[PATCH] device_cgroup: fix RCU imbalance in error case

[Jann Horn]

When dev_exception_add() returns an error (due to a failed memory
allocation), make sure that we move the RCU preemption count back to where
it was before we were called. We dropped the RCU read lock inside the loop
body, so we can't just "break".

sparse complains about this, too:

$ make -s C=2 security/device_cgroup.o
./include/linux/rcupdate.h:647:9: warning: context imbalance in
'propagate_exception' - unexpected unlock

Fixes: d591fb56618f ("device_cgroup: simplify cgroup tree walk in propagate_exception()")
Cc: stable@xxxxxxxxxxxxxxx
Signed-off-by: Jann Horn <jannh@xxxxxxxxxx>
---
Compile-tested only.

I'm not entirely sure who's supposed to be the maintainer for this thing.
The sign-offs on the commits to this file come from Tejun, but MAINTAINERS
claims it's part of security/, so I'm just sending this to both the
security folks and the cgroup folks, you can figure out whose tree you want
to take this through. :P
If the cgroup folks feel responsible for this file, maybe you could fix up
MAINTAINERS?

 security/device_cgroup.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/security/device_cgroup.c b/security/device_cgroup.c
index cd97929fac66..dc28914fa72e 100644
--- a/security/device_cgroup.c
+++ b/security/device_cgroup.c
@@ -560,7 +560,7 @@ static int propagate_exception(struct dev_cgroup *devcg_root,
 		    devcg->behavior == DEVCG_DEFAULT_ALLOW) {
 			rc = dev_exception_add(devcg, ex);
 			if (rc)
-				break;
+				return rc;
 		} else {
 			/*
 			 * in the other possible cases:
-- 
2.21.0.225.g810b269d1ac-goog