[PATCH v3 1/2] Implement IP_UNICAST_IF socket option.
From: Erich E. Hoover <hidden>
Date: 2012-02-06 17:58:01
Subsystem:
networking [general], networking [ipv4/ipv6], the rest, user datagram protocol (udp) · Maintainers:
"David S. Miller", Eric Dumazet, Jakub Kicinski, Paolo Abeni, David Ahern, Ido Schimmel, Linus Torvalds, Willem de Bruijn
The IP_UNICAST_IF feature is needed by the Wine project. This patch implements the feature by setting the outgoing interface in a similar fashion to that of IP_PKTINFO. A separate option is needed to handle this feature since the existing options do not provide all of the characteristics required by IP_UNICAST_IF, a summary is provided below. SO_BINDTODEVICE: * SO_BINDTODEVICE requires administrative privileges, IP_UNICAST_IF does not. From reading some old mailing list articles my understanding is that SO_BINDTODEVICE requires administrative privileges because it can override the administrator's routing settings. * The SO_BINDTODEVICE option restricts both outbound and inbound traffic, IP_UNICAST_IF only impacts outbound traffic. IP_PKTINFO: * Since IP_PKTINFO and IP_UNICAST_IF are independent options, implementing IP_UNICAST_IF with IP_PKTINFO will likely break some applications. * Implementing IP_UNICAST_IF on top of IP_PKTINFO significantly complicates the Wine codebase and reduces the socket performance (doing this requires a lot of extra communication between the "server" and "user" layers). bind(): * bind() does not work on broadcast packets, IP_UNICAST_IF is specifically intended to work with broadcast packets. Signed-off-by: Erich E. Hoover <redacted> --- include/linux/in.h | 1 + include/net/inet_sock.h | 2 + include/net/ip.h | 1 + net/ipv4/af_inet.c | 2 + net/ipv4/ip_sockglue.c | 48 +++++++++++++++++++++++++++++++++++++++++++++++ net/ipv4/ping.c | 2 +- net/ipv4/raw.c | 2 +- net/ipv4/udp.c | 2 +- net/ipv6/af_inet6.c | 2 + 9 files changed, 59 insertions(+), 3 deletions(-)
diff --git a/include/linux/in.h b/include/linux/in.h
index 01129c0..89f6682 100644
--- a/include/linux/in.h
+++ b/include/linux/in.h@@ -86,6 +86,7 @@ struct in_addr { #define IP_MINTTL 21 #define IP_NODEFRAG 22 +#define IP_UNICAST_IF 23 /* IP_MTU_DISCOVER values */ #define IP_PMTUDISC_DONT 0 /* Never send DF frames */
diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index e3e4051..ad517f5 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h@@ -132,6 +132,7 @@ struct rtable; * @tos - TOS * @mc_ttl - Multicasting TTL * @is_icsk - is this an inet_connection_sock? + * @outif_index - Outgoing device index * @mc_index - Multicast device index * @mc_list - Group array * @cork - info to build ip hdr on each ip frag while socket is corked
@@ -167,6 +168,7 @@ struct inet_sock { transparent:1, mc_all:1, nodefrag:1; + int outif_index; int mc_index; __be32 mc_addr; struct ip_mc_socklist __rcu *mc_list;
diff --git a/include/net/ip.h b/include/net/ip.h
index 775009f..05aa269 100644
--- a/include/net/ip.h
+++ b/include/net/ip.h@@ -452,6 +452,7 @@ extern int ip_options_rcv_srr(struct sk_buff *skb); extern void ipv4_pktinfo_prepare(struct sk_buff *skb); extern void ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb); +extern int ip_default_ifindex(const struct sock *sk); extern int ip_cmsg_send(struct net *net, struct msghdr *msg, struct ipcm_cookie *ipc); extern int ip_setsockopt(struct sock *sk, int level, int optname, char __user *optval, unsigned int optlen);
diff --git a/net/ipv4/af_inet.c b/net/ipv4/af_inet.c
index f7b5670..a5855cd 100644
--- a/net/ipv4/af_inet.c
+++ b/net/ipv4/af_inet.c@@ -375,6 +375,8 @@ lookup_protocol: sk->sk_protocol = protocol; sk->sk_backlog_rcv = sk->sk_prot->backlog_rcv; + inet->outif_index = 0; + inet->uc_ttl = -1; inet->mc_loop = 1; inet->mc_ttl = 1;
diff --git a/net/ipv4/ip_sockglue.c b/net/ipv4/ip_sockglue.c
index 8aa87c1..87646e4 100644
--- a/net/ipv4/ip_sockglue.c
+++ b/net/ipv4/ip_sockglue.c@@ -186,6 +186,21 @@ void ip_cmsg_recv(struct msghdr *msg, struct sk_buff *skb) } EXPORT_SYMBOL(ip_cmsg_recv); +int ip_default_ifindex(const struct sock *sk) +{ + struct inet_sock *inet = inet_sk(sk); + int ifindex = sk->sk_bound_dev_if; + + /* + * If not bound to a specific interface then set the outgoing interface + * to the value from the IP_UNICAST_IF socket option. + */ + if (!ifindex) + ifindex = inet->outif_index; + + return ifindex; +} + int ip_cmsg_send(struct net *net, struct msghdr *msg, struct ipcm_cookie *ipc) { int err;
@@ -469,6 +484,7 @@ static int do_ip_setsockopt(struct sock *sk, int level, (1<<IP_ROUTER_ALERT) | (1<<IP_FREEBIND) | (1<<IP_PASSSEC) | (1<<IP_TRANSPARENT) | (1<<IP_MINTTL) | (1<<IP_NODEFRAG))) || + optname == IP_UNICAST_IF || optname == IP_MULTICAST_TTL || optname == IP_MULTICAST_ALL || optname == IP_MULTICAST_LOOP ||
@@ -628,6 +644,35 @@ static int do_ip_setsockopt(struct sock *sk, int level, goto e_inval; inet->mc_loop = !!val; break; + case IP_UNICAST_IF: + { + struct net_device *dev = NULL; + int ifindex; + + if (optlen != sizeof(int)) + goto e_inval; + + ifindex = (__force int)ntohl((__force __be32)val); + if (ifindex == 0) { + inet->outif_index = 0; + err = 0; + break; + } + + dev = dev_get_by_index(sock_net(sk), ifindex); + err = -EADDRNOTAVAIL; + if (!dev) + break; + dev_put(dev); + + err = -EINVAL; + if (sk->sk_bound_dev_if && ifindex != sk->sk_bound_dev_if) + break; + + inet->outif_index = ifindex; + err = 0; + break; + } case IP_MULTICAST_IF: { struct ip_mreqn mreq;
@@ -1178,6 +1223,9 @@ static int do_ip_getsockopt(struct sock *sk, int level, int optname, case IP_MULTICAST_LOOP: val = inet->mc_loop; break; + case IP_UNICAST_IF: + val = (__force int)htonl((__u32) inet->outif_index); + break; case IP_MULTICAST_IF: { struct in_addr addr;
diff --git a/net/ipv4/ping.c b/net/ipv4/ping.c
index aea5a19..abeb454 100644
--- a/net/ipv4/ping.c
+++ b/net/ipv4/ping.c@@ -510,7 +510,7 @@ static int ping_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.addr = inet->inet_saddr; ipc.opt = NULL; - ipc.oif = sk->sk_bound_dev_if; + ipc.oif = ip_default_ifindex(sk); ipc.tx_flags = 0; err = sock_tx_timestamp(sk, &ipc.tx_flags); if (err)
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 3ccda5a..000d9fb 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c@@ -515,7 +515,7 @@ static int raw_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, ipc.addr = inet->inet_saddr; ipc.opt = NULL; ipc.tx_flags = 0; - ipc.oif = sk->sk_bound_dev_if; + ipc.oif = ip_default_ifindex(sk); if (msg->msg_controllen) { err = ip_cmsg_send(sock_net(sk), msg, &ipc);
diff --git a/net/ipv4/udp.c b/net/ipv4/udp.c
index 5d075b5..651eb62 100644
--- a/net/ipv4/udp.c
+++ b/net/ipv4/udp.c@@ -869,7 +869,7 @@ int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg, } ipc.addr = inet->inet_saddr; - ipc.oif = sk->sk_bound_dev_if; + ipc.oif = ip_default_ifindex(sk); err = sock_tx_timestamp(sk, &ipc.tx_flags); if (err) return err;
diff --git a/net/ipv6/af_inet6.c b/net/ipv6/af_inet6.c
index 273f48d..13b84bc 100644
--- a/net/ipv6/af_inet6.c
+++ b/net/ipv6/af_inet6.c@@ -210,6 +210,8 @@ lookup_protocol: */ inet->uc_ttl = -1; + inet->outif_index = 0; + inet->mc_loop = 1; inet->mc_ttl = 1; inet->mc_index = 0;
--
1.7.5.4