xfs: fix ->write_inode return values
authorChristoph Hellwig <hch@infradead.org>
Sat, 27 Aug 2011 05:57:55 +0000 (05:57 +0000)
committerAlex Elder <aelder@sgi.com>
Thu, 1 Sep 2011 14:46:11 +0000 (09:46 -0500)
commit58d84c4ee0389ddeb86238d5d8359a982c9f7a5b
treec43d68e71407f45b4ecde6c2602d979ca1d03c34
parent866e4ed77448a0c311e1b055eb72ea05423fd799
xfs: fix ->write_inode return values

Currently we always redirty an inode that was attempted to be written out
synchronously but has been cleaned by an AIL pushed internall, which is
rather bogus.  Fix that by doing the i_update_core check early on and
return 0 for it.  Also include async calls for it, as doing any work for
those is just as pointless.  While we're at it also fix the sign for the
EIO return in case of a filesystem shutdown, and fix the completely
non-sensical locking around xfs_log_inode.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Signed-off-by: Alex Elder <aelder@sgi.com>
(cherry picked from commit 297db93bb74cf687510313eb235a7aec14d67e97)

Signed-off-by: Alex Elder <aelder@sgi.com>
fs/xfs/xfs_super.c