Kirill A. Shutemov 8c0a853770 fs: push rcu_barrier() from deactivate_locked_super() to filesystems
There's no reason to call rcu_barrier() on every
deactivate_locked_super().  We only need to make sure that all delayed rcu
free inodes are flushed before we destroy related cache.

Removing rcu_barrier() from deactivate_locked_super() affects some fast
paths.  E.g.  on my machine exit_group() of a last process in IPC
namespace takes 0.07538s.  rcu_barrier() takes 0.05188s of that time.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
2012-10-02 21:35:55 -04:00
..
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2011-07-25 15:03:13 -05:00
2011-07-13 13:43:48 +02:00
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2011-03-07 10:09:35 +11:00
2012-05-29 23:28:33 -04:00
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2012-07-29 16:23:51 -05:00
2012-07-29 16:23:51 -05:00
2012-05-14 16:20:54 -05:00
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2010-12-16 16:05:51 -06:00
2012-07-30 13:37:53 -07:00
2011-08-12 16:21:35 -05:00
2012-06-21 14:22:27 -05:00
2012-06-21 14:22:27 -05:00
2012-05-14 16:21:00 -05:00
2012-03-13 17:01:15 -05:00
2011-08-12 16:21:35 -05:00
2011-08-12 16:21:35 -05:00
2012-05-14 16:21:00 -05:00
2012-07-22 10:38:32 -05:00
2012-07-31 09:45:48 +04:00
2012-01-03 22:55:00 -05:00
2011-08-12 16:21:35 -05:00