[linux-2.6-block.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 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;
	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 = skb->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[])
{
	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 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 = "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:
	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);
}

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;
	48
	49	static 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;
	91	struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4;
d55c670c AD	92	u32 orig_mark = skb->mark;
d55c670c AD	93	int ret;
df3893c1 SK	94
df3893c1 SK	95	if (!tunnel)
1181412c	96	return 1;
df3893c1 SK	97
	98	dev = tunnel->dev;
	99
	100	if (err) {
	101	dev->stats.rx_errors++;
	102	dev->stats.rx_dropped++;
	103
	104	return 0;
1181412c	105	}
1181412c	106
df3893c1 SK	107	x = xfrm_input_state(skb);
	108	family = x->inner_mode->afinfo->family;
	109
d55c670c AD	110	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	111	ret = xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family);
	112	skb->mark = orig_mark;
	113
	114	if (!ret)
df3893c1 SK	115	return -EPERM;
	116
	117	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	118	skb->dev = dev;
	119
	120	tstats = this_cpu_ptr(dev->tstats);
	121
	122	u64_stats_update_begin(&tstats->syncp);
	123	tstats->rx_packets++;
	124	tstats->rx_bytes += skb->len;
	125	u64_stats_update_end(&tstats->syncp);
	126
	127	return 0;
1181412c S	128	}
1181412c S	129
6e2de802 SK	130	static bool vti_state_check(const struct xfrm_state *x, __be32 dst, __be32 src)
	131	{
	132	xfrm_address_t daddr = (xfrm_address_t )&dst;
	133	xfrm_address_t saddr = (xfrm_address_t )&src;
	134
	135	/* if there is no transform then this tunnel is not functional.
	136	* Or if the xfrm is not mode tunnel.
	137	*/
	138	if (!x \|\| x->props.mode != XFRM_MODE_TUNNEL \|\|
	139	x->props.family != AF_INET)
	140	return false;
	141
	142	if (!dst)
	143	return xfrm_addr_equal(saddr, &x->props.saddr, AF_INET);
	144
	145	if (!xfrm_state_addr_check(x, daddr, saddr, AF_INET))
	146	return false;
	147
	148	return true;
	149	}
	150
78a010cc SK	151	static netdev_tx_t vti_xmit(struct sk_buff skb, struct net_device dev,
78a010cc SK	152	struct flowi *fl)
1181412c S	153	{
1181412c S	154	struct ip_tunnel *tunnel = netdev_priv(dev);
6e2de802	155	struct ip_tunnel_parm *parms = &tunnel->parms;
a34cd4f3	156	struct dst_entry *dst = skb_dst(skb);
1181412c	157	struct net_device tdev; / Device to other host */
b9959fd3	158	int err;
d6af1a31	159	int mtu;
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
d6af1a31 SK	196	mtu = dst_mtu(dst);
	197	if (skb->len > mtu) {
	198	skb_dst(skb)->ops->update_pmtu(skb_dst(skb), NULL, skb, mtu);
	199	if (skb->protocol == htons(ETH_P_IP)) {
	200	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
	201	htonl(mtu));
	202	} else {
	203	if (mtu < IPV6_MIN_MTU)
	204	mtu = IPV6_MIN_MTU;
	205
	206	icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
	207	}
	208
	209	dst_release(dst);
	210	goto tx_error;
	211	}
	212
df3893c1	213	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
a34cd4f3	214	skb_dst_set(skb, dst);
1181412c S	215	skb->dev = skb_dst(skb)->dev;
1181412c S	216
13206b6b	217	err = dst_output(tunnel->net, skb->sk, skb);
b9959fd3 AW	218	if (net_xmit_eval(err) == 0)
b9959fd3 AW	219	err = skb->len;
039f5062	220	iptunnel_xmit_stats(dev, err);
1181412c S	221	return NETDEV_TX_OK;
	222
	223	tx_error_icmp:
	224	dst_link_failure(skb);
	225	tx_error:
	226	dev->stats.tx_errors++;
3acfa1e7	227	kfree_skb(skb);
1181412c S	228	return NETDEV_TX_OK;
	229	}
	230
78a010cc SK	231	/* This function assumes it is being called from dev_queue_xmit()
	232	* and that skb is filled properly by that function.
	233	*/
	234	static netdev_tx_t vti_tunnel_xmit(struct sk_buff skb, struct net_device dev)
	235	{
	236	struct ip_tunnel *tunnel = netdev_priv(dev);
	237	struct flowi fl;
	238
	239	memset(&fl, 0, sizeof(fl));
	240
78a010cc SK	241	switch (skb->protocol) {
	242	case htons(ETH_P_IP):
	243	xfrm_decode_session(skb, &fl, AF_INET);
	244	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	245	break;
	246	case htons(ETH_P_IPV6):
	247	xfrm_decode_session(skb, &fl, AF_INET6);
	248	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
	249	break;
	250	default:
	251	dev->stats.tx_errors++;
	252	dev_kfree_skb(skb);
	253	return NETDEV_TX_OK;
	254	}
	255
cd5279c1 AD	256	/* override mark with tunnel output key */
	257	fl.flowi_mark = be32_to_cpu(tunnel->parms.o_key);
	258
78a010cc SK	259	return vti_xmit(skb, dev, &fl);
	260	}
	261
df3893c1 SK	262	static int vti4_err(struct sk_buff *skb, u32 info)
	263	{
	264	__be32 spi;
6d004d6c	265	__u32 mark;
df3893c1 SK	266	struct xfrm_state *x;
	267	struct ip_tunnel *tunnel;
	268	struct ip_esp_hdr *esph;
	269	struct ip_auth_hdr *ah ;
	270	struct ip_comp_hdr *ipch;
	271	struct net *net = dev_net(skb->dev);
	272	const struct iphdr iph = (const struct iphdr )skb->data;
	273	int protocol = iph->protocol;
	274	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	275
	276	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
	277	iph->daddr, iph->saddr, 0);
	278	if (!tunnel)
	279	return -1;
	280
6d004d6c SK	281	mark = be32_to_cpu(tunnel->parms.o_key);
6d004d6c SK	282
df3893c1 SK	283	switch (protocol) {
	284	case IPPROTO_ESP:
	285	esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
	286	spi = esph->spi;
	287	break;
	288	case IPPROTO_AH:
	289	ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
	290	spi = ah->spi;
	291	break;
	292	case IPPROTO_COMP:
	293	ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	294	spi = htonl(ntohs(ipch->cpi));
	295	break;
	296	default:
	297	return 0;
	298	}
	299
	300	switch (icmp_hdr(skb)->type) {
	301	case ICMP_DEST_UNREACH:
	302	if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
	303	return 0;
	304	case ICMP_REDIRECT:
	305	break;
	306	default:
	307	return 0;
	308	}
	309
6d004d6c	310	x = xfrm_state_lookup(net, mark, (const xfrm_address_t *)&iph->daddr,
df3893c1 SK	311	spi, protocol, AF_INET);
	312	if (!x)
	313	return 0;
	314
	315	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
	316	ipv4_update_pmtu(skb, net, info, 0, 0, protocol, 0);
	317	else
	318	ipv4_redirect(skb, net, 0, 0, protocol, 0);
	319	xfrm_state_put(x);
	320
	321	return 0;
	322	}
	323
1181412c S	324	static int
	325	vti_tunnel_ioctl(struct net_device dev, struct ifreq ifr, int cmd)
	326	{
	327	int err = 0;
	328	struct ip_tunnel_parm p;
1181412c	329
b9959fd3 AW	330	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
b9959fd3 AW	331	return -EFAULT;
1181412c	332
b9959fd3	333	if (cmd == SIOCADDTUNNEL \|\| cmd == SIOCCHGTUNNEL) {
1181412c S	334	if (p.iph.version != 4 \|\| p.iph.protocol != IPPROTO_IPIP \|\|
1181412c S	335	p.iph.ihl != 5)
b9959fd3 AW	336	return -EINVAL;
b9959fd3 AW	337	}
1181412c	338
7c8e6b9c DP	339	if (!(p.i_flags & GRE_KEY))
	340	p.i_key = 0;
	341	if (!(p.o_flags & GRE_KEY))
	342	p.o_key = 0;
	343
	344	p.i_flags = VTI_ISVTI;
	345
b9959fd3 AW	346	err = ip_tunnel_ioctl(dev, &p, cmd);
	347	if (err)
	348	return err;
1181412c	349
b9959fd3	350	if (cmd != SIOCDELTUNNEL) {
df3893c1	351	p.i_flags \|= GRE_KEY;
b9959fd3	352	p.o_flags \|= GRE_KEY;
1181412c S	353	}
1181412c S	354
b9959fd3 AW	355	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
b9959fd3 AW	356	return -EFAULT;
1181412c S	357	return 0;
	358	}
	359
	360	static const struct net_device_ops vti_netdev_ops = {
	361	.ndo_init = vti_tunnel_init,
b9959fd3	362	.ndo_uninit = ip_tunnel_uninit,
1181412c S	363	.ndo_start_xmit = vti_tunnel_xmit,
1181412c S	364	.ndo_do_ioctl = vti_tunnel_ioctl,
b9959fd3	365	.ndo_change_mtu = ip_tunnel_change_mtu,
f61dd388	366	.ndo_get_stats64 = ip_tunnel_get_stats64,
1e99584b	367	.ndo_get_iflink = ip_tunnel_get_iflink,
1181412c S	368	};
1181412c S	369
b9959fd3	370	static void vti_tunnel_setup(struct net_device *dev)
1181412c	371	{
b9959fd3	372	dev->netdev_ops = &vti_netdev_ops;
8d89dcdf	373	dev->type = ARPHRD_TUNNEL;
b9959fd3	374	ip_tunnel_setup(dev, vti_net_id);
1181412c S	375	}
1181412c S	376
b9959fd3	377	static int vti_tunnel_init(struct net_device *dev)
1181412c	378	{
b9959fd3 AW	379	struct ip_tunnel *tunnel = netdev_priv(dev);
	380	struct iphdr *iph = &tunnel->parms.iph;
	381
	382	memcpy(dev->dev_addr, &iph->saddr, 4);
	383	memcpy(dev->broadcast, &iph->daddr, 4);
1181412c	384
1181412c S	385	dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr);
	386	dev->mtu = ETH_DATA_LEN;
	387	dev->flags = IFF_NOARP;
1181412c	388	dev->addr_len = 4;
1181412c	389	dev->features \|= NETIF_F_LLTX;
02875878	390	netif_keep_dst(dev);
1181412c	391
b9959fd3	392	return ip_tunnel_init(dev);
1181412c S	393	}
1181412c S	394
b9959fd3	395	static void __net_init vti_fb_tunnel_init(struct net_device *dev)
1181412c S	396	{
	397	struct ip_tunnel *tunnel = netdev_priv(dev);
	398	struct iphdr *iph = &tunnel->parms.iph;
1181412c	399
1181412c S	400	iph->version = 4;
	401	iph->protocol = IPPROTO_IPIP;
	402	iph->ihl = 5;
1181412c S	403	}
1181412c S	404
df3893c1	405	static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
1181412c	406	.handler = vti_rcv,
df3893c1 SK	407	.input_handler = vti_input,
	408	.cb_handler = vti_rcv_cb,
	409	.err_handler = vti4_err,
	410	.priority = 100,
	411	};
	412
	413	static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
	414	.handler = vti_rcv,
	415	.input_handler = vti_input,
	416	.cb_handler = vti_rcv_cb,
	417	.err_handler = vti4_err,
	418	.priority = 100,
	419	};
	420
	421	static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
	422	.handler = vti_rcv,
	423	.input_handler = vti_input,
	424	.cb_handler = vti_rcv_cb,
	425	.err_handler = vti4_err,
	426	.priority = 100,
1181412c S	427	};
1181412c S	428
1181412c S	429	static int __net_init vti_init_net(struct net *net)
	430	{
	431	int err;
b9959fd3	432	struct ip_tunnel_net *itn;
1181412c	433
b9959fd3	434	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
1181412c	435	if (err)
b9959fd3 AW	436	return err;
	437	itn = net_generic(net, vti_net_id);
	438	vti_fb_tunnel_init(itn->fb_tunnel_dev);
1181412c	439	return 0;
1181412c S	440	}
	441
	442	static void __net_exit vti_exit_net(struct net *net)
	443	{
b9959fd3	444	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
6c742e71	445	ip_tunnel_delete_net(itn, &vti_link_ops);
1181412c S	446	}
	447
	448	static struct pernet_operations vti_net_ops = {
	449	.init = vti_init_net,
	450	.exit = vti_exit_net,
	451	.id = &vti_net_id,
b9959fd3	452	.size = sizeof(struct ip_tunnel_net),
1181412c S	453	};
	454
	455	static int vti_tunnel_validate(struct nlattr tb[], struct nlattr data[])
	456	{
	457	return 0;
	458	}
	459
	460	static void vti_netlink_parms(struct nlattr *data[],
	461	struct ip_tunnel_parm *parms)
	462	{
	463	memset(parms, 0, sizeof(*parms));
	464
	465	parms->iph.protocol = IPPROTO_IPIP;
	466
	467	if (!data)
	468	return;
	469
df3893c1 SK	470	parms->i_flags = VTI_ISVTI;
df3893c1 SK	471
1181412c S	472	if (data[IFLA_VTI_LINK])
	473	parms->link = nla_get_u32(data[IFLA_VTI_LINK]);
	474
	475	if (data[IFLA_VTI_IKEY])
	476	parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);
	477
	478	if (data[IFLA_VTI_OKEY])
	479	parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);
	480
	481	if (data[IFLA_VTI_LOCAL])
67b61f6c	482	parms->iph.saddr = nla_get_in_addr(data[IFLA_VTI_LOCAL]);
1181412c S	483
1181412c S	484	if (data[IFLA_VTI_REMOTE])
67b61f6c	485	parms->iph.daddr = nla_get_in_addr(data[IFLA_VTI_REMOTE]);
1181412c S	486
	487	}
	488
	489	static int vti_newlink(struct net src_net, struct net_device dev,
	490	struct nlattr tb[], struct nlattr data[])
	491	{
b9959fd3	492	struct ip_tunnel_parm parms;
1181412c	493
b9959fd3 AW	494	vti_netlink_parms(data, &parms);
b9959fd3 AW	495	return ip_tunnel_newlink(dev, tb, &parms);
1181412c S	496	}
	497
	498	static int vti_changelink(struct net_device dev, struct nlattr tb[],
	499	struct nlattr *data[])
	500	{
1181412c	501	struct ip_tunnel_parm p;
1181412c	502
1181412c	503	vti_netlink_parms(data, &p);
b9959fd3	504	return ip_tunnel_changelink(dev, tb, &p);
1181412c S	505	}
	506
	507	static size_t vti_get_size(const struct net_device *dev)
	508	{
	509	return
	510	/* IFLA_VTI_LINK */
	511	nla_total_size(4) +
	512	/* IFLA_VTI_IKEY */
	513	nla_total_size(4) +
	514	/* IFLA_VTI_OKEY */
	515	nla_total_size(4) +
	516	/* IFLA_VTI_LOCAL */
	517	nla_total_size(4) +
	518	/* IFLA_VTI_REMOTE */
	519	nla_total_size(4) +
	520	0;
	521	}
	522
	523	static int vti_fill_info(struct sk_buff skb, const struct net_device dev)
	524	{
	525	struct ip_tunnel *t = netdev_priv(dev);
	526	struct ip_tunnel_parm *p = &t->parms;
	527
	528	nla_put_u32(skb, IFLA_VTI_LINK, p->link);
	529	nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key);
	530	nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key);
930345ea JB	531	nla_put_in_addr(skb, IFLA_VTI_LOCAL, p->iph.saddr);
930345ea JB	532	nla_put_in_addr(skb, IFLA_VTI_REMOTE, p->iph.daddr);
1181412c S	533
	534	return 0;
	535	}
	536
	537	static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	538	[IFLA_VTI_LINK] = { .type = NLA_U32 },
	539	[IFLA_VTI_IKEY] = { .type = NLA_U32 },
	540	[IFLA_VTI_OKEY] = { .type = NLA_U32 },
	541	[IFLA_VTI_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	542	[IFLA_VTI_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
	543	};
	544
	545	static struct rtnl_link_ops vti_link_ops __read_mostly = {
	546	.kind = "vti",
	547	.maxtype = IFLA_VTI_MAX,
	548	.policy = vti_policy,
	549	.priv_size = sizeof(struct ip_tunnel),
	550	.setup = vti_tunnel_setup,
	551	.validate = vti_tunnel_validate,
	552	.newlink = vti_newlink,
	553	.changelink = vti_changelink,
20ea60ca	554	.dellink = ip_tunnel_dellink,
1181412c S	555	.get_size = vti_get_size,
1181412c S	556	.fill_info = vti_fill_info,
1728d4fa	557	.get_link_net = ip_tunnel_get_link_net,
1181412c S	558	};
	559
	560	static int __init vti_init(void)
	561	{
1990e4f8	562	const char *msg;
1181412c S	563	int err;
1181412c S	564
1990e4f8	565	pr_info("IPv4 over IPsec tunneling driver\n");
1181412c	566
1990e4f8	567	msg = "tunnel device";
1181412c S	568	err = register_pernet_device(&vti_net_ops);
1181412c S	569	if (err < 0)
1990e4f8	570	goto pernet_dev_failed;
df3893c1	571
1990e4f8 MK	572	msg = "tunnel protocols";
	573	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	574	if (err < 0)
	575	goto xfrm_proto_esp_failed;
df3893c1	576	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8 MK	577	if (err < 0)
1990e4f8 MK	578	goto xfrm_proto_ah_failed;
df3893c1	579	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8 MK	580	if (err < 0)
1990e4f8 MK	581	goto xfrm_proto_comp_failed;
1181412c	582
1990e4f8	583	msg = "netlink interface";
1181412c S	584	err = rtnl_link_register(&vti_link_ops);
	585	if (err < 0)
	586	goto rtnl_link_failed;
	587
	588	return err;
	589
	590	rtnl_link_failed:
df3893c1	591	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
1990e4f8	592	xfrm_proto_comp_failed:
df3893c1	593	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
1990e4f8	594	xfrm_proto_ah_failed:
df3893c1	595	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1990e4f8	596	xfrm_proto_esp_failed:
1181412c	597	unregister_pernet_device(&vti_net_ops);
1990e4f8 MK	598	pernet_dev_failed:
1990e4f8 MK	599	pr_err("vti init: failed to register %s\n", msg);
1181412c S	600	return err;
	601	}
	602
	603	static void __exit vti_fini(void)
	604	{
	605	rtnl_link_unregister(&vti_link_ops);
1990e4f8 MK	606	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	607	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	608	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c S	609	unregister_pernet_device(&vti_net_ops);
	610	}
	611
	612	module_init(vti_init);
	613	module_exit(vti_fini);
	614	MODULE_LICENSE("GPL");
	615	MODULE_ALIAS_RTNL_LINK("vti");
	616	MODULE_ALIAS_NETDEV("ip_vti0");