[GitHub/exynos8895/android_kernel_samsung_universal8895.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/mroute.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 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 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);
	family = x->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 err;

	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;
	}

	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 = skb->len;
	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
	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[])
{
	return 0;
}

static void vti_netlink_parms(struct nlattr *data[],
			      struct ip_tunnel_parm *parms)
{
	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]);

}

static int vti_newlink(struct net *src_net, struct net_device *dev,
		       struct nlattr *tb[], struct nlattr *data[])
{
	struct ip_tunnel_parm parms;

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

static int vti_changelink(struct net_device *dev, struct nlattr *tb[],
			  struct nlattr *data[])
{
	struct ip_tunnel_parm p;

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

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) +
		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;

	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);

	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) },
};

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>
	33	#include <linux/mroute.h>
	34	#include <linux/init.h>
	35	#include <linux/netfilter_ipv4.h>
	36	#include <linux/if_ether.h>
78a010cc	37	#include <linux/icmpv6.h>
1181412c S	38
	39	#include <net/sock.h>
	40	#include <net/ip.h>
	41	#include <net/icmp.h>
c5441932	42	#include <net/ip_tunnels.h>
1181412c S	43	#include <net/inet_ecn.h>
	44	#include <net/xfrm.h>
	45	#include <net/net_namespace.h>
	46	#include <net/netns/generic.h>
	47
1181412c S	48	static struct rtnl_link_ops vti_link_ops __read_mostly;
	49
	50	static int vti_net_id __read_mostly;
1181412c	51	static int vti_tunnel_init(struct net_device *dev);
1181412c	52
df3893c1 SK	53	static int vti_input(struct sk_buff *skb, int nexthdr, __be32 spi,
df3893c1 SK	54	int encap_type)
1181412c S	55	{
	56	struct ip_tunnel *tunnel;
	57	const struct iphdr *iph = ip_hdr(skb);
b9959fd3 AW	58	struct net *net = dev_net(skb->dev);
b9959fd3 AW	59	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
1181412c	60
b9959fd3 AW	61	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
b9959fd3 AW	62	iph->saddr, iph->daddr, 0);
00db4124	63	if (tunnel) {
df3893c1 SK	64	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
	65	goto drop;
	66
	67	XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = tunnel;
df3893c1 SK	68
	69	return xfrm_input(skb, nexthdr, spi, encap_type);
	70	}
	71
	72	return -EINVAL;
	73	drop:
	74	kfree_skb(skb);
	75	return 0;
	76	}
	77
	78	static int vti_rcv(struct sk_buff *skb)
	79	{
	80	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
	81	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
	82
	83	return vti_input(skb, ip_hdr(skb)->protocol, 0, 0);
	84	}
	85
	86	static int vti_rcv_cb(struct sk_buff *skb, int err)
	87	{
	88	unsigned short family;
	89	struct net_device *dev;
	90	struct pcpu_sw_netstats *tstats;
	91	struct xfrm_state *x;
	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 SK	108	x = xfrm_input_state(skb);
	109	family = x->inner_mode->afinfo->family;
	110
d55c670c AD	111	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	112	ret = xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family);
	113	skb->mark = orig_mark;
	114
	115	if (!ret)
df3893c1 SK	116	return -EPERM;
	117
	118	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	119	skb->dev = dev;
	120
	121	tstats = this_cpu_ptr(dev->tstats);
	122
	123	u64_stats_update_begin(&tstats->syncp);
	124	tstats->rx_packets++;
	125	tstats->rx_bytes += skb->len;
	126	u64_stats_update_end(&tstats->syncp);
	127
	128	return 0;
1181412c S	129	}
1181412c S	130
6e2de802 SK	131	static bool vti_state_check(const struct xfrm_state *x, __be32 dst, __be32 src)
	132	{
	133	xfrm_address_t daddr = (xfrm_address_t )&dst;
	134	xfrm_address_t saddr = (xfrm_address_t )&src;
	135
	136	/* if there is no transform then this tunnel is not functional.
	137	* Or if the xfrm is not mode tunnel.
	138	*/
	139	if (!x \|\| x->props.mode != XFRM_MODE_TUNNEL \|\|
	140	x->props.family != AF_INET)
	141	return false;
	142
	143	if (!dst)
	144	return xfrm_addr_equal(saddr, &x->props.saddr, AF_INET);
	145
	146	if (!xfrm_state_addr_check(x, daddr, saddr, AF_INET))
	147	return false;
	148
	149	return true;
	150	}
	151
78a010cc SK	152	static netdev_tx_t vti_xmit(struct sk_buff skb, struct net_device dev,
78a010cc SK	153	struct flowi *fl)
1181412c S	154	{
1181412c S	155	struct ip_tunnel *tunnel = netdev_priv(dev);
6e2de802	156	struct ip_tunnel_parm *parms = &tunnel->parms;
a34cd4f3	157	struct dst_entry *dst = skb_dst(skb);
1181412c	158	struct net_device tdev; / Device to other host */
b9959fd3	159	int err;
1181412c	160
a34cd4f3	161	if (!dst) {
df3893c1 SK	162	dev->stats.tx_carrier_errors++;
	163	goto tx_error_icmp;
	164	}
1181412c	165
a34cd4f3	166	dst_hold(dst);
78a010cc	167	dst = xfrm_lookup(tunnel->net, dst, fl, NULL, 0);
a34cd4f3	168	if (IS_ERR(dst)) {
1181412c S	169	dev->stats.tx_carrier_errors++;
	170	goto tx_error_icmp;
	171	}
df3893c1	172
6e2de802	173	if (!vti_state_check(dst->xfrm, parms->iph.daddr, parms->iph.saddr)) {
1181412c	174	dev->stats.tx_carrier_errors++;
a34cd4f3	175	dst_release(dst);
1181412c S	176	goto tx_error_icmp;
1181412c S	177	}
6e2de802	178
a34cd4f3	179	tdev = dst->dev;
1181412c S	180
1181412c S	181	if (tdev == dev) {
a34cd4f3	182	dst_release(dst);
1181412c S	183	dev->stats.collisions++;
	184	goto tx_error;
	185	}
	186
	187	if (tunnel->err_count > 0) {
	188	if (time_before(jiffies,
	189	tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
	190	tunnel->err_count--;
	191	dst_link_failure(skb);
	192	} else
	193	tunnel->err_count = 0;
	194	}
	195
df3893c1	196	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
a34cd4f3	197	skb_dst_set(skb, dst);
1181412c S	198	skb->dev = skb_dst(skb)->dev;
1181412c S	199
13206b6b	200	err = dst_output(tunnel->net, skb->sk, skb);
b9959fd3 AW	201	if (net_xmit_eval(err) == 0)
	202	err = skb->len;
	203	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
1181412c S	204	return NETDEV_TX_OK;
	205
	206	tx_error_icmp:
	207	dst_link_failure(skb);
	208	tx_error:
	209	dev->stats.tx_errors++;
3acfa1e7	210	kfree_skb(skb);
1181412c S	211	return NETDEV_TX_OK;
	212	}
	213
78a010cc SK	214	/* This function assumes it is being called from dev_queue_xmit()
	215	* and that skb is filled properly by that function.
	216	*/
	217	static netdev_tx_t vti_tunnel_xmit(struct sk_buff skb, struct net_device dev)
	218	{
	219	struct ip_tunnel *tunnel = netdev_priv(dev);
	220	struct flowi fl;
	221
	222	memset(&fl, 0, sizeof(fl));
	223
78a010cc SK	224	switch (skb->protocol) {
	225	case htons(ETH_P_IP):
	226	xfrm_decode_session(skb, &fl, AF_INET);
	227	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	228	break;
	229	case htons(ETH_P_IPV6):
	230	xfrm_decode_session(skb, &fl, AF_INET6);
	231	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
	232	break;
	233	default:
	234	dev->stats.tx_errors++;
	235	dev_kfree_skb(skb);
	236	return NETDEV_TX_OK;
	237	}
	238
cd5279c1 AD	239	/* override mark with tunnel output key */
	240	fl.flowi_mark = be32_to_cpu(tunnel->parms.o_key);
	241
78a010cc SK	242	return vti_xmit(skb, dev, &fl);
	243	}
	244
df3893c1 SK	245	static int vti4_err(struct sk_buff *skb, u32 info)
	246	{
	247	__be32 spi;
6d004d6c	248	__u32 mark;
df3893c1 SK	249	struct xfrm_state *x;
	250	struct ip_tunnel *tunnel;
	251	struct ip_esp_hdr *esph;
	252	struct ip_auth_hdr *ah ;
	253	struct ip_comp_hdr *ipch;
	254	struct net *net = dev_net(skb->dev);
	255	const struct iphdr iph = (const struct iphdr )skb->data;
	256	int protocol = iph->protocol;
	257	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	258
	259	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
	260	iph->daddr, iph->saddr, 0);
	261	if (!tunnel)
	262	return -1;
	263
6d004d6c SK	264	mark = be32_to_cpu(tunnel->parms.o_key);
6d004d6c SK	265
df3893c1 SK	266	switch (protocol) {
	267	case IPPROTO_ESP:
	268	esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
	269	spi = esph->spi;
	270	break;
	271	case IPPROTO_AH:
	272	ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
	273	spi = ah->spi;
	274	break;
	275	case IPPROTO_COMP:
	276	ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	277	spi = htonl(ntohs(ipch->cpi));
	278	break;
	279	default:
	280	return 0;
	281	}
	282
	283	switch (icmp_hdr(skb)->type) {
	284	case ICMP_DEST_UNREACH:
	285	if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
	286	return 0;
	287	case ICMP_REDIRECT:
	288	break;
	289	default:
	290	return 0;
	291	}
	292
6d004d6c	293	x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr,
df3893c1 SK	294	spi, protocol, AF_INET);
	295	if (!x)
	296	return 0;
	297
	298	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
	299	ipv4_update_pmtu(skb, net, info, 0, 0, protocol, 0);
	300	else
	301	ipv4_redirect(skb, net, 0, 0, protocol, 0);
	302	xfrm_state_put(x);
	303
	304	return 0;
	305	}
	306
1181412c S	307	static int
	308	vti_tunnel_ioctl(struct net_device dev, struct ifreq ifr, int cmd)
	309	{
	310	int err = 0;
	311	struct ip_tunnel_parm p;
1181412c	312
b9959fd3 AW	313	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
b9959fd3 AW	314	return -EFAULT;
1181412c	315
b9959fd3	316	if (cmd == SIOCADDTUNNEL \|\| cmd == SIOCCHGTUNNEL) {
1181412c S	317	if (p.iph.version != 4 \|\| p.iph.protocol != IPPROTO_IPIP \|\|
1181412c S	318	p.iph.ihl != 5)
b9959fd3 AW	319	return -EINVAL;
b9959fd3 AW	320	}
1181412c	321
7c8e6b9c DP	322	if (!(p.i_flags & GRE_KEY))
	323	p.i_key = 0;
	324	if (!(p.o_flags & GRE_KEY))
	325	p.o_key = 0;
	326
	327	p.i_flags = VTI_ISVTI;
	328
b9959fd3 AW	329	err = ip_tunnel_ioctl(dev, &p, cmd);
	330	if (err)
	331	return err;
1181412c	332
b9959fd3	333	if (cmd != SIOCDELTUNNEL) {
df3893c1	334	p.i_flags \|= GRE_KEY;
b9959fd3	335	p.o_flags \|= GRE_KEY;
1181412c S	336	}
1181412c S	337
b9959fd3 AW	338	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
b9959fd3 AW	339	return -EFAULT;
1181412c S	340	return 0;
	341	}
	342
	343	static const struct net_device_ops vti_netdev_ops = {
	344	.ndo_init = vti_tunnel_init,
b9959fd3	345	.ndo_uninit = ip_tunnel_uninit,
1181412c S	346	.ndo_start_xmit = vti_tunnel_xmit,
1181412c S	347	.ndo_do_ioctl = vti_tunnel_ioctl,
b9959fd3	348	.ndo_change_mtu = ip_tunnel_change_mtu,
f61dd388	349	.ndo_get_stats64 = ip_tunnel_get_stats64,
1e99584b	350	.ndo_get_iflink = ip_tunnel_get_iflink,
1181412c S	351	};
1181412c S	352
b9959fd3	353	static void vti_tunnel_setup(struct net_device *dev)
1181412c	354	{
b9959fd3	355	dev->netdev_ops = &vti_netdev_ops;
8d89dcdf	356	dev->type = ARPHRD_TUNNEL;
b9959fd3	357	ip_tunnel_setup(dev, vti_net_id);
1181412c S	358	}
1181412c S	359
b9959fd3	360	static int vti_tunnel_init(struct net_device *dev)
1181412c	361	{
b9959fd3 AW	362	struct ip_tunnel *tunnel = netdev_priv(dev);
	363	struct iphdr *iph = &tunnel->parms.iph;
	364
	365	memcpy(dev->dev_addr, &iph->saddr, 4);
	366	memcpy(dev->broadcast, &iph->daddr, 4);
1181412c	367
1181412c S	368	dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr);
	369	dev->mtu = ETH_DATA_LEN;
	370	dev->flags = IFF_NOARP;
1181412c	371	dev->addr_len = 4;
1181412c	372	dev->features \|= NETIF_F_LLTX;
02875878	373	netif_keep_dst(dev);
1181412c	374
b9959fd3	375	return ip_tunnel_init(dev);
1181412c S	376	}
1181412c S	377
b9959fd3	378	static void __net_init vti_fb_tunnel_init(struct net_device *dev)
1181412c S	379	{
	380	struct ip_tunnel *tunnel = netdev_priv(dev);
	381	struct iphdr *iph = &tunnel->parms.iph;
1181412c	382
1181412c S	383	iph->version = 4;
	384	iph->protocol = IPPROTO_IPIP;
	385	iph->ihl = 5;
1181412c S	386	}
1181412c S	387
df3893c1	388	static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
1181412c	389	.handler = vti_rcv,
df3893c1 SK	390	.input_handler = vti_input,
	391	.cb_handler = vti_rcv_cb,
	392	.err_handler = vti4_err,
	393	.priority = 100,
	394	};
	395
	396	static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
	397	.handler = vti_rcv,
	398	.input_handler = vti_input,
	399	.cb_handler = vti_rcv_cb,
	400	.err_handler = vti4_err,
	401	.priority = 100,
	402	};
	403
	404	static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
	405	.handler = vti_rcv,
	406	.input_handler = vti_input,
	407	.cb_handler = vti_rcv_cb,
	408	.err_handler = vti4_err,
	409	.priority = 100,
1181412c S	410	};
1181412c S	411
1181412c S	412	static int __net_init vti_init_net(struct net *net)
	413	{
	414	int err;
b9959fd3	415	struct ip_tunnel_net *itn;
1181412c	416
b9959fd3	417	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
1181412c	418	if (err)
b9959fd3 AW	419	return err;
	420	itn = net_generic(net, vti_net_id);
	421	vti_fb_tunnel_init(itn->fb_tunnel_dev);
1181412c	422	return 0;
1181412c S	423	}
	424
	425	static void __net_exit vti_exit_net(struct net *net)
	426	{
b9959fd3	427	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
6c742e71	428	ip_tunnel_delete_net(itn, &vti_link_ops);
1181412c S	429	}
	430
	431	static struct pernet_operations vti_net_ops = {
	432	.init = vti_init_net,
	433	.exit = vti_exit_net,
	434	.id = &vti_net_id,
b9959fd3	435	.size = sizeof(struct ip_tunnel_net),
1181412c S	436	};
	437
	438	static int vti_tunnel_validate(struct nlattr tb[], struct nlattr data[])
	439	{
	440	return 0;
	441	}
	442
	443	static void vti_netlink_parms(struct nlattr *data[],
	444	struct ip_tunnel_parm *parms)
	445	{
	446	memset(parms, 0, sizeof(*parms));
	447
	448	parms->iph.protocol = IPPROTO_IPIP;
	449
	450	if (!data)
	451	return;
	452
df3893c1 SK	453	parms->i_flags = VTI_ISVTI;
df3893c1 SK	454
1181412c S	455	if (data[IFLA_VTI_LINK])
	456	parms->link = nla_get_u32(data[IFLA_VTI_LINK]);
	457
	458	if (data[IFLA_VTI_IKEY])
	459	parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);
	460
	461	if (data[IFLA_VTI_OKEY])
	462	parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);
	463
	464	if (data[IFLA_VTI_LOCAL])
67b61f6c	465	parms->iph.saddr = nla_get_in_addr(data[IFLA_VTI_LOCAL]);
1181412c S	466
1181412c S	467	if (data[IFLA_VTI_REMOTE])
67b61f6c	468	parms->iph.daddr = nla_get_in_addr(data[IFLA_VTI_REMOTE]);
1181412c S	469
	470	}
	471
	472	static int vti_newlink(struct net src_net, struct net_device dev,
	473	struct nlattr tb[], struct nlattr data[])
	474	{
b9959fd3	475	struct ip_tunnel_parm parms;
1181412c	476
b9959fd3 AW	477	vti_netlink_parms(data, &parms);
b9959fd3 AW	478	return ip_tunnel_newlink(dev, tb, &parms);
1181412c S	479	}
	480
	481	static int vti_changelink(struct net_device dev, struct nlattr tb[],
	482	struct nlattr *data[])
	483	{
1181412c	484	struct ip_tunnel_parm p;
1181412c	485
1181412c	486	vti_netlink_parms(data, &p);
b9959fd3	487	return ip_tunnel_changelink(dev, tb, &p);
1181412c S	488	}
	489
	490	static size_t vti_get_size(const struct net_device *dev)
	491	{
	492	return
	493	/* IFLA_VTI_LINK */
	494	nla_total_size(4) +
	495	/* IFLA_VTI_IKEY */
	496	nla_total_size(4) +
	497	/* IFLA_VTI_OKEY */
	498	nla_total_size(4) +
	499	/* IFLA_VTI_LOCAL */
	500	nla_total_size(4) +
	501	/* IFLA_VTI_REMOTE */
	502	nla_total_size(4) +
	503	0;
	504	}
	505
	506	static int vti_fill_info(struct sk_buff skb, const struct net_device dev)
	507	{
	508	struct ip_tunnel *t = netdev_priv(dev);
	509	struct ip_tunnel_parm *p = &t->parms;
	510
	511	nla_put_u32(skb, IFLA_VTI_LINK, p->link);
	512	nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key);
	513	nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key);
930345ea JB	514	nla_put_in_addr(skb, IFLA_VTI_LOCAL, p->iph.saddr);
930345ea JB	515	nla_put_in_addr(skb, IFLA_VTI_REMOTE, p->iph.daddr);
1181412c S	516
	517	return 0;
	518	}
	519
	520	static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	521	[IFLA_VTI_LINK] = { .type = NLA_U32 },
	522	[IFLA_VTI_IKEY] = { .type = NLA_U32 },
	523	[IFLA_VTI_OKEY] = { .type = NLA_U32 },
	524	[IFLA_VTI_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	525	[IFLA_VTI_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
	526	};
	527
	528	static struct rtnl_link_ops vti_link_ops __read_mostly = {
	529	.kind = "vti",
	530	.maxtype = IFLA_VTI_MAX,
	531	.policy = vti_policy,
	532	.priv_size = sizeof(struct ip_tunnel),
	533	.setup = vti_tunnel_setup,
	534	.validate = vti_tunnel_validate,
	535	.newlink = vti_newlink,
	536	.changelink = vti_changelink,
20ea60ca	537	.dellink = ip_tunnel_dellink,
1181412c S	538	.get_size = vti_get_size,
1181412c S	539	.fill_info = vti_fill_info,
1728d4fa	540	.get_link_net = ip_tunnel_get_link_net,
1181412c S	541	};
1181412c S	542
0bb225a0 LR	543	static bool is_vti_tunnel(const struct net_device *dev)
	544	{
	545	return dev->netdev_ops == &vti_netdev_ops;
	546	}
	547
	548	static int vti_device_event(struct notifier_block *unused,
	549	unsigned long event, void *ptr)
	550	{
	551	struct net_device *dev = netdev_notifier_info_to_dev(ptr);
	552	struct ip_tunnel *tunnel = netdev_priv(dev);
	553
	554	if (!is_vti_tunnel(dev))
	555	return NOTIFY_DONE;
	556
	557	switch (event) {
	558	case NETDEV_DOWN:
	559	if (!net_eq(tunnel->net, dev_net(dev)))
	560	xfrm_garbage_collect(tunnel->net);
	561	break;
	562	}
	563	return NOTIFY_DONE;
	564	}
	565
	566	static struct notifier_block vti_notifier_block __read_mostly = {
	567	.notifier_call = vti_device_event,
	568	};
	569
1181412c S	570	static int __init vti_init(void)
1181412c S	571	{
1990e4f8	572	const char *msg;
1181412c S	573	int err;
1181412c S	574
1990e4f8	575	pr_info("IPv4 over IPsec tunneling driver\n");
1181412c	576
0bb225a0 LR	577	register_netdevice_notifier(&vti_notifier_block);
0bb225a0 LR	578
1990e4f8	579	msg = "tunnel device";
1181412c S	580	err = register_pernet_device(&vti_net_ops);
1181412c S	581	if (err < 0)
1990e4f8	582	goto pernet_dev_failed;
df3893c1	583
1990e4f8 MK	584	msg = "tunnel protocols";
	585	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	586	if (err < 0)
	587	goto xfrm_proto_esp_failed;
df3893c1	588	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8 MK	589	if (err < 0)
1990e4f8 MK	590	goto xfrm_proto_ah_failed;
df3893c1	591	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8 MK	592	if (err < 0)
1990e4f8 MK	593	goto xfrm_proto_comp_failed;
1181412c	594
1990e4f8	595	msg = "netlink interface";
1181412c S	596	err = rtnl_link_register(&vti_link_ops);
	597	if (err < 0)
	598	goto rtnl_link_failed;
	599
	600	return err;
	601
	602	rtnl_link_failed:
df3893c1	603	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8	604	xfrm_proto_comp_failed:
df3893c1	605	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8	606	xfrm_proto_ah_failed:
df3893c1	607	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1990e4f8	608	xfrm_proto_esp_failed:
1181412c	609	unregister_pernet_device(&vti_net_ops);
1990e4f8	610	pernet_dev_failed:
0bb225a0	611	unregister_netdevice_notifier(&vti_notifier_block);
1990e4f8	612	pr_err("vti init: failed to register %s\n", msg);
1181412c S	613	return err;
	614	}
	615
	616	static void __exit vti_fini(void)
	617	{
	618	rtnl_link_unregister(&vti_link_ops);
1990e4f8 MK	619	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	620	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	621	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c	622	unregister_pernet_device(&vti_net_ops);
0bb225a0	623	unregister_netdevice_notifier(&vti_notifier_block);
1181412c S	624	}
	625
	626	module_init(vti_init);
	627	module_exit(vti_fini);
	628	MODULE_LICENSE("GPL");
	629	MODULE_ALIAS_RTNL_LINK("vti");
	630	MODULE_ALIAS_NETDEV("ip_vti0");