From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay1.corp.sgi.com [137.38.102.111]) by oss.sgi.com (Postfix) with ESMTP id 82D367F4E for ; Thu, 12 Sep 2013 01:27:32 -0500 (CDT) Received: from cuda.sgi.com (cuda1.sgi.com [192.48.157.11]) by relay1.corp.sgi.com (Postfix) with ESMTP id 55C3A8F8039 for ; Wed, 11 Sep 2013 23:27:29 -0700 (PDT) Received: from userp1040.oracle.com (userp1040.oracle.com [156.151.31.81]) by cuda.sgi.com with ESMTP id k7VULt6MtBs0OPDo (version=TLSv1 cipher=AES256-SHA bits=256 verify=NO) for ; Wed, 11 Sep 2013 23:27:28 -0700 (PDT) Message-ID: <52315EF0.1070804@oracle.com> Date: Thu, 12 Sep 2013 14:28:00 +0800 From: Jeff Liu MIME-Version: 1.0 Subject: Re: [deadlock] AGI vs AGF ordering deadlocks References: <20130910073629.GA19103@dastard> <522ED124.4080502@oracle.com> In-Reply-To: <522ED124.4080502@oracle.com> 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 Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: Dave Chinner Cc: xfs@oss.sgi.com On 09/10/2013 03:58 PM, Jeff Liu wrote: > On 09/10/2013 03:36 PM, Dave Chinner wrote: > >> FOlks, >> >> I just got confirmation of a deadlock I suspected has existed for >> some time. A concurrent 16-way create and 16-way unlink just locked >> up with two threads looking like this: >> >> fs_mark D ffff88021bd931c0 3656 7204 7117 0x00000000 >> ffff8801e75293a8 0000000000000086 ffff88012c6d0000 ffff8801e7529fd8 >> ffff8801e7529fd8 ffff8801e7529fd8 ffff8802d32aae40 ffff88012c6d0000 >> ffff8801a2f79d40 7fffffffffffffff ffff8801ee733bb0 0000000000000002 >> Call Trace: >> [] schedule+0x29/0x70 >> [] schedule_timeout+0x149/0x1f0 >> [] __down_common+0x91/0xe8 >> [] __down+0x1d/0x1f >> [] down+0x41/0x50 >> [] xfs_buf_lock+0x40/0xf0 >> [] _xfs_buf_find+0x1d1/0x4d0 >> [] xfs_buf_get_map+0x35/0x180 >> [] xfs_buf_read_map+0x37/0x110 >> [] xfs_trans_read_buf_map+0x379/0x600 >> [] xfs_read_agf+0xa8/0x100 >> [] xfs_alloc_read_agf+0x6a/0x250 >> [] xfs_alloc_fix_freelist+0x4f0/0x5a0 >> [] xfs_alloc_vextent+0x440/0x840 >> [] xfs_ialloc_ag_alloc+0x13f/0x520 >> [] xfs_dialloc+0x121/0x2d0 >> [] xfs_ialloc+0x5b/0x7c0 >> [] xfs_dir_ialloc+0x9a/0x2f0 >> [] xfs_create+0x47d/0x6a0 >> [] xfs_vn_mknod+0xba/0x1c0 >> [] xfs_vn_create+0x13/0x20 >> [] vfs_create+0xb5/0xf0 >> [] do_last.isra.56+0x760/0xd10 >> [] path_openat+0xbe/0x620 >> [] do_filp_open+0x43/0xa0 >> [] do_sys_open+0x13c/0x230 >> [] SyS_open+0x22/0x30 >> [] system_call_fastpath+0x16/0x1b >> >> That a thread holding an AGI and blocking trying to get the AGF to >> do an inode chunk allocation. >> >> rm D ffff88021bd931c0 3048 7073 7063 0x00000000 >> ffff8802bc66d998 0000000000000086 ffff8802d32aae40 ffff8802bc66dfd8 >> ffff8802bc66dfd8 ffff8802bc66dfd8 ffff88012c6d5c80 ffff8802d32aae40 >> ffff8804091b2b00 7fffffffffffffff ffff8801b943c570 0000000000000002 >> Call Trace: >> [] schedule+0x29/0x70 >> [] schedule_timeout+0x149/0x1f0 >> [] __down_common+0x91/0xe8 >> [] __down+0x1d/0x1f >> [] down+0x41/0x50 >> [] xfs_buf_lock+0x40/0xf0 >> [] _xfs_buf_find+0x1d1/0x4d0 >> [] xfs_buf_get_map+0x35/0x180 >> [] xfs_buf_read_map+0x37/0x110 >> [] xfs_trans_read_buf_map+0x379/0x600 >> [] xfs_read_agi+0xaa/0x100 >> [] xfs_iunlink+0x8e/0x260 >> [] xfs_droplink+0x78/0x80 >> [] xfs_remove+0x331/0x420 >> [] xfs_vn_unlink+0x52/0xa0 >> [] vfs_unlink+0x9e/0x110 >> [] do_unlinkat+0x1a1/0x230 >> [] SyS_unlinkat+0x1b/0x40 >> >> And that's a thread that has just freed a directory block and so >> holds an AGF lock, and is trying to take the AGI lock to add the >> inode to the unlinked list. Everything else is now stuck waiting >> for log space because one of the two buffers we've deadlocked on >> here pins the tail of the log. >> >> The solution is to place the inode on the unlinked list before we >> remove the directory entry so that we keep the same locking order as >> inode allocation. >> >> I don't have time to look at this for at least a week, so if someone >> could work up solution that'd be wonderful... > > Although I can reproduce it for now, but it looks interesting to me. Sorry, s/can/can not/. > I'll take care of this problem. Still no luck to reproduce it on my poor laptop, so I have to release this for someone who can reproduce it and be interesting enough in fix it. :) Thanks, -Jeff _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs