[linux-2.6-block.git] / net / ipv4 / ip_vti.c

// SPDX-License-Identifier: GPL-2.0-or-later
/*
 *	Linux NET3: IP/IP protocol decoder modified to support
 *		    virtual tunnel interface
 *
 *	Authors:
 *		Saurabh Mohan (saurabh.mohan@vyatta.com) 05/07/2012
 */

/*
   This version of net/ipv4/ip_vti.c is cloned of net/ipv4/ipip.c

   For comments look at net/ipv4/ip_gre.c --ANK
 */


#include <linux/capability.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/uaccess.h>
#include <linux/skbuff.h>
#include <linux/netdevice.h>
#include <linux/in.h>
#include <linux/tcp.h>
#include <linux/udp.h>
#include <linux/if_arp.h>
#include <linux/init.h>
#include <linux/netfilter_ipv4.h>
#include <linux/if_ether.h>
#include <linux/icmpv6.h>

#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
#include <net/ip_tunnels.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
#include <net/netns/generic.h>

static struct rtnl_link_ops vti_link_ops __read_mostly;

static unsigned int vti_net_id __read_mostly;
static int vti_tunnel_init(struct net_device *dev);

static int vti_input(struct sk_buff *skb, int nexthdr, __be32 spi,
		     int encap_type, bool update_skb_dev)
{
	struct ip_tunnel *tunnel;
	const struct iphdr *iph = ip_hdr(skb);
	struct net *net = dev_net(skb->dev);
	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);

	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
				  iph->saddr, iph->daddr, 0);
	if (tunnel) {
		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
			goto drop;

		XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = tunnel;

		if (update_skb_dev)
			skb->dev = tunnel->dev;

		return xfrm_input(skb, nexthdr, spi, encap_type);
	}

	return -EINVAL;
drop:
	kfree_skb(skb);
	return 0;
}

static int vti_input_proto(struct sk_buff *skb, int nexthdr, __be32 spi,
			   int encap_type)
{
	return vti_input(skb, nexthdr, spi, encap_type, false);
}

static int vti_rcv(struct sk_buff *skb, __be32 spi, bool update_skb_dev)
{
	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);

	return vti_input(skb, ip_hdr(skb)->protocol, spi, 0, update_skb_dev);
}

static int vti_rcv_proto(struct sk_buff *skb)
{
	return vti_rcv(skb, 0, false);
}

static int vti_rcv_tunnel(struct sk_buff *skb)
{
	return vti_rcv(skb, ip_hdr(skb)->saddr, true);
}

static int vti_rcv_cb(struct sk_buff *skb, int err)
{
	unsigned short family;
	struct net_device *dev;
	struct pcpu_sw_netstats *tstats;
	struct xfrm_state *x;
	const struct xfrm_mode *inner_mode;
	struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4;
	u32 orig_mark = skb->mark;
	int ret;

	if (!tunnel)
		return 1;

	dev = tunnel->dev;

	if (err) {
		dev->stats.rx_errors++;
		dev->stats.rx_dropped++;

		return 0;
	}

	x = xfrm_input_state(skb);

	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;
		}
	}

	family = inner_mode->family;

	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	ret = xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family);
	skb->mark = orig_mark;

	if (!ret)
		return -EPERM;

	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	skb->dev = dev;

	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 bool vti_state_check(const struct xfrm_state *x, __be32 dst, __be32 src)
{
	xfrm_address_t *daddr = (xfrm_address_t *)&dst;
	xfrm_address_t *saddr = (xfrm_address_t *)&src;

	/* if there is no transform then this tunnel is not functional.
	 * Or if the xfrm is not mode tunnel.
	 */
	if (!x || x->props.mode != XFRM_MODE_TUNNEL ||
	    x->props.family != AF_INET)
		return false;

	if (!dst)
		return xfrm_addr_equal(saddr, &x->props.saddr, AF_INET);

	if (!xfrm_state_addr_check(x, daddr, saddr, AF_INET))
		return false;

	return true;
}

static netdev_tx_t vti_xmit(struct sk_buff *skb, struct net_device *dev,
			    struct flowi *fl)
{
	struct ip_tunnel *tunnel = netdev_priv(dev);
	struct ip_tunnel_parm *parms = &tunnel->parms;
	struct dst_entry *dst = skb_dst(skb);
	struct net_device *tdev;	/* Device to other host */
	int pkt_len = skb->len;
	int err;
	int mtu;

	if (!dst) {
		dev->stats.tx_carrier_errors++;
		goto tx_error_icmp;
	}

	dst_hold(dst);
	dst = xfrm_lookup(tunnel->net, dst, fl, NULL, 0);
	if (IS_ERR(dst)) {
		dev->stats.tx_carrier_errors++;
		goto tx_error_icmp;
	}

	if (!vti_state_check(dst->xfrm, parms->iph.daddr, parms->iph.saddr)) {
		dev->stats.tx_carrier_errors++;
		dst_release(dst);
		goto tx_error_icmp;
	}

	tdev = dst->dev;

	if (tdev == dev) {
		dst_release(dst);
		dev->stats.collisions++;
		goto tx_error;
	}

	mtu = dst_mtu(dst);
	if (skb->len > mtu) {
		skb_dst_update_pmtu(skb, mtu);
		if (skb->protocol == htons(ETH_P_IP)) {
			icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
				  htonl(mtu));
		} else {
			if (mtu < IPV6_MIN_MTU)
				mtu = IPV6_MIN_MTU;

			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
		}

		dst_release(dst);
		goto tx_error;
	}

	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
	skb_dst_set(skb, dst);
	skb->dev = skb_dst(skb)->dev;

	err = dst_output(tunnel->net, skb->sk, skb);
	if (net_xmit_eval(err) == 0)
		err = pkt_len;
	iptunnel_xmit_stats(dev, err);
	return NETDEV_TX_OK;

tx_error_icmp:
	dst_link_failure(skb);
tx_error:
	dev->stats.tx_errors++;
	kfree_skb(skb);
	return NETDEV_TX_OK;
}

/* This function assumes it is being called from dev_queue_xmit()
 * and that skb is filled properly by that function.
 */
static netdev_tx_t vti_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
{
	struct ip_tunnel *tunnel = netdev_priv(dev);
	struct flowi fl;

	if (!pskb_inet_may_pull(skb))
		goto tx_err;

	memset(&fl, 0, sizeof(fl));

	switch (skb->protocol) {
	case htons(ETH_P_IP):
		xfrm_decode_session(skb, &fl, AF_INET);
		memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
		break;
	case htons(ETH_P_IPV6):
		xfrm_decode_session(skb, &fl, AF_INET6);
		memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
		break;
	default:
		goto tx_err;
	}

	/* override mark with tunnel output key */
	fl.flowi_mark = be32_to_cpu(tunnel->parms.o_key);

	return vti_xmit(skb, dev, &fl);

tx_err:
	dev->stats.tx_errors++;
	kfree_skb(skb);
	return NETDEV_TX_OK;
}

static int vti4_err(struct sk_buff *skb, u32 info)
{
	__be32 spi;
	__u32 mark;
	struct xfrm_state *x;
	struct ip_tunnel *tunnel;
	struct ip_esp_hdr *esph;
	struct ip_auth_hdr *ah ;
	struct ip_comp_hdr *ipch;
	struct net *net = dev_net(skb->dev);
	const struct iphdr *iph = (const struct iphdr *)skb->data;
	int protocol = iph->protocol;
	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);

	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
				  iph->daddr, iph->saddr, 0);
	if (!tunnel)
		return -1;

	mark = be32_to_cpu(tunnel->parms.o_key);

	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, mark, (const xfrm_address_t *)&iph->daddr,
			      spi, protocol, AF_INET);
	if (!x)
		return 0;

	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
		ipv4_update_pmtu(skb, net, info, 0, protocol);
	else
		ipv4_redirect(skb, net, 0, protocol);
	xfrm_state_put(x);

	return 0;
}

static int
vti_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
{
	int err = 0;
	struct ip_tunnel_parm p;

	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
		return -EFAULT;

	if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) {
		if (p.iph.version != 4 || p.iph.protocol != IPPROTO_IPIP ||
		    p.iph.ihl != 5)
			return -EINVAL;
	}

	if (!(p.i_flags & GRE_KEY))
		p.i_key = 0;
	if (!(p.o_flags & GRE_KEY))
		p.o_key = 0;

	p.i_flags = VTI_ISVTI;

	err = ip_tunnel_ioctl(dev, &p, cmd);
	if (err)
		return err;

	if (cmd != SIOCDELTUNNEL) {
		p.i_flags |= GRE_KEY;
		p.o_flags |= GRE_KEY;
	}

	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
		return -EFAULT;
	return 0;
}

static const struct net_device_ops vti_netdev_ops = {
	.ndo_init	= vti_tunnel_init,
	.ndo_uninit	= ip_tunnel_uninit,
	.ndo_start_xmit	= vti_tunnel_xmit,
	.ndo_do_ioctl	= vti_tunnel_ioctl,
	.ndo_change_mtu	= ip_tunnel_change_mtu,
	.ndo_get_stats64 = ip_tunnel_get_stats64,
	.ndo_get_iflink = ip_tunnel_get_iflink,
};

static void vti_tunnel_setup(struct net_device *dev)
{
	dev->netdev_ops		= &vti_netdev_ops;
	dev->type		= ARPHRD_TUNNEL;
	ip_tunnel_setup(dev, vti_net_id);
}

static int vti_tunnel_init(struct net_device *dev)
{
	struct ip_tunnel *tunnel = netdev_priv(dev);
	struct iphdr *iph = &tunnel->parms.iph;

	memcpy(dev->dev_addr, &iph->saddr, 4);
	memcpy(dev->broadcast, &iph->daddr, 4);

	dev->flags		= IFF_NOARP;
	dev->addr_len		= 4;
	dev->features		|= NETIF_F_LLTX;
	netif_keep_dst(dev);

	return ip_tunnel_init(dev);
}

static void __net_init vti_fb_tunnel_init(struct net_device *dev)
{
	struct ip_tunnel *tunnel = netdev_priv(dev);
	struct iphdr *iph = &tunnel->parms.iph;

	iph->version		= 4;
	iph->protocol		= IPPROTO_IPIP;
	iph->ihl		= 5;
}

static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
	.handler	=	vti_rcv_proto,
	.input_handler	=	vti_input_proto,
	.cb_handler	=	vti_rcv_cb,
	.err_handler	=	vti4_err,
	.priority	=	100,
};

static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
	.handler	=	vti_rcv_proto,
	.input_handler	=	vti_input_proto,
	.cb_handler	=	vti_rcv_cb,
	.err_handler	=	vti4_err,
	.priority	=	100,
};

static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
	.handler	=	vti_rcv_proto,
	.input_handler	=	vti_input_proto,
	.cb_handler	=	vti_rcv_cb,
	.err_handler	=	vti4_err,
	.priority	=	100,
};

static struct xfrm_tunnel ipip_handler __read_mostly = {
	.handler	=	vti_rcv_tunnel,
	.err_handler	=	vti4_err,
	.priority	=	0,
};

static int __net_init vti_init_net(struct net *net)
{
	int err;
	struct ip_tunnel_net *itn;

	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
	if (err)
		return err;
	itn = net_generic(net, vti_net_id);
	if (itn->fb_tunnel_dev)
		vti_fb_tunnel_init(itn->fb_tunnel_dev);
	return 0;
}

static void __net_exit vti_exit_batch_net(struct list_head *list_net)
{
	ip_tunnel_delete_nets(list_net, vti_net_id, &vti_link_ops);
}

static struct pernet_operations vti_net_ops = {
	.init = vti_init_net,
	.exit_batch = vti_exit_batch_net,
	.id   = &vti_net_id,
	.size = sizeof(struct ip_tunnel_net),
};

static int vti_tunnel_validate(struct nlattr *tb[], struct nlattr *data[],
			       struct netlink_ext_ack *extack)
{
	return 0;
}

static void vti_netlink_parms(struct nlattr *data[],
			      struct ip_tunnel_parm *parms,
			      __u32 *fwmark)
{
	memset(parms, 0, sizeof(*parms));

	parms->iph.protocol = IPPROTO_IPIP;

	if (!data)
		return;

	parms->i_flags = VTI_ISVTI;

	if (data[IFLA_VTI_LINK])
		parms->link = nla_get_u32(data[IFLA_VTI_LINK]);

	if (data[IFLA_VTI_IKEY])
		parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);

	if (data[IFLA_VTI_OKEY])
		parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);

	if (data[IFLA_VTI_LOCAL])
		parms->iph.saddr = nla_get_in_addr(data[IFLA_VTI_LOCAL]);

	if (data[IFLA_VTI_REMOTE])
		parms->iph.daddr = nla_get_in_addr(data[IFLA_VTI_REMOTE]);

	if (data[IFLA_VTI_FWMARK])
		*fwmark = nla_get_u32(data[IFLA_VTI_FWMARK]);
}

static int vti_newlink(struct net *src_net, struct net_device *dev,
		       struct nlattr *tb[], struct nlattr *data[],
		       struct netlink_ext_ack *extack)
{
	struct ip_tunnel_parm parms;
	__u32 fwmark = 0;

	vti_netlink_parms(data, &parms, &fwmark);
	return ip_tunnel_newlink(dev, tb, &parms, fwmark);
}

static int vti_changelink(struct net_device *dev, struct nlattr *tb[],
			  struct nlattr *data[],
			  struct netlink_ext_ack *extack)
{
	struct ip_tunnel *t = netdev_priv(dev);
	__u32 fwmark = t->fwmark;
	struct ip_tunnel_parm p;

	vti_netlink_parms(data, &p, &fwmark);
	return ip_tunnel_changelink(dev, tb, &p, fwmark);
}

static size_t vti_get_size(const struct net_device *dev)
{
	return
		/* IFLA_VTI_LINK */
		nla_total_size(4) +
		/* IFLA_VTI_IKEY */
		nla_total_size(4) +
		/* IFLA_VTI_OKEY */
		nla_total_size(4) +
		/* IFLA_VTI_LOCAL */
		nla_total_size(4) +
		/* IFLA_VTI_REMOTE */
		nla_total_size(4) +
		/* IFLA_VTI_FWMARK */
		nla_total_size(4) +
		0;
}

static int vti_fill_info(struct sk_buff *skb, const struct net_device *dev)
{
	struct ip_tunnel *t = netdev_priv(dev);
	struct ip_tunnel_parm *p = &t->parms;

	if (nla_put_u32(skb, IFLA_VTI_LINK, p->link) ||
	    nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key) ||
	    nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key) ||
	    nla_put_in_addr(skb, IFLA_VTI_LOCAL, p->iph.saddr) ||
	    nla_put_in_addr(skb, IFLA_VTI_REMOTE, p->iph.daddr) ||
	    nla_put_u32(skb, IFLA_VTI_FWMARK, t->fwmark))
		return -EMSGSIZE;

	return 0;
}

static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	[IFLA_VTI_LINK]		= { .type = NLA_U32 },
	[IFLA_VTI_IKEY]		= { .type = NLA_U32 },
	[IFLA_VTI_OKEY]		= { .type = NLA_U32 },
	[IFLA_VTI_LOCAL]	= { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	[IFLA_VTI_REMOTE]	= { .len = FIELD_SIZEOF(struct iphdr, daddr) },
	[IFLA_VTI_FWMARK]	= { .type = NLA_U32 },
};

static struct rtnl_link_ops vti_link_ops __read_mostly = {
	.kind		= "vti",
	.maxtype	= IFLA_VTI_MAX,
	.policy		= vti_policy,
	.priv_size	= sizeof(struct ip_tunnel),
	.setup		= vti_tunnel_setup,
	.validate	= vti_tunnel_validate,
	.newlink	= vti_newlink,
	.changelink	= vti_changelink,
	.dellink        = ip_tunnel_dellink,
	.get_size	= vti_get_size,
	.fill_info	= vti_fill_info,
	.get_link_net	= ip_tunnel_get_link_net,
};

static int __init vti_init(void)
{
	const char *msg;
	int err;

	pr_info("IPv4 over IPsec tunneling driver\n");

	msg = "tunnel device";
	err = register_pernet_device(&vti_net_ops);
	if (err < 0)
		goto pernet_dev_failed;

	msg = "tunnel protocols";
	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	if (err < 0)
		goto xfrm_proto_esp_failed;
	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
	if (err < 0)
		goto xfrm_proto_ah_failed;
	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
	if (err < 0)
		goto xfrm_proto_comp_failed;

	msg = "ipip tunnel";
	err = xfrm4_tunnel_register(&ipip_handler, AF_INET);
	if (err < 0)
		goto xfrm_tunnel_failed;

	msg = "netlink interface";
	err = rtnl_link_register(&vti_link_ops);
	if (err < 0)
		goto rtnl_link_failed;

	return err;

rtnl_link_failed:
	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
xfrm_tunnel_failed:
	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
xfrm_proto_comp_failed:
	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
xfrm_proto_ah_failed:
	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
xfrm_proto_esp_failed:
	unregister_pernet_device(&vti_net_ops);
pernet_dev_failed:
	pr_err("vti init: failed to register %s\n", msg);
	return err;
}

static void __exit vti_fini(void)
{
	rtnl_link_unregister(&vti_link_ops);
	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
	unregister_pernet_device(&vti_net_ops);
}

module_init(vti_init);
module_exit(vti_fini);
MODULE_LICENSE("GPL");
MODULE_ALIAS_RTNL_LINK("vti");
MODULE_ALIAS_NETDEV("ip_vti0");
Commit	Line	Data
2874c5fd	1	// SPDX-License-Identifier: GPL-2.0-or-later
1181412c S	2	/*
	3	* Linux NET3: IP/IP protocol decoder modified to support
	4	* virtual tunnel interface
	5	*
	6	* Authors:
	7	* Saurabh Mohan (saurabh.mohan@vyatta.com) 05/07/2012
1181412c S	8	*/
	9
	10	/*
	11	This version of net/ipv4/ip_vti.c is cloned of net/ipv4/ipip.c
	12
	13	For comments look at net/ipv4/ip_gre.c --ANK
	14	*/
	15
	16
	17	#include <linux/capability.h>
	18	#include <linux/module.h>
	19	#include <linux/types.h>
	20	#include <linux/kernel.h>
	21	#include <linux/uaccess.h>
	22	#include <linux/skbuff.h>
	23	#include <linux/netdevice.h>
	24	#include <linux/in.h>
	25	#include <linux/tcp.h>
	26	#include <linux/udp.h>
	27	#include <linux/if_arp.h>
1181412c S	28	#include <linux/init.h>
	29	#include <linux/netfilter_ipv4.h>
	30	#include <linux/if_ether.h>
78a010cc	31	#include <linux/icmpv6.h>
1181412c S	32
	33	#include <net/sock.h>
	34	#include <net/ip.h>
	35	#include <net/icmp.h>
c5441932	36	#include <net/ip_tunnels.h>
1181412c S	37	#include <net/inet_ecn.h>
	38	#include <net/xfrm.h>
	39	#include <net/net_namespace.h>
	40	#include <net/netns/generic.h>
	41
1181412c S	42	static struct rtnl_link_ops vti_link_ops __read_mostly;
1181412c S	43
c7d03a00	44	static unsigned int vti_net_id __read_mostly;
1181412c	45	static int vti_tunnel_init(struct net_device *dev);
1181412c	46
df3893c1	47	static int vti_input(struct sk_buff *skb, int nexthdr, __be32 spi,
f981c57f	48	int encap_type, bool update_skb_dev)
1181412c S	49	{
	50	struct ip_tunnel *tunnel;
	51	const struct iphdr *iph = ip_hdr(skb);
b9959fd3 AW	52	struct net *net = dev_net(skb->dev);
b9959fd3 AW	53	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
1181412c	54
b9959fd3 AW	55	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
b9959fd3 AW	56	iph->saddr, iph->daddr, 0);
00db4124	57	if (tunnel) {
df3893c1 SK	58	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
	59	goto drop;
	60
	61	XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = tunnel;
df3893c1	62
f981c57f JS	63	if (update_skb_dev)
	64	skb->dev = tunnel->dev;
	65
df3893c1 SK	66	return xfrm_input(skb, nexthdr, spi, encap_type);
	67	}
	68
	69	return -EINVAL;
	70	drop:
	71	kfree_skb(skb);
	72	return 0;
	73	}
	74
f981c57f JS	75	static int vti_input_proto(struct sk_buff *skb, int nexthdr, __be32 spi,
f981c57f JS	76	int encap_type)
dd9ee344	77	{
f981c57f	78	return vti_input(skb, nexthdr, spi, encap_type, false);
dd9ee344 SY	79	}
dd9ee344 SY	80
f981c57f	81	static int vti_rcv(struct sk_buff *skb, __be32 spi, bool update_skb_dev)
df3893c1 SK	82	{
	83	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
	84	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
	85
f981c57f	86	return vti_input(skb, ip_hdr(skb)->protocol, spi, 0, update_skb_dev);
df3893c1 SK	87	}
df3893c1 SK	88
f981c57f	89	static int vti_rcv_proto(struct sk_buff *skb)
dd9ee344	90	{
f981c57f JS	91	return vti_rcv(skb, 0, false);
f981c57f JS	92	}
dd9ee344	93
f981c57f JS	94	static int vti_rcv_tunnel(struct sk_buff *skb)
	95	{
	96	return vti_rcv(skb, ip_hdr(skb)->saddr, true);
dd9ee344 SY	97	}
dd9ee344 SY	98
df3893c1 SK	99	static int vti_rcv_cb(struct sk_buff *skb, int err)
	100	{
	101	unsigned short family;
	102	struct net_device *dev;
	103	struct pcpu_sw_netstats *tstats;
	104	struct xfrm_state *x;
4c145dce	105	const struct xfrm_mode *inner_mode;
df3893c1	106	struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4;
d55c670c AD	107	u32 orig_mark = skb->mark;
d55c670c AD	108	int ret;
df3893c1 SK	109
df3893c1 SK	110	if (!tunnel)
1181412c	111	return 1;
df3893c1 SK	112
	113	dev = tunnel->dev;
	114
	115	if (err) {
	116	dev->stats.rx_errors++;
	117	dev->stats.rx_dropped++;
	118
	119	return 0;
1181412c	120	}
1181412c	121
df3893c1	122	x = xfrm_input_state(skb);
1fb81e09	123
c9500d7b	124	inner_mode = &x->inner_mode;
1fb81e09	125
	126	if (x->sel.family == AF_UNSPEC) {
	127	inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol);
	128	if (inner_mode == NULL) {
	129	XFRM_INC_STATS(dev_net(skb->dev),
	130	LINUX_MIB_XFRMINSTATEMODEERROR);
	131	return -EINVAL;
	132	}
	133	}
	134
b45714b1	135	family = inner_mode->family;
df3893c1	136
d55c670c AD	137	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	138	ret = xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family);
	139	skb->mark = orig_mark;
	140
	141	if (!ret)
df3893c1 SK	142	return -EPERM;
	143
	144	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	145	skb->dev = dev;
	146
	147	tstats = this_cpu_ptr(dev->tstats);
	148
	149	u64_stats_update_begin(&tstats->syncp);
	150	tstats->rx_packets++;
	151	tstats->rx_bytes += skb->len;
	152	u64_stats_update_end(&tstats->syncp);
	153
	154	return 0;
1181412c S	155	}
1181412c S	156
6e2de802 SK	157	static bool vti_state_check(const struct xfrm_state *x, __be32 dst, __be32 src)
	158	{
	159	xfrm_address_t daddr = (xfrm_address_t )&dst;
	160	xfrm_address_t saddr = (xfrm_address_t )&src;
	161
	162	/* if there is no transform then this tunnel is not functional.
	163	* Or if the xfrm is not mode tunnel.
	164	*/
	165	if (!x \|\| x->props.mode != XFRM_MODE_TUNNEL \|\|
	166	x->props.family != AF_INET)
	167	return false;
	168
	169	if (!dst)
	170	return xfrm_addr_equal(saddr, &x->props.saddr, AF_INET);
	171
	172	if (!xfrm_state_addr_check(x, daddr, saddr, AF_INET))
	173	return false;
	174
	175	return true;
	176	}
	177
78a010cc SK	178	static netdev_tx_t vti_xmit(struct sk_buff skb, struct net_device dev,
78a010cc SK	179	struct flowi *fl)
1181412c S	180	{
1181412c S	181	struct ip_tunnel *tunnel = netdev_priv(dev);
6e2de802	182	struct ip_tunnel_parm *parms = &tunnel->parms;
a34cd4f3	183	struct dst_entry *dst = skb_dst(skb);
1181412c	184	struct net_device tdev; / Device to other host */
36f6ee22	185	int pkt_len = skb->len;
b9959fd3	186	int err;
d6af1a31	187	int mtu;
1181412c	188
a34cd4f3	189	if (!dst) {
df3893c1 SK	190	dev->stats.tx_carrier_errors++;
	191	goto tx_error_icmp;
	192	}
1181412c	193
a34cd4f3	194	dst_hold(dst);
78a010cc	195	dst = xfrm_lookup(tunnel->net, dst, fl, NULL, 0);
a34cd4f3	196	if (IS_ERR(dst)) {
1181412c S	197	dev->stats.tx_carrier_errors++;
	198	goto tx_error_icmp;
	199	}
df3893c1	200
6e2de802	201	if (!vti_state_check(dst->xfrm, parms->iph.daddr, parms->iph.saddr)) {
1181412c	202	dev->stats.tx_carrier_errors++;
a34cd4f3	203	dst_release(dst);
1181412c S	204	goto tx_error_icmp;
1181412c S	205	}
6e2de802	206
a34cd4f3	207	tdev = dst->dev;
1181412c S	208
1181412c S	209	if (tdev == dev) {
a34cd4f3	210	dst_release(dst);
1181412c S	211	dev->stats.collisions++;
	212	goto tx_error;
	213	}
	214
d6af1a31 SK	215	mtu = dst_mtu(dst);
d6af1a31 SK	216	if (skb->len > mtu) {
f15ca723	217	skb_dst_update_pmtu(skb, mtu);
d6af1a31 SK	218	if (skb->protocol == htons(ETH_P_IP)) {
	219	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
	220	htonl(mtu));
	221	} else {
	222	if (mtu < IPV6_MIN_MTU)
	223	mtu = IPV6_MIN_MTU;
	224
	225	icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
	226	}
	227
	228	dst_release(dst);
	229	goto tx_error;
	230	}
	231
df3893c1	232	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
a34cd4f3	233	skb_dst_set(skb, dst);
1181412c S	234	skb->dev = skb_dst(skb)->dev;
1181412c S	235
13206b6b	236	err = dst_output(tunnel->net, skb->sk, skb);
b9959fd3	237	if (net_xmit_eval(err) == 0)
36f6ee22	238	err = pkt_len;
039f5062	239	iptunnel_xmit_stats(dev, err);
1181412c S	240	return NETDEV_TX_OK;
	241
	242	tx_error_icmp:
	243	dst_link_failure(skb);
	244	tx_error:
	245	dev->stats.tx_errors++;
3acfa1e7	246	kfree_skb(skb);
1181412c S	247	return NETDEV_TX_OK;
	248	}
	249
78a010cc SK	250	/* This function assumes it is being called from dev_queue_xmit()
	251	* and that skb is filled properly by that function.
	252	*/
	253	static netdev_tx_t vti_tunnel_xmit(struct sk_buff skb, struct net_device dev)
	254	{
	255	struct ip_tunnel *tunnel = netdev_priv(dev);
	256	struct flowi fl;
	257
cb9f1b78 WB	258	if (!pskb_inet_may_pull(skb))
	259	goto tx_err;
	260
78a010cc SK	261	memset(&fl, 0, sizeof(fl));
78a010cc SK	262
78a010cc SK	263	switch (skb->protocol) {
	264	case htons(ETH_P_IP):
	265	xfrm_decode_session(skb, &fl, AF_INET);
	266	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	267	break;
	268	case htons(ETH_P_IPV6):
	269	xfrm_decode_session(skb, &fl, AF_INET6);
	270	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
	271	break;
	272	default:
cb9f1b78	273	goto tx_err;
78a010cc SK	274	}
78a010cc SK	275
cd5279c1 AD	276	/* override mark with tunnel output key */
	277	fl.flowi_mark = be32_to_cpu(tunnel->parms.o_key);
	278
78a010cc	279	return vti_xmit(skb, dev, &fl);
cb9f1b78 WB	280
	281	tx_err:
	282	dev->stats.tx_errors++;
	283	kfree_skb(skb);
	284	return NETDEV_TX_OK;
78a010cc SK	285	}
78a010cc SK	286
df3893c1 SK	287	static int vti4_err(struct sk_buff *skb, u32 info)
	288	{
	289	__be32 spi;
6d004d6c	290	__u32 mark;
df3893c1 SK	291	struct xfrm_state *x;
	292	struct ip_tunnel *tunnel;
	293	struct ip_esp_hdr *esph;
	294	struct ip_auth_hdr *ah ;
	295	struct ip_comp_hdr *ipch;
	296	struct net *net = dev_net(skb->dev);
	297	const struct iphdr iph = (const struct iphdr )skb->data;
	298	int protocol = iph->protocol;
	299	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	300
	301	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
	302	iph->daddr, iph->saddr, 0);
	303	if (!tunnel)
	304	return -1;
	305
6d004d6c SK	306	mark = be32_to_cpu(tunnel->parms.o_key);
6d004d6c SK	307
df3893c1 SK	308	switch (protocol) {
	309	case IPPROTO_ESP:
	310	esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
	311	spi = esph->spi;
	312	break;
	313	case IPPROTO_AH:
	314	ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
	315	spi = ah->spi;
	316	break;
	317	case IPPROTO_COMP:
	318	ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	319	spi = htonl(ntohs(ipch->cpi));
	320	break;
	321	default:
	322	return 0;
	323	}
	324
	325	switch (icmp_hdr(skb)->type) {
	326	case ICMP_DEST_UNREACH:
	327	if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
	328	return 0;
	329	case ICMP_REDIRECT:
	330	break;
	331	default:
	332	return 0;
	333	}
	334
6d004d6c	335	x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr,
df3893c1 SK	336	spi, protocol, AF_INET);
	337	if (!x)
	338	return 0;
	339
	340	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
d888f396	341	ipv4_update_pmtu(skb, net, info, 0, protocol);
df3893c1	342	else
1042caa7	343	ipv4_redirect(skb, net, 0, protocol);
df3893c1 SK	344	xfrm_state_put(x);
	345
	346	return 0;
	347	}
	348
1181412c S	349	static int
	350	vti_tunnel_ioctl(struct net_device dev, struct ifreq ifr, int cmd)
	351	{
	352	int err = 0;
	353	struct ip_tunnel_parm p;
1181412c	354
b9959fd3 AW	355	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
b9959fd3 AW	356	return -EFAULT;
1181412c	357
b9959fd3	358	if (cmd == SIOCADDTUNNEL \|\| cmd == SIOCCHGTUNNEL) {
1181412c S	359	if (p.iph.version != 4 \|\| p.iph.protocol != IPPROTO_IPIP \|\|
1181412c S	360	p.iph.ihl != 5)
b9959fd3 AW	361	return -EINVAL;
b9959fd3 AW	362	}
1181412c	363
7c8e6b9c DP	364	if (!(p.i_flags & GRE_KEY))
	365	p.i_key = 0;
	366	if (!(p.o_flags & GRE_KEY))
	367	p.o_key = 0;
	368
	369	p.i_flags = VTI_ISVTI;
	370
b9959fd3 AW	371	err = ip_tunnel_ioctl(dev, &p, cmd);
	372	if (err)
	373	return err;
1181412c	374
b9959fd3	375	if (cmd != SIOCDELTUNNEL) {
df3893c1	376	p.i_flags \|= GRE_KEY;
b9959fd3	377	p.o_flags \|= GRE_KEY;
1181412c S	378	}
1181412c S	379
b9959fd3 AW	380	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
b9959fd3 AW	381	return -EFAULT;
1181412c S	382	return 0;
	383	}
	384
	385	static const struct net_device_ops vti_netdev_ops = {
	386	.ndo_init = vti_tunnel_init,
b9959fd3	387	.ndo_uninit = ip_tunnel_uninit,
1181412c S	388	.ndo_start_xmit = vti_tunnel_xmit,
1181412c S	389	.ndo_do_ioctl = vti_tunnel_ioctl,
b9959fd3	390	.ndo_change_mtu = ip_tunnel_change_mtu,
f61dd388	391	.ndo_get_stats64 = ip_tunnel_get_stats64,
1e99584b	392	.ndo_get_iflink = ip_tunnel_get_iflink,
1181412c S	393	};
1181412c S	394
b9959fd3	395	static void vti_tunnel_setup(struct net_device *dev)
1181412c	396	{
b9959fd3	397	dev->netdev_ops = &vti_netdev_ops;
8d89dcdf	398	dev->type = ARPHRD_TUNNEL;
b9959fd3	399	ip_tunnel_setup(dev, vti_net_id);
1181412c S	400	}
1181412c S	401
b9959fd3	402	static int vti_tunnel_init(struct net_device *dev)
1181412c	403	{
b9959fd3 AW	404	struct ip_tunnel *tunnel = netdev_priv(dev);
	405	struct iphdr *iph = &tunnel->parms.iph;
	406
	407	memcpy(dev->dev_addr, &iph->saddr, 4);
	408	memcpy(dev->broadcast, &iph->daddr, 4);
1181412c	409
1181412c	410	dev->flags = IFF_NOARP;
1181412c	411	dev->addr_len = 4;
1181412c	412	dev->features \|= NETIF_F_LLTX;
02875878	413	netif_keep_dst(dev);
1181412c	414
b9959fd3	415	return ip_tunnel_init(dev);
1181412c S	416	}
1181412c S	417
b9959fd3	418	static void __net_init vti_fb_tunnel_init(struct net_device *dev)
1181412c S	419	{
	420	struct ip_tunnel *tunnel = netdev_priv(dev);
	421	struct iphdr *iph = &tunnel->parms.iph;
1181412c	422
1181412c S	423	iph->version = 4;
	424	iph->protocol = IPPROTO_IPIP;
	425	iph->ihl = 5;
1181412c S	426	}
1181412c S	427
df3893c1	428	static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
f981c57f JS	429	.handler = vti_rcv_proto,
f981c57f JS	430	.input_handler = vti_input_proto,
df3893c1 SK	431	.cb_handler = vti_rcv_cb,
	432	.err_handler = vti4_err,
	433	.priority = 100,
	434	};
	435
	436	static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
f981c57f JS	437	.handler = vti_rcv_proto,
f981c57f JS	438	.input_handler = vti_input_proto,
df3893c1 SK	439	.cb_handler = vti_rcv_cb,
	440	.err_handler = vti4_err,
	441	.priority = 100,
	442	};
	443
	444	static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
f981c57f JS	445	.handler = vti_rcv_proto,
f981c57f JS	446	.input_handler = vti_input_proto,
df3893c1 SK	447	.cb_handler = vti_rcv_cb,
	448	.err_handler = vti4_err,
	449	.priority = 100,
1181412c S	450	};
1181412c S	451
dd9ee344	452	static struct xfrm_tunnel ipip_handler __read_mostly = {
f981c57f	453	.handler = vti_rcv_tunnel,
dd9ee344 SY	454	.err_handler = vti4_err,
	455	.priority = 0,
	456	};
	457
1181412c S	458	static int __net_init vti_init_net(struct net *net)
	459	{
	460	int err;
b9959fd3	461	struct ip_tunnel_net *itn;
1181412c	462
b9959fd3	463	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
1181412c	464	if (err)
b9959fd3 AW	465	return err;
b9959fd3 AW	466	itn = net_generic(net, vti_net_id);
cd1aa9c2 HY	467	if (itn->fb_tunnel_dev)
cd1aa9c2 HY	468	vti_fb_tunnel_init(itn->fb_tunnel_dev);
1181412c	469	return 0;
1181412c S	470	}
1181412c S	471
64bc1781	472	static void __net_exit vti_exit_batch_net(struct list_head *list_net)
1181412c	473	{
64bc1781	474	ip_tunnel_delete_nets(list_net, vti_net_id, &vti_link_ops);
1181412c S	475	}
	476
	477	static struct pernet_operations vti_net_ops = {
	478	.init = vti_init_net,
64bc1781	479	.exit_batch = vti_exit_batch_net,
1181412c	480	.id = &vti_net_id,
b9959fd3	481	.size = sizeof(struct ip_tunnel_net),
1181412c S	482	};
1181412c S	483
a8b8a889 MS	484	static int vti_tunnel_validate(struct nlattr tb[], struct nlattr data[],
a8b8a889 MS	485	struct netlink_ext_ack *extack)
1181412c S	486	{
	487	return 0;
	488	}
	489
	490	static void vti_netlink_parms(struct nlattr *data[],
9830ad4c CG	491	struct ip_tunnel_parm *parms,
9830ad4c CG	492	__u32 *fwmark)
1181412c S	493	{
	494	memset(parms, 0, sizeof(*parms));
	495
	496	parms->iph.protocol = IPPROTO_IPIP;
	497
	498	if (!data)
	499	return;
	500
df3893c1 SK	501	parms->i_flags = VTI_ISVTI;
df3893c1 SK	502
1181412c S	503	if (data[IFLA_VTI_LINK])
	504	parms->link = nla_get_u32(data[IFLA_VTI_LINK]);
	505
	506	if (data[IFLA_VTI_IKEY])
	507	parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);
	508
	509	if (data[IFLA_VTI_OKEY])
	510	parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);
	511
	512	if (data[IFLA_VTI_LOCAL])
67b61f6c	513	parms->iph.saddr = nla_get_in_addr(data[IFLA_VTI_LOCAL]);
1181412c S	514
1181412c S	515	if (data[IFLA_VTI_REMOTE])
67b61f6c	516	parms->iph.daddr = nla_get_in_addr(data[IFLA_VTI_REMOTE]);
1181412c	517
9830ad4c CG	518	if (data[IFLA_VTI_FWMARK])
9830ad4c CG	519	*fwmark = nla_get_u32(data[IFLA_VTI_FWMARK]);
1181412c S	520	}
	521
	522	static int vti_newlink(struct net src_net, struct net_device dev,
7a3f4a18 MS	523	struct nlattr tb[], struct nlattr data[],
7a3f4a18 MS	524	struct netlink_ext_ack *extack)
1181412c	525	{
b9959fd3	526	struct ip_tunnel_parm parms;
9830ad4c	527	__u32 fwmark = 0;
1181412c	528
9830ad4c CG	529	vti_netlink_parms(data, &parms, &fwmark);
9830ad4c CG	530	return ip_tunnel_newlink(dev, tb, &parms, fwmark);
1181412c S	531	}
	532
	533	static int vti_changelink(struct net_device dev, struct nlattr tb[],
ad744b22 MS	534	struct nlattr *data[],
ad744b22 MS	535	struct netlink_ext_ack *extack)
1181412c	536	{
9830ad4c CG	537	struct ip_tunnel *t = netdev_priv(dev);
9830ad4c CG	538	__u32 fwmark = t->fwmark;
1181412c	539	struct ip_tunnel_parm p;
1181412c	540
9830ad4c CG	541	vti_netlink_parms(data, &p, &fwmark);
9830ad4c CG	542	return ip_tunnel_changelink(dev, tb, &p, fwmark);
1181412c S	543	}
	544
	545	static size_t vti_get_size(const struct net_device *dev)
	546	{
	547	return
	548	/* IFLA_VTI_LINK */
	549	nla_total_size(4) +
	550	/* IFLA_VTI_IKEY */
	551	nla_total_size(4) +
	552	/* IFLA_VTI_OKEY */
	553	nla_total_size(4) +
	554	/* IFLA_VTI_LOCAL */
	555	nla_total_size(4) +
	556	/* IFLA_VTI_REMOTE */
	557	nla_total_size(4) +
9830ad4c CG	558	/* IFLA_VTI_FWMARK */
9830ad4c CG	559	nla_total_size(4) +
1181412c S	560	0;
	561	}
	562
	563	static int vti_fill_info(struct sk_buff skb, const struct net_device dev)
	564	{
	565	struct ip_tunnel *t = netdev_priv(dev);
	566	struct ip_tunnel_parm *p = &t->parms;
	567
8ed508fd HL	568	if (nla_put_u32(skb, IFLA_VTI_LINK, p->link) \|\|
	569	nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key) \|\|
	570	nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key) \|\|
	571	nla_put_in_addr(skb, IFLA_VTI_LOCAL, p->iph.saddr) \|\|
	572	nla_put_in_addr(skb, IFLA_VTI_REMOTE, p->iph.daddr) \|\|
	573	nla_put_u32(skb, IFLA_VTI_FWMARK, t->fwmark))
	574	return -EMSGSIZE;
1181412c S	575
	576	return 0;
	577	}
	578
	579	static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	580	[IFLA_VTI_LINK] = { .type = NLA_U32 },
	581	[IFLA_VTI_IKEY] = { .type = NLA_U32 },
	582	[IFLA_VTI_OKEY] = { .type = NLA_U32 },
	583	[IFLA_VTI_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	584	[IFLA_VTI_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
9830ad4c	585	[IFLA_VTI_FWMARK] = { .type = NLA_U32 },
1181412c S	586	};
	587
	588	static struct rtnl_link_ops vti_link_ops __read_mostly = {
	589	.kind = "vti",
	590	.maxtype = IFLA_VTI_MAX,
	591	.policy = vti_policy,
	592	.priv_size = sizeof(struct ip_tunnel),
	593	.setup = vti_tunnel_setup,
	594	.validate = vti_tunnel_validate,
	595	.newlink = vti_newlink,
	596	.changelink = vti_changelink,
20ea60ca	597	.dellink = ip_tunnel_dellink,
1181412c S	598	.get_size = vti_get_size,
1181412c S	599	.fill_info = vti_fill_info,
1728d4fa	600	.get_link_net = ip_tunnel_get_link_net,
1181412c S	601	};
	602
	603	static int __init vti_init(void)
	604	{
1990e4f8	605	const char *msg;
1181412c S	606	int err;
1181412c S	607
1990e4f8	608	pr_info("IPv4 over IPsec tunneling driver\n");
1181412c	609
1990e4f8	610	msg = "tunnel device";
1181412c S	611	err = register_pernet_device(&vti_net_ops);
1181412c S	612	if (err < 0)
1990e4f8	613	goto pernet_dev_failed;
df3893c1	614
1990e4f8 MK	615	msg = "tunnel protocols";
	616	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	617	if (err < 0)
	618	goto xfrm_proto_esp_failed;
df3893c1	619	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8 MK	620	if (err < 0)
1990e4f8 MK	621	goto xfrm_proto_ah_failed;
df3893c1	622	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8 MK	623	if (err < 0)
1990e4f8 MK	624	goto xfrm_proto_comp_failed;
1181412c	625
dd9ee344 SY	626	msg = "ipip tunnel";
dd9ee344 SY	627	err = xfrm4_tunnel_register(&ipip_handler, AF_INET);
01ce31c5	628	if (err < 0)
dd9ee344	629	goto xfrm_tunnel_failed;
dd9ee344	630
1990e4f8	631	msg = "netlink interface";
1181412c S	632	err = rtnl_link_register(&vti_link_ops);
	633	if (err < 0)
	634	goto rtnl_link_failed;
	635
	636	return err;
	637
	638	rtnl_link_failed:
dd9ee344	639	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
5483844c JS	640	xfrm_tunnel_failed:
5483844c JS	641	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8	642	xfrm_proto_comp_failed:
df3893c1	643	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8	644	xfrm_proto_ah_failed:
df3893c1	645	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1990e4f8	646	xfrm_proto_esp_failed:
1181412c	647	unregister_pernet_device(&vti_net_ops);
1990e4f8 MK	648	pernet_dev_failed:
1990e4f8 MK	649	pr_err("vti init: failed to register %s\n", msg);
1181412c S	650	return err;
	651	}
	652
	653	static void __exit vti_fini(void)
	654	{
	655	rtnl_link_unregister(&vti_link_ops);
5483844c	656	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
1990e4f8 MK	657	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	658	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	659	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c S	660	unregister_pernet_device(&vti_net_ops);
	661	}
	662
	663	module_init(vti_init);
	664	module_exit(vti_fini);
	665	MODULE_LICENSE("GPL");
	666	MODULE_ALIAS_RTNL_LINK("vti");
	667	MODULE_ALIAS_NETDEV("ip_vti0");