[PATCH net-next 11/18] ipv4: Add helpers for neigh lookup for nexthop

[David Ahern <dsahern@kernel.org>]

From: David Ahern <dsahern@gmail.com>

A common them in the output path is looking up a neigh entry for a
nexthop, either the gateway in an rtable or a fallback to the daddr
in the skb:

        nexthop = (__force u32)rt_nexthop(rt, ip_hdr(skb)->daddr);
        neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
        if (unlikely(!neigh))
                neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);

To allow the nexthop to be an IPv6 address we need to consider the
family of the nexthop and then call __ipv*_neigh_lookup_noref based
on it.

To make this simpler, add a ip_neigh_gw4 helper similar to ip_neigh_gw6
which handles:

        neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
        if (unlikely(!neigh))
                neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);

And then add a second one, ip_neigh_for_gw, that calls either
ip_neigh_gw4 or ip_neigh_gw6 based on the address family of the gateway.

Update the output paths in the VRF driver and core v4 code to use
ip_neigh_for_gw simplifying the family based lookup and making both
ready for a v6 nexthop.

ipv4_neigh_lookup has a different need - the potential to resolve a
passed in address in addition to any gateway in the rtable or skb. Since
this is a one-off, add ip_neigh_gw4 and ip_neigh_gw6 diectly. The
difference between __neigh_create used by the helpers and neigh_create
called by ipv4_neigh_lookup is taking a refcount, so add rcu_read_lock_bh
and bump the refcnt on the neigh entry.

Signed-off-by: David Ahern <dsahern@gmail.com>
---
 drivers/net/vrf.c    | 10 ++++------
 include/net/route.h  | 32 ++++++++++++++++++++++++++++++++
 net/ipv4/ip_output.c | 11 ++++-------
 net/ipv4/route.c     | 29 +++++++++++++++++++----------
 4 files changed, 59 insertions(+), 23 deletions(-)

diff --git a/drivers/net/vrf.c b/drivers/net/vrf.c
index 60c36d06bb4f..abca05cb856a 100644
--- a/drivers/net/vrf.c
+++ b/drivers/net/vrf.c
@@ -549,7 +549,7 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
 	struct net_device *dev = dst->dev;
 	unsigned int hh_len = LL_RESERVED_SPACE(dev);
 	struct neighbour *neigh;
-	u32 nexthop;
+	bool is_v6gw = false;
 	int ret = -EINVAL;
 
 	nf_reset(skb);
@@ -572,13 +572,11 @@ static int vrf_finish_output(struct net *net, struct sock *sk, struct sk_buff *s
 
 	rcu_read_lock_bh();
 
-	nexthop = (__force u32)rt_nexthop(rt, ip_hdr(skb)->daddr);
-	neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
-	if (unlikely(!neigh))
-		neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);
+	neigh = ip_neigh_for_gw(rt, skb, &is_v6gw);
 	if (!IS_ERR(neigh)) {
 		sock_confirm_neigh(skb, neigh);
-		ret = neigh_output(neigh, skb, false);
+		/* if crossing protocols, can not used the cached header */
+		ret = neigh_output(neigh, skb, is_v6gw);
 		rcu_read_unlock_bh();
 		return ret;
 	}
diff --git a/include/net/route.h b/include/net/route.h
index 5d28a2509b58..96f6c9ae33c2 100644
--- a/include/net/route.h
+++ b/include/net/route.h
@@ -29,6 +29,8 @@
 #include <net/flow.h>
 #include <net/inet_sock.h>
 #include <net/ip_fib.h>
+#include <net/arp.h>
+#include <net/ndisc.h>
 #include <linux/in_route.h>
 #include <linux/rtnetlink.h>
 #include <linux/rcupdate.h>
@@ -350,4 +352,34 @@ static inline int ip4_dst_hoplimit(const struct dst_entry *dst)
 	return hoplimit;
 }
 
+static inline struct neighbour *ip_neigh_gw4(struct net_device *dev,
+					     __be32 daddr)
+{
+	struct neighbour *neigh;
+
+	neigh = __ipv4_neigh_lookup_noref(dev, daddr);
+	if (unlikely(!neigh))
+		neigh = __neigh_create(&arp_tbl, &daddr, dev, false);
+
+	return neigh;
+}
+
+static inline struct neighbour *ip_neigh_for_gw(struct rtable *rt,
+						struct sk_buff *skb,
+						bool *is_v6gw)
+{
+	struct net_device *dev = rt->dst.dev;
+	struct neighbour *neigh;
+
+	if (likely(rt->rt_gw_family == AF_INET)) {
+		neigh = ip_neigh_gw4(dev, rt->rt_gw4);
+	} else if (rt->rt_gw_family == AF_INET6) {
+		neigh = ip_neigh_gw6(dev, &rt->rt_gw6);
+		*is_v6gw = true;
+	} else {
+		neigh = ip_neigh_gw4(dev, ip_hdr(skb)->daddr);
+	}
+	return neigh;
+}
+
 #endif	/* _ROUTE_H */
diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index 8cf8b0118f70..e5b8d7726737 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -188,7 +188,7 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s
 	struct net_device *dev = dst->dev;
 	unsigned int hh_len = LL_RESERVED_SPACE(dev);
 	struct neighbour *neigh;
-	u32 nexthop;
+	bool is_v6gw = false;
 
 	if (rt->rt_type == RTN_MULTICAST) {
 		IP_UPD_PO_STATS(net, IPSTATS_MIB_OUTMCAST, skb->len);
@@ -218,16 +218,13 @@ static int ip_finish_output2(struct net *net, struct sock *sk, struct sk_buff *s
 	}
 
 	rcu_read_lock_bh();
-	nexthop = (__force u32) rt_nexthop(rt, ip_hdr(skb)->daddr);
-	neigh = __ipv4_neigh_lookup_noref(dev, nexthop);
-	if (unlikely(!neigh))
-		neigh = __neigh_create(&arp_tbl, &nexthop, dev, false);
+	neigh = ip_neigh_for_gw(rt, skb, &is_v6gw);
 	if (!IS_ERR(neigh)) {
 		int res;
 
 		sock_confirm_neigh(skb, neigh);
-		res = neigh_output(neigh, skb, false);
-
+		/* if crossing protocols, can not used the cached header */
+		res = neigh_output(neigh, skb, is_v6gw);
 		rcu_read_unlock_bh();
 		return res;
 	}
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 6e58acf0a87b..32ecb4c1c7e3 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -436,18 +436,27 @@ static struct neighbour *ipv4_neigh_lookup(const struct dst_entry *dst,
 {
 	const struct rtable *rt = container_of(dst, struct rtable, dst);
 	struct net_device *dev = dst->dev;
-	const __be32 *pkey = daddr;
 	struct neighbour *n;
 
-	if (rt->rt_gw_family == AF_INET)
-		pkey = (const __be32 *) &rt->rt_gw4;
-	else if (skb)
-		pkey = &ip_hdr(skb)->daddr;
-
-	n = __ipv4_neigh_lookup(dev, *(__force u32 *)pkey);
-	if (n)
-		return n;
-	return neigh_create(&arp_tbl, pkey, dev);
+	rcu_read_lock_bh();
+
+	if (likely(rt->rt_gw_family == AF_INET)) {
+		n = ip_neigh_gw4(dev, rt->rt_gw4);
+	} else if (rt->rt_gw_family == AF_INET6) {
+		n = ip_neigh_gw6(dev, &rt->rt_gw6);
+        } else {
+		__be32 pkey;
+
+		pkey = skb ? ip_hdr(skb)->daddr : *((__be32 *) daddr);
+		n = ip_neigh_gw4(dev, pkey);
+	}
+
+	if (n && !refcount_inc_not_zero(&n->refcnt))
+		n = NULL;
+
+	rcu_read_unlock_bh();
+
+	return n;
 }
 
 static void ipv4_confirm_neigh(const struct dst_entry *dst, const void *daddr)
-- 
2.11.0