xfs: don't defer metadata allocation to the workqueue
authorDave Chinner <dchinner@redhat.com>
Wed, 11 Jul 2012 21:40:43 +0000 (07:40 +1000)
committerBen Myers <bpm@sgi.com>
Fri, 13 Jul 2012 18:09:27 +0000 (13:09 -0500)
commitaa292847b9fc6e187547110de833a7d3131bbddf
treedf2976e85569c2f04531733508606ffa969661da
parente3a746f5aab71f2dd0a83116772922fb37ae29d6
xfs: don't defer metadata allocation to the workqueue

Almost all metadata allocations come from shallow stack usage
situations. Avoid the overhead of switching the allocation to a
workqueue as we are not in danger of running out of stack when
making these allocations. Metadata allocations are already marked
through the args that are passed down, so this is trivial to do.

Signed-off-by: Dave Chinner <dchinner@redhat.com>
Reported-by: Mel Gorman <mgorman@suse.de>
Tested-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Ben Myers <bpm@sgi.com>
fs/xfs/xfs_alloc.c