sit.c 21.3 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5
/*
 *	IPv6 over IPv4 tunnel device - Simple Internet Transition (SIT)
 *	Linux INET6 implementation
 *
 *	Authors:
6
 *	Pedro Roque		<roque@di.fc.ul.pt>
L
Linus Torvalds 已提交
7 8 9 10 11 12 13 14 15 16 17 18
 *	Alexey Kuznetsov	<kuznet@ms2.inr.ac.ru>
 *
 *	$Id: sit.c,v 1.53 2001/09/25 05:09:53 davem Exp $
 *
 *	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.
 *
 *	Changes:
 * Roger Venning <r.venning@telstra.com>:	6to4 support
 * Nate Thompson <nate@thebog.net>:		6to4 support
F
Fred L. Templin 已提交
19
 * Fred L. Templin <fltemplin@acm.org>:		isatap support
L
Linus Torvalds 已提交
20 21 22
 */

#include <linux/module.h>
23
#include <linux/capability.h>
L
Linus Torvalds 已提交
24 25 26 27 28 29 30 31 32 33 34 35
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/in6.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/icmp.h>
#include <asm/uaccess.h>
#include <linux/init.h>
#include <linux/netfilter_ipv4.h>
36
#include <linux/if_ether.h>
L
Linus Torvalds 已提交
37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62

#include <net/sock.h>
#include <net/snmp.h>

#include <net/ipv6.h>
#include <net/protocol.h>
#include <net/transp_v6.h>
#include <net/ip6_fib.h>
#include <net/ip6_route.h>
#include <net/ndisc.h>
#include <net/addrconf.h>
#include <net/ip.h>
#include <net/udp.h>
#include <net/icmp.h>
#include <net/ipip.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/dsfield.h>

/*
   This version of net/ipv6/sit.c is cloned of net/ipv4/ip_gre.c

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

#define HASH_SIZE  16
A
Al Viro 已提交
63
#define HASH(addr) (((__force u32)addr^((__force u32)addr>>4))&0xF)
L
Linus Torvalds 已提交
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78

static int ipip6_fb_tunnel_init(struct net_device *dev);
static int ipip6_tunnel_init(struct net_device *dev);
static void ipip6_tunnel_setup(struct net_device *dev);

static struct net_device *ipip6_fb_tunnel_dev;

static struct ip_tunnel *tunnels_r_l[HASH_SIZE];
static struct ip_tunnel *tunnels_r[HASH_SIZE];
static struct ip_tunnel *tunnels_l[HASH_SIZE];
static struct ip_tunnel *tunnels_wc[1];
static struct ip_tunnel **tunnels[4] = { tunnels_wc, tunnels_l, tunnels_r, tunnels_r_l };

static DEFINE_RWLOCK(ipip6_lock);

A
Al Viro 已提交
79
static struct ip_tunnel * ipip6_tunnel_lookup(__be32 remote, __be32 local)
L
Linus Torvalds 已提交
80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102
{
	unsigned h0 = HASH(remote);
	unsigned h1 = HASH(local);
	struct ip_tunnel *t;

	for (t = tunnels_r_l[h0^h1]; t; t = t->next) {
		if (local == t->parms.iph.saddr &&
		    remote == t->parms.iph.daddr && (t->dev->flags&IFF_UP))
			return t;
	}
	for (t = tunnels_r[h0]; t; t = t->next) {
		if (remote == t->parms.iph.daddr && (t->dev->flags&IFF_UP))
			return t;
	}
	for (t = tunnels_l[h1]; t; t = t->next) {
		if (local == t->parms.iph.saddr && (t->dev->flags&IFF_UP))
			return t;
	}
	if ((t = tunnels_wc[0]) != NULL && (t->dev->flags&IFF_UP))
		return t;
	return NULL;
}

103
static struct ip_tunnel **__ipip6_bucket(struct ip_tunnel_parm *parms)
L
Linus Torvalds 已提交
104
{
105 106
	__be32 remote = parms->iph.daddr;
	__be32 local = parms->iph.saddr;
L
Linus Torvalds 已提交
107 108 109 110 111 112 113 114 115 116 117 118 119 120
	unsigned h = 0;
	int prio = 0;

	if (remote) {
		prio |= 2;
		h ^= HASH(remote);
	}
	if (local) {
		prio |= 1;
		h ^= HASH(local);
	}
	return &tunnels[prio][h];
}

121 122 123 124 125
static inline struct ip_tunnel **ipip6_bucket(struct ip_tunnel *t)
{
	return __ipip6_bucket(&t->parms);
}

L
Linus Torvalds 已提交
126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151
static void ipip6_tunnel_unlink(struct ip_tunnel *t)
{
	struct ip_tunnel **tp;

	for (tp = ipip6_bucket(t); *tp; tp = &(*tp)->next) {
		if (t == *tp) {
			write_lock_bh(&ipip6_lock);
			*tp = t->next;
			write_unlock_bh(&ipip6_lock);
			break;
		}
	}
}

static void ipip6_tunnel_link(struct ip_tunnel *t)
{
	struct ip_tunnel **tp = ipip6_bucket(t);

	t->next = *tp;
	write_lock_bh(&ipip6_lock);
	*tp = t;
	write_unlock_bh(&ipip6_lock);
}

static struct ip_tunnel * ipip6_tunnel_locate(struct ip_tunnel_parm *parms, int create)
{
A
Al Viro 已提交
152 153
	__be32 remote = parms->iph.daddr;
	__be32 local = parms->iph.saddr;
L
Linus Torvalds 已提交
154 155 156 157
	struct ip_tunnel *t, **tp, *nt;
	struct net_device *dev;
	char name[IFNAMSIZ];

158
	for (tp = __ipip6_bucket(parms); (t = *tp) != NULL; tp = &t->next) {
L
Linus Torvalds 已提交
159 160 161 162 163 164 165 166
		if (local == t->parms.iph.saddr && remote == t->parms.iph.daddr)
			return t;
	}
	if (!create)
		goto failed;

	if (parms->name[0])
		strlcpy(name, parms->name, IFNAMSIZ);
167 168
	else
		sprintf(name, "sit%%d");
L
Linus Torvalds 已提交
169 170 171 172 173

	dev = alloc_netdev(sizeof(*t), name, ipip6_tunnel_setup);
	if (dev == NULL)
		return NULL;

174 175 176 177 178
	if (strchr(name, '%')) {
		if (dev_alloc_name(dev, name) < 0)
			goto failed_free;
	}

179
	nt = netdev_priv(dev);
L
Linus Torvalds 已提交
180 181 182
	dev->init = ipip6_tunnel_init;
	nt->parms = *parms;

F
Fred L. Templin 已提交
183 184 185
	if (parms->i_flags & SIT_ISATAP)
		dev->priv_flags |= IFF_ISATAP;

186 187
	if (register_netdevice(dev) < 0)
		goto failed_free;
L
Linus Torvalds 已提交
188 189 190 191 192 193

	dev_hold(dev);

	ipip6_tunnel_link(nt);
	return nt;

194 195
failed_free:
	free_netdev(dev);
L
Linus Torvalds 已提交
196 197 198 199 200 201 202 203 204 205 206 207
failed:
	return NULL;
}

static void ipip6_tunnel_uninit(struct net_device *dev)
{
	if (dev == ipip6_fb_tunnel_dev) {
		write_lock_bh(&ipip6_lock);
		tunnels_wc[0] = NULL;
		write_unlock_bh(&ipip6_lock);
		dev_put(dev);
	} else {
208
		ipip6_tunnel_unlink(netdev_priv(dev));
L
Linus Torvalds 已提交
209 210 211 212 213
		dev_put(dev);
	}
}


214
static int ipip6_err(struct sk_buff *skb, u32 info)
L
Linus Torvalds 已提交
215 216 217 218 219 220 221 222
{
#ifndef I_WISH_WORLD_WERE_PERFECT

/* It is not :-( 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.
 */
	struct iphdr *iph = (struct iphdr*)skb->data;
223 224
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
L
Linus Torvalds 已提交
225
	struct ip_tunnel *t;
226
	int err;
L
Linus Torvalds 已提交
227 228 229 230

	switch (type) {
	default:
	case ICMP_PARAMETERPROB:
231
		return 0;
L
Linus Torvalds 已提交
232 233 234 235 236 237

	case ICMP_DEST_UNREACH:
		switch (code) {
		case ICMP_SR_FAILED:
		case ICMP_PORT_UNREACH:
			/* Impossible event. */
238
			return 0;
L
Linus Torvalds 已提交
239 240
		case ICMP_FRAG_NEEDED:
			/* Soft state for pmtu is maintained by IP core. */
241
			return 0;
L
Linus Torvalds 已提交
242 243 244 245 246 247 248 249 250 251
		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)
252
			return 0;
L
Linus Torvalds 已提交
253 254 255
		break;
	}

256 257
	err = -ENOENT;

L
Linus Torvalds 已提交
258 259 260 261
	read_lock(&ipip6_lock);
	t = ipip6_tunnel_lookup(iph->daddr, iph->saddr);
	if (t == NULL || t->parms.iph.daddr == 0)
		goto out;
262 263

	err = 0;
L
Linus Torvalds 已提交
264 265 266 267 268 269 270 271 272 273
	if (t->parms.iph.ttl == 0 && type == ICMP_TIME_EXCEEDED)
		goto out;

	if (jiffies - t->err_time < IPTUNNEL_ERR_TIMEO)
		t->err_count++;
	else
		t->err_count = 1;
	t->err_time = jiffies;
out:
	read_unlock(&ipip6_lock);
274
	return err;
L
Linus Torvalds 已提交
275 276 277 278
#else
	struct iphdr *iph = (struct iphdr*)dp;
	int hlen = iph->ihl<<2;
	struct ipv6hdr *iph6;
279 280
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
L
Linus Torvalds 已提交
281 282 283 284 285 286 287 288 289 290 291 292 293 294
	int rel_type = 0;
	int rel_code = 0;
	int rel_info = 0;
	struct sk_buff *skb2;
	struct rt6_info *rt6i;

	if (len < hlen + sizeof(struct ipv6hdr))
		return;
	iph6 = (struct ipv6hdr*)(dp + hlen);

	switch (type) {
	default:
		return;
	case ICMP_PARAMETERPROB:
295
		if (icmp_hdr(skb)->un.gateway < hlen)
L
Linus Torvalds 已提交
296 297 298 299 300 301
			return;

		/* So... This guy found something strange INSIDE encapsulated
		   packet. Well, he is fool, but what can we do ?
		 */
		rel_type = ICMPV6_PARAMPROB;
302
		rel_info = icmp_hdr(skb)->un.gateway - hlen;
L
Linus Torvalds 已提交
303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334
		break;

	case ICMP_DEST_UNREACH:
		switch (code) {
		case ICMP_SR_FAILED:
		case ICMP_PORT_UNREACH:
			/* Impossible event. */
			return;
		case ICMP_FRAG_NEEDED:
			/* Too complicated case ... */
			return;
		default:
			/* All others are translated to HOST_UNREACH.
			   rfc2003 contains "deep thoughts" about NET_UNREACH,
			   I believe, it is just ether pollution. --ANK
			 */
			rel_type = ICMPV6_DEST_UNREACH;
			rel_code = ICMPV6_ADDR_UNREACH;
			break;
		}
		break;
	case ICMP_TIME_EXCEEDED:
		if (code != ICMP_EXC_TTL)
			return;
		rel_type = ICMPV6_TIME_EXCEED;
		rel_code = ICMPV6_EXC_HOPLIMIT;
		break;
	}

	/* Prepare fake skb to feed it to icmpv6_send */
	skb2 = skb_clone(skb, GFP_ATOMIC);
	if (skb2 == NULL)
335
		return 0;
L
Linus Torvalds 已提交
336 337 338
	dst_release(skb2->dst);
	skb2->dst = NULL;
	skb_pull(skb2, skb->data - (u8*)iph6);
339
	skb_reset_network_header(skb2);
L
Linus Torvalds 已提交
340 341 342 343 344 345 346 347 348

	/* Try to guess incoming interface */
	rt6i = rt6_lookup(&iph6->saddr, NULL, NULL, 0);
	if (rt6i && rt6i->rt6i_dev) {
		skb2->dev = rt6i->rt6i_dev;

		rt6i = rt6_lookup(&iph6->daddr, &iph6->saddr, NULL, 0);

		if (rt6i && rt6i->rt6i_dev && rt6i->rt6i_dev->type == ARPHRD_SIT) {
349
			struct ip_tunnel *t = netdev_priv(rt6i->rt6i_dev);
L
Linus Torvalds 已提交
350 351 352 353 354 355 356 357
			if (rel_type == ICMPV6_TIME_EXCEED && t->parms.iph.ttl) {
				rel_type = ICMPV6_DEST_UNREACH;
				rel_code = ICMPV6_ADDR_UNREACH;
			}
			icmpv6_send(skb2, rel_type, rel_code, rel_info, skb2->dev);
		}
	}
	kfree_skb(skb2);
358
	return 0;
L
Linus Torvalds 已提交
359 360 361 362 363 364
#endif
}

static inline void ipip6_ecn_decapsulate(struct iphdr *iph, struct sk_buff *skb)
{
	if (INET_ECN_is_ce(iph->tos))
365
		IP6_ECN_set_ce(ipv6_hdr(skb));
L
Linus Torvalds 已提交
366 367
}

F
Fred L. Templin 已提交
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409
/* ISATAP (RFC4214) - check source address */
static int
isatap_srcok(struct sk_buff *skb, struct iphdr *iph, struct net_device *dev)
{
	struct neighbour *neigh;
	struct dst_entry *dst;
	struct rt6_info *rt;
	struct flowi fl;
	struct in6_addr *addr6;
	struct in6_addr rtr;
	struct ipv6hdr *iph6;
	int ok = 0;

	/* from onlink default router */
	ipv6_addr_set(&rtr,  htonl(0xFE800000), 0, 0, 0);
	ipv6_isatap_eui64(rtr.s6_addr + 8, iph->saddr);
	if ((rt = rt6_get_dflt_router(&rtr, dev))) {
		dst_release(&rt->u.dst);
		return 1;
	}

	iph6 = ipv6_hdr(skb);
	memset(&fl, 0, sizeof(fl));
	fl.proto = iph6->nexthdr;
	ipv6_addr_copy(&fl.fl6_dst, &iph6->saddr);
	fl.oif = dev->ifindex;
	security_skb_classify_flow(skb, &fl);

	dst = ip6_route_output(NULL, &fl);
	if (!dst->error && (dst->dev == dev) && (neigh = dst->neighbour)) {

		addr6 = (struct in6_addr*)&neigh->primary_key;

		/* from correct previous hop */
		if (ipv6_addr_is_isatap(addr6) &&
		    (addr6->s6_addr32[3] == iph->saddr))
			ok = 1;
	}
	dst_release(dst);
	return ok;
}

L
Linus Torvalds 已提交
410 411 412 413 414 415 416 417
static int ipip6_rcv(struct sk_buff *skb)
{
	struct iphdr *iph;
	struct ip_tunnel *tunnel;

	if (!pskb_may_pull(skb, sizeof(struct ipv6hdr)))
		goto out;

418
	iph = ip_hdr(skb);
L
Linus Torvalds 已提交
419 420 421 422

	read_lock(&ipip6_lock);
	if ((tunnel = ipip6_tunnel_lookup(iph->saddr, iph->daddr)) != NULL) {
		secpath_reset(skb);
423
		skb->mac_header = skb->network_header;
424
		skb_reset_network_header(skb);
425
		IPCB(skb)->flags = 0;
L
Linus Torvalds 已提交
426 427
		skb->protocol = htons(ETH_P_IPV6);
		skb->pkt_type = PACKET_HOST;
F
Fred L. Templin 已提交
428 429 430 431 432 433 434 435

		if ((tunnel->dev->priv_flags & IFF_ISATAP) &&
		    !isatap_srcok(skb, iph, tunnel->dev)) {
			tunnel->stat.rx_errors++;
			read_unlock(&ipip6_lock);
			kfree_skb(skb);
			return 0;
		}
L
Linus Torvalds 已提交
436 437 438 439 440 441 442 443 444 445 446 447
		tunnel->stat.rx_packets++;
		tunnel->stat.rx_bytes += skb->len;
		skb->dev = tunnel->dev;
		dst_release(skb->dst);
		skb->dst = NULL;
		nf_reset(skb);
		ipip6_ecn_decapsulate(iph, skb);
		netif_rx(skb);
		read_unlock(&ipip6_lock);
		return 0;
	}

448
	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);
L
Linus Torvalds 已提交
449 450 451 452 453 454 455 456 457
	kfree_skb(skb);
	read_unlock(&ipip6_lock);
out:
	return 0;
}

/* Returns the embedded IPv4 address if the IPv6 address
   comes from 6to4 (RFC 3056) addr space */

A
Al Viro 已提交
458
static inline __be32 try_6to4(struct in6_addr *v6dst)
L
Linus Torvalds 已提交
459
{
A
Al Viro 已提交
460
	__be32 dst = 0;
L
Linus Torvalds 已提交
461 462

	if (v6dst->s6_addr16[0] == htons(0x2002)) {
463
		/* 6to4 v6 addr has 16 bits prefix, 32 v4addr, 16 SLA, ... */
L
Linus Torvalds 已提交
464 465 466 467 468 469 470 471 472 473 474 475
		memcpy(&dst, &v6dst->s6_addr16[1], 4);
	}
	return dst;
}

/*
 *	This function assumes it is being called from dev_queue_xmit()
 *	and that skb is filled properly by that function.
 */

static int ipip6_tunnel_xmit(struct sk_buff *skb, struct net_device *dev)
{
476
	struct ip_tunnel *tunnel = netdev_priv(dev);
L
Linus Torvalds 已提交
477 478
	struct net_device_stats *stats = &tunnel->stat;
	struct iphdr  *tiph = &tunnel->parms.iph;
479
	struct ipv6hdr *iph6 = ipv6_hdr(skb);
L
Linus Torvalds 已提交
480 481 482 483
	u8     tos = tunnel->parms.iph.tos;
	struct rtable *rt;     			/* Route to the other host */
	struct net_device *tdev;			/* Device to other host */
	struct iphdr  *iph;			/* Our new IP header */
484
	unsigned int max_headroom;		/* The extra header space needed */
A
Al Viro 已提交
485
	__be32 dst = tiph->daddr;
L
Linus Torvalds 已提交
486
	int    mtu;
487
	struct in6_addr *addr6;
L
Linus Torvalds 已提交
488 489 490 491 492 493 494 495 496 497
	int addr_type;

	if (tunnel->recursion++) {
		tunnel->stat.collisions++;
		goto tx_error;
	}

	if (skb->protocol != htons(ETH_P_IPV6))
		goto tx_error;

F
Fred L. Templin 已提交
498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520
	/* ISATAP (RFC4214) - must come before 6to4 */
	if (dev->priv_flags & IFF_ISATAP) {
		struct neighbour *neigh = NULL;

		if (skb->dst)
			neigh = skb->dst->neighbour;

		if (neigh == NULL) {
			if (net_ratelimit())
				printk(KERN_DEBUG "sit: nexthop == NULL\n");
			goto tx_error;
		}

		addr6 = (struct in6_addr*)&neigh->primary_key;
		addr_type = ipv6_addr_type(addr6);

		if ((addr_type & IPV6_ADDR_UNICAST) &&
		     ipv6_addr_is_isatap(addr6))
			dst = addr6->s6_addr32[3];
		else
			goto tx_error;
	}

L
Linus Torvalds 已提交
521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539
	if (!dst)
		dst = try_6to4(&iph6->daddr);

	if (!dst) {
		struct neighbour *neigh = NULL;

		if (skb->dst)
			neigh = skb->dst->neighbour;

		if (neigh == NULL) {
			if (net_ratelimit())
				printk(KERN_DEBUG "sit: nexthop == NULL\n");
			goto tx_error;
		}

		addr6 = (struct in6_addr*)&neigh->primary_key;
		addr_type = ipv6_addr_type(addr6);

		if (addr_type == IPV6_ADDR_ANY) {
540
			addr6 = &ipv6_hdr(skb)->daddr;
L
Linus Torvalds 已提交
541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
			addr_type = ipv6_addr_type(addr6);
		}

		if ((addr_type & IPV6_ADDR_COMPATv4) == 0)
			goto tx_error_icmp;

		dst = addr6->s6_addr32[3];
	}

	{
		struct flowi fl = { .nl_u = { .ip4_u =
					      { .daddr = dst,
						.saddr = tiph->saddr,
						.tos = RT_TOS(tos) } },
				    .oif = tunnel->parms.link,
				    .proto = IPPROTO_IPV6 };
557
		if (ip_route_output_key(&init_net, &rt, &fl)) {
L
Linus Torvalds 已提交
558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608
			tunnel->stat.tx_carrier_errors++;
			goto tx_error_icmp;
		}
	}
	if (rt->rt_type != RTN_UNICAST) {
		ip_rt_put(rt);
		tunnel->stat.tx_carrier_errors++;
		goto tx_error_icmp;
	}
	tdev = rt->u.dst.dev;

	if (tdev == dev) {
		ip_rt_put(rt);
		tunnel->stat.collisions++;
		goto tx_error;
	}

	if (tiph->frag_off)
		mtu = dst_mtu(&rt->u.dst) - sizeof(struct iphdr);
	else
		mtu = skb->dst ? dst_mtu(skb->dst) : dev->mtu;

	if (mtu < 68) {
		tunnel->stat.collisions++;
		ip_rt_put(rt);
		goto tx_error;
	}
	if (mtu < IPV6_MIN_MTU)
		mtu = IPV6_MIN_MTU;
	if (tunnel->parms.iph.daddr && skb->dst)
		skb->dst->ops->update_pmtu(skb->dst, mtu);

	if (skb->len > mtu) {
		icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu, dev);
		ip_rt_put(rt);
		goto tx_error;
	}

	if (tunnel->err_count > 0) {
		if (jiffies - tunnel->err_time < IPTUNNEL_ERR_TIMEO) {
			tunnel->err_count--;
			dst_link_failure(skb);
		} else
			tunnel->err_count = 0;
	}

	/*
	 * Okay, now see if we can stuff it in the buffer as-is.
	 */
	max_headroom = LL_RESERVED_SPACE(tdev)+sizeof(struct iphdr);

609 610
	if (skb_headroom(skb) < max_headroom || skb_shared(skb) ||
	    (skb_cloned(skb) && !skb_clone_writable(skb, 0))) {
L
Linus Torvalds 已提交
611 612 613
		struct sk_buff *new_skb = skb_realloc_headroom(skb, max_headroom);
		if (!new_skb) {
			ip_rt_put(rt);
614
			stats->tx_dropped++;
L
Linus Torvalds 已提交
615 616 617 618 619 620 621 622
			dev_kfree_skb(skb);
			tunnel->recursion--;
			return 0;
		}
		if (skb->sk)
			skb_set_owner_w(new_skb, skb->sk);
		dev_kfree_skb(skb);
		skb = new_skb;
623
		iph6 = ipv6_hdr(skb);
L
Linus Torvalds 已提交
624 625
	}

626
	skb->transport_header = skb->network_header;
627 628
	skb_push(skb, sizeof(struct iphdr));
	skb_reset_network_header(skb);
L
Linus Torvalds 已提交
629
	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
630
	IPCB(skb)->flags = 0;
L
Linus Torvalds 已提交
631 632 633 634 635 636 637
	dst_release(skb->dst);
	skb->dst = &rt->u.dst;

	/*
	 *	Push down and install the IPIP header.
	 */

638
	iph 			=	ip_hdr(skb);
L
Linus Torvalds 已提交
639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668
	iph->version		=	4;
	iph->ihl		=	sizeof(struct iphdr)>>2;
	if (mtu > IPV6_MIN_MTU)
		iph->frag_off	=	htons(IP_DF);
	else
		iph->frag_off	=	0;

	iph->protocol		=	IPPROTO_IPV6;
	iph->tos		=	INET_ECN_encapsulate(tos, ipv6_get_dsfield(iph6));
	iph->daddr		=	rt->rt_dst;
	iph->saddr		=	rt->rt_src;

	if ((iph->ttl = tiph->ttl) == 0)
		iph->ttl	=	iph6->hop_limit;

	nf_reset(skb);

	IPTUNNEL_XMIT();
	tunnel->recursion--;
	return 0;

tx_error_icmp:
	dst_link_failure(skb);
tx_error:
	stats->tx_errors++;
	dev_kfree_skb(skb);
	tunnel->recursion--;
	return 0;
}

669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685
static void ipip6_tunnel_bind_dev(struct net_device *dev)
{
	struct net_device *tdev = NULL;
	struct ip_tunnel *tunnel;
	struct iphdr *iph;

	tunnel = netdev_priv(dev);
	iph = &tunnel->parms.iph;

	if (iph->daddr) {
		struct flowi fl = { .nl_u = { .ip4_u =
					      { .daddr = iph->daddr,
						.saddr = iph->saddr,
						.tos = RT_TOS(iph->tos) } },
				    .oif = tunnel->parms.link,
				    .proto = IPPROTO_IPV6 };
		struct rtable *rt;
686
		if (!ip_route_output_key(&init_net, &rt, &fl)) {
687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704
			tdev = rt->u.dst.dev;
			ip_rt_put(rt);
		}
		dev->flags |= IFF_POINTOPOINT;
	}

	if (!tdev && tunnel->parms.link)
		tdev = __dev_get_by_index(&init_net, tunnel->parms.link);

	if (tdev) {
		dev->hard_header_len = tdev->hard_header_len + sizeof(struct iphdr);
		dev->mtu = tdev->mtu - sizeof(struct iphdr);
		if (dev->mtu < IPV6_MIN_MTU)
			dev->mtu = IPV6_MIN_MTU;
	}
	dev->iflink = tunnel->parms.link;
}

L
Linus Torvalds 已提交
705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722
static int
ipip6_tunnel_ioctl (struct net_device *dev, struct ifreq *ifr, int cmd)
{
	int err = 0;
	struct ip_tunnel_parm p;
	struct ip_tunnel *t;

	switch (cmd) {
	case SIOCGETTUNNEL:
		t = NULL;
		if (dev == ipip6_fb_tunnel_dev) {
			if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) {
				err = -EFAULT;
				break;
			}
			t = ipip6_tunnel_locate(&p, 0);
		}
		if (t == NULL)
723
			t = netdev_priv(dev);
L
Linus Torvalds 已提交
724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759
		memcpy(&p, &t->parms, sizeof(p));
		if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
			err = -EFAULT;
		break;

	case SIOCADDTUNNEL:
	case SIOCCHGTUNNEL:
		err = -EPERM;
		if (!capable(CAP_NET_ADMIN))
			goto done;

		err = -EFAULT;
		if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
			goto done;

		err = -EINVAL;
		if (p.iph.version != 4 || p.iph.protocol != IPPROTO_IPV6 ||
		    p.iph.ihl != 5 || (p.iph.frag_off&htons(~IP_DF)))
			goto done;
		if (p.iph.ttl)
			p.iph.frag_off |= htons(IP_DF);

		t = ipip6_tunnel_locate(&p, cmd == SIOCADDTUNNEL);

		if (dev != ipip6_fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
			if (t != NULL) {
				if (t->dev != dev) {
					err = -EEXIST;
					break;
				}
			} else {
				if (((dev->flags&IFF_POINTOPOINT) && !p.iph.daddr) ||
				    (!(dev->flags&IFF_POINTOPOINT) && p.iph.daddr)) {
					err = -EINVAL;
					break;
				}
760
				t = netdev_priv(dev);
L
Linus Torvalds 已提交
761 762 763 764 765 766 767 768 769 770 771 772 773 774 775
				ipip6_tunnel_unlink(t);
				t->parms.iph.saddr = p.iph.saddr;
				t->parms.iph.daddr = p.iph.daddr;
				memcpy(dev->dev_addr, &p.iph.saddr, 4);
				memcpy(dev->broadcast, &p.iph.daddr, 4);
				ipip6_tunnel_link(t);
				netdev_state_change(dev);
			}
		}

		if (t) {
			err = 0;
			if (cmd == SIOCCHGTUNNEL) {
				t->parms.iph.ttl = p.iph.ttl;
				t->parms.iph.tos = p.iph.tos;
776 777 778 779 780
				if (t->parms.link != p.link) {
					t->parms.link = p.link;
					ipip6_tunnel_bind_dev(dev);
					netdev_state_change(dev);
				}
L
Linus Torvalds 已提交
781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800
			}
			if (copy_to_user(ifr->ifr_ifru.ifru_data, &t->parms, sizeof(p)))
				err = -EFAULT;
		} else
			err = (cmd == SIOCADDTUNNEL ? -ENOBUFS : -ENOENT);
		break;

	case SIOCDELTUNNEL:
		err = -EPERM;
		if (!capable(CAP_NET_ADMIN))
			goto done;

		if (dev == ipip6_fb_tunnel_dev) {
			err = -EFAULT;
			if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
				goto done;
			err = -ENOENT;
			if ((t = ipip6_tunnel_locate(&p, 0)) == NULL)
				goto done;
			err = -EPERM;
801
			if (t == netdev_priv(ipip6_fb_tunnel_dev))
L
Linus Torvalds 已提交
802 803 804
				goto done;
			dev = t->dev;
		}
805 806
		unregister_netdevice(dev);
		err = 0;
L
Linus Torvalds 已提交
807 808 809 810 811 812 813 814 815 816 817 818
		break;

	default:
		err = -EINVAL;
	}

done:
	return err;
}

static struct net_device_stats *ipip6_tunnel_get_stats(struct net_device *dev)
{
819
	return &(((struct ip_tunnel*)netdev_priv(dev))->stat);
L
Linus Torvalds 已提交
820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840
}

static int ipip6_tunnel_change_mtu(struct net_device *dev, int new_mtu)
{
	if (new_mtu < IPV6_MIN_MTU || new_mtu > 0xFFF8 - sizeof(struct iphdr))
		return -EINVAL;
	dev->mtu = new_mtu;
	return 0;
}

static void ipip6_tunnel_setup(struct net_device *dev)
{
	dev->uninit		= ipip6_tunnel_uninit;
	dev->destructor 	= free_netdev;
	dev->hard_start_xmit	= ipip6_tunnel_xmit;
	dev->get_stats		= ipip6_tunnel_get_stats;
	dev->do_ioctl		= ipip6_tunnel_ioctl;
	dev->change_mtu		= ipip6_tunnel_change_mtu;

	dev->type		= ARPHRD_SIT;
	dev->hard_header_len 	= LL_MAX_HEADER + sizeof(struct iphdr);
841
	dev->mtu		= ETH_DATA_LEN - sizeof(struct iphdr);
L
Linus Torvalds 已提交
842 843 844 845 846 847 848 849 850
	dev->flags		= IFF_NOARP;
	dev->iflink		= 0;
	dev->addr_len		= 4;
}

static int ipip6_tunnel_init(struct net_device *dev)
{
	struct ip_tunnel *tunnel;

851
	tunnel = netdev_priv(dev);
L
Linus Torvalds 已提交
852 853 854 855 856 857 858

	tunnel->dev = dev;
	strcpy(tunnel->parms.name, dev->name);

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

859
	ipip6_tunnel_bind_dev(dev);
L
Linus Torvalds 已提交
860 861 862 863

	return 0;
}

864
static int __init ipip6_fb_tunnel_init(struct net_device *dev)
L
Linus Torvalds 已提交
865
{
866
	struct ip_tunnel *tunnel = netdev_priv(dev);
L
Linus Torvalds 已提交
867 868 869 870 871 872 873 874 875 876 877 878 879 880 881
	struct iphdr *iph = &tunnel->parms.iph;

	tunnel->dev = dev;
	strcpy(tunnel->parms.name, dev->name);

	iph->version		= 4;
	iph->protocol		= IPPROTO_IPV6;
	iph->ihl		= 5;
	iph->ttl		= 64;

	dev_hold(dev);
	tunnels_wc[0]		= tunnel;
	return 0;
}

882
static struct xfrm_tunnel sit_handler = {
L
Linus Torvalds 已提交
883 884
	.handler	=	ipip6_rcv,
	.err_handler	=	ipip6_err,
885
	.priority	=	1,
L
Linus Torvalds 已提交
886 887
};

888 889 890 891 892 893 894 895 896 897 898 899 900 901
static void __exit sit_destroy_tunnels(void)
{
	int prio;

	for (prio = 1; prio < 4; prio++) {
		int h;
		for (h = 0; h < HASH_SIZE; h++) {
			struct ip_tunnel *t;
			while ((t = tunnels[prio][h]) != NULL)
				unregister_netdevice(t->dev);
		}
	}
}

902
static void __exit sit_cleanup(void)
L
Linus Torvalds 已提交
903
{
904
	xfrm4_tunnel_deregister(&sit_handler, AF_INET6);
905 906 907 908 909

	rtnl_lock();
	sit_destroy_tunnels();
	unregister_netdevice(ipip6_fb_tunnel_dev);
	rtnl_unlock();
L
Linus Torvalds 已提交
910 911
}

912
static int __init sit_init(void)
L
Linus Torvalds 已提交
913 914 915 916 917
{
	int err;

	printk(KERN_INFO "IPv6 over IPv4 tunneling driver\n");

918
	if (xfrm4_tunnel_register(&sit_handler, AF_INET6) < 0) {
L
Linus Torvalds 已提交
919 920 921 922
		printk(KERN_INFO "sit init: Can't add protocol\n");
		return -EAGAIN;
	}

923
	ipip6_fb_tunnel_dev = alloc_netdev(sizeof(struct ip_tunnel), "sit0",
L
Linus Torvalds 已提交
924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939
					   ipip6_tunnel_setup);
	if (!ipip6_fb_tunnel_dev) {
		err = -ENOMEM;
		goto err1;
	}

	ipip6_fb_tunnel_dev->init = ipip6_fb_tunnel_init;

	if ((err =  register_netdev(ipip6_fb_tunnel_dev)))
		goto err2;

 out:
	return err;
 err2:
	free_netdev(ipip6_fb_tunnel_dev);
 err1:
940
	xfrm4_tunnel_deregister(&sit_handler, AF_INET6);
L
Linus Torvalds 已提交
941 942
	goto out;
}
943 944 945

module_init(sit_init);
module_exit(sit_cleanup);
946
MODULE_LICENSE("GPL");
947
MODULE_ALIAS("sit0");