From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755138AbYEPJCa (ORCPT ); Fri, 16 May 2008 05:02:30 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1750845AbYEPJCW (ORCPT ); Fri, 16 May 2008 05:02:22 -0400 Received: from mail.numalliance.com ([213.41.233.134]:65133 "EHLO mail.numalliance.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750770AbYEPJCU (ORCPT ); Fri, 16 May 2008 05:02:20 -0400 Message-ID: <482D4CC8.2050501@free.fr> Date: Fri, 16 May 2008 10:58:48 +0200 From: =?ISO-8859-1?Q?St=E9phane_ANCELOT?= User-Agent: Thunderbird 2.0.0.14 (X11/20080513) MIME-Version: 1.0 To: Christoph Lameter Cc: Pekka Enberg , linux-kernel@vger.kernel.org Subject: Re: detecting kernel mem leak References: <1209462094.4816ed4e93977@imp.free.fr> <84144f020804290546h666983aci77fbb08eb507e777@mail.gmail.com> <4821C526.8030901@free.fr> <4829566B.7090804@free.fr> In-Reply-To: X-Enigmail-Version: 0.95.6 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Christoph Lameter a écrit : > On Tue, 13 May 2008, Stéphane ANCELOT wrote: > > >> I kept my kernel running with few applications for 5 days , doing >> nothing more than backing up few kb of data on disk and refresh few X apps. >> >> Ater five days the global memory available go down from 24Mb to 8Mb ... >> > > That is normal. Linux tries to put all memory to use and will free on > demand. > > >> The are some signifiant changes in slabinfo but now, I do not know where >> to search ? >> > > Compile the slabinfo tool. > > gcc -o slabinfo linux/Documentation/vm/slabinfo.c > > Then you can do > > slabinfo -T > > to get an overview of how much is used by slabs. But I do not see that > slabs are using an excessive amount. So toying around with slabinfo is > not going to get you anywhere. > > 1) slabinfo tells me SYSFS support for SLUB not active In the kernel, there is SLAB or SLUB , my kernel is at this time configured for SLAB allocator. it is documented SLUB minimizes cache line usage. Do you think I have to switch to SLUB ? 2) regarding memory debugging, your reply and some mesages told it was normal the memory was growing (with ext3 buffer_heads...) and released on demand. This sounds to me it becomes VERY VERY difficult telling if my system is STABLE or NOT. Is there a way to bypass it ? I assume I have to do some kind of small program trying to allocate almost the full remaining memory available at startup to empty caches ? Best Regards Steph