Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755540Ab0DEQkc (ORCPT ); Mon, 5 Apr 2010 12:40:32 -0400 Received: from e1.ny.us.ibm.com ([32.97.182.141]:43394 "EHLO e1.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755289Ab0DEQk0 (ORCPT ); Mon, 5 Apr 2010 12:40:26 -0400 Date: Mon, 5 Apr 2010 09:40:23 -0700 From: "Paul E. McKenney" To: Arjan van de Ven Cc: Dominik Brodowski , Alan Stern , linux-kernel@vger.kernel.org, Thomas Gleixner , Ingo Molnar , Peter Zijlstra , Dmitry Torokhov Subject: Re: A few questions and issues with dynticks, NOHZ and powertop Message-ID: <20100405164023.GD2525@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20100404163924.GA18428@comet.dominikbrodowski.net> <20100404204725.GC2644@linux.vnet.ibm.com> <20100404233702.GA24102@linux.vnet.ibm.com> <4BB95C85.80205@linux.intel.com> <20100405042222.GD2644@linux.vnet.ibm.com> <4BB9F657.4050901@linux.intel.com> <20100405151401.GA2525@linux.vnet.ibm.com> <4BBA0AC5.1080905@linux.intel.com> <20100405162249.GB2525@linux.vnet.ibm.com> <4BBA0E74.2090803@linux.intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4BBA0E74.2090803@linux.intel.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1532 Lines: 33 On Mon, Apr 05, 2010 at 09:23:16AM -0700, Arjan van de Ven wrote: > On 4/5/2010 9:22, Paul E. McKenney wrote: > >On Mon, Apr 05, 2010 at 09:07:33AM -0700, Arjan van de Ven wrote: > >>On 4/5/2010 8:14, Paul E. McKenney wrote: > >>>So the main issue is that for many workloads, it is best to run full bore > >>>and get done quickly, thus allowing the entire machine to be powered down? > >> > >>yep > >> > >>>If so, it seems likely that there would be some workloads that were sometimes > >>>unable to use all the CPUs, in which case shutting down (idling, offlining, > >>>dyntick-idling, whatever) the excess CPUs might nevertheless be the right > >>>thing to do. > >> > >>but the point is that the normal scheduler + idle behavior gives you exactly that > >>in a natural way ! > >>If you don't have enough work (tasks) to keep all cores busy, the others are and stay idle. > > > >So your earlier objection was not to dyntick-idle as such, but rather > >to artificially constraining the scheduler to induce dyntick-idle? > > my objection was against the notion that offlining cpus helps power/energy ;-) Fair enough, at least in general. I should hasten to add that Lai's patch also helps in the case where NR_CPUS is greater than the number of CPUs on the system. Thanx, Paul -- 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/