From foo@baz Tue 01 Oct 2019 03:27:55 PM CEST
From: Ka-Cheong Poon <ka-cheong.poon@oracle.com>
Date: Tue, 24 Sep 2019 08:51:16 -0700
Subject: net/rds: Check laddr_check before calling it

From: Ka-Cheong Poon <ka-cheong.poon@oracle.com>

[ Upstream commit 05733434ee9ae6548723a808647248583e347cca ]

In rds_bind(), laddr_check is called without checking if it is NULL or
not.  And rs_transport should be reset if rds_add_bound() fails.

Fixes: c5c1a030a7db ("net/rds: An rds_sock is added too early to the hash table")
Reported-by: syzbot+fae39afd2101a17ec624@syzkaller.appspotmail.com
Signed-off-by: Ka-Cheong Poon <ka-cheong.poon@oracle.com>
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/rds/bind.c |    5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)

--- a/net/rds/bind.c
+++ b/net/rds/bind.c
@@ -244,7 +244,8 @@ int rds_bind(struct socket *sock, struct
 	 */
 	if (rs->rs_transport) {
 		trans = rs->rs_transport;
-		if (trans->laddr_check(sock_net(sock->sk),
+		if (!trans->laddr_check ||
+		    trans->laddr_check(sock_net(sock->sk),
 				       binding_addr, scope_id) != 0) {
 			ret = -ENOPROTOOPT;
 			goto out;
@@ -263,6 +264,8 @@ int rds_bind(struct socket *sock, struct
 
 	sock_set_flag(sk, SOCK_RCU_FREE);
 	ret = rds_add_bound(rs, binding_addr, &port, scope_id);
+	if (ret)
+		rs->rs_transport = NULL;
 
 out:
 	release_sock(sk);