On Fri, 2011-07-01 at 05:43 -0400, Christoph Hellwig wrote:
> The following script from Wu Fengguang shows very bad behaviour in XFS
> when aggressively dirtying data during a sync on XFS, with sync times
> up to almost 10 times as long as ext4.
(Note that I skipped over patch 7 for the time being,
trying to skip ahead to simpler changes to review.)
I think the change looks fine but the description doesn't
completely match it (unless I'm missing something).
> A large part of the issue is that XFS writes data out itself two times
> in the ->sync_fs method, overriding the lifelock protection in the core
> writeback code, and another issue is the lock-less xfs_ioend_wait call,
> which doesn't prevent new ioend from beeing queue up while waiting for
> the count to reach zero.
The change affects only the first thing you mention here, not
the second.
Also, if you plan to update the description--some typo's:
- "in the face of" in the subject
- "livelock protection" above
- "beeing" -> "being"
> This patch removes the XFS-internal sync calls and relies on the VFS
> to do it's work just like all other filesystems do. Note that the
> i_iocount wait which is rather suboptimal is simply removed here.
> We already do it in ->write_inode, which keeps the current supoptimal
> behaviour. We'll eventually need to remove that as well, but that's
> material for a separate commit.
The i_iocount wait is not affected by your patch.
> ------------------------------ snip ------------------------------
> #!/bin/sh
>
> umount /dev/sda7
> mkfs.xfs -f /dev/sda7
> # mkfs.ext4 /dev/sda7
> # mkfs.btrfs /dev/sda7
> mount /dev/sda7 /fs
>
> echo $((50<<20)) > /proc/sys/vm/dirty_bytes
>
> pid=
> for i in `seq 10`
> do
> dd if=/dev/zero of=/fs/zero-$i bs=1M count=1000 &
> pid="$pid $!"
> done
>
> sleep 1
>
> tic=$(date +'%s')
> sync
> tac=$(date +'%s')
>
> echo
> echo sync time: $((tac-tic))
> egrep '(Dirty|Writeback|NFS_Unstable)' /proc/meminfo
>
> pidof dd > /dev/null && { kill -9 $pid; echo sync NOT livelocked; }
> ------------------------------ snip ------------------------------
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> Reported-by: Wu Fengguang <fengguang.wu@xxxxxxxxx>
> Reviewed-by: Dave Chinner <dchinner@xxxxxxxxxx>
I'm OK with the change, but really prefer to have
the description not include stuff that just isn't
there. If you want me to commit this as-is, just
say so and I will. Otherwise, post an update and
I'll use that. In any case, you can consider this
reviewed by me.
Reviewed-by: Alex Elder <aelder@xxxxxxx>
> Index: xfs/fs/xfs/linux-2.6/xfs_sync.c
> ===================================================================
> --- xfs.orig/fs/xfs/linux-2.6/xfs_sync.c 2011-06-29 11:26:14.109219361
> +0200
> +++ xfs/fs/xfs/linux-2.6/xfs_sync.c 2011-06-29 11:37:20.642275110 +0200
> @@ -359,14 +359,12 @@ xfs_quiesce_data(
> {
> int error, error2 = 0;
>
> - /* push non-blocking */
> - xfs_sync_data(mp, 0);
> xfs_qm_sync(mp, SYNC_TRYLOCK);
> -
> - /* push and block till complete */
> - xfs_sync_data(mp, SYNC_WAIT);
> xfs_qm_sync(mp, SYNC_WAIT);
>
> + /* force out the newly dirtied log buffers */
> + xfs_log_force(mp, XFS_LOG_SYNC);
> +
> /* write superblock and hoover up shutdown errors */
> error = xfs_sync_fsdata(mp);
>
>
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs
|