audit: fix info leak in AUDIT_GET requests
authorMathias Krause <[email protected]>
Mon, 30 Sep 2013 20:04:24 +0000 (22:04 +0200)
committerEric Paris <[email protected]>
Tue, 5 Nov 2013 16:08:30 +0000 (11:08 -0500)
We leak 4 bytes of kernel stack in response to an AUDIT_GET request as
we miss to initialize the mask member of status_set. Fix that.

Cc: Al Viro <[email protected]>
Cc: Eric Paris <[email protected]>
Cc: [email protected] # v2.6.6+
Signed-off-by: Mathias Krause <[email protected]>
Signed-off-by: Richard Guy Briggs <[email protected]>
Signed-off-by: Eric Paris <[email protected]>
kernel/audit.c

index f3f36f5eb4ae4938c129ae225a95e1cf44da813f..e0f7767e4a3d6bf02426836796b8756d5452153a 100644 (file)
@@ -759,6 +759,7 @@ static int audit_receive_msg(struct sk_buff *skb, struct nlmsghdr *nlh)
 
        switch (msg_type) {
        case AUDIT_GET:
+               status_set.mask          = 0;
                status_set.enabled       = audit_enabled;
                status_set.failure       = audit_failure;
                status_set.pid           = audit_pid;