--- zzzz-none-000/linux-3.10.107/net/ipv4/ip_gre.c	2017-06-27 09:49:32.000000000 +0000
+++ scorpion-7490-727/linux-3.10.107/net/ipv4/ip_gre.c	2021-02-04 17:41:59.000000000 +0000
@@ -25,6 +25,7 @@
 #include <linux/udp.h>
 #include <linux/if_arp.h>
 #include <linux/mroute.h>
+#include <linux/if_vlan.h>
 #include <linux/init.h>
 #include <linux/in6.h>
 #include <linux/inetdevice.h>
@@ -47,6 +48,7 @@
 #include <net/netns/generic.h>
 #include <net/rtnetlink.h>
 #include <net/gre.h>
+#include <net/dst_metadata.h>
 
 #if IS_ENABLED(CONFIG_IPV6)
 #include <net/ipv6.h>
@@ -121,86 +123,111 @@
 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)
+	if (o_flags & TUNNEL_CSUM)
 		addend += 4;
-	if (o_flags&TUNNEL_KEY)
+	if (o_flags & TUNNEL_KEY)
 		addend += 4;
-	if (o_flags&TUNNEL_SEQ)
+	if (o_flags & TUNNEL_SEQ)
 		addend += 4;
 	return addend;
 }
 
+static __be16 gre_flags_to_tnl_flags(__be16 flags)
+{
+	__be16 tflags = 0;
+
+	if (flags & GRE_CSUM)
+		tflags |= TUNNEL_CSUM;
+	if (flags & GRE_ROUTING)
+		tflags |= TUNNEL_ROUTING;
+	if (flags & GRE_KEY)
+		tflags |= TUNNEL_KEY;
+	if (flags & GRE_SEQ)
+		tflags |= TUNNEL_SEQ;
+	if (flags & GRE_STRICT)
+		tflags |= TUNNEL_STRICT;
+	if (flags & GRE_REC)
+		tflags |= TUNNEL_REC;
+	if (flags & GRE_VERSION)
+		tflags |= TUNNEL_VERSION;
+
+	return tflags;
+}
+
+static __be16 tnl_flags_to_gre_flags(__be16 tflags)
+{
+	__be16 flags = 0;
+
+	if (tflags & TUNNEL_CSUM)
+		flags |= GRE_CSUM;
+	if (tflags & TUNNEL_ROUTING)
+		flags |= GRE_ROUTING;
+	if (tflags & TUNNEL_KEY)
+		flags |= GRE_KEY;
+	if (tflags & TUNNEL_SEQ)
+		flags |= GRE_SEQ;
+	if (tflags & TUNNEL_STRICT)
+		flags |= GRE_STRICT;
+	if (tflags & TUNNEL_REC)
+		flags |= GRE_REC;
+	if (tflags & TUNNEL_VERSION)
+		flags |= GRE_VERSION;
+
+	return flags;
+}
+
+/* Fills in tpi and returns header length to be pulled. */
 static int parse_gre_header(struct sk_buff *skb, struct tnl_ptk_info *tpi,
-			    bool *csum_err, int *hdr_len)
+			    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);
+	greh = (struct gre_base_hdr *)skb_transport_header(skb);
 	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);
+	hdr_len = ip_gre_calc_hlen(tpi->flags);
 
-	if (!pskb_may_pull(skb, *hdr_len))
+	if (!pskb_may_pull(skb, hdr_len))
 		return -EINVAL;
 
-	greh = (struct gre_base_hdr *)(skb_network_header(skb) + ip_hlen);
-
+	greh = (struct gre_base_hdr *)skb_transport_header(skb);
 	tpi->proto = greh->protocol;
 
 	options = (__be32 *)(greh + 1);
 	if (greh->flags & GRE_CSUM) {
-		if (check_checksum(skb)) {
+		if (skb_checksum_simple_validate(skb)) {
 			*csum_err = true;
 			return -EINVAL;
 		}
+
+		skb_checksum_try_convert(skb, IPPROTO_GRE, 0,
+					 null_compute_pseudo);
 		options++;
 	}
 
 	if (greh->flags & GRE_KEY) {
 		tpi->key = *options;
 		options++;
-	} else
+	} else {
 		tpi->key = 0;
-
+	}
 	if (unlikely(greh->flags & GRE_SEQ)) {
 		tpi->seq = *options;
 		options++;
-	} else
+	} 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
@@ -208,16 +235,16 @@
 	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))
+			hdr_len += 4;
+			if (!pskb_may_pull(skb, hdr_len))
 				return -EINVAL;
 		}
 	}
-
-	return 0;
+	return hdr_len;
 }
 
-static void ipgre_err(struct sk_buff *skb, u32 info)
+static void ipgre_err(struct sk_buff *skb, u32 info,
+		      const struct tnl_ptk_info *tpi)
 {
 
 	/* All the routers (except for Linux) return only
@@ -239,14 +266,6 @@
 	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:
@@ -267,6 +286,7 @@
 			break;
 		}
 		break;
+
 	case ICMP_TIME_EXCEEDED:
 		if (code != ICMP_EXC_TTL)
 			return;
@@ -276,28 +296,18 @@
 		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);
+	iph = (const struct iphdr *)(icmp_hdr(skb) + 1);
+	t = ip_tunnel_lookup(itn, skb->dev->ifindex, tpi->flags,
+			     iph->daddr, iph->saddr, tpi->key);
 
-	if (t == NULL)
+	if (!t)
 		return;
 
-	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;
@@ -312,96 +322,161 @@
 	t->err_time = jiffies;
 }
 
-static int ipgre_rcv(struct sk_buff *skb)
+static void gre_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;
+
+	if (parse_gre_header(skb, &tpi, &csum_err) < 0) {
+		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;
+	}
+
+	ipgre_err(skb, info, &tpi);
+}
+
+static __be64 key_to_tunnel_id(__be32 key)
+{
+#ifdef __BIG_ENDIAN
+	return (__force __be64)((__force u32)key);
+#else
+	return (__force __be64)((__force u64)key << 32);
+#endif
+}
+
+/* Returns the least-significant 32 bits of a __be64. */
+static __be32 tunnel_id_to_key(__be64 x)
+{
+#ifdef __BIG_ENDIAN
+	return (__force __be32)x;
+#else
+	return (__force __be32)((__force u64)x >> 32);
+#endif
+}
+
+static int ipgre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi)
 {
 	struct net *net = dev_net(skb->dev);
+	struct metadata_dst *tun_dst = NULL;
 	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) {
 		skb_pop_mac_header(skb);
-		ip_tunnel_rcv(tunnel, skb, &tpi, hdr_len, log_ecn_error);
-		return 0;
+		if (tunnel->collect_md) {
+			__be16 flags;
+			__be64 tun_id;
+
+			flags = tpi->flags & (TUNNEL_CSUM | TUNNEL_KEY);
+			tun_id = key_to_tunnel_id(tpi->key);
+			tun_dst = ip_tun_rx_dst(skb, flags, tun_id, 0);
+			if (!tun_dst)
+				return PACKET_REJECT;
+		}
+
+		ip_tunnel_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error);
+		return PACKET_RCVD;
 	}
-	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
-drop:
-	kfree_skb(skb);
-	return 0;
+	return PACKET_REJECT;
 }
 
-static struct sk_buff *handle_offloads(struct ip_tunnel *tunnel, struct sk_buff *skb)
+static int gre_rcv(struct sk_buff *skb)
 {
-	int err;
+	struct tnl_ptk_info tpi;
+	bool csum_err = false;
+	int hdr_len;
 
-	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;
+#ifdef CONFIG_NET_IPGRE_BROADCAST
+	if (ipv4_is_multicast(ip_hdr(skb)->daddr)) {
+		/* Looped back packet, drop it! */
+		if (rt_is_output_route(skb_rtable(skb)))
+			goto drop;
+	}
+#endif
+
+	hdr_len = parse_gre_header(skb, &tpi, &csum_err);
+	if (hdr_len < 0)
+		goto drop;
+	if (iptunnel_pull_header(skb, hdr_len, tpi.proto) < 0)
+		goto drop;
 
-	return skb;
+	if (ipgre_rcv(skb, &tpi) == PACKET_RCVD)
+		return 0;
 
-error:
+	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
+drop:
 	kfree_skb(skb);
-	return ERR_PTR(err);
+	return 0;
 }
 
-static struct sk_buff *gre_build_header(struct sk_buff *skb,
-					const struct tnl_ptk_info *tpi,
-					int hdr_len)
+static void build_header(struct sk_buff *skb, int hdr_len, __be16 flags,
+			 __be16 proto, __be32 key, __be32 seq)
 {
 	struct gre_base_hdr *greh;
 
 	skb_push(skb, hdr_len);
 
+	skb_reset_transport_header(skb);
 	greh = (struct gre_base_hdr *)skb->data;
-	greh->flags = tnl_flags_to_gre_flags(tpi->flags);
-	greh->protocol = tpi->proto;
+	greh->flags = tnl_flags_to_gre_flags(flags);
+	greh->protocol = proto;
 
-	if (tpi->flags&(TUNNEL_KEY|TUNNEL_CSUM|TUNNEL_SEQ)) {
+	if (flags & (TUNNEL_KEY | TUNNEL_CSUM | TUNNEL_SEQ)) {
 		__be32 *ptr = (__be32 *)(((u8 *)greh) + hdr_len - 4);
 
-		if (tpi->flags&TUNNEL_SEQ) {
-			*ptr = tpi->seq;
+		if (flags & TUNNEL_SEQ) {
+			*ptr = seq;
 			ptr--;
 		}
-		if (tpi->flags&TUNNEL_KEY) {
-			*ptr = tpi->key;
+		if (flags & TUNNEL_KEY) {
+			*ptr = key;
 			ptr--;
 		}
-		if (tpi->flags&TUNNEL_CSUM &&
-		    !(skb_shinfo(skb)->gso_type & SKB_GSO_GRE)) {
-			*(__sum16 *)ptr = 0;
+		if (flags & TUNNEL_CSUM &&
+		    !(skb_shinfo(skb)->gso_type &
+		      (SKB_GSO_GRE | SKB_GSO_GRE_CSUM))) {
+			*ptr = 0;
 			*(__sum16 *)ptr = csum_fold(skb_checksum(skb, 0,
 								 skb->len, 0));
 		}
 	}
-
-	return skb;
 }
 
 static void __gre_xmit(struct sk_buff *skb, struct net_device *dev,
@@ -409,28 +484,118 @@
 		       __be16 proto)
 {
 	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;
 	if (tunnel->parms.o_flags & TUNNEL_SEQ)
 		tunnel->o_seqno++;
-	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;
+	build_header(skb, tunnel->tun_hlen, tunnel->parms.o_flags,
+		     proto, tunnel->parms.o_key, htonl(tunnel->o_seqno));
+
+	skb_set_inner_protocol(skb, proto);
+	ip_tunnel_xmit(skb, dev, tnl_params, tnl_params->protocol);
+}
+
+static struct sk_buff *gre_handle_offloads(struct sk_buff *skb,
+					   bool csum)
+{
+	return iptunnel_handle_offloads(skb, csum,
+					csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
+}
+
+static struct rtable *gre_get_rt(struct sk_buff *skb,
+				 struct net_device *dev,
+				 struct flowi4 *fl,
+				 const struct ip_tunnel_key *key)
+{
+	struct net *net = dev_net(dev);
+
+	memset(fl, 0, sizeof(*fl));
+	fl->daddr = key->u.ipv4.dst;
+	fl->saddr = key->u.ipv4.src;
+	fl->flowi4_tos = RT_TOS(key->tos);
+	fl->flowi4_mark = skb->mark;
+	fl->flowi4_proto = IPPROTO_GRE;
+
+	return ip_route_output_key(net, fl);
+}
+
+static void gre_fb_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct ip_tunnel_info *tun_info;
+	const struct ip_tunnel_key *key;
+	struct flowi4 fl;
+	struct rtable *rt;
+	int min_headroom;
+	int tunnel_hlen;
+	__be16 df, flags;
+	int err;
+
+	tun_info = skb_tunnel_info(skb);
+	if (unlikely(!tun_info || !(tun_info->mode & IP_TUNNEL_INFO_TX) ||
+		     ip_tunnel_info_af(tun_info) != AF_INET))
+		goto err_free_skb;
+
+	key = &tun_info->key;
+	rt = gre_get_rt(skb, dev, &fl, key);
+	if (IS_ERR(rt))
+		goto err_free_skb;
+
+	tunnel_hlen = ip_gre_calc_hlen(key->tun_flags);
+
+	min_headroom = LL_RESERVED_SPACE(rt->dst.dev) + rt->dst.header_len
+			+ tunnel_hlen + sizeof(struct iphdr);
+	if (skb_headroom(skb) < min_headroom || skb_header_cloned(skb)) {
+		int head_delta = SKB_DATA_ALIGN(min_headroom -
+						skb_headroom(skb) +
+						16);
+		err = pskb_expand_head(skb, max_t(int, head_delta, 0),
+				       0, GFP_ATOMIC);
+		if (unlikely(err))
+			goto err_free_rt;
+	}
+
+	/* Push Tunnel header. */
+	skb = gre_handle_offloads(skb, !!(tun_info->key.tun_flags & TUNNEL_CSUM));
+	if (IS_ERR(skb)) {
+		skb = NULL;
+		goto err_free_rt;
 	}
 
-	ip_tunnel_xmit(skb, dev, tnl_params);
+	flags = tun_info->key.tun_flags & (TUNNEL_CSUM | TUNNEL_KEY);
+	build_header(skb, tunnel_hlen, flags, htons(ETH_P_TEB),
+		     tunnel_id_to_key(tun_info->key.tun_id), 0);
+
+	df = key->tun_flags & TUNNEL_DONT_FRAGMENT ?  htons(IP_DF) : 0;
+	err = iptunnel_xmit(skb->sk, rt, skb, fl.saddr,
+			    key->u.ipv4.dst, IPPROTO_GRE,
+			    key->tos, key->ttl, df, false);
+	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
+	return;
+
+err_free_rt:
+	ip_rt_put(rt);
+err_free_skb:
+	kfree_skb(skb);
+	dev->stats.tx_dropped++;
+}
+
+static int gre_fill_metadata_dst(struct net_device *dev, struct sk_buff *skb)
+{
+	struct ip_tunnel_info *info = skb_tunnel_info(skb);
+	struct rtable *rt;
+	struct flowi4 fl4;
+
+	if (ip_tunnel_info_af(info) != AF_INET)
+		return -EINVAL;
+
+	rt = gre_get_rt(skb, dev, &fl4, &info->key);
+	if (IS_ERR(rt))
+		return PTR_ERR(rt);
+
+	ip_rt_put(rt);
+	info->key.u.ipv4.src = fl4.saddr;
+	return 0;
 }
 
 static netdev_tx_t ipgre_xmit(struct sk_buff *skb,
@@ -439,9 +604,10 @@
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 	const struct iphdr *tnl_params;
 
-	skb = handle_offloads(tunnel, skb);
-	if (IS_ERR(skb))
-		goto out;
+	if (tunnel->collect_md) {
+		gre_fb_xmit(skb, dev);
+		return NETDEV_TX_OK;
+	}
 
 	if (dev->header_ops) {
 		/* Need space for new headers */
@@ -455,6 +621,7 @@
 		 * to gre header.
 		 */
 		skb_pull(skb, tunnel->hlen + sizeof(struct iphdr));
+		skb_reset_mac_header(skb);
 	} else {
 		if (skb_cow_head(skb, dev->needed_headroom))
 			goto free_skb;
@@ -462,12 +629,17 @@
 		tnl_params = &tunnel->parms.iph;
 	}
 
-	__gre_xmit(skb, dev, tnl_params, skb->protocol);
+	skb = gre_handle_offloads(skb, !!(tunnel->parms.o_flags&TUNNEL_CSUM));
+	if (IS_ERR(skb))
+		goto out;
 
+	skb->skb_iif = dev->ifindex;
+
+	__gre_xmit(skb, dev, tnl_params, skb->protocol);
 	return NETDEV_TX_OK;
 
 free_skb:
-	dev_kfree_skb(skb);
+	kfree_skb(skb);
 out:
 	dev->stats.tx_dropped++;
 	return NETDEV_TX_OK;
@@ -478,19 +650,25 @@
 {
 	struct ip_tunnel *tunnel = netdev_priv(dev);
 
-	skb = handle_offloads(tunnel, skb);
+	if (tunnel->collect_md) {
+		gre_fb_xmit(skb, dev);
+		return NETDEV_TX_OK;
+	}
+
+	skb = gre_handle_offloads(skb, !!(tunnel->parms.o_flags&TUNNEL_CSUM));
 	if (IS_ERR(skb))
 		goto out;
 
 	if (skb_cow_head(skb, dev->needed_headroom))
 		goto free_skb;
 
-	__gre_xmit(skb, dev, &tunnel->parms.iph, htons(ETH_P_TEB));
+	skb->skb_iif = dev->ifindex;
 
+	__gre_xmit(skb, dev, &tunnel->parms.iph, htons(ETH_P_TEB));
 	return NETDEV_TX_OK;
 
 free_skb:
-	dev_kfree_skb(skb);
+	kfree_skb(skb);
 out:
 	dev->stats.tx_dropped++;
 	return NETDEV_TX_OK;
@@ -499,7 +677,7 @@
 static int ipgre_tunnel_ioctl(struct net_device *dev,
 			      struct ifreq *ifr, int cmd)
 {
-	int err = 0;
+	int err;
 	struct ip_tunnel_parm p;
 
 	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
@@ -529,7 +707,6 @@
    It allows to construct virtual multiprotocol broadcast "LAN"
    over the Internet, provided multicast routing is tuned.
 
-
    I have no idea was this bicycle invented before me,
    so that I had to set ARPHRD_IPGRE to a random value.
    I have an impression, that Cisco could make something similar,
@@ -599,7 +776,7 @@
 		struct flowi4 fl4;
 		struct rtable *rt;
 
-		rt = ip_route_output_gre(dev_net(dev), &fl4,
+		rt = ip_route_output_gre(t->net, &fl4,
 					 t->parms.iph.daddr,
 					 t->parms.iph.saddr,
 					 t->parms.o_key,
@@ -609,7 +786,7 @@
 			return -EADDRNOTAVAIL;
 		dev = rt->dst.dev;
 		ip_rt_put(rt);
-		if (__in_dev_get_rtnl(dev) == NULL)
+		if (!__in_dev_get_rtnl(dev))
 			return -EADDRNOTAVAIL;
 		t->mlink = dev->ifindex;
 		ip_mc_inc_group(__in_dev_get_rtnl(dev), t->parms.iph.daddr);
@@ -623,7 +800,7 @@
 
 	if (ipv4_is_multicast(t->parms.iph.daddr) && t->mlink) {
 		struct in_device *in_dev;
-		in_dev = inetdev_by_index(dev_net(dev), t->mlink);
+		in_dev = inetdev_by_index(t->net, t->mlink);
 		if (in_dev)
 			ip_mc_dec_group(in_dev, t->parms.iph.daddr);
 	}
@@ -642,6 +819,7 @@
 	.ndo_do_ioctl		= ipgre_tunnel_ioctl,
 	.ndo_change_mtu		= ip_tunnel_change_mtu,
 	.ndo_get_stats64	= ip_tunnel_get_stats64,
+	.ndo_get_iflink		= ip_tunnel_get_iflink,
 };
 
 #define GRE_FEATURES (NETIF_F_SG |		\
@@ -659,15 +837,20 @@
 static void __gre_tunnel_init(struct net_device *dev)
 {
 	struct ip_tunnel *tunnel;
+	int t_hlen;
 
 	tunnel = netdev_priv(dev);
-	tunnel->hlen = ip_gre_calc_hlen(tunnel->parms.o_flags);
+	tunnel->tun_hlen = ip_gre_calc_hlen(tunnel->parms.o_flags);
 	tunnel->parms.iph.protocol = IPPROTO_GRE;
 
-	dev->needed_headroom	= LL_MAX_HEADER + sizeof(struct iphdr) + 4;
-	dev->mtu		= ETH_DATA_LEN - sizeof(struct iphdr) - 4;
+	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
+
+	t_hlen = tunnel->hlen + sizeof(struct iphdr);
+
+	dev->needed_headroom	= LL_MAX_HEADER + t_hlen + 4;
+	dev->mtu		= ETH_DATA_LEN - t_hlen - 4;
 
-	dev->features		|= NETIF_F_NETNS_LOCAL | GRE_FEATURES;
+	dev->features		|= GRE_FEATURES;
 	dev->hw_features	|= GRE_FEATURES;
 
 	if (!(tunnel->parms.o_flags & TUNNEL_SEQ)) {
@@ -692,7 +875,7 @@
 	memcpy(dev->broadcast, &iph->daddr, 4);
 
 	dev->flags		= IFF_NOARP;
-	dev->priv_flags		&= ~IFF_XMIT_DST_RELEASE;
+	netif_keep_dst(dev);
 	dev->addr_len		= 4;
 
 	if (iph->daddr) {
@@ -711,8 +894,8 @@
 }
 
 static const struct gre_protocol ipgre_protocol = {
-	.handler     = ipgre_rcv,
-	.err_handler = ipgre_err,
+	.handler     = gre_rcv,
+	.err_handler = gre_err,
 };
 
 static int __net_init ipgre_init_net(struct net *net)
@@ -723,7 +906,7 @@
 static void __net_exit ipgre_exit_net(struct net *net)
 {
 	struct ip_tunnel_net *itn = net_generic(net, ipgre_net_id);
-	ip_tunnel_delete_net(itn);
+	ip_tunnel_delete_net(itn, &ipgre_link_ops);
 }
 
 static struct pernet_operations ipgre_net_ops = {
@@ -775,8 +958,10 @@
 	return ipgre_tunnel_validate(tb, data);
 }
 
-static void ipgre_netlink_parms(struct nlattr *data[], struct nlattr *tb[],
-			       struct ip_tunnel_parm *parms)
+static void ipgre_netlink_parms(struct net_device *dev,
+				struct nlattr *data[],
+				struct nlattr *tb[],
+				struct ip_tunnel_parm *parms)
 {
 	memset(parms, 0, sizeof(*parms));
 
@@ -801,10 +986,10 @@
 		parms->o_key = nla_get_be32(data[IFLA_GRE_OKEY]);
 
 	if (data[IFLA_GRE_LOCAL])
-		parms->iph.saddr = nla_get_be32(data[IFLA_GRE_LOCAL]);
+		parms->iph.saddr = nla_get_in_addr(data[IFLA_GRE_LOCAL]);
 
 	if (data[IFLA_GRE_REMOTE])
-		parms->iph.daddr = nla_get_be32(data[IFLA_GRE_REMOTE]);
+		parms->iph.daddr = nla_get_in_addr(data[IFLA_GRE_REMOTE]);
 
 	if (data[IFLA_GRE_TTL])
 		parms->iph.ttl = nla_get_u8(data[IFLA_GRE_TTL]);
@@ -814,11 +999,52 @@
 
 	if (!data[IFLA_GRE_PMTUDISC] || nla_get_u8(data[IFLA_GRE_PMTUDISC]))
 		parms->iph.frag_off = htons(IP_DF);
+
+	if (data[IFLA_GRE_COLLECT_METADATA]) {
+		struct ip_tunnel *t = netdev_priv(dev);
+
+		t->collect_md = true;
+	}
+}
+
+/* This function returns true when ENCAP attributes are present in the nl msg */
+static bool ipgre_netlink_encap_parms(struct nlattr *data[],
+				      struct ip_tunnel_encap *ipencap)
+{
+	bool ret = false;
+
+	memset(ipencap, 0, sizeof(*ipencap));
+
+	if (!data)
+		return ret;
+
+	if (data[IFLA_GRE_ENCAP_TYPE]) {
+		ret = true;
+		ipencap->type = nla_get_u16(data[IFLA_GRE_ENCAP_TYPE]);
+	}
+
+	if (data[IFLA_GRE_ENCAP_FLAGS]) {
+		ret = true;
+		ipencap->flags = nla_get_u16(data[IFLA_GRE_ENCAP_FLAGS]);
+	}
+
+	if (data[IFLA_GRE_ENCAP_SPORT]) {
+		ret = true;
+		ipencap->sport = nla_get_be16(data[IFLA_GRE_ENCAP_SPORT]);
+	}
+
+	if (data[IFLA_GRE_ENCAP_DPORT]) {
+		ret = true;
+		ipencap->dport = nla_get_be16(data[IFLA_GRE_ENCAP_DPORT]);
+	}
+
+	return ret;
 }
 
 static int gre_tap_init(struct net_device *dev)
 {
 	__gre_tunnel_init(dev);
+	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
 
 	return ip_tunnel_init(dev);
 }
@@ -831,12 +1057,15 @@
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_change_mtu		= ip_tunnel_change_mtu,
 	.ndo_get_stats64	= ip_tunnel_get_stats64,
+	.ndo_get_iflink		= ip_tunnel_get_iflink,
+	.ndo_fill_metadata_dst	= gre_fill_metadata_dst,
 };
 
 static void ipgre_tap_setup(struct net_device *dev)
 {
 	ether_setup(dev);
 	dev->netdev_ops		= &gre_tap_netdev_ops;
+	dev->priv_flags		|= IFF_LIVE_ADDR_CHANGE | IFF_GRE_V4_TAP;
 	ip_tunnel_setup(dev, gre_tap_net_id);
 }
 
@@ -844,8 +1073,17 @@
 			 struct nlattr *tb[], struct nlattr *data[])
 {
 	struct ip_tunnel_parm p;
+	struct ip_tunnel_encap ipencap;
+
+	if (ipgre_netlink_encap_parms(data, &ipencap)) {
+		struct ip_tunnel *t = netdev_priv(dev);
+		int err = ip_tunnel_encap_setup(t, &ipencap);
 
-	ipgre_netlink_parms(data, tb, &p);
+		if (err < 0)
+			return err;
+	}
+
+	ipgre_netlink_parms(dev, data, tb, &p);
 	return ip_tunnel_newlink(dev, tb, &p);
 }
 
@@ -853,8 +1091,17 @@
 			    struct nlattr *data[])
 {
 	struct ip_tunnel_parm p;
+	struct ip_tunnel_encap ipencap;
+
+	if (ipgre_netlink_encap_parms(data, &ipencap)) {
+		struct ip_tunnel *t = netdev_priv(dev);
+		int err = ip_tunnel_encap_setup(t, &ipencap);
+
+		if (err < 0)
+			return err;
+	}
 
-	ipgre_netlink_parms(data, tb, &p);
+	ipgre_netlink_parms(dev, data, tb, &p);
 	return ip_tunnel_changelink(dev, tb, &p);
 }
 
@@ -881,6 +1128,16 @@
 		nla_total_size(1) +
 		/* IFLA_GRE_PMTUDISC */
 		nla_total_size(1) +
+		/* IFLA_GRE_ENCAP_TYPE */
+		nla_total_size(2) +
+		/* IFLA_GRE_ENCAP_FLAGS */
+		nla_total_size(2) +
+		/* IFLA_GRE_ENCAP_SPORT */
+		nla_total_size(2) +
+		/* IFLA_GRE_ENCAP_DPORT */
+		nla_total_size(2) +
+		/* IFLA_GRE_COLLECT_METADATA */
+		nla_total_size(0) +
 		0;
 }
 
@@ -894,13 +1151,29 @@
 	    nla_put_be16(skb, IFLA_GRE_OFLAGS, tnl_flags_to_gre_flags(p->o_flags)) ||
 	    nla_put_be32(skb, IFLA_GRE_IKEY, p->i_key) ||
 	    nla_put_be32(skb, IFLA_GRE_OKEY, p->o_key) ||
-	    nla_put_be32(skb, IFLA_GRE_LOCAL, p->iph.saddr) ||
-	    nla_put_be32(skb, IFLA_GRE_REMOTE, p->iph.daddr) ||
+	    nla_put_in_addr(skb, IFLA_GRE_LOCAL, p->iph.saddr) ||
+	    nla_put_in_addr(skb, IFLA_GRE_REMOTE, p->iph.daddr) ||
 	    nla_put_u8(skb, IFLA_GRE_TTL, p->iph.ttl) ||
 	    nla_put_u8(skb, IFLA_GRE_TOS, p->iph.tos) ||
 	    nla_put_u8(skb, IFLA_GRE_PMTUDISC,
 		       !!(p->iph.frag_off & htons(IP_DF))))
 		goto nla_put_failure;
+
+	if (nla_put_u16(skb, IFLA_GRE_ENCAP_TYPE,
+			t->encap.type) ||
+	    nla_put_be16(skb, IFLA_GRE_ENCAP_SPORT,
+			 t->encap.sport) ||
+	    nla_put_be16(skb, IFLA_GRE_ENCAP_DPORT,
+			 t->encap.dport) ||
+	    nla_put_u16(skb, IFLA_GRE_ENCAP_FLAGS,
+			t->encap.flags))
+		goto nla_put_failure;
+
+	if (t->collect_md) {
+		if (nla_put_flag(skb, IFLA_GRE_COLLECT_METADATA))
+			goto nla_put_failure;
+	}
+
 	return 0;
 
 nla_put_failure:
@@ -918,6 +1191,11 @@
 	[IFLA_GRE_TTL]		= { .type = NLA_U8 },
 	[IFLA_GRE_TOS]		= { .type = NLA_U8 },
 	[IFLA_GRE_PMTUDISC]	= { .type = NLA_U8 },
+	[IFLA_GRE_ENCAP_TYPE]	= { .type = NLA_U16 },
+	[IFLA_GRE_ENCAP_FLAGS]	= { .type = NLA_U16 },
+	[IFLA_GRE_ENCAP_SPORT]	= { .type = NLA_U16 },
+	[IFLA_GRE_ENCAP_DPORT]	= { .type = NLA_U16 },
+	[IFLA_GRE_COLLECT_METADATA]	= { .type = NLA_FLAG },
 };
 
 static struct rtnl_link_ops ipgre_link_ops __read_mostly = {
@@ -932,6 +1210,7 @@
 	.dellink	= ip_tunnel_dellink,
 	.get_size	= ipgre_get_size,
 	.fill_info	= ipgre_fill_info,
+	.get_link_net	= ip_tunnel_get_link_net,
 };
 
 static struct rtnl_link_ops ipgre_tap_ops __read_mostly = {
@@ -946,17 +1225,55 @@
 	.dellink	= ip_tunnel_dellink,
 	.get_size	= ipgre_get_size,
 	.fill_info	= ipgre_fill_info,
+	.get_link_net	= ip_tunnel_get_link_net,
 };
 
+struct net_device *gretap_fb_dev_create(struct net *net, const char *name,
+					u8 name_assign_type)
+{
+	struct nlattr *tb[IFLA_MAX + 1];
+	struct net_device *dev;
+	struct ip_tunnel *t;
+	int err;
+
+	memset(&tb, 0, sizeof(tb));
+
+	dev = rtnl_create_link(net, name, name_assign_type,
+			       &ipgre_tap_ops, tb);
+	if (IS_ERR(dev))
+		return dev;
+
+	/* Configure flow based GRE device. */
+	t = netdev_priv(dev);
+	t->collect_md = true;
+
+	err = ipgre_newlink(net, dev, tb, NULL);
+	if (err < 0)
+		goto out;
+
+	/* openvswitch users expect packet sizes to be unrestricted,
+	 * so set the largest MTU we can.
+	 */
+	err = __ip_tunnel_change_mtu(dev, IP_MAX_MTU, false);
+	if (err)
+		goto out;
+
+	return dev;
+out:
+	free_netdev(dev);
+	return ERR_PTR(err);
+}
+EXPORT_SYMBOL_GPL(gretap_fb_dev_create);
+
 static int __net_init ipgre_tap_init_net(struct net *net)
 {
-	return ip_tunnel_init_net(net, gre_tap_net_id, &ipgre_tap_ops, NULL);
+	return ip_tunnel_init_net(net, gre_tap_net_id, &ipgre_tap_ops, "gretap0");
 }
 
 static void __net_exit ipgre_tap_exit_net(struct net *net)
 {
 	struct ip_tunnel_net *itn = net_generic(net, gre_tap_net_id);
-	ip_tunnel_delete_net(itn);
+	ip_tunnel_delete_net(itn, &ipgre_tap_ops);
 }
 
 static struct pernet_operations ipgre_tap_net_ops = {
@@ -1011,8 +1328,7 @@
 {
 	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_del_protocol(&ipgre_protocol, GREPROTO_CISCO);
 	unregister_pernet_device(&ipgre_tap_net_ops);
 	unregister_pernet_device(&ipgre_net_ops);
 }