From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from pop-mail.india.tejasnetworks.com (india.tejasnetworks.com [164.164.94.83]) by ozlabs.org (Postfix) with SMTP id 5AFBF67A65 for ; Fri, 11 Feb 2005 17:29:44 +1100 (EST) Content-Type: text/plain; charset="iso-8859-1" From: robin To: Gerhard Jaeger , linuxppc-embedded@ozlabs.org, etux@embeddedtux.org Date: Fri, 11 Feb 2005 11:59:04 +0530 References: <20050204192117.4F29CC108D@atlas.denx.de> <200502101137.07711.robin@india.tejasnetworks.com> <200502100840.38232.g.jaeger@sysgo.com> In-Reply-To: <200502100840.38232.g.jaeger@sysgo.com> MIME-Version: 1.0 Message-Id: <200502111159.04265.robin@india.tejasnetworks.com> Subject: Memory leak problem List-Id: Linux on Embedded PowerPC Developers Mail List List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Hi=20 Thanks for your help.=20 We found the memory leak in the syncppp module in linux. 1. Did somebody encounter any problems(leaks) with the=20 syncppp module in denx 2.4.20 kernel??=20 2. Are there any known issues with this module?? The scenario is we are ftp-ing continuosly between=20 two nodes using syncppp module. The leak is ocuring at=20 both the nodes. Regards, Robin On Thursday 10 Feb 2005 1:10 pm, Gerhard Jaeger wrote: > On Thursday 10 February 2005 07:07, robin wrote: > > Hi > > We have an application running on denx linux kernel 2.4.20. > > > > Apparently there is a memory leak that we are not able to find. > > The application is giving out of memory error. Even if we restart > > the application without rebooting the kernel, (killing all applicatio= n > > processes and restarting), it gives the same out of memory error. > > Rebooting the kernel solves the problem. > > > > It seems that the memory leak is not at the application level > > 'cos restarting the application doesnt solve the problem.. > > > > 1. Any ideas where I should look for the leak?? > > Check /proc/slabinfo and see if there are some "supicious" buffer, > that only increase and never decrease. From the kind of buffer > showing this behaviour you might find out where to dig... > Check it before running the application and afterwards. > > > 2. Is there any known memory leak problems with the kernel?? > > There are enough places, where a kernel could leak, but in general > such leaks in the main part are detected really fast. Undiscovered > leaks are often found in the drivers itself. > > What's your kernel config? > Which HW are you using? > What does this application? > > Ciao, > Gerhard --=20 And I hate redundancy, and having different functions for the same thing. =09- Linus Torvalds on linux-kernel