From: Wu Fengguang <fengguang.wu@intel.com>
To: Dave Chinner <david@fromorbit.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
Jan Kara <jack@suse.cz>, Christoph Hellwig <hch@infradead.org>,
"linux-fsdevel@vger.kernel.org" <linux-fsdevel@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH 02/17] writeback: update dirtied_when for synced inode to prevent livelock
Date: Fri, 13 May 2011 11:08:10 +0800 [thread overview]
Message-ID: <20110513030810.GB8016@localhost> (raw)
In-Reply-To: <20110512224211.GI19446@dastard>
On Fri, May 13, 2011 at 06:42:11AM +0800, Dave Chinner wrote:
> On Thu, May 12, 2011 at 09:57:08PM +0800, Wu Fengguang wrote:
> > Explicitly update .dirtied_when on synced inodes, so that they are no
> > longer considered for writeback in the next round.
> >
> > We'll do 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") will
> > no longer be enough to stop sync livelock.
> >
> > Reviewed-by: Jan Kara <jack@suse.cz>
> > Signed-off-by: Wu Fengguang <fengguang.wu@intel.com>
> > ---
> > 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] [<ffffffff8109be73>] ? lock_release_holdtime+0xa3/0xab
> > [ 3581.184392] [<ffffffff8108cc0d>] ? prepare_to_wait+0x6c/0x79
> > [ 3581.184756] [<ffffffff8108cc0d>] ? prepare_to_wait+0x6c/0x79
> > [ 3581.185120] [<ffffffff812ed520>] xfs_ioend_wait+0x87/0x9f
> > [ 3581.185474] [<ffffffff8108c97a>] ? wake_up_bit+0x2a/0x2a
> > [ 3581.185827] [<ffffffff812f742a>] xfs_sync_inode_data+0x92/0x9d
> > [ 3581.186198] [<ffffffff812f76e2>] xfs_inode_ag_walk+0x1a5/0x287
> > [ 3581.186569] [<ffffffff812f779b>] ? xfs_inode_ag_walk+0x25e/0x287
> > [ 3581.186946] [<ffffffff812f7398>] ? xfs_sync_worker+0x69/0x69
> > [ 3581.187311] [<ffffffff812e2354>] ? xfs_perag_get+0x68/0xd0
> > [ 3581.187669] [<ffffffff81092175>] ? local_clock+0x41/0x5a
> > [ 3581.188020] [<ffffffff8109be73>] ? lock_release_holdtime+0xa3/0xab
> > [ 3581.188403] [<ffffffff812e22ec>] ? xfs_check_sizes+0x160/0x160
> > [ 3581.188773] [<ffffffff812e2354>] ? xfs_perag_get+0x68/0xd0
> > [ 3581.189130] [<ffffffff812e236c>] ? xfs_perag_get+0x80/0xd0
> > [ 3581.189488] [<ffffffff812e22ec>] ? xfs_check_sizes+0x160/0x160
> > [ 3581.189858] [<ffffffff812f7831>] ? xfs_inode_ag_iterator+0x6d/0x8f
> > [ 3581.190241] [<ffffffff812f7398>] ? xfs_sync_worker+0x69/0x69
> > [ 3581.190606] [<ffffffff812f780b>] xfs_inode_ag_iterator+0x47/0x8f
> > [ 3581.190982] [<ffffffff811611f5>] ? __sync_filesystem+0x7a/0x7a
> > [ 3581.191352] [<ffffffff812f7877>] xfs_sync_data+0x24/0x43
> > [ 3581.191703] [<ffffffff812f7911>] xfs_quiesce_data+0x2c/0x88
> > [ 3581.192065] [<ffffffff812f5556>] xfs_fs_sync_fs+0x21/0x48
> > [ 3581.192419] [<ffffffff811611e1>] __sync_filesystem+0x66/0x7a
> > [ 3581.192783] [<ffffffff8116120b>] sync_one_sb+0x16/0x18
> > [ 3581.193128] [<ffffffff8113e3e3>] iterate_supers+0x72/0xce
> > [ 3581.193482] [<ffffffff81161140>] sync_filesystems+0x20/0x22
> > [ 3581.193842] [<ffffffff8116127e>] sys_sync+0x21/0x33
> > [ 3581.194177] [<ffffffff819016c2>] system_call_fastpath+0x16/0x1b
> >
> > --- linux-next.orig/fs/fs-writeback.c 2011-05-10 09:50:07.000000000 +0800
> > +++ linux-next/fs/fs-writeback.c 2011-05-10 10:03:00.000000000 +0800
> > @@ -419,6 +419,15 @@ writeback_single_inode(struct inode *ino
> > spin_lock(&inode->i_lock);
> > inode->i_state &= ~I_SYNC;
> > if (!(inode->i_state & I_FREEING)) {
> > + /*
> > + * Sync livelock prevention. Each inode is tagged and synced in
> > + * one shot, so we can unconditionally update its dirty time to
> > + * prevent syncing it again. Note that time ordering of b_dirty
> > + * list will be kept because the following code either removes
> > + * the inode from b_dirty or calls redirty_tail().
> > + */
> > + if (wbc->sync_mode == WB_SYNC_ALL || wbc->tagged_sync)
> > + inode->dirtied_when = jiffies;
>
> Shouldn't this update only ocur if the inode is still dirty?
Yeah, that would be better even though the current form won't lead to
errors.
Let's add one more test (inode->i_state & I_DIRTY)?
(I was actually aware of the trade offs and didn't bother to add it..)
Thanks,
Fengguang
next prev parent reply other threads:[~2011-05-13 3:08 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2011-05-12 13:57 [PATCH 00/17] writeback fixes and cleanups for 2.6.40 (v2) Wu Fengguang
2011-05-12 13:57 ` [PATCH 01/17] writeback: introduce .tagged_sync for the WB_SYNC_NONE sync stage Wu Fengguang
2011-05-12 22:40 ` Dave Chinner
2011-05-13 2:56 ` Wu Fengguang
2011-05-13 10:17 ` Christoph Hellwig
2011-05-15 23:43 ` Dave Chinner
2011-05-16 5:39 ` Wu Fengguang
2011-05-19 21:17 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 02/17] writeback: update dirtied_when for synced inode to prevent livelock Wu Fengguang
2011-05-12 22:42 ` Dave Chinner
2011-05-13 3:08 ` Wu Fengguang [this message]
2011-05-19 21:31 ` Wu Fengguang
2011-05-23 13:14 ` Jan Kara
2011-05-24 3:03 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 03/17] writeback: introduce writeback_control.inodes_cleaned Wu Fengguang
2011-05-12 22:44 ` Dave Chinner
2011-05-13 3:36 ` Wu Fengguang
2011-05-15 23:50 ` Dave Chinner
2011-05-16 10:40 ` Christoph Hellwig
2011-05-16 11:14 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 04/17] writeback: try more writeback as long as something was written Wu Fengguang
2011-05-12 13:57 ` [PATCH 05/17] writeback: the kupdate expire timestamp should be a moving target Wu Fengguang
2011-05-12 13:57 ` [PATCH 06/17] writeback: sync expired inodes first in background writeback Wu Fengguang
2011-05-12 22:55 ` Dave Chinner
2011-05-16 13:00 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 07/17] writeback: refill b_io iff empty Wu Fengguang
2011-05-12 13:57 ` [PATCH 08/17] writeback: split inode_wb_list_lock into bdi_writeback.list_lock Wu Fengguang
2011-05-12 13:57 ` [PATCH 09/17] writeback: elevate queue_io() into wb_writeback() Wu Fengguang
2011-05-12 13:57 ` [PATCH 10/17] writeback: avoid extra sync work at enqueue time Wu Fengguang
2011-05-12 13:57 ` [PATCH 11/17] writeback: add bdi_dirty_limit() kernel-doc Wu Fengguang
2011-05-12 13:57 ` [PATCH 12/17] writeback: skip balance_dirty_pages() for in-memory fs Wu Fengguang
2011-05-16 10:43 ` Christoph Hellwig
2011-05-16 10:49 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 13/17] writeback: remove writeback_control.more_io Wu Fengguang
2011-05-12 14:25 ` Minchan Kim
2011-05-12 23:04 ` Dave Chinner
2011-05-13 5:03 ` Wu Fengguang
2011-05-15 23:54 ` Dave Chinner
2011-05-12 13:57 ` [PATCH 14/17] writeback: make writeback_control.nr_to_write straight Wu Fengguang
2011-05-12 14:56 ` Jan Kara
2011-05-12 23:18 ` Dave Chinner
2011-05-13 5:28 ` Wu Fengguang
2011-05-16 0:12 ` Dave Chinner
2011-05-16 12:05 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 15/17] writeback: remove .nonblocking and .encountered_congestion Wu Fengguang
2011-05-12 13:57 ` [PATCH 16/17] writeback: trace event writeback_single_inode Wu Fengguang
2011-05-12 23:20 ` Dave Chinner
2011-05-13 5:37 ` Wu Fengguang
2011-05-16 0:14 ` Dave Chinner
2011-05-16 12:21 ` Wu Fengguang
2011-05-12 13:57 ` [PATCH 17/17] writeback: trace event writeback_queue_io Wu Fengguang
-- strict thread matches above, loose matches on Subject: below --
2011-05-06 3:08 [PATCH 00/17] writeback fixes and cleanups for 2.6.40 Wu Fengguang
2011-05-06 3:08 ` [PATCH 02/17] writeback: update dirtied_when for synced inode to prevent livelock Wu Fengguang
2011-05-06 16:33 ` Jan Kara
2011-05-10 2:14 ` Wu Fengguang
2011-05-10 12:05 ` Jan Kara
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20110513030810.GB8016@localhost \
--to=fengguang.wu@intel.com \
--cc=akpm@linux-foundation.org \
--cc=david@fromorbit.com \
--cc=hch@infradead.org \
--cc=jack@suse.cz \
--cc=linux-fsdevel@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).