udp.c 54.9 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7
/*
 * INET		An implementation of the TCP/IP protocol suite for the LINUX
 *		operating system.  INET is implemented using the  BSD Socket
 *		interface as the means of communication with the user level.
 *
 *		The User Datagram Protocol (UDP).
 *
8
 * Authors:	Ross Biro
L
Linus Torvalds 已提交
9 10
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
11
 *		Alan Cox, <alan@lxorguk.ukuu.org.uk>
L
Linus Torvalds 已提交
12 13 14 15 16 17 18 19 20
 *		Hirokazu Takahashi, <taka@valinux.co.jp>
 *
 * Fixes:
 *		Alan Cox	:	verify_area() calls
 *		Alan Cox	: 	stopped close while in use off icmp
 *					messages. Not a fix but a botch that
 *					for udp at least is 'valid'.
 *		Alan Cox	:	Fixed icmp handling properly
 *		Alan Cox	: 	Correct error for oversized datagrams
21 22
 *		Alan Cox	:	Tidied select() semantics.
 *		Alan Cox	:	udp_err() fixed properly, also now
L
Linus Torvalds 已提交
23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56
 *					select and read wake correctly on errors
 *		Alan Cox	:	udp_send verify_area moved to avoid mem leak
 *		Alan Cox	:	UDP can count its memory
 *		Alan Cox	:	send to an unknown connection causes
 *					an ECONNREFUSED off the icmp, but
 *					does NOT close.
 *		Alan Cox	:	Switched to new sk_buff handlers. No more backlog!
 *		Alan Cox	:	Using generic datagram code. Even smaller and the PEEK
 *					bug no longer crashes it.
 *		Fred Van Kempen	: 	Net2e support for sk->broadcast.
 *		Alan Cox	:	Uses skb_free_datagram
 *		Alan Cox	:	Added get/set sockopt support.
 *		Alan Cox	:	Broadcasting without option set returns EACCES.
 *		Alan Cox	:	No wakeup calls. Instead we now use the callbacks.
 *		Alan Cox	:	Use ip_tos and ip_ttl
 *		Alan Cox	:	SNMP Mibs
 *		Alan Cox	:	MSG_DONTROUTE, and 0.0.0.0 support.
 *		Matt Dillon	:	UDP length checks.
 *		Alan Cox	:	Smarter af_inet used properly.
 *		Alan Cox	:	Use new kernel side addressing.
 *		Alan Cox	:	Incorrect return on truncated datagram receive.
 *	Arnt Gulbrandsen 	:	New udp_send and stuff
 *		Alan Cox	:	Cache last socket
 *		Alan Cox	:	Route cache
 *		Jon Peatfield	:	Minor efficiency fix to sendto().
 *		Mike Shaver	:	RFC1122 checks.
 *		Alan Cox	:	Nonblocking error fix.
 *	Willy Konynenberg	:	Transparent proxying support.
 *		Mike McLagan	:	Routing by source
 *		David S. Miller	:	New socket lookup architecture.
 *					Last socket cache retained as it
 *					does have a high hit rate.
 *		Olaf Kirch	:	Don't linearise iovec on sendmsg.
 *		Andi Kleen	:	Some cleanups, cache destination entry
57
 *					for connect.
L
Linus Torvalds 已提交
58 59 60 61 62 63 64 65 66 67 68 69 70
 *	Vitaly E. Lavrov	:	Transparent proxy revived after year coma.
 *		Melvin Smith	:	Check msg_name not msg_namelen in sendto(),
 *					return ENOTCONN for unconnected sockets (POSIX)
 *		Janos Farkas	:	don't deliver multi/broadcasts to a different
 *					bound-to-device socket
 *	Hirokazu Takahashi	:	HW checksumming for outgoing UDP
 *					datagrams.
 *	Hirokazu Takahashi	:	sendfile() on UDP works now.
 *		Arnaldo C. Melo :	convert /proc/net/udp to seq_file
 *	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.
 *	Derek Atkins <derek@ihtfp.com>: Add Encapulation Support
71
 *	James Chapman		:	Add L2TP encapsulation type.
L
Linus Torvalds 已提交
72 73 74 75 76 77 78
 *
 *
 *		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.
 */
79

L
Linus Torvalds 已提交
80 81 82
#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>
H
Hideo Aoki 已提交
83
#include <linux/bootmem.h>
84 85
#include <linux/highmem.h>
#include <linux/swap.h>
L
Linus Torvalds 已提交
86 87 88 89 90
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/module.h>
#include <linux/socket.h>
#include <linux/sockios.h>
91
#include <linux/igmp.h>
L
Linus Torvalds 已提交
92 93 94 95 96 97
#include <linux/in.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
98
#include <net/tcp_states.h>
L
Linus Torvalds 已提交
99 100 101
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
102
#include <net/net_namespace.h>
L
Linus Torvalds 已提交
103 104 105 106
#include <net/icmp.h>
#include <net/route.h>
#include <net/checksum.h>
#include <net/xfrm.h>
107
#include "udp_impl.h"
L
Linus Torvalds 已提交
108

109
struct udp_table udp_table __read_mostly;
110
EXPORT_SYMBOL(udp_table);
L
Linus Torvalds 已提交
111

H
Hideo Aoki 已提交
112 113
int sysctl_udp_mem[3] __read_mostly;
EXPORT_SYMBOL(sysctl_udp_mem);
E
Eric Dumazet 已提交
114 115

int sysctl_udp_rmem_min __read_mostly;
H
Hideo Aoki 已提交
116
EXPORT_SYMBOL(sysctl_udp_rmem_min);
E
Eric Dumazet 已提交
117 118

int sysctl_udp_wmem_min __read_mostly;
H
Hideo Aoki 已提交
119 120 121 122 123
EXPORT_SYMBOL(sysctl_udp_wmem_min);

atomic_t udp_memory_allocated;
EXPORT_SYMBOL(udp_memory_allocated);

124 125
#define MAX_UDP_PORTS 65536
#define PORTS_PER_CHAIN (MAX_UDP_PORTS / UDP_HTABLE_SIZE_MIN)
126

127
static int udp_lib_lport_inuse(struct net *net, __u16 num,
128
			       const struct udp_hslot *hslot,
129
			       unsigned long *bitmap,
130 131
			       struct sock *sk,
			       int (*saddr_comp)(const struct sock *sk1,
132 133
						 const struct sock *sk2),
			       unsigned int log)
L
Linus Torvalds 已提交
134
{
135
	struct sock *sk2;
136
	struct hlist_nulls_node *node;
137

138
	sk_nulls_for_each(sk2, node, &hslot->head)
139 140
		if (net_eq(sock_net(sk2), net) &&
		    sk2 != sk &&
141
		    (bitmap || udp_sk(sk2)->udp_port_hash == num) &&
142 143 144
		    (!sk2->sk_reuse || !sk->sk_reuse) &&
		    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if ||
		     sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
145 146
		    (*saddr_comp)(sk, sk2)) {
			if (bitmap)
147 148
				__set_bit(udp_sk(sk2)->udp_port_hash >> log,
					  bitmap);
149 150 151
			else
				return 1;
		}
152 153 154
	return 0;
}

E
Eric Dumazet 已提交
155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170
/*
 * Note: we still hold spinlock of primary hash chain, so no other writer
 * can insert/delete a socket with local_port == num
 */
static int udp_lib_lport_inuse2(struct net *net, __u16 num,
			       struct udp_hslot *hslot2,
			       struct sock *sk,
			       int (*saddr_comp)(const struct sock *sk1,
						 const struct sock *sk2))
{
	struct sock *sk2;
	struct hlist_nulls_node *node;
	int res = 0;

	spin_lock(&hslot2->lock);
	udp_portaddr_for_each_entry(sk2, node, &hslot2->head)
171 172 173 174 175 176
		if (net_eq(sock_net(sk2), net) &&
		    sk2 != sk &&
		    (udp_sk(sk2)->udp_port_hash == num) &&
		    (!sk2->sk_reuse || !sk->sk_reuse) &&
		    (!sk2->sk_bound_dev_if || !sk->sk_bound_dev_if ||
		     sk2->sk_bound_dev_if == sk->sk_bound_dev_if) &&
E
Eric Dumazet 已提交
177 178 179 180 181 182 183 184
		    (*saddr_comp)(sk, sk2)) {
			res = 1;
			break;
		}
	spin_unlock(&hslot2->lock);
	return res;
}

185
/**
186
 *  udp_lib_get_port  -  UDP/-Lite port lookup for IPv4 and IPv6
187 188 189
 *
 *  @sk:          socket struct in question
 *  @snum:        port number to look up
190
 *  @saddr_comp:  AF-dependent comparison of bound local IP addresses
E
Eric Dumazet 已提交
191 192
 *  @hash2_nulladdr: AF-dependant hash value in secondary hash chains,
 *                   with NULL address
193
 */
194
int udp_lib_get_port(struct sock *sk, unsigned short snum,
195
		       int (*saddr_comp)(const struct sock *sk1,
E
Eric Dumazet 已提交
196 197
					 const struct sock *sk2),
		     unsigned int hash2_nulladdr)
198
{
199
	struct udp_hslot *hslot, *hslot2;
200
	struct udp_table *udptable = sk->sk_prot->h.udp_table;
201
	int    error = 1;
202
	struct net *net = sock_net(sk);
L
Linus Torvalds 已提交
203

204
	if (!snum) {
E
Eric Dumazet 已提交
205 206
		int low, high, remaining;
		unsigned rand;
207 208
		unsigned short first, last;
		DECLARE_BITMAP(bitmap, PORTS_PER_CHAIN);
209

210
		inet_get_local_port_range(&low, &high);
211
		remaining = (high - low) + 1;
212

E
Eric Dumazet 已提交
213
		rand = net_random();
214 215 216 217
		first = (((u64)rand * remaining) >> 32) + low;
		/*
		 * force rand to be an odd multiple of UDP_HTABLE_SIZE
		 */
218
		rand = (rand | 1) * (udptable->mask + 1);
E
Eric Dumazet 已提交
219 220
		last = first + udptable->mask + 1;
		do {
221
			hslot = udp_hashslot(udptable, net, first);
222
			bitmap_zero(bitmap, PORTS_PER_CHAIN);
223
			spin_lock_bh(&hslot->lock);
224
			udp_lib_lport_inuse(net, snum, hslot, bitmap, sk,
225
					    saddr_comp, udptable->log);
226 227 228 229 230 231 232

			snum = first;
			/*
			 * Iterate on all possible values of snum for this hash.
			 * Using steps of an odd multiple of UDP_HTABLE_SIZE
			 * give us randomization and full range coverage.
			 */
E
Eric Dumazet 已提交
233
			do {
234
				if (low <= snum && snum <= high &&
235
				    !test_bit(snum >> udptable->log, bitmap))
236 237 238 239
					goto found;
				snum += rand;
			} while (snum != first);
			spin_unlock_bh(&hslot->lock);
E
Eric Dumazet 已提交
240
		} while (++first != last);
241
		goto fail;
242
	} else {
243
		hslot = udp_hashslot(udptable, net, snum);
244
		spin_lock_bh(&hslot->lock);
E
Eric Dumazet 已提交
245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268
		if (hslot->count > 10) {
			int exist;
			unsigned int slot2 = udp_sk(sk)->udp_portaddr_hash ^ snum;

			slot2          &= udptable->mask;
			hash2_nulladdr &= udptable->mask;

			hslot2 = udp_hashslot2(udptable, slot2);
			if (hslot->count < hslot2->count)
				goto scan_primary_hash;

			exist = udp_lib_lport_inuse2(net, snum, hslot2,
						     sk, saddr_comp);
			if (!exist && (hash2_nulladdr != slot2)) {
				hslot2 = udp_hashslot2(udptable, hash2_nulladdr);
				exist = udp_lib_lport_inuse2(net, snum, hslot2,
							     sk, saddr_comp);
			}
			if (exist)
				goto fail_unlock;
			else
				goto found;
		}
scan_primary_hash:
269 270
		if (udp_lib_lport_inuse(net, snum, hslot, NULL, sk,
					saddr_comp, 0))
271 272
			goto fail_unlock;
	}
273
found:
E
Eric Dumazet 已提交
274
	inet_sk(sk)->inet_num = snum;
275 276
	udp_sk(sk)->udp_port_hash = snum;
	udp_sk(sk)->udp_portaddr_hash ^= snum;
L
Linus Torvalds 已提交
277
	if (sk_unhashed(sk)) {
278
		sk_nulls_add_node_rcu(sk, &hslot->head);
E
Eric Dumazet 已提交
279
		hslot->count++;
280
		sock_prot_inuse_add(sock_net(sk), sk->sk_prot, 1);
281 282 283 284 285 286 287

		hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash);
		spin_lock(&hslot2->lock);
		hlist_nulls_add_head_rcu(&udp_sk(sk)->udp_portaddr_node,
					 &hslot2->head);
		hslot2->count++;
		spin_unlock(&hslot2->lock);
L
Linus Torvalds 已提交
288
	}
289
	error = 0;
290 291
fail_unlock:
	spin_unlock_bh(&hslot->lock);
L
Linus Torvalds 已提交
292
fail:
293 294
	return error;
}
E
Eric Dumazet 已提交
295
EXPORT_SYMBOL(udp_lib_get_port);
296

297
static int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
298 299 300
{
	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);

E
Eric Dumazet 已提交
301
	return 	(!ipv6_only_sock(sk2)  &&
E
Eric Dumazet 已提交
302 303
		 (!inet1->inet_rcv_saddr || !inet2->inet_rcv_saddr ||
		   inet1->inet_rcv_saddr == inet2->inet_rcv_saddr));
304 305
}

306 307 308 309 310 311
static unsigned int udp4_portaddr_hash(struct net *net, __be32 saddr,
				       unsigned int port)
{
	return jhash_1word(saddr, net_hash_mix(net)) ^ port;
}

312
int udp_v4_get_port(struct sock *sk, unsigned short snum)
313
{
E
Eric Dumazet 已提交
314 315 316 317 318
	unsigned int hash2_nulladdr =
		udp4_portaddr_hash(sock_net(sk), INADDR_ANY, snum);
	unsigned int hash2_partial =
		udp4_portaddr_hash(sock_net(sk), inet_sk(sk)->inet_rcv_saddr, 0);

319
	/* precompute partial secondary hash */
E
Eric Dumazet 已提交
320 321
	udp_sk(sk)->udp_portaddr_hash = hash2_partial;
	return udp_lib_get_port(sk, snum, ipv4_rcv_saddr_equal, hash2_nulladdr);
322 323
}

324 325 326 327 328 329
static inline int compute_score(struct sock *sk, struct net *net, __be32 saddr,
			 unsigned short hnum,
			 __be16 sport, __be32 daddr, __be16 dport, int dif)
{
	int score = -1;

330
	if (net_eq(sock_net(sk), net) && udp_sk(sk)->udp_port_hash == hnum &&
331 332 333 334
			!ipv6_only_sock(sk)) {
		struct inet_sock *inet = inet_sk(sk);

		score = (sk->sk_family == PF_INET ? 1 : 0);
E
Eric Dumazet 已提交
335 336
		if (inet->inet_rcv_saddr) {
			if (inet->inet_rcv_saddr != daddr)
337 338 339
				return -1;
			score += 2;
		}
E
Eric Dumazet 已提交
340 341
		if (inet->inet_daddr) {
			if (inet->inet_daddr != saddr)
342 343 344
				return -1;
			score += 2;
		}
E
Eric Dumazet 已提交
345 346
		if (inet->inet_dport) {
			if (inet->inet_dport != sport)
347 348 349 350 351 352 353 354 355 356 357 358
				return -1;
			score += 2;
		}
		if (sk->sk_bound_dev_if) {
			if (sk->sk_bound_dev_if != dif)
				return -1;
			score += 2;
		}
	}
	return score;
}

E
Eric Dumazet 已提交
359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441
/*
 * In this second variant, we check (daddr, dport) matches (inet_rcv_sadd, inet_num)
 */
#define SCORE2_MAX (1 + 2 + 2 + 2)
static inline int compute_score2(struct sock *sk, struct net *net,
				 __be32 saddr, __be16 sport,
				 __be32 daddr, unsigned int hnum, int dif)
{
	int score = -1;

	if (net_eq(sock_net(sk), net) && !ipv6_only_sock(sk)) {
		struct inet_sock *inet = inet_sk(sk);

		if (inet->inet_rcv_saddr != daddr)
			return -1;
		if (inet->inet_num != hnum)
			return -1;

		score = (sk->sk_family == PF_INET ? 1 : 0);
		if (inet->inet_daddr) {
			if (inet->inet_daddr != saddr)
				return -1;
			score += 2;
		}
		if (inet->inet_dport) {
			if (inet->inet_dport != sport)
				return -1;
			score += 2;
		}
		if (sk->sk_bound_dev_if) {
			if (sk->sk_bound_dev_if != dif)
				return -1;
			score += 2;
		}
	}
	return score;
}


/* called with read_rcu_lock() */
static struct sock *udp4_lib_lookup2(struct net *net,
		__be32 saddr, __be16 sport,
		__be32 daddr, unsigned int hnum, int dif,
		struct udp_hslot *hslot2, unsigned int slot2)
{
	struct sock *sk, *result;
	struct hlist_nulls_node *node;
	int score, badness;

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;
			if (score == SCORE2_MAX)
				goto exact_match;
		}
	}
	/*
	 * 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:
		if (unlikely(!atomic_inc_not_zero(&result->sk_refcnt)))
			result = NULL;
		else if (unlikely(compute_score2(result, net, saddr, sport,
				  daddr, hnum, dif) < badness)) {
			sock_put(result);
			goto begin;
		}
	}
	return result;
}

442 443 444 445 446
/* UDP is nearly always wildcards out the wazoo, it makes no sense to try
 * harder than this. -DaveM
 */
static struct sock *__udp4_lib_lookup(struct net *net, __be32 saddr,
		__be16 sport, __be32 daddr, __be16 dport,
447
		int dif, struct udp_table *udptable)
448
{
449
	struct sock *sk, *result;
450
	struct hlist_nulls_node *node;
451
	unsigned short hnum = ntohs(dport);
E
Eric Dumazet 已提交
452 453
	unsigned int hash2, slot2, slot = udp_hashfn(net, hnum, udptable->mask);
	struct udp_hslot *hslot2, *hslot = &udptable->hash[slot];
454
	int score, badness;
455

456
	rcu_read_lock();
E
Eric Dumazet 已提交
457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480
	if (hslot->count > 10) {
		hash2 = udp4_portaddr_hash(net, daddr, hnum);
		slot2 = hash2 & udptable->mask;
		hslot2 = &udptable->hash2[slot2];
		if (hslot->count < hslot2->count)
			goto begin;

		result = udp4_lib_lookup2(net, saddr, sport,
					  daddr, hnum, dif,
					  hslot2, slot2);
		if (!result) {
			hash2 = udp4_portaddr_hash(net, INADDR_ANY, hnum);
			slot2 = hash2 & udptable->mask;
			hslot2 = &udptable->hash2[slot2];
			if (hslot->count < hslot2->count)
				goto begin;

			result = udp4_lib_lookup2(net, INADDR_ANY, sport,
						  daddr, hnum, dif,
						  hslot2, slot2);
		}
		rcu_read_unlock();
		return result;
	}
481 482 483
begin:
	result = NULL;
	badness = -1;
484
	sk_nulls_for_each_rcu(sk, node, &hslot->head) {
485 486 487 488 489
		score = compute_score(sk, net, saddr, hnum, sport,
				      daddr, dport, dif);
		if (score > badness) {
			result = sk;
			badness = score;
490 491
		}
	}
492 493 494 495 496
	/*
	 * 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
497
	if (get_nulls_value(node) != slot)
498 499
		goto begin;

500 501 502 503 504 505 506 507 508 509
	if (result) {
		if (unlikely(!atomic_inc_not_zero(&result->sk_refcnt)))
			result = NULL;
		else if (unlikely(compute_score(result, net, saddr, hnum, sport,
				  daddr, dport, dif) < badness)) {
			sock_put(result);
			goto begin;
		}
	}
	rcu_read_unlock();
510 511 512
	return result;
}

513 514
static inline struct sock *__udp4_lib_lookup_skb(struct sk_buff *skb,
						 __be16 sport, __be16 dport,
515
						 struct udp_table *udptable)
516
{
517
	struct sock *sk;
518 519
	const struct iphdr *iph = ip_hdr(skb);

520 521 522
	if (unlikely(sk = skb_steal_sock(skb)))
		return sk;
	else
E
Eric Dumazet 已提交
523
		return __udp4_lib_lookup(dev_net(skb_dst(skb)->dev), iph->saddr, sport,
524 525
					 iph->daddr, dport, inet_iif(skb),
					 udptable);
526 527
}

528 529 530
struct sock *udp4_lib_lookup(struct net *net, __be32 saddr, __be16 sport,
			     __be32 daddr, __be16 dport, int dif)
{
531
	return __udp4_lib_lookup(net, saddr, sport, daddr, dport, dif, &udp_table);
532 533 534
}
EXPORT_SYMBOL_GPL(udp4_lib_lookup);

535
static inline struct sock *udp_v4_mcast_next(struct net *net, struct sock *sk,
536 537 538 539
					     __be16 loc_port, __be32 loc_addr,
					     __be16 rmt_port, __be32 rmt_addr,
					     int dif)
{
540
	struct hlist_nulls_node *node;
541 542 543
	struct sock *s = sk;
	unsigned short hnum = ntohs(loc_port);

544
	sk_nulls_for_each_from(s, node) {
545 546
		struct inet_sock *inet = inet_sk(s);

547 548 549 550 551 552 553
		if (!net_eq(sock_net(s), net) ||
		    udp_sk(s)->udp_port_hash != hnum ||
		    (inet->inet_daddr && inet->inet_daddr != rmt_addr) ||
		    (inet->inet_dport != rmt_port && inet->inet_dport) ||
		    (inet->inet_rcv_saddr &&
		     inet->inet_rcv_saddr != loc_addr) ||
		    ipv6_only_sock(s) ||
554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575
		    (s->sk_bound_dev_if && s->sk_bound_dev_if != dif))
			continue;
		if (!ip_mc_sf_allow(s, loc_addr, rmt_addr, dif))
			continue;
		goto found;
	}
	s = NULL;
found:
	return s;
}

/*
 * This routine is called by the ICMP module when it gets some
 * sort of error condition.  If err < 0 then the socket should
 * be closed and the error returned to the user.  If err > 0
 * it's just the icmp type << 8 | icmp code.
 * Header points to the ip header of the error packet. We move
 * on past this. Then (as it used to claim before adjustment)
 * header points to the first 8 bytes of the udp header.  We need
 * to find the appropriate port.
 */

576
void __udp4_lib_err(struct sk_buff *skb, u32 info, struct udp_table *udptable)
577 578
{
	struct inet_sock *inet;
E
Eric Dumazet 已提交
579 580
	struct iphdr *iph = (struct iphdr *)skb->data;
	struct udphdr *uh = (struct udphdr *)(skb->data+(iph->ihl<<2));
581 582 583 584 585
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
	struct sock *sk;
	int harderr;
	int err;
586
	struct net *net = dev_net(skb->dev);
587

588
	sk = __udp4_lib_lookup(net, iph->daddr, uh->dest,
589 590
			iph->saddr, uh->source, skb->dev->ifindex, udptable);
	if (sk == NULL) {
591
		ICMP_INC_STATS_BH(net, ICMP_MIB_INERRORS);
592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634
		return;	/* No socket for error */
	}

	err = 0;
	harderr = 0;
	inet = inet_sk(sk);

	switch (type) {
	default:
	case ICMP_TIME_EXCEEDED:
		err = EHOSTUNREACH;
		break;
	case ICMP_SOURCE_QUENCH:
		goto out;
	case ICMP_PARAMETERPROB:
		err = EPROTO;
		harderr = 1;
		break;
	case ICMP_DEST_UNREACH:
		if (code == ICMP_FRAG_NEEDED) { /* Path MTU discovery */
			if (inet->pmtudisc != IP_PMTUDISC_DONT) {
				err = EMSGSIZE;
				harderr = 1;
				break;
			}
			goto out;
		}
		err = EHOSTUNREACH;
		if (code <= NR_ICMP_UNREACH) {
			harderr = icmp_err_convert[code].fatal;
			err = icmp_err_convert[code].errno;
		}
		break;
	}

	/*
	 *      RFC1122: OK.  Passes ICMP errors back to application, as per
	 *	4.1.3.3.
	 */
	if (!inet->recverr) {
		if (!harderr || sk->sk_state != TCP_ESTABLISHED)
			goto out;
	} else {
E
Eric Dumazet 已提交
635
		ip_icmp_error(sk, skb, err, uh->dest, info, (u8 *)(uh+1));
636 637 638 639 640 641 642 643 644
	}
	sk->sk_err = err;
	sk->sk_error_report(sk);
out:
	sock_put(sk);
}

void udp_err(struct sk_buff *skb, u32 info)
{
645
	__udp4_lib_err(skb, info, &udp_table);
646 647 648 649 650
}

/*
 * Throw away all pending data and cancel the corking. Socket is locked.
 */
651
void udp_flush_pending_frames(struct sock *sk)
652 653 654 655 656 657 658 659 660
{
	struct udp_sock *up = udp_sk(sk);

	if (up->pending) {
		up->len = 0;
		up->pending = 0;
		ip_flush_pending_frames(sk);
	}
}
661
EXPORT_SYMBOL(udp_flush_pending_frames);
662 663 664 665 666 667 668 669

/**
 * 	udp4_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 udp4_hwcsum_outgoing(struct sock *sk, struct sk_buff *skb,
E
Eric Dumazet 已提交
670
				 __be32 src, __be32 dst, int len)
671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740
{
	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_tcpudp_magic(src, dst, 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_tcpudp_magic(src, dst, len, IPPROTO_UDP, csum);
		if (uh->check == 0)
			uh->check = CSUM_MANGLED_0;
	}
}

/*
 * Push out all pending data as one UDP datagram. Socket is locked.
 */
static int udp_push_pending_frames(struct sock *sk)
{
	struct udp_sock  *up = udp_sk(sk);
	struct inet_sock *inet = inet_sk(sk);
	struct flowi *fl = &inet->cork.fl;
	struct sk_buff *skb;
	struct udphdr *uh;
	int err = 0;
	int is_udplite = IS_UDPLITE(sk);
	__wsum csum = 0;

	/* Grab the skbuff where UDP header space exists. */
	if ((skb = skb_peek(&sk->sk_write_queue)) == NULL)
		goto out;

	/*
	 * Create a UDP header
	 */
	uh = udp_hdr(skb);
	uh->source = fl->fl_ip_sport;
	uh->dest = fl->fl_ip_dport;
	uh->len = htons(up->len);
	uh->check = 0;

	if (is_udplite)  				 /*     UDP-Lite      */
		csum  = udplite_csum_outgoing(sk, skb);

	else if (sk->sk_no_check == UDP_CSUM_NOXMIT) {   /* UDP csum disabled */

		skb->ip_summed = CHECKSUM_NONE;
		goto send;

	} else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */

E
Eric Dumazet 已提交
741
		udp4_hwcsum_outgoing(sk, skb, fl->fl4_src, fl->fl4_dst, up->len);
742 743 744 745 746 747 748
		goto send;

	} else						 /*   `normal' UDP    */
		csum = udp_csum_outgoing(sk, skb);

	/* add protocol-dependent pseudo-header */
	uh->check = csum_tcpudp_magic(fl->fl4_src, fl->fl4_dst, up->len,
E
Eric Dumazet 已提交
749
				      sk->sk_protocol, csum);
750 751 752 753 754
	if (uh->check == 0)
		uh->check = CSUM_MANGLED_0;

send:
	err = ip_push_pending_frames(sk);
E
Eric Dumazet 已提交
755 756 757 758 759 760 761 762 763
	if (err) {
		if (err == -ENOBUFS && !inet->recverr) {
			UDP_INC_STATS_USER(sock_net(sk),
					   UDP_MIB_SNDBUFERRORS, is_udplite);
			err = 0;
		}
	} else
		UDP_INC_STATS_USER(sock_net(sk),
				   UDP_MIB_OUTDATAGRAMS, is_udplite);
764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793
out:
	up->len = 0;
	up->pending = 0;
	return err;
}

int udp_sendmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
		size_t len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct udp_sock *up = udp_sk(sk);
	int ulen = len;
	struct ipcm_cookie ipc;
	struct rtable *rt = NULL;
	int free = 0;
	int connected = 0;
	__be32 daddr, faddr, saddr;
	__be16 dport;
	u8  tos;
	int err, is_udplite = IS_UDPLITE(sk);
	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);

	if (len > 0xFFFF)
		return -EMSGSIZE;

	/*
	 *	Check the flags.
	 */

E
Eric Dumazet 已提交
794
	if (msg->msg_flags & MSG_OOB) /* Mirror BSD error message compatibility */
795 796 797
		return -EOPNOTSUPP;

	ipc.opt = NULL;
798
	ipc.shtx.flags = 0;
799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820

	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_INET)) {
				release_sock(sk);
				return -EINVAL;
			}
			goto do_append_data;
		}
		release_sock(sk);
	}
	ulen += sizeof(struct udphdr);

	/*
	 *	Get and verify the address.
	 */
	if (msg->msg_name) {
E
Eric Dumazet 已提交
821
		struct sockaddr_in * usin = (struct sockaddr_in *)msg->msg_name;
822 823 824 825 826 827 828 829 830 831 832 833 834 835
		if (msg->msg_namelen < sizeof(*usin))
			return -EINVAL;
		if (usin->sin_family != AF_INET) {
			if (usin->sin_family != AF_UNSPEC)
				return -EAFNOSUPPORT;
		}

		daddr = usin->sin_addr.s_addr;
		dport = usin->sin_port;
		if (dport == 0)
			return -EINVAL;
	} else {
		if (sk->sk_state != TCP_ESTABLISHED)
			return -EDESTADDRREQ;
E
Eric Dumazet 已提交
836 837
		daddr = inet->inet_daddr;
		dport = inet->inet_dport;
838 839 840 841 842
		/* Open fast path for connected socket.
		   Route will not be used, if at least one option is set.
		 */
		connected = 1;
	}
E
Eric Dumazet 已提交
843
	ipc.addr = inet->inet_saddr;
844 845

	ipc.oif = sk->sk_bound_dev_if;
846 847 848
	err = sock_tx_timestamp(msg, sk, &ipc.shtx);
	if (err)
		return err;
849
	if (msg->msg_controllen) {
850
		err = ip_cmsg_send(sock_net(sk), msg, &ipc);
851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885
		if (err)
			return err;
		if (ipc.opt)
			free = 1;
		connected = 0;
	}
	if (!ipc.opt)
		ipc.opt = inet->opt;

	saddr = ipc.addr;
	ipc.addr = faddr = daddr;

	if (ipc.opt && ipc.opt->srr) {
		if (!daddr)
			return -EINVAL;
		faddr = ipc.opt->faddr;
		connected = 0;
	}
	tos = RT_TOS(inet->tos);
	if (sock_flag(sk, SOCK_LOCALROUTE) ||
	    (msg->msg_flags & MSG_DONTROUTE) ||
	    (ipc.opt && ipc.opt->is_strictroute)) {
		tos |= RTO_ONLINK;
		connected = 0;
	}

	if (ipv4_is_multicast(daddr)) {
		if (!ipc.oif)
			ipc.oif = inet->mc_index;
		if (!saddr)
			saddr = inet->mc_addr;
		connected = 0;
	}

	if (connected)
E
Eric Dumazet 已提交
886
		rt = (struct rtable *)sk_dst_check(sk, 0);
887 888 889

	if (rt == NULL) {
		struct flowi fl = { .oif = ipc.oif,
890
				    .mark = sk->sk_mark,
891 892 893 894 895
				    .nl_u = { .ip4_u =
					      { .daddr = faddr,
						.saddr = saddr,
						.tos = tos } },
				    .proto = sk->sk_protocol,
896
				    .flags = inet_sk_flowi_flags(sk),
897
				    .uli_u = { .ports =
E
Eric Dumazet 已提交
898
					       { .sport = inet->inet_sport,
899
						 .dport = dport } } };
900 901
		struct net *net = sock_net(sk);

902
		security_sk_classify_flow(sk, &fl);
903
		err = ip_route_output_flow(net, &rt, &fl, sk, 1);
904 905
		if (err) {
			if (err == -ENETUNREACH)
906
				IP_INC_STATS_BH(net, IPSTATS_MIB_OUTNOROUTES);
907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941
			goto out;
		}

		err = -EACCES;
		if ((rt->rt_flags & RTCF_BROADCAST) &&
		    !sock_flag(sk, SOCK_BROADCAST))
			goto out;
		if (connected)
			sk_dst_set(sk, dst_clone(&rt->u.dst));
	}

	if (msg->msg_flags&MSG_CONFIRM)
		goto do_confirm;
back_from_confirm:

	saddr = rt->rt_src;
	if (!ipc.addr)
		daddr = ipc.addr = rt->rt_dst;

	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);

		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
		err = -EINVAL;
		goto out;
	}
	/*
	 *	Now cork the socket to pend data.
	 */
	inet->cork.fl.fl4_dst = daddr;
	inet->cork.fl.fl_ip_dport = dport;
	inet->cork.fl.fl4_src = saddr;
E
Eric Dumazet 已提交
942
	inet->cork.fl.fl_ip_sport = inet->inet_sport;
943 944 945 946 947 948
	up->pending = AF_INET;

do_append_data:
	up->len += ulen;
	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
	err = ip_append_data(sk, getfrag, msg->msg_iov, ulen,
949
			sizeof(struct udphdr), &ipc, &rt,
950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972
			corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
	if (err)
		udp_flush_pending_frames(sk);
	else if (!corkreq)
		err = udp_push_pending_frames(sk);
	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
		up->pending = 0;
	release_sock(sk);

out:
	ip_rt_put(rt);
	if (free)
		kfree(ipc.opt);
	if (!err)
		return len;
	/*
	 * 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)) {
973 974
		UDP_INC_STATS_USER(sock_net(sk),
				UDP_MIB_SNDBUFERRORS, is_udplite);
975 976 977 978 979 980 981 982 983 984
	}
	return err;

do_confirm:
	dst_confirm(&rt->u.dst);
	if (!(msg->msg_flags&MSG_PROBE) || len)
		goto back_from_confirm;
	err = 0;
	goto out;
}
E
Eric Dumazet 已提交
985
EXPORT_SYMBOL(udp_sendmsg);
986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034

int udp_sendpage(struct sock *sk, struct page *page, int offset,
		 size_t size, int flags)
{
	struct udp_sock *up = udp_sk(sk);
	int ret;

	if (!up->pending) {
		struct msghdr msg = {	.msg_flags = flags|MSG_MORE };

		/* Call udp_sendmsg to specify destination address which
		 * sendpage interface can't pass.
		 * This will succeed only when the socket is connected.
		 */
		ret = udp_sendmsg(NULL, sk, &msg, 0);
		if (ret < 0)
			return ret;
	}

	lock_sock(sk);

	if (unlikely(!up->pending)) {
		release_sock(sk);

		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 3\n");
		return -EINVAL;
	}

	ret = ip_append_page(sk, page, offset, size, flags);
	if (ret == -EOPNOTSUPP) {
		release_sock(sk);
		return sock_no_sendpage(sk->sk_socket, page, offset,
					size, flags);
	}
	if (ret < 0) {
		udp_flush_pending_frames(sk);
		goto out;
	}

	up->len += size;
	if (!(up->corkflag || (flags&MSG_MORE)))
		ret = udp_push_pending_frames(sk);
	if (!ret)
		ret = size;
out:
	release_sock(sk);
	return ret;
}

E
Eric Dumazet 已提交
1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055

/**
 *	first_packet_length	- return length of first packet in receive queue
 *	@sk: socket
 *
 *	Drops all bad checksum frames, until a valid one is found.
 *	Returns the length of found skb, or 0 if none is found.
 */
static unsigned int first_packet_length(struct sock *sk)
{
	struct sk_buff_head list_kill, *rcvq = &sk->sk_receive_queue;
	struct sk_buff *skb;
	unsigned int res;

	__skb_queue_head_init(&list_kill);

	spin_lock_bh(&rcvq->lock);
	while ((skb = skb_peek(rcvq)) != NULL &&
		udp_lib_checksum_complete(skb)) {
		UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS,
				 IS_UDPLITE(sk));
E
Eric Dumazet 已提交
1056
		atomic_inc(&sk->sk_drops);
E
Eric Dumazet 已提交
1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071
		__skb_unlink(skb, rcvq);
		__skb_queue_tail(&list_kill, skb);
	}
	res = skb ? skb->len : 0;
	spin_unlock_bh(&rcvq->lock);

	if (!skb_queue_empty(&list_kill)) {
		lock_sock(sk);
		__skb_queue_purge(&list_kill);
		sk_mem_reclaim_partial(sk);
		release_sock(sk);
	}
	return res;
}

L
Linus Torvalds 已提交
1072 1073 1074
/*
 *	IOCTL requests applicable to the UDP protocol
 */
1075

L
Linus Torvalds 已提交
1076 1077
int udp_ioctl(struct sock *sk, int cmd, unsigned long arg)
{
1078 1079
	switch (cmd) {
	case SIOCOUTQ:
L
Linus Torvalds 已提交
1080
	{
1081 1082
		int amount = sk_wmem_alloc_get(sk);

1083 1084
		return put_user(amount, (int __user *)arg);
	}
L
Linus Torvalds 已提交
1085

1086 1087
	case SIOCINQ:
	{
E
Eric Dumazet 已提交
1088
		unsigned int amount = first_packet_length(sk);
1089

E
Eric Dumazet 已提交
1090
		if (amount)
1091 1092 1093 1094 1095
			/*
			 * We will only return the amount
			 * of this packet since that is all
			 * that will be read.
			 */
E
Eric Dumazet 已提交
1096 1097
			amount -= sizeof(struct udphdr);

1098 1099
		return put_user(amount, (int __user *)arg);
	}
L
Linus Torvalds 已提交
1100

1101 1102
	default:
		return -ENOIOCTLCMD;
L
Linus Torvalds 已提交
1103
	}
1104 1105

	return 0;
L
Linus Torvalds 已提交
1106
}
E
Eric Dumazet 已提交
1107
EXPORT_SYMBOL(udp_ioctl);
L
Linus Torvalds 已提交
1108

1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119
/*
 * 	This should be easy, if there is something there we
 * 	return it, otherwise we block.
 */

int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
		size_t len, int noblock, int flags, int *addr_len)
{
	struct inet_sock *inet = inet_sk(sk);
	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
	struct sk_buff *skb;
G
Gerrit Renker 已提交
1120
	unsigned int ulen;
1121 1122 1123 1124 1125 1126 1127 1128
	int peeked;
	int err;
	int is_udplite = IS_UDPLITE(sk);

	/*
	 *	Check any passed addresses
	 */
	if (addr_len)
E
Eric Dumazet 已提交
1129
		*addr_len = sizeof(*sin);
1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140

	if (flags & MSG_ERRQUEUE)
		return ip_recv_error(sk, msg, len);

try_again:
	skb = __skb_recv_datagram(sk, flags | (noblock ? MSG_DONTWAIT : 0),
				  &peeked, &err);
	if (!skb)
		goto out;

	ulen = skb->len - sizeof(struct udphdr);
G
Gerrit Renker 已提交
1141 1142 1143
	if (len > ulen)
		len = ulen;
	else if (len < ulen)
1144 1145 1146 1147 1148 1149 1150 1151
		msg->msg_flags |= MSG_TRUNC;

	/*
	 * 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.
	 */

G
Gerrit Renker 已提交
1152
	if (len < ulen || UDP_SKB_CB(skb)->partial_cov) {
1153 1154 1155 1156 1157 1158
		if (udp_lib_checksum_complete(skb))
			goto csum_copy_err;
	}

	if (skb_csum_unnecessary(skb))
		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
G
Gerrit Renker 已提交
1159
					      msg->msg_iov, len);
1160
	else {
E
Eric Dumazet 已提交
1161 1162 1163
		err = skb_copy_and_csum_datagram_iovec(skb,
						       sizeof(struct udphdr),
						       msg->msg_iov);
1164 1165 1166 1167 1168 1169 1170 1171 1172

		if (err == -EINVAL)
			goto csum_copy_err;
	}

	if (err)
		goto out_free;

	if (!peeked)
1173 1174
		UDP_INC_STATS_USER(sock_net(sk),
				UDP_MIB_INDATAGRAMS, is_udplite);
1175

1176
	sock_recv_ts_and_drops(msg, sk, skb);
1177 1178

	/* Copy the address. */
E
Eric Dumazet 已提交
1179
	if (sin) {
1180 1181 1182 1183 1184 1185 1186 1187
		sin->sin_family = AF_INET;
		sin->sin_port = udp_hdr(skb)->source;
		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
	}
	if (inet->cmsg_flags)
		ip_cmsg_recv(msg, skb);

G
Gerrit Renker 已提交
1188
	err = len;
1189 1190 1191 1192
	if (flags & MSG_TRUNC)
		err = ulen;

out_free:
1193
	skb_free_datagram_locked(sk, skb);
1194 1195 1196 1197 1198 1199
out:
	return err;

csum_copy_err:
	lock_sock(sk);
	if (!skb_kill_datagram(sk, skb, flags))
1200
		UDP_INC_STATS_USER(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
1201 1202 1203 1204 1205 1206 1207 1208
	release_sock(sk);

	if (noblock)
		return -EAGAIN;
	goto try_again;
}


L
Linus Torvalds 已提交
1209 1210 1211 1212 1213 1214
int udp_disconnect(struct sock *sk, int flags)
{
	struct inet_sock *inet = inet_sk(sk);
	/*
	 *	1003.1g - break association.
	 */
1215

L
Linus Torvalds 已提交
1216
	sk->sk_state = TCP_CLOSE;
E
Eric Dumazet 已提交
1217 1218
	inet->inet_daddr = 0;
	inet->inet_dport = 0;
L
Linus Torvalds 已提交
1219 1220 1221 1222 1223 1224
	sk->sk_bound_dev_if = 0;
	if (!(sk->sk_userlocks & SOCK_BINDADDR_LOCK))
		inet_reset_saddr(sk);

	if (!(sk->sk_userlocks & SOCK_BINDPORT_LOCK)) {
		sk->sk_prot->unhash(sk);
E
Eric Dumazet 已提交
1225
		inet->inet_sport = 0;
L
Linus Torvalds 已提交
1226 1227 1228 1229
	}
	sk_dst_reset(sk);
	return 0;
}
E
Eric Dumazet 已提交
1230
EXPORT_SYMBOL(udp_disconnect);
L
Linus Torvalds 已提交
1231

1232 1233
void udp_lib_unhash(struct sock *sk)
{
1234 1235
	if (sk_hashed(sk)) {
		struct udp_table *udptable = sk->sk_prot->h.udp_table;
1236 1237 1238 1239 1240
		struct udp_hslot *hslot, *hslot2;

		hslot  = udp_hashslot(udptable, sock_net(sk),
				      udp_sk(sk)->udp_port_hash);
		hslot2 = udp_hashslot2(udptable, udp_sk(sk)->udp_portaddr_hash);
1241

1242 1243
		spin_lock_bh(&hslot->lock);
		if (sk_nulls_del_node_init_rcu(sk)) {
E
Eric Dumazet 已提交
1244
			hslot->count--;
E
Eric Dumazet 已提交
1245
			inet_sk(sk)->inet_num = 0;
1246
			sock_prot_inuse_add(sock_net(sk), sk->sk_prot, -1);
1247 1248 1249 1250 1251

			spin_lock(&hslot2->lock);
			hlist_nulls_del_init_rcu(&udp_sk(sk)->udp_portaddr_node);
			hslot2->count--;
			spin_unlock(&hslot2->lock);
1252 1253
		}
		spin_unlock_bh(&hslot->lock);
1254 1255 1256 1257
	}
}
EXPORT_SYMBOL(udp_lib_unhash);

H
Herbert Xu 已提交
1258 1259
static int __udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
{
E
Eric Dumazet 已提交
1260 1261 1262 1263
	int rc = sock_queue_rcv_skb(sk, skb);

	if (rc < 0) {
		int is_udplite = IS_UDPLITE(sk);
H
Herbert Xu 已提交
1264 1265

		/* Note that an ENOMEM error is charged twice */
E
Eric Dumazet 已提交
1266
		if (rc == -ENOMEM)
H
Herbert Xu 已提交
1267 1268
			UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_RCVBUFERRORS,
					 is_udplite);
E
Eric Dumazet 已提交
1269 1270 1271
		UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
		kfree_skb(skb);
		return -1;
H
Herbert Xu 已提交
1272 1273 1274 1275 1276 1277
	}

	return 0;

}

1278 1279 1280 1281 1282 1283 1284 1285
/* returns:
 *  -1: error
 *   0: success
 *  >0: "udp encap" protocol resubmission
 *
 * Note that in the success and error cases, the skb is assumed to
 * have either been requeued or freed.
 */
E
Eric Dumazet 已提交
1286
int udp_queue_rcv_skb(struct sock *sk, struct sk_buff *skb)
1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317
{
	struct udp_sock *up = udp_sk(sk);
	int rc;
	int is_udplite = IS_UDPLITE(sk);

	/*
	 *	Charge it to the socket, dropping if the queue is full.
	 */
	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
		goto drop;
	nf_reset(skb);

	if (up->encap_type) {
		/*
		 * 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 */
		if (skb->len > sizeof(struct udphdr) &&
		    up->encap_rcv != NULL) {
			int ret;

			ret = (*up->encap_rcv)(sk, skb);
			if (ret <= 0) {
1318 1319
				UDP_INC_STATS_BH(sock_net(sk),
						 UDP_MIB_INDATAGRAMS,
1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368
						 is_udplite);
				return -ret;
			}
		}

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

	/*
	 * 	UDP-Lite specific tests, ignored on UDP sockets
	 */
	if ((is_udplite & UDPLITE_RECV_CC)  &&  UDP_SKB_CB(skb)->partial_cov) {

		/*
		 * MIB statistics other than incrementing the error count are
		 * disabled for the following two types of errors: these depend
		 * on the application settings, not on the functioning of the
		 * protocol stack as such.
		 *
		 * RFC 3828 here recommends (sec 3.3): "There should also be a
		 * way ... to ... at least let the receiving application block
		 * delivery of packets with coverage values less than a value
		 * provided by the application."
		 */
		if (up->pcrlen == 0) {          /* full coverage was set  */
			LIMIT_NETDEBUG(KERN_WARNING "UDPLITE: partial coverage "
				"%d while full coverage %d requested\n",
				UDP_SKB_CB(skb)->cscov, skb->len);
			goto drop;
		}
		/* The next case involves violating the min. coverage requested
		 * by the receiver. This is subtle: if receiver wants x and x is
		 * greater than the buffersize/MTU then receiver will complain
		 * that it wants x while sender emits packets of smaller size y.
		 * Therefore the above ...()->partial_cov statement is essential.
		 */
		if (UDP_SKB_CB(skb)->cscov  <  up->pcrlen) {
			LIMIT_NETDEBUG(KERN_WARNING
				"UDPLITE: coverage %d too small, need min %d\n",
				UDP_SKB_CB(skb)->cscov, up->pcrlen);
			goto drop;
		}
	}

	if (sk->sk_filter) {
		if (udp_lib_checksum_complete(skb))
			goto drop;
	}

H
Herbert Xu 已提交
1369
	rc = 0;
1370

H
Herbert Xu 已提交
1371 1372 1373 1374 1375 1376 1377 1378
	bh_lock_sock(sk);
	if (!sock_owned_by_user(sk))
		rc = __udp_queue_rcv_skb(sk, skb);
	else
		sk_add_backlog(sk, skb);
	bh_unlock_sock(sk);

	return rc;
1379 1380

drop:
1381
	UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS, is_udplite);
E
Eric Dumazet 已提交
1382
	atomic_inc(&sk->sk_drops);
1383 1384 1385 1386
	kfree_skb(skb);
	return -1;
}

1387 1388 1389 1390 1391 1392

static void flush_stack(struct sock **stack, unsigned int count,
			struct sk_buff *skb, unsigned int final)
{
	unsigned int i;
	struct sk_buff *skb1 = NULL;
1393
	struct sock *sk;
1394 1395

	for (i = 0; i < count; i++) {
1396
		sk = stack[i];
1397 1398 1399
		if (likely(skb1 == NULL))
			skb1 = (i == final) ? skb : skb_clone(skb, GFP_ATOMIC);

1400 1401 1402 1403 1404 1405 1406 1407 1408
		if (!skb1) {
			atomic_inc(&sk->sk_drops);
			UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_RCVBUFERRORS,
					 IS_UDPLITE(sk));
			UDP_INC_STATS_BH(sock_net(sk), UDP_MIB_INERRORS,
					 IS_UDPLITE(sk));
		}

		if (skb1 && udp_queue_rcv_skb(sk, skb1) <= 0)
1409 1410 1411 1412 1413 1414
			skb1 = NULL;
	}
	if (unlikely(skb1))
		kfree_skb(skb1);
}

1415 1416 1417
/*
 *	Multicasts and broadcasts go to each listener.
 *
1418
 *	Note: called only from the BH handler context.
1419
 */
1420
static int __udp4_lib_mcast_deliver(struct net *net, struct sk_buff *skb,
1421 1422
				    struct udphdr  *uh,
				    __be32 saddr, __be32 daddr,
1423
				    struct udp_table *udptable)
1424
{
1425
	struct sock *sk, *stack[256 / sizeof(struct sock *)];
1426
	struct udp_hslot *hslot = udp_hashslot(udptable, net, ntohs(uh->dest));
1427
	int dif;
1428
	unsigned int i, count = 0;
1429

1430
	spin_lock(&hslot->lock);
1431
	sk = sk_nulls_head(&hslot->head);
1432
	dif = skb->dev->ifindex;
1433
	sk = udp_v4_mcast_next(net, sk, uh->dest, daddr, uh->source, saddr, dif);
1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450
	while (sk) {
		stack[count++] = sk;
		sk = udp_v4_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;
		}
	}
	/*
	 * before releasing chain lock, we must take a reference on sockets
	 */
	for (i = 0; i < count; i++)
		sock_hold(stack[i]);

1451
	spin_unlock(&hslot->lock);
1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463

	/*
	 * do the slow work with no lock held
	 */
	if (count) {
		flush_stack(stack, count, skb, count - 1);

		for (i = 0; i < count; i++)
			sock_put(stack[i]);
	} else {
		kfree_skb(skb);
	}
1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490
	return 0;
}

/* Initialize UDP checksum. If exited with zero value (success),
 * CHECKSUM_UNNECESSARY means, that no more checks are required.
 * Otherwise, csum completion requires chacksumming packet body,
 * including udp header and folding it to skb->csum.
 */
static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,
				 int proto)
{
	const struct iphdr *iph;
	int err;

	UDP_SKB_CB(skb)->partial_cov = 0;
	UDP_SKB_CB(skb)->cscov = skb->len;

	if (proto == IPPROTO_UDPLITE) {
		err = udplite_checksum_init(skb, uh);
		if (err)
			return err;
	}

	iph = ip_hdr(skb);
	if (uh->check == 0) {
		skb->ip_summed = CHECKSUM_UNNECESSARY;
	} else if (skb->ip_summed == CHECKSUM_COMPLETE) {
E
Eric Dumazet 已提交
1491
		if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508
				      proto, skb->csum))
			skb->ip_summed = CHECKSUM_UNNECESSARY;
	}
	if (!skb_csum_unnecessary(skb))
		skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
					       skb->len, proto, 0);
	/* Probably, we should checksum udp header (it should be in cache
	 * in any case) and data in tiny packets (< rx copybreak).
	 */

	return 0;
}

/*
 *	All we need to do is get the socket, and then do a checksum.
 */

1509
int __udp4_lib_rcv(struct sk_buff *skb, struct udp_table *udptable,
1510 1511 1512
		   int proto)
{
	struct sock *sk;
1513
	struct udphdr *uh;
1514
	unsigned short ulen;
E
Eric Dumazet 已提交
1515
	struct rtable *rt = skb_rtable(skb);
1516
	__be32 saddr, daddr;
1517
	struct net *net = dev_net(skb->dev);
1518 1519 1520 1521 1522 1523 1524

	/*
	 *  Validate the packet.
	 */
	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
		goto drop;		/* No space for header. */

1525
	uh   = udp_hdr(skb);
1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539
	ulen = ntohs(uh->len);
	if (ulen > skb->len)
		goto short_packet;

	if (proto == IPPROTO_UDP) {
		/* UDP validates ulen. */
		if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
			goto short_packet;
		uh = udp_hdr(skb);
	}

	if (udp4_csum_init(skb, uh, proto))
		goto csum_error;

1540 1541 1542
	saddr = ip_hdr(skb)->saddr;
	daddr = ip_hdr(skb)->daddr;

1543
	if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
1544 1545
		return __udp4_lib_mcast_deliver(net, skb, uh,
				saddr, daddr, udptable);
1546

1547
	sk = __udp4_lib_lookup_skb(skb, uh->source, uh->dest, udptable);
1548 1549

	if (sk != NULL) {
H
Herbert Xu 已提交
1550
		int ret = udp_queue_rcv_skb(sk, skb);
1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568
		sock_put(sk);

		/* a return value > 0 means to resubmit the input, but
		 * it wants the return to be -protocol, or 0
		 */
		if (ret > 0)
			return -ret;
		return 0;
	}

	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
		goto drop;
	nf_reset(skb);

	/* No socket. Drop packet silently, if checksum is wrong */
	if (udp_lib_checksum_complete(skb))
		goto csum_error;

1569
	UDP_INC_STATS_BH(net, UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
1570 1571 1572 1573 1574 1575 1576 1577 1578 1579
	icmp_send(skb, ICMP_DEST_UNREACH, ICMP_PORT_UNREACH, 0);

	/*
	 * Hmm.  We got an UDP packet to a port to which we
	 * don't wanna listen.  Ignore it.
	 */
	kfree_skb(skb);
	return 0;

short_packet:
1580
	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %pI4:%u %d/%d to %pI4:%u\n",
1581
		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
1582
		       &saddr,
1583 1584 1585
		       ntohs(uh->source),
		       ulen,
		       skb->len,
1586
		       &daddr,
1587 1588 1589 1590 1591 1592 1593 1594
		       ntohs(uh->dest));
	goto drop;

csum_error:
	/*
	 * RFC1122: OK.  Discards the bad packet silently (as far as
	 * the network is concerned, anyway) as per 4.1.3.4 (MUST).
	 */
1595
	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %pI4:%u to %pI4:%u ulen %d\n",
1596
		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
1597
		       &saddr,
1598
		       ntohs(uh->source),
1599
		       &daddr,
1600 1601 1602
		       ntohs(uh->dest),
		       ulen);
drop:
1603
	UDP_INC_STATS_BH(net, UDP_MIB_INERRORS, proto == IPPROTO_UDPLITE);
1604 1605 1606 1607 1608 1609
	kfree_skb(skb);
	return 0;
}

int udp_rcv(struct sk_buff *skb)
{
1610
	return __udp4_lib_rcv(skb, &udp_table, IPPROTO_UDP);
1611 1612
}

1613
void udp_destroy_sock(struct sock *sk)
1614 1615 1616 1617 1618 1619
{
	lock_sock(sk);
	udp_flush_pending_frames(sk);
	release_sock(sk);
}

L
Linus Torvalds 已提交
1620 1621 1622
/*
 *	Socket option code for UDP
 */
1623
int udp_lib_setsockopt(struct sock *sk, int level, int optname,
1624
		       char __user *optval, unsigned int optlen,
1625
		       int (*push_pending_frames)(struct sock *))
L
Linus Torvalds 已提交
1626 1627 1628 1629
{
	struct udp_sock *up = udp_sk(sk);
	int val;
	int err = 0;
W
Wang Chen 已提交
1630
	int is_udplite = IS_UDPLITE(sk);
L
Linus Torvalds 已提交
1631

E
Eric Dumazet 已提交
1632
	if (optlen < sizeof(int))
L
Linus Torvalds 已提交
1633 1634 1635 1636 1637
		return -EINVAL;

	if (get_user(val, (int __user *)optval))
		return -EFAULT;

1638
	switch (optname) {
L
Linus Torvalds 已提交
1639 1640 1641 1642 1643 1644
	case UDP_CORK:
		if (val != 0) {
			up->corkflag = 1;
		} else {
			up->corkflag = 0;
			lock_sock(sk);
1645
			(*push_pending_frames)(sk);
L
Linus Torvalds 已提交
1646 1647 1648
			release_sock(sk);
		}
		break;
1649

L
Linus Torvalds 已提交
1650 1651 1652 1653 1654
	case UDP_ENCAP:
		switch (val) {
		case 0:
		case UDP_ENCAP_ESPINUDP:
		case UDP_ENCAP_ESPINUDP_NON_IKE:
1655 1656
			up->encap_rcv = xfrm4_udp_encap_rcv;
			/* FALLTHROUGH */
1657
		case UDP_ENCAP_L2TPINUDP:
L
Linus Torvalds 已提交
1658 1659 1660 1661 1662 1663 1664 1665
			up->encap_type = val;
			break;
		default:
			err = -ENOPROTOOPT;
			break;
		}
		break;

1666 1667 1668 1669 1670 1671
	/*
	 * 	UDP-Lite's partial checksum coverage (RFC 3828).
	 */
	/* The sender sets actual checksum coverage length via this option.
	 * The case coverage > packet length is handled by send module. */
	case UDPLITE_SEND_CSCOV:
W
Wang Chen 已提交
1672
		if (!is_udplite)         /* Disable the option on UDP sockets */
1673 1674 1675
			return -ENOPROTOOPT;
		if (val != 0 && val < 8) /* Illegal coverage: use default (8) */
			val = 8;
1676 1677
		else if (val > USHORT_MAX)
			val = USHORT_MAX;
1678 1679 1680 1681
		up->pcslen = val;
		up->pcflag |= UDPLITE_SEND_CC;
		break;

1682 1683
	/* The receiver specifies a minimum checksum coverage value. To make
	 * sense, this should be set to at least 8 (as done below). If zero is
1684 1685
	 * used, this again means full checksum coverage.                     */
	case UDPLITE_RECV_CSCOV:
W
Wang Chen 已提交
1686
		if (!is_udplite)         /* Disable the option on UDP sockets */
1687 1688 1689
			return -ENOPROTOOPT;
		if (val != 0 && val < 8) /* Avoid silly minimal values.       */
			val = 8;
1690 1691
		else if (val > USHORT_MAX)
			val = USHORT_MAX;
1692 1693 1694 1695
		up->pcrlen = val;
		up->pcflag |= UDPLITE_RECV_CC;
		break;

L
Linus Torvalds 已提交
1696 1697 1698
	default:
		err = -ENOPROTOOPT;
		break;
1699
	}
L
Linus Torvalds 已提交
1700 1701 1702

	return err;
}
E
Eric Dumazet 已提交
1703
EXPORT_SYMBOL(udp_lib_setsockopt);
L
Linus Torvalds 已提交
1704

1705
int udp_setsockopt(struct sock *sk, int level, int optname,
1706
		   char __user *optval, unsigned int optlen)
1707 1708 1709 1710 1711 1712 1713 1714 1715
{
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
					  udp_push_pending_frames);
	return ip_setsockopt(sk, level, optname, optval, optlen);
}

#ifdef CONFIG_COMPAT
int compat_udp_setsockopt(struct sock *sk, int level, int optname,
1716
			  char __user *optval, unsigned int optlen)
1717 1718 1719 1720 1721 1722 1723 1724
{
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
		return udp_lib_setsockopt(sk, level, optname, optval, optlen,
					  udp_push_pending_frames);
	return compat_ip_setsockopt(sk, level, optname, optval, optlen);
}
#endif

1725 1726
int udp_lib_getsockopt(struct sock *sk, int level, int optname,
		       char __user *optval, int __user *optlen)
L
Linus Torvalds 已提交
1727 1728 1729 1730
{
	struct udp_sock *up = udp_sk(sk);
	int val, len;

E
Eric Dumazet 已提交
1731
	if (get_user(len, optlen))
L
Linus Torvalds 已提交
1732 1733 1734
		return -EFAULT;

	len = min_t(unsigned int, len, sizeof(int));
1735

1736
	if (len < 0)
L
Linus Torvalds 已提交
1737 1738
		return -EINVAL;

1739
	switch (optname) {
L
Linus Torvalds 已提交
1740 1741 1742 1743 1744 1745 1746 1747
	case UDP_CORK:
		val = up->corkflag;
		break;

	case UDP_ENCAP:
		val = up->encap_type;
		break;

1748 1749 1750 1751 1752 1753 1754 1755 1756 1757
	/* The following two cannot be changed on UDP sockets, the return is
	 * always 0 (which corresponds to the full checksum coverage of UDP). */
	case UDPLITE_SEND_CSCOV:
		val = up->pcslen;
		break;

	case UDPLITE_RECV_CSCOV:
		val = up->pcrlen;
		break;

L
Linus Torvalds 已提交
1758 1759
	default:
		return -ENOPROTOOPT;
1760
	}
L
Linus Torvalds 已提交
1761

1762
	if (put_user(len, optlen))
1763
		return -EFAULT;
E
Eric Dumazet 已提交
1764
	if (copy_to_user(optval, &val, len))
L
Linus Torvalds 已提交
1765
		return -EFAULT;
1766
	return 0;
L
Linus Torvalds 已提交
1767
}
E
Eric Dumazet 已提交
1768
EXPORT_SYMBOL(udp_lib_getsockopt);
L
Linus Torvalds 已提交
1769

1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786
int udp_getsockopt(struct sock *sk, int level, int optname,
		   char __user *optval, int __user *optlen)
{
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
	return ip_getsockopt(sk, level, optname, optval, optlen);
}

#ifdef CONFIG_COMPAT
int compat_udp_getsockopt(struct sock *sk, int level, int optname,
				 char __user *optval, int __user *optlen)
{
	if (level == SOL_UDP  ||  level == SOL_UDPLITE)
		return udp_lib_getsockopt(sk, level, optname, optval, optlen);
	return compat_ip_getsockopt(sk, level, optname, optval, optlen);
}
#endif
L
Linus Torvalds 已提交
1787 1788 1789 1790 1791 1792
/**
 * 	udp_poll - wait for a UDP event.
 *	@file - file struct
 *	@sock - socket
 *	@wait - poll table
 *
1793
 *	This is same as datagram poll, except for the special case of
L
Linus Torvalds 已提交
1794 1795 1796 1797 1798 1799 1800 1801 1802 1803
 *	blocking sockets. If application is using a blocking fd
 *	and a packet with checksum error is in the queue;
 *	then it could get return from select indicating data available
 *	but then block when reading it. Add special case code
 *	to work around these arguably broken applications.
 */
unsigned int udp_poll(struct file *file, struct socket *sock, poll_table *wait)
{
	unsigned int mask = datagram_poll(file, sock, wait);
	struct sock *sk = sock->sk;
1804

L
Linus Torvalds 已提交
1805
	/* Check for false positives due to checksum errors */
E
Eric Dumazet 已提交
1806 1807 1808
	if ((mask & POLLRDNORM) && !(file->f_flags & O_NONBLOCK) &&
	    !(sk->sk_shutdown & RCV_SHUTDOWN) && !first_packet_length(sk))
		mask &= ~(POLLIN | POLLRDNORM);
L
Linus Torvalds 已提交
1809 1810

	return mask;
1811

L
Linus Torvalds 已提交
1812
}
E
Eric Dumazet 已提交
1813
EXPORT_SYMBOL(udp_poll);
L
Linus Torvalds 已提交
1814

1815 1816 1817 1818 1819 1820 1821 1822 1823 1824 1825 1826 1827
struct proto udp_prot = {
	.name		   = "UDP",
	.owner		   = THIS_MODULE,
	.close		   = udp_lib_close,
	.connect	   = ip4_datagram_connect,
	.disconnect	   = udp_disconnect,
	.ioctl		   = udp_ioctl,
	.destroy	   = udp_destroy_sock,
	.setsockopt	   = udp_setsockopt,
	.getsockopt	   = udp_getsockopt,
	.sendmsg	   = udp_sendmsg,
	.recvmsg	   = udp_recvmsg,
	.sendpage	   = udp_sendpage,
H
Herbert Xu 已提交
1828
	.backlog_rcv	   = __udp_queue_rcv_skb,
1829 1830 1831 1832 1833 1834 1835 1836
	.hash		   = udp_lib_hash,
	.unhash		   = udp_lib_unhash,
	.get_port	   = udp_v4_get_port,
	.memory_allocated  = &udp_memory_allocated,
	.sysctl_mem	   = sysctl_udp_mem,
	.sysctl_wmem	   = &sysctl_udp_wmem_min,
	.sysctl_rmem	   = &sysctl_udp_rmem_min,
	.obj_size	   = sizeof(struct udp_sock),
1837
	.slab_flags	   = SLAB_DESTROY_BY_RCU,
1838
	.h.udp_table	   = &udp_table,
1839 1840 1841 1842 1843
#ifdef CONFIG_COMPAT
	.compat_setsockopt = compat_udp_setsockopt,
	.compat_getsockopt = compat_udp_getsockopt,
#endif
};
E
Eric Dumazet 已提交
1844
EXPORT_SYMBOL(udp_prot);
L
Linus Torvalds 已提交
1845 1846 1847 1848

/* ------------------------------------------------------------------------ */
#ifdef CONFIG_PROC_FS

1849
static struct sock *udp_get_first(struct seq_file *seq, int start)
L
Linus Torvalds 已提交
1850 1851 1852
{
	struct sock *sk;
	struct udp_iter_state *state = seq->private;
1853
	struct net *net = seq_file_net(seq);
L
Linus Torvalds 已提交
1854

1855 1856
	for (state->bucket = start; state->bucket <= state->udp_table->mask;
	     ++state->bucket) {
1857
		struct hlist_nulls_node *node;
1858
		struct udp_hslot *hslot = &state->udp_table->hash[state->bucket];
1859 1860 1861 1862

		if (hlist_nulls_empty(&hslot->head))
			continue;

1863
		spin_lock_bh(&hslot->lock);
1864
		sk_nulls_for_each(sk, node, &hslot->head) {
1865
			if (!net_eq(sock_net(sk), net))
1866
				continue;
L
Linus Torvalds 已提交
1867 1868 1869
			if (sk->sk_family == state->family)
				goto found;
		}
1870
		spin_unlock_bh(&hslot->lock);
L
Linus Torvalds 已提交
1871 1872 1873 1874 1875 1876 1877 1878 1879
	}
	sk = NULL;
found:
	return sk;
}

static struct sock *udp_get_next(struct seq_file *seq, struct sock *sk)
{
	struct udp_iter_state *state = seq->private;
1880
	struct net *net = seq_file_net(seq);
L
Linus Torvalds 已提交
1881 1882

	do {
1883
		sk = sk_nulls_next(sk);
1884
	} while (sk && (!net_eq(sock_net(sk), net) || sk->sk_family != state->family));
L
Linus Torvalds 已提交
1885

1886
	if (!sk) {
1887
		if (state->bucket <= state->udp_table->mask)
1888
			spin_unlock_bh(&state->udp_table->hash[state->bucket].lock);
1889
		return udp_get_first(seq, state->bucket + 1);
L
Linus Torvalds 已提交
1890 1891 1892 1893 1894 1895
	}
	return sk;
}

static struct sock *udp_get_idx(struct seq_file *seq, loff_t pos)
{
1896
	struct sock *sk = udp_get_first(seq, 0);
L
Linus Torvalds 已提交
1897 1898

	if (sk)
1899
		while (pos && (sk = udp_get_next(seq, sk)) != NULL)
L
Linus Torvalds 已提交
1900 1901 1902 1903 1904 1905
			--pos;
	return pos ? NULL : sk;
}

static void *udp_seq_start(struct seq_file *seq, loff_t *pos)
{
1906
	struct udp_iter_state *state = seq->private;
1907
	state->bucket = MAX_UDP_PORTS;
1908

1909
	return *pos ? udp_get_idx(seq, *pos-1) : SEQ_START_TOKEN;
L
Linus Torvalds 已提交
1910 1911 1912 1913 1914 1915
}

static void *udp_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	struct sock *sk;

1916
	if (v == SEQ_START_TOKEN)
L
Linus Torvalds 已提交
1917 1918 1919 1920 1921 1922 1923 1924 1925 1926
		sk = udp_get_idx(seq, 0);
	else
		sk = udp_get_next(seq, v);

	++*pos;
	return sk;
}

static void udp_seq_stop(struct seq_file *seq, void *v)
{
1927 1928
	struct udp_iter_state *state = seq->private;

1929
	if (state->bucket <= state->udp_table->mask)
1930
		spin_unlock_bh(&state->udp_table->hash[state->bucket].lock);
L
Linus Torvalds 已提交
1931 1932 1933 1934 1935
}

static int udp_seq_open(struct inode *inode, struct file *file)
{
	struct udp_seq_afinfo *afinfo = PDE(inode)->data;
1936 1937
	struct udp_iter_state *s;
	int err;
1938

1939 1940 1941 1942
	err = seq_open_net(inode, file, &afinfo->seq_ops,
			   sizeof(struct udp_iter_state));
	if (err < 0)
		return err;
1943

1944
	s = ((struct seq_file *)file->private_data)->private;
L
Linus Torvalds 已提交
1945
	s->family		= afinfo->family;
1946
	s->udp_table		= afinfo->udp_table;
1947
	return err;
1948 1949
}

L
Linus Torvalds 已提交
1950
/* ------------------------------------------------------------------------ */
1951
int udp_proc_register(struct net *net, struct udp_seq_afinfo *afinfo)
L
Linus Torvalds 已提交
1952 1953 1954 1955
{
	struct proc_dir_entry *p;
	int rc = 0;

1956 1957 1958 1959
	afinfo->seq_fops.open		= udp_seq_open;
	afinfo->seq_fops.read		= seq_read;
	afinfo->seq_fops.llseek		= seq_lseek;
	afinfo->seq_fops.release	= seq_release_net;
L
Linus Torvalds 已提交
1960

1961 1962 1963 1964
	afinfo->seq_ops.start		= udp_seq_start;
	afinfo->seq_ops.next		= udp_seq_next;
	afinfo->seq_ops.stop		= udp_seq_stop;

1965 1966 1967
	p = proc_create_data(afinfo->name, S_IRUGO, net->proc_net,
			     &afinfo->seq_fops, afinfo);
	if (!p)
L
Linus Torvalds 已提交
1968 1969 1970
		rc = -ENOMEM;
	return rc;
}
E
Eric Dumazet 已提交
1971
EXPORT_SYMBOL(udp_proc_register);
L
Linus Torvalds 已提交
1972

1973
void udp_proc_unregister(struct net *net, struct udp_seq_afinfo *afinfo)
L
Linus Torvalds 已提交
1974
{
1975
	proc_net_remove(net, afinfo->name);
L
Linus Torvalds 已提交
1976
}
E
Eric Dumazet 已提交
1977
EXPORT_SYMBOL(udp_proc_unregister);
1978 1979

/* ------------------------------------------------------------------------ */
1980 1981
static void udp4_format_sock(struct sock *sp, struct seq_file *f,
		int bucket, int *len)
1982 1983
{
	struct inet_sock *inet = inet_sk(sp);
E
Eric Dumazet 已提交
1984 1985 1986 1987
	__be32 dest = inet->inet_daddr;
	__be32 src  = inet->inet_rcv_saddr;
	__u16 destp	  = ntohs(inet->inet_dport);
	__u16 srcp	  = ntohs(inet->inet_sport);
1988

1989
	seq_printf(f, "%5d: %08X:%04X %08X:%04X"
E
Eric Dumazet 已提交
1990
		" %02X %08X:%08X %02X:%08lX %08X %5d %8d %lu %d %p %d%n",
1991
		bucket, src, srcp, dest, destp, sp->sk_state,
1992 1993
		sk_wmem_alloc_get(sp),
		sk_rmem_alloc_get(sp),
1994
		0, 0L, 0, sock_i_uid(sp), 0, sock_i_ino(sp),
E
Eric Dumazet 已提交
1995 1996
		atomic_read(&sp->sk_refcnt), sp,
		atomic_read(&sp->sk_drops), len);
1997 1998 1999 2000 2001 2002 2003 2004
}

int udp4_seq_show(struct seq_file *seq, void *v)
{
	if (v == SEQ_START_TOKEN)
		seq_printf(seq, "%-127s\n",
			   "  sl  local_address rem_address   st tx_queue "
			   "rx_queue tr tm->when retrnsmt   uid  timeout "
E
Eric Dumazet 已提交
2005
			   "inode ref pointer drops");
2006 2007
	else {
		struct udp_iter_state *state = seq->private;
2008
		int len;
2009

2010
		udp4_format_sock(v, seq, state->bucket, &len);
E
Eric Dumazet 已提交
2011
		seq_printf(seq, "%*s\n", 127 - len, "");
2012 2013 2014 2015 2016 2017 2018 2019
	}
	return 0;
}

/* ------------------------------------------------------------------------ */
static struct udp_seq_afinfo udp4_seq_afinfo = {
	.name		= "udp",
	.family		= AF_INET,
2020
	.udp_table	= &udp_table,
2021 2022 2023
	.seq_fops	= {
		.owner	=	THIS_MODULE,
	},
2024 2025 2026
	.seq_ops	= {
		.show		= udp4_seq_show,
	},
2027 2028
};

2029
static int __net_init udp4_proc_init_net(struct net *net)
2030 2031 2032 2033
{
	return udp_proc_register(net, &udp4_seq_afinfo);
}

2034
static void __net_exit udp4_proc_exit_net(struct net *net)
2035 2036 2037 2038 2039 2040 2041 2042 2043
{
	udp_proc_unregister(net, &udp4_seq_afinfo);
}

static struct pernet_operations udp4_net_ops = {
	.init = udp4_proc_init_net,
	.exit = udp4_proc_exit_net,
};

2044 2045
int __init udp4_proc_init(void)
{
2046
	return register_pernet_subsys(&udp4_net_ops);
2047 2048 2049 2050
}

void udp4_proc_exit(void)
{
2051
	unregister_pernet_subsys(&udp4_net_ops);
2052
}
L
Linus Torvalds 已提交
2053 2054
#endif /* CONFIG_PROC_FS */

2055 2056
static __initdata unsigned long uhash_entries;
static int __init set_uhash_entries(char *str)
2057
{
2058 2059 2060 2061 2062 2063 2064 2065
	if (!str)
		return 0;
	uhash_entries = simple_strtoul(str, &str, 0);
	if (uhash_entries && uhash_entries < UDP_HTABLE_SIZE_MIN)
		uhash_entries = UDP_HTABLE_SIZE_MIN;
	return 1;
}
__setup("uhash_entries=", set_uhash_entries);
2066

2067 2068 2069 2070 2071 2072
void __init udp_table_init(struct udp_table *table, const char *name)
{
	unsigned int i;

	if (!CONFIG_BASE_SMALL)
		table->hash = alloc_large_system_hash(name,
2073
			2 * sizeof(struct udp_hslot),
2074 2075 2076 2077 2078 2079 2080 2081 2082 2083 2084
			uhash_entries,
			21, /* one slot per 2 MB */
			0,
			&table->log,
			&table->mask,
			64 * 1024);
	/*
	 * Make sure hash table has the minimum size
	 */
	if (CONFIG_BASE_SMALL || table->mask < UDP_HTABLE_SIZE_MIN - 1) {
		table->hash = kmalloc(UDP_HTABLE_SIZE_MIN *
2085
				      2 * sizeof(struct udp_hslot), GFP_KERNEL);
2086 2087 2088 2089 2090
		if (!table->hash)
			panic(name);
		table->log = ilog2(UDP_HTABLE_SIZE_MIN);
		table->mask = UDP_HTABLE_SIZE_MIN - 1;
	}
2091
	table->hash2 = table->hash + (table->mask + 1);
2092
	for (i = 0; i <= table->mask; i++) {
2093
		INIT_HLIST_NULLS_HEAD(&table->hash[i].head, i);
E
Eric Dumazet 已提交
2094
		table->hash[i].count = 0;
2095 2096
		spin_lock_init(&table->hash[i].lock);
	}
2097 2098 2099 2100 2101
	for (i = 0; i <= table->mask; i++) {
		INIT_HLIST_NULLS_HEAD(&table->hash2[i].head, i);
		table->hash2[i].count = 0;
		spin_lock_init(&table->hash2[i].lock);
	}
2102 2103
}

H
Hideo Aoki 已提交
2104 2105
void __init udp_init(void)
{
2106
	unsigned long nr_pages, limit;
H
Hideo Aoki 已提交
2107

2108
	udp_table_init(&udp_table, "UDP");
H
Hideo Aoki 已提交
2109 2110 2111 2112
	/* Set the pressure threshold up by the same strategy of TCP. It is a
	 * fraction of global memory that is up to 1/2 at 256 MB, decreasing
	 * toward zero with the amount of memory, with a floor of 128 pages.
	 */
2113 2114 2115
	nr_pages = totalram_pages - totalhigh_pages;
	limit = min(nr_pages, 1UL<<(28-PAGE_SHIFT)) >> (20-PAGE_SHIFT);
	limit = (limit * (nr_pages >> (20-PAGE_SHIFT))) >> (PAGE_SHIFT-11);
H
Hideo Aoki 已提交
2116 2117 2118 2119 2120 2121 2122 2123 2124
	limit = max(limit, 128UL);
	sysctl_udp_mem[0] = limit / 4 * 3;
	sysctl_udp_mem[1] = limit;
	sysctl_udp_mem[2] = sysctl_udp_mem[0] * 2;

	sysctl_udp_rmem_min = SK_MEM_QUANTUM;
	sysctl_udp_wmem_min = SK_MEM_QUANTUM;
}

2125 2126 2127 2128 2129 2130 2131 2132 2133 2134 2135 2136 2137 2138 2139 2140 2141 2142 2143 2144 2145 2146 2147 2148 2149 2150 2151 2152 2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172
int udp4_ufo_send_check(struct sk_buff *skb)
{
	const struct iphdr *iph;
	struct udphdr *uh;

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

	iph = ip_hdr(skb);
	uh = udp_hdr(skb);

	uh->check = ~csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
				       IPPROTO_UDP, 0);
	skb->csum_start = skb_transport_header(skb) - skb->head;
	skb->csum_offset = offsetof(struct udphdr, check);
	skb->ip_summed = CHECKSUM_PARTIAL;
	return 0;
}

struct sk_buff *udp4_ufo_fragment(struct sk_buff *skb, int features)
{
	struct sk_buff *segs = ERR_PTR(-EINVAL);
	unsigned int mss;
	int offset;
	__wsum csum;

	mss = skb_shinfo(skb)->gso_size;
	if (unlikely(skb->len <= mss))
		goto out;

	if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) {
		/* Packet is from an untrusted source, reset gso_segs. */
		int type = skb_shinfo(skb)->gso_type;

		if (unlikely(type & ~(SKB_GSO_UDP | SKB_GSO_DODGY) ||
			     !(type & (SKB_GSO_UDP))))
			goto out;

		skb_shinfo(skb)->gso_segs = DIV_ROUND_UP(skb->len, mss);

		segs = NULL;
		goto out;
	}

	/* Do software UFO. Complete and fill in the UDP checksum as HW cannot
	 * do checksum of UDP packets sent as multiple IP fragments.
	 */
	offset = skb->csum_start - skb_headroom(skb);
E
Eric Dumazet 已提交
2173
	csum = skb_checksum(skb, offset, skb->len - offset, 0);
2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185
	offset += skb->csum_offset;
	*(__sum16 *)(skb->data + offset) = csum_fold(csum);
	skb->ip_summed = CHECKSUM_NONE;

	/* Fragment the skb. IP headers of the fragments are updated in
	 * inet_gso_segment()
	 */
	segs = skb_segment(skb, features);
out:
	return segs;
}