netlink: hold nl_sock_hash_lock during diag dump

Although RCU protection would be possible during diag dump, doing
so allows for concurrent table mutations which can render the
in-table offset between individual Netlink messages invalid and
thus cause legitimate sockets to be skipped in the dump.

Since the diag dump is relatively low volume and consistency is
more important than performance, the table mutex is held during
dump.

Reported-by: Andrey Wagin <avagin@gmail.com>
Signed-off-by: Thomas Graf <tgraf@suug.ch>
Fixes: e341694 ("netlink: Convert netlink_lookup() to use RCU protected hash table")
Signed-off-by: David S. Miller <davem@davemloft.net>

net/netlink/af_netlink.c

@@ -104,6 +104,7 @@ static atomic_t nl_table_users = ATOMIC_INIT(0);
 
 /* Protects netlink socket hash table mutations */
 DEFINE_MUTEX(nl_sk_hash_lock);
+EXPORT_SYMBOL_GPL(nl_sk_hash_lock);
 
 static int lockdep_nl_sk_hash_is_held(void)
 {

net/netlink/af_netlink.h

@@ -73,5 +73,6 @@ struct netlink_table {
 
 extern struct netlink_table *nl_table;
 extern rwlock_t nl_table_lock;
+extern struct mutex nl_sk_hash_lock;
 
 #endif

net/netlink/diag.c

@@ -170,6 +170,7 @@ static int netlink_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
 
 	req = nlmsg_data(cb->nlh);
 
+	mutex_lock(&nl_sk_hash_lock);
 	read_lock(&nl_table_lock);
 
 	if (req->sdiag_protocol == NDIAG_PROTO_ALL) {
@@ -183,13 +184,15 @@ static int netlink_diag_dump(struct sk_buff *skb, struct netlink_callback *cb)
 	} else {
 		if (req->sdiag_protocol >= MAX_LINKS) {
 			read_unlock(&nl_table_lock);
+			mutex_unlock(&nl_sk_hash_lock);
 			return -ENOENT;
 		}
 
 		__netlink_diag_dump(skb, cb, req->sdiag_protocol, s_num);
 	}
 
 	read_unlock(&nl_table_lock);
+	mutex_unlock(&nl_sk_hash_lock);
 
 	return skb->len;
 }