On Tue, May 08, 2012 at 11:29:42AM -0500, Ben Myers wrote:
> On Fri, Apr 27, 2012 at 07:45:22PM +1000, Dave Chinner wrote:
> > From: Dave Chinner <dchinner@xxxxxxxxxx>
> >
> > Often mounting small filesystem with small logs will emit a warning
> > such as:
> >
> > XFS (vdb): Invalid block length (0x2000) for buffer
> >
> > during log recovery. This causes tests to randomly fail because this
> > output causes the clean filesystem checks on test completion to
> > think the filesystem is inconsistent.
> >
> > The cause of the error is simply that log recovery is asking for a
> > buffer size that is larger than the log when zeroing the tail. This
> > is because the buffer size is rounded up, and if the right head and
> > tail conditions exist then the buffer size can be larger than the log.
> > Limit the variable size xlog_get_bp() callers to requesting buffers
> > smaller than the log.
> >
> > Signed-off-by: Dave Chinner <dchinner@xxxxxxxxxx>
> > ---
> > fs/xfs/xfs_log_recover.c | 4 ++++
> > 1 file changed, 4 insertions(+)
> >
> > diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
> > index d7abe5f..ca38690 100644
> > --- a/fs/xfs/xfs_log_recover.c
> > +++ b/fs/xfs/xfs_log_recover.c
> > @@ -441,6 +441,8 @@ xlog_find_verify_cycle(
> > * a log sector, or we're out of luck.
> > */
> > bufblks = 1 << ffs(nbblks);
> > + while (bufblks > log->l_logBBsize)
> > + bufblks >>= 1;
>
> AFAICS you don't need a loop here. The following would be sufficient to make
> xlog_buf_bbcount_valid return 0.
>
> if (bufblks > log->l_logBBsize)
> bufblks = log->l_logBBsize;
Yes, I could do that, but then there is a different set of boundary
conditions to test. I know that the >>=1 logic works, but I have no
idea what new corner cases occur when bufblks == log->l_logBBsize.
> It is a bit more obviously correct.
It may be to read, but it's certainly more different from a
verification point of view. Given how long and arduous the process
was to find the source of the problem, I am very wary of changing
logic to run in ways that are different and very difficult to
actually test....
Cheers,
Dave.
--
Dave Chinner
david@xxxxxxxxxxxxx
|