gfs2: use workqueue instead of slow-work
authorTejun Heo <tj@kernel.org>
Tue, 20 Jul 2010 20:09:02 +0000 (22:09 +0200)
committerTejun Heo <tj@kernel.org>
Fri, 23 Jul 2010 11:14:25 +0000 (13:14 +0200)
Workqueue can now handle high concurrency.  Convert gfs to use
workqueue instead of slow-work.

* Steven pointed out that recovery path might be run from allocation
  path and thus requires forward progress guarantee without memory
  allocation.  Create and use gfs_recovery_wq with rescuer.  Please
  note that forward progress wasn't guaranteed with slow-work.

* Updated to use non-reentrant workqueue.

Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Steven Whitehouse <swhiteho@redhat.com>

No differences found