udp.c 36.8 KB
Newer Older
L
Linus Torvalds 已提交
1 2
/*
 *	UDP over IPv6
3
 *	Linux INET6 implementation
L
Linus Torvalds 已提交
4 5
 *
 *	Authors:
6
 *	Pedro Roque		<roque@di.fc.ul.pt>
L
Linus Torvalds 已提交
7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
 *
 *	Based on linux/ipv4/udp.c
 *
 *	Fixes:
 *	Hideaki YOSHIFUJI	:	sin6_scope_id support
 *	YOSHIFUJI Hideaki @USAGI and:	Support IPV6_V6ONLY socket option, which
 *	Alexey Kuznetsov		allow both IPv4 and IPv6 sockets to bind
 *					a single port at the same time.
 *      Kazunori MIYAZAWA @USAGI:       change process style to use ip6_append_data
 *      YOSHIFUJI Hideaki @USAGI:	convert /proc/net/udp6 to seq_file.
 *
 *	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/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/in6.h>
#include <linux/netdevice.h>
#include <linux/if_arp.h>
#include <linux/ipv6.h>
#include <linux/icmpv6.h>
#include <linux/init.h>
35
#include <linux/module.h>
36
#include <linux/skbuff.h>
37
#include <linux/slab.h>
L
Linus Torvalds 已提交
38 39 40 41 42 43 44
#include <asm/uaccess.h>

#include <net/ndisc.h>
#include <net/protocol.h>
#include <net/transp_v6.h>
#include <net/ip6_route.h>
#include <net/raw.h>
45
#include <net/tcp_states.h>
L
Linus Torvalds 已提交
46 47
#include <net/ip6_checksum.h>
#include <net/xfrm.h>
48
#include <net/inet6_hashtables.h>
49
#include <net/ll_poll.h>
L
Linus Torvalds 已提交
50 51 52

#include <linux/proc_fs.h>
#include <linux/seq_file.h>
53
#include <trace/events/skb.h>
54
#include "udp_impl.h"
L
Linus Torvalds 已提交
55

56 57 58 59
int ipv6_rcv_saddr_equal(const struct sock *sk, const struct sock *sk2)
{
	const struct in6_addr *sk_rcv_saddr6 = &inet6_sk(sk)->rcv_saddr;
	const struct in6_addr *sk2_rcv_saddr6 = inet6_rcv_saddr(sk2);
60 61
	__be32 sk1_rcv_saddr = sk_rcv_saddr(sk);
	__be32 sk2_rcv_saddr = sk_rcv_saddr(sk2);
62 63 64 65 66 67 68
	int sk_ipv6only = ipv6_only_sock(sk);
	int sk2_ipv6only = inet_v6_ipv6only(sk2);
	int addr_type = ipv6_addr_type(sk_rcv_saddr6);
	int addr_type2 = sk2_rcv_saddr6 ? ipv6_addr_type(sk2_rcv_saddr6) : IPV6_ADDR_MAPPED;

	/* if both are mapped, treat as IPv4 */
	if (addr_type == IPV6_ADDR_MAPPED && addr_type2 == IPV6_ADDR_MAPPED)
69
		return (!sk2_ipv6only &&
E
Eric Dumazet 已提交
70 71
			(!sk1_rcv_saddr || !sk2_rcv_saddr ||
			  sk1_rcv_saddr == sk2_rcv_saddr));
72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87

	if (addr_type2 == IPV6_ADDR_ANY &&
	    !(sk2_ipv6only && addr_type == IPV6_ADDR_MAPPED))
		return 1;

	if (addr_type == IPV6_ADDR_ANY &&
	    !(sk_ipv6only && addr_type2 == IPV6_ADDR_MAPPED))
		return 1;

	if (sk2_rcv_saddr6 &&
	    ipv6_addr_equal(sk_rcv_saddr6, sk2_rcv_saddr6))
		return 1;

	return 0;
}

88 89 90 91 92 93 94 95
static unsigned int udp6_portaddr_hash(struct net *net,
				       const struct in6_addr *addr6,
				       unsigned int port)
{
	unsigned int hash, mix = net_hash_mix(net);

	if (ipv6_addr_any(addr6))
		hash = jhash_1word(0, mix);
B
Brian Haley 已提交
96
	else if (ipv6_addr_v4mapped(addr6))
97
		hash = jhash_1word((__force u32)addr6->s6_addr32[3], mix);
98
	else
99
		hash = jhash2((__force u32 *)addr6->s6_addr32, 4, mix);
100 101 102 103 104

	return hash ^ port;
}


105
int udp_v6_get_port(struct sock *sk, unsigned short snum)
L
Linus Torvalds 已提交
106
{
E
Eric Dumazet 已提交
107 108
	unsigned int hash2_nulladdr =
		udp6_portaddr_hash(sock_net(sk), &in6addr_any, snum);
109
	unsigned int hash2_partial =
E
Eric Dumazet 已提交
110 111
		udp6_portaddr_hash(sock_net(sk), &inet6_sk(sk)->rcv_saddr, 0);

112
	/* precompute partial secondary hash */
E
Eric Dumazet 已提交
113 114
	udp_sk(sk)->udp_portaddr_hash = hash2_partial;
	return udp_lib_get_port(sk, snum, ipv6_rcv_saddr_equal, hash2_nulladdr);
L
Linus Torvalds 已提交
115 116
}

E
Eric Dumazet 已提交
117 118 119 120 121 122 123 124 125
static void udp_v6_rehash(struct sock *sk)
{
	u16 new_hash = udp6_portaddr_hash(sock_net(sk),
					  &inet6_sk(sk)->rcv_saddr,
					  inet_sk(sk)->inet_num);

	udp_lib_rehash(sk, new_hash);
}

126 127
static inline int compute_score(struct sock *sk, struct net *net,
				unsigned short hnum,
128 129
				const struct in6_addr *saddr, __be16 sport,
				const struct in6_addr *daddr, __be16 dport,
130 131 132 133
				int dif)
{
	int score = -1;

134
	if (net_eq(sock_net(sk), net) && udp_sk(sk)->udp_port_hash == hnum &&
135 136 137 138 139
			sk->sk_family == PF_INET6) {
		struct ipv6_pinfo *np = inet6_sk(sk);
		struct inet_sock *inet = inet_sk(sk);

		score = 0;
E
Eric Dumazet 已提交
140 141
		if (inet->inet_dport) {
			if (inet->inet_dport != sport)
142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163
				return -1;
			score++;
		}
		if (!ipv6_addr_any(&np->rcv_saddr)) {
			if (!ipv6_addr_equal(&np->rcv_saddr, daddr))
				return -1;
			score++;
		}
		if (!ipv6_addr_any(&np->daddr)) {
			if (!ipv6_addr_equal(&np->daddr, saddr))
				return -1;
			score++;
		}
		if (sk->sk_bound_dev_if) {
			if (sk->sk_bound_dev_if != dif)
				return -1;
			score++;
		}
	}
	return score;
}

E
Eric Dumazet 已提交
164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207
#define SCORE2_MAX (1 + 1 + 1)
static inline int compute_score2(struct sock *sk, struct net *net,
				const struct in6_addr *saddr, __be16 sport,
				const struct in6_addr *daddr, unsigned short hnum,
				int dif)
{
	int score = -1;

	if (net_eq(sock_net(sk), net) && udp_sk(sk)->udp_port_hash == hnum &&
			sk->sk_family == PF_INET6) {
		struct ipv6_pinfo *np = inet6_sk(sk);
		struct inet_sock *inet = inet_sk(sk);

		if (!ipv6_addr_equal(&np->rcv_saddr, daddr))
			return -1;
		score = 0;
		if (inet->inet_dport) {
			if (inet->inet_dport != sport)
				return -1;
			score++;
		}
		if (!ipv6_addr_any(&np->daddr)) {
			if (!ipv6_addr_equal(&np->daddr, saddr))
				return -1;
			score++;
		}
		if (sk->sk_bound_dev_if) {
			if (sk->sk_bound_dev_if != dif)
				return -1;
			score++;
		}
	}
	return score;
}


/* called with read_rcu_lock() */
static struct sock *udp6_lib_lookup2(struct net *net,
		const struct in6_addr *saddr, __be16 sport,
		const struct in6_addr *daddr, unsigned int hnum, int dif,
		struct udp_hslot *hslot2, unsigned int slot2)
{
	struct sock *sk, *result;
	struct hlist_nulls_node *node;
208 209
	int score, badness, matches = 0, reuseport = 0;
	u32 hash = 0;
E
Eric Dumazet 已提交
210 211 212 213 214 215 216 217 218 219

begin:
	result = NULL;
	badness = -1;
	udp_portaddr_for_each_entry_rcu(sk, node, &hslot2->head) {
		score = compute_score2(sk, net, saddr, sport,
				      daddr, hnum, dif);
		if (score > badness) {
			result = sk;
			badness = score;
220 221 222 223 224 225
			reuseport = sk->sk_reuseport;
			if (reuseport) {
				hash = inet6_ehashfn(net, daddr, hnum,
						     saddr, sport);
				matches = 1;
			} else if (score == SCORE2_MAX)
E
Eric Dumazet 已提交
226
				goto exact_match;
227 228 229 230 231
		} else if (score == badness && reuseport) {
			matches++;
			if (((u64)hash * matches) >> 32 == 0)
				result = sk;
			hash = next_pseudo_random32(hash);
E
Eric Dumazet 已提交
232 233 234 235 236 237 238 239 240 241 242 243
		}
	}
	/*
	 * if the nulls value we got at the end of this lookup is
	 * not the expected one, we must restart lookup.
	 * We probably met an item that was moved to another chain.
	 */
	if (get_nulls_value(node) != slot2)
		goto begin;

	if (result) {
exact_match:
E
Eric Dumazet 已提交
244
		if (unlikely(!atomic_inc_not_zero_hint(&result->sk_refcnt, 2)))
E
Eric Dumazet 已提交
245 246 247 248 249 250 251 252 253 254
			result = NULL;
		else if (unlikely(compute_score2(result, net, saddr, sport,
				  daddr, hnum, dif) < badness)) {
			sock_put(result);
			goto begin;
		}
	}
	return result;
}

255
struct sock *__udp6_lib_lookup(struct net *net,
256 257
				      const struct in6_addr *saddr, __be16 sport,
				      const struct in6_addr *daddr, __be16 dport,
258
				      int dif, struct udp_table *udptable)
L
Linus Torvalds 已提交
259
{
260
	struct sock *sk, *result;
261
	struct hlist_nulls_node *node;
L
Linus Torvalds 已提交
262
	unsigned short hnum = ntohs(dport);
E
Eric Dumazet 已提交
263 264
	unsigned int hash2, slot2, slot = udp_hashfn(net, hnum, udptable->mask);
	struct udp_hslot *hslot2, *hslot = &udptable->hash[slot];
265 266
	int score, badness, matches = 0, reuseport = 0;
	u32 hash = 0;
267

268
	rcu_read_lock();
E
Eric Dumazet 已提交
269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
	if (hslot->count > 10) {
		hash2 = udp6_portaddr_hash(net, daddr, hnum);
		slot2 = hash2 & udptable->mask;
		hslot2 = &udptable->hash2[slot2];
		if (hslot->count < hslot2->count)
			goto begin;

		result = udp6_lib_lookup2(net, saddr, sport,
					  daddr, hnum, dif,
					  hslot2, slot2);
		if (!result) {
			hash2 = udp6_portaddr_hash(net, &in6addr_any, hnum);
			slot2 = hash2 & udptable->mask;
			hslot2 = &udptable->hash2[slot2];
			if (hslot->count < hslot2->count)
				goto begin;

286 287
			result = udp6_lib_lookup2(net, saddr, sport,
						  &in6addr_any, hnum, dif,
E
Eric Dumazet 已提交
288 289 290 291 292
						  hslot2, slot2);
		}
		rcu_read_unlock();
		return result;
	}
293 294 295
begin:
	result = NULL;
	badness = -1;
296
	sk_nulls_for_each_rcu(sk, node, &hslot->head) {
297 298 299 300
		score = compute_score(sk, net, hnum, saddr, sport, daddr, dport, dif);
		if (score > badness) {
			result = sk;
			badness = score;
301 302 303 304 305 306 307 308 309 310 311
			reuseport = sk->sk_reuseport;
			if (reuseport) {
				hash = inet6_ehashfn(net, daddr, hnum,
						     saddr, sport);
				matches = 1;
			}
		} else if (score == badness && reuseport) {
			matches++;
			if (((u64)hash * matches) >> 32 == 0)
				result = sk;
			hash = next_pseudo_random32(hash);
L
Linus Torvalds 已提交
312 313
		}
	}
314 315 316 317 318
	/*
	 * if the nulls value we got at the end of this lookup is
	 * not the expected one, we must restart lookup.
	 * We probably met an item that was moved to another chain.
	 */
E
Eric Dumazet 已提交
319
	if (get_nulls_value(node) != slot)
320 321
		goto begin;

322
	if (result) {
E
Eric Dumazet 已提交
323
		if (unlikely(!atomic_inc_not_zero_hint(&result->sk_refcnt, 2)))
324 325 326 327 328 329 330 331
			result = NULL;
		else if (unlikely(compute_score(result, net, hnum, saddr, sport,
					daddr, dport, dif) < badness)) {
			sock_put(result);
			goto begin;
		}
	}
	rcu_read_unlock();
L
Linus Torvalds 已提交
332 333
	return result;
}
334
EXPORT_SYMBOL_GPL(__udp6_lib_lookup);
L
Linus Torvalds 已提交
335

336 337
static struct sock *__udp6_lib_lookup_skb(struct sk_buff *skb,
					  __be16 sport, __be16 dport,
338
					  struct udp_table *udptable)
339
{
340
	struct sock *sk;
341
	const struct ipv6hdr *iph = ipv6_hdr(skb);
342

343 344
	if (unlikely(sk = skb_steal_sock(skb)))
		return sk;
E
Eric Dumazet 已提交
345 346 347
	return __udp6_lib_lookup(dev_net(skb_dst(skb)->dev), &iph->saddr, sport,
				 &iph->daddr, dport, inet6_iif(skb),
				 udptable);
348 349
}

350 351 352 353 354 355 356 357
struct sock *udp6_lib_lookup(struct net *net, const struct in6_addr *saddr, __be16 sport,
			     const struct in6_addr *daddr, __be16 dport, int dif)
{
	return __udp6_lib_lookup(net, saddr, sport, daddr, dport, dif, &udp_table);
}
EXPORT_SYMBOL_GPL(udp6_lib_lookup);


L
Linus Torvalds 已提交
358 359 360 361 362
/*
 * 	This should be easy, if there is something there we
 * 	return it, otherwise we block.
 */

363
int udpv6_recvmsg(struct kiocb *iocb, struct sock *sk,
L
Linus Torvalds 已提交
364 365 366 367 368
		  struct msghdr *msg, size_t len,
		  int noblock, int flags, int *addr_len)
{
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct inet_sock *inet = inet_sk(sk);
369
	struct sk_buff *skb;
370
	unsigned int ulen, copied;
371
	int peeked, off = 0;
372 373
	int err;
	int is_udplite = IS_UDPLITE(sk);
374
	int is_udp4;
375
	bool slow;
L
Linus Torvalds 已提交
376

377
	if (addr_len)
378
		*addr_len = sizeof(struct sockaddr_in6);
379

L
Linus Torvalds 已提交
380 381 382
	if (flags & MSG_ERRQUEUE)
		return ipv6_recv_error(sk, msg, len);

383 384 385
	if (np->rxpmtu && np->rxopt.bits.rxpmtu)
		return ipv6_recv_rxpmtu(sk, msg, len);

L
Linus Torvalds 已提交
386
try_again:
387
	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
388
				  &peeked, &off, &err);
L
Linus Torvalds 已提交
389 390 391
	if (!skb)
		goto out;

392
	ulen = skb->len - sizeof(struct udphdr);
393 394 395 396
	copied = len;
	if (copied > ulen)
		copied = ulen;
	else if (copied < ulen)
397
		msg->msg_flags |= MSG_TRUNC;
L
Linus Torvalds 已提交
398

399 400
	is_udp4 = (skb->protocol == htons(ETH_P_IP));

401
	/*
402 403 404
	 * If checksum is needed at all, try to do it while copying the
	 * data.  If the data is truncated, or if we only want a partial
	 * coverage checksum (UDP-Lite), do it before the copy.
405 406
	 */

407
	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
408
		if (udp_lib_checksum_complete(skb))
L
Linus Torvalds 已提交
409
			goto csum_copy_err;
410 411
	}

412
	if (skb_csum_unnecessary(skb))
413
		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
414
					      msg->msg_iov, copied);
415
	else {
L
Linus Torvalds 已提交
416 417 418 419
		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);
		if (err == -EINVAL)
			goto csum_copy_err;
	}
420 421
	if (unlikely(err)) {
		trace_kfree_skb(skb, udpv6_recvmsg);
422 423 424 425 426 427 428 429 430 431 432
		if (!peeked) {
			atomic_inc(&sk->sk_drops);
			if (is_udp4)
				UDP_INC_STATS_USER(sock_net(sk),
						   UDP_MIB_INERRORS,
						   is_udplite);
			else
				UDP6_INC_STATS_USER(sock_net(sk),
						    UDP_MIB_INERRORS,
						    is_udplite);
		}
L
Linus Torvalds 已提交
433
		goto out_free;
434
	}
435 436 437 438 439 440 441 442
	if (!peeked) {
		if (is_udp4)
			UDP_INC_STATS_USER(sock_net(sk),
					UDP_MIB_INDATAGRAMS, is_udplite);
		else
			UDP6_INC_STATS_USER(sock_net(sk),
					UDP_MIB_INDATAGRAMS, is_udplite);
	}
443

444
	sock_recv_ts_and_drops(msg, sk, skb);
L
Linus Torvalds 已提交
445 446 447 448

	/* Copy the address. */
	if (msg->msg_name) {
		struct sockaddr_in6 *sin6;
449

L
Linus Torvalds 已提交
450 451
		sin6 = (struct sockaddr_in6 *) msg->msg_name;
		sin6->sin6_family = AF_INET6;
452
		sin6->sin6_port = udp_hdr(skb)->source;
L
Linus Torvalds 已提交
453 454
		sin6->sin6_flowinfo = 0;

455
		if (is_udp4) {
B
Brian Haley 已提交
456 457
			ipv6_addr_set_v4mapped(ip_hdr(skb)->saddr,
					       &sin6->sin6_addr);
458 459
			sin6->sin6_scope_id = 0;
		} else {
A
Alexey Dobriyan 已提交
460
			sin6->sin6_addr = ipv6_hdr(skb)->saddr;
461 462 463
			sin6->sin6_scope_id =
				ipv6_iface_scope_id(&sin6->sin6_addr,
						    IP6CB(skb)->iif);
L
Linus Torvalds 已提交
464 465 466
		}

	}
467
	if (is_udp4) {
L
Linus Torvalds 已提交
468 469 470 471
		if (inet->cmsg_flags)
			ip_cmsg_recv(msg, skb);
	} else {
		if (np->rxopt.all)
472
			ip6_datagram_recv_ctl(sk, msg, skb);
473
	}
L
Linus Torvalds 已提交
474

475
	err = copied;
L
Linus Torvalds 已提交
476
	if (flags & MSG_TRUNC)
477
		err = ulen;
L
Linus Torvalds 已提交
478 479

out_free:
480
	skb_free_datagram_locked(sk, skb);
L
Linus Torvalds 已提交
481 482 483 484
out:
	return err;

csum_copy_err:
485
	slow = lock_sock_fast(sk);
486
	if (!skb_kill_datagram(sk, skb, flags)) {
487 488 489
		if (is_udp4) {
			UDP_INC_STATS_USER(sock_net(sk),
					UDP_MIB_CSUMERRORS, is_udplite);
490 491
			UDP_INC_STATS_USER(sock_net(sk),
					UDP_MIB_INERRORS, is_udplite);
492 493 494
		} else {
			UDP6_INC_STATS_USER(sock_net(sk),
					UDP_MIB_CSUMERRORS, is_udplite);
495 496
			UDP6_INC_STATS_USER(sock_net(sk),
					UDP_MIB_INERRORS, is_udplite);
497
		}
498
	}
499
	unlock_sock_fast(sk, slow);
L
Linus Torvalds 已提交
500

501
	if (noblock)
L
Linus Torvalds 已提交
502
		return -EAGAIN;
503 504 505

	/* starting over for a new packet */
	msg->msg_flags &= ~MSG_TRUNC;
L
Linus Torvalds 已提交
506 507 508
	goto try_again;
}

509
void __udp6_lib_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
510
		    u8 type, u8 code, int offset, __be32 info,
511
		    struct udp_table *udptable)
L
Linus Torvalds 已提交
512 513
{
	struct ipv6_pinfo *np;
514 515 516
	const struct ipv6hdr *hdr = (const struct ipv6hdr *)skb->data;
	const struct in6_addr *saddr = &hdr->saddr;
	const struct in6_addr *daddr = &hdr->daddr;
L
Linus Torvalds 已提交
517 518 519 520
	struct udphdr *uh = (struct udphdr*)(skb->data+offset);
	struct sock *sk;
	int err;

521
	sk = __udp6_lib_lookup(dev_net(skb->dev), daddr, uh->dest,
522
			       saddr, uh->source, inet6_iif(skb), udptable);
L
Linus Torvalds 已提交
523 524 525
	if (sk == NULL)
		return;

526 527
	if (type == ICMPV6_PKT_TOOBIG)
		ip6_sk_update_pmtu(skb, sk, info);
528 529
	if (type == NDISC_REDIRECT)
		ip6_sk_redirect(skb, sk);
530

L
Linus Torvalds 已提交
531 532 533 534 535 536 537 538
	np = inet6_sk(sk);

	if (!icmpv6_err_convert(type, code, &err) && !np->recverr)
		goto out;

	if (sk->sk_state != TCP_ESTABLISHED && !np->recverr)
		goto out;

539
	if (np->recverr)
L
Linus Torvalds 已提交
540
		ipv6_icmp_error(sk, skb, err, uh->dest, ntohl(info), (u8 *)(uh+1));
541

L
Linus Torvalds 已提交
542 543 544 545 546 547
	sk->sk_err = err;
	sk->sk_error_report(sk);
out:
	sock_put(sk);
}

548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569
static int __udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
{
	int rc;

	if (!ipv6_addr_any(&inet6_sk(sk)->daddr))
		sock_rps_save_rxhash(sk, skb);

	rc = sock_queue_rcv_skb(sk, skb);
	if (rc < 0) {
		int is_udplite = IS_UDPLITE(sk);

		/* Note that an ENOMEM error is charged twice */
		if (rc == -ENOMEM)
			UDP6_INC_STATS_BH(sock_net(sk),
					UDP_MIB_RCVBUFERRORS, is_udplite);
		UDP6_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
		kfree_skb(skb);
		return -1;
	}
	return 0;
}

570
static __inline__ void udpv6_err(struct sk_buff *skb,
571 572
				 struct inet6_skb_parm *opt, u8 type,
				 u8 code, int offset, __be32 info     )
573
{
574
	__udp6_lib_err(skb, opt, type, code, offset, info, &udp_table);
575 576
}

577 578 579 580 581 582 583 584
static struct static_key udpv6_encap_needed __read_mostly;
void udpv6_encap_enable(void)
{
	if (!static_key_enabled(&udpv6_encap_needed))
		static_key_slow_inc(&udpv6_encap_needed);
}
EXPORT_SYMBOL(udpv6_encap_enable);

585
int udpv6_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
L
Linus Torvalds 已提交
586
{
587
	struct udp_sock *up = udp_sk(sk);
588
	int rc;
W
Wang Chen 已提交
589
	int is_udplite = IS_UDPLITE(sk);
590

591 592
	if (!xfrm6_policy_check(sk, XFRM_POLICY_IN, skb))
		goto drop;
L
Linus Torvalds 已提交
593

594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624
	if (static_key_false(&udpv6_encap_needed) && up->encap_type) {
		int (*encap_rcv)(struct sock *sk, struct sk_buff *skb);

		/*
		 * This is an encapsulation socket so pass the skb to
		 * the socket's udp_encap_rcv() hook. Otherwise, just
		 * fall through and pass this up the UDP socket.
		 * up->encap_rcv() returns the following value:
		 * =0 if skb was successfully passed to the encap
		 *    handler or was discarded by it.
		 * >0 if skb should be passed on to UDP.
		 * <0 if skb should be resubmitted as proto -N
		 */

		/* if we're overly short, let UDP handle it */
		encap_rcv = ACCESS_ONCE(up->encap_rcv);
		if (skb->len > sizeof(struct udphdr) && encap_rcv != NULL) {
			int ret;

			ret = encap_rcv(sk, skb);
			if (ret <= 0) {
				UDP_INC_STATS_BH(sock_net(sk),
						 UDP_MIB_INDATAGRAMS,
						 is_udplite);
				return -ret;
			}
		}

		/* FALLTHROUGH -- it's a UDP Packet */
	}

625 626 627
	/*
	 * UDP-Lite specific tests, ignored on UDP sockets (see net/ipv4/udp.c).
	 */
W
Wang Chen 已提交
628
	if ((is_udplite & UDPLITE_RECV_CC)  &&  UDP_SKB_CB(skb)->partial_cov) {
629 630 631 632 633 634 635 636 637 638 639 640 641

		if (up->pcrlen == 0) {          /* full coverage was set  */
			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: partial coverage"
				" %d while full coverage %d requested\n",
				UDP_SKB_CB(skb)->cscov, skb->len);
			goto drop;
		}
		if (UDP_SKB_CB(skb)->cscov  <  up->pcrlen) {
			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE6: coverage %d "
						    "too small, need min %d\n",
				       UDP_SKB_CB(skb)->cscov, up->pcrlen);
			goto drop;
		}
L
Linus Torvalds 已提交
642 643
	}

644
	if (rcu_access_pointer(sk->sk_filter)) {
645
		if (udp_lib_checksum_complete(skb))
646
			goto csum_error;
647
	}
648

649 650 651
	if (sk_rcvqueues_full(sk, skb, sk->sk_rcvbuf))
		goto drop;

652
	skb_dst_drop(skb);
653

654 655 656 657 658 659 660 661 662
	bh_lock_sock(sk);
	rc = 0;
	if (!sock_owned_by_user(sk))
		rc = __udpv6_queue_rcv_skb(sk, skb);
	else if (sk_add_backlog(sk, skb, sk->sk_rcvbuf)) {
		bh_unlock_sock(sk);
		goto drop;
	}
	bh_unlock_sock(sk);
663 664

	return rc;
665 666
csum_error:
	UDP6_INC_STATS_BH(sock_net(sk), UDP_MIB_CSUMERRORS, is_udplite);
667
drop:
668
	UDP6_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
669
	atomic_inc(&sk->sk_drops);
670 671
	kfree_skb(skb);
	return -1;
L
Linus Torvalds 已提交
672 673
}

674
static struct sock *udp_v6_mcast_next(struct net *net, struct sock *sk,
675 676
				      __be16 loc_port, const struct in6_addr *loc_addr,
				      __be16 rmt_port, const struct in6_addr *rmt_addr,
L
Linus Torvalds 已提交
677 678
				      int dif)
{
679
	struct hlist_nulls_node *node;
L
Linus Torvalds 已提交
680 681 682
	struct sock *s = sk;
	unsigned short num = ntohs(loc_port);

683
	sk_nulls_for_each_from(s, node) {
L
Linus Torvalds 已提交
684 685
		struct inet_sock *inet = inet_sk(s);

686
		if (!net_eq(sock_net(s), net))
687 688
			continue;

689 690
		if (udp_sk(s)->udp_port_hash == num &&
		    s->sk_family == PF_INET6) {
L
Linus Torvalds 已提交
691
			struct ipv6_pinfo *np = inet6_sk(s);
E
Eric Dumazet 已提交
692 693
			if (inet->inet_dport) {
				if (inet->inet_dport != rmt_port)
L
Linus Torvalds 已提交
694 695 696 697 698 699 700 701 702 703
					continue;
			}
			if (!ipv6_addr_any(&np->daddr) &&
			    !ipv6_addr_equal(&np->daddr, rmt_addr))
				continue;

			if (s->sk_bound_dev_if && s->sk_bound_dev_if != dif)
				continue;

			if (!ipv6_addr_any(&np->rcv_saddr)) {
704 705
				if (!ipv6_addr_equal(&np->rcv_saddr, loc_addr))
					continue;
L
Linus Torvalds 已提交
706
			}
S
Stephen Hemminger 已提交
707
			if (!inet6_mc_check(s, loc_addr, rmt_addr))
L
Linus Torvalds 已提交
708 709 710 711 712 713 714
				continue;
			return s;
		}
	}
	return NULL;
}

715 716 717
static void flush_stack(struct sock **stack, unsigned int count,
			struct sk_buff *skb, unsigned int final)
{
718
	struct sk_buff *skb1 = NULL;
719
	struct sock *sk;
720
	unsigned int i;
721 722

	for (i = 0; i < count; i++) {
723
		sk = stack[i];
724 725 726 727 728 729 730 731
		if (likely(skb1 == NULL))
			skb1 = (i == final) ? skb : skb_clone(skb, GFP_ATOMIC);
		if (!skb1) {
			atomic_inc(&sk->sk_drops);
			UDP6_INC_STATS_BH(sock_net(sk), UDP_MIB_RCVBUFERRORS,
					  IS_UDPLITE(sk));
			UDP6_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS,
					  IS_UDPLITE(sk));
732
		}
733 734 735

		if (skb1 && udpv6_queue_rcv_skb(sk, skb1) <= 0)
			skb1 = NULL;
736
	}
737 738
	if (unlikely(skb1))
		kfree_skb(skb1);
739
}
L
Linus Torvalds 已提交
740 741 742 743
/*
 * Note: called only from the BH handler context,
 * so we don't need to lock the hashes.
 */
744
static int __udp6_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
745
		const struct in6_addr *saddr, const struct in6_addr *daddr,
746
		struct udp_table *udptable)
L
Linus Torvalds 已提交
747
{
748
	struct sock *sk, *stack[256 / sizeof(struct sock *)];
749
	const struct udphdr *uh = udp_hdr(skb);
750
	struct udp_hslot *hslot = udp_hashslot(udptable, net, ntohs(uh->dest));
L
Linus Torvalds 已提交
751
	int dif;
752
	unsigned int i, count = 0;
L
Linus Torvalds 已提交
753

754
	spin_lock(&hslot->lock);
755
	sk = sk_nulls_head(&hslot->head);
756
	dif = inet6_iif(skb);
757
	sk = udp_v6_mcast_next(net, sk, uh->dest, daddr, uh->source, saddr, dif);
758 759 760 761 762 763 764 765 766
	while (sk) {
		stack[count++] = sk;
		sk = udp_v6_mcast_next(net, sk_nulls_next(sk), uh->dest, daddr,
				       uh->source, saddr, dif);
		if (unlikely(count == ARRAY_SIZE(stack))) {
			if (!sk)
				break;
			flush_stack(stack, count, skb, ~0);
			count = 0;
H
Hideo Aoki 已提交
767
		}
L
Linus Torvalds 已提交
768
	}
769 770 771 772 773 774
	/*
	 * before releasing the lock, we must take reference on sockets
	 */
	for (i = 0; i < count; i++)
		sock_hold(stack[i]);

775
	spin_unlock(&hslot->lock);
776 777 778 779 780 781 782 783 784

	if (count) {
		flush_stack(stack, count, skb, count - 1);

		for (i = 0; i < count; i++)
			sock_put(stack[i]);
	} else {
		kfree_skb(skb);
	}
785
	return 0;
L
Linus Torvalds 已提交
786 787
}

788
int __udp6_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
789
		   int proto)
L
Linus Torvalds 已提交
790
{
791
	struct net *net = dev_net(skb->dev);
L
Linus Torvalds 已提交
792
	struct sock *sk;
793
	struct udphdr *uh;
794
	const struct in6_addr *saddr, *daddr;
L
Linus Torvalds 已提交
795 796 797
	u32 ulen = 0;

	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
798
		goto discard;
L
Linus Torvalds 已提交
799

800 801
	saddr = &ipv6_hdr(skb)->saddr;
	daddr = &ipv6_hdr(skb)->daddr;
802
	uh = udp_hdr(skb);
L
Linus Torvalds 已提交
803 804

	ulen = ntohs(uh->len);
805 806
	if (ulen > skb->len)
		goto short_packet;
L
Linus Torvalds 已提交
807

808 809
	if (proto == IPPROTO_UDP) {
		/* UDP validates ulen. */
L
Linus Torvalds 已提交
810

811 812 813
		/* Check for jumbo payload */
		if (ulen == 0)
			ulen = skb->len;
L
Linus Torvalds 已提交
814

815 816
		if (ulen < sizeof(*uh))
			goto short_packet;
L
Linus Torvalds 已提交
817

818 819 820
		if (ulen < skb->len) {
			if (pskb_trim_rcsum(skb, ulen))
				goto short_packet;
821 822
			saddr = &ipv6_hdr(skb)->saddr;
			daddr = &ipv6_hdr(skb)->daddr;
823
			uh = udp_hdr(skb);
824 825
		}
	}
L
Linus Torvalds 已提交
826

827
	if (udp6_csum_init(skb, uh, proto))
828
		goto csum_error;
829

830 831
	/*
	 *	Multicast receive code
L
Linus Torvalds 已提交
832
	 */
833
	if (ipv6_addr_is_multicast(daddr))
834 835
		return __udp6_lib_mcast_deliver(net, skb,
				saddr, daddr, udptable);
L
Linus Torvalds 已提交
836 837 838

	/* Unicast */

839
	/*
L
Linus Torvalds 已提交
840 841 842
	 * check socket cache ... must talk to Alan about his plans
	 * for sock caches... i'll skip this for now.
	 */
843
	sk = __udp6_lib_lookup_skb(skb, uh->source, uh->dest, udptable);
844
	if (sk != NULL) {
845 846 847 848
		int ret;

		sk_mark_ll(sk, skb);
		ret = udpv6_queue_rcv_skb(sk, skb);
849
		sock_put(sk);
L
Linus Torvalds 已提交
850

851 852 853 854 855
		/* a return value > 0 means to resubmit the input, but
		 * it wants the return to be -protocol, or 0
		 */
		if (ret > 0)
			return -ret;
L
Linus Torvalds 已提交
856

S
Stephen Hemminger 已提交
857
		return 0;
L
Linus Torvalds 已提交
858
	}
859

860
	if (!xfrm6_policy_check(NULL, XFRM_POLICY_IN, skb))
861
		goto discard;
862 863

	if (udp_lib_checksum_complete(skb))
864
		goto csum_error;
865 866 867 868 869

	UDP6_INC_STATS_BH(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
	icmpv6_send(skb, ICMPV6_DEST_UNREACH, ICMPV6_PORT_UNREACH, 0);

	kfree_skb(skb);
S
Stephen Hemminger 已提交
870
	return 0;
L
Linus Torvalds 已提交
871

872
short_packet:
873
	LIMIT_NETDEBUG(KERN_DEBUG "UDP%sv6: short packet: From [%pI6c]:%u %d/%d to [%pI6c]:%u\n",
874
		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
875 876 877 878 879 880
		       saddr,
		       ntohs(uh->source),
		       ulen,
		       skb->len,
		       daddr,
		       ntohs(uh->dest));
881 882 883
	goto discard;
csum_error:
	UDP6_INC_STATS_BH(net, UDP_MIB_CSUMERRORS, proto == IPPROTO_UDPLITE);
L
Linus Torvalds 已提交
884
discard:
885
	UDP6_INC_STATS_BH(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
L
Linus Torvalds 已提交
886
	kfree_skb(skb);
S
Stephen Hemminger 已提交
887
	return 0;
L
Linus Torvalds 已提交
888
}
889

890
static __inline__ int udpv6_rcv(struct sk_buff *skb)
891
{
892
	return __udp6_lib_rcv(skb, &udp_table, IPPROTO_UDP);
893 894
}

L
Linus Torvalds 已提交
895 896 897 898 899 900 901
/*
 * Throw away all pending data and cancel the corking. Socket is locked.
 */
static void udp_v6_flush_pending_frames(struct sock *sk)
{
	struct udp_sock *up = udp_sk(sk);

902 903 904
	if (up->pending == AF_INET)
		udp_flush_pending_frames(sk);
	else if (up->pending) {
L
Linus Torvalds 已提交
905 906 907
		up->len = 0;
		up->pending = 0;
		ip6_flush_pending_frames(sk);
908
	}
L
Linus Torvalds 已提交
909 910
}

911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951
/**
 * 	udp6_hwcsum_outgoing  -  handle outgoing HW checksumming
 * 	@sk: 	socket we are sending on
 * 	@skb: 	sk_buff containing the filled-in UDP header
 * 	        (checksum field must be zeroed out)
 */
static void udp6_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb,
				 const struct in6_addr *saddr,
				 const struct in6_addr *daddr, int len)
{
	unsigned int offset;
	struct udphdr *uh = udp_hdr(skb);
	__wsum csum = 0;

	if (skb_queue_len(&sk->sk_write_queue) == 1) {
		/* Only one fragment on the socket.  */
		skb->csum_start = skb_transport_header(skb) - skb->head;
		skb->csum_offset = offsetof(struct udphdr, check);
		uh->check = ~csum_ipv6_magic(saddr, daddr, len, IPPROTO_UDP, 0);
	} else {
		/*
		 * HW-checksum won't work as there are two or more
		 * fragments on the socket so that all csums of sk_buffs
		 * should be together
		 */
		offset = skb_transport_offset(skb);
		skb->csum = skb_checksum(skb, offset, skb->len - offset, 0);

		skb->ip_summed = CHECKSUM_NONE;

		skb_queue_walk(&sk->sk_write_queue, skb) {
			csum = csum_add(csum, skb->csum);
		}

		uh->check = csum_ipv6_magic(saddr, daddr, len, IPPROTO_UDP,
					    csum);
		if (uh->check == 0)
			uh->check = CSUM_MANGLED_0;
	}
}

L
Linus Torvalds 已提交
952 953 954 955
/*
 *	Sending
 */

956
static int udp_v6_push_pending_frames(struct sock *sk)
L
Linus Torvalds 已提交
957 958 959
{
	struct sk_buff *skb;
	struct udphdr *uh;
960
	struct udp_sock  *up = udp_sk(sk);
L
Linus Torvalds 已提交
961
	struct inet_sock *inet = inet_sk(sk);
962
	struct flowi6 *fl6;
L
Linus Torvalds 已提交
963
	int err = 0;
W
Wang Chen 已提交
964
	int is_udplite = IS_UDPLITE(sk);
965
	__wsum csum = 0;
L
Linus Torvalds 已提交
966

967 968 969 970 971
	if (up->pending == AF_INET)
		return udp_push_pending_frames(sk);

	fl6 = &inet->cork.fl.u.ip6;

L
Linus Torvalds 已提交
972 973 974 975 976 977 978
	/* Grab the skbuff where UDP header space exists. */
	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
		goto out;

	/*
	 * Create a UDP header
	 */
979
	uh = udp_hdr(skb);
980 981
	uh->source = fl6->fl6_sport;
	uh->dest = fl6->fl6_dport;
L
Linus Torvalds 已提交
982 983 984
	uh->len = htons(up->len);
	uh->check = 0;

W
Wang Chen 已提交
985
	if (is_udplite)
986
		csum = udplite_csum_outgoing(sk, skb);
987
	else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */
988
		udp6_hwcsum_outgoing(sk, skb, &fl6->saddr, &fl6->daddr,
989 990 991
				     up->len);
		goto send;
	} else
992
		csum = udp_csum_outgoing(sk, skb);
L
Linus Torvalds 已提交
993

994
	/* add protocol-dependent pseudo-header */
995 996
	uh->check = csum_ipv6_magic(&fl6->saddr, &fl6->daddr,
				    up->len, fl6->flowi6_proto, csum);
L
Linus Torvalds 已提交
997
	if (uh->check == 0)
998
		uh->check = CSUM_MANGLED_0;
L
Linus Torvalds 已提交
999

1000
send:
L
Linus Torvalds 已提交
1001
	err = ip6_push_pending_frames(sk);
E
Eric Dumazet 已提交
1002 1003 1004 1005 1006 1007 1008 1009 1010
	if (err) {
		if (err == -ENOBUFS && !inet6_sk(sk)->recverr) {
			UDP6_INC_STATS_USER(sock_net(sk),
					    UDP_MIB_SNDBUFERRORS, is_udplite);
			err = 0;
		}
	} else
		UDP6_INC_STATS_USER(sock_net(sk),
				    UDP_MIB_OUTDATAGRAMS, is_udplite);
L
Linus Torvalds 已提交
1011 1012 1013 1014 1015 1016
out:
	up->len = 0;
	up->pending = 0;
	return err;
}

1017
int udpv6_sendmsg(struct kiocb *iocb, struct sock *sk,
L
Linus Torvalds 已提交
1018 1019 1020 1021 1022 1023 1024
		  struct msghdr *msg, size_t len)
{
	struct ipv6_txoptions opt_space;
	struct udp_sock *up = udp_sk(sk);
	struct inet_sock *inet = inet_sk(sk);
	struct ipv6_pinfo *np = inet6_sk(sk);
	struct sockaddr_in6 *sin6 = (struct sockaddr_in6 *) msg->msg_name;
1025
	struct in6_addr *daddr, *final_p, final;
L
Linus Torvalds 已提交
1026 1027
	struct ipv6_txoptions *opt = NULL;
	struct ip6_flowlabel *flowlabel = NULL;
1028
	struct flowi6 fl6;
L
Linus Torvalds 已提交
1029 1030 1031 1032
	struct dst_entry *dst;
	int addr_len = msg->msg_namelen;
	int ulen = len;
	int hlimit = -1;
1033
	int tclass = -1;
1034
	int dontfrag = -1;
L
Linus Torvalds 已提交
1035 1036
	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
	int err;
1037
	int connected = 0;
W
Wang Chen 已提交
1038
	int is_udplite = IS_UDPLITE(sk);
1039
	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
L
Linus Torvalds 已提交
1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065

	/* destination address check */
	if (sin6) {
		if (addr_len < offsetof(struct sockaddr, sa_data))
			return -EINVAL;

		switch (sin6->sin6_family) {
		case AF_INET6:
			if (addr_len < SIN6_LEN_RFC2133)
				return -EINVAL;
			daddr = &sin6->sin6_addr;
			break;
		case AF_INET:
			goto do_udp_sendmsg;
		case AF_UNSPEC:
			msg->msg_name = sin6 = NULL;
			msg->msg_namelen = addr_len = 0;
			daddr = NULL;
			break;
		default:
			return -EINVAL;
		}
	} else if (!up->pending) {
		if (sk->sk_state != TCP_ESTABLISHED)
			return -EDESTADDRREQ;
		daddr = &np->daddr;
1066
	} else
L
Linus Torvalds 已提交
1067 1068 1069
		daddr = NULL;

	if (daddr) {
B
Brian Haley 已提交
1070
		if (ipv6_addr_v4mapped(daddr)) {
L
Linus Torvalds 已提交
1071 1072
			struct sockaddr_in sin;
			sin.sin_family = AF_INET;
E
Eric Dumazet 已提交
1073
			sin.sin_port = sin6 ? sin6->sin6_port : inet->inet_dport;
L
Linus Torvalds 已提交
1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087
			sin.sin_addr.s_addr = daddr->s6_addr32[3];
			msg->msg_name = &sin;
			msg->msg_namelen = sizeof(sin);
do_udp_sendmsg:
			if (__ipv6_only_sock(sk))
				return -ENETUNREACH;
			return udp_sendmsg(iocb, sk, msg, len);
		}
	}

	if (up->pending == AF_INET)
		return udp_sendmsg(iocb, sk, msg, len);

	/* Rough check on arithmetic overflow,
1088
	   better check is made in ip6_append_data().
L
Linus Torvalds 已提交
1089 1090 1091
	   */
	if (len > INT_MAX - sizeof(struct udphdr))
		return -EMSGSIZE;
1092

L
Linus Torvalds 已提交
1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110
	if (up->pending) {
		/*
		 * There are pending frames.
		 * The socket lock must be held while it's corked.
		 */
		lock_sock(sk);
		if (likely(up->pending)) {
			if (unlikely(up->pending != AF_INET6)) {
				release_sock(sk);
				return -EAFNOSUPPORT;
			}
			dst = NULL;
			goto do_append_data;
		}
		release_sock(sk);
	}
	ulen += sizeof(struct udphdr);

1111
	memset(&fl6, 0, sizeof(fl6));
L
Linus Torvalds 已提交
1112 1113 1114 1115 1116

	if (sin6) {
		if (sin6->sin6_port == 0)
			return -EINVAL;

1117
		fl6.fl6_dport = sin6->sin6_port;
L
Linus Torvalds 已提交
1118 1119 1120
		daddr = &sin6->sin6_addr;

		if (np->sndflow) {
1121 1122 1123
			fl6.flowlabel = sin6->sin6_flowinfo&IPV6_FLOWINFO_MASK;
			if (fl6.flowlabel&IPV6_FLOWLABEL_MASK) {
				flowlabel = fl6_sock_lookup(sk, fl6.flowlabel);
L
Linus Torvalds 已提交
1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139
				if (flowlabel == NULL)
					return -EINVAL;
				daddr = &flowlabel->dst;
			}
		}

		/*
		 * Otherwise it will be difficult to maintain
		 * sk->sk_dst_cache.
		 */
		if (sk->sk_state == TCP_ESTABLISHED &&
		    ipv6_addr_equal(daddr, &np->daddr))
			daddr = &np->daddr;

		if (addr_len >= sizeof(struct sockaddr_in6) &&
		    sin6->sin6_scope_id &&
1140
		    __ipv6_addr_needs_scope_id(__ipv6_addr_type(daddr)))
1141
			fl6.flowi6_oif = sin6->sin6_scope_id;
L
Linus Torvalds 已提交
1142 1143 1144 1145
	} else {
		if (sk->sk_state != TCP_ESTABLISHED)
			return -EDESTADDRREQ;

1146
		fl6.fl6_dport = inet->inet_dport;
L
Linus Torvalds 已提交
1147
		daddr = &np->daddr;
1148
		fl6.flowlabel = np->flow_label;
1149
		connected = 1;
L
Linus Torvalds 已提交
1150 1151
	}

1152 1153
	if (!fl6.flowi6_oif)
		fl6.flowi6_oif = sk->sk_bound_dev_if;
L
Linus Torvalds 已提交
1154

1155 1156
	if (!fl6.flowi6_oif)
		fl6.flowi6_oif = np->sticky_pktinfo.ipi6_ifindex;
1157

1158
	fl6.flowi6_mark = sk->sk_mark;
B
Brian Haley 已提交
1159

L
Linus Torvalds 已提交
1160 1161 1162 1163 1164
	if (msg->msg_controllen) {
		opt = &opt_space;
		memset(opt, 0, sizeof(struct ipv6_txoptions));
		opt->tot_len = sizeof(*opt);

1165 1166
		err = ip6_datagram_send_ctl(sock_net(sk), sk, msg, &fl6, opt,
					    &hlimit, &tclass, &dontfrag);
L
Linus Torvalds 已提交
1167 1168 1169 1170
		if (err < 0) {
			fl6_sock_release(flowlabel);
			return err;
		}
1171 1172
		if ((fl6.flowlabel&IPV6_FLOWLABEL_MASK) && !flowlabel) {
			flowlabel = fl6_sock_lookup(sk, fl6.flowlabel);
L
Linus Torvalds 已提交
1173 1174 1175 1176 1177
			if (flowlabel == NULL)
				return -EINVAL;
		}
		if (!(opt->opt_nflen|opt->opt_flen))
			opt = NULL;
1178
		connected = 0;
L
Linus Torvalds 已提交
1179 1180 1181
	}
	if (opt == NULL)
		opt = np->opt;
1182 1183 1184
	if (flowlabel)
		opt = fl6_merge_options(&opt_space, flowlabel, opt);
	opt = ipv6_fixup_options(&opt_space, opt);
L
Linus Torvalds 已提交
1185

1186
	fl6.flowi6_proto = sk->sk_protocol;
1187
	if (!ipv6_addr_any(daddr))
A
Alexey Dobriyan 已提交
1188
		fl6.daddr = *daddr;
1189
	else
1190 1191
		fl6.daddr.s6_addr[15] = 0x1; /* :: means loopback (BSD'ism) */
	if (ipv6_addr_any(&fl6.saddr) && !ipv6_addr_any(&np->saddr))
A
Alexey Dobriyan 已提交
1192
		fl6.saddr = np->saddr;
1193
	fl6.fl6_sport = inet->inet_sport;
1194

1195
	final_p = fl6_update_dst(&fl6, opt, &final);
1196
	if (final_p)
1197
		connected = 0;
L
Linus Torvalds 已提交
1198

1199 1200
	if (!fl6.flowi6_oif && ipv6_addr_is_multicast(&fl6.daddr)) {
		fl6.flowi6_oif = np->mcast_oif;
1201
		connected = 0;
1202 1203
	} else if (!fl6.flowi6_oif)
		fl6.flowi6_oif = np->ucast_oif;
L
Linus Torvalds 已提交
1204

1205
	security_sk_classify_flow(sk, flowi6_to_flowi(&fl6));
V
Venkat Yekkirala 已提交
1206

1207
	dst = ip6_sk_dst_lookup_flow(sk, &fl6, final_p, true);
1208 1209 1210
	if (IS_ERR(dst)) {
		err = PTR_ERR(dst);
		dst = NULL;
L
Linus Torvalds 已提交
1211
		goto out;
1212
	}
L
Linus Torvalds 已提交
1213 1214

	if (hlimit < 0) {
1215
		if (ipv6_addr_is_multicast(&fl6.daddr))
L
Linus Torvalds 已提交
1216 1217 1218 1219
			hlimit = np->mcast_hops;
		else
			hlimit = np->hop_limit;
		if (hlimit < 0)
1220
			hlimit = ip6_dst_hoplimit(dst);
L
Linus Torvalds 已提交
1221 1222
	}

G
Gerrit Renker 已提交
1223
	if (tclass < 0)
1224 1225
		tclass = np->tclass;

1226 1227 1228
	if (dontfrag < 0)
		dontfrag = np->dontfrag;

L
Linus Torvalds 已提交
1229 1230 1231 1232 1233 1234 1235 1236 1237 1238
	if (msg->msg_flags&MSG_CONFIRM)
		goto do_confirm;
back_from_confirm:

	lock_sock(sk);
	if (unlikely(up->pending)) {
		/* The socket is already corked while preparing it. */
		/* ... which is an evident application bug. --ANK */
		release_sock(sk);

1239
		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
L
Linus Torvalds 已提交
1240 1241 1242 1243 1244 1245 1246 1247
		err = -EINVAL;
		goto out;
	}

	up->pending = AF_INET6;

do_append_data:
	up->len += ulen;
1248 1249
	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
	err = ip6_append_data(sk, getfrag, msg->msg_iov, ulen,
1250
		sizeof(struct udphdr), hlimit, tclass, opt, &fl6,
1251
		(struct rt6_info*)dst,
1252
		corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags, dontfrag);
L
Linus Torvalds 已提交
1253 1254 1255
	if (err)
		udp_v6_flush_pending_frames(sk);
	else if (!corkreq)
1256
		err = udp_v6_push_pending_frames(sk);
H
Herbert Xu 已提交
1257 1258
	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
		up->pending = 0;
L
Linus Torvalds 已提交
1259

1260 1261 1262
	if (dst) {
		if (connected) {
			ip6_dst_store(sk, dst,
1263
				      ipv6_addr_equal(&fl6.daddr, &np->daddr) ?
1264 1265
				      &np->daddr : NULL,
#ifdef CONFIG_IPV6_SUBTREES
1266
				      ipv6_addr_equal(&fl6.saddr, &np->saddr) ?
1267 1268 1269
				      &np->saddr :
#endif
				      NULL);
1270 1271 1272
		} else {
			dst_release(dst);
		}
1273
		dst = NULL;
1274 1275
	}

L
Linus Torvalds 已提交
1276 1277 1278 1279
	if (err > 0)
		err = np->recverr ? net_xmit_errno(err) : 0;
	release_sock(sk);
out:
1280
	dst_release(dst);
L
Linus Torvalds 已提交
1281
	fl6_sock_release(flowlabel);
1282
	if (!err)
L
Linus Torvalds 已提交
1283
		return len;
1284 1285 1286 1287 1288 1289 1290 1291
	/*
	 * ENOBUFS = no kernel mem, SOCK_NOSPACE = no sndbuf space.  Reporting
	 * ENOBUFS might not be good (it's not tunable per se), but otherwise
	 * we don't have a good statistic (IpOutDiscards but it can be too many
	 * things).  We could add another new stat but at least for now that
	 * seems like overkill.
	 */
	if (err == -ENOBUFS || test_bit(SOCK_NOSPACE, &sk->sk_socket->flags)) {
1292 1293
		UDP6_INC_STATS_USER(sock_net(sk),
				UDP_MIB_SNDBUFERRORS, is_udplite);
1294
	}
L
Linus Torvalds 已提交
1295 1296 1297 1298 1299 1300 1301 1302 1303 1304
	return err;

do_confirm:
	dst_confirm(dst);
	if (!(msg->msg_flags&MSG_PROBE) || len)
		goto back_from_confirm;
	err = 0;
	goto out;
}

1305
void udpv6_destroy_sock(struct sock *sk)
L
Linus Torvalds 已提交
1306
{
T
Tom Parkin 已提交
1307
	struct udp_sock *up = udp_sk(sk);
L
Linus Torvalds 已提交
1308 1309 1310 1311
	lock_sock(sk);
	udp_v6_flush_pending_frames(sk);
	release_sock(sk);

T
Tom Parkin 已提交
1312 1313 1314 1315 1316 1317 1318
	if (static_key_false(&udpv6_encap_needed) && up->encap_type) {
		void (*encap_destroy)(struct sock *sk);
		encap_destroy = ACCESS_ONCE(up->encap_destroy);
		if (encap_destroy)
			encap_destroy(sk);
	}

L
Linus Torvalds 已提交
1319 1320 1321 1322 1323 1324
	inet6_destroy_sock(sk);
}

/*
 *	Socket option code for UDP
 */
1325
int udpv6_setsockopt(struct sock *sk, int level, int optname,
1326
		     char __user *optval, unsigned int optlen)
1327
{
1328
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
1329 1330
		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
					  udp_v6_push_pending_frames);
1331
	return ipv6_setsockopt(sk, level, optname, optval, optlen);
1332 1333 1334
}

#ifdef CONFIG_COMPAT
1335
int compat_udpv6_setsockopt(struct sock *sk, int level, int optname,
1336
			    char __user *optval, unsigned int optlen)
1337
{
1338
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
1339 1340
		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
					  udp_v6_push_pending_frames);
1341
	return compat_ipv6_setsockopt(sk, level, optname, optval, optlen);
1342 1343 1344
}
#endif

1345 1346
int udpv6_getsockopt(struct sock *sk, int level, int optname,
		     char __user *optval, int __user *optlen)
1347
{
1348
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
1349
		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
1350
	return ipv6_getsockopt(sk, level, optname, optval, optlen);
1351 1352 1353
}

#ifdef CONFIG_COMPAT
1354 1355
int compat_udpv6_getsockopt(struct sock *sk, int level, int optname,
			    char __user *optval, int __user *optlen)
1356
{
1357
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
1358
		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
1359
	return compat_ipv6_getsockopt(sk, level, optname, optval, optlen);
1360 1361 1362
}
#endif

1363
static const struct inet6_protocol udpv6_protocol = {
L
Linus Torvalds 已提交
1364 1365 1366 1367 1368 1369 1370
	.handler	=	udpv6_rcv,
	.err_handler	=	udpv6_err,
	.flags		=	INET6_PROTO_NOPOLICY|INET6_PROTO_FINAL,
};

/* ------------------------------------------------------------------------ */
#ifdef CONFIG_PROC_FS
1371
int udp6_seq_show(struct seq_file *seq, void *v)
L
Linus Torvalds 已提交
1372
{
1373 1374 1375 1376 1377 1378 1379 1380 1381
	if (v == SEQ_START_TOKEN) {
		seq_puts(seq, IPV6_SEQ_DGRAM_HEADER);
	} else {
		int bucket = ((struct udp_iter_state *)seq->private)->bucket;
		struct inet_sock *inet = inet_sk(v);
		__u16 srcp = ntohs(inet->inet_sport);
		__u16 destp = ntohs(inet->inet_dport);
		ip6_dgram_sock_seq_show(seq, v, srcp, destp, bucket);
	}
L
Linus Torvalds 已提交
1382 1383 1384
	return 0;
}

1385 1386 1387 1388 1389 1390 1391 1392
static const struct file_operations udp6_afinfo_seq_fops = {
	.owner    = THIS_MODULE,
	.open     = udp_seq_open,
	.read     = seq_read,
	.llseek   = seq_lseek,
	.release  = seq_release_net
};

L
Linus Torvalds 已提交
1393 1394 1395
static struct udp_seq_afinfo udp6_seq_afinfo = {
	.name		= "udp6",
	.family		= AF_INET6,
1396
	.udp_table	= &udp_table,
1397
	.seq_fops	= &udp6_afinfo_seq_fops,
1398 1399 1400
	.seq_ops	= {
		.show		= udp6_seq_show,
	},
L
Linus Torvalds 已提交
1401 1402
};

1403
int __net_init udp6_proc_init(struct net *net)
L
Linus Torvalds 已提交
1404
{
1405
	return udp_proc_register(net, &udp6_seq_afinfo);
L
Linus Torvalds 已提交
1406 1407
}

1408 1409
void udp6_proc_exit(struct net *net) {
	udp_proc_unregister(net, &udp6_seq_afinfo);
L
Linus Torvalds 已提交
1410 1411 1412
}
#endif /* CONFIG_PROC_FS */

E
Eric Dumazet 已提交
1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423
void udp_v6_clear_sk(struct sock *sk, int size)
{
	struct inet_sock *inet = inet_sk(sk);

	/* we do not want to clear pinet6 field, because of RCU lookups */
	sk_prot_clear_portaddr_nulls(sk, offsetof(struct inet_sock, pinet6));

	size -= offsetof(struct inet_sock, pinet6) + sizeof(inet->pinet6);
	memset(&inet->pinet6 + 1, 0, size);
}

L
Linus Torvalds 已提交
1424 1425 1426
/* ------------------------------------------------------------------------ */

struct proto udpv6_prot = {
1427 1428
	.name		   = "UDPv6",
	.owner		   = THIS_MODULE,
1429
	.close		   = udp_lib_close,
1430 1431 1432 1433 1434 1435 1436 1437
	.connect	   = ip6_datagram_connect,
	.disconnect	   = udp_disconnect,
	.ioctl		   = udp_ioctl,
	.destroy	   = udpv6_destroy_sock,
	.setsockopt	   = udpv6_setsockopt,
	.getsockopt	   = udpv6_getsockopt,
	.sendmsg	   = udpv6_sendmsg,
	.recvmsg	   = udpv6_recvmsg,
1438
	.backlog_rcv	   = __udpv6_queue_rcv_skb,
1439 1440
	.hash		   = udp_lib_hash,
	.unhash		   = udp_lib_unhash,
E
Eric Dumazet 已提交
1441
	.rehash		   = udp_v6_rehash,
1442
	.get_port	   = udp_v6_get_port,
H
Hideo Aoki 已提交
1443 1444 1445 1446
	.memory_allocated  = &udp_memory_allocated,
	.sysctl_mem	   = sysctl_udp_mem,
	.sysctl_wmem	   = &sysctl_udp_wmem_min,
	.sysctl_rmem	   = &sysctl_udp_rmem_min,
1447
	.obj_size	   = sizeof(struct udp6_sock),
1448
	.slab_flags	   = SLAB_DESTROY_BY_RCU,
1449
	.h.udp_table	   = &udp_table,
1450
#ifdef CONFIG_COMPAT
1451 1452
	.compat_setsockopt = compat_udpv6_setsockopt,
	.compat_getsockopt = compat_udpv6_getsockopt,
1453
#endif
E
Eric Dumazet 已提交
1454
	.clear_sk	   = udp_v6_clear_sk,
L
Linus Torvalds 已提交
1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466
};

static struct inet_protosw udpv6_protosw = {
	.type =      SOCK_DGRAM,
	.protocol =  IPPROTO_UDP,
	.prot =      &udpv6_prot,
	.ops =       &inet6_dgram_ops,
	.no_check =  UDP_CSUM_DEFAULT,
	.flags =     INET_PROTOSW_PERMANENT,
};


1467
int __init udpv6_init(void)
L
Linus Torvalds 已提交
1468
{
1469 1470
	int ret;

1471 1472
	ret = inet6_add_protocol(&udpv6_protocol, IPPROTO_UDP);
	if (ret)
1473
		goto out;
1474

1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485
	ret = inet6_register_protosw(&udpv6_protosw);
	if (ret)
		goto out_udpv6_protocol;
out:
	return ret;

out_udpv6_protocol:
	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
	goto out;
}

1486
void udpv6_exit(void)
1487 1488 1489
{
	inet6_unregister_protosw(&udpv6_protosw);
	inet6_del_protocol(&udpv6_protocol, IPPROTO_UDP);
L
Linus Torvalds 已提交
1490
}