CGroups: CGroups: Hierarchy locking/refcount changes

These patches present an alternative to some of the other cgroups
locking/refcount patches that have been proposed on LKML recently.

Several of these patches have been to address the race opened by
moving the calls to pre_destroy() callbacks outside of cgroup_mutex;
rather than continuing to patch up the holes caused by that change,
these patches introduce new locking/refcount rules to ultimately allow
the previous atomicity of cgroup_rmdir() to be restored.

These three patches give:

1/3 - introduce a per-subsystem hierarchy_mutex which a subsystem can
      use to prevent changes to its own cgroup tree

2/3 - use hierarchy_mutex in place of calling cgroup_lock() in the
      memory controller

3/3 - introduce a css_tryget() function similar to the one proposed by
      Kamezawa, but avoiding spurious refcount failures in the event
      of a race between a css_tryget() and an unsuccessful cgroup_rmdir()

Future patches will likely involve:

- using hierarchy mutex in place of cgroup_lock() in more subsystems
  where appropriate

- restoring the atomicity of cgroup_rmdir() with respect to cgroup_create()

Signed-off-by: Paul Menage <menage@google.com>