xfs: don't flush inodes from background inode reclaim
authorChristoph Hellwig <hch@infradead.org>
Mon, 23 Apr 2012 05:58:35 +0000 (15:58 +1000)
committerBen Myers <bpm@sgi.com>
Mon, 14 May 2012 21:20:28 +0000 (16:20 -0500)
commit8a48088f6439249019b5e17f6391e710656879d9
treeb898076f1f2245dc435fa8e64b1ddac2b8424c7e
parent211e4d434bd737be38aabad0247ce3da9964370e
xfs: don't flush inodes from background inode reclaim

We already flush dirty inodes throug the AIL regularly, there is no reason
to have second thread compete with it and disturb the I/O pattern.  We still
do write inodes when doing a synchronous reclaim from the shrinker or during
unmount for now.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Dave Chinner <dchinner@redhat.com>
Reviewed-by: Mark Tinguely <tinguely@sgi.com>
Signed-off-by: Ben Myers <bpm@sgi.com>
fs/xfs/xfs_sync.c