From ed45b3440ffe515828a38fce2e397e85514eb93b Mon Sep 17 00:00:00 2001
From: Xiu Jianfeng <xiujianfeng@huawei.com>
Date: Sat, 12 Oct 2024 07:22:46 +0000
Subject: [PATCH] cgroup: Fix potential overflow issue when checking max_depth
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[ Upstream commit 3cc4e13bb1617f6a13e5e6882465984148743cf4 ]

cgroup.max.depth is the maximum allowed descent depth below the current
cgroup. If the actual descent depth is equal or larger, an attempt to
create a new child cgroup will fail. However due to the cgroup->max_depth
is of int type and having the default value INT_MAX, the condition
'level > cgroup->max_depth' will never be satisfied, and it will cause
an overflow of the level after it reaches to INT_MAX.

Fix it by starting the level from 0 and using '>=' instead.

It's worth mentioning that this issue is unlikely to occur in reality,
as it's impossible to have a depth of INT_MAX hierarchy, but should be
be avoided logically.

Fixes: 1a926e0bbab8 ("cgroup: implement hierarchy limits")
Signed-off-by: Xiu Jianfeng <xiujianfeng@huawei.com>
Reviewed-by: Michal Koutný <mkoutny@suse.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 kernel/cgroup/cgroup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 8d93b8ec1..3f96d0f9e 100755
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -5497,7 +5497,7 @@ static bool cgroup_check_hierarchy_limits(struct cgroup *parent)
 {
 	struct cgroup *cgroup;
 	int ret = false;
-	int level = 1;
+	int level = 0;
 
 	lockdep_assert_held(&cgroup_mutex);
 
@@ -5505,7 +5505,7 @@ static bool cgroup_check_hierarchy_limits(struct cgroup *parent)
 		if (cgroup->nr_descendants >= cgroup->max_descendants)
 			goto fail;
 
-		if (level > cgroup->max_depth)
+		if (level >= cgroup->max_depth)
 			goto fail;
 
 		level++;