transport.c 16.1 KB
Newer Older
L
Linus Torvalds 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59
/* SCTP kernel reference Implementation
 * 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 La Monte H.P. Yarroll
 *
 * This file is part of the SCTP kernel reference Implementation
 *
 * This module provides the abstraction for an SCTP tranport representing
 * a remote transport address.  For local transport addresses, we just use
 * union sctp_addr.
 *
 * The SCTP reference implementation 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, or (at your option)
 * any later version.
 *
 * The SCTP reference implementation is distributed in the hope that it
 * 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
 * along with GNU CC; see the file COPYING.  If not, write to
 * the Free Software Foundation, 59 Temple Place - Suite 330,
 * Boston, MA 02111-1307, USA.
 *
 * Please send any bug reports or fixes you make to the
 * email address(es):
 *    lksctp developers <lksctp-developers@lists.sourceforge.net>
 *
 * Or submit a bug report through the following website:
 *    http://www.sf.net/projects/lksctp
 *
 * Written or modified by:
 *    La Monte H.P. Yarroll <piggy@acm.org>
 *    Karl Knutson          <karl@athena.chicago.il.us>
 *    Jon Grimm             <jgrimm@us.ibm.com>
 *    Xingang Guo           <xingang.guo@intel.com>
 *    Hui Huang             <hui.huang@nokia.com>
 *    Sridhar Samudrala	    <sri@us.ibm.com>
 *    Ardelle Fan	    <ardelle.fan@intel.com>
 *
 * Any bugs reported given to us we will try to fix... any fixes shared will
 * be incorporated into the next SCTP release.
 */

#include <linux/types.h>
#include <net/sctp/sctp.h>
#include <net/sctp/sm.h>

/* 1st Level Abstractions.  */

/* Initialize a new transport from provided memory.  */
static struct sctp_transport *sctp_transport_init(struct sctp_transport *peer,
						  const union sctp_addr *addr,
A
Al Viro 已提交
60
						  gfp_t gfp)
L
Linus Torvalds 已提交
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
{
	/* Copy in the address.  */
	peer->ipaddr = *addr;
	peer->af_specific = sctp_get_af_specific(addr->sa.sa_family);
	peer->asoc = NULL;

	peer->dst = NULL;
	memset(&peer->saddr, 0, sizeof(union sctp_addr));

	/* From 6.3.1 RTO Calculation:
	 *
	 * C1) Until an RTT measurement has been made for a packet sent to the
	 * given destination transport address, set RTO to the protocol
	 * parameter 'RTO.Initial'.
	 */
	peer->rtt = 0;
	peer->rto = sctp_rto_initial;
	peer->rttvar = 0;
	peer->srtt = 0;
	peer->rto_pending = 0;

	peer->last_time_heard = jiffies;
	peer->last_time_used = jiffies;
	peer->last_time_ecne_reduced = jiffies;

F
Frank Filz 已提交
86 87 88
	peer->init_sent_count = 0;

	peer->state = SCTP_ACTIVE;
89 90 91 92
	peer->param_flags = SPP_HB_DISABLE |
			    SPP_PMTUD_ENABLE |
			    SPP_SACKDELAY_ENABLE;
	peer->hbinterval  = 0;
L
Linus Torvalds 已提交
93 94

	/* Initialize the default path max_retrans.  */
95
	peer->pathmaxrxt  = sctp_max_retrans_path;
L
Linus Torvalds 已提交
96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126
	peer->error_count = 0;

	INIT_LIST_HEAD(&peer->transmitted);
	INIT_LIST_HEAD(&peer->send_ready);
	INIT_LIST_HEAD(&peer->transports);

	/* Set up the retransmission timer.  */
	init_timer(&peer->T3_rtx_timer);
	peer->T3_rtx_timer.function = sctp_generate_t3_rtx_event;
	peer->T3_rtx_timer.data = (unsigned long)peer;

	/* Set up the heartbeat timer. */
	init_timer(&peer->hb_timer);
	peer->hb_timer.function = sctp_generate_heartbeat_event;
	peer->hb_timer.data = (unsigned long)peer;

	atomic_set(&peer->refcnt, 1);
	peer->dead = 0;

	peer->malloced = 0;

	/* Initialize the state information for SFR-CACC */
	peer->cacc.changeover_active = 0;
	peer->cacc.cycling_changeover = 0;
	peer->cacc.next_tsn_at_change = 0;
	peer->cacc.cacc_saw_newack = 0;

	return peer;
}

/* Allocate and initialize a new transport.  */
A
Alexey Dobriyan 已提交
127
struct sctp_transport *sctp_transport_new(const union sctp_addr *addr,
A
Al Viro 已提交
128
					  gfp_t gfp)
L
Linus Torvalds 已提交
129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234
{
        struct sctp_transport *transport;

        transport = t_new(struct sctp_transport, gfp);
	if (!transport)
		goto fail;

	if (!sctp_transport_init(transport, addr, gfp))
		goto fail_init;

	transport->malloced = 1;
	SCTP_DBG_OBJCNT_INC(transport);

	return transport;

fail_init:
	kfree(transport);

fail:
	return NULL;
}

/* This transport is no longer needed.  Free up if possible, or
 * delay until it last reference count.
 */
void sctp_transport_free(struct sctp_transport *transport)
{
	transport->dead = 1;

	/* Try to delete the heartbeat timer.  */
	if (del_timer(&transport->hb_timer))
		sctp_transport_put(transport);

	/* Delete the T3_rtx timer if it's active.
	 * There is no point in not doing this now and letting
	 * structure hang around in memory since we know
	 * the tranport is going away.
	 */
	if (timer_pending(&transport->T3_rtx_timer) &&
	    del_timer(&transport->T3_rtx_timer))
		sctp_transport_put(transport);


	sctp_transport_put(transport);
}

/* Destroy the transport data structure.
 * Assumes there are no more users of this structure.
 */
static void sctp_transport_destroy(struct sctp_transport *transport)
{
	SCTP_ASSERT(transport->dead, "Transport is not dead", return);

	if (transport->asoc)
		sctp_association_put(transport->asoc);

        sctp_packet_free(&transport->packet);

	dst_release(transport->dst);
	kfree(transport);
	SCTP_DBG_OBJCNT_DEC(transport);
}

/* Start T3_rtx timer if it is not already running and update the heartbeat
 * timer.  This routine is called every time a DATA chunk is sent.
 */
void sctp_transport_reset_timers(struct sctp_transport *transport)
{
	/* RFC 2960 6.3.2 Retransmission Timer Rules
	 *
	 * R1) Every time a DATA chunk is sent to any address(including a
	 * retransmission), if the T3-rtx timer of that address is not running
	 * start it running so that it will expire after the RTO of that
	 * address.
	 */

	if (!timer_pending(&transport->T3_rtx_timer))
		if (!mod_timer(&transport->T3_rtx_timer,
			       jiffies + transport->rto))
			sctp_transport_hold(transport);

	/* When a data chunk is sent, reset the heartbeat interval.  */
	if (!mod_timer(&transport->hb_timer,
		       sctp_transport_timeout(transport)))
	    sctp_transport_hold(transport);
}

/* This transport has been assigned to an association.
 * Initialize fields from the association or from the sock itself.
 * Register the reference count in the association.
 */
void sctp_transport_set_owner(struct sctp_transport *transport,
			      struct sctp_association *asoc)
{
	transport->asoc = asoc;
	sctp_association_hold(asoc);
}

/* Initialize the pmtu of a transport. */
void sctp_transport_pmtu(struct sctp_transport *transport)
{
	struct dst_entry *dst;

	dst = transport->af_specific->get_dst(NULL, &transport->ipaddr, NULL);

	if (dst) {
235
		transport->pathmtu = dst_mtu(dst);
L
Linus Torvalds 已提交
236 237
		dst_release(dst);
	} else
238
		transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT;
L
Linus Torvalds 已提交
239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259
}

/* Caches the dst entry and source address for a transport's destination
 * address.
 */
void sctp_transport_route(struct sctp_transport *transport,
			  union sctp_addr *saddr, struct sctp_sock *opt)
{
	struct sctp_association *asoc = transport->asoc;
	struct sctp_af *af = transport->af_specific;
	union sctp_addr *daddr = &transport->ipaddr;
	struct dst_entry *dst;

	dst = af->get_dst(asoc, daddr, saddr);

	if (saddr)
		memcpy(&transport->saddr, saddr, sizeof(union sctp_addr));
	else
		af->get_saddr(asoc, dst, daddr, &transport->saddr);

	transport->dst = dst;
260 261 262
	if ((transport->param_flags & SPP_PMTUD_DISABLE) && transport->pathmtu) {
		return;
	}
L
Linus Torvalds 已提交
263
	if (dst) {
264
		transport->pathmtu = dst_mtu(dst);
L
Linus Torvalds 已提交
265 266 267 268 269

		/* Initialize sk->sk_rcv_saddr, if the transport is the
		 * association's active path for getsockname().
		 */ 
		if (asoc && (transport == asoc->peer.active_path))
270 271
			opt->pf->af->to_sk_saddr(&transport->saddr,
						 asoc->base.sk);
L
Linus Torvalds 已提交
272
	} else
273
		transport->pathmtu = SCTP_DEFAULT_MAXSEGMENT;
L
Linus Torvalds 已提交
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352
}

/* Hold a reference to a transport.  */
void sctp_transport_hold(struct sctp_transport *transport)
{
	atomic_inc(&transport->refcnt);
}

/* Release a reference to a transport and clean up
 * if there are no more references.
 */
void sctp_transport_put(struct sctp_transport *transport)
{
	if (atomic_dec_and_test(&transport->refcnt))
		sctp_transport_destroy(transport);
}

/* Update transport's RTO based on the newly calculated RTT. */
void sctp_transport_update_rto(struct sctp_transport *tp, __u32 rtt)
{
	/* Check for valid transport.  */
	SCTP_ASSERT(tp, "NULL transport", return);

	/* We should not be doing any RTO updates unless rto_pending is set.  */
	SCTP_ASSERT(tp->rto_pending, "rto_pending not set", return);

	if (tp->rttvar || tp->srtt) {
		/* 6.3.1 C3) When a new RTT measurement R' is made, set
		 * RTTVAR <- (1 - RTO.Beta) * RTTVAR + RTO.Beta * |SRTT - R'|
		 * SRTT <- (1 - RTO.Alpha) * SRTT + RTO.Alpha * R'
		 */

		/* Note:  The above algorithm has been rewritten to
		 * express rto_beta and rto_alpha as inverse powers
		 * of two.
		 * For example, assuming the default value of RTO.Alpha of
		 * 1/8, rto_alpha would be expressed as 3.
		 */
		tp->rttvar = tp->rttvar - (tp->rttvar >> sctp_rto_beta)
			+ ((abs(tp->srtt - rtt)) >> sctp_rto_beta);
		tp->srtt = tp->srtt - (tp->srtt >> sctp_rto_alpha)
			+ (rtt >> sctp_rto_alpha);
	} else {
		/* 6.3.1 C2) When the first RTT measurement R is made, set
		 * SRTT <- R, RTTVAR <- R/2.
		 */
		tp->srtt = rtt;
		tp->rttvar = rtt >> 1;
	}

	/* 6.3.1 G1) Whenever RTTVAR is computed, if RTTVAR = 0, then
	 * adjust RTTVAR <- G, where G is the CLOCK GRANULARITY.
	 */
	if (tp->rttvar == 0)
		tp->rttvar = SCTP_CLOCK_GRANULARITY;

	/* 6.3.1 C3) After the computation, update RTO <- SRTT + 4 * RTTVAR. */
	tp->rto = tp->srtt + (tp->rttvar << 2);

	/* 6.3.1 C6) Whenever RTO is computed, if it is less than RTO.Min
	 * seconds then it is rounded up to RTO.Min seconds.
	 */
	if (tp->rto < tp->asoc->rto_min)
		tp->rto = tp->asoc->rto_min;

	/* 6.3.1 C7) A maximum value may be placed on RTO provided it is
	 * at least RTO.max seconds.
	 */
	if (tp->rto > tp->asoc->rto_max)
		tp->rto = tp->asoc->rto_max;

	tp->rtt = rtt;

	/* Reset rto_pending so that a new RTT measurement is started when a
	 * new data chunk is sent.
	 */
	tp->rto_pending = 0;

	SCTP_DEBUG_PRINTK("%s: transport: %p, rtt: %d, srtt: %d "
353
			  "rttvar: %d, rto: %ld\n", __FUNCTION__,
L
Linus Torvalds 已提交
354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377
			  tp, rtt, tp->srtt, tp->rttvar, tp->rto);
}

/* This routine updates the transport's cwnd and partial_bytes_acked
 * parameters based on the bytes acked in the received SACK.
 */
void sctp_transport_raise_cwnd(struct sctp_transport *transport,
			       __u32 sack_ctsn, __u32 bytes_acked)
{
	__u32 cwnd, ssthresh, flight_size, pba, pmtu;

	cwnd = transport->cwnd;
	flight_size = transport->flight_size;

	/* The appropriate cwnd increase algorithm is performed if, and only
	 * if the cumulative TSN has advanced and the congestion window is
	 * being fully utilized.
	 */
	if ((transport->asoc->ctsn_ack_point >= sack_ctsn) ||
	    (flight_size < cwnd))
		return;

	ssthresh = transport->ssthresh;
	pba = transport->partial_bytes_acked;
378
	pmtu = transport->asoc->pathmtu;
L
Linus Torvalds 已提交
379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449

	if (cwnd <= ssthresh) {
		/* RFC 2960 7.2.1, sctpimpguide-05 2.14.2 When cwnd is less
		 * than or equal to ssthresh an SCTP endpoint MUST use the
		 * slow start algorithm to increase cwnd only if the current
		 * congestion window is being fully utilized and an incoming
		 * SACK advances the Cumulative TSN Ack Point. Only when these
		 * two conditions are met can the cwnd be increased otherwise
		 * the cwnd MUST not be increased. If these conditions are met
		 * then cwnd MUST be increased by at most the lesser of
		 * 1) the total size of the previously outstanding DATA
		 * chunk(s) acknowledged, and 2) the destination's path MTU.
		 */
		if (bytes_acked > pmtu)
			cwnd += pmtu;
		else
			cwnd += bytes_acked;
		SCTP_DEBUG_PRINTK("%s: SLOW START: transport: %p, "
				  "bytes_acked: %d, cwnd: %d, ssthresh: %d, "
				  "flight_size: %d, pba: %d\n",
				  __FUNCTION__,
				  transport, bytes_acked, cwnd,
				  ssthresh, flight_size, pba);
	} else {
		/* RFC 2960 7.2.2 Whenever cwnd is greater than ssthresh,
		 * upon each SACK arrival that advances the Cumulative TSN Ack
		 * Point, increase partial_bytes_acked by the total number of
		 * bytes of all new chunks acknowledged in that SACK including
		 * chunks acknowledged by the new Cumulative TSN Ack and by
		 * Gap Ack Blocks.
		 *
		 * When partial_bytes_acked is equal to or greater than cwnd
		 * and before the arrival of the SACK the sender had cwnd or
		 * more bytes of data outstanding (i.e., before arrival of the
		 * SACK, flightsize was greater than or equal to cwnd),
		 * increase cwnd by MTU, and reset partial_bytes_acked to
		 * (partial_bytes_acked - cwnd).
		 */
		pba += bytes_acked;
		if (pba >= cwnd) {
			cwnd += pmtu;
			pba = ((cwnd < pba) ? (pba - cwnd) : 0);
		}
		SCTP_DEBUG_PRINTK("%s: CONGESTION AVOIDANCE: "
				  "transport: %p, bytes_acked: %d, cwnd: %d, "
				  "ssthresh: %d, flight_size: %d, pba: %d\n",
				  __FUNCTION__,
				  transport, bytes_acked, cwnd,
				  ssthresh, flight_size, pba);
	}

	transport->cwnd = cwnd;
	transport->partial_bytes_acked = pba;
}

/* This routine is used to lower the transport's cwnd when congestion is
 * detected.
 */
void sctp_transport_lower_cwnd(struct sctp_transport *transport,
			       sctp_lower_cwnd_t reason)
{
	switch (reason) {
	case SCTP_LOWER_CWND_T3_RTX:
		/* RFC 2960 Section 7.2.3, sctpimpguide
		 * When the T3-rtx timer expires on an address, SCTP should
		 * perform slow start by:
		 *      ssthresh = max(cwnd/2, 4*MTU)
		 *      cwnd = 1*MTU
		 *      partial_bytes_acked = 0
		 */
		transport->ssthresh = max(transport->cwnd/2,
450 451
					  4*transport->asoc->pathmtu);
		transport->cwnd = transport->asoc->pathmtu;
L
Linus Torvalds 已提交
452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467
		break;

	case SCTP_LOWER_CWND_FAST_RTX:
		/* RFC 2960 7.2.4 Adjust the ssthresh and cwnd of the
		 * destination address(es) to which the missing DATA chunks
		 * were last sent, according to the formula described in
		 * Section 7.2.3.
	 	 *
	 	 * RFC 2960 7.2.3, sctpimpguide Upon detection of packet
		 * losses from SACK (see Section 7.2.4), An endpoint
		 * should do the following:
		 *      ssthresh = max(cwnd/2, 4*MTU)
		 *      cwnd = ssthresh
		 *      partial_bytes_acked = 0
		 */
		transport->ssthresh = max(transport->cwnd/2,
468
					  4*transport->asoc->pathmtu);
L
Linus Torvalds 已提交
469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487
		transport->cwnd = transport->ssthresh;
		break;

	case SCTP_LOWER_CWND_ECNE:
		/* RFC 2481 Section 6.1.2.
		 * If the sender receives an ECN-Echo ACK packet
		 * then the sender knows that congestion was encountered in the
		 * network on the path from the sender to the receiver. The
		 * indication of congestion should be treated just as a
		 * congestion loss in non-ECN Capable TCP. That is, the TCP
		 * source halves the congestion window "cwnd" and reduces the
		 * slow start threshold "ssthresh".
		 * A critical condition is that TCP does not react to
		 * congestion indications more than once every window of
		 * data (or more loosely more than once every round-trip time).
		 */
		if ((jiffies - transport->last_time_ecne_reduced) >
		    transport->rtt) {
			transport->ssthresh = max(transport->cwnd/2,
488
					  	  4*transport->asoc->pathmtu);
L
Linus Torvalds 已提交
489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504
			transport->cwnd = transport->ssthresh;
			transport->last_time_ecne_reduced = jiffies;
		}
		break;

	case SCTP_LOWER_CWND_INACTIVE:
		/* RFC 2960 Section 7.2.1, sctpimpguide
		 * When the endpoint does not transmit data on a given
		 * transport address, the cwnd of the transport address
		 * should be adjusted to max(cwnd/2, 4*MTU) per RTO.
		 * NOTE: Although the draft recommends that this check needs
		 * to be done every RTO interval, we do it every hearbeat
		 * interval.
		 */
		if ((jiffies - transport->last_time_used) > transport->rto)
			transport->cwnd = max(transport->cwnd/2,
505
						 4*transport->asoc->pathmtu);
L
Linus Torvalds 已提交
506 507 508 509 510 511 512 513 514 515 516 517 518 519
		break;
	};

	transport->partial_bytes_acked = 0;
	SCTP_DEBUG_PRINTK("%s: transport: %p reason: %d cwnd: "
			  "%d ssthresh: %d\n", __FUNCTION__,
			  transport, reason,
			  transport->cwnd, transport->ssthresh);
}

/* What is the next timeout value for this transport? */
unsigned long sctp_transport_timeout(struct sctp_transport *t)
{
	unsigned long timeout;
520
	timeout = t->hbinterval + t->rto + sctp_jitter(t->rto);
L
Linus Torvalds 已提交
521 522 523
	timeout += jiffies;
	return timeout;
}