ip6_gre.c 38.7 KB
Newer Older
X
xeb@mail.ru 已提交
1 2 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 36 37 38 39
/*
 *	GRE over IPv6 protocol decoder.
 *
 *	Authors: Dmitry Kozlov (xeb@mail.ru)
 *
 *	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.
 *
 */

#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt

#include <linux/capability.h>
#include <linux/module.h>
#include <linux/types.h>
#include <linux/kernel.h>
#include <linux/slab.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/init.h>
#include <linux/in6.h>
#include <linux/inetdevice.h>
#include <linux/igmp.h>
#include <linux/netfilter_ipv4.h>
#include <linux/etherdevice.h>
#include <linux/if_ether.h>
#include <linux/hash.h>
#include <linux/if_tunnel.h>
#include <linux/ip6_tunnel.h>

#include <net/sock.h>
#include <net/ip.h>
40
#include <net/ip_tunnels.h>
X
xeb@mail.ru 已提交
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
#include <net/icmp.h>
#include <net/protocol.h>
#include <net/addrconf.h>
#include <net/arp.h>
#include <net/checksum.h>
#include <net/dsfield.h>
#include <net/inet_ecn.h>
#include <net/xfrm.h>
#include <net/net_namespace.h>
#include <net/netns/generic.h>
#include <net/rtnetlink.h>

#include <net/ipv6.h>
#include <net/ip6_fib.h>
#include <net/ip6_route.h>
#include <net/ip6_tunnel.h>
57
#include <net/gre.h>
X
xeb@mail.ru 已提交
58 59


60 61 62 63
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");

64 65
#define IP6_GRE_HASH_SIZE_SHIFT  5
#define IP6_GRE_HASH_SIZE (1 << IP6_GRE_HASH_SIZE_SHIFT)
X
xeb@mail.ru 已提交
66

67
static unsigned int ip6gre_net_id __read_mostly;
X
xeb@mail.ru 已提交
68
struct ip6gre_net {
69
	struct ip6_tnl __rcu *tunnels[4][IP6_GRE_HASH_SIZE];
X
xeb@mail.ru 已提交
70 71 72 73 74

	struct net_device *fb_tunnel_dev;
};

static struct rtnl_link_ops ip6gre_link_ops __read_mostly;
N
Nicolas Dichtel 已提交
75
static struct rtnl_link_ops ip6gre_tap_ops __read_mostly;
X
xeb@mail.ru 已提交
76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
static int ip6gre_tunnel_init(struct net_device *dev);
static void ip6gre_tunnel_setup(struct net_device *dev);
static void ip6gre_tunnel_link(struct ip6gre_net *ign, struct ip6_tnl *t);
static void ip6gre_tnl_link_config(struct ip6_tnl *t, int set_mtu);

/* Tunnel hash table */

/*
   4 hash tables:

   3: (remote,local)
   2: (remote,*)
   1: (*,local)
   0: (*,*)

   We require exact key match i.e. if a key is present in packet
   it will match only tunnel with the same key; if it is not present,
   it will match only keyless tunnel.

   All keysless packets, if not matched configured keyless tunnels
   will match fallback tunnel.
 */

99
#define HASH_KEY(key) (((__force u32)key^((__force u32)key>>4))&(IP6_GRE_HASH_SIZE - 1))
X
xeb@mail.ru 已提交
100 101 102 103
static u32 HASH_ADDR(const struct in6_addr *addr)
{
	u32 hash = ipv6_addr_hash(addr);

104
	return hash_32(hash, IP6_GRE_HASH_SIZE_SHIFT);
X
xeb@mail.ru 已提交
105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127
}

#define tunnels_r_l	tunnels[3]
#define tunnels_r	tunnels[2]
#define tunnels_l	tunnels[1]
#define tunnels_wc	tunnels[0]

/* Given src, dst and key, find appropriate for input tunnel. */

static struct ip6_tnl *ip6gre_tunnel_lookup(struct net_device *dev,
		const struct in6_addr *remote, const struct in6_addr *local,
		__be32 key, __be16 gre_proto)
{
	struct net *net = dev_net(dev);
	int link = dev->ifindex;
	unsigned int h0 = HASH_ADDR(remote);
	unsigned int h1 = HASH_KEY(key);
	struct ip6_tnl *t, *cand = NULL;
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
	int dev_type = (gre_proto == htons(ETH_P_TEB)) ?
		       ARPHRD_ETHER : ARPHRD_IP6GRE;
	int score, cand_score = 4;

A
Amerigo Wang 已提交
128
	for_each_ip_tunnel_rcu(t, ign->tunnels_r_l[h0 ^ h1]) {
X
xeb@mail.ru 已提交
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152
		if (!ipv6_addr_equal(local, &t->parms.laddr) ||
		    !ipv6_addr_equal(remote, &t->parms.raddr) ||
		    key != t->parms.i_key ||
		    !(t->dev->flags & IFF_UP))
			continue;

		if (t->dev->type != ARPHRD_IP6GRE &&
		    t->dev->type != dev_type)
			continue;

		score = 0;
		if (t->parms.link != link)
			score |= 1;
		if (t->dev->type != dev_type)
			score |= 2;
		if (score == 0)
			return t;

		if (score < cand_score) {
			cand = t;
			cand_score = score;
		}
	}

A
Amerigo Wang 已提交
153
	for_each_ip_tunnel_rcu(t, ign->tunnels_r[h0 ^ h1]) {
X
xeb@mail.ru 已提交
154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
		if (!ipv6_addr_equal(remote, &t->parms.raddr) ||
		    key != t->parms.i_key ||
		    !(t->dev->flags & IFF_UP))
			continue;

		if (t->dev->type != ARPHRD_IP6GRE &&
		    t->dev->type != dev_type)
			continue;

		score = 0;
		if (t->parms.link != link)
			score |= 1;
		if (t->dev->type != dev_type)
			score |= 2;
		if (score == 0)
			return t;

		if (score < cand_score) {
			cand = t;
			cand_score = score;
		}
	}

A
Amerigo Wang 已提交
177
	for_each_ip_tunnel_rcu(t, ign->tunnels_l[h1]) {
X
xeb@mail.ru 已提交
178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202
		if ((!ipv6_addr_equal(local, &t->parms.laddr) &&
			  (!ipv6_addr_equal(local, &t->parms.raddr) ||
				 !ipv6_addr_is_multicast(local))) ||
		    key != t->parms.i_key ||
		    !(t->dev->flags & IFF_UP))
			continue;

		if (t->dev->type != ARPHRD_IP6GRE &&
		    t->dev->type != dev_type)
			continue;

		score = 0;
		if (t->parms.link != link)
			score |= 1;
		if (t->dev->type != dev_type)
			score |= 2;
		if (score == 0)
			return t;

		if (score < cand_score) {
			cand = t;
			cand_score = score;
		}
	}

A
Amerigo Wang 已提交
203
	for_each_ip_tunnel_rcu(t, ign->tunnels_wc[h1]) {
X
xeb@mail.ru 已提交
204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225
		if (t->parms.i_key != key ||
		    !(t->dev->flags & IFF_UP))
			continue;

		if (t->dev->type != ARPHRD_IP6GRE &&
		    t->dev->type != dev_type)
			continue;

		score = 0;
		if (t->parms.link != link)
			score |= 1;
		if (t->dev->type != dev_type)
			score |= 2;
		if (score == 0)
			return t;

		if (score < cand_score) {
			cand = t;
			cand_score = score;
		}
	}

226
	if (cand)
X
xeb@mail.ru 已提交
227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316
		return cand;

	dev = ign->fb_tunnel_dev;
	if (dev->flags & IFF_UP)
		return netdev_priv(dev);

	return NULL;
}

static struct ip6_tnl __rcu **__ip6gre_bucket(struct ip6gre_net *ign,
		const struct __ip6_tnl_parm *p)
{
	const struct in6_addr *remote = &p->raddr;
	const struct in6_addr *local = &p->laddr;
	unsigned int h = HASH_KEY(p->i_key);
	int prio = 0;

	if (!ipv6_addr_any(local))
		prio |= 1;
	if (!ipv6_addr_any(remote) && !ipv6_addr_is_multicast(remote)) {
		prio |= 2;
		h ^= HASH_ADDR(remote);
	}

	return &ign->tunnels[prio][h];
}

static inline struct ip6_tnl __rcu **ip6gre_bucket(struct ip6gre_net *ign,
		const struct ip6_tnl *t)
{
	return __ip6gre_bucket(ign, &t->parms);
}

static void ip6gre_tunnel_link(struct ip6gre_net *ign, struct ip6_tnl *t)
{
	struct ip6_tnl __rcu **tp = ip6gre_bucket(ign, t);

	rcu_assign_pointer(t->next, rtnl_dereference(*tp));
	rcu_assign_pointer(*tp, t);
}

static void ip6gre_tunnel_unlink(struct ip6gre_net *ign, struct ip6_tnl *t)
{
	struct ip6_tnl __rcu **tp;
	struct ip6_tnl *iter;

	for (tp = ip6gre_bucket(ign, t);
	     (iter = rtnl_dereference(*tp)) != NULL;
	     tp = &iter->next) {
		if (t == iter) {
			rcu_assign_pointer(*tp, t->next);
			break;
		}
	}
}

static struct ip6_tnl *ip6gre_tunnel_find(struct net *net,
					   const struct __ip6_tnl_parm *parms,
					   int type)
{
	const struct in6_addr *remote = &parms->raddr;
	const struct in6_addr *local = &parms->laddr;
	__be32 key = parms->i_key;
	int link = parms->link;
	struct ip6_tnl *t;
	struct ip6_tnl __rcu **tp;
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);

	for (tp = __ip6gre_bucket(ign, parms);
	     (t = rtnl_dereference(*tp)) != NULL;
	     tp = &t->next)
		if (ipv6_addr_equal(local, &t->parms.laddr) &&
		    ipv6_addr_equal(remote, &t->parms.raddr) &&
		    key == t->parms.i_key &&
		    link == t->parms.link &&
		    type == t->dev->type)
			break;

	return t;
}

static struct ip6_tnl *ip6gre_tunnel_locate(struct net *net,
		const struct __ip6_tnl_parm *parms, int create)
{
	struct ip6_tnl *t, *nt;
	struct net_device *dev;
	char name[IFNAMSIZ];
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);

	t = ip6gre_tunnel_find(net, parms, ARPHRD_IP6GRE);
317 318
	if (t && create)
		return NULL;
X
xeb@mail.ru 已提交
319 320 321 322 323 324 325 326
	if (t || !create)
		return t;

	if (parms->name[0])
		strlcpy(name, parms->name, IFNAMSIZ);
	else
		strcpy(name, "ip6gre%d");

327 328
	dev = alloc_netdev(sizeof(*t), name, NET_NAME_UNKNOWN,
			   ip6gre_tunnel_setup);
X
xeb@mail.ru 已提交
329 330 331 332 333 334 335 336 337 338
	if (!dev)
		return NULL;

	dev_net_set(dev, net);

	nt = netdev_priv(dev);
	nt->parms = *parms;
	dev->rtnl_link_ops = &ip6gre_link_ops;

	nt->dev = dev;
N
Nicolas Dichtel 已提交
339
	nt->net = dev_net(dev);
X
xeb@mail.ru 已提交
340 341 342 343 344 345
	ip6gre_tnl_link_config(nt, 1);

	if (register_netdevice(dev) < 0)
		goto failed_free;

	/* Can use a lockless transmit, unless we generate output sequences */
346
	if (!(nt->parms.o_flags & TUNNEL_SEQ))
X
xeb@mail.ru 已提交
347 348 349 350 351 352 353 354 355 356 357 358 359
		dev->features |= NETIF_F_LLTX;

	dev_hold(dev);
	ip6gre_tunnel_link(ign, nt);
	return nt;

failed_free:
	free_netdev(dev);
	return NULL;
}

static void ip6gre_tunnel_uninit(struct net_device *dev)
{
N
Nicolas Dichtel 已提交
360 361
	struct ip6_tnl *t = netdev_priv(dev);
	struct ip6gre_net *ign = net_generic(t->net, ip6gre_net_id);
X
xeb@mail.ru 已提交
362

N
Nicolas Dichtel 已提交
363
	ip6gre_tunnel_unlink(ign, t);
364
	dst_cache_reset(&t->dst_cache);
X
xeb@mail.ru 已提交
365 366 367 368 369
	dev_put(dev);
}


static void ip6gre_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
370
		       u8 type, u8 code, int offset, __be32 info)
X
xeb@mail.ru 已提交
371
{
372 373 374
	const struct gre_base_hdr *greh;
	const struct ipv6hdr *ipv6h;
	int grehlen = sizeof(*greh);
X
xeb@mail.ru 已提交
375
	struct ip6_tnl *t;
376
	int key_off = 0;
X
xeb@mail.ru 已提交
377
	__be16 flags;
378
	__be32 key;
X
xeb@mail.ru 已提交
379

380 381 382 383 384 385 386 387 388 389 390
	if (!pskb_may_pull(skb, offset + grehlen))
		return;
	greh = (const struct gre_base_hdr *)(skb->data + offset);
	flags = greh->flags;
	if (flags & (GRE_VERSION | GRE_ROUTING))
		return;
	if (flags & GRE_CSUM)
		grehlen += 4;
	if (flags & GRE_KEY) {
		key_off = grehlen + offset;
		grehlen += 4;
X
xeb@mail.ru 已提交
391 392
	}

393
	if (!pskb_may_pull(skb, offset + grehlen))
X
xeb@mail.ru 已提交
394
		return;
E
Eric Dumazet 已提交
395
	ipv6h = (const struct ipv6hdr *)skb->data;
396 397
	greh = (const struct gre_base_hdr *)(skb->data + offset);
	key = key_off ? *(__be32 *)(skb->data + key_off) : 0;
X
xeb@mail.ru 已提交
398 399

	t = ip6gre_tunnel_lookup(skb->dev, &ipv6h->daddr, &ipv6h->saddr,
400
				 key, greh->protocol);
401
	if (!t)
402
		return;
X
xeb@mail.ru 已提交
403 404 405 406 407 408

	switch (type) {
		__u32 teli;
		struct ipv6_tlv_tnl_enc_lim *tel;
		__u32 mtu;
	case ICMPV6_DEST_UNREACH:
409 410
		net_dbg_ratelimited("%s: Path to destination invalid or inactive!\n",
				    t->parms.name);
X
xeb@mail.ru 已提交
411 412 413
		break;
	case ICMPV6_TIME_EXCEED:
		if (code == ICMPV6_EXC_HOPLIMIT) {
414 415
			net_dbg_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n",
					    t->parms.name);
X
xeb@mail.ru 已提交
416 417 418 419 420 421 422
		}
		break;
	case ICMPV6_PARAMPROB:
		teli = 0;
		if (code == ICMPV6_HDR_FIELD)
			teli = ip6_tnl_parse_tlv_enc_lim(skb, skb->data);

423
		if (teli && teli == be32_to_cpu(info) - 2) {
X
xeb@mail.ru 已提交
424 425
			tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli];
			if (tel->encap_limit == 0) {
426 427
				net_dbg_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n",
						    t->parms.name);
X
xeb@mail.ru 已提交
428 429
			}
		} else {
430 431
			net_dbg_ratelimited("%s: Recipient unable to parse tunneled packet!\n",
					    t->parms.name);
X
xeb@mail.ru 已提交
432 433 434
		}
		break;
	case ICMPV6_PKT_TOOBIG:
435
		mtu = be32_to_cpu(info) - offset;
X
xeb@mail.ru 已提交
436 437 438 439 440 441 442 443 444 445 446 447 448
		if (mtu < IPV6_MIN_MTU)
			mtu = IPV6_MIN_MTU;
		t->dev->mtu = mtu;
		break;
	}

	if (time_before(jiffies, t->err_time + IP6TUNNEL_ERR_TIMEO))
		t->err_count++;
	else
		t->err_count = 1;
	t->err_time = jiffies;
}

449
static int ip6gre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi)
X
xeb@mail.ru 已提交
450 451 452 453 454 455
{
	const struct ipv6hdr *ipv6h;
	struct ip6_tnl *tunnel;

	ipv6h = ipv6_hdr(skb);
	tunnel = ip6gre_tunnel_lookup(skb->dev,
456 457
				      &ipv6h->saddr, &ipv6h->daddr, tpi->key,
				      tpi->proto);
X
xeb@mail.ru 已提交
458
	if (tunnel) {
459
		ip6_tnl_rcv(tunnel, skb, tpi, NULL, false);
X
xeb@mail.ru 已提交
460

461 462
		return PACKET_RCVD;
	}
463

464 465
	return PACKET_REJECT;
}
466

467 468 469 470 471
static int gre_rcv(struct sk_buff *skb)
{
	struct tnl_ptk_info tpi;
	bool csum_err = false;
	int hdr_len;
472

E
Eric Dumazet 已提交
473
	hdr_len = gre_parse_header(skb, &tpi, &csum_err, htons(ETH_P_IPV6), 0);
474
	if (hdr_len < 0)
475
		goto drop;
X
xeb@mail.ru 已提交
476

477 478
	if (iptunnel_pull_header(skb, hdr_len, tpi.proto, false))
		goto drop;
X
xeb@mail.ru 已提交
479

480
	if (ip6gre_rcv(skb, &tpi) == PACKET_RCVD)
X
xeb@mail.ru 已提交
481 482
		return 0;

483
	icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);
X
xeb@mail.ru 已提交
484 485 486 487 488
drop:
	kfree_skb(skb);
	return 0;
}

489
static int gre_handle_offloads(struct sk_buff *skb, bool csum)
X
xeb@mail.ru 已提交
490
{
491 492
	return iptunnel_handle_offloads(skb,
					csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
X
xeb@mail.ru 已提交
493 494
}

495 496 497 498
static netdev_tx_t __gre6_xmit(struct sk_buff *skb,
			       struct net_device *dev, __u8 dsfield,
			       struct flowi6 *fl6, int encap_limit,
			       __u32 *pmtu, __be16 proto)
X
xeb@mail.ru 已提交
499 500
{
	struct ip6_tnl *tunnel = netdev_priv(dev);
501 502
	__be16 protocol = (dev->type == ARPHRD_ETHER) ?
			  htons(ETH_P_TEB) : proto;
X
xeb@mail.ru 已提交
503 504 505 506

	if (dev->type == ARPHRD_ETHER)
		IPCB(skb)->flags = 0;

507 508 509
	if (dev->header_ops && dev->type == ARPHRD_IP6GRE)
		fl6->daddr = ((struct ipv6hdr *)skb->data)->daddr;
	else
X
xeb@mail.ru 已提交
510 511
		fl6->daddr = tunnel->parms.raddr;

512 513
	if (tunnel->parms.o_flags & TUNNEL_SEQ)
		tunnel->o_seqno++;
X
xeb@mail.ru 已提交
514

515 516 517
	/* Push GRE header. */
	gre_build_header(skb, tunnel->tun_hlen, tunnel->parms.o_flags,
			 protocol, tunnel->parms.o_key, htonl(tunnel->o_seqno));
X
xeb@mail.ru 已提交
518

519 520
	return ip6_tnl_xmit(skb, dev, dsfield, fl6, encap_limit, pmtu,
			    NEXTHDR_GRE);
X
xeb@mail.ru 已提交
521 522 523 524 525 526 527 528 529 530 531 532
}

static inline int ip6gre_xmit_ipv4(struct sk_buff *skb, struct net_device *dev)
{
	struct ip6_tnl *t = netdev_priv(dev);
	const struct iphdr  *iph = ip_hdr(skb);
	int encap_limit = -1;
	struct flowi6 fl6;
	__u8 dsfield;
	__u32 mtu;
	int err;

533 534
	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));

X
xeb@mail.ru 已提交
535 536 537 538 539 540 541 542 543 544 545 546
	if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
		encap_limit = t->parms.encap_limit;

	memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));

	dsfield = ipv4_get_dsfield(iph);

	if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)
		fl6.flowlabel |= htonl((__u32)iph->tos << IPV6_TCLASS_SHIFT)
					  & IPV6_TCLASS_MASK;
	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FWMARK)
		fl6.flowi6_mark = skb->mark;
547 548
	else
		fl6.flowi6_mark = t->parms.fwmark;
X
xeb@mail.ru 已提交
549

550 551
	fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);

552 553 554 555 556 557
	err = gre_handle_offloads(skb, !!(t->parms.o_flags & TUNNEL_CSUM));
	if (err)
		return -1;

	err = __gre6_xmit(skb, dev, dsfield, &fl6, encap_limit, &mtu,
			  skb->protocol);
X
xeb@mail.ru 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
	if (err != 0) {
		/* XXX: send ICMP error even if DF is not set. */
		if (err == -EMSGSIZE)
			icmp_send(skb, ICMP_DEST_UNREACH, ICMP_FRAG_NEEDED,
				  htonl(mtu));
		return -1;
	}

	return 0;
}

static inline int ip6gre_xmit_ipv6(struct sk_buff *skb, struct net_device *dev)
{
	struct ip6_tnl *t = netdev_priv(dev);
	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
	int encap_limit = -1;
	__u16 offset;
	struct flowi6 fl6;
	__u8 dsfield;
	__u32 mtu;
	int err;

	if (ipv6_addr_equal(&t->parms.raddr, &ipv6h->saddr))
		return -1;

	offset = ip6_tnl_parse_tlv_enc_lim(skb, skb_network_header(skb));
584 585 586
	/* ip6_tnl_parse_tlv_enc_lim() might have reallocated skb->head */
	ipv6h = ipv6_hdr(skb);

X
xeb@mail.ru 已提交
587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604
	if (offset > 0) {
		struct ipv6_tlv_tnl_enc_lim *tel;
		tel = (struct ipv6_tlv_tnl_enc_lim *)&skb_network_header(skb)[offset];
		if (tel->encap_limit == 0) {
			icmpv6_send(skb, ICMPV6_PARAMPROB,
				    ICMPV6_HDR_FIELD, offset + 2);
			return -1;
		}
		encap_limit = tel->encap_limit - 1;
	} else if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
		encap_limit = t->parms.encap_limit;

	memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));

	dsfield = ipv6_get_dsfield(ipv6h);
	if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)
		fl6.flowlabel |= (*(__be32 *) ipv6h & IPV6_TCLASS_MASK);
	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FLOWLABEL)
F
Florent Fourcot 已提交
605
		fl6.flowlabel |= ip6_flowlabel(ipv6h);
X
xeb@mail.ru 已提交
606 607
	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FWMARK)
		fl6.flowi6_mark = skb->mark;
608 609
	else
		fl6.flowi6_mark = t->parms.fwmark;
X
xeb@mail.ru 已提交
610

611 612
	fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);

613 614 615 616 617
	if (gre_handle_offloads(skb, !!(t->parms.o_flags & TUNNEL_CSUM)))
		return -1;

	err = __gre6_xmit(skb, dev, dsfield, &fl6, encap_limit,
			  &mtu, skb->protocol);
X
xeb@mail.ru 已提交
618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659
	if (err != 0) {
		if (err == -EMSGSIZE)
			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
		return -1;
	}

	return 0;
}

/**
 * ip6_tnl_addr_conflict - compare packet addresses to tunnel's own
 *   @t: the outgoing tunnel device
 *   @hdr: IPv6 header from the incoming packet
 *
 * Description:
 *   Avoid trivial tunneling loop by checking that tunnel exit-point
 *   doesn't match source of incoming packet.
 *
 * Return:
 *   1 if conflict,
 *   0 else
 **/

static inline bool ip6gre_tnl_addr_conflict(const struct ip6_tnl *t,
	const struct ipv6hdr *hdr)
{
	return ipv6_addr_equal(&t->parms.raddr, &hdr->saddr);
}

static int ip6gre_xmit_other(struct sk_buff *skb, struct net_device *dev)
{
	struct ip6_tnl *t = netdev_priv(dev);
	int encap_limit = -1;
	struct flowi6 fl6;
	__u32 mtu;
	int err;

	if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
		encap_limit = t->parms.encap_limit;

	memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));

660 661 662 663 664
	err = gre_handle_offloads(skb, !!(t->parms.o_flags & TUNNEL_CSUM));
	if (err)
		return err;

	err = __gre6_xmit(skb, dev, 0, &fl6, encap_limit, &mtu, skb->protocol);
X
xeb@mail.ru 已提交
665 666 667 668 669 670 671 672 673 674 675

	return err;
}

static netdev_tx_t ip6gre_tunnel_xmit(struct sk_buff *skb,
	struct net_device *dev)
{
	struct ip6_tnl *t = netdev_priv(dev);
	struct net_device_stats *stats = &t->dev->stats;
	int ret;

676
	if (!ip6_tnl_xmit_ctl(t, &t->parms.laddr, &t->parms.raddr))
677
		goto tx_err;
X
xeb@mail.ru 已提交
678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707

	switch (skb->protocol) {
	case htons(ETH_P_IP):
		ret = ip6gre_xmit_ipv4(skb, dev);
		break;
	case htons(ETH_P_IPV6):
		ret = ip6gre_xmit_ipv6(skb, dev);
		break;
	default:
		ret = ip6gre_xmit_other(skb, dev);
		break;
	}

	if (ret < 0)
		goto tx_err;

	return NETDEV_TX_OK;

tx_err:
	stats->tx_errors++;
	stats->tx_dropped++;
	kfree_skb(skb);
	return NETDEV_TX_OK;
}

static void ip6gre_tnl_link_config(struct ip6_tnl *t, int set_mtu)
{
	struct net_device *dev = t->dev;
	struct __ip6_tnl_parm *p = &t->parms;
	struct flowi6 *fl6 = &t->fl.u.ip6;
T
Tom Herbert 已提交
708
	int t_hlen;
X
xeb@mail.ru 已提交
709 710 711 712 713 714 715 716 717 718 719

	if (dev->type != ARPHRD_ETHER) {
		memcpy(dev->dev_addr, &p->laddr, sizeof(struct in6_addr));
		memcpy(dev->broadcast, &p->raddr, sizeof(struct in6_addr));
	}

	/* Set up flowi template */
	fl6->saddr = p->laddr;
	fl6->daddr = p->raddr;
	fl6->flowi6_oif = p->link;
	fl6->flowlabel = 0;
720
	fl6->flowi6_proto = IPPROTO_GRE;
X
xeb@mail.ru 已提交
721 722 723 724 725 726 727 728 729 730 731 732 733 734 735

	if (!(p->flags&IP6_TNL_F_USE_ORIG_TCLASS))
		fl6->flowlabel |= IPV6_TCLASS_MASK & p->flowinfo;
	if (!(p->flags&IP6_TNL_F_USE_ORIG_FLOWLABEL))
		fl6->flowlabel |= IPV6_FLOWLABEL_MASK & p->flowinfo;

	p->flags &= ~(IP6_TNL_F_CAP_XMIT|IP6_TNL_F_CAP_RCV|IP6_TNL_F_CAP_PER_PACKET);
	p->flags |= ip6_tnl_get_cap(t, &p->laddr, &p->raddr);

	if (p->flags&IP6_TNL_F_CAP_XMIT &&
			p->flags&IP6_TNL_F_CAP_RCV && dev->type != ARPHRD_ETHER)
		dev->flags |= IFF_POINTOPOINT;
	else
		dev->flags &= ~IFF_POINTOPOINT;

T
Tom Herbert 已提交
736 737
	t->tun_hlen = gre_calc_hlen(t->parms.o_flags);

738
	t->hlen = t->encap_hlen + t->tun_hlen;
T
Tom Herbert 已提交
739 740

	t_hlen = t->hlen + sizeof(struct ipv6hdr);
X
xeb@mail.ru 已提交
741 742 743 744 745

	if (p->flags & IP6_TNL_F_CAP_XMIT) {
		int strict = (ipv6_addr_type(&p->raddr) &
			      (IPV6_ADDR_MULTICAST|IPV6_ADDR_LINKLOCAL));

N
Nicolas Dichtel 已提交
746
		struct rt6_info *rt = rt6_lookup(t->net,
X
xeb@mail.ru 已提交
747 748 749
						 &p->raddr, &p->laddr,
						 p->link, strict);

750
		if (!rt)
X
xeb@mail.ru 已提交
751 752 753
			return;

		if (rt->dst.dev) {
T
Tom Herbert 已提交
754 755
			dev->hard_header_len = rt->dst.dev->hard_header_len +
					       t_hlen;
X
xeb@mail.ru 已提交
756 757

			if (set_mtu) {
T
Tom Herbert 已提交
758
				dev->mtu = rt->dst.dev->mtu - t_hlen;
X
xeb@mail.ru 已提交
759 760
				if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
					dev->mtu -= 8;
761 762
				if (dev->type == ARPHRD_ETHER)
					dev->mtu -= ETH_HLEN;
X
xeb@mail.ru 已提交
763 764 765 766 767

				if (dev->mtu < IPV6_MIN_MTU)
					dev->mtu = IPV6_MIN_MTU;
			}
		}
A
Amerigo Wang 已提交
768
		ip6_rt_put(rt);
X
xeb@mail.ru 已提交
769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786
	}
}

static int ip6gre_tnl_change(struct ip6_tnl *t,
	const struct __ip6_tnl_parm *p, int set_mtu)
{
	t->parms.laddr = p->laddr;
	t->parms.raddr = p->raddr;
	t->parms.flags = p->flags;
	t->parms.hop_limit = p->hop_limit;
	t->parms.encap_limit = p->encap_limit;
	t->parms.flowinfo = p->flowinfo;
	t->parms.link = p->link;
	t->parms.proto = p->proto;
	t->parms.i_key = p->i_key;
	t->parms.o_key = p->o_key;
	t->parms.i_flags = p->i_flags;
	t->parms.o_flags = p->o_flags;
787
	t->parms.fwmark = p->fwmark;
788
	dst_cache_reset(&t->dst_cache);
X
xeb@mail.ru 已提交
789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804
	ip6gre_tnl_link_config(t, set_mtu);
	return 0;
}

static void ip6gre_tnl_parm_from_user(struct __ip6_tnl_parm *p,
	const struct ip6_tnl_parm2 *u)
{
	p->laddr = u->laddr;
	p->raddr = u->raddr;
	p->flags = u->flags;
	p->hop_limit = u->hop_limit;
	p->encap_limit = u->encap_limit;
	p->flowinfo = u->flowinfo;
	p->link = u->link;
	p->i_key = u->i_key;
	p->o_key = u->o_key;
T
Tom Herbert 已提交
805 806
	p->i_flags = gre_flags_to_tnl_flags(u->i_flags);
	p->o_flags = gre_flags_to_tnl_flags(u->o_flags);
X
xeb@mail.ru 已提交
807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822
	memcpy(p->name, u->name, sizeof(u->name));
}

static void ip6gre_tnl_parm_to_user(struct ip6_tnl_parm2 *u,
	const struct __ip6_tnl_parm *p)
{
	u->proto = IPPROTO_GRE;
	u->laddr = p->laddr;
	u->raddr = p->raddr;
	u->flags = p->flags;
	u->hop_limit = p->hop_limit;
	u->encap_limit = p->encap_limit;
	u->flowinfo = p->flowinfo;
	u->link = p->link;
	u->i_key = p->i_key;
	u->o_key = p->o_key;
T
Tom Herbert 已提交
823 824
	u->i_flags = gre_tnl_flags_to_gre_flags(p->i_flags);
	u->o_flags = gre_tnl_flags_to_gre_flags(p->o_flags);
X
xeb@mail.ru 已提交
825 826 827 828 829 830 831 832 833
	memcpy(u->name, p->name, sizeof(u->name));
}

static int ip6gre_tunnel_ioctl(struct net_device *dev,
	struct ifreq *ifr, int cmd)
{
	int err = 0;
	struct ip6_tnl_parm2 p;
	struct __ip6_tnl_parm p1;
N
Nicolas Dichtel 已提交
834 835
	struct ip6_tnl *t = netdev_priv(dev);
	struct net *net = t->net;
X
xeb@mail.ru 已提交
836 837
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);

838 839
	memset(&p1, 0, sizeof(p1));

X
xeb@mail.ru 已提交
840 841 842 843 844 845 846 847 848
	switch (cmd) {
	case SIOCGETTUNNEL:
		if (dev == ign->fb_tunnel_dev) {
			if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p))) {
				err = -EFAULT;
				break;
			}
			ip6gre_tnl_parm_from_user(&p1, &p);
			t = ip6gre_tunnel_locate(net, &p1, 0);
849
			if (!t)
N
Nicolas Dichtel 已提交
850
				t = netdev_priv(dev);
X
xeb@mail.ru 已提交
851
		}
852
		memset(&p, 0, sizeof(p));
X
xeb@mail.ru 已提交
853 854 855 856 857 858 859 860
		ip6gre_tnl_parm_to_user(&p, &t->parms);
		if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
			err = -EFAULT;
		break;

	case SIOCADDTUNNEL:
	case SIOCCHGTUNNEL:
		err = -EPERM;
861
		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
X
xeb@mail.ru 已提交
862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880
			goto done;

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

		err = -EINVAL;
		if ((p.i_flags|p.o_flags)&(GRE_VERSION|GRE_ROUTING))
			goto done;

		if (!(p.i_flags&GRE_KEY))
			p.i_key = 0;
		if (!(p.o_flags&GRE_KEY))
			p.o_key = 0;

		ip6gre_tnl_parm_from_user(&p1, &p);
		t = ip6gre_tunnel_locate(net, &p1, cmd == SIOCADDTUNNEL);

		if (dev != ign->fb_tunnel_dev && cmd == SIOCCHGTUNNEL) {
881
			if (t) {
X
xeb@mail.ru 已提交
882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899
				if (t->dev != dev) {
					err = -EEXIST;
					break;
				}
			} else {
				t = netdev_priv(dev);

				ip6gre_tunnel_unlink(ign, t);
				synchronize_net();
				ip6gre_tnl_change(t, &p1, 1);
				ip6gre_tunnel_link(ign, t);
				netdev_state_change(dev);
			}
		}

		if (t) {
			err = 0;

900
			memset(&p, 0, sizeof(p));
X
xeb@mail.ru 已提交
901 902 903 904 905 906 907 908 909
			ip6gre_tnl_parm_to_user(&p, &t->parms);
			if (copy_to_user(ifr->ifr_ifru.ifru_data, &p, sizeof(p)))
				err = -EFAULT;
		} else
			err = (cmd == SIOCADDTUNNEL ? -ENOBUFS : -ENOENT);
		break;

	case SIOCDELTUNNEL:
		err = -EPERM;
910
		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
X
xeb@mail.ru 已提交
911 912 913 914 915 916 917 918 919
			goto done;

		if (dev == ign->fb_tunnel_dev) {
			err = -EFAULT;
			if (copy_from_user(&p, ifr->ifr_ifru.ifru_data, sizeof(p)))
				goto done;
			err = -ENOENT;
			ip6gre_tnl_parm_from_user(&p1, &p);
			t = ip6gre_tunnel_locate(net, &p1, 0);
920
			if (!t)
X
xeb@mail.ru 已提交
921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946
				goto done;
			err = -EPERM;
			if (t == netdev_priv(ign->fb_tunnel_dev))
				goto done;
			dev = t->dev;
		}
		unregister_netdevice(dev);
		err = 0;
		break;

	default:
		err = -EINVAL;
	}

done:
	return err;
}

static int ip6gre_header(struct sk_buff *skb, struct net_device *dev,
			unsigned short type,
			const void *daddr, const void *saddr, unsigned int len)
{
	struct ip6_tnl *t = netdev_priv(dev);
	struct ipv6hdr *ipv6h = (struct ipv6hdr *)skb_push(skb, t->hlen);
	__be16 *p = (__be16 *)(ipv6h+1);

947 948
	ip6_flow_hdr(ipv6h, 0,
		     ip6_make_flowlabel(dev_net(dev), skb,
949
					t->fl.u.ip6.flowlabel, true,
950
					&t->fl.u.ip6));
X
xeb@mail.ru 已提交
951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981
	ipv6h->hop_limit = t->parms.hop_limit;
	ipv6h->nexthdr = NEXTHDR_GRE;
	ipv6h->saddr = t->parms.laddr;
	ipv6h->daddr = t->parms.raddr;

	p[0]		= t->parms.o_flags;
	p[1]		= htons(type);

	/*
	 *	Set the source hardware address.
	 */

	if (saddr)
		memcpy(&ipv6h->saddr, saddr, sizeof(struct in6_addr));
	if (daddr)
		memcpy(&ipv6h->daddr, daddr, sizeof(struct in6_addr));
	if (!ipv6_addr_any(&ipv6h->daddr))
		return t->hlen;

	return -t->hlen;
}

static const struct header_ops ip6gre_header_ops = {
	.create	= ip6gre_header,
};

static const struct net_device_ops ip6gre_netdev_ops = {
	.ndo_init		= ip6gre_tunnel_init,
	.ndo_uninit		= ip6gre_tunnel_uninit,
	.ndo_start_xmit		= ip6gre_tunnel_xmit,
	.ndo_do_ioctl		= ip6gre_tunnel_ioctl,
982
	.ndo_change_mtu		= ip6_tnl_change_mtu,
983
	.ndo_get_stats64	= ip_tunnel_get_stats64,
984
	.ndo_get_iflink		= ip6_tnl_get_iflink,
X
xeb@mail.ru 已提交
985 986 987 988
};

static void ip6gre_dev_free(struct net_device *dev)
{
989 990
	struct ip6_tnl *t = netdev_priv(dev);

991
	dst_cache_destroy(&t->dst_cache);
X
xeb@mail.ru 已提交
992 993 994 995 996 997 998 999 1000 1001
	free_percpu(dev->tstats);
	free_netdev(dev);
}

static void ip6gre_tunnel_setup(struct net_device *dev)
{
	dev->netdev_ops = &ip6gre_netdev_ops;
	dev->destructor = ip6gre_dev_free;

	dev->type = ARPHRD_IP6GRE;
1002

X
xeb@mail.ru 已提交
1003 1004
	dev->flags |= IFF_NOARP;
	dev->addr_len = sizeof(struct in6_addr);
1005
	netif_keep_dst(dev);
1006 1007 1008
	/* This perm addr will be used as interface identifier by IPv6 */
	dev->addr_assign_type = NET_ADDR_RANDOM;
	eth_random_addr(dev->perm_addr);
X
xeb@mail.ru 已提交
1009 1010
}

1011
static int ip6gre_tunnel_init_common(struct net_device *dev)
X
xeb@mail.ru 已提交
1012 1013
{
	struct ip6_tnl *tunnel;
1014
	int ret;
1015
	int t_hlen;
X
xeb@mail.ru 已提交
1016 1017 1018 1019

	tunnel = netdev_priv(dev);

	tunnel->dev = dev;
N
Nicolas Dichtel 已提交
1020
	tunnel->net = dev_net(dev);
X
xeb@mail.ru 已提交
1021 1022
	strcpy(tunnel->parms.name, dev->name);

1023 1024 1025 1026
	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
	if (!dev->tstats)
		return -ENOMEM;

1027
	ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
1028 1029 1030 1031 1032 1033
	if (ret) {
		free_percpu(dev->tstats);
		dev->tstats = NULL;
		return ret;
	}

1034
	tunnel->tun_hlen = gre_calc_hlen(tunnel->parms.o_flags);
1035
	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
T
Tom Herbert 已提交
1036
	t_hlen = tunnel->hlen + sizeof(struct ipv6hdr);
1037

T
Tom Herbert 已提交
1038 1039
	dev->hard_header_len = LL_MAX_HEADER + t_hlen;
	dev->mtu = ETH_DATA_LEN - t_hlen;
1040 1041
	if (dev->type == ARPHRD_ETHER)
		dev->mtu -= ETH_HLEN;
1042 1043 1044
	if (!(tunnel->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
		dev->mtu -= 8;

1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058
	return 0;
}

static int ip6gre_tunnel_init(struct net_device *dev)
{
	struct ip6_tnl *tunnel;
	int ret;

	ret = ip6gre_tunnel_init_common(dev);
	if (ret)
		return ret;

	tunnel = netdev_priv(dev);

X
xeb@mail.ru 已提交
1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072
	memcpy(dev->dev_addr, &tunnel->parms.laddr, sizeof(struct in6_addr));
	memcpy(dev->broadcast, &tunnel->parms.raddr, sizeof(struct in6_addr));

	if (ipv6_addr_any(&tunnel->parms.raddr))
		dev->header_ops = &ip6gre_header_ops;

	return 0;
}

static void ip6gre_fb_tunnel_init(struct net_device *dev)
{
	struct ip6_tnl *tunnel = netdev_priv(dev);

	tunnel->dev = dev;
N
Nicolas Dichtel 已提交
1073
	tunnel->net = dev_net(dev);
X
xeb@mail.ru 已提交
1074 1075 1076 1077 1078 1079 1080 1081 1082
	strcpy(tunnel->parms.name, dev->name);

	tunnel->hlen		= sizeof(struct ipv6hdr) + 4;

	dev_hold(dev);
}


static struct inet6_protocol ip6gre_protocol __read_mostly = {
1083
	.handler     = gre_rcv,
X
xeb@mail.ru 已提交
1084 1085 1086 1087
	.err_handler = ip6gre_err,
	.flags       = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
};

N
Nicolas Dichtel 已提交
1088
static void ip6gre_destroy_tunnels(struct net *net, struct list_head *head)
X
xeb@mail.ru 已提交
1089
{
N
Nicolas Dichtel 已提交
1090 1091
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
	struct net_device *dev, *aux;
X
xeb@mail.ru 已提交
1092 1093
	int prio;

N
Nicolas Dichtel 已提交
1094 1095 1096 1097 1098
	for_each_netdev_safe(net, dev, aux)
		if (dev->rtnl_link_ops == &ip6gre_link_ops ||
		    dev->rtnl_link_ops == &ip6gre_tap_ops)
			unregister_netdevice_queue(dev, head);

X
xeb@mail.ru 已提交
1099 1100
	for (prio = 0; prio < 4; prio++) {
		int h;
1101
		for (h = 0; h < IP6_GRE_HASH_SIZE; h++) {
X
xeb@mail.ru 已提交
1102 1103 1104 1105
			struct ip6_tnl *t;

			t = rtnl_dereference(ign->tunnels[prio][h]);

1106
			while (t) {
N
Nicolas Dichtel 已提交
1107 1108 1109 1110 1111 1112
				/* If dev is in the same netns, it has already
				 * been added to the list by the previous loop.
				 */
				if (!net_eq(dev_net(t->dev), net))
					unregister_netdevice_queue(t->dev,
								   head);
X
xeb@mail.ru 已提交
1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124
				t = rtnl_dereference(t->next);
			}
		}
	}
}

static int __net_init ip6gre_init_net(struct net *net)
{
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
	int err;

	ign->fb_tunnel_dev = alloc_netdev(sizeof(struct ip6_tnl), "ip6gre0",
1125 1126
					  NET_NAME_UNKNOWN,
					  ip6gre_tunnel_setup);
X
xeb@mail.ru 已提交
1127 1128 1129 1130 1131
	if (!ign->fb_tunnel_dev) {
		err = -ENOMEM;
		goto err_alloc_dev;
	}
	dev_net_set(ign->fb_tunnel_dev, net);
N
Nicolas Dichtel 已提交
1132 1133 1134 1135 1136
	/* FB netdevice is special: we have one, and only one per netns.
	 * Allowing to move it to another netns is clearly unsafe.
	 */
	ign->fb_tunnel_dev->features |= NETIF_F_NETNS_LOCAL;

X
xeb@mail.ru 已提交
1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159

	ip6gre_fb_tunnel_init(ign->fb_tunnel_dev);
	ign->fb_tunnel_dev->rtnl_link_ops = &ip6gre_link_ops;

	err = register_netdev(ign->fb_tunnel_dev);
	if (err)
		goto err_reg_dev;

	rcu_assign_pointer(ign->tunnels_wc[0],
			   netdev_priv(ign->fb_tunnel_dev));
	return 0;

err_reg_dev:
	ip6gre_dev_free(ign->fb_tunnel_dev);
err_alloc_dev:
	return err;
}

static void __net_exit ip6gre_exit_net(struct net *net)
{
	LIST_HEAD(list);

	rtnl_lock();
N
Nicolas Dichtel 已提交
1160
	ip6gre_destroy_tunnels(net, &list);
X
xeb@mail.ru 已提交
1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204
	unregister_netdevice_many(&list);
	rtnl_unlock();
}

static struct pernet_operations ip6gre_net_ops = {
	.init = ip6gre_init_net,
	.exit = ip6gre_exit_net,
	.id   = &ip6gre_net_id,
	.size = sizeof(struct ip6gre_net),
};

static int ip6gre_tunnel_validate(struct nlattr *tb[], struct nlattr *data[])
{
	__be16 flags;

	if (!data)
		return 0;

	flags = 0;
	if (data[IFLA_GRE_IFLAGS])
		flags |= nla_get_be16(data[IFLA_GRE_IFLAGS]);
	if (data[IFLA_GRE_OFLAGS])
		flags |= nla_get_be16(data[IFLA_GRE_OFLAGS]);
	if (flags & (GRE_VERSION|GRE_ROUTING))
		return -EINVAL;

	return 0;
}

static int ip6gre_tap_validate(struct nlattr *tb[], struct nlattr *data[])
{
	struct in6_addr daddr;

	if (tb[IFLA_ADDRESS]) {
		if (nla_len(tb[IFLA_ADDRESS]) != ETH_ALEN)
			return -EINVAL;
		if (!is_valid_ether_addr(nla_data(tb[IFLA_ADDRESS])))
			return -EADDRNOTAVAIL;
	}

	if (!data)
		goto out;

	if (data[IFLA_GRE_REMOTE]) {
1205
		daddr = nla_get_in6_addr(data[IFLA_GRE_REMOTE]);
X
xeb@mail.ru 已提交
1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226
		if (ipv6_addr_any(&daddr))
			return -EINVAL;
	}

out:
	return ip6gre_tunnel_validate(tb, data);
}


static void ip6gre_netlink_parms(struct nlattr *data[],
				struct __ip6_tnl_parm *parms)
{
	memset(parms, 0, sizeof(*parms));

	if (!data)
		return;

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

	if (data[IFLA_GRE_IFLAGS])
T
Tom Herbert 已提交
1227 1228
		parms->i_flags = gre_flags_to_tnl_flags(
				nla_get_be16(data[IFLA_GRE_IFLAGS]));
X
xeb@mail.ru 已提交
1229 1230

	if (data[IFLA_GRE_OFLAGS])
T
Tom Herbert 已提交
1231 1232
		parms->o_flags = gre_flags_to_tnl_flags(
				nla_get_be16(data[IFLA_GRE_OFLAGS]));
X
xeb@mail.ru 已提交
1233 1234 1235 1236 1237 1238 1239 1240

	if (data[IFLA_GRE_IKEY])
		parms->i_key = nla_get_be32(data[IFLA_GRE_IKEY]);

	if (data[IFLA_GRE_OKEY])
		parms->o_key = nla_get_be32(data[IFLA_GRE_OKEY]);

	if (data[IFLA_GRE_LOCAL])
1241
		parms->laddr = nla_get_in6_addr(data[IFLA_GRE_LOCAL]);
X
xeb@mail.ru 已提交
1242 1243

	if (data[IFLA_GRE_REMOTE])
1244
		parms->raddr = nla_get_in6_addr(data[IFLA_GRE_REMOTE]);
X
xeb@mail.ru 已提交
1245 1246 1247 1248 1249 1250 1251 1252

	if (data[IFLA_GRE_TTL])
		parms->hop_limit = nla_get_u8(data[IFLA_GRE_TTL]);

	if (data[IFLA_GRE_ENCAP_LIMIT])
		parms->encap_limit = nla_get_u8(data[IFLA_GRE_ENCAP_LIMIT]);

	if (data[IFLA_GRE_FLOWINFO])
1253
		parms->flowinfo = nla_get_be32(data[IFLA_GRE_FLOWINFO]);
X
xeb@mail.ru 已提交
1254 1255 1256

	if (data[IFLA_GRE_FLAGS])
		parms->flags = nla_get_u32(data[IFLA_GRE_FLAGS]);
1257 1258 1259

	if (data[IFLA_GRE_FWMARK])
		parms->fwmark = nla_get_u32(data[IFLA_GRE_FWMARK]);
X
xeb@mail.ru 已提交
1260 1261 1262 1263 1264
}

static int ip6gre_tap_init(struct net_device *dev)
{
	struct ip6_tnl *tunnel;
1265
	int ret;
X
xeb@mail.ru 已提交
1266

1267 1268 1269
	ret = ip6gre_tunnel_init_common(dev);
	if (ret)
		return ret;
X
xeb@mail.ru 已提交
1270

1271 1272
	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;

1273
	tunnel = netdev_priv(dev);
X
xeb@mail.ru 已提交
1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285

	ip6gre_tnl_link_config(tunnel, 1);

	return 0;
}

static const struct net_device_ops ip6gre_tap_netdev_ops = {
	.ndo_init = ip6gre_tap_init,
	.ndo_uninit = ip6gre_tunnel_uninit,
	.ndo_start_xmit = ip6gre_tunnel_xmit,
	.ndo_set_mac_address = eth_mac_addr,
	.ndo_validate_addr = eth_validate_addr,
1286
	.ndo_change_mtu = ip6_tnl_change_mtu,
1287
	.ndo_get_stats64 = ip_tunnel_get_stats64,
1288
	.ndo_get_iflink = ip6_tnl_get_iflink,
X
xeb@mail.ru 已提交
1289 1290
};

1291 1292 1293 1294 1295
#define GRE6_FEATURES (NETIF_F_SG |		\
		       NETIF_F_FRAGLIST |	\
		       NETIF_F_HIGHDMA |		\
		       NETIF_F_HW_CSUM)

X
xeb@mail.ru 已提交
1296 1297 1298 1299 1300 1301 1302 1303 1304
static void ip6gre_tap_setup(struct net_device *dev)
{

	ether_setup(dev);

	dev->netdev_ops = &ip6gre_tap_netdev_ops;
	dev->destructor = ip6gre_dev_free;

	dev->features |= NETIF_F_NETNS_LOCAL;
J
Jiri Benc 已提交
1305
	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
1306
	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
X
xeb@mail.ru 已提交
1307 1308
}

1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341
static bool ip6gre_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_GRE_ENCAP_TYPE]) {
		ret = true;
		ipencap->type = nla_get_u16(data[IFLA_GRE_ENCAP_TYPE]);
	}

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

	if (data[IFLA_GRE_ENCAP_SPORT]) {
		ret = true;
		ipencap->sport = nla_get_be16(data[IFLA_GRE_ENCAP_SPORT]);
	}

	if (data[IFLA_GRE_ENCAP_DPORT]) {
		ret = true;
		ipencap->dport = nla_get_be16(data[IFLA_GRE_ENCAP_DPORT]);
	}

	return ret;
}

X
xeb@mail.ru 已提交
1342 1343 1344 1345 1346 1347
static int ip6gre_newlink(struct net *src_net, struct net_device *dev,
	struct nlattr *tb[], struct nlattr *data[])
{
	struct ip6_tnl *nt;
	struct net *net = dev_net(dev);
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
1348
	struct ip_tunnel_encap ipencap;
X
xeb@mail.ru 已提交
1349 1350 1351
	int err;

	nt = netdev_priv(dev);
1352 1353 1354 1355 1356 1357 1358 1359

	if (ip6gre_netlink_encap_parms(data, &ipencap)) {
		int err = ip6_tnl_encap_setup(nt, &ipencap);

		if (err < 0)
			return err;
	}

X
xeb@mail.ru 已提交
1360 1361 1362 1363 1364 1365 1366 1367 1368
	ip6gre_netlink_parms(data, &nt->parms);

	if (ip6gre_tunnel_find(net, &nt->parms, dev->type))
		return -EEXIST;

	if (dev->type == ARPHRD_ETHER && !tb[IFLA_ADDRESS])
		eth_hw_addr_random(dev);

	nt->dev = dev;
N
Nicolas Dichtel 已提交
1369
	nt->net = dev_net(dev);
X
xeb@mail.ru 已提交
1370 1371
	ip6gre_tnl_link_config(nt, !tb[IFLA_MTU]);

1372 1373 1374
	dev->features		|= GRE6_FEATURES;
	dev->hw_features	|= GRE6_FEATURES;

1375
	if (!(nt->parms.o_flags & TUNNEL_SEQ)) {
1376 1377 1378
		/* TCP offload with GRE SEQ is not supported, nor
		 * can we support 2 levels of outer headers requiring
		 * an update.
A
Alexander Duyck 已提交
1379
		 */
1380 1381 1382 1383 1384
		if (!(nt->parms.o_flags & TUNNEL_CSUM) ||
		    (nt->encap.type == TUNNEL_ENCAP_NONE)) {
			dev->features    |= NETIF_F_GSO_SOFTWARE;
			dev->hw_features |= NETIF_F_GSO_SOFTWARE;
		}
A
Alexander Duyck 已提交
1385 1386 1387 1388

		/* Can use a lockless transmit, unless we generate
		 * output sequences
		 */
X
xeb@mail.ru 已提交
1389
		dev->features |= NETIF_F_LLTX;
A
Alexander Duyck 已提交
1390
	}
X
xeb@mail.ru 已提交
1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405

	err = register_netdevice(dev);
	if (err)
		goto out;

	dev_hold(dev);
	ip6gre_tunnel_link(ign, nt);

out:
	return err;
}

static int ip6gre_changelink(struct net_device *dev, struct nlattr *tb[],
			    struct nlattr *data[])
{
N
Nicolas Dichtel 已提交
1406 1407
	struct ip6_tnl *t, *nt = netdev_priv(dev);
	struct net *net = nt->net;
X
xeb@mail.ru 已提交
1408 1409
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
	struct __ip6_tnl_parm p;
1410
	struct ip_tunnel_encap ipencap;
X
xeb@mail.ru 已提交
1411 1412 1413 1414

	if (dev == ign->fb_tunnel_dev)
		return -EINVAL;

1415 1416 1417 1418 1419 1420 1421
	if (ip6gre_netlink_encap_parms(data, &ipencap)) {
		int err = ip6_tnl_encap_setup(nt, &ipencap);

		if (err < 0)
			return err;
	}

X
xeb@mail.ru 已提交
1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432
	ip6gre_netlink_parms(data, &p);

	t = ip6gre_tunnel_locate(net, &p, 0);

	if (t) {
		if (t->dev != dev)
			return -EEXIST;
	} else {
		t = nt;
	}

1433 1434 1435
	ip6gre_tunnel_unlink(ign, t);
	ip6gre_tnl_change(t, &p, !tb[IFLA_MTU]);
	ip6gre_tunnel_link(ign, t);
X
xeb@mail.ru 已提交
1436 1437 1438
	return 0;
}

1439 1440 1441 1442 1443 1444 1445 1446 1447
static void ip6gre_dellink(struct net_device *dev, struct list_head *head)
{
	struct net *net = dev_net(dev);
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);

	if (dev != ign->fb_tunnel_dev)
		unregister_netdevice_queue(dev, head);
}

X
xeb@mail.ru 已提交
1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461
static size_t ip6gre_get_size(const struct net_device *dev)
{
	return
		/* IFLA_GRE_LINK */
		nla_total_size(4) +
		/* IFLA_GRE_IFLAGS */
		nla_total_size(2) +
		/* IFLA_GRE_OFLAGS */
		nla_total_size(2) +
		/* IFLA_GRE_IKEY */
		nla_total_size(4) +
		/* IFLA_GRE_OKEY */
		nla_total_size(4) +
		/* IFLA_GRE_LOCAL */
N
Nicolas Dichtel 已提交
1462
		nla_total_size(sizeof(struct in6_addr)) +
X
xeb@mail.ru 已提交
1463
		/* IFLA_GRE_REMOTE */
N
Nicolas Dichtel 已提交
1464
		nla_total_size(sizeof(struct in6_addr)) +
X
xeb@mail.ru 已提交
1465 1466 1467 1468 1469 1470 1471 1472
		/* IFLA_GRE_TTL */
		nla_total_size(1) +
		/* IFLA_GRE_ENCAP_LIMIT */
		nla_total_size(1) +
		/* IFLA_GRE_FLOWINFO */
		nla_total_size(4) +
		/* IFLA_GRE_FLAGS */
		nla_total_size(4) +
1473 1474 1475 1476 1477 1478 1479 1480
		/* IFLA_GRE_ENCAP_TYPE */
		nla_total_size(2) +
		/* IFLA_GRE_ENCAP_FLAGS */
		nla_total_size(2) +
		/* IFLA_GRE_ENCAP_SPORT */
		nla_total_size(2) +
		/* IFLA_GRE_ENCAP_DPORT */
		nla_total_size(2) +
1481 1482
		/* IFLA_GRE_FWMARK */
		nla_total_size(4) +
X
xeb@mail.ru 已提交
1483 1484 1485 1486 1487 1488 1489 1490 1491
		0;
}

static int ip6gre_fill_info(struct sk_buff *skb, const struct net_device *dev)
{
	struct ip6_tnl *t = netdev_priv(dev);
	struct __ip6_tnl_parm *p = &t->parms;

	if (nla_put_u32(skb, IFLA_GRE_LINK, p->link) ||
T
Tom Herbert 已提交
1492 1493 1494 1495
	    nla_put_be16(skb, IFLA_GRE_IFLAGS,
			 gre_tnl_flags_to_gre_flags(p->i_flags)) ||
	    nla_put_be16(skb, IFLA_GRE_OFLAGS,
			 gre_tnl_flags_to_gre_flags(p->o_flags)) ||
X
xeb@mail.ru 已提交
1496 1497
	    nla_put_be32(skb, IFLA_GRE_IKEY, p->i_key) ||
	    nla_put_be32(skb, IFLA_GRE_OKEY, p->o_key) ||
1498 1499
	    nla_put_in6_addr(skb, IFLA_GRE_LOCAL, &p->laddr) ||
	    nla_put_in6_addr(skb, IFLA_GRE_REMOTE, &p->raddr) ||
X
xeb@mail.ru 已提交
1500 1501 1502
	    nla_put_u8(skb, IFLA_GRE_TTL, p->hop_limit) ||
	    nla_put_u8(skb, IFLA_GRE_ENCAP_LIMIT, p->encap_limit) ||
	    nla_put_be32(skb, IFLA_GRE_FLOWINFO, p->flowinfo) ||
1503 1504
	    nla_put_u32(skb, IFLA_GRE_FLAGS, p->flags) ||
	    nla_put_u32(skb, IFLA_GRE_FWMARK, p->fwmark))
X
xeb@mail.ru 已提交
1505
		goto nla_put_failure;
1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516

	if (nla_put_u16(skb, IFLA_GRE_ENCAP_TYPE,
			t->encap.type) ||
	    nla_put_be16(skb, IFLA_GRE_ENCAP_SPORT,
			 t->encap.sport) ||
	    nla_put_be16(skb, IFLA_GRE_ENCAP_DPORT,
			 t->encap.dport) ||
	    nla_put_u16(skb, IFLA_GRE_ENCAP_FLAGS,
			t->encap.flags))
		goto nla_put_failure;

X
xeb@mail.ru 已提交
1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534
	return 0;

nla_put_failure:
	return -EMSGSIZE;
}

static const struct nla_policy ip6gre_policy[IFLA_GRE_MAX + 1] = {
	[IFLA_GRE_LINK]        = { .type = NLA_U32 },
	[IFLA_GRE_IFLAGS]      = { .type = NLA_U16 },
	[IFLA_GRE_OFLAGS]      = { .type = NLA_U16 },
	[IFLA_GRE_IKEY]        = { .type = NLA_U32 },
	[IFLA_GRE_OKEY]        = { .type = NLA_U32 },
	[IFLA_GRE_LOCAL]       = { .len = FIELD_SIZEOF(struct ipv6hdr, saddr) },
	[IFLA_GRE_REMOTE]      = { .len = FIELD_SIZEOF(struct ipv6hdr, daddr) },
	[IFLA_GRE_TTL]         = { .type = NLA_U8 },
	[IFLA_GRE_ENCAP_LIMIT] = { .type = NLA_U8 },
	[IFLA_GRE_FLOWINFO]    = { .type = NLA_U32 },
	[IFLA_GRE_FLAGS]       = { .type = NLA_U32 },
1535 1536 1537 1538
	[IFLA_GRE_ENCAP_TYPE]   = { .type = NLA_U16 },
	[IFLA_GRE_ENCAP_FLAGS]  = { .type = NLA_U16 },
	[IFLA_GRE_ENCAP_SPORT]  = { .type = NLA_U16 },
	[IFLA_GRE_ENCAP_DPORT]  = { .type = NLA_U16 },
1539
	[IFLA_GRE_FWMARK]       = { .type = NLA_U32 },
X
xeb@mail.ru 已提交
1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550
};

static struct rtnl_link_ops ip6gre_link_ops __read_mostly = {
	.kind		= "ip6gre",
	.maxtype	= IFLA_GRE_MAX,
	.policy		= ip6gre_policy,
	.priv_size	= sizeof(struct ip6_tnl),
	.setup		= ip6gre_tunnel_setup,
	.validate	= ip6gre_tunnel_validate,
	.newlink	= ip6gre_newlink,
	.changelink	= ip6gre_changelink,
1551
	.dellink	= ip6gre_dellink,
X
xeb@mail.ru 已提交
1552 1553
	.get_size	= ip6gre_get_size,
	.fill_info	= ip6gre_fill_info,
1554
	.get_link_net	= ip6_tnl_get_link_net,
X
xeb@mail.ru 已提交
1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567
};

static struct rtnl_link_ops ip6gre_tap_ops __read_mostly = {
	.kind		= "ip6gretap",
	.maxtype	= IFLA_GRE_MAX,
	.policy		= ip6gre_policy,
	.priv_size	= sizeof(struct ip6_tnl),
	.setup		= ip6gre_tap_setup,
	.validate	= ip6gre_tap_validate,
	.newlink	= ip6gre_newlink,
	.changelink	= ip6gre_changelink,
	.get_size	= ip6gre_get_size,
	.fill_info	= ip6gre_fill_info,
1568
	.get_link_net	= ip6_tnl_get_link_net,
X
xeb@mail.ru 已提交
1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624
};

/*
 *	And now the modules code and kernel interface.
 */

static int __init ip6gre_init(void)
{
	int err;

	pr_info("GRE over IPv6 tunneling driver\n");

	err = register_pernet_device(&ip6gre_net_ops);
	if (err < 0)
		return err;

	err = inet6_add_protocol(&ip6gre_protocol, IPPROTO_GRE);
	if (err < 0) {
		pr_info("%s: can't add protocol\n", __func__);
		goto add_proto_failed;
	}

	err = rtnl_link_register(&ip6gre_link_ops);
	if (err < 0)
		goto rtnl_link_failed;

	err = rtnl_link_register(&ip6gre_tap_ops);
	if (err < 0)
		goto tap_ops_failed;

out:
	return err;

tap_ops_failed:
	rtnl_link_unregister(&ip6gre_link_ops);
rtnl_link_failed:
	inet6_del_protocol(&ip6gre_protocol, IPPROTO_GRE);
add_proto_failed:
	unregister_pernet_device(&ip6gre_net_ops);
	goto out;
}

static void __exit ip6gre_fini(void)
{
	rtnl_link_unregister(&ip6gre_tap_ops);
	rtnl_link_unregister(&ip6gre_link_ops);
	inet6_del_protocol(&ip6gre_protocol, IPPROTO_GRE);
	unregister_pernet_device(&ip6gre_net_ops);
}

module_init(ip6gre_init);
module_exit(ip6gre_fini);
MODULE_LICENSE("GPL");
MODULE_AUTHOR("D. Kozlov (xeb@mail.ru)");
MODULE_DESCRIPTION("GRE over IPv6 tunneling device");
MODULE_ALIAS_RTNL_LINK("ip6gre");
1625
MODULE_ALIAS_RTNL_LINK("ip6gretap");
X
xeb@mail.ru 已提交
1626
MODULE_ALIAS_NETDEV("ip6gre0");