[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/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 != NULL) {
		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
			goto drop;

		XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4 = tunnel;
		skb->mark = be32_to_cpu(tunnel->parms.i_key);

		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;

	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;

	if (!xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family))
		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(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));

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

	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;
	}

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

static int vti4_err(struct sk_buff *skb, u32 info)
{
	__be32 spi;
	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;

	switch (protocol) {
	case IPPROTO_ESP:
		esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
		spi = esph->spi;
		break;
	case IPPROTO_AH:
		ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
		spi = ah->spi;
		break;
	case IPPROTO_COMP:
		ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
		spi = htonl(ntohs(ipch->cpi));
		break;
	default:
		return 0;
	}

	switch (icmp_hdr(skb)->type) {
	case ICMP_DEST_UNREACH:
		if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
			return 0;
	case ICMP_REDIRECT:
		break;
	default:
		return 0;
	}

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

	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;
	}

	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,
};

static void vti_tunnel_setup(struct net_device *dev)
{
	dev->netdev_ops		= &vti_netdev_ops;
	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->type		= ARPHRD_TUNNEL;
	dev->mtu		= ETH_DATA_LEN;
	dev->flags		= IFF_NOARP;
	dev->iflink		= 0;
	dev->addr_len		= 4;
	dev->features		|= NETIF_F_LLTX;
	dev->priv_flags		&= ~IFF_XMIT_DST_RELEASE;

	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_be32(data[IFLA_VTI_LOCAL]);

	if (data[IFLA_VTI_REMOTE])
		parms->iph.daddr = nla_get_be32(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_be32(skb, IFLA_VTI_LOCAL, p->iph.saddr);
	nla_put_be32(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,
	.get_size	= vti_get_size,
	.fill_info	= vti_fill_info,
};

static int __init vti_init(void)
{
	int err;

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

	err = register_pernet_device(&vti_net_ops);
	if (err < 0)
		return err;
	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	if (err < 0) {
		unregister_pernet_device(&vti_net_ops);
		pr_info("vti init: can't register tunnel\n");

		return err;
	}

	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
	if (err < 0) {
		xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
		unregister_pernet_device(&vti_net_ops);
		pr_info("vti init: can't register tunnel\n");

		return err;
	}

	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
	if (err < 0) {
		xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
		xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
		unregister_pernet_device(&vti_net_ops);
		pr_info("vti init: can't register tunnel\n");

		return err;
	}

	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);
	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
	unregister_pernet_device(&vti_net_ops);
	return err;
}

static void __exit vti_fini(void)
{
	rtnl_link_unregister(&vti_link_ops);
	if (xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP))
		pr_info("vti close: can't deregister tunnel\n");
	if (xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH))
		pr_info("vti close: can't deregister tunnel\n");
	if (xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP))
		pr_info("vti close: can't deregister tunnel\n");


	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>
	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);
1181412c	63	if (tunnel != NULL) {
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;
	68	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	69
	70	return xfrm_input(skb, nexthdr, spi, encap_type);
	71	}
	72
	73	return -EINVAL;
	74	drop:
	75	kfree_skb(skb);
	76	return 0;
	77	}
	78
	79	static int vti_rcv(struct sk_buff *skb)
	80	{
	81	XFRM_SPI_SKB_CB(skb)->family = AF_INET;
	82	XFRM_SPI_SKB_CB(skb)->daddroff = offsetof(struct iphdr, daddr);
	83
	84	return vti_input(skb, ip_hdr(skb)->protocol, 0, 0);
	85	}
	86
	87	static int vti_rcv_cb(struct sk_buff *skb, int err)
	88	{
	89	unsigned short family;
	90	struct net_device *dev;
	91	struct pcpu_sw_netstats *tstats;
	92	struct xfrm_state *x;
	93	struct ip_tunnel *tunnel = XFRM_TUNNEL_SKB_CB(skb)->tunnel.ip4;
	94
	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
	110	if (!xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family))
	111	return -EPERM;
	112
	113	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	114	skb->dev = dev;
	115
	116	tstats = this_cpu_ptr(dev->tstats);
	117
	118	u64_stats_update_begin(&tstats->syncp);
	119	tstats->rx_packets++;
	120	tstats->rx_bytes += skb->len;
	121	u64_stats_update_end(&tstats->syncp);
	122
	123	return 0;
1181412c S	124	}
1181412c S	125
6e2de802 SK	126	static bool vti_state_check(const struct xfrm_state *x, __be32 dst, __be32 src)
	127	{
	128	xfrm_address_t daddr = (xfrm_address_t )&dst;
	129	xfrm_address_t saddr = (xfrm_address_t )&src;
	130
	131	/* if there is no transform then this tunnel is not functional.
	132	* Or if the xfrm is not mode tunnel.
	133	*/
	134	if (!x \|\| x->props.mode != XFRM_MODE_TUNNEL \|\|
	135	x->props.family != AF_INET)
	136	return false;
	137
	138	if (!dst)
	139	return xfrm_addr_equal(saddr, &x->props.saddr, AF_INET);
	140
	141	if (!xfrm_state_addr_check(x, daddr, saddr, AF_INET))
	142	return false;
	143
	144	return true;
	145	}
	146
78a010cc SK	147	static netdev_tx_t vti_xmit(struct sk_buff skb, struct net_device dev,
78a010cc SK	148	struct flowi *fl)
1181412c S	149	{
1181412c S	150	struct ip_tunnel *tunnel = netdev_priv(dev);
6e2de802	151	struct ip_tunnel_parm *parms = &tunnel->parms;
a34cd4f3	152	struct dst_entry *dst = skb_dst(skb);
1181412c	153	struct net_device tdev; / Device to other host */
b9959fd3	154	int err;
1181412c	155
a34cd4f3	156	if (!dst) {
df3893c1 SK	157	dev->stats.tx_carrier_errors++;
	158	goto tx_error_icmp;
	159	}
1181412c	160
a34cd4f3	161	dst_hold(dst);
78a010cc	162	dst = xfrm_lookup(tunnel->net, dst, fl, NULL, 0);
a34cd4f3	163	if (IS_ERR(dst)) {
1181412c S	164	dev->stats.tx_carrier_errors++;
	165	goto tx_error_icmp;
	166	}
df3893c1	167
6e2de802	168	if (!vti_state_check(dst->xfrm, parms->iph.daddr, parms->iph.saddr)) {
1181412c	169	dev->stats.tx_carrier_errors++;
a34cd4f3	170	dst_release(dst);
1181412c S	171	goto tx_error_icmp;
1181412c S	172	}
6e2de802	173
a34cd4f3	174	tdev = dst->dev;
1181412c S	175
1181412c S	176	if (tdev == dev) {
a34cd4f3	177	dst_release(dst);
1181412c S	178	dev->stats.collisions++;
	179	goto tx_error;
	180	}
	181
	182	if (tunnel->err_count > 0) {
	183	if (time_before(jiffies,
	184	tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
	185	tunnel->err_count--;
	186	dst_link_failure(skb);
	187	} else
	188	tunnel->err_count = 0;
	189	}
	190
df3893c1	191	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
a34cd4f3	192	skb_dst_set(skb, dst);
1181412c S	193	skb->dev = skb_dst(skb)->dev;
1181412c S	194
b9959fd3 AW	195	err = dst_output(skb);
	196	if (net_xmit_eval(err) == 0)
	197	err = skb->len;
	198	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
1181412c S	199	return NETDEV_TX_OK;
	200
	201	tx_error_icmp:
	202	dst_link_failure(skb);
	203	tx_error:
	204	dev->stats.tx_errors++;
3acfa1e7	205	kfree_skb(skb);
1181412c S	206	return NETDEV_TX_OK;
	207	}
	208
78a010cc SK	209	/* This function assumes it is being called from dev_queue_xmit()
	210	* and that skb is filled properly by that function.
	211	*/
	212	static netdev_tx_t vti_tunnel_xmit(struct sk_buff skb, struct net_device dev)
	213	{
	214	struct ip_tunnel *tunnel = netdev_priv(dev);
	215	struct flowi fl;
	216
	217	memset(&fl, 0, sizeof(fl));
	218
	219	skb->mark = be32_to_cpu(tunnel->parms.o_key);
	220
	221	switch (skb->protocol) {
	222	case htons(ETH_P_IP):
	223	xfrm_decode_session(skb, &fl, AF_INET);
	224	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	225	break;
	226	case htons(ETH_P_IPV6):
	227	xfrm_decode_session(skb, &fl, AF_INET6);
	228	memset(IP6CB(skb), 0, sizeof(*IP6CB(skb)));
	229	break;
	230	default:
	231	dev->stats.tx_errors++;
	232	dev_kfree_skb(skb);
	233	return NETDEV_TX_OK;
	234	}
	235
	236	return vti_xmit(skb, dev, &fl);
	237	}
	238
df3893c1 SK	239	static int vti4_err(struct sk_buff *skb, u32 info)
	240	{
	241	__be32 spi;
	242	struct xfrm_state *x;
	243	struct ip_tunnel *tunnel;
	244	struct ip_esp_hdr *esph;
	245	struct ip_auth_hdr *ah ;
	246	struct ip_comp_hdr *ipch;
	247	struct net *net = dev_net(skb->dev);
	248	const struct iphdr iph = (const struct iphdr )skb->data;
	249	int protocol = iph->protocol;
	250	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	251
	252	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
	253	iph->daddr, iph->saddr, 0);
	254	if (!tunnel)
	255	return -1;
	256
	257	switch (protocol) {
	258	case IPPROTO_ESP:
	259	esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
	260	spi = esph->spi;
	261	break;
	262	case IPPROTO_AH:
	263	ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
	264	spi = ah->spi;
	265	break;
	266	case IPPROTO_COMP:
	267	ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	268	spi = htonl(ntohs(ipch->cpi));
	269	break;
	270	default:
	271	return 0;
	272	}
	273
	274	switch (icmp_hdr(skb)->type) {
	275	case ICMP_DEST_UNREACH:
	276	if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
	277	return 0;
	278	case ICMP_REDIRECT:
	279	break;
	280	default:
	281	return 0;
	282	}
	283
	284	x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr,
	285	spi, protocol, AF_INET);
	286	if (!x)
	287	return 0;
	288
	289	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
	290	ipv4_update_pmtu(skb, net, info, 0, 0, protocol, 0);
	291	else
	292	ipv4_redirect(skb, net, 0, 0, protocol, 0);
	293	xfrm_state_put(x);
	294
	295	return 0;
	296	}
	297
1181412c S	298	static int
	299	vti_tunnel_ioctl(struct net_device dev, struct ifreq ifr, int cmd)
	300	{
	301	int err = 0;
	302	struct ip_tunnel_parm p;
1181412c	303
b9959fd3 AW	304	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
b9959fd3 AW	305	return -EFAULT;
1181412c	306
b9959fd3	307	if (cmd == SIOCADDTUNNEL \|\| cmd == SIOCCHGTUNNEL) {
1181412c S	308	if (p.iph.version != 4 \|\| p.iph.protocol != IPPROTO_IPIP \|\|
1181412c S	309	p.iph.ihl != 5)
b9959fd3 AW	310	return -EINVAL;
b9959fd3 AW	311	}
1181412c	312
df3893c1	313	p.i_flags \|= VTI_ISVTI;
b9959fd3 AW	314	err = ip_tunnel_ioctl(dev, &p, cmd);
	315	if (err)
	316	return err;
1181412c	317
b9959fd3	318	if (cmd != SIOCDELTUNNEL) {
df3893c1	319	p.i_flags \|= GRE_KEY;
b9959fd3	320	p.o_flags \|= GRE_KEY;
1181412c S	321	}
1181412c S	322
b9959fd3 AW	323	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
b9959fd3 AW	324	return -EFAULT;
1181412c S	325	return 0;
	326	}
	327
	328	static const struct net_device_ops vti_netdev_ops = {
	329	.ndo_init = vti_tunnel_init,
b9959fd3	330	.ndo_uninit = ip_tunnel_uninit,
1181412c S	331	.ndo_start_xmit = vti_tunnel_xmit,
1181412c S	332	.ndo_do_ioctl = vti_tunnel_ioctl,
b9959fd3	333	.ndo_change_mtu = ip_tunnel_change_mtu,
f61dd388	334	.ndo_get_stats64 = ip_tunnel_get_stats64,
1181412c S	335	};
1181412c S	336
b9959fd3	337	static void vti_tunnel_setup(struct net_device *dev)
1181412c	338	{
b9959fd3 AW	339	dev->netdev_ops = &vti_netdev_ops;
b9959fd3 AW	340	ip_tunnel_setup(dev, vti_net_id);
1181412c S	341	}
1181412c S	342
b9959fd3	343	static int vti_tunnel_init(struct net_device *dev)
1181412c	344	{
b9959fd3 AW	345	struct ip_tunnel *tunnel = netdev_priv(dev);
	346	struct iphdr *iph = &tunnel->parms.iph;
	347
	348	memcpy(dev->dev_addr, &iph->saddr, 4);
	349	memcpy(dev->broadcast, &iph->daddr, 4);
1181412c S	350
1181412c S	351	dev->type = ARPHRD_TUNNEL;
1181412c S	352	dev->mtu = ETH_DATA_LEN;
	353	dev->flags = IFF_NOARP;
	354	dev->iflink = 0;
	355	dev->addr_len = 4;
1181412c S	356	dev->features \|= NETIF_F_LLTX;
1181412c S	357	dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
1181412c	358
b9959fd3	359	return ip_tunnel_init(dev);
1181412c S	360	}
1181412c S	361
b9959fd3	362	static void __net_init vti_fb_tunnel_init(struct net_device *dev)
1181412c S	363	{
	364	struct ip_tunnel *tunnel = netdev_priv(dev);
	365	struct iphdr *iph = &tunnel->parms.iph;
1181412c	366
1181412c S	367	iph->version = 4;
	368	iph->protocol = IPPROTO_IPIP;
	369	iph->ihl = 5;
1181412c S	370	}
1181412c S	371
df3893c1	372	static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
1181412c	373	.handler = vti_rcv,
df3893c1 SK	374	.input_handler = vti_input,
	375	.cb_handler = vti_rcv_cb,
	376	.err_handler = vti4_err,
	377	.priority = 100,
	378	};
	379
	380	static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
	381	.handler = vti_rcv,
	382	.input_handler = vti_input,
	383	.cb_handler = vti_rcv_cb,
	384	.err_handler = vti4_err,
	385	.priority = 100,
	386	};
	387
	388	static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
	389	.handler = vti_rcv,
	390	.input_handler = vti_input,
	391	.cb_handler = vti_rcv_cb,
	392	.err_handler = vti4_err,
	393	.priority = 100,
1181412c S	394	};
1181412c S	395
1181412c S	396	static int __net_init vti_init_net(struct net *net)
	397	{
	398	int err;
b9959fd3	399	struct ip_tunnel_net *itn;
1181412c	400
b9959fd3	401	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
1181412c	402	if (err)
b9959fd3 AW	403	return err;
	404	itn = net_generic(net, vti_net_id);
	405	vti_fb_tunnel_init(itn->fb_tunnel_dev);
1181412c	406	return 0;
1181412c S	407	}
	408
	409	static void __net_exit vti_exit_net(struct net *net)
	410	{
b9959fd3	411	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
6c742e71	412	ip_tunnel_delete_net(itn, &vti_link_ops);
1181412c S	413	}
	414
	415	static struct pernet_operations vti_net_ops = {
	416	.init = vti_init_net,
	417	.exit = vti_exit_net,
	418	.id = &vti_net_id,
b9959fd3	419	.size = sizeof(struct ip_tunnel_net),
1181412c S	420	};
	421
	422	static int vti_tunnel_validate(struct nlattr tb[], struct nlattr data[])
	423	{
	424	return 0;
	425	}
	426
	427	static void vti_netlink_parms(struct nlattr *data[],
	428	struct ip_tunnel_parm *parms)
	429	{
	430	memset(parms, 0, sizeof(*parms));
	431
	432	parms->iph.protocol = IPPROTO_IPIP;
	433
	434	if (!data)
	435	return;
	436
df3893c1 SK	437	parms->i_flags = VTI_ISVTI;
df3893c1 SK	438
1181412c S	439	if (data[IFLA_VTI_LINK])
	440	parms->link = nla_get_u32(data[IFLA_VTI_LINK]);
	441
	442	if (data[IFLA_VTI_IKEY])
	443	parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);
	444
	445	if (data[IFLA_VTI_OKEY])
	446	parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);
	447
	448	if (data[IFLA_VTI_LOCAL])
	449	parms->iph.saddr = nla_get_be32(data[IFLA_VTI_LOCAL]);
	450
	451	if (data[IFLA_VTI_REMOTE])
	452	parms->iph.daddr = nla_get_be32(data[IFLA_VTI_REMOTE]);
	453
	454	}
	455
	456	static int vti_newlink(struct net src_net, struct net_device dev,
	457	struct nlattr tb[], struct nlattr data[])
	458	{
b9959fd3	459	struct ip_tunnel_parm parms;
1181412c	460
b9959fd3 AW	461	vti_netlink_parms(data, &parms);
b9959fd3 AW	462	return ip_tunnel_newlink(dev, tb, &parms);
1181412c S	463	}
	464
	465	static int vti_changelink(struct net_device dev, struct nlattr tb[],
	466	struct nlattr *data[])
	467	{
1181412c	468	struct ip_tunnel_parm p;
1181412c	469
1181412c	470	vti_netlink_parms(data, &p);
b9959fd3	471	return ip_tunnel_changelink(dev, tb, &p);
1181412c S	472	}
	473
	474	static size_t vti_get_size(const struct net_device *dev)
	475	{
	476	return
	477	/* IFLA_VTI_LINK */
	478	nla_total_size(4) +
	479	/* IFLA_VTI_IKEY */
	480	nla_total_size(4) +
	481	/* IFLA_VTI_OKEY */
	482	nla_total_size(4) +
	483	/* IFLA_VTI_LOCAL */
	484	nla_total_size(4) +
	485	/* IFLA_VTI_REMOTE */
	486	nla_total_size(4) +
	487	0;
	488	}
	489
	490	static int vti_fill_info(struct sk_buff skb, const struct net_device dev)
	491	{
	492	struct ip_tunnel *t = netdev_priv(dev);
	493	struct ip_tunnel_parm *p = &t->parms;
	494
	495	nla_put_u32(skb, IFLA_VTI_LINK, p->link);
	496	nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key);
	497	nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key);
	498	nla_put_be32(skb, IFLA_VTI_LOCAL, p->iph.saddr);
	499	nla_put_be32(skb, IFLA_VTI_REMOTE, p->iph.daddr);
	500
	501	return 0;
	502	}
	503
	504	static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	505	[IFLA_VTI_LINK] = { .type = NLA_U32 },
	506	[IFLA_VTI_IKEY] = { .type = NLA_U32 },
	507	[IFLA_VTI_OKEY] = { .type = NLA_U32 },
	508	[IFLA_VTI_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	509	[IFLA_VTI_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
	510	};
	511
	512	static struct rtnl_link_ops vti_link_ops __read_mostly = {
	513	.kind = "vti",
	514	.maxtype = IFLA_VTI_MAX,
	515	.policy = vti_policy,
	516	.priv_size = sizeof(struct ip_tunnel),
	517	.setup = vti_tunnel_setup,
	518	.validate = vti_tunnel_validate,
	519	.newlink = vti_newlink,
	520	.changelink = vti_changelink,
	521	.get_size = vti_get_size,
	522	.fill_info = vti_fill_info,
	523	};
	524
	525	static int __init vti_init(void)
	526	{
	527	int err;
	528
	529	pr_info("IPv4 over IPSec tunneling driver\n");
	530
	531	err = register_pernet_device(&vti_net_ops);
	532	if (err < 0)
	533	return err;
df3893c1 SK	534	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	535	if (err < 0) {
	536	unregister_pernet_device(&vti_net_ops);
	537	pr_info("vti init: can't register tunnel\n");
	538
	539	return err;
	540	}
	541
	542	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
	543	if (err < 0) {
	544	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
	545	unregister_pernet_device(&vti_net_ops);
	546	pr_info("vti init: can't register tunnel\n");
	547
	548	return err;
	549	}
	550
	551	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
1181412c	552	if (err < 0) {
df3893c1 SK	553	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
df3893c1 SK	554	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c	555	unregister_pernet_device(&vti_net_ops);
b9959fd3	556	pr_info("vti init: can't register tunnel\n");
df3893c1 SK	557
df3893c1 SK	558	return err;
1181412c S	559	}
	560
	561	err = rtnl_link_register(&vti_link_ops);
	562	if (err < 0)
	563	goto rtnl_link_failed;
	564
	565	return err;
	566
	567	rtnl_link_failed:
df3893c1 SK	568	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	569	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	570	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c S	571	unregister_pernet_device(&vti_net_ops);
	572	return err;
	573	}
	574
	575	static void __exit vti_fini(void)
	576	{
	577	rtnl_link_unregister(&vti_link_ops);
df3893c1 SK	578	if (xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP))
	579	pr_info("vti close: can't deregister tunnel\n");
	580	if (xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH))
1181412c	581	pr_info("vti close: can't deregister tunnel\n");
df3893c1 SK	582	if (xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP))
	583	pr_info("vti close: can't deregister tunnel\n");
	584
1181412c S	585
	586	unregister_pernet_device(&vti_net_ops);
	587	}
	588
	589	module_init(vti_init);
	590	module_exit(vti_fini);
	591	MODULE_LICENSE("GPL");
	592	MODULE_ALIAS_RTNL_LINK("vti");
	593	MODULE_ALIAS_NETDEV("ip_vti0");