From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754156AbXGFXnZ (ORCPT ); Fri, 6 Jul 2007 19:43:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1752467AbXGFXnS (ORCPT ); Fri, 6 Jul 2007 19:43:18 -0400 Received: from bc.sympatico.ca ([209.226.175.184]:39042 "EHLO tomts22-srv.bellnexxia.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752401AbXGFXnR (ORCPT ); Fri, 6 Jul 2007 19:43:17 -0400 Date: Fri, 6 Jul 2007 19:43:15 -0400 From: Mathieu Desnoyers To: Adrian Bunk Cc: Chuck Ebbert , Andi Kleen , Andrew Morton , Alexey Dobriyan , linux-kernel@vger.kernel.org, mbligh@google.com Subject: Re: [patch 10/10] Scheduler profiling - Use immediate values Message-ID: <20070706234315.GA19804@Krystal> References: <20070703164046.645090494@polymtl.ca> <20070703164516.377240547@polymtl.ca> <20070703181151.GB5800@martell.zuzino.mipt.ru> <20070703185748.GA4047@Krystal> <20070705132120.8edbc1f3.akpm@linux-foundation.org> <468EBEB2.4070605@redhat.com> <20070706232843.GT3492@stusta.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <20070706232843.GT3492@stusta.de> X-Editor: vi X-Info: http://krystal.dyndns.org:8080 X-Operating-System: Linux/2.6.21.3-grsec (i686) X-Uptime: 19:40:28 up 5 days, 18:23, 6 users, load average: 0.02, 0.17, 0.19 User-Agent: Mutt/1.5.13 (2006-08-11) Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org * Adrian Bunk (bunk@stusta.de) wrote: > On Fri, Jul 06, 2007 at 06:14:10PM -0400, Chuck Ebbert wrote: > > On 07/06/2007 07:44 AM, Andi Kleen wrote: > > > I think the optimization is a good idea, although i dislike it > > > that it is complicated for the dynamic markers. If it was just > > > static it would be much simpler. > > > > Another thing to consider is that there might be hundreds of these > > probes/tracepoints active in an instrumented kernel. The overhead > > adds up fast, so the gain may be worth all the pain. > > Only if you want to squeeze the last bit of performance out of > _debugging_ functionality. > > You avoid all the pain if you simply don't use debugging functionality > on production systems. > Adrian, Please have a look at my markers posts, especially: http://www.ussg.iu.edu/hypermail/linux/kernel/0707.0/0669.html And also look into OLS 2007 proceedings for Martin Bligh's paper on Debugging Google sized clusters. It basically makes the case for adding functionnality to debug _user space_ problems on production systems that can be turned on dynamically. Mathieu > cu > Adrian > > -- > > "Is there not promise of rain?" Ling Tan asked suddenly out > of the darkness. There had been need of rain for many days. > "Only a promise," Lao Er said. > Pearl S. Buck - Dragon Seed > -- Mathieu Desnoyers Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F BA06 3F25 A8FE 3BAE 9A68