[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 <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;
}

/* 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 dst_entry *dst = skb_dst(skb);
	struct net_device *tdev;	/* Device to other host */
	struct flowi fl;
	int err;

	if (skb->protocol != htons(ETH_P_IP))
		goto tx_error;

	memset(&fl, 0, sizeof(fl));
	skb->mark = be32_to_cpu(tunnel->parms.o_key);
	xfrm_decode_session(skb, &fl, AF_INET);

	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 there is no transform then this tunnel is not functional.
	 * Or if the xfrm is not mode tunnel.
	 */
	if (!dst->xfrm ||
	    dst->xfrm->props.mode != XFRM_MODE_TUNNEL) {
		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;
	}

	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
	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;
}

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->hard_header_len	= LL_MAX_HEADER + sizeof(struct iphdr);
	dev->mtu		= ETH_DATA_LEN;
	dev->flags		= IFF_NOARP;
	dev->iflink		= 0;
	dev->addr_len		= 4;
	dev->features		|= NETIF_F_NETNS_LOCAL;
	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>
	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);
1181412c	62	if (tunnel != NULL) {
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;
	67	skb->mark = be32_to_cpu(tunnel->parms.i_key);
	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;
	93
	94	if (!tunnel)
1181412c	95	return 1;
df3893c1 SK	96
	97	dev = tunnel->dev;
	98
	99	if (err) {
	100	dev->stats.rx_errors++;
	101	dev->stats.rx_dropped++;
	102
	103	return 0;
1181412c	104	}
1181412c	105
df3893c1 SK	106	x = xfrm_input_state(skb);
	107	family = x->inner_mode->afinfo->family;
	108
	109	if (!xfrm_policy_check(NULL, XFRM_POLICY_IN, skb, family))
	110	return -EPERM;
	111
	112	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(skb->dev)));
	113	skb->dev = dev;
	114
	115	tstats = this_cpu_ptr(dev->tstats);
	116
	117	u64_stats_update_begin(&tstats->syncp);
	118	tstats->rx_packets++;
	119	tstats->rx_bytes += skb->len;
	120	u64_stats_update_end(&tstats->syncp);
	121
	122	return 0;
1181412c S	123	}
	124
	125	/* This function assumes it is being called from dev_queue_xmit()
	126	* and that skb is filled properly by that function.
	127	*/
1181412c S	128	static netdev_tx_t vti_tunnel_xmit(struct sk_buff skb, struct net_device dev)
	129	{
	130	struct ip_tunnel *tunnel = netdev_priv(dev);
a34cd4f3	131	struct dst_entry *dst = skb_dst(skb);
1181412c	132	struct net_device tdev; / Device to other host */
df3893c1	133	struct flowi fl;
b9959fd3	134	int err;
1181412c S	135
	136	if (skb->protocol != htons(ETH_P_IP))
	137	goto tx_error;
	138
df3893c1 SK	139	memset(&fl, 0, sizeof(fl));
	140	skb->mark = be32_to_cpu(tunnel->parms.o_key);
	141	xfrm_decode_session(skb, &fl, AF_INET);
	142
a34cd4f3	143	if (!dst) {
df3893c1 SK	144	dev->stats.tx_carrier_errors++;
	145	goto tx_error_icmp;
	146	}
1181412c	147
a34cd4f3 SK	148	dst_hold(dst);
	149	dst = xfrm_lookup(tunnel->net, dst, &fl, NULL, 0);
	150	if (IS_ERR(dst)) {
1181412c S	151	dev->stats.tx_carrier_errors++;
	152	goto tx_error_icmp;
	153	}
df3893c1	154
1181412c S	155	/* if there is no transform then this tunnel is not functional.
	156	* Or if the xfrm is not mode tunnel.
	157	*/
a34cd4f3 SK	158	if (!dst->xfrm \|\|
a34cd4f3 SK	159	dst->xfrm->props.mode != XFRM_MODE_TUNNEL) {
1181412c	160	dev->stats.tx_carrier_errors++;
a34cd4f3	161	dst_release(dst);
1181412c S	162	goto tx_error_icmp;
1181412c S	163	}
a34cd4f3	164	tdev = dst->dev;
1181412c S	165
1181412c S	166	if (tdev == dev) {
a34cd4f3	167	dst_release(dst);
1181412c S	168	dev->stats.collisions++;
	169	goto tx_error;
	170	}
	171
	172	if (tunnel->err_count > 0) {
	173	if (time_before(jiffies,
	174	tunnel->err_time + IPTUNNEL_ERR_TIMEO)) {
	175	tunnel->err_count--;
	176	dst_link_failure(skb);
	177	} else
	178	tunnel->err_count = 0;
	179	}
	180
baafc77b	181	memset(IPCB(skb), 0, sizeof(*IPCB(skb)));
df3893c1	182	skb_scrub_packet(skb, !net_eq(tunnel->net, dev_net(dev)));
a34cd4f3	183	skb_dst_set(skb, dst);
1181412c S	184	skb->dev = skb_dst(skb)->dev;
1181412c S	185
b9959fd3 AW	186	err = dst_output(skb);
	187	if (net_xmit_eval(err) == 0)
	188	err = skb->len;
	189	iptunnel_xmit_stats(err, &dev->stats, dev->tstats);
1181412c S	190	return NETDEV_TX_OK;
	191
	192	tx_error_icmp:
	193	dst_link_failure(skb);
	194	tx_error:
	195	dev->stats.tx_errors++;
3acfa1e7	196	kfree_skb(skb);
1181412c S	197	return NETDEV_TX_OK;
	198	}
	199
df3893c1 SK	200	static int vti4_err(struct sk_buff *skb, u32 info)
	201	{
	202	__be32 spi;
	203	struct xfrm_state *x;
	204	struct ip_tunnel *tunnel;
	205	struct ip_esp_hdr *esph;
	206	struct ip_auth_hdr *ah ;
	207	struct ip_comp_hdr *ipch;
	208	struct net *net = dev_net(skb->dev);
	209	const struct iphdr iph = (const struct iphdr )skb->data;
	210	int protocol = iph->protocol;
	211	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
	212
	213	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
	214	iph->daddr, iph->saddr, 0);
	215	if (!tunnel)
	216	return -1;
	217
	218	switch (protocol) {
	219	case IPPROTO_ESP:
	220	esph = (struct ip_esp_hdr *)(skb->data+(iph->ihl<<2));
	221	spi = esph->spi;
	222	break;
	223	case IPPROTO_AH:
	224	ah = (struct ip_auth_hdr *)(skb->data+(iph->ihl<<2));
	225	spi = ah->spi;
	226	break;
	227	case IPPROTO_COMP:
	228	ipch = (struct ip_comp_hdr *)(skb->data+(iph->ihl<<2));
	229	spi = htonl(ntohs(ipch->cpi));
	230	break;
	231	default:
	232	return 0;
	233	}
	234
	235	switch (icmp_hdr(skb)->type) {
	236	case ICMP_DEST_UNREACH:
	237	if (icmp_hdr(skb)->code != ICMP_FRAG_NEEDED)
	238	return 0;
	239	case ICMP_REDIRECT:
	240	break;
	241	default:
	242	return 0;
	243	}
	244
	245	x = xfrm_state_lookup(net, skb->mark, (const xfrm_address_t *)&iph->daddr,
	246	spi, protocol, AF_INET);
	247	if (!x)
	248	return 0;
	249
	250	if (icmp_hdr(skb)->type == ICMP_DEST_UNREACH)
	251	ipv4_update_pmtu(skb, net, info, 0, 0, protocol, 0);
	252	else
	253	ipv4_redirect(skb, net, 0, 0, protocol, 0);
	254	xfrm_state_put(x);
	255
	256	return 0;
	257	}
	258
1181412c S	259	static int
	260	vti_tunnel_ioctl(struct net_device dev, struct ifreq ifr, int cmd)
	261	{
	262	int err = 0;
	263	struct ip_tunnel_parm p;
1181412c	264
b9959fd3 AW	265	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
b9959fd3 AW	266	return -EFAULT;
1181412c	267
b9959fd3	268	if (cmd == SIOCADDTUNNEL \|\| cmd == SIOCCHGTUNNEL) {
1181412c S	269	if (p.iph.version != 4 \|\| p.iph.protocol != IPPROTO_IPIP \|\|
1181412c S	270	p.iph.ihl != 5)
b9959fd3 AW	271	return -EINVAL;
b9959fd3 AW	272	}
1181412c	273
df3893c1	274	p.i_flags \|= VTI_ISVTI;
b9959fd3 AW	275	err = ip_tunnel_ioctl(dev, &p, cmd);
	276	if (err)
	277	return err;
1181412c	278
b9959fd3	279	if (cmd != SIOCDELTUNNEL) {
df3893c1	280	p.i_flags \|= GRE_KEY;
b9959fd3	281	p.o_flags \|= GRE_KEY;
1181412c S	282	}
1181412c S	283
b9959fd3 AW	284	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
b9959fd3 AW	285	return -EFAULT;
1181412c S	286	return 0;
	287	}
	288
	289	static const struct net_device_ops vti_netdev_ops = {
	290	.ndo_init = vti_tunnel_init,
b9959fd3	291	.ndo_uninit = ip_tunnel_uninit,
1181412c S	292	.ndo_start_xmit = vti_tunnel_xmit,
1181412c S	293	.ndo_do_ioctl = vti_tunnel_ioctl,
b9959fd3	294	.ndo_change_mtu = ip_tunnel_change_mtu,
f61dd388	295	.ndo_get_stats64 = ip_tunnel_get_stats64,
1181412c S	296	};
1181412c S	297
b9959fd3	298	static void vti_tunnel_setup(struct net_device *dev)
1181412c	299	{
b9959fd3 AW	300	dev->netdev_ops = &vti_netdev_ops;
b9959fd3 AW	301	ip_tunnel_setup(dev, vti_net_id);
1181412c S	302	}
1181412c S	303
b9959fd3	304	static int vti_tunnel_init(struct net_device *dev)
1181412c	305	{
b9959fd3 AW	306	struct ip_tunnel *tunnel = netdev_priv(dev);
	307	struct iphdr *iph = &tunnel->parms.iph;
	308
	309	memcpy(dev->dev_addr, &iph->saddr, 4);
	310	memcpy(dev->broadcast, &iph->daddr, 4);
1181412c S	311
	312	dev->type = ARPHRD_TUNNEL;
	313	dev->hard_header_len = LL_MAX_HEADER + sizeof(struct iphdr);
	314	dev->mtu = ETH_DATA_LEN;
	315	dev->flags = IFF_NOARP;
	316	dev->iflink = 0;
	317	dev->addr_len = 4;
	318	dev->features \|= NETIF_F_NETNS_LOCAL;
	319	dev->features \|= NETIF_F_LLTX;
	320	dev->priv_flags &= ~IFF_XMIT_DST_RELEASE;
1181412c	321
b9959fd3	322	return ip_tunnel_init(dev);
1181412c S	323	}
1181412c S	324
b9959fd3	325	static void __net_init vti_fb_tunnel_init(struct net_device *dev)
1181412c S	326	{
	327	struct ip_tunnel *tunnel = netdev_priv(dev);
	328	struct iphdr *iph = &tunnel->parms.iph;
1181412c	329
1181412c S	330	iph->version = 4;
	331	iph->protocol = IPPROTO_IPIP;
	332	iph->ihl = 5;
1181412c S	333	}
1181412c S	334
df3893c1	335	static struct xfrm4_protocol vti_esp4_protocol __read_mostly = {
1181412c	336	.handler = vti_rcv,
df3893c1 SK	337	.input_handler = vti_input,
	338	.cb_handler = vti_rcv_cb,
	339	.err_handler = vti4_err,
	340	.priority = 100,
	341	};
	342
	343	static struct xfrm4_protocol vti_ah4_protocol __read_mostly = {
	344	.handler = vti_rcv,
	345	.input_handler = vti_input,
	346	.cb_handler = vti_rcv_cb,
	347	.err_handler = vti4_err,
	348	.priority = 100,
	349	};
	350
	351	static struct xfrm4_protocol vti_ipcomp4_protocol __read_mostly = {
	352	.handler = vti_rcv,
	353	.input_handler = vti_input,
	354	.cb_handler = vti_rcv_cb,
	355	.err_handler = vti4_err,
	356	.priority = 100,
1181412c S	357	};
1181412c S	358
1181412c S	359	static int __net_init vti_init_net(struct net *net)
	360	{
	361	int err;
b9959fd3	362	struct ip_tunnel_net *itn;
1181412c	363
b9959fd3	364	err = ip_tunnel_init_net(net, vti_net_id, &vti_link_ops, "ip_vti0");
1181412c	365	if (err)
b9959fd3 AW	366	return err;
	367	itn = net_generic(net, vti_net_id);
	368	vti_fb_tunnel_init(itn->fb_tunnel_dev);
1181412c	369	return 0;
1181412c S	370	}
	371
	372	static void __net_exit vti_exit_net(struct net *net)
	373	{
b9959fd3	374	struct ip_tunnel_net *itn = net_generic(net, vti_net_id);
6c742e71	375	ip_tunnel_delete_net(itn, &vti_link_ops);
1181412c S	376	}
	377
	378	static struct pernet_operations vti_net_ops = {
	379	.init = vti_init_net,
	380	.exit = vti_exit_net,
	381	.id = &vti_net_id,
b9959fd3	382	.size = sizeof(struct ip_tunnel_net),
1181412c S	383	};
	384
	385	static int vti_tunnel_validate(struct nlattr tb[], struct nlattr data[])
	386	{
	387	return 0;
	388	}
	389
	390	static void vti_netlink_parms(struct nlattr *data[],
	391	struct ip_tunnel_parm *parms)
	392	{
	393	memset(parms, 0, sizeof(*parms));
	394
	395	parms->iph.protocol = IPPROTO_IPIP;
	396
	397	if (!data)
	398	return;
	399
df3893c1 SK	400	parms->i_flags = VTI_ISVTI;
df3893c1 SK	401
1181412c S	402	if (data[IFLA_VTI_LINK])
	403	parms->link = nla_get_u32(data[IFLA_VTI_LINK]);
	404
	405	if (data[IFLA_VTI_IKEY])
	406	parms->i_key = nla_get_be32(data[IFLA_VTI_IKEY]);
	407
	408	if (data[IFLA_VTI_OKEY])
	409	parms->o_key = nla_get_be32(data[IFLA_VTI_OKEY]);
	410
	411	if (data[IFLA_VTI_LOCAL])
	412	parms->iph.saddr = nla_get_be32(data[IFLA_VTI_LOCAL]);
	413
	414	if (data[IFLA_VTI_REMOTE])
	415	parms->iph.daddr = nla_get_be32(data[IFLA_VTI_REMOTE]);
	416
	417	}
	418
	419	static int vti_newlink(struct net src_net, struct net_device dev,
	420	struct nlattr tb[], struct nlattr data[])
	421	{
b9959fd3	422	struct ip_tunnel_parm parms;
1181412c	423
b9959fd3 AW	424	vti_netlink_parms(data, &parms);
b9959fd3 AW	425	return ip_tunnel_newlink(dev, tb, &parms);
1181412c S	426	}
	427
	428	static int vti_changelink(struct net_device dev, struct nlattr tb[],
	429	struct nlattr *data[])
	430	{
1181412c	431	struct ip_tunnel_parm p;
1181412c	432
1181412c	433	vti_netlink_parms(data, &p);
b9959fd3	434	return ip_tunnel_changelink(dev, tb, &p);
1181412c S	435	}
	436
	437	static size_t vti_get_size(const struct net_device *dev)
	438	{
	439	return
	440	/* IFLA_VTI_LINK */
	441	nla_total_size(4) +
	442	/* IFLA_VTI_IKEY */
	443	nla_total_size(4) +
	444	/* IFLA_VTI_OKEY */
	445	nla_total_size(4) +
	446	/* IFLA_VTI_LOCAL */
	447	nla_total_size(4) +
	448	/* IFLA_VTI_REMOTE */
	449	nla_total_size(4) +
	450	0;
	451	}
	452
	453	static int vti_fill_info(struct sk_buff skb, const struct net_device dev)
	454	{
	455	struct ip_tunnel *t = netdev_priv(dev);
	456	struct ip_tunnel_parm *p = &t->parms;
	457
	458	nla_put_u32(skb, IFLA_VTI_LINK, p->link);
	459	nla_put_be32(skb, IFLA_VTI_IKEY, p->i_key);
	460	nla_put_be32(skb, IFLA_VTI_OKEY, p->o_key);
	461	nla_put_be32(skb, IFLA_VTI_LOCAL, p->iph.saddr);
	462	nla_put_be32(skb, IFLA_VTI_REMOTE, p->iph.daddr);
	463
	464	return 0;
	465	}
	466
	467	static const struct nla_policy vti_policy[IFLA_VTI_MAX + 1] = {
	468	[IFLA_VTI_LINK] = { .type = NLA_U32 },
	469	[IFLA_VTI_IKEY] = { .type = NLA_U32 },
	470	[IFLA_VTI_OKEY] = { .type = NLA_U32 },
	471	[IFLA_VTI_LOCAL] = { .len = FIELD_SIZEOF(struct iphdr, saddr) },
	472	[IFLA_VTI_REMOTE] = { .len = FIELD_SIZEOF(struct iphdr, daddr) },
	473	};
	474
	475	static struct rtnl_link_ops vti_link_ops __read_mostly = {
	476	.kind = "vti",
	477	.maxtype = IFLA_VTI_MAX,
	478	.policy = vti_policy,
	479	.priv_size = sizeof(struct ip_tunnel),
	480	.setup = vti_tunnel_setup,
	481	.validate = vti_tunnel_validate,
	482	.newlink = vti_newlink,
	483	.changelink = vti_changelink,
	484	.get_size = vti_get_size,
	485	.fill_info = vti_fill_info,
	486	};
	487
	488	static int __init vti_init(void)
	489	{
	490	int err;
	491
	492	pr_info("IPv4 over IPSec tunneling driver\n");
	493
	494	err = register_pernet_device(&vti_net_ops);
	495	if (err < 0)
	496	return err;
df3893c1 SK	497	err = xfrm4_protocol_register(&vti_esp4_protocol, IPPROTO_ESP);
	498	if (err < 0) {
	499	unregister_pernet_device(&vti_net_ops);
	500	pr_info("vti init: can't register tunnel\n");
	501
	502	return err;
	503	}
	504
	505	err = xfrm4_protocol_register(&vti_ah4_protocol, IPPROTO_AH);
	506	if (err < 0) {
	507	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
	508	unregister_pernet_device(&vti_net_ops);
	509	pr_info("vti init: can't register tunnel\n");
	510
	511	return err;
	512	}
	513
	514	err = xfrm4_protocol_register(&vti_ipcomp4_protocol, IPPROTO_COMP);
1181412c	515	if (err < 0) {
df3893c1 SK	516	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
df3893c1 SK	517	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c	518	unregister_pernet_device(&vti_net_ops);
b9959fd3	519	pr_info("vti init: can't register tunnel\n");
df3893c1 SK	520
df3893c1 SK	521	return err;
1181412c S	522	}
	523
	524	err = rtnl_link_register(&vti_link_ops);
	525	if (err < 0)
	526	goto rtnl_link_failed;
	527
	528	return err;
	529
	530	rtnl_link_failed:
df3893c1 SK	531	xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP);
	532	xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH);
	533	xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP);
1181412c S	534	unregister_pernet_device(&vti_net_ops);
	535	return err;
	536	}
	537
	538	static void __exit vti_fini(void)
	539	{
	540	rtnl_link_unregister(&vti_link_ops);
df3893c1 SK	541	if (xfrm4_protocol_deregister(&vti_ipcomp4_protocol, IPPROTO_COMP))
	542	pr_info("vti close: can't deregister tunnel\n");
	543	if (xfrm4_protocol_deregister(&vti_ah4_protocol, IPPROTO_AH))
1181412c	544	pr_info("vti close: can't deregister tunnel\n");
df3893c1 SK	545	if (xfrm4_protocol_deregister(&vti_esp4_protocol, IPPROTO_ESP))
	546	pr_info("vti close: can't deregister tunnel\n");
	547
1181412c S	548
	549	unregister_pernet_device(&vti_net_ops);
	550	}
	551
	552	module_init(vti_init);
	553	module_exit(vti_fini);
	554	MODULE_LICENSE("GPL");
	555	MODULE_ALIAS_RTNL_LINK("vti");
	556	MODULE_ALIAS_NETDEV("ip_vti0");