Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S263998AbUDVMek (ORCPT ); Thu, 22 Apr 2004 08:34:40 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264000AbUDVMek (ORCPT ); Thu, 22 Apr 2004 08:34:40 -0400 Received: from atrey.karlin.mff.cuni.cz ([195.113.31.123]:8145 "EHLO atrey.karlin.mff.cuni.cz") by vger.kernel.org with ESMTP id S263998AbUDVMei (ORCPT ); Thu, 22 Apr 2004 08:34:38 -0400 Date: Thu, 22 Apr 2004 14:34:38 +0200 From: Jan Kara To: torvalds@osdl.org Cc: akpm@osdl.org, linux-kernel@vger.kernel.org Subject: Fix for ext3 journalled quota Message-ID: <20040422123438.GM3895@atrey.karlin.mff.cuni.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1148 Lines: 28 Hi! I've attached a fix for a problem in ext3 journalled quota patch - the problem is that detecting whether dqput() sleeps was wrong and so we could possibly schedule when holding a spinlock. If you applied the patch for journalled quota please apply also this one. Andrew, can you please apply the patch to your kernel? Thanks Honza diff -ruX /home/jack/.kerndiffexclude linux-2.6.5-2-jquota/fs/dquot.c linux-2.6.5-3-jquota2/fs/dquot.c --- linux-2.6.5-2-jquota/fs/dquot.c 2004-04-21 10:45:30.000000000 +0200 +++ linux-2.6.5-3-jquota2/fs/dquot.c 2004-04-21 10:50:13.000000000 +0200 @@ -642,7 +642,7 @@ /* Return 0 if dqput() won't block (note that 1 doesn't necessarily mean blocking) */ static inline int dqput_blocks(struct dquot *dquot) { - if (atomic_read(&dquot->dq_count) <= 1 && dquot_dirty(dquot)) + if (atomic_read(&dquot->dq_count) <= 1) return 1; return 0; } - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/