Merge branch 'l3mdev-accept'

David Ahern says:

====================
net: Allow accepted sockets to be bound to l3mdev domain

Allow accepted sockets to derive their sk_bound_dev_if setting from the
l3mdev domain in which the packets originated. This version adds a sysctl
to control whether the setting is inherited, making the functionality
similar to sk_mark and its sysctl_tcp_fwmark_accept setting.

This effectively allow a process to have a "VRF-global" listen socket,
with child sockets bound to the VRF device in which the packet originated.
A similar behavior can be achieved using sk_mark, but a solution using marks
is incomplete as it does not handle duplicate addresses in different L3
domains/VRFs. Allowing sockets to inherit the sk_bound_dev_if from l3mdev
domain provides a complete solution.
====================

Signed-off-by: David S. Miller <davem@davemloft.net>

3 files changed, 40 insertions, 0 deletions

diff --git a/include/net/inet_sock.h b/include/net/inet_sock.h
index 625bdf95d673..012b1f91f3ec 100644
--- a/include/net/inet_sock.h
+++ b/include/net/inet_sock.h
@@ -28,6 +28,7 @@
 #include <net/request_sock.h>
 #include <net/netns/hash.h>
 #include <net/tcp_states.h>
+#include <net/l3mdev.h>
 
 /** struct ip_options - IP Options
  *
@@ -113,6 +114,19 @@ static inline u32 inet_request_mark(const struct sock *sk, struct sk_buff *skb)
 	return sk->sk_mark;
 }
 
+static inline int inet_request_bound_dev_if(const struct sock *sk,
+					    struct sk_buff *skb)
+{
+#ifdef CONFIG_NET_L3_MASTER_DEV
+	struct net *net = sock_net(sk);
+
+	if (!sk->sk_bound_dev_if && net->ipv4.sysctl_tcp_l3mdev_accept)
+		return l3mdev_master_ifindex_by_index(net, skb->skb_iif);
+#endif
+
+	return sk->sk_bound_dev_if;
+}
+
 struct inet_cork {
 	unsigned int		flags;
 	__be32			addr;
diff --git a/include/net/l3mdev.h b/include/net/l3mdev.h
index 774d85b2d5d9..786226f8e77b 100644
--- a/include/net/l3mdev.h
+++ b/include/net/l3mdev.h
@@ -51,6 +51,24 @@ static inline int l3mdev_master_ifindex(struct net_device *dev)
 	return ifindex;
 }
 
+static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex)
+{
+	struct net_device *dev;
+	int rc = 0;
+
+	if (likely(ifindex)) {
+		rcu_read_lock();
+
+		dev = dev_get_by_index_rcu(net, ifindex);
+		if (dev)
+			rc = l3mdev_master_ifindex_rcu(dev);
+
+		rcu_read_unlock();
+	}
+
+	return rc;
+}
+
 /* get index of an interface to use for FIB lookups. For devices
  * enslaved to an L3 master device FIB lookups are based on the
  * master index
@@ -167,6 +185,11 @@ static inline int l3mdev_master_ifindex(struct net_device *dev)
 	return 0;
 }
 
+static inline int l3mdev_master_ifindex_by_index(struct net *net, int ifindex)
+{
+	return 0;
+}
+
 static inline int l3mdev_fib_oif_rcu(struct net_device *dev)
 {
 	return dev ? dev->ifindex : 0;
diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
index c68926b4899c..d75be32650ba 100644
--- a/include/net/netns/ipv4.h
+++ b/include/net/netns/ipv4.h
@@ -86,6 +86,9 @@ struct netns_ipv4 {
 
 	int sysctl_fwmark_reflect;
 	int sysctl_tcp_fwmark_accept;
+#ifdef CONFIG_NET_L3_MASTER_DEV
+	int sysctl_tcp_l3mdev_accept;
+#endif
 	int sysctl_tcp_mtu_probing;
 	int sysctl_tcp_base_mss;
 	int sysctl_tcp_probe_threshold;