[mirror_ubuntu-artful-kernel.git] / net / ipv4 / ip_vti.c

/*
 *	Linux NET3: IP/IP protocol decoder modified to support
 *		    virtual tunnel interface
 *
 *	Authors:
 *		Saurabh Mohan (saurabh.mohan@vyatta.com) 05/07/2012
 *
 *	This program is free software; you can redistribute it and/or
 *	modify it under the terms of the GNU General Public License
 *	as published by the Free Software Foundation; either version
 *	2 of the License, or (at your option) any later version.
 *
 */

/*
   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)
{
	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;

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

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

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

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;
	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->afinfo->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;
	}

	if (tunnel->err_count > 0) {
		if (time_before(jiffies,
				tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
			tunnel->err_count--;
			dst_link_failure(skb);
		} else
			tunnel->err_count = 0;
	}

	mtu = dst_mtu(dst);
	if (skb->len > mtu) {
		skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, 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;

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

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

	return vti_xmit(skb, dev, &fl);
}

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, 0, protocol, 0);
	else
		ipv4_redirect(skb, net, 0, 0, protocol, 0);
	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->hard_header_len	= LL_MAX_HEADER + sizeof(struct iphdr);
	dev->mtu		= ETH_DATA_LEN;
	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,
	.input_handler	=	vti_input,
	.cb_handler	=	vti_rcv_cb,
	.err_handler	=	vti4_err,
	.priority	=	100,
};

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

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

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);
	vti_fb_tunnel_init(itn->fb_tunnel_dev);
	return 0;
}

static void __net_exit vti_exit_net(struct net *net)
{
	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	ip_tunnel_delete_net(itn, &vti_link_ops);
}

static struct pernet_operations vti_net_ops = {
	.init = vti_init_net,
	.exit = vti_exit_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 bool is_vti_tunnel(const struct net_device *dev)
{
	return dev->netdev_ops == &vti_netdev_ops;
}

static int vti_device_event(struct notifier_block *unused,
			    unsigned long event, void *ptr)
{
	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	struct ip_tunnel *tunnel = netdev_priv(dev);

	if (!is_vti_tunnel(dev))
		return NOTIFY_DONE;

	switch (event) {
	case NETDEV_DOWN:
		if (!net_eq(tunnel->net, dev_net(dev)))
			xfrm_garbage_collect(tunnel->net);
		break;
	}
	return NOTIFY_DONE;
}

static struct notifier_block vti_notifier_block __read_mostly = {
	.notifier_call = vti_device_event,
};

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

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

	register_netdevice_notifier(&vti_notifier_block);

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

	return err;

rtnl_link_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:
	unregister_netdevice_notifier(&vti_notifier_block);
	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_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);
	unregister_netdevice_notifier(&vti_notifier_block);
}

module_init(vti_init);
module_exit(vti_fini);
MODULE_LICENSE("GPL");
MODULE_ALIAS_RTNL_LINK("vti");
MODULE_ALIAS_NETDEV("ip_vti0");
Commit	Line	Data
1181412c S	1	/*
	2	* Linux NET3: IP/IP protocol decoder modified to support
	3	* virtual tunnel interface
	4	*
	5	* Authors:
	6	* Saurabh Mohan (saurabh.mohan@vyatta.com) 05/07/2012
	7	*
	8	* This program is free software; you can redistribute it and/or
	9	* modify it under the terms of the GNU General Public License
	10	* as published by the Free Software Foundation; either version
	11	* 2 of the License, or (at your option) any later version.
	12	*
	13	*/
	14
	15	/*
	16	This version of net/ipv4/ip_vti.c is cloned of net/ipv4/ipip.c
	17
	18	For comments look at net/ipv4/ip_gre.c --ANK
	19	*/
	20
	21
	22	#include <linux/capability.h>
	23	#include <linux/module.h>
	24	#include <linux/types.h>
	25	#include <linux/kernel.h>
	26	#include <linux/uaccess.h>
	27	#include <linux/skbuff.h>
	28	#include <linux/netdevice.h>
	29	#include <linux/in.h>
	30	#include <linux/tcp.h>
	31	#include <linux/udp.h>
	32	#include <linux/if_arp.h>
1181412c S	33	#include <linux/init.h>
	34	#include <linux/netfilter_ipv4.h>
	35	#include <linux/if_ether.h>
78a010cc	36	#include <linux/icmpv6.h>
1181412c S	37
	38	#include <net/sock.h>
	39	#include <net/ip.h>
	40	#include <net/icmp.h>
c5441932	41	#include <net/ip_tunnels.h>
1181412c S	42	#include <net/inet_ecn.h>
	43	#include <net/xfrm.h>
	44	#include <net/net_namespace.h>
	45	#include <net/netns/generic.h>
	46
1181412c S	47	static struct rtnl_link_ops vti_link_ops __read_mostly;
1181412c S	48
c7d03a00	49	static unsigned int vti_net_id __read_mostly;
1181412c	50	static int vti_tunnel_init(struct net_device *dev);
1181412c	51
df3893c1 SK	52	static int vti_input(struct sk_buff *skb, int nexthdr, __be32 spi,
df3893c1 SK	53	int encap_type)
1181412c S	54	{
	55	struct ip_tunnel *tunnel;
	56	const struct iphdr *iph = ip_hdr(skb);
b9959fd3 AW	57	struct net *net = dev_net(skb->dev);
b9959fd3 AW	58	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
1181412c	59
b9959fd3 AW	60	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
b9959fd3 AW	61	iph->saddr, iph->daddr, 0);
00db4124	62	if (tunnel) {
df3893c1 SK	63	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
	64	goto drop;
	65
	66	XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = tunnel;
df3893c1 SK	67
	68	return xfrm_input(skb, nexthdr, spi, encap_type);
	69	}
	70
	71	return -EINVAL;
	72	drop:
	73	kfree_skb(skb);
	74	return 0;
	75	}
	76
	77	static int vti_rcv(struct sk_buff *skb)
	78	{
	79	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
	80	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
	81
	82	return vti_input(skb, ip_hdr(skb)->protocol, 0, 0);
	83	}
	84
	85	static int vti_rcv_cb(struct sk_buff *skb, int err)
	86	{
	87	unsigned short family;
	88	struct net_device *dev;
	89	struct pcpu_sw_netstats *tstats;
	90	struct xfrm_state *x;
1fb81e09	91	struct xfrm_mode *inner_mode;
df3893c1	92	struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4;
d55c670c AD	93	u32 orig_mark = skb->mark;
d55c670c AD	94	int ret;
df3893c1 SK	95
df3893c1 SK	96	if (!tunnel)
1181412c	97	return 1;
df3893c1 SK	98
	99	dev = tunnel->dev;
	100
	101	if (err) {
	102	dev->stats.rx_errors++;
	103	dev->stats.rx_dropped++;
	104
	105	return 0;
1181412c	106	}
1181412c	107
df3893c1	108	x = xfrm_input_state(skb);
1fb81e09	109
	110	inner_mode = x->inner_mode;
	111
	112	if (x->sel.family == AF_UNSPEC) {
	113	inner_mode = xfrm_ip2inner_mode(x, XFRM_MODE_SKB_CB(skb)->protocol);
	114	if (inner_mode == NULL) {
	115	XFRM_INC_STATS(dev_net(skb->dev),
	116	LINUX_MIB_XFRMINSTATEMODEERROR);
	117	return -EINVAL;
	118	}
	119	}
	120
	121	family = inner_mode->afinfo->family;
df3893c1	122
d55c670c AD	123	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	124	ret = xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family);
	125	skb->mark = orig_mark;
	126
	127	if (!ret)
df3893c1 SK	128	return -EPERM;
	129
	130	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	131	skb->dev = dev;
	132
	133	tstats = this_cpu_ptr(dev->tstats);
	134
	135	u64_stats_update_begin(&tstats->syncp);
	136	tstats->rx_packets++;
	137	tstats->rx_bytes += skb->len;
	138	u64_stats_update_end(&tstats->syncp);
	139
	140	return 0;
1181412c S	141	}
1181412c S	142
6e2de802 SK	143	static bool vti_state_check(const struct xfrm_state *x, __be32 dst, __be32 src)
	144	{
	145	xfrm_address_t daddr = (xfrm_address_t )&dst;
	146	xfrm_address_t saddr = (xfrm_address_t )&src;
	147
	148	/* if there is no transform then this tunnel is not functional.
	149	* Or if the xfrm is not mode tunnel.
	150	*/
	151	if (!x \|\| x->props.mode != XFRM_MODE_TUNNEL \|\|
	152	x->props.family != AF_INET)
	153	return false;
	154
	155	if (!dst)
	156	return xfrm_addr_equal(saddr, &x->props.saddr, AF_INET);
	157
	158	if (!xfrm_state_addr_check(x, daddr, saddr, AF_INET))
	159	return false;
	160
	161	return true;
	162	}
	163
78a010cc SK	164	static netdev_tx_t vti_xmit(struct sk_buff skb, struct net_device dev,
78a010cc SK	165	struct flowi *fl)
1181412c S	166	{
1181412c S	167	struct ip_tunnel *tunnel = netdev_priv(dev);
6e2de802	168	struct ip_tunnel_parm *parms = &tunnel->parms;
a34cd4f3	169	struct dst_entry *dst = skb_dst(skb);
1181412c	170	struct net_device tdev; / Device to other host */
23220f84	171	int pkt_len = skb->len;
b9959fd3	172	int err;
d6af1a31	173	int mtu;
1181412c	174
a34cd4f3	175	if (!dst) {
df3893c1 SK	176	dev->stats.tx_carrier_errors++;
	177	goto tx_error_icmp;
	178	}
1181412c	179
a34cd4f3	180	dst_hold(dst);
78a010cc	181	dst = xfrm_lookup(tunnel->net, dst, fl, NULL, 0);
a34cd4f3	182	if (IS_ERR(dst)) {
1181412c S	183	dev->stats.tx_carrier_errors++;
	184	goto tx_error_icmp;
	185	}
df3893c1	186
6e2de802	187	if (!vti_state_check(dst->xfrm, parms->iph.daddr, parms->iph.saddr)) {
1181412c	188	dev->stats.tx_carrier_errors++;
a34cd4f3	189	dst_release(dst);
1181412c S	190	goto tx_error_icmp;
1181412c S	191	}
6e2de802	192
a34cd4f3	193	tdev = dst->dev;
1181412c S	194
1181412c S	195	if (tdev == dev) {
a34cd4f3	196	dst_release(dst);
1181412c S	197	dev->stats.collisions++;
	198	goto tx_error;
	199	}
	200
	201	if (tunnel->err_count > 0) {
	202	if (time_before(jiffies,
	203	tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
	204	tunnel->err_count--;
	205	dst_link_failure(skb);
	206	} else
	207	tunnel->err_count = 0;
	208	}
	209
d6af1a31 SK	210	mtu = dst_mtu(dst);
	211	if (skb->len > mtu) {
	212	skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, skb, mtu);
	213	if (skb->protocol == htons(ETH_P_IP)) {
	214	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
	215	htonl(mtu));
	216	} else {
	217	if (mtu < IPV6_MIN_MTU)
	218	mtu = IPV6_MIN_MTU;
	219
	220	icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
	221	}
	222
	223	dst_release(dst);
	224	goto tx_error;
	225	}
	226
df3893c1	227	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
a34cd4f3	228	skb_dst_set(skb, dst);
1181412c S	229	skb->dev = skb_dst(skb)->dev;
1181412c S	230
13206b6b	231	err = dst_output(tunnel->net, skb->sk, skb);
b9959fd3	232	if (net_xmit_eval(err) == 0)
23220f84	233	err = pkt_len;
039f5062	234	iptunnel_xmit_stats(dev, err);
1181412c S	235	return NETDEV_TX_OK;
	236
	237	tx_error_icmp:
	238	dst_link_failure(skb);
	239	tx_error:
	240	dev->stats.tx_errors++;
3acfa1e7	241	kfree_skb(skb);
1181412c S	242	return NETDEV_TX_OK;
	243	}
	244
78a010cc SK	245	/* This function assumes it is being called from dev_queue_xmit()
	246	* and that skb is filled properly by that function.
	247	*/
	248	static netdev_tx_t vti_tunnel_xmit(struct sk_buff skb, struct net_device dev)
	249	{
	250	struct ip_tunnel *tunnel = netdev_priv(dev);
	251	struct flowi fl;
	252
	253	memset(&fl, 0, sizeof(fl));
	254
78a010cc SK	255	switch (skb->protocol) {
	256	case htons(ETH_P_IP):
	257	xfrm_decode_session(skb, &fl, AF_INET);
	258	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	259	break;
	260	case htons(ETH_P_IPV6):
	261	xfrm_decode_session(skb, &fl, AF_INET6);
	262	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
	263	break;
	264	default:
	265	dev->stats.tx_errors++;
	266	dev_kfree_skb(skb);
	267	return NETDEV_TX_OK;
	268	}
	269
cd5279c1 AD	270	/* override mark with tunnel output key */
	271	fl.flowi_mark = be32_to_cpu(tunnel->parms.o_key);
	272
78a010cc SK	273	return vti_xmit(skb, dev, &fl);
	274	}
	275
df3893c1 SK	276	static int vti4_err(struct sk_buff *skb, u32 info)
	277	{
	278	__be32 spi;
6d004d6c	279	__u32 mark;
df3893c1 SK	280	struct xfrm_state *x;
	281	struct ip_tunnel *tunnel;
	282	struct ip_esp_hdr *esph;
	283	struct ip_auth_hdr *ah ;
	284	struct ip_comp_hdr *ipch;
	285	struct net *net = dev_net(skb->dev);
	286	const struct iphdr iph = (const struct iphdr )skb->data;
	287	int protocol = iph->protocol;
	288	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	289
	290	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
	291	iph->daddr, iph->saddr, 0);
	292	if (!tunnel)
	293	return -1;
	294
6d004d6c SK	295	mark = be32_to_cpu(tunnel->parms.o_key);
6d004d6c SK	296
df3893c1 SK	297	switch (protocol) {
	298	case IPPROTO_ESP:
	299	esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
	300	spi = esph->spi;
	301	break;
	302	case IPPROTO_AH:
	303	ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
	304	spi = ah->spi;
	305	break;
	306	case IPPROTO_COMP:
	307	ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	308	spi = htonl(ntohs(ipch->cpi));
	309	break;
	310	default:
	311	return 0;
	312	}
	313
	314	switch (icmp_hdr(skb)->type) {
	315	case ICMP_DEST_UNREACH:
	316	if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
	317	return 0;
	318	case ICMP_REDIRECT:
	319	break;
	320	default:
	321	return 0;
	322	}
	323
6d004d6c	324	x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr,
df3893c1 SK	325	spi, protocol, AF_INET);
	326	if (!x)
	327	return 0;
	328
	329	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
	330	ipv4_update_pmtu(skb, net, info, 0, 0, protocol, 0);
	331	else
	332	ipv4_redirect(skb, net, 0, 0, protocol, 0);
	333	xfrm_state_put(x);
	334
	335	return 0;
	336	}
	337
1181412c S	338	static int
	339	vti_tunnel_ioctl(struct net_device dev, struct ifreq ifr, int cmd)
	340	{
	341	int err = 0;
	342	struct ip_tunnel_parm p;
1181412c	343
b9959fd3 AW	344	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
b9959fd3 AW	345	return -EFAULT;
1181412c	346
b9959fd3	347	if (cmd == SIOCADDTUNNEL \|\| cmd == SIOCCHGTUNNEL) {
1181412c S	348	if (p.iph.version != 4 \|\| p.iph.protocol != IPPROTO_IPIP \|\|
1181412c S	349	p.iph.ihl != 5)
b9959fd3 AW	350	return -EINVAL;
b9959fd3 AW	351	}
1181412c	352
7c8e6b9c DP	353	if (!(p.i_flags & GRE_KEY))
	354	p.i_key = 0;
	355	if (!(p.o_flags & GRE_KEY))
	356	p.o_key = 0;
	357
	358	p.i_flags = VTI_ISVTI;
	359
b9959fd3 AW	360	err = ip_tunnel_ioctl(dev, &p, cmd);
	361	if (err)
	362	return err;
1181412c	363
b9959fd3	364	if (cmd != SIOCDELTUNNEL) {
df3893c1	365	p.i_flags \|= GRE_KEY;
b9959fd3	366	p.o_flags \|= GRE_KEY;
1181412c S	367	}
1181412c S	368
b9959fd3 AW	369	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
b9959fd3 AW	370	return -EFAULT;
1181412c S	371	return 0;
	372	}
	373
	374	static const struct net_device_ops vti_netdev_ops = {
	375	.ndo_init = vti_tunnel_init,
b9959fd3	376	.ndo_uninit = ip_tunnel_uninit,
1181412c S	377	.ndo_start_xmit = vti_tunnel_xmit,
1181412c S	378	.ndo_do_ioctl = vti_tunnel_ioctl,
b9959fd3	379	.ndo_change_mtu = ip_tunnel_change_mtu,
f61dd388	380	.ndo_get_stats64 = ip_tunnel_get_stats64,
1e99584b	381	.ndo_get_iflink = ip_tunnel_get_iflink,
1181412c S	382	};
1181412c S	383
b9959fd3	384	static void vti_tunnel_setup(struct net_device *dev)
1181412c	385	{
b9959fd3	386	dev->netdev_ops = &vti_netdev_ops;
8d89dcdf	387	dev->type = ARPHRD_TUNNEL;
b9959fd3	388	ip_tunnel_setup(dev, vti_net_id);
1181412c S	389	}
1181412c S	390
b9959fd3	391	static int vti_tunnel_init(struct net_device *dev)
1181412c	392	{
b9959fd3 AW	393	struct ip_tunnel *tunnel = netdev_priv(dev);
	394	struct iphdr *iph = &tunnel->parms.iph;
	395
	396	memcpy(dev->dev_addr, &iph->saddr, 4);
	397	memcpy(dev->broadcast, &iph->daddr, 4);
1181412c	398
1181412c S	399	dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr);
	400	dev->mtu = ETH_DATA_LEN;
	401	dev->flags = IFF_NOARP;
1181412c	402	dev->addr_len = 4;
1181412c	403	dev->features \|= NETIF_F_LLTX;
02875878	404	netif_keep_dst(dev);
1181412c	405
b9959fd3	406	return ip_tunnel_init(dev);
1181412c S	407	}
1181412c S	408
b9959fd3	409	static void __net_init vti_fb_tunnel_init(struct net_device *dev)
1181412c S	410	{
	411	struct ip_tunnel *tunnel = netdev_priv(dev);
	412	struct iphdr *iph = &tunnel->parms.iph;
1181412c	413
1181412c S	414	iph->version = 4;
	415	iph->protocol = IPPROTO_IPIP;
	416	iph->ihl = 5;
1181412c S	417	}
1181412c S	418
df3893c1	419	static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
1181412c	420	.handler = vti_rcv,
df3893c1 SK	421	.input_handler = vti_input,
	422	.cb_handler = vti_rcv_cb,
	423	.err_handler = vti4_err,
	424	.priority = 100,
	425	};
	426
	427	static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
	428	.handler = vti_rcv,
	429	.input_handler = vti_input,
	430	.cb_handler = vti_rcv_cb,
	431	.err_handler = vti4_err,
	432	.priority = 100,
	433	};
	434
	435	static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
	436	.handler = vti_rcv,
	437	.input_handler = vti_input,
	438	.cb_handler = vti_rcv_cb,
	439	.err_handler = vti4_err,
	440	.priority = 100,
1181412c S	441	};
1181412c S	442
1181412c S	443	static int __net_init vti_init_net(struct net *net)
	444	{
	445	int err;
b9959fd3	446	struct ip_tunnel_net *itn;
1181412c	447
b9959fd3	448	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
1181412c	449	if (err)
b9959fd3 AW	450	return err;
	451	itn = net_generic(net, vti_net_id);
	452	vti_fb_tunnel_init(itn->fb_tunnel_dev);
1181412c	453	return 0;
1181412c S	454	}
	455
	456	static void __net_exit vti_exit_net(struct net *net)
	457	{
b9959fd3	458	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
6c742e71	459	ip_tunnel_delete_net(itn, &vti_link_ops);
1181412c S	460	}
	461
	462	static struct pernet_operations vti_net_ops = {
	463	.init = vti_init_net,
	464	.exit = vti_exit_net,
	465	.id = &vti_net_id,
b9959fd3	466	.size = sizeof(struct ip_tunnel_net),
1181412c S	467	};
1181412c S	468
a8b8a889 MS	469	static int vti_tunnel_validate(struct nlattr tb[], struct nlattr data[],
a8b8a889 MS	470	struct netlink_ext_ack *extack)
1181412c S	471	{
	472	return 0;
	473	}
	474
	475	static void vti_netlink_parms(struct nlattr *data[],
9830ad4c CG	476	struct ip_tunnel_parm *parms,
9830ad4c CG	477	__u32 *fwmark)
1181412c S	478	{
	479	memset(parms, 0, sizeof(*parms));
	480
	481	parms->iph.protocol = IPPROTO_IPIP;
	482
	483	if (!data)
	484	return;
	485
df3893c1 SK	486	parms->i_flags = VTI_ISVTI;
df3893c1 SK	487
1181412c S	488	if (data[IFLA_VTI_LINK])
	489	parms->link = nla_get_u32(data[IFLA_VTI_LINK]);
	490
	491	if (data[IFLA_VTI_IKEY])
	492	parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);
	493
	494	if (data[IFLA_VTI_OKEY])
	495	parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);
	496
	497	if (data[IFLA_VTI_LOCAL])
67b61f6c	498	parms->iph.saddr = nla_get_in_addr(data[IFLA_VTI_LOCAL]);
1181412c S	499
1181412c S	500	if (data[IFLA_VTI_REMOTE])
67b61f6c	501	parms->iph.daddr = nla_get_in_addr(data[IFLA_VTI_REMOTE]);
1181412c	502
9830ad4c CG	503	if (data[IFLA_VTI_FWMARK])
9830ad4c CG	504	*fwmark = nla_get_u32(data[IFLA_VTI_FWMARK]);
1181412c S	505	}
	506
	507	static int vti_newlink(struct net src_net, struct net_device dev,
7a3f4a18 MS	508	struct nlattr tb[], struct nlattr data[],
7a3f4a18 MS	509	struct netlink_ext_ack *extack)
1181412c	510	{
b9959fd3	511	struct ip_tunnel_parm parms;
9830ad4c	512	__u32 fwmark = 0;
1181412c	513
9830ad4c CG	514	vti_netlink_parms(data, &parms, &fwmark);
9830ad4c CG	515	return ip_tunnel_newlink(dev, tb, &parms, fwmark);
1181412c S	516	}
	517
	518	static int vti_changelink(struct net_device dev, struct nlattr tb[],
ad744b22 MS	519	struct nlattr *data[],
ad744b22 MS	520	struct netlink_ext_ack *extack)
1181412c	521	{
9830ad4c CG	522	struct ip_tunnel *t = netdev_priv(dev);
9830ad4c CG	523	__u32 fwmark = t->fwmark;
1181412c	524	struct ip_tunnel_parm p;
1181412c	525
9830ad4c CG	526	vti_netlink_parms(data, &p, &fwmark);
9830ad4c CG	527	return ip_tunnel_changelink(dev, tb, &p, fwmark);
1181412c S	528	}
	529
	530	static size_t vti_get_size(const struct net_device *dev)
	531	{
	532	return
	533	/* IFLA_VTI_LINK */
	534	nla_total_size(4) +
	535	/* IFLA_VTI_IKEY */
	536	nla_total_size(4) +
	537	/* IFLA_VTI_OKEY */
	538	nla_total_size(4) +
	539	/* IFLA_VTI_LOCAL */
	540	nla_total_size(4) +
	541	/* IFLA_VTI_REMOTE */
	542	nla_total_size(4) +
9830ad4c CG	543	/* IFLA_VTI_FWMARK */
9830ad4c CG	544	nla_total_size(4) +
1181412c S	545	0;
	546	}
	547
	548	static int vti_fill_info(struct sk_buff skb, const struct net_device dev)
	549	{
	550	struct ip_tunnel *t = netdev_priv(dev);
	551	struct ip_tunnel_parm *p = &t->parms;
	552
8ed508fd HL	553	if (nla_put_u32(skb, IFLA_VTI_LINK, p->link) \|\|
	554	nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key) \|\|
	555	nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key) \|\|
	556	nla_put_in_addr(skb, IFLA_VTI_LOCAL, p->iph.saddr) \|\|
	557	nla_put_in_addr(skb, IFLA_VTI_REMOTE, p->iph.daddr) \|\|
	558	nla_put_u32(skb, IFLA_VTI_FWMARK, t->fwmark))
	559	return -EMSGSIZE;
1181412c S	560
	561	return 0;
	562	}
	563
	564	static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	565	[IFLA_VTI_LINK] = { .type = NLA_U32 },
	566	[IFLA_VTI_IKEY] = { .type = NLA_U32 },
	567	[IFLA_VTI_OKEY] = { .type = NLA_U32 },
	568	[IFLA_VTI_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	569	[IFLA_VTI_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
9830ad4c	570	[IFLA_VTI_FWMARK] = { .type = NLA_U32 },
1181412c S	571	};
	572
	573	static struct rtnl_link_ops vti_link_ops __read_mostly = {
	574	.kind = "vti",
	575	.maxtype = IFLA_VTI_MAX,
	576	.policy = vti_policy,
	577	.priv_size = sizeof(struct ip_tunnel),
	578	.setup = vti_tunnel_setup,
	579	.validate = vti_tunnel_validate,
	580	.newlink = vti_newlink,
	581	.changelink = vti_changelink,
20ea60ca	582	.dellink = ip_tunnel_dellink,
1181412c S	583	.get_size = vti_get_size,
1181412c S	584	.fill_info = vti_fill_info,
1728d4fa	585	.get_link_net = ip_tunnel_get_link_net,
1181412c S	586	};
1181412c S	587
a5d0dc81 LR	588	static bool is_vti_tunnel(const struct net_device *dev)
	589	{
	590	return dev->netdev_ops == &vti_netdev_ops;
	591	}
	592
	593	static int vti_device_event(struct notifier_block *unused,
	594	unsigned long event, void *ptr)
	595	{
	596	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	597	struct ip_tunnel *tunnel = netdev_priv(dev);
	598
	599	if (!is_vti_tunnel(dev))
	600	return NOTIFY_DONE;
	601
	602	switch (event) {
	603	case NETDEV_DOWN:
	604	if (!net_eq(tunnel->net, dev_net(dev)))
	605	xfrm_garbage_collect(tunnel->net);
	606	break;
	607	}
	608	return NOTIFY_DONE;
	609	}
	610
	611	static struct notifier_block vti_notifier_block __read_mostly = {
	612	.notifier_call = vti_device_event,
	613	};
	614
1181412c S	615	static int __init vti_init(void)
1181412c S	616	{
1990e4f8	617	const char *msg;
1181412c S	618	int err;
1181412c S	619
1990e4f8	620	pr_info("IPv4 over IPsec tunneling driver\n");
1181412c	621
a5d0dc81 LR	622	register_netdevice_notifier(&vti_notifier_block);
a5d0dc81 LR	623
1990e4f8	624	msg = "tunnel device";
1181412c S	625	err = register_pernet_device(&vti_net_ops);
1181412c S	626	if (err < 0)
1990e4f8	627	goto pernet_dev_failed;
df3893c1	628
1990e4f8 MK	629	msg = "tunnel protocols";
	630	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	631	if (err < 0)
	632	goto xfrm_proto_esp_failed;
df3893c1	633	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8 MK	634	if (err < 0)
1990e4f8 MK	635	goto xfrm_proto_ah_failed;
df3893c1	636	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8 MK	637	if (err < 0)
1990e4f8 MK	638	goto xfrm_proto_comp_failed;
1181412c	639
1990e4f8	640	msg = "netlink interface";
1181412c S	641	err = rtnl_link_register(&vti_link_ops);
	642	if (err < 0)
	643	goto rtnl_link_failed;
	644
	645	return err;
	646
	647	rtnl_link_failed:
df3893c1	648	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8	649	xfrm_proto_comp_failed:
df3893c1	650	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8	651	xfrm_proto_ah_failed:
df3893c1	652	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1990e4f8	653	xfrm_proto_esp_failed:
1181412c	654	unregister_pernet_device(&vti_net_ops);
1990e4f8	655	pernet_dev_failed:
a5d0dc81	656	unregister_netdevice_notifier(&vti_notifier_block);
1990e4f8	657	pr_err("vti init: failed to register %s\n", msg);
1181412c S	658	return err;
	659	}
	660
	661	static void __exit vti_fini(void)
	662	{
	663	rtnl_link_unregister(&vti_link_ops);
1990e4f8 MK	664	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	665	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	666	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c	667	unregister_pernet_device(&vti_net_ops);
a5d0dc81	668	unregister_netdevice_notifier(&vti_notifier_block);
1181412c S	669	}
	670
	671	module_init(vti_init);
	672	module_exit(vti_fini);
	673	MODULE_LICENSE("GPL");
	674	MODULE_ALIAS_RTNL_LINK("vti");
	675	MODULE_ALIAS_NETDEV("ip_vti0");