xfs
[Top] [All Lists]

Re: do_sync() and XFSQA test 182 failures....

To: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Subject: Re: do_sync() and XFSQA test 182 failures....
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Sat, 1 Nov 2008 09:03:49 +1100
Cc: xfs@xxxxxxxxxxx
In-reply-to: <20081031203716.GB11514@xxxxxxxxxxxxx>
Mail-followup-to: Christoph Hellwig <hch@xxxxxxxxxxxxx>, xfs@xxxxxxxxxxx
References: <20081030085020.GP17077@disturbed> <20081030224625.GA18690@xxxxxxxxxxxxx> <20081031001249.GM4985@disturbed> <20081031004814.GN4985@disturbed> <20081031203716.GB11514@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.18 (2008-05-17)
On Fri, Oct 31, 2008 at 04:37:16PM -0400, Christoph Hellwig wrote:
> On Fri, Oct 31, 2008 at 11:48:14AM +1100, Dave Chinner wrote:
> > @@ -1118,8 +1118,7 @@ xfs_fs_write_super(
> >     struct super_block      *sb)
> >  {
> >     if (!(sb->s_flags & MS_RDONLY))
> > +           xfs_sync_fsdata(XFS_M(sb), SYNC_WAIT);
> 
> Not commenting on the rest yet, but I'm sure you noticed that before
> your patch we ever called xfs_sync_fsdata with SYNC_WAIT at all.  I
> think with the current codebase there is nothing to guarantee we ever
> write the superblock out to disk as xfs_wait_buftarg specificly
> excludes the superblock (XBF_FS_MANAGED)

Almost. unmount calls xfs_unmountfs_writesb() which does a
synchronous write of the superblock. But in the current code sync(1)
certainly never, ever triggers a superblock write we wait for...

Cheers,

Dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

<Prev in Thread] Current Thread [Next in Thread>