net/ipv6/udp_tunnel: prefer SO_BINDTOIFINDEX over SO_BINDTODEVICE
authorDavid Herrmann <dh.herrmann@gmail.com>
Tue, 15 Jan 2019 13:42:16 +0000 (14:42 +0100)
committerDavid S. Miller <davem@davemloft.net>
Thu, 17 Jan 2019 22:55:52 +0000 (14:55 -0800)
The udp-tunnel setup allows binding sockets to a network device. Prefer
the new SO_BINDTOIFINDEX to avoid temporarily resolving the device-name
just to look it up in the ioctl again.

Reviewed-by: Tom Gundersen <teg@jklm.no>
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
net/ipv6/ip6_udp_tunnel.c

index ad1a9ccd4b44a1079a09115216e82c1c3f07c709..25430c991cea34e952a47f01326f400556b08d1e 100644 (file)
@@ -32,18 +32,9 @@ int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
                        goto error;
        }
        if (cfg->bind_ifindex) {
-               struct net_device *dev;
-
-               dev = dev_get_by_index(net, cfg->bind_ifindex);
-               if (!dev) {
-                       err = -ENODEV;
-                       goto error;
-               }
-
-               err = kernel_setsockopt(sock, SOL_SOCKET, SO_BINDTODEVICE,
-                                       dev->name, strlen(dev->name) + 1);
-               dev_put(dev);
-
+               err = kernel_setsockopt(sock, SOL_SOCKET, SO_BINDTOIFINDEX,
+                                       (void *)&cfg->bind_ifindex,
+                                       sizeof(cfg->bind_ifindex));
                if (err < 0)
                        goto error;
        }