From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1756534Ab1I2OIw (ORCPT ); Thu, 29 Sep 2011 10:08:52 -0400 Received: from smtp105.prem.mail.ac4.yahoo.com ([76.13.13.44]:41662 "HELO smtp105.prem.mail.ac4.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1754590Ab1I2OIv (ORCPT ); Thu, 29 Sep 2011 10:08:51 -0400 X-Yahoo-Newman-Property: ymail-3 X-YMail-OSG: 49so3.EVM1nmUimaeOhFLJfUz_sCeVi74bjSMyvImKIfYM1 VBkmpJh8F34vJ1Qkfe4xRAxKj_VuZnci9NAVM.Bnm9fx.z11eykByLdcbMMu vgKNtiwW.KWp25KlcE0Ki2atfkKMSj8yHigM8v9qXKSFkJbdylJ.BgXi7PgG L4uYIRlgK7cCIvjcQM1Sa4OepcAsNS5sMgVmCZ.eFgI3pJSHX.geeNytfMt5 RjIYhzBM0S8lXdSuYE9Q7uf8jTdLg8BNHQRUImFEpQ9AoIrTJBqF_.3lwMyC 5ebx2s4Jft8D9qtl0QHyOAawa3cKvGVqY7DzgJz114IDB20r9tbUkdMApMde 2guXLKAgOiYpils3yVs4KTEPQK3MWk61f5Q3UyivTVrVY3KQQNUAzhIIWFBS 9hqafVLLjZgFWRLct4pApXsw93fejRuctxkfLtcu7OiNWb_GK X-Yahoo-SMTP: _Dag8S.swBC1p4FJKLCXbs8NQzyse1SYSgnAbY0- Date: Thu, 29 Sep 2011 09:08:47 -0500 (CDT) From: Christoph Lameter X-X-Sender: cl@router.home To: Catalin Marinas cc: Eric Dumazet , Huajun Li , "linux-mm@kvack.org" , netdev , linux-kernel , Tejun Heo Subject: Re: Question about memory leak detector giving false positive report for net/core/flow.c In-Reply-To: <20110928172342.GH23559@e102109-lin.cambridge.arm.com> Message-ID: References: <1317054774.6363.9.camel@edumazet-HP-Compaq-6005-Pro-SFF-PC> <20110926165024.GA21617@e102109-lin.cambridge.arm.com> <1317066395.2796.11.camel@edumazet-laptop> <20110928172342.GH23559@e102109-lin.cambridge.arm.com> User-Agent: Alpine 2.00 (DEB 1167 2008-08-23) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Wed, 28 Sep 2011, Catalin Marinas wrote: > I tried this but it's tricky. The problem is that the percpu pointer > returned by alloc_percpu() does not directly point to the per-cpu chunks > and kmemleak would report most percpu allocations as leaks. So far the > workaround is to simply mark the alloc_percpu() objects as never leaking > and at least we avoid false positives in other areas. See the patch > below (note that you have to increase the CONFIG_KMEMLEAK_EARLY_LOG_SIZE > as there are many alloc_percpu() calls before kmemleak is fully > initialised): Seems that kernel.org is out and so tejon wont be seeing these.