br_multicast.c 51.1 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13
/*
 * 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>
14
#include <linux/export.h>
15 16 17 18
#include <linux/if_ether.h>
#include <linux/igmp.h>
#include <linux/jhash.h>
#include <linux/kernel.h>
19
#include <linux/log2.h>
20 21 22 23 24 25 26
#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>
27
#include <linux/inetdevice.h>
28
#include <net/ip.h>
E
Eric Dumazet 已提交
29
#if IS_ENABLED(CONFIG_IPV6)
30 31
#include <net/ipv6.h>
#include <net/mld.h>
32
#include <net/ip6_checksum.h>
33
#include <net/addrconf.h>
34
#endif
35 36 37

#include "br_private.h"

38
static void br_multicast_start_querier(struct net_bridge *br,
39
				       struct bridge_mcast_own_query *query);
C
Cong Wang 已提交
40
unsigned int br_mdb_rehash_seq;
41

42 43 44 45
static inline int br_ip_equal(const struct br_ip *a, const struct br_ip *b)
{
	if (a->proto != b->proto)
		return 0;
46 47
	if (a->vid != b->vid)
		return 0;
48 49 50
	switch (a->proto) {
	case htons(ETH_P_IP):
		return a->u.ip4 == b->u.ip4;
E
Eric Dumazet 已提交
51
#if IS_ENABLED(CONFIG_IPV6)
52 53 54
	case htons(ETH_P_IPV6):
		return ipv6_addr_equal(&a->u.ip6, &b->u.ip6);
#endif
55 56 57 58
	}
	return 0;
}

59 60
static inline int __br_ip4_hash(struct net_bridge_mdb_htable *mdb, __be32 ip,
				__u16 vid)
61
{
62
	return jhash_2words((__force u32)ip, vid, mdb->secret) & (mdb->max - 1);
63 64
}

E
Eric Dumazet 已提交
65
#if IS_ENABLED(CONFIG_IPV6)
66
static inline int __br_ip6_hash(struct net_bridge_mdb_htable *mdb,
67 68
				const struct in6_addr *ip,
				__u16 vid)
69
{
70 71
	return jhash_2words(ipv6_addr_hash(ip), vid,
			    mdb->secret) & (mdb->max - 1);
72 73 74
}
#endif

75 76 77 78 79
static inline int br_ip_hash(struct net_bridge_mdb_htable *mdb,
			     struct br_ip *ip)
{
	switch (ip->proto) {
	case htons(ETH_P_IP):
80
		return __br_ip4_hash(mdb, ip->u.ip4, ip->vid);
E
Eric Dumazet 已提交
81
#if IS_ENABLED(CONFIG_IPV6)
82
	case htons(ETH_P_IPV6):
83
		return __br_ip6_hash(mdb, &ip->u.ip6, ip->vid);
84
#endif
85 86
	}
	return 0;
87 88 89
}

static struct net_bridge_mdb_entry *__br_mdb_ip_get(
90
	struct net_bridge_mdb_htable *mdb, struct br_ip *dst, int hash)
91 92 93
{
	struct net_bridge_mdb_entry *mp;

94
	hlist_for_each_entry_rcu(mp, &mdb->mhash[hash], hlist[mdb->ver]) {
95
		if (br_ip_equal(&mp->addr, dst))
96 97 98 99 100 101
			return mp;
	}

	return NULL;
}

102 103
struct net_bridge_mdb_entry *br_mdb_ip_get(struct net_bridge_mdb_htable *mdb,
					   struct br_ip *dst)
104 105 106 107 108 109 110
{
	if (!mdb)
		return NULL;

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

111
static struct net_bridge_mdb_entry *br_mdb_ip4_get(
112
	struct net_bridge_mdb_htable *mdb, __be32 dst, __u16 vid)
113
{
114 115 116 117
	struct br_ip br_dst;

	br_dst.u.ip4 = dst;
	br_dst.proto = htons(ETH_P_IP);
118
	br_dst.vid = vid;
119

120
	return br_mdb_ip_get(mdb, &br_dst);
121 122
}

E
Eric Dumazet 已提交
123
#if IS_ENABLED(CONFIG_IPV6)
124
static struct net_bridge_mdb_entry *br_mdb_ip6_get(
125 126
	struct net_bridge_mdb_htable *mdb, const struct in6_addr *dst,
	__u16 vid)
127 128
{
	struct br_ip br_dst;
129

A
Alexey Dobriyan 已提交
130
	br_dst.u.ip6 = *dst;
131
	br_dst.proto = htons(ETH_P_IPV6);
132
	br_dst.vid = vid;
133

134
	return br_mdb_ip_get(mdb, &br_dst);
135 136 137
}
#endif

138
struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br,
139
					struct sk_buff *skb, u16 vid)
140
{
141
	struct net_bridge_mdb_htable *mdb = rcu_dereference(br->mdb);
142 143
	struct br_ip ip;

144
	if (br->multicast_disabled)
145 146
		return NULL;

147
	if (BR_INPUT_SKB_CB(skb)->igmp)
148 149
		return NULL;

150
	ip.proto = skb->protocol;
151
	ip.vid = vid;
152

153 154
	switch (skb->protocol) {
	case htons(ETH_P_IP):
155 156
		ip.u.ip4 = ip_hdr(skb)->daddr;
		break;
E
Eric Dumazet 已提交
157
#if IS_ENABLED(CONFIG_IPV6)
158
	case htons(ETH_P_IPV6):
A
Alexey Dobriyan 已提交
159
		ip.u.ip6 = ipv6_hdr(skb)->daddr;
160 161
		break;
#endif
162 163
	default:
		return NULL;
164 165
	}

166
	return br_mdb_ip_get(mdb, &ip);
167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189
}

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++)
190
		hlist_for_each_entry(mp, &old->mhash[i], hlist[old->ver])
191
			hlist_add_head(&mp->hlist[new->ver],
192
				       &new->mhash[br_ip_hash(new, &mp->addr)]);
193 194 195 196 197 198 199

	if (!elasticity)
		return 0;

	maxlen = 0;
	for (i = 0; i < new->max; i++) {
		len = 0;
200
		hlist_for_each_entry(mp, &new->mhash[i], hlist[new->ver])
201 202 203 204 205 206 207 208
			len++;
		if (len > maxlen)
			maxlen = len;
	}

	return maxlen > elasticity ? -EINVAL : 0;
}

209
void br_multicast_free_pg(struct rcu_head *head)
210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
{
	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;

235
	mp->mglist = false;
236 237 238 239

	if (mp->ports)
		goto out;

240 241
	mdb = mlock_dereference(br->mdb, br);

242 243 244 245 246 247 248 249 250 251 252 253
	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)
{
254
	struct net_bridge_mdb_htable *mdb;
255 256
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
257 258 259
	struct net_bridge_port_group __rcu **pp;

	mdb = mlock_dereference(br->mdb, br);
260

261
	mp = br_mdb_ip_get(mdb, &pg->addr);
262 263 264
	if (WARN_ON(!mp))
		return;

265 266 267
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
268 269 270
		if (p != pg)
			continue;

271
		rcu_assign_pointer(*pp, p->next);
272 273 274 275
		hlist_del_init(&p->mglist);
		del_timer(&p->timer);
		call_rcu_bh(&p->rcu, br_multicast_free_pg);

276
		if (!mp->ports && !mp->mglist &&
277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292
		    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) ||
293
	    hlist_unhashed(&pg->mglist) || pg->state & MDB_PERMANENT)
294 295 296 297 298 299 300 301
		goto out;

	br_multicast_del_pg(br, pg);

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

302
static int br_mdb_rehash(struct net_bridge_mdb_htable __rcu **mdbp, int max,
303 304
			 int elasticity)
{
305
	struct net_bridge_mdb_htable *old = rcu_dereference_protected(*mdbp, 1);
306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339
	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
340
	br_mdb_rehash_seq++;
341 342 343 344 345 346 347 348
	call_rcu_bh(&mdb->rcu, br_mdb_free);

out:
	rcu_assign_pointer(*mdbp, mdb);

	return 0;
}

349 350
static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br,
						    __be32 group)
351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366
{
	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);

367
	ether_addr_copy(eth->h_source, br->dev->dev_addr);
368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387
	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;
388 389
	iph->saddr = br->multicast_query_use_ifaddr ?
		     inet_select_addr(br->dev, 0, RT_SCOPE_LINK) : 0;
390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414
	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
415
#if IS_ENABLED(CONFIG_IPV6)
416
static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br,
417
						    const struct in6_addr *group)
418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436
{
	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);

437
	ether_addr_copy(eth->h_source, br->dev->dev_addr);
438 439 440 441 442 443 444 445
	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);
446
	ip6h->payload_len = htons(8 + sizeof(*mldq));
447 448
	ip6h->nexthdr = IPPROTO_HOPOPTS;
	ip6h->hop_limit = 1;
449
	ipv6_addr_set(&ip6h->daddr, htonl(0xff020000), 0, 0, htonl(1));
450 451 452 453 454
	if (ipv6_dev_get_saddr(dev_net(br->dev), br->dev, &ip6h->daddr, 0,
			       &ip6h->saddr)) {
		kfree_skb(skb);
		return NULL;
	}
455
	ipv6_eth_mc_map(&ip6h->daddr, eth->h_dest);
456 457 458 459 460 461 462 463

	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;
464 465
	hopopt[6] = IPV6_TLV_PAD1;		/* Pad1 */
	hopopt[7] = IPV6_TLV_PAD1;		/* Pad1 */
466 467 468 469 470 471 472

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

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

473 474 475
	interval = ipv6_addr_any(group) ?
			br->multicast_query_response_interval :
			br->multicast_last_member_interval;
476 477 478 479 480 481

	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
482
	mldq->mld_mca = *group;
483 484 485 486 487 488 489 490 491 492 493 494 495 496 497

	/* 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

498 499 500 501 502 503
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
504
#if IS_ENABLED(CONFIG_IPV6)
505 506 507
	case htons(ETH_P_IPV6):
		return br_ip6_multicast_alloc_query(br, &addr->u.ip6);
#endif
508 509 510 511
	}
	return NULL;
}

512
static struct net_bridge_mdb_entry *br_multicast_get_group(
513 514
	struct net_bridge *br, struct net_bridge_port *port,
	struct br_ip *group, int hash)
515
{
516
	struct net_bridge_mdb_htable *mdb;
517
	struct net_bridge_mdb_entry *mp;
518 519
	unsigned int count = 0;
	unsigned int max;
520 521 522
	int elasticity;
	int err;

523
	mdb = rcu_dereference_protected(br->mdb, 1);
524
	hlist_for_each_entry(mp, &mdb->mhash[hash], hlist[mdb->ver]) {
525
		count++;
526
		if (unlikely(br_ip_equal(group, &mp->addr)))
527 528 529 530 531 532 533 534
			return mp;
	}

	elasticity = 0;
	max = mdb->max;

	if (unlikely(count > br->hash_elasticity && count)) {
		if (net_ratelimit())
535 536 537
			br_info(br, "Multicast hash table "
				"chain limit reached: %s\n",
				port ? port->dev->name : br->dev->name);
538 539 540 541 542 543

		elasticity = br->hash_elasticity;
	}

	if (mdb->size >= max) {
		max *= 2;
544 545 546 547 548
		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);
549 550 551 552 553 554 555 556 557 558
			err = -E2BIG;
disable:
			br->multicast_disabled = 1;
			goto err;
		}
	}

	if (max > mdb->max || elasticity) {
		if (mdb->old) {
			if (net_ratelimit())
559 560 561
				br_info(br, "Multicast hash table "
					"on fire: %s\n",
					port ? port->dev->name : br->dev->name);
562 563 564 565 566 567
			err = -EEXIST;
			goto err;
		}

		err = br_mdb_rehash(&br->mdb, max, elasticity);
		if (err) {
568 569 570 571
			br_warn(br, "Cannot rehash multicast "
				"hash table, disabling snooping: %s, %d, %d\n",
				port ? port->dev->name : br->dev->name,
				mdb->size, err);
572 573 574 575 576 577 578 579 580 581 582 583 584 585
			goto disable;
		}

		err = -EAGAIN;
		goto err;
	}

	return NULL;

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

586 587
struct net_bridge_mdb_entry *br_multicast_new_group(struct net_bridge *br,
	struct net_bridge_port *port, struct br_ip *group)
588
{
589
	struct net_bridge_mdb_htable *mdb;
590 591
	struct net_bridge_mdb_entry *mp;
	int hash;
592
	int err;
593

594
	mdb = rcu_dereference_protected(br->mdb, 1);
595
	if (!mdb) {
596 597 598
		err = br_mdb_rehash(&br->mdb, BR_HASH_SIZE, 0);
		if (err)
			return ERR_PTR(err);
599 600 601 602 603 604 605 606 607 608 609
		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:
610
		mdb = rcu_dereference_protected(br->mdb, 1);
611 612 613 614 615 616 617 618 619
		hash = br_ip_hash(mdb, group);
		break;

	default:
		goto out;
	}

	mp = kzalloc(sizeof(*mp), GFP_ATOMIC);
	if (unlikely(!mp))
620
		return ERR_PTR(-ENOMEM);
621 622

	mp->br = br;
623
	mp->addr = *group;
624 625 626
	setup_timer(&mp->timer, br_multicast_group_expired,
		    (unsigned long)mp);

627 628 629 630 631 632 633
	hlist_add_head_rcu(&mp->hlist[mdb->ver], &mdb->mhash[hash]);
	mdb->size++;

out:
	return mp;
}

634 635 636
struct net_bridge_port_group *br_multicast_new_port_group(
			struct net_bridge_port *port,
			struct br_ip *group,
637 638
			struct net_bridge_port_group __rcu *next,
			unsigned char state)
639 640 641 642 643 644 645 646 647
{
	struct net_bridge_port_group *p;

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

	p->addr = *group;
	p->port = port;
648
	p->state = state;
649
	rcu_assign_pointer(p->next, next);
650 651 652 653 654 655
	hlist_add_head(&p->mglist, &port->mglist);
	setup_timer(&p->timer, br_multicast_port_group_expired,
		    (unsigned long)p);
	return p;
}

656
static int br_multicast_add_group(struct net_bridge *br,
657 658
				  struct net_bridge_port *port,
				  struct br_ip *group)
659 660 661
{
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
662
	struct net_bridge_port_group __rcu **pp;
663
	unsigned long now = jiffies;
664 665 666 667 668 669 670 671 672
	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);
673
	if (IS_ERR(mp))
674 675 676
		goto err;

	if (!port) {
677
		mp->mglist = true;
678
		mod_timer(&mp->timer, now + br->multicast_membership_interval);
679 680 681
		goto out;
	}

682 683 684
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
685
		if (p->port == port)
686
			goto found;
687 688 689 690
		if ((unsigned long)p->port < (unsigned long)port)
			break;
	}

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

697 698
found:
	mod_timer(&p->timer, now + br->multicast_membership_interval);
699 700 701 702 703 704 705 706
out:
	err = 0;

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

707 708
static int br_ip4_multicast_add_group(struct net_bridge *br,
				      struct net_bridge_port *port,
709 710
				      __be32 group,
				      __u16 vid)
711 712 713 714 715 716 717 718
{
	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);
719
	br_group.vid = vid;
720 721 722 723

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

E
Eric Dumazet 已提交
724
#if IS_ENABLED(CONFIG_IPV6)
725 726
static int br_ip6_multicast_add_group(struct net_bridge *br,
				      struct net_bridge_port *port,
727 728
				      const struct in6_addr *group,
				      __u16 vid)
729 730 731
{
	struct br_ip br_group;

732
	if (ipv6_addr_is_ll_all_nodes(group))
733 734
		return 0;

A
Alexey Dobriyan 已提交
735
	br_group.u.ip6 = *group;
736
	br_group.proto = htons(ETH_P_IPV6);
737
	br_group.vid = vid;
738 739 740 741 742

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

743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763
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)
{
}

764
static void br_multicast_querier_expired(struct net_bridge *br,
765
					 struct bridge_mcast_own_query *query)
766 767 768 769 770
{
	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) || br->multicast_disabled)
		goto out;

771
	br_multicast_start_querier(br, query);
772 773 774 775 776

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

777 778 779 780
static void br_ip4_multicast_querier_expired(unsigned long data)
{
	struct net_bridge *br = (void *)data;

781
	br_multicast_querier_expired(br, &br->ip4_own_query);
782 783 784 785 786 787 788
}

#if IS_ENABLED(CONFIG_IPV6)
static void br_ip6_multicast_querier_expired(unsigned long data)
{
	struct net_bridge *br = (void *)data;

789
	br_multicast_querier_expired(br, &br->ip6_own_query);
790 791 792
}
#endif

793 794 795 796 797 798 799 800 801 802 803 804
static void br_multicast_select_own_querier(struct net_bridge *br,
					    struct br_ip *ip,
					    struct sk_buff *skb)
{
	if (ip->proto == htons(ETH_P_IP))
		br->ip4_querier.addr.u.ip4 = ip_hdr(skb)->saddr;
#if IS_ENABLED(CONFIG_IPV6)
	else
		br->ip6_querier.addr.u.ip6 = ipv6_hdr(skb)->saddr;
#endif
}

805 806 807
static void __br_multicast_send_query(struct net_bridge *br,
				      struct net_bridge_port *port,
				      struct br_ip *ip)
808 809 810
{
	struct sk_buff *skb;

811
	skb = br_multicast_alloc_query(br, ip);
812
	if (!skb)
813
		return;
814 815 816 817

	if (port) {
		__skb_push(skb, sizeof(struct ethhdr));
		skb->dev = port->dev;
818
		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
819
			dev_queue_xmit);
820 821
	} else {
		br_multicast_select_own_querier(br, ip, skb);
822
		netif_rx(skb);
823
	}
824 825 826
}

static void br_multicast_send_query(struct net_bridge *br,
827
				    struct net_bridge_port *port,
828
				    struct bridge_mcast_own_query *own_query)
829 830 831
{
	unsigned long time;
	struct br_ip br_group;
832
	struct bridge_mcast_other_query *other_query = NULL;
833 834

	if (!netif_running(br->dev) || br->multicast_disabled ||
835
	    !br->multicast_querier)
836 837
		return;

838 839
	memset(&br_group.u, 0, sizeof(br_group.u));

840 841 842
	if (port ? (own_query == &port->ip4_own_query) :
		   (own_query == &br->ip4_own_query)) {
		other_query = &br->ip4_other_query;
843
		br_group.proto = htons(ETH_P_IP);
E
Eric Dumazet 已提交
844
#if IS_ENABLED(CONFIG_IPV6)
845
	} else {
846
		other_query = &br->ip6_other_query;
847
		br_group.proto = htons(ETH_P_IPV6);
848
#endif
849 850
	}

851
	if (!other_query || timer_pending(&other_query->timer))
852 853 854
		return;

	__br_multicast_send_query(br, port, &br_group);
855 856

	time = jiffies;
857
	time += own_query->startup_sent < br->multicast_startup_query_count ?
858 859
		br->multicast_startup_query_interval :
		br->multicast_query_interval;
860
	mod_timer(&own_query->timer, time);
861 862
}

863 864 865
static void
br_multicast_port_query_expired(struct net_bridge_port *port,
				struct bridge_mcast_own_query *query)
866 867 868 869
{
	struct net_bridge *br = port->br;

	spin_lock(&br->multicast_lock);
870 871
	if (port->state == BR_STATE_DISABLED ||
	    port->state == BR_STATE_BLOCKING)
872 873
		goto out;

874 875
	if (query->startup_sent < br->multicast_startup_query_count)
		query->startup_sent++;
876

877
	br_multicast_send_query(port->br, port, query);
878 879 880 881 882

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

883 884 885 886
static void br_ip4_multicast_port_query_expired(unsigned long data)
{
	struct net_bridge_port *port = (void *)data;

887
	br_multicast_port_query_expired(port, &port->ip4_own_query);
888 889 890 891 892 893 894
}

#if IS_ENABLED(CONFIG_IPV6)
static void br_ip6_multicast_port_query_expired(unsigned long data)
{
	struct net_bridge_port *port = (void *)data;

895
	br_multicast_port_query_expired(port, &port->ip6_own_query);
896 897 898
}
#endif

899 900 901 902 903 904
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);
905 906
	setup_timer(&port->ip4_own_query.timer,
		    br_ip4_multicast_port_query_expired, (unsigned long)port);
907
#if IS_ENABLED(CONFIG_IPV6)
908 909
	setup_timer(&port->ip6_own_query.timer,
		    br_ip6_multicast_port_query_expired, (unsigned long)port);
910
#endif
911 912 913 914 915 916 917
}

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

918
static void br_multicast_enable(struct bridge_mcast_own_query *query)
919
{
920
	query->startup_sent = 0;
921

922 923 924
	if (try_to_del_timer_sync(&query->timer) >= 0 ||
	    del_timer(&query->timer))
		mod_timer(&query->timer, jiffies);
925 926
}

927 928 929 930 931 932 933 934
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;

935
	br_multicast_enable(&port->ip4_own_query);
936
#if IS_ENABLED(CONFIG_IPV6)
937
	br_multicast_enable(&port->ip6_own_query);
938
#endif
939 940 941 942 943 944 945 946 947

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;
948
	struct hlist_node *n;
949 950

	spin_lock(&br->multicast_lock);
951
	hlist_for_each_entry_safe(pg, n, &port->mglist, mglist)
952 953 954 955 956
		br_multicast_del_pg(br, pg);

	if (!hlist_unhashed(&port->rlist))
		hlist_del_init_rcu(&port->rlist);
	del_timer(&port->multicast_router_timer);
957
	del_timer(&port->ip4_own_query.timer);
958
#if IS_ENABLED(CONFIG_IPV6)
959
	del_timer(&port->ip6_own_query.timer);
960
#endif
961 962 963
	spin_unlock(&br->multicast_lock);
}

964 965
static int br_ip4_multicast_igmp3_report(struct net_bridge *br,
					 struct net_bridge_port *port,
966 967
					 struct sk_buff *skb,
					 u16 vid)
968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989
{
	struct igmpv3_report *ih;
	struct igmpv3_grec *grec;
	int i;
	int len;
	int num;
	int type;
	int err = 0;
	__be32 group;

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

	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
990
		grec = (void *)(skb->data + len - sizeof(*grec));
991 992 993
		group = grec->grec_mca;
		type = grec->grec_type;

E
Eric Dumazet 已提交
994
		len += ntohs(grec->grec_nsrcs) * 4;
995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
		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;
		}

1012
		err = br_ip4_multicast_add_group(br, port, group, vid);
1013 1014 1015 1016 1017 1018 1019
		if (err)
			break;
	}

	return err;
}

E
Eric Dumazet 已提交
1020
#if IS_ENABLED(CONFIG_IPV6)
1021 1022
static int br_ip6_multicast_mld2_report(struct net_bridge *br,
					struct net_bridge_port *port,
1023 1024
					struct sk_buff *skb,
					u16 vid)
1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044
{
	struct icmp6hdr *icmp6h;
	struct mld2_grec *grec;
	int i;
	int len;
	int num;
	int err = 0;

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

	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,
1045
							  grec_nsrcs),
1046 1047 1048 1049 1050 1051
					   sizeof(_nsrcs), &_nsrcs);
		if (!nsrcs)
			return -EINVAL;

		if (!pskb_may_pull(skb,
				   len + sizeof(*grec) +
1052
				   sizeof(struct in6_addr) * ntohs(*nsrcs)))
1053 1054 1055
			return -EINVAL;

		grec = (struct mld2_grec *)(skb->data + len);
1056 1057
		len += sizeof(*grec) +
		       sizeof(struct in6_addr) * ntohs(*nsrcs);
1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072

		/* 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;
		}

1073 1074
		err = br_ip6_multicast_add_group(br, port, &grec->grec_mca,
						 vid);
1075 1076 1077 1078 1079 1080 1081 1082
		if (!err)
			break;
	}

	return err;
}
#endif

1083
static bool br_ip4_multicast_select_querier(struct net_bridge *br,
1084
					    struct net_bridge_port *port,
1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101
					    __be32 saddr)
{
	if (!timer_pending(&br->ip4_own_query.timer) &&
	    !timer_pending(&br->ip4_other_query.timer))
		goto update;

	if (!br->ip4_querier.addr.u.ip4)
		goto update;

	if (ntohl(saddr) <= ntohl(br->ip4_querier.addr.u.ip4))
		goto update;

	return false;

update:
	br->ip4_querier.addr.u.ip4 = saddr;

1102 1103 1104
	/* update protected by general multicast_lock by caller */
	rcu_assign_pointer(br->ip4_querier.port, port);

1105 1106 1107 1108 1109
	return true;
}

#if IS_ENABLED(CONFIG_IPV6)
static bool br_ip6_multicast_select_querier(struct net_bridge *br,
1110
					    struct net_bridge_port *port,
1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124
					    struct in6_addr *saddr)
{
	if (!timer_pending(&br->ip6_own_query.timer) &&
	    !timer_pending(&br->ip6_other_query.timer))
		goto update;

	if (ipv6_addr_cmp(saddr, &br->ip6_querier.addr.u.ip6) <= 0)
		goto update;

	return false;

update:
	br->ip6_querier.addr.u.ip6 = *saddr;

1125 1126 1127
	/* update protected by general multicast_lock by caller */
	rcu_assign_pointer(br->ip6_querier.port, port);

1128 1129 1130 1131 1132
	return true;
}
#endif

static bool br_multicast_select_querier(struct net_bridge *br,
1133
					struct net_bridge_port *port,
1134 1135 1136 1137
					struct br_ip *saddr)
{
	switch (saddr->proto) {
	case htons(ETH_P_IP):
1138
		return br_ip4_multicast_select_querier(br, port, saddr->u.ip4);
1139 1140
#if IS_ENABLED(CONFIG_IPV6)
	case htons(ETH_P_IPV6):
1141
		return br_ip6_multicast_select_querier(br, port, &saddr->u.ip6);
1142 1143 1144 1145 1146 1147
#endif
	}

	return false;
}

1148
static void
1149 1150 1151
br_multicast_update_query_timer(struct net_bridge *br,
				struct bridge_mcast_other_query *query,
				unsigned long max_delay)
1152
{
1153 1154
	if (!timer_pending(&query->timer))
		query->delay_time = jiffies + max_delay;
1155

1156
	mod_timer(&query->timer, jiffies + br->multicast_querier_interval);
1157 1158
}

1159
/*
C
Cong Wang 已提交
1160
 * Add port to router_list
1161 1162 1163
 *  list is maintained ordered by pointer value
 *  and locked by br->multicast_lock and RCU
 */
1164 1165 1166
static void br_multicast_add_router(struct net_bridge *br,
				    struct net_bridge_port *port)
{
1167
	struct net_bridge_port *p;
1168
	struct hlist_node *slot = NULL;
1169

1170
	hlist_for_each_entry(p, &br->router_list, rlist) {
1171 1172
		if ((unsigned long) port >= (unsigned long) p)
			break;
1173
		slot = &p->rlist;
1174 1175
	}

1176
	if (slot)
1177
		hlist_add_behind_rcu(&port->rlist, slot);
1178 1179
	else
		hlist_add_head_rcu(&port->rlist, &br->router_list);
1180 1181
}

1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199
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;

1200
	br_multicast_add_router(br, port);
1201 1202 1203 1204 1205 1206 1207 1208

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,
1209
					struct bridge_mcast_other_query *query,
1210
					struct br_ip *saddr,
1211
					unsigned long max_delay)
1212
{
1213
	if (!br_multicast_select_querier(br, port, saddr))
1214 1215
		return;

1216
	br_multicast_update_query_timer(br, query, max_delay);
1217 1218 1219
	br_multicast_mark_router(br, port);
}

1220 1221
static int br_ip4_multicast_query(struct net_bridge *br,
				  struct net_bridge_port *port,
1222 1223
				  struct sk_buff *skb,
				  u16 vid)
1224
{
1225
	const struct iphdr *iph = ip_hdr(skb);
1226 1227 1228 1229
	struct igmphdr *ih = igmp_hdr(skb);
	struct net_bridge_mdb_entry *mp;
	struct igmpv3_query *ih3;
	struct net_bridge_port_group *p;
1230
	struct net_bridge_port_group __rcu **pp;
1231
	struct br_ip saddr;
1232 1233 1234
	unsigned long max_delay;
	unsigned long now = jiffies;
	__be32 group;
1235
	int err = 0;
1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251

	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 {
1252 1253 1254 1255
		if (!pskb_may_pull(skb, sizeof(struct igmpv3_query))) {
			err = -EINVAL;
			goto out;
		}
1256 1257 1258

		ih3 = igmpv3_query_hdr(skb);
		if (ih3->nsrcs)
1259
			goto out;
1260

1261 1262
		max_delay = ih3->code ?
			    IGMPV3_MRC(ih3->code) * (HZ / IGMP_TIMER_SCALE) : 1;
1263 1264
	}

1265 1266 1267 1268 1269 1270 1271 1272
	/* RFC2236+RFC3376 (IGMPv2+IGMPv3) require the multicast link layer
	 * all-systems destination addresses (224.0.0.1) for general queries
	 */
	if (!group && iph->daddr != htonl(INADDR_ALLHOSTS_GROUP)) {
		err = -EINVAL;
		goto out;
	}

1273 1274 1275
	if (!group) {
		saddr.proto = htons(ETH_P_IP);
		saddr.u.ip4 = iph->saddr;
1276

1277 1278
		br_multicast_query_received(br, port, &br->ip4_other_query,
					    &saddr, max_delay);
1279
		goto out;
1280
	}
1281

1282
	mp = br_mdb_ip4_get(mlock_dereference(br->mdb, br), group, vid);
1283 1284 1285 1286 1287
	if (!mp)
		goto out;

	max_delay *= br->multicast_last_member_count;

1288
	if (mp->mglist &&
1289 1290 1291 1292 1293
	    (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);

1294 1295 1296
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
1297 1298 1299
		if (timer_pending(&p->timer) ?
		    time_after(p->timer.expires, now + max_delay) :
		    try_to_del_timer_sync(&p->timer) >= 0)
1300
			mod_timer(&p->timer, now + max_delay);
1301 1302 1303 1304
	}

out:
	spin_unlock(&br->multicast_lock);
1305
	return err;
1306 1307
}

E
Eric Dumazet 已提交
1308
#if IS_ENABLED(CONFIG_IPV6)
1309 1310
static int br_ip6_multicast_query(struct net_bridge *br,
				  struct net_bridge_port *port,
1311 1312
				  struct sk_buff *skb,
				  u16 vid)
1313
{
1314
	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
1315
	struct mld_msg *mld;
1316 1317
	struct net_bridge_mdb_entry *mp;
	struct mld2_query *mld2q;
1318 1319
	struct net_bridge_port_group *p;
	struct net_bridge_port_group __rcu **pp;
1320
	struct br_ip saddr;
1321 1322
	unsigned long max_delay;
	unsigned long now = jiffies;
1323
	const struct in6_addr *group = NULL;
1324
	bool is_general_query;
1325 1326 1327 1328 1329 1330 1331
	int err = 0;

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

1332 1333 1334 1335 1336 1337
	/* RFC2710+RFC3810 (MLDv1+MLDv2) require link-local source addresses */
	if (!(ipv6_addr_type(&ip6h->saddr) & IPV6_ADDR_LINKLOCAL)) {
		err = -EINVAL;
		goto out;
	}

1338 1339 1340 1341 1342 1343
	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
1344
		max_delay = msecs_to_jiffies(ntohs(mld->mld_maxdelay));
1345 1346
		if (max_delay)
			group = &mld->mld_mca;
1347
	} else {
1348 1349 1350 1351 1352 1353 1354
		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;
1355 1356

		max_delay = max(msecs_to_jiffies(mldv2_mrc(mld2q)), 1UL);
1357 1358
	}

1359 1360 1361 1362 1363 1364 1365 1366 1367 1368
	is_general_query = group && ipv6_addr_any(group);

	/* RFC2710+RFC3810 (MLDv1+MLDv2) require the multicast link layer
	 * all-nodes destination address (ff02::1) for general queries
	 */
	if (is_general_query && !ipv6_addr_is_ll_all_nodes(&ip6h->daddr)) {
		err = -EINVAL;
		goto out;
	}

1369 1370 1371
	if (is_general_query) {
		saddr.proto = htons(ETH_P_IPV6);
		saddr.u.ip6 = ip6h->saddr;
1372

1373 1374
		br_multicast_query_received(br, port, &br->ip6_other_query,
					    &saddr, max_delay);
1375
		goto out;
1376 1377
	} else if (!group) {
		goto out;
1378
	}
1379

1380
	mp = br_mdb_ip6_get(mlock_dereference(br->mdb, br), group, vid);
1381 1382 1383 1384
	if (!mp)
		goto out;

	max_delay *= br->multicast_last_member_count;
1385
	if (mp->mglist &&
1386 1387 1388 1389 1390
	    (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);

1391 1392 1393
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
1394 1395 1396
		if (timer_pending(&p->timer) ?
		    time_after(p->timer.expires, now + max_delay) :
		    try_to_del_timer_sync(&p->timer) >= 0)
1397
			mod_timer(&p->timer, now + max_delay);
1398 1399 1400 1401 1402 1403 1404 1405
	}

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

1406 1407 1408 1409 1410 1411
static void
br_multicast_leave_group(struct net_bridge *br,
			 struct net_bridge_port *port,
			 struct br_ip *group,
			 struct bridge_mcast_other_query *other_query,
			 struct bridge_mcast_own_query *own_query)
1412 1413 1414 1415 1416 1417 1418 1419 1420 1421
{
	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) ||
1422
	    timer_pending(&other_query->timer))
1423 1424
		goto out;

1425
	mdb = mlock_dereference(br->mdb, br);
1426 1427 1428 1429
	mp = br_mdb_ip_get(mdb, group);
	if (!mp)
		goto out;

1430
	if (br->multicast_querier) {
1431 1432 1433 1434
		__br_multicast_send_query(br, port, &mp->addr);

		time = jiffies + br->multicast_last_member_count *
				 br->multicast_last_member_interval;
1435

1436
		mod_timer(&own_query->timer, time);
1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454

		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;
		}
	}

1455
	if (port && (port->flags & BR_MULTICAST_FAST_LEAVE)) {
1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467
		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
1468
			br_mdb_notify(br->dev, port, group, RTM_DELMDB);
1469

1470
			if (!mp->ports && !mp->mglist &&
1471 1472 1473 1474 1475 1476
			    netif_running(br->dev))
				mod_timer(&mp->timer, jiffies);
		}
		goto out;
	}

1477 1478 1479 1480 1481
	now = jiffies;
	time = now + br->multicast_last_member_count *
		     br->multicast_last_member_interval;

	if (!port) {
1482
		if (mp->mglist &&
1483 1484 1485 1486 1487
		    (timer_pending(&mp->timer) ?
		     time_after(mp->timer.expires, time) :
		     try_to_del_timer_sync(&mp->timer) >= 0)) {
			mod_timer(&mp->timer, time);
		}
1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505

		goto out;
	}

	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;
1506 1507 1508 1509 1510
	}
out:
	spin_unlock(&br->multicast_lock);
}

1511 1512
static void br_ip4_multicast_leave_group(struct net_bridge *br,
					 struct net_bridge_port *port,
1513 1514
					 __be32 group,
					 __u16 vid)
1515 1516
{
	struct br_ip br_group;
1517
	struct bridge_mcast_own_query *own_query;
1518 1519 1520 1521

	if (ipv4_is_local_multicast(group))
		return;

1522 1523
	own_query = port ? &port->ip4_own_query : &br->ip4_own_query;

1524 1525
	br_group.u.ip4 = group;
	br_group.proto = htons(ETH_P_IP);
1526
	br_group.vid = vid;
1527

1528 1529
	br_multicast_leave_group(br, port, &br_group, &br->ip4_other_query,
				 own_query);
1530 1531
}

E
Eric Dumazet 已提交
1532
#if IS_ENABLED(CONFIG_IPV6)
1533 1534
static void br_ip6_multicast_leave_group(struct net_bridge *br,
					 struct net_bridge_port *port,
1535 1536
					 const struct in6_addr *group,
					 __u16 vid)
1537 1538
{
	struct br_ip br_group;
1539
	struct bridge_mcast_own_query *own_query;
1540

1541
	if (ipv6_addr_is_ll_all_nodes(group))
1542 1543
		return;

1544 1545
	own_query = port ? &port->ip6_own_query : &br->ip6_own_query;

A
Alexey Dobriyan 已提交
1546
	br_group.u.ip6 = *group;
1547
	br_group.proto = htons(ETH_P_IPV6);
1548
	br_group.vid = vid;
1549

1550 1551
	br_multicast_leave_group(br, port, &br_group, &br->ip6_other_query,
				 own_query);
1552 1553
}
#endif
1554

1555 1556
static int br_multicast_ipv4_rcv(struct net_bridge *br,
				 struct net_bridge_port *port,
1557 1558
				 struct sk_buff *skb,
				 u16 vid)
1559 1560
{
	struct sk_buff *skb2 = skb;
1561
	const struct iphdr *iph;
1562
	struct igmphdr *ih;
1563 1564
	unsigned int len;
	unsigned int offset;
1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583
	int err;

	/* 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;

1584
	if (iph->protocol != IPPROTO_IGMP) {
1585
		if (!ipv4_is_local_multicast(iph->daddr))
1586
			BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1587
		return 0;
1588
	}
1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600

	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)
1601
			goto err_out;
1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620
	}

	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))
1621
			goto out;
1622 1623 1624 1625 1626 1627 1628 1629 1630 1631
	}

	err = 0;

	BR_INPUT_SKB_CB(skb)->igmp = 1;
	ih = igmp_hdr(skb2);

	switch (ih->type) {
	case IGMP_HOST_MEMBERSHIP_REPORT:
	case IGMPV2_HOST_MEMBERSHIP_REPORT:
1632
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1633
		err = br_ip4_multicast_add_group(br, port, ih->group, vid);
1634 1635
		break;
	case IGMPV3_HOST_MEMBERSHIP_REPORT:
1636
		err = br_ip4_multicast_igmp3_report(br, port, skb2, vid);
1637 1638
		break;
	case IGMP_HOST_MEMBERSHIP_QUERY:
1639
		err = br_ip4_multicast_query(br, port, skb2, vid);
1640 1641
		break;
	case IGMP_HOST_LEAVE_MESSAGE:
1642
		br_ip4_multicast_leave_group(br, port, ih->group, vid);
1643 1644 1645 1646 1647
		break;
	}

out:
	__skb_push(skb2, offset);
1648
err_out:
1649 1650 1651 1652 1653
	if (skb2 != skb)
		kfree_skb(skb2);
	return err;
}

E
Eric Dumazet 已提交
1654
#if IS_ENABLED(CONFIG_IPV6)
1655 1656
static int br_multicast_ipv6_rcv(struct net_bridge *br,
				 struct net_bridge_port *port,
1657 1658
				 struct sk_buff *skb,
				 u16 vid)
1659
{
1660
	struct sk_buff *skb2;
1661
	const struct ipv6hdr *ip6h;
1662
	u8 icmp6_type;
1663
	u8 nexthdr;
1664
	__be16 frag_off;
1665
	unsigned int len;
K
Kulikov Vasiliy 已提交
1666
	int offset;
1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679
	int err;

	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.
	 */
1680 1681 1682 1683
	if (ip6h->version != 6)
		return 0;

	/* Prevent flooding this packet if there is no listener present */
1684
	if (!ipv6_addr_is_ll_all_nodes(&ip6h->daddr))
1685 1686 1687
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;

	if (ip6h->nexthdr != IPPROTO_HOPOPTS ||
1688 1689 1690
	    ip6h->payload_len == 0)
		return 0;

1691
	len = ntohs(ip6h->payload_len) + sizeof(*ip6h);
1692 1693 1694 1695
	if (skb->len < len)
		return -EINVAL;

	nexthdr = ip6h->nexthdr;
1696
	offset = ipv6_skip_exthdr(skb, sizeof(*ip6h), &nexthdr, &frag_off);
1697 1698 1699 1700 1701 1702 1703 1704 1705

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

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

1706 1707 1708 1709
	err = -EINVAL;
	if (!pskb_may_pull(skb2, offset + sizeof(struct icmp6hdr)))
		goto out;

1710 1711 1712 1713
	len -= offset - skb_network_offset(skb2);

	__skb_pull(skb2, offset);
	skb_reset_transport_header(skb2);
1714 1715
	skb_postpull_rcsum(skb2, skb_network_header(skb2),
			   skb_network_header_len(skb2));
1716

1717
	icmp6_type = icmp6_hdr(skb2)->icmp6_type;
1718

1719
	switch (icmp6_type) {
1720 1721 1722 1723 1724 1725 1726 1727 1728 1729 1730 1731 1732 1733 1734
	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;
1735
		err = -EINVAL;
1736 1737
	}

1738 1739
	ip6h = ipv6_hdr(skb2);

1740 1741
	switch (skb2->ip_summed) {
	case CHECKSUM_COMPLETE:
1742 1743
		if (!csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, skb2->len,
					IPPROTO_ICMPV6, skb2->csum))
1744 1745 1746
			break;
		/*FALLTHROUGH*/
	case CHECKSUM_NONE:
1747 1748 1749 1750 1751
		skb2->csum = ~csum_unfold(csum_ipv6_magic(&ip6h->saddr,
							&ip6h->daddr,
							skb2->len,
							IPPROTO_ICMPV6, 0));
		if (__skb_checksum_complete(skb2))
1752 1753 1754 1755 1756 1757 1758
			goto out;
	}

	err = 0;

	BR_INPUT_SKB_CB(skb)->igmp = 1;

1759
	switch (icmp6_type) {
1760 1761
	case ICMPV6_MGM_REPORT:
	    {
1762 1763 1764 1765 1766 1767
		struct mld_msg *mld;
		if (!pskb_may_pull(skb2, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *)skb_transport_header(skb2);
1768
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1769
		err = br_ip6_multicast_add_group(br, port, &mld->mld_mca, vid);
1770 1771 1772
		break;
	    }
	case ICMPV6_MLD2_REPORT:
1773
		err = br_ip6_multicast_mld2_report(br, port, skb2, vid);
1774 1775
		break;
	case ICMPV6_MGM_QUERY:
1776
		err = br_ip6_multicast_query(br, port, skb2, vid);
1777 1778 1779
		break;
	case ICMPV6_MGM_REDUCTION:
	    {
1780 1781 1782 1783 1784 1785
		struct mld_msg *mld;
		if (!pskb_may_pull(skb2, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *)skb_transport_header(skb2);
1786
		br_ip6_multicast_leave_group(br, port, &mld->mld_mca, vid);
1787 1788 1789 1790
	    }
	}

out:
1791
	kfree_skb(skb2);
1792 1793 1794 1795
	return err;
}
#endif

1796
int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port,
1797
		     struct sk_buff *skb, u16 vid)
1798
{
1799 1800 1801
	BR_INPUT_SKB_CB(skb)->igmp = 0;
	BR_INPUT_SKB_CB(skb)->mrouters_only = 0;

1802 1803 1804 1805 1806
	if (br->multicast_disabled)
		return 0;

	switch (skb->protocol) {
	case htons(ETH_P_IP):
1807
		return br_multicast_ipv4_rcv(br, port, skb, vid);
E
Eric Dumazet 已提交
1808
#if IS_ENABLED(CONFIG_IPV6)
1809
	case htons(ETH_P_IPV6):
1810
		return br_multicast_ipv6_rcv(br, port, skb, vid);
1811
#endif
1812 1813 1814 1815 1816
	}

	return 0;
}

1817
static void br_multicast_query_expired(struct net_bridge *br,
1818 1819
				       struct bridge_mcast_own_query *query,
				       struct bridge_mcast_querier *querier)
1820 1821 1822 1823 1824
{
	spin_lock(&br->multicast_lock);
	if (query->startup_sent < br->multicast_startup_query_count)
		query->startup_sent++;

1825
	rcu_assign_pointer(querier, NULL);
1826 1827 1828 1829 1830
	br_multicast_send_query(br, NULL, query);
	spin_unlock(&br->multicast_lock);
}

static void br_ip4_multicast_query_expired(unsigned long data)
1831 1832 1833
{
	struct net_bridge *br = (void *)data;

1834
	br_multicast_query_expired(br, &br->ip4_own_query, &br->ip4_querier);
1835
}
1836

1837 1838 1839 1840
#if IS_ENABLED(CONFIG_IPV6)
static void br_ip6_multicast_query_expired(unsigned long data)
{
	struct net_bridge *br = (void *)data;
1841

1842
	br_multicast_query_expired(br, &br->ip6_own_query, &br->ip6_querier);
1843
}
1844
#endif
1845 1846 1847 1848 1849 1850 1851

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

	br->multicast_router = 1;
1852
	br->multicast_querier = 0;
1853
	br->multicast_query_use_ifaddr = 0;
1854 1855 1856 1857 1858 1859 1860 1861 1862 1863
	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;

1864
	br->ip4_other_query.delay_time = 0;
1865
	br->ip4_querier.port = NULL;
1866
#if IS_ENABLED(CONFIG_IPV6)
1867
	br->ip6_other_query.delay_time = 0;
1868
	br->ip6_querier.port = NULL;
1869
#endif
1870

1871 1872 1873
	spin_lock_init(&br->multicast_lock);
	setup_timer(&br->multicast_router_timer,
		    br_multicast_local_router_expired, 0);
1874 1875 1876
	setup_timer(&br->ip4_other_query.timer,
		    br_ip4_multicast_querier_expired, (unsigned long)br);
	setup_timer(&br->ip4_own_query.timer, br_ip4_multicast_query_expired,
1877
		    (unsigned long)br);
1878
#if IS_ENABLED(CONFIG_IPV6)
1879 1880 1881
	setup_timer(&br->ip6_other_query.timer,
		    br_ip6_multicast_querier_expired, (unsigned long)br);
	setup_timer(&br->ip6_own_query.timer, br_ip6_multicast_query_expired,
1882 1883
		    (unsigned long)br);
#endif
1884 1885
}

1886
static void __br_multicast_open(struct net_bridge *br,
1887
				struct bridge_mcast_own_query *query)
1888
{
1889
	query->startup_sent = 0;
1890 1891 1892 1893

	if (br->multicast_disabled)
		return;

1894 1895 1896 1897 1898
	mod_timer(&query->timer, jiffies);
}

void br_multicast_open(struct net_bridge *br)
{
1899
	__br_multicast_open(br, &br->ip4_own_query);
1900
#if IS_ENABLED(CONFIG_IPV6)
1901
	__br_multicast_open(br, &br->ip6_own_query);
1902
#endif
1903 1904 1905 1906 1907 1908
}

void br_multicast_stop(struct net_bridge *br)
{
	struct net_bridge_mdb_htable *mdb;
	struct net_bridge_mdb_entry *mp;
1909
	struct hlist_node *n;
1910 1911 1912 1913
	u32 ver;
	int i;

	del_timer_sync(&br->multicast_router_timer);
1914 1915
	del_timer_sync(&br->ip4_other_query.timer);
	del_timer_sync(&br->ip4_own_query.timer);
1916
#if IS_ENABLED(CONFIG_IPV6)
1917 1918
	del_timer_sync(&br->ip6_other_query.timer);
	del_timer_sync(&br->ip6_own_query.timer);
1919
#endif
1920 1921

	spin_lock_bh(&br->multicast_lock);
1922
	mdb = mlock_dereference(br->mdb, br);
1923 1924 1925 1926 1927 1928 1929
	if (!mdb)
		goto out;

	br->mdb = NULL;

	ver = mdb->ver;
	for (i = 0; i < mdb->max; i++) {
1930
		hlist_for_each_entry_safe(mp, n, &mdb->mhash[i],
1931 1932 1933 1934 1935 1936 1937 1938
					  hlist[ver]) {
			del_timer(&mp->timer);
			call_rcu_bh(&mp->rcu, br_multicast_free_group);
		}
	}

	if (mdb->old) {
		spin_unlock_bh(&br->multicast_lock);
1939
		rcu_barrier_bh();
1940 1941 1942 1943 1944 1945 1946 1947 1948 1949
		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);
}
1950 1951 1952 1953 1954 1955 1956 1957 1958 1959 1960 1961 1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019

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;
}
2020

2021
static void br_multicast_start_querier(struct net_bridge *br,
2022
				       struct bridge_mcast_own_query *query)
2023 2024
{
	struct net_bridge_port *port;
2025

2026
	__br_multicast_open(br, query);
2027 2028 2029 2030 2031 2032

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

2033 2034
		if (query == &br->ip4_own_query)
			br_multicast_enable(&port->ip4_own_query);
2035 2036
#if IS_ENABLED(CONFIG_IPV6)
		else
2037
			br_multicast_enable(&port->ip6_own_query);
2038
#endif
2039 2040 2041 2042 2043
	}
}

int br_multicast_toggle(struct net_bridge *br, unsigned long val)
{
2044
	int err = 0;
2045
	struct net_bridge_mdb_htable *mdb;
2046

2047
	spin_lock_bh(&br->multicast_lock);
2048 2049 2050 2051 2052 2053 2054
	if (br->multicast_disabled == !val)
		goto unlock;

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

2055 2056 2057
	if (!netif_running(br->dev))
		goto unlock;

2058 2059 2060
	mdb = mlock_dereference(br->mdb, br);
	if (mdb) {
		if (mdb->old) {
2061 2062 2063 2064 2065 2066
			err = -EEXIST;
rollback:
			br->multicast_disabled = !!val;
			goto unlock;
		}

2067
		err = br_mdb_rehash(&br->mdb, mdb->max,
2068 2069 2070 2071 2072
				    br->hash_elasticity);
		if (err)
			goto rollback;
	}

2073
	br_multicast_start_querier(br, &br->ip4_own_query);
2074
#if IS_ENABLED(CONFIG_IPV6)
2075
	br_multicast_start_querier(br, &br->ip6_own_query);
2076
#endif
2077 2078

unlock:
2079
	spin_unlock_bh(&br->multicast_lock);
2080 2081 2082

	return err;
}
2083

2084 2085
int br_multicast_set_querier(struct net_bridge *br, unsigned long val)
{
2086 2087
	unsigned long max_delay;

2088 2089 2090 2091 2092 2093 2094
	val = !!val;

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

	br->multicast_querier = val;
2095 2096 2097 2098 2099
	if (!val)
		goto unlock;

	max_delay = br->multicast_query_response_interval;

2100 2101
	if (!timer_pending(&br->ip4_other_query.timer))
		br->ip4_other_query.delay_time = jiffies + max_delay;
2102

2103
	br_multicast_start_querier(br, &br->ip4_own_query);
2104 2105

#if IS_ENABLED(CONFIG_IPV6)
2106 2107
	if (!timer_pending(&br->ip6_other_query.timer))
		br->ip6_other_query.delay_time = jiffies + max_delay;
2108

2109
	br_multicast_start_querier(br, &br->ip6_own_query);
2110
#endif
2111 2112 2113 2114 2115 2116 2117

unlock:
	spin_unlock_bh(&br->multicast_lock);

	return 0;
}

2118 2119 2120 2121
int br_multicast_set_hash_max(struct net_bridge *br, unsigned long val)
{
	int err = -ENOENT;
	u32 old;
2122
	struct net_bridge_mdb_htable *mdb;
2123

2124
	spin_lock_bh(&br->multicast_lock);
2125 2126 2127 2128 2129 2130
	if (!netif_running(br->dev))
		goto unlock;

	err = -EINVAL;
	if (!is_power_of_2(val))
		goto unlock;
2131 2132 2133

	mdb = mlock_dereference(br->mdb, br);
	if (mdb && val < mdb->size)
2134 2135 2136 2137 2138 2139 2140
		goto unlock;

	err = 0;

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

2141 2142
	if (mdb) {
		if (mdb->old) {
2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155
			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:
2156
	spin_unlock_bh(&br->multicast_lock);
2157 2158 2159

	return err;
}
2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211 2212 2213 2214 2215 2216

/**
 * br_multicast_list_adjacent - Returns snooped multicast addresses
 * @dev:	The bridge port adjacent to which to retrieve addresses
 * @br_ip_list:	The list to store found, snooped multicast IP addresses in
 *
 * Creates a list of IP addresses (struct br_ip_list) sensed by the multicast
 * snooping feature on all bridge ports of dev's bridge device, excluding
 * the addresses from dev itself.
 *
 * Returns the number of items added to br_ip_list.
 *
 * Notes:
 * - br_ip_list needs to be initialized by caller
 * - br_ip_list might contain duplicates in the end
 *   (needs to be taken care of by caller)
 * - br_ip_list needs to be freed by caller
 */
int br_multicast_list_adjacent(struct net_device *dev,
			       struct list_head *br_ip_list)
{
	struct net_bridge *br;
	struct net_bridge_port *port;
	struct net_bridge_port_group *group;
	struct br_ip_list *entry;
	int count = 0;

	rcu_read_lock();
	if (!br_ip_list || !br_port_exists(dev))
		goto unlock;

	port = br_port_get_rcu(dev);
	if (!port || !port->br)
		goto unlock;

	br = port->br;

	list_for_each_entry_rcu(port, &br->port_list, list) {
		if (!port->dev || port->dev == dev)
			continue;

		hlist_for_each_entry_rcu(group, &port->mglist, mglist) {
			entry = kmalloc(sizeof(*entry), GFP_ATOMIC);
			if (!entry)
				goto unlock;

			entry->addr = group->addr;
			list_add(&entry->list, br_ip_list);
			count++;
		}
	}

unlock:
	rcu_read_unlock();
	return count;
}
EXPORT_SYMBOL_GPL(br_multicast_list_adjacent);
2217

2218 2219 2220 2221 2222 2223 2224 2225 2226 2227 2228 2229 2230 2231 2232 2233 2234 2235 2236 2237 2238 2239 2240 2241 2242 2243 2244 2245 2246 2247 2248 2249 2250 2251 2252 2253 2254
/**
 * br_multicast_has_querier_anywhere - Checks for a querier on a bridge
 * @dev: The bridge port providing the bridge on which to check for a querier
 * @proto: The protocol family to check for: IGMP -> ETH_P_IP, MLD -> ETH_P_IPV6
 *
 * Checks whether the given interface has a bridge on top and if so returns
 * true if a valid querier exists anywhere on the bridged link layer.
 * Otherwise returns false.
 */
bool br_multicast_has_querier_anywhere(struct net_device *dev, int proto)
{
	struct net_bridge *br;
	struct net_bridge_port *port;
	struct ethhdr eth;
	bool ret = false;

	rcu_read_lock();
	if (!br_port_exists(dev))
		goto unlock;

	port = br_port_get_rcu(dev);
	if (!port || !port->br)
		goto unlock;

	br = port->br;

	memset(&eth, 0, sizeof(eth));
	eth.h_proto = htons(proto);

	ret = br_multicast_querier_exists(br, &eth);

unlock:
	rcu_read_unlock();
	return ret;
}
EXPORT_SYMBOL_GPL(br_multicast_has_querier_anywhere);

2255 2256 2257 2258 2259 2260 2261 2262 2263 2264 2265 2266 2267 2268 2269 2270 2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285
/**
 * br_multicast_has_querier_adjacent - Checks for a querier behind a bridge port
 * @dev: The bridge port adjacent to which to check for a querier
 * @proto: The protocol family to check for: IGMP -> ETH_P_IP, MLD -> ETH_P_IPV6
 *
 * Checks whether the given interface has a bridge on top and if so returns
 * true if a selected querier is behind one of the other ports of this
 * bridge. Otherwise returns false.
 */
bool br_multicast_has_querier_adjacent(struct net_device *dev, int proto)
{
	struct net_bridge *br;
	struct net_bridge_port *port;
	bool ret = false;

	rcu_read_lock();
	if (!br_port_exists(dev))
		goto unlock;

	port = br_port_get_rcu(dev);
	if (!port || !port->br)
		goto unlock;

	br = port->br;

	switch (proto) {
	case ETH_P_IP:
		if (!timer_pending(&br->ip4_other_query.timer) ||
		    rcu_dereference(br->ip4_querier.port) == port)
			goto unlock;
		break;
2286
#if IS_ENABLED(CONFIG_IPV6)
2287 2288 2289 2290 2291
	case ETH_P_IPV6:
		if (!timer_pending(&br->ip6_other_query.timer) ||
		    rcu_dereference(br->ip6_querier.port) == port)
			goto unlock;
		break;
2292
#endif
2293 2294 2295 2296 2297 2298 2299 2300 2301 2302
	default:
		goto unlock;
	}

	ret = true;
unlock:
	rcu_read_unlock();
	return ret;
}
EXPORT_SYMBOL_GPL(br_multicast_has_querier_adjacent);