br_multicast.c 41.6 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
/*
 * Bridge multicast support.
 *
 * Copyright (c) 2010 Herbert Xu <herbert@gondor.apana.org.au>
 *
 * 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.
 *
 */

#include <linux/err.h>
#include <linux/if_ether.h>
#include <linux/igmp.h>
#include <linux/jhash.h>
#include <linux/kernel.h>
18
#include <linux/log2.h>
19 20 21 22 23 24 25
#include <linux/netdevice.h>
#include <linux/netfilter_bridge.h>
#include <linux/random.h>
#include <linux/rculist.h>
#include <linux/skbuff.h>
#include <linux/slab.h>
#include <linux/timer.h>
26
#include <linux/inetdevice.h>
27
#include <net/ip.h>
E
Eric Dumazet 已提交
28
#if IS_ENABLED(CONFIG_IPV6)
29 30
#include <net/ipv6.h>
#include <net/mld.h>
31
#include <net/ip6_checksum.h>
32
#endif
33 34 35

#include "br_private.h"

36
static void br_multicast_start_querier(struct net_bridge *br);
C
Cong Wang 已提交
37
unsigned int br_mdb_rehash_seq;
38

39 40 41 42
static inline int br_ip_equal(const struct br_ip *a, const struct br_ip *b)
{
	if (a->proto != b->proto)
		return 0;
43 44
	if (a->vid != b->vid)
		return 0;
45 46 47
	switch (a->proto) {
	case htons(ETH_P_IP):
		return a->u.ip4 == b->u.ip4;
E
Eric Dumazet 已提交
48
#if IS_ENABLED(CONFIG_IPV6)
49 50 51
	case htons(ETH_P_IPV6):
		return ipv6_addr_equal(&a->u.ip6, &b->u.ip6);
#endif
52 53 54 55
	}
	return 0;
}

56 57
static inline int __br_ip4_hash(struct net_bridge_mdb_htable *mdb, __be32 ip,
				__u16 vid)
58
{
59
	return jhash_2words((__force u32)ip, vid, mdb->secret) & (mdb->max - 1);
60 61
}

E
Eric Dumazet 已提交
62
#if IS_ENABLED(CONFIG_IPV6)
63
static inline int __br_ip6_hash(struct net_bridge_mdb_htable *mdb,
64 65
				const struct in6_addr *ip,
				__u16 vid)
66
{
67 68
	return jhash_2words(ipv6_addr_hash(ip), vid,
			    mdb->secret) & (mdb->max - 1);
69 70 71
}
#endif

72 73 74 75 76
static inline int br_ip_hash(struct net_bridge_mdb_htable *mdb,
			     struct br_ip *ip)
{
	switch (ip->proto) {
	case htons(ETH_P_IP):
77
		return __br_ip4_hash(mdb, ip->u.ip4, ip->vid);
E
Eric Dumazet 已提交
78
#if IS_ENABLED(CONFIG_IPV6)
79
	case htons(ETH_P_IPV6):
80
		return __br_ip6_hash(mdb, &ip->u.ip6, ip->vid);
81
#endif
82 83
	}
	return 0;
84 85 86
}

static struct net_bridge_mdb_entry *__br_mdb_ip_get(
87
	struct net_bridge_mdb_htable *mdb, struct br_ip *dst, int hash)
88 89 90
{
	struct net_bridge_mdb_entry *mp;

91
	hlist_for_each_entry_rcu(mp, &mdb->mhash[hash], hlist[mdb->ver]) {
92
		if (br_ip_equal(&mp->addr, dst))
93 94 95 96 97 98
			return mp;
	}

	return NULL;
}

99 100
struct net_bridge_mdb_entry *br_mdb_ip_get(struct net_bridge_mdb_htable *mdb,
					   struct br_ip *dst)
101 102 103 104 105 106 107
{
	if (!mdb)
		return NULL;

	return __br_mdb_ip_get(mdb, dst, br_ip_hash(mdb, dst));
}

108
static struct net_bridge_mdb_entry *br_mdb_ip4_get(
109
	struct net_bridge_mdb_htable *mdb, __be32 dst, __u16 vid)
110
{
111 112 113 114
	struct br_ip br_dst;

	br_dst.u.ip4 = dst;
	br_dst.proto = htons(ETH_P_IP);
115
	br_dst.vid = vid;
116

117
	return br_mdb_ip_get(mdb, &br_dst);
118 119
}

E
Eric Dumazet 已提交
120
#if IS_ENABLED(CONFIG_IPV6)
121
static struct net_bridge_mdb_entry *br_mdb_ip6_get(
122 123
	struct net_bridge_mdb_htable *mdb, const struct in6_addr *dst,
	__u16 vid)
124 125
{
	struct br_ip br_dst;
126

A
Alexey Dobriyan 已提交
127
	br_dst.u.ip6 = *dst;
128
	br_dst.proto = htons(ETH_P_IPV6);
129
	br_dst.vid = vid;
130

131
	return br_mdb_ip_get(mdb, &br_dst);
132 133 134
}
#endif

135
struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br,
136
					struct sk_buff *skb, u16 vid)
137
{
138
	struct net_bridge_mdb_htable *mdb = rcu_dereference(br->mdb);
139 140
	struct br_ip ip;

141
	if (br->multicast_disabled)
142 143
		return NULL;

144
	if (BR_INPUT_SKB_CB(skb)->igmp)
145 146
		return NULL;

147
	ip.proto = skb->protocol;
148
	ip.vid = vid;
149

150 151
	switch (skb->protocol) {
	case htons(ETH_P_IP):
152 153
		ip.u.ip4 = ip_hdr(skb)->daddr;
		break;
E
Eric Dumazet 已提交
154
#if IS_ENABLED(CONFIG_IPV6)
155
	case htons(ETH_P_IPV6):
A
Alexey Dobriyan 已提交
156
		ip.u.ip6 = ipv6_hdr(skb)->daddr;
157 158
		break;
#endif
159 160
	default:
		return NULL;
161 162
	}

163
	return br_mdb_ip_get(mdb, &ip);
164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186
}

static void br_mdb_free(struct rcu_head *head)
{
	struct net_bridge_mdb_htable *mdb =
		container_of(head, struct net_bridge_mdb_htable, rcu);
	struct net_bridge_mdb_htable *old = mdb->old;

	mdb->old = NULL;
	kfree(old->mhash);
	kfree(old);
}

static int br_mdb_copy(struct net_bridge_mdb_htable *new,
		       struct net_bridge_mdb_htable *old,
		       int elasticity)
{
	struct net_bridge_mdb_entry *mp;
	int maxlen;
	int len;
	int i;

	for (i = 0; i < old->max; i++)
187
		hlist_for_each_entry(mp, &old->mhash[i], hlist[old->ver])
188
			hlist_add_head(&mp->hlist[new->ver],
189
				       &new->mhash[br_ip_hash(new, &mp->addr)]);
190 191 192 193 194 195 196

	if (!elasticity)
		return 0;

	maxlen = 0;
	for (i = 0; i < new->max; i++) {
		len = 0;
197
		hlist_for_each_entry(mp, &new->mhash[i], hlist[new->ver])
198 199 200 201 202 203 204 205
			len++;
		if (len > maxlen)
			maxlen = len;
	}

	return maxlen > elasticity ? -EINVAL : 0;
}

206
void br_multicast_free_pg(struct rcu_head *head)
207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
{
	struct net_bridge_port_group *p =
		container_of(head, struct net_bridge_port_group, rcu);

	kfree(p);
}

static void br_multicast_free_group(struct rcu_head *head)
{
	struct net_bridge_mdb_entry *mp =
		container_of(head, struct net_bridge_mdb_entry, rcu);

	kfree(mp);
}

static void br_multicast_group_expired(unsigned long data)
{
	struct net_bridge_mdb_entry *mp = (void *)data;
	struct net_bridge *br = mp->br;
	struct net_bridge_mdb_htable *mdb;

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) || timer_pending(&mp->timer))
		goto out;

232
	mp->mglist = false;
233 234 235 236

	if (mp->ports)
		goto out;

237 238
	mdb = mlock_dereference(br->mdb, br);

239 240 241 242 243 244 245 246 247 248 249 250
	hlist_del_rcu(&mp->hlist[mdb->ver]);
	mdb->size--;

	call_rcu_bh(&mp->rcu, br_multicast_free_group);

out:
	spin_unlock(&br->multicast_lock);
}

static void br_multicast_del_pg(struct net_bridge *br,
				struct net_bridge_port_group *pg)
{
251
	struct net_bridge_mdb_htable *mdb;
252 253
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
254 255 256
	struct net_bridge_port_group __rcu **pp;

	mdb = mlock_dereference(br->mdb, br);
257

258
	mp = br_mdb_ip_get(mdb, &pg->addr);
259 260 261
	if (WARN_ON(!mp))
		return;

262 263 264
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
265 266 267
		if (p != pg)
			continue;

268
		rcu_assign_pointer(*pp, p->next);
269 270 271 272
		hlist_del_init(&p->mglist);
		del_timer(&p->timer);
		call_rcu_bh(&p->rcu, br_multicast_free_pg);

273
		if (!mp->ports && !mp->mglist && mp->timer_armed &&
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289
		    netif_running(br->dev))
			mod_timer(&mp->timer, jiffies);

		return;
	}

	WARN_ON(1);
}

static void br_multicast_port_group_expired(unsigned long data)
{
	struct net_bridge_port_group *pg = (void *)data;
	struct net_bridge *br = pg->port->br;

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) || timer_pending(&pg->timer) ||
290
	    hlist_unhashed(&pg->mglist) || pg->state & MDB_PERMANENT)
291 292 293 294 295 296 297 298
		goto out;

	br_multicast_del_pg(br, pg);

out:
	spin_unlock(&br->multicast_lock);
}

299
static int br_mdb_rehash(struct net_bridge_mdb_htable __rcu **mdbp, int max,
300 301
			 int elasticity)
{
302
	struct net_bridge_mdb_htable *old = rcu_dereference_protected(*mdbp, 1);
303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336
	struct net_bridge_mdb_htable *mdb;
	int err;

	mdb = kmalloc(sizeof(*mdb), GFP_ATOMIC);
	if (!mdb)
		return -ENOMEM;

	mdb->max = max;
	mdb->old = old;

	mdb->mhash = kzalloc(max * sizeof(*mdb->mhash), GFP_ATOMIC);
	if (!mdb->mhash) {
		kfree(mdb);
		return -ENOMEM;
	}

	mdb->size = old ? old->size : 0;
	mdb->ver = old ? old->ver ^ 1 : 0;

	if (!old || elasticity)
		get_random_bytes(&mdb->secret, sizeof(mdb->secret));
	else
		mdb->secret = old->secret;

	if (!old)
		goto out;

	err = br_mdb_copy(mdb, old, elasticity);
	if (err) {
		kfree(mdb->mhash);
		kfree(mdb);
		return err;
	}

C
Cong Wang 已提交
337
	br_mdb_rehash_seq++;
338 339 340 341 342 343 344 345
	call_rcu_bh(&mdb->rcu, br_mdb_free);

out:
	rcu_assign_pointer(*mdbp, mdb);

	return 0;
}

346 347
static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br,
						    __be32 group)
348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384
{
	struct sk_buff *skb;
	struct igmphdr *ih;
	struct ethhdr *eth;
	struct iphdr *iph;

	skb = netdev_alloc_skb_ip_align(br->dev, sizeof(*eth) + sizeof(*iph) +
						 sizeof(*ih) + 4);
	if (!skb)
		goto out;

	skb->protocol = htons(ETH_P_IP);

	skb_reset_mac_header(skb);
	eth = eth_hdr(skb);

	memcpy(eth->h_source, br->dev->dev_addr, 6);
	eth->h_dest[0] = 1;
	eth->h_dest[1] = 0;
	eth->h_dest[2] = 0x5e;
	eth->h_dest[3] = 0;
	eth->h_dest[4] = 0;
	eth->h_dest[5] = 1;
	eth->h_proto = htons(ETH_P_IP);
	skb_put(skb, sizeof(*eth));

	skb_set_network_header(skb, skb->len);
	iph = ip_hdr(skb);

	iph->version = 4;
	iph->ihl = 6;
	iph->tos = 0xc0;
	iph->tot_len = htons(sizeof(*iph) + sizeof(*ih) + 4);
	iph->id = 0;
	iph->frag_off = htons(IP_DF);
	iph->ttl = 1;
	iph->protocol = IPPROTO_IGMP;
385 386
	iph->saddr = br->multicast_query_use_ifaddr ?
		     inet_select_addr(br->dev, 0, RT_SCOPE_LINK) : 0;
387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411
	iph->daddr = htonl(INADDR_ALLHOSTS_GROUP);
	((u8 *)&iph[1])[0] = IPOPT_RA;
	((u8 *)&iph[1])[1] = 4;
	((u8 *)&iph[1])[2] = 0;
	((u8 *)&iph[1])[3] = 0;
	ip_send_check(iph);
	skb_put(skb, 24);

	skb_set_transport_header(skb, skb->len);
	ih = igmp_hdr(skb);
	ih->type = IGMP_HOST_MEMBERSHIP_QUERY;
	ih->code = (group ? br->multicast_last_member_interval :
			    br->multicast_query_response_interval) /
		   (HZ / IGMP_TIMER_SCALE);
	ih->group = group;
	ih->csum = 0;
	ih->csum = ip_compute_csum((void *)ih, sizeof(struct igmphdr));
	skb_put(skb, sizeof(*ih));

	__skb_pull(skb, sizeof(*eth));

out:
	return skb;
}

E
Eric Dumazet 已提交
412
#if IS_ENABLED(CONFIG_IPV6)
413
static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br,
414
						    const struct in6_addr *group)
415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442
{
	struct sk_buff *skb;
	struct ipv6hdr *ip6h;
	struct mld_msg *mldq;
	struct ethhdr *eth;
	u8 *hopopt;
	unsigned long interval;

	skb = netdev_alloc_skb_ip_align(br->dev, sizeof(*eth) + sizeof(*ip6h) +
						 8 + sizeof(*mldq));
	if (!skb)
		goto out;

	skb->protocol = htons(ETH_P_IPV6);

	/* Ethernet header */
	skb_reset_mac_header(skb);
	eth = eth_hdr(skb);

	memcpy(eth->h_source, br->dev->dev_addr, 6);
	eth->h_proto = htons(ETH_P_IPV6);
	skb_put(skb, sizeof(*eth));

	/* IPv6 header + HbH option */
	skb_set_network_header(skb, skb->len);
	ip6h = ipv6_hdr(skb);

	*(__force __be32 *)ip6h = htonl(0x60000000);
443
	ip6h->payload_len = htons(8 + sizeof(*mldq));
444 445
	ip6h->nexthdr = IPPROTO_HOPOPTS;
	ip6h->hop_limit = 1;
446
	ipv6_addr_set(&ip6h->daddr, htonl(0xff020000), 0, 0, htonl(1));
447 448 449 450 451
	if (ipv6_dev_get_saddr(dev_net(br->dev), br->dev, &ip6h->daddr, 0,
			       &ip6h->saddr)) {
		kfree_skb(skb);
		return NULL;
	}
452
	ipv6_eth_mc_map(&ip6h->daddr, eth->h_dest);
453 454 455 456 457 458 459 460

	hopopt = (u8 *)(ip6h + 1);
	hopopt[0] = IPPROTO_ICMPV6;		/* next hdr */
	hopopt[1] = 0;				/* length of HbH */
	hopopt[2] = IPV6_TLV_ROUTERALERT;	/* Router Alert */
	hopopt[3] = 2;				/* Length of RA Option */
	hopopt[4] = 0;				/* Type = 0x0000 (MLD) */
	hopopt[5] = 0;
461 462
	hopopt[6] = IPV6_TLV_PAD1;		/* Pad1 */
	hopopt[7] = IPV6_TLV_PAD1;		/* Pad1 */
463 464 465 466 467 468 469

	skb_put(skb, sizeof(*ip6h) + 8);

	/* ICMPv6 */
	skb_set_transport_header(skb, skb->len);
	mldq = (struct mld_msg *) icmp6_hdr(skb);

470 471 472
	interval = ipv6_addr_any(group) ?
			br->multicast_query_response_interval :
			br->multicast_last_member_interval;
473 474 475 476 477 478

	mldq->mld_type = ICMPV6_MGM_QUERY;
	mldq->mld_code = 0;
	mldq->mld_cksum = 0;
	mldq->mld_maxdelay = htons((u16)jiffies_to_msecs(interval));
	mldq->mld_reserved = 0;
A
Alexey Dobriyan 已提交
479
	mldq->mld_mca = *group;
480 481 482 483 484 485 486 487 488 489 490 491 492 493 494

	/* checksum */
	mldq->mld_cksum = csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr,
					  sizeof(*mldq), IPPROTO_ICMPV6,
					  csum_partial(mldq,
						       sizeof(*mldq), 0));
	skb_put(skb, sizeof(*mldq));

	__skb_pull(skb, sizeof(*eth));

out:
	return skb;
}
#endif

495 496 497 498 499 500
static struct sk_buff *br_multicast_alloc_query(struct net_bridge *br,
						struct br_ip *addr)
{
	switch (addr->proto) {
	case htons(ETH_P_IP):
		return br_ip4_multicast_alloc_query(br, addr->u.ip4);
E
Eric Dumazet 已提交
501
#if IS_ENABLED(CONFIG_IPV6)
502 503 504
	case htons(ETH_P_IPV6):
		return br_ip6_multicast_alloc_query(br, &addr->u.ip6);
#endif
505 506 507 508
	}
	return NULL;
}

509
static struct net_bridge_mdb_entry *br_multicast_get_group(
510 511
	struct net_bridge *br, struct net_bridge_port *port,
	struct br_ip *group, int hash)
512
{
513
	struct net_bridge_mdb_htable *mdb;
514
	struct net_bridge_mdb_entry *mp;
515 516
	unsigned int count = 0;
	unsigned int max;
517 518 519
	int elasticity;
	int err;

520
	mdb = rcu_dereference_protected(br->mdb, 1);
521
	hlist_for_each_entry(mp, &mdb->mhash[hash], hlist[mdb->ver]) {
522
		count++;
523
		if (unlikely(br_ip_equal(group, &mp->addr)))
524 525 526 527 528 529 530 531
			return mp;
	}

	elasticity = 0;
	max = mdb->max;

	if (unlikely(count > br->hash_elasticity && count)) {
		if (net_ratelimit())
532 533 534
			br_info(br, "Multicast hash table "
				"chain limit reached: %s\n",
				port ? port->dev->name : br->dev->name);
535 536 537 538 539 540

		elasticity = br->hash_elasticity;
	}

	if (mdb->size >= max) {
		max *= 2;
541 542 543 544 545
		if (unlikely(max > br->hash_max)) {
			br_warn(br, "Multicast hash table maximum of %d "
				"reached, disabling snooping: %s\n",
				br->hash_max,
				port ? port->dev->name : br->dev->name);
546 547 548 549 550 551 552 553 554 555
			err = -E2BIG;
disable:
			br->multicast_disabled = 1;
			goto err;
		}
	}

	if (max > mdb->max || elasticity) {
		if (mdb->old) {
			if (net_ratelimit())
556 557 558
				br_info(br, "Multicast hash table "
					"on fire: %s\n",
					port ? port->dev->name : br->dev->name);
559 560 561 562 563 564
			err = -EEXIST;
			goto err;
		}

		err = br_mdb_rehash(&br->mdb, max, elasticity);
		if (err) {
565 566 567 568
			br_warn(br, "Cannot rehash multicast "
				"hash table, disabling snooping: %s, %d, %d\n",
				port ? port->dev->name : br->dev->name,
				mdb->size, err);
569 570 571 572 573 574 575 576 577 578 579 580 581 582
			goto disable;
		}

		err = -EAGAIN;
		goto err;
	}

	return NULL;

err:
	mp = ERR_PTR(err);
	return mp;
}

583 584
struct net_bridge_mdb_entry *br_multicast_new_group(struct net_bridge *br,
	struct net_bridge_port *port, struct br_ip *group)
585
{
586
	struct net_bridge_mdb_htable *mdb;
587 588
	struct net_bridge_mdb_entry *mp;
	int hash;
589
	int err;
590

591
	mdb = rcu_dereference_protected(br->mdb, 1);
592
	if (!mdb) {
593 594 595
		err = br_mdb_rehash(&br->mdb, BR_HASH_SIZE, 0);
		if (err)
			return ERR_PTR(err);
596 597 598 599 600 601 602 603 604 605 606
		goto rehash;
	}

	hash = br_ip_hash(mdb, group);
	mp = br_multicast_get_group(br, port, group, hash);
	switch (PTR_ERR(mp)) {
	case 0:
		break;

	case -EAGAIN:
rehash:
607
		mdb = rcu_dereference_protected(br->mdb, 1);
608 609 610 611 612 613 614 615 616
		hash = br_ip_hash(mdb, group);
		break;

	default:
		goto out;
	}

	mp = kzalloc(sizeof(*mp), GFP_ATOMIC);
	if (unlikely(!mp))
617
		return ERR_PTR(-ENOMEM);
618 619

	mp->br = br;
620
	mp->addr = *group;
621

622 623 624
	setup_timer(&mp->timer, br_multicast_group_expired,
		    (unsigned long)mp);

625 626 627 628 629 630 631
	hlist_add_head_rcu(&mp->hlist[mdb->ver], &mdb->mhash[hash]);
	mdb->size++;

out:
	return mp;
}

632 633 634
struct net_bridge_port_group *br_multicast_new_port_group(
			struct net_bridge_port *port,
			struct br_ip *group,
635 636
			struct net_bridge_port_group __rcu *next,
			unsigned char state)
637 638 639 640 641 642 643 644 645
{
	struct net_bridge_port_group *p;

	p = kzalloc(sizeof(*p), GFP_ATOMIC);
	if (unlikely(!p))
		return NULL;

	p->addr = *group;
	p->port = port;
646
	p->state = state;
647
	rcu_assign_pointer(p->next, next);
648 649 650 651 652 653
	hlist_add_head(&p->mglist, &port->mglist);
	setup_timer(&p->timer, br_multicast_port_group_expired,
		    (unsigned long)p);
	return p;
}

654
static int br_multicast_add_group(struct net_bridge *br,
655 656
				  struct net_bridge_port *port,
				  struct br_ip *group)
657 658 659
{
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
660
	struct net_bridge_port_group __rcu **pp;
661 662 663 664 665 666 667 668 669
	int err;

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) ||
	    (port && port->state == BR_STATE_DISABLED))
		goto out;

	mp = br_multicast_new_group(br, port, group);
	err = PTR_ERR(mp);
670
	if (IS_ERR(mp))
671 672 673
		goto err;

	if (!port) {
674
		mp->mglist = true;
675 676 677
		goto out;
	}

678 679 680
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
681
		if (p->port == port)
682
			goto out;
683 684 685 686
		if ((unsigned long)p->port < (unsigned long)port)
			break;
	}

687
	p = br_multicast_new_port_group(port, group, *pp, MDB_TEMPORARY);
688 689 690
	if (unlikely(!p))
		goto err;
	rcu_assign_pointer(*pp, p);
C
Cong Wang 已提交
691
	br_mdb_notify(br->dev, port, group, RTM_NEWMDB);
692 693 694 695 696 697 698 699 700

out:
	err = 0;

err:
	spin_unlock(&br->multicast_lock);
	return err;
}

701 702
static int br_ip4_multicast_add_group(struct net_bridge *br,
				      struct net_bridge_port *port,
703 704
				      __be32 group,
				      __u16 vid)
705 706 707 708 709 710 711 712
{
	struct br_ip br_group;

	if (ipv4_is_local_multicast(group))
		return 0;

	br_group.u.ip4 = group;
	br_group.proto = htons(ETH_P_IP);
713
	br_group.vid = vid;
714 715 716 717

	return br_multicast_add_group(br, port, &br_group);
}

E
Eric Dumazet 已提交
718
#if IS_ENABLED(CONFIG_IPV6)
719 720
static int br_ip6_multicast_add_group(struct net_bridge *br,
				      struct net_bridge_port *port,
721 722
				      const struct in6_addr *group,
				      __u16 vid)
723 724 725
{
	struct br_ip br_group;

726
	if (!ipv6_is_transient_multicast(group))
727 728
		return 0;

A
Alexey Dobriyan 已提交
729
	br_group.u.ip6 = *group;
730
	br_group.proto = htons(ETH_P_IPV6);
731
	br_group.vid = vid;
732 733 734 735 736

	return br_multicast_add_group(br, port, &br_group);
}
#endif

737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757
static void br_multicast_router_expired(unsigned long data)
{
	struct net_bridge_port *port = (void *)data;
	struct net_bridge *br = port->br;

	spin_lock(&br->multicast_lock);
	if (port->multicast_router != 1 ||
	    timer_pending(&port->multicast_router_timer) ||
	    hlist_unhashed(&port->rlist))
		goto out;

	hlist_del_init_rcu(&port->rlist);

out:
	spin_unlock(&br->multicast_lock);
}

static void br_multicast_local_router_expired(unsigned long data)
{
}

758 759
static void br_multicast_querier_expired(unsigned long data)
{
760
	struct net_bridge *br = (void *)data;
761 762 763 764 765 766 767 768 769 770 771

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) || br->multicast_disabled)
		goto out;

	br_multicast_start_querier(br);

out:
	spin_unlock(&br->multicast_lock);
}

772 773 774
static void __br_multicast_send_query(struct net_bridge *br,
				      struct net_bridge_port *port,
				      struct br_ip *ip)
775 776 777
{
	struct sk_buff *skb;

778
	skb = br_multicast_alloc_query(br, ip);
779
	if (!skb)
780
		return;
781 782 783 784

	if (port) {
		__skb_push(skb, sizeof(struct ethhdr));
		skb->dev = port->dev;
785
		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
786 787 788
			dev_queue_xmit);
	} else
		netif_rx(skb);
789 790 791 792 793 794 795 796 797
}

static void br_multicast_send_query(struct net_bridge *br,
				    struct net_bridge_port *port, u32 sent)
{
	unsigned long time;
	struct br_ip br_group;

	if (!netif_running(br->dev) || br->multicast_disabled ||
798
	    !br->multicast_querier ||
799 800 801
	    timer_pending(&br->multicast_querier_timer))
		return;

802 803
	memset(&br_group.u, 0, sizeof(br_group.u));

804
	br_group.proto = htons(ETH_P_IP);
805
	__br_multicast_send_query(br, port, &br_group);
806

E
Eric Dumazet 已提交
807
#if IS_ENABLED(CONFIG_IPV6)
808
	br_group.proto = htons(ETH_P_IPV6);
809
	__br_multicast_send_query(br, port, &br_group);
810
#endif
811 812 813 814 815 816 817 818 819 820 821 822 823 824 825

	time = jiffies;
	time += sent < br->multicast_startup_query_count ?
		br->multicast_startup_query_interval :
		br->multicast_query_interval;
	mod_timer(port ? &port->multicast_query_timer :
			 &br->multicast_query_timer, time);
}

static void br_multicast_port_query_expired(unsigned long data)
{
	struct net_bridge_port *port = (void *)data;
	struct net_bridge *br = port->br;

	spin_lock(&br->multicast_lock);
826 827
	if (port->state == BR_STATE_DISABLED ||
	    port->state == BR_STATE_BLOCKING)
828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855
		goto out;

	if (port->multicast_startup_queries_sent <
	    br->multicast_startup_query_count)
		port->multicast_startup_queries_sent++;

	br_multicast_send_query(port->br, port,
				port->multicast_startup_queries_sent);

out:
	spin_unlock(&br->multicast_lock);
}

void br_multicast_add_port(struct net_bridge_port *port)
{
	port->multicast_router = 1;

	setup_timer(&port->multicast_router_timer, br_multicast_router_expired,
		    (unsigned long)port);
	setup_timer(&port->multicast_query_timer,
		    br_multicast_port_query_expired, (unsigned long)port);
}

void br_multicast_del_port(struct net_bridge_port *port)
{
	del_timer_sync(&port->multicast_router_timer);
}

856 857 858 859 860 861 862 863 864
static void __br_multicast_enable_port(struct net_bridge_port *port)
{
	port->multicast_startup_queries_sent = 0;

	if (try_to_del_timer_sync(&port->multicast_query_timer) >= 0 ||
	    del_timer(&port->multicast_query_timer))
		mod_timer(&port->multicast_query_timer, jiffies);
}

865 866 867 868 869 870 871 872
void br_multicast_enable_port(struct net_bridge_port *port)
{
	struct net_bridge *br = port->br;

	spin_lock(&br->multicast_lock);
	if (br->multicast_disabled || !netif_running(br->dev))
		goto out;

873
	__br_multicast_enable_port(port);
874 875 876 877 878 879 880 881 882

out:
	spin_unlock(&br->multicast_lock);
}

void br_multicast_disable_port(struct net_bridge_port *port)
{
	struct net_bridge *br = port->br;
	struct net_bridge_port_group *pg;
883
	struct hlist_node *n;
884 885

	spin_lock(&br->multicast_lock);
886
	hlist_for_each_entry_safe(pg, n, &port->mglist, mglist)
887 888 889 890 891 892 893 894 895
		br_multicast_del_pg(br, pg);

	if (!hlist_unhashed(&port->rlist))
		hlist_del_init_rcu(&port->rlist);
	del_timer(&port->multicast_router_timer);
	del_timer(&port->multicast_query_timer);
	spin_unlock(&br->multicast_lock);
}

896 897 898
static int br_ip4_multicast_igmp3_report(struct net_bridge *br,
					 struct net_bridge_port *port,
					 struct sk_buff *skb)
899 900 901 902 903 904 905 906 907
{
	struct igmpv3_report *ih;
	struct igmpv3_grec *grec;
	int i;
	int len;
	int num;
	int type;
	int err = 0;
	__be32 group;
908
	u16 vid = 0;
909 910 911 912

	if (!pskb_may_pull(skb, sizeof(*ih)))
		return -EINVAL;

913
	br_vlan_get_tag(skb, &vid);
914 915 916 917 918 919 920 921 922
	ih = igmpv3_report_hdr(skb);
	num = ntohs(ih->ngrec);
	len = sizeof(*ih);

	for (i = 0; i < num; i++) {
		len += sizeof(*grec);
		if (!pskb_may_pull(skb, len))
			return -EINVAL;

H
Herbert Xu 已提交
923
		grec = (void *)(skb->data + len - sizeof(*grec));
924 925 926
		group = grec->grec_mca;
		type = grec->grec_type;

E
Eric Dumazet 已提交
927
		len += ntohs(grec->grec_nsrcs) * 4;
928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944
		if (!pskb_may_pull(skb, len))
			return -EINVAL;

		/* We treat this as an IGMPv2 report for now. */
		switch (type) {
		case IGMPV3_MODE_IS_INCLUDE:
		case IGMPV3_MODE_IS_EXCLUDE:
		case IGMPV3_CHANGE_TO_INCLUDE:
		case IGMPV3_CHANGE_TO_EXCLUDE:
		case IGMPV3_ALLOW_NEW_SOURCES:
		case IGMPV3_BLOCK_OLD_SOURCES:
			break;

		default:
			continue;
		}

945
		err = br_ip4_multicast_add_group(br, port, group, vid);
946 947 948 949 950 951 952
		if (err)
			break;
	}

	return err;
}

E
Eric Dumazet 已提交
953
#if IS_ENABLED(CONFIG_IPV6)
954 955 956 957 958 959 960 961 962 963
static int br_ip6_multicast_mld2_report(struct net_bridge *br,
					struct net_bridge_port *port,
					struct sk_buff *skb)
{
	struct icmp6hdr *icmp6h;
	struct mld2_grec *grec;
	int i;
	int len;
	int num;
	int err = 0;
964
	u16 vid = 0;
965 966 967 968

	if (!pskb_may_pull(skb, sizeof(*icmp6h)))
		return -EINVAL;

969
	br_vlan_get_tag(skb, &vid);
970 971 972 973 974 975 976 977 978
	icmp6h = icmp6_hdr(skb);
	num = ntohs(icmp6h->icmp6_dataun.un_data16[1]);
	len = sizeof(*icmp6h);

	for (i = 0; i < num; i++) {
		__be16 *nsrcs, _nsrcs;

		nsrcs = skb_header_pointer(skb,
					   len + offsetof(struct mld2_grec,
979
							  grec_nsrcs),
980 981 982 983 984 985
					   sizeof(_nsrcs), &_nsrcs);
		if (!nsrcs)
			return -EINVAL;

		if (!pskb_may_pull(skb,
				   len + sizeof(*grec) +
986
				   sizeof(struct in6_addr) * ntohs(*nsrcs)))
987 988 989
			return -EINVAL;

		grec = (struct mld2_grec *)(skb->data + len);
990 991
		len += sizeof(*grec) +
		       sizeof(struct in6_addr) * ntohs(*nsrcs);
992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006

		/* We treat these as MLDv1 reports for now. */
		switch (grec->grec_type) {
		case MLD2_MODE_IS_INCLUDE:
		case MLD2_MODE_IS_EXCLUDE:
		case MLD2_CHANGE_TO_INCLUDE:
		case MLD2_CHANGE_TO_EXCLUDE:
		case MLD2_ALLOW_NEW_SOURCES:
		case MLD2_BLOCK_OLD_SOURCES:
			break;

		default:
			continue;
		}

1007 1008
		err = br_ip6_multicast_add_group(br, port, &grec->grec_mca,
						 vid);
1009 1010 1011 1012 1013 1014 1015 1016
		if (!err)
			break;
	}

	return err;
}
#endif

1017 1018 1019 1020 1021 1022 1023 1024 1025 1026
static void br_multicast_update_querier_timer(struct net_bridge *br,
					      unsigned long max_delay)
{
	if (!timer_pending(&br->multicast_querier_timer))
		br->multicast_querier_delay_time = jiffies + max_delay;

	mod_timer(&br->multicast_querier_timer,
		  jiffies + br->multicast_querier_interval);
}

1027
/*
C
Cong Wang 已提交
1028
 * Add port to router_list
1029 1030 1031
 *  list is maintained ordered by pointer value
 *  and locked by br->multicast_lock and RCU
 */
1032 1033 1034
static void br_multicast_add_router(struct net_bridge *br,
				    struct net_bridge_port *port)
{
1035
	struct net_bridge_port *p;
1036
	struct hlist_node *slot = NULL;
1037

1038
	hlist_for_each_entry(p, &br->router_list, rlist) {
1039 1040
		if ((unsigned long) port >= (unsigned long) p)
			break;
1041
		slot = &p->rlist;
1042 1043
	}

1044 1045
	if (slot)
		hlist_add_after_rcu(slot, &port->rlist);
1046 1047
	else
		hlist_add_head_rcu(&port->rlist, &br->router_list);
1048 1049
}

1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067
static void br_multicast_mark_router(struct net_bridge *br,
				     struct net_bridge_port *port)
{
	unsigned long now = jiffies;

	if (!port) {
		if (br->multicast_router == 1)
			mod_timer(&br->multicast_router_timer,
				  now + br->multicast_querier_interval);
		return;
	}

	if (port->multicast_router != 1)
		return;

	if (!hlist_unhashed(&port->rlist))
		goto timer;

1068
	br_multicast_add_router(br, port);
1069 1070 1071 1072 1073 1074 1075 1076

timer:
	mod_timer(&port->multicast_router_timer,
		  now + br->multicast_querier_interval);
}

static void br_multicast_query_received(struct net_bridge *br,
					struct net_bridge_port *port,
1077 1078
					int saddr,
					unsigned long max_delay)
1079 1080
{
	if (saddr)
1081
		br_multicast_update_querier_timer(br, max_delay);
1082 1083 1084 1085 1086 1087
	else if (timer_pending(&br->multicast_querier_timer))
		return;

	br_multicast_mark_router(br, port);
}

1088 1089 1090
static int br_ip4_multicast_query(struct net_bridge *br,
				  struct net_bridge_port *port,
				  struct sk_buff *skb)
1091
{
1092
	const struct iphdr *iph = ip_hdr(skb);
1093 1094 1095 1096
	struct igmphdr *ih = igmp_hdr(skb);
	struct net_bridge_mdb_entry *mp;
	struct igmpv3_query *ih3;
	struct net_bridge_port_group *p;
1097
	struct net_bridge_port_group __rcu **pp;
1098 1099 1100
	unsigned long max_delay;
	unsigned long now = jiffies;
	__be32 group;
1101
	int err = 0;
1102
	u16 vid = 0;
1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) ||
	    (port && port->state == BR_STATE_DISABLED))
		goto out;

	group = ih->group;

	if (skb->len == sizeof(*ih)) {
		max_delay = ih->code * (HZ / IGMP_TIMER_SCALE);

		if (!max_delay) {
			max_delay = 10 * HZ;
			group = 0;
		}
	} else {
1119 1120 1121 1122
		if (!pskb_may_pull(skb, sizeof(struct igmpv3_query))) {
			err = -EINVAL;
			goto out;
		}
1123 1124 1125

		ih3 = igmpv3_query_hdr(skb);
		if (ih3->nsrcs)
1126
			goto out;
1127

1128 1129
		max_delay = ih3->code ?
			    IGMPV3_MRC(ih3->code) * (HZ / IGMP_TIMER_SCALE) : 1;
1130 1131
	}

1132 1133
	br_multicast_query_received(br, port, !!iph->saddr, max_delay);

1134 1135 1136
	if (!group)
		goto out;

1137 1138
	br_vlan_get_tag(skb, &vid);
	mp = br_mdb_ip4_get(mlock_dereference(br->mdb, br), group, vid);
1139 1140 1141
	if (!mp)
		goto out;

1142 1143 1144
	mod_timer(&mp->timer, now + br->multicast_membership_interval);
	mp->timer_armed = true;

1145 1146
	max_delay *= br->multicast_last_member_count;

1147
	if (mp->mglist &&
1148 1149 1150 1151 1152
	    (timer_pending(&mp->timer) ?
	     time_after(mp->timer.expires, now + max_delay) :
	     try_to_del_timer_sync(&mp->timer) >= 0))
		mod_timer(&mp->timer, now + max_delay);

1153 1154 1155
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
1156 1157 1158
		if (timer_pending(&p->timer) ?
		    time_after(p->timer.expires, now + max_delay) :
		    try_to_del_timer_sync(&p->timer) >= 0)
1159
			mod_timer(&p->timer, now + max_delay);
1160 1161 1162 1163
	}

out:
	spin_unlock(&br->multicast_lock);
1164
	return err;
1165 1166
}

E
Eric Dumazet 已提交
1167
#if IS_ENABLED(CONFIG_IPV6)
1168 1169 1170 1171
static int br_ip6_multicast_query(struct net_bridge *br,
				  struct net_bridge_port *port,
				  struct sk_buff *skb)
{
1172
	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
1173
	struct mld_msg *mld;
1174 1175
	struct net_bridge_mdb_entry *mp;
	struct mld2_query *mld2q;
1176 1177
	struct net_bridge_port_group *p;
	struct net_bridge_port_group __rcu **pp;
1178 1179
	unsigned long max_delay;
	unsigned long now = jiffies;
1180
	const struct in6_addr *group = NULL;
1181
	int err = 0;
1182
	u16 vid = 0;
1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) ||
	    (port && port->state == BR_STATE_DISABLED))
		goto out;

	if (skb->len == sizeof(*mld)) {
		if (!pskb_may_pull(skb, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *) icmp6_hdr(skb);
L
Li RongQing 已提交
1195
		max_delay = msecs_to_jiffies(ntohs(mld->mld_maxdelay));
1196 1197
		if (max_delay)
			group = &mld->mld_mca;
1198
	} else {
1199 1200 1201 1202 1203 1204 1205
		if (!pskb_may_pull(skb, sizeof(*mld2q))) {
			err = -EINVAL;
			goto out;
		}
		mld2q = (struct mld2_query *)icmp6_hdr(skb);
		if (!mld2q->mld2q_nsrcs)
			group = &mld2q->mld2q_mca;
1206 1207

		max_delay = max(msecs_to_jiffies(mldv2_mrc(mld2q)), 1UL);
1208 1209
	}

1210 1211 1212
	br_multicast_query_received(br, port, !ipv6_addr_any(&ip6h->saddr),
				    max_delay);

1213 1214 1215
	if (!group)
		goto out;

1216 1217
	br_vlan_get_tag(skb, &vid);
	mp = br_mdb_ip6_get(mlock_dereference(br->mdb, br), group, vid);
1218 1219 1220
	if (!mp)
		goto out;

1221 1222 1223
	mod_timer(&mp->timer, now + br->multicast_membership_interval);
	mp->timer_armed = true;

1224
	max_delay *= br->multicast_last_member_count;
1225
	if (mp->mglist &&
1226 1227 1228 1229 1230
	    (timer_pending(&mp->timer) ?
	     time_after(mp->timer.expires, now + max_delay) :
	     try_to_del_timer_sync(&mp->timer) >= 0))
		mod_timer(&mp->timer, now + max_delay);

1231 1232 1233
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
1234 1235 1236
		if (timer_pending(&p->timer) ?
		    time_after(p->timer.expires, now + max_delay) :
		    try_to_del_timer_sync(&p->timer) >= 0)
1237
			mod_timer(&p->timer, now + max_delay);
1238 1239 1240 1241 1242 1243 1244 1245
	}

out:
	spin_unlock(&br->multicast_lock);
	return err;
}
#endif

1246 1247
static void br_multicast_leave_group(struct net_bridge *br,
				     struct net_bridge_port *port,
1248
				     struct br_ip *group)
1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261
{
	struct net_bridge_mdb_htable *mdb;
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
	unsigned long now;
	unsigned long time;

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) ||
	    (port && port->state == BR_STATE_DISABLED) ||
	    timer_pending(&br->multicast_querier_timer))
		goto out;

1262
	mdb = mlock_dereference(br->mdb, br);
1263 1264 1265 1266
	mp = br_mdb_ip_get(mdb, group);
	if (!mp)
		goto out;

1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292
	if (br->multicast_querier &&
	    !timer_pending(&br->multicast_querier_timer)) {
		__br_multicast_send_query(br, port, &mp->addr);

		time = jiffies + br->multicast_last_member_count *
				 br->multicast_last_member_interval;
		mod_timer(port ? &port->multicast_query_timer :
				 &br->multicast_query_timer, time);

		for (p = mlock_dereference(mp->ports, br);
		     p != NULL;
		     p = mlock_dereference(p->next, br)) {
			if (p->port != port)
				continue;

			if (!hlist_unhashed(&p->mglist) &&
			    (timer_pending(&p->timer) ?
			     time_after(p->timer.expires, time) :
			     try_to_del_timer_sync(&p->timer) >= 0)) {
				mod_timer(&p->timer, time);
			}

			break;
		}
	}

1293
	if (port && (port->flags & BR_MULTICAST_FAST_LEAVE)) {
1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305
		struct net_bridge_port_group __rcu **pp;

		for (pp = &mp->ports;
		     (p = mlock_dereference(*pp, br)) != NULL;
		     pp = &p->next) {
			if (p->port != port)
				continue;

			rcu_assign_pointer(*pp, p->next);
			hlist_del_init(&p->mglist);
			del_timer(&p->timer);
			call_rcu_bh(&p->rcu, br_multicast_free_pg);
C
Cong Wang 已提交
1306
			br_mdb_notify(br->dev, port, group, RTM_DELMDB);
1307

1308
			if (!mp->ports && !mp->mglist && mp->timer_armed &&
1309 1310 1311 1312 1313 1314
			    netif_running(br->dev))
				mod_timer(&mp->timer, jiffies);
		}
		goto out;
	}

1315 1316 1317 1318 1319
	now = jiffies;
	time = now + br->multicast_last_member_count *
		     br->multicast_last_member_interval;

	if (!port) {
1320
		if (mp->mglist && mp->timer_armed &&
1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331
		    (timer_pending(&mp->timer) ?
		     time_after(mp->timer.expires, time) :
		     try_to_del_timer_sync(&mp->timer) >= 0)) {
			mod_timer(&mp->timer, time);
		}
	}

out:
	spin_unlock(&br->multicast_lock);
}

1332 1333
static void br_ip4_multicast_leave_group(struct net_bridge *br,
					 struct net_bridge_port *port,
1334 1335
					 __be32 group,
					 __u16 vid)
1336 1337 1338 1339 1340 1341 1342 1343
{
	struct br_ip br_group;

	if (ipv4_is_local_multicast(group))
		return;

	br_group.u.ip4 = group;
	br_group.proto = htons(ETH_P_IP);
1344
	br_group.vid = vid;
1345 1346 1347 1348

	br_multicast_leave_group(br, port, &br_group);
}

E
Eric Dumazet 已提交
1349
#if IS_ENABLED(CONFIG_IPV6)
1350 1351
static void br_ip6_multicast_leave_group(struct net_bridge *br,
					 struct net_bridge_port *port,
1352 1353
					 const struct in6_addr *group,
					 __u16 vid)
1354 1355 1356
{
	struct br_ip br_group;

1357
	if (!ipv6_is_transient_multicast(group))
1358 1359
		return;

A
Alexey Dobriyan 已提交
1360
	br_group.u.ip6 = *group;
1361
	br_group.proto = htons(ETH_P_IPV6);
1362
	br_group.vid = vid;
1363 1364 1365 1366

	br_multicast_leave_group(br, port, &br_group);
}
#endif
1367

1368 1369 1370 1371 1372
static int br_multicast_ipv4_rcv(struct net_bridge *br,
				 struct net_bridge_port *port,
				 struct sk_buff *skb)
{
	struct sk_buff *skb2 = skb;
1373
	const struct iphdr *iph;
1374
	struct igmphdr *ih;
1375 1376
	unsigned int len;
	unsigned int offset;
1377
	int err;
1378
	u16 vid = 0;
1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396

	/* We treat OOM as packet loss for now. */
	if (!pskb_may_pull(skb, sizeof(*iph)))
		return -EINVAL;

	iph = ip_hdr(skb);

	if (iph->ihl < 5 || iph->version != 4)
		return -EINVAL;

	if (!pskb_may_pull(skb, ip_hdrlen(skb)))
		return -EINVAL;

	iph = ip_hdr(skb);

	if (unlikely(ip_fast_csum((u8 *)iph, iph->ihl)))
		return -EINVAL;

1397
	if (iph->protocol != IPPROTO_IGMP) {
1398
		if (!ipv4_is_local_multicast(iph->daddr))
1399
			BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1400
		return 0;
1401
	}
1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413

	len = ntohs(iph->tot_len);
	if (skb->len < len || len < ip_hdrlen(skb))
		return -EINVAL;

	if (skb->len > len) {
		skb2 = skb_clone(skb, GFP_ATOMIC);
		if (!skb2)
			return -ENOMEM;

		err = pskb_trim_rcsum(skb2, len);
		if (err)
1414
			goto err_out;
1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433
	}

	len -= ip_hdrlen(skb2);
	offset = skb_network_offset(skb2) + ip_hdrlen(skb2);
	__skb_pull(skb2, offset);
	skb_reset_transport_header(skb2);

	err = -EINVAL;
	if (!pskb_may_pull(skb2, sizeof(*ih)))
		goto out;

	switch (skb2->ip_summed) {
	case CHECKSUM_COMPLETE:
		if (!csum_fold(skb2->csum))
			break;
		/* fall through */
	case CHECKSUM_NONE:
		skb2->csum = 0;
		if (skb_checksum_complete(skb2))
1434
			goto out;
1435 1436 1437 1438
	}

	err = 0;

1439
	br_vlan_get_tag(skb2, &vid);
1440 1441 1442 1443 1444 1445
	BR_INPUT_SKB_CB(skb)->igmp = 1;
	ih = igmp_hdr(skb2);

	switch (ih->type) {
	case IGMP_HOST_MEMBERSHIP_REPORT:
	case IGMPV2_HOST_MEMBERSHIP_REPORT:
1446
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1447
		err = br_ip4_multicast_add_group(br, port, ih->group, vid);
1448 1449
		break;
	case IGMPV3_HOST_MEMBERSHIP_REPORT:
1450
		err = br_ip4_multicast_igmp3_report(br, port, skb2);
1451 1452
		break;
	case IGMP_HOST_MEMBERSHIP_QUERY:
1453
		err = br_ip4_multicast_query(br, port, skb2);
1454 1455
		break;
	case IGMP_HOST_LEAVE_MESSAGE:
1456
		br_ip4_multicast_leave_group(br, port, ih->group, vid);
1457 1458 1459 1460 1461
		break;
	}

out:
	__skb_push(skb2, offset);
1462
err_out:
1463 1464 1465 1466 1467
	if (skb2 != skb)
		kfree_skb(skb2);
	return err;
}

E
Eric Dumazet 已提交
1468
#if IS_ENABLED(CONFIG_IPV6)
1469 1470 1471 1472
static int br_multicast_ipv6_rcv(struct net_bridge *br,
				 struct net_bridge_port *port,
				 struct sk_buff *skb)
{
1473
	struct sk_buff *skb2;
1474
	const struct ipv6hdr *ip6h;
1475
	u8 icmp6_type;
1476
	u8 nexthdr;
1477
	__be16 frag_off;
1478
	unsigned int len;
K
Kulikov Vasiliy 已提交
1479
	int offset;
1480
	int err;
1481
	u16 vid = 0;
1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493

	if (!pskb_may_pull(skb, sizeof(*ip6h)))
		return -EINVAL;

	ip6h = ipv6_hdr(skb);

	/*
	 * We're interested in MLD messages only.
	 *  - Version is 6
	 *  - MLD has always Router Alert hop-by-hop option
	 *  - But we do not support jumbrograms.
	 */
1494 1495 1496 1497 1498 1499 1500 1501
	if (ip6h->version != 6)
		return 0;

	/* Prevent flooding this packet if there is no listener present */
	if (ipv6_is_transient_multicast(&ip6h->daddr))
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;

	if (ip6h->nexthdr != IPPROTO_HOPOPTS ||
1502 1503 1504
	    ip6h->payload_len == 0)
		return 0;

1505
	len = ntohs(ip6h->payload_len) + sizeof(*ip6h);
1506 1507 1508 1509
	if (skb->len < len)
		return -EINVAL;

	nexthdr = ip6h->nexthdr;
1510
	offset = ipv6_skip_exthdr(skb, sizeof(*ip6h), &nexthdr, &frag_off);
1511 1512 1513 1514 1515 1516 1517 1518 1519

	if (offset < 0 || nexthdr != IPPROTO_ICMPV6)
		return 0;

	/* Okay, we found ICMPv6 header */
	skb2 = skb_clone(skb, GFP_ATOMIC);
	if (!skb2)
		return -ENOMEM;

1520 1521 1522 1523
	err = -EINVAL;
	if (!pskb_may_pull(skb2, offset + sizeof(struct icmp6hdr)))
		goto out;

1524 1525 1526 1527
	len -= offset - skb_network_offset(skb2);

	__skb_pull(skb2, offset);
	skb_reset_transport_header(skb2);
1528 1529
	skb_postpull_rcsum(skb2, skb_network_header(skb2),
			   skb_network_header_len(skb2));
1530

1531
	icmp6_type = icmp6_hdr(skb2)->icmp6_type;
1532

1533
	switch (icmp6_type) {
1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548
	case ICMPV6_MGM_QUERY:
	case ICMPV6_MGM_REPORT:
	case ICMPV6_MGM_REDUCTION:
	case ICMPV6_MLD2_REPORT:
		break;
	default:
		err = 0;
		goto out;
	}

	/* Okay, we found MLD message. Check further. */
	if (skb2->len > len) {
		err = pskb_trim_rcsum(skb2, len);
		if (err)
			goto out;
1549
		err = -EINVAL;
1550 1551
	}

1552 1553
	ip6h = ipv6_hdr(skb2);

1554 1555
	switch (skb2->ip_summed) {
	case CHECKSUM_COMPLETE:
1556 1557
		if (!csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, skb2->len,
					IPPROTO_ICMPV6, skb2->csum))
1558 1559 1560
			break;
		/*FALLTHROUGH*/
	case CHECKSUM_NONE:
1561 1562 1563 1564 1565
		skb2->csum = ~csum_unfold(csum_ipv6_magic(&ip6h->saddr,
							&ip6h->daddr,
							skb2->len,
							IPPROTO_ICMPV6, 0));
		if (__skb_checksum_complete(skb2))
1566 1567 1568 1569 1570
			goto out;
	}

	err = 0;

1571
	br_vlan_get_tag(skb, &vid);
1572 1573
	BR_INPUT_SKB_CB(skb)->igmp = 1;

1574
	switch (icmp6_type) {
1575 1576
	case ICMPV6_MGM_REPORT:
	    {
1577 1578 1579 1580 1581 1582
		struct mld_msg *mld;
		if (!pskb_may_pull(skb2, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *)skb_transport_header(skb2);
1583
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1584
		err = br_ip6_multicast_add_group(br, port, &mld->mld_mca, vid);
1585 1586 1587 1588 1589 1590 1591 1592 1593 1594
		break;
	    }
	case ICMPV6_MLD2_REPORT:
		err = br_ip6_multicast_mld2_report(br, port, skb2);
		break;
	case ICMPV6_MGM_QUERY:
		err = br_ip6_multicast_query(br, port, skb2);
		break;
	case ICMPV6_MGM_REDUCTION:
	    {
1595 1596 1597 1598 1599 1600
		struct mld_msg *mld;
		if (!pskb_may_pull(skb2, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *)skb_transport_header(skb2);
1601
		br_ip6_multicast_leave_group(br, port, &mld->mld_mca, vid);
1602 1603 1604 1605
	    }
	}

out:
1606
	kfree_skb(skb2);
1607 1608 1609 1610
	return err;
}
#endif

1611 1612 1613
int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port,
		     struct sk_buff *skb)
{
1614 1615 1616
	BR_INPUT_SKB_CB(skb)->igmp = 0;
	BR_INPUT_SKB_CB(skb)->mrouters_only = 0;

1617 1618 1619 1620 1621 1622
	if (br->multicast_disabled)
		return 0;

	switch (skb->protocol) {
	case htons(ETH_P_IP):
		return br_multicast_ipv4_rcv(br, port, skb);
E
Eric Dumazet 已提交
1623
#if IS_ENABLED(CONFIG_IPV6)
1624 1625 1626
	case htons(ETH_P_IPV6):
		return br_multicast_ipv6_rcv(br, port, skb);
#endif
1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650 1651
	}

	return 0;
}

static void br_multicast_query_expired(unsigned long data)
{
	struct net_bridge *br = (void *)data;

	spin_lock(&br->multicast_lock);
	if (br->multicast_startup_queries_sent <
	    br->multicast_startup_query_count)
		br->multicast_startup_queries_sent++;

	br_multicast_send_query(br, NULL, br->multicast_startup_queries_sent);

	spin_unlock(&br->multicast_lock);
}

void br_multicast_init(struct net_bridge *br)
{
	br->hash_elasticity = 4;
	br->hash_max = 512;

	br->multicast_router = 1;
1652
	br->multicast_querier = 0;
1653
	br->multicast_query_use_ifaddr = 0;
1654 1655 1656 1657 1658 1659 1660 1661 1662 1663
	br->multicast_last_member_count = 2;
	br->multicast_startup_query_count = 2;

	br->multicast_last_member_interval = HZ;
	br->multicast_query_response_interval = 10 * HZ;
	br->multicast_startup_query_interval = 125 * HZ / 4;
	br->multicast_query_interval = 125 * HZ;
	br->multicast_querier_interval = 255 * HZ;
	br->multicast_membership_interval = 260 * HZ;

1664 1665
	br->multicast_querier_delay_time = 0;

1666 1667 1668 1669
	spin_lock_init(&br->multicast_lock);
	setup_timer(&br->multicast_router_timer,
		    br_multicast_local_router_expired, 0);
	setup_timer(&br->multicast_querier_timer,
1670
		    br_multicast_querier_expired, (unsigned long)br);
1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688
	setup_timer(&br->multicast_query_timer, br_multicast_query_expired,
		    (unsigned long)br);
}

void br_multicast_open(struct net_bridge *br)
{
	br->multicast_startup_queries_sent = 0;

	if (br->multicast_disabled)
		return;

	mod_timer(&br->multicast_query_timer, jiffies);
}

void br_multicast_stop(struct net_bridge *br)
{
	struct net_bridge_mdb_htable *mdb;
	struct net_bridge_mdb_entry *mp;
1689
	struct hlist_node *n;
1690 1691 1692 1693 1694 1695 1696 1697
	u32 ver;
	int i;

	del_timer_sync(&br->multicast_router_timer);
	del_timer_sync(&br->multicast_querier_timer);
	del_timer_sync(&br->multicast_query_timer);

	spin_lock_bh(&br->multicast_lock);
1698
	mdb = mlock_dereference(br->mdb, br);
1699 1700 1701 1702 1703 1704 1705
	if (!mdb)
		goto out;

	br->mdb = NULL;

	ver = mdb->ver;
	for (i = 0; i < mdb->max; i++) {
1706
		hlist_for_each_entry_safe(mp, n, &mdb->mhash[i],
1707 1708
					  hlist[ver]) {
			del_timer(&mp->timer);
1709
			mp->timer_armed = false;
1710 1711 1712 1713 1714 1715
			call_rcu_bh(&mp->rcu, br_multicast_free_group);
		}
	}

	if (mdb->old) {
		spin_unlock_bh(&br->multicast_lock);
1716
		rcu_barrier_bh();
1717 1718 1719 1720 1721 1722 1723 1724 1725 1726
		spin_lock_bh(&br->multicast_lock);
		WARN_ON(mdb->old);
	}

	mdb->old = mdb;
	call_rcu_bh(&mdb->rcu, br_mdb_free);

out:
	spin_unlock_bh(&br->multicast_lock);
}
1727 1728 1729 1730 1731 1732 1733 1734 1735 1736 1737 1738 1739 1740 1741 1742 1743 1744 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 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796

int br_multicast_set_router(struct net_bridge *br, unsigned long val)
{
	int err = -ENOENT;

	spin_lock_bh(&br->multicast_lock);
	if (!netif_running(br->dev))
		goto unlock;

	switch (val) {
	case 0:
	case 2:
		del_timer(&br->multicast_router_timer);
		/* fall through */
	case 1:
		br->multicast_router = val;
		err = 0;
		break;

	default:
		err = -EINVAL;
		break;
	}

unlock:
	spin_unlock_bh(&br->multicast_lock);

	return err;
}

int br_multicast_set_port_router(struct net_bridge_port *p, unsigned long val)
{
	struct net_bridge *br = p->br;
	int err = -ENOENT;

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) || p->state == BR_STATE_DISABLED)
		goto unlock;

	switch (val) {
	case 0:
	case 1:
	case 2:
		p->multicast_router = val;
		err = 0;

		if (val < 2 && !hlist_unhashed(&p->rlist))
			hlist_del_init_rcu(&p->rlist);

		if (val == 1)
			break;

		del_timer(&p->multicast_router_timer);

		if (val == 0)
			break;

		br_multicast_add_router(br, p);
		break;

	default:
		err = -EINVAL;
		break;
	}

unlock:
	spin_unlock(&br->multicast_lock);

	return err;
}
1797

1798
static void br_multicast_start_querier(struct net_bridge *br)
1799 1800
{
	struct net_bridge_port *port;
1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812 1813 1814

	br_multicast_open(br);

	list_for_each_entry(port, &br->port_list, list) {
		if (port->state == BR_STATE_DISABLED ||
		    port->state == BR_STATE_BLOCKING)
			continue;

		__br_multicast_enable_port(port);
	}
}

int br_multicast_toggle(struct net_bridge *br, unsigned long val)
{
1815
	int err = 0;
1816
	struct net_bridge_mdb_htable *mdb;
1817

1818
	spin_lock_bh(&br->multicast_lock);
1819 1820 1821 1822 1823 1824 1825
	if (br->multicast_disabled == !val)
		goto unlock;

	br->multicast_disabled = !val;
	if (br->multicast_disabled)
		goto unlock;

1826 1827 1828
	if (!netif_running(br->dev))
		goto unlock;

1829 1830 1831
	mdb = mlock_dereference(br->mdb, br);
	if (mdb) {
		if (mdb->old) {
1832 1833 1834 1835 1836 1837
			err = -EEXIST;
rollback:
			br->multicast_disabled = !!val;
			goto unlock;
		}

1838
		err = br_mdb_rehash(&br->mdb, mdb->max,
1839 1840 1841 1842 1843
				    br->hash_elasticity);
		if (err)
			goto rollback;
	}

1844
	br_multicast_start_querier(br);
1845 1846

unlock:
1847
	spin_unlock_bh(&br->multicast_lock);
1848 1849 1850

	return err;
}
1851

1852 1853
int br_multicast_set_querier(struct net_bridge *br, unsigned long val)
{
1854 1855
	unsigned long max_delay;

1856 1857 1858 1859 1860 1861 1862
	val = !!val;

	spin_lock_bh(&br->multicast_lock);
	if (br->multicast_querier == val)
		goto unlock;

	br->multicast_querier = val;
1863 1864 1865 1866 1867 1868 1869 1870
	if (!val)
		goto unlock;

	max_delay = br->multicast_query_response_interval;
	if (!timer_pending(&br->multicast_querier_timer))
		br->multicast_querier_delay_time = jiffies + max_delay;

	br_multicast_start_querier(br);
1871 1872 1873 1874 1875 1876 1877

unlock:
	spin_unlock_bh(&br->multicast_lock);

	return 0;
}

1878 1879 1880 1881
int br_multicast_set_hash_max(struct net_bridge *br, unsigned long val)
{
	int err = -ENOENT;
	u32 old;
1882
	struct net_bridge_mdb_htable *mdb;
1883 1884 1885 1886 1887 1888 1889 1890

	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev))
		goto unlock;

	err = -EINVAL;
	if (!is_power_of_2(val))
		goto unlock;
1891 1892 1893

	mdb = mlock_dereference(br->mdb, br);
	if (mdb && val < mdb->size)
1894 1895 1896 1897 1898 1899 1900
		goto unlock;

	err = 0;

	old = br->hash_max;
	br->hash_max = val;

1901 1902
	if (mdb) {
		if (mdb->old) {
1903 1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919
			err = -EEXIST;
rollback:
			br->hash_max = old;
			goto unlock;
		}

		err = br_mdb_rehash(&br->mdb, br->hash_max,
				    br->hash_elasticity);
		if (err)
			goto rollback;
	}

unlock:
	spin_unlock(&br->multicast_lock);

	return err;
}