br_multicast.c 44.1 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 37
static void br_multicast_start_querier(struct net_bridge *br,
				       struct bridge_mcast_query *query);
C
Cong Wang 已提交
38
unsigned int br_mdb_rehash_seq;
39

40 41 42 43
static inline int br_ip_equal(const struct br_ip *a, const struct br_ip *b)
{
	if (a->proto != b->proto)
		return 0;
44 45
	if (a->vid != b->vid)
		return 0;
46 47 48
	switch (a->proto) {
	case htons(ETH_P_IP):
		return a->u.ip4 == b->u.ip4;
E
Eric Dumazet 已提交
49
#if IS_ENABLED(CONFIG_IPV6)
50 51 52
	case htons(ETH_P_IPV6):
		return ipv6_addr_equal(&a->u.ip6, &b->u.ip6);
#endif
53 54 55 56
	}
	return 0;
}

57 58
static inline int __br_ip4_hash(struct net_bridge_mdb_htable *mdb, __be32 ip,
				__u16 vid)
59
{
60
	return jhash_2words((__force u32)ip, vid, mdb->secret) & (mdb->max - 1);
61 62
}

E
Eric Dumazet 已提交
63
#if IS_ENABLED(CONFIG_IPV6)
64
static inline int __br_ip6_hash(struct net_bridge_mdb_htable *mdb,
65 66
				const struct in6_addr *ip,
				__u16 vid)
67
{
68 69
	return jhash_2words(ipv6_addr_hash(ip), vid,
			    mdb->secret) & (mdb->max - 1);
70 71 72
}
#endif

73 74 75 76 77
static inline int br_ip_hash(struct net_bridge_mdb_htable *mdb,
			     struct br_ip *ip)
{
	switch (ip->proto) {
	case htons(ETH_P_IP):
78
		return __br_ip4_hash(mdb, ip->u.ip4, ip->vid);
E
Eric Dumazet 已提交
79
#if IS_ENABLED(CONFIG_IPV6)
80
	case htons(ETH_P_IPV6):
81
		return __br_ip6_hash(mdb, &ip->u.ip6, ip->vid);
82
#endif
83 84
	}
	return 0;
85 86 87
}

static struct net_bridge_mdb_entry *__br_mdb_ip_get(
88
	struct net_bridge_mdb_htable *mdb, struct br_ip *dst, int hash)
89 90 91
{
	struct net_bridge_mdb_entry *mp;

92
	hlist_for_each_entry_rcu(mp, &mdb->mhash[hash], hlist[mdb->ver]) {
93
		if (br_ip_equal(&mp->addr, dst))
94 95 96 97 98 99
			return mp;
	}

	return NULL;
}

100 101
struct net_bridge_mdb_entry *br_mdb_ip_get(struct net_bridge_mdb_htable *mdb,
					   struct br_ip *dst)
102 103 104 105 106 107 108
{
	if (!mdb)
		return NULL;

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

109
static struct net_bridge_mdb_entry *br_mdb_ip4_get(
110
	struct net_bridge_mdb_htable *mdb, __be32 dst, __u16 vid)
111
{
112 113 114 115
	struct br_ip br_dst;

	br_dst.u.ip4 = dst;
	br_dst.proto = htons(ETH_P_IP);
116
	br_dst.vid = vid;
117

118
	return br_mdb_ip_get(mdb, &br_dst);
119 120
}

E
Eric Dumazet 已提交
121
#if IS_ENABLED(CONFIG_IPV6)
122
static struct net_bridge_mdb_entry *br_mdb_ip6_get(
123 124
	struct net_bridge_mdb_htable *mdb, const struct in6_addr *dst,
	__u16 vid)
125 126
{
	struct br_ip br_dst;
127

A
Alexey Dobriyan 已提交
128
	br_dst.u.ip6 = *dst;
129
	br_dst.proto = htons(ETH_P_IPV6);
130
	br_dst.vid = vid;
131

132
	return br_mdb_ip_get(mdb, &br_dst);
133 134 135
}
#endif

136
struct net_bridge_mdb_entry *br_mdb_get(struct net_bridge *br,
137
					struct sk_buff *skb, u16 vid)
138
{
139
	struct net_bridge_mdb_htable *mdb = rcu_dereference(br->mdb);
140 141
	struct br_ip ip;

142
	if (br->multicast_disabled)
143 144
		return NULL;

145
	if (BR_INPUT_SKB_CB(skb)->igmp)
146 147
		return NULL;

148
	ip.proto = skb->protocol;
149
	ip.vid = vid;
150

151 152
	switch (skb->protocol) {
	case htons(ETH_P_IP):
153 154
		ip.u.ip4 = ip_hdr(skb)->daddr;
		break;
E
Eric Dumazet 已提交
155
#if IS_ENABLED(CONFIG_IPV6)
156
	case htons(ETH_P_IPV6):
A
Alexey Dobriyan 已提交
157
		ip.u.ip6 = ipv6_hdr(skb)->daddr;
158 159
		break;
#endif
160 161
	default:
		return NULL;
162 163
	}

164
	return br_mdb_ip_get(mdb, &ip);
165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187
}

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++)
188
		hlist_for_each_entry(mp, &old->mhash[i], hlist[old->ver])
189
			hlist_add_head(&mp->hlist[new->ver],
190
				       &new->mhash[br_ip_hash(new, &mp->addr)]);
191 192 193 194 195 196 197

	if (!elasticity)
		return 0;

	maxlen = 0;
	for (i = 0; i < new->max; i++) {
		len = 0;
198
		hlist_for_each_entry(mp, &new->mhash[i], hlist[new->ver])
199 200 201 202 203 204 205 206
			len++;
		if (len > maxlen)
			maxlen = len;
	}

	return maxlen > elasticity ? -EINVAL : 0;
}

207
void br_multicast_free_pg(struct rcu_head *head)
208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
{
	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;

233
	mp->mglist = false;
234 235 236 237

	if (mp->ports)
		goto out;

238 239
	mdb = mlock_dereference(br->mdb, br);

240 241 242 243 244 245 246 247 248 249 250 251
	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)
{
252
	struct net_bridge_mdb_htable *mdb;
253 254
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
255 256 257
	struct net_bridge_port_group __rcu **pp;

	mdb = mlock_dereference(br->mdb, br);
258

259
	mp = br_mdb_ip_get(mdb, &pg->addr);
260 261 262
	if (WARN_ON(!mp))
		return;

263 264 265
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
266 267 268
		if (p != pg)
			continue;

269
		rcu_assign_pointer(*pp, p->next);
270 271 272 273
		hlist_del_init(&p->mglist);
		del_timer(&p->timer);
		call_rcu_bh(&p->rcu, br_multicast_free_pg);

274
		if (!mp->ports && !mp->mglist && mp->timer_armed &&
275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290
		    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) ||
291
	    hlist_unhashed(&pg->mglist) || pg->state & MDB_PERMANENT)
292 293 294 295 296 297 298 299
		goto out;

	br_multicast_del_pg(br, pg);

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

300
static int br_mdb_rehash(struct net_bridge_mdb_htable __rcu **mdbp, int max,
301 302
			 int elasticity)
{
303
	struct net_bridge_mdb_htable *old = rcu_dereference_protected(*mdbp, 1);
304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337
	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
338
	br_mdb_rehash_seq++;
339 340 341 342 343 344 345 346
	call_rcu_bh(&mdb->rcu, br_mdb_free);

out:
	rcu_assign_pointer(*mdbp, mdb);

	return 0;
}

347 348
static struct sk_buff *br_ip4_multicast_alloc_query(struct net_bridge *br,
						    __be32 group)
349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385
{
	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;
386 387
	iph->saddr = br->multicast_query_use_ifaddr ?
		     inet_select_addr(br->dev, 0, RT_SCOPE_LINK) : 0;
388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412
	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
413
#if IS_ENABLED(CONFIG_IPV6)
414
static struct sk_buff *br_ip6_multicast_alloc_query(struct net_bridge *br,
415
						    const struct in6_addr *group)
416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443
{
	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);
444
	ip6h->payload_len = htons(8 + sizeof(*mldq));
445 446
	ip6h->nexthdr = IPPROTO_HOPOPTS;
	ip6h->hop_limit = 1;
447
	ipv6_addr_set(&ip6h->daddr, htonl(0xff020000), 0, 0, htonl(1));
448 449 450 451 452
	if (ipv6_dev_get_saddr(dev_net(br->dev), br->dev, &ip6h->daddr, 0,
			       &ip6h->saddr)) {
		kfree_skb(skb);
		return NULL;
	}
453
	ipv6_eth_mc_map(&ip6h->daddr, eth->h_dest);
454 455 456 457 458 459 460 461

	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;
462 463
	hopopt[6] = IPV6_TLV_PAD1;		/* Pad1 */
	hopopt[7] = IPV6_TLV_PAD1;		/* Pad1 */
464 465 466 467 468 469 470

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

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

471 472 473
	interval = ipv6_addr_any(group) ?
			br->multicast_query_response_interval :
			br->multicast_last_member_interval;
474 475 476 477 478 479

	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
480
	mldq->mld_mca = *group;
481 482 483 484 485 486 487 488 489 490 491 492 493 494 495

	/* 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

496 497 498 499 500 501
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
502
#if IS_ENABLED(CONFIG_IPV6)
503 504 505
	case htons(ETH_P_IPV6):
		return br_ip6_multicast_alloc_query(br, &addr->u.ip6);
#endif
506 507 508 509
	}
	return NULL;
}

510
static struct net_bridge_mdb_entry *br_multicast_get_group(
511 512
	struct net_bridge *br, struct net_bridge_port *port,
	struct br_ip *group, int hash)
513
{
514
	struct net_bridge_mdb_htable *mdb;
515
	struct net_bridge_mdb_entry *mp;
516 517
	unsigned int count = 0;
	unsigned int max;
518 519 520
	int elasticity;
	int err;

521
	mdb = rcu_dereference_protected(br->mdb, 1);
522
	hlist_for_each_entry(mp, &mdb->mhash[hash], hlist[mdb->ver]) {
523
		count++;
524
		if (unlikely(br_ip_equal(group, &mp->addr)))
525 526 527 528 529 530 531 532
			return mp;
	}

	elasticity = 0;
	max = mdb->max;

	if (unlikely(count > br->hash_elasticity && count)) {
		if (net_ratelimit())
533 534 535
			br_info(br, "Multicast hash table "
				"chain limit reached: %s\n",
				port ? port->dev->name : br->dev->name);
536 537 538 539 540 541

		elasticity = br->hash_elasticity;
	}

	if (mdb->size >= max) {
		max *= 2;
542 543 544 545 546
		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);
547 548 549 550 551 552 553 554 555 556
			err = -E2BIG;
disable:
			br->multicast_disabled = 1;
			goto err;
		}
	}

	if (max > mdb->max || elasticity) {
		if (mdb->old) {
			if (net_ratelimit())
557 558 559
				br_info(br, "Multicast hash table "
					"on fire: %s\n",
					port ? port->dev->name : br->dev->name);
560 561 562 563 564 565
			err = -EEXIST;
			goto err;
		}

		err = br_mdb_rehash(&br->mdb, max, elasticity);
		if (err) {
566 567 568 569
			br_warn(br, "Cannot rehash multicast "
				"hash table, disabling snooping: %s, %d, %d\n",
				port ? port->dev->name : br->dev->name,
				mdb->size, err);
570 571 572 573 574 575 576 577 578 579 580 581 582 583
			goto disable;
		}

		err = -EAGAIN;
		goto err;
	}

	return NULL;

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

584 585
struct net_bridge_mdb_entry *br_multicast_new_group(struct net_bridge *br,
	struct net_bridge_port *port, struct br_ip *group)
586
{
587
	struct net_bridge_mdb_htable *mdb;
588 589
	struct net_bridge_mdb_entry *mp;
	int hash;
590
	int err;
591

592
	mdb = rcu_dereference_protected(br->mdb, 1);
593
	if (!mdb) {
594 595 596
		err = br_mdb_rehash(&br->mdb, BR_HASH_SIZE, 0);
		if (err)
			return ERR_PTR(err);
597 598 599 600 601 602 603 604 605 606 607
		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:
608
		mdb = rcu_dereference_protected(br->mdb, 1);
609 610 611 612 613 614 615 616 617
		hash = br_ip_hash(mdb, group);
		break;

	default:
		goto out;
	}

	mp = kzalloc(sizeof(*mp), GFP_ATOMIC);
	if (unlikely(!mp))
618
		return ERR_PTR(-ENOMEM);
619 620

	mp->br = br;
621
	mp->addr = *group;
622

623 624 625
	setup_timer(&mp->timer, br_multicast_group_expired,
		    (unsigned long)mp);

626 627 628 629 630 631 632
	hlist_add_head_rcu(&mp->hlist[mdb->ver], &mdb->mhash[hash]);
	mdb->size++;

out:
	return mp;
}

633 634 635
struct net_bridge_port_group *br_multicast_new_port_group(
			struct net_bridge_port *port,
			struct br_ip *group,
636 637
			struct net_bridge_port_group __rcu *next,
			unsigned char state)
638 639 640 641 642 643 644 645 646
{
	struct net_bridge_port_group *p;

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

	p->addr = *group;
	p->port = port;
647
	p->state = state;
648
	rcu_assign_pointer(p->next, next);
649 650 651 652 653 654
	hlist_add_head(&p->mglist, &port->mglist);
	setup_timer(&p->timer, br_multicast_port_group_expired,
		    (unsigned long)p);
	return p;
}

655
static int br_multicast_add_group(struct net_bridge *br,
656 657
				  struct net_bridge_port *port,
				  struct br_ip *group)
658 659 660
{
	struct net_bridge_mdb_entry *mp;
	struct net_bridge_port_group *p;
661
	struct net_bridge_port_group __rcu **pp;
662 663 664 665 666 667 668 669 670
	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);
671
	if (IS_ERR(mp))
672 673 674
		goto err;

	if (!port) {
675
		mp->mglist = true;
676 677 678
		goto out;
	}

679 680 681
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
682
		if (p->port == port)
683
			goto out;
684 685 686 687
		if ((unsigned long)p->port < (unsigned long)port)
			break;
	}

688
	p = br_multicast_new_port_group(port, group, *pp, MDB_TEMPORARY);
689 690 691
	if (unlikely(!p))
		goto err;
	rcu_assign_pointer(*pp, p);
C
Cong Wang 已提交
692
	br_mdb_notify(br->dev, port, group, RTM_NEWMDB);
693 694 695 696 697 698 699 700 701

out:
	err = 0;

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

702 703
static int br_ip4_multicast_add_group(struct net_bridge *br,
				      struct net_bridge_port *port,
704 705
				      __be32 group,
				      __u16 vid)
706 707 708 709 710 711 712 713
{
	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);
714
	br_group.vid = vid;
715 716 717 718

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

E
Eric Dumazet 已提交
719
#if IS_ENABLED(CONFIG_IPV6)
720 721
static int br_ip6_multicast_add_group(struct net_bridge *br,
				      struct net_bridge_port *port,
722 723
				      const struct in6_addr *group,
				      __u16 vid)
724 725 726
{
	struct br_ip br_group;

727
	if (!ipv6_is_transient_multicast(group))
728 729
		return 0;

A
Alexey Dobriyan 已提交
730
	br_group.u.ip6 = *group;
731
	br_group.proto = htons(ETH_P_IPV6);
732
	br_group.vid = vid;
733 734 735 736 737

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

738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758
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)
{
}

759 760
static void br_multicast_querier_expired(struct net_bridge *br,
					 struct bridge_mcast_query *query)
761 762 763 764 765
{
	spin_lock(&br->multicast_lock);
	if (!netif_running(br->dev) || br->multicast_disabled)
		goto out;

766
	br_multicast_start_querier(br, query);
767 768 769 770 771

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

772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787
static void br_ip4_multicast_querier_expired(unsigned long data)
{
	struct net_bridge *br = (void *)data;

	br_multicast_querier_expired(br, &br->ip4_query);
}

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

	br_multicast_querier_expired(br, &br->ip6_query);
}
#endif

788 789 790
static void __br_multicast_send_query(struct net_bridge *br,
				      struct net_bridge_port *port,
				      struct br_ip *ip)
791 792 793
{
	struct sk_buff *skb;

794
	skb = br_multicast_alloc_query(br, ip);
795
	if (!skb)
796
		return;
797 798 799 800

	if (port) {
		__skb_push(skb, sizeof(struct ethhdr));
		skb->dev = port->dev;
801
		NF_HOOK(NFPROTO_BRIDGE, NF_BR_LOCAL_OUT, skb, NULL, skb->dev,
802 803 804
			dev_queue_xmit);
	} else
		netif_rx(skb);
805 806 807
}

static void br_multicast_send_query(struct net_bridge *br,
808 809
				    struct net_bridge_port *port,
				    struct bridge_mcast_query *query)
810 811 812
{
	unsigned long time;
	struct br_ip br_group;
813
	struct bridge_mcast_querier *querier = NULL;
814 815

	if (!netif_running(br->dev) || br->multicast_disabled ||
816
	    !br->multicast_querier)
817 818
		return;

819 820
	memset(&br_group.u, 0, sizeof(br_group.u));

821 822 823 824
	if (port ? (query == &port->ip4_query) :
		   (query == &br->ip4_query)) {
		querier = &br->ip4_querier;
		br_group.proto = htons(ETH_P_IP);
E
Eric Dumazet 已提交
825
#if IS_ENABLED(CONFIG_IPV6)
826 827 828
	} else {
		querier = &br->ip6_querier;
		br_group.proto = htons(ETH_P_IPV6);
829
#endif
830 831 832 833 834 835
	}

	if (!querier || timer_pending(&querier->timer))
		return;

	__br_multicast_send_query(br, port, &br_group);
836 837

	time = jiffies;
838
	time += query->startup_sent < br->multicast_startup_query_count ?
839 840
		br->multicast_startup_query_interval :
		br->multicast_query_interval;
841
	mod_timer(&query->timer, time);
842 843
}

844 845
static void br_multicast_port_query_expired(struct net_bridge_port *port,
					    struct bridge_mcast_query *query)
846 847 848 849
{
	struct net_bridge *br = port->br;

	spin_lock(&br->multicast_lock);
850 851
	if (port->state == BR_STATE_DISABLED ||
	    port->state == BR_STATE_BLOCKING)
852 853
		goto out;

854 855
	if (query->startup_sent < br->multicast_startup_query_count)
		query->startup_sent++;
856

857
	br_multicast_send_query(port->br, port, query);
858 859 860 861 862

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

863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878
static void br_ip4_multicast_port_query_expired(unsigned long data)
{
	struct net_bridge_port *port = (void *)data;

	br_multicast_port_query_expired(port, &port->ip4_query);
}

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

	br_multicast_port_query_expired(port, &port->ip6_query);
}
#endif

879 880 881 882 883 884
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);
885 886 887 888 889 890
	setup_timer(&port->ip4_query.timer, br_ip4_multicast_port_query_expired,
		    (unsigned long)port);
#if IS_ENABLED(CONFIG_IPV6)
	setup_timer(&port->ip6_query.timer, br_ip6_multicast_port_query_expired,
		    (unsigned long)port);
#endif
891 892 893 894 895 896 897
}

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

898
static void br_multicast_enable(struct bridge_mcast_query *query)
899
{
900
	query->startup_sent = 0;
901

902 903 904
	if (try_to_del_timer_sync(&query->timer) >= 0 ||
	    del_timer(&query->timer))
		mod_timer(&query->timer, jiffies);
905 906
}

907 908 909 910 911 912 913 914
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;

915 916 917 918
	br_multicast_enable(&port->ip4_query);
#if IS_ENABLED(CONFIG_IPV6)
	br_multicast_enable(&port->ip6_query);
#endif
919 920 921 922 923 924 925 926 927

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;
928
	struct hlist_node *n;
929 930

	spin_lock(&br->multicast_lock);
931
	hlist_for_each_entry_safe(pg, n, &port->mglist, mglist)
932 933 934 935 936
		br_multicast_del_pg(br, pg);

	if (!hlist_unhashed(&port->rlist))
		hlist_del_init_rcu(&port->rlist);
	del_timer(&port->multicast_router_timer);
937 938 939 940
	del_timer(&port->ip4_query.timer);
#if IS_ENABLED(CONFIG_IPV6)
	del_timer(&port->ip6_query.timer);
#endif
941 942 943
	spin_unlock(&br->multicast_lock);
}

944 945 946
static int br_ip4_multicast_igmp3_report(struct net_bridge *br,
					 struct net_bridge_port *port,
					 struct sk_buff *skb)
947 948 949 950 951 952 953 954 955
{
	struct igmpv3_report *ih;
	struct igmpv3_grec *grec;
	int i;
	int len;
	int num;
	int type;
	int err = 0;
	__be32 group;
956
	u16 vid = 0;
957 958 959 960

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

961
	br_vlan_get_tag(skb, &vid);
962 963 964 965 966 967 968 969 970
	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
971
		grec = (void *)(skb->data + len - sizeof(*grec));
972 973 974
		group = grec->grec_mca;
		type = grec->grec_type;

E
Eric Dumazet 已提交
975
		len += ntohs(grec->grec_nsrcs) * 4;
976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992
		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;
		}

993
		err = br_ip4_multicast_add_group(br, port, group, vid);
994 995 996 997 998 999 1000
		if (err)
			break;
	}

	return err;
}

E
Eric Dumazet 已提交
1001
#if IS_ENABLED(CONFIG_IPV6)
1002 1003 1004 1005 1006 1007 1008 1009 1010 1011
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;
1012
	u16 vid = 0;
1013 1014 1015 1016

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

1017
	br_vlan_get_tag(skb, &vid);
1018 1019 1020 1021 1022 1023 1024 1025 1026
	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,
1027
							  grec_nsrcs),
1028 1029 1030 1031 1032 1033
					   sizeof(_nsrcs), &_nsrcs);
		if (!nsrcs)
			return -EINVAL;

		if (!pskb_may_pull(skb,
				   len + sizeof(*grec) +
1034
				   sizeof(struct in6_addr) * ntohs(*nsrcs)))
1035 1036 1037
			return -EINVAL;

		grec = (struct mld2_grec *)(skb->data + len);
1038 1039
		len += sizeof(*grec) +
		       sizeof(struct in6_addr) * ntohs(*nsrcs);
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054

		/* 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;
		}

1055 1056
		err = br_ip6_multicast_add_group(br, port, &grec->grec_mca,
						 vid);
1057 1058 1059 1060 1061 1062 1063 1064
		if (!err)
			break;
	}

	return err;
}
#endif

1065 1066 1067 1068
static void
br_multicast_update_querier_timer(struct net_bridge *br,
				  struct bridge_mcast_querier *querier,
				  unsigned long max_delay)
1069
{
1070 1071
	if (!timer_pending(&querier->timer))
		querier->delay_time = jiffies + max_delay;
1072

1073
	mod_timer(&querier->timer, jiffies + br->multicast_querier_interval);
1074 1075
}

1076
/*
C
Cong Wang 已提交
1077
 * Add port to router_list
1078 1079 1080
 *  list is maintained ordered by pointer value
 *  and locked by br->multicast_lock and RCU
 */
1081 1082 1083
static void br_multicast_add_router(struct net_bridge *br,
				    struct net_bridge_port *port)
{
1084
	struct net_bridge_port *p;
1085
	struct hlist_node *slot = NULL;
1086

1087
	hlist_for_each_entry(p, &br->router_list, rlist) {
1088 1089
		if ((unsigned long) port >= (unsigned long) p)
			break;
1090
		slot = &p->rlist;
1091 1092
	}

1093 1094
	if (slot)
		hlist_add_after_rcu(slot, &port->rlist);
1095 1096
	else
		hlist_add_head_rcu(&port->rlist, &br->router_list);
1097 1098
}

1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116
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;

1117
	br_multicast_add_router(br, port);
1118 1119 1120 1121 1122 1123 1124 1125

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,
1126
					struct bridge_mcast_querier *querier,
1127 1128
					int saddr,
					unsigned long max_delay)
1129 1130
{
	if (saddr)
1131 1132
		br_multicast_update_querier_timer(br, querier, max_delay);
	else if (timer_pending(&querier->timer))
1133 1134 1135 1136 1137
		return;

	br_multicast_mark_router(br, port);
}

1138 1139 1140
static int br_ip4_multicast_query(struct net_bridge *br,
				  struct net_bridge_port *port,
				  struct sk_buff *skb)
1141
{
1142
	const struct iphdr *iph = ip_hdr(skb);
1143 1144 1145 1146
	struct igmphdr *ih = igmp_hdr(skb);
	struct net_bridge_mdb_entry *mp;
	struct igmpv3_query *ih3;
	struct net_bridge_port_group *p;
1147
	struct net_bridge_port_group __rcu **pp;
1148 1149 1150
	unsigned long max_delay;
	unsigned long now = jiffies;
	__be32 group;
1151
	int err = 0;
1152
	u16 vid = 0;
1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168

	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 {
1169 1170 1171 1172
		if (!pskb_may_pull(skb, sizeof(struct igmpv3_query))) {
			err = -EINVAL;
			goto out;
		}
1173 1174 1175

		ih3 = igmpv3_query_hdr(skb);
		if (ih3->nsrcs)
1176
			goto out;
1177

1178 1179
		max_delay = ih3->code ?
			    IGMPV3_MRC(ih3->code) * (HZ / IGMP_TIMER_SCALE) : 1;
1180 1181
	}

1182 1183
	br_multicast_query_received(br, port, &br->ip4_querier, !!iph->saddr,
				    max_delay);
1184

1185 1186 1187
	if (!group)
		goto out;

1188 1189
	br_vlan_get_tag(skb, &vid);
	mp = br_mdb_ip4_get(mlock_dereference(br->mdb, br), group, vid);
1190 1191 1192
	if (!mp)
		goto out;

1193 1194 1195
	mod_timer(&mp->timer, now + br->multicast_membership_interval);
	mp->timer_armed = true;

1196 1197
	max_delay *= br->multicast_last_member_count;

1198
	if (mp->mglist &&
1199 1200 1201 1202 1203
	    (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);

1204 1205 1206
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
1207 1208 1209
		if (timer_pending(&p->timer) ?
		    time_after(p->timer.expires, now + max_delay) :
		    try_to_del_timer_sync(&p->timer) >= 0)
1210
			mod_timer(&p->timer, now + max_delay);
1211 1212 1213 1214
	}

out:
	spin_unlock(&br->multicast_lock);
1215
	return err;
1216 1217
}

E
Eric Dumazet 已提交
1218
#if IS_ENABLED(CONFIG_IPV6)
1219 1220 1221 1222
static int br_ip6_multicast_query(struct net_bridge *br,
				  struct net_bridge_port *port,
				  struct sk_buff *skb)
{
1223
	const struct ipv6hdr *ip6h = ipv6_hdr(skb);
1224
	struct mld_msg *mld;
1225 1226
	struct net_bridge_mdb_entry *mp;
	struct mld2_query *mld2q;
1227 1228
	struct net_bridge_port_group *p;
	struct net_bridge_port_group __rcu **pp;
1229 1230
	unsigned long max_delay;
	unsigned long now = jiffies;
1231
	const struct in6_addr *group = NULL;
1232
	int err = 0;
1233
	u16 vid = 0;
1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245

	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
1246
		max_delay = msecs_to_jiffies(ntohs(mld->mld_maxdelay));
1247 1248
		if (max_delay)
			group = &mld->mld_mca;
1249
	} else {
1250 1251 1252 1253 1254 1255 1256
		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;
1257 1258

		max_delay = max(msecs_to_jiffies(MLDV2_MRC(ntohs(mld2q->mld2q_mrc))), 1UL);
1259 1260
	}

1261 1262
	br_multicast_query_received(br, port, &br->ip6_querier,
				    !ipv6_addr_any(&ip6h->saddr), max_delay);
1263

1264 1265 1266
	if (!group)
		goto out;

1267 1268
	br_vlan_get_tag(skb, &vid);
	mp = br_mdb_ip6_get(mlock_dereference(br->mdb, br), group, vid);
1269 1270 1271
	if (!mp)
		goto out;

1272 1273 1274
	mod_timer(&mp->timer, now + br->multicast_membership_interval);
	mp->timer_armed = true;

1275
	max_delay *= br->multicast_last_member_count;
1276
	if (mp->mglist &&
1277 1278 1279 1280 1281
	    (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);

1282 1283 1284
	for (pp = &mp->ports;
	     (p = mlock_dereference(*pp, br)) != NULL;
	     pp = &p->next) {
1285 1286 1287
		if (timer_pending(&p->timer) ?
		    time_after(p->timer.expires, now + max_delay) :
		    try_to_del_timer_sync(&p->timer) >= 0)
1288
			mod_timer(&p->timer, now + max_delay);
1289 1290 1291 1292 1293 1294 1295 1296
	}

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

1297 1298
static void br_multicast_leave_group(struct net_bridge *br,
				     struct net_bridge_port *port,
1299 1300 1301
				     struct br_ip *group,
				     struct bridge_mcast_querier *querier,
				     struct bridge_mcast_query *query)
1302 1303 1304 1305 1306 1307 1308 1309 1310 1311
{
	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) ||
1312
	    timer_pending(&querier->timer))
1313 1314
		goto out;

1315
	mdb = mlock_dereference(br->mdb, br);
1316 1317 1318 1319
	mp = br_mdb_ip_get(mdb, group);
	if (!mp)
		goto out;

1320
	if (br->multicast_querier) {
1321 1322 1323 1324
		__br_multicast_send_query(br, port, &mp->addr);

		time = jiffies + br->multicast_last_member_count *
				 br->multicast_last_member_interval;
1325 1326

		mod_timer(&query->timer, time);
1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344

		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;
		}
	}

1345
	if (port && (port->flags & BR_MULTICAST_FAST_LEAVE)) {
1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357
		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
1358
			br_mdb_notify(br->dev, port, group, RTM_DELMDB);
1359

1360
			if (!mp->ports && !mp->mglist && mp->timer_armed &&
1361 1362 1363 1364 1365 1366
			    netif_running(br->dev))
				mod_timer(&mp->timer, jiffies);
		}
		goto out;
	}

1367 1368 1369 1370 1371
	now = jiffies;
	time = now + br->multicast_last_member_count *
		     br->multicast_last_member_interval;

	if (!port) {
1372
		if (mp->mglist && mp->timer_armed &&
1373 1374 1375 1376 1377 1378 1379 1380 1381 1382
		    (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);
}

1383 1384
static void br_ip4_multicast_leave_group(struct net_bridge *br,
					 struct net_bridge_port *port,
1385 1386
					 __be32 group,
					 __u16 vid)
1387 1388
{
	struct br_ip br_group;
1389 1390
	struct bridge_mcast_query *query = port ? &port->ip4_query :
						  &br->ip4_query;
1391 1392 1393 1394 1395 1396

	if (ipv4_is_local_multicast(group))
		return;

	br_group.u.ip4 = group;
	br_group.proto = htons(ETH_P_IP);
1397
	br_group.vid = vid;
1398

1399
	br_multicast_leave_group(br, port, &br_group, &br->ip4_querier, query);
1400 1401
}

E
Eric Dumazet 已提交
1402
#if IS_ENABLED(CONFIG_IPV6)
1403 1404
static void br_ip6_multicast_leave_group(struct net_bridge *br,
					 struct net_bridge_port *port,
1405 1406
					 const struct in6_addr *group,
					 __u16 vid)
1407 1408
{
	struct br_ip br_group;
1409 1410 1411
	struct bridge_mcast_query *query = port ? &port->ip6_query :
						  &br->ip6_query;

1412

1413
	if (!ipv6_is_transient_multicast(group))
1414 1415
		return;

A
Alexey Dobriyan 已提交
1416
	br_group.u.ip6 = *group;
1417
	br_group.proto = htons(ETH_P_IPV6);
1418
	br_group.vid = vid;
1419

1420
	br_multicast_leave_group(br, port, &br_group, &br->ip6_querier, query);
1421 1422
}
#endif
1423

1424 1425 1426 1427 1428
static int br_multicast_ipv4_rcv(struct net_bridge *br,
				 struct net_bridge_port *port,
				 struct sk_buff *skb)
{
	struct sk_buff *skb2 = skb;
1429
	const struct iphdr *iph;
1430
	struct igmphdr *ih;
1431 1432
	unsigned int len;
	unsigned int offset;
1433
	int err;
1434
	u16 vid = 0;
1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452

	/* 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;

1453
	if (iph->protocol != IPPROTO_IGMP) {
1454
		if (!ipv4_is_local_multicast(iph->daddr))
1455
			BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1456
		return 0;
1457
	}
1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469

	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)
1470
			goto err_out;
1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489
	}

	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))
1490
			goto out;
1491 1492 1493 1494
	}

	err = 0;

1495
	br_vlan_get_tag(skb2, &vid);
1496 1497 1498 1499 1500 1501
	BR_INPUT_SKB_CB(skb)->igmp = 1;
	ih = igmp_hdr(skb2);

	switch (ih->type) {
	case IGMP_HOST_MEMBERSHIP_REPORT:
	case IGMPV2_HOST_MEMBERSHIP_REPORT:
1502
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1503
		err = br_ip4_multicast_add_group(br, port, ih->group, vid);
1504 1505
		break;
	case IGMPV3_HOST_MEMBERSHIP_REPORT:
1506
		err = br_ip4_multicast_igmp3_report(br, port, skb2);
1507 1508
		break;
	case IGMP_HOST_MEMBERSHIP_QUERY:
1509
		err = br_ip4_multicast_query(br, port, skb2);
1510 1511
		break;
	case IGMP_HOST_LEAVE_MESSAGE:
1512
		br_ip4_multicast_leave_group(br, port, ih->group, vid);
1513 1514 1515 1516 1517
		break;
	}

out:
	__skb_push(skb2, offset);
1518
err_out:
1519 1520 1521 1522 1523
	if (skb2 != skb)
		kfree_skb(skb2);
	return err;
}

E
Eric Dumazet 已提交
1524
#if IS_ENABLED(CONFIG_IPV6)
1525 1526 1527 1528
static int br_multicast_ipv6_rcv(struct net_bridge *br,
				 struct net_bridge_port *port,
				 struct sk_buff *skb)
{
1529
	struct sk_buff *skb2;
1530
	const struct ipv6hdr *ip6h;
1531
	u8 icmp6_type;
1532
	u8 nexthdr;
1533
	__be16 frag_off;
1534
	unsigned int len;
K
Kulikov Vasiliy 已提交
1535
	int offset;
1536
	int err;
1537
	u16 vid = 0;
1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554

	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.
	 */
	if (ip6h->version != 6 ||
	    ip6h->nexthdr != IPPROTO_HOPOPTS ||
	    ip6h->payload_len == 0)
		return 0;

1555
	len = ntohs(ip6h->payload_len) + sizeof(*ip6h);
1556 1557 1558 1559
	if (skb->len < len)
		return -EINVAL;

	nexthdr = ip6h->nexthdr;
1560
	offset = ipv6_skip_exthdr(skb, sizeof(*ip6h), &nexthdr, &frag_off);
1561 1562 1563 1564 1565 1566 1567 1568 1569

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

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

1570 1571 1572 1573
	err = -EINVAL;
	if (!pskb_may_pull(skb2, offset + sizeof(struct icmp6hdr)))
		goto out;

1574 1575 1576 1577
	len -= offset - skb_network_offset(skb2);

	__skb_pull(skb2, offset);
	skb_reset_transport_header(skb2);
1578 1579
	skb_postpull_rcsum(skb2, skb_network_header(skb2),
			   skb_network_header_len(skb2));
1580

1581
	icmp6_type = icmp6_hdr(skb2)->icmp6_type;
1582

1583
	switch (icmp6_type) {
1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598
	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;
1599
		err = -EINVAL;
1600 1601
	}

1602 1603
	ip6h = ipv6_hdr(skb2);

1604 1605
	switch (skb2->ip_summed) {
	case CHECKSUM_COMPLETE:
1606 1607
		if (!csum_ipv6_magic(&ip6h->saddr, &ip6h->daddr, skb2->len,
					IPPROTO_ICMPV6, skb2->csum))
1608 1609 1610
			break;
		/*FALLTHROUGH*/
	case CHECKSUM_NONE:
1611 1612 1613 1614 1615
		skb2->csum = ~csum_unfold(csum_ipv6_magic(&ip6h->saddr,
							&ip6h->daddr,
							skb2->len,
							IPPROTO_ICMPV6, 0));
		if (__skb_checksum_complete(skb2))
1616 1617 1618 1619 1620
			goto out;
	}

	err = 0;

1621
	br_vlan_get_tag(skb, &vid);
1622 1623
	BR_INPUT_SKB_CB(skb)->igmp = 1;

1624
	switch (icmp6_type) {
1625 1626
	case ICMPV6_MGM_REPORT:
	    {
1627 1628 1629 1630 1631 1632
		struct mld_msg *mld;
		if (!pskb_may_pull(skb2, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *)skb_transport_header(skb2);
1633
		BR_INPUT_SKB_CB(skb)->mrouters_only = 1;
1634
		err = br_ip6_multicast_add_group(br, port, &mld->mld_mca, vid);
1635 1636 1637 1638 1639 1640 1641 1642 1643 1644
		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:
	    {
1645 1646 1647 1648 1649 1650
		struct mld_msg *mld;
		if (!pskb_may_pull(skb2, sizeof(*mld))) {
			err = -EINVAL;
			goto out;
		}
		mld = (struct mld_msg *)skb_transport_header(skb2);
1651
		br_ip6_multicast_leave_group(br, port, &mld->mld_mca, vid);
1652 1653 1654 1655
	    }
	}

out:
1656
	kfree_skb(skb2);
1657 1658 1659 1660
	return err;
}
#endif

1661 1662 1663
int br_multicast_rcv(struct net_bridge *br, struct net_bridge_port *port,
		     struct sk_buff *skb)
{
1664 1665 1666
	BR_INPUT_SKB_CB(skb)->igmp = 0;
	BR_INPUT_SKB_CB(skb)->mrouters_only = 0;

1667 1668 1669 1670 1671 1672
	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
1673
#if IS_ENABLED(CONFIG_IPV6)
1674 1675 1676
	case htons(ETH_P_IPV6):
		return br_multicast_ipv6_rcv(br, port, skb);
#endif
1677 1678 1679 1680 1681
	}

	return 0;
}

1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693
static void br_multicast_query_expired(struct net_bridge *br,
				       struct bridge_mcast_query *query)
{
	spin_lock(&br->multicast_lock);
	if (query->startup_sent < br->multicast_startup_query_count)
		query->startup_sent++;

	br_multicast_send_query(br, NULL, query);
	spin_unlock(&br->multicast_lock);
}

static void br_ip4_multicast_query_expired(unsigned long data)
1694 1695 1696
{
	struct net_bridge *br = (void *)data;

1697 1698
	br_multicast_query_expired(br, &br->ip4_query);
}
1699

1700 1701 1702 1703
#if IS_ENABLED(CONFIG_IPV6)
static void br_ip6_multicast_query_expired(unsigned long data)
{
	struct net_bridge *br = (void *)data;
1704

1705
	br_multicast_query_expired(br, &br->ip6_query);
1706
}
1707
#endif
1708 1709 1710 1711 1712 1713 1714

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

	br->multicast_router = 1;
1715
	br->multicast_querier = 0;
1716
	br->multicast_query_use_ifaddr = 0;
1717 1718 1719 1720 1721 1722 1723 1724 1725 1726
	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;

1727 1728 1729 1730
	br->ip4_querier.delay_time = 0;
#if IS_ENABLED(CONFIG_IPV6)
	br->ip6_querier.delay_time = 0;
#endif
1731

1732 1733 1734
	spin_lock_init(&br->multicast_lock);
	setup_timer(&br->multicast_router_timer,
		    br_multicast_local_router_expired, 0);
1735 1736 1737
	setup_timer(&br->ip4_querier.timer, br_ip4_multicast_querier_expired,
		    (unsigned long)br);
	setup_timer(&br->ip4_query.timer, br_ip4_multicast_query_expired,
1738
		    (unsigned long)br);
1739 1740 1741 1742 1743 1744
#if IS_ENABLED(CONFIG_IPV6)
	setup_timer(&br->ip6_querier.timer, br_ip6_multicast_querier_expired,
		    (unsigned long)br);
	setup_timer(&br->ip6_query.timer, br_ip6_multicast_query_expired,
		    (unsigned long)br);
#endif
1745 1746
}

1747 1748
static void __br_multicast_open(struct net_bridge *br,
				struct bridge_mcast_query *query)
1749
{
1750
	query->startup_sent = 0;
1751 1752 1753 1754

	if (br->multicast_disabled)
		return;

1755 1756 1757 1758 1759 1760 1761 1762 1763
	mod_timer(&query->timer, jiffies);
}

void br_multicast_open(struct net_bridge *br)
{
	__br_multicast_open(br, &br->ip4_query);
#if IS_ENABLED(CONFIG_IPV6)
	__br_multicast_open(br, &br->ip6_query);
#endif
1764 1765 1766 1767 1768 1769
}

void br_multicast_stop(struct net_bridge *br)
{
	struct net_bridge_mdb_htable *mdb;
	struct net_bridge_mdb_entry *mp;
1770
	struct hlist_node *n;
1771 1772 1773 1774
	u32 ver;
	int i;

	del_timer_sync(&br->multicast_router_timer);
1775 1776 1777 1778 1779 1780
	del_timer_sync(&br->ip4_querier.timer);
	del_timer_sync(&br->ip4_query.timer);
#if IS_ENABLED(CONFIG_IPV6)
	del_timer_sync(&br->ip6_querier.timer);
	del_timer_sync(&br->ip6_query.timer);
#endif
1781 1782

	spin_lock_bh(&br->multicast_lock);
1783
	mdb = mlock_dereference(br->mdb, br);
1784 1785 1786 1787 1788 1789 1790
	if (!mdb)
		goto out;

	br->mdb = NULL;

	ver = mdb->ver;
	for (i = 0; i < mdb->max; i++) {
1791
		hlist_for_each_entry_safe(mp, n, &mdb->mhash[i],
1792 1793
					  hlist[ver]) {
			del_timer(&mp->timer);
1794
			mp->timer_armed = false;
1795 1796 1797 1798 1799 1800
			call_rcu_bh(&mp->rcu, br_multicast_free_group);
		}
	}

	if (mdb->old) {
		spin_unlock_bh(&br->multicast_lock);
1801
		rcu_barrier_bh();
1802 1803 1804 1805 1806 1807 1808 1809 1810 1811
		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);
}
1812 1813 1814 1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827 1828 1829 1830 1831 1832 1833 1834 1835 1836 1837 1838 1839 1840 1841 1842 1843 1844 1845 1846 1847 1848 1849 1850 1851 1852 1853 1854 1855 1856 1857 1858 1859 1860 1861 1862 1863 1864 1865 1866 1867 1868 1869 1870 1871 1872 1873 1874 1875 1876 1877 1878 1879 1880 1881

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;
}
1882

1883 1884
static void br_multicast_start_querier(struct net_bridge *br,
				       struct bridge_mcast_query *query)
1885 1886
{
	struct net_bridge_port *port;
1887

1888
	__br_multicast_open(br, query);
1889 1890 1891 1892 1893 1894

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

1895 1896 1897 1898 1899 1900
		if (query == &br->ip4_query)
			br_multicast_enable(&port->ip4_query);
#if IS_ENABLED(CONFIG_IPV6)
		else
			br_multicast_enable(&port->ip6_query);
#endif
1901 1902 1903 1904 1905
	}
}

int br_multicast_toggle(struct net_bridge *br, unsigned long val)
{
1906
	int err = 0;
1907
	struct net_bridge_mdb_htable *mdb;
1908

1909
	spin_lock_bh(&br->multicast_lock);
1910 1911 1912 1913 1914 1915 1916
	if (br->multicast_disabled == !val)
		goto unlock;

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

1917 1918 1919
	if (!netif_running(br->dev))
		goto unlock;

1920 1921 1922
	mdb = mlock_dereference(br->mdb, br);
	if (mdb) {
		if (mdb->old) {
1923 1924 1925 1926 1927 1928
			err = -EEXIST;
rollback:
			br->multicast_disabled = !!val;
			goto unlock;
		}

1929
		err = br_mdb_rehash(&br->mdb, mdb->max,
1930 1931 1932 1933 1934
				    br->hash_elasticity);
		if (err)
			goto rollback;
	}

1935 1936 1937 1938
	br_multicast_start_querier(br, &br->ip4_query);
#if IS_ENABLED(CONFIG_IPV6)
	br_multicast_start_querier(br, &br->ip6_query);
#endif
1939 1940

unlock:
1941
	spin_unlock_bh(&br->multicast_lock);
1942 1943 1944

	return err;
}
1945

1946 1947
int br_multicast_set_querier(struct net_bridge *br, unsigned long val)
{
1948 1949
	unsigned long max_delay;

1950 1951 1952 1953 1954 1955 1956
	val = !!val;

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

	br->multicast_querier = val;
1957 1958 1959 1960 1961
	if (!val)
		goto unlock;

	max_delay = br->multicast_query_response_interval;

1962 1963 1964 1965 1966 1967 1968 1969 1970 1971 1972
	if (!timer_pending(&br->ip4_querier.timer))
		br->ip4_querier.delay_time = jiffies + max_delay;

	br_multicast_start_querier(br, &br->ip4_query);

#if IS_ENABLED(CONFIG_IPV6)
	if (!timer_pending(&br->ip6_querier.timer))
		br->ip6_querier.delay_time = jiffies + max_delay;

	br_multicast_start_querier(br, &br->ip6_query);
#endif
1973 1974 1975 1976 1977 1978 1979

unlock:
	spin_unlock_bh(&br->multicast_lock);

	return 0;
}

1980 1981 1982 1983
int br_multicast_set_hash_max(struct net_bridge *br, unsigned long val)
{
	int err = -ENOENT;
	u32 old;
1984
	struct net_bridge_mdb_htable *mdb;
1985 1986 1987 1988 1989 1990 1991 1992

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

	err = -EINVAL;
	if (!is_power_of_2(val))
		goto unlock;
1993 1994 1995

	mdb = mlock_dereference(br->mdb, br);
	if (mdb && val < mdb->size)
1996 1997 1998 1999 2000 2001 2002
		goto unlock;

	err = 0;

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

2003 2004
	if (mdb) {
		if (mdb->old) {
2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021
			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;
}