netlink: use kfree_rcu() in netlink_release()

On some suspend/resume operations involving wimax device, we have
noticed some intermittent memory corruptions in netlink code.

Stéphane Marchesin tracked this corruption in netlink_update_listeners()
and suggested a patch.

It appears netlink_release() should use kfree_rcu() instead of kfree()
for the listeners structure as it may be used by other cpus using RCU
protection.

netlink_release() must set to NULL the listeners pointer when
it is about to be freed.

Also have to protect netlink_update_listeners() and
netlink_has_listeners() if listeners is NULL.

Add a nl_deref_protected() lockdep helper to properly document which
locks protects us.

Reported-by: Jonathan Kliegman <kliegs@google.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Stéphane Marchesin <marcheu@google.com>
Cc: Sam Leffler <sleffler@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/netlink/af_netlink.c

@@ -138,6 +138,8 @@ static int netlink_dump(struct sock *sk);
 static DEFINE_RWLOCK(nl_table_lock);
 static atomic_t nl_table_users = ATOMIC_INIT(0);
 
+#define nl_deref_protected(X) rcu_dereference_protected(X, lockdep_is_held(&nl_table_lock));
+
 static ATOMIC_NOTIFIER_HEAD(netlink_chain);
 
 static inline u32 netlink_group_mask(u32 group)
@@ -345,14 +347,19 @@ netlink_update_listeners(struct sock *sk)
 	struct hlist_node *node;
 	unsigned long mask;
 	unsigned int i;
+	struct listeners *listeners;
+
+	listeners = nl_deref_protected(tbl->listeners);
+	if (!listeners)
+		return;
 
 	for (i = 0; i < NLGRPLONGS(tbl->groups); i++) {
 		mask = 0;
 		sk_for_each_bound(sk, node, &tbl->mc_list) {
 			if (i < NLGRPLONGS(nlk_sk(sk)->ngroups))
 				mask |= nlk_sk(sk)->groups[i];
 		}
-		tbl->listeners->masks[i] = mask;
+		listeners->masks[i] = mask;
 	}
 	/* this function is only called with the netlink table "grabbed", which
 	 * makes sure updates are visible before bind or setsockopt return. */
@@ -536,7 +543,11 @@ static int netlink_release(struct socket *sock)
 	if (netlink_is_kernel(sk)) {
 		BUG_ON(nl_table[sk->sk_protocol].registered == 0);
 		if (--nl_table[sk->sk_protocol].registered == 0) {
-			kfree(nl_table[sk->sk_protocol].listeners);
+			struct listeners *old;
+
+			old = nl_deref_protected(nl_table[sk->sk_protocol].listeners);
+			RCU_INIT_POINTER(nl_table[sk->sk_protocol].listeners, NULL);
+			kfree_rcu(old, rcu);
 			nl_table[sk->sk_protocol].module = NULL;
 			nl_table[sk->sk_protocol].bind = NULL;
 			nl_table[sk->sk_protocol].flags = 0;
@@ -982,7 +993,7 @@ int netlink_has_listeners(struct sock *sk, unsigned int group)
 	rcu_read_lock();
 	listeners = rcu_dereference(nl_table[sk->sk_protocol].listeners);
 
-	if (group - 1 < nl_table[sk->sk_protocol].groups)
+	if (listeners && group - 1 < nl_table[sk->sk_protocol].groups)
 		res = test_bit(group - 1, listeners->masks);
 
 	rcu_read_unlock();
@@ -1625,7 +1636,7 @@ int __netlink_change_ngroups(struct sock *sk, unsigned int groups)
 		new = kzalloc(sizeof(*new) + NLGRPSZ(groups), GFP_ATOMIC);
 		if (!new)
 			return -ENOMEM;
-		old = rcu_dereference_protected(tbl->listeners, 1);
+		old = nl_deref_protected(tbl->listeners);
 		memcpy(new->masks, old->masks, NLGRPSZ(tbl->groups));
 		rcu_assign_pointer(tbl->listeners, new);