ip6_gre.c 52.2 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>
58
#include <net/erspan.h>
59
#include <net/dst_metadata.h>
X
xeb@mail.ru 已提交
60 61


62 63 64 65
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");

66 67
#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
68

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

73
	struct ip6_tnl __rcu *collect_md_tun;
X
xeb@mail.ru 已提交
74 75 76 77
	struct net_device *fb_tunnel_dev;
};

static struct rtnl_link_ops ip6gre_link_ops __read_mostly;
N
Nicolas Dichtel 已提交
78
static struct rtnl_link_ops ip6gre_tap_ops __read_mostly;
79
static struct rtnl_link_ops ip6erspan_tap_ops __read_mostly;
X
xeb@mail.ru 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
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.
 */

103
#define HASH_KEY(key) (((__force u32)key^((__force u32)key>>4))&(IP6_GRE_HASH_SIZE - 1))
X
xeb@mail.ru 已提交
104 105 106 107
static u32 HASH_ADDR(const struct in6_addr *addr)
{
	u32 hash = ipv6_addr_hash(addr);

108
	return hash_32(hash, IP6_GRE_HASH_SIZE_SHIFT);
X
xeb@mail.ru 已提交
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);
128
	int dev_type = (gre_proto == htons(ETH_P_TEB) ||
129 130
			gre_proto == htons(ETH_P_ERSPAN) ||
			gre_proto == htons(ETH_P_ERSPAN2)) ?
X
xeb@mail.ru 已提交
131 132 133
		       ARPHRD_ETHER : ARPHRD_IP6GRE;
	int score, cand_score = 4;

A
Amerigo Wang 已提交
134
	for_each_ip_tunnel_rcu(t, ign->tunnels_r_l[h0 ^ h1]) {
X
xeb@mail.ru 已提交
135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158
		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
159
	for_each_ip_tunnel_rcu(t, ign->tunnels_r[h0 ^ h1]) {
X
xeb@mail.ru 已提交
160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182
		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
183
	for_each_ip_tunnel_rcu(t, ign->tunnels_l[h1]) {
X
xeb@mail.ru 已提交
184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208
		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
209
	for_each_ip_tunnel_rcu(t, ign->tunnels_wc[h1]) {
X
xeb@mail.ru 已提交
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
		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;
		}
	}

232
	if (cand)
X
xeb@mail.ru 已提交
233 234
		return cand;

235 236 237 238
	t = rcu_dereference(ign->collect_md_tun);
	if (t && t->dev->flags & IFF_UP)
		return t;

X
xeb@mail.ru 已提交
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
	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);

274 275 276
	if (t->parms.collect_md)
		rcu_assign_pointer(ign->collect_md_tun, t);

X
xeb@mail.ru 已提交
277 278 279 280 281 282 283 284 285
	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;

286 287 288
	if (t->parms.collect_md)
		rcu_assign_pointer(ign->collect_md_tun, NULL);

X
xeb@mail.ru 已提交
289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332
	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);
333 334
	if (t && create)
		return NULL;
X
xeb@mail.ru 已提交
335 336 337 338 339 340 341 342
	if (t || !create)
		return t;

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

343 344
	dev = alloc_netdev(sizeof(*t), name, NET_NAME_UNKNOWN,
			   ip6gre_tunnel_setup);
X
xeb@mail.ru 已提交
345 346 347 348 349 350 351 352 353 354
	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
355
	nt->net = dev_net(dev);
X
xeb@mail.ru 已提交
356 357 358 359

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

360 361
	ip6gre_tnl_link_config(nt, 1);

X
xeb@mail.ru 已提交
362
	/* Can use a lockless transmit, unless we generate output sequences */
363
	if (!(nt->parms.o_flags & TUNNEL_SEQ))
X
xeb@mail.ru 已提交
364 365 366 367 368 369 370 371 372 373 374 375 376
		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
377 378
	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
379

N
Nicolas Dichtel 已提交
380
	ip6gre_tunnel_unlink(ign, t);
381
	dst_cache_reset(&t->dst_cache);
X
xeb@mail.ru 已提交
382 383 384 385 386
	dev_put(dev);
}


static void ip6gre_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
387
		       u8 type, u8 code, int offset, __be32 info)
X
xeb@mail.ru 已提交
388
{
389
	struct net *net = dev_net(skb->dev);
390 391 392
	const struct gre_base_hdr *greh;
	const struct ipv6hdr *ipv6h;
	int grehlen = sizeof(*greh);
X
xeb@mail.ru 已提交
393
	struct ip6_tnl *t;
394
	int key_off = 0;
X
xeb@mail.ru 已提交
395
	__be16 flags;
396
	__be32 key;
X
xeb@mail.ru 已提交
397

398 399 400 401 402 403 404 405 406 407 408
	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
409 410
	}

411
	if (!pskb_may_pull(skb, offset + grehlen))
X
xeb@mail.ru 已提交
412
		return;
E
Eric Dumazet 已提交
413
	ipv6h = (const struct ipv6hdr *)skb->data;
414 415
	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
416 417

	t = ip6gre_tunnel_lookup(skb->dev, &ipv6h->daddr, &ipv6h->saddr,
418
				 key, greh->protocol);
419
	if (!t)
420
		return;
X
xeb@mail.ru 已提交
421 422 423

	switch (type) {
		struct ipv6_tlv_tnl_enc_lim *tel;
X
Xin Long 已提交
424
		__u32 teli;
X
xeb@mail.ru 已提交
425
	case ICMPV6_DEST_UNREACH:
426 427
		net_dbg_ratelimited("%s: Path to destination invalid or inactive!\n",
				    t->parms.name);
428 429 430
		if (code != ICMPV6_PORT_UNREACH)
			break;
		return;
X
xeb@mail.ru 已提交
431 432
	case ICMPV6_TIME_EXCEED:
		if (code == ICMPV6_EXC_HOPLIMIT) {
433 434
			net_dbg_ratelimited("%s: Too small hop limit or routing loop in tunnel!\n",
					    t->parms.name);
435
			break;
X
xeb@mail.ru 已提交
436
		}
437
		return;
X
xeb@mail.ru 已提交
438 439 440 441 442
	case ICMPV6_PARAMPROB:
		teli = 0;
		if (code == ICMPV6_HDR_FIELD)
			teli = ip6_tnl_parse_tlv_enc_lim(skb, skb->data);

443
		if (teli && teli == be32_to_cpu(info) - 2) {
X
xeb@mail.ru 已提交
444 445
			tel = (struct ipv6_tlv_tnl_enc_lim *) &skb->data[teli];
			if (tel->encap_limit == 0) {
446 447
				net_dbg_ratelimited("%s: Too small encapsulation limit or routing loop in tunnel!\n",
						    t->parms.name);
X
xeb@mail.ru 已提交
448 449
			}
		} else {
450 451
			net_dbg_ratelimited("%s: Recipient unable to parse tunneled packet!\n",
					    t->parms.name);
X
xeb@mail.ru 已提交
452
		}
453
		return;
X
xeb@mail.ru 已提交
454
	case ICMPV6_PKT_TOOBIG:
X
Xin Long 已提交
455
		ip6_update_pmtu(skb, net, info, 0, 0, sock_net_uid(net, NULL));
456
		return;
457 458 459 460
	case NDISC_REDIRECT:
		ip6_redirect(skb, net, skb->dev->ifindex, 0,
			     sock_net_uid(net, NULL));
		return;
X
xeb@mail.ru 已提交
461 462 463 464 465 466 467 468 469
	}

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

470
static int ip6gre_rcv(struct sk_buff *skb, const struct tnl_ptk_info *tpi)
X
xeb@mail.ru 已提交
471 472 473 474 475 476
{
	const struct ipv6hdr *ipv6h;
	struct ip6_tnl *tunnel;

	ipv6h = ipv6_hdr(skb);
	tunnel = ip6gre_tunnel_lookup(skb->dev,
477 478
				      &ipv6h->saddr, &ipv6h->daddr, tpi->key,
				      tpi->proto);
X
xeb@mail.ru 已提交
479
	if (tunnel) {
480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495
		if (tunnel->parms.collect_md) {
			struct metadata_dst *tun_dst;
			__be64 tun_id;
			__be16 flags;

			flags = tpi->flags;
			tun_id = key32_to_tunnel_id(tpi->key);

			tun_dst = ipv6_tun_rx_dst(skb, flags, tun_id, 0);
			if (!tun_dst)
				return PACKET_REJECT;

			ip6_tnl_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error);
		} else {
			ip6_tnl_rcv(tunnel, skb, tpi, NULL, log_ecn_error);
		}
X
xeb@mail.ru 已提交
496

497 498
		return PACKET_RCVD;
	}
499

500 501
	return PACKET_REJECT;
}
502

503 504 505
static int ip6erspan_rcv(struct sk_buff *skb, int gre_hdr_len,
			 struct tnl_ptk_info *tpi)
{
506 507
	struct erspan_base_hdr *ershdr;
	struct erspan_metadata *pkt_md;
508
	const struct ipv6hdr *ipv6h;
509
	struct erspan_md2 *md2;
510
	struct ip6_tnl *tunnel;
511
	u8 ver;
512 513 514 515

	if (unlikely(!pskb_may_pull(skb, sizeof(*ershdr))))
		return PACKET_REJECT;

516 517
	ipv6h = ipv6_hdr(skb);
	ershdr = (struct erspan_base_hdr *)skb->data;
518 519
	ver = ershdr->ver;
	tpi->key = cpu_to_be32(get_session_id(ershdr));
520 521 522 523 524

	tunnel = ip6gre_tunnel_lookup(skb->dev,
				      &ipv6h->saddr, &ipv6h->daddr, tpi->key,
				      tpi->proto);
	if (tunnel) {
525 526 527
		int len = erspan_hdr_len(ver);

		if (unlikely(!pskb_may_pull(skb, len)))
W
William Tu 已提交
528
			return PACKET_REJECT;
529

530 531 532
		ershdr = (struct erspan_base_hdr *)skb->data;
		pkt_md = (struct erspan_metadata *)(ershdr + 1);

533
		if (__iptunnel_pull_header(skb, len,
534 535 536 537
					   htons(ETH_P_TEB),
					   false, false) < 0)
			return PACKET_REJECT;

538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555
		if (tunnel->parms.collect_md) {
			struct metadata_dst *tun_dst;
			struct ip_tunnel_info *info;
			struct erspan_metadata *md;
			__be64 tun_id;
			__be16 flags;

			tpi->flags |= TUNNEL_KEY;
			flags = tpi->flags;
			tun_id = key32_to_tunnel_id(tpi->key);

			tun_dst = ipv6_tun_rx_dst(skb, flags, tun_id,
						  sizeof(*md));
			if (!tun_dst)
				return PACKET_REJECT;

			info = &tun_dst->u.tun_info;
			md = ip_tunnel_info_opts(info);
W
William Tu 已提交
556
			md->version = ver;
557 558 559
			md2 = &md->u.md2;
			memcpy(md2, pkt_md, ver == 1 ? ERSPAN_V1_MDSIZE :
						       ERSPAN_V2_MDSIZE);
560 561 562 563 564 565 566 567
			info->key.tun_flags |= TUNNEL_ERSPAN_OPT;
			info->options_len = sizeof(*md);

			ip6_tnl_rcv(tunnel, skb, tpi, tun_dst, log_ecn_error);

		} else {
			ip6_tnl_rcv(tunnel, skb, tpi, NULL, log_ecn_error);
		}
568 569 570 571 572 573 574

		return PACKET_RCVD;
	}

	return PACKET_REJECT;
}

575 576 577 578 579
static int gre_rcv(struct sk_buff *skb)
{
	struct tnl_ptk_info tpi;
	bool csum_err = false;
	int hdr_len;
580

E
Eric Dumazet 已提交
581
	hdr_len = gre_parse_header(skb, &tpi, &csum_err, htons(ETH_P_IPV6), 0);
582
	if (hdr_len < 0)
583
		goto drop;
X
xeb@mail.ru 已提交
584

585 586
	if (iptunnel_pull_header(skb, hdr_len, tpi.proto, false))
		goto drop;
X
xeb@mail.ru 已提交
587

W
William Tu 已提交
588 589
	if (unlikely(tpi.proto == htons(ETH_P_ERSPAN) ||
		     tpi.proto == htons(ETH_P_ERSPAN2))) {
590 591
		if (ip6erspan_rcv(skb, hdr_len, &tpi) == PACKET_RCVD)
			return 0;
592
		goto out;
593 594
	}

595
	if (ip6gre_rcv(skb, &tpi) == PACKET_RCVD)
X
xeb@mail.ru 已提交
596 597
		return 0;

598
out:
599
	icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);
X
xeb@mail.ru 已提交
600 601 602 603 604
drop:
	kfree_skb(skb);
	return 0;
}

605
static int gre_handle_offloads(struct sk_buff *skb, bool csum)
X
xeb@mail.ru 已提交
606
{
607 608
	return iptunnel_handle_offloads(skb,
					csum ? SKB_GSO_GRE_CSUM : SKB_GSO_GRE);
X
xeb@mail.ru 已提交
609 610
}

W
William Tu 已提交
611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682
static void prepare_ip6gre_xmit_ipv4(struct sk_buff *skb,
				     struct net_device *dev,
				     struct flowi6 *fl6, __u8 *dsfield,
				     int *encap_limit)
{
	const struct iphdr *iph = ip_hdr(skb);
	struct ip6_tnl *t = netdev_priv(dev);

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

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

	if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)
		*dsfield = ipv4_get_dsfield(iph);
	else
		*dsfield = ip6_tclass(t->parms.flowinfo);

	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FWMARK)
		fl6->flowi6_mark = skb->mark;
	else
		fl6->flowi6_mark = t->parms.fwmark;

	fl6->flowi6_uid = sock_net_uid(dev_net(dev), NULL);
}

static int prepare_ip6gre_xmit_ipv6(struct sk_buff *skb,
				    struct net_device *dev,
				    struct flowi6 *fl6, __u8 *dsfield,
				    int *encap_limit)
{
	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
	struct ip6_tnl *t = netdev_priv(dev);
	__u16 offset;

	offset = ip6_tnl_parse_tlv_enc_lim(skb, skb_network_header(skb));
	/* ip6_tnl_parse_tlv_enc_lim() might have reallocated skb->head */

	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));

	if (t->parms.flags & IP6_TNL_F_USE_ORIG_TCLASS)
		*dsfield = ipv6_get_dsfield(ipv6h);
	else
		*dsfield = ip6_tclass(t->parms.flowinfo);

	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FLOWLABEL)
		fl6->flowlabel |= ip6_flowlabel(ipv6h);

	if (t->parms.flags & IP6_TNL_F_USE_ORIG_FWMARK)
		fl6->flowi6_mark = skb->mark;
	else
		fl6->flowi6_mark = t->parms.fwmark;

	fl6->flowi6_uid = sock_net_uid(dev_net(dev), NULL);

	return 0;
}

683 684 685 686
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
687 688
{
	struct ip6_tnl *tunnel = netdev_priv(dev);
689
	__be16 protocol;
X
xeb@mail.ru 已提交
690 691 692 693

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

694 695 696
	if (dev->header_ops && dev->type == ARPHRD_IP6GRE)
		fl6->daddr = ((struct ipv6hdr *)skb->data)->daddr;
	else
X
xeb@mail.ru 已提交
697 698
		fl6->daddr = tunnel->parms.raddr;

699 700
	if (tunnel->parms.o_flags & TUNNEL_SEQ)
		tunnel->o_seqno++;
X
xeb@mail.ru 已提交
701

702
	/* Push GRE header. */
703
	protocol = (dev->type == ARPHRD_ETHER) ? htons(ETH_P_TEB) : proto;
704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735

	if (tunnel->parms.collect_md) {
		struct ip_tunnel_info *tun_info;
		const struct ip_tunnel_key *key;
		__be16 flags;

		tun_info = skb_tunnel_info(skb);
		if (unlikely(!tun_info ||
			     !(tun_info->mode & IP_TUNNEL_INFO_TX) ||
			     ip_tunnel_info_af(tun_info) != AF_INET6))
			return -EINVAL;

		key = &tun_info->key;
		memset(fl6, 0, sizeof(*fl6));
		fl6->flowi6_proto = IPPROTO_GRE;
		fl6->daddr = key->u.ipv6.dst;
		fl6->flowlabel = key->label;
		fl6->flowi6_uid = sock_net_uid(dev_net(dev), NULL);

		dsfield = key->tos;
		flags = key->tun_flags & (TUNNEL_CSUM | TUNNEL_KEY);
		tunnel->tun_hlen = gre_calc_hlen(flags);

		gre_build_header(skb, tunnel->tun_hlen,
				 flags, protocol,
				 tunnel_id_to_key32(tun_info->key.tun_id), 0);

	} else {
		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
736

737 738
	return ip6_tnl_xmit(skb, dev, dsfield, fl6, encap_limit, pmtu,
			    NEXTHDR_GRE);
X
xeb@mail.ru 已提交
739 740 741 742 743 744 745
}

static inline int ip6gre_xmit_ipv4(struct sk_buff *skb, struct net_device *dev)
{
	struct ip6_tnl *t = netdev_priv(dev);
	int encap_limit = -1;
	struct flowi6 fl6;
746
	__u8 dsfield = 0;
X
xeb@mail.ru 已提交
747 748 749
	__u32 mtu;
	int err;

750 751
	memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));

752 753 754
	if (!t->parms.collect_md)
		prepare_ip6gre_xmit_ipv4(skb, dev, &fl6,
					 &dsfield, &encap_limit);
755

756 757 758 759 760 761
	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
762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778
	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;
	struct flowi6 fl6;
779
	__u8 dsfield = 0;
X
xeb@mail.ru 已提交
780 781 782 783 784 785
	__u32 mtu;
	int err;

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

786 787
	if (!t->parms.collect_md &&
	    prepare_ip6gre_xmit_ipv6(skb, dev, &fl6, &dsfield, &encap_limit))
W
William Tu 已提交
788
		return -1;
789

790 791 792 793 794
	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
795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834
	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;

835 836
	if (!t->parms.collect_md)
		memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));
X
xeb@mail.ru 已提交
837

838 839 840 841 842
	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
843 844 845 846 847 848 849 850 851 852 853

	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;

854
	if (!ip6_tnl_xmit_ctl(t, &t->parms.laddr, &t->parms.raddr))
855
		goto tx_err;
X
xeb@mail.ru 已提交
856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880

	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;
}

881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907
static netdev_tx_t ip6erspan_tunnel_xmit(struct sk_buff *skb,
					 struct net_device *dev)
{
	struct ipv6hdr *ipv6h = ipv6_hdr(skb);
	struct ip6_tnl *t = netdev_priv(dev);
	struct dst_entry *dst = skb_dst(skb);
	struct net_device_stats *stats;
	bool truncate = false;
	int encap_limit = -1;
	__u8 dsfield = false;
	struct flowi6 fl6;
	int err = -EINVAL;
	__u32 mtu;

	if (!ip6_tnl_xmit_ctl(t, &t->parms.laddr, &t->parms.raddr))
		goto tx_err;

	if (gre_handle_offloads(skb, false))
		goto tx_err;

	if (skb->len > dev->mtu + dev->hard_header_len) {
		pskb_trim(skb, dev->mtu + dev->hard_header_len);
		truncate = true;
	}

	t->parms.o_flags &= ~TUNNEL_KEY;
	IPCB(skb)->flags = 0;
908 909 910 911 912 913 914 915

	/* For collect_md mode, derive fl6 from the tunnel key,
	 * for native mode, call prepare_ip6gre_xmit_{ipv4,ipv6}.
	 */
	if (t->parms.collect_md) {
		struct ip_tunnel_info *tun_info;
		const struct ip_tunnel_key *key;
		struct erspan_metadata *md;
916
		__be32 tun_id;
917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935

		tun_info = skb_tunnel_info(skb);
		if (unlikely(!tun_info ||
			     !(tun_info->mode & IP_TUNNEL_INFO_TX) ||
			     ip_tunnel_info_af(tun_info) != AF_INET6))
			return -EINVAL;

		key = &tun_info->key;
		memset(&fl6, 0, sizeof(fl6));
		fl6.flowi6_proto = IPPROTO_GRE;
		fl6.daddr = key->u.ipv6.dst;
		fl6.flowlabel = key->label;
		fl6.flowi6_uid = sock_net_uid(dev_net(dev), NULL);

		dsfield = key->tos;
		md = ip_tunnel_info_opts(tun_info);
		if (!md)
			goto tx_err;

936
		tun_id = tunnel_id_to_key32(key->tun_id);
W
William Tu 已提交
937 938
		if (md->version == 1) {
			erspan_build_header(skb,
939
					    ntohl(tun_id),
W
William Tu 已提交
940 941 942 943
					    ntohl(md->u.index), truncate,
					    false);
		} else if (md->version == 2) {
			erspan_build_header_v2(skb,
944 945 946 947
					       ntohl(tun_id),
					       md->u.md2.dir,
					       get_hwid(&md->u.md2),
					       truncate, false);
W
William Tu 已提交
948
		}
949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967
	} else {
		switch (skb->protocol) {
		case htons(ETH_P_IP):
			memset(&(IPCB(skb)->opt), 0, sizeof(IPCB(skb)->opt));
			prepare_ip6gre_xmit_ipv4(skb, dev, &fl6,
						 &dsfield, &encap_limit);
			break;
		case htons(ETH_P_IPV6):
			if (ipv6_addr_equal(&t->parms.raddr, &ipv6h->saddr))
				goto tx_err;
			if (prepare_ip6gre_xmit_ipv6(skb, dev, &fl6,
						     &dsfield, &encap_limit))
				goto tx_err;
			break;
		default:
			memcpy(&fl6, &t->fl.u.ip6, sizeof(fl6));
			break;
		}

W
William Tu 已提交
968
		if (t->parms.erspan_ver == 1)
969
			erspan_build_header(skb, ntohl(t->parms.o_key),
W
William Tu 已提交
970 971 972
					    t->parms.index,
					    truncate, false);
		else
973
			erspan_build_header_v2(skb, ntohl(t->parms.o_key),
W
William Tu 已提交
974 975 976
					       t->parms.dir,
					       t->parms.hwid,
					       truncate, false);
977 978
		fl6.daddr = t->parms.raddr;
	}
979 980 981 982 983 984

	/* Push GRE header. */
	gre_build_header(skb, 8, TUNNEL_SEQ,
			 htons(ETH_P_ERSPAN), 0, htonl(t->o_seqno++));

	/* TooBig packet may have updated dst->dev's mtu */
985
	if (!t->parms.collect_md && dst && dst_mtu(dst) > dst->dev->mtu)
986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
		dst->ops->update_pmtu(dst, NULL, skb, dst->dev->mtu);

	err = ip6_tnl_xmit(skb, dev, dsfield, &fl6, encap_limit, &mtu,
			   NEXTHDR_GRE);
	if (err != 0) {
		/* XXX: send ICMP error even if DF is not set. */
		if (err == -EMSGSIZE) {
			if (skb->protocol == htons(ETH_P_IP))
				icmp_send(skb, ICMP_DEST_UNREACH,
					  ICMP_FRAG_NEEDED, htonl(mtu));
			else
				icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
		}

		goto tx_err;
	}
	return NETDEV_TX_OK;

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

X
xeb@mail.ru 已提交
1012 1013 1014 1015 1016
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
1017
	int t_hlen;
X
xeb@mail.ru 已提交
1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028

	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;
1029
	fl6->flowi6_proto = IPPROTO_GRE;
X
xeb@mail.ru 已提交
1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044

	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
1045 1046
	t->tun_hlen = gre_calc_hlen(t->parms.o_flags);

1047
	t->hlen = t->encap_hlen + t->tun_hlen;
T
Tom Herbert 已提交
1048 1049

	t_hlen = t->hlen + sizeof(struct ipv6hdr);
X
xeb@mail.ru 已提交
1050 1051 1052 1053 1054

	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
1055
		struct rt6_info *rt = rt6_lookup(t->net,
X
xeb@mail.ru 已提交
1056 1057 1058
						 &p->raddr, &p->laddr,
						 p->link, strict);

1059
		if (!rt)
X
xeb@mail.ru 已提交
1060 1061 1062
			return;

		if (rt->dst.dev) {
T
Tom Herbert 已提交
1063 1064
			dev->hard_header_len = rt->dst.dev->hard_header_len +
					       t_hlen;
X
xeb@mail.ru 已提交
1065 1066

			if (set_mtu) {
T
Tom Herbert 已提交
1067
				dev->mtu = rt->dst.dev->mtu - t_hlen;
X
xeb@mail.ru 已提交
1068 1069
				if (!(t->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
					dev->mtu -= 8;
1070 1071
				if (dev->type == ARPHRD_ETHER)
					dev->mtu -= ETH_HLEN;
X
xeb@mail.ru 已提交
1072 1073 1074 1075 1076

				if (dev->mtu < IPV6_MIN_MTU)
					dev->mtu = IPV6_MIN_MTU;
			}
		}
A
Amerigo Wang 已提交
1077
		ip6_rt_put(rt);
X
xeb@mail.ru 已提交
1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095
	}
}

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;
1096
	t->parms.fwmark = p->fwmark;
1097
	dst_cache_reset(&t->dst_cache);
X
xeb@mail.ru 已提交
1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113
	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
1114 1115
	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
1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131
	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
1132 1133
	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
1134 1135 1136 1137 1138 1139 1140 1141 1142
	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
1143 1144
	struct ip6_tnl *t = netdev_priv(dev);
	struct net *net = t->net;
X
xeb@mail.ru 已提交
1145 1146
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);

1147 1148
	memset(&p1, 0, sizeof(p1));

X
xeb@mail.ru 已提交
1149 1150 1151 1152 1153 1154 1155 1156 1157
	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);
1158
			if (!t)
N
Nicolas Dichtel 已提交
1159
				t = netdev_priv(dev);
X
xeb@mail.ru 已提交
1160
		}
1161
		memset(&p, 0, sizeof(p));
X
xeb@mail.ru 已提交
1162 1163 1164 1165 1166 1167 1168 1169
		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;
1170
		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
X
xeb@mail.ru 已提交
1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189
			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) {
1190
			if (t) {
X
xeb@mail.ru 已提交
1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208
				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;

1209
			memset(&p, 0, sizeof(p));
X
xeb@mail.ru 已提交
1210 1211 1212 1213 1214 1215 1216 1217 1218
			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;
1219
		if (!ns_capable(net->user_ns, CAP_NET_ADMIN))
X
xeb@mail.ru 已提交
1220 1221 1222 1223 1224 1225 1226 1227 1228
			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);
1229
			if (!t)
X
xeb@mail.ru 已提交
1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248
				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,
1249 1250
			 unsigned short type, const void *daddr,
			 const void *saddr, unsigned int len)
X
xeb@mail.ru 已提交
1251 1252
{
	struct ip6_tnl *t = netdev_priv(dev);
1253 1254
	struct ipv6hdr *ipv6h;
	__be16 *p;
X
xeb@mail.ru 已提交
1255

1256 1257 1258 1259
	ipv6h = skb_push(skb, t->hlen + sizeof(*ipv6h));
	ip6_flow_hdr(ipv6h, 0, ip6_make_flowlabel(dev_net(dev), skb,
						  t->fl.u.ip6.flowlabel,
						  true, &t->fl.u.ip6));
X
xeb@mail.ru 已提交
1260 1261 1262 1263 1264
	ipv6h->hop_limit = t->parms.hop_limit;
	ipv6h->nexthdr = NEXTHDR_GRE;
	ipv6h->saddr = t->parms.laddr;
	ipv6h->daddr = t->parms.raddr;

1265 1266 1267
	p = (__be16 *)(ipv6h + 1);
	p[0] = t->parms.o_flags;
	p[1] = htons(type);
X
xeb@mail.ru 已提交
1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291

	/*
	 *	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,
1292
	.ndo_change_mtu		= ip6_tnl_change_mtu,
1293
	.ndo_get_stats64	= ip_tunnel_get_stats64,
1294
	.ndo_get_iflink		= ip6_tnl_get_iflink,
X
xeb@mail.ru 已提交
1295 1296 1297 1298
};

static void ip6gre_dev_free(struct net_device *dev)
{
1299 1300
	struct ip6_tnl *t = netdev_priv(dev);

1301
	dst_cache_destroy(&t->dst_cache);
X
xeb@mail.ru 已提交
1302 1303 1304 1305 1306 1307
	free_percpu(dev->tstats);
}

static void ip6gre_tunnel_setup(struct net_device *dev)
{
	dev->netdev_ops = &ip6gre_netdev_ops;
1308 1309
	dev->needs_free_netdev = true;
	dev->priv_destructor = ip6gre_dev_free;
X
xeb@mail.ru 已提交
1310 1311

	dev->type = ARPHRD_IP6GRE;
1312

X
xeb@mail.ru 已提交
1313 1314
	dev->flags |= IFF_NOARP;
	dev->addr_len = sizeof(struct in6_addr);
1315
	netif_keep_dst(dev);
1316 1317 1318
	/* 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
1319 1320
}

1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350
#define GRE6_FEATURES (NETIF_F_SG |		\
		       NETIF_F_FRAGLIST |	\
		       NETIF_F_HIGHDMA |	\
		       NETIF_F_HW_CSUM)

static void ip6gre_tnl_init_features(struct net_device *dev)
{
	struct ip6_tnl *nt = netdev_priv(dev);

	dev->features		|= GRE6_FEATURES;
	dev->hw_features	|= GRE6_FEATURES;

	if (!(nt->parms.o_flags & TUNNEL_SEQ)) {
		/* TCP offload with GRE SEQ is not supported, nor
		 * can we support 2 levels of outer headers requiring
		 * an update.
		 */
		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;
		}

		/* Can use a lockless transmit, unless we generate
		 * output sequences
		 */
		dev->features |= NETIF_F_LLTX;
	}
}

1351
static int ip6gre_tunnel_init_common(struct net_device *dev)
X
xeb@mail.ru 已提交
1352 1353
{
	struct ip6_tnl *tunnel;
1354
	int ret;
1355
	int t_hlen;
X
xeb@mail.ru 已提交
1356 1357 1358 1359

	tunnel = netdev_priv(dev);

	tunnel->dev = dev;
N
Nicolas Dichtel 已提交
1360
	tunnel->net = dev_net(dev);
X
xeb@mail.ru 已提交
1361 1362
	strcpy(tunnel->parms.name, dev->name);

1363 1364 1365 1366
	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
	if (!dev->tstats)
		return -ENOMEM;

1367
	ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
1368 1369 1370 1371 1372 1373
	if (ret) {
		free_percpu(dev->tstats);
		dev->tstats = NULL;
		return ret;
	}

1374
	tunnel->tun_hlen = gre_calc_hlen(tunnel->parms.o_flags);
1375
	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
T
Tom Herbert 已提交
1376
	t_hlen = tunnel->hlen + sizeof(struct ipv6hdr);
1377

T
Tom Herbert 已提交
1378 1379
	dev->hard_header_len = LL_MAX_HEADER + t_hlen;
	dev->mtu = ETH_DATA_LEN - t_hlen;
1380 1381
	if (dev->type == ARPHRD_ETHER)
		dev->mtu -= ETH_HLEN;
1382 1383 1384
	if (!(tunnel->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
		dev->mtu -= 8;

1385 1386 1387 1388
	if (tunnel->parms.collect_md) {
		dev->features |= NETIF_F_NETNS_LOCAL;
		netif_keep_dst(dev);
	}
1389
	ip6gre_tnl_init_features(dev);
1390

1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404
	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);

1405 1406 1407
	if (tunnel->parms.collect_md)
		return 0;

X
xeb@mail.ru 已提交
1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421
	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
1422
	tunnel->net = dev_net(dev);
X
xeb@mail.ru 已提交
1423 1424 1425 1426 1427 1428 1429 1430
	strcpy(tunnel->parms.name, dev->name);

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

	dev_hold(dev);
}

static struct inet6_protocol ip6gre_protocol __read_mostly = {
1431
	.handler     = gre_rcv,
X
xeb@mail.ru 已提交
1432 1433 1434 1435
	.err_handler = ip6gre_err,
	.flags       = INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
};

N
Nicolas Dichtel 已提交
1436
static void ip6gre_destroy_tunnels(struct net *net, struct list_head *head)
X
xeb@mail.ru 已提交
1437
{
N
Nicolas Dichtel 已提交
1438 1439
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
	struct net_device *dev, *aux;
X
xeb@mail.ru 已提交
1440 1441
	int prio;

N
Nicolas Dichtel 已提交
1442 1443
	for_each_netdev_safe(net, dev, aux)
		if (dev->rtnl_link_ops == &ip6gre_link_ops ||
1444 1445
		    dev->rtnl_link_ops == &ip6gre_tap_ops ||
		    dev->rtnl_link_ops == &ip6erspan_tap_ops)
N
Nicolas Dichtel 已提交
1446 1447
			unregister_netdevice_queue(dev, head);

X
xeb@mail.ru 已提交
1448 1449
	for (prio = 0; prio < 4; prio++) {
		int h;
1450
		for (h = 0; h < IP6_GRE_HASH_SIZE; h++) {
X
xeb@mail.ru 已提交
1451 1452 1453 1454
			struct ip6_tnl *t;

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

1455
			while (t) {
N
Nicolas Dichtel 已提交
1456 1457 1458 1459 1460 1461
				/* 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
1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473
				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",
1474 1475
					  NET_NAME_UNKNOWN,
					  ip6gre_tunnel_setup);
X
xeb@mail.ru 已提交
1476 1477 1478 1479 1480
	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
1481 1482 1483 1484 1485
	/* 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
1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498

	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:
1499
	free_netdev(ign->fb_tunnel_dev);
X
xeb@mail.ru 已提交
1500 1501 1502 1503
err_alloc_dev:
	return err;
}

1504
static void __net_exit ip6gre_exit_batch_net(struct list_head *net_list)
X
xeb@mail.ru 已提交
1505
{
1506
	struct net *net;
X
xeb@mail.ru 已提交
1507 1508 1509
	LIST_HEAD(list);

	rtnl_lock();
1510 1511
	list_for_each_entry(net, net_list, exit_list)
		ip6gre_destroy_tunnels(net, &list);
X
xeb@mail.ru 已提交
1512 1513 1514 1515 1516 1517
	unregister_netdevice_many(&list);
	rtnl_unlock();
}

static struct pernet_operations ip6gre_net_ops = {
	.init = ip6gre_init_net,
1518
	.exit_batch = ip6gre_exit_batch_net,
X
xeb@mail.ru 已提交
1519 1520 1521 1522
	.id   = &ip6gre_net_id,
	.size = sizeof(struct ip6gre_net),
};

1523 1524
static int ip6gre_tunnel_validate(struct nlattr *tb[], struct nlattr *data[],
				  struct netlink_ext_ack *extack)
X
xeb@mail.ru 已提交
1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541
{
	__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;
}

1542 1543
static int ip6gre_tap_validate(struct nlattr *tb[], struct nlattr *data[],
			       struct netlink_ext_ack *extack)
X
xeb@mail.ru 已提交
1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557
{
	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]) {
1558
		daddr = nla_get_in6_addr(data[IFLA_GRE_REMOTE]);
X
xeb@mail.ru 已提交
1559 1560 1561 1562 1563
		if (ipv6_addr_any(&daddr))
			return -EINVAL;
	}

out:
1564
	return ip6gre_tunnel_validate(tb, data, extack);
X
xeb@mail.ru 已提交
1565 1566
}

1567 1568 1569 1570
static int ip6erspan_tap_validate(struct nlattr *tb[], struct nlattr *data[],
				  struct netlink_ext_ack *extack)
{
	__be16 flags = 0;
W
William Tu 已提交
1571
	int ret, ver = 0;
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

	if (!data)
		return 0;

	ret = ip6gre_tap_validate(tb, data, extack);
	if (ret)
		return ret;

	/* ERSPAN should only have GRE sequence and key flag */
	if (data[IFLA_GRE_OFLAGS])
		flags |= nla_get_be16(data[IFLA_GRE_OFLAGS]);
	if (data[IFLA_GRE_IFLAGS])
		flags |= nla_get_be16(data[IFLA_GRE_IFLAGS]);
	if (!data[IFLA_GRE_COLLECT_METADATA] &&
	    flags != (GRE_SEQ | GRE_KEY))
		return -EINVAL;

	/* ERSPAN Session ID only has 10-bit. Since we reuse
	 * 32-bit key field as ID, check it's range.
	 */
	if (data[IFLA_GRE_IKEY] &&
	    (ntohl(nla_get_be32(data[IFLA_GRE_IKEY])) & ~ID_MASK))
		return -EINVAL;

	if (data[IFLA_GRE_OKEY] &&
	    (ntohl(nla_get_be32(data[IFLA_GRE_OKEY])) & ~ID_MASK))
		return -EINVAL;

W
William Tu 已提交
1600 1601 1602
	if (data[IFLA_GRE_ERSPAN_VER]) {
		ver = nla_get_u8(data[IFLA_GRE_ERSPAN_VER]);
		if (ver != 1 && ver != 2)
1603 1604
			return -EINVAL;
	}
W
William Tu 已提交
1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628

	if (ver == 1) {
		if (data[IFLA_GRE_ERSPAN_INDEX]) {
			u32 index = nla_get_u32(data[IFLA_GRE_ERSPAN_INDEX]);

			if (index & ~INDEX_MASK)
				return -EINVAL;
		}
	} else if (ver == 2) {
		if (data[IFLA_GRE_ERSPAN_DIR]) {
			u16 dir = nla_get_u8(data[IFLA_GRE_ERSPAN_DIR]);

			if (dir & ~(DIR_MASK >> DIR_OFFSET))
				return -EINVAL;
		}

		if (data[IFLA_GRE_ERSPAN_HWID]) {
			u16 hwid = nla_get_u16(data[IFLA_GRE_ERSPAN_HWID]);

			if (hwid & ~(HWID_MASK >> HWID_OFFSET))
				return -EINVAL;
		}
	}

1629 1630
	return 0;
}
X
xeb@mail.ru 已提交
1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643

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
1644 1645
		parms->i_flags = gre_flags_to_tnl_flags(
				nla_get_be16(data[IFLA_GRE_IFLAGS]));
X
xeb@mail.ru 已提交
1646 1647

	if (data[IFLA_GRE_OFLAGS])
T
Tom Herbert 已提交
1648 1649
		parms->o_flags = gre_flags_to_tnl_flags(
				nla_get_be16(data[IFLA_GRE_OFLAGS]));
X
xeb@mail.ru 已提交
1650 1651 1652 1653 1654 1655 1656 1657

	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])
1658
		parms->laddr = nla_get_in6_addr(data[IFLA_GRE_LOCAL]);
X
xeb@mail.ru 已提交
1659 1660

	if (data[IFLA_GRE_REMOTE])
1661
		parms->raddr = nla_get_in6_addr(data[IFLA_GRE_REMOTE]);
X
xeb@mail.ru 已提交
1662 1663 1664 1665 1666 1667 1668 1669

	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])
1670
		parms->flowinfo = nla_get_be32(data[IFLA_GRE_FLOWINFO]);
X
xeb@mail.ru 已提交
1671 1672 1673

	if (data[IFLA_GRE_FLAGS])
		parms->flags = nla_get_u32(data[IFLA_GRE_FLAGS]);
1674 1675 1676

	if (data[IFLA_GRE_FWMARK])
		parms->fwmark = nla_get_u32(data[IFLA_GRE_FWMARK]);
1677

1678 1679
	if (data[IFLA_GRE_COLLECT_METADATA])
		parms->collect_md = true;
W
William Tu 已提交
1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692

	if (data[IFLA_GRE_ERSPAN_VER])
		parms->erspan_ver = nla_get_u8(data[IFLA_GRE_ERSPAN_VER]);

	if (parms->erspan_ver == 1) {
		if (data[IFLA_GRE_ERSPAN_INDEX])
			parms->index = nla_get_u32(data[IFLA_GRE_ERSPAN_INDEX]);
	} else if (parms->erspan_ver == 2) {
		if (data[IFLA_GRE_ERSPAN_DIR])
			parms->dir = nla_get_u8(data[IFLA_GRE_ERSPAN_DIR]);
		if (data[IFLA_GRE_ERSPAN_HWID])
			parms->hwid = nla_get_u16(data[IFLA_GRE_ERSPAN_HWID]);
	}
X
xeb@mail.ru 已提交
1693 1694 1695 1696
}

static int ip6gre_tap_init(struct net_device *dev)
{
1697
	int ret;
X
xeb@mail.ru 已提交
1698

1699 1700 1701
	ret = ip6gre_tunnel_init_common(dev);
	if (ret)
		return ret;
X
xeb@mail.ru 已提交
1702

1703 1704
	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;

X
xeb@mail.ru 已提交
1705 1706 1707 1708 1709 1710 1711 1712 1713
	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,
1714
	.ndo_change_mtu = ip6_tnl_change_mtu,
1715
	.ndo_get_stats64 = ip_tunnel_get_stats64,
1716
	.ndo_get_iflink = ip6_tnl_get_iflink,
X
xeb@mail.ru 已提交
1717 1718
};

1719 1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743
static int ip6erspan_tap_init(struct net_device *dev)
{
	struct ip6_tnl *tunnel;
	int t_hlen;
	int ret;

	tunnel = netdev_priv(dev);

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

	dev->tstats = netdev_alloc_pcpu_stats(struct pcpu_sw_netstats);
	if (!dev->tstats)
		return -ENOMEM;

	ret = dst_cache_init(&tunnel->dst_cache, GFP_KERNEL);
	if (ret) {
		free_percpu(dev->tstats);
		dev->tstats = NULL;
		return ret;
	}

	tunnel->tun_hlen = 8;
	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen +
W
William Tu 已提交
1744
		       erspan_hdr_len(tunnel->parms.erspan_ver);
1745 1746 1747 1748 1749 1750 1751 1752 1753 1754 1755 1756 1757 1758 1759 1760 1761 1762 1763 1764 1765 1766 1767 1768 1769 1770 1771
	t_hlen = tunnel->hlen + sizeof(struct ipv6hdr);

	dev->hard_header_len = LL_MAX_HEADER + t_hlen;
	dev->mtu = ETH_DATA_LEN - t_hlen;
	if (dev->type == ARPHRD_ETHER)
		dev->mtu -= ETH_HLEN;
	if (!(tunnel->parms.flags & IP6_TNL_F_IGN_ENCAP_LIMIT))
		dev->mtu -= 8;

	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
	tunnel = netdev_priv(dev);
	ip6gre_tnl_link_config(tunnel, 1);

	return 0;
}

static const struct net_device_ops ip6erspan_netdev_ops = {
	.ndo_init =		ip6erspan_tap_init,
	.ndo_uninit =		ip6gre_tunnel_uninit,
	.ndo_start_xmit =	ip6erspan_tunnel_xmit,
	.ndo_set_mac_address =	eth_mac_addr,
	.ndo_validate_addr =	eth_validate_addr,
	.ndo_change_mtu =	ip6_tnl_change_mtu,
	.ndo_get_stats64 =	ip_tunnel_get_stats64,
	.ndo_get_iflink =	ip6_tnl_get_iflink,
};

X
xeb@mail.ru 已提交
1772 1773 1774 1775 1776
static void ip6gre_tap_setup(struct net_device *dev)
{

	ether_setup(dev);

1777
	dev->max_mtu = 0;
X
xeb@mail.ru 已提交
1778
	dev->netdev_ops = &ip6gre_tap_netdev_ops;
1779 1780
	dev->needs_free_netdev = true;
	dev->priv_destructor = ip6gre_dev_free;
X
xeb@mail.ru 已提交
1781 1782

	dev->features |= NETIF_F_NETNS_LOCAL;
J
Jiri Benc 已提交
1783
	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
1784
	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
1785
	netif_keep_dst(dev);
X
xeb@mail.ru 已提交
1786 1787
}

1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800 1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814 1815 1816 1817 1818 1819 1820
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
1821
static int ip6gre_newlink(struct net *src_net, struct net_device *dev,
1822 1823
			  struct nlattr *tb[], struct nlattr *data[],
			  struct netlink_ext_ack *extack)
X
xeb@mail.ru 已提交
1824 1825 1826 1827
{
	struct ip6_tnl *nt;
	struct net *net = dev_net(dev);
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
1828
	struct ip_tunnel_encap ipencap;
X
xeb@mail.ru 已提交
1829 1830 1831
	int err;

	nt = netdev_priv(dev);
1832 1833 1834 1835 1836 1837 1838 1839

	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
1840 1841
	ip6gre_netlink_parms(data, &nt->parms);

1842 1843 1844 1845 1846 1847 1848
	if (nt->parms.collect_md) {
		if (rtnl_dereference(ign->collect_md_tun))
			return -EEXIST;
	} else {
		if (ip6gre_tunnel_find(net, &nt->parms, dev->type))
			return -EEXIST;
	}
X
xeb@mail.ru 已提交
1849 1850 1851 1852 1853

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

	nt->dev = dev;
N
Nicolas Dichtel 已提交
1854
	nt->net = dev_net(dev);
X
xeb@mail.ru 已提交
1855 1856 1857 1858 1859

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

1860 1861 1862 1863 1864
	ip6gre_tnl_link_config(nt, !tb[IFLA_MTU]);

	if (tb[IFLA_MTU])
		ip6_tnl_change_mtu(dev, nla_get_u32(tb[IFLA_MTU]));

X
xeb@mail.ru 已提交
1865 1866 1867 1868 1869 1870 1871 1872
	dev_hold(dev);
	ip6gre_tunnel_link(ign, nt);

out:
	return err;
}

static int ip6gre_changelink(struct net_device *dev, struct nlattr *tb[],
1873 1874
			     struct nlattr *data[],
			     struct netlink_ext_ack *extack)
X
xeb@mail.ru 已提交
1875
{
N
Nicolas Dichtel 已提交
1876 1877
	struct ip6_tnl *t, *nt = netdev_priv(dev);
	struct net *net = nt->net;
X
xeb@mail.ru 已提交
1878 1879
	struct ip6gre_net *ign = net_generic(net, ip6gre_net_id);
	struct __ip6_tnl_parm p;
1880
	struct ip_tunnel_encap ipencap;
X
xeb@mail.ru 已提交
1881 1882 1883 1884

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

1885 1886 1887 1888 1889 1890 1891
	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
1892 1893 1894 1895 1896 1897 1898 1899 1900 1901 1902
	ip6gre_netlink_parms(data, &p);

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

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

1903 1904 1905
	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
1906 1907 1908
	return 0;
}

1909 1910 1911 1912 1913 1914 1915 1916 1917
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
1918 1919 1920 1921 1922 1923 1924 1925 1926 1927 1928 1929 1930 1931
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
1932
		nla_total_size(sizeof(struct in6_addr)) +
X
xeb@mail.ru 已提交
1933
		/* IFLA_GRE_REMOTE */
N
Nicolas Dichtel 已提交
1934
		nla_total_size(sizeof(struct in6_addr)) +
X
xeb@mail.ru 已提交
1935 1936 1937 1938 1939 1940 1941 1942
		/* 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) +
1943 1944 1945 1946 1947 1948 1949 1950
		/* 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) +
1951 1952
		/* IFLA_GRE_COLLECT_METADATA */
		nla_total_size(0) +
1953 1954
		/* IFLA_GRE_FWMARK */
		nla_total_size(4) +
1955 1956
		/* IFLA_GRE_ERSPAN_INDEX */
		nla_total_size(4) +
X
xeb@mail.ru 已提交
1957 1958 1959 1960 1961 1962 1963 1964 1965
		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
1966 1967 1968 1969
	    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
1970 1971
	    nla_put_be32(skb, IFLA_GRE_IKEY, p->i_key) ||
	    nla_put_be32(skb, IFLA_GRE_OKEY, p->o_key) ||
1972 1973
	    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
1974 1975 1976
	    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) ||
1977
	    nla_put_u32(skb, IFLA_GRE_FLAGS, p->flags) ||
1978 1979
	    nla_put_u32(skb, IFLA_GRE_FWMARK, p->fwmark) ||
	    nla_put_u32(skb, IFLA_GRE_ERSPAN_INDEX, p->index))
X
xeb@mail.ru 已提交
1980
		goto nla_put_failure;
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991

	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;

1992 1993 1994 1995 1996
	if (p->collect_md) {
		if (nla_put_flag(skb, IFLA_GRE_COLLECT_METADATA))
			goto nla_put_failure;
	}

W
William Tu 已提交
1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
	if (nla_put_u8(skb, IFLA_GRE_ERSPAN_VER, p->erspan_ver))
		goto nla_put_failure;

	if (p->erspan_ver == 1) {
		if (nla_put_u32(skb, IFLA_GRE_ERSPAN_INDEX, p->index))
			goto nla_put_failure;
	} else if (p->erspan_ver == 2) {
		if (nla_put_u8(skb, IFLA_GRE_ERSPAN_DIR, p->dir))
			goto nla_put_failure;
		if (nla_put_u16(skb, IFLA_GRE_ERSPAN_HWID, p->hwid))
			goto nla_put_failure;
	}

X
xeb@mail.ru 已提交
2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027
	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 },
2028 2029 2030 2031
	[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 },
2032
	[IFLA_GRE_COLLECT_METADATA] = { .type = NLA_FLAG },
2033
	[IFLA_GRE_FWMARK]       = { .type = NLA_U32 },
2034
	[IFLA_GRE_ERSPAN_INDEX] = { .type = NLA_U32 },
W
William Tu 已提交
2035 2036 2037
	[IFLA_GRE_ERSPAN_VER]	= { .type = NLA_U8 },
	[IFLA_GRE_ERSPAN_DIR]	= { .type = NLA_U8 },
	[IFLA_GRE_ERSPAN_HWID]	= { .type = NLA_U16 },
X
xeb@mail.ru 已提交
2038 2039
};

2040 2041 2042 2043 2044 2045 2046 2047 2048 2049 2050 2051 2052 2053
static void ip6erspan_tap_setup(struct net_device *dev)
{
	ether_setup(dev);

	dev->netdev_ops = &ip6erspan_netdev_ops;
	dev->needs_free_netdev = true;
	dev->priv_destructor = ip6gre_dev_free;

	dev->features |= NETIF_F_NETNS_LOCAL;
	dev->priv_flags &= ~IFF_TX_SKB_SHARING;
	dev->priv_flags |= IFF_LIVE_ADDR_CHANGE;
	netif_keep_dst(dev);
}

X
xeb@mail.ru 已提交
2054 2055 2056 2057 2058 2059 2060 2061 2062
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,
2063
	.dellink	= ip6gre_dellink,
X
xeb@mail.ru 已提交
2064 2065
	.get_size	= ip6gre_get_size,
	.fill_info	= ip6gre_fill_info,
2066
	.get_link_net	= ip6_tnl_get_link_net,
X
xeb@mail.ru 已提交
2067 2068 2069 2070 2071 2072 2073 2074 2075 2076 2077 2078 2079
};

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,
2080
	.get_link_net	= ip6_tnl_get_link_net,
X
xeb@mail.ru 已提交
2081 2082
};

2083 2084 2085 2086 2087 2088 2089 2090 2091 2092 2093 2094 2095 2096
static struct rtnl_link_ops ip6erspan_tap_ops __read_mostly = {
	.kind		= "ip6erspan",
	.maxtype	= IFLA_GRE_MAX,
	.policy		= ip6gre_policy,
	.priv_size	= sizeof(struct ip6_tnl),
	.setup		= ip6erspan_tap_setup,
	.validate	= ip6erspan_tap_validate,
	.newlink	= ip6gre_newlink,
	.changelink	= ip6gre_changelink,
	.get_size	= ip6gre_get_size,
	.fill_info	= ip6gre_fill_info,
	.get_link_net	= ip6_tnl_get_link_net,
};

X
xeb@mail.ru 已提交
2097 2098 2099 2100 2101 2102 2103 2104 2105 2106 2107 2108 2109 2110 2111 2112 2113 2114 2115 2116 2117 2118 2119 2120 2121 2122 2123 2124
/*
 *	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;

2125 2126 2127 2128
	err = rtnl_link_register(&ip6erspan_tap_ops);
	if (err < 0)
		goto erspan_link_failed;

X
xeb@mail.ru 已提交
2129 2130 2131
out:
	return err;

2132 2133
erspan_link_failed:
	rtnl_link_unregister(&ip6gre_tap_ops);
X
xeb@mail.ru 已提交
2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146
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);
2147
	rtnl_link_unregister(&ip6erspan_tap_ops);
X
xeb@mail.ru 已提交
2148 2149 2150 2151 2152 2153 2154 2155 2156 2157
	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");
2158
MODULE_ALIAS_RTNL_LINK("ip6gretap");
W
William Tu 已提交
2159
MODULE_ALIAS_RTNL_LINK("ip6erspan");
X
xeb@mail.ru 已提交
2160
MODULE_ALIAS_NETDEV("ip6gre0");