From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by oss.sgi.com (8.14.3/8.14.3/SuSE Linux 0.8) with ESMTP id n2FMLBmE042296 for ; Sun, 15 Mar 2009 17:21:31 -0500 Received: from ipmail05.adl2.internode.on.net (localhost [127.0.0.1]) by cuda.sgi.com (Spam Firewall) with ESMTP id 8B0781C448AE for ; Sun, 15 Mar 2009 15:20:48 -0700 (PDT) Received: from ipmail05.adl2.internode.on.net (ipmail05.adl2.internode.on.net [203.16.214.145]) by cuda.sgi.com with ESMTP id mlO7bf1sMpNR8IfF for ; Sun, 15 Mar 2009 15:20:48 -0700 (PDT) Date: Mon, 16 Mar 2009 09:19:21 +1100 From: Dave Chinner Subject: Re: XFS lock warning, 2.6.29-rc8 Message-ID: <20090315221921.GY26138@disturbed> References: <200903151459.01320.denys@visp.net.lb> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <200903151459.01320.denys@visp.net.lb> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: xfs-bounces@oss.sgi.com Errors-To: xfs-bounces@oss.sgi.com To: Denys Fedoryschenko Cc: linux-kernel@vger.kernel.org, xfs@oss.sgi.com [ changed cc to xfs@oss.sgi.com. linux-xfs@vger.kernel.org does not exist. ] On Sun, Mar 15, 2009 at 02:59:01PM +0200, Denys Fedoryschenko wrote: > hi > > Here is what i got suddently. But seems everything is working. > If any other information required, please let me know. Known issue. > [39993.560234] the existing dependency chain (in reverse order) is: > [39993.560236] > [39993.560236] -> #1 (&mm->mmap_sem){----}: > [39993.560240] [] __lock_acquire+0x12b0/0x161a > [39993.560244] [] lock_acquire+0x55/0x71 > [39993.560247] [] down_read+0x34/0x41 > [39993.560252] [] do_page_fault+0x3fa/0x848 > [39993.560256] [] page_fault+0x1f/0x30 > [39993.560258] [] generic_file_aio_read+0x35f/0x5da > [39993.560268] [] xfs_read+0x17d/0x1f5 > [39993.560272] [] xfs_file_aio_read+0x5f/0x61 > [39993.560275] [] do_sync_read+0xe7/0x12d > [39993.560284] [] vfs_read+0xab/0x134 > [39993.560287] [] sys_read+0x47/0x6f > [39993.560289] [] system_call_fastpath+0x16/0x1b > [39993.560293] [] 0xffffffffffffffff > [39993.560302] > [39993.560302] -> #0 (&(&ip->i_iolock)->mr_lock){----}: > [39993.560306] [] __lock_acquire+0xf7f/0x161a > [39993.560309] [] lock_acquire+0x55/0x71 > [39993.560312] [] down_write_nested+0x37/0x46 > [39993.560316] [] xfs_ilock+0x27/0x79 > [39993.560318] [] xfs_inactive+0x14d/0x411 > [39993.560322] [] xfs_fs_clear_inode+0x83/0x85 > [39993.560325] [] clear_inode+0x8a/0xe3 > [39993.560329] [] generic_delete_inode+0xf8/0x173 > [39993.560332] [] generic_drop_inode+0x17/0x1d5 > [39993.560335] [] iput+0x61/0x65 > [39993.560338] [] dentry_iput+0xa9/0xc1 > [39993.560341] [] d_kill+0x40/0x60 > [39993.560344] [] dput+0x141/0x14e > [39993.560347] [] __fput+0x173/0x198 > [39993.560350] [] fput+0x18/0x1a > [39993.560353] [] remove_vma+0x3e/0x63 > [39993.560355] [] do_munmap+0x2e9/0x30b > [39993.560358] [] sys_munmap+0x40/0x59 > [39993.560361] [] system_call_fastpath+0x16/0x1b > [39993.560363] [] 0xffffffffffffffff This is a VM problem where it calls fput() with the mmap_sem() held in remove_vma(). It makes the incorrect assumption that filesystems will never use the same lock in the IO path and the inode release path. This can deadlock if you are really unlucky. Cheers, Dave. -- Dave Chinner david@fromorbit.com _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759521AbZCOWVB (ORCPT ); Sun, 15 Mar 2009 18:21:01 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754963AbZCOWUw (ORCPT ); Sun, 15 Mar 2009 18:20:52 -0400 Received: from ipmail05.adl2.internode.on.net ([203.16.214.145]:55053 "EHLO ipmail05.adl2.internode.on.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752475AbZCOWUv (ORCPT ); Sun, 15 Mar 2009 18:20:51 -0400 Date: Mon, 16 Mar 2009 09:19:21 +1100 From: Dave Chinner To: Denys Fedoryschenko Cc: xfs@oss.sgi.com, linux-kernel@vger.kernel.org Subject: Re: XFS lock warning, 2.6.29-rc8 Message-ID: <20090315221921.GY26138@disturbed> Mail-Followup-To: Denys Fedoryschenko , xfs@oss.sgi.com, linux-kernel@vger.kernel.org References: <200903151459.01320.denys@visp.net.lb> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200903151459.01320.denys@visp.net.lb> User-Agent: Mutt/1.5.18 (2008-05-17) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org [ changed cc to xfs@oss.sgi.com. linux-xfs@vger.kernel.org does not exist. ] On Sun, Mar 15, 2009 at 02:59:01PM +0200, Denys Fedoryschenko wrote: > hi > > Here is what i got suddently. But seems everything is working. > If any other information required, please let me know. Known issue. > [39993.560234] the existing dependency chain (in reverse order) is: > [39993.560236] > [39993.560236] -> #1 (&mm->mmap_sem){----}: > [39993.560240] [] __lock_acquire+0x12b0/0x161a > [39993.560244] [] lock_acquire+0x55/0x71 > [39993.560247] [] down_read+0x34/0x41 > [39993.560252] [] do_page_fault+0x3fa/0x848 > [39993.560256] [] page_fault+0x1f/0x30 > [39993.560258] [] generic_file_aio_read+0x35f/0x5da > [39993.560268] [] xfs_read+0x17d/0x1f5 > [39993.560272] [] xfs_file_aio_read+0x5f/0x61 > [39993.560275] [] do_sync_read+0xe7/0x12d > [39993.560284] [] vfs_read+0xab/0x134 > [39993.560287] [] sys_read+0x47/0x6f > [39993.560289] [] system_call_fastpath+0x16/0x1b > [39993.560293] [] 0xffffffffffffffff > [39993.560302] > [39993.560302] -> #0 (&(&ip->i_iolock)->mr_lock){----}: > [39993.560306] [] __lock_acquire+0xf7f/0x161a > [39993.560309] [] lock_acquire+0x55/0x71 > [39993.560312] [] down_write_nested+0x37/0x46 > [39993.560316] [] xfs_ilock+0x27/0x79 > [39993.560318] [] xfs_inactive+0x14d/0x411 > [39993.560322] [] xfs_fs_clear_inode+0x83/0x85 > [39993.560325] [] clear_inode+0x8a/0xe3 > [39993.560329] [] generic_delete_inode+0xf8/0x173 > [39993.560332] [] generic_drop_inode+0x17/0x1d5 > [39993.560335] [] iput+0x61/0x65 > [39993.560338] [] dentry_iput+0xa9/0xc1 > [39993.560341] [] d_kill+0x40/0x60 > [39993.560344] [] dput+0x141/0x14e > [39993.560347] [] __fput+0x173/0x198 > [39993.560350] [] fput+0x18/0x1a > [39993.560353] [] remove_vma+0x3e/0x63 > [39993.560355] [] do_munmap+0x2e9/0x30b > [39993.560358] [] sys_munmap+0x40/0x59 > [39993.560361] [] system_call_fastpath+0x16/0x1b > [39993.560363] [] 0xffffffffffffffff This is a VM problem where it calls fput() with the mmap_sem() held in remove_vma(). It makes the incorrect assumption that filesystems will never use the same lock in the IO path and the inode release path. This can deadlock if you are really unlucky. Cheers, Dave. -- Dave Chinner david@fromorbit.com