From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760322AbXGERMZ (ORCPT ); Thu, 5 Jul 2007 13:12:25 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1757212AbXGERMR (ORCPT ); Thu, 5 Jul 2007 13:12:17 -0400 Received: from agminet01.oracle.com ([141.146.126.228]:22762 "EHLO agminet01.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757160AbXGERMQ (ORCPT ); Thu, 5 Jul 2007 13:12:16 -0400 In-Reply-To: <1183602311.29435.1.camel@dyn9047017100.beaverton.ibm.com> References: <46832499.40807@us.ibm.com> <5A6608DF-EDC3-43C1-BFB7-58ACBF94DBCB@oracle.com> <46859DE2.9040604@us.ibm.com> <20070703222855.GA6293@mami.zabbo.net> <1183602311.29435.1.camel@dyn9047017100.beaverton.ibm.com> Mime-Version: 1.0 (Apple Message framework v752.3) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <3439F83E-9FBD-4743-B2A9-63009B7087FD@oracle.com> Cc: Linus Torvalds , Andrew Morton , linux-fsdevel , lkml Content-Transfer-Encoding: 7bit From: Zach Brown Subject: Re: [PATCH] dio: remove bogus refcounting BUG_ON Date: Thu, 5 Jul 2007 10:11:20 -0700 To: Badari Pulavarty X-Mailer: Apple Mail (2.752.3) X-Brightmail-Tracker: AAAAAQAAAAI= X-Whitelist: TRUE Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > the BUG_ON(). But unfortunately, our perf. team is able reproduce the > problem. What are they doing to reproduce it? How much setup does it take? > Debug indicated that, the ret2 == 1 :( That could be consistent with the theory that we're racing with the dio struct being freed and reused before it's tested in the BUG_ON() condition. Suparna's suggestion to sample dio->is_async before releasing the refcount and using that in the BUG_ON condition is a good one. - z