1 /* $NetBSD: tcp_congctl.c,v 1.15 2008/04/28 20:24:09 martin Exp $ */ 2 3 /*- 4 * Copyright (c) 1997, 1998, 1999, 2001, 2005, 2006 The NetBSD Foundation, Inc. 5 * All rights reserved. 6 * 7 * This code is derived from software contributed to The NetBSD Foundation 8 * by Jason R. Thorpe and Kevin M. Lahey of the Numerical Aerospace Simulation 9 * Facility, NASA Ames Research Center. 10 * This code is derived from software contributed to The NetBSD Foundation 11 * by Charles M. Hannum. 12 * This code is derived from software contributed to The NetBSD Foundation 13 * by Rui Paulo. 14 * 15 * Redistribution and use in source and binary forms, with or without 16 * modification, are permitted provided that the following conditions 17 * are met: 18 * 1. Redistributions of source code must retain the above copyright 19 * notice, this list of conditions and the following disclaimer. 20 * 2. Redistributions in binary form must reproduce the above copyright 21 * notice, this list of conditions and the following disclaimer in the 22 * documentation and/or other materials provided with the distribution. 23 * 24 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 25 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 26 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 27 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 28 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 29 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 30 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 31 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 32 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 33 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 34 * POSSIBILITY OF SUCH DAMAGE. 35 */ 36 37 /* 38 * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. 39 * All rights reserved. 40 * 41 * Redistribution and use in source and binary forms, with or without 42 * modification, are permitted provided that the following conditions 43 * are met: 44 * 1. Redistributions of source code must retain the above copyright 45 * notice, this list of conditions and the following disclaimer. 46 * 2. Redistributions in binary form must reproduce the above copyright 47 * notice, this list of conditions and the following disclaimer in the 48 * documentation and/or other materials provided with the distribution. 49 * 3. Neither the name of the project nor the names of its contributors 50 * may be used to endorse or promote products derived from this software 51 * without specific prior written permission. 52 * 53 * THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND 54 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 55 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 56 * ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE 57 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 58 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 59 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 60 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 61 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 62 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 63 * SUCH DAMAGE. 64 */ 65 66 /* 67 * @(#)COPYRIGHT 1.1 (NRL) 17 January 1995 68 * 69 * NRL grants permission for redistribution and use in source and binary 70 * forms, with or without modification, of the software and documentation 71 * created at NRL provided that the following conditions are met: 72 * 73 * 1. Redistributions of source code must retain the above copyright 74 * notice, this list of conditions and the following disclaimer. 75 * 2. Redistributions in binary form must reproduce the above copyright 76 * notice, this list of conditions and the following disclaimer in the 77 * documentation and/or other materials provided with the distribution. 78 * 3. All advertising materials mentioning features or use of this software 79 * must display the following acknowledgements: 80 * This product includes software developed by the University of 81 * California, Berkeley and its contributors. 82 * This product includes software developed at the Information 83 * Technology Division, US Naval Research Laboratory. 84 * 4. Neither the name of the NRL nor the names of its contributors 85 * may be used to endorse or promote products derived from this software 86 * without specific prior written permission. 87 * 88 * THE SOFTWARE PROVIDED BY NRL IS PROVIDED BY NRL AND CONTRIBUTORS ``AS 89 * IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 90 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A 91 * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NRL OR 92 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, 93 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 94 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR 95 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 96 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING 97 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 98 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 99 * 100 * The views and conclusions contained in the software and documentation 101 * are those of the authors and should not be interpreted as representing 102 * official policies, either expressed or implied, of the US Naval 103 * Research Laboratory (NRL). 104 */ 105 106 /* 107 * Copyright (c) 1982, 1986, 1988, 1990, 1993, 1994, 1995 108 * The Regents of the University of California. All rights reserved. 109 * 110 * Redistribution and use in source and binary forms, with or without 111 * modification, are permitted provided that the following conditions 112 * are met: 113 * 1. Redistributions of source code must retain the above copyright 114 * notice, this list of conditions and the following disclaimer. 115 * 2. Redistributions in binary form must reproduce the above copyright 116 * notice, this list of conditions and the following disclaimer in the 117 * documentation and/or other materials provided with the distribution. 118 * 3. Neither the name of the University nor the names of its contributors 119 * may be used to endorse or promote products derived from this software 120 * without specific prior written permission. 121 * 122 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 123 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 124 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 125 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 126 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 127 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 128 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 129 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 130 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 131 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 132 * SUCH DAMAGE. 133 * 134 * @(#)tcp_input.c 8.12 (Berkeley) 5/24/95 135 */ 136 137 #include <sys/cdefs.h> 138 __KERNEL_RCSID(0, "$NetBSD: tcp_congctl.c,v 1.15 2008/04/28 20:24:09 martin Exp $"); 139 140 #include "opt_inet.h" 141 #include "opt_tcp_debug.h" 142 #include "opt_tcp_congctl.h" 143 144 #include <sys/param.h> 145 #include <sys/systm.h> 146 #include <sys/malloc.h> 147 #include <sys/mbuf.h> 148 #include <sys/protosw.h> 149 #include <sys/socket.h> 150 #include <sys/socketvar.h> 151 #include <sys/errno.h> 152 #include <sys/syslog.h> 153 #include <sys/pool.h> 154 #include <sys/domain.h> 155 #include <sys/kernel.h> 156 #include <sys/mutex.h> 157 158 #include <net/if.h> 159 #include <net/route.h> 160 161 #include <netinet/in.h> 162 #include <netinet/in_systm.h> 163 #include <netinet/ip.h> 164 #include <netinet/in_pcb.h> 165 #include <netinet/in_var.h> 166 #include <netinet/ip_var.h> 167 168 #ifdef INET6 169 #ifndef INET 170 #include <netinet/in.h> 171 #endif 172 #include <netinet/ip6.h> 173 #include <netinet6/ip6_var.h> 174 #include <netinet6/in6_pcb.h> 175 #include <netinet6/ip6_var.h> 176 #include <netinet6/in6_var.h> 177 #include <netinet/icmp6.h> 178 #include <netinet6/nd6.h> 179 #endif 180 181 #include <netinet/tcp.h> 182 #include <netinet/tcp_fsm.h> 183 #include <netinet/tcp_seq.h> 184 #include <netinet/tcp_timer.h> 185 #include <netinet/tcp_var.h> 186 #include <netinet/tcpip.h> 187 #include <netinet/tcp_congctl.h> 188 #ifdef TCP_DEBUG 189 #include <netinet/tcp_debug.h> 190 #endif 191 192 /* 193 * TODO: 194 * consider separating the actual implementations in another file. 195 */ 196 197 static int tcp_reno_fast_retransmit(struct tcpcb *, const struct tcphdr *); 198 static void tcp_reno_slow_retransmit(struct tcpcb *); 199 static void tcp_reno_fast_retransmit_newack(struct tcpcb *, 200 const struct tcphdr *); 201 static void tcp_reno_newack(struct tcpcb *, const struct tcphdr *); 202 static void tcp_reno_congestion_exp(struct tcpcb *tp); 203 204 static int tcp_newreno_fast_retransmit(struct tcpcb *, const struct tcphdr *); 205 static void tcp_newreno_fast_retransmit_newack(struct tcpcb *, 206 const struct tcphdr *); 207 static void tcp_newreno_newack(struct tcpcb *, const struct tcphdr *); 208 209 210 static void tcp_congctl_fillnames(void); 211 212 extern int tcprexmtthresh; 213 214 MALLOC_DEFINE(M_TCPCONGCTL, "tcpcongctl", "TCP congestion control structures"); 215 216 /* currently selected global congestion control */ 217 char tcp_congctl_global_name[TCPCC_MAXLEN]; 218 219 /* available global congestion control algorithms */ 220 char tcp_congctl_avail[10 * TCPCC_MAXLEN]; 221 222 /* 223 * Used to list the available congestion control algorithms. 224 */ 225 TAILQ_HEAD(, tcp_congctlent) tcp_congctlhd = 226 TAILQ_HEAD_INITIALIZER(tcp_congctlhd); 227 228 static struct tcp_congctlent * tcp_congctl_global; 229 230 static kmutex_t tcp_congctl_mtx; 231 232 void 233 tcp_congctl_init(void) 234 { 235 int r; 236 237 mutex_init(&tcp_congctl_mtx, MUTEX_DEFAULT, IPL_NONE); 238 239 /* Base algorithms. */ 240 r = tcp_congctl_register("reno", &tcp_reno_ctl); 241 KASSERT(r == 0); 242 r = tcp_congctl_register("newreno", &tcp_newreno_ctl); 243 KASSERT(r == 0); 244 245 /* NewReno is the default. */ 246 #ifndef TCP_CONGCTL_DEFAULT 247 #define TCP_CONGCTL_DEFAULT "newreno" 248 #endif 249 250 r = tcp_congctl_select(NULL, TCP_CONGCTL_DEFAULT); 251 KASSERT(r == 0); 252 } 253 254 /* 255 * Register a congestion algorithm and select it if we have none. 256 */ 257 int 258 tcp_congctl_register(const char *name, const struct tcp_congctl *tcc) 259 { 260 struct tcp_congctlent *ntcc, *tccp; 261 262 TAILQ_FOREACH(tccp, &tcp_congctlhd, congctl_ent) 263 if (!strcmp(name, tccp->congctl_name)) { 264 /* name already registered */ 265 return EEXIST; 266 } 267 268 ntcc = malloc(sizeof(*ntcc), M_TCPCONGCTL, M_WAITOK|M_ZERO); 269 270 strlcpy(ntcc->congctl_name, name, sizeof(ntcc->congctl_name) - 1); 271 ntcc->congctl_ctl = tcc; 272 273 TAILQ_INSERT_TAIL(&tcp_congctlhd, ntcc, congctl_ent); 274 tcp_congctl_fillnames(); 275 276 if (TAILQ_FIRST(&tcp_congctlhd) == ntcc) 277 tcp_congctl_select(NULL, name); 278 279 return 0; 280 } 281 282 int 283 tcp_congctl_unregister(const char *name) 284 { 285 struct tcp_congctlent *tccp, *rtccp; 286 unsigned int size; 287 288 rtccp = NULL; 289 size = 0; 290 TAILQ_FOREACH(tccp, &tcp_congctlhd, congctl_ent) { 291 if (!strcmp(name, tccp->congctl_name)) 292 rtccp = tccp; 293 size++; 294 } 295 296 if (!rtccp) 297 return ENOENT; 298 299 if (size <= 1 || tcp_congctl_global == rtccp || rtccp->congctl_refcnt) 300 return EBUSY; 301 302 TAILQ_REMOVE(&tcp_congctlhd, rtccp, congctl_ent); 303 free(rtccp, M_TCPCONGCTL); 304 tcp_congctl_fillnames(); 305 306 return 0; 307 } 308 309 /* 310 * Select a congestion algorithm by name. 311 */ 312 int 313 tcp_congctl_select(struct tcpcb *tp, const char *name) 314 { 315 struct tcp_congctlent *tccp, *old_tccp, *new_tccp; 316 bool old_found, new_found; 317 318 KASSERT(name); 319 320 old_found = (tp == NULL || tp->t_congctl == NULL); 321 old_tccp = NULL; 322 new_found = false; 323 new_tccp = NULL; 324 325 TAILQ_FOREACH(tccp, &tcp_congctlhd, congctl_ent) { 326 if (!old_found && tccp->congctl_ctl == tp->t_congctl) { 327 old_tccp = tccp; 328 old_found = true; 329 } 330 331 if (!new_found && !strcmp(name, tccp->congctl_name)) { 332 new_tccp = tccp; 333 new_found = true; 334 } 335 336 if (new_found && old_found) { 337 if (tp) { 338 mutex_enter(&tcp_congctl_mtx); 339 if (old_tccp) 340 old_tccp->congctl_refcnt--; 341 tp->t_congctl = new_tccp->congctl_ctl; 342 new_tccp->congctl_refcnt++; 343 mutex_exit(&tcp_congctl_mtx); 344 } else { 345 tcp_congctl_global = new_tccp; 346 strlcpy(tcp_congctl_global_name, 347 new_tccp->congctl_name, 348 sizeof(tcp_congctl_global_name) - 1); 349 } 350 return 0; 351 } 352 } 353 354 return EINVAL; 355 } 356 357 void 358 tcp_congctl_release(struct tcpcb *tp) 359 { 360 struct tcp_congctlent *tccp; 361 362 KASSERT(tp->t_congctl); 363 364 TAILQ_FOREACH(tccp, &tcp_congctlhd, congctl_ent) { 365 if (tccp->congctl_ctl == tp->t_congctl) { 366 tccp->congctl_refcnt--; 367 return; 368 } 369 } 370 } 371 372 /* 373 * Returns the name of a congestion algorithm. 374 */ 375 const char * 376 tcp_congctl_bystruct(const struct tcp_congctl *tcc) 377 { 378 struct tcp_congctlent *tccp; 379 380 KASSERT(tcc); 381 382 TAILQ_FOREACH(tccp, &tcp_congctlhd, congctl_ent) 383 if (tccp->congctl_ctl == tcc) 384 return tccp->congctl_name; 385 386 return NULL; 387 } 388 389 static void 390 tcp_congctl_fillnames(void) 391 { 392 struct tcp_congctlent *tccp; 393 const char *delim = " "; 394 395 tcp_congctl_avail[0] = '\0'; 396 TAILQ_FOREACH(tccp, &tcp_congctlhd, congctl_ent) { 397 strlcat(tcp_congctl_avail, tccp->congctl_name, 398 sizeof(tcp_congctl_avail) - 1); 399 if (TAILQ_NEXT(tccp, congctl_ent)) 400 strlcat(tcp_congctl_avail, delim, 401 sizeof(tcp_congctl_avail) - 1); 402 } 403 404 } 405 406 /* ------------------------------------------------------------------------ */ 407 408 /* 409 * TCP/Reno congestion control. 410 */ 411 static void 412 tcp_reno_congestion_exp(struct tcpcb *tp) 413 { 414 u_int win; 415 416 /* 417 * Halve the congestion window and reduce the 418 * slow start threshold. 419 */ 420 win = min(tp->snd_wnd, tp->snd_cwnd) / 2 / tp->t_segsz; 421 if (win < 2) 422 win = 2; 423 424 tp->snd_ssthresh = win * tp->t_segsz; 425 tp->snd_recover = tp->snd_max; 426 tp->snd_cwnd = tp->snd_ssthresh; 427 428 /* 429 * When using TCP ECN, notify the peer that 430 * we reduced the cwnd. 431 */ 432 if (TCP_ECN_ALLOWED(tp)) 433 tp->t_flags |= TF_ECN_SND_CWR; 434 } 435 436 437 438 static int 439 tcp_reno_fast_retransmit(struct tcpcb *tp, const struct tcphdr *th) 440 { 441 /* 442 * We know we're losing at the current 443 * window size so do congestion avoidance 444 * (set ssthresh to half the current window 445 * and pull our congestion window back to 446 * the new ssthresh). 447 * 448 * Dup acks mean that packets have left the 449 * network (they're now cached at the receiver) 450 * so bump cwnd by the amount in the receiver 451 * to keep a constant cwnd packets in the 452 * network. 453 * 454 * If we are using TCP/SACK, then enter 455 * Fast Recovery if the receiver SACKs 456 * data that is tcprexmtthresh * MSS 457 * bytes past the last ACKed segment, 458 * irrespective of the number of DupAcks. 459 */ 460 461 tcp_seq onxt; 462 463 onxt = tp->snd_nxt; 464 tcp_reno_congestion_exp(tp); 465 tp->t_partialacks = 0; 466 TCP_TIMER_DISARM(tp, TCPT_REXMT); 467 tp->t_rtttime = 0; 468 if (TCP_SACK_ENABLED(tp)) { 469 tp->t_dupacks = tcprexmtthresh; 470 tp->sack_newdata = tp->snd_nxt; 471 tp->snd_cwnd = tp->t_segsz; 472 (void) tcp_output(tp); 473 return 0; 474 } 475 tp->snd_nxt = th->th_ack; 476 tp->snd_cwnd = tp->t_segsz; 477 (void) tcp_output(tp); 478 tp->snd_cwnd = tp->snd_ssthresh + tp->t_segsz * tp->t_dupacks; 479 if (SEQ_GT(onxt, tp->snd_nxt)) 480 tp->snd_nxt = onxt; 481 482 return 0; 483 } 484 485 static void 486 tcp_reno_slow_retransmit(struct tcpcb *tp) 487 { 488 u_int win; 489 490 /* 491 * Close the congestion window down to one segment 492 * (we'll open it by one segment for each ack we get). 493 * Since we probably have a window's worth of unacked 494 * data accumulated, this "slow start" keeps us from 495 * dumping all that data as back-to-back packets (which 496 * might overwhelm an intermediate gateway). 497 * 498 * There are two phases to the opening: Initially we 499 * open by one mss on each ack. This makes the window 500 * size increase exponentially with time. If the 501 * window is larger than the path can handle, this 502 * exponential growth results in dropped packet(s) 503 * almost immediately. To get more time between 504 * drops but still "push" the network to take advantage 505 * of improving conditions, we switch from exponential 506 * to linear window opening at some threshhold size. 507 * For a threshhold, we use half the current window 508 * size, truncated to a multiple of the mss. 509 * 510 * (the minimum cwnd that will give us exponential 511 * growth is 2 mss. We don't allow the threshhold 512 * to go below this.) 513 */ 514 515 win = min(tp->snd_wnd, tp->snd_cwnd) / 2 / tp->t_segsz; 516 if (win < 2) 517 win = 2; 518 /* Loss Window MUST be one segment. */ 519 tp->snd_cwnd = tp->t_segsz; 520 tp->snd_ssthresh = win * tp->t_segsz; 521 tp->t_partialacks = -1; 522 tp->t_dupacks = 0; 523 tp->t_bytes_acked = 0; 524 } 525 526 static void 527 tcp_reno_fast_retransmit_newack(struct tcpcb *tp, 528 const struct tcphdr *th) 529 { 530 if (tp->t_partialacks < 0) { 531 /* 532 * We were not in fast recovery. Reset the duplicate ack 533 * counter. 534 */ 535 tp->t_dupacks = 0; 536 } else { 537 /* 538 * Clamp the congestion window to the crossover point and 539 * exit fast recovery. 540 */ 541 if (tp->snd_cwnd > tp->snd_ssthresh) 542 tp->snd_cwnd = tp->snd_ssthresh; 543 tp->t_partialacks = -1; 544 tp->t_dupacks = 0; 545 tp->t_bytes_acked = 0; 546 } 547 } 548 549 static void 550 tcp_reno_newack(struct tcpcb *tp, const struct tcphdr *th) 551 { 552 /* 553 * When new data is acked, open the congestion window. 554 */ 555 556 u_int cw = tp->snd_cwnd; 557 u_int incr = tp->t_segsz; 558 559 if (tcp_do_abc) { 560 561 /* 562 * RFC 3465 Appropriate Byte Counting (ABC) 563 */ 564 565 int acked = th->th_ack - tp->snd_una; 566 567 if (cw >= tp->snd_ssthresh) { 568 tp->t_bytes_acked += acked; 569 if (tp->t_bytes_acked >= cw) { 570 /* Time to increase the window. */ 571 tp->t_bytes_acked -= cw; 572 } else { 573 /* No need to increase yet. */ 574 incr = 0; 575 } 576 } else { 577 /* 578 * use 2*SMSS or 1*SMSS for the "L" param, 579 * depending on sysctl setting. 580 * 581 * (See RFC 3465 2.3 Choosing the Limit) 582 */ 583 u_int abc_lim; 584 585 abc_lim = (tcp_abc_aggressive == 0 || 586 tp->snd_nxt != tp->snd_max) ? incr : incr * 2; 587 incr = min(acked, abc_lim); 588 } 589 } else { 590 591 /* 592 * If the window gives us less than ssthresh packets 593 * in flight, open exponentially (segsz per packet). 594 * Otherwise open linearly: segsz per window 595 * (segsz^2 / cwnd per packet). 596 */ 597 598 if (cw >= tp->snd_ssthresh) { 599 incr = incr * incr / cw; 600 } 601 } 602 603 tp->snd_cwnd = min(cw + incr, TCP_MAXWIN << tp->snd_scale); 604 } 605 606 const struct tcp_congctl tcp_reno_ctl = { 607 .fast_retransmit = tcp_reno_fast_retransmit, 608 .slow_retransmit = tcp_reno_slow_retransmit, 609 .fast_retransmit_newack = tcp_reno_fast_retransmit_newack, 610 .newack = tcp_reno_newack, 611 .cong_exp = tcp_reno_congestion_exp, 612 }; 613 614 /* 615 * TCP/NewReno Congestion control. 616 */ 617 static int 618 tcp_newreno_fast_retransmit(struct tcpcb *tp, const struct tcphdr *th) 619 { 620 if (SEQ_LT(th->th_ack, tp->snd_high)) { 621 /* 622 * False fast retransmit after timeout. 623 * Do not enter fast recovery 624 */ 625 tp->t_dupacks = 0; 626 return 1; 627 } else { 628 /* 629 * Fast retransmit is same as reno. 630 */ 631 return tcp_reno_fast_retransmit(tp, th); 632 } 633 634 return 0; 635 } 636 637 /* 638 * Implement the NewReno response to a new ack, checking for partial acks in 639 * fast recovery. 640 */ 641 static void 642 tcp_newreno_fast_retransmit_newack(struct tcpcb *tp, const struct tcphdr *th) 643 { 644 if (tp->t_partialacks < 0) { 645 /* 646 * We were not in fast recovery. Reset the duplicate ack 647 * counter. 648 */ 649 tp->t_dupacks = 0; 650 } else if (SEQ_LT(th->th_ack, tp->snd_recover)) { 651 /* 652 * This is a partial ack. Retransmit the first unacknowledged 653 * segment and deflate the congestion window by the amount of 654 * acknowledged data. Do not exit fast recovery. 655 */ 656 tcp_seq onxt = tp->snd_nxt; 657 u_long ocwnd = tp->snd_cwnd; 658 659 /* 660 * snd_una has not yet been updated and the socket's send 661 * buffer has not yet drained off the ACK'd data, so we 662 * have to leave snd_una as it was to get the correct data 663 * offset in tcp_output(). 664 */ 665 if (++tp->t_partialacks == 1) 666 TCP_TIMER_DISARM(tp, TCPT_REXMT); 667 tp->t_rtttime = 0; 668 tp->snd_nxt = th->th_ack; 669 /* 670 * Set snd_cwnd to one segment beyond ACK'd offset. snd_una 671 * is not yet updated when we're called. 672 */ 673 tp->snd_cwnd = tp->t_segsz + (th->th_ack - tp->snd_una); 674 (void) tcp_output(tp); 675 tp->snd_cwnd = ocwnd; 676 if (SEQ_GT(onxt, tp->snd_nxt)) 677 tp->snd_nxt = onxt; 678 /* 679 * Partial window deflation. Relies on fact that tp->snd_una 680 * not updated yet. 681 */ 682 tp->snd_cwnd -= (th->th_ack - tp->snd_una - tp->t_segsz); 683 } else { 684 /* 685 * Complete ack. Inflate the congestion window to ssthresh 686 * and exit fast recovery. 687 * 688 * Window inflation should have left us with approx. 689 * snd_ssthresh outstanding data. But in case we 690 * would be inclined to send a burst, better to do 691 * it via the slow start mechanism. 692 */ 693 if (SEQ_SUB(tp->snd_max, th->th_ack) < tp->snd_ssthresh) 694 tp->snd_cwnd = SEQ_SUB(tp->snd_max, th->th_ack) 695 + tp->t_segsz; 696 else 697 tp->snd_cwnd = tp->snd_ssthresh; 698 tp->t_partialacks = -1; 699 tp->t_dupacks = 0; 700 tp->t_bytes_acked = 0; 701 } 702 } 703 704 static void 705 tcp_newreno_newack(struct tcpcb *tp, const struct tcphdr *th) 706 { 707 /* 708 * If we are still in fast recovery (meaning we are using 709 * NewReno and we have only received partial acks), do not 710 * inflate the window yet. 711 */ 712 if (tp->t_partialacks < 0) 713 tcp_reno_newack(tp, th); 714 } 715 716 717 const struct tcp_congctl tcp_newreno_ctl = { 718 .fast_retransmit = tcp_newreno_fast_retransmit, 719 .slow_retransmit = tcp_reno_slow_retransmit, 720 .fast_retransmit_newack = tcp_newreno_fast_retransmit_newack, 721 .newack = tcp_newreno_newack, 722 .cong_exp = tcp_reno_congestion_exp, 723 }; 724 725 726