Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/klassert/ipsec-next

Steffen Klassert says:

====================
pull request (net-next): ipsec-next 2018-07-27

1) Extend the output_mark to also support the input direction
   and masking the mark values before applying to the skb.

2) Add a new lookup key for the upcomming xfrm interfaces.

3) Extend the xfrm lookups to match xfrm interface IDs.

4) Add virtual xfrm interfaces. The purpose of these interfaces
   is to overcome the design limitations that the existing
   VTI devices have.

  The main limitations that we see with the current VTI are the
  following:

  VTI interfaces are L3 tunnels with configurable endpoints.
  For xfrm, the tunnel endpoint are already determined by the SA.
  So the VTI tunnel endpoints must be either the same as on the
  SA or wildcards. In case VTI tunnel endpoints are same as on
  the SA, we get a one to one correlation between the SA and
  the tunnel. So each SA needs its own tunnel interface.

  On the other hand, we can have only one VTI tunnel with
  wildcard src/dst tunnel endpoints in the system because the
  lookup is based on the tunnel endpoints. The existing tunnel
  lookup won't work with multiple tunnels with wildcard
  tunnel endpoints. Some usecases require more than on
  VTI tunnel of this type, for example if somebody has multiple
  namespaces and every namespace requires such a VTI.

  VTI needs separate interfaces for IPv4 and IPv6 tunnels.
  So when routing to a VTI, we have to know to which address
  family this traffic class is going to be encapsulated.
  This is a lmitation because it makes routing more complex
  and it is not always possible to know what happens behind the
  VTI, e.g. when the VTI is move to some namespace.

  VTI works just with tunnel mode SAs. We need generic interfaces
  that ensures transfomation, regardless of the xfrm mode and
  the encapsulated address family.

  VTI is configured with a combination GRE keys and xfrm marks.
  With this we have to deal with some extra cases in the generic
  tunnel lookup because the GRE keys on the VTI are actually
  not GRE keys, the GRE keys were just reused for something else.
  All extensions to the VTI interfaces would require to add
  even more complexity to the generic tunnel lookup.

  So to overcome this, we developed xfrm interfaces with the
  following design goal:

  It should be possible to tunnel IPv4 and IPv6 through the same
  interface.

  No limitation on xfrm mode (tunnel, transport and beet).

  Should be a generic virtual interface that ensures IPsec
  transformation, no need to know what happens behind the
  interface.

  Interfaces should be configured with a new key that must match a
  new policy/SA lookup key.

  The lookup logic should stay in the xfrm codebase, no need to
  change or extend generic routing and tunnel lookups.

  Should be possible to use IPsec hardware offloads of the underlying
  interface.

5) Remove xfrm pcpu policy cache. This was added after the flowcache
   removal, but it turned out to make things even worse.
   From Florian Westphal.

6) Allow to update the set mark on SA updates.
   From Nathan Harold.

7) Convert some timestamps to time64_t.
   From Arnd Bergmann.

8) Don't check the offload_handle in xfrm code,
   it is an opaque data cookie for the driver.
   From Shannon Nelson.

9) Remove xfrmi interface ID from flowi. After this pach
   no generic code is touched anymore to do xfrm interface
   lookups. From Benedict Wong.

10) Allow to update the xfrm interface ID on SA updates.
    From Nathan Harold.

11) Don't pass zero to ERR_PTR() in xfrm_resolve_and_create_bundle.
    From YueHaibing.

12) Return more detailed errors on xfrm interface creation.
    From Benedict Wong.

13) Use PTR_ERR_OR_ZERO instead of IS_ERR + PTR_ERR.
    From the kbuild test robot.
====================

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

9 files changed, 1244 insertions, 223 deletions

diff --git a/net/xfrm/Kconfig b/net/xfrm/Kconfig
index eab952cca7d0..4a9ee2d83158 100644
--- a/net/xfrm/Kconfig
+++ b/net/xfrm/Kconfig
@@ -25,6 +25,14 @@ config XFRM_USER
 
 	  If unsure, say Y.
 
+config XFRM_INTERFACE
+	tristate "Transformation virtual interface"
+	depends on XFRM && IPV6
+	---help---
+	  This provides a virtual interface to route IPsec traffic.
+
+	  If unsure, say N.
+
 config XFRM_SUB_POLICY
 	bool "Transformation sub policy support"
 	depends on XFRM
diff --git a/net/xfrm/Makefile b/net/xfrm/Makefile
index 0bd2465a8c5a..fbc4552d17b8 100644
--- a/net/xfrm/Makefile
+++ b/net/xfrm/Makefile
@@ -10,3 +10,4 @@ obj-$(CONFIG_XFRM_STATISTICS) += xfrm_proc.o
 obj-$(CONFIG_XFRM_ALGO) += xfrm_algo.o
 obj-$(CONFIG_XFRM_USER) += xfrm_user.o
 obj-$(CONFIG_XFRM_IPCOMP) += xfrm_ipcomp.o
+obj-$(CONFIG_XFRM_INTERFACE) += xfrm_interface.o
diff --git a/net/xfrm/xfrm_device.c b/net/xfrm/xfrm_device.c
index 175941e15a6e..5611b7521020 100644
--- a/net/xfrm/xfrm_device.c
+++ b/net/xfrm/xfrm_device.c
@@ -56,7 +56,7 @@ struct sk_buff *validate_xmit_xfrm(struct sk_buff *skb, netdev_features_t featur
 	if (skb_is_gso(skb)) {
 		struct net_device *dev = skb->dev;
 
-		if (unlikely(!x->xso.offload_handle || (x->xso.dev != dev))) {
+		if (unlikely(x->xso.dev != dev)) {
 			struct sk_buff *segs;
 
 			/* Packet got rerouted, fixup features and segment it. */
@@ -162,7 +162,8 @@ int xfrm_dev_state_add(struct net *net, struct xfrm_state *x,
 		}
 
 		dst = __xfrm_dst_lookup(net, 0, 0, saddr, daddr,
-					x->props.family, x->props.output_mark);
+					x->props.family,
+					xfrm_smark_get(0, x));
 		if (IS_ERR(dst))
 			return 0;
 
@@ -210,8 +211,8 @@ bool xfrm_dev_offload_ok(struct sk_buff *skb, struct xfrm_state *x)
 	if (!x->type_offload || x->encap)
 		return false;
 
-	if ((!dev || (x->xso.offload_handle && (dev == xfrm_dst_path(dst)->dev))) &&
-	     (!xdst->child->xfrm && x->type->get_mtu)) {
+	if ((!dev || (dev == xfrm_dst_path(dst)->dev)) &&
+	    (!xdst->child->xfrm && x->type->get_mtu)) {
 		mtu = x->type->get_mtu(x, xdst->child_mtu_cached);
 
 		if (skb->len <= mtu)
@@ -306,12 +307,6 @@ static int xfrm_dev_register(struct net_device *dev)
 	return xfrm_api_check(dev);
 }
 
-static int xfrm_dev_unregister(struct net_device *dev)
-{
-	xfrm_policy_cache_flush();
-	return NOTIFY_DONE;
-}
-
 static int xfrm_dev_feat_change(struct net_device *dev)
 {
 	return xfrm_api_check(dev);
@@ -322,7 +317,6 @@ static int xfrm_dev_down(struct net_device *dev)
 	if (dev->features & NETIF_F_HW_ESP)
 		xfrm_dev_state_flush(dev_net(dev), dev, true);
 
-	xfrm_policy_cache_flush();
 	return NOTIFY_DONE;
 }
 
@@ -334,9 +328,6 @@ static int xfrm_dev_event(struct notifier_block *this, unsigned long event, void
 	case NETDEV_REGISTER:
 		return xfrm_dev_register(dev);
 
-	case NETDEV_UNREGISTER:
-		return xfrm_dev_unregister(dev);
-
 	case NETDEV_FEAT_CHANGE:
 		return xfrm_dev_feat_change(dev);
 
diff --git a/net/xfrm/xfrm_input.c b/net/xfrm/xfrm_input.c
index 352abca2605f..b89c9c7f8c5c 100644
--- a/net/xfrm/xfrm_input.c
+++ b/net/xfrm/xfrm_input.c
@@ -320,6 +320,7 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 
 	seq = 0;
 	if (!spi && (err = xfrm_parse_spi(skb, nexthdr, &spi, &seq)) != 0) {
+		secpath_reset(skb);
 		XFRM_INC_STATS(net, LINUX_MIB_XFRMINHDRERROR);
 		goto drop;
 	}
@@ -328,17 +329,21 @@ int xfrm_input(struct sk_buff *skb, int nexthdr, __be32 spi, int encap_type)
 				   XFRM_SPI_SKB_CB(skb)->daddroff);
 	do {
 		if (skb->sp->len == XFRM_MAX_DEPTH) {
+			secpath_reset(skb);
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINBUFFERERROR);
 			goto drop;
 		}
 
 		x = xfrm_state_lookup(net, mark, daddr, spi, nexthdr, family);
 		if (x == NULL) {
+			secpath_reset(skb);
 			XFRM_INC_STATS(net, LINUX_MIB_XFRMINNOSTATES);
 			xfrm_audit_state_notfound(skb, family, spi, seq);
 			goto drop;
 		}
 
+		skb->mark = xfrm_smark_get(skb->mark, x);
+
 		skb->sp->xvec[skb->sp->len++] = x;
 
 lock:
diff --git a/net/xfrm/xfrm_interface.c b/net/xfrm/xfrm_interface.c
new file mode 100644
index 000000000000..31acc6f33d98
--- /dev/null
+++ b/net/xfrm/xfrm_interface.c
@@ -0,0 +1,975 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ *	XFRM virtual interface
+ *
+ *	Copyright (C) 2018 secunet Security Networks AG
+ *
+ *	Author:
+ *	Steffen Klassert <steffen.klassert@secunet.com>
+ */
+
+#include <linux/module.h>
+#include <linux/capability.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/sockios.h>
+#include <linux/icmp.h>
+#include <linux/if.h>
+#include <linux/in.h>
+#include <linux/ip.h>
+#include <linux/net.h>
+#include <linux/in6.h>
+#include <linux/netdevice.h>
+#include <linux/if_link.h>
+#include <linux/if_arp.h>
+#include <linux/icmpv6.h>
+#include <linux/init.h>
+#include <linux/route.h>
+#include <linux/rtnetlink.h>
+#include <linux/netfilter_ipv6.h>
+#include <linux/slab.h>
+#include <linux/hash.h>
+
+#include <linux/uaccess.h>
+#include <linux/atomic.h>
+
+#include <net/icmp.h>
+#include <net/ip.h>
+#include <net/ipv6.h>
+#include <net/ip6_route.h>
+#include <net/addrconf.h>
+#include <net/xfrm.h>
+#include <net/net_namespace.h>
+#include <net/netns/generic.h>
+#include <linux/etherdevice.h>
+
+static int xfrmi_dev_init(struct net_device *dev);
+static void xfrmi_dev_setup(struct net_device *dev);
+static struct rtnl_link_ops xfrmi_link_ops __read_mostly;
+static unsigned int xfrmi_net_id __read_mostly;
+
+struct xfrmi_net {
+	/* lists for storing interfaces in use */
+	struct xfrm_if __rcu *xfrmi[1];
+};
+
+#define for_each_xfrmi_rcu(start, xi) \
+	for (xi = rcu_dereference(start); xi; xi = rcu_dereference(xi->next))
+
+static struct xfrm_if *xfrmi_lookup(struct net *net, struct xfrm_state *x)
+{
+	struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
+	struct xfrm_if *xi;
+
+	for_each_xfrmi_rcu(xfrmn->xfrmi[0], xi) {
+		if (x->if_id == xi->p.if_id &&
+		    (xi->dev->flags & IFF_UP))
+			return xi;
+	}
+
+	return NULL;
+}
+
+static struct xfrm_if *xfrmi_decode_session(struct sk_buff *skb)
+{
+	struct xfrmi_net *xfrmn;
+	int ifindex;
+	struct xfrm_if *xi;
+
+	if (!skb->dev)
+		return NULL;
+
+	xfrmn = net_generic(dev_net(skb->dev), xfrmi_net_id);
+	ifindex = skb->dev->ifindex;
+
+	for_each_xfrmi_rcu(xfrmn->xfrmi[0], xi) {
+		if (ifindex == xi->dev->ifindex &&
+			(xi->dev->flags & IFF_UP))
+				return xi;
+	}
+
+	return NULL;
+}
+
+static void xfrmi_link(struct xfrmi_net *xfrmn, struct xfrm_if *xi)
+{
+	struct xfrm_if __rcu **xip = &xfrmn->xfrmi[0];
+
+	rcu_assign_pointer(xi->next , rtnl_dereference(*xip));
+	rcu_assign_pointer(*xip, xi);
+}
+
+static void xfrmi_unlink(struct xfrmi_net *xfrmn, struct xfrm_if *xi)
+{
+	struct xfrm_if __rcu **xip;
+	struct xfrm_if *iter;
+
+	for (xip = &xfrmn->xfrmi[0];
+	     (iter = rtnl_dereference(*xip)) != NULL;
+	     xip = &iter->next) {
+		if (xi == iter) {
+			rcu_assign_pointer(*xip, xi->next);
+			break;
+		}
+	}
+}
+
+static void xfrmi_dev_free(struct net_device *dev)
+{
+	free_percpu(dev->tstats);
+}
+
+static int xfrmi_create2(struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct net *net = dev_net(dev);
+	struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
+	int err;
+
+	dev->rtnl_link_ops = &xfrmi_link_ops;
+	err = register_netdevice(dev);
+	if (err < 0)
+		goto out;
+
+	strcpy(xi->p.name, dev->name);
+
+	dev_hold(dev);
+	xfrmi_link(xfrmn, xi);
+
+	return 0;
+
+out:
+	return err;
+}
+
+static struct xfrm_if *xfrmi_create(struct net *net, struct xfrm_if_parms *p)
+{
+	struct net_device *dev;
+	struct xfrm_if *xi;
+	char name[IFNAMSIZ];
+	int err;
+
+	if (p->name[0]) {
+		strlcpy(name, p->name, IFNAMSIZ);
+	} else {
+		err = -EINVAL;
+		goto failed;
+	}
+
+	dev = alloc_netdev(sizeof(*xi), name, NET_NAME_UNKNOWN, xfrmi_dev_setup);
+	if (!dev) {
+		err = -EAGAIN;
+		goto failed;
+	}
+
+	dev_net_set(dev, net);
+
+	xi = netdev_priv(dev);
+	xi->p = *p;
+	xi->net = net;
+	xi->dev = dev;
+	xi->phydev = dev_get_by_index(net, p->link);
+	if (!xi->phydev) {
+		err = -ENODEV;
+		goto failed_free;
+	}
+
+	err = xfrmi_create2(dev);
+	if (err < 0)
+		goto failed_dev_put;
+
+	return xi;
+
+failed_dev_put:
+	dev_put(xi->phydev);
+failed_free:
+	free_netdev(dev);
+failed:
+	return ERR_PTR(err);
+}
+
+static struct xfrm_if *xfrmi_locate(struct net *net, struct xfrm_if_parms *p,
+				   int create)
+{
+	struct xfrm_if __rcu **xip;
+	struct xfrm_if *xi;
+	struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
+
+	for (xip = &xfrmn->xfrmi[0];
+	     (xi = rtnl_dereference(*xip)) != NULL;
+	     xip = &xi->next) {
+		if (xi->p.if_id == p->if_id) {
+			if (create)
+				return ERR_PTR(-EEXIST);
+
+			return xi;
+		}
+	}
+	if (!create)
+		return ERR_PTR(-ENODEV);
+	return xfrmi_create(net, p);
+}
+
+static void xfrmi_dev_uninit(struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct xfrmi_net *xfrmn = net_generic(xi->net, xfrmi_net_id);
+
+	xfrmi_unlink(xfrmn, xi);
+	dev_put(xi->phydev);
+	dev_put(dev);
+}
+
+static void xfrmi_scrub_packet(struct sk_buff *skb, bool xnet)
+{
+	skb->tstamp = 0;
+	skb->pkt_type = PACKET_HOST;
+	skb->skb_iif = 0;
+	skb->ignore_df = 0;
+	skb_dst_drop(skb);
+	nf_reset(skb);
+	nf_reset_trace(skb);
+
+	if (!xnet)
+		return;
+
+	ipvs_reset(skb);
+	secpath_reset(skb);
+	skb_orphan(skb);
+	skb->mark = 0;
+}
+
+static int xfrmi_rcv_cb(struct sk_buff *skb, int err)
+{
+	struct pcpu_sw_netstats *tstats;
+	struct xfrm_mode *inner_mode;
+	struct net_device *dev;
+	struct xfrm_state *x;
+	struct xfrm_if *xi;
+	bool xnet;
+
+	if (err && !skb->sp)
+		return 0;
+
+	x = xfrm_input_state(skb);
+
+	xi = xfrmi_lookup(xs_net(x), x);
+	if (!xi)
+		return 1;
+
+	dev = xi->dev;
+	skb->dev = dev;
+
+	if (err) {
+		dev->stats.rx_errors++;
+		dev->stats.rx_dropped++;
+
+		return 0;
+	}
+
+	xnet = !net_eq(xi->net, dev_net(skb->dev));
+
+	if (xnet) {
+		inner_mode = x->inner_mode;
+
+		if (x->sel.family == AF_UNSPEC) {
+			inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol);
+			if (inner_mode == NULL) {
+				XFRM_INC_STATS(dev_net(skb->dev),
+					       LINUX_MIB_XFRMINSTATEMODEERROR);
+				return -EINVAL;
+			}
+		}
+
+		if (!xfrm_policy_check(NULL, XFRM_POLICY_IN, skb,
+				       inner_mode->afinfo->family))
+			return -EPERM;
+	}
+
+	xfrmi_scrub_packet(skb, xnet);
+
+	tstats = this_cpu_ptr(dev->tstats);
+
+	u64_stats_update_begin(&tstats->syncp);
+	tstats->rx_packets++;
+	tstats->rx_bytes += skb->len;
+	u64_stats_update_end(&tstats->syncp);
+
+	return 0;
+}
+
+static int
+xfrmi_xmit2(struct sk_buff *skb, struct net_device *dev, struct flowi *fl)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct net_device_stats *stats = &xi->dev->stats;
+	struct dst_entry *dst = skb_dst(skb);
+	unsigned int length = skb->len;
+	struct net_device *tdev;
+	struct xfrm_state *x;
+	int err = -1;
+	int mtu;
+
+	if (!dst)
+		goto tx_err_link_failure;
+
+	dst_hold(dst);
+	dst = xfrm_lookup_with_ifid(xi->net, dst, fl, NULL, 0, xi->p.if_id);
+	if (IS_ERR(dst)) {
+		err = PTR_ERR(dst);
+		dst = NULL;
+		goto tx_err_link_failure;
+	}
+
+	x = dst->xfrm;
+	if (!x)
+		goto tx_err_link_failure;
+
+	if (x->if_id != xi->p.if_id)
+		goto tx_err_link_failure;
+
+	tdev = dst->dev;
+
+	if (tdev == dev) {
+		stats->collisions++;
+		net_warn_ratelimited("%s: Local routing loop detected!\n",
+				     xi->p.name);
+		goto tx_err_dst_release;
+	}
+
+	mtu = dst_mtu(dst);
+	if (!skb->ignore_df && skb->len > mtu) {
+		skb_dst_update_pmtu(skb, mtu);
+
+		if (skb->protocol == htons(ETH_P_IPV6)) {
+			if (mtu < IPV6_MIN_MTU)
+				mtu = IPV6_MIN_MTU;
+
+			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
+		} else {
+			icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
+				  htonl(mtu));
+		}
+
+		dst_release(dst);
+		return -EMSGSIZE;
+	}
+
+	xfrmi_scrub_packet(skb, !net_eq(xi->net, dev_net(dev)));
+	skb_dst_set(skb, dst);
+	skb->dev = tdev;
+
+	err = dst_output(xi->net, skb->sk, skb);
+	if (net_xmit_eval(err) == 0) {
+		struct pcpu_sw_netstats *tstats = this_cpu_ptr(dev->tstats);
+
+		u64_stats_update_begin(&tstats->syncp);
+		tstats->tx_bytes += length;
+		tstats->tx_packets++;
+		u64_stats_update_end(&tstats->syncp);
+	} else {
+		stats->tx_errors++;
+		stats->tx_aborted_errors++;
+	}
+
+	return 0;
+tx_err_link_failure:
+	stats->tx_carrier_errors++;
+	dst_link_failure(skb);
+tx_err_dst_release:
+	dst_release(dst);
+	return err;
+}
+
+static netdev_tx_t xfrmi_xmit(struct sk_buff *skb, struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct net_device_stats *stats = &xi->dev->stats;
+	struct flowi fl;
+	int ret;
+
+	memset(&fl, 0, sizeof(fl));
+
+	switch (skb->protocol) {
+	case htons(ETH_P_IPV6):
+		xfrm_decode_session(skb, &fl, AF_INET6);
+		memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
+		break;
+	case htons(ETH_P_IP):
+		xfrm_decode_session(skb, &fl, AF_INET);
+		memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
+		break;
+	default:
+		goto tx_err;
+	}
+
+	fl.flowi_oif = xi->phydev->ifindex;
+
+	ret = xfrmi_xmit2(skb, dev, &fl);
+	if (ret < 0)
+		goto tx_err;
+
+	return NETDEV_TX_OK;
+
+tx_err:
+	stats->tx_errors++;
+	stats->tx_dropped++;
+	kfree_skb(skb);
+	return NETDEV_TX_OK;
+}
+
+static int xfrmi4_err(struct sk_buff *skb, u32 info)
+{
+	const struct iphdr *iph = (const struct iphdr *)skb->data;
+	struct net *net = dev_net(skb->dev);
+	int protocol = iph->protocol;
+	struct ip_comp_hdr *ipch;
+	struct ip_esp_hdr *esph;
+	struct ip_auth_hdr *ah ;
+	struct xfrm_state *x;
+	struct xfrm_if *xi;
+	__be32 spi;
+
+	switch (protocol) {
+	case IPPROTO_ESP:
+		esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
+		spi = esph->spi;
+		break;
+	case IPPROTO_AH:
+		ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
+		spi = ah->spi;
+		break;
+	case IPPROTO_COMP:
+		ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
+		spi = htonl(ntohs(ipch->cpi));
+		break;
+	default:
+		return 0;
+	}
+
+	switch (icmp_hdr(skb)->type) {
+	case ICMP_DEST_UNREACH:
+		if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
+			return 0;
+	case ICMP_REDIRECT:
+		break;
+	default:
+		return 0;
+	}
+
+	x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr,
+			      spi, protocol, AF_INET);
+	if (!x)
+		return 0;
+
+	xi = xfrmi_lookup(net, x);
+	if (!xi) {
+		xfrm_state_put(x);
+		return -1;
+	}
+
+	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
+		ipv4_update_pmtu(skb, net, info, 0, 0, protocol, 0);
+	else
+		ipv4_redirect(skb, net, 0, 0, protocol, 0);
+	xfrm_state_put(x);
+
+	return 0;
+}
+
+static int xfrmi6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
+		    u8 type, u8 code, int offset, __be32 info)
+{
+	const struct ipv6hdr *iph = (const struct ipv6hdr *)skb->data;
+	struct net *net = dev_net(skb->dev);
+	int protocol = iph->nexthdr;
+	struct ip_comp_hdr *ipch;
+	struct ip_esp_hdr *esph;
+	struct ip_auth_hdr *ah;
+	struct xfrm_state *x;
+	struct xfrm_if *xi;
+	__be32 spi;
+
+	switch (protocol) {
+	case IPPROTO_ESP:
+		esph = (struct ip_esp_hdr *)(skb->data + offset);
+		spi = esph->spi;
+		break;
+	case IPPROTO_AH:
+		ah = (struct ip_auth_hdr *)(skb->data + offset);
+		spi = ah->spi;
+		break;
+	case IPPROTO_COMP:
+		ipch = (struct ip_comp_hdr *)(skb->data + offset);
+		spi = htonl(ntohs(ipch->cpi));
+		break;
+	default:
+		return 0;
+	}
+
+	if (type != ICMPV6_PKT_TOOBIG &&
+	    type != NDISC_REDIRECT)
+		return 0;
+
+	x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr,
+			      spi, protocol, AF_INET6);
+	if (!x)
+		return 0;
+
+	xi = xfrmi_lookup(net, x);
+	if (!xi) {
+		xfrm_state_put(x);
+		return -1;
+	}
+
+	if (type == NDISC_REDIRECT)
+		ip6_redirect(skb, net, skb->dev->ifindex, 0,
+			     sock_net_uid(net, NULL));
+	else
+		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
+	xfrm_state_put(x);
+
+	return 0;
+}
+
+static int xfrmi_change(struct xfrm_if *xi, const struct xfrm_if_parms *p)
+{
+	if (xi->p.link != p->link)
+		return -EINVAL;
+
+	xi->p.if_id = p->if_id;
+
+	return 0;
+}
+
+static int xfrmi_update(struct xfrm_if *xi, struct xfrm_if_parms *p)
+{
+	struct net *net = dev_net(xi->dev);
+	struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
+	int err;
+
+	xfrmi_unlink(xfrmn, xi);
+	synchronize_net();
+	err = xfrmi_change(xi, p);
+	xfrmi_link(xfrmn, xi);
+	netdev_state_change(xi->dev);
+	return err;
+}
+
+static void xfrmi_get_stats64(struct net_device *dev,
+			       struct rtnl_link_stats64 *s)
+{
+	int cpu;
+
+	if (!dev->tstats)
+		return;
+
+	for_each_possible_cpu(cpu) {
+		struct pcpu_sw_netstats *stats;
+		struct pcpu_sw_netstats tmp;
+		int start;
+
+		stats = per_cpu_ptr(dev->tstats, cpu);
+		do {
+			start = u64_stats_fetch_begin_irq(&stats->syncp);
+			tmp.rx_packets = stats->rx_packets;
+			tmp.rx_bytes   = stats->rx_bytes;
+			tmp.tx_packets = stats->tx_packets;
+			tmp.tx_bytes   = stats->tx_bytes;
+		} while (u64_stats_fetch_retry_irq(&stats->syncp, start));
+
+		s->rx_packets += tmp.rx_packets;
+		s->rx_bytes   += tmp.rx_bytes;
+		s->tx_packets += tmp.tx_packets;
+		s->tx_bytes   += tmp.tx_bytes;
+	}
+
+	s->rx_dropped = dev->stats.rx_dropped;
+	s->tx_dropped = dev->stats.tx_dropped;
+}
+
+static int xfrmi_get_iflink(const struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+
+	return xi->phydev->ifindex;
+}
+
+
+static const struct net_device_ops xfrmi_netdev_ops = {
+	.ndo_init	= xfrmi_dev_init,
+	.ndo_uninit	= xfrmi_dev_uninit,
+	.ndo_start_xmit = xfrmi_xmit,
+	.ndo_get_stats64 = xfrmi_get_stats64,
+	.ndo_get_iflink = xfrmi_get_iflink,
+};
+
+static void xfrmi_dev_setup(struct net_device *dev)
+{
+	dev->netdev_ops 	= &xfrmi_netdev_ops;
+	dev->type		= ARPHRD_NONE;
+	dev->hard_header_len 	= ETH_HLEN;
+	dev->min_header_len	= ETH_HLEN;
+	dev->mtu		= ETH_DATA_LEN;
+	dev->min_mtu		= ETH_MIN_MTU;
+	dev->max_mtu		= ETH_DATA_LEN;
+	dev->addr_len		= ETH_ALEN;
+	dev->flags 		= IFF_NOARP;
+	dev->needs_free_netdev	= true;
+	dev->priv_destructor	= xfrmi_dev_free;
+	netif_keep_dst(dev);
+}
+
+static int xfrmi_dev_init(struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct net_device *phydev = xi->phydev;
+	int err;
+
+	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
+	if (!dev->tstats)
+		return -ENOMEM;
+
+	err = gro_cells_init(&xi->gro_cells, dev);
+	if (err) {
+		free_percpu(dev->tstats);
+		return err;
+	}
+
+	dev->features |= NETIF_F_LLTX;
+
+	dev->needed_headroom = phydev->needed_headroom;
+	dev->needed_tailroom = phydev->needed_tailroom;
+
+	if (is_zero_ether_addr(dev->dev_addr))
+		eth_hw_addr_inherit(dev, phydev);
+	if (is_zero_ether_addr(dev->broadcast))
+		memcpy(dev->broadcast, phydev->broadcast, dev->addr_len);
+
+	return 0;
+}
+
+static int xfrmi_validate(struct nlattr *tb[], struct nlattr *data[],
+			 struct netlink_ext_ack *extack)
+{
+	return 0;
+}
+
+static void xfrmi_netlink_parms(struct nlattr *data[],
+			       struct xfrm_if_parms *parms)
+{
+	memset(parms, 0, sizeof(*parms));
+
+	if (!data)
+		return;
+
+	if (data[IFLA_XFRM_LINK])
+		parms->link = nla_get_u32(data[IFLA_XFRM_LINK]);
+
+	if (data[IFLA_XFRM_IF_ID])
+		parms->if_id = nla_get_u32(data[IFLA_XFRM_IF_ID]);
+}
+
+static int xfrmi_newlink(struct net *src_net, struct net_device *dev,
+			struct nlattr *tb[], struct nlattr *data[],
+			struct netlink_ext_ack *extack)
+{
+	struct net *net = dev_net(dev);
+	struct xfrm_if_parms *p;
+	struct xfrm_if *xi;
+
+	xi = netdev_priv(dev);
+	p = &xi->p;
+
+	xfrmi_netlink_parms(data, p);
+
+	if (!tb[IFLA_IFNAME])
+		return -EINVAL;
+
+	nla_strlcpy(p->name, tb[IFLA_IFNAME], IFNAMSIZ);
+
+	xi = xfrmi_locate(net, p, 1);
+	return PTR_ERR_OR_ZERO(xi);
+}
+
+static void xfrmi_dellink(struct net_device *dev, struct list_head *head)
+{
+	unregister_netdevice_queue(dev, head);
+}
+
+static int xfrmi_changelink(struct net_device *dev, struct nlattr *tb[],
+			   struct nlattr *data[],
+			   struct netlink_ext_ack *extack)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct net *net = dev_net(dev);
+
+	xfrmi_netlink_parms(data, &xi->p);
+
+	xi = xfrmi_locate(net, &xi->p, 0);
+
+	if (IS_ERR_OR_NULL(xi)) {
+		xi = netdev_priv(dev);
+	} else {
+		if (xi->dev != dev)
+			return -EEXIST;
+	}
+
+	return xfrmi_update(xi, &xi->p);
+}
+
+static size_t xfrmi_get_size(const struct net_device *dev)
+{
+	return
+		/* IFLA_XFRM_LINK */
+		nla_total_size(4) +
+		/* IFLA_XFRM_IF_ID */
+		nla_total_size(4) +
+		0;
+}
+
+static int xfrmi_fill_info(struct sk_buff *skb, const struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+	struct xfrm_if_parms *parm = &xi->p;
+
+	if (nla_put_u32(skb, IFLA_XFRM_LINK, parm->link) ||
+	    nla_put_u32(skb, IFLA_XFRM_IF_ID, parm->if_id))
+		goto nla_put_failure;
+	return 0;
+
+nla_put_failure:
+	return -EMSGSIZE;
+}
+
+struct net *xfrmi_get_link_net(const struct net_device *dev)
+{
+	struct xfrm_if *xi = netdev_priv(dev);
+
+	return dev_net(xi->phydev);
+}
+
+static const struct nla_policy xfrmi_policy[IFLA_XFRM_MAX + 1] = {
+	[IFLA_XFRM_LINK]	= { .type = NLA_U32 },
+	[IFLA_XFRM_IF_ID]	= { .type = NLA_U32 },
+};
+
+static struct rtnl_link_ops xfrmi_link_ops __read_mostly = {
+	.kind		= "xfrm",
+	.maxtype	= IFLA_XFRM_MAX,
+	.policy		= xfrmi_policy,
+	.priv_size	= sizeof(struct xfrm_if),
+	.setup		= xfrmi_dev_setup,
+	.validate	= xfrmi_validate,
+	.newlink	= xfrmi_newlink,
+	.dellink	= xfrmi_dellink,
+	.changelink	= xfrmi_changelink,
+	.get_size	= xfrmi_get_size,
+	.fill_info	= xfrmi_fill_info,
+	.get_link_net	= xfrmi_get_link_net,
+};
+
+static void __net_exit xfrmi_destroy_interfaces(struct xfrmi_net *xfrmn)
+{
+	struct xfrm_if *xi;
+	LIST_HEAD(list);
+
+	xi = rtnl_dereference(xfrmn->xfrmi[0]);
+	if (!xi)
+		return;
+
+	unregister_netdevice_queue(xi->dev, &list);
+	unregister_netdevice_many(&list);
+}
+
+static int __net_init xfrmi_init_net(struct net *net)
+{
+	return 0;
+}
+
+static void __net_exit xfrmi_exit_net(struct net *net)
+{
+	struct xfrmi_net *xfrmn = net_generic(net, xfrmi_net_id);
+
+	rtnl_lock();
+	xfrmi_destroy_interfaces(xfrmn);
+	rtnl_unlock();
+}
+
+static struct pernet_operations xfrmi_net_ops = {
+	.init = xfrmi_init_net,
+	.exit = xfrmi_exit_net,
+	.id   = &xfrmi_net_id,
+	.size = sizeof(struct xfrmi_net),
+};
+
+static struct xfrm6_protocol xfrmi_esp6_protocol __read_mostly = {
+	.handler	=	xfrm6_rcv,
+	.cb_handler	=	xfrmi_rcv_cb,
+	.err_handler	=	xfrmi6_err,
+	.priority	=	10,
+};
+
+static struct xfrm6_protocol xfrmi_ah6_protocol __read_mostly = {
+	.handler	=	xfrm6_rcv,
+	.cb_handler	=	xfrmi_rcv_cb,
+	.err_handler	=	xfrmi6_err,
+	.priority	=	10,
+};
+
+static struct xfrm6_protocol xfrmi_ipcomp6_protocol __read_mostly = {
+	.handler	=	xfrm6_rcv,
+	.cb_handler	=	xfrmi_rcv_cb,
+	.err_handler	=	xfrmi6_err,
+	.priority	=	10,
+};
+
+static struct xfrm4_protocol xfrmi_esp4_protocol __read_mostly = {
+	.handler	=	xfrm4_rcv,
+	.input_handler	=	xfrm_input,
+	.cb_handler	=	xfrmi_rcv_cb,
+	.err_handler	=	xfrmi4_err,
+	.priority	=	10,
+};
+
+static struct xfrm4_protocol xfrmi_ah4_protocol __read_mostly = {
+	.handler	=	xfrm4_rcv,
+	.input_handler	=	xfrm_input,
+	.cb_handler	=	xfrmi_rcv_cb,
+	.err_handler	=	xfrmi4_err,
+	.priority	=	10,
+};
+
+static struct xfrm4_protocol xfrmi_ipcomp4_protocol __read_mostly = {
+	.handler	=	xfrm4_rcv,
+	.input_handler	=	xfrm_input,
+	.cb_handler	=	xfrmi_rcv_cb,
+	.err_handler	=	xfrmi4_err,
+	.priority	=	10,
+};
+
+static int __init xfrmi4_init(void)
+{
+	int err;
+
+	err = xfrm4_protocol_register(&xfrmi_esp4_protocol, IPPROTO_ESP);
+	if (err < 0)
+		goto xfrm_proto_esp_failed;
+	err = xfrm4_protocol_register(&xfrmi_ah4_protocol, IPPROTO_AH);
+	if (err < 0)
+		goto xfrm_proto_ah_failed;
+	err = xfrm4_protocol_register(&xfrmi_ipcomp4_protocol, IPPROTO_COMP);
+	if (err < 0)
+		goto xfrm_proto_comp_failed;
+
+	return 0;
+
+xfrm_proto_comp_failed:
+	xfrm4_protocol_deregister(&xfrmi_ah4_protocol, IPPROTO_AH);
+xfrm_proto_ah_failed:
+	xfrm4_protocol_deregister(&xfrmi_esp4_protocol, IPPROTO_ESP);
+xfrm_proto_esp_failed:
+	return err;
+}
+
+static void xfrmi4_fini(void)
+{
+	xfrm4_protocol_deregister(&xfrmi_ipcomp4_protocol, IPPROTO_COMP);
+	xfrm4_protocol_deregister(&xfrmi_ah4_protocol, IPPROTO_AH);
+	xfrm4_protocol_deregister(&xfrmi_esp4_protocol, IPPROTO_ESP);
+}
+
+static int __init xfrmi6_init(void)
+{
+	int err;
+
+	err = xfrm6_protocol_register(&xfrmi_esp6_protocol, IPPROTO_ESP);
+	if (err < 0)
+		goto xfrm_proto_esp_failed;
+	err = xfrm6_protocol_register(&xfrmi_ah6_protocol, IPPROTO_AH);
+	if (err < 0)
+		goto xfrm_proto_ah_failed;
+	err = xfrm6_protocol_register(&xfrmi_ipcomp6_protocol, IPPROTO_COMP);
+	if (err < 0)
+		goto xfrm_proto_comp_failed;
+
+	return 0;
+
+xfrm_proto_comp_failed:
+	xfrm6_protocol_deregister(&xfrmi_ah6_protocol, IPPROTO_AH);
+xfrm_proto_ah_failed:
+	xfrm6_pro