gue: Implement direction IP encapsulation

This patch implements direct encapsulation of IPv4 and IPv6 packets
in UDP. This is done a version "1" of GUE and as explained in I-D
draft-ietf-nvo3-gue-03.

Changes here are only in the receive path, fou with IPxIPx already
supports the transmit side. Both the normal receive path and
GRO path are modified to check for GUE version and check for
IP version in the case that GUE version is "1".

Tested:

IPIP with direct GUE encap
  1 TCP_STREAM
    4530 Mbps
  200 TCP_RR
    1297625 tps
    135/232/444 90/95/99% latencies

IP4IP6 with direct GUE encap
  1 TCP_STREAM
    4903 Mbps
  200 TCP_RR
    1184481 tps
    149/253/473 90/95/99% latencies

IP6IP6 direct GUE encap
  1 TCP_STREAM
   5146 Mbps
  200 TCP_RR
    1202879 tps
    146/251/472 90/95/99% latencies

SIT with direct GUE encap
  1 TCP_STREAM
    6111 Mbps
  200 TCP_RR
    1250337 tps
    139/241/467 90/95/99% latencies

Signed-off-by: Tom Herbert <tom@herbertland.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
diff --git a/net/ipv4/fou.c b/net/ipv4/fou.c
index 5f9207c..321d57f 100644
--- a/net/ipv4/fou.c
+++ b/net/ipv4/fou.c
@@ -129,6 +129,36 @@
 
 	guehdr = (struct guehdr *)&udp_hdr(skb)[1];
 
+	switch (guehdr->version) {
+	case 0: /* Full GUE header present */
+		break;
+
+	case 1: {
+		/* Direct encasulation of IPv4 or IPv6 */
+
+		int prot;
+
+		switch (((struct iphdr *)guehdr)->version) {
+		case 4:
+			prot = IPPROTO_IPIP;
+			break;
+		case 6:
+			prot = IPPROTO_IPV6;
+			break;
+		default:
+			goto drop;
+		}
+
+		if (fou_recv_pull(skb, fou, sizeof(struct udphdr)))
+			goto drop;
+
+		return -prot;
+	}
+
+	default: /* Undefined version */
+		goto drop;
+	}
+
 	optlen = guehdr->hlen << 2;
 	len += optlen;
 
@@ -289,6 +319,7 @@
 	int flush = 1;
 	struct fou *fou = fou_from_sock(sk);
 	struct gro_remcsum grc;
+	u8 proto;
 
 	skb_gro_remcsum_init(&grc);
 
@@ -302,6 +333,25 @@
 			goto out;
 	}
 
+	switch (guehdr->version) {
+	case 0:
+		break;
+	case 1:
+		switch (((struct iphdr *)guehdr)->version) {
+		case 4:
+			proto = IPPROTO_IPIP;
+			break;
+		case 6:
+			proto = IPPROTO_IPV6;
+			break;
+		default:
+			goto out;
+		}
+		goto next_proto;
+	default:
+		goto out;
+	}
+
 	optlen = guehdr->hlen << 2;
 	len += optlen;
 
@@ -370,6 +420,10 @@
 		}
 	}
 
+	proto = guehdr->proto_ctype;
+
+next_proto:
+
 	/* We can clear the encap_mark for GUE as we are essentially doing
 	 * one of two possible things.  We are either adding an L4 tunnel
 	 * header to the outer L3 tunnel header, or we are are simply
@@ -383,7 +437,7 @@
 
 	rcu_read_lock();
 	offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads;
-	ops = rcu_dereference(offloads[guehdr->proto_ctype]);
+	ops = rcu_dereference(offloads[proto]);
 	if (WARN_ON_ONCE(!ops || !ops->callbacks.gro_receive))
 		goto out_unlock;
 
@@ -404,13 +458,30 @@
 	const struct net_offload **offloads;
 	struct guehdr *guehdr = (struct guehdr *)(skb->data + nhoff);
 	const struct net_offload *ops;
-	unsigned int guehlen;
+	unsigned int guehlen = 0;
 	u8 proto;
 	int err = -ENOENT;
 
-	proto = guehdr->proto_ctype;
-
-	guehlen = sizeof(*guehdr) + (guehdr->hlen << 2);
+	switch (guehdr->version) {
+	case 0:
+		proto = guehdr->proto_ctype;
+		guehlen = sizeof(*guehdr) + (guehdr->hlen << 2);
+		break;
+	case 1:
+		switch (((struct iphdr *)guehdr)->version) {
+		case 4:
+			proto = IPPROTO_IPIP;
+			break;
+		case 6:
+			proto = IPPROTO_IPV6;
+			break;
+		default:
+			return err;
+		}
+		break;
+	default:
+		return err;
+	}
 
 	rcu_read_lock();
 	offloads = NAPI_GRO_CB(skb)->is_ipv6 ? inet6_offloads : inet_offloads;