From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754578AbYJaCow (ORCPT ); Thu, 30 Oct 2008 22:44:52 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1753189AbYJaCon (ORCPT ); Thu, 30 Oct 2008 22:44:43 -0400 Received: from e6.ny.us.ibm.com ([32.97.182.146]:37293 "EHLO e6.ny.us.ibm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753174AbYJaCom (ORCPT ); Thu, 30 Oct 2008 22:44:42 -0400 Date: Thu, 30 Oct 2008 08:52:18 -0700 From: "Paul E. McKenney" To: Lai Jiangshan Cc: linux-kernel@vger.kernel.org, mingo@elte.hu, rjw@sisk.pl, dipankar@in.ibm.com, tglx@linuxtronix.de, andi@firstfloor.org Subject: Re: [PATCH] v3 rudimentary tracing for Classic RCU Message-ID: <20081030155218.GA6162@linux.vnet.ibm.com> Reply-To: paulmck@linux.vnet.ibm.com References: <20081007212215.GN6384@linux.vnet.ibm.com> <20081009010846.GA10188@linux.vnet.ibm.com> <20081010034421.GA12449@linux.vnet.ibm.com> <20081013230914.GA27467@linux.vnet.ibm.com> <49005C09.6010805@cn.fujitsu.com> <20081026215930.GC6236@linux.vnet.ibm.com> <20081027215029.GA31664@linux.vnet.ibm.com> <20081027235748.GA3647@linux.vnet.ibm.com> <20081029011633.GA21940@linux.vnet.ibm.com> <4907BCFB.8090600@cn.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4907BCFB.8090600@cn.fujitsu.com> User-Agent: Mutt/1.5.15+20070412 (2007-04-11) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, Oct 29, 2008 at 09:31:39AM +0800, Lai Jiangshan wrote: > Paul E. McKenney wrote: > > > Ported to 2.6.27 to ease testing, and fixed up the formatting a bit. > > I remember quickly scanning through one-line-per-CPU data to find > > problems with rcutree.c, but the expanded names are quite nice -- I > > had to brutally abbreviate to fit each CPU on a single line. > > And I agree that seqfile does work nicely for this. > > > > I am strongly tempted to suggest CSV format output with headers to > > allow the data to be pulled into gnumeric or oocalc -- this would > > make handling large numbers of CPUs much easier. (Quickly donning > > the asbestos suit with titanium pinstripes...) > > Hi, Paul, > My patch is like an example, you do not need keep the > codes that I have wrote. you can rewrite all of it. these files > are only accessed by RCU guys, not users. and some file in /proc > is very simple but easy to be read: /proc/stat, /proc/loadavg. Hello, Jiangshan, Good point -- I do very much appreciate the seqfile tutorial!!! Thanx, Paul