Andrew and Li reviewed and found that we need to check for val being 1 or 0
for the root container as well. use_hierarchy's type is changed to bool.
We still continue to use the ease of write_X64 for writing to it and then
check if the values are sane.
Signed-off-by: Balbir Singh <balbir@linux.vnet.ibm.com>
---
mm/memcontrol.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff -puN mm/memcontrol.c~memcg-fix-add-hierarchy-selector mm/memcontrol.c
--- linux-2.6.28-rc4/mm/memcontrol.c~memcg-fix-add-hierarchy-selector 2008-11-19 11:07:21.000000000 +0530
+++ linux-2.6.28-rc4-balbir/mm/memcontrol.c 2008-11-19 14:11:56.000000000 +0530
@@ -151,7 +151,7 @@ struct mem_cgroup {
/*
* Should the accounting and control be hierarchical, per subtree?
*/
- unsigned long use_hierarchy;
+ bool use_hierarchy;
int obsolete;
atomic_t refcnt;
@@ -1556,8 +1556,8 @@ static int mem_cgroup_hierarchy_write(st
* For the root cgroup, parent_mem is NULL, we allow value to be
* set if there are no children.
*/
- if (!parent_mem || (!parent_mem->use_hierarchy &&
- (val == 1 || val == 0))) {
+ if ((!parent_mem || !parent_mem->use_hierarchy) &&
+ (val == 1 || val == 0)) {
if (list_empty(&cont->children))
mem->use_hierarchy = val;
else
_
--
Balbir
--
