Merge git://git.kernel.org/pub/scm/linux/kernel/git/davem/net

The new route handling in ip_mc_finish_output() from 'net' overlapped
with the new support for returning congestion notifications from BPF
programs.

In order to handle this I had to take the dev_loopback_xmit() calls
out of the switch statement.

The aquantia driver conflicts were simple overlapping changes.

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

3 files changed, 55 insertions, 5 deletions

diff --git a/net/ipv4/ip_output.c b/net/ipv4/ip_output.c
index cdd6c3418b9e..cc7ef0d05bbd 100644
--- a/net/ipv4/ip_output.c
+++ b/net/ipv4/ip_output.c
@@ -327,18 +327,35 @@ static int ip_finish_output(struct net *net, struct sock *sk, struct sk_buff *sk
 static int ip_mc_finish_output(struct net *net, struct sock *sk,
 			       struct sk_buff *skb)
 {
-	int ret;
+	struct rtable *new_rt;
+	bool do_cn = false;
+	int ret, err;
 
 	ret = BPF_CGROUP_RUN_PROG_INET_EGRESS(sk, skb);
 	switch (ret) {
-	case NET_XMIT_SUCCESS:
-		return dev_loopback_xmit(net, sk, skb);
 	case NET_XMIT_CN:
-		return dev_loopback_xmit(net, sk, skb) ? : ret;
+		do_cn = true;
+		/* fall through */
+	case NET_XMIT_SUCCESS:
+		break;
 	default:
 		kfree_skb(skb);
 		return ret;
 	}
+
+	/* Reset rt_iif so that inet_iif() will return skb->skb_iif. Setting
+	 * this to non-zero causes ipi_ifindex in in_pktinfo to be overwritten,
+	 * see ipv4_pktinfo_prepare().
+	 */
+	new_rt = rt_dst_clone(net->loopback_dev, skb_rtable(skb));
+	if (new_rt) {
+		new_rt->rt_iif = 0;
+		skb_dst_drop(skb);
+		skb_dst_set(skb, &new_rt->dst);
+	}
+
+	err = dev_loopback_xmit(net, sk, skb);
+	return (do_cn && err) ? ret : err;
 }
 
 int ip_mc_output(struct net *net, struct sock *sk, struct sk_buff *skb)
diff --git a/net/ipv4/raw.c b/net/ipv4/raw.c
index 0b8e06ca75d6..40a6abbc9cf6 100644
--- a/net/ipv4/raw.c
+++ b/net/ipv4/raw.c
@@ -197,7 +197,7 @@ static int raw_v4_input(struct sk_buff *skb, const struct iphdr *iph, int hash)
 		}
 		sk = __raw_v4_lookup(net, sk_next(sk), iph->protocol,
 				     iph->saddr, iph->daddr,
-				     skb->dev->ifindex, sdif);
+				     dif, sdif);
 	}
 out:
 	read_unlock(&raw_v4_hashinfo.lock);
diff --git a/net/ipv4/route.c b/net/ipv4/route.c
index 59670fafcd26..a3e466b6a60c 100644
--- a/net/ipv4/route.c
+++ b/net/ipv4/route.c
@@ -1648,6 +1648,39 @@ struct rtable *rt_dst_alloc(struct net_device *dev,
 }
 EXPORT_SYMBOL(rt_dst_alloc);
 
+struct rtable *rt_dst_clone(struct net_device *dev, struct rtable *rt)
+{
+	struct rtable *new_rt;
+
+	new_rt = dst_alloc(&ipv4_dst_ops, dev, 1, DST_OBSOLETE_FORCE_CHK,
+			   rt->dst.flags);
+
+	if (new_rt) {
+		new_rt->rt_genid = rt_genid_ipv4(dev_net(dev));
+		new_rt->rt_flags = rt->rt_flags;
+		new_rt->rt_type = rt->rt_type;
+		new_rt->rt_is_input = rt->rt_is_input;
+		new_rt->rt_iif = rt->rt_iif;
+		new_rt->rt_pmtu = rt->rt_pmtu;
+		new_rt->rt_mtu_locked = rt->rt_mtu_locked;
+		new_rt->rt_gw_family = rt->rt_gw_family;
+		if (rt->rt_gw_family == AF_INET)
+			new_rt->rt_gw4 = rt->rt_gw4;
+		else if (rt->rt_gw_family == AF_INET6)
+			new_rt->rt_gw6 = rt->rt_gw6;
+		INIT_LIST_HEAD(&new_rt->rt_uncached);
+
+		new_rt->dst.flags |= DST_HOST;
+		new_rt->dst.input = rt->dst.input;
+		new_rt->dst.output = rt->dst.output;
+		new_rt->dst.error = rt->dst.error;
+		new_rt->dst.lastuse = jiffies;
+		new_rt->dst.lwtstate = lwtstate_get(rt->dst.lwtstate);
+	}
+	return new_rt;
+}
+EXPORT_SYMBOL(rt_dst_clone);
+
 /* called in rcu_read_lock() section */
 int ip_mc_validate_source(struct sk_buff *skb, __be32 daddr, __be32 saddr,
 			  u8 tos, struct net_device *dev,