From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752578AbZIYVPM (ORCPT ); Fri, 25 Sep 2009 17:15:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752469AbZIYVPM (ORCPT ); Fri, 25 Sep 2009 17:15:12 -0400 Received: from gate.crashing.org ([63.228.1.57]:41032 "EHLO gate.crashing.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751823AbZIYVPL (ORCPT ); Fri, 25 Sep 2009 17:15:11 -0400 Subject: Re: [PATCH v3 0/3] cpu: pseries: Cpu offline states framework From: Benjamin Herrenschmidt To: Peter Zijlstra Cc: Gautham R Shenoy , Joel Schopp , Balbir Singh , Venkatesh Pallipadi , Dipankar Sarma , Vaidyanathan Srinivasan , Arun R Bharadwaj , linuxppc-dev@lists.ozlabs.org, linux-kernel@vger.kernel.org, "Darrick J. Wong" In-Reply-To: <1253890120.18939.189.camel@laptop> References: <20090915120629.20523.79019.stgit@sofia.in.ibm.com> <1253016701.5506.73.camel@laptop> <1253753501.7103.358.camel@pasglop> <1253890120.18939.189.camel@laptop> Content-Type: text/plain Date: Sat, 26 Sep 2009 07:12:48 +1000 Message-Id: <1253913169.7103.529.camel@pasglop> Mime-Version: 1.0 X-Mailer: Evolution 2.26.1 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, 2009-09-25 at 16:48 +0200, Peter Zijlstra wrote: > On Thu, 2009-09-24 at 10:51 +1000, Benjamin Herrenschmidt wrote: > > On Tue, 2009-09-15 at 14:11 +0200, Peter Zijlstra wrote: > > > I still think its a layering violation... its the hypervisor manager > > > that should be bothered in what state an off-lined cpu is in. > > > > > That's not how our hypervisor works. > > Then fix it? Are you serious ? :-) > CPU hotplug is terribly invasive and expensive to the kernel, doing > hotplug on a minute basis is just plain crazy. > > If you want a CPU in a keep it near and don't hand it back to the HV > state, why not use cpusets to isolate it and simply not run tasks on it? > > cpusets don't use stopmachine and are much nicer to the rest of the > kernel over-all. Gautham, what is the different in term of power saving between having it idle for long periods of time (which could do H_CEDE and with NO_HZ, probably wouln't need to wake up that often) and having it unplugged in a H_CEDE loop ? Ben. From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from gate.crashing.org (gate.crashing.org [63.228.1.57]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (Client did not present a certificate) by ozlabs.org (Postfix) with ESMTPS id 0B178B7BC6 for ; Sat, 26 Sep 2009 07:15:10 +1000 (EST) Subject: Re: [PATCH v3 0/3] cpu: pseries: Cpu offline states framework From: Benjamin Herrenschmidt To: Peter Zijlstra In-Reply-To: <1253890120.18939.189.camel@laptop> References: <20090915120629.20523.79019.stgit@sofia.in.ibm.com> <1253016701.5506.73.camel@laptop> <1253753501.7103.358.camel@pasglop> <1253890120.18939.189.camel@laptop> Content-Type: text/plain Date: Sat, 26 Sep 2009 07:12:48 +1000 Message-Id: <1253913169.7103.529.camel@pasglop> Mime-Version: 1.0 Cc: Gautham R Shenoy , linux-kernel@vger.kernel.org, Venkatesh Pallipadi , Arun R Bharadwaj , linuxppc-dev@lists.ozlabs.org, "Darrick J. Wong" List-Id: Linux on PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Fri, 2009-09-25 at 16:48 +0200, Peter Zijlstra wrote: > On Thu, 2009-09-24 at 10:51 +1000, Benjamin Herrenschmidt wrote: > > On Tue, 2009-09-15 at 14:11 +0200, Peter Zijlstra wrote: > > > I still think its a layering violation... its the hypervisor manager > > > that should be bothered in what state an off-lined cpu is in. > > > > > That's not how our hypervisor works. > > Then fix it? Are you serious ? :-) > CPU hotplug is terribly invasive and expensive to the kernel, doing > hotplug on a minute basis is just plain crazy. > > If you want a CPU in a keep it near and don't hand it back to the HV > state, why not use cpusets to isolate it and simply not run tasks on it? > > cpusets don't use stopmachine and are much nicer to the rest of the > kernel over-all. Gautham, what is the different in term of power saving between having it idle for long periods of time (which could do H_CEDE and with NO_HZ, probably wouln't need to wake up that often) and having it unplugged in a H_CEDE loop ? Ben.