ipip.c 18.7 KB
Newer Older
L
Linus Torvalds 已提交
1
/*
2
 *	Linux NET3:	IP/IP protocol decoder.
L
Linus Torvalds 已提交
3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
 *
 *	Authors:
 *		Sam Lantinga (slouken@cs.ucdavis.edu)  02/01/95
 *
 *	Fixes:
 *		Alan Cox	:	Merged and made usable non modular (its so tiny its silly as
 *					a module taking up 2 pages).
 *		Alan Cox	: 	Fixed bug with 1.3.18 and IPIP not working (now needs to set skb->h.iph)
 *					to keep ip_forward happy.
 *		Alan Cox	:	More fixes for 1.3.21, and firewall fix. Maybe this will work soon 8).
 *		Kai Schulte	:	Fixed #defines for IP_FIREWALL->FIREWALL
 *              David Woodhouse :       Perform some basic ICMP handling.
 *                                      IPIP Routing without decapsulation.
 *              Carlos Picoto   :       GRE over IP support
 *		Alexey Kuznetsov:	Reworked. Really, now it is truncated version of ipv4/ip_gre.c.
 *					I do not want to merge them together.
 *
 *	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.
 *
 */

/* tunnel.c: an IP tunnel driver

	The purpose of this driver is to provide an IP tunnel through
	which you can tunnel network traffic transparently across subnets.

	This was written by looking at Nick Holloway's dummy driver
	Thanks for the great code!

		-Sam Lantinga	(slouken@cs.ucdavis.edu)  02/01/95
36

L
Linus Torvalds 已提交
37 38 39 40 41 42
	Minor tweaks:
		Cleaned up the code a little and added some pre-1.3.0 tweaks.
		dev->hard_header/hard_header_len changed to use no headers.
		Comments/bracketing tweaked.
		Made the tunnels use dev->name not tunnel: when error reporting.
		Added tx_dropped stat
43

44
		-Alan Cox	(alan@lxorguk.ukuu.org.uk) 21 March 95
L
Linus Torvalds 已提交
45 46 47 48 49 50 51 52

	Reworked:
		Changed to tunnel to destination gateway in addition to the
			tunnel's pointopoint address
		Almost completely rewritten
		Note:  There is currently no firewall or ICMP handling done.

		-Sam Lantinga	(slouken@cs.ucdavis.edu) 02/13/96
53

L
Linus Torvalds 已提交
54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75
*/

/* Things I wish I had known when writing the tunnel driver:

	When the tunnel_xmit() function is called, the skb contains the
	packet to be sent (plus a great deal of extra info), and dev
	contains the tunnel device that _we_ are.

	When we are passed a packet, we are expected to fill in the
	source address with our source IP address.

	What is the proper way to allocate, copy and free a buffer?
	After you allocate it, it is a "0 length" chunk of memory
	starting at zero.  If you want to add headers to the buffer
	later, you'll have to call "skb_reserve(skb, amount)" with
	the amount of memory you want reserved.  Then, you call
	"skb_put(skb, amount)" with the amount of space you want in
	the buffer.  skb_put() returns a pointer to the top (#0) of
	that buffer.  skb->len is set to the amount of space you have
	"allocated" with skb_put().  You can then write up to skb->len
	bytes to that buffer.  If you need more, you can call skb_put()
	again with the additional amount of space you need.  You can
76
	find out how much more space you can allocate by calling
L
Linus Torvalds 已提交
77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92
	"skb_tailroom(skb)".
	Now, to add header space, call "skb_push(skb, header_len)".
	This creates space at the beginning of the buffer and returns
	a pointer to this new space.  If later you need to strip a
	header from a buffer, call "skb_pull(skb, header_len)".
	skb_headroom() will return how much space is left at the top
	of the buffer (before the main data).  Remember, this headroom
	space must be reserved before the skb_put() function is called.
	*/

/*
   This version of net/ipv4/ipip.c is cloned of net/ipv4/ip_gre.c

   For comments look at net/ipv4/ip_gre.c --ANK
 */

93

94
#include <linux/capability.h>
L
Linus Torvalds 已提交
95 96 97
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
98
#include <linux/slab.h>
99
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
100 101 102 103 104 105 106 107
#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>
108
#include <linux/if_ether.h>
L
Linus Torvalds 已提交
109 110 111 112

#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
113
#include <net/ip_tunnels.h>
L
Linus Torvalds 已提交
114 115
#include <net/inet_ecn.h>
#include <net/xfrm.h>
116 117
#include <net/net_namespace.h>
#include <net/netns/generic.h>
118
#include <net/dst_metadata.h>
L
Linus Torvalds 已提交
119

120 121 122 123
static bool log_ecn_error = true;
module_param(log_ecn_error, bool, 0644);
MODULE_PARM_DESC(log_ecn_error, "Log packets received with corrupted ECN");

124
static unsigned int ipip_net_id __read_mostly;
125

E
Eric Dumazet 已提交
126
static int ipip_tunnel_init(struct net_device *dev);
127
static struct rtnl_link_ops ipip_link_ops __read_mostly;
L
Linus Torvalds 已提交
128

H
Herbert Xu 已提交
129
static int ipip_err(struct sk_buff *skb, u32 info)
L
Linus Torvalds 已提交
130
{
131 132 133 134
	/* All the routers (except for Linux) return only
	 * 8 bytes of packet payload. It means, that precise relaying of
	 * ICMP in the real Internet is absolutely infeasible.
	 */
P
Pravin B Shelar 已提交
135 136
	struct net *net = dev_net(skb->dev);
	struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
137
	const struct iphdr *iph = (const struct iphdr *)skb->data;
P
Pravin B Shelar 已提交
138 139
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
140 141 142
	struct ip_tunnel *t;
	int err = 0;

143 144 145 146 147 148 149
	t = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
			     iph->daddr, iph->saddr, 0);
	if (!t) {
		err = -ENOENT;
		goto out;
	}

150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175
	switch (type) {
	case ICMP_DEST_UNREACH:
		switch (code) {
		case ICMP_SR_FAILED:
			/* Impossible event. */
			goto out;
		default:
			/* All others are translated to HOST_UNREACH.
			 * rfc2003 contains "deep thoughts" about NET_UNREACH,
			 * I believe they are just ether pollution. --ANK
			 */
			break;
		}
		break;

	case ICMP_TIME_EXCEEDED:
		if (code != ICMP_EXC_TTL)
			goto out;
		break;

	case ICMP_REDIRECT:
		break;

	default:
		goto out;
	}
L
Linus Torvalds 已提交
176

177
	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
178
		ipv4_update_pmtu(skb, net, info, t->parms.link, iph->protocol);
179 180 181
		goto out;
	}

182
	if (type == ICMP_REDIRECT) {
183
		ipv4_redirect(skb, net, t->parms.link, iph->protocol);
184 185 186
		goto out;
	}

187 188
	if (t->parms.iph.daddr == 0) {
		err = -ENOENT;
L
Linus Torvalds 已提交
189
		goto out;
190
	}
H
Herbert Xu 已提交
191

L
Linus Torvalds 已提交
192 193 194
	if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
		goto out;

195
	if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
L
Linus Torvalds 已提交
196 197 198 199
		t->err_count++;
	else
		t->err_count = 1;
	t->err_time = jiffies;
200

P
Pravin B Shelar 已提交
201
out:
H
Herbert Xu 已提交
202
	return err;
L
Linus Torvalds 已提交
203 204
}

S
Simon Horman 已提交
205
static const struct tnl_ptk_info ipip_tpi = {
P
Pravin B Shelar 已提交
206 207 208 209
	/* no tunnel info required for ipip. */
	.proto = htons(ETH_P_IP),
};

S
Simon Horman 已提交
210 211 212 213 214 215 216 217
#if IS_ENABLED(CONFIG_MPLS)
static const struct tnl_ptk_info mplsip_tpi = {
	/* no tunnel info required for mplsip. */
	.proto = htons(ETH_P_MPLS_UC),
};
#endif

static int ipip_tunnel_rcv(struct sk_buff *skb, u8 ipproto)
L
Linus Torvalds 已提交
218
{
P
Pravin B Shelar 已提交
219 220
	struct net *net = dev_net(skb->dev);
	struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
221
	struct metadata_dst *tun_dst = NULL;
L
Linus Torvalds 已提交
222
	struct ip_tunnel *tunnel;
223
	const struct iphdr *iph;
E
Eric Dumazet 已提交
224

225
	iph = ip_hdr(skb);
P
Pravin B Shelar 已提交
226 227 228
	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
			iph->saddr, iph->daddr, 0);
	if (tunnel) {
S
Simon Horman 已提交
229 230 231 232 233 234
		const struct tnl_ptk_info *tpi;

		if (tunnel->parms.iph.protocol != ipproto &&
		    tunnel->parms.iph.protocol != 0)
			goto drop;

235 236
		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
			goto drop;
S
Simon Horman 已提交
237 238 239 240 241 242 243
#if IS_ENABLED(CONFIG_MPLS)
		if (ipproto == IPPROTO_MPLS)
			tpi = &mplsip_tpi;
		else
#endif
			tpi = &ipip_tpi;
		if (iptunnel_pull_header(skb, 0, tpi->proto, false))
244
			goto drop;
245 246 247 248 249 250
		if (tunnel->collect_md) {
			tun_dst = ip_tun_rx_dst(skb, 0, 0, 0);
			if (!tun_dst)
				return 0;
		}
		return ip_tunnel_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error);
L
Linus Torvalds 已提交
251 252 253
	}

	return -1;
254 255 256 257

drop:
	kfree_skb(skb);
	return 0;
L
Linus Torvalds 已提交
258 259
}

S
Simon Horman 已提交
260 261 262 263 264 265 266 267 268 269 270 271
static int ipip_rcv(struct sk_buff *skb)
{
	return ipip_tunnel_rcv(skb, IPPROTO_IPIP);
}

#if IS_ENABLED(CONFIG_MPLS)
static int mplsip_rcv(struct sk_buff *skb)
{
	return ipip_tunnel_rcv(skb, IPPROTO_MPLS);
}
#endif

L
Linus Torvalds 已提交
272 273 274 275
/*
 *	This function assumes it is being called from dev_queue_xmit()
 *	and that skb is filled properly by that function.
 */
S
Simon Horman 已提交
276 277
static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb,
				    struct net_device *dev)
L
Linus Torvalds 已提交
278
{
279
	struct ip_tunnel *tunnel = netdev_priv(dev);
280
	const struct iphdr  *tiph = &tunnel->parms.iph;
S
Simon Horman 已提交
281 282 283 284 285 286 287 288 289 290 291 292 293 294
	u8 ipproto;

	switch (skb->protocol) {
	case htons(ETH_P_IP):
		ipproto = IPPROTO_IPIP;
		break;
#if IS_ENABLED(CONFIG_MPLS)
	case htons(ETH_P_MPLS_UC):
		ipproto = IPPROTO_MPLS;
		break;
#endif
	default:
		goto tx_error;
	}
L
Linus Torvalds 已提交
295

S
Simon Horman 已提交
296
	if (tiph->protocol != ipproto && tiph->protocol != 0)
L
Linus Torvalds 已提交
297 298
		goto tx_error;

299
	if (iptunnel_handle_offloads(skb, SKB_GSO_IPXIP4))
300
		goto tx_error;
301

S
Simon Horman 已提交
302
	skb_set_inner_ipproto(skb, ipproto);
T
Tom Herbert 已提交
303

304
	if (tunnel->collect_md)
305
		ip_md_tunnel_xmit(skb, dev, ipproto, 0);
306 307
	else
		ip_tunnel_xmit(skb, dev, tiph, ipproto);
308
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
309 310

tx_error:
E
Eric Dumazet 已提交
311
	kfree_skb(skb);
312

E
Eric Dumazet 已提交
313
	dev->stats.tx_errors++;
314
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
315 316
}

S
Simon Horman 已提交
317 318 319 320 321 322 323 324 325 326 327 328 329 330
static bool ipip_tunnel_ioctl_verify_protocol(u8 ipproto)
{
	switch (ipproto) {
	case 0:
	case IPPROTO_IPIP:
#if IS_ENABLED(CONFIG_MPLS)
	case IPPROTO_MPLS:
#endif
		return true;
	}

	return false;
}

L
Linus Torvalds 已提交
331
static int
P
Pravin B Shelar 已提交
332
ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
L
Linus Torvalds 已提交
333 334 335 336
{
	int err = 0;
	struct ip_tunnel_parm p;

P
Pravin B Shelar 已提交
337 338
	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
		return -EFAULT;
L
Linus Torvalds 已提交
339

C
Cong Wang 已提交
340
	if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) {
S
Simon Horman 已提交
341 342
		if (p.iph.version != 4 ||
		    !ipip_tunnel_ioctl_verify_protocol(p.iph.protocol) ||
C
Cong Wang 已提交
343 344 345 346
		    p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)))
			return -EINVAL;
	}

E
Eric Dumazet 已提交
347 348
	p.i_key = p.o_key = 0;
	p.i_flags = p.o_flags = 0;
P
Pravin B Shelar 已提交
349 350 351 352 353 354 355
	err = ip_tunnel_ioctl(dev, &p, cmd);
	if (err)
		return err;

	if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
		return -EFAULT;

L
Linus Torvalds 已提交
356 357 358
	return 0;
}

359
static const struct net_device_ops ipip_netdev_ops = {
P
Pravin B Shelar 已提交
360 361
	.ndo_init       = ipip_tunnel_init,
	.ndo_uninit     = ip_tunnel_uninit,
362 363
	.ndo_start_xmit	= ipip_tunnel_xmit,
	.ndo_do_ioctl	= ipip_tunnel_ioctl,
P
Pravin B Shelar 已提交
364 365
	.ndo_change_mtu = ip_tunnel_change_mtu,
	.ndo_get_stats64 = ip_tunnel_get_stats64,
366
	.ndo_get_iflink = ip_tunnel_get_iflink,
367 368
};

E
Eric Dumazet 已提交
369 370 371
#define IPIP_FEATURES (NETIF_F_SG |		\
		       NETIF_F_FRAGLIST |	\
		       NETIF_F_HIGHDMA |	\
E
Eric Dumazet 已提交
372
		       NETIF_F_GSO_SOFTWARE |	\
E
Eric Dumazet 已提交
373 374
		       NETIF_F_HW_CSUM)

L
Linus Torvalds 已提交
375 376
static void ipip_tunnel_setup(struct net_device *dev)
{
377
	dev->netdev_ops		= &ipip_netdev_ops;
L
Linus Torvalds 已提交
378 379 380 381

	dev->type		= ARPHRD_TUNNEL;
	dev->flags		= IFF_NOARP;
	dev->addr_len		= 4;
E
Eric Dumazet 已提交
382
	dev->features		|= NETIF_F_LLTX;
383
	netif_keep_dst(dev);
E
Eric Dumazet 已提交
384 385 386

	dev->features		|= IPIP_FEATURES;
	dev->hw_features	|= IPIP_FEATURES;
P
Pravin B Shelar 已提交
387
	ip_tunnel_setup(dev, ipip_net_id);
L
Linus Torvalds 已提交
388 389
}

E
Eric Dumazet 已提交
390
static int ipip_tunnel_init(struct net_device *dev)
L
Linus Torvalds 已提交
391
{
392
	struct ip_tunnel *tunnel = netdev_priv(dev);
L
Linus Torvalds 已提交
393 394 395 396

	memcpy(dev->dev_addr, &tunnel->parms.iph.saddr, 4);
	memcpy(dev->broadcast, &tunnel->parms.iph.daddr, 4);

397 398
	tunnel->tun_hlen = 0;
	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
P
Pravin B Shelar 已提交
399
	return ip_tunnel_init(dev);
L
Linus Torvalds 已提交
400 401
}

402 403
static int ipip_tunnel_validate(struct nlattr *tb[], struct nlattr *data[],
				struct netlink_ext_ack *extack)
S
Simon Horman 已提交
404 405 406 407 408 409 410 411 412 413 414 415 416
{
	u8 proto;

	if (!data || !data[IFLA_IPTUN_PROTO])
		return 0;

	proto = nla_get_u8(data[IFLA_IPTUN_PROTO]);
	if (proto != IPPROTO_IPIP && proto != IPPROTO_MPLS && proto != 0)
		return -EINVAL;

	return 0;
}

417
static void ipip_netlink_parms(struct nlattr *data[],
418 419
			       struct ip_tunnel_parm *parms, bool *collect_md,
			       __u32 *fwmark)
420 421 422 423 424 425
{
	memset(parms, 0, sizeof(*parms));

	parms->iph.version = 4;
	parms->iph.protocol = IPPROTO_IPIP;
	parms->iph.ihl = 5;
426
	*collect_md = false;
427 428 429 430 431 432 433 434

	if (!data)
		return;

	if (data[IFLA_IPTUN_LINK])
		parms->link = nla_get_u32(data[IFLA_IPTUN_LINK]);

	if (data[IFLA_IPTUN_LOCAL])
435
		parms->iph.saddr = nla_get_in_addr(data[IFLA_IPTUN_LOCAL]);
436 437

	if (data[IFLA_IPTUN_REMOTE])
438
		parms->iph.daddr = nla_get_in_addr(data[IFLA_IPTUN_REMOTE]);
439 440 441 442 443 444 445 446 447 448

	if (data[IFLA_IPTUN_TTL]) {
		parms->iph.ttl = nla_get_u8(data[IFLA_IPTUN_TTL]);
		if (parms->iph.ttl)
			parms->iph.frag_off = htons(IP_DF);
	}

	if (data[IFLA_IPTUN_TOS])
		parms->iph.tos = nla_get_u8(data[IFLA_IPTUN_TOS]);

S
Simon Horman 已提交
449 450 451
	if (data[IFLA_IPTUN_PROTO])
		parms->iph.protocol = nla_get_u8(data[IFLA_IPTUN_PROTO]);

452 453
	if (!data[IFLA_IPTUN_PMTUDISC] || nla_get_u8(data[IFLA_IPTUN_PMTUDISC]))
		parms->iph.frag_off = htons(IP_DF);
454 455 456

	if (data[IFLA_IPTUN_COLLECT_METADATA])
		*collect_md = true;
457 458 459

	if (data[IFLA_IPTUN_FWMARK])
		*fwmark = nla_get_u32(data[IFLA_IPTUN_FWMARK]);
460 461
}

462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484
/* This function returns true when ENCAP attributes are present in the nl msg */
static bool ipip_netlink_encap_parms(struct nlattr *data[],
				     struct ip_tunnel_encap *ipencap)
{
	bool ret = false;

	memset(ipencap, 0, sizeof(*ipencap));

	if (!data)
		return ret;

	if (data[IFLA_IPTUN_ENCAP_TYPE]) {
		ret = true;
		ipencap->type = nla_get_u16(data[IFLA_IPTUN_ENCAP_TYPE]);
	}

	if (data[IFLA_IPTUN_ENCAP_FLAGS]) {
		ret = true;
		ipencap->flags = nla_get_u16(data[IFLA_IPTUN_ENCAP_FLAGS]);
	}

	if (data[IFLA_IPTUN_ENCAP_SPORT]) {
		ret = true;
485
		ipencap->sport = nla_get_be16(data[IFLA_IPTUN_ENCAP_SPORT]);
486 487 488 489
	}

	if (data[IFLA_IPTUN_ENCAP_DPORT]) {
		ret = true;
490
		ipencap->dport = nla_get_be16(data[IFLA_IPTUN_ENCAP_DPORT]);
491 492 493 494 495
	}

	return ret;
}

496
static int ipip_newlink(struct net *src_net, struct net_device *dev,
497 498
			struct nlattr *tb[], struct nlattr *data[],
			struct netlink_ext_ack *extack)
499
{
500
	struct ip_tunnel *t = netdev_priv(dev);
P
Pravin B Shelar 已提交
501
	struct ip_tunnel_parm p;
502
	struct ip_tunnel_encap ipencap;
503
	__u32 fwmark = 0;
504 505 506 507 508 509 510

	if (ipip_netlink_encap_parms(data, &ipencap)) {
		int err = ip_tunnel_encap_setup(t, &ipencap);

		if (err < 0)
			return err;
	}
511

512 513
	ipip_netlink_parms(data, &p, &t->collect_md, &fwmark);
	return ip_tunnel_newlink(dev, tb, &p, fwmark);
514 515 516
}

static int ipip_changelink(struct net_device *dev, struct nlattr *tb[],
517 518
			   struct nlattr *data[],
			   struct netlink_ext_ack *extack)
519
{
520
	struct ip_tunnel *t = netdev_priv(dev);
521
	struct ip_tunnel_parm p;
522
	struct ip_tunnel_encap ipencap;
523
	bool collect_md;
524
	__u32 fwmark = t->fwmark;
525 526 527 528 529 530 531

	if (ipip_netlink_encap_parms(data, &ipencap)) {
		int err = ip_tunnel_encap_setup(t, &ipencap);

		if (err < 0)
			return err;
	}
532

533
	ipip_netlink_parms(data, &p, &collect_md, &fwmark);
534 535
	if (collect_md)
		return -EINVAL;
536 537 538 539 540

	if (((dev->flags & IFF_POINTOPOINT) && !p.iph.daddr) ||
	    (!(dev->flags & IFF_POINTOPOINT) && p.iph.daddr))
		return -EINVAL;

541
	return ip_tunnel_changelink(dev, tb, &p, fwmark);
542 543
}

544 545 546 547 548 549 550 551 552 553 554 555 556
static size_t ipip_get_size(const struct net_device *dev)
{
	return
		/* IFLA_IPTUN_LINK */
		nla_total_size(4) +
		/* IFLA_IPTUN_LOCAL */
		nla_total_size(4) +
		/* IFLA_IPTUN_REMOTE */
		nla_total_size(4) +
		/* IFLA_IPTUN_TTL */
		nla_total_size(1) +
		/* IFLA_IPTUN_TOS */
		nla_total_size(1) +
S
Simon Horman 已提交
557 558
		/* IFLA_IPTUN_PROTO */
		nla_total_size(1) +
559 560
		/* IFLA_IPTUN_PMTUDISC */
		nla_total_size(1) +
561 562 563 564 565 566 567 568
		/* IFLA_IPTUN_ENCAP_TYPE */
		nla_total_size(2) +
		/* IFLA_IPTUN_ENCAP_FLAGS */
		nla_total_size(2) +
		/* IFLA_IPTUN_ENCAP_SPORT */
		nla_total_size(2) +
		/* IFLA_IPTUN_ENCAP_DPORT */
		nla_total_size(2) +
569 570
		/* IFLA_IPTUN_COLLECT_METADATA */
		nla_total_size(0) +
571 572
		/* IFLA_IPTUN_FWMARK */
		nla_total_size(4) +
573 574 575 576 577 578 579 580 581
		0;
}

static int ipip_fill_info(struct sk_buff *skb, const struct net_device *dev)
{
	struct ip_tunnel *tunnel = netdev_priv(dev);
	struct ip_tunnel_parm *parm = &tunnel->parms;

	if (nla_put_u32(skb, IFLA_IPTUN_LINK, parm->link) ||
582 583
	    nla_put_in_addr(skb, IFLA_IPTUN_LOCAL, parm->iph.saddr) ||
	    nla_put_in_addr(skb, IFLA_IPTUN_REMOTE, parm->iph.daddr) ||
584
	    nla_put_u8(skb, IFLA_IPTUN_TTL, parm->iph.ttl) ||
585
	    nla_put_u8(skb, IFLA_IPTUN_TOS, parm->iph.tos) ||
S
Simon Horman 已提交
586
	    nla_put_u8(skb, IFLA_IPTUN_PROTO, parm->iph.protocol) ||
587
	    nla_put_u8(skb, IFLA_IPTUN_PMTUDISC,
588 589
		       !!(parm->iph.frag_off & htons(IP_DF))) ||
	    nla_put_u32(skb, IFLA_IPTUN_FWMARK, tunnel->fwmark))
590
		goto nla_put_failure;
591 592 593

	if (nla_put_u16(skb, IFLA_IPTUN_ENCAP_TYPE,
			tunnel->encap.type) ||
594 595 596 597
	    nla_put_be16(skb, IFLA_IPTUN_ENCAP_SPORT,
			 tunnel->encap.sport) ||
	    nla_put_be16(skb, IFLA_IPTUN_ENCAP_DPORT,
			 tunnel->encap.dport) ||
598
	    nla_put_u16(skb, IFLA_IPTUN_ENCAP_FLAGS,
599
			tunnel->encap.flags))
600 601
		goto nla_put_failure;

602 603 604
	if (tunnel->collect_md)
		if (nla_put_flag(skb, IFLA_IPTUN_COLLECT_METADATA))
			goto nla_put_failure;
605 606 607 608 609 610
	return 0;

nla_put_failure:
	return -EMSGSIZE;
}

611 612 613 614 615 616
static const struct nla_policy ipip_policy[IFLA_IPTUN_MAX + 1] = {
	[IFLA_IPTUN_LINK]		= { .type = NLA_U32 },
	[IFLA_IPTUN_LOCAL]		= { .type = NLA_U32 },
	[IFLA_IPTUN_REMOTE]		= { .type = NLA_U32 },
	[IFLA_IPTUN_TTL]		= { .type = NLA_U8 },
	[IFLA_IPTUN_TOS]		= { .type = NLA_U8 },
S
Simon Horman 已提交
617
	[IFLA_IPTUN_PROTO]		= { .type = NLA_U8 },
618
	[IFLA_IPTUN_PMTUDISC]		= { .type = NLA_U8 },
619 620 621 622
	[IFLA_IPTUN_ENCAP_TYPE]		= { .type = NLA_U16 },
	[IFLA_IPTUN_ENCAP_FLAGS]	= { .type = NLA_U16 },
	[IFLA_IPTUN_ENCAP_SPORT]	= { .type = NLA_U16 },
	[IFLA_IPTUN_ENCAP_DPORT]	= { .type = NLA_U16 },
623
	[IFLA_IPTUN_COLLECT_METADATA]	= { .type = NLA_FLAG },
624
	[IFLA_IPTUN_FWMARK]		= { .type = NLA_U32 },
625 626
};

627 628 629
static struct rtnl_link_ops ipip_link_ops __read_mostly = {
	.kind		= "ipip",
	.maxtype	= IFLA_IPTUN_MAX,
630
	.policy		= ipip_policy,
631
	.priv_size	= sizeof(struct ip_tunnel),
632
	.setup		= ipip_tunnel_setup,
S
Simon Horman 已提交
633
	.validate	= ipip_tunnel_validate,
634 635
	.newlink	= ipip_newlink,
	.changelink	= ipip_changelink,
P
Pravin B Shelar 已提交
636
	.dellink	= ip_tunnel_dellink,
637 638
	.get_size	= ipip_get_size,
	.fill_info	= ipip_fill_info,
639
	.get_link_net	= ip_tunnel_get_link_net,
640 641
};

642
static struct xfrm_tunnel ipip_handler __read_mostly = {
L
Linus Torvalds 已提交
643 644
	.handler	=	ipip_rcv,
	.err_handler	=	ipip_err,
H
Herbert Xu 已提交
645
	.priority	=	1,
L
Linus Torvalds 已提交
646 647
};

S
Simon Horman 已提交
648 649 650 651 652 653 654 655
#if IS_ENABLED(CONFIG_MPLS)
static struct xfrm_tunnel mplsip_handler __read_mostly = {
	.handler	=	mplsip_rcv,
	.err_handler	=	ipip_err,
	.priority	=	1,
};
#endif

656
static int __net_init ipip_init_net(struct net *net)
657
{
P
Pravin B Shelar 已提交
658
	return ip_tunnel_init_net(net, ipip_net_id, &ipip_link_ops, "tunl0");
659 660
}

661
static void __net_exit ipip_exit_batch_net(struct list_head *list_net)
662
{
663
	ip_tunnel_delete_nets(list_net, ipip_net_id, &ipip_link_ops);
664 665 666 667
}

static struct pernet_operations ipip_net_ops = {
	.init = ipip_init_net,
668
	.exit_batch = ipip_exit_batch_net,
669
	.id   = &ipip_net_id,
P
Pravin B Shelar 已提交
670
	.size = sizeof(struct ip_tunnel_net),
671 672
};

L
Linus Torvalds 已提交
673 674 675 676
static int __init ipip_init(void)
{
	int err;

S
Simon Horman 已提交
677
	pr_info("ipip: IPv4 and MPLS over IPv4 tunneling driver\n");
L
Linus Torvalds 已提交
678

679 680 681 682 683
	err = register_pernet_device(&ipip_net_ops);
	if (err < 0)
		return err;
	err = xfrm4_tunnel_register(&ipip_handler, AF_INET);
	if (err < 0) {
J
Joe Perches 已提交
684
		pr_info("%s: can't register tunnel\n", __func__);
S
Simon Horman 已提交
685 686 687 688 689 690 691
		goto xfrm_tunnel_ipip_failed;
	}
#if IS_ENABLED(CONFIG_MPLS)
	err = xfrm4_tunnel_register(&mplsip_handler, AF_MPLS);
	if (err < 0) {
		pr_info("%s: can't register tunnel\n", __func__);
		goto xfrm_tunnel_mplsip_failed;
L
Linus Torvalds 已提交
692
	}
S
Simon Horman 已提交
693
#endif
694 695 696 697 698
	err = rtnl_link_register(&ipip_link_ops);
	if (err < 0)
		goto rtnl_link_failed;

out:
L
Linus Torvalds 已提交
699
	return err;
700 701

rtnl_link_failed:
S
Simon Horman 已提交
702 703 704 705 706
#if IS_ENABLED(CONFIG_MPLS)
	xfrm4_tunnel_deregister(&mplsip_handler, AF_INET);
xfrm_tunnel_mplsip_failed:

#endif
707
	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
S
Simon Horman 已提交
708
xfrm_tunnel_ipip_failed:
709 710
	unregister_pernet_device(&ipip_net_ops);
	goto out;
L
Linus Torvalds 已提交
711 712 713 714
}

static void __exit ipip_fini(void)
{
715
	rtnl_link_unregister(&ipip_link_ops);
716
	if (xfrm4_tunnel_deregister(&ipip_handler, AF_INET))
J
Joe Perches 已提交
717
		pr_info("%s: can't deregister tunnel\n", __func__);
S
Simon Horman 已提交
718 719 720 721
#if IS_ENABLED(CONFIG_MPLS)
	if (xfrm4_tunnel_deregister(&mplsip_handler, AF_MPLS))
		pr_info("%s: can't deregister tunnel\n", __func__);
#endif
722
	unregister_pernet_device(&ipip_net_ops);
L
Linus Torvalds 已提交
723 724 725 726 727
}

module_init(ipip_init);
module_exit(ipip_fini);
MODULE_LICENSE("GPL");
728
MODULE_ALIAS_RTNL_LINK("ipip");
729
MODULE_ALIAS_NETDEV("tunl0");