6 files changed, 464 insertions, 304 deletions

diff --git a/net/ipv4/Makefile b/net/ipv4/Makefile
index 7fcf8101d85..86ded0bac9c 100644
--- a/net/ipv4/Makefile
+++ b/net/ipv4/Makefile
@@ -11,7 +11,7 @@ obj-y     := route.o inetpeer.o protocol.o \
 	     tcp_offload.o datagram.o raw.o udp.o udplite.o \
 	     udp_offload.o arp.o icmp.o devinet.o af_inet.o igmp.o \
 	     fib_frontend.o fib_semantics.o fib_trie.o \
-	     inet_fragment.o ping.o
+	     inet_fragment.o ping.o ip_tunnel_core.o
 
 obj-$(CONFIG_NET_IP_TUNNEL) += ip_tunnel.o
 obj-$(CONFIG_SYSCTL) += sysctl_net_ipv4.o
diff --git a/net/ipv4/gre.c b/net/ipv4/gre.c
index b2e805af9b8..ba4803e609b 100644
--- a/net/ipv4/gre.c
+++ b/net/ipv4/gre.c
@@ -13,6 +13,8 @@
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
 #include <linux/module.h>
+#include <linux/if.h>
+#include <linux/icmp.h>
 #include <linux/kernel.h>
 #include <linux/kmod.h>
 #include <linux/skbuff.h>
@@ -24,51 +26,270 @@
 #include <net/protocol.h>
 #include <net/gre.h>
 
+#include <net/icmp.h>
+#include <net/route.h>
+#include <net/xfrm.h>
 
 static const struct gre_protocol __rcu *gre_proto[GREPROTO_MAX] __read_mostly;
-static DEFINE_SPINLOCK(gre_proto_lock);
+static struct gre_cisco_protocol __rcu *gre_cisco_proto_list[GRE_IP_PROTO_MAX];
 
 int gre_add_protocol(const struct gre_protocol *proto, u8 version)
 {
 	if (version >= GREPROTO_MAX)
-		goto err_out;
-
-	spin_lock(&gre_proto_lock);
-	if (gre_proto[version])
-		goto err_out_unlock;
-
-	RCU_INIT_POINTER(gre_proto[version], proto);
-	spin_unlock(&gre_proto_lock);
-	return 0;
+		return -EINVAL;
 
-err_out_unlock:
-	spin_unlock(&gre_proto_lock);
-err_out:
-	return -1;
+	return (cmpxchg((const struct gre_protocol **)&gre_proto[version], NULL, proto) == NULL) ?
+		0 : -EBUSY;
 }
 EXPORT_SYMBOL_GPL(gre_add_protocol);
 
 int gre_del_protocol(const struct gre_protocol *proto, u8 version)
 {
+	int ret;
+
 	if (version >= GREPROTO_MAX)
-		goto err_out;
-
-	spin_lock(&gre_proto_lock);
-	if (rcu_dereference_protected(gre_proto[version],
-			lockdep_is_held(&gre_proto_lock)) != proto)
-		goto err_out_unlock;
-	RCU_INIT_POINTER(gre_proto[version], NULL);
-	spin_unlock(&gre_proto_lock);
+		return -EINVAL;
+
+	ret = (cmpxchg((const struct gre_protocol **)&gre_proto[version], proto, NULL) == proto) ?
+		0 : -EBUSY;
+
+	if (ret)
+		return ret;
+
 	synchronize_rcu();
 	return 0;
-
-err_out_unlock:
-	spin_unlock(&gre_proto_lock);
-err_out:
-	return -1;
 }
 EXPORT_SYMBOL_GPL(gre_del_protocol);
 
+void gre_build_header(struct sk_buff *skb, const struct tnl_ptk_info *tpi,
+		      int hdr_len)
+{
+	struct gre_base_hdr *greh;
+
+	skb_push(skb, hdr_len);
+
+	greh = (struct gre_base_hdr *)skb->data;
+	greh->flags = tnl_flags_to_gre_flags(tpi->flags);
+	greh->protocol = tpi->proto;
+
+	if (tpi->flags&(TUNNEL_KEY|TUNNEL_CSUM|TUNNEL_SEQ)) {
+		__be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4);
+
+		if (tpi->flags&TUNNEL_SEQ) {
+			*ptr = tpi->seq;
+			ptr--;
+		}
+		if (tpi->flags&TUNNEL_KEY) {
+			*ptr = tpi->key;
+			ptr--;
+		}
+		if (tpi->flags&TUNNEL_CSUM &&
+		    !(skb_shinfo(skb)->gso_type & SKB_GSO_GRE)) {
+			*ptr = 0;
+			*(__sum16 *)ptr = csum_fold(skb_checksum(skb, 0,
+								 skb->len, 0));
+		}
+	}
+}
+EXPORT_SYMBOL_GPL(gre_build_header);
+
+struct sk_buff *gre_handle_offloads(struct sk_buff *skb, bool gre_csum)
+{
+	int err;
+
+	if (likely(!skb->encapsulation)) {
+		skb_reset_inner_headers(skb);
+		skb->encapsulation = 1;
+	}
+
+	if (skb_is_gso(skb)) {
+		err = skb_unclone(skb, GFP_ATOMIC);
+		if (unlikely(err))
+			goto error;
+		skb_shinfo(skb)->gso_type |= SKB_GSO_GRE;
+		return skb;
+	} else if (skb->ip_summed == CHECKSUM_PARTIAL && gre_csum) {
+		err = skb_checksum_help(skb);
+		if (unlikely(err))
+			goto error;
+	} else if (skb->ip_summed != CHECKSUM_PARTIAL)
+		skb->ip_summed = CHECKSUM_NONE;
+
+	return skb;
+error:
+	kfree_skb(skb);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(gre_handle_offloads);
+
+static __sum16 check_checksum(struct sk_buff *skb)
+{
+	__sum16 csum = 0;
+
+	switch (skb->ip_summed) {
+	case CHECKSUM_COMPLETE:
+		csum = csum_fold(skb->csum);
+
+		if (!csum)
+			break;
+		/* Fall through. */
+
+	case CHECKSUM_NONE:
+		skb->csum = 0;
+		csum = __skb_checksum_complete(skb);
+		skb->ip_summed = CHECKSUM_COMPLETE;
+		break;
+	}
+
+	return csum;
+}
+
+static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi,
+			    bool *csum_err)
+{
+	unsigned int ip_hlen = ip_hdrlen(skb);
+	const struct gre_base_hdr *greh;
+	__be32 *options;
+	int hdr_len;
+
+	if (unlikely(!pskb_may_pull(skb, sizeof(struct gre_base_hdr))))
+		return -EINVAL;
+
+	greh = (struct gre_base_hdr *)(skb_network_header(skb) + ip_hlen);
+	if (unlikely(greh->flags & (GRE_VERSION | GRE_ROUTING)))
+		return -EINVAL;
+
+	tpi->flags = gre_flags_to_tnl_flags(greh->flags);
+	hdr_len = ip_gre_calc_hlen(tpi->flags);
+
+	if (!pskb_may_pull(skb, hdr_len))
+		return -EINVAL;
+
+	greh = (struct gre_base_hdr *)(skb_network_header(skb) + ip_hlen);
+	tpi->proto = greh->protocol;
+
+	options = (__be32 *)(greh + 1);
+	if (greh->flags & GRE_CSUM) {
+		if (check_checksum(skb)) {
+			*csum_err = true;
+			return -EINVAL;
+		}
+		options++;
+	}
+
+	if (greh->flags & GRE_KEY) {
+		tpi->key = *options;
+		options++;
+	} else
+		tpi->key = 0;
+
+	if (unlikely(greh->flags & GRE_SEQ)) {
+		tpi->seq = *options;
+		options++;
+	} else
+		tpi->seq = 0;
+
+	/* WCCP version 1 and 2 protocol decoding.
+	 * - Change protocol to IP
+	 * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
+	 */
+	if (greh->flags == 0 && tpi->proto == htons(ETH_P_WCCP)) {
+		tpi->proto = htons(ETH_P_IP);
+		if ((*(u8 *)options & 0xF0) != 0x40) {
+			hdr_len += 4;
+			if (!pskb_may_pull(skb, hdr_len))
+				return -EINVAL;
+		}
+	}
+
+	return iptunnel_pull_header(skb, hdr_len, tpi->proto);
+}
+
+static int gre_cisco_rcv(struct sk_buff *skb)
+{
+	struct tnl_ptk_info tpi;
+	int i;
+	bool csum_err = false;
+
+	if (parse_gre_header(skb, &tpi, &csum_err) < 0)
+		goto drop;
+
+	rcu_read_lock();
+	for (i = 0; i < GRE_IP_PROTO_MAX; i++) {
+		struct gre_cisco_protocol *proto;
+		int ret;
+
+		proto = rcu_dereference(gre_cisco_proto_list[i]);
+		if (!proto)
+			continue;
+		ret = proto->handler(skb, &tpi);
+		if (ret == PACKET_RCVD) {
+			rcu_read_unlock();
+			return 0;
+		}
+	}
+	rcu_read_unlock();
+
+	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
+drop:
+	kfree_skb(skb);
+	return 0;
+}
+
+static void gre_cisco_err(struct sk_buff *skb, u32 info)
+{
+	/* All the routers (except for Linux) return only
+	 * 8 bytes of packet payload. It means, that precise relaying of
+	 * ICMP in the real Internet is absolutely infeasible.
+	 *
+	 * Moreover, Cisco "wise men" put GRE key to the third word
+	 * in GRE header. It makes impossible maintaining even soft
+	 * state for keyed
+	 * GRE tunnels with enabled checksum. Tell them "thank you".
+	 *
+	 * Well, I wonder, rfc1812 was written by Cisco employee,
+	 * what the hell these idiots break standards established
+	 * by themselves???
+	 */
+
+	const int type = icmp_hdr(skb)->type;
+	const int code = icmp_hdr(skb)->code;
+	struct tnl_ptk_info tpi;
+	bool csum_err = false;
+	int i;
+
+	if (parse_gre_header(skb, &tpi, &csum_err)) {
+		if (!csum_err)		/* ignore csum errors. */
+			return;
+	}
+
+	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
+		ipv4_update_pmtu(skb, dev_net(skb->dev), info,
+				skb->dev->ifindex, 0, IPPROTO_GRE, 0);
+		return;
+	}
+	if (type == ICMP_REDIRECT) {
+		ipv4_redirect(skb, dev_net(skb->dev), skb->dev->ifindex, 0,
+				IPPROTO_GRE, 0);
+		return;
+	}
+
+	rcu_read_lock();
+	for (i = 0; i < GRE_IP_PROTO_MAX; i++) {
+		struct gre_cisco_protocol *proto;
+
+		proto = rcu_dereference(gre_cisco_proto_list[i]);
+		if (!proto)
+			continue;
+
+		if (proto->err_handler(skb, info, &tpi) == PACKET_RCVD)
+			goto out;
+
+	}
+out:
+	rcu_read_unlock();
+}
+
 static int gre_rcv(struct sk_buff *skb)
 {
 	const struct gre_protocol *proto;
@@ -220,27 +441,68 @@ static const struct net_offload gre_offload = {
 	},
 };
 
+static const struct gre_protocol ipgre_protocol = {
+	.handler     = gre_cisco_rcv,
+	.err_handler = gre_cisco_err,
+};
+
+int gre_cisco_register(struct gre_cisco_protocol *newp)
+{
+	struct gre_cisco_protocol **proto = (struct gre_cisco_protocol **)
+					    &gre_cisco_proto_list[newp->priority];
+
+	return (cmpxchg(proto, NULL, newp) == NULL) ? 0 : -EBUSY;
+}
+EXPORT_SYMBOL_GPL(gre_cisco_register);
+
+int gre_cisco_unregister(struct gre_cisco_protocol *del_proto)
+{
+	struct gre_cisco_protocol **proto = (struct gre_cisco_protocol **)
+					    &gre_cisco_proto_list[del_proto->priority];
+	int ret;
+
+	ret = (cmpxchg(proto, del_proto, NULL) == del_proto) ? 0 : -EINVAL;
+
+	if (ret)
+		return ret;
+
+	synchronize_net();
+	return 0;
+}
+EXPORT_SYMBOL_GPL(gre_cisco_unregister);
+
 static int __init gre_init(void)
 {
 	pr_info("GRE over IPv4 demultiplexor driver\n");
 
 	if (inet_add_protocol(&net_gre_protocol, IPPROTO_GRE) < 0) {
 		pr_err("can't add protocol\n");
-		return -EAGAIN;
+		goto err;
+	}
+
+	if (gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0) {
+		pr_info("%s: can't add ipgre handler\n", __func__);
+		goto err_gre;
 	}
 
 	if (inet_add_offload(&gre_offload, IPPROTO_GRE)) {
 		pr_err("can't add protocol offload\n");
-		inet_del_protocol(&net_gre_protocol, IPPROTO_GRE);
-		return -EAGAIN;
+		goto err_gso;
 	}
 
 	return 0;
+err_gso:
+	gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
+err_gre:
+	inet_del_protocol(&net_gre_protocol, IPPROTO_GRE);
+err:
+	return -EAGAIN;
 }
 
 static void __exit gre_exit(void)
 {
 	inet_del_offload(&gre_offload, IPPROTO_GRE);
+	gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
 	inet_del_protocol(&net_gre_protocol, IPPROTO_GRE);
 }
 
@@ -250,4 +512,3 @@ module_exit(gre_exit);
 MODULE_DESCRIPTION("GRE over IPv4 demultiplexer driver");
 MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)");
 MODULE_LICENSE("GPL");
-
diff --git a/net/ipv4/ip_gre.c b/net/ipv4/ip_gre.c
index a982657d05e..c326e869993 100644
--- a/net/ipv4/ip_gre.c
+++ b/net/ipv4/ip_gre.c
@@ -121,103 +121,8 @@ static int ipgre_tunnel_init(struct net_device *dev);
 static int ipgre_net_id __read_mostly;
 static int gre_tap_net_id __read_mostly;
 
-static __sum16 check_checksum(struct sk_buff *skb)
-{
-	__sum16 csum = 0;
-
-	switch (skb->ip_summed) {
-	case CHECKSUM_COMPLETE:
-		csum = csum_fold(skb->csum);
-
-		if (!csum)
-			break;
-		/* Fall through. */
-
-	case CHECKSUM_NONE:
-		skb->csum = 0;
-		csum = __skb_checksum_complete(skb);
-		skb->ip_summed = CHECKSUM_COMPLETE;
-		break;
-	}
-
-	return csum;
-}
-
-static int ip_gre_calc_hlen(__be16 o_flags)
-{
-	int addend = 4;
-
-	if (o_flags&TUNNEL_CSUM)
-		addend += 4;
-	if (o_flags&TUNNEL_KEY)
-		addend += 4;
-	if (o_flags&TUNNEL_SEQ)
-		addend += 4;
-	return addend;
-}
-
-static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi,
-			    bool *csum_err, int *hdr_len)
-{
-	unsigned int ip_hlen = ip_hdrlen(skb);
-	const struct gre_base_hdr *greh;
-	__be32 *options;
-
-	if (unlikely(!pskb_may_pull(skb, sizeof(struct gre_base_hdr))))
-		return -EINVAL;
-
-	greh = (struct gre_base_hdr *)(skb_network_header(skb) + ip_hlen);
-	if (unlikely(greh->flags & (GRE_VERSION | GRE_ROUTING)))
-		return -EINVAL;
-
-	tpi->flags = gre_flags_to_tnl_flags(greh->flags);
-	*hdr_len = ip_gre_calc_hlen(tpi->flags);
-
-	if (!pskb_may_pull(skb, *hdr_len))
-		return -EINVAL;
-
-	greh = (struct gre_base_hdr *)(skb_network_header(skb) + ip_hlen);
-
-	tpi->proto = greh->protocol;
-
-	options = (__be32 *)(greh + 1);
-	if (greh->flags & GRE_CSUM) {
-		if (check_checksum(skb)) {
-			*csum_err = true;
-			return -EINVAL;
-		}
-		options++;
-	}
-
-	if (greh->flags & GRE_KEY) {
-		tpi->key = *options;
-		options++;
-	} else
-		tpi->key = 0;
-
-	if (unlikely(greh->flags & GRE_SEQ)) {
-		tpi->seq = *options;
-		options++;
-	} else
-		tpi->seq = 0;
-
-	/* WCCP version 1 and 2 protocol decoding.
-	 * - Change protocol to IP
-	 * - When dealing with WCCPv2, Skip extra 4 bytes in GRE header
-	 */
-	if (greh->flags == 0 && tpi->proto == htons(ETH_P_WCCP)) {
-		tpi->proto = htons(ETH_P_IP);
-		if ((*(u8 *)options & 0xF0) != 0x40) {
-			*hdr_len += 4;
-			if (!pskb_may_pull(skb, *hdr_len))
-				return -EINVAL;
-		}
-	}
-
-	return 0;
-}
-
-static void ipgre_err(struct sk_buff *skb, u32 info)
+static int ipgre_err(struct sk_buff *skb, u32 info,
+		     const struct tnl_ptk_info *tpi)
 {
 
 	/* All the routers (except for Linux) return only
@@ -239,26 +144,18 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
 	const int type = icmp_hdr(skb)->type;
 	const int code = icmp_hdr(skb)->code;
 	struct ip_tunnel *t;
-	struct tnl_ptk_info tpi;
-	int hdr_len;
-	bool csum_err = false;
-
-	if (parse_gre_header(skb, &tpi, &csum_err, &hdr_len)) {
-		if (!csum_err)          /* ignore csum errors. */
-			return;
-	}
 
 	switch (type) {
 	default:
 	case ICMP_PARAMETERPROB:
-		return;
+		return PACKET_RCVD;
 
 	case ICMP_DEST_UNREACH:
 		switch (code) {
 		case ICMP_SR_FAILED:
 		case ICMP_PORT_UNREACH:
 			/* Impossible event. */
-			return;
+			return PACKET_RCVD;
 		default:
 			/* All others are translated to HOST_UNREACH.
 			   rfc2003 contains "deep thoughts" about NET_UNREACH,
@@ -269,138 +166,61 @@ static void ipgre_err(struct sk_buff *skb, u32 info)
 		break;
 	case ICMP_TIME_EXCEEDED:
 		if (code != ICMP_EXC_TTL)
-			return;
+			return PACKET_RCVD;
 		break;
 
 	case ICMP_REDIRECT:
 		break;
 	}
 
-	if (tpi.proto == htons(ETH_P_TEB))
+	if (tpi->proto == htons(ETH_P_TEB))
 		itn = net_generic(net, gre_tap_net_id);
 	else
 		itn = net_generic(net, ipgre_net_id);
 
 	iph = (const struct iphdr *)skb->data;
-	t = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi.flags,
-			     iph->daddr, iph->saddr, tpi.key);
+	t = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi->flags,
+			     iph->daddr, iph->saddr, tpi->key);
 
 	if (t == NULL)
-		return;
+		return PACKET_REJECT;
 
-	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
-		ipv4_update_pmtu(skb, dev_net(skb->dev), info,
-				 t->parms.link, 0, IPPROTO_GRE, 0);
-		return;
-	}
-	if (type == ICMP_REDIRECT) {
-		ipv4_redirect(skb, dev_net(skb->dev), t->parms.link, 0,
-			      IPPROTO_GRE, 0);
-		return;
-	}
 	if (t->parms.iph.daddr == 0 ||
 	    ipv4_is_multicast(t->parms.iph.daddr))
-		return;
+		return PACKET_RCVD;
 
 	if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
-		return;
+		return PACKET_RCVD;
 
 	if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
 		t->err_count++;
 	else
 		t->err_count = 1;
 	t->err_time = jiffies;
+	return PACKET_RCVD;
 }
 
-static int ipgre_rcv(struct sk_buff *skb)
+static int ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi)
 {
 	struct net *net = dev_net(skb->dev);
 	struct ip_tunnel_net *itn;
 	const struct iphdr *iph;
 	struct ip_tunnel *tunnel;
-	struct tnl_ptk_info tpi;
-	int hdr_len;
-	bool csum_err = false;
-
-	if (parse_gre_header(skb, &tpi, &csum_err, &hdr_len) < 0)
-		goto drop;
 
-	if (tpi.proto == htons(ETH_P_TEB))
+	if (tpi->proto == htons(ETH_P_TEB))
 		itn = net_generic(net, gre_tap_net_id);
 	else
 		itn = net_generic(net, ipgre_net_id);
 
 	iph = ip_hdr(skb);
-	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi.flags,
-				  iph->saddr, iph->daddr, tpi.key);
+	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi->flags,
+				  iph->saddr, iph->daddr, tpi->key);
 
 	if (tunnel) {
-		ip_tunnel_rcv(tunnel, skb, &tpi, log_ecn_error);
-		return 0;
+		ip_tunnel_rcv(tunnel, skb, tpi, log_ecn_error);
+		return PACKET_RCVD;
 	}
-	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
-drop:
-	kfree_skb(skb);
-	return 0;
-}
-
-static struct sk_buff *handle_offloads(struct ip_tunnel *tunnel, struct sk_buff *skb)
-{
-	int err;
-
-	if (skb_is_gso(skb)) {
-		err = skb_unclone(skb, GFP_ATOMIC);
-		if (unlikely(err))
-			goto error;
-		skb_shinfo(skb)->gso_type |= SKB_GSO_GRE;
-		return skb;
-	} else if (skb->ip_summed == CHECKSUM_PARTIAL &&
-		   tunnel->parms.o_flags&TUNNEL_CSUM) {
-		err = skb_checksum_help(skb);
-		if (unlikely(err))
-			goto error;
-	} else if (skb->ip_summed != CHECKSUM_PARTIAL)
-		skb->ip_summed = CHECKSUM_NONE;
-
-	return skb;
-
-error:
-	kfree_skb(skb);
-	return ERR_PTR(err);
-}
-
-static struct sk_buff *gre_build_header(struct sk_buff *skb,
-					const struct tnl_ptk_info *tpi,
-					int hdr_len)
-{
-	struct gre_base_hdr *greh;
-
-	skb_push(skb, hdr_len);
-
-	greh = (struct gre_base_hdr *)skb->data;
-	greh->flags = tnl_flags_to_gre_flags(tpi->flags);
-	greh->protocol = tpi->proto;
-
-	if (tpi->flags&(TUNNEL_KEY|TUNNEL_CSUM|TUNNEL_SEQ)) {
-		__be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4);
-
-		if (tpi->flags&TUNNEL_SEQ) {
-			*ptr = tpi->seq;
-			ptr--;
-		}
-		if (tpi->flags&TUNNEL_KEY) {
-			*ptr = tpi->key;
-			ptr--;
-		}
-		if (tpi->flags&TUNNEL_CSUM &&
-		    !(skb_shinfo(skb)->gso_type & SKB_GSO_GRE)) {
-			*(__sum16 *)ptr = 0;
-			*(__sum16 *)ptr = csum_fold(skb_checksum(skb, 0,
-								 skb->len, 0));
-		}
-	}
-
-	return skb;
+	return PACKET_REJECT;
 }
 
 static void __gre_xmit(struct sk_buff *skb, struct net_device *dev,
@@ -410,11 +230,6 @@ static void __gre_xmit(struct sk_buff *skb, struct net_device *dev,
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	struct tnl_ptk_info tpi;
 
-	if (likely(!skb->encapsulation)) {
-		skb_reset_inner_headers(skb);
-		skb->encapsulation = 1;
-	}
-
 	tpi.flags = tunnel->parms.o_flags;
 	tpi.proto = proto;
 	tpi.key = tunnel->parms.o_key;
@@ -423,11 +238,7 @@ static void __gre_xmit(struct sk_buff *skb, struct net_device *dev,
 	tpi.seq = htonl(tunnel->o_seqno);
 
 	/* Push GRE header. */
-	skb = gre_build_header(skb, &tpi, tunnel->hlen);
-	if (unlikely(!skb)) {
-		dev->stats.tx_dropped++;
-		return;
-	}
+	gre_build_header(skb, &tpi, tunnel->hlen);
 
 	ip_tunnel_xmit(skb, dev, tnl_params, tnl_params->protocol);
 }
@@ -438,7 +249,7 @@ static netdev_tx_t ipgre_xmit(struct sk_buff *skb,
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	const struct iphdr *tnl_params;
 
-	skb = handle_offloads(tunnel, skb);
+	skb = gre_handle_offloads(skb, !!(tunnel->parms.o_flags&TUNNEL_CSUM));
 	if (IS_ERR(skb))
 		goto out;
 
@@ -477,7 +288,7 @@ static netdev_tx_t gre_tap_xmit(struct sk_buff *skb,
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 
-	skb = handle_offloads(tunnel, skb);
+	skb = gre_handle_offloads(skb, !!(tunnel->parms.o_flags&TUNNEL_CSUM));
 	if (IS_ERR(skb))
 		goto out;
 
@@ -708,9 +519,10 @@ static int ipgre_tunnel_init(struct net_device *dev)
 	return ip_tunnel_init(dev);
 }
 
-static const struct gre_protocol ipgre_protocol = {
-	.handler     = ipgre_rcv,
-	.err_handler = ipgre_err,
+static struct gre_cisco_protocol ipgre_protocol = {
+	.handler        = ipgre_rcv,
+	.err_handler    = ipgre_err,
+	.priority       = 0,
 };
 
 static int __net_init ipgre_init_net(struct net *net)
@@ -978,7 +790,7 @@ static int __init ipgre_init(void)
 	if (err < 0)
 		goto pnet_tap_faied;
 
-	err = gre_add_protocol(&ipgre_protocol, GREPROTO_CISCO);
+	err = gre_cisco_register(&ipgre_protocol);
 	if (err < 0) {
 		pr_info("%s: can't add protocol\n", __func__);
 		goto add_proto_failed;
@@ -997,7 +809,7 @@ static int __init ipgre_init(void)
 tap_ops_failed:
 	rtnl_link_unregister(&ipgre_link_ops);
 rtnl_link_failed:
-	gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
+	gre_cisco_unregister(&ipgre_protocol);
 add_proto_failed:
 	unregister_pernet_device(&ipgre_tap_net_ops);
 pnet_tap_faied:
@@ -1009,8 +821,7 @@ static void __exit ipgre_fini(void)
 {
 	rtnl_link_unregister(&ipgre_tap_ops);
 	rtnl_link_unregister(&ipgre_link_ops);
-	if (gre_del_protocol(&ipgre_protocol, GREPROTO_CISCO) < 0)
-		pr_info("%s: can't remove protocol\n", __func__);
+	gre_cisco_unregister(&ipgre_protocol);
 	unregister_pernet_device(&ipgre_tap_net_ops);
 	unregister_pernet_device(&ipgre_net_ops);
 }
diff --git a/net/ipv4/ip_tunnel.c b/net/ipv4/ip_tunnel.c
index e189db409b0..bd227e5ea9d 100644
--- a/net/ipv4/ip_tunnel.c
+++ b/net/ipv4/ip_tunnel.c
@@ -408,13 +408,6 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 	const struct iphdr *iph = ip_hdr(skb);
 	int err;
 
-	secpath_reset(skb);
-
-	skb->protocol = tpi->proto;
-
-	skb->mac_header = skb->network_header;
-	__pskb_pull(skb, tunnel->hlen);
-	skb_postpull_rcsum(skb, skb_transport_header(skb), tunnel->hlen);
 #ifdef CONFIG_NET_IPGRE_BROADCAST
 	if (ipv4_is_multicast(iph->daddr)) {
 		/* Looped back packet, drop it! */
@@ -442,23 +435,6 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 		tunnel->i_seqno = ntohl(tpi->seq) + 1;
 	}
 
-	/* Warning: All skb pointers will be invalidated! */
-	if (tunnel->dev->type == ARPHRD_ETHER) {
-		if (!pskb_may_pull(skb, ETH_HLEN)) {
-			tunnel->dev->stats.rx_length_errors++;
-			tunnel->dev->stats.rx_errors++;
-			goto drop;
-		}
-
-		iph = ip_hdr(skb);
-		skb->protocol = eth_type_trans(skb, tunnel->dev);
-		skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
-	}
-
-	skb->pkt_type = PACKET_HOST;
-	__skb_tunnel_rx(skb, tunnel->dev);
-
-	skb_reset_network_header(skb);
 	err = IP_ECN_decapsulate(iph, skb);
 	if (unlikely(err)) {
 		if (log_ecn_error)
@@ -477,6 +453,12 @@ int ip_tunnel_rcv(struct ip_tunnel *tunnel, struct sk_buff *skb,
 	tstats->rx_bytes += skb->len;
 	u64_stats_update_end(&tstats->syncp);
 
+	if (tunnel->dev->type == ARPHRD_ETHER) {
+		skb->protocol = eth_type_trans(skb, tunnel->dev);
+		skb_postpull_rcsum(skb, eth_hdr(skb), ETH_HLEN);
+	} else {
+		skb->dev = tunnel->dev;
+	}
 	gro_cells_receive(&tunnel->gro_cells, skb);
 	return 0;
 
@@ -491,19 +473,17 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	const struct iphdr *inner_iph;
-	struct iphdr *iph;
 	struct flowi4 fl4;
 	u8     tos, ttl;
 	__be16 df;
 	struct rtable *rt;		/* Route to the other host */
-	struct net_device *tdev;	/* Device to other host */
 	unsigned int max_headroom;	/* The extra header space needed */
 	__be32 dst;
 	int mtu;
+	int err;
 
 	inner_iph = (const struct iphdr *)skb_inner_network_header(skb);
 
-	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
 	dst = tnl_params->daddr;
 	if (dst == 0) {
 		/* NBMA tunnel */
@@ -571,14 +551,11 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 		dev->stats.tx_carrier_errors++;
 		goto tx_error;
 	}
-	tdev = rt->dst.dev;
-
-	if (tdev == dev) {
+	if (rt->dst.dev == dev) {
 		ip_rt_put(rt);
 		dev->stats.collisions++;
 		goto tx_error;
 	}
-
 	df = tnl_params->frag_off;
 
 	if (df)
@@ -596,6 +573,7 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 		if (!skb_is_gso(skb) &&
 		    (inner_iph->frag_off&htons(IP_DF)) &&
 		     mtu < ntohs(inner_iph->tot_len)) {
+			memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
 			icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED, htonl(mtu));
 			ip_rt_put(rt);
 			goto tx_error;
@@ -646,8 +624,8 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 			ttl = ip4_dst_hoplimit(&rt->dst);
 	}
 
-	max_headroom = LL_RESERVED_SPACE(tdev) + sizeof(struct iphdr)
-					       + rt->dst.header_len;
+	max_headroom = LL_RESERVED_SPACE(rt->dst.dev) + sizeof(struct iphdr)
+			+ rt->dst.header_len;
 	if (max_headroom > dev->needed_headroom) {
 		dev->needed_headroom = max_headroom;
 		if (skb_cow_head(skb, dev->needed_headroom)) {
@@ -657,27 +635,11 @@ void ip_tunnel_xmit(struct sk_buff *skb, struct net_device *dev,
 		}
 	}
 
-	skb_dst_drop(skb);
-	skb_dst_set(skb, &rt->dst);
-
-	/* Push down and install the IP header. */
-	skb_push(skb, sizeof(struct iphdr));
-	skb_reset_network_header(skb);
-
-	iph = ip_hdr(skb);
-	inner_iph = (const struct iphdr *)skb_inner_network_header(skb);
+	err = iptunnel_xmit(dev_net(dev), rt, skb,
+			    fl4.saddr, fl4.daddr, protocol,
+			    ip_tunnel_ecn_encap(tos, inner_iph, skb), ttl, df);
+	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
 
-	iph->version	=	4;
-	iph->ihl	=	sizeof(struct iphdr) >> 2;
-	iph->frag_off	=	df;
-	iph->protocol	=	protocol;
-	iph->tos	=	ip_tunnel_ecn_encap(tos, inner_iph, skb);
-	iph->daddr	=	fl4.daddr;
-	iph->saddr	=	fl4.saddr;
-	iph->ttl	=	ttl;
-	tunnel_ip_select_ident(skb, inner_iph, &rt->dst);
-
-	iptunnel_xmit(skb, dev);
 	return;
 
 #if IS_ENABLED(CONFIG_IPV6)
diff --git a/net/ipv4/ip_tunnel_core.c b/net/ipv4/ip_tunnel_core.c
new file mode 100644
index 00000000000..7167b08977d
--- /dev/null
+++ b/net/ipv4/ip_tunnel_core.c
@@ -0,0 +1,122 @@
+/*
+ * Copyright (c) 2013 Nicira, Inc.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of version 2 of the GNU General Public
+ * License as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
+ * General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
+ * 02110-1301, USA
+ */
+
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/skbuff.h>
+#include <linux/netdevice.h>
+#include <linux/in.h>
+#include <linux/if_arp.h>
+#include <linux/mroute.h>
+#include <linux/init.h>
+#include <linux/in6.h>
+#include <linux/inetdevice.h>
+#include <linux/netfilter_ipv4.h>
+#include <linux/etherdevice.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+
+#include <net/ip.h>
+#include <net/icmp.h>
+#include <net/protocol.h>
+#include <net/ip_tunnels.h>
+#include <net/arp.h>
+#include <net/checksum.h>
+#include <net/dsfield.h>
+#include <net/inet_ecn.h>
+#include <net/xfrm.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <net/rtnetlink.h>
+
+int iptunnel_xmit(struct net *net, struct rtable *rt,
+		  struct sk_buff *skb,
+		  __be32 src, __be32 dst, __u8 proto,
+		  __u8 tos, __u8 ttl, __be16 df)
+{
+	int pkt_len = skb->len;
+	struct iphdr *iph;
+	int err;
+
+	nf_reset(skb);
+	secpath_reset(skb);
+	skb->rxhash = 0;
+	skb_dst_drop(skb);
+	skb_dst_set(skb, &rt->dst);
+	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
+
+	/* Push down and install the IP header. */
+	__skb_push(skb, sizeof(struct iphdr));
+	skb_reset_network_header(skb);
+
+	iph = ip_hdr(skb);
+
+	iph->version	=	4;
+	iph->ihl	=	sizeof(struct iphdr) >> 2;
+	iph->frag_off	=	df;
+	iph->protocol	=	proto;
+	iph->tos	=	tos;
+	iph->daddr	=	dst;
+	iph->saddr	=	src;
+	iph->ttl	=	ttl;
+	tunnel_ip_select_ident(skb,
+			       (const struct iphdr *)skb_inner_network_header(skb),
+			       &rt->dst);
+
+	err = ip_local_out(skb);
+	if (unlikely(net_xmit_eval(err)))
+		pkt_len = 0;
+	return pkt_len;
+}
+EXPORT_SYMBOL_GPL(iptunnel_xmit);
+
+int iptunnel_pull_header(struct sk_buff *skb, int hdr_len, __be16 inner_proto)
+{
+	if (unlikely(!pskb_may_pull(skb, hdr_len)))
+		return -ENOMEM;
+
+	skb_pull_rcsum(skb, hdr_len);
+
+	if (inner_proto == htons(ETH_P_TEB)) {
+		struct ethhdr *eh = (struct ethhdr *)skb->data;
+
+		if (unlikely(!pskb_may_pull(skb, ETH_HLEN)))
+			return -ENOMEM;
+
+		if (likely(ntohs(eh->h_proto) >= ETH_P_802_3_MIN))
+			skb->protocol = eh->h_proto;
+		else
+			skb->protocol = htons(ETH_P_802_2);
+
+	} else {
+		skb->protocol = inner_proto;
+	}
+
+	nf_reset(skb);
+	secpath_reset(skb);
+	if (!skb->l4_rxhash)
+		skb->rxhash = 0;
+	skb_dst_drop(skb);
+	skb->vlan_tci = 0;
+	skb_set_queue_mapping(skb, 0);
+	skb->pkt_type = PACKET_HOST;
+	return 0;
+}
+EXPORT_SYMBOL_GPL(iptunnel_pull_header);
diff --git a/net/ipv4/ipip.c b/net/ipv4/ipip.c
index 9df7ecd393f..e6905fbda2a 100644
--- a/net/ipv4/ipip.c
+++ b/net/ipv4/ipip.c
@@ -188,8 +188,12 @@ static int ipip_rcv(struct sk_buff *skb)
 	struct net *net = dev_net(skb->dev);
 	struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
 	struct ip_tunnel *tunnel;
-	const struct iphdr *iph = ip_hdr(skb);
+	const struct iphdr *iph;
 
+	if (iptunnel_pull_header(skb, 0, tpi.proto))
+		goto drop;
+
+	iph = ip_hdr(skb);
 	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
 			iph->saddr, iph->daddr, 0);
 	if (tunnel) {