l2tp: fix a sock refcnt leak in l2tp_tunnel_register

This issue happens when trying to add an existent tunnel. It
doesn't call sock_put() before returning -EEXIST to release
the sock refcnt that was held by calling sock_hold() before
the existence check.

This patch is to fix it by holding the sock after doing the
existence check.

Fixes: f6cd651b ("l2tp: fix race in duplicate tunnel detection")
Reported-by: Jianlin Shi <jishi@redhat.com>
Signed-off-by: Xin Long <lucien.xin@gmail.com>
Reviewed-by: Guillaume Nault <g.nault@alphalink.fr>
Signed-off-by: David S. Miller <davem@davemloft.net>

net/l2tp/l2tp_core.c

@@ -1490,12 +1490,7 @@ int l2tp_tunnel_register(struct l2tp_tunnel *tunnel, struct net *net,
 			goto err_sock;
 	}
 
-	sk = sock->sk;
-
-	sock_hold(sk);
-	tunnel->sock = sk;
 	tunnel->l2tp_net = net;
-
 	pn = l2tp_pernet(net);
 
 	spin_lock_bh(&pn->l2tp_tunnel_list_lock);
@@ -1510,6 +1505,10 @@ int l2tp_tunnel_register(struct l2tp_tunnel *tunnel, struct net *net,
 	list_add_rcu(&tunnel->list, &pn->l2tp_tunnel_list);
 	spin_unlock_bh(&pn->l2tp_tunnel_list_lock);
 
+	sk = sock->sk;
+	sock_hold(sk);
+	tunnel->sock = sk;
+
 	if (tunnel->encap == L2TP_ENCAPTYPE_UDP) {
 		struct udp_tunnel_sock_cfg udp_cfg = {
 			.sk_user_data = tunnel,