tcp_input.c revision 1.163
1 /* $NetBSD: tcp_input.c,v 1.163 2003/03/01 04:40:27 thorpej Exp $ */ 2 3 /* 4 * Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project. 5 * All rights reserved. 6 * 7 * Redistribution and use in source and binary forms, with or without 8 * modification, are permitted provided that the following conditions 9 * are met: 10 * 1. Redistributions of source code must retain the above copyright 11 * notice, this list of conditions and the following disclaimer. 12 * 2. Redistributions in binary form must reproduce the above copyright 13 * notice, this list of conditions and the following disclaimer in the 14 * documentation and/or other materials provided with the distribution. 15 * 3. Neither the name of the project nor the names of its contributors 16 * may be used to endorse or promote products derived from this software 17 * without specific prior written permission. 18 * 19 * THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND 20 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 21 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 22 * ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE 23 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 24 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 25 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 26 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 27 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 28 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 29 * SUCH DAMAGE. 30 */ 31 32 /* 33 * @(#)COPYRIGHT 1.1 (NRL) 17 January 1995 34 * 35 * NRL grants permission for redistribution and use in source and binary 36 * forms, with or without modification, of the software and documentation 37 * created at NRL provided that the following conditions are met: 38 * 39 * 1. Redistributions of source code must retain the above copyright 40 * notice, this list of conditions and the following disclaimer. 41 * 2. Redistributions in binary form must reproduce the above copyright 42 * notice, this list of conditions and the following disclaimer in the 43 * documentation and/or other materials provided with the distribution. 44 * 3. All advertising materials mentioning features or use of this software 45 * must display the following acknowledgements: 46 * This product includes software developed by the University of 47 * California, Berkeley and its contributors. 48 * This product includes software developed at the Information 49 * Technology Division, US Naval Research Laboratory. 50 * 4. Neither the name of the NRL nor the names of its contributors 51 * may be used to endorse or promote products derived from this software 52 * without specific prior written permission. 53 * 54 * THE SOFTWARE PROVIDED BY NRL IS PROVIDED BY NRL AND CONTRIBUTORS ``AS 55 * IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 56 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A 57 * PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NRL OR 58 * CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, 59 * EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, 60 * PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR 61 * PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF 62 * LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING 63 * NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 64 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 65 * 66 * The views and conclusions contained in the software and documentation 67 * are those of the authors and should not be interpreted as representing 68 * official policies, either expressed or implied, of the US Naval 69 * Research Laboratory (NRL). 70 */ 71 72 /*- 73 * Copyright (c) 1997, 1998, 1999, 2001 The NetBSD Foundation, Inc. 74 * All rights reserved. 75 * 76 * This code is derived from software contributed to The NetBSD Foundation 77 * by Jason R. Thorpe and Kevin M. Lahey of the Numerical Aerospace Simulation 78 * Facility, NASA Ames Research Center. 79 * 80 * Redistribution and use in source and binary forms, with or without 81 * modification, are permitted provided that the following conditions 82 * are met: 83 * 1. Redistributions of source code must retain the above copyright 84 * notice, this list of conditions and the following disclaimer. 85 * 2. Redistributions in binary form must reproduce the above copyright 86 * notice, this list of conditions and the following disclaimer in the 87 * documentation and/or other materials provided with the distribution. 88 * 3. All advertising materials mentioning features or use of this software 89 * must display the following acknowledgement: 90 * This product includes software developed by the NetBSD 91 * Foundation, Inc. and its contributors. 92 * 4. Neither the name of The NetBSD Foundation nor the names of its 93 * contributors may be used to endorse or promote products derived 94 * from this software without specific prior written permission. 95 * 96 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 97 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 98 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 99 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 100 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 101 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 102 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 103 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 104 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 105 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 106 * POSSIBILITY OF SUCH DAMAGE. 107 */ 108 109 /* 110 * Copyright (c) 1982, 1986, 1988, 1990, 1993, 1994, 1995 111 * The Regents of the University of California. All rights reserved. 112 * 113 * Redistribution and use in source and binary forms, with or without 114 * modification, are permitted provided that the following conditions 115 * are met: 116 * 1. Redistributions of source code must retain the above copyright 117 * notice, this list of conditions and the following disclaimer. 118 * 2. Redistributions in binary form must reproduce the above copyright 119 * notice, this list of conditions and the following disclaimer in the 120 * documentation and/or other materials provided with the distribution. 121 * 3. All advertising materials mentioning features or use of this software 122 * must display the following acknowledgement: 123 * This product includes software developed by the University of 124 * California, Berkeley and its contributors. 125 * 4. Neither the name of the University nor the names of its contributors 126 * may be used to endorse or promote products derived from this software 127 * without specific prior written permission. 128 * 129 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 130 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 131 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 132 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 133 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 134 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 135 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 136 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 137 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 138 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 139 * SUCH DAMAGE. 140 * 141 * @(#)tcp_input.c 8.12 (Berkeley) 5/24/95 142 */ 143 144 /* 145 * TODO list for SYN cache stuff: 146 * 147 * Find room for a "state" field, which is needed to keep a 148 * compressed state for TIME_WAIT TCBs. It's been noted already 149 * that this is fairly important for very high-volume web and 150 * mail servers, which use a large number of short-lived 151 * connections. 152 */ 153 154 #include <sys/cdefs.h> 155 __KERNEL_RCSID(0, "$NetBSD: tcp_input.c,v 1.163 2003/03/01 04:40:27 thorpej Exp $"); 156 157 #include "opt_inet.h" 158 #include "opt_ipsec.h" 159 #include "opt_inet_csum.h" 160 #include "opt_tcp_debug.h" 161 162 #include <sys/param.h> 163 #include <sys/systm.h> 164 #include <sys/malloc.h> 165 #include <sys/mbuf.h> 166 #include <sys/protosw.h> 167 #include <sys/socket.h> 168 #include <sys/socketvar.h> 169 #include <sys/errno.h> 170 #include <sys/syslog.h> 171 #include <sys/pool.h> 172 #include <sys/domain.h> 173 #include <sys/kernel.h> 174 175 #include <net/if.h> 176 #include <net/route.h> 177 #include <net/if_types.h> 178 179 #include <netinet/in.h> 180 #include <netinet/in_systm.h> 181 #include <netinet/ip.h> 182 #include <netinet/in_pcb.h> 183 #include <netinet/ip_var.h> 184 185 #ifdef INET6 186 #ifndef INET 187 #include <netinet/in.h> 188 #endif 189 #include <netinet/ip6.h> 190 #include <netinet6/ip6_var.h> 191 #include <netinet6/in6_pcb.h> 192 #include <netinet6/ip6_var.h> 193 #include <netinet6/in6_var.h> 194 #include <netinet/icmp6.h> 195 #include <netinet6/nd6.h> 196 #endif 197 198 #ifdef PULLDOWN_TEST 199 #ifndef INET6 200 /* always need ip6.h for IP6_EXTHDR_GET */ 201 #include <netinet/ip6.h> 202 #endif 203 #endif 204 205 #include <netinet/tcp.h> 206 #include <netinet/tcp_fsm.h> 207 #include <netinet/tcp_seq.h> 208 #include <netinet/tcp_timer.h> 209 #include <netinet/tcp_var.h> 210 #include <netinet/tcpip.h> 211 #include <netinet/tcp_debug.h> 212 213 #include <machine/stdarg.h> 214 215 #ifdef IPSEC 216 #include <netinet6/ipsec.h> 217 #include <netkey/key.h> 218 #endif /*IPSEC*/ 219 #ifdef INET6 220 #include "faith.h" 221 #if defined(NFAITH) && NFAITH > 0 222 #include <net/if_faith.h> 223 #endif 224 #endif 225 226 int tcprexmtthresh = 3; 227 int tcp_log_refused; 228 229 static int tcp_rst_ppslim_count = 0; 230 static struct timeval tcp_rst_ppslim_last; 231 232 #define TCP_PAWS_IDLE (24 * 24 * 60 * 60 * PR_SLOWHZ) 233 234 /* for modulo comparisons of timestamps */ 235 #define TSTMP_LT(a,b) ((int)((a)-(b)) < 0) 236 #define TSTMP_GEQ(a,b) ((int)((a)-(b)) >= 0) 237 238 /* 239 * Neighbor Discovery, Neighbor Unreachability Detection Upper layer hint. 240 */ 241 #ifdef INET6 242 #define ND6_HINT(tp) \ 243 do { \ 244 if (tp && tp->t_in6pcb && tp->t_family == AF_INET6 \ 245 && tp->t_in6pcb->in6p_route.ro_rt) { \ 246 nd6_nud_hint(tp->t_in6pcb->in6p_route.ro_rt, NULL, 0); \ 247 } \ 248 } while (/*CONSTCOND*/ 0) 249 #else 250 #define ND6_HINT(tp) 251 #endif 252 253 /* 254 * Macro to compute ACK transmission behavior. Delay the ACK unless 255 * we have already delayed an ACK (must send an ACK every two segments). 256 * We also ACK immediately if we received a PUSH and the ACK-on-PUSH 257 * option is enabled. 258 */ 259 #define TCP_SETUP_ACK(tp, th) \ 260 do { \ 261 if ((tp)->t_flags & TF_DELACK || \ 262 (tcp_ack_on_push && (th)->th_flags & TH_PUSH)) \ 263 tp->t_flags |= TF_ACKNOW; \ 264 else \ 265 TCP_SET_DELACK(tp); \ 266 } while (/*CONSTCOND*/ 0) 267 268 /* 269 * Convert TCP protocol fields to host order for easier processing. 270 */ 271 #define TCP_FIELDS_TO_HOST(th) \ 272 do { \ 273 NTOHL((th)->th_seq); \ 274 NTOHL((th)->th_ack); \ 275 NTOHS((th)->th_win); \ 276 NTOHS((th)->th_urp); \ 277 } while (/*CONSTCOND*/ 0) 278 279 /* 280 * ... and reverse the above. 281 */ 282 #define TCP_FIELDS_TO_NET(th) \ 283 do { \ 284 HTONL((th)->th_seq); \ 285 HTONL((th)->th_ack); \ 286 HTONS((th)->th_win); \ 287 HTONS((th)->th_urp); \ 288 } while (/*CONSTCOND*/ 0) 289 290 #ifdef TCP_CSUM_COUNTERS 291 #include <sys/device.h> 292 293 extern struct evcnt tcp_hwcsum_ok; 294 extern struct evcnt tcp_hwcsum_bad; 295 extern struct evcnt tcp_hwcsum_data; 296 extern struct evcnt tcp_swcsum; 297 298 #define TCP_CSUM_COUNTER_INCR(ev) (ev)->ev_count++ 299 300 #else 301 302 #define TCP_CSUM_COUNTER_INCR(ev) /* nothing */ 303 304 #endif /* TCP_CSUM_COUNTERS */ 305 306 #ifdef TCP_REASS_COUNTERS 307 #include <sys/device.h> 308 309 extern struct evcnt tcp_reass_; 310 extern struct evcnt tcp_reass_empty; 311 extern struct evcnt tcp_reass_iteration[8]; 312 extern struct evcnt tcp_reass_prependfirst; 313 extern struct evcnt tcp_reass_prepend; 314 extern struct evcnt tcp_reass_insert; 315 extern struct evcnt tcp_reass_inserttail; 316 extern struct evcnt tcp_reass_append; 317 extern struct evcnt tcp_reass_appendtail; 318 extern struct evcnt tcp_reass_overlaptail; 319 extern struct evcnt tcp_reass_overlapfront; 320 extern struct evcnt tcp_reass_segdup; 321 extern struct evcnt tcp_reass_fragdup; 322 323 #define TCP_REASS_COUNTER_INCR(ev) (ev)->ev_count++ 324 325 #else 326 327 #define TCP_REASS_COUNTER_INCR(ev) /* nothing */ 328 329 #endif /* TCP_REASS_COUNTERS */ 330 331 #ifdef INET 332 static void tcp4_log_refused __P((const struct ip *, const struct tcphdr *)); 333 #endif 334 #ifdef INET6 335 static void tcp6_log_refused 336 __P((const struct ip6_hdr *, const struct tcphdr *)); 337 #endif 338 339 int 340 tcp_reass(tp, th, m, tlen) 341 struct tcpcb *tp; 342 struct tcphdr *th; 343 struct mbuf *m; 344 int *tlen; 345 { 346 struct ipqent *p, *q, *nq, *tiqe = NULL; 347 struct socket *so = NULL; 348 int pkt_flags; 349 tcp_seq pkt_seq; 350 unsigned pkt_len; 351 u_long rcvpartdupbyte = 0; 352 u_long rcvoobyte; 353 #ifdef TCP_REASS_COUNTERS 354 u_int count = 0; 355 #endif 356 357 if (tp->t_inpcb) 358 so = tp->t_inpcb->inp_socket; 359 #ifdef INET6 360 else if (tp->t_in6pcb) 361 so = tp->t_in6pcb->in6p_socket; 362 #endif 363 364 TCP_REASS_LOCK_CHECK(tp); 365 366 /* 367 * Call with th==0 after become established to 368 * force pre-ESTABLISHED data up to user socket. 369 */ 370 if (th == 0) 371 goto present; 372 373 rcvoobyte = *tlen; 374 /* 375 * Copy these to local variables because the tcpiphdr 376 * gets munged while we are collapsing mbufs. 377 */ 378 pkt_seq = th->th_seq; 379 pkt_len = *tlen; 380 pkt_flags = th->th_flags; 381 382 TCP_REASS_COUNTER_INCR(&tcp_reass_); 383 384 if ((p = TAILQ_LAST(&tp->segq, ipqehead)) != NULL) { 385 /* 386 * When we miss a packet, the vast majority of time we get 387 * packets that follow it in order. So optimize for that. 388 */ 389 if (pkt_seq == p->ipqe_seq + p->ipqe_len) { 390 p->ipqe_len += pkt_len; 391 p->ipqe_flags |= pkt_flags; 392 m_cat(p->ipqe_m, m); 393 tiqe = p; 394 TAILQ_REMOVE(&tp->timeq, p, ipqe_timeq); 395 TCP_REASS_COUNTER_INCR(&tcp_reass_appendtail); 396 goto skip_replacement; 397 } 398 /* 399 * While we're here, if the pkt is completely beyond 400 * anything we have, just insert it at the tail. 401 */ 402 if (SEQ_GT(pkt_seq, p->ipqe_seq + p->ipqe_len)) { 403 TCP_REASS_COUNTER_INCR(&tcp_reass_inserttail); 404 goto insert_it; 405 } 406 } 407 408 q = TAILQ_FIRST(&tp->segq); 409 410 if (q != NULL) { 411 /* 412 * If this segment immediately precedes the first out-of-order 413 * block, simply slap the segment in front of it and (mostly) 414 * skip the complicated logic. 415 */ 416 if (pkt_seq + pkt_len == q->ipqe_seq) { 417 q->ipqe_seq = pkt_seq; 418 q->ipqe_len += pkt_len; 419 q->ipqe_flags |= pkt_flags; 420 m_cat(m, q->ipqe_m); 421 q->ipqe_m = m; 422 tiqe = q; 423 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 424 TCP_REASS_COUNTER_INCR(&tcp_reass_prependfirst); 425 goto skip_replacement; 426 } 427 } else { 428 TCP_REASS_COUNTER_INCR(&tcp_reass_empty); 429 } 430 431 /* 432 * Find a segment which begins after this one does. 433 */ 434 for (p = NULL; q != NULL; q = nq) { 435 nq = TAILQ_NEXT(q, ipqe_q); 436 #ifdef TCP_REASS_COUNTERS 437 count++; 438 #endif 439 /* 440 * If the received segment is just right after this 441 * fragment, merge the two together and then check 442 * for further overlaps. 443 */ 444 if (q->ipqe_seq + q->ipqe_len == pkt_seq) { 445 #ifdef TCPREASS_DEBUG 446 printf("tcp_reass[%p]: concat %u:%u(%u) to %u:%u(%u)\n", 447 tp, pkt_seq, pkt_seq + pkt_len, pkt_len, 448 q->ipqe_seq, q->ipqe_seq + q->ipqe_len, q->ipqe_len); 449 #endif 450 pkt_len += q->ipqe_len; 451 pkt_flags |= q->ipqe_flags; 452 pkt_seq = q->ipqe_seq; 453 m_cat(q->ipqe_m, m); 454 m = q->ipqe_m; 455 TCP_REASS_COUNTER_INCR(&tcp_reass_append); 456 goto free_ipqe; 457 } 458 /* 459 * If the received segment is completely past this 460 * fragment, we need to go the next fragment. 461 */ 462 if (SEQ_LT(q->ipqe_seq + q->ipqe_len, pkt_seq)) { 463 p = q; 464 continue; 465 } 466 /* 467 * If the fragment is past the received segment, 468 * it (or any following) can't be concatenated. 469 */ 470 if (SEQ_GT(q->ipqe_seq, pkt_seq + pkt_len)) { 471 TCP_REASS_COUNTER_INCR(&tcp_reass_insert); 472 break; 473 } 474 475 /* 476 * We've received all the data in this segment before. 477 * mark it as a duplicate and return. 478 */ 479 if (SEQ_LEQ(q->ipqe_seq, pkt_seq) && 480 SEQ_GEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) { 481 tcpstat.tcps_rcvduppack++; 482 tcpstat.tcps_rcvdupbyte += pkt_len; 483 m_freem(m); 484 if (tiqe != NULL) 485 pool_put(&ipqent_pool, tiqe); 486 TCP_REASS_COUNTER_INCR(&tcp_reass_segdup); 487 return (0); 488 } 489 /* 490 * Received segment completely overlaps this fragment 491 * so we drop the fragment (this keeps the temporal 492 * ordering of segments correct). 493 */ 494 if (SEQ_GEQ(q->ipqe_seq, pkt_seq) && 495 SEQ_LEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) { 496 rcvpartdupbyte += q->ipqe_len; 497 m_freem(q->ipqe_m); 498 TCP_REASS_COUNTER_INCR(&tcp_reass_fragdup); 499 goto free_ipqe; 500 } 501 /* 502 * RX'ed segment extends past the end of the 503 * fragment. Drop the overlapping bytes. Then 504 * merge the fragment and segment then treat as 505 * a longer received packet. 506 */ 507 if (SEQ_LT(q->ipqe_seq, pkt_seq) 508 && SEQ_GT(q->ipqe_seq + q->ipqe_len, pkt_seq)) { 509 int overlap = q->ipqe_seq + q->ipqe_len - pkt_seq; 510 #ifdef TCPREASS_DEBUG 511 printf("tcp_reass[%p]: trim starting %d bytes of %u:%u(%u)\n", 512 tp, overlap, 513 pkt_seq, pkt_seq + pkt_len, pkt_len); 514 #endif 515 m_adj(m, overlap); 516 rcvpartdupbyte += overlap; 517 m_cat(q->ipqe_m, m); 518 m = q->ipqe_m; 519 pkt_seq = q->ipqe_seq; 520 pkt_len += q->ipqe_len - overlap; 521 rcvoobyte -= overlap; 522 TCP_REASS_COUNTER_INCR(&tcp_reass_overlaptail); 523 goto free_ipqe; 524 } 525 /* 526 * RX'ed segment extends past the front of the 527 * fragment. Drop the overlapping bytes on the 528 * received packet. The packet will then be 529 * contatentated with this fragment a bit later. 530 */ 531 if (SEQ_GT(q->ipqe_seq, pkt_seq) 532 && SEQ_LT(q->ipqe_seq, pkt_seq + pkt_len)) { 533 int overlap = pkt_seq + pkt_len - q->ipqe_seq; 534 #ifdef TCPREASS_DEBUG 535 printf("tcp_reass[%p]: trim trailing %d bytes of %u:%u(%u)\n", 536 tp, overlap, 537 pkt_seq, pkt_seq + pkt_len, pkt_len); 538 #endif 539 m_adj(m, -overlap); 540 pkt_len -= overlap; 541 rcvpartdupbyte += overlap; 542 TCP_REASS_COUNTER_INCR(&tcp_reass_overlapfront); 543 rcvoobyte -= overlap; 544 } 545 /* 546 * If the received segment immediates precedes this 547 * fragment then tack the fragment onto this segment 548 * and reinsert the data. 549 */ 550 if (q->ipqe_seq == pkt_seq + pkt_len) { 551 #ifdef TCPREASS_DEBUG 552 printf("tcp_reass[%p]: append %u:%u(%u) to %u:%u(%u)\n", 553 tp, q->ipqe_seq, q->ipqe_seq + q->ipqe_len, q->ipqe_len, 554 pkt_seq, pkt_seq + pkt_len, pkt_len); 555 #endif 556 pkt_len += q->ipqe_len; 557 pkt_flags |= q->ipqe_flags; 558 m_cat(m, q->ipqe_m); 559 TAILQ_REMOVE(&tp->segq, q, ipqe_q); 560 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 561 if (tiqe == NULL) { 562 tiqe = q; 563 } else { 564 pool_put(&ipqent_pool, q); 565 } 566 TCP_REASS_COUNTER_INCR(&tcp_reass_prepend); 567 break; 568 } 569 /* 570 * If the fragment is before the segment, remember it. 571 * When this loop is terminated, p will contain the 572 * pointer to fragment that is right before the received 573 * segment. 574 */ 575 if (SEQ_LEQ(q->ipqe_seq, pkt_seq)) 576 p = q; 577 578 continue; 579 580 /* 581 * This is a common operation. It also will allow 582 * to save doing a malloc/free in most instances. 583 */ 584 free_ipqe: 585 TAILQ_REMOVE(&tp->segq, q, ipqe_q); 586 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 587 if (tiqe == NULL) { 588 tiqe = q; 589 } else { 590 pool_put(&ipqent_pool, q); 591 } 592 } 593 594 #ifdef TCP_REASS_COUNTERS 595 if (count > 7) 596 TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[0]); 597 else if (count > 0) 598 TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[count]); 599 #endif 600 601 insert_it: 602 603 /* 604 * Allocate a new queue entry since the received segment did not 605 * collapse onto any other out-of-order block; thus we are allocating 606 * a new block. If it had collapsed, tiqe would not be NULL and 607 * we would be reusing it. 608 * XXX If we can't, just drop the packet. XXX 609 */ 610 if (tiqe == NULL) { 611 tiqe = pool_get(&ipqent_pool, PR_NOWAIT); 612 if (tiqe == NULL) { 613 tcpstat.tcps_rcvmemdrop++; 614 m_freem(m); 615 return (0); 616 } 617 } 618 619 /* 620 * Update the counters. 621 */ 622 tcpstat.tcps_rcvoopack++; 623 tcpstat.tcps_rcvoobyte += rcvoobyte; 624 if (rcvpartdupbyte) { 625 tcpstat.tcps_rcvpartduppack++; 626 tcpstat.tcps_rcvpartdupbyte += rcvpartdupbyte; 627 } 628 629 /* 630 * Insert the new fragment queue entry into both queues. 631 */ 632 tiqe->ipqe_m = m; 633 tiqe->ipqe_seq = pkt_seq; 634 tiqe->ipqe_len = pkt_len; 635 tiqe->ipqe_flags = pkt_flags; 636 if (p == NULL) { 637 TAILQ_INSERT_HEAD(&tp->segq, tiqe, ipqe_q); 638 #ifdef TCPREASS_DEBUG 639 if (tiqe->ipqe_seq != tp->rcv_nxt) 640 printf("tcp_reass[%p]: insert %u:%u(%u) at front\n", 641 tp, pkt_seq, pkt_seq + pkt_len, pkt_len); 642 #endif 643 } else { 644 TAILQ_INSERT_AFTER(&tp->segq, p, tiqe, ipqe_q); 645 #ifdef TCPREASS_DEBUG 646 printf("tcp_reass[%p]: insert %u:%u(%u) after %u:%u(%u)\n", 647 tp, pkt_seq, pkt_seq + pkt_len, pkt_len, 648 p->ipqe_seq, p->ipqe_seq + p->ipqe_len, p->ipqe_len); 649 #endif 650 } 651 652 skip_replacement: 653 654 TAILQ_INSERT_HEAD(&tp->timeq, tiqe, ipqe_timeq); 655 656 present: 657 /* 658 * Present data to user, advancing rcv_nxt through 659 * completed sequence space. 660 */ 661 if (TCPS_HAVEESTABLISHED(tp->t_state) == 0) 662 return (0); 663 q = TAILQ_FIRST(&tp->segq); 664 if (q == NULL || q->ipqe_seq != tp->rcv_nxt) 665 return (0); 666 if (tp->t_state == TCPS_SYN_RECEIVED && q->ipqe_len) 667 return (0); 668 669 tp->rcv_nxt += q->ipqe_len; 670 pkt_flags = q->ipqe_flags & TH_FIN; 671 ND6_HINT(tp); 672 673 TAILQ_REMOVE(&tp->segq, q, ipqe_q); 674 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 675 if (so->so_state & SS_CANTRCVMORE) 676 m_freem(q->ipqe_m); 677 else 678 sbappendstream(&so->so_rcv, q->ipqe_m); 679 pool_put(&ipqent_pool, q); 680 sorwakeup(so); 681 return (pkt_flags); 682 } 683 684 #ifdef INET6 685 int 686 tcp6_input(mp, offp, proto) 687 struct mbuf **mp; 688 int *offp, proto; 689 { 690 struct mbuf *m = *mp; 691 692 /* 693 * draft-itojun-ipv6-tcp-to-anycast 694 * better place to put this in? 695 */ 696 if (m->m_flags & M_ANYCAST6) { 697 struct ip6_hdr *ip6; 698 if (m->m_len < sizeof(struct ip6_hdr)) { 699 if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) { 700 tcpstat.tcps_rcvshort++; 701 return IPPROTO_DONE; 702 } 703 } 704 ip6 = mtod(m, struct ip6_hdr *); 705 icmp6_error(m, ICMP6_DST_UNREACH, ICMP6_DST_UNREACH_ADDR, 706 (caddr_t)&ip6->ip6_dst - (caddr_t)ip6); 707 return IPPROTO_DONE; 708 } 709 710 tcp_input(m, *offp, proto); 711 return IPPROTO_DONE; 712 } 713 #endif 714 715 #ifdef INET 716 static void 717 tcp4_log_refused(ip, th) 718 const struct ip *ip; 719 const struct tcphdr *th; 720 { 721 char src[4*sizeof "123"]; 722 char dst[4*sizeof "123"]; 723 724 if (ip) { 725 strcpy(src, inet_ntoa(ip->ip_src)); 726 strcpy(dst, inet_ntoa(ip->ip_dst)); 727 } 728 else { 729 strcpy(src, "(unknown)"); 730 strcpy(dst, "(unknown)"); 731 } 732 log(LOG_INFO, 733 "Connection attempt to TCP %s:%d from %s:%d\n", 734 dst, ntohs(th->th_dport), 735 src, ntohs(th->th_sport)); 736 } 737 #endif 738 739 #ifdef INET6 740 static void 741 tcp6_log_refused(ip6, th) 742 const struct ip6_hdr *ip6; 743 const struct tcphdr *th; 744 { 745 char src[INET6_ADDRSTRLEN]; 746 char dst[INET6_ADDRSTRLEN]; 747 748 if (ip6) { 749 strcpy(src, ip6_sprintf(&ip6->ip6_src)); 750 strcpy(dst, ip6_sprintf(&ip6->ip6_dst)); 751 } 752 else { 753 strcpy(src, "(unknown v6)"); 754 strcpy(dst, "(unknown v6)"); 755 } 756 log(LOG_INFO, 757 "Connection attempt to TCP [%s]:%d from [%s]:%d\n", 758 dst, ntohs(th->th_dport), 759 src, ntohs(th->th_sport)); 760 } 761 #endif 762 763 /* 764 * TCP input routine, follows pages 65-76 of the 765 * protocol specification dated September, 1981 very closely. 766 */ 767 void 768 #if __STDC__ 769 tcp_input(struct mbuf *m, ...) 770 #else 771 tcp_input(m, va_alist) 772 struct mbuf *m; 773 #endif 774 { 775 struct tcphdr *th; 776 struct ip *ip; 777 struct inpcb *inp; 778 #ifdef INET6 779 struct ip6_hdr *ip6; 780 struct in6pcb *in6p; 781 #endif 782 u_int8_t *optp = NULL; 783 int optlen = 0; 784 int len, tlen, toff, hdroptlen = 0; 785 struct tcpcb *tp = 0; 786 int tiflags; 787 struct socket *so = NULL; 788 int todrop, acked, ourfinisacked, needoutput = 0; 789 #ifdef TCP_DEBUG 790 short ostate = 0; 791 #endif 792 int iss = 0; 793 u_long tiwin; 794 struct tcp_opt_info opti; 795 int off, iphlen; 796 va_list ap; 797 int af; /* af on the wire */ 798 struct mbuf *tcp_saveti = NULL; 799 800 MCLAIM(m, &tcp_rx_mowner); 801 va_start(ap, m); 802 toff = va_arg(ap, int); 803 (void)va_arg(ap, int); /* ignore value, advance ap */ 804 va_end(ap); 805 806 tcpstat.tcps_rcvtotal++; 807 808 bzero(&opti, sizeof(opti)); 809 opti.ts_present = 0; 810 opti.maxseg = 0; 811 812 /* 813 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN. 814 * 815 * TCP is, by definition, unicast, so we reject all 816 * multicast outright. 817 * 818 * Note, there are additional src/dst address checks in 819 * the AF-specific code below. 820 */ 821 if (m->m_flags & (M_BCAST|M_MCAST)) { 822 /* XXX stat */ 823 goto drop; 824 } 825 #ifdef INET6 826 if (m->m_flags & M_ANYCAST6) { 827 /* XXX stat */ 828 goto drop; 829 } 830 #endif 831 832 /* 833 * Get IP and TCP header together in first mbuf. 834 * Note: IP leaves IP header in first mbuf. 835 */ 836 ip = mtod(m, struct ip *); 837 #ifdef INET6 838 ip6 = NULL; 839 #endif 840 switch (ip->ip_v) { 841 #ifdef INET 842 case 4: 843 af = AF_INET; 844 iphlen = sizeof(struct ip); 845 #ifndef PULLDOWN_TEST 846 /* would like to get rid of this... */ 847 if (toff > sizeof (struct ip)) { 848 ip_stripoptions(m, (struct mbuf *)0); 849 toff = sizeof(struct ip); 850 } 851 if (m->m_len < toff + sizeof (struct tcphdr)) { 852 if ((m = m_pullup(m, toff + sizeof (struct tcphdr))) == 0) { 853 tcpstat.tcps_rcvshort++; 854 return; 855 } 856 } 857 ip = mtod(m, struct ip *); 858 th = (struct tcphdr *)(mtod(m, caddr_t) + toff); 859 #else 860 ip = mtod(m, struct ip *); 861 IP6_EXTHDR_GET(th, struct tcphdr *, m, toff, 862 sizeof(struct tcphdr)); 863 if (th == NULL) { 864 tcpstat.tcps_rcvshort++; 865 return; 866 } 867 #endif 868 /* We do the checksum after PCB lookup... */ 869 len = ntohs(ip->ip_len); 870 tlen = len - toff; 871 break; 872 #endif 873 #ifdef INET6 874 case 6: 875 ip = NULL; 876 iphlen = sizeof(struct ip6_hdr); 877 af = AF_INET6; 878 #ifndef PULLDOWN_TEST 879 if (m->m_len < toff + sizeof(struct tcphdr)) { 880 m = m_pullup(m, toff + sizeof(struct tcphdr)); /*XXX*/ 881 if (m == NULL) { 882 tcpstat.tcps_rcvshort++; 883 return; 884 } 885 } 886 ip6 = mtod(m, struct ip6_hdr *); 887 th = (struct tcphdr *)(mtod(m, caddr_t) + toff); 888 #else 889 ip6 = mtod(m, struct ip6_hdr *); 890 IP6_EXTHDR_GET(th, struct tcphdr *, m, toff, 891 sizeof(struct tcphdr)); 892 if (th == NULL) { 893 tcpstat.tcps_rcvshort++; 894 return; 895 } 896 #endif 897 898 /* Be proactive about malicious use of IPv4 mapped address */ 899 if (IN6_IS_ADDR_V4MAPPED(&ip6->ip6_src) || 900 IN6_IS_ADDR_V4MAPPED(&ip6->ip6_dst)) { 901 /* XXX stat */ 902 goto drop; 903 } 904 905 /* 906 * Be proactive about unspecified IPv6 address in source. 907 * As we use all-zero to indicate unbounded/unconnected pcb, 908 * unspecified IPv6 address can be used to confuse us. 909 * 910 * Note that packets with unspecified IPv6 destination is 911 * already dropped in ip6_input. 912 */ 913 if (IN6_IS_ADDR_UNSPECIFIED(&ip6->ip6_src)) { 914 /* XXX stat */ 915 goto drop; 916 } 917 918 /* 919 * Make sure destination address is not multicast. 920 * Source address checked in ip6_input(). 921 */ 922 if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) { 923 /* XXX stat */ 924 goto drop; 925 } 926 927 /* We do the checksum after PCB lookup... */ 928 len = m->m_pkthdr.len; 929 tlen = len - toff; 930 break; 931 #endif 932 default: 933 m_freem(m); 934 return; 935 } 936 937 KASSERT(TCP_HDR_ALIGNED_P(th)); 938 939 /* 940 * Check that TCP offset makes sense, 941 * pull out TCP options and adjust length. XXX 942 */ 943 off = th->th_off << 2; 944 if (off < sizeof (struct tcphdr) || off > tlen) { 945 tcpstat.tcps_rcvbadoff++; 946 goto drop; 947 } 948 tlen -= off; 949 950 /* 951 * tcp_input() has been modified to use tlen to mean the TCP data 952 * length throughout the function. Other functions can use 953 * m->m_pkthdr.len as the basis for calculating the TCP data length. 954 * rja 955 */ 956 957 if (off > sizeof (struct tcphdr)) { 958 #ifndef PULLDOWN_TEST 959 if (m->m_len < toff + off) { 960 if ((m = m_pullup(m, toff + off)) == 0) { 961 tcpstat.tcps_rcvshort++; 962 return; 963 } 964 switch (af) { 965 #ifdef INET 966 case AF_INET: 967 ip = mtod(m, struct ip *); 968 break; 969 #endif 970 #ifdef INET6 971 case AF_INET6: 972 ip6 = mtod(m, struct ip6_hdr *); 973 break; 974 #endif 975 } 976 th = (struct tcphdr *)(mtod(m, caddr_t) + toff); 977 } 978 #else 979 IP6_EXTHDR_GET(th, struct tcphdr *, m, toff, off); 980 if (th == NULL) { 981 tcpstat.tcps_rcvshort++; 982 return; 983 } 984 /* 985 * NOTE: ip/ip6 will not be affected by m_pulldown() 986 * (as they're before toff) and we don't need to update those. 987 */ 988 #endif 989 KASSERT(TCP_HDR_ALIGNED_P(th)); 990 optlen = off - sizeof (struct tcphdr); 991 optp = ((u_int8_t *)th) + sizeof(struct tcphdr); 992 /* 993 * Do quick retrieval of timestamp options ("options 994 * prediction?"). If timestamp is the only option and it's 995 * formatted as recommended in RFC 1323 appendix A, we 996 * quickly get the values now and not bother calling 997 * tcp_dooptions(), etc. 998 */ 999 if ((optlen == TCPOLEN_TSTAMP_APPA || 1000 (optlen > TCPOLEN_TSTAMP_APPA && 1001 optp[TCPOLEN_TSTAMP_APPA] == TCPOPT_EOL)) && 1002 *(u_int32_t *)optp == htonl(TCPOPT_TSTAMP_HDR) && 1003 (th->th_flags & TH_SYN) == 0) { 1004 opti.ts_present = 1; 1005 opti.ts_val = ntohl(*(u_int32_t *)(optp + 4)); 1006 opti.ts_ecr = ntohl(*(u_int32_t *)(optp + 8)); 1007 optp = NULL; /* we've parsed the options */ 1008 } 1009 } 1010 tiflags = th->th_flags; 1011 1012 /* 1013 * Locate pcb for segment. 1014 */ 1015 findpcb: 1016 inp = NULL; 1017 #ifdef INET6 1018 in6p = NULL; 1019 #endif 1020 switch (af) { 1021 #ifdef INET 1022 case AF_INET: 1023 inp = in_pcblookup_connect(&tcbtable, ip->ip_src, th->th_sport, 1024 ip->ip_dst, th->th_dport); 1025 if (inp == 0) { 1026 ++tcpstat.tcps_pcbhashmiss; 1027 inp = in_pcblookup_bind(&tcbtable, ip->ip_dst, th->th_dport); 1028 } 1029 #ifdef INET6 1030 if (inp == 0) { 1031 struct in6_addr s, d; 1032 1033 /* mapped addr case */ 1034 bzero(&s, sizeof(s)); 1035 s.s6_addr16[5] = htons(0xffff); 1036 bcopy(&ip->ip_src, &s.s6_addr32[3], sizeof(ip->ip_src)); 1037 bzero(&d, sizeof(d)); 1038 d.s6_addr16[5] = htons(0xffff); 1039 bcopy(&ip->ip_dst, &d.s6_addr32[3], sizeof(ip->ip_dst)); 1040 in6p = in6_pcblookup_connect(&tcb6, &s, th->th_sport, 1041 &d, th->th_dport, 0); 1042 if (in6p == 0) { 1043 ++tcpstat.tcps_pcbhashmiss; 1044 in6p = in6_pcblookup_bind(&tcb6, &d, 1045 th->th_dport, 0); 1046 } 1047 } 1048 #endif 1049 #ifndef INET6 1050 if (inp == 0) 1051 #else 1052 if (inp == 0 && in6p == 0) 1053 #endif 1054 { 1055 ++tcpstat.tcps_noport; 1056 if (tcp_log_refused && 1057 (tiflags & (TH_RST|TH_ACK|TH_SYN)) == TH_SYN) { 1058 tcp4_log_refused(ip, th); 1059 } 1060 TCP_FIELDS_TO_HOST(th); 1061 goto dropwithreset_ratelim; 1062 } 1063 #ifdef IPSEC 1064 if (inp && ipsec4_in_reject(m, inp)) { 1065 ipsecstat.in_polvio++; 1066 goto drop; 1067 } 1068 #ifdef INET6 1069 else if (in6p && ipsec4_in_reject_so(m, in6p->in6p_socket)) { 1070 ipsecstat.in_polvio++; 1071 goto drop; 1072 } 1073 #endif 1074 #endif /*IPSEC*/ 1075 break; 1076 #endif /*INET*/ 1077 #ifdef INET6 1078 case AF_INET6: 1079 { 1080 int faith; 1081 1082 #if defined(NFAITH) && NFAITH > 0 1083 faith = faithprefix(&ip6->ip6_dst); 1084 #else 1085 faith = 0; 1086 #endif 1087 in6p = in6_pcblookup_connect(&tcb6, &ip6->ip6_src, th->th_sport, 1088 &ip6->ip6_dst, th->th_dport, faith); 1089 if (in6p == NULL) { 1090 ++tcpstat.tcps_pcbhashmiss; 1091 in6p = in6_pcblookup_bind(&tcb6, &ip6->ip6_dst, 1092 th->th_dport, faith); 1093 } 1094 if (in6p == NULL) { 1095 ++tcpstat.tcps_noport; 1096 if (tcp_log_refused && 1097 (tiflags & (TH_RST|TH_ACK|TH_SYN)) == TH_SYN) { 1098 tcp6_log_refused(ip6, th); 1099 } 1100 TCP_FIELDS_TO_HOST(th); 1101 goto dropwithreset_ratelim; 1102 } 1103 #ifdef IPSEC 1104 if (ipsec6_in_reject(m, in6p)) { 1105 ipsec6stat.in_polvio++; 1106 goto drop; 1107 } 1108 #endif /*IPSEC*/ 1109 break; 1110 } 1111 #endif 1112 } 1113 1114 /* 1115 * If the state is CLOSED (i.e., TCB does not exist) then 1116 * all data in the incoming segment is discarded. 1117 * If the TCB exists but is in CLOSED state, it is embryonic, 1118 * but should either do a listen or a connect soon. 1119 */ 1120 tp = NULL; 1121 so = NULL; 1122 if (inp) { 1123 tp = intotcpcb(inp); 1124 so = inp->inp_socket; 1125 } 1126 #ifdef INET6 1127 else if (in6p) { 1128 tp = in6totcpcb(in6p); 1129 so = in6p->in6p_socket; 1130 } 1131 #endif 1132 if (tp == 0) { 1133 TCP_FIELDS_TO_HOST(th); 1134 goto dropwithreset_ratelim; 1135 } 1136 if (tp->t_state == TCPS_CLOSED) 1137 goto drop; 1138 1139 /* 1140 * Checksum extended TCP header and data. 1141 */ 1142 switch (af) { 1143 #ifdef INET 1144 case AF_INET: 1145 switch (m->m_pkthdr.csum_flags & 1146 ((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_TCPv4) | 1147 M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) { 1148 case M_CSUM_TCPv4|M_CSUM_TCP_UDP_BAD: 1149 TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_bad); 1150 goto badcsum; 1151 1152 case M_CSUM_TCPv4|M_CSUM_DATA: 1153 TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_data); 1154 if ((m->m_pkthdr.csum_data ^ 0xffff) != 0) 1155 goto badcsum; 1156 break; 1157 1158 case M_CSUM_TCPv4: 1159 /* Checksum was okay. */ 1160 TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_ok); 1161 break; 1162 1163 default: 1164 /* Must compute it ourselves. */ 1165 TCP_CSUM_COUNTER_INCR(&tcp_swcsum); 1166 #ifndef PULLDOWN_TEST 1167 { 1168 struct ipovly *ipov; 1169 ipov = (struct ipovly *)ip; 1170 bzero(ipov->ih_x1, sizeof ipov->ih_x1); 1171 ipov->ih_len = htons(tlen + off); 1172 1173 if (in_cksum(m, len) != 0) 1174 goto badcsum; 1175 } 1176 #else 1177 if (in4_cksum(m, IPPROTO_TCP, toff, tlen + off) != 0) 1178 goto badcsum; 1179 #endif /* ! PULLDOWN_TEST */ 1180 break; 1181 } 1182 break; 1183 #endif /* INET4 */ 1184 1185 #ifdef INET6 1186 case AF_INET6: 1187 if (in6_cksum(m, IPPROTO_TCP, toff, tlen + off) != 0) 1188 goto badcsum; 1189 break; 1190 #endif /* INET6 */ 1191 } 1192 1193 TCP_FIELDS_TO_HOST(th); 1194 1195 /* Unscale the window into a 32-bit value. */ 1196 if ((tiflags & TH_SYN) == 0) 1197 tiwin = th->th_win << tp->snd_scale; 1198 else 1199 tiwin = th->th_win; 1200 1201 #ifdef INET6 1202 /* save packet options if user wanted */ 1203 if (in6p && (in6p->in6p_flags & IN6P_CONTROLOPTS)) { 1204 if (in6p->in6p_options) { 1205 m_freem(in6p->in6p_options); 1206 in6p->in6p_options = 0; 1207 } 1208 ip6_savecontrol(in6p, &in6p->in6p_options, ip6, m); 1209 } 1210 #endif 1211 1212 if (so->so_options & (SO_DEBUG|SO_ACCEPTCONN)) { 1213 union syn_cache_sa src; 1214 union syn_cache_sa dst; 1215 1216 bzero(&src, sizeof(src)); 1217 bzero(&dst, sizeof(dst)); 1218 switch (af) { 1219 #ifdef INET 1220 case AF_INET: 1221 src.sin.sin_len = sizeof(struct sockaddr_in); 1222 src.sin.sin_family = AF_INET; 1223 src.sin.sin_addr = ip->ip_src; 1224 src.sin.sin_port = th->th_sport; 1225 1226 dst.sin.sin_len = sizeof(struct sockaddr_in); 1227 dst.sin.sin_family = AF_INET; 1228 dst.sin.sin_addr = ip->ip_dst; 1229 dst.sin.sin_port = th->th_dport; 1230 break; 1231 #endif 1232 #ifdef INET6 1233 case AF_INET6: 1234 src.sin6.sin6_len = sizeof(struct sockaddr_in6); 1235 src.sin6.sin6_family = AF_INET6; 1236 src.sin6.sin6_addr = ip6->ip6_src; 1237 src.sin6.sin6_port = th->th_sport; 1238 1239 dst.sin6.sin6_len = sizeof(struct sockaddr_in6); 1240 dst.sin6.sin6_family = AF_INET6; 1241 dst.sin6.sin6_addr = ip6->ip6_dst; 1242 dst.sin6.sin6_port = th->th_dport; 1243 break; 1244 #endif /* INET6 */ 1245 default: 1246 goto badsyn; /*sanity*/ 1247 } 1248 1249 if (so->so_options & SO_DEBUG) { 1250 #ifdef TCP_DEBUG 1251 ostate = tp->t_state; 1252 #endif 1253 1254 tcp_saveti = NULL; 1255 if (iphlen + sizeof(struct tcphdr) > MHLEN) 1256 goto nosave; 1257 1258 if (m->m_len > iphlen && (m->m_flags & M_EXT) == 0) { 1259 tcp_saveti = m_copym(m, 0, iphlen, M_DONTWAIT); 1260 if (!tcp_saveti) 1261 goto nosave; 1262 } else { 1263 MGETHDR(tcp_saveti, M_DONTWAIT, MT_HEADER); 1264 if (!tcp_saveti) 1265 goto nosave; 1266 MCLAIM(m, &tcp_mowner); 1267 tcp_saveti->m_len = iphlen; 1268 m_copydata(m, 0, iphlen, 1269 mtod(tcp_saveti, caddr_t)); 1270 } 1271 1272 if (M_TRAILINGSPACE(tcp_saveti) < sizeof(struct tcphdr)) { 1273 m_freem(tcp_saveti); 1274 tcp_saveti = NULL; 1275 } else { 1276 tcp_saveti->m_len += sizeof(struct tcphdr); 1277 bcopy(th, mtod(tcp_saveti, caddr_t) + iphlen, 1278 sizeof(struct tcphdr)); 1279 } 1280 if (tcp_saveti) { 1281 /* 1282 * need to recover version # field, which was 1283 * overwritten on ip_cksum computation. 1284 */ 1285 struct ip *sip; 1286 sip = mtod(tcp_saveti, struct ip *); 1287 switch (af) { 1288 #ifdef INET 1289 case AF_INET: 1290 sip->ip_v = 4; 1291 break; 1292 #endif 1293 #ifdef INET6 1294 case AF_INET6: 1295 sip->ip_v = 6; 1296 break; 1297 #endif 1298 } 1299 } 1300 nosave:; 1301 } 1302 if (so->so_options & SO_ACCEPTCONN) { 1303 if ((tiflags & (TH_RST|TH_ACK|TH_SYN)) != TH_SYN) { 1304 if (tiflags & TH_RST) { 1305 syn_cache_reset(&src.sa, &dst.sa, th); 1306 } else if ((tiflags & (TH_ACK|TH_SYN)) == 1307 (TH_ACK|TH_SYN)) { 1308 /* 1309 * Received a SYN,ACK. This should 1310 * never happen while we are in 1311 * LISTEN. Send an RST. 1312 */ 1313 goto badsyn; 1314 } else if (tiflags & TH_ACK) { 1315 so = syn_cache_get(&src.sa, &dst.sa, 1316 th, toff, tlen, so, m); 1317 if (so == NULL) { 1318 /* 1319 * We don't have a SYN for 1320 * this ACK; send an RST. 1321 */ 1322 goto badsyn; 1323 } else if (so == 1324 (struct socket *)(-1)) { 1325 /* 1326 * We were unable to create 1327 * the connection. If the 1328 * 3-way handshake was 1329 * completed, and RST has 1330 * been sent to the peer. 1331 * Since the mbuf might be 1332 * in use for the reply, 1333 * do not free it. 1334 */ 1335 m = NULL; 1336 } else { 1337 /* 1338 * We have created a 1339 * full-blown connection. 1340 */ 1341 tp = NULL; 1342 inp = NULL; 1343 #ifdef INET6 1344 in6p = NULL; 1345 #endif 1346 switch (so->so_proto->pr_domain->dom_family) { 1347 #ifdef INET 1348 case AF_INET: 1349 inp = sotoinpcb(so); 1350 tp = intotcpcb(inp); 1351 break; 1352 #endif 1353 #ifdef INET6 1354 case AF_INET6: 1355 in6p = sotoin6pcb(so); 1356 tp = in6totcpcb(in6p); 1357 break; 1358 #endif 1359 } 1360 if (tp == NULL) 1361 goto badsyn; /*XXX*/ 1362 tiwin <<= tp->snd_scale; 1363 goto after_listen; 1364 } 1365 } else { 1366 /* 1367 * None of RST, SYN or ACK was set. 1368 * This is an invalid packet for a 1369 * TCB in LISTEN state. Send a RST. 1370 */ 1371 goto badsyn; 1372 } 1373 } else { 1374 /* 1375 * Received a SYN. 1376 */ 1377 1378 #ifdef INET6 1379 /* 1380 * If deprecated address is forbidden, we do 1381 * not accept SYN to deprecated interface 1382 * address to prevent any new inbound 1383 * connection from getting established. 1384 * When we do not accept SYN, we send a TCP 1385 * RST, with deprecated source address (instead 1386 * of dropping it). We compromise it as it is 1387 * much better for peer to send a RST, and 1388 * RST will be the final packet for the 1389 * exchange. 1390 * 1391 * If we do not forbid deprecated addresses, we 1392 * accept the SYN packet. RFC2462 does not 1393 * suggest dropping SYN in this case. 1394 * If we decipher RFC2462 5.5.4, it says like 1395 * this: 1396 * 1. use of deprecated addr with existing 1397 * communication is okay - "SHOULD continue 1398 * to be used" 1399 * 2. use of it with new communication: 1400 * (2a) "SHOULD NOT be used if alternate 1401 * address with sufficient scope is 1402 * available" 1403 * (2b) nothing mentioned otherwise. 1404 * Here we fall into (2b) case as we have no 1405 * choice in our source address selection - we 1406 * must obey the peer. 1407 * 1408 * The wording in RFC2462 is confusing, and 1409 * there are multiple description text for 1410 * deprecated address handling - worse, they 1411 * are not exactly the same. I believe 5.5.4 1412 * is the best one, so we follow 5.5.4. 1413 */ 1414 if (af == AF_INET6 && !ip6_use_deprecated) { 1415 struct in6_ifaddr *ia6; 1416 if ((ia6 = in6ifa_ifpwithaddr(m->m_pkthdr.rcvif, 1417 &ip6->ip6_dst)) && 1418 (ia6->ia6_flags & IN6_IFF_DEPRECATED)) { 1419 tp = NULL; 1420 goto dropwithreset; 1421 } 1422 } 1423 #endif 1424 1425 /* 1426 * LISTEN socket received a SYN 1427 * from itself? This can't possibly 1428 * be valid; drop the packet. 1429 */ 1430 if (th->th_sport == th->th_dport) { 1431 int i; 1432 1433 switch (af) { 1434 #ifdef INET 1435 case AF_INET: 1436 i = in_hosteq(ip->ip_src, ip->ip_dst); 1437 break; 1438 #endif 1439 #ifdef INET6 1440 case AF_INET6: 1441 i = IN6_ARE_ADDR_EQUAL(&ip6->ip6_src, &ip6->ip6_dst); 1442 break; 1443 #endif 1444 default: 1445 i = 1; 1446 } 1447 if (i) { 1448 tcpstat.tcps_badsyn++; 1449 goto drop; 1450 } 1451 } 1452 1453 /* 1454 * SYN looks ok; create compressed TCP 1455 * state for it. 1456 */ 1457 if (so->so_qlen <= so->so_qlimit && 1458 syn_cache_add(&src.sa, &dst.sa, th, tlen, 1459 so, m, optp, optlen, &opti)) 1460 m = NULL; 1461 } 1462 goto drop; 1463 } 1464 } 1465 1466 after_listen: 1467 #ifdef DIAGNOSTIC 1468 /* 1469 * Should not happen now that all embryonic connections 1470 * are handled with compressed state. 1471 */ 1472 if (tp->t_state == TCPS_LISTEN) 1473 panic("tcp_input: TCPS_LISTEN"); 1474 #endif 1475 1476 /* 1477 * Segment received on connection. 1478 * Reset idle time and keep-alive timer. 1479 */ 1480 tp->t_rcvtime = tcp_now; 1481 if (TCPS_HAVEESTABLISHED(tp->t_state)) 1482 TCP_TIMER_ARM(tp, TCPT_KEEP, tcp_keepidle); 1483 1484 /* 1485 * Process options. 1486 */ 1487 if (optp) 1488 tcp_dooptions(tp, optp, optlen, th, &opti); 1489 1490 /* 1491 * Header prediction: check for the two common cases 1492 * of a uni-directional data xfer. If the packet has 1493 * no control flags, is in-sequence, the window didn't 1494 * change and we're not retransmitting, it's a 1495 * candidate. If the length is zero and the ack moved 1496 * forward, we're the sender side of the xfer. Just 1497 * free the data acked & wake any higher level process 1498 * that was blocked waiting for space. If the length 1499 * is non-zero and the ack didn't move, we're the 1500 * receiver side. If we're getting packets in-order 1501 * (the reassembly queue is empty), add the data to 1502 * the socket buffer and note that we need a delayed ack. 1503 */ 1504 if (tp->t_state == TCPS_ESTABLISHED && 1505 (tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ACK)) == TH_ACK && 1506 (!opti.ts_present || TSTMP_GEQ(opti.ts_val, tp->ts_recent)) && 1507 th->th_seq == tp->rcv_nxt && 1508 tiwin && tiwin == tp->snd_wnd && 1509 tp->snd_nxt == tp->snd_max) { 1510 1511 /* 1512 * If last ACK falls within this segment's sequence numbers, 1513 * record the timestamp. 1514 */ 1515 if (opti.ts_present && 1516 SEQ_LEQ(th->th_seq, tp->last_ack_sent) && 1517 SEQ_LT(tp->last_ack_sent, th->th_seq + tlen)) { 1518 tp->ts_recent_age = TCP_TIMESTAMP(tp); 1519 tp->ts_recent = opti.ts_val; 1520 } 1521 1522 if (tlen == 0) { 1523 if (SEQ_GT(th->th_ack, tp->snd_una) && 1524 SEQ_LEQ(th->th_ack, tp->snd_max) && 1525 tp->snd_cwnd >= tp->snd_wnd && 1526 tp->t_dupacks < tcprexmtthresh) { 1527 /* 1528 * this is a pure ack for outstanding data. 1529 */ 1530 ++tcpstat.tcps_predack; 1531 if (opti.ts_present && opti.ts_ecr) 1532 tcp_xmit_timer(tp, 1533 TCP_TIMESTAMP(tp) - opti.ts_ecr + 1); 1534 else if (tp->t_rtttime && 1535 SEQ_GT(th->th_ack, tp->t_rtseq)) 1536 tcp_xmit_timer(tp, 1537 tcp_now - tp->t_rtttime); 1538 acked = th->th_ack - tp->snd_una; 1539 tcpstat.tcps_rcvackpack++; 1540 tcpstat.tcps_rcvackbyte += acked; 1541 ND6_HINT(tp); 1542 sbdrop(&so->so_snd, acked); 1543 /* 1544 * We want snd_recover to track snd_una to 1545 * avoid sequence wraparound problems for 1546 * very large transfers. 1547 */ 1548 tp->snd_una = tp->snd_recover = th->th_ack; 1549 m_freem(m); 1550 1551 /* 1552 * If all outstanding data are acked, stop 1553 * retransmit timer, otherwise restart timer 1554 * using current (possibly backed-off) value. 1555 * If process is waiting for space, 1556 * wakeup/selwakeup/signal. If data 1557 * are ready to send, let tcp_output 1558 * decide between more output or persist. 1559 */ 1560 if (tp->snd_una == tp->snd_max) 1561 TCP_TIMER_DISARM(tp, TCPT_REXMT); 1562 else if (TCP_TIMER_ISARMED(tp, 1563 TCPT_PERSIST) == 0) 1564 TCP_TIMER_ARM(tp, TCPT_REXMT, 1565 tp->t_rxtcur); 1566 1567 sowwakeup(so); 1568 if (so->so_snd.sb_cc) 1569 (void) tcp_output(tp); 1570 if (tcp_saveti) 1571 m_freem(tcp_saveti); 1572 return; 1573 } 1574 } else if (th->th_ack == tp->snd_una && 1575 TAILQ_FIRST(&tp->segq) == NULL && 1576 tlen <= sbspace(&so->so_rcv)) { 1577 /* 1578 * this is a pure, in-sequence data packet 1579 * with nothing on the reassembly queue and 1580 * we have enough buffer space to take it. 1581 */ 1582 ++tcpstat.tcps_preddat; 1583 tp->rcv_nxt += tlen; 1584 tcpstat.tcps_rcvpack++; 1585 tcpstat.tcps_rcvbyte += tlen; 1586 ND6_HINT(tp); 1587 /* 1588 * Drop TCP, IP headers and TCP options then add data 1589 * to socket buffer. 1590 */ 1591 if (so->so_state & SS_CANTRCVMORE) 1592 m_freem(m); 1593 else { 1594 m_adj(m, toff + off); 1595 sbappendstream(&so->so_rcv, m); 1596 } 1597 sorwakeup(so); 1598 TCP_SETUP_ACK(tp, th); 1599 if (tp->t_flags & TF_ACKNOW) 1600 (void) tcp_output(tp); 1601 if (tcp_saveti) 1602 m_freem(tcp_saveti); 1603 return; 1604 } 1605 } 1606 1607 /* 1608 * Compute mbuf offset to TCP data segment. 1609 */ 1610 hdroptlen = toff + off; 1611 1612 /* 1613 * Calculate amount of space in receive window, 1614 * and then do TCP input processing. 1615 * Receive window is amount of space in rcv queue, 1616 * but not less than advertised window. 1617 */ 1618 { int win; 1619 1620 win = sbspace(&so->so_rcv); 1621 if (win < 0) 1622 win = 0; 1623 tp->rcv_wnd = imax(win, (int)(tp->rcv_adv - tp->rcv_nxt)); 1624 } 1625 1626 switch (tp->t_state) { 1627 case TCPS_LISTEN: 1628 /* 1629 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN 1630 */ 1631 if (m->m_flags & (M_BCAST|M_MCAST)) 1632 goto drop; 1633 switch (af) { 1634 #ifdef INET6 1635 case AF_INET6: 1636 if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) 1637 goto drop; 1638 break; 1639 #endif /* INET6 */ 1640 case AF_INET: 1641 if (IN_MULTICAST(ip->ip_dst.s_addr) || 1642 in_broadcast(ip->ip_dst, m->m_pkthdr.rcvif)) 1643 goto drop; 1644 break; 1645 } 1646 break; 1647 1648 /* 1649 * If the state is SYN_SENT: 1650 * if seg contains an ACK, but not for our SYN, drop the input. 1651 * if seg contains a RST, then drop the connection. 1652 * if seg does not contain SYN, then drop it. 1653 * Otherwise this is an acceptable SYN segment 1654 * initialize tp->rcv_nxt and tp->irs 1655 * if seg contains ack then advance tp->snd_una 1656 * if SYN has been acked change to ESTABLISHED else SYN_RCVD state 1657 * arrange for segment to be acked (eventually) 1658 * continue processing rest of data/controls, beginning with URG 1659 */ 1660 case TCPS_SYN_SENT: 1661 if ((tiflags & TH_ACK) && 1662 (SEQ_LEQ(th->th_ack, tp->iss) || 1663 SEQ_GT(th->th_ack, tp->snd_max))) 1664 goto dropwithreset; 1665 if (tiflags & TH_RST) { 1666 if (tiflags & TH_ACK) 1667 tp = tcp_drop(tp, ECONNREFUSED); 1668 goto drop; 1669 } 1670 if ((tiflags & TH_SYN) == 0) 1671 goto drop; 1672 if (tiflags & TH_ACK) { 1673 tp->snd_una = tp->snd_recover = th->th_ack; 1674 if (SEQ_LT(tp->snd_nxt, tp->snd_una)) 1675 tp->snd_nxt = tp->snd_una; 1676 TCP_TIMER_DISARM(tp, TCPT_REXMT); 1677 } 1678 tp->irs = th->th_seq; 1679 tcp_rcvseqinit(tp); 1680 tp->t_flags |= TF_ACKNOW; 1681 tcp_mss_from_peer(tp, opti.maxseg); 1682 1683 /* 1684 * Initialize the initial congestion window. If we 1685 * had to retransmit the SYN, we must initialize cwnd 1686 * to 1 segment (i.e. the Loss Window). 1687 */ 1688 if (tp->t_flags & TF_SYN_REXMT) 1689 tp->snd_cwnd = tp->t_peermss; 1690 else { 1691 int ss = tcp_init_win; 1692 #ifdef INET 1693 if (inp != NULL && in_localaddr(inp->inp_faddr)) 1694 ss = tcp_init_win_local; 1695 #endif 1696 #ifdef INET6 1697 if (in6p != NULL && in6_localaddr(&in6p->in6p_faddr)) 1698 ss = tcp_init_win_local; 1699 #endif 1700 tp->snd_cwnd = TCP_INITIAL_WINDOW(ss, tp->t_peermss); 1701 } 1702 1703 tcp_rmx_rtt(tp); 1704 if (tiflags & TH_ACK) { 1705 tcpstat.tcps_connects++; 1706 soisconnected(so); 1707 tcp_established(tp); 1708 /* Do window scaling on this connection? */ 1709 if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) == 1710 (TF_RCVD_SCALE|TF_REQ_SCALE)) { 1711 tp->snd_scale = tp->requested_s_scale; 1712 tp->rcv_scale = tp->request_r_scale; 1713 } 1714 TCP_REASS_LOCK(tp); 1715 (void) tcp_reass(tp, NULL, (struct mbuf *)0, &tlen); 1716 TCP_REASS_UNLOCK(tp); 1717 /* 1718 * if we didn't have to retransmit the SYN, 1719 * use its rtt as our initial srtt & rtt var. 1720 */ 1721 if (tp->t_rtttime) 1722 tcp_xmit_timer(tp, tcp_now - tp->t_rtttime); 1723 } else 1724 tp->t_state = TCPS_SYN_RECEIVED; 1725 1726 /* 1727 * Advance th->th_seq to correspond to first data byte. 1728 * If data, trim to stay within window, 1729 * dropping FIN if necessary. 1730 */ 1731 th->th_seq++; 1732 if (tlen > tp->rcv_wnd) { 1733 todrop = tlen - tp->rcv_wnd; 1734 m_adj(m, -todrop); 1735 tlen = tp->rcv_wnd; 1736 tiflags &= ~TH_FIN; 1737 tcpstat.tcps_rcvpackafterwin++; 1738 tcpstat.tcps_rcvbyteafterwin += todrop; 1739 } 1740 tp->snd_wl1 = th->th_seq - 1; 1741 tp->rcv_up = th->th_seq; 1742 goto step6; 1743 1744 /* 1745 * If the state is SYN_RECEIVED: 1746 * If seg contains an ACK, but not for our SYN, drop the input 1747 * and generate an RST. See page 36, rfc793 1748 */ 1749 case TCPS_SYN_RECEIVED: 1750 if ((tiflags & TH_ACK) && 1751 (SEQ_LEQ(th->th_ack, tp->iss) || 1752 SEQ_GT(th->th_ack, tp->snd_max))) 1753 goto dropwithreset; 1754 break; 1755 } 1756 1757 /* 1758 * States other than LISTEN or SYN_SENT. 1759 * First check timestamp, if present. 1760 * Then check that at least some bytes of segment are within 1761 * receive window. If segment begins before rcv_nxt, 1762 * drop leading data (and SYN); if nothing left, just ack. 1763 * 1764 * RFC 1323 PAWS: If we have a timestamp reply on this segment 1765 * and it's less than ts_recent, drop it. 1766 */ 1767 if (opti.ts_present && (tiflags & TH_RST) == 0 && tp->ts_recent && 1768 TSTMP_LT(opti.ts_val, tp->ts_recent)) { 1769 1770 /* Check to see if ts_recent is over 24 days old. */ 1771 if ((int)(TCP_TIMESTAMP(tp) - tp->ts_recent_age) > 1772 TCP_PAWS_IDLE) { 1773 /* 1774 * Invalidate ts_recent. If this segment updates 1775 * ts_recent, the age will be reset later and ts_recent 1776 * will get a valid value. If it does not, setting 1777 * ts_recent to zero will at least satisfy the 1778 * requirement that zero be placed in the timestamp 1779 * echo reply when ts_recent isn't valid. The 1780 * age isn't reset until we get a valid ts_recent 1781 * because we don't want out-of-order segments to be 1782 * dropped when ts_recent is old. 1783 */ 1784 tp->ts_recent = 0; 1785 } else { 1786 tcpstat.tcps_rcvduppack++; 1787 tcpstat.tcps_rcvdupbyte += tlen; 1788 tcpstat.tcps_pawsdrop++; 1789 goto dropafterack; 1790 } 1791 } 1792 1793 todrop = tp->rcv_nxt - th->th_seq; 1794 if (todrop > 0) { 1795 if (tiflags & TH_SYN) { 1796 tiflags &= ~TH_SYN; 1797 th->th_seq++; 1798 if (th->th_urp > 1) 1799 th->th_urp--; 1800 else { 1801 tiflags &= ~TH_URG; 1802 th->th_urp = 0; 1803 } 1804 todrop--; 1805 } 1806 if (todrop > tlen || 1807 (todrop == tlen && (tiflags & TH_FIN) == 0)) { 1808 /* 1809 * Any valid FIN must be to the left of the window. 1810 * At this point the FIN must be a duplicate or 1811 * out of sequence; drop it. 1812 */ 1813 tiflags &= ~TH_FIN; 1814 /* 1815 * Send an ACK to resynchronize and drop any data. 1816 * But keep on processing for RST or ACK. 1817 */ 1818 tp->t_flags |= TF_ACKNOW; 1819 todrop = tlen; 1820 tcpstat.tcps_rcvdupbyte += todrop; 1821 tcpstat.tcps_rcvduppack++; 1822 } else { 1823 tcpstat.tcps_rcvpartduppack++; 1824 tcpstat.tcps_rcvpartdupbyte += todrop; 1825 } 1826 hdroptlen += todrop; /*drop from head afterwards*/ 1827 th->th_seq += todrop; 1828 tlen -= todrop; 1829 if (th->th_urp > todrop) 1830 th->th_urp -= todrop; 1831 else { 1832 tiflags &= ~TH_URG; 1833 th->th_urp = 0; 1834 } 1835 } 1836 1837 /* 1838 * If new data are received on a connection after the 1839 * user processes are gone, then RST the other end. 1840 */ 1841 if ((so->so_state & SS_NOFDREF) && 1842 tp->t_state > TCPS_CLOSE_WAIT && tlen) { 1843 tp = tcp_close(tp); 1844 tcpstat.tcps_rcvafterclose++; 1845 goto dropwithreset; 1846 } 1847 1848 /* 1849 * If segment ends after window, drop trailing data 1850 * (and PUSH and FIN); if nothing left, just ACK. 1851 */ 1852 todrop = (th->th_seq + tlen) - (tp->rcv_nxt+tp->rcv_wnd); 1853 if (todrop > 0) { 1854 tcpstat.tcps_rcvpackafterwin++; 1855 if (todrop >= tlen) { 1856 tcpstat.tcps_rcvbyteafterwin += tlen; 1857 /* 1858 * If a new connection request is received 1859 * while in TIME_WAIT, drop the old connection 1860 * and start over if the sequence numbers 1861 * are above the previous ones. 1862 * 1863 * NOTE: We will checksum the packet again, and 1864 * so we need to put the header fields back into 1865 * network order! 1866 * XXX This kind of sucks, but we don't expect 1867 * XXX this to happen very often, so maybe it 1868 * XXX doesn't matter so much. 1869 */ 1870 if (tiflags & TH_SYN && 1871 tp->t_state == TCPS_TIME_WAIT && 1872 SEQ_GT(th->th_seq, tp->rcv_nxt)) { 1873 iss = tcp_new_iss(tp, tp->snd_nxt); 1874 tp = tcp_close(tp); 1875 TCP_FIELDS_TO_NET(th); 1876 goto findpcb; 1877 } 1878 /* 1879 * If window is closed can only take segments at 1880 * window edge, and have to drop data and PUSH from 1881 * incoming segments. Continue processing, but 1882 * remember to ack. Otherwise, drop segment 1883 * and ack. 1884 */ 1885 if (tp->rcv_wnd == 0 && th->th_seq == tp->rcv_nxt) { 1886 tp->t_flags |= TF_ACKNOW; 1887 tcpstat.tcps_rcvwinprobe++; 1888 } else 1889 goto dropafterack; 1890 } else 1891 tcpstat.tcps_rcvbyteafterwin += todrop; 1892 m_adj(m, -todrop); 1893 tlen -= todrop; 1894 tiflags &= ~(TH_PUSH|TH_FIN); 1895 } 1896 1897 /* 1898 * If last ACK falls within this segment's sequence numbers, 1899 * and the timestamp is newer, record it. 1900 */ 1901 if (opti.ts_present && TSTMP_GEQ(opti.ts_val, tp->ts_recent) && 1902 SEQ_LEQ(th->th_seq, tp->last_ack_sent) && 1903 SEQ_LT(tp->last_ack_sent, th->th_seq + tlen + 1904 ((tiflags & (TH_SYN|TH_FIN)) != 0))) { 1905 tp->ts_recent_age = TCP_TIMESTAMP(tp); 1906 tp->ts_recent = opti.ts_val; 1907 } 1908 1909 /* 1910 * If the RST bit is set examine the state: 1911 * SYN_RECEIVED STATE: 1912 * If passive open, return to LISTEN state. 1913 * If active open, inform user that connection was refused. 1914 * ESTABLISHED, FIN_WAIT_1, FIN_WAIT2, CLOSE_WAIT STATES: 1915 * Inform user that connection was reset, and close tcb. 1916 * CLOSING, LAST_ACK, TIME_WAIT STATES 1917 * Close the tcb. 1918 */ 1919 if (tiflags&TH_RST) switch (tp->t_state) { 1920 1921 case TCPS_SYN_RECEIVED: 1922 so->so_error = ECONNREFUSED; 1923 goto close; 1924 1925 case TCPS_ESTABLISHED: 1926 case TCPS_FIN_WAIT_1: 1927 case TCPS_FIN_WAIT_2: 1928 case TCPS_CLOSE_WAIT: 1929 so->so_error = ECONNRESET; 1930 close: 1931 tp->t_state = TCPS_CLOSED; 1932 tcpstat.tcps_drops++; 1933 tp = tcp_close(tp); 1934 goto drop; 1935 1936 case TCPS_CLOSING: 1937 case TCPS_LAST_ACK: 1938 case TCPS_TIME_WAIT: 1939 tp = tcp_close(tp); 1940 goto drop; 1941 } 1942 1943 /* 1944 * If a SYN is in the window, then this is an 1945 * error and we send an RST and drop the connection. 1946 */ 1947 if (tiflags & TH_SYN) { 1948 tp = tcp_drop(tp, ECONNRESET); 1949 goto dropwithreset; 1950 } 1951 1952 /* 1953 * If the ACK bit is off we drop the segment and return. 1954 */ 1955 if ((tiflags & TH_ACK) == 0) { 1956 if (tp->t_flags & TF_ACKNOW) 1957 goto dropafterack; 1958 else 1959 goto drop; 1960 } 1961 1962 /* 1963 * Ack processing. 1964 */ 1965 switch (tp->t_state) { 1966 1967 /* 1968 * In SYN_RECEIVED state if the ack ACKs our SYN then enter 1969 * ESTABLISHED state and continue processing, otherwise 1970 * send an RST. 1971 */ 1972 case TCPS_SYN_RECEIVED: 1973 if (SEQ_GT(tp->snd_una, th->th_ack) || 1974 SEQ_GT(th->th_ack, tp->snd_max)) 1975 goto dropwithreset; 1976 tcpstat.tcps_connects++; 1977 soisconnected(so); 1978 tcp_established(tp); 1979 /* Do window scaling? */ 1980 if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) == 1981 (TF_RCVD_SCALE|TF_REQ_SCALE)) { 1982 tp->snd_scale = tp->requested_s_scale; 1983 tp->rcv_scale = tp->request_r_scale; 1984 } 1985 TCP_REASS_LOCK(tp); 1986 (void) tcp_reass(tp, NULL, (struct mbuf *)0, &tlen); 1987 TCP_REASS_UNLOCK(tp); 1988 tp->snd_wl1 = th->th_seq - 1; 1989 /* fall into ... */ 1990 1991 /* 1992 * In ESTABLISHED state: drop duplicate ACKs; ACK out of range 1993 * ACKs. If the ack is in the range 1994 * tp->snd_una < th->th_ack <= tp->snd_max 1995 * then advance tp->snd_una to th->th_ack and drop 1996 * data from the retransmission queue. If this ACK reflects 1997 * more up to date window information we update our window information. 1998 */ 1999 case TCPS_ESTABLISHED: 2000 case TCPS_FIN_WAIT_1: 2001 case TCPS_FIN_WAIT_2: 2002 case TCPS_CLOSE_WAIT: 2003 case TCPS_CLOSING: 2004 case TCPS_LAST_ACK: 2005 case TCPS_TIME_WAIT: 2006 2007 if (SEQ_LEQ(th->th_ack, tp->snd_una)) { 2008 if (tlen == 0 && tiwin == tp->snd_wnd) { 2009 tcpstat.tcps_rcvdupack++; 2010 /* 2011 * If we have outstanding data (other than 2012 * a window probe), this is a completely 2013 * duplicate ack (ie, window info didn't 2014 * change), the ack is the biggest we've 2015 * seen and we've seen exactly our rexmt 2016 * threshhold of them, assume a packet 2017 * has been dropped and retransmit it. 2018 * Kludge snd_nxt & the congestion 2019 * window so we send only this one 2020 * packet. 2021 * 2022 * We know we're losing at the current 2023 * window size so do congestion avoidance 2024 * (set ssthresh to half the current window 2025 * and pull our congestion window back to 2026 * the new ssthresh). 2027 * 2028 * Dup acks mean that packets have left the 2029 * network (they're now cached at the receiver) 2030 * so bump cwnd by the amount in the receiver 2031 * to keep a constant cwnd packets in the 2032 * network. 2033 */ 2034 if (TCP_TIMER_ISARMED(tp, TCPT_REXMT) == 0 || 2035 th->th_ack != tp->snd_una) 2036 tp->t_dupacks = 0; 2037 else if (++tp->t_dupacks == tcprexmtthresh) { 2038 tcp_seq onxt = tp->snd_nxt; 2039 u_int win = 2040 min(tp->snd_wnd, tp->snd_cwnd) / 2041 2 / tp->t_segsz; 2042 if (tcp_do_newreno && SEQ_LT(th->th_ack, 2043 tp->snd_recover)) { 2044 /* 2045 * False fast retransmit after 2046 * timeout. Do not cut window. 2047 */ 2048 tp->snd_cwnd += tp->t_segsz; 2049 tp->t_dupacks = 0; 2050 (void) tcp_output(tp); 2051 goto drop; 2052 } 2053 2054 if (win < 2) 2055 win = 2; 2056 tp->snd_ssthresh = win * tp->t_segsz; 2057 tp->snd_recover = tp->snd_max; 2058 TCP_TIMER_DISARM(tp, TCPT_REXMT); 2059 tp->t_rtttime = 0; 2060 tp->snd_nxt = th->th_ack; 2061 tp->snd_cwnd = tp->t_segsz; 2062 (void) tcp_output(tp); 2063 tp->snd_cwnd = tp->snd_ssthresh + 2064 tp->t_segsz * tp->t_dupacks; 2065 if (SEQ_GT(onxt, tp->snd_nxt)) 2066 tp->snd_nxt = onxt; 2067 goto drop; 2068 } else if (tp->t_dupacks > tcprexmtthresh) { 2069 tp->snd_cwnd += tp->t_segsz; 2070 (void) tcp_output(tp); 2071 goto drop; 2072 } 2073 } else 2074 tp->t_dupacks = 0; 2075 break; 2076 } 2077 /* 2078 * If the congestion window was inflated to account 2079 * for the other side's cached packets, retract it. 2080 */ 2081 if (tcp_do_newreno == 0) { 2082 if (tp->t_dupacks >= tcprexmtthresh && 2083 tp->snd_cwnd > tp->snd_ssthresh) 2084 tp->snd_cwnd = tp->snd_ssthresh; 2085 tp->t_dupacks = 0; 2086 } else if (tp->t_dupacks >= tcprexmtthresh && 2087 tcp_newreno(tp, th) == 0) { 2088 tp->snd_cwnd = tp->snd_ssthresh; 2089 /* 2090 * Window inflation should have left us with approx. 2091 * snd_ssthresh outstanding data. But in case we 2092 * would be inclined to send a burst, better to do 2093 * it via the slow start mechanism. 2094 */ 2095 if (SEQ_SUB(tp->snd_max, th->th_ack) < tp->snd_ssthresh) 2096 tp->snd_cwnd = SEQ_SUB(tp->snd_max, th->th_ack) 2097 + tp->t_segsz; 2098 tp->t_dupacks = 0; 2099 } 2100 if (SEQ_GT(th->th_ack, tp->snd_max)) { 2101 tcpstat.tcps_rcvacktoomuch++; 2102 goto dropafterack; 2103 } 2104 acked = th->th_ack - tp->snd_una; 2105 tcpstat.tcps_rcvackpack++; 2106 tcpstat.tcps_rcvackbyte += acked; 2107 2108 /* 2109 * If we have a timestamp reply, update smoothed 2110 * round trip time. If no timestamp is present but 2111 * transmit timer is running and timed sequence 2112 * number was acked, update smoothed round trip time. 2113 * Since we now have an rtt measurement, cancel the 2114 * timer backoff (cf., Phil Karn's retransmit alg.). 2115 * Recompute the initial retransmit timer. 2116 */ 2117 if (opti.ts_present && opti.ts_ecr) 2118 tcp_xmit_timer(tp, TCP_TIMESTAMP(tp) - opti.ts_ecr + 1); 2119 else if (tp->t_rtttime && SEQ_GT(th->th_ack, tp->t_rtseq)) 2120 tcp_xmit_timer(tp, tcp_now - tp->t_rtttime); 2121 2122 /* 2123 * If all outstanding data is acked, stop retransmit 2124 * timer and remember to restart (more output or persist). 2125 * If there is more data to be acked, restart retransmit 2126 * timer, using current (possibly backed-off) value. 2127 */ 2128 if (th->th_ack == tp->snd_max) { 2129 TCP_TIMER_DISARM(tp, TCPT_REXMT); 2130 needoutput = 1; 2131 } else if (TCP_TIMER_ISARMED(tp, TCPT_PERSIST) == 0) 2132 TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur); 2133 /* 2134 * When new data is acked, open the congestion window. 2135 * If the window gives us less than ssthresh packets 2136 * in flight, open exponentially (segsz per packet). 2137 * Otherwise open linearly: segsz per window 2138 * (segsz^2 / cwnd per packet), plus a constant 2139 * fraction of a packet (segsz/8) to help larger windows 2140 * open quickly enough. 2141 */ 2142 { 2143 u_int cw = tp->snd_cwnd; 2144 u_int incr = tp->t_segsz; 2145 2146 if (cw > tp->snd_ssthresh) 2147 incr = incr * incr / cw; 2148 if (tcp_do_newreno == 0 || SEQ_GEQ(th->th_ack, tp->snd_recover)) 2149 tp->snd_cwnd = min(cw + incr, 2150 TCP_MAXWIN << tp->snd_scale); 2151 } 2152 ND6_HINT(tp); 2153 if (acked > so->so_snd.sb_cc) { 2154 tp->snd_wnd -= so->so_snd.sb_cc; 2155 sbdrop(&so->so_snd, (int)so->so_snd.sb_cc); 2156 ourfinisacked = 1; 2157 } else { 2158 sbdrop(&so->so_snd, acked); 2159 tp->snd_wnd -= acked; 2160 ourfinisacked = 0; 2161 } 2162 sowwakeup(so); 2163 /* 2164 * We want snd_recover to track snd_una to 2165 * avoid sequence wraparound problems for 2166 * very large transfers. 2167 */ 2168 tp->snd_una = tp->snd_recover = th->th_ack; 2169 if (SEQ_LT(tp->snd_nxt, tp->snd_una)) 2170 tp->snd_nxt = tp->snd_una; 2171 2172 switch (tp->t_state) { 2173 2174 /* 2175 * In FIN_WAIT_1 STATE in addition to the processing 2176 * for the ESTABLISHED state if our FIN is now acknowledged 2177 * then enter FIN_WAIT_2. 2178 */ 2179 case TCPS_FIN_WAIT_1: 2180 if (ourfinisacked) { 2181 /* 2182 * If we can't receive any more 2183 * data, then closing user can proceed. 2184 * Starting the timer is contrary to the 2185 * specification, but if we don't get a FIN 2186 * we'll hang forever. 2187 */ 2188 if (so->so_state & SS_CANTRCVMORE) { 2189 soisdisconnected(so); 2190 if (tcp_maxidle > 0) 2191 TCP_TIMER_ARM(tp, TCPT_2MSL, 2192 tcp_maxidle); 2193 } 2194 tp->t_state = TCPS_FIN_WAIT_2; 2195 } 2196 break; 2197 2198 /* 2199 * In CLOSING STATE in addition to the processing for 2200 * the ESTABLISHED state if the ACK acknowledges our FIN 2201 * then enter the TIME-WAIT state, otherwise ignore 2202 * the segment. 2203 */ 2204 case TCPS_CLOSING: 2205 if (ourfinisacked) { 2206 tp->t_state = TCPS_TIME_WAIT; 2207 tcp_canceltimers(tp); 2208 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2209 soisdisconnected(so); 2210 } 2211 break; 2212 2213 /* 2214 * In LAST_ACK, we may still be waiting for data to drain 2215 * and/or to be acked, as well as for the ack of our FIN. 2216 * If our FIN is now acknowledged, delete the TCB, 2217 * enter the closed state and return. 2218 */ 2219 case TCPS_LAST_ACK: 2220 if (ourfinisacked) { 2221 tp = tcp_close(tp); 2222 goto drop; 2223 } 2224 break; 2225 2226 /* 2227 * In TIME_WAIT state the only thing that should arrive 2228 * is a retransmission of the remote FIN. Acknowledge 2229 * it and restart the finack timer. 2230 */ 2231 case TCPS_TIME_WAIT: 2232 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2233 goto dropafterack; 2234 } 2235 } 2236 2237 step6: 2238 /* 2239 * Update window information. 2240 * Don't look at window if no ACK: TAC's send garbage on first SYN. 2241 */ 2242 if ((tiflags & TH_ACK) && (SEQ_LT(tp->snd_wl1, th->th_seq) || 2243 (tp->snd_wl1 == th->th_seq && SEQ_LT(tp->snd_wl2, th->th_ack)) || 2244 (tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd))) { 2245 /* keep track of pure window updates */ 2246 if (tlen == 0 && 2247 tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd) 2248 tcpstat.tcps_rcvwinupd++; 2249 tp->snd_wnd = tiwin; 2250 tp->snd_wl1 = th->th_seq; 2251 tp->snd_wl2 = th->th_ack; 2252 if (tp->snd_wnd > tp->max_sndwnd) 2253 tp->max_sndwnd = tp->snd_wnd; 2254 needoutput = 1; 2255 } 2256 2257 /* 2258 * Process segments with URG. 2259 */ 2260 if ((tiflags & TH_URG) && th->th_urp && 2261 TCPS_HAVERCVDFIN(tp->t_state) == 0) { 2262 /* 2263 * This is a kludge, but if we receive and accept 2264 * random urgent pointers, we'll crash in 2265 * soreceive. It's hard to imagine someone 2266 * actually wanting to send this much urgent data. 2267 */ 2268 if (th->th_urp + so->so_rcv.sb_cc > sb_max) { 2269 th->th_urp = 0; /* XXX */ 2270 tiflags &= ~TH_URG; /* XXX */ 2271 goto dodata; /* XXX */ 2272 } 2273 /* 2274 * If this segment advances the known urgent pointer, 2275 * then mark the data stream. This should not happen 2276 * in CLOSE_WAIT, CLOSING, LAST_ACK or TIME_WAIT STATES since 2277 * a FIN has been received from the remote side. 2278 * In these states we ignore the URG. 2279 * 2280 * According to RFC961 (Assigned Protocols), 2281 * the urgent pointer points to the last octet 2282 * of urgent data. We continue, however, 2283 * to consider it to indicate the first octet 2284 * of data past the urgent section as the original 2285 * spec states (in one of two places). 2286 */ 2287 if (SEQ_GT(th->th_seq+th->th_urp, tp->rcv_up)) { 2288 tp->rcv_up = th->th_seq + th->th_urp; 2289 so->so_oobmark = so->so_rcv.sb_cc + 2290 (tp->rcv_up - tp->rcv_nxt) - 1; 2291 if (so->so_oobmark == 0) 2292 so->so_state |= SS_RCVATMARK; 2293 sohasoutofband(so); 2294 tp->t_oobflags &= ~(TCPOOB_HAVEDATA | TCPOOB_HADDATA); 2295 } 2296 /* 2297 * Remove out of band data so doesn't get presented to user. 2298 * This can happen independent of advancing the URG pointer, 2299 * but if two URG's are pending at once, some out-of-band 2300 * data may creep in... ick. 2301 */ 2302 if (th->th_urp <= (u_int16_t) tlen 2303 #ifdef SO_OOBINLINE 2304 && (so->so_options & SO_OOBINLINE) == 0 2305 #endif 2306 ) 2307 tcp_pulloutofband(so, th, m, hdroptlen); 2308 } else 2309 /* 2310 * If no out of band data is expected, 2311 * pull receive urgent pointer along 2312 * with the receive window. 2313 */ 2314 if (SEQ_GT(tp->rcv_nxt, tp->rcv_up)) 2315 tp->rcv_up = tp->rcv_nxt; 2316 dodata: /* XXX */ 2317 2318 /* 2319 * Process the segment text, merging it into the TCP sequencing queue, 2320 * and arranging for acknowledgement of receipt if necessary. 2321 * This process logically involves adjusting tp->rcv_wnd as data 2322 * is presented to the user (this happens in tcp_usrreq.c, 2323 * case PRU_RCVD). If a FIN has already been received on this 2324 * connection then we just ignore the text. 2325 */ 2326 if ((tlen || (tiflags & TH_FIN)) && 2327 TCPS_HAVERCVDFIN(tp->t_state) == 0) { 2328 /* 2329 * Insert segment ti into reassembly queue of tcp with 2330 * control block tp. Return TH_FIN if reassembly now includes 2331 * a segment with FIN. The macro form does the common case 2332 * inline (segment is the next to be received on an 2333 * established connection, and the queue is empty), 2334 * avoiding linkage into and removal from the queue and 2335 * repetition of various conversions. 2336 * Set DELACK for segments received in order, but ack 2337 * immediately when segments are out of order 2338 * (so fast retransmit can work). 2339 */ 2340 /* NOTE: this was TCP_REASS() macro, but used only once */ 2341 TCP_REASS_LOCK(tp); 2342 if (th->th_seq == tp->rcv_nxt && 2343 TAILQ_FIRST(&tp->segq) == NULL && 2344 tp->t_state == TCPS_ESTABLISHED) { 2345 TCP_SETUP_ACK(tp, th); 2346 tp->rcv_nxt += tlen; 2347 tiflags = th->th_flags & TH_FIN; 2348 tcpstat.tcps_rcvpack++; 2349 tcpstat.tcps_rcvbyte += tlen; 2350 ND6_HINT(tp); 2351 if (so->so_state & SS_CANTRCVMORE) 2352 m_freem(m); 2353 else { 2354 m_adj(m, hdroptlen); 2355 sbappendstream(&(so)->so_rcv, m); 2356 } 2357 sorwakeup(so); 2358 } else { 2359 m_adj(m, hdroptlen); 2360 tiflags = tcp_reass(tp, th, m, &tlen); 2361 tp->t_flags |= TF_ACKNOW; 2362 } 2363 TCP_REASS_UNLOCK(tp); 2364 2365 /* 2366 * Note the amount of data that peer has sent into 2367 * our window, in order to estimate the sender's 2368 * buffer size. 2369 */ 2370 len = so->so_rcv.sb_hiwat - (tp->rcv_adv - tp->rcv_nxt); 2371 } else { 2372 m_freem(m); 2373 m = NULL; 2374 tiflags &= ~TH_FIN; 2375 } 2376 2377 /* 2378 * If FIN is received ACK the FIN and let the user know 2379 * that the connection is closing. Ignore a FIN received before 2380 * the connection is fully established. 2381 */ 2382 if ((tiflags & TH_FIN) && TCPS_HAVEESTABLISHED(tp->t_state)) { 2383 if (TCPS_HAVERCVDFIN(tp->t_state) == 0) { 2384 socantrcvmore(so); 2385 tp->t_flags |= TF_ACKNOW; 2386 tp->rcv_nxt++; 2387 } 2388 switch (tp->t_state) { 2389 2390 /* 2391 * In ESTABLISHED STATE enter the CLOSE_WAIT state. 2392 */ 2393 case TCPS_ESTABLISHED: 2394 tp->t_state = TCPS_CLOSE_WAIT; 2395 break; 2396 2397 /* 2398 * If still in FIN_WAIT_1 STATE FIN has not been acked so 2399 * enter the CLOSING state. 2400 */ 2401 case TCPS_FIN_WAIT_1: 2402 tp->t_state = TCPS_CLOSING; 2403 break; 2404 2405 /* 2406 * In FIN_WAIT_2 state enter the TIME_WAIT state, 2407 * starting the time-wait timer, turning off the other 2408 * standard timers. 2409 */ 2410 case TCPS_FIN_WAIT_2: 2411 tp->t_state = TCPS_TIME_WAIT; 2412 tcp_canceltimers(tp); 2413 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2414 soisdisconnected(so); 2415 break; 2416 2417 /* 2418 * In TIME_WAIT state restart the 2 MSL time_wait timer. 2419 */ 2420 case TCPS_TIME_WAIT: 2421 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2422 break; 2423 } 2424 } 2425 #ifdef TCP_DEBUG 2426 if (so->so_options & SO_DEBUG) 2427 tcp_trace(TA_INPUT, ostate, tp, tcp_saveti, 0); 2428 #endif 2429 2430 /* 2431 * Return any desired output. 2432 */ 2433 if (needoutput || (tp->t_flags & TF_ACKNOW)) 2434 (void) tcp_output(tp); 2435 if (tcp_saveti) 2436 m_freem(tcp_saveti); 2437 return; 2438 2439 badsyn: 2440 /* 2441 * Received a bad SYN. Increment counters and dropwithreset. 2442 */ 2443 tcpstat.tcps_badsyn++; 2444 tp = NULL; 2445 goto dropwithreset; 2446 2447 dropafterack: 2448 /* 2449 * Generate an ACK dropping incoming segment if it occupies 2450 * sequence space, where the ACK reflects our state. 2451 */ 2452 if (tiflags & TH_RST) 2453 goto drop; 2454 m_freem(m); 2455 tp->t_flags |= TF_ACKNOW; 2456 (void) tcp_output(tp); 2457 if (tcp_saveti) 2458 m_freem(tcp_saveti); 2459 return; 2460 2461 dropwithreset_ratelim: 2462 /* 2463 * We may want to rate-limit RSTs in certain situations, 2464 * particularly if we are sending an RST in response to 2465 * an attempt to connect to or otherwise communicate with 2466 * a port for which we have no socket. 2467 */ 2468 if (ppsratecheck(&tcp_rst_ppslim_last, &tcp_rst_ppslim_count, 2469 tcp_rst_ppslim) == 0) { 2470 /* XXX stat */ 2471 goto drop; 2472 } 2473 /* ...fall into dropwithreset... */ 2474 2475 dropwithreset: 2476 /* 2477 * Generate a RST, dropping incoming segment. 2478 * Make ACK acceptable to originator of segment. 2479 */ 2480 if (tiflags & TH_RST) 2481 goto drop; 2482 2483 switch (af) { 2484 #ifdef INET6 2485 case AF_INET6: 2486 /* For following calls to tcp_respond */ 2487 if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) 2488 goto drop; 2489 break; 2490 #endif /* INET6 */ 2491 case AF_INET: 2492 if (IN_MULTICAST(ip->ip_dst.s_addr) || 2493 in_broadcast(ip->ip_dst, m->m_pkthdr.rcvif)) 2494 goto drop; 2495 } 2496 2497 { 2498 /* 2499 * need to recover version # field, which was overwritten on 2500 * ip_cksum computation. 2501 */ 2502 struct ip *sip; 2503 sip = mtod(m, struct ip *); 2504 switch (af) { 2505 #ifdef INET 2506 case AF_INET: 2507 sip->ip_v = 4; 2508 break; 2509 #endif 2510 #ifdef INET6 2511 case AF_INET6: 2512 sip->ip_v = 6; 2513 break; 2514 #endif 2515 } 2516 } 2517 if (tiflags & TH_ACK) 2518 (void)tcp_respond(tp, m, m, th, (tcp_seq)0, th->th_ack, TH_RST); 2519 else { 2520 if (tiflags & TH_SYN) 2521 tlen++; 2522 (void)tcp_respond(tp, m, m, th, th->th_seq + tlen, (tcp_seq)0, 2523 TH_RST|TH_ACK); 2524 } 2525 if (tcp_saveti) 2526 m_freem(tcp_saveti); 2527 return; 2528 2529 badcsum: 2530 tcpstat.tcps_rcvbadsum++; 2531 drop: 2532 /* 2533 * Drop space held by incoming segment and return. 2534 */ 2535 if (tp) { 2536 if (tp->t_inpcb) 2537 so = tp->t_inpcb->inp_socket; 2538 #ifdef INET6 2539 else if (tp->t_in6pcb) 2540 so = tp->t_in6pcb->in6p_socket; 2541 #endif 2542 else 2543 so = NULL; 2544 #ifdef TCP_DEBUG 2545 if (so && (so->so_options & SO_DEBUG) != 0) 2546 tcp_trace(TA_DROP, ostate, tp, tcp_saveti, 0); 2547 #endif 2548 } 2549 if (tcp_saveti) 2550 m_freem(tcp_saveti); 2551 m_freem(m); 2552 return; 2553 } 2554 2555 void 2556 tcp_dooptions(tp, cp, cnt, th, oi) 2557 struct tcpcb *tp; 2558 u_char *cp; 2559 int cnt; 2560 struct tcphdr *th; 2561 struct tcp_opt_info *oi; 2562 { 2563 u_int16_t mss; 2564 int opt, optlen; 2565 2566 for (; cnt > 0; cnt -= optlen, cp += optlen) { 2567 opt = cp[0]; 2568 if (opt == TCPOPT_EOL) 2569 break; 2570 if (opt == TCPOPT_NOP) 2571 optlen = 1; 2572 else { 2573 if (cnt < 2) 2574 break; 2575 optlen = cp[1]; 2576 if (optlen < 2 || optlen > cnt) 2577 break; 2578 } 2579 switch (opt) { 2580 2581 default: 2582 continue; 2583 2584 case TCPOPT_MAXSEG: 2585 if (optlen != TCPOLEN_MAXSEG) 2586 continue; 2587 if (!(th->th_flags & TH_SYN)) 2588 continue; 2589 bcopy(cp + 2, &mss, sizeof(mss)); 2590 oi->maxseg = ntohs(mss); 2591 break; 2592 2593 case TCPOPT_WINDOW: 2594 if (optlen != TCPOLEN_WINDOW) 2595 continue; 2596 if (!(th->th_flags & TH_SYN)) 2597 continue; 2598 tp->t_flags |= TF_RCVD_SCALE; 2599 tp->requested_s_scale = cp[2]; 2600 if (tp->requested_s_scale > TCP_MAX_WINSHIFT) { 2601 #if 0 /*XXX*/ 2602 char *p; 2603 2604 if (ip) 2605 p = ntohl(ip->ip_src); 2606 #ifdef INET6 2607 else if (ip6) 2608 p = ip6_sprintf(&ip6->ip6_src); 2609 #endif 2610 else 2611 p = "(unknown)"; 2612 log(LOG_ERR, "TCP: invalid wscale %d from %s, " 2613 "assuming %d\n", 2614 tp->requested_s_scale, p, 2615 TCP_MAX_WINSHIFT); 2616 #else 2617 log(LOG_ERR, "TCP: invalid wscale %d, " 2618 "assuming %d\n", 2619 tp->requested_s_scale, 2620 TCP_MAX_WINSHIFT); 2621 #endif 2622 tp->requested_s_scale = TCP_MAX_WINSHIFT; 2623 } 2624 break; 2625 2626 case TCPOPT_TIMESTAMP: 2627 if (optlen != TCPOLEN_TIMESTAMP) 2628 continue; 2629 oi->ts_present = 1; 2630 bcopy(cp + 2, &oi->ts_val, sizeof(oi->ts_val)); 2631 NTOHL(oi->ts_val); 2632 bcopy(cp + 6, &oi->ts_ecr, sizeof(oi->ts_ecr)); 2633 NTOHL(oi->ts_ecr); 2634 2635 /* 2636 * A timestamp received in a SYN makes 2637 * it ok to send timestamp requests and replies. 2638 */ 2639 if (th->th_flags & TH_SYN) { 2640 tp->t_flags |= TF_RCVD_TSTMP; 2641 tp->ts_recent = oi->ts_val; 2642 tp->ts_recent_age = TCP_TIMESTAMP(tp); 2643 } 2644 break; 2645 case TCPOPT_SACK_PERMITTED: 2646 if (optlen != TCPOLEN_SACK_PERMITTED) 2647 continue; 2648 if (!(th->th_flags & TH_SYN)) 2649 continue; 2650 tp->t_flags &= ~TF_CANT_TXSACK; 2651 break; 2652 2653 case TCPOPT_SACK: 2654 if (tp->t_flags & TF_IGNR_RXSACK) 2655 continue; 2656 if (optlen % 8 != 2 || optlen < 10) 2657 continue; 2658 cp += 2; 2659 optlen -= 2; 2660 for (; optlen > 0; cp -= 8, optlen -= 8) { 2661 tcp_seq lwe, rwe; 2662 bcopy((char *)cp, (char *) &lwe, sizeof(lwe)); 2663 NTOHL(lwe); 2664 bcopy((char *)cp, (char *) &rwe, sizeof(rwe)); 2665 NTOHL(rwe); 2666 /* tcp_mark_sacked(tp, lwe, rwe); */ 2667 } 2668 break; 2669 } 2670 } 2671 } 2672 2673 /* 2674 * Pull out of band byte out of a segment so 2675 * it doesn't appear in the user's data queue. 2676 * It is still reflected in the segment length for 2677 * sequencing purposes. 2678 */ 2679 void 2680 tcp_pulloutofband(so, th, m, off) 2681 struct socket *so; 2682 struct tcphdr *th; 2683 struct mbuf *m; 2684 int off; 2685 { 2686 int cnt = off + th->th_urp - 1; 2687 2688 while (cnt >= 0) { 2689 if (m->m_len > cnt) { 2690 char *cp = mtod(m, caddr_t) + cnt; 2691 struct tcpcb *tp = sototcpcb(so); 2692 2693 tp->t_iobc = *cp; 2694 tp->t_oobflags |= TCPOOB_HAVEDATA; 2695 bcopy(cp+1, cp, (unsigned)(m->m_len - cnt - 1)); 2696 m->m_len--; 2697 return; 2698 } 2699 cnt -= m->m_len; 2700 m = m->m_next; 2701 if (m == 0) 2702 break; 2703 } 2704 panic("tcp_pulloutofband"); 2705 } 2706 2707 /* 2708 * Collect new round-trip time estimate 2709 * and update averages and current timeout. 2710 */ 2711 void 2712 tcp_xmit_timer(tp, rtt) 2713 struct tcpcb *tp; 2714 uint32_t rtt; 2715 { 2716 int32_t delta; 2717 2718 tcpstat.tcps_rttupdated++; 2719 if (tp->t_srtt != 0) { 2720 /* 2721 * srtt is stored as fixed point with 3 bits after the 2722 * binary point (i.e., scaled by 8). The following magic 2723 * is equivalent to the smoothing algorithm in rfc793 with 2724 * an alpha of .875 (srtt = rtt/8 + srtt*7/8 in fixed 2725 * point). Adjust rtt to origin 0. 2726 */ 2727 delta = (rtt << 2) - (tp->t_srtt >> TCP_RTT_SHIFT); 2728 if ((tp->t_srtt += delta) <= 0) 2729 tp->t_srtt = 1 << 2; 2730 /* 2731 * We accumulate a smoothed rtt variance (actually, a 2732 * smoothed mean difference), then set the retransmit 2733 * timer to smoothed rtt + 4 times the smoothed variance. 2734 * rttvar is stored as fixed point with 2 bits after the 2735 * binary point (scaled by 4). The following is 2736 * equivalent to rfc793 smoothing with an alpha of .75 2737 * (rttvar = rttvar*3/4 + |delta| / 4). This replaces 2738 * rfc793's wired-in beta. 2739 */ 2740 if (delta < 0) 2741 delta = -delta; 2742 delta -= (tp->t_rttvar >> TCP_RTTVAR_SHIFT); 2743 if ((tp->t_rttvar += delta) <= 0) 2744 tp->t_rttvar = 1 << 2; 2745 } else { 2746 /* 2747 * No rtt measurement yet - use the unsmoothed rtt. 2748 * Set the variance to half the rtt (so our first 2749 * retransmit happens at 3*rtt). 2750 */ 2751 tp->t_srtt = rtt << (TCP_RTT_SHIFT + 2); 2752 tp->t_rttvar = rtt << (TCP_RTTVAR_SHIFT + 2 - 1); 2753 } 2754 tp->t_rtttime = 0; 2755 tp->t_rxtshift = 0; 2756 2757 /* 2758 * the retransmit should happen at rtt + 4 * rttvar. 2759 * Because of the way we do the smoothing, srtt and rttvar 2760 * will each average +1/2 tick of bias. When we compute 2761 * the retransmit timer, we want 1/2 tick of rounding and 2762 * 1 extra tick because of +-1/2 tick uncertainty in the 2763 * firing of the timer. The bias will give us exactly the 2764 * 1.5 tick we need. But, because the bias is 2765 * statistical, we have to test that we don't drop below 2766 * the minimum feasible timer (which is 2 ticks). 2767 */ 2768 TCPT_RANGESET(tp->t_rxtcur, TCP_REXMTVAL(tp), 2769 max(tp->t_rttmin, rtt + 2), TCPTV_REXMTMAX); 2770 2771 /* 2772 * We received an ack for a packet that wasn't retransmitted; 2773 * it is probably safe to discard any error indications we've 2774 * received recently. This isn't quite right, but close enough 2775 * for now (a route might have failed after we sent a segment, 2776 * and the return path might not be symmetrical). 2777 */ 2778 tp->t_softerror = 0; 2779 } 2780 2781 /* 2782 * Checks for partial ack. If partial ack arrives, force the retransmission 2783 * of the next unacknowledged segment, do not clear tp->t_dupacks, and return 2784 * 1. By setting snd_nxt to th_ack, this forces retransmission timer to 2785 * be started again. If the ack advances at least to tp->snd_recover, return 0. 2786 */ 2787 int 2788 tcp_newreno(tp, th) 2789 struct tcpcb *tp; 2790 struct tcphdr *th; 2791 { 2792 tcp_seq onxt = tp->snd_nxt; 2793 u_long ocwnd = tp->snd_cwnd; 2794 2795 if (SEQ_LT(th->th_ack, tp->snd_recover)) { 2796 /* 2797 * snd_una has not yet been updated and the socket's send 2798 * buffer has not yet drained off the ACK'd data, so we 2799 * have to leave snd_una as it was to get the correct data 2800 * offset in tcp_output(). 2801 */ 2802 TCP_TIMER_DISARM(tp, TCPT_REXMT); 2803 tp->t_rtttime = 0; 2804 tp->snd_nxt = th->th_ack; 2805 /* 2806 * Set snd_cwnd to one segment beyond ACK'd offset. snd_una 2807 * is not yet updated when we're called. 2808 */ 2809 tp->snd_cwnd = tp->t_segsz + (th->th_ack - tp->snd_una); 2810 (void) tcp_output(tp); 2811 tp->snd_cwnd = ocwnd; 2812 if (SEQ_GT(onxt, tp->snd_nxt)) 2813 tp->snd_nxt = onxt; 2814 /* 2815 * Partial window deflation. Relies on fact that tp->snd_una 2816 * not updated yet. 2817 */ 2818 tp->snd_cwnd -= (th->th_ack - tp->snd_una - tp->t_segsz); 2819 return 1; 2820 } 2821 return 0; 2822 } 2823 2824 2825 /* 2826 * TCP compressed state engine. Currently used to hold compressed 2827 * state for SYN_RECEIVED. 2828 */ 2829 2830 u_long syn_cache_count; 2831 u_int32_t syn_hash1, syn_hash2; 2832 2833 #define SYN_HASH(sa, sp, dp) \ 2834 ((((sa)->s_addr^syn_hash1)*(((((u_int32_t)(dp))<<16) + \ 2835 ((u_int32_t)(sp)))^syn_hash2))) 2836 #ifndef INET6 2837 #define SYN_HASHALL(hash, src, dst) \ 2838 do { \ 2839 hash = SYN_HASH(&((struct sockaddr_in *)(src))->sin_addr, \ 2840 ((struct sockaddr_in *)(src))->sin_port, \ 2841 ((struct sockaddr_in *)(dst))->sin_port); \ 2842 } while (/*CONSTCOND*/ 0) 2843 #else 2844 #define SYN_HASH6(sa, sp, dp) \ 2845 ((((sa)->s6_addr32[0] ^ (sa)->s6_addr32[3] ^ syn_hash1) * \ 2846 (((((u_int32_t)(dp))<<16) + ((u_int32_t)(sp)))^syn_hash2)) \ 2847 & 0x7fffffff) 2848 2849 #define SYN_HASHALL(hash, src, dst) \ 2850 do { \ 2851 switch ((src)->sa_family) { \ 2852 case AF_INET: \ 2853 hash = SYN_HASH(&((struct sockaddr_in *)(src))->sin_addr, \ 2854 ((struct sockaddr_in *)(src))->sin_port, \ 2855 ((struct sockaddr_in *)(dst))->sin_port); \ 2856 break; \ 2857 case AF_INET6: \ 2858 hash = SYN_HASH6(&((struct sockaddr_in6 *)(src))->sin6_addr, \ 2859 ((struct sockaddr_in6 *)(src))->sin6_port, \ 2860 ((struct sockaddr_in6 *)(dst))->sin6_port); \ 2861 break; \ 2862 default: \ 2863 hash = 0; \ 2864 } \ 2865 } while (/*CONSTCOND*/0) 2866 #endif /* INET6 */ 2867 2868 #define SYN_CACHE_RM(sc) \ 2869 do { \ 2870 TAILQ_REMOVE(&tcp_syn_cache[(sc)->sc_bucketidx].sch_bucket, \ 2871 (sc), sc_bucketq); \ 2872 (sc)->sc_tp = NULL; \ 2873 LIST_REMOVE((sc), sc_tpq); \ 2874 tcp_syn_cache[(sc)->sc_bucketidx].sch_length--; \ 2875 callout_stop(&(sc)->sc_timer); \ 2876 syn_cache_count--; \ 2877 } while (/*CONSTCOND*/0) 2878 2879 #define SYN_CACHE_PUT(sc) \ 2880 do { \ 2881 if ((sc)->sc_ipopts) \ 2882 (void) m_free((sc)->sc_ipopts); \ 2883 if ((sc)->sc_route4.ro_rt != NULL) \ 2884 RTFREE((sc)->sc_route4.ro_rt); \ 2885 pool_put(&syn_cache_pool, (sc)); \ 2886 } while (/*CONSTCOND*/0) 2887 2888 struct pool syn_cache_pool; 2889 2890 /* 2891 * We don't estimate RTT with SYNs, so each packet starts with the default 2892 * RTT and each timer step has a fixed timeout value. 2893 */ 2894 #define SYN_CACHE_TIMER_ARM(sc) \ 2895 do { \ 2896 TCPT_RANGESET((sc)->sc_rxtcur, \ 2897 TCPTV_SRTTDFLT * tcp_backoff[(sc)->sc_rxtshift], TCPTV_MIN, \ 2898 TCPTV_REXMTMAX); \ 2899 callout_reset(&(sc)->sc_timer, \ 2900 (sc)->sc_rxtcur * (hz / PR_SLOWHZ), syn_cache_timer, (sc)); \ 2901 } while (/*CONSTCOND*/0) 2902 2903 #define SYN_CACHE_TIMESTAMP(sc) (tcp_now - (sc)->sc_timebase) 2904 2905 void 2906 syn_cache_init() 2907 { 2908 int i; 2909 2910 /* Initialize the hash buckets. */ 2911 for (i = 0; i < tcp_syn_cache_size; i++) 2912 TAILQ_INIT(&tcp_syn_cache[i].sch_bucket); 2913 2914 /* Initialize the syn cache pool. */ 2915 pool_init(&syn_cache_pool, sizeof(struct syn_cache), 0, 0, 0, 2916 "synpl", NULL); 2917 } 2918 2919 void 2920 syn_cache_insert(sc, tp) 2921 struct syn_cache *sc; 2922 struct tcpcb *tp; 2923 { 2924 struct syn_cache_head *scp; 2925 struct syn_cache *sc2; 2926 int s; 2927 2928 /* 2929 * If there are no entries in the hash table, reinitialize 2930 * the hash secrets. 2931 */ 2932 if (syn_cache_count == 0) { 2933 struct timeval tv; 2934 microtime(&tv); 2935 syn_hash1 = arc4random() ^ (u_long)≻ 2936 syn_hash2 = arc4random() ^ tv.tv_usec; 2937 } 2938 2939 SYN_HASHALL(sc->sc_hash, &sc->sc_src.sa, &sc->sc_dst.sa); 2940 sc->sc_bucketidx = sc->sc_hash % tcp_syn_cache_size; 2941 scp = &tcp_syn_cache[sc->sc_bucketidx]; 2942 2943 /* 2944 * Make sure that we don't overflow the per-bucket 2945 * limit or the total cache size limit. 2946 */ 2947 s = splsoftnet(); 2948 if (scp->sch_length >= tcp_syn_bucket_limit) { 2949 tcpstat.tcps_sc_bucketoverflow++; 2950 /* 2951 * The bucket is full. Toss the oldest element in the 2952 * bucket. This will be the first entry in the bucket. 2953 */ 2954 sc2 = TAILQ_FIRST(&scp->sch_bucket); 2955 #ifdef DIAGNOSTIC 2956 /* 2957 * This should never happen; we should always find an 2958 * entry in our bucket. 2959 */ 2960 if (sc2 == NULL) 2961 panic("syn_cache_insert: bucketoverflow: impossible"); 2962 #endif 2963 SYN_CACHE_RM(sc2); 2964 SYN_CACHE_PUT(sc2); 2965 } else if (syn_cache_count >= tcp_syn_cache_limit) { 2966 struct syn_cache_head *scp2, *sce; 2967 2968 tcpstat.tcps_sc_overflowed++; 2969 /* 2970 * The cache is full. Toss the oldest entry in the 2971 * first non-empty bucket we can find. 2972 * 2973 * XXX We would really like to toss the oldest 2974 * entry in the cache, but we hope that this 2975 * condition doesn't happen very often. 2976 */ 2977 scp2 = scp; 2978 if (TAILQ_EMPTY(&scp2->sch_bucket)) { 2979 sce = &tcp_syn_cache[tcp_syn_cache_size]; 2980 for (++scp2; scp2 != scp; scp2++) { 2981 if (scp2 >= sce) 2982 scp2 = &tcp_syn_cache[0]; 2983 if (! TAILQ_EMPTY(&scp2->sch_bucket)) 2984 break; 2985 } 2986 #ifdef DIAGNOSTIC 2987 /* 2988 * This should never happen; we should always find a 2989 * non-empty bucket. 2990 */ 2991 if (scp2 == scp) 2992 panic("syn_cache_insert: cacheoverflow: " 2993 "impossible"); 2994 #endif 2995 } 2996 sc2 = TAILQ_FIRST(&scp2->sch_bucket); 2997 SYN_CACHE_RM(sc2); 2998 SYN_CACHE_PUT(sc2); 2999 } 3000 3001 /* 3002 * Initialize the entry's timer. 3003 */ 3004 sc->sc_rxttot = 0; 3005 sc->sc_rxtshift = 0; 3006 SYN_CACHE_TIMER_ARM(sc); 3007 3008 /* Link it from tcpcb entry */ 3009 LIST_INSERT_HEAD(&tp->t_sc, sc, sc_tpq); 3010 3011 /* Put it into the bucket. */ 3012 TAILQ_INSERT_TAIL(&scp->sch_bucket, sc, sc_bucketq); 3013 scp->sch_length++; 3014 syn_cache_count++; 3015 3016 tcpstat.tcps_sc_added++; 3017 splx(s); 3018 } 3019 3020 /* 3021 * Walk the timer queues, looking for SYN,ACKs that need to be retransmitted. 3022 * If we have retransmitted an entry the maximum number of times, expire 3023 * that entry. 3024 */ 3025 void 3026 syn_cache_timer(void *arg) 3027 { 3028 struct syn_cache *sc = arg; 3029 int s; 3030 3031 s = splsoftnet(); 3032 3033 if (__predict_false(sc->sc_rxtshift == TCP_MAXRXTSHIFT)) { 3034 /* Drop it -- too many retransmissions. */ 3035 goto dropit; 3036 } 3037 3038 /* 3039 * Compute the total amount of time this entry has 3040 * been on a queue. If this entry has been on longer 3041 * than the keep alive timer would allow, expire it. 3042 */ 3043 sc->sc_rxttot += sc->sc_rxtcur; 3044 if (sc->sc_rxttot >= TCPTV_KEEP_INIT) 3045 goto dropit; 3046 3047 tcpstat.tcps_sc_retransmitted++; 3048 (void) syn_cache_respond(sc, NULL); 3049 3050 /* Advance the timer back-off. */ 3051 sc->sc_rxtshift++; 3052 SYN_CACHE_TIMER_ARM(sc); 3053 3054 splx(s); 3055 return; 3056 3057 dropit: 3058 tcpstat.tcps_sc_timed_out++; 3059 SYN_CACHE_RM(sc); 3060 SYN_CACHE_PUT(sc); 3061 splx(s); 3062 } 3063 3064 /* 3065 * Remove syn cache created by the specified tcb entry, 3066 * because this does not make sense to keep them 3067 * (if there's no tcb entry, syn cache entry will never be used) 3068 */ 3069 void 3070 syn_cache_cleanup(tp) 3071 struct tcpcb *tp; 3072 { 3073 struct syn_cache *sc, *nsc; 3074 int s; 3075 3076 s = splsoftnet(); 3077 3078 for (sc = LIST_FIRST(&tp->t_sc); sc != NULL; sc = nsc) { 3079 nsc = LIST_NEXT(sc, sc_tpq); 3080 3081 #ifdef DIAGNOSTIC 3082 if (sc->sc_tp != tp) 3083 panic("invalid sc_tp in syn_cache_cleanup"); 3084 #endif 3085 SYN_CACHE_RM(sc); 3086 SYN_CACHE_PUT(sc); 3087 } 3088 /* just for safety */ 3089 LIST_INIT(&tp->t_sc); 3090 3091 splx(s); 3092 } 3093 3094 /* 3095 * Find an entry in the syn cache. 3096 */ 3097 struct syn_cache * 3098 syn_cache_lookup(src, dst, headp) 3099 struct sockaddr *src; 3100 struct sockaddr *dst; 3101 struct syn_cache_head **headp; 3102 { 3103 struct syn_cache *sc; 3104 struct syn_cache_head *scp; 3105 u_int32_t hash; 3106 int s; 3107 3108 SYN_HASHALL(hash, src, dst); 3109 3110 scp = &tcp_syn_cache[hash % tcp_syn_cache_size]; 3111 *headp = scp; 3112 s = splsoftnet(); 3113 for (sc = TAILQ_FIRST(&scp->sch_bucket); sc != NULL; 3114 sc = TAILQ_NEXT(sc, sc_bucketq)) { 3115 if (sc->sc_hash != hash) 3116 continue; 3117 if (!bcmp(&sc->sc_src, src, src->sa_len) && 3118 !bcmp(&sc->sc_dst, dst, dst->sa_len)) { 3119 splx(s); 3120 return (sc); 3121 } 3122 } 3123 splx(s); 3124 return (NULL); 3125 } 3126 3127 /* 3128 * This function gets called when we receive an ACK for a 3129 * socket in the LISTEN state. We look up the connection 3130 * in the syn cache, and if its there, we pull it out of 3131 * the cache and turn it into a full-blown connection in 3132 * the SYN-RECEIVED state. 3133 * 3134 * The return values may not be immediately obvious, and their effects 3135 * can be subtle, so here they are: 3136 * 3137 * NULL SYN was not found in cache; caller should drop the 3138 * packet and send an RST. 3139 * 3140 * -1 We were unable to create the new connection, and are 3141 * aborting it. An ACK,RST is being sent to the peer 3142 * (unless we got screwey sequence numbners; see below), 3143 * because the 3-way handshake has been completed. Caller 3144 * should not free the mbuf, since we may be using it. If 3145 * we are not, we will free it. 3146 * 3147 * Otherwise, the return value is a pointer to the new socket 3148 * associated with the connection. 3149 */ 3150 struct socket * 3151 syn_cache_get(src, dst, th, hlen, tlen, so, m) 3152 struct sockaddr *src; 3153 struct sockaddr *dst; 3154 struct tcphdr *th; 3155 unsigned int hlen, tlen; 3156 struct socket *so; 3157 struct mbuf *m; 3158 { 3159 struct syn_cache *sc; 3160 struct syn_cache_head *scp; 3161 struct inpcb *inp = NULL; 3162 #ifdef INET6 3163 struct in6pcb *in6p = NULL; 3164 #endif 3165 struct tcpcb *tp = 0; 3166 struct mbuf *am; 3167 int s; 3168 struct socket *oso; 3169 3170 s = splsoftnet(); 3171 if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) { 3172 splx(s); 3173 return (NULL); 3174 } 3175 3176 /* 3177 * Verify the sequence and ack numbers. Try getting the correct 3178 * response again. 3179 */ 3180 if ((th->th_ack != sc->sc_iss + 1) || 3181 SEQ_LEQ(th->th_seq, sc->sc_irs) || 3182 SEQ_GT(th->th_seq, sc->sc_irs + 1 + sc->sc_win)) { 3183 (void) syn_cache_respond(sc, m); 3184 splx(s); 3185 return ((struct socket *)(-1)); 3186 } 3187 3188 /* Remove this cache entry */ 3189 SYN_CACHE_RM(sc); 3190 splx(s); 3191 3192 /* 3193 * Ok, create the full blown connection, and set things up 3194 * as they would have been set up if we had created the 3195 * connection when the SYN arrived. If we can't create 3196 * the connection, abort it. 3197 */ 3198 /* 3199 * inp still has the OLD in_pcb stuff, set the 3200 * v6-related flags on the new guy, too. This is 3201 * done particularly for the case where an AF_INET6 3202 * socket is bound only to a port, and a v4 connection 3203 * comes in on that port. 3204 * we also copy the flowinfo from the original pcb 3205 * to the new one. 3206 */ 3207 oso = so; 3208 so = sonewconn(so, SS_ISCONNECTED); 3209 if (so == NULL) 3210 goto resetandabort; 3211 3212 switch (so->so_proto->pr_domain->dom_family) { 3213 #ifdef INET 3214 case AF_INET: 3215 inp = sotoinpcb(so); 3216 break; 3217 #endif 3218 #ifdef INET6 3219 case AF_INET6: 3220 in6p = sotoin6pcb(so); 3221 break; 3222 #endif 3223 } 3224 switch (src->sa_family) { 3225 #ifdef INET 3226 case AF_INET: 3227 if (inp) { 3228 inp->inp_laddr = ((struct sockaddr_in *)dst)->sin_addr; 3229 inp->inp_lport = ((struct sockaddr_in *)dst)->sin_port; 3230 inp->inp_options = ip_srcroute(); 3231 in_pcbstate(inp, INP_BOUND); 3232 if (inp->inp_options == NULL) { 3233 inp->inp_options = sc->sc_ipopts; 3234 sc->sc_ipopts = NULL; 3235 } 3236 } 3237 #ifdef INET6 3238 else if (in6p) { 3239 /* IPv4 packet to AF_INET6 socket */ 3240 bzero(&in6p->in6p_laddr, sizeof(in6p->in6p_laddr)); 3241 in6p->in6p_laddr.s6_addr16[5] = htons(0xffff); 3242 bcopy(&((struct sockaddr_in *)dst)->sin_addr, 3243 &in6p->in6p_laddr.s6_addr32[3], 3244 sizeof(((struct sockaddr_in *)dst)->sin_addr)); 3245 in6p->in6p_lport = ((struct sockaddr_in *)dst)->sin_port; 3246 in6totcpcb(in6p)->t_family = AF_INET; 3247 } 3248 #endif 3249 break; 3250 #endif 3251 #ifdef INET6 3252 case AF_INET6: 3253 if (in6p) { 3254 in6p->in6p_laddr = ((struct sockaddr_in6 *)dst)->sin6_addr; 3255 in6p->in6p_lport = ((struct sockaddr_in6 *)dst)->sin6_port; 3256 #if 0 3257 in6p->in6p_flowinfo = ip6->ip6_flow & IPV6_FLOWINFO_MASK; 3258 /*inp->inp_options = ip6_srcroute();*/ /* soon. */ 3259 #endif 3260 } 3261 break; 3262 #endif 3263 } 3264 #ifdef INET6 3265 if (in6p && in6totcpcb(in6p)->t_family == AF_INET6 && sotoinpcb(oso)) { 3266 struct in6pcb *oin6p = sotoin6pcb(oso); 3267 /* inherit socket options from the listening socket */ 3268 in6p->in6p_flags |= (oin6p->in6p_flags & IN6P_CONTROLOPTS); 3269 if (in6p->in6p_flags & IN6P_CONTROLOPTS) { 3270 m_freem(in6p->in6p_options); 3271 in6p->in6p_options = 0; 3272 } 3273 ip6_savecontrol(in6p, &in6p->in6p_options, 3274 mtod(m, struct ip6_hdr *), m); 3275 } 3276 #endif 3277 3278 #ifdef IPSEC 3279 /* 3280 * we make a copy of policy, instead of sharing the policy, 3281 * for better behavior in terms of SA lookup and dead SA removal. 3282 */ 3283 if (inp) { 3284 /* copy old policy into new socket's */ 3285 if (ipsec_copy_pcbpolicy(sotoinpcb(oso)->inp_sp, inp->inp_sp)) 3286 printf("tcp_input: could not copy policy\n"); 3287 } 3288 #ifdef INET6 3289 else if (in6p) { 3290 /* copy old policy into new socket's */ 3291 if (ipsec_copy_pcbpolicy(sotoin6pcb(oso)->in6p_sp, 3292 in6p->in6p_sp)) 3293 printf("tcp_input: could not copy policy\n"); 3294 } 3295 #endif 3296 #endif 3297 3298 /* 3299 * Give the new socket our cached route reference. 3300 */ 3301 if (inp) 3302 inp->inp_route = sc->sc_route4; /* struct assignment */ 3303 #ifdef INET6 3304 else 3305 in6p->in6p_route = sc->sc_route6; 3306 #endif 3307 sc->sc_route4.ro_rt = NULL; 3308 3309 am = m_get(M_DONTWAIT, MT_SONAME); /* XXX */ 3310 if (am == NULL) 3311 goto resetandabort; 3312 MCLAIM(am, &tcp_mowner); 3313 am->m_len = src->sa_len; 3314 bcopy(src, mtod(am, caddr_t), src->sa_len); 3315 if (inp) { 3316 if (in_pcbconnect(inp, am)) { 3317 (void) m_free(am); 3318 goto resetandabort; 3319 } 3320 } 3321 #ifdef INET6 3322 else if (in6p) { 3323 if (src->sa_family == AF_INET) { 3324 /* IPv4 packet to AF_INET6 socket */ 3325 struct sockaddr_in6 *sin6; 3326 sin6 = mtod(am, struct sockaddr_in6 *); 3327 am->m_len = sizeof(*sin6); 3328 bzero(sin6, sizeof(*sin6)); 3329 sin6->sin6_family = AF_INET6; 3330 sin6->sin6_len = sizeof(*sin6); 3331 sin6->sin6_port = ((struct sockaddr_in *)src)->sin_port; 3332 sin6->sin6_addr.s6_addr16[5] = htons(0xffff); 3333 bcopy(&((struct sockaddr_in *)src)->sin_addr, 3334 &sin6->sin6_addr.s6_addr32[3], 3335 sizeof(sin6->sin6_addr.s6_addr32[3])); 3336 } 3337 if (in6_pcbconnect(in6p, am)) { 3338 (void) m_free(am); 3339 goto resetandabort; 3340 } 3341 } 3342 #endif 3343 else { 3344 (void) m_free(am); 3345 goto resetandabort; 3346 } 3347 (void) m_free(am); 3348 3349 if (inp) 3350 tp = intotcpcb(inp); 3351 #ifdef INET6 3352 else if (in6p) 3353 tp = in6totcpcb(in6p); 3354 #endif 3355 else 3356 tp = NULL; 3357 tp->t_flags = sototcpcb(oso)->t_flags & TF_NODELAY; 3358 if (sc->sc_request_r_scale != 15) { 3359 tp->requested_s_scale = sc->sc_requested_s_scale; 3360 tp->request_r_scale = sc->sc_request_r_scale; 3361 tp->snd_scale = sc->sc_requested_s_scale; 3362 tp->rcv_scale = sc->sc_request_r_scale; 3363 tp->t_flags |= TF_REQ_SCALE|TF_RCVD_SCALE; 3364 } 3365 if (sc->sc_flags & SCF_TIMESTAMP) 3366 tp->t_flags |= TF_REQ_TSTMP|TF_RCVD_TSTMP; 3367 tp->ts_timebase = sc->sc_timebase; 3368 3369 tp->t_template = tcp_template(tp); 3370 if (tp->t_template == 0) { 3371 tp = tcp_drop(tp, ENOBUFS); /* destroys socket */ 3372 so = NULL; 3373 m_freem(m); 3374 goto abort; 3375 } 3376 3377 tp->iss = sc->sc_iss; 3378 tp->irs = sc->sc_irs; 3379 tcp_sendseqinit(tp); 3380 tcp_rcvseqinit(tp); 3381 tp->t_state = TCPS_SYN_RECEIVED; 3382 TCP_TIMER_ARM(tp, TCPT_KEEP, TCPTV_KEEP_INIT); 3383 tcpstat.tcps_accepts++; 3384 3385 /* Initialize tp->t_ourmss before we deal with the peer's! */ 3386 tp->t_ourmss = sc->sc_ourmaxseg; 3387 tcp_mss_from_peer(tp, sc->sc_peermaxseg); 3388 3389 /* 3390 * Initialize the initial congestion window. If we 3391 * had to retransmit the SYN,ACK, we must initialize cwnd 3392 * to 1 segment (i.e. the Loss Window). 3393 */ 3394 if (sc->sc_rxtshift) 3395 tp->snd_cwnd = tp->t_peermss; 3396 else { 3397 int ss = tcp_init_win; 3398 #ifdef INET 3399 if (inp != NULL && in_localaddr(inp->inp_faddr)) 3400 ss = tcp_init_win_local; 3401 #endif 3402 #ifdef INET6 3403 if (in6p != NULL && in6_localaddr(&in6p->in6p_faddr)) 3404 ss = tcp_init_win_local; 3405 #endif 3406 tp->snd_cwnd = TCP_INITIAL_WINDOW(ss, tp->t_peermss); 3407 } 3408 3409 tcp_rmx_rtt(tp); 3410 tp->snd_wl1 = sc->sc_irs; 3411 tp->rcv_up = sc->sc_irs + 1; 3412 3413 /* 3414 * This is what whould have happened in tcp_output() when 3415 * the SYN,ACK was sent. 3416 */ 3417 tp->snd_up = tp->snd_una; 3418 tp->snd_max = tp->snd_nxt = tp->iss+1; 3419 TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur); 3420 if (sc->sc_win > 0 && SEQ_GT(tp->rcv_nxt + sc->sc_win, tp->rcv_adv)) 3421 tp->rcv_adv = tp->rcv_nxt + sc->sc_win; 3422 tp->last_ack_sent = tp->rcv_nxt; 3423 3424 tcpstat.tcps_sc_completed++; 3425 SYN_CACHE_PUT(sc); 3426 return (so); 3427 3428 resetandabort: 3429 (void) tcp_respond(NULL, m, m, th, 3430 th->th_seq + tlen, (tcp_seq)0, TH_RST|TH_ACK); 3431 abort: 3432 if (so != NULL) 3433 (void) soabort(so); 3434 SYN_CACHE_PUT(sc); 3435 tcpstat.tcps_sc_aborted++; 3436 return ((struct socket *)(-1)); 3437 } 3438 3439 /* 3440 * This function is called when we get a RST for a 3441 * non-existent connection, so that we can see if the 3442 * connection is in the syn cache. If it is, zap it. 3443 */ 3444 3445 void 3446 syn_cache_reset(src, dst, th) 3447 struct sockaddr *src; 3448 struct sockaddr *dst; 3449 struct tcphdr *th; 3450 { 3451 struct syn_cache *sc; 3452 struct syn_cache_head *scp; 3453 int s = splsoftnet(); 3454 3455 if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) { 3456 splx(s); 3457 return; 3458 } 3459 if (SEQ_LT(th->th_seq, sc->sc_irs) || 3460 SEQ_GT(th->th_seq, sc->sc_irs+1)) { 3461 splx(s); 3462 return; 3463 } 3464 SYN_CACHE_RM(sc); 3465 splx(s); 3466 tcpstat.tcps_sc_reset++; 3467 SYN_CACHE_PUT(sc); 3468 } 3469 3470 void 3471 syn_cache_unreach(src, dst, th) 3472 struct sockaddr *src; 3473 struct sockaddr *dst; 3474 struct tcphdr *th; 3475 { 3476 struct syn_cache *sc; 3477 struct syn_cache_head *scp; 3478 int s; 3479 3480 s = splsoftnet(); 3481 if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) { 3482 splx(s); 3483 return; 3484 } 3485 /* If the sequence number != sc_iss, then it's a bogus ICMP msg */ 3486 if (ntohl (th->th_seq) != sc->sc_iss) { 3487 splx(s); 3488 return; 3489 } 3490 3491 /* 3492 * If we've rertransmitted 3 times and this is our second error, 3493 * we remove the entry. Otherwise, we allow it to continue on. 3494 * This prevents us from incorrectly nuking an entry during a 3495 * spurious network outage. 3496 * 3497 * See tcp_notify(). 3498 */ 3499 if ((sc->sc_flags & SCF_UNREACH) == 0 || sc->sc_rxtshift < 3) { 3500 sc->sc_flags |= SCF_UNREACH; 3501 splx(s); 3502 return; 3503 } 3504 3505 SYN_CACHE_RM(sc); 3506 splx(s); 3507 tcpstat.tcps_sc_unreach++; 3508 SYN_CACHE_PUT(sc); 3509 } 3510 3511 /* 3512 * Given a LISTEN socket and an inbound SYN request, add 3513 * this to the syn cache, and send back a segment: 3514 * <SEQ=ISS><ACK=RCV_NXT><CTL=SYN,ACK> 3515 * to the source. 3516 * 3517 * IMPORTANT NOTE: We do _NOT_ ACK data that might accompany the SYN. 3518 * Doing so would require that we hold onto the data and deliver it 3519 * to the application. However, if we are the target of a SYN-flood 3520 * DoS attack, an attacker could send data which would eventually 3521 * consume all available buffer space if it were ACKed. By not ACKing 3522 * the data, we avoid this DoS scenario. 3523 */ 3524 3525 int 3526 syn_cache_add(src, dst, th, hlen, so, m, optp, optlen, oi) 3527 struct sockaddr *src; 3528 struct sockaddr *dst; 3529 struct tcphdr *th; 3530 unsigned int hlen; 3531 struct socket *so; 3532 struct mbuf *m; 3533 u_char *optp; 3534 int optlen; 3535 struct tcp_opt_info *oi; 3536 { 3537 struct tcpcb tb, *tp; 3538 long win; 3539 struct syn_cache *sc; 3540 struct syn_cache_head *scp; 3541 struct mbuf *ipopts; 3542 3543 tp = sototcpcb(so); 3544 3545 /* 3546 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN 3547 * 3548 * Note this check is performed in tcp_input() very early on. 3549 */ 3550 3551 /* 3552 * Initialize some local state. 3553 */ 3554 win = sbspace(&so->so_rcv); 3555 if (win > TCP_MAXWIN) 3556 win = TCP_MAXWIN; 3557 3558 switch (src->sa_family) { 3559 #ifdef INET 3560 case AF_INET: 3561 /* 3562 * Remember the IP options, if any. 3563 */ 3564 ipopts = ip_srcroute(); 3565 break; 3566 #endif 3567 default: 3568 ipopts = NULL; 3569 } 3570 3571 if (optp) { 3572 tb.t_flags = tcp_do_rfc1323 ? (TF_REQ_SCALE|TF_REQ_TSTMP) : 0; 3573 tcp_dooptions(&tb, optp, optlen, th, oi); 3574 } else 3575 tb.t_flags = 0; 3576 3577 /* 3578 * See if we already have an entry for this connection. 3579 * If we do, resend the SYN,ACK. We do not count this 3580 * as a retransmission (XXX though maybe we should). 3581 */ 3582 if ((sc = syn_cache_lookup(src, dst, &scp)) != NULL) { 3583 tcpstat.tcps_sc_dupesyn++; 3584 if (ipopts) { 3585 /* 3586 * If we were remembering a previous source route, 3587 * forget it and use the new one we've been given. 3588 */ 3589 if (sc->sc_ipopts) 3590 (void) m_free(sc->sc_ipopts); 3591 sc->sc_ipopts = ipopts; 3592 } 3593 sc->sc_timestamp = tb.ts_recent; 3594 if (syn_cache_respond(sc, m) == 0) { 3595 tcpstat.tcps_sndacks++; 3596 tcpstat.tcps_sndtotal++; 3597 } 3598 return (1); 3599 } 3600 3601 sc = pool_get(&syn_cache_pool, PR_NOWAIT); 3602 if (sc == NULL) { 3603 if (ipopts) 3604 (void) m_free(ipopts); 3605 return (0); 3606 } 3607 3608 /* 3609 * Fill in the cache, and put the necessary IP and TCP 3610 * options into the reply. 3611 */ 3612 bzero(sc, sizeof(struct syn_cache)); 3613 callout_init(&sc->sc_timer); 3614 bcopy(src, &sc->sc_src, src->sa_len); 3615 bcopy(dst, &sc->sc_dst, dst->sa_len); 3616 sc->sc_flags = 0; 3617 sc->sc_ipopts = ipopts; 3618 sc->sc_irs = th->th_seq; 3619 switch (src->sa_family) { 3620 #ifdef INET 3621 case AF_INET: 3622 { 3623 struct sockaddr_in *srcin = (void *) src; 3624 struct sockaddr_in *dstin = (void *) dst; 3625 3626 sc->sc_iss = tcp_new_iss1(&dstin->sin_addr, 3627 &srcin->sin_addr, dstin->sin_port, 3628 srcin->sin_port, sizeof(dstin->sin_addr), 0); 3629 break; 3630 } 3631 #endif /* INET */ 3632 #ifdef INET6 3633 case AF_INET6: 3634 { 3635 struct sockaddr_in6 *srcin6 = (void *) src; 3636 struct sockaddr_in6 *dstin6 = (void *) dst; 3637 3638 sc->sc_iss = tcp_new_iss1(&dstin6->sin6_addr, 3639 &srcin6->sin6_addr, dstin6->sin6_port, 3640 srcin6->sin6_port, sizeof(dstin6->sin6_addr), 0); 3641 break; 3642 } 3643 #endif /* INET6 */ 3644 } 3645 sc->sc_peermaxseg = oi->maxseg; 3646 sc->sc_ourmaxseg = tcp_mss_to_advertise(m->m_flags & M_PKTHDR ? 3647 m->m_pkthdr.rcvif : NULL, 3648 sc->sc_src.sa.sa_family); 3649 sc->sc_win = win; 3650 sc->sc_timebase = tcp_now; /* see tcp_newtcpcb() */ 3651 sc->sc_timestamp = tb.ts_recent; 3652 if ((tb.t_flags & (TF_REQ_TSTMP|TF_RCVD_TSTMP)) == 3653 (TF_REQ_TSTMP|TF_RCVD_TSTMP)) 3654 sc->sc_flags |= SCF_TIMESTAMP; 3655 if ((tb.t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) == 3656 (TF_RCVD_SCALE|TF_REQ_SCALE)) { 3657 sc->sc_requested_s_scale = tb.requested_s_scale; 3658 sc->sc_request_r_scale = 0; 3659 while (sc->sc_request_r_scale < TCP_MAX_WINSHIFT && 3660 TCP_MAXWIN << sc->sc_request_r_scale < 3661 so->so_rcv.sb_hiwat) 3662 sc->sc_request_r_scale++; 3663 } else { 3664 sc->sc_requested_s_scale = 15; 3665 sc->sc_request_r_scale = 15; 3666 } 3667 sc->sc_tp = tp; 3668 if (syn_cache_respond(sc, m) == 0) { 3669 syn_cache_insert(sc, tp); 3670 tcpstat.tcps_sndacks++; 3671 tcpstat.tcps_sndtotal++; 3672 } else { 3673 SYN_CACHE_PUT(sc); 3674 tcpstat.tcps_sc_dropped++; 3675 } 3676 return (1); 3677 } 3678 3679 int 3680 syn_cache_respond(sc, m) 3681 struct syn_cache *sc; 3682 struct mbuf *m; 3683 { 3684 struct route *ro; 3685 u_int8_t *optp; 3686 int optlen, error; 3687 u_int16_t tlen; 3688 struct ip *ip = NULL; 3689 #ifdef INET6 3690 struct ip6_hdr *ip6 = NULL; 3691 #endif 3692 struct tcphdr *th; 3693 u_int hlen; 3694 3695 switch (sc->sc_src.sa.sa_family) { 3696 case AF_INET: 3697 hlen = sizeof(struct ip); 3698 ro = &sc->sc_route4; 3699 break; 3700 #ifdef INET6 3701 case AF_INET6: 3702 hlen = sizeof(struct ip6_hdr); 3703 ro = (struct route *)&sc->sc_route6; 3704 break; 3705 #endif 3706 default: 3707 if (m) 3708 m_freem(m); 3709 return EAFNOSUPPORT; 3710 } 3711 3712 /* Compute the size of the TCP options. */ 3713 optlen = 4 + (sc->sc_request_r_scale != 15 ? 4 : 0) + 3714 ((sc->sc_flags & SCF_TIMESTAMP) ? TCPOLEN_TSTAMP_APPA : 0); 3715 3716 tlen = hlen + sizeof(struct tcphdr) + optlen; 3717 3718 /* 3719 * Create the IP+TCP header from scratch. 3720 */ 3721 if (m) 3722 m_freem(m); 3723 #ifdef DIAGNOSTIC 3724 if (max_linkhdr + tlen > MCLBYTES) 3725 return (ENOBUFS); 3726 #endif 3727 MGETHDR(m, M_DONTWAIT, MT_DATA); 3728 if (m && tlen > MHLEN) { 3729 MCLGET(m, M_DONTWAIT); 3730 if ((m->m_flags & M_EXT) == 0) { 3731 m_freem(m); 3732 m = NULL; 3733 } 3734 } 3735 if (m == NULL) 3736 return (ENOBUFS); 3737 MCLAIM(m, &tcp_tx_mowner); 3738 3739 /* Fixup the mbuf. */ 3740 m->m_data += max_linkhdr; 3741 m->m_len = m->m_pkthdr.len = tlen; 3742 #ifdef IPSEC 3743 if (sc->sc_tp) { 3744 struct tcpcb *tp; 3745 struct socket *so; 3746 3747 tp = sc->sc_tp; 3748 if (tp->t_inpcb) 3749 so = tp->t_inpcb->inp_socket; 3750 #ifdef INET6 3751 else if (tp->t_in6pcb) 3752 so = tp->t_in6pcb->in6p_socket; 3753 #endif 3754 else 3755 so = NULL; 3756 /* use IPsec policy on listening socket, on SYN ACK */ 3757 if (ipsec_setsocket(m, so) != 0) { 3758 m_freem(m); 3759 return ENOBUFS; 3760 } 3761 } 3762 #endif 3763 m->m_pkthdr.rcvif = NULL; 3764 memset(mtod(m, u_char *), 0, tlen); 3765 3766 switch (sc->sc_src.sa.sa_family) { 3767 case AF_INET: 3768 ip = mtod(m, struct ip *); 3769 ip->ip_dst = sc->sc_src.sin.sin_addr; 3770 ip->ip_src = sc->sc_dst.sin.sin_addr; 3771 ip->ip_p = IPPROTO_TCP; 3772 th = (struct tcphdr *)(ip + 1); 3773 th->th_dport = sc->sc_src.sin.sin_port; 3774 th->th_sport = sc->sc_dst.sin.sin_port; 3775 break; 3776 #ifdef INET6 3777 case AF_INET6: 3778 ip6 = mtod(m, struct ip6_hdr *); 3779 ip6->ip6_dst = sc->sc_src.sin6.sin6_addr; 3780 ip6->ip6_src = sc->sc_dst.sin6.sin6_addr; 3781 ip6->ip6_nxt = IPPROTO_TCP; 3782 /* ip6_plen will be updated in ip6_output() */ 3783 th = (struct tcphdr *)(ip6 + 1); 3784 th->th_dport = sc->sc_src.sin6.sin6_port; 3785 th->th_sport = sc->sc_dst.sin6.sin6_port; 3786 break; 3787 #endif 3788 default: 3789 th = NULL; 3790 } 3791 3792 th->th_seq = htonl(sc->sc_iss); 3793 th->th_ack = htonl(sc->sc_irs + 1); 3794 th->th_off = (sizeof(struct tcphdr) + optlen) >> 2; 3795 th->th_flags = TH_SYN|TH_ACK; 3796 th->th_win = htons(sc->sc_win); 3797 /* th_sum already 0 */ 3798 /* th_urp already 0 */ 3799 3800 /* Tack on the TCP options. */ 3801 optp = (u_int8_t *)(th + 1); 3802 *optp++ = TCPOPT_MAXSEG; 3803 *optp++ = 4; 3804 *optp++ = (sc->sc_ourmaxseg >> 8) & 0xff; 3805 *optp++ = sc->sc_ourmaxseg & 0xff; 3806 3807 if (sc->sc_request_r_scale != 15) { 3808 *((u_int32_t *)optp) = htonl(TCPOPT_NOP << 24 | 3809 TCPOPT_WINDOW << 16 | TCPOLEN_WINDOW << 8 | 3810 sc->sc_request_r_scale); 3811 optp += 4; 3812 } 3813 3814 if (sc->sc_flags & SCF_TIMESTAMP) { 3815 u_int32_t *lp = (u_int32_t *)(optp); 3816 /* Form timestamp option as shown in appendix A of RFC 1323. */ 3817 *lp++ = htonl(TCPOPT_TSTAMP_HDR); 3818 *lp++ = htonl(SYN_CACHE_TIMESTAMP(sc)); 3819 *lp = htonl(sc->sc_timestamp); 3820 optp += TCPOLEN_TSTAMP_APPA; 3821 } 3822 3823 /* Compute the packet's checksum. */ 3824 switch (sc->sc_src.sa.sa_family) { 3825 case AF_INET: 3826 ip->ip_len = htons(tlen - hlen); 3827 th->th_sum = 0; 3828 th->th_sum = in_cksum(m, tlen); 3829 break; 3830 #ifdef INET6 3831 case AF_INET6: 3832 ip6->ip6_plen = htons(tlen - hlen); 3833 th->th_sum = 0; 3834 th->th_sum = in6_cksum(m, IPPROTO_TCP, hlen, tlen - hlen); 3835 break; 3836 #endif 3837 } 3838 3839 /* 3840 * Fill in some straggling IP bits. Note the stack expects 3841 * ip_len to be in host order, for convenience. 3842 */ 3843 switch (sc->sc_src.sa.sa_family) { 3844 #ifdef INET 3845 case AF_INET: 3846 ip->ip_len = htons(tlen); 3847 ip->ip_ttl = ip_defttl; 3848 /* XXX tos? */ 3849 break; 3850 #endif 3851 #ifdef INET6 3852 case AF_INET6: 3853 ip6->ip6_vfc &= ~IPV6_VERSION_MASK; 3854 ip6->ip6_vfc |= IPV6_VERSION; 3855 ip6->ip6_plen = htons(tlen - hlen); 3856 /* ip6_hlim will be initialized afterwards */ 3857 /* XXX flowlabel? */ 3858 break; 3859 #endif 3860 } 3861 3862 switch (sc->sc_src.sa.sa_family) { 3863 #ifdef INET 3864 case AF_INET: 3865 error = ip_output(m, sc->sc_ipopts, ro, 3866 (ip_mtudisc ? IP_MTUDISC : 0), 3867 NULL); 3868 break; 3869 #endif 3870 #ifdef INET6 3871 case AF_INET6: 3872 ip6->ip6_hlim = in6_selecthlim(NULL, 3873 ro->ro_rt ? ro->ro_rt->rt_ifp : NULL); 3874 3875 error = ip6_output(m, NULL /*XXX*/, (struct route_in6 *)ro, 3876 0, NULL, NULL); 3877 break; 3878 #endif 3879 default: 3880 error = EAFNOSUPPORT; 3881 break; 3882 } 3883 return (error); 3884 } 3885
Indexes created Thu Sep 25 08:09:54 GMT 2025