From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S2992875AbXDDLuB (ORCPT ); Wed, 4 Apr 2007 07:50:01 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S2992876AbXDDLuB (ORCPT ); Wed, 4 Apr 2007 07:50:01 -0400 Received: from mx3.mail.elte.hu ([157.181.1.138]:51471 "EHLO mx3.mail.elte.hu" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S2992875AbXDDLuA (ORCPT ); Wed, 4 Apr 2007 07:50:00 -0400 Date: Wed, 4 Apr 2007 13:49:01 +0200 From: Ingo Molnar To: Gautham R Shenoy Cc: Srivatsa Vaddagiri , akpm@linux-foundation.org, paulmck@us.ibm.com, torvalds@linux-foundation.org, linux-kernel@vger.kernel.org, Oleg Nesterov , "Rafael J. Wysocki" , dipankar@in.ibm.com, dino@in.ibm.com, masami.hiramatsu.pt@hitachi.com Subject: Re: [RFC] Cpu-hotplug: Using the Process Freezer (try2) Message-ID: <20070404114901.GA8261@elte.hu> References: <20070402061612.GA7072@elte.hu> <20070402092818.GE2456@in.ibm.com> <20070402111828.GA14771@elte.hu> <20070402124200.GA9566@in.ibm.com> <20070402185607.GA2081@elte.hu> <20070403125619.GA32444@in.ibm.com> <20070403141516.GB14900@in.ibm.com> <20070404031502.GA23591@in.ibm.com> <20070404100431.GB19120@elte.hu> <20070404104123.GA14825@in.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070404104123.GA14825@in.ibm.com> User-Agent: Mutt/1.4.2.2i X-ELTE-VirusStatus: clean X-ELTE-SpamScore: -2.0 X-ELTE-SpamLevel: X-ELTE-SpamCheck: no X-ELTE-SpamVersion: ELTE 2.0 X-ELTE-SpamCheck-Details: score=-2.0 required=5.9 tests=BAYES_00 autolearn=no SpamAssassin version=3.1.7 -2.0 BAYES_00 BODY: Bayesian spam probability is 0 to 1% [score: 0.0000] Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org * Gautham R Shenoy wrote: > On Wed, Apr 04, 2007 at 12:04:31PM +0200, Ingo Molnar wrote: > > > > yeah, i think you are right - and the TASK_UNINTERRUPTIBLE thing is > > not only quite complex, it also breaks the symmetry of freezer use > > (sw-suspend obviously cannot freeze uninterruptible tasks). We > > should watch whether the current latency of freezing is good enough > > in practice. > > Ok. Do you have any specific tests in mind which I can run and post > the numbers? > > As of now, I've been stressing the system with (kernbench + ondemand > governor) and timing the hotplug operation. i suspect a fork-intensive application like kernbench should be close to the worst-case already. A more IO-intensive workload would maximize the uninterruptible-sleep latencies perhaps? Ingo