From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755705Ab1EDVKx (ORCPT ); Wed, 4 May 2011 17:10:53 -0400 Received: from cantor2.suse.de ([195.135.220.15]:36064 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753217Ab1EDVKw (ORCPT ); Wed, 4 May 2011 17:10:52 -0400 Date: Wed, 4 May 2011 23:10:50 +0200 From: Jan Kara To: Wu Fengguang Cc: Andrew Morton , Jan Kara , Dave Chinner , Christoph Hellwig , LKML , linux-fsdevel@vger.kernel.org Subject: Re: [PATCH 2/3] writeback: update dirtied_when for synced inode to prevent livelock Message-ID: <20110504211050.GH6968@quack.suse.cz> References: <20110502031750.135798606@intel.com> <20110502033035.664444896@intel.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20110502033035.664444896@intel.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon 02-05-11 11:17:52, Wu Fengguang wrote: > With the more aggressive "keep writeback as long as we wrote something" > logic in wb_writeback(), the "use LONG_MAX .nr_to_write" trick in commit > b9543dac5bbc ("writeback: avoid livelocking WB_SYNC_ALL writeback") is > no longer enough to stop sync livelock. > > The fix is to explicitly update .dirtied_when on synced inodes, so that > they are no longer considered for writeback in the next round. Looks good. But shouldn't you also update i_dirtied_when in the case where !mapping_tagged(mapping, PAGECACHE_TAG_DIRTY) but inode->i_state & I_DIRTY? Mostly for consistency since the chances of livelock because of this are really close to zero. Honza > > CC: Jan Kara > Signed-off-by: Wu Fengguang > --- > fs/fs-writeback.c | 9 +++++++++ > 1 file changed, 9 insertions(+) > > ext3/ext4 are working fine now, however tests show that XFS may still > livelock inside the XFS routines: > > [ 3581.181253] sync D ffff8800b6ca15d8 4560 4403 4392 0x00000000 > [ 3581.181734] ffff88006f775bc8 0000000000000046 ffff8800b6ca12b8 00000001b6ca1938 > [ 3581.182411] ffff88006f774000 00000000001d2e40 00000000001d2e40 ffff8800b6ca1280 > [ 3581.183088] 00000000001d2e40 ffff88006f775fd8 00000340af111ef2 00000000001d2e40 > [ 3581.183765] Call Trace: > [ 3581.184008] [] ? lock_release_holdtime+0xa3/0xab > [ 3581.184392] [] ? prepare_to_wait+0x6c/0x79 > [ 3581.184756] [] ? prepare_to_wait+0x6c/0x79 > [ 3581.185120] [] xfs_ioend_wait+0x87/0x9f > [ 3581.185474] [] ? wake_up_bit+0x2a/0x2a > [ 3581.185827] [] xfs_sync_inode_data+0x92/0x9d > [ 3581.186198] [] xfs_inode_ag_walk+0x1a5/0x287 > [ 3581.186569] [] ? xfs_inode_ag_walk+0x25e/0x287 > [ 3581.186946] [] ? xfs_sync_worker+0x69/0x69 > [ 3581.187311] [] ? xfs_perag_get+0x68/0xd0 > [ 3581.187669] [] ? local_clock+0x41/0x5a > [ 3581.188020] [] ? lock_release_holdtime+0xa3/0xab > [ 3581.188403] [] ? xfs_check_sizes+0x160/0x160 > [ 3581.188773] [] ? xfs_perag_get+0x68/0xd0 > [ 3581.189130] [] ? xfs_perag_get+0x80/0xd0 > [ 3581.189488] [] ? xfs_check_sizes+0x160/0x160 > [ 3581.189858] [] ? xfs_inode_ag_iterator+0x6d/0x8f > [ 3581.190241] [] ? xfs_sync_worker+0x69/0x69 > [ 3581.190606] [] xfs_inode_ag_iterator+0x47/0x8f > [ 3581.190982] [] ? __sync_filesystem+0x7a/0x7a > [ 3581.191352] [] xfs_sync_data+0x24/0x43 > [ 3581.191703] [] xfs_quiesce_data+0x2c/0x88 > [ 3581.192065] [] xfs_fs_sync_fs+0x21/0x48 > [ 3581.192419] [] __sync_filesystem+0x66/0x7a > [ 3581.192783] [] sync_one_sb+0x16/0x18 > [ 3581.193128] [] iterate_supers+0x72/0xce > [ 3581.193482] [] sync_filesystems+0x20/0x22 > [ 3581.193842] [] sys_sync+0x21/0x33 > [ 3581.194177] [] system_call_fastpath+0x16/0x1b > > --- linux-next.orig/fs/fs-writeback.c 2011-05-02 11:16:50.000000000 +0800 > +++ linux-next/fs/fs-writeback.c 2011-05-02 11:17:24.000000000 +0800 > @@ -431,6 +431,15 @@ writeback_single_inode(struct inode *ino > requeue_io(inode, wb); > } else { > /* > + * sync livelock prevention: each inode is > + * tagged and synced in one shot. If still > + * dirty, move it back to s_dirty with updated > + * dirty time to prevent syncing it again. > + */ > + if (wbc->sync_mode == WB_SYNC_ALL || > + wbc->tagged_sync) > + inode->dirtied_when = jiffies; > + /* > * Writeback blocked by something other than > * congestion. Delay the inode for some time to > * avoid spinning on the CPU (100% iowait) > > -- Jan Kara SUSE Labs, CR