input.c 33.4 KB
Newer Older
1
/* SCTP kernel implementation
Linus Torvalds's avatar
Linus Torvalds committed
2 3 4 5 6 7 8
 * Copyright (c) 1999-2000 Cisco, Inc.
 * Copyright (c) 1999-2001 Motorola, Inc.
 * Copyright (c) 2001-2003 International Business Machines, Corp.
 * Copyright (c) 2001 Intel Corp.
 * Copyright (c) 2001 Nokia, Inc.
 * Copyright (c) 2001 La Monte H.P. Yarroll
 *
9
 * This file is part of the SCTP kernel implementation
Linus Torvalds's avatar
Linus Torvalds committed
10 11 12
 *
 * These functions handle all input from the IP layer into SCTP.
 *
13
 * This SCTP implementation is free software;
Linus Torvalds's avatar
Linus Torvalds committed
14 15 16 17 18
 * 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, or (at your option)
 * any later version.
 *
19
 * This SCTP implementation is distributed in the hope that it
Linus Torvalds's avatar
Linus Torvalds committed
20 21 22 23 24 25
 * will be useful, but WITHOUT ANY WARRANTY; without even the implied
 *                 ************************
 * warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
 * See the GNU General Public License for more details.
 *
 * You should have received a copy of the GNU General Public License
26 27
 * along with GNU CC; see the file COPYING.  If not, see
 * <http://www.gnu.org/licenses/>.
Linus Torvalds's avatar
Linus Torvalds committed
28 29 30
 *
 * Please send any bug reports or fixes you make to the
 * email address(es):
31
 *    lksctp developers <linux-sctp@vger.kernel.org>
Linus Torvalds's avatar
Linus Torvalds committed
32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
 *
 * Written or modified by:
 *    La Monte H.P. Yarroll <piggy@acm.org>
 *    Karl Knutson <karl@athena.chicago.il.us>
 *    Xingang Guo <xingang.guo@intel.com>
 *    Jon Grimm <jgrimm@us.ibm.com>
 *    Hui Huang <hui.huang@nokia.com>
 *    Daisy Chang <daisyc@us.ibm.com>
 *    Sridhar Samudrala <sri@us.ibm.com>
 *    Ardelle Fan <ardelle.fan@intel.com>
 */

#include <linux/types.h>
#include <linux/list.h> /* For struct list_head */
#include <linux/socket.h>
#include <linux/ip.h>
#include <linux/time.h> /* For struct timeval */
49
#include <linux/slab.h>
Linus Torvalds's avatar
Linus Torvalds committed
50 51 52 53 54 55 56
#include <net/ip.h>
#include <net/icmp.h>
#include <net/snmp.h>
#include <net/sock.h>
#include <net/xfrm.h>
#include <net/sctp/sctp.h>
#include <net/sctp/sm.h>
57
#include <net/sctp/checksum.h>
58
#include <net/net_namespace.h>
Linus Torvalds's avatar
Linus Torvalds committed
59 60 61

/* Forward declarations for internal helpers. */
static int sctp_rcv_ootb(struct sk_buff *);
62 63
static struct sctp_association *__sctp_rcv_lookup(struct net *net,
				      struct sk_buff *skb,
Linus Torvalds's avatar
Linus Torvalds committed
64
				      const union sctp_addr *paddr,
65
				      const union sctp_addr *laddr,
Linus Torvalds's avatar
Linus Torvalds committed
66
				      struct sctp_transport **transportp);
67 68
static struct sctp_endpoint *__sctp_rcv_lookup_endpoint(struct net *net,
						const union sctp_addr *laddr);
Linus Torvalds's avatar
Linus Torvalds committed
69
static struct sctp_association *__sctp_lookup_association(
70
					struct net *net,
Linus Torvalds's avatar
Linus Torvalds committed
71 72 73 74
					const union sctp_addr *local,
					const union sctp_addr *peer,
					struct sctp_transport **pt);

Zhu Yi's avatar
Zhu Yi committed
75
static int sctp_add_backlog(struct sock *sk, struct sk_buff *skb);
76

Linus Torvalds's avatar
Linus Torvalds committed
77 78

/* Calculate the SCTP checksum of an SCTP packet.  */
79
static inline int sctp_rcv_checksum(struct net *net, struct sk_buff *skb)
Linus Torvalds's avatar
Linus Torvalds committed
80
{
81
	struct sctphdr *sh = sctp_hdr(skb);
82
	__le32 cmp = sh->checksum;
83
	__le32 val = sctp_compute_cksum(skb, 0);
Linus Torvalds's avatar
Linus Torvalds committed
84 85 86

	if (val != cmp) {
		/* CRC failure, dump it. */
87
		__SCTP_INC_STATS(net, SCTP_MIB_CHECKSUMERRORS);
Linus Torvalds's avatar
Linus Torvalds committed
88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107
		return -1;
	}
	return 0;
}

/*
 * This is the routine which IP calls when receiving an SCTP packet.
 */
int sctp_rcv(struct sk_buff *skb)
{
	struct sock *sk;
	struct sctp_association *asoc;
	struct sctp_endpoint *ep = NULL;
	struct sctp_ep_common *rcvr;
	struct sctp_transport *transport = NULL;
	struct sctp_chunk *chunk;
	union sctp_addr src;
	union sctp_addr dest;
	int family;
	struct sctp_af *af;
108
	struct net *net = dev_net(skb->dev);
Linus Torvalds's avatar
Linus Torvalds committed
109

110
	if (skb->pkt_type != PACKET_HOST)
Linus Torvalds's avatar
Linus Torvalds committed
111 112
		goto discard_it;

113
	__SCTP_INC_STATS(net, SCTP_MIB_INSCTPPACKS);
Linus Torvalds's avatar
Linus Torvalds committed
114

115 116 117 118 119
	/* If packet is too small to contain a single chunk, let's not
	 * waste time on it anymore.
	 */
	if (skb->len < sizeof(struct sctphdr) + sizeof(struct sctp_chunkhdr) +
		       skb_transport_offset(skb))
120 121
		goto discard_it;

122 123 124 125 126 127 128
	/* If the packet is fragmented and we need to do crc checking,
	 * it's better to just linearize it otherwise crc computing
	 * takes longer.
	 */
	if ((!(skb_shinfo(skb)->gso_type & SKB_GSO_SCTP) &&
	     skb_linearize(skb)) ||
	    !pskb_may_pull(skb, sizeof(struct sctphdr)))
129
		goto discard_it;
Linus Torvalds's avatar
Linus Torvalds committed
130

131
	/* Pull up the IP header. */
132
	__skb_pull(skb, skb_transport_offset(skb));
133 134 135 136

	skb->csum_valid = 0; /* Previous value not applicable */
	if (skb_csum_unnecessary(skb))
		__skb_decr_checksum_unnecessary(skb);
137 138 139
	else if (!sctp_checksum_disable &&
		 !(skb_shinfo(skb)->gso_type & SKB_GSO_SCTP) &&
		 sctp_rcv_checksum(net, skb) < 0)
Linus Torvalds's avatar
Linus Torvalds committed
140
		goto discard_it;
141
	skb->csum_valid = 1;
Linus Torvalds's avatar
Linus Torvalds committed
142

143
	__skb_pull(skb, sizeof(struct sctphdr));
Linus Torvalds's avatar
Linus Torvalds committed
144

145
	family = ipver2af(ip_hdr(skb)->version);
Linus Torvalds's avatar
Linus Torvalds committed
146 147 148
	af = sctp_get_af_specific(family);
	if (unlikely(!af))
		goto discard_it;
149
	SCTP_INPUT_CB(skb)->af = af;
Linus Torvalds's avatar
Linus Torvalds committed
150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165

	/* Initialize local addresses for lookups. */
	af->from_skb(&src, skb, 1);
	af->from_skb(&dest, skb, 0);

	/* If the packet is to or from a non-unicast address,
	 * silently discard the packet.
	 *
	 * This is not clearly defined in the RFC except in section
	 * 8.4 - OOTB handling.  However, based on the book "Stream Control
	 * Transmission Protocol" 2.1, "It is important to note that the
	 * IP address of an SCTP transport address must be a routable
	 * unicast address.  In other words, IP multicast addresses and
	 * IP broadcast addresses cannot be used in an SCTP transport
	 * address."
	 */
166 167
	if (!af->addr_valid(&src, NULL, skb) ||
	    !af->addr_valid(&dest, NULL, skb))
Linus Torvalds's avatar
Linus Torvalds committed
168 169
		goto discard_it;

170
	asoc = __sctp_rcv_lookup(net, skb, &src, &dest, &transport);
171

172
	if (!asoc)
173
		ep = __sctp_rcv_lookup_endpoint(net, &dest);
174 175 176 177 178 179 180 181 182

	/* Retrieve the common input handling substructure. */
	rcvr = asoc ? &asoc->base : &ep->base;
	sk = rcvr->sk;

	/*
	 * If a frame arrives on an interface and the receiving socket is
	 * bound to another interface, via SO_BINDTODEVICE, treat it as OOTB
	 */
183
	if (sk->sk_bound_dev_if && (sk->sk_bound_dev_if != af->skb_iif(skb))) {
184 185
		if (transport) {
			sctp_transport_put(transport);
186
			asoc = NULL;
187
			transport = NULL;
188 189 190 191
		} else {
			sctp_endpoint_put(ep);
			ep = NULL;
		}
192
		sk = net->sctp.ctl_sock;
193 194 195 196 197
		ep = sctp_sk(sk)->ep;
		sctp_endpoint_hold(ep);
		rcvr = &ep->base;
	}

Linus Torvalds's avatar
Linus Torvalds committed
198 199 200 201 202 203 204 205 206 207
	/*
	 * RFC 2960, 8.4 - Handle "Out of the blue" Packets.
	 * An SCTP packet is called an "out of the blue" (OOTB)
	 * packet if it is correctly formed, i.e., passed the
	 * receiver's checksum check, but the receiver is not
	 * able to identify the association to which this
	 * packet belongs.
	 */
	if (!asoc) {
		if (sctp_rcv_ootb(skb)) {
208
			__SCTP_INC_STATS(net, SCTP_MIB_OUTOFBLUES);
Linus Torvalds's avatar
Linus Torvalds committed
209 210 211 212 213 214
			goto discard_release;
		}
	}

	if (!xfrm_policy_check(sk, XFRM_POLICY_IN, skb, family))
		goto discard_release;
215
	nf_reset(skb);
Linus Torvalds's avatar
Linus Torvalds committed
216

217
	if (sk_filter(sk, skb))
218
		goto discard_release;
Linus Torvalds's avatar
Linus Torvalds committed
219 220

	/* Create an SCTP packet structure. */
221
	chunk = sctp_chunkify(skb, asoc, sk, GFP_ATOMIC);
222
	if (!chunk)
Linus Torvalds's avatar
Linus Torvalds committed
223
		goto discard_release;
224
	SCTP_INPUT_CB(skb)->chunk = chunk;
Linus Torvalds's avatar
Linus Torvalds committed
225 226 227 228 229

	/* Remember what endpoint is to handle this packet. */
	chunk->rcvr = rcvr;

	/* Remember the SCTP header. */
230
	chunk->sctp_hdr = sctp_hdr(skb);
Linus Torvalds's avatar
Linus Torvalds committed
231 232

	/* Set the source and destination addresses of the incoming chunk.  */
233
	sctp_init_addrs(chunk, &src, &dest);
Linus Torvalds's avatar
Linus Torvalds committed
234 235 236 237 238 239 240 241

	/* Remember where we came from.  */
	chunk->transport = transport;

	/* Acquire access to the sock lock. Note: We are safe from other
	 * bottom halves on this lock, but a user may be in the lock too,
	 * so check if it is busy.
	 */
242
	bh_lock_sock(sk);
Linus Torvalds's avatar
Linus Torvalds committed
243

244 245 246 247 248 249 250 251
	if (sk != rcvr->sk) {
		/* Our cached sk is different from the rcvr->sk.  This is
		 * because migrate()/accept() may have moved the association
		 * to a new socket and released all the sockets.  So now we
		 * are holding a lock on the old socket while the user may
		 * be doing something with the new socket.  Switch our veiw
		 * of the current sk.
		 */
252
		bh_unlock_sock(sk);
253
		sk = rcvr->sk;
254
		bh_lock_sock(sk);
255 256
	}

257
	if (sock_owned_by_user(sk)) {
Zhu Yi's avatar
Zhu Yi committed
258
		if (sctp_add_backlog(sk, skb)) {
259
			bh_unlock_sock(sk);
Zhu Yi's avatar
Zhu Yi committed
260 261 262 263
			sctp_chunk_free(chunk);
			skb = NULL; /* sctp_chunk_free already freed the skb */
			goto discard_release;
		}
264
		__SCTP_INC_STATS(net, SCTP_MIB_IN_PKT_BACKLOG);
265
	} else {
266
		__SCTP_INC_STATS(net, SCTP_MIB_IN_PKT_SOFTIRQ);
267
		sctp_inq_push(&chunk->rcvr->inqueue, chunk);
268
	}
Linus Torvalds's avatar
Linus Torvalds committed
269

270
	bh_unlock_sock(sk);
271 272

	/* Release the asoc/ep ref we took in the lookup calls. */
273 274
	if (transport)
		sctp_transport_put(transport);
275 276
	else
		sctp_endpoint_put(ep);
277

278
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
279 280

discard_it:
281
	__SCTP_INC_STATS(net, SCTP_MIB_IN_PKT_DISCARDS);
Linus Torvalds's avatar
Linus Torvalds committed
282
	kfree_skb(skb);
283
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
284 285

discard_release:
286
	/* Release the asoc/ep ref we took in the lookup calls. */
287 288
	if (transport)
		sctp_transport_put(transport);
289
	else
Linus Torvalds's avatar
Linus Torvalds committed
290 291 292 293 294
		sctp_endpoint_put(ep);

	goto discard_it;
}

295 296 297 298
/* Process the backlog queue of the socket.  Every skb on
 * the backlog holds a ref on an association or endpoint.
 * We hold this ref throughout the state machine to make
 * sure that the structure we need is still around.
Linus Torvalds's avatar
Linus Torvalds committed
299 300 301
 */
int sctp_backlog_rcv(struct sock *sk, struct sk_buff *skb)
{
302
	struct sctp_chunk *chunk = SCTP_INPUT_CB(skb)->chunk;
303
	struct sctp_inq *inqueue = &chunk->rcvr->inqueue;
304
	struct sctp_transport *t = chunk->transport;
305
	struct sctp_ep_common *rcvr = NULL;
306
	int backloged = 0;
307

308
	rcvr = chunk->rcvr;
309

310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331
	/* If the rcvr is dead then the association or endpoint
	 * has been deleted and we can safely drop the chunk
	 * and refs that we are holding.
	 */
	if (rcvr->dead) {
		sctp_chunk_free(chunk);
		goto done;
	}

	if (unlikely(rcvr->sk != sk)) {
		/* In this case, the association moved from one socket to
		 * another.  We are currently sitting on the backlog of the
		 * old socket, so we need to move.
		 * However, since we are here in the process context we
		 * need to take make sure that the user doesn't own
		 * the new socket when we process the packet.
		 * If the new socket is user-owned, queue the chunk to the
		 * backlog of the new socket without dropping any refs.
		 * Otherwise, we can safely push the chunk on the inqueue.
		 */

		sk = rcvr->sk;
332
		local_bh_disable();
333
		bh_lock_sock(sk);
334 335

		if (sock_owned_by_user(sk)) {
336
			if (sk_add_backlog(sk, skb, sk->sk_rcvbuf))
Zhu Yi's avatar
Zhu Yi committed
337 338 339
				sctp_chunk_free(chunk);
			else
				backloged = 1;
340 341 342
		} else
			sctp_inq_push(inqueue, chunk);

343
		bh_unlock_sock(sk);
344
		local_bh_enable();
345 346 347 348 349 350 351 352 353 354 355

		/* If the chunk was backloged again, don't drop refs */
		if (backloged)
			return 0;
	} else {
		sctp_inq_push(inqueue, chunk);
	}

done:
	/* Release the refs we took in sctp_add_backlog */
	if (SCTP_EP_TYPE_ASSOCIATION == rcvr->type)
356
		sctp_transport_put(t);
357 358 359 360 361
	else if (SCTP_EP_TYPE_SOCKET == rcvr->type)
		sctp_endpoint_put(sctp_ep(rcvr));
	else
		BUG();

362
	return 0;
Linus Torvalds's avatar
Linus Torvalds committed
363 364
}

Zhu Yi's avatar
Zhu Yi committed
365
static int sctp_add_backlog(struct sock *sk, struct sk_buff *skb)
366
{
367
	struct sctp_chunk *chunk = SCTP_INPUT_CB(skb)->chunk;
368
	struct sctp_transport *t = chunk->transport;
369
	struct sctp_ep_common *rcvr = chunk->rcvr;
Zhu Yi's avatar
Zhu Yi committed
370
	int ret;
371

372
	ret = sk_add_backlog(sk, skb, sk->sk_rcvbuf);
Zhu Yi's avatar
Zhu Yi committed
373 374 375 376 377 378
	if (!ret) {
		/* Hold the assoc/ep while hanging on the backlog queue.
		 * This way, we know structures we need will not disappear
		 * from us
		 */
		if (SCTP_EP_TYPE_ASSOCIATION == rcvr->type)
379
			sctp_transport_hold(t);
Zhu Yi's avatar
Zhu Yi committed
380 381 382 383 384 385
		else if (SCTP_EP_TYPE_SOCKET == rcvr->type)
			sctp_endpoint_hold(sctp_ep(rcvr));
		else
			BUG();
	}
	return ret;
386

387 388
}

Linus Torvalds's avatar
Linus Torvalds committed
389 390 391 392
/* Handle icmp frag needed error. */
void sctp_icmp_frag_needed(struct sock *sk, struct sctp_association *asoc,
			   struct sctp_transport *t, __u32 pmtu)
{
393
	if (!t || (t->pathmtu <= pmtu))
394 395
		return;

396 397 398 399 400 401
	if (sock_owned_by_user(sk)) {
		asoc->pmtu_pending = 1;
		t->pmtu_pending = 1;
		return;
	}

402
	if (t->param_flags & SPP_PMTUD_ENABLE) {
403
		/* Update transports view of the MTU */
404
		sctp_transport_update_pmtu(sk, t, pmtu);
Linus Torvalds's avatar
Linus Torvalds committed
405

406
		/* Update association pmtu. */
407
		sctp_assoc_sync_pmtu(sk, asoc);
Linus Torvalds's avatar
Linus Torvalds committed
408
	}
409 410 411 412 413

	/* Retransmit with the new pmtu setting.
	 * Normally, if PMTU discovery is disabled, an ICMP Fragmentation
	 * Needed will never be sent, but if a message was sent before
	 * PMTU discovery was disabled that was larger than the PMTU, it
414
	 * would not be fragmented, so it must be re-transmitted fragmented.
415 416
	 */
	sctp_retransmit(&asoc->outqueue, t, SCTP_RTXR_PMTUD);
Linus Torvalds's avatar
Linus Torvalds committed
417 418
}

419 420
void sctp_icmp_redirect(struct sock *sk, struct sctp_transport *t,
			struct sk_buff *skb)
421 422 423 424 425 426
{
	struct dst_entry *dst;

	if (!t)
		return;
	dst = sctp_transport_dst_check(t);
427
	if (dst)
428
		dst->ops->redirect(dst, sk, skb);
429 430
}

Linus Torvalds's avatar
Linus Torvalds committed
431 432 433 434 435 436 437 438 439 440 441 442
/*
 * SCTP Implementer's Guide, 2.37 ICMP handling procedures
 *
 * ICMP8) If the ICMP code is a "Unrecognized next header type encountered"
 *        or a "Protocol Unreachable" treat this message as an abort
 *        with the T bit set.
 *
 * This function sends an event to the state machine, which will abort the
 * association.
 *
 */
void sctp_icmp_proto_unreachable(struct sock *sk,
443 444
			   struct sctp_association *asoc,
			   struct sctp_transport *t)
Linus Torvalds's avatar
Linus Torvalds committed
445
{
446 447 448 449 450 451 452 453 454
	if (sock_owned_by_user(sk)) {
		if (timer_pending(&t->proto_unreach_timer))
			return;
		else {
			if (!mod_timer(&t->proto_unreach_timer,
						jiffies + (HZ/20)))
				sctp_association_hold(asoc);
		}
	} else {
455 456
		struct net *net = sock_net(sk);

457 458 459
		pr_debug("%s: unrecognized next header type "
			 "encountered!\n", __func__);

460
		if (del_timer(&t->proto_unreach_timer))
461
			sctp_association_put(asoc);
Linus Torvalds's avatar
Linus Torvalds committed
462

463
		sctp_do_sm(net, SCTP_EVENT_T_OTHER,
464 465 466 467
			   SCTP_ST_OTHER(SCTP_EVENT_ICMP_PROTO_UNREACH),
			   asoc->state, asoc->ep, asoc, t,
			   GFP_ATOMIC);
	}
Linus Torvalds's avatar
Linus Torvalds committed
468 469 470
}

/* Common lookup code for icmp/icmpv6 error handler. */
471
struct sock *sctp_err_lookup(struct net *net, int family, struct sk_buff *skb,
Linus Torvalds's avatar
Linus Torvalds committed
472 473 474 475 476 477 478 479
			     struct sctphdr *sctphdr,
			     struct sctp_association **app,
			     struct sctp_transport **tpp)
{
	union sctp_addr saddr;
	union sctp_addr daddr;
	struct sctp_af *af;
	struct sock *sk = NULL;
480
	struct sctp_association *asoc;
Linus Torvalds's avatar
Linus Torvalds committed
481
	struct sctp_transport *transport = NULL;
482 483 484
	struct sctp_init_chunk *chunkhdr;
	__u32 vtag = ntohl(sctphdr->vtag);
	int len = skb->len - ((void *)sctphdr - (void *)skb->data);
Linus Torvalds's avatar
Linus Torvalds committed
485

486
	*app = NULL; *tpp = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
487 488 489 490 491 492 493 494 495 496 497 498 499

	af = sctp_get_af_specific(family);
	if (unlikely(!af)) {
		return NULL;
	}

	/* Initialize local addresses for lookups. */
	af->from_skb(&saddr, skb, 1);
	af->from_skb(&daddr, skb, 0);

	/* Look for an association that matches the incoming ICMP error
	 * packet.
	 */
500
	asoc = __sctp_lookup_association(net, &saddr, &daddr, &transport);
501 502
	if (!asoc)
		return NULL;
Linus Torvalds's avatar
Linus Torvalds committed
503

504
	sk = asoc->base.sk;
Linus Torvalds's avatar
Linus Torvalds committed
505

506 507 508 509 510 511 512 513 514 515 516 517 518
	/* RFC 4960, Appendix C. ICMP Handling
	 *
	 * ICMP6) An implementation MUST validate that the Verification Tag
	 * contained in the ICMP message matches the Verification Tag of
	 * the peer.  If the Verification Tag is not 0 and does NOT
	 * match, discard the ICMP message.  If it is 0 and the ICMP
	 * message contains enough bytes to verify that the chunk type is
	 * an INIT chunk and that the Initiate Tag matches the tag of the
	 * peer, continue with ICMP7.  If the ICMP message is too short
	 * or the chunk type or the Initiate Tag does not match, silently
	 * discard the packet.
	 */
	if (vtag == 0) {
Joe Perches's avatar
Joe Perches committed
519
		chunkhdr = (void *)sctphdr + sizeof(struct sctphdr);
520 521 522 523 524 525 526
		if (len < sizeof(struct sctphdr) + sizeof(sctp_chunkhdr_t)
			  + sizeof(__be32) ||
		    chunkhdr->chunk_hdr.type != SCTP_CID_INIT ||
		    ntohl(chunkhdr->init_hdr.init_tag) != asoc->c.my_vtag) {
			goto out;
		}
	} else if (vtag != asoc->c.peer_vtag) {
527 528
		goto out;
	}
Linus Torvalds's avatar
Linus Torvalds committed
529

530
	bh_lock_sock(sk);
Linus Torvalds's avatar
Linus Torvalds committed
531 532 533 534 535

	/* If too many ICMPs get dropped on busy
	 * servers this needs to be solved differently.
	 */
	if (sock_owned_by_user(sk))
536
		__NET_INC_STATS(net, LINUX_MIB_LOCKDROPPEDICMPS);
Linus Torvalds's avatar
Linus Torvalds committed
537 538 539 540 541 542

	*app = asoc;
	*tpp = transport;
	return sk;

out:
543
	sctp_transport_put(transport);
Linus Torvalds's avatar
Linus Torvalds committed
544 545 546 547
	return NULL;
}

/* Common cleanup code for icmp/icmpv6 error handler. */
548
void sctp_err_finish(struct sock *sk, struct sctp_transport *t)
Linus Torvalds's avatar
Linus Torvalds committed
549
{
550
	bh_unlock_sock(sk);
551
	sctp_transport_put(t);
Linus Torvalds's avatar
Linus Torvalds committed
552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570
}

/*
 * 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.  After adjustment
 * header points to the first 8 bytes of the sctp header.  We need
 * to find the appropriate port.
 *
 * The locking strategy used here is very "optimistic". When
 * someone else accesses the socket the ICMP is just dropped
 * and for some paths there is no check at all.
 * A more general error queue to queue errors for later handling
 * is probably better.
 *
 */
void sctp_v4_err(struct sk_buff *skb, __u32 info)
{
571
	const struct iphdr *iph = (const struct iphdr *)skb->data;
572
	const int ihlen = iph->ihl * 4;
573 574
	const int type = icmp_hdr(skb)->type;
	const int code = icmp_hdr(skb)->code;
Linus Torvalds's avatar
Linus Torvalds committed
575
	struct sock *sk;
576
	struct sctp_association *asoc = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
577 578
	struct sctp_transport *transport;
	struct inet_sock *inet;
579
	__u16 saveip, savesctp;
Linus Torvalds's avatar
Linus Torvalds committed
580
	int err;
581
	struct net *net = dev_net(skb->dev);
Linus Torvalds's avatar
Linus Torvalds committed
582 583

	/* Fix up skb to look at the embedded net header. */
584 585
	saveip = skb->network_header;
	savesctp = skb->transport_header;
586
	skb_reset_network_header(skb);
587
	skb_set_transport_header(skb, ihlen);
588
	sk = sctp_err_lookup(net, AF_INET, skb, sctp_hdr(skb), &asoc, &transport);
589
	/* Put back, the original values. */
590 591
	skb->network_header = saveip;
	skb->transport_header = savesctp;
Linus Torvalds's avatar
Linus Torvalds committed
592
	if (!sk) {
Eric Dumazet's avatar
Eric Dumazet committed
593
		__ICMP_INC_STATS(net, ICMP_MIB_INERRORS);
Linus Torvalds's avatar
Linus Torvalds committed
594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609
		return;
	}
	/* Warning:  The sock lock is held.  Remember to call
	 * sctp_err_finish!
	 */

	switch (type) {
	case ICMP_PARAMETERPROB:
		err = EPROTO;
		break;
	case ICMP_DEST_UNREACH:
		if (code > NR_ICMP_UNREACH)
			goto out_unlock;

		/* PMTU discovery (RFC1191) */
		if (ICMP_FRAG_NEEDED == code) {
610
			sctp_icmp_frag_needed(sk, asoc, transport,
611
					      SCTP_TRUNC4(info));
Linus Torvalds's avatar
Linus Torvalds committed
612
			goto out_unlock;
613
		} else {
Linus Torvalds's avatar
Linus Torvalds committed
614
			if (ICMP_PROT_UNREACH == code) {
615
				sctp_icmp_proto_unreachable(sk, asoc,
Linus Torvalds's avatar
Linus Torvalds committed
616 617 618 619 620 621 622 623 624 625 626 627 628 629 630
							    transport);
				goto out_unlock;
			}
		}
		err = icmp_err_convert[code].errno;
		break;
	case ICMP_TIME_EXCEEDED:
		/* Ignore any time exceeded errors due to fragment reassembly
		 * timeouts.
		 */
		if (ICMP_EXC_FRAGTIME == code)
			goto out_unlock;

		err = EHOSTUNREACH;
		break;
631 632
	case ICMP_REDIRECT:
		sctp_icmp_redirect(sk, transport, skb);
633
		/* Fall through to out_unlock. */
Linus Torvalds's avatar
Linus Torvalds committed
634 635 636 637 638 639 640 641 642 643 644 645 646
	default:
		goto out_unlock;
	}

	inet = inet_sk(sk);
	if (!sock_owned_by_user(sk) && inet->recverr) {
		sk->sk_err = err;
		sk->sk_error_report(sk);
	} else {  /* Only an error on timeout */
		sk->sk_err_soft = err;
	}

out_unlock:
647
	sctp_err_finish(sk, transport);
Linus Torvalds's avatar
Linus Torvalds committed
648 649 650 651 652 653 654 655 656 657 658 659 660 661
}

/*
 * RFC 2960, 8.4 - Handle "Out of the blue" Packets.
 *
 * This function scans all the chunks in the OOTB packet to determine if
 * the packet should be discarded right away.  If a response might be needed
 * for this packet, or, if further processing is possible, the packet will
 * be queued to a proper inqueue for the next phase of handling.
 *
 * Output:
 * Return 0 - If further processing is needed.
 * Return 1 - If the packet can be discarded right away.
 */
662
static int sctp_rcv_ootb(struct sk_buff *skb)
Linus Torvalds's avatar
Linus Torvalds committed
663
{
664 665
	sctp_chunkhdr_t *ch, _ch;
	int ch_end, offset = 0;
Linus Torvalds's avatar
Linus Torvalds committed
666 667

	/* Scan through all the chunks in the packet.  */
668
	do {
669 670 671 672 673 674
		/* Make sure we have at least the header there */
		if (offset + sizeof(sctp_chunkhdr_t) > skb->len)
			break;

		ch = skb_header_pointer(skb, offset, sizeof(*ch), &_ch);

675 676 677 678
		/* Break out if chunk length is less then minimal. */
		if (ntohs(ch->length) < sizeof(sctp_chunkhdr_t))
			break;

679
		ch_end = offset + SCTP_PAD4(ntohs(ch->length));
680
		if (ch_end > skb->len)
681
			break;
Linus Torvalds's avatar
Linus Torvalds committed
682 683 684 685 686 687 688 689 690 691 692 693 694 695 696

		/* RFC 8.4, 2) If the OOTB packet contains an ABORT chunk, the
		 * receiver MUST silently discard the OOTB packet and take no
		 * further action.
		 */
		if (SCTP_CID_ABORT == ch->type)
			goto discard;

		/* RFC 8.4, 6) If the packet contains a SHUTDOWN COMPLETE
		 * chunk, the receiver should silently discard the packet
		 * and take no further action.
		 */
		if (SCTP_CID_SHUTDOWN_COMPLETE == ch->type)
			goto discard;

697 698 699 700 701 702 703 704
		/* RFC 4460, 2.11.2
		 * This will discard packets with INIT chunk bundled as
		 * subsequent chunks in the packet.  When INIT is first,
		 * the normal INIT processing will discard the chunk.
		 */
		if (SCTP_CID_INIT == ch->type && (void *)ch != skb->data)
			goto discard;

705 706
		offset = ch_end;
	} while (ch_end < skb->len);
Linus Torvalds's avatar
Linus Torvalds committed
707 708 709 710 711 712 713 714 715 716

	return 0;

discard:
	return 1;
}

/* Insert endpoint into the hash table.  */
static void __sctp_hash_endpoint(struct sctp_endpoint *ep)
{
717
	struct net *net = sock_net(ep->base.sk);
Linus Torvalds's avatar
Linus Torvalds committed
718 719 720 721 722
	struct sctp_ep_common *epb;
	struct sctp_hashbucket *head;

	epb = &ep->base;

723
	epb->hashent = sctp_ep_hashfn(net, epb->bind_addr.port);
Linus Torvalds's avatar
Linus Torvalds committed
724 725
	head = &sctp_ep_hashtable[epb->hashent];

726
	write_lock(&head->lock);
727
	hlist_add_head(&epb->node, &head->chain);
728
	write_unlock(&head->lock);
Linus Torvalds's avatar
Linus Torvalds committed
729 730 731 732 733
}

/* Add an endpoint to the hash. Local BH-safe. */
void sctp_hash_endpoint(struct sctp_endpoint *ep)
{
734
	local_bh_disable();
Linus Torvalds's avatar
Linus Torvalds committed
735
	__sctp_hash_endpoint(ep);
736
	local_bh_enable();
Linus Torvalds's avatar
Linus Torvalds committed
737 738 739 740 741
}

/* Remove endpoint from the hash table.  */
static void __sctp_unhash_endpoint(struct sctp_endpoint *ep)
{
742
	struct net *net = sock_net(ep->base.sk);
Linus Torvalds's avatar
Linus Torvalds committed
743 744 745 746 747
	struct sctp_hashbucket *head;
	struct sctp_ep_common *epb;

	epb = &ep->base;

748
	epb->hashent = sctp_ep_hashfn(net, epb->bind_addr.port);
Linus Torvalds's avatar
Linus Torvalds committed
749 750 751

	head = &sctp_ep_hashtable[epb->hashent];

752
	write_lock(&head->lock);
753
	hlist_del_init(&epb->node);
754
	write_unlock(&head->lock);
Linus Torvalds's avatar
Linus Torvalds committed
755 756 757 758 759
}

/* Remove endpoint from the hash.  Local BH-safe. */
void sctp_unhash_endpoint(struct sctp_endpoint *ep)
{
760
	local_bh_disable();
Linus Torvalds's avatar
Linus Torvalds committed
761
	__sctp_unhash_endpoint(ep);
762
	local_bh_enable();
Linus Torvalds's avatar
Linus Torvalds committed
763 764 765
}

/* Look up an endpoint. */
766 767
static struct sctp_endpoint *__sctp_rcv_lookup_endpoint(struct net *net,
						const union sctp_addr *laddr)
Linus Torvalds's avatar
Linus Torvalds committed
768 769 770 771 772 773
{
	struct sctp_hashbucket *head;
	struct sctp_ep_common *epb;
	struct sctp_endpoint *ep;
	int hash;

774
	hash = sctp_ep_hashfn(net, ntohs(laddr->v4.sin_port));
Linus Torvalds's avatar
Linus Torvalds committed
775 776
	head = &sctp_ep_hashtable[hash];
	read_lock(&head->lock);
777
	sctp_for_each_hentry(epb, &head->chain) {
Linus Torvalds's avatar
Linus Torvalds committed
778
		ep = sctp_ep(epb);
779
		if (sctp_endpoint_is_match(ep, net, laddr))
Linus Torvalds's avatar
Linus Torvalds committed
780 781 782
			goto hit;
	}

783
	ep = sctp_sk(net->sctp.ctl_sock)->ep;
Linus Torvalds's avatar
Linus Torvalds committed
784 785 786 787 788 789 790

hit:
	sctp_endpoint_hold(ep);
	read_unlock(&head->lock);
	return ep;
}

791 792
/* rhashtable for transport */
struct sctp_hash_cmp_arg {
793 794 795
	const union sctp_addr	*paddr;
	const struct net	*net;
	u16			lport;
796 797 798 799 800
};

static inline int sctp_hash_cmp(struct rhashtable_compare_arg *arg,
				const void *ptr)
{
801
	struct sctp_transport *t = (struct sctp_transport *)ptr;
802
	const struct sctp_hash_cmp_arg *x = arg->key;
803
	int err = 1;
804 805

	if (!sctp_cmp_addr_exact(&t->ipaddr, x->paddr))
806 807 808 809
		return err;
	if (!sctp_transport_hold(t))
		return err;

810 811 812
	if (!net_eq(sock_net(t->asoc->base.sk), x->net))
		goto out;
	if (x->lport != htons(t->asoc->base.bind_addr.port))
813
		goto out;
814

815 816 817 818
	err = 0;
out:
	sctp_transport_put(t);
	return err;
819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842
}

static inline u32 sctp_hash_obj(const void *data, u32 len, u32 seed)
{
	const struct sctp_transport *t = data;
	const union sctp_addr *paddr = &t->ipaddr;
	const struct net *net = sock_net(t->asoc->base.sk);
	u16 lport = htons(t->asoc->base.bind_addr.port);
	u32 addr;

	if (paddr->sa.sa_family == AF_INET6)
		addr = jhash(&paddr->v6.sin6_addr, 16, seed);
	else
		addr = paddr->v4.sin_addr.s_addr;

	return  jhash_3words(addr, ((__u32)paddr->v4.sin_port) << 16 |
			     (__force __u32)lport, net_hash_mix(net), seed);
}

static inline u32 sctp_hash_key(const void *data, u32 len, u32 seed)
{
	const struct sctp_hash_cmp_arg *x = data;
	const union sctp_addr *paddr = x->paddr;
	const struct net *net = x->net;
843
	u16 lport = x->lport;
844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864
	u32 addr;

	if (paddr->sa.sa_family == AF_INET6)
		addr = jhash(&paddr->v6.sin6_addr, 16, seed);
	else
		addr = paddr->v4.sin_addr.s_addr;

	return  jhash_3words(addr, ((__u32)paddr->v4.sin_port) << 16 |
			     (__force __u32)lport, net_hash_mix(net), seed);
}

static const struct rhashtable_params sctp_hash_params = {
	.head_offset		= offsetof(struct sctp_transport, node),
	.hashfn			= sctp_hash_key,
	.obj_hashfn		= sctp_hash_obj,
	.obj_cmpfn		= sctp_hash_cmp,
	.automatic_shrinking	= true,
};

int sctp_transport_hashtable_init(void)
{
865
	return rhltable_init(&sctp_transport_hashtable, &sctp_hash_params);
866 867 868 869
}

void sctp_transport_hashtable_destroy(void)
{
870
	rhltable_destroy(&sctp_transport_hashtable);
871 872
}

873
int sctp_hash_transport(struct sctp_transport *t)
874 875
{
	struct sctp_hash_cmp_arg arg;
876
	int err;
877

878
	if (t->asoc->temp)
879
		return 0;
880

881
	arg.net   = sock_net(t->asoc->base.sk);
882 883
	arg.paddr = &t->ipaddr;
	arg.lport = htons(t->asoc->base.bind_addr.port);
884

885 886 887 888 889 890
	err = rhltable_insert_key(&sctp_transport_hashtable, &arg,
				  &t->node, sctp_hash_params);
	if (err)
		pr_err_once("insert transport fail, errno %d\n", err);

	return err;
891 892 893 894
}

void sctp_unhash_transport(struct sctp_transport *t)
{
895 896 897
	if (t->asoc->temp)
		return;

898 899
	rhltable_remove(&sctp_transport_hashtable, &t->node,
			sctp_hash_params);
900 901
}

902
/* return a transport with holding it */
903 904 905 906 907
struct sctp_transport *sctp_addrs_lookup_transport(
				struct net *net,
				const union sctp_addr *laddr,
				const union sctp_addr *paddr)
{
908 909
	struct rhlist_head *tmp, *list;
	struct sctp_transport *t;
910 911 912
	struct sctp_hash_cmp_arg arg = {
		.paddr = paddr,
		.net   = net,
913
		.lport = laddr->v4.sin_port,
914 915
	};

916 917 918 919 920 921 922 923 924 925 926 927 928 929
	list = rhltable_lookup(&sctp_transport_hashtable, &arg,
			       sctp_hash_params);

	rhl_for_each_entry_rcu(t, tmp, list, node) {
		if (!sctp_transport_hold(t))
			continue;

		if (sctp_bind_addr_match(&t->asoc->base.bind_addr,
					 laddr, sctp_sk(t->asoc->base.sk)))
			return t;
		sctp_transport_put(t);
	}

	return NULL;
930 931
}

932
/* return a transport without holding it, as it's only used under sock lock */
933 934 935 936 937
struct sctp_transport *sctp_epaddr_lookup_transport(
				const struct sctp_endpoint *ep,
				const union sctp_addr *paddr)
{
	struct net *net = sock_net(ep->base.sk);
938 939
	struct rhlist_head *tmp, *list;
	struct sctp_transport *t;
940 941 942
	struct sctp_hash_cmp_arg arg = {
		.paddr = paddr,
		.net   = net,
943
		.lport = htons(ep->base.bind_addr.port),
944
	};
945

946 947 948 949 950 951 952 953
	list = rhltable_lookup(&sctp_transport_hashtable, &arg,
			       sctp_hash_params);

	rhl_for_each_entry_rcu(t, tmp, list, node)
		if (ep == t->asoc->ep)
			return t;

	return NULL;
954 955
}

Linus Torvalds's avatar
Linus Torvalds committed
956 957
/* Look up an association. */
static struct sctp_association *__sctp_lookup_association(
958
					struct net *net,
Linus Torvalds's avatar
Linus Torvalds committed
959 960 961 962
					const union sctp_addr *local,
					const union sctp_addr *peer,
					struct sctp_transport **pt)
{
963
	struct sctp_transport *t;
964
	struct sctp_association *asoc = NULL;
Linus Torvalds's avatar
Linus Torvalds committed
965

966
	t = sctp_addrs_lookup_transport(net, local, peer);
967
	if (!t)
968
		goto out;
Linus Torvalds's avatar
Linus Torvalds committed
969

970
	asoc = t->asoc;
971
	*pt = t;
Linus Torvalds's avatar
Linus Torvalds committed
972

973 974
out:
	return asoc;
Linus Torvalds's avatar
Linus Torvalds committed
975 976
}

977
/* Look up an association. protected by RCU read lock */
978
static
979 980
struct sctp_association *sctp_lookup_association(struct net *net,
						 const union sctp_addr *laddr,
Linus Torvalds's avatar
Linus Torvalds committed
981
						 const union sctp_addr *paddr,
982
						 struct sctp_transport **transportp)
Linus Torvalds's avatar
Linus Torvalds committed
983 984 985
{
	struct sctp_association *asoc;

986
	rcu_read_lock();
987
	asoc = __sctp_lookup_association(net, laddr, paddr, transportp);
988
	rcu_read_unlock();
Linus Torvalds's avatar
Linus Torvalds committed
989 990 991 992 993

	return asoc;
}

/* Is there an association matching the given local and peer addresses? */
994 995
int sctp_has_association(struct net *net,
			 const union sctp_addr *laddr,
Linus Torvalds's avatar
Linus Torvalds committed
996 997 998 999 1000
			 const union sctp_addr *paddr)
{
	struct sctp_association *asoc;
	struct sctp_transport *transport;

1001
	if ((asoc = sctp_lookup_association(net, laddr, paddr, &transport))) {
1002
		sctp_transport_put(transport);
Linus Torvalds's avatar
Linus Torvalds committed
1003 1004 1005