 AM >To: Len Brown >Cc: linux-pm@lists.linux-foundation.org; Pavel Machek; LKML; >Adam Belay; Andi Kleen; Lee Revell >Subject: Re: [linux-pm] [PATCH] cpuidle: avoid singing capacitors > >On Wed, 12 Mar 2008 15:11:17 -0400 >Len Brown wrote: > >> > >> > So for now, the only viable workaround is processor.max_cstate.... >> >> yup, that's why we put it in. Is it insufficient? >> > >For some value of insufficient. It is sub-optimal. I'm hoping >there is a way to enter C3 in a pattern that avoids the noise >and still gives a reduced power usage. > Can you try using /dev/cpu_dma_latency from an user space application. Refer Documentation/pm_qos_interface.txt for usage. With this interface you can change menu governor behavior from userspace, by dynamically allowing it to use C3 or not. I am thinking that something like allowing C3 for few seconds and blocking it for few seconds may help. Thanks, Venki