From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756571AbZB1Ama (ORCPT ); Fri, 27 Feb 2009 19:42:30 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754296AbZB1AmW (ORCPT ); Fri, 27 Feb 2009 19:42:22 -0500 Received: from fgwmail6.fujitsu.co.jp ([192.51.44.36]:46585 "EHLO fgwmail6.fujitsu.co.jp" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753770AbZB1AmV (ORCPT ); Fri, 27 Feb 2009 19:42:21 -0500 Message-ID: In-Reply-To: <661fca240cb094bcf4d261ccebab9848.squirrel@webmail-b.css.fujitsu.com> References: <20090227140537.390905c4.kamezawa.hiroyu@jp.fujitsu.com> <20090227162928.6225fc80.kamezawa.hiroyu@jp.fujitsu.com> <661fca240cb094bcf4d261ccebab9848.squirrel@webmail-b.css.fujitsu.com> Date: Sat, 28 Feb 2009 09:42:17 +0900 (JST) Subject: Re: [PATCH] change cpuacct usage percpu format v2 From: "KAMEZAWA Hiroyuki" To: "KAMEZAWA Hiroyuki" Cc: "Ken Chen" , "KAMEZAWA Hiroyuki" , "linux-kernel@vger.kernel.org" , "menage@google.com" , "lizf@cn.fujitsu.com" , "balbir@linux.vnet.ibm.com" , mingo@elte.hu, "akpm@linux-foundation.org" User-Agent: SquirrelMail/1.4.16 MIME-Version: 1.0 Content-Type: text/plain;charset=iso-2022-jp Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) Importance: Normal Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org KAMEZAWA Hiroyuki wrote： > Ken Chen wrote： >> On Fri, Feb 27, 2009 at 3:34 PM, KAMEZAWA Hiroyuki >> wrote: >>> "a lot of" ? I talking about cpu hotplug and reading another file as >>> /sys/devices/system/cpu/present every time before reading this file >>> gives much much much more overhead ;) >> >> yes, really a lot. CPU hotplug is an uncommon event. It happens >> perhaps once a day? maybe once an hour? >> > Are you saying the software should have hotplug script and send SIGHUP or > some to reload the present map ? > >> User monitoring process usually reads usage_percpu at fairly high >> rate, say once a sec. At each pass it will need to parse N number of >> CPU index. The overhead is N_CPU * T, where T is time in second >> between cpu hotplug event. Assume T = one day, on a moderate sized >> 64-CPU size machine, the overhead is: >> >> 64 * 86400 : 1, that's like 5.5 million to 1 ratio. To me that is >> *high* overhead. >> > Sounds strange. I can't catch hat you want to say. > Ignore above, I caught, at last. I'll add text to documenation. BTW, current interface to reset cpuacct (write ops) just reset specified level of cpuacct and will not clear other hierarchical levels. Doesn't this behavior confuse software ? Thanks, -Kame