From mboxrd@z Thu Jan 1 00:00:00 1970 From: Peter Zijlstra Subject: Re: [v7 PATCH 0/7]: cpuidle/x86/POWER: Cleanup idle power management code in x86, cleanup drivers/cpuidle/cpuidle.c and introduce cpuidle to POWER. Date: Wed, 07 Oct 2009 15:05:11 +0200 Message-ID: <1254920711.26976.243.camel@twins> References: <20091006152421.GA7278@linux.vnet.ibm.com> <20091006163521.GA10425@linux.vnet.ibm.com> <1254852279.17055.2.camel@laptop> <20091007112648.GC7646@dirshya.in.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20091007112648.GC7646@dirshya.in.ibm.com> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: linuxppc-dev-bounces+glppd-linuxppc64-dev=m.gmane.org@lists.ozlabs.org Errors-To: linuxppc-dev-bounces+glppd-linuxppc64-dev=m.gmane.org@lists.ozlabs.org To: svaidy@linux.vnet.ibm.com Cc: linux-arch@vger.kernel.org, Gautham R Shenoy , Venkatesh Pallipadi , linux-kernel@vger.kernel.org, Paul Mackerras , arun@linux.vnet.ibm.com, Ingo Molnar , linuxppc-dev@lists.ozlabs.org, Arjan van de Ven List-Id: linux-arch.vger.kernel.org On Wed, 2009-10-07 at 16:56 +0530, Vaidyanathan Srinivasan wrote: > * Peter Zijlstra [2009-10-06 20:04:39]: > > > On Tue, 2009-10-06 at 22:05 +0530, Arun R Bharadwaj wrote: > > > > > Also, the per-cpu nature of registration/unregistration of cpuidle > > > has been maintained as ACPI needs this. > > > > Right, so can't we ditch that and have acpi default to the lowest > common > > C-state and warn when various cpus report different C-states? > > Hi Peter, > > As Arjan mentioned previously, the per-cpu registration has to stay > for x86 for now due to legacy ACPI compatibility. Breaking that may > break lot of existing users and we do not have a clean fallback > method. >From what I understood some broken ass bioses report different C state availability on different CPUs in the same SMP system. I'm suggesting to work around that by limiting all CPUs to the subset of C states reported on all CPUs, instead of the current mess. I haven't heard anybody tell me why that wouldn't be possible on x86 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from viefep13-int.chello.at ([62.179.121.33]:62486 "EHLO viefep13-int.chello.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933335AbZJGNQG (ORCPT ); Wed, 7 Oct 2009 09:16:06 -0400 Subject: Re: [v7 PATCH 0/7]: cpuidle/x86/POWER: Cleanup idle power management code in x86, cleanup drivers/cpuidle/cpuidle.c and introduce cpuidle to POWER. From: Peter Zijlstra In-Reply-To: <20091007112648.GC7646@dirshya.in.ibm.com> References: <20091006152421.GA7278@linux.vnet.ibm.com> <20091006163521.GA10425@linux.vnet.ibm.com> <1254852279.17055.2.camel@laptop> <20091007112648.GC7646@dirshya.in.ibm.com> Content-Type: text/plain Date: Wed, 07 Oct 2009 15:05:11 +0200 Message-ID: <1254920711.26976.243.camel@twins> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: linux-arch-owner@vger.kernel.org List-ID: To: svaidy@linux.vnet.ibm.com Cc: Arjan van de Ven , arun@linux.vnet.ibm.com, Joel Schopp , Benjamin Herrenschmidt , Paul Mackerras , Ingo Molnar , Dipankar Sarma , Balbir Singh , Gautham R Shenoy , Venkatesh Pallipadi , linux-kernel@vger.kernel.org, linuxppc-dev@lists.ozlabs.org, linux-arch@vger.kernel.org Message-ID: <20091007130511.C48hJsNzLimTE-ZdWeKi3UkdvFKnfTLlKJu2FMuSoTc@z> On Wed, 2009-10-07 at 16:56 +0530, Vaidyanathan Srinivasan wrote: > * Peter Zijlstra [2009-10-06 20:04:39]: > > > On Tue, 2009-10-06 at 22:05 +0530, Arun R Bharadwaj wrote: > > > > > Also, the per-cpu nature of registration/unregistration of cpuidle > > > has been maintained as ACPI needs this. > > > > Right, so can't we ditch that and have acpi default to the lowest > common > > C-state and warn when various cpus report different C-states? > > Hi Peter, > > As Arjan mentioned previously, the per-cpu registration has to stay > for x86 for now due to legacy ACPI compatibility. Breaking that may > break lot of existing users and we do not have a clean fallback > method. >From what I understood some broken ass bioses report different C state availability on different CPUs in the same SMP system. I'm suggesting to work around that by limiting all CPUs to the subset of C states reported on all CPUs, instead of the current mess. I haven't heard anybody tell me why that wouldn't be possible on x86