udp.c 45.2 KB
Newer Older
Linus Torvalds's avatar
Linus Torvalds committed
1
2
3
4
5
6
7
8
9
/*
 * 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).
 *
 * Version:	$Id: udp.c,v 1.102 2002/02/01 22:01:04 davem Exp $
 *
10
 * Authors:	Ross Biro
Linus Torvalds's avatar
Linus Torvalds committed
11
12
13
14
15
16
17
18
19
20
21
22
 *		Fred N. van Kempen, <waltje@uWalt.NL.Mugnet.ORG>
 *		Arnt Gulbrandsen, <agulbra@nvg.unit.no>
 *		Alan Cox, <Alan.Cox@linux.org>
 *		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
23
24
 *		Alan Cox	:	Tidied select() semantics.
 *		Alan Cox	:	udp_err() fixed properly, also now
Linus Torvalds's avatar
Linus Torvalds committed
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
57
58
 *					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
59
 *					for connect.
Linus Torvalds's avatar
Linus Torvalds committed
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
 *	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
 *
 *
 *		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.
 */
80

Linus Torvalds's avatar
Linus Torvalds committed
81
82
83
84
85
86
87
88
#include <asm/system.h>
#include <asm/uaccess.h>
#include <asm/ioctls.h>
#include <linux/types.h>
#include <linux/fcntl.h>
#include <linux/module.h>
#include <linux/socket.h>
#include <linux/sockios.h>
89
#include <linux/igmp.h>
Linus Torvalds's avatar
Linus Torvalds committed
90
91
92
93
94
95
#include <linux/in.h>
#include <linux/errno.h>
#include <linux/timer.h>
#include <linux/mm.h>
#include <linux/inet.h>
#include <linux/netdevice.h>
96
#include <net/tcp_states.h>
Linus Torvalds's avatar
Linus Torvalds committed
97
98
99
100
101
102
103
#include <linux/skbuff.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <net/icmp.h>
#include <net/route.h>
#include <net/checksum.h>
#include <net/xfrm.h>
104
#include "udp_impl.h"
Linus Torvalds's avatar
Linus Torvalds committed
105
106
107
108
109

/*
 *	Snmp MIB for the UDP layer
 */

110
DEFINE_SNMP_STAT(struct udp_mib, udp_statistics) __read_mostly;
Linus Torvalds's avatar
Linus Torvalds committed
111
112
113
114

struct hlist_head udp_hash[UDP_HTABLE_SIZE];
DEFINE_RWLOCK(udp_hash_lock);

115
static int udp_port_rover;
Linus Torvalds's avatar
Linus Torvalds committed
116

117
118
119
120
121
122
123
124
125
126
127
128
129
/*
 * Note about this hash function :
 * Typical use is probably daddr = 0, only dport is going to vary hash
 */
static inline unsigned int hash_port_and_addr(__u16 port, __be32 addr)
{
	addr ^= addr >> 16;
	addr ^= addr >> 8;
	return port ^ addr;
}

static inline int __udp_lib_port_inuse(unsigned int hash, int port,
	__be32 daddr, struct hlist_head udptable[])
Linus Torvalds's avatar
Linus Torvalds committed
130
{
131
	struct sock *sk;
Linus Torvalds's avatar
Linus Torvalds committed
132
	struct hlist_node *node;
133
	struct inet_sock *inet;
134

135
136
137
138
139
140
141
	sk_for_each(sk, node, &udptable[hash & (UDP_HTABLE_SIZE - 1)]) {
		if (sk->sk_hash != hash)
			continue;
		inet = inet_sk(sk);
		if (inet->num != port)
			continue;
		if (inet->rcv_saddr == daddr)
142
			return 1;
143
	}
144
145
146
147
	return 0;
}

/**
148
 *  __udp_lib_get_port  -  UDP/-Lite port lookup for IPv4 and IPv6
149
150
151
 *
 *  @sk:          socket struct in question
 *  @snum:        port number to look up
152
153
 *  @udptable:    hash list table, must be of UDP_HTABLE_SIZE
 *  @port_rover:  pointer to record of last unallocated port
154
155
 *  @saddr_comp:  AF-dependent comparison of bound local IP addresses
 */
156
157
158
159
int __udp_lib_get_port(struct sock *sk, unsigned short snum,
		       struct hlist_head udptable[], int *port_rover,
		       int (*saddr_comp)(const struct sock *sk1,
					 const struct sock *sk2 )    )
160
161
162
{
	struct hlist_node *node;
	struct hlist_head *head;
Linus Torvalds's avatar
Linus Torvalds committed
163
	struct sock *sk2;
164
	unsigned int hash;
165
	int    error = 1;
Linus Torvalds's avatar
Linus Torvalds committed
166
167
168
169
170

	write_lock_bh(&udp_hash_lock);
	if (snum == 0) {
		int best_size_so_far, best, result, i;

171
172
173
		if (*port_rover > sysctl_local_port_range[1] ||
		    *port_rover < sysctl_local_port_range[0])
			*port_rover = sysctl_local_port_range[0];
Linus Torvalds's avatar
Linus Torvalds committed
174
		best_size_so_far = 32767;
175
		best = result = *port_rover;
Linus Torvalds's avatar
Linus Torvalds committed
176
177
178
		for (i = 0; i < UDP_HTABLE_SIZE; i++, result++) {
			int size;

179
180
181
			hash = hash_port_and_addr(result,
					inet_sk(sk)->rcv_saddr);
			head = &udptable[hash & (UDP_HTABLE_SIZE - 1)];
182
			if (hlist_empty(head)) {
Linus Torvalds's avatar
Linus Torvalds committed
183
184
185
186
187
188
189
				if (result > sysctl_local_port_range[1])
					result = sysctl_local_port_range[0] +
						((result - sysctl_local_port_range[0]) &
						 (UDP_HTABLE_SIZE - 1));
				goto gotit;
			}
			size = 0;
190
191
192
193
194
195
196
197
			sk_for_each(sk2, node, head) {
				if (++size >= best_size_so_far)
					goto next;
			}
			best_size_so_far = size;
			best = result;
		next:
			;
Linus Torvalds's avatar
Linus Torvalds committed
198
199
		}
		result = best;
200
201
		for (i = 0; i < (1 << 16) / UDP_HTABLE_SIZE;
		     i++, result += UDP_HTABLE_SIZE) {
Linus Torvalds's avatar
Linus Torvalds committed
202
203
204
205
			if (result > sysctl_local_port_range[1])
				result = sysctl_local_port_range[0]
					+ ((result - sysctl_local_port_range[0]) &
					   (UDP_HTABLE_SIZE - 1));
206
207
208
209
210
211
212
			hash = hash_port_and_addr(result, 0);
			if (__udp_lib_port_inuse(hash, result,
						 0, udptable))
				continue;
			if (!inet_sk(sk)->rcv_saddr)
				break;

213
214
215
216
			hash = hash_port_and_addr(result,
					inet_sk(sk)->rcv_saddr);
			if (! __udp_lib_port_inuse(hash, result,
				inet_sk(sk)->rcv_saddr, udptable))
Linus Torvalds's avatar
Linus Torvalds committed
217
218
219
220
221
				break;
		}
		if (i >= (1 << 16) / UDP_HTABLE_SIZE)
			goto fail;
gotit:
222
		*port_rover = snum = result;
Linus Torvalds's avatar
Linus Torvalds committed
223
	} else {
224
		hash = hash_port_and_addr(snum, 0);
225
		head = &udptable[hash & (UDP_HTABLE_SIZE - 1)];
226
227

		sk_for_each(sk2, node, head)
228
229
230
231
232
233
234
			if (sk2->sk_hash == hash &&
			    sk2 != sk &&
			    inet_sk(sk2)->num == snum &&
			    (!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) &&
			    (*saddr_comp)(sk, sk2))
Linus Torvalds's avatar
Linus Torvalds committed
235
				goto fail;
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253

		if (inet_sk(sk)->rcv_saddr) {
			hash = hash_port_and_addr(snum,
						  inet_sk(sk)->rcv_saddr);
			head = &udptable[hash & (UDP_HTABLE_SIZE - 1)];

			sk_for_each(sk2, node, head)
				if (sk2->sk_hash == hash &&
				    sk2 != sk &&
				    inet_sk(sk2)->num == snum &&
				    (!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) &&
				    (*saddr_comp)(sk, sk2))
					goto fail;
		}
Linus Torvalds's avatar
Linus Torvalds committed
254
	}
255
	inet_sk(sk)->num = snum;
256
	sk->sk_hash = hash;
Linus Torvalds's avatar
Linus Torvalds committed
257
	if (sk_unhashed(sk)) {
258
		head = &udptable[hash & (UDP_HTABLE_SIZE - 1)];
259
		sk_add_node(sk, head);
Linus Torvalds's avatar
Linus Torvalds committed
260
261
		sock_prot_inc_use(sk->sk_prot);
	}
262
	error = 0;
Linus Torvalds's avatar
Linus Torvalds committed
263
264
fail:
	write_unlock_bh(&udp_hash_lock);
265
266
267
	return error;
}

Stephen Hemminger's avatar
Stephen Hemminger committed
268
int udp_get_port(struct sock *sk, unsigned short snum,
269
270
271
272
273
			int (*scmp)(const struct sock *, const struct sock *))
{
	return  __udp_lib_get_port(sk, snum, udp_hash, &udp_port_rover, scmp);
}

Stephen Hemminger's avatar
Stephen Hemminger committed
274
int ipv4_rcv_saddr_equal(const struct sock *sk1, const struct sock *sk2)
275
276
277
278
279
280
281
282
283
284
285
{
	struct inet_sock *inet1 = inet_sk(sk1), *inet2 = inet_sk(sk2);

	return 	( !ipv6_only_sock(sk2)  &&
		  (!inet1->rcv_saddr || !inet2->rcv_saddr ||
		   inet1->rcv_saddr == inet2->rcv_saddr      ));
}

static inline int udp_v4_get_port(struct sock *sk, unsigned short snum)
{
	return udp_get_port(sk, snum, ipv4_rcv_saddr_equal);
Linus Torvalds's avatar
Linus Torvalds committed
286
287
288
289
290
}

/* UDP is nearly always wildcards out the wazoo, it makes no sense to try
 * harder than this. -DaveM
 */
291
292
293
static struct sock *__udp4_lib_lookup(__be32 saddr, __be16 sport,
				      __be32 daddr, __be16 dport,
				      int dif, struct hlist_head udptable[])
Linus Torvalds's avatar
Linus Torvalds committed
294
295
296
{
	struct sock *sk, *result = NULL;
	struct hlist_node *node;
297
	unsigned int hash, hashwild;
298
	int score, best = -1, hport = ntohs(dport);
299

300
301
 	hash = hash_port_and_addr(hport, daddr);
 	hashwild = hash_port_and_addr(hport, 0);
Linus Torvalds's avatar
Linus Torvalds committed
302

303
	read_lock(&udp_hash_lock);
304
305
306
307

lookup:

	sk_for_each(sk, node, &udptable[hash & (UDP_HTABLE_SIZE - 1)]) {
Linus Torvalds's avatar
Linus Torvalds committed
308
309
		struct inet_sock *inet = inet_sk(sk);

310
		if (sk->sk_hash != hash || ipv6_only_sock(sk) ||
311
			inet->num != hport)
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
			continue;

		score = (sk->sk_family == PF_INET ? 1 : 0);
		if (inet->rcv_saddr) {
			if (inet->rcv_saddr != daddr)
				continue;
			score+=2;
		}
		if (inet->daddr) {
			if (inet->daddr != saddr)
				continue;
			score+=2;
		}
		if (inet->dport) {
			if (inet->dport != sport)
				continue;
			score+=2;
		}
		if (sk->sk_bound_dev_if) {
			if (sk->sk_bound_dev_if != dif)
				continue;
			score+=2;
		}
		if (score == 9) {
			result = sk;
			goto found;
		} else if (score > best) {
			result = sk;
			best = score;
Linus Torvalds's avatar
Linus Torvalds committed
341
342
		}
	}
343
344
345
346
347
348

	if (hash != hashwild) {
		hash = hashwild;
		goto lookup;
	}
found:
349
350
	if (result)
		sock_hold(result);
Linus Torvalds's avatar
Linus Torvalds committed
351
	read_unlock(&udp_hash_lock);
352
	return result;
Linus Torvalds's avatar
Linus Torvalds committed
353
354
}

355
356
357
358
static inline struct sock *udp_v4_mcast_next(struct sock *sk, unsigned int hnum,
					     int hport, __be32 loc_addr,
					     __be16 rmt_port, __be32 rmt_addr,
					     int dif)
Linus Torvalds's avatar
Linus Torvalds committed
359
360
361
362
363
364
365
{
	struct hlist_node *node;
	struct sock *s = sk;

	sk_for_each_from(s, node) {
		struct inet_sock *inet = inet_sk(s);

366
		if (s->sk_hash != hnum					||
367
		    inet->num != hport					||
Linus Torvalds's avatar
Linus Torvalds committed
368
369
370
371
372
373
374
375
376
		    (inet->daddr && inet->daddr != rmt_addr)		||
		    (inet->dport != rmt_port && inet->dport)		||
		    (inet->rcv_saddr && inet->rcv_saddr != loc_addr)	||
		    ipv6_only_sock(s)					||
		    (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;
377
	}
Linus Torvalds's avatar
Linus Torvalds committed
378
379
	s = NULL;
found:
380
	return s;
Linus Torvalds's avatar
Linus Torvalds committed
381
382
383
384
385
386
}

/*
 * 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
387
 * it's just the icmp type << 8 | icmp code.
Linus Torvalds's avatar
Linus Torvalds committed
388
389
390
391
392
393
 * 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.
 */

394
void __udp4_lib_err(struct sk_buff *skb, u32 info, struct hlist_head udptable[])
Linus Torvalds's avatar
Linus Torvalds committed
395
396
397
398
{
	struct inet_sock *inet;
	struct iphdr *iph = (struct iphdr*)skb->data;
	struct udphdr *uh = (struct udphdr*)(skb->data+(iph->ihl<<2));
399
400
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
Linus Torvalds's avatar
Linus Torvalds committed
401
402
403
404
	struct sock *sk;
	int harderr;
	int err;

405
406
	sk = __udp4_lib_lookup(iph->daddr, uh->dest, iph->saddr, uh->source,
			       skb->dev->ifindex, udptable		    );
Linus Torvalds's avatar
Linus Torvalds committed
407
408
	if (sk == NULL) {
		ICMP_INC_STATS_BH(ICMP_MIB_INERRORS);
409
		return;	/* No socket for error */
Linus Torvalds's avatar
Linus Torvalds committed
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
442
443
444
	}

	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;
	}

	/*
445
	 *      RFC1122: OK.  Passes ICMP errors back to application, as per
Linus Torvalds's avatar
Linus Torvalds committed
446
447
448
449
450
451
452
453
454
455
456
457
458
459
	 *	4.1.3.3.
	 */
	if (!inet->recverr) {
		if (!harderr || sk->sk_state != TCP_ESTABLISHED)
			goto out;
	} else {
		ip_icmp_error(sk, skb, err, uh->dest, info, (u8*)(uh+1));
	}
	sk->sk_err = err;
	sk->sk_error_report(sk);
out:
	sock_put(sk);
}

Stephen Hemminger's avatar
Stephen Hemminger committed
460
void udp_err(struct sk_buff *skb, u32 info)
461
462
463
464
{
	return __udp4_lib_err(skb, info, udp_hash);
}

Linus Torvalds's avatar
Linus Torvalds committed
465
466
467
468
469
470
471
472
473
474
475
476
477
478
/*
 * Throw away all pending data and cancel the corking. Socket is locked.
 */
static void udp_flush_pending_frames(struct sock *sk)
{
	struct udp_sock *up = udp_sk(sk);

	if (up->pending) {
		up->len = 0;
		up->pending = 0;
		ip_flush_pending_frames(sk);
	}
}

479
480
481
482
483
484
485
486
487
/**
 * 	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,
				 __be32 src, __be32 dst, int len      )
{
488
	unsigned int offset;
489
	struct udphdr *uh = udp_hdr(skb);
490
	__wsum csum = 0;
491
492
493
494
495

	if (skb_queue_len(&sk->sk_write_queue) == 1) {
		/*
		 * Only one fragment on the socket.
		 */
496
		skb->csum_start = skb_transport_header(skb) - skb->head;
Al Viro's avatar
Al Viro committed
497
		skb->csum_offset = offsetof(struct udphdr, check);
498
499
500
501
502
503
504
		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
		 */
505
		offset = skb_transport_offset(skb);
506
507
508
509
510
511
512
513
514
515
		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)
516
			uh->check = CSUM_MANGLED_0;
517
518
519
	}
}

Linus Torvalds's avatar
Linus Torvalds committed
520
521
522
/*
 * Push out all pending data as one UDP datagram. Socket is locked.
 */
523
static int udp_push_pending_frames(struct sock *sk)
Linus Torvalds's avatar
Linus Torvalds committed
524
{
525
	struct udp_sock  *up = udp_sk(sk);
Linus Torvalds's avatar
Linus Torvalds committed
526
527
528
529
530
	struct inet_sock *inet = inet_sk(sk);
	struct flowi *fl = &inet->cork.fl;
	struct sk_buff *skb;
	struct udphdr *uh;
	int err = 0;
531
	__wsum csum = 0;
Linus Torvalds's avatar
Linus Torvalds committed
532
533
534
535
536
537
538
539

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

	/*
	 * Create a UDP header
	 */
540
	uh = udp_hdr(skb);
Linus Torvalds's avatar
Linus Torvalds committed
541
542
543
544
545
	uh->source = fl->fl_ip_sport;
	uh->dest = fl->fl_ip_dport;
	uh->len = htons(up->len);
	uh->check = 0;

546
547
548
549
550
	if (up->pcflag)  				 /*     UDP-Lite      */
		csum  = udplite_csum_outgoing(sk, skb);

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

Linus Torvalds's avatar
Linus Torvalds committed
551
552
553
		skb->ip_summed = CHECKSUM_NONE;
		goto send;

554
	} else if (skb->ip_summed == CHECKSUM_PARTIAL) { /* UDP hardware csum */
Linus Torvalds's avatar
Linus Torvalds committed
555

556
557
558
559
560
561
562
563
564
565
		udp4_hwcsum_outgoing(sk, skb, fl->fl4_src,fl->fl4_dst, up->len);
		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,
				      sk->sk_protocol, csum             );
	if (uh->check == 0)
566
		uh->check = CSUM_MANGLED_0;
Linus Torvalds's avatar
Linus Torvalds committed
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585

send:
	err = ip_push_pending_frames(sk);
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;
586
	__be32 daddr, faddr, saddr;
587
	__be16 dport;
Linus Torvalds's avatar
Linus Torvalds committed
588
	u8  tos;
589
	int err, is_udplite = up->pcflag;
Linus Torvalds's avatar
Linus Torvalds committed
590
	int corkreq = up->corkflag || msg->msg_flags&MSG_MORE;
591
	int (*getfrag)(void *, char *, int, int, int, struct sk_buff *);
Linus Torvalds's avatar
Linus Torvalds committed
592
593
594
595

	if (len > 0xFFFF)
		return -EMSGSIZE;

596
	/*
Linus Torvalds's avatar
Linus Torvalds committed
597
598
599
600
601
602
603
604
605
606
607
	 *	Check the flags.
	 */

	if (msg->msg_flags&MSG_OOB)	/* Mirror BSD error message compatibility */
		return -EOPNOTSUPP;

	ipc.opt = NULL;

	if (up->pending) {
		/*
		 * There are pending frames.
608
		 * The socket lock must be held while it's corked.
Linus Torvalds's avatar
Linus Torvalds committed
609
610
611
612
613
614
615
		 */
		lock_sock(sk);
		if (likely(up->pending)) {
			if (unlikely(up->pending != AF_INET)) {
				release_sock(sk);
				return -EINVAL;
			}
616
			goto do_append_data;
Linus Torvalds's avatar
Linus Torvalds committed
617
618
619
620
621
622
		}
		release_sock(sk);
	}
	ulen += sizeof(struct udphdr);

	/*
623
	 *	Get and verify the address.
Linus Torvalds's avatar
Linus Torvalds committed
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
	 */
	if (msg->msg_name) {
		struct sockaddr_in * usin = (struct sockaddr_in*)msg->msg_name;
		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;
		daddr = inet->daddr;
		dport = inet->dport;
		/* Open fast path for connected socket.
		   Route will not be used, if at least one option is set.
		 */
		connected = 1;
647
	}
Linus Torvalds's avatar
Linus Torvalds committed
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
	ipc.addr = inet->saddr;

	ipc.oif = sk->sk_bound_dev_if;
	if (msg->msg_controllen) {
		err = ip_cmsg_send(msg, &ipc);
		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) ||
673
	    (msg->msg_flags & MSG_DONTROUTE) ||
Linus Torvalds's avatar
Linus Torvalds committed
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
	    (ipc.opt && ipc.opt->is_strictroute)) {
		tos |= RTO_ONLINK;
		connected = 0;
	}

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

	if (connected)
		rt = (struct rtable*)sk_dst_check(sk, 0);

	if (rt == NULL) {
		struct flowi fl = { .oif = ipc.oif,
				    .nl_u = { .ip4_u =
					      { .daddr = faddr,
						.saddr = saddr,
						.tos = tos } },
696
				    .proto = sk->sk_protocol,
Linus Torvalds's avatar
Linus Torvalds committed
697
698
699
				    .uli_u = { .ports =
					       { .sport = inet->sport,
						 .dport = dport } } };
700
		security_sk_classify_flow(sk, &fl);
701
		err = ip_route_output_flow(&rt, &fl, sk, 1);
Linus Torvalds's avatar
Linus Torvalds committed
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
		if (err)
			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);

727
		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 2\n");
Linus Torvalds's avatar
Linus Torvalds committed
728
729
730
731
732
733
734
735
736
737
738
739
740
741
		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;
	inet->cork.fl.fl_ip_sport = inet->sport;
	up->pending = AF_INET;

do_append_data:
	up->len += ulen;
742
743
744
	getfrag  =  is_udplite ?  udplite_getfrag : ip_generic_getfrag;
	err = ip_append_data(sk, getfrag, msg->msg_iov, ulen,
			sizeof(struct udphdr), &ipc, rt,
Linus Torvalds's avatar
Linus Torvalds committed
745
746
747
748
			corkreq ? msg->msg_flags|MSG_MORE : msg->msg_flags);
	if (err)
		udp_flush_pending_frames(sk);
	else if (!corkreq)
749
		err = udp_push_pending_frames(sk);
Herbert Xu's avatar
Herbert Xu committed
750
751
	else if (unlikely(skb_queue_empty(&sk->sk_write_queue)))
		up->pending = 0;
Linus Torvalds's avatar
Linus Torvalds committed
752
753
754
755
756
757
758
	release_sock(sk);

out:
	ip_rt_put(rt);
	if (free)
		kfree(ipc.opt);
	if (!err) {
759
		UDP_INC_STATS_USER(UDP_MIB_OUTDATAGRAMS, is_udplite);
Linus Torvalds's avatar
Linus Torvalds committed
760
761
		return len;
	}
762
763
764
765
766
767
768
769
	/*
	 * 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)) {
770
		UDP_INC_STATS_USER(UDP_MIB_SNDBUFERRORS, is_udplite);
771
	}
Linus Torvalds's avatar
Linus Torvalds committed
772
773
774
775
776
777
778
779
780
781
	return err;

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

782
783
int udp_sendpage(struct sock *sk, struct page *page, int offset,
		 size_t size, int flags)
Linus Torvalds's avatar
Linus Torvalds committed
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
{
	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);

805
		LIMIT_NETDEBUG(KERN_DEBUG "udp cork app bug 3\n");
Linus Torvalds's avatar
Linus Torvalds committed
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
		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)))
822
		ret = udp_push_pending_frames(sk);
Linus Torvalds's avatar
Linus Torvalds committed
823
824
825
826
827
828
829
830
831
832
	if (!ret)
		ret = size;
out:
	release_sock(sk);
	return ret;
}

/*
 *	IOCTL requests applicable to the UDP protocol
 */
833

Linus Torvalds's avatar
Linus Torvalds committed
834
835
int udp_ioctl(struct sock *sk, int cmd, unsigned long arg)
{
836
837
	switch (cmd) {
	case SIOCOUTQ:
Linus Torvalds's avatar
Linus Torvalds committed
838
	{
839
840
841
		int amount = atomic_read(&sk->sk_wmem_alloc);
		return put_user(amount, (int __user *)arg);
	}
Linus Torvalds's avatar
Linus Torvalds committed
842

843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
	case SIOCINQ:
	{
		struct sk_buff *skb;
		unsigned long amount;

		amount = 0;
		spin_lock_bh(&sk->sk_receive_queue.lock);
		skb = skb_peek(&sk->sk_receive_queue);
		if (skb != NULL) {
			/*
			 * We will only return the amount
			 * of this packet since that is all
			 * that will be read.
			 */
			amount = skb->len - sizeof(struct udphdr);
Linus Torvalds's avatar
Linus Torvalds committed
858
		}
859
860
861
		spin_unlock_bh(&sk->sk_receive_queue.lock);
		return put_user(amount, (int __user *)arg);
	}
Linus Torvalds's avatar
Linus Torvalds committed
862

863
864
	default:
		return -ENOIOCTLCMD;
Linus Torvalds's avatar
Linus Torvalds committed
865
	}
866
867

	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
868
869
870
871
872
873
874
}

/*
 * 	This should be easy, if there is something there we
 * 	return it, otherwise we block.
 */

875
int udp_recvmsg(struct kiocb *iocb, struct sock *sk, struct msghdr *msg,
876
		size_t len, int noblock, int flags, int *addr_len)
Linus Torvalds's avatar
Linus Torvalds committed
877
878
{
	struct inet_sock *inet = inet_sk(sk);
879
880
	struct sockaddr_in *sin = (struct sockaddr_in *)msg->msg_name;
	struct sk_buff *skb;
881
882
883
	unsigned int ulen, copied;
	int err;
	int is_udplite = IS_UDPLITE(sk);
Linus Torvalds's avatar
Linus Torvalds committed
884
885
886
887
888
889
890
891
892
893
894
895
896
897

	/*
	 *	Check any passed addresses
	 */
	if (addr_len)
		*addr_len=sizeof(*sin);

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

try_again:
	skb = skb_recv_datagram(sk, flags, noblock, &err);
	if (!skb)
		goto out;
898

899
900
901
902
903
	ulen = skb->len - sizeof(struct udphdr);
	copied = len;
	if (copied > ulen)
		copied = ulen;
	else if (copied < ulen)
Linus Torvalds's avatar
Linus Torvalds committed
904
905
		msg->msg_flags |= MSG_TRUNC;

906
	/*
907
908
909
	 * 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.
910
911
	 */

912
913
	if (copied < ulen || UDP_SKB_CB(skb)->partial_cov) {
		if (udp_lib_checksum_complete(skb))
Linus Torvalds's avatar
Linus Torvalds committed
914
			goto csum_copy_err;
915
916
	}

917
	if (skb_csum_unnecessary(skb))
918
919
920
		err = skb_copy_datagram_iovec(skb, sizeof(struct udphdr),
					      msg->msg_iov, copied       );
	else {
Linus Torvalds's avatar
Linus Torvalds committed
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
		err = skb_copy_and_csum_datagram_iovec(skb, sizeof(struct udphdr), msg->msg_iov);

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

	if (err)
		goto out_free;

	sock_recv_timestamp(msg, sk, skb);

	/* Copy the address. */
	if (sin)
	{
		sin->sin_family = AF_INET;
936
		sin->sin_port = udp_hdr(skb)->source;
937
		sin->sin_addr.s_addr = ip_hdr(skb)->saddr;
Linus Torvalds's avatar
Linus Torvalds committed
938
		memset(sin->sin_zero, 0, sizeof(sin->sin_zero));
939
	}
Linus Torvalds's avatar
Linus Torvalds committed
940
941
942
943
944
	if (inet->cmsg_flags)
		ip_cmsg_recv(msg, skb);

	err = copied;
	if (flags & MSG_TRUNC)
945
		err = ulen;
946

Linus Torvalds's avatar
Linus Torvalds committed
947
out_free:
948
	skb_free_datagram(sk, skb);
Linus Torvalds's avatar
Linus Torvalds committed
949
out:
950
	return err;
Linus Torvalds's avatar
Linus Torvalds committed
951
952

csum_copy_err:
953
	UDP_INC_STATS_BH(UDP_MIB_INERRORS, is_udplite);
Linus Torvalds's avatar
Linus Torvalds committed
954

955
	skb_kill_datagram(sk, skb, flags);
Linus Torvalds's avatar
Linus Torvalds committed
956
957

	if (noblock)
958
		return -EAGAIN;
Linus Torvalds's avatar
Linus Torvalds committed
959
960
961
962
963
964
965
966
967
968
	goto try_again;
}


int udp_disconnect(struct sock *sk, int flags)
{
	struct inet_sock *inet = inet_sk(sk);
	/*
	 *	1003.1g - break association.
	 */
969

Linus Torvalds's avatar
Linus Torvalds committed
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
	sk->sk_state = TCP_CLOSE;
	inet->daddr = 0;
	inet->dport = 0;
	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);
		inet->sport = 0;
	}
	sk_dst_reset(sk);
	return 0;
}

/* return:
 * 	1  if the the UDP system should process it
 *	0  if we should drop this packet
 * 	-1 if it should get processed by xfrm4_rcv_encap
 */
static int udp_encap_rcv(struct sock * sk, struct sk_buff *skb)
{
#ifndef CONFIG_XFRM
993
	return 1;
Linus Torvalds's avatar
Linus Torvalds committed
994
995
#else
	struct udp_sock *up = udp_sk(sk);
996
	struct udphdr *uh;
Linus Torvalds's avatar
Linus Torvalds committed
997
998
	struct iphdr *iph;
	int iphlen, len;
999

1000
1001
	__u8 *udpdata;
	__be32 *udpdata32;
Linus Torvalds's avatar
Linus Torvalds committed
1002
1003
1004
	__u16 encap_type = up->encap_type;

	/* if we're overly short, let UDP handle it */
1005
1006
	len = skb->len - sizeof(struct udphdr);
	if (len <= 0)
Linus Torvalds's avatar
Linus Torvalds committed
1007
1008
1009
1010
1011
1012
		return 1;

	/* if this is not encapsulated socket, then just return now */
	if (!encap_type)
		return 1;

1013
1014
1015
1016
1017
1018
	/* If this is a paged skb, make sure we pull up
	 * whatever data we need to look at. */
	if (!pskb_may_pull(skb, sizeof(struct udphdr) + min(len, 8)))
		return 1;

	/* Now we can get the pointers */
1019
	uh = udp_hdr(skb);
1020
1021
	udpdata = (__u8 *)uh + sizeof(struct udphdr);
	udpdata32 = (__be32 *)udpdata;
Linus Torvalds's avatar
Linus Torvalds committed
1022
1023
1024
1025
1026
1027
1028

	switch (encap_type) {
	default:
	case UDP_ENCAP_ESPINUDP:
		/* Check if this is a keepalive packet.  If so, eat it. */
		if (len == 1 && udpdata[0] == 0xff) {
			return 0;
1029
		} else if (len > sizeof(struct ip_esp_hdr) && udpdata32[0] != 0) {
Linus Torvalds's avatar
Linus Torvalds committed
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
			/* ESP Packet without Non-ESP header */
			len = sizeof(struct udphdr);
		} else
			/* Must be an IKE packet.. pass it through */
			return 1;
		break;
	case UDP_ENCAP_ESPINUDP_NON_IKE:
		/* Check if this is a keepalive packet.  If so, eat it. */
		if (len == 1 && udpdata[0] == 0xff) {
			return 0;
		} else if (len > 2 * sizeof(u32) + sizeof(struct ip_esp_hdr) &&
			   udpdata32[0] == 0 && udpdata32[1] == 0) {
1042

Linus Torvalds's avatar
Linus Torvalds committed
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
			/* ESP Packet with Non-IKE marker */
			len = sizeof(struct udphdr) + 2 * sizeof(u32);
		} else
			/* Must be an IKE packet.. pass it through */
			return 1;
		break;
	}

	/* At this point we are sure that this is an ESPinUDP packet,
	 * so we need to remove 'len' bytes from the packet (the UDP
	 * header and optional ESP marker bytes) and then modify the
	 * protocol to ESP, and then call into the transform receiver.
	 */
1056
1057
	if (skb_cloned(skb) && pskb_expand_head(skb, 0, 0, GFP_ATOMIC))
		return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1058
1059

	/* Now we can update and verify the packet length... */
1060
	iph = ip_hdr(skb);
Linus Torvalds's avatar
Linus Torvalds committed
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
	iphlen = iph->ihl << 2;
	iph->tot_len = htons(ntohs(iph->tot_len) - len);
	if (skb->len < iphlen + len) {
		/* packet is too small!?! */
		return 0;
	}

	/* pull the data buffer up to the ESP header and set the
	 * transport header to point to ESP.  Keep UDP on the stack
	 * for later.
	 */
1072
1073
	__skb_pull(skb, len);
	skb_reset_transport_header(skb);
Linus Torvalds's avatar
Linus Torvalds committed
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090

	/* modify the protocol (it's ESP!) */
	iph->protocol = IPPROTO_ESP;

	/* and let the caller know to send this into the ESP processor... */
	return -1;
#endif
}

/* 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.
 */
1091
int udp_queue_rcv_skb(struct sock * sk, struct sk_buff *skb)
Linus Torvalds's avatar
Linus Torvalds committed
1092
1093
{
	struct udp_sock *up = udp_sk(sk);
1094
	int rc;
Linus Torvalds's avatar
Linus Torvalds committed
1095
1096
1097
1098

	/*
	 *	Charge it to the socket, dropping if the queue is full.
	 */
1099
1100
	if (!xfrm4_policy_check(sk, XFRM_POLICY_IN, skb))
		goto drop;
1101
	nf_reset(skb);
Linus Torvalds's avatar
Linus Torvalds committed
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123

	if (up->encap_type) {
		/*
		 * This is an encapsulation socket, so let's see if this is
		 * an encapsulated packet.
		 * If it's a keepalive packet, then just eat it.
		 * If it's an encapsulateed packet, then pass it to the
		 * IPsec xfrm input and return the response
		 * appropriately.  Otherwise, just fall through and
		 * pass this up the UDP socket.
		 */
		int ret;

		ret = udp_encap_rcv(sk, skb);
		if (ret == 0) {
			/* Eat the packet .. */
			kfree_skb(skb);
			return 0;
		}
		if (ret < 0) {
			/* process the ESP packet */
			ret = xfrm4_rcv_encap(skb, up->encap_type);
1124
			UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag);
Linus Torvalds's avatar
Linus Torvalds committed
1125
1126
1127
1128
1129
			return -ret;
		}
		/* FALLTHROUGH -- it's a UDP Packet */
	}

1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
	/*
	 * 	UDP-Lite specific tests, ignored on UDP sockets
	 */
	if ((up->pcflag & 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;
Linus Torvalds's avatar
Linus Torvalds committed
1151
		}
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
		/* 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;
		}
	}

1166
1167
	if (sk->sk_filter) {
		if (udp_lib_checksum_complete(skb))
1168
			goto drop;
Linus Torvalds's avatar
Linus Torvalds committed
1169
1170
	}

1171
1172
1173
	if ((rc = sock_queue_rcv_skb(sk,skb)) < 0) {
		/* Note that an ENOMEM error is charged twice */
		if (rc == -ENOMEM)
1174
1175
			UDP_INC_STATS_BH(UDP_MIB_RCVBUFERRORS, up->pcflag);
		goto drop;
Linus Torvalds's avatar
Linus Torvalds committed
1176
	}
1177
1178

	UDP_INC_STATS_BH(UDP_MIB_INDATAGRAMS, up->pcflag);
Linus Torvalds's avatar
Linus Torvalds committed
1179
	return 0;
1180
1181
1182
1183
1184

drop:
	UDP_INC_STATS_BH(UDP_MIB_INERRORS, up->pcflag);
	kfree_skb(skb);
	return -1;
Linus Torvalds's avatar
Linus Torvalds committed
1185
1186
1187
1188
1189
1190
1191
1192
}

/*
 *	Multicasts and broadcasts go to each listener.
 *
 *	Note: called only from the BH handler context,
 *	so we don't need to lock the hashes.
 */
1193
1194
1195
1196
static int __udp4_lib_mcast_deliver(struct sk_buff *skb,
				    struct udphdr  *uh,
				    __be32 saddr, __be32 daddr,
				    struct hlist_head udptable[])
Linus Torvalds's avatar
Linus Torvalds committed
1197
{
1198
	struct sock *sk, *skw, *sknext;
Linus Torvalds's avatar
Linus Torvalds committed
1199
	int dif;
1200
1201
1202
	int hport = ntohs(uh->dest);
	unsigned int hash = hash_port_and_addr(hport, daddr);
	unsigned int hashwild = hash_port_and_addr(hport, 0);
Linus Torvalds's avatar
Linus Torvalds committed
1203
1204
1205

	dif = skb->dev->ifindex;

1206
1207
1208
1209
1210
	read_lock(&udp_hash_lock);

	sk = sk_head(&udptable[hash & (UDP_HTABLE_SIZE - 1)]);
	skw = sk_head(&udptable[hashwild & (UDP_HTABLE_SIZE - 1)]);

1211
	sk = udp_v4_mcast_next(sk, hash, hport, daddr, uh->source, saddr, dif);
1212
1213
	if (!sk) {
		hash = hashwild;
1214
		sk = udp_v4_mcast_next(skw, hash, hport, daddr, uh->source,
1215
1216
1217
			saddr, dif);
	}
	if (sk) {
Linus Torvalds's avatar
Linus Torvalds committed
1218
1219
		do {
			struct sk_buff *skb1 = skb;
1220
1221
			sknext = udp_v4_mcast_next(sk_next(sk), hash, hport,
						daddr, uh->source, saddr, dif);
1222
1223
			if (!sknext && hash != hashwild) {
				hash = hashwild;
1224
				sknext = udp_v4_mcast_next(skw, hash, hport,
1225
1226
					daddr, uh->source, saddr, dif);
			}
1227
			if (sknext)
Linus Torvalds's avatar
Linus Torvalds committed
1228
1229
				skb1 = skb_clone(skb, GFP_ATOMIC);

1230
			if (skb1) {
Linus Torvalds's avatar
Linus Torvalds committed
1231
1232
				int ret = udp_queue_rcv_skb(sk, skb1);
				if (ret > 0)
1233
1234
1235
1236
					/*
					 * we should probably re-process
					 * instead of dropping packets here.
					 */
Linus Torvalds's avatar
Linus Torvalds committed
1237
1238
1239
					kfree_skb(skb1);
			}
			sk = sknext;
1240
		} while (sknext);
Linus Torvalds's avatar
Linus Torvalds committed
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
	} else
		kfree_skb(skb);
	read_unlock(&udp_hash_lock);
	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.
 */
1252
1253
static inline int udp4_csum_init(struct sk_buff *skb, struct udphdr *uh,
				 int proto)
Linus Torvalds's avatar
Linus Torvalds committed
1254
{
1255
	const struct iphdr *iph;
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
	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;
	}

1267
	iph = ip_hdr(skb);
Linus Torvalds's avatar
Linus Torvalds committed
1268
1269
	if (uh->check == 0) {
		skb->ip_summed = CHECKSUM_UNNECESSARY;
1270
	} else if (skb->ip_summed == CHECKSUM_COMPLETE) {
1271
1272
	       if (!csum_tcpudp_magic(iph->saddr, iph->daddr, skb->len,
				      proto, skb->csum))
1273
			skb->ip_summed = CHECKSUM_UNNECESSARY;
Linus Torvalds's avatar
Linus Torvalds committed
1274
	}
1275
	if (!skb_csum_unnecessary(skb))
1276
		skb->csum = csum_tcpudp_nofold(iph->saddr, iph->daddr,
1277
					       skb->len, proto, 0);
Linus Torvalds's avatar
Linus Torvalds committed
1278
1279
1280
	/* Probably, we should checksum udp header (it should be in cache
	 * in any case) and data in tiny packets (< rx copybreak).
	 */
1281

1282
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1283
1284
1285
}

/*
1286
 *	All we need to do is get the socket, and then do a checksum.
Linus Torvalds's avatar
Linus Torvalds committed
1287
 */
1288

1289
int __udp4_lib_rcv(struct sk_buff *skb, struct hlist_head udptable[],
1290
		   int proto)
Linus Torvalds's avatar
Linus Torvalds committed
1291
{
1292
	struct sock *sk;
1293
	struct udphdr *uh = udp_hdr(skb);
Linus Torvalds's avatar
Linus Torvalds committed
1294
1295
	unsigned short ulen;
	struct rtable *rt = (struct rtable*)skb->dst;
1296
1297
	__be32 saddr = ip_hdr(skb)->saddr;
	__be32 daddr = ip_hdr(skb)->daddr;
Linus Torvalds's avatar
Linus Torvalds committed
1298
1299

	/*
1300
	 *  Validate the packet.
Linus Torvalds's avatar
Linus Torvalds committed
1301
1302
	 */
	if (!pskb_may_pull(skb, sizeof(struct udphdr)))
1303
		goto drop;		/* No space for header. */
Linus Torvalds's avatar
Linus Torvalds committed
1304
1305

	ulen = ntohs(uh->len);
1306
	if (ulen > skb->len)
Linus Torvalds's avatar
Linus Torvalds committed
1307
1308
		goto short_packet;

1309
1310
	if (proto == IPPROTO_UDP) {
		/* UDP validates ulen. */
1311
1312
		if (ulen < sizeof(*uh) || pskb_trim_rcsum(skb, ulen))
			goto short_packet;
1313
		uh = udp_hdr(skb);
1314
	}
Linus Torvalds's avatar
Linus Torvalds committed
1315

1316
1317
1318
	if (udp4_csum_init(skb, uh, proto))
		goto csum_error;

1319
	if (rt->rt_flags & (RTCF_BROADCAST|RTCF_MULTICAST))
1320
		return __udp4_lib_mcast_deliver(skb, uh, saddr, daddr, udptable);
Linus Torvalds's avatar
Linus Torvalds committed
1321

1322
	sk = __udp4_lib_lookup(saddr, uh->source, daddr, uh->dest,
1323
			       skb->dev->ifindex, udptable);
Linus Torvalds's avatar
Linus Torvalds committed
1324
1325
1326
1327
1328
1329

	if (sk != NULL) {
		int ret = udp_queue_rcv_skb(sk, skb);
		sock_put(sk);

		/* a return value > 0 means to resubmit the input, but
1330
		 * it wants the return to be -protocol, or 0
Linus Torvalds's avatar
Linus Torvalds committed
1331
1332
1333
1334
1335
1336
1337
1338
		 */
		if (ret > 0)
			return -ret;
		return 0;
	}

	if (!xfrm4_policy_check(NULL, XFRM_POLICY_IN, skb))
		goto drop;
1339
	nf_reset(skb);
Linus Torvalds's avatar
Linus Torvalds committed
1340
1341

	/* No socket. Drop packet silently, if checksum is wrong */
1342
	if (udp_lib_checksum_complete(skb))
Linus Torvalds's avatar
Linus Torvalds committed
1343
1344
		goto csum_error;

1345
	UDP_INC_STATS_BH(UDP_MIB_NOPORTS, proto == IPPROTO_UDPLITE);
Linus Torvalds's avatar
Linus Torvalds committed
1346
1347
1348
1349
1350
1351
1352
	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);
1353
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
1354
1355

short_packet:
1356
	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: short packet: From %u.%u.%u.%u:%u %d/%d to %u.%u.%u.%u:%u\n",
1357
		       proto == IPPROTO_UDPLITE ? "-Lite" : "",
1358
1359
1360
		       NIPQUAD(saddr),
		       ntohs(uh->source),
		       ulen,
1361
		       skb->len,
1362
1363
		       NIPQUAD(daddr),
		       ntohs(uh->dest));
1364
	goto drop;
Linus Torvalds's avatar
Linus Torvalds committed
1365
1366

csum_error:
1367
1368
1369
	/*
	 * RFC1122: OK.  Discards the bad packet silently (as far as
	 * the network is concerned, anyway) as per 4.1.3.4 (MUST).
Linus Torvalds's avatar
Linus Torvalds committed
1370
	 */
1371
	LIMIT_NETDEBUG(KERN_DEBUG "UDP%s: bad checksum. From %d.%d.%d.%d:%d to %d.%d.%d.%d:%d ulen %d\n",
1372
		       proto == IPPROTO_UDPLITE ? "-Lite" : "",