From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1751798Ab2A3KBM (ORCPT ); Mon, 30 Jan 2012 05:01:12 -0500 Received: from plane.gmane.org ([80.91.229.3]:34463 "EHLO plane.gmane.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751097Ab2A3KBL (ORCPT ); Mon, 30 Jan 2012 05:01:11 -0500 X-Injected-Via-Gmane: http://gmane.org/ To: linux-kernel@vger.kernel.org From: =?UTF-8?B?SsO2cmctVm9sa2VyIFBlZXR6?= Subject: Re: [PATCH v2] proc: speedup /proc/stat handling Date: Mon, 30 Jan 2012 11:00:58 +0100 Message-ID: <4F266A5A.6050207@web.de> References: <1327075164.12389.31.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> <1327449683.14373.12.camel@edumazet-laptop> <20120125091818.45e00b3c.kamezawa.hiroyu@jp.fujitsu.com> <1327451195.14373.28.camel@edumazet-laptop> <4F264F75.7040601@web.de> <1327915555.2288.18.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-Complaints-To: usenet@dough.gmane.org Cc: linux-kernel@vger.kernel.org, KAMEZAWA Hiroyuki , Andrew Morton , Glauber Costa , Peter Zijlstra , Ingo Molnar , Russell King - ARM Linux , Paul Tuner X-Gmane-NNTP-Posting-Host: p5b3718ae.dip.t-dialin.net User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:8.0) Gecko/20120104 Icedove/8.0 In-Reply-To: <1327915555.2288.18.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Eric Dumazet wrote, on 01/30/12 10:25: > Le lundi 30 janvier 2012 à 09:06 +0100, Jörg-Volker Peetz a écrit : >> Eric Dumazet wrote, on 01/25/12 01:26: >>> Le mercredi 25 janvier 2012 à 09:18 +0900, KAMEZAWA Hiroyuki a écrit : >>> >>>> BTW, what is the reason of this change ? >>>> >>>>> - unsigned size = 4096 * (1 + num_possible_cpus() / 32); >>>>> + unsigned size = 1024 + 128 * num_possible_cpus(); >>>> >>>> I think size of buffer is affected by the number of online cpus. >>>> (Maybe 128 is enough but please add comment why 128 ?) >>>> >>> >>> There is no change, as 4096/32 is 128 bytes per cpu. >>> >> >> Wrong math, only num_possible_cpus() is divided by 32. Thus, >> >> - unsigned size = 4096 * (1 + num_possible_cpus() / 32); >> + unsigned size = 4096 + 128 * num_possible_cpus(); >> >> > > > It is good math, once you take the time to think a bit about it. > > The original question was about the 128 * num_possible_cpus() > > 4096/32 is 128 as I said. > > The 4096 -> 1024 is just taking into account fact that once you do the > correct computations, you dont need initial 4096 value, and 1024 is more > than enough. Thanks for your explanation, that makes sense now. At first, I was mislead by your comment "There is no change" which was apparently related to the second term in the formula. > > Example on a dual core machine : > > # dmesg|grep nr_irq > [ 0.000000] nr_irqs_gsi: 40 > [ 0.000000] NR_IRQS:2304 nr_irqs:712 16 > > size = 1024 + 2*128 + 2*712 = 2704 bytes (rounded to 4096 by kmalloc()) > > # wc -c /proc/stat > 1767 /proc/stat > > Problem with original math was that for a machine with 16 cpus or a > machine with 1 cpu, we ended with the same 4096 value. That was a real > problem. > > If we instead use "unsigned size = 4096 + 128 * num_possible_cpus();" as > you suggest, we would always allocate 2 pages of memory, this is not > needed at all for typical 1/2/4 way machines. > -- Best regards, Jörg-Volker.