ipip.c 18.6 KB
Newer Older
1
// SPDX-License-Identifier: GPL-2.0-or-later
L
Linus Torvalds 已提交
2
/*
3
 *	Linux NET3:	IP/IP protocol decoder.
L
Linus Torvalds 已提交
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
 *
 *	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.
 */

/* 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
31

L
Linus Torvalds 已提交
32 33 34 35 36 37
	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
38

39
		-Alan Cox	(alan@lxorguk.ukuu.org.uk) 21 March 95
L
Linus Torvalds 已提交
40 41 42 43 44 45 46 47

	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
48

L
Linus Torvalds 已提交
49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70
*/

/* 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
71
	find out how much more space you can allocate by calling
L
Linus Torvalds 已提交
72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87
	"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
 */

88

89
#include <linux/capability.h>
L
Linus Torvalds 已提交
90 91 92
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
93
#include <linux/slab.h>
94
#include <linux/uaccess.h>
L
Linus Torvalds 已提交
95 96 97 98 99 100 101 102
#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>
103
#include <linux/if_ether.h>
L
Linus Torvalds 已提交
104 105 106 107

#include <net/sock.h>
#include <net/ip.h>
#include <net/icmp.h>
108
#include <net/ip_tunnels.h>
L
Linus Torvalds 已提交
109 110
#include <net/inet_ecn.h>
#include <net/xfrm.h>
111 112
#include <net/net_namespace.h>
#include <net/netns/generic.h>
113
#include <net/dst_metadata.h>
L
Linus Torvalds 已提交
114

115 116 117 118
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");

119
static unsigned int ipip_net_id __read_mostly;
120

E
Eric Dumazet 已提交
121
static int ipip_tunnel_init(struct net_device *dev);
122
static struct rtnl_link_ops ipip_link_ops __read_mostly;
L
Linus Torvalds 已提交
123

H
Herbert Xu 已提交
124
static int ipip_err(struct sk_buff *skb, u32 info)
L
Linus Torvalds 已提交
125
{
126 127 128 129
	/* 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
130 131
	struct net *net = dev_net(skb->dev);
	struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
132
	const struct iphdr *iph = (const struct iphdr *)skb->data;
P
Pravin B Shelar 已提交
133 134
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
135 136 137
	struct ip_tunnel *t;
	int err = 0;

138 139 140 141 142 143 144
	t = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
			     iph->daddr, iph->saddr, 0);
	if (!t) {
		err = -ENOENT;
		goto out;
	}

145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
	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
171

172
	if (type == ICMP_DEST_UNREACH && code == ICMP_FRAG_NEEDED) {
173
		ipv4_update_pmtu(skb, net, info, t->parms.link, iph->protocol);
174 175 176
		goto out;
	}

177
	if (type == ICMP_REDIRECT) {
178
		ipv4_redirect(skb, net, t->parms.link, iph->protocol);
179 180 181
		goto out;
	}

182 183
	if (t->parms.iph.daddr == 0) {
		err = -ENOENT;
L
Linus Torvalds 已提交
184
		goto out;
185
	}
H
Herbert Xu 已提交
186

L
Linus Torvalds 已提交
187 188 189
	if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
		goto out;

190
	if (time_before(jiffies, t->err_time + IPTUNNEL_ERR_TIMEO))
L
Linus Torvalds 已提交
191 192 193 194
		t->err_count++;
	else
		t->err_count = 1;
	t->err_time = jiffies;
195

P
Pravin B Shelar 已提交
196
out:
H
Herbert Xu 已提交
197
	return err;
L
Linus Torvalds 已提交
198 199
}

S
Simon Horman 已提交
200
static const struct tnl_ptk_info ipip_tpi = {
P
Pravin B Shelar 已提交
201 202 203 204
	/* no tunnel info required for ipip. */
	.proto = htons(ETH_P_IP),
};

S
Simon Horman 已提交
205 206 207 208 209 210 211 212
#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
213
{
P
Pravin B Shelar 已提交
214 215
	struct net *net = dev_net(skb->dev);
	struct ip_tunnel_net *itn = net_generic(net, ipip_net_id);
216
	struct metadata_dst *tun_dst = NULL;
L
Linus Torvalds 已提交
217
	struct ip_tunnel *tunnel;
218
	const struct iphdr *iph;
E
Eric Dumazet 已提交
219

220
	iph = ip_hdr(skb);
P
Pravin B Shelar 已提交
221 222 223
	tunnel = ip_tunnel_lookup(itn, skb->dev->ifindex, TUNNEL_NO_KEY,
			iph->saddr, iph->daddr, 0);
	if (tunnel) {
S
Simon Horman 已提交
224 225 226 227 228 229
		const struct tnl_ptk_info *tpi;

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

230 231
		if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
			goto drop;
S
Simon Horman 已提交
232 233 234 235 236 237 238
#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))
239
			goto drop;
240 241 242 243 244 245
		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
246 247 248
	}

	return -1;
249 250 251 252

drop:
	kfree_skb(skb);
	return 0;
L
Linus Torvalds 已提交
253 254
}

S
Simon Horman 已提交
255 256 257 258 259 260 261 262 263 264 265 266
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
267 268 269 270
/*
 *	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
271 272
static netdev_tx_t ipip_tunnel_xmit(struct sk_buff *skb,
				    struct net_device *dev)
L
Linus Torvalds 已提交
273
{
274
	struct ip_tunnel *tunnel = netdev_priv(dev);
275
	const struct iphdr  *tiph = &tunnel->parms.iph;
S
Simon Horman 已提交
276 277
	u8 ipproto;

278 279 280
	if (!pskb_inet_may_pull(skb))
		goto tx_error;

S
Simon Horman 已提交
281 282 283 284 285 286 287 288 289 290 291 292
	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
293

S
Simon Horman 已提交
294
	if (tiph->protocol != ipproto && tiph->protocol != 0)
L
Linus Torvalds 已提交
295 296
		goto tx_error;

297
	if (iptunnel_handle_offloads(skb, SKB_GSO_IPXIP4))
298
		goto tx_error;
299

S
Simon Horman 已提交
300
	skb_set_inner_ipproto(skb, ipproto);
T
Tom Herbert 已提交
301

302
	if (tunnel->collect_md)
303
		ip_md_tunnel_xmit(skb, dev, ipproto, 0);
304 305
	else
		ip_tunnel_xmit(skb, dev, tiph, ipproto);
306
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
307 308

tx_error:
E
Eric Dumazet 已提交
309
	kfree_skb(skb);
310

E
Eric Dumazet 已提交
311
	dev->stats.tx_errors++;
312
	return NETDEV_TX_OK;
L
Linus Torvalds 已提交
313 314
}

S
Simon Horman 已提交
315 316 317 318 319 320 321 322 323 324 325 326 327 328
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
329
static int
P
Pravin B Shelar 已提交
330
ipip_tunnel_ioctl(struct net_device *dev, struct ifreq *ifr, int cmd)
L
Linus Torvalds 已提交
331 332 333 334
{
	int err = 0;
	struct ip_tunnel_parm p;

P
Pravin B Shelar 已提交
335 336
	if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
		return -EFAULT;
L
Linus Torvalds 已提交
337

C
Cong Wang 已提交
338
	if (cmd == SIOCADDTUNNEL || cmd == SIOCCHGTUNNEL) {
S
Simon Horman 已提交
339 340
		if (p.iph.version != 4 ||
		    !ipip_tunnel_ioctl_verify_protocol(p.iph.protocol) ||
C
Cong Wang 已提交
341 342 343 344
		    p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)))
			return -EINVAL;
	}

E
Eric Dumazet 已提交
345 346
	p.i_key = p.o_key = 0;
	p.i_flags = p.o_flags = 0;
P
Pravin B Shelar 已提交
347 348 349 350 351 352 353
	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
354 355 356
	return 0;
}

357
static const struct net_device_ops ipip_netdev_ops = {
P
Pravin B Shelar 已提交
358 359
	.ndo_init       = ipip_tunnel_init,
	.ndo_uninit     = ip_tunnel_uninit,
360 361
	.ndo_start_xmit	= ipip_tunnel_xmit,
	.ndo_do_ioctl	= ipip_tunnel_ioctl,
P
Pravin B Shelar 已提交
362 363
	.ndo_change_mtu = ip_tunnel_change_mtu,
	.ndo_get_stats64 = ip_tunnel_get_stats64,
364
	.ndo_get_iflink = ip_tunnel_get_iflink,
365 366
};

E
Eric Dumazet 已提交
367 368 369
#define IPIP_FEATURES (NETIF_F_SG |		\
		       NETIF_F_FRAGLIST |	\
		       NETIF_F_HIGHDMA |	\
E
Eric Dumazet 已提交
370
		       NETIF_F_GSO_SOFTWARE |	\
E
Eric Dumazet 已提交
371 372
		       NETIF_F_HW_CSUM)

L
Linus Torvalds 已提交
373 374
static void ipip_tunnel_setup(struct net_device *dev)
{
375
	dev->netdev_ops		= &ipip_netdev_ops;
L
Linus Torvalds 已提交
376 377 378 379

	dev->type		= ARPHRD_TUNNEL;
	dev->flags		= IFF_NOARP;
	dev->addr_len		= 4;
E
Eric Dumazet 已提交
380
	dev->features		|= NETIF_F_LLTX;
381
	netif_keep_dst(dev);
E
Eric Dumazet 已提交
382 383 384

	dev->features		|= IPIP_FEATURES;
	dev->hw_features	|= IPIP_FEATURES;
P
Pravin B Shelar 已提交
385
	ip_tunnel_setup(dev, ipip_net_id);
L
Linus Torvalds 已提交
386 387
}

E
Eric Dumazet 已提交
388
static int ipip_tunnel_init(struct net_device *dev)
L
Linus Torvalds 已提交
389
{
390
	struct ip_tunnel *tunnel = netdev_priv(dev);
L
Linus Torvalds 已提交
391 392 393 394

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

395 396
	tunnel->tun_hlen = 0;
	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
P
Pravin B Shelar 已提交
397
	return ip_tunnel_init(dev);
L
Linus Torvalds 已提交
398 399
}

400 401
static int ipip_tunnel_validate(struct nlattr *tb[], struct nlattr *data[],
				struct netlink_ext_ack *extack)
S
Simon Horman 已提交
402 403 404 405 406 407 408 409 410 411 412 413 414
{
	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;
}

415
static void ipip_netlink_parms(struct nlattr *data[],
416 417
			       struct ip_tunnel_parm *parms, bool *collect_md,
			       __u32 *fwmark)
418 419 420 421 422 423
{
	memset(parms, 0, sizeof(*parms));

	parms->iph.version = 4;
	parms->iph.protocol = IPPROTO_IPIP;
	parms->iph.ihl = 5;
424
	*collect_md = false;
425 426 427 428 429 430 431 432

	if (!data)
		return;

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

	if (data[IFLA_IPTUN_LOCAL])
433
		parms->iph.saddr = nla_get_in_addr(data[IFLA_IPTUN_LOCAL]);
434 435

	if (data[IFLA_IPTUN_REMOTE])
436
		parms->iph.daddr = nla_get_in_addr(data[IFLA_IPTUN_REMOTE]);
437 438 439 440 441 442 443 444 445 446

	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
447 448 449
	if (data[IFLA_IPTUN_PROTO])
		parms->iph.protocol = nla_get_u8(data[IFLA_IPTUN_PROTO]);

450 451
	if (!data[IFLA_IPTUN_PMTUDISC] || nla_get_u8(data[IFLA_IPTUN_PMTUDISC]))
		parms->iph.frag_off = htons(IP_DF);
452 453 454

	if (data[IFLA_IPTUN_COLLECT_METADATA])
		*collect_md = true;
455 456 457

	if (data[IFLA_IPTUN_FWMARK])
		*fwmark = nla_get_u32(data[IFLA_IPTUN_FWMARK]);
458 459
}

460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482
/* 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;
483
		ipencap->sport = nla_get_be16(data[IFLA_IPTUN_ENCAP_SPORT]);
484 485 486 487
	}

	if (data[IFLA_IPTUN_ENCAP_DPORT]) {
		ret = true;
488
		ipencap->dport = nla_get_be16(data[IFLA_IPTUN_ENCAP_DPORT]);
489 490 491 492 493
	}

	return ret;
}

494
static int ipip_newlink(struct net *src_net, struct net_device *dev,
495 496
			struct nlattr *tb[], struct nlattr *data[],
			struct netlink_ext_ack *extack)
497
{
498
	struct ip_tunnel *t = netdev_priv(dev);
P
Pravin B Shelar 已提交
499
	struct ip_tunnel_parm p;
500
	struct ip_tunnel_encap ipencap;
501
	__u32 fwmark = 0;
502 503 504 505 506 507 508

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

		if (err < 0)
			return err;
	}
509

510 511
	ipip_netlink_parms(data, &p, &t->collect_md, &fwmark);
	return ip_tunnel_newlink(dev, tb, &p, fwmark);
512 513 514
}

static int ipip_changelink(struct net_device *dev, struct nlattr *tb[],
515 516
			   struct nlattr *data[],
			   struct netlink_ext_ack *extack)
517
{
518
	struct ip_tunnel *t = netdev_priv(dev);
519
	struct ip_tunnel_parm p;
520
	struct ip_tunnel_encap ipencap;
521
	bool collect_md;
522
	__u32 fwmark = t->fwmark;
523 524 525 526 527 528 529

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

		if (err < 0)
			return err;
	}
530

531
	ipip_netlink_parms(data, &p, &collect_md, &fwmark);
532 533
	if (collect_md)
		return -EINVAL;
534 535 536 537 538

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

539
	return ip_tunnel_changelink(dev, tb, &p, fwmark);
540 541
}

542 543 544 545 546 547 548 549 550 551 552 553 554
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
555 556
		/* IFLA_IPTUN_PROTO */
		nla_total_size(1) +
557 558
		/* IFLA_IPTUN_PMTUDISC */
		nla_total_size(1) +
559 560 561 562 563 564 565 566
		/* 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) +
567 568
		/* IFLA_IPTUN_COLLECT_METADATA */
		nla_total_size(0) +
569 570
		/* IFLA_IPTUN_FWMARK */
		nla_total_size(4) +
571 572 573 574 575 576 577 578 579
		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) ||
580 581
	    nla_put_in_addr(skb, IFLA_IPTUN_LOCAL, parm->iph.saddr) ||
	    nla_put_in_addr(skb, IFLA_IPTUN_REMOTE, parm->iph.daddr) ||
582
	    nla_put_u8(skb, IFLA_IPTUN_TTL, parm->iph.ttl) ||
583
	    nla_put_u8(skb, IFLA_IPTUN_TOS, parm->iph.tos) ||
S
Simon Horman 已提交
584
	    nla_put_u8(skb, IFLA_IPTUN_PROTO, parm->iph.protocol) ||
585
	    nla_put_u8(skb, IFLA_IPTUN_PMTUDISC,
586 587
		       !!(parm->iph.frag_off & htons(IP_DF))) ||
	    nla_put_u32(skb, IFLA_IPTUN_FWMARK, tunnel->fwmark))
588
		goto nla_put_failure;
589 590 591

	if (nla_put_u16(skb, IFLA_IPTUN_ENCAP_TYPE,
			tunnel->encap.type) ||
592 593 594 595
	    nla_put_be16(skb, IFLA_IPTUN_ENCAP_SPORT,
			 tunnel->encap.sport) ||
	    nla_put_be16(skb, IFLA_IPTUN_ENCAP_DPORT,
			 tunnel->encap.dport) ||
596
	    nla_put_u16(skb, IFLA_IPTUN_ENCAP_FLAGS,
597
			tunnel->encap.flags))
598 599
		goto nla_put_failure;

600 601 602
	if (tunnel->collect_md)
		if (nla_put_flag(skb, IFLA_IPTUN_COLLECT_METADATA))
			goto nla_put_failure;
603 604 605 606 607 608
	return 0;

nla_put_failure:
	return -EMSGSIZE;
}

609 610 611 612 613 614
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
615
	[IFLA_IPTUN_PROTO]		= { .type = NLA_U8 },
616
	[IFLA_IPTUN_PMTUDISC]		= { .type = NLA_U8 },
617 618 619 620
	[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 },
621
	[IFLA_IPTUN_COLLECT_METADATA]	= { .type = NLA_FLAG },
622
	[IFLA_IPTUN_FWMARK]		= { .type = NLA_U32 },
623 624
};

625 626 627
static struct rtnl_link_ops ipip_link_ops __read_mostly = {
	.kind		= "ipip",
	.maxtype	= IFLA_IPTUN_MAX,
628
	.policy		= ipip_policy,
629
	.priv_size	= sizeof(struct ip_tunnel),
630
	.setup		= ipip_tunnel_setup,
S
Simon Horman 已提交
631
	.validate	= ipip_tunnel_validate,
632 633
	.newlink	= ipip_newlink,
	.changelink	= ipip_changelink,
P
Pravin B Shelar 已提交
634
	.dellink	= ip_tunnel_dellink,
635 636
	.get_size	= ipip_get_size,
	.fill_info	= ipip_fill_info,
637
	.get_link_net	= ip_tunnel_get_link_net,
638 639
};

640
static struct xfrm_tunnel ipip_handler __read_mostly = {
L
Linus Torvalds 已提交
641 642
	.handler	=	ipip_rcv,
	.err_handler	=	ipip_err,
H
Herbert Xu 已提交
643
	.priority	=	1,
L
Linus Torvalds 已提交
644 645
};

S
Simon Horman 已提交
646 647 648 649 650 651 652 653
#if IS_ENABLED(CONFIG_MPLS)
static struct xfrm_tunnel mplsip_handler __read_mostly = {
	.handler	=	mplsip_rcv,
	.err_handler	=	ipip_err,
	.priority	=	1,
};
#endif

654
static int __net_init ipip_init_net(struct net *net)
655
{
P
Pravin B Shelar 已提交
656
	return ip_tunnel_init_net(net, ipip_net_id, &ipip_link_ops, "tunl0");
657 658
}

659
static void __net_exit ipip_exit_batch_net(struct list_head *list_net)
660
{
661
	ip_tunnel_delete_nets(list_net, ipip_net_id, &ipip_link_ops);
662 663 664 665
}

static struct pernet_operations ipip_net_ops = {
	.init = ipip_init_net,
666
	.exit_batch = ipip_exit_batch_net,
667
	.id   = &ipip_net_id,
P
Pravin B Shelar 已提交
668
	.size = sizeof(struct ip_tunnel_net),
669 670
};

L
Linus Torvalds 已提交
671 672 673 674
static int __init ipip_init(void)
{
	int err;

S
Simon Horman 已提交
675
	pr_info("ipip: IPv4 and MPLS over IPv4 tunneling driver\n");
L
Linus Torvalds 已提交
676

677 678 679 680 681
	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
682
		pr_info("%s: can't register tunnel\n", __func__);
S
Simon Horman 已提交
683 684 685 686 687 688 689
		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
690
	}
S
Simon Horman 已提交
691
#endif
692 693 694 695 696
	err = rtnl_link_register(&ipip_link_ops);
	if (err < 0)
		goto rtnl_link_failed;

out:
L
Linus Torvalds 已提交
697
	return err;
698 699

rtnl_link_failed:
S
Simon Horman 已提交
700 701 702 703 704
#if IS_ENABLED(CONFIG_MPLS)
	xfrm4_tunnel_deregister(&mplsip_handler, AF_INET);
xfrm_tunnel_mplsip_failed:

#endif
705
	xfrm4_tunnel_deregister(&ipip_handler, AF_INET);
S
Simon Horman 已提交
706
xfrm_tunnel_ipip_failed:
707 708
	unregister_pernet_device(&ipip_net_ops);
	goto out;
L
Linus Torvalds 已提交
709 710 711 712
}

static void __exit ipip_fini(void)
{
713
	rtnl_link_unregister(&ipip_link_ops);
714
	if (xfrm4_tunnel_deregister(&ipip_handler, AF_INET))
J
Joe Perches 已提交
715
		pr_info("%s: can't deregister tunnel\n", __func__);
S
Simon Horman 已提交
716 717 718 719
#if IS_ENABLED(CONFIG_MPLS)
	if (xfrm4_tunnel_deregister(&mplsip_handler, AF_MPLS))
		pr_info("%s: can't deregister tunnel\n", __func__);
#endif
720
	unregister_pernet_device(&ipip_net_ops);
L
Linus Torvalds 已提交
721 722 723 724 725
}

module_init(ipip_init);
module_exit(ipip_fini);
MODULE_LICENSE("GPL");
726
MODULE_ALIAS_RTNL_LINK("ipip");
727
MODULE_ALIAS_NETDEV("tunl0");