tcp_input.c revision 1.298
1 /* $NetBSD: tcp_input.c,v 1.298 2009/07/18 23:09:53 minskim 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, 2005, 2006 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 * This code is derived from software contributed to The NetBSD Foundation 80 * by Charles M. Hannum. 81 * This code is derived from software contributed to The NetBSD Foundation 82 * by Rui Paulo. 83 * 84 * Redistribution and use in source and binary forms, with or without 85 * modification, are permitted provided that the following conditions 86 * are met: 87 * 1. Redistributions of source code must retain the above copyright 88 * notice, this list of conditions and the following disclaimer. 89 * 2. Redistributions in binary form must reproduce the above copyright 90 * notice, this list of conditions and the following disclaimer in the 91 * documentation and/or other materials provided with the distribution. 92 * 93 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS 94 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED 95 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR 96 * PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS 97 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR 98 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF 99 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS 100 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN 101 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) 102 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE 103 * POSSIBILITY OF SUCH DAMAGE. 104 */ 105 106 /* 107 * Copyright (c) 1982, 1986, 1988, 1990, 1993, 1994, 1995 108 * The Regents of the University of California. All rights reserved. 109 * 110 * Redistribution and use in source and binary forms, with or without 111 * modification, are permitted provided that the following conditions 112 * are met: 113 * 1. Redistributions of source code must retain the above copyright 114 * notice, this list of conditions and the following disclaimer. 115 * 2. Redistributions in binary form must reproduce the above copyright 116 * notice, this list of conditions and the following disclaimer in the 117 * documentation and/or other materials provided with the distribution. 118 * 3. Neither the name of the University nor the names of its contributors 119 * may be used to endorse or promote products derived from this software 120 * without specific prior written permission. 121 * 122 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND 123 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE 124 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE 125 * ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE 126 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL 127 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS 128 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) 129 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT 130 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY 131 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF 132 * SUCH DAMAGE. 133 * 134 * @(#)tcp_input.c 8.12 (Berkeley) 5/24/95 135 */ 136 137 /* 138 * TODO list for SYN cache stuff: 139 * 140 * Find room for a "state" field, which is needed to keep a 141 * compressed state for TIME_WAIT TCBs. It's been noted already 142 * that this is fairly important for very high-volume web and 143 * mail servers, which use a large number of short-lived 144 * connections. 145 */ 146 147 #include <sys/cdefs.h> 148 __KERNEL_RCSID(0, "$NetBSD: tcp_input.c,v 1.298 2009/07/18 23:09:53 minskim Exp $"); 149 150 #include "opt_inet.h" 151 #include "opt_ipsec.h" 152 #include "opt_inet_csum.h" 153 #include "opt_tcp_debug.h" 154 155 #include <sys/param.h> 156 #include <sys/systm.h> 157 #include <sys/malloc.h> 158 #include <sys/mbuf.h> 159 #include <sys/protosw.h> 160 #include <sys/socket.h> 161 #include <sys/socketvar.h> 162 #include <sys/errno.h> 163 #include <sys/syslog.h> 164 #include <sys/pool.h> 165 #include <sys/domain.h> 166 #include <sys/kernel.h> 167 #ifdef TCP_SIGNATURE 168 #include <sys/md5.h> 169 #endif 170 #include <sys/lwp.h> /* for lwp0 */ 171 172 #include <net/if.h> 173 #include <net/route.h> 174 #include <net/if_types.h> 175 176 #include <netinet/in.h> 177 #include <netinet/in_systm.h> 178 #include <netinet/ip.h> 179 #include <netinet/in_pcb.h> 180 #include <netinet/in_var.h> 181 #include <netinet/ip_var.h> 182 #include <netinet/in_offload.h> 183 184 #ifdef INET6 185 #ifndef INET 186 #include <netinet/in.h> 187 #endif 188 #include <netinet/ip6.h> 189 #include <netinet6/ip6_var.h> 190 #include <netinet6/in6_pcb.h> 191 #include <netinet6/ip6_var.h> 192 #include <netinet6/in6_var.h> 193 #include <netinet/icmp6.h> 194 #include <netinet6/nd6.h> 195 #ifdef TCP_SIGNATURE 196 #include <netinet6/scope6_var.h> 197 #endif 198 #endif 199 200 #ifndef INET6 201 /* always need ip6.h for IP6_EXTHDR_GET */ 202 #include <netinet/ip6.h> 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/tcp_private.h> 211 #include <netinet/tcpip.h> 212 #include <netinet/tcp_congctl.h> 213 #include <netinet/tcp_debug.h> 214 215 #include <machine/stdarg.h> 216 217 #ifdef IPSEC 218 #include <netinet6/ipsec.h> 219 #include <netinet6/ipsec_private.h> 220 #include <netkey/key.h> 221 #endif /*IPSEC*/ 222 #ifdef INET6 223 #include "faith.h" 224 #if defined(NFAITH) && NFAITH > 0 225 #include <net/if_faith.h> 226 #endif 227 #endif /* IPSEC */ 228 229 #ifdef FAST_IPSEC 230 #include <netipsec/ipsec.h> 231 #include <netipsec/ipsec_var.h> 232 #include <netipsec/ipsec_private.h> 233 #include <netipsec/key.h> 234 #ifdef INET6 235 #include <netipsec/ipsec6.h> 236 #endif 237 #endif /* FAST_IPSEC*/ 238 239 int tcprexmtthresh = 3; 240 int tcp_log_refused; 241 242 int tcp_do_autorcvbuf = 0; 243 int tcp_autorcvbuf_inc = 16 * 1024; 244 int tcp_autorcvbuf_max = 256 * 1024; 245 246 static int tcp_rst_ppslim_count = 0; 247 static struct timeval tcp_rst_ppslim_last; 248 static int tcp_ackdrop_ppslim_count = 0; 249 static struct timeval tcp_ackdrop_ppslim_last; 250 251 #define TCP_PAWS_IDLE (24U * 24 * 60 * 60 * PR_SLOWHZ) 252 253 /* for modulo comparisons of timestamps */ 254 #define TSTMP_LT(a,b) ((int)((a)-(b)) < 0) 255 #define TSTMP_GEQ(a,b) ((int)((a)-(b)) >= 0) 256 257 /* 258 * Neighbor Discovery, Neighbor Unreachability Detection Upper layer hint. 259 */ 260 #ifdef INET6 261 static inline void 262 nd6_hint(struct tcpcb *tp) 263 { 264 struct rtentry *rt; 265 266 if (tp != NULL && tp->t_in6pcb != NULL && tp->t_family == AF_INET6 && 267 (rt = rtcache_validate(&tp->t_in6pcb->in6p_route)) != NULL) 268 nd6_nud_hint(rt, NULL, 0); 269 } 270 #else 271 static inline void 272 nd6_hint(struct tcpcb *tp) 273 { 274 } 275 #endif 276 277 /* 278 * Compute ACK transmission behavior. Delay the ACK unless 279 * we have already delayed an ACK (must send an ACK every two segments). 280 * We also ACK immediately if we received a PUSH and the ACK-on-PUSH 281 * option is enabled. 282 */ 283 static void 284 tcp_setup_ack(struct tcpcb *tp, const struct tcphdr *th) 285 { 286 287 if (tp->t_flags & TF_DELACK || 288 (tcp_ack_on_push && th->th_flags & TH_PUSH)) 289 tp->t_flags |= TF_ACKNOW; 290 else 291 TCP_SET_DELACK(tp); 292 } 293 294 static void 295 icmp_check(struct tcpcb *tp, const struct tcphdr *th, int acked) 296 { 297 298 /* 299 * If we had a pending ICMP message that refers to data that have 300 * just been acknowledged, disregard the recorded ICMP message. 301 */ 302 if ((tp->t_flags & TF_PMTUD_PEND) && 303 SEQ_GT(th->th_ack, tp->t_pmtud_th_seq)) 304 tp->t_flags &= ~TF_PMTUD_PEND; 305 306 /* 307 * Keep track of the largest chunk of data 308 * acknowledged since last PMTU update 309 */ 310 if (tp->t_pmtud_mss_acked < acked) 311 tp->t_pmtud_mss_acked = acked; 312 } 313 314 /* 315 * Convert TCP protocol fields to host order for easier processing. 316 */ 317 static void 318 tcp_fields_to_host(struct tcphdr *th) 319 { 320 321 NTOHL(th->th_seq); 322 NTOHL(th->th_ack); 323 NTOHS(th->th_win); 324 NTOHS(th->th_urp); 325 } 326 327 /* 328 * ... and reverse the above. 329 */ 330 static void 331 tcp_fields_to_net(struct tcphdr *th) 332 { 333 334 HTONL(th->th_seq); 335 HTONL(th->th_ack); 336 HTONS(th->th_win); 337 HTONS(th->th_urp); 338 } 339 340 #ifdef TCP_CSUM_COUNTERS 341 #include <sys/device.h> 342 343 #if defined(INET) 344 extern struct evcnt tcp_hwcsum_ok; 345 extern struct evcnt tcp_hwcsum_bad; 346 extern struct evcnt tcp_hwcsum_data; 347 extern struct evcnt tcp_swcsum; 348 #endif /* defined(INET) */ 349 #if defined(INET6) 350 extern struct evcnt tcp6_hwcsum_ok; 351 extern struct evcnt tcp6_hwcsum_bad; 352 extern struct evcnt tcp6_hwcsum_data; 353 extern struct evcnt tcp6_swcsum; 354 #endif /* defined(INET6) */ 355 356 #define TCP_CSUM_COUNTER_INCR(ev) (ev)->ev_count++ 357 358 #else 359 360 #define TCP_CSUM_COUNTER_INCR(ev) /* nothing */ 361 362 #endif /* TCP_CSUM_COUNTERS */ 363 364 #ifdef TCP_REASS_COUNTERS 365 #include <sys/device.h> 366 367 extern struct evcnt tcp_reass_; 368 extern struct evcnt tcp_reass_empty; 369 extern struct evcnt tcp_reass_iteration[8]; 370 extern struct evcnt tcp_reass_prependfirst; 371 extern struct evcnt tcp_reass_prepend; 372 extern struct evcnt tcp_reass_insert; 373 extern struct evcnt tcp_reass_inserttail; 374 extern struct evcnt tcp_reass_append; 375 extern struct evcnt tcp_reass_appendtail; 376 extern struct evcnt tcp_reass_overlaptail; 377 extern struct evcnt tcp_reass_overlapfront; 378 extern struct evcnt tcp_reass_segdup; 379 extern struct evcnt tcp_reass_fragdup; 380 381 #define TCP_REASS_COUNTER_INCR(ev) (ev)->ev_count++ 382 383 #else 384 385 #define TCP_REASS_COUNTER_INCR(ev) /* nothing */ 386 387 #endif /* TCP_REASS_COUNTERS */ 388 389 static int tcp_reass(struct tcpcb *, const struct tcphdr *, struct mbuf *, 390 int *); 391 static int tcp_dooptions(struct tcpcb *, const u_char *, int, 392 struct tcphdr *, struct mbuf *, int, struct tcp_opt_info *); 393 394 #ifdef INET 395 static void tcp4_log_refused(const struct ip *, const struct tcphdr *); 396 #endif 397 #ifdef INET6 398 static void tcp6_log_refused(const struct ip6_hdr *, const struct tcphdr *); 399 #endif 400 401 #define TRAVERSE(x) while ((x)->m_next) (x) = (x)->m_next 402 403 #if defined(MBUFTRACE) 404 struct mowner tcp_reass_mowner = MOWNER_INIT("tcp", "reass"); 405 #endif /* defined(MBUFTRACE) */ 406 407 static struct pool tcpipqent_pool; 408 409 void 410 tcpipqent_init(void) 411 { 412 413 pool_init(&tcpipqent_pool, sizeof(struct ipqent), 0, 0, 0, "tcpipqepl", 414 NULL, IPL_VM); 415 } 416 417 struct ipqent * 418 tcpipqent_alloc(void) 419 { 420 struct ipqent *ipqe; 421 int s; 422 423 s = splvm(); 424 ipqe = pool_get(&tcpipqent_pool, PR_NOWAIT); 425 splx(s); 426 427 return ipqe; 428 } 429 430 void 431 tcpipqent_free(struct ipqent *ipqe) 432 { 433 int s; 434 435 s = splvm(); 436 pool_put(&tcpipqent_pool, ipqe); 437 splx(s); 438 } 439 440 static int 441 tcp_reass(struct tcpcb *tp, const struct tcphdr *th, struct mbuf *m, int *tlen) 442 { 443 struct ipqent *p, *q, *nq, *tiqe = NULL; 444 struct socket *so = NULL; 445 int pkt_flags; 446 tcp_seq pkt_seq; 447 unsigned pkt_len; 448 u_long rcvpartdupbyte = 0; 449 u_long rcvoobyte; 450 #ifdef TCP_REASS_COUNTERS 451 u_int count = 0; 452 #endif 453 uint64_t *tcps; 454 455 if (tp->t_inpcb) 456 so = tp->t_inpcb->inp_socket; 457 #ifdef INET6 458 else if (tp->t_in6pcb) 459 so = tp->t_in6pcb->in6p_socket; 460 #endif 461 462 TCP_REASS_LOCK_CHECK(tp); 463 464 /* 465 * Call with th==0 after become established to 466 * force pre-ESTABLISHED data up to user socket. 467 */ 468 if (th == 0) 469 goto present; 470 471 m_claimm(m, &tcp_reass_mowner); 472 473 rcvoobyte = *tlen; 474 /* 475 * Copy these to local variables because the tcpiphdr 476 * gets munged while we are collapsing mbufs. 477 */ 478 pkt_seq = th->th_seq; 479 pkt_len = *tlen; 480 pkt_flags = th->th_flags; 481 482 TCP_REASS_COUNTER_INCR(&tcp_reass_); 483 484 if ((p = TAILQ_LAST(&tp->segq, ipqehead)) != NULL) { 485 /* 486 * When we miss a packet, the vast majority of time we get 487 * packets that follow it in order. So optimize for that. 488 */ 489 if (pkt_seq == p->ipqe_seq + p->ipqe_len) { 490 p->ipqe_len += pkt_len; 491 p->ipqe_flags |= pkt_flags; 492 m_cat(p->ipre_mlast, m); 493 TRAVERSE(p->ipre_mlast); 494 m = NULL; 495 tiqe = p; 496 TAILQ_REMOVE(&tp->timeq, p, ipqe_timeq); 497 TCP_REASS_COUNTER_INCR(&tcp_reass_appendtail); 498 goto skip_replacement; 499 } 500 /* 501 * While we're here, if the pkt is completely beyond 502 * anything we have, just insert it at the tail. 503 */ 504 if (SEQ_GT(pkt_seq, p->ipqe_seq + p->ipqe_len)) { 505 TCP_REASS_COUNTER_INCR(&tcp_reass_inserttail); 506 goto insert_it; 507 } 508 } 509 510 q = TAILQ_FIRST(&tp->segq); 511 512 if (q != NULL) { 513 /* 514 * If this segment immediately precedes the first out-of-order 515 * block, simply slap the segment in front of it and (mostly) 516 * skip the complicated logic. 517 */ 518 if (pkt_seq + pkt_len == q->ipqe_seq) { 519 q->ipqe_seq = pkt_seq; 520 q->ipqe_len += pkt_len; 521 q->ipqe_flags |= pkt_flags; 522 m_cat(m, q->ipqe_m); 523 q->ipqe_m = m; 524 q->ipre_mlast = m; /* last mbuf may have changed */ 525 TRAVERSE(q->ipre_mlast); 526 tiqe = q; 527 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 528 TCP_REASS_COUNTER_INCR(&tcp_reass_prependfirst); 529 goto skip_replacement; 530 } 531 } else { 532 TCP_REASS_COUNTER_INCR(&tcp_reass_empty); 533 } 534 535 /* 536 * Find a segment which begins after this one does. 537 */ 538 for (p = NULL; q != NULL; q = nq) { 539 nq = TAILQ_NEXT(q, ipqe_q); 540 #ifdef TCP_REASS_COUNTERS 541 count++; 542 #endif 543 /* 544 * If the received segment is just right after this 545 * fragment, merge the two together and then check 546 * for further overlaps. 547 */ 548 if (q->ipqe_seq + q->ipqe_len == pkt_seq) { 549 #ifdef TCPREASS_DEBUG 550 printf("tcp_reass[%p]: concat %u:%u(%u) to %u:%u(%u)\n", 551 tp, pkt_seq, pkt_seq + pkt_len, pkt_len, 552 q->ipqe_seq, q->ipqe_seq + q->ipqe_len, q->ipqe_len); 553 #endif 554 pkt_len += q->ipqe_len; 555 pkt_flags |= q->ipqe_flags; 556 pkt_seq = q->ipqe_seq; 557 m_cat(q->ipre_mlast, m); 558 TRAVERSE(q->ipre_mlast); 559 m = q->ipqe_m; 560 TCP_REASS_COUNTER_INCR(&tcp_reass_append); 561 goto free_ipqe; 562 } 563 /* 564 * If the received segment is completely past this 565 * fragment, we need to go the next fragment. 566 */ 567 if (SEQ_LT(q->ipqe_seq + q->ipqe_len, pkt_seq)) { 568 p = q; 569 continue; 570 } 571 /* 572 * If the fragment is past the received segment, 573 * it (or any following) can't be concatenated. 574 */ 575 if (SEQ_GT(q->ipqe_seq, pkt_seq + pkt_len)) { 576 TCP_REASS_COUNTER_INCR(&tcp_reass_insert); 577 break; 578 } 579 580 /* 581 * We've received all the data in this segment before. 582 * mark it as a duplicate and return. 583 */ 584 if (SEQ_LEQ(q->ipqe_seq, pkt_seq) && 585 SEQ_GEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) { 586 tcps = TCP_STAT_GETREF(); 587 tcps[TCP_STAT_RCVDUPPACK]++; 588 tcps[TCP_STAT_RCVDUPBYTE] += pkt_len; 589 TCP_STAT_PUTREF(); 590 tcp_new_dsack(tp, pkt_seq, pkt_len); 591 m_freem(m); 592 if (tiqe != NULL) { 593 tcpipqent_free(tiqe); 594 } 595 TCP_REASS_COUNTER_INCR(&tcp_reass_segdup); 596 return (0); 597 } 598 /* 599 * Received segment completely overlaps this fragment 600 * so we drop the fragment (this keeps the temporal 601 * ordering of segments correct). 602 */ 603 if (SEQ_GEQ(q->ipqe_seq, pkt_seq) && 604 SEQ_LEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) { 605 rcvpartdupbyte += q->ipqe_len; 606 m_freem(q->ipqe_m); 607 TCP_REASS_COUNTER_INCR(&tcp_reass_fragdup); 608 goto free_ipqe; 609 } 610 /* 611 * RX'ed segment extends past the end of the 612 * fragment. Drop the overlapping bytes. Then 613 * merge the fragment and segment then treat as 614 * a longer received packet. 615 */ 616 if (SEQ_LT(q->ipqe_seq, pkt_seq) && 617 SEQ_GT(q->ipqe_seq + q->ipqe_len, pkt_seq)) { 618 int overlap = q->ipqe_seq + q->ipqe_len - pkt_seq; 619 #ifdef TCPREASS_DEBUG 620 printf("tcp_reass[%p]: trim starting %d bytes of %u:%u(%u)\n", 621 tp, overlap, 622 pkt_seq, pkt_seq + pkt_len, pkt_len); 623 #endif 624 m_adj(m, overlap); 625 rcvpartdupbyte += overlap; 626 m_cat(q->ipre_mlast, m); 627 TRAVERSE(q->ipre_mlast); 628 m = q->ipqe_m; 629 pkt_seq = q->ipqe_seq; 630 pkt_len += q->ipqe_len - overlap; 631 rcvoobyte -= overlap; 632 TCP_REASS_COUNTER_INCR(&tcp_reass_overlaptail); 633 goto free_ipqe; 634 } 635 /* 636 * RX'ed segment extends past the front of the 637 * fragment. Drop the overlapping bytes on the 638 * received packet. The packet will then be 639 * contatentated with this fragment a bit later. 640 */ 641 if (SEQ_GT(q->ipqe_seq, pkt_seq) && 642 SEQ_LT(q->ipqe_seq, pkt_seq + pkt_len)) { 643 int overlap = pkt_seq + pkt_len - q->ipqe_seq; 644 #ifdef TCPREASS_DEBUG 645 printf("tcp_reass[%p]: trim trailing %d bytes of %u:%u(%u)\n", 646 tp, overlap, 647 pkt_seq, pkt_seq + pkt_len, pkt_len); 648 #endif 649 m_adj(m, -overlap); 650 pkt_len -= overlap; 651 rcvpartdupbyte += overlap; 652 TCP_REASS_COUNTER_INCR(&tcp_reass_overlapfront); 653 rcvoobyte -= overlap; 654 } 655 /* 656 * If the received segment immediates precedes this 657 * fragment then tack the fragment onto this segment 658 * and reinsert the data. 659 */ 660 if (q->ipqe_seq == pkt_seq + pkt_len) { 661 #ifdef TCPREASS_DEBUG 662 printf("tcp_reass[%p]: append %u:%u(%u) to %u:%u(%u)\n", 663 tp, q->ipqe_seq, q->ipqe_seq + q->ipqe_len, q->ipqe_len, 664 pkt_seq, pkt_seq + pkt_len, pkt_len); 665 #endif 666 pkt_len += q->ipqe_len; 667 pkt_flags |= q->ipqe_flags; 668 m_cat(m, q->ipqe_m); 669 TAILQ_REMOVE(&tp->segq, q, ipqe_q); 670 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 671 tp->t_segqlen--; 672 KASSERT(tp->t_segqlen >= 0); 673 KASSERT(tp->t_segqlen != 0 || 674 (TAILQ_EMPTY(&tp->segq) && 675 TAILQ_EMPTY(&tp->timeq))); 676 if (tiqe == NULL) { 677 tiqe = q; 678 } else { 679 tcpipqent_free(q); 680 } 681 TCP_REASS_COUNTER_INCR(&tcp_reass_prepend); 682 break; 683 } 684 /* 685 * If the fragment is before the segment, remember it. 686 * When this loop is terminated, p will contain the 687 * pointer to fragment that is right before the received 688 * segment. 689 */ 690 if (SEQ_LEQ(q->ipqe_seq, pkt_seq)) 691 p = q; 692 693 continue; 694 695 /* 696 * This is a common operation. It also will allow 697 * to save doing a malloc/free in most instances. 698 */ 699 free_ipqe: 700 TAILQ_REMOVE(&tp->segq, q, ipqe_q); 701 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 702 tp->t_segqlen--; 703 KASSERT(tp->t_segqlen >= 0); 704 KASSERT(tp->t_segqlen != 0 || 705 (TAILQ_EMPTY(&tp->segq) && TAILQ_EMPTY(&tp->timeq))); 706 if (tiqe == NULL) { 707 tiqe = q; 708 } else { 709 tcpipqent_free(q); 710 } 711 } 712 713 #ifdef TCP_REASS_COUNTERS 714 if (count > 7) 715 TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[0]); 716 else if (count > 0) 717 TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[count]); 718 #endif 719 720 insert_it: 721 722 /* 723 * Allocate a new queue entry since the received segment did not 724 * collapse onto any other out-of-order block; thus we are allocating 725 * a new block. If it had collapsed, tiqe would not be NULL and 726 * we would be reusing it. 727 * XXX If we can't, just drop the packet. XXX 728 */ 729 if (tiqe == NULL) { 730 tiqe = tcpipqent_alloc(); 731 if (tiqe == NULL) { 732 TCP_STATINC(TCP_STAT_RCVMEMDROP); 733 m_freem(m); 734 return (0); 735 } 736 } 737 738 /* 739 * Update the counters. 740 */ 741 tcps = TCP_STAT_GETREF(); 742 tcps[TCP_STAT_RCVOOPACK]++; 743 tcps[TCP_STAT_RCVOOBYTE] += rcvoobyte; 744 if (rcvpartdupbyte) { 745 tcps[TCP_STAT_RCVPARTDUPPACK]++; 746 tcps[TCP_STAT_RCVPARTDUPBYTE] += rcvpartdupbyte; 747 } 748 TCP_STAT_PUTREF(); 749 750 /* 751 * Insert the new fragment queue entry into both queues. 752 */ 753 tiqe->ipqe_m = m; 754 tiqe->ipre_mlast = m; 755 tiqe->ipqe_seq = pkt_seq; 756 tiqe->ipqe_len = pkt_len; 757 tiqe->ipqe_flags = pkt_flags; 758 if (p == NULL) { 759 TAILQ_INSERT_HEAD(&tp->segq, tiqe, ipqe_q); 760 #ifdef TCPREASS_DEBUG 761 if (tiqe->ipqe_seq != tp->rcv_nxt) 762 printf("tcp_reass[%p]: insert %u:%u(%u) at front\n", 763 tp, pkt_seq, pkt_seq + pkt_len, pkt_len); 764 #endif 765 } else { 766 TAILQ_INSERT_AFTER(&tp->segq, p, tiqe, ipqe_q); 767 #ifdef TCPREASS_DEBUG 768 printf("tcp_reass[%p]: insert %u:%u(%u) after %u:%u(%u)\n", 769 tp, pkt_seq, pkt_seq + pkt_len, pkt_len, 770 p->ipqe_seq, p->ipqe_seq + p->ipqe_len, p->ipqe_len); 771 #endif 772 } 773 tp->t_segqlen++; 774 775 skip_replacement: 776 777 TAILQ_INSERT_HEAD(&tp->timeq, tiqe, ipqe_timeq); 778 779 present: 780 /* 781 * Present data to user, advancing rcv_nxt through 782 * completed sequence space. 783 */ 784 if (TCPS_HAVEESTABLISHED(tp->t_state) == 0) 785 return (0); 786 q = TAILQ_FIRST(&tp->segq); 787 if (q == NULL || q->ipqe_seq != tp->rcv_nxt) 788 return (0); 789 if (tp->t_state == TCPS_SYN_RECEIVED && q->ipqe_len) 790 return (0); 791 792 tp->rcv_nxt += q->ipqe_len; 793 pkt_flags = q->ipqe_flags & TH_FIN; 794 nd6_hint(tp); 795 796 TAILQ_REMOVE(&tp->segq, q, ipqe_q); 797 TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq); 798 tp->t_segqlen--; 799 KASSERT(tp->t_segqlen >= 0); 800 KASSERT(tp->t_segqlen != 0 || 801 (TAILQ_EMPTY(&tp->segq) && TAILQ_EMPTY(&tp->timeq))); 802 if (so->so_state & SS_CANTRCVMORE) 803 m_freem(q->ipqe_m); 804 else 805 sbappendstream(&so->so_rcv, q->ipqe_m); 806 tcpipqent_free(q); 807 sorwakeup(so); 808 return (pkt_flags); 809 } 810 811 #ifdef INET6 812 int 813 tcp6_input(struct mbuf **mp, int *offp, int proto) 814 { 815 struct mbuf *m = *mp; 816 817 /* 818 * draft-itojun-ipv6-tcp-to-anycast 819 * better place to put this in? 820 */ 821 if (m->m_flags & M_ANYCAST6) { 822 struct ip6_hdr *ip6; 823 if (m->m_len < sizeof(struct ip6_hdr)) { 824 if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) { 825 TCP_STATINC(TCP_STAT_RCVSHORT); 826 return IPPROTO_DONE; 827 } 828 } 829 ip6 = mtod(m, struct ip6_hdr *); 830 icmp6_error(m, ICMP6_DST_UNREACH, ICMP6_DST_UNREACH_ADDR, 831 (char *)&ip6->ip6_dst - (char *)ip6); 832 return IPPROTO_DONE; 833 } 834 835 tcp_input(m, *offp, proto); 836 return IPPROTO_DONE; 837 } 838 #endif 839 840 #ifdef INET 841 static void 842 tcp4_log_refused(const struct ip *ip, const struct tcphdr *th) 843 { 844 char src[4*sizeof "123"]; 845 char dst[4*sizeof "123"]; 846 847 if (ip) { 848 strlcpy(src, inet_ntoa(ip->ip_src), sizeof(src)); 849 strlcpy(dst, inet_ntoa(ip->ip_dst), sizeof(dst)); 850 } 851 else { 852 strlcpy(src, "(unknown)", sizeof(src)); 853 strlcpy(dst, "(unknown)", sizeof(dst)); 854 } 855 log(LOG_INFO, 856 "Connection attempt to TCP %s:%d from %s:%d\n", 857 dst, ntohs(th->th_dport), 858 src, ntohs(th->th_sport)); 859 } 860 #endif 861 862 #ifdef INET6 863 static void 864 tcp6_log_refused(const struct ip6_hdr *ip6, const struct tcphdr *th) 865 { 866 char src[INET6_ADDRSTRLEN]; 867 char dst[INET6_ADDRSTRLEN]; 868 869 if (ip6) { 870 strlcpy(src, ip6_sprintf(&ip6->ip6_src), sizeof(src)); 871 strlcpy(dst, ip6_sprintf(&ip6->ip6_dst), sizeof(dst)); 872 } 873 else { 874 strlcpy(src, "(unknown v6)", sizeof(src)); 875 strlcpy(dst, "(unknown v6)", sizeof(dst)); 876 } 877 log(LOG_INFO, 878 "Connection attempt to TCP [%s]:%d from [%s]:%d\n", 879 dst, ntohs(th->th_dport), 880 src, ntohs(th->th_sport)); 881 } 882 #endif 883 884 /* 885 * Checksum extended TCP header and data. 886 */ 887 int 888 tcp_input_checksum(int af, struct mbuf *m, const struct tcphdr *th, 889 int toff, int off, int tlen) 890 { 891 892 /* 893 * XXX it's better to record and check if this mbuf is 894 * already checked. 895 */ 896 897 switch (af) { 898 #ifdef INET 899 case AF_INET: 900 switch (m->m_pkthdr.csum_flags & 901 ((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_TCPv4) | 902 M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) { 903 case M_CSUM_TCPv4|M_CSUM_TCP_UDP_BAD: 904 TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_bad); 905 goto badcsum; 906 907 case M_CSUM_TCPv4|M_CSUM_DATA: { 908 u_int32_t hw_csum = m->m_pkthdr.csum_data; 909 910 TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_data); 911 if (m->m_pkthdr.csum_flags & M_CSUM_NO_PSEUDOHDR) { 912 const struct ip *ip = 913 mtod(m, const struct ip *); 914 915 hw_csum = in_cksum_phdr(ip->ip_src.s_addr, 916 ip->ip_dst.s_addr, 917 htons(hw_csum + tlen + off + IPPROTO_TCP)); 918 } 919 if ((hw_csum ^ 0xffff) != 0) 920 goto badcsum; 921 break; 922 } 923 924 case M_CSUM_TCPv4: 925 /* Checksum was okay. */ 926 TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_ok); 927 break; 928 929 default: 930 /* 931 * Must compute it ourselves. Maybe skip checksum 932 * on loopback interfaces. 933 */ 934 if (__predict_true(!(m->m_pkthdr.rcvif->if_flags & 935 IFF_LOOPBACK) || 936 tcp_do_loopback_cksum)) { 937 TCP_CSUM_COUNTER_INCR(&tcp_swcsum); 938 if (in4_cksum(m, IPPROTO_TCP, toff, 939 tlen + off) != 0) 940 goto badcsum; 941 } 942 break; 943 } 944 break; 945 #endif /* INET4 */ 946 947 #ifdef INET6 948 case AF_INET6: 949 switch (m->m_pkthdr.csum_flags & 950 ((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_TCPv6) | 951 M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) { 952 case M_CSUM_TCPv6|M_CSUM_TCP_UDP_BAD: 953 TCP_CSUM_COUNTER_INCR(&tcp6_hwcsum_bad); 954 goto badcsum; 955 956 #if 0 /* notyet */ 957 case M_CSUM_TCPv6|M_CSUM_DATA: 958 #endif 959 960 case M_CSUM_TCPv6: 961 /* Checksum was okay. */ 962 TCP_CSUM_COUNTER_INCR(&tcp6_hwcsum_ok); 963 break; 964 965 default: 966 /* 967 * Must compute it ourselves. Maybe skip checksum 968 * on loopback interfaces. 969 */ 970 if (__predict_true((m->m_flags & M_LOOP) == 0 || 971 tcp_do_loopback_cksum)) { 972 TCP_CSUM_COUNTER_INCR(&tcp6_swcsum); 973 if (in6_cksum(m, IPPROTO_TCP, toff, 974 tlen + off) != 0) 975 goto badcsum; 976 } 977 } 978 break; 979 #endif /* INET6 */ 980 } 981 982 return 0; 983 984 badcsum: 985 TCP_STATINC(TCP_STAT_RCVBADSUM); 986 return -1; 987 } 988 989 /* 990 * TCP input routine, follows pages 65-76 of RFC 793 very closely. 991 */ 992 void 993 tcp_input(struct mbuf *m, ...) 994 { 995 struct tcphdr *th; 996 struct ip *ip; 997 struct inpcb *inp; 998 #ifdef INET6 999 struct ip6_hdr *ip6; 1000 struct in6pcb *in6p; 1001 #endif 1002 u_int8_t *optp = NULL; 1003 int optlen = 0; 1004 int len, tlen, toff, hdroptlen = 0; 1005 struct tcpcb *tp = 0; 1006 int tiflags; 1007 struct socket *so = NULL; 1008 int todrop, dupseg, acked, ourfinisacked, needoutput = 0; 1009 #ifdef TCP_DEBUG 1010 short ostate = 0; 1011 #endif 1012 u_long tiwin; 1013 struct tcp_opt_info opti; 1014 int off, iphlen; 1015 va_list ap; 1016 int af; /* af on the wire */ 1017 struct mbuf *tcp_saveti = NULL; 1018 uint32_t ts_rtt; 1019 uint8_t iptos; 1020 uint64_t *tcps; 1021 1022 MCLAIM(m, &tcp_rx_mowner); 1023 va_start(ap, m); 1024 toff = va_arg(ap, int); 1025 (void)va_arg(ap, int); /* ignore value, advance ap */ 1026 va_end(ap); 1027 1028 TCP_STATINC(TCP_STAT_RCVTOTAL); 1029 1030 memset(&opti, 0, sizeof(opti)); 1031 opti.ts_present = 0; 1032 opti.maxseg = 0; 1033 1034 /* 1035 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN. 1036 * 1037 * TCP is, by definition, unicast, so we reject all 1038 * multicast outright. 1039 * 1040 * Note, there are additional src/dst address checks in 1041 * the AF-specific code below. 1042 */ 1043 if (m->m_flags & (M_BCAST|M_MCAST)) { 1044 /* XXX stat */ 1045 goto drop; 1046 } 1047 #ifdef INET6 1048 if (m->m_flags & M_ANYCAST6) { 1049 /* XXX stat */ 1050 goto drop; 1051 } 1052 #endif 1053 1054 /* 1055 * Get IP and TCP header. 1056 * Note: IP leaves IP header in first mbuf. 1057 */ 1058 ip = mtod(m, struct ip *); 1059 #ifdef INET6 1060 ip6 = NULL; 1061 #endif 1062 switch (ip->ip_v) { 1063 #ifdef INET 1064 case 4: 1065 af = AF_INET; 1066 iphlen = sizeof(struct ip); 1067 ip = mtod(m, struct ip *); 1068 IP6_EXTHDR_GET(th, struct tcphdr *, m, toff, 1069 sizeof(struct tcphdr)); 1070 if (th == NULL) { 1071 TCP_STATINC(TCP_STAT_RCVSHORT); 1072 return; 1073 } 1074 /* We do the checksum after PCB lookup... */ 1075 len = ntohs(ip->ip_len); 1076 tlen = len - toff; 1077 iptos = ip->ip_tos; 1078 break; 1079 #endif 1080 #ifdef INET6 1081 case 6: 1082 ip = NULL; 1083 iphlen = sizeof(struct ip6_hdr); 1084 af = AF_INET6; 1085 ip6 = mtod(m, struct ip6_hdr *); 1086 IP6_EXTHDR_GET(th, struct tcphdr *, m, toff, 1087 sizeof(struct tcphdr)); 1088 if (th == NULL) { 1089 TCP_STATINC(TCP_STAT_RCVSHORT); 1090 return; 1091 } 1092 1093 /* Be proactive about malicious use of IPv4 mapped address */ 1094 if (IN6_IS_ADDR_V4MAPPED(&ip6->ip6_src) || 1095 IN6_IS_ADDR_V4MAPPED(&ip6->ip6_dst)) { 1096 /* XXX stat */ 1097 goto drop; 1098 } 1099 1100 /* 1101 * Be proactive about unspecified IPv6 address in source. 1102 * As we use all-zero to indicate unbounded/unconnected pcb, 1103 * unspecified IPv6 address can be used to confuse us. 1104 * 1105 * Note that packets with unspecified IPv6 destination is 1106 * already dropped in ip6_input. 1107 */ 1108 if (IN6_IS_ADDR_UNSPECIFIED(&ip6->ip6_src)) { 1109 /* XXX stat */ 1110 goto drop; 1111 } 1112 1113 /* 1114 * Make sure destination address is not multicast. 1115 * Source address checked in ip6_input(). 1116 */ 1117 if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) { 1118 /* XXX stat */ 1119 goto drop; 1120 } 1121 1122 /* We do the checksum after PCB lookup... */ 1123 len = m->m_pkthdr.len; 1124 tlen = len - toff; 1125 iptos = (ntohl(ip6->ip6_flow) >> 20) & 0xff; 1126 break; 1127 #endif 1128 default: 1129 m_freem(m); 1130 return; 1131 } 1132 1133 KASSERT(TCP_HDR_ALIGNED_P(th)); 1134 1135 /* 1136 * Check that TCP offset makes sense, 1137 * pull out TCP options and adjust length. XXX 1138 */ 1139 off = th->th_off << 2; 1140 if (off < sizeof (struct tcphdr) || off > tlen) { 1141 TCP_STATINC(TCP_STAT_RCVBADOFF); 1142 goto drop; 1143 } 1144 tlen -= off; 1145 1146 /* 1147 * tcp_input() has been modified to use tlen to mean the TCP data 1148 * length throughout the function. Other functions can use 1149 * m->m_pkthdr.len as the basis for calculating the TCP data length. 1150 * rja 1151 */ 1152 1153 if (off > sizeof (struct tcphdr)) { 1154 IP6_EXTHDR_GET(th, struct tcphdr *, m, toff, off); 1155 if (th == NULL) { 1156 TCP_STATINC(TCP_STAT_RCVSHORT); 1157 return; 1158 } 1159 /* 1160 * NOTE: ip/ip6 will not be affected by m_pulldown() 1161 * (as they're before toff) and we don't need to update those. 1162 */ 1163 KASSERT(TCP_HDR_ALIGNED_P(th)); 1164 optlen = off - sizeof (struct tcphdr); 1165 optp = ((u_int8_t *)th) + sizeof(struct tcphdr); 1166 /* 1167 * Do quick retrieval of timestamp options ("options 1168 * prediction?"). If timestamp is the only option and it's 1169 * formatted as recommended in RFC 1323 appendix A, we 1170 * quickly get the values now and not bother calling 1171 * tcp_dooptions(), etc. 1172 */ 1173 if ((optlen == TCPOLEN_TSTAMP_APPA || 1174 (optlen > TCPOLEN_TSTAMP_APPA && 1175 optp[TCPOLEN_TSTAMP_APPA] == TCPOPT_EOL)) && 1176 *(u_int32_t *)optp == htonl(TCPOPT_TSTAMP_HDR) && 1177 (th->th_flags & TH_SYN) == 0) { 1178 opti.ts_present = 1; 1179 opti.ts_val = ntohl(*(u_int32_t *)(optp + 4)); 1180 opti.ts_ecr = ntohl(*(u_int32_t *)(optp + 8)); 1181 optp = NULL; /* we've parsed the options */ 1182 } 1183 } 1184 tiflags = th->th_flags; 1185 1186 /* 1187 * Locate pcb for segment. 1188 */ 1189 findpcb: 1190 inp = NULL; 1191 #ifdef INET6 1192 in6p = NULL; 1193 #endif 1194 switch (af) { 1195 #ifdef INET 1196 case AF_INET: 1197 inp = in_pcblookup_connect(&tcbtable, ip->ip_src, th->th_sport, 1198 ip->ip_dst, th->th_dport); 1199 if (inp == 0) { 1200 TCP_STATINC(TCP_STAT_PCBHASHMISS); 1201 inp = in_pcblookup_bind(&tcbtable, ip->ip_dst, th->th_dport); 1202 } 1203 #ifdef INET6 1204 if (inp == 0) { 1205 struct in6_addr s, d; 1206 1207 /* mapped addr case */ 1208 memset(&s, 0, sizeof(s)); 1209 s.s6_addr16[5] = htons(0xffff); 1210 bcopy(&ip->ip_src, &s.s6_addr32[3], sizeof(ip->ip_src)); 1211 memset(&d, 0, sizeof(d)); 1212 d.s6_addr16[5] = htons(0xffff); 1213 bcopy(&ip->ip_dst, &d.s6_addr32[3], sizeof(ip->ip_dst)); 1214 in6p = in6_pcblookup_connect(&tcbtable, &s, 1215 th->th_sport, &d, th->th_dport, 0); 1216 if (in6p == 0) { 1217 TCP_STATINC(TCP_STAT_PCBHASHMISS); 1218 in6p = in6_pcblookup_bind(&tcbtable, &d, 1219 th->th_dport, 0); 1220 } 1221 } 1222 #endif 1223 #ifndef INET6 1224 if (inp == 0) 1225 #else 1226 if (inp == 0 && in6p == 0) 1227 #endif 1228 { 1229 TCP_STATINC(TCP_STAT_NOPORT); 1230 if (tcp_log_refused && 1231 (tiflags & (TH_RST|TH_ACK|TH_SYN)) == TH_SYN) { 1232 tcp4_log_refused(ip, th); 1233 } 1234 tcp_fields_to_host(th); 1235 goto dropwithreset_ratelim; 1236 } 1237 #if defined(IPSEC) || defined(FAST_IPSEC) 1238 if (inp && (inp->inp_socket->so_options & SO_ACCEPTCONN) == 0 && 1239 ipsec4_in_reject(m, inp)) { 1240 IPSEC_STATINC(IPSEC_STAT_IN_POLVIO); 1241 goto drop; 1242 } 1243 #ifdef INET6 1244 else if (in6p && 1245 (in6p->in6p_socket->so_options & SO_ACCEPTCONN) == 0 && 1246 ipsec6_in_reject_so(m, in6p->in6p_socket)) { 1247 IPSEC_STATINC(IPSEC_STAT_IN_POLVIO); 1248 goto drop; 1249 } 1250 #endif 1251 #endif /*IPSEC*/ 1252 break; 1253 #endif /*INET*/ 1254 #ifdef INET6 1255 case AF_INET6: 1256 { 1257 int faith; 1258 1259 #if defined(NFAITH) && NFAITH > 0 1260 faith = faithprefix(&ip6->ip6_dst); 1261 #else 1262 faith = 0; 1263 #endif 1264 in6p = in6_pcblookup_connect(&tcbtable, &ip6->ip6_src, 1265 th->th_sport, &ip6->ip6_dst, th->th_dport, faith); 1266 if (in6p == NULL) { 1267 TCP_STATINC(TCP_STAT_PCBHASHMISS); 1268 in6p = in6_pcblookup_bind(&tcbtable, &ip6->ip6_dst, 1269 th->th_dport, faith); 1270 } 1271 if (in6p == NULL) { 1272 TCP_STATINC(TCP_STAT_NOPORT); 1273 if (tcp_log_refused && 1274 (tiflags & (TH_RST|TH_ACK|TH_SYN)) == TH_SYN) { 1275 tcp6_log_refused(ip6, th); 1276 } 1277 tcp_fields_to_host(th); 1278 goto dropwithreset_ratelim; 1279 } 1280 #if defined(IPSEC) || defined(FAST_IPSEC) 1281 if ((in6p->in6p_socket->so_options & SO_ACCEPTCONN) == 0 && 1282 ipsec6_in_reject(m, in6p)) { 1283 IPSEC6_STATINC(IPSEC_STAT_IN_POLVIO); 1284 goto drop; 1285 } 1286 #endif /*IPSEC*/ 1287 break; 1288 } 1289 #endif 1290 } 1291 1292 /* 1293 * If the state is CLOSED (i.e., TCB does not exist) then 1294 * all data in the incoming segment is discarded. 1295 * If the TCB exists but is in CLOSED state, it is embryonic, 1296 * but should either do a listen or a connect soon. 1297 */ 1298 tp = NULL; 1299 so = NULL; 1300 if (inp) { 1301 /* Check the minimum TTL for socket. */ 1302 if (ip->ip_ttl < inp->inp_ip_minttl) 1303 goto drop; 1304 1305 tp = intotcpcb(inp); 1306 so = inp->inp_socket; 1307 } 1308 #ifdef INET6 1309 else if (in6p) { 1310 tp = in6totcpcb(in6p); 1311 so = in6p->in6p_socket; 1312 } 1313 #endif 1314 if (tp == 0) { 1315 tcp_fields_to_host(th); 1316 goto dropwithreset_ratelim; 1317 } 1318 if (tp->t_state == TCPS_CLOSED) 1319 goto drop; 1320 1321 KASSERT(so->so_lock == softnet_lock); 1322 KASSERT(solocked(so)); 1323 1324 /* 1325 * Checksum extended TCP header and data. 1326 */ 1327 if (tcp_input_checksum(af, m, th, toff, off, tlen)) 1328 goto badcsum; 1329 1330 tcp_fields_to_host(th); 1331 1332 /* Unscale the window into a 32-bit value. */ 1333 if ((tiflags & TH_SYN) == 0) 1334 tiwin = th->th_win << tp->snd_scale; 1335 else 1336 tiwin = th->th_win; 1337 1338 #ifdef INET6 1339 /* save packet options if user wanted */ 1340 if (in6p && (in6p->in6p_flags & IN6P_CONTROLOPTS)) { 1341 if (in6p->in6p_options) { 1342 m_freem(in6p->in6p_options); 1343 in6p->in6p_options = 0; 1344 } 1345 KASSERT(ip6 != NULL); 1346 ip6_savecontrol(in6p, &in6p->in6p_options, ip6, m); 1347 } 1348 #endif 1349 1350 if (so->so_options & (SO_DEBUG|SO_ACCEPTCONN)) { 1351 union syn_cache_sa src; 1352 union syn_cache_sa dst; 1353 1354 memset(&src, 0, sizeof(src)); 1355 memset(&dst, 0, sizeof(dst)); 1356 switch (af) { 1357 #ifdef INET 1358 case AF_INET: 1359 src.sin.sin_len = sizeof(struct sockaddr_in); 1360 src.sin.sin_family = AF_INET; 1361 src.sin.sin_addr = ip->ip_src; 1362 src.sin.sin_port = th->th_sport; 1363 1364 dst.sin.sin_len = sizeof(struct sockaddr_in); 1365 dst.sin.sin_family = AF_INET; 1366 dst.sin.sin_addr = ip->ip_dst; 1367 dst.sin.sin_port = th->th_dport; 1368 break; 1369 #endif 1370 #ifdef INET6 1371 case AF_INET6: 1372 src.sin6.sin6_len = sizeof(struct sockaddr_in6); 1373 src.sin6.sin6_family = AF_INET6; 1374 src.sin6.sin6_addr = ip6->ip6_src; 1375 src.sin6.sin6_port = th->th_sport; 1376 1377 dst.sin6.sin6_len = sizeof(struct sockaddr_in6); 1378 dst.sin6.sin6_family = AF_INET6; 1379 dst.sin6.sin6_addr = ip6->ip6_dst; 1380 dst.sin6.sin6_port = th->th_dport; 1381 break; 1382 #endif /* INET6 */ 1383 default: 1384 goto badsyn; /*sanity*/ 1385 } 1386 1387 if (so->so_options & SO_DEBUG) { 1388 #ifdef TCP_DEBUG 1389 ostate = tp->t_state; 1390 #endif 1391 1392 tcp_saveti = NULL; 1393 if (iphlen + sizeof(struct tcphdr) > MHLEN) 1394 goto nosave; 1395 1396 if (m->m_len > iphlen && (m->m_flags & M_EXT) == 0) { 1397 tcp_saveti = m_copym(m, 0, iphlen, M_DONTWAIT); 1398 if (!tcp_saveti) 1399 goto nosave; 1400 } else { 1401 MGETHDR(tcp_saveti, M_DONTWAIT, MT_HEADER); 1402 if (!tcp_saveti) 1403 goto nosave; 1404 MCLAIM(m, &tcp_mowner); 1405 tcp_saveti->m_len = iphlen; 1406 m_copydata(m, 0, iphlen, 1407 mtod(tcp_saveti, void *)); 1408 } 1409 1410 if (M_TRAILINGSPACE(tcp_saveti) < sizeof(struct tcphdr)) { 1411 m_freem(tcp_saveti); 1412 tcp_saveti = NULL; 1413 } else { 1414 tcp_saveti->m_len += sizeof(struct tcphdr); 1415 memcpy(mtod(tcp_saveti, char *) + iphlen, th, 1416 sizeof(struct tcphdr)); 1417 } 1418 nosave:; 1419 } 1420 if (so->so_options & SO_ACCEPTCONN) { 1421 if ((tiflags & (TH_RST|TH_ACK|TH_SYN)) != TH_SYN) { 1422 if (tiflags & TH_RST) { 1423 syn_cache_reset(&src.sa, &dst.sa, th); 1424 } else if ((tiflags & (TH_ACK|TH_SYN)) == 1425 (TH_ACK|TH_SYN)) { 1426 /* 1427 * Received a SYN,ACK. This should 1428 * never happen while we are in 1429 * LISTEN. Send an RST. 1430 */ 1431 goto badsyn; 1432 } else if (tiflags & TH_ACK) { 1433 so = syn_cache_get(&src.sa, &dst.sa, 1434 th, toff, tlen, so, m); 1435 if (so == NULL) { 1436 /* 1437 * We don't have a SYN for 1438 * this ACK; send an RST. 1439 */ 1440 goto badsyn; 1441 } else if (so == 1442 (struct socket *)(-1)) { 1443 /* 1444 * We were unable to create 1445 * the connection. If the 1446 * 3-way handshake was 1447 * completed, and RST has 1448 * been sent to the peer. 1449 * Since the mbuf might be 1450 * in use for the reply, 1451 * do not free it. 1452 */ 1453 m = NULL; 1454 } else { 1455 /* 1456 * We have created a 1457 * full-blown connection. 1458 */ 1459 tp = NULL; 1460 inp = NULL; 1461 #ifdef INET6 1462 in6p = NULL; 1463 #endif 1464 switch (so->so_proto->pr_domain->dom_family) { 1465 #ifdef INET 1466 case AF_INET: 1467 inp = sotoinpcb(so); 1468 tp = intotcpcb(inp); 1469 break; 1470 #endif 1471 #ifdef INET6 1472 case AF_INET6: 1473 in6p = sotoin6pcb(so); 1474 tp = in6totcpcb(in6p); 1475 break; 1476 #endif 1477 } 1478 if (tp == NULL) 1479 goto badsyn; /*XXX*/ 1480 tiwin <<= tp->snd_scale; 1481 goto after_listen; 1482 } 1483 } else { 1484 /* 1485 * None of RST, SYN or ACK was set. 1486 * This is an invalid packet for a 1487 * TCB in LISTEN state. Send a RST. 1488 */ 1489 goto badsyn; 1490 } 1491 } else { 1492 /* 1493 * Received a SYN. 1494 * 1495 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN 1496 */ 1497 if (m->m_flags & (M_BCAST|M_MCAST)) 1498 goto drop; 1499 1500 switch (af) { 1501 #ifdef INET6 1502 case AF_INET6: 1503 if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) 1504 goto drop; 1505 break; 1506 #endif /* INET6 */ 1507 case AF_INET: 1508 if (IN_MULTICAST(ip->ip_dst.s_addr) || 1509 in_broadcast(ip->ip_dst, m->m_pkthdr.rcvif)) 1510 goto drop; 1511 break; 1512 } 1513 1514 #ifdef INET6 1515 /* 1516 * If deprecated address is forbidden, we do 1517 * not accept SYN to deprecated interface 1518 * address to prevent any new inbound 1519 * connection from getting established. 1520 * When we do not accept SYN, we send a TCP 1521 * RST, with deprecated source address (instead 1522 * of dropping it). We compromise it as it is 1523 * much better for peer to send a RST, and 1524 * RST will be the final packet for the 1525 * exchange. 1526 * 1527 * If we do not forbid deprecated addresses, we 1528 * accept the SYN packet. RFC2462 does not 1529 * suggest dropping SYN in this case. 1530 * If we decipher RFC2462 5.5.4, it says like 1531 * this: 1532 * 1. use of deprecated addr with existing 1533 * communication is okay - "SHOULD continue 1534 * to be used" 1535 * 2. use of it with new communication: 1536 * (2a) "SHOULD NOT be used if alternate 1537 * address with sufficient scope is 1538 * available" 1539 * (2b) nothing mentioned otherwise. 1540 * Here we fall into (2b) case as we have no 1541 * choice in our source address selection - we 1542 * must obey the peer. 1543 * 1544 * The wording in RFC2462 is confusing, and 1545 * there are multiple description text for 1546 * deprecated address handling - worse, they 1547 * are not exactly the same. I believe 5.5.4 1548 * is the best one, so we follow 5.5.4. 1549 */ 1550 if (af == AF_INET6 && !ip6_use_deprecated) { 1551 struct in6_ifaddr *ia6; 1552 if ((ia6 = in6ifa_ifpwithaddr(m->m_pkthdr.rcvif, 1553 &ip6->ip6_dst)) && 1554 (ia6->ia6_flags & IN6_IFF_DEPRECATED)) { 1555 tp = NULL; 1556 goto dropwithreset; 1557 } 1558 } 1559 #endif 1560 1561 #if defined(IPSEC) || defined(FAST_IPSEC) 1562 switch (af) { 1563 #ifdef INET 1564 case AF_INET: 1565 if (ipsec4_in_reject_so(m, so)) { 1566 IPSEC_STATINC(IPSEC_STAT_IN_POLVIO); 1567 tp = NULL; 1568 goto dropwithreset; 1569 } 1570 break; 1571 #endif 1572 #ifdef INET6 1573 case AF_INET6: 1574 if (ipsec6_in_reject_so(m, so)) { 1575 IPSEC6_STATINC(IPSEC_STAT_IN_POLVIO); 1576 tp = NULL; 1577 goto dropwithreset; 1578 } 1579 break; 1580 #endif /*INET6*/ 1581 } 1582 #endif /*IPSEC*/ 1583 1584 /* 1585 * LISTEN socket received a SYN 1586 * from itself? This can't possibly 1587 * be valid; drop the packet. 1588 */ 1589 if (th->th_sport == th->th_dport) { 1590 int i; 1591 1592 switch (af) { 1593 #ifdef INET 1594 case AF_INET: 1595 i = in_hosteq(ip->ip_src, ip->ip_dst); 1596 break; 1597 #endif 1598 #ifdef INET6 1599 case AF_INET6: 1600 i = IN6_ARE_ADDR_EQUAL(&ip6->ip6_src, &ip6->ip6_dst); 1601 break; 1602 #endif 1603 default: 1604 i = 1; 1605 } 1606 if (i) { 1607 TCP_STATINC(TCP_STAT_BADSYN); 1608 goto drop; 1609 } 1610 } 1611 1612 /* 1613 * SYN looks ok; create compressed TCP 1614 * state for it. 1615 */ 1616 if (so->so_qlen <= so->so_qlimit && 1617 syn_cache_add(&src.sa, &dst.sa, th, tlen, 1618 so, m, optp, optlen, &opti)) 1619 m = NULL; 1620 } 1621 goto drop; 1622 } 1623 } 1624 1625 after_listen: 1626 #ifdef DIAGNOSTIC 1627 /* 1628 * Should not happen now that all embryonic connections 1629 * are handled with compressed state. 1630 */ 1631 if (tp->t_state == TCPS_LISTEN) 1632 panic("tcp_input: TCPS_LISTEN"); 1633 #endif 1634 1635 /* 1636 * Segment received on connection. 1637 * Reset idle time and keep-alive timer. 1638 */ 1639 tp->t_rcvtime = tcp_now; 1640 if (TCPS_HAVEESTABLISHED(tp->t_state)) 1641 TCP_TIMER_ARM(tp, TCPT_KEEP, tp->t_keepidle); 1642 1643 /* 1644 * Process options. 1645 */ 1646 #ifdef TCP_SIGNATURE 1647 if (optp || (tp->t_flags & TF_SIGNATURE)) 1648 #else 1649 if (optp) 1650 #endif 1651 if (tcp_dooptions(tp, optp, optlen, th, m, toff, &opti) < 0) 1652 goto drop; 1653 1654 if (TCP_SACK_ENABLED(tp)) { 1655 tcp_del_sackholes(tp, th); 1656 } 1657 1658 if (TCP_ECN_ALLOWED(tp)) { 1659 switch (iptos & IPTOS_ECN_MASK) { 1660 case IPTOS_ECN_CE: 1661 tp->t_flags |= TF_ECN_SND_ECE; 1662 TCP_STATINC(TCP_STAT_ECN_CE); 1663 break; 1664 case IPTOS_ECN_ECT0: 1665 TCP_STATINC(TCP_STAT_ECN_ECT); 1666 break; 1667 case IPTOS_ECN_ECT1: 1668 /* XXX: ignore for now -- rpaulo */ 1669 break; 1670 } 1671 1672 if (tiflags & TH_CWR) 1673 tp->t_flags &= ~TF_ECN_SND_ECE; 1674 1675 /* 1676 * Congestion experienced. 1677 * Ignore if we are already trying to recover. 1678 */ 1679 if ((tiflags & TH_ECE) && SEQ_GEQ(tp->snd_una, tp->snd_recover)) 1680 tp->t_congctl->cong_exp(tp); 1681 } 1682 1683 if (opti.ts_present && opti.ts_ecr) { 1684 /* 1685 * Calculate the RTT from the returned time stamp and the 1686 * connection's time base. If the time stamp is later than 1687 * the current time, or is extremely old, fall back to non-1323 1688 * RTT calculation. Since ts_ecr is unsigned, we can test both 1689 * at the same time. 1690 */ 1691 ts_rtt = TCP_TIMESTAMP(tp) - opti.ts_ecr + 1; 1692 if (ts_rtt > TCP_PAWS_IDLE) 1693 ts_rtt = 0; 1694 } else { 1695 ts_rtt = 0; 1696 } 1697 1698 /* 1699 * Header prediction: check for the two common cases 1700 * of a uni-directional data xfer. If the packet has 1701 * no control flags, is in-sequence, the window didn't 1702 * change and we're not retransmitting, it's a 1703 * candidate. If the length is zero and the ack moved 1704 * forward, we're the sender side of the xfer. Just 1705 * free the data acked & wake any higher level process 1706 * that was blocked waiting for space. If the length 1707 * is non-zero and the ack didn't move, we're the 1708 * receiver side. If we're getting packets in-order 1709 * (the reassembly queue is empty), add the data to 1710 * the socket buffer and note that we need a delayed ack. 1711 */ 1712 if (tp->t_state == TCPS_ESTABLISHED && 1713 (tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ECE|TH_CWR|TH_ACK)) 1714 == TH_ACK && 1715 (!opti.ts_present || TSTMP_GEQ(opti.ts_val, tp->ts_recent)) && 1716 th->th_seq == tp->rcv_nxt && 1717 tiwin && tiwin == tp->snd_wnd && 1718 tp->snd_nxt == tp->snd_max) { 1719 1720 /* 1721 * If last ACK falls within this segment's sequence numbers, 1722 * record the timestamp. 1723 * NOTE that the test is modified according to the latest 1724 * proposal of the tcplw (at) cray.com list (Braden 1993/04/26). 1725 * 1726 * note that we already know 1727 * TSTMP_GEQ(opti.ts_val, tp->ts_recent) 1728 */ 1729 if (opti.ts_present && 1730 SEQ_LEQ(th->th_seq, tp->last_ack_sent)) { 1731 tp->ts_recent_age = tcp_now; 1732 tp->ts_recent = opti.ts_val; 1733 } 1734 1735 if (tlen == 0) { 1736 /* Ack prediction. */ 1737 if (SEQ_GT(th->th_ack, tp->snd_una) && 1738 SEQ_LEQ(th->th_ack, tp->snd_max) && 1739 tp->snd_cwnd >= tp->snd_wnd && 1740 tp->t_partialacks < 0) { 1741 /* 1742 * this is a pure ack for outstanding data. 1743 */ 1744 if (ts_rtt) 1745 tcp_xmit_timer(tp, ts_rtt); 1746 else if (tp->t_rtttime && 1747 SEQ_GT(th->th_ack, tp->t_rtseq)) 1748 tcp_xmit_timer(tp, 1749 tcp_now - tp->t_rtttime); 1750 acked = th->th_ack - tp->snd_una; 1751 tcps = TCP_STAT_GETREF(); 1752 tcps[TCP_STAT_PREDACK]++; 1753 tcps[TCP_STAT_RCVACKPACK]++; 1754 tcps[TCP_STAT_RCVACKBYTE] += acked; 1755 TCP_STAT_PUTREF(); 1756 nd6_hint(tp); 1757 1758 if (acked > (tp->t_lastoff - tp->t_inoff)) 1759 tp->t_lastm = NULL; 1760 sbdrop(&so->so_snd, acked); 1761 tp->t_lastoff -= acked; 1762 1763 icmp_check(tp, th, acked); 1764 1765 tp->snd_una = th->th_ack; 1766 tp->snd_fack = tp->snd_una; 1767 if (SEQ_LT(tp->snd_high, tp->snd_una)) 1768 tp->snd_high = tp->snd_una; 1769 m_freem(m); 1770 1771 /* 1772 * If all outstanding data are acked, stop 1773 * retransmit timer, otherwise restart timer 1774 * using current (possibly backed-off) value. 1775 * If process is waiting for space, 1776 * wakeup/selnotify/signal. If data 1777 * are ready to send, let tcp_output 1778 * decide between more output or persist. 1779 */ 1780 if (tp->snd_una == tp->snd_max) 1781 TCP_TIMER_DISARM(tp, TCPT_REXMT); 1782 else if (TCP_TIMER_ISARMED(tp, 1783 TCPT_PERSIST) == 0) 1784 TCP_TIMER_ARM(tp, TCPT_REXMT, 1785 tp->t_rxtcur); 1786 1787 sowwakeup(so); 1788 if (so->so_snd.sb_cc) 1789 (void) tcp_output(tp); 1790 if (tcp_saveti) 1791 m_freem(tcp_saveti); 1792 return; 1793 } 1794 } else if (th->th_ack == tp->snd_una && 1795 TAILQ_FIRST(&tp->segq) == NULL && 1796 tlen <= sbspace(&so->so_rcv)) { 1797 int newsize = 0; /* automatic sockbuf scaling */ 1798 1799 /* 1800 * this is a pure, in-sequence data packet 1801 * with nothing on the reassembly queue and 1802 * we have enough buffer space to take it. 1803 */ 1804 tp->rcv_nxt += tlen; 1805 tcps = TCP_STAT_GETREF(); 1806 tcps[TCP_STAT_PREDDAT]++; 1807 tcps[TCP_STAT_RCVPACK]++; 1808 tcps[TCP_STAT_RCVBYTE] += tlen; 1809 TCP_STAT_PUTREF(); 1810 nd6_hint(tp); 1811 1812 /* 1813 * Automatic sizing enables the performance of large buffers 1814 * and most of the efficiency of small ones by only allocating 1815 * space when it is needed. 1816 * 1817 * On the receive side the socket buffer memory is only rarely 1818 * used to any significant extent. This allows us to be much 1819 * more aggressive in scaling the receive socket buffer. For 1820 * the case that the buffer space is actually used to a large 1821 * extent and we run out of kernel memory we can simply drop 1822 * the new segments; TCP on the sender will just retransmit it 1823 * later. Setting the buffer size too big may only consume too 1824 * much kernel memory if the application doesn't read() from 1825 * the socket or packet loss or reordering makes use of the 1826 * reassembly queue. 1827 * 1828 * The criteria to step up the receive buffer one notch are: 1829 * 1. the number of bytes received during the time it takes 1830 * one timestamp to be reflected back to us (the RTT); 1831 * 2. received bytes per RTT is within seven eighth of the 1832 * current socket buffer size; 1833 * 3. receive buffer size has not hit maximal automatic size; 1834 * 1835 * This algorithm does one step per RTT at most and only if 1836 * we receive a bulk stream w/o packet losses or reorderings. 1837 * Shrinking the buffer during idle times is not necessary as 1838 * it doesn't consume any memory when idle. 1839 * 1840 * TODO: Only step up if the application is actually serving 1841 * the buffer to better manage the socket buffer resources. 1842 */ 1843 if (tcp_do_autorcvbuf && 1844 opti.ts_ecr && 1845 (so->so_rcv.sb_flags & SB_AUTOSIZE)) { 1846 if (opti.ts_ecr > tp->rfbuf_ts && 1847 opti.ts_ecr - tp->rfbuf_ts < PR_SLOWHZ) { 1848 if (tp->rfbuf_cnt > 1849 (so->so_rcv.sb_hiwat / 8 * 7) && 1850 so->so_rcv.sb_hiwat < 1851 tcp_autorcvbuf_max) { 1852 newsize = 1853 min(so->so_rcv.sb_hiwat + 1854 tcp_autorcvbuf_inc, 1855 tcp_autorcvbuf_max); 1856 } 1857 /* Start over with next RTT. */ 1858 tp->rfbuf_ts = 0; 1859 tp->rfbuf_cnt = 0; 1860 } else 1861 tp->rfbuf_cnt += tlen; /* add up */ 1862 } 1863 1864 /* 1865 * Drop TCP, IP headers and TCP options then add data 1866 * to socket buffer. 1867 */ 1868 if (so->so_state & SS_CANTRCVMORE) 1869 m_freem(m); 1870 else { 1871 /* 1872 * Set new socket buffer size. 1873 * Give up when limit is reached. 1874 */ 1875 if (newsize) 1876 if (!sbreserve(&so->so_rcv, 1877 newsize, so)) 1878 so->so_rcv.sb_flags &= ~SB_AUTOSIZE; 1879 m_adj(m, toff + off); 1880 sbappendstream(&so->so_rcv, m); 1881 } 1882 sorwakeup(so); 1883 tcp_setup_ack(tp, th); 1884 if (tp->t_flags & TF_ACKNOW) 1885 (void) tcp_output(tp); 1886 if (tcp_saveti) 1887 m_freem(tcp_saveti); 1888 return; 1889 } 1890 } 1891 1892 /* 1893 * Compute mbuf offset to TCP data segment. 1894 */ 1895 hdroptlen = toff + off; 1896 1897 /* 1898 * Calculate amount of space in receive window, 1899 * and then do TCP input processing. 1900 * Receive window is amount of space in rcv queue, 1901 * but not less than advertised window. 1902 */ 1903 { int win; 1904 1905 win = sbspace(&so->so_rcv); 1906 if (win < 0) 1907 win = 0; 1908 tp->rcv_wnd = imax(win, (int)(tp->rcv_adv - tp->rcv_nxt)); 1909 } 1910 1911 /* Reset receive buffer auto scaling when not in bulk receive mode. */ 1912 tp->rfbuf_ts = 0; 1913 tp->rfbuf_cnt = 0; 1914 1915 switch (tp->t_state) { 1916 /* 1917 * If the state is SYN_SENT: 1918 * if seg contains an ACK, but not for our SYN, drop the input. 1919 * if seg contains a RST, then drop the connection. 1920 * if seg does not contain SYN, then drop it. 1921 * Otherwise this is an acceptable SYN segment 1922 * initialize tp->rcv_nxt and tp->irs 1923 * if seg contains ack then advance tp->snd_una 1924 * if seg contains a ECE and ECN support is enabled, the stream 1925 * is ECN capable. 1926 * if SYN has been acked change to ESTABLISHED else SYN_RCVD state 1927 * arrange for segment to be acked (eventually) 1928 * continue processing rest of data/controls, beginning with URG 1929 */ 1930 case TCPS_SYN_SENT: 1931 if ((tiflags & TH_ACK) && 1932 (SEQ_LEQ(th->th_ack, tp->iss) || 1933 SEQ_GT(th->th_ack, tp->snd_max))) 1934 goto dropwithreset; 1935 if (tiflags & TH_RST) { 1936 if (tiflags & TH_ACK) 1937 tp = tcp_drop(tp, ECONNREFUSED); 1938 goto drop; 1939 } 1940 if ((tiflags & TH_SYN) == 0) 1941 goto drop; 1942 if (tiflags & TH_ACK) { 1943 tp->snd_una = th->th_ack; 1944 if (SEQ_LT(tp->snd_nxt, tp->snd_una)) 1945 tp->snd_nxt = tp->snd_una; 1946 if (SEQ_LT(tp->snd_high, tp->snd_una)) 1947 tp->snd_high = tp->snd_una; 1948 TCP_TIMER_DISARM(tp, TCPT_REXMT); 1949 1950 if ((tiflags & TH_ECE) && tcp_do_ecn) { 1951 tp->t_flags |= TF_ECN_PERMIT; 1952 TCP_STATINC(TCP_STAT_ECN_SHS); 1953 } 1954 1955 } 1956 tp->irs = th->th_seq; 1957 tcp_rcvseqinit(tp); 1958 tp->t_flags |= TF_ACKNOW; 1959 tcp_mss_from_peer(tp, opti.maxseg); 1960 1961 /* 1962 * Initialize the initial congestion window. If we 1963 * had to retransmit the SYN, we must initialize cwnd 1964 * to 1 segment (i.e. the Loss Window). 1965 */ 1966 if (tp->t_flags & TF_SYN_REXMT) 1967 tp->snd_cwnd = tp->t_peermss; 1968 else { 1969 int ss = tcp_init_win; 1970 #ifdef INET 1971 if (inp != NULL && in_localaddr(inp->inp_faddr)) 1972 ss = tcp_init_win_local; 1973 #endif 1974 #ifdef INET6 1975 if (in6p != NULL && in6_localaddr(&in6p->in6p_faddr)) 1976 ss = tcp_init_win_local; 1977 #endif 1978 tp->snd_cwnd = TCP_INITIAL_WINDOW(ss, tp->t_peermss); 1979 } 1980 1981 tcp_rmx_rtt(tp); 1982 if (tiflags & TH_ACK) { 1983 TCP_STATINC(TCP_STAT_CONNECTS); 1984 soisconnected(so); 1985 tcp_established(tp); 1986 /* Do window scaling on this connection? */ 1987 if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) == 1988 (TF_RCVD_SCALE|TF_REQ_SCALE)) { 1989 tp->snd_scale = tp->requested_s_scale; 1990 tp->rcv_scale = tp->request_r_scale; 1991 } 1992 TCP_REASS_LOCK(tp); 1993 (void) tcp_reass(tp, NULL, (struct mbuf *)0, &tlen); 1994 TCP_REASS_UNLOCK(tp); 1995 /* 1996 * if we didn't have to retransmit the SYN, 1997 * use its rtt as our initial srtt & rtt var. 1998 */ 1999 if (tp->t_rtttime) 2000 tcp_xmit_timer(tp, tcp_now - tp->t_rtttime); 2001 } else 2002 tp->t_state = TCPS_SYN_RECEIVED; 2003 2004 /* 2005 * Advance th->th_seq to correspond to first data byte. 2006 * If data, trim to stay within window, 2007 * dropping FIN if necessary. 2008 */ 2009 th->th_seq++; 2010 if (tlen > tp->rcv_wnd) { 2011 todrop = tlen - tp->rcv_wnd; 2012 m_adj(m, -todrop); 2013 tlen = tp->rcv_wnd; 2014 tiflags &= ~TH_FIN; 2015 tcps = TCP_STAT_GETREF(); 2016 tcps[TCP_STAT_RCVPACKAFTERWIN]++; 2017 tcps[TCP_STAT_RCVBYTEAFTERWIN] += todrop; 2018 TCP_STAT_PUTREF(); 2019 } 2020 tp->snd_wl1 = th->th_seq - 1; 2021 tp->rcv_up = th->th_seq; 2022 goto step6; 2023 2024 /* 2025 * If the state is SYN_RECEIVED: 2026 * If seg contains an ACK, but not for our SYN, drop the input 2027 * and generate an RST. See page 36, rfc793 2028 */ 2029 case TCPS_SYN_RECEIVED: 2030 if ((tiflags & TH_ACK) && 2031 (SEQ_LEQ(th->th_ack, tp->iss) || 2032 SEQ_GT(th->th_ack, tp->snd_max))) 2033 goto dropwithreset; 2034 break; 2035 } 2036 2037 /* 2038 * States other than LISTEN or SYN_SENT. 2039 * First check timestamp, if present. 2040 * Then check that at least some bytes of segment are within 2041 * receive window. If segment begins before rcv_nxt, 2042 * drop leading data (and SYN); if nothing left, just ack. 2043 * 2044 * RFC 1323 PAWS: If we have a timestamp reply on this segment 2045 * and it's less than ts_recent, drop it. 2046 */ 2047 if (opti.ts_present && (tiflags & TH_RST) == 0 && tp->ts_recent && 2048 TSTMP_LT(opti.ts_val, tp->ts_recent)) { 2049 2050 /* Check to see if ts_recent is over 24 days old. */ 2051 if (tcp_now - tp->ts_recent_age > TCP_PAWS_IDLE) { 2052 /* 2053 * Invalidate ts_recent. If this segment updates 2054 * ts_recent, the age will be reset later and ts_recent 2055 * will get a valid value. If it does not, setting 2056 * ts_recent to zero will at least satisfy the 2057 * requirement that zero be placed in the timestamp 2058 * echo reply when ts_recent isn't valid. The 2059 * age isn't reset until we get a valid ts_recent 2060 * because we don't want out-of-order segments to be 2061 * dropped when ts_recent is old. 2062 */ 2063 tp->ts_recent = 0; 2064 } else { 2065 tcps = TCP_STAT_GETREF(); 2066 tcps[TCP_STAT_RCVDUPPACK]++; 2067 tcps[TCP_STAT_RCVDUPBYTE] += tlen; 2068 tcps[TCP_STAT_PAWSDROP]++; 2069 TCP_STAT_PUTREF(); 2070 tcp_new_dsack(tp, th->th_seq, tlen); 2071 goto dropafterack; 2072 } 2073 } 2074 2075 todrop = tp->rcv_nxt - th->th_seq; 2076 dupseg = false; 2077 if (todrop > 0) { 2078 if (tiflags & TH_SYN) { 2079 tiflags &= ~TH_SYN; 2080 th->th_seq++; 2081 if (th->th_urp > 1) 2082 th->th_urp--; 2083 else { 2084 tiflags &= ~TH_URG; 2085 th->th_urp = 0; 2086 } 2087 todrop--; 2088 } 2089 if (todrop > tlen || 2090 (todrop == tlen && (tiflags & TH_FIN) == 0)) { 2091 /* 2092 * Any valid FIN or RST must be to the left of the 2093 * window. At this point the FIN or RST must be a 2094 * duplicate or out of sequence; drop it. 2095 */ 2096 if (tiflags & TH_RST) 2097 goto drop; 2098 tiflags &= ~(TH_FIN|TH_RST); 2099 /* 2100 * Send an ACK to resynchronize and drop any data. 2101 * But keep on processing for RST or ACK. 2102 */ 2103 tp->t_flags |= TF_ACKNOW; 2104 todrop = tlen; 2105 dupseg = true; 2106 tcps = TCP_STAT_GETREF(); 2107 tcps[TCP_STAT_RCVDUPPACK]++; 2108 tcps[TCP_STAT_RCVDUPBYTE] += todrop; 2109 TCP_STAT_PUTREF(); 2110 } else if ((tiflags & TH_RST) && 2111 th->th_seq != tp->rcv_nxt) { 2112 /* 2113 * Test for reset before adjusting the sequence 2114 * number for overlapping data. 2115 */ 2116 goto dropafterack_ratelim; 2117 } else { 2118 tcps = TCP_STAT_GETREF(); 2119 tcps[TCP_STAT_RCVPARTDUPPACK]++; 2120 tcps[TCP_STAT_RCVPARTDUPBYTE] += todrop; 2121 TCP_STAT_PUTREF(); 2122 } 2123 tcp_new_dsack(tp, th->th_seq, todrop); 2124 hdroptlen += todrop; /*drop from head afterwards*/ 2125 th->th_seq += todrop; 2126 tlen -= todrop; 2127 if (th->th_urp > todrop) 2128 th->th_urp -= todrop; 2129 else { 2130 tiflags &= ~TH_URG; 2131 th->th_urp = 0; 2132 } 2133 } 2134 2135 /* 2136 * If new data are received on a connection after the 2137 * user processes are gone, then RST the other end. 2138 */ 2139 if ((so->so_state & SS_NOFDREF) && 2140 tp->t_state > TCPS_CLOSE_WAIT && tlen) { 2141 tp = tcp_close(tp); 2142 TCP_STATINC(TCP_STAT_RCVAFTERCLOSE); 2143 goto dropwithreset; 2144 } 2145 2146 /* 2147 * If segment ends after window, drop trailing data 2148 * (and PUSH and FIN); if nothing left, just ACK. 2149 */ 2150 todrop = (th->th_seq + tlen) - (tp->rcv_nxt+tp->rcv_wnd); 2151 if (todrop > 0) { 2152 TCP_STATINC(TCP_STAT_RCVPACKAFTERWIN); 2153 if (todrop >= tlen) { 2154 /* 2155 * The segment actually starts after the window. 2156 * th->th_seq + tlen - tp->rcv_nxt - tp->rcv_wnd >= tlen 2157 * th->th_seq - tp->rcv_nxt - tp->rcv_wnd >= 0 2158 * th->th_seq >= tp->rcv_nxt + tp->rcv_wnd 2159 */ 2160 TCP_STATADD(TCP_STAT_RCVBYTEAFTERWIN, tlen); 2161 /* 2162 * If a new connection request is received 2163 * while in TIME_WAIT, drop the old connection 2164 * and start over if the sequence numbers 2165 * are above the previous ones. 2166 * 2167 * NOTE: We will checksum the packet again, and 2168 * so we need to put the header fields back into 2169 * network order! 2170 * XXX This kind of sucks, but we don't expect 2171 * XXX this to happen very often, so maybe it 2172 * XXX doesn't matter so much. 2173 */ 2174 if (tiflags & TH_SYN && 2175 tp->t_state == TCPS_TIME_WAIT && 2176 SEQ_GT(th->th_seq, tp->rcv_nxt)) { 2177 tp = tcp_close(tp); 2178 tcp_fields_to_net(th); 2179 goto findpcb; 2180 } 2181 /* 2182 * If window is closed can only take segments at 2183 * window edge, and have to drop data and PUSH from 2184 * incoming segments. Continue processing, but 2185 * remember to ack. Otherwise, drop segment 2186 * and (if not RST) ack. 2187 */ 2188 if (tp->rcv_wnd == 0 && th->th_seq == tp->rcv_nxt) { 2189 tp->t_flags |= TF_ACKNOW; 2190 TCP_STATINC(TCP_STAT_RCVWINPROBE); 2191 } else 2192 goto dropafterack; 2193 } else 2194 TCP_STATADD(TCP_STAT_RCVBYTEAFTERWIN, todrop); 2195 m_adj(m, -todrop); 2196 tlen -= todrop; 2197 tiflags &= ~(TH_PUSH|TH_FIN); 2198 } 2199 2200 /* 2201 * If last ACK falls within this segment's sequence numbers, 2202 * record the timestamp. 2203 * NOTE: 2204 * 1) That the test incorporates suggestions from the latest 2205 * proposal of the tcplw (at) cray.com list (Braden 1993/04/26). 2206 * 2) That updating only on newer timestamps interferes with 2207 * our earlier PAWS tests, so this check should be solely 2208 * predicated on the sequence space of this segment. 2209 * 3) That we modify the segment boundary check to be 2210 * Last.ACK.Sent <= SEG.SEQ + SEG.Len 2211 * instead of RFC1323's 2212 * Last.ACK.Sent < SEG.SEQ + SEG.Len, 2213 * This modified check allows us to overcome RFC1323's 2214 * limitations as described in Stevens TCP/IP Illustrated 2215 * Vol. 2 p.869. In such cases, we can still calculate the 2216 * RTT correctly when RCV.NXT == Last.ACK.Sent. 2217 */ 2218 if (opti.ts_present && 2219 SEQ_LEQ(th->th_seq, tp->last_ack_sent) && 2220 SEQ_LEQ(tp->last_ack_sent, th->th_seq + tlen + 2221 ((tiflags & (TH_SYN|TH_FIN)) != 0))) { 2222 tp->ts_recent_age = tcp_now; 2223 tp->ts_recent = opti.ts_val; 2224 } 2225 2226 /* 2227 * If the RST bit is set examine the state: 2228 * SYN_RECEIVED STATE: 2229 * If passive open, return to LISTEN state. 2230 * If active open, inform user that connection was refused. 2231 * ESTABLISHED, FIN_WAIT_1, FIN_WAIT2, CLOSE_WAIT STATES: 2232 * Inform user that connection was reset, and close tcb. 2233 * CLOSING, LAST_ACK, TIME_WAIT STATES 2234 * Close the tcb. 2235 */ 2236 if (tiflags & TH_RST) { 2237 if (th->th_seq != tp->rcv_nxt) 2238 goto dropafterack_ratelim; 2239 2240 switch (tp->t_state) { 2241 case TCPS_SYN_RECEIVED: 2242 so->so_error = ECONNREFUSED; 2243 goto close; 2244 2245 case TCPS_ESTABLISHED: 2246 case TCPS_FIN_WAIT_1: 2247 case TCPS_FIN_WAIT_2: 2248 case TCPS_CLOSE_WAIT: 2249 so->so_error = ECONNRESET; 2250 close: 2251 tp->t_state = TCPS_CLOSED; 2252 TCP_STATINC(TCP_STAT_DROPS); 2253 tp = tcp_close(tp); 2254 goto drop; 2255 2256 case TCPS_CLOSING: 2257 case TCPS_LAST_ACK: 2258 case TCPS_TIME_WAIT: 2259 tp = tcp_close(tp); 2260 goto drop; 2261 } 2262 } 2263 2264 /* 2265 * Since we've covered the SYN-SENT and SYN-RECEIVED states above 2266 * we must be in a synchronized state. RFC791 states (under RST 2267 * generation) that any unacceptable segment (an out-of-order SYN 2268 * qualifies) received in a synchronized state must elicit only an 2269 * empty acknowledgment segment ... and the connection remains in 2270 * the same state. 2271 */ 2272 if (tiflags & TH_SYN) { 2273 if (tp->rcv_nxt == th->th_seq) { 2274 tcp_respond(tp, m, m, th, (tcp_seq)0, th->th_ack - 1, 2275 TH_ACK); 2276 if (tcp_saveti) 2277 m_freem(tcp_saveti); 2278 return; 2279 } 2280 2281 goto dropafterack_ratelim; 2282 } 2283 2284 /* 2285 * If the ACK bit is off we drop the segment and return. 2286 */ 2287 if ((tiflags & TH_ACK) == 0) { 2288 if (tp->t_flags & TF_ACKNOW) 2289 goto dropafterack; 2290 else 2291 goto drop; 2292 } 2293 2294 /* 2295 * Ack processing. 2296 */ 2297 switch (tp->t_state) { 2298 2299 /* 2300 * In SYN_RECEIVED state if the ack ACKs our SYN then enter 2301 * ESTABLISHED state and continue processing, otherwise 2302 * send an RST. 2303 */ 2304 case TCPS_SYN_RECEIVED: 2305 if (SEQ_GT(tp->snd_una, th->th_ack) || 2306 SEQ_GT(th->th_ack, tp->snd_max)) 2307 goto dropwithreset; 2308 TCP_STATINC(TCP_STAT_CONNECTS); 2309 soisconnected(so); 2310 tcp_established(tp); 2311 /* Do window scaling? */ 2312 if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) == 2313 (TF_RCVD_SCALE|TF_REQ_SCALE)) { 2314 tp->snd_scale = tp->requested_s_scale; 2315 tp->rcv_scale = tp->request_r_scale; 2316 } 2317 TCP_REASS_LOCK(tp); 2318 (void) tcp_reass(tp, NULL, (struct mbuf *)0, &tlen); 2319 TCP_REASS_UNLOCK(tp); 2320 tp->snd_wl1 = th->th_seq - 1; 2321 /* fall into ... */ 2322 2323 /* 2324 * In ESTABLISHED state: drop duplicate ACKs; ACK out of range 2325 * ACKs. If the ack is in the range 2326 * tp->snd_una < th->th_ack <= tp->snd_max 2327 * then advance tp->snd_una to th->th_ack and drop 2328 * data from the retransmission queue. If this ACK reflects 2329 * more up to date window information we update our window information. 2330 */ 2331 case TCPS_ESTABLISHED: 2332 case TCPS_FIN_WAIT_1: 2333 case TCPS_FIN_WAIT_2: 2334 case TCPS_CLOSE_WAIT: 2335 case TCPS_CLOSING: 2336 case TCPS_LAST_ACK: 2337 case TCPS_TIME_WAIT: 2338 2339 if (SEQ_LEQ(th->th_ack, tp->snd_una)) { 2340 if (tlen == 0 && !dupseg && tiwin == tp->snd_wnd) { 2341 TCP_STATINC(TCP_STAT_RCVDUPPACK); 2342 /* 2343 * If we have outstanding data (other than 2344 * a window probe), this is a completely 2345 * duplicate ack (ie, window info didn't 2346 * change), the ack is the biggest we've 2347 * seen and we've seen exactly our rexmt 2348 * threshhold of them, assume a packet 2349 * has been dropped and retransmit it. 2350 * Kludge snd_nxt & the congestion 2351 * window so we send only this one 2352 * packet. 2353 */ 2354 if (TCP_TIMER_ISARMED(tp, TCPT_REXMT) == 0 || 2355 th->th_ack != tp->snd_una) 2356 tp->t_dupacks = 0; 2357 else if (tp->t_partialacks < 0 && 2358 (++tp->t_dupacks == tcprexmtthresh || 2359 TCP_FACK_FASTRECOV(tp))) { 2360 /* 2361 * Do the fast retransmit, and adjust 2362 * congestion control paramenters. 2363 */ 2364 if (tp->t_congctl->fast_retransmit(tp, th)) { 2365 /* False fast retransmit */ 2366 break; 2367 } else 2368 goto drop; 2369 } else if (tp->t_dupacks > tcprexmtthresh) { 2370 tp->snd_cwnd += tp->t_segsz; 2371 (void) tcp_output(tp); 2372 goto drop; 2373 } 2374 } else { 2375 /* 2376 * If the ack appears to be very old, only 2377 * allow data that is in-sequence. This 2378 * makes it somewhat more difficult to insert 2379 * forged data by guessing sequence numbers. 2380 * Sent an ack to try to update the send 2381 * sequence number on the other side. 2382 */ 2383 if (tlen && th->th_seq != tp->rcv_nxt && 2384 SEQ_LT(th->th_ack, 2385 tp->snd_una - tp->max_sndwnd)) 2386 goto dropafterack; 2387 } 2388 break; 2389 } 2390 /* 2391 * If the congestion window was inflated to account 2392 * for the other side's cached packets, retract it. 2393 */ 2394 /* XXX: make SACK have his own congestion control 2395 * struct -- rpaulo */ 2396 if (TCP_SACK_ENABLED(tp)) 2397 tcp_sack_newack(tp, th); 2398 else 2399 tp->t_congctl->fast_retransmit_newack(tp, th); 2400 if (SEQ_GT(th->th_ack, tp->snd_max)) { 2401 TCP_STATINC(TCP_STAT_RCVACKTOOMUCH); 2402 goto dropafterack; 2403 } 2404 acked = th->th_ack - tp->snd_una; 2405 tcps = TCP_STAT_GETREF(); 2406 tcps[TCP_STAT_RCVACKPACK]++; 2407 tcps[TCP_STAT_RCVACKBYTE] += acked; 2408 TCP_STAT_PUTREF(); 2409 2410 /* 2411 * If we have a timestamp reply, update smoothed 2412 * round trip time. If no timestamp is present but 2413 * transmit timer is running and timed sequence 2414 * number was acked, update smoothed round trip time. 2415 * Since we now have an rtt measurement, cancel the 2416 * timer backoff (cf., Phil Karn's retransmit alg.). 2417 * Recompute the initial retransmit timer. 2418 */ 2419 if (ts_rtt) 2420 tcp_xmit_timer(tp, ts_rtt); 2421 else if (tp->t_rtttime && SEQ_GT(th->th_ack, tp->t_rtseq)) 2422 tcp_xmit_timer(tp, tcp_now - tp->t_rtttime); 2423 2424 /* 2425 * If all outstanding data is acked, stop retransmit 2426 * timer and remember to restart (more output or persist). 2427 * If there is more data to be acked, restart retransmit 2428 * timer, using current (possibly backed-off) value. 2429 */ 2430 if (th->th_ack == tp->snd_max) { 2431 TCP_TIMER_DISARM(tp, TCPT_REXMT); 2432 needoutput = 1; 2433 } else if (TCP_TIMER_ISARMED(tp, TCPT_PERSIST) == 0) 2434 TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur); 2435 2436 /* 2437 * New data has been acked, adjust the congestion window. 2438 */ 2439 tp->t_congctl->newack(tp, th); 2440 2441 nd6_hint(tp); 2442 if (acked > so->so_snd.sb_cc) { 2443 tp->snd_wnd -= so->so_snd.sb_cc; 2444 sbdrop(&so->so_snd, (int)so->so_snd.sb_cc); 2445 ourfinisacked = 1; 2446 } else { 2447 if (acked > (tp->t_lastoff - tp->t_inoff)) 2448 tp->t_lastm = NULL; 2449 sbdrop(&so->so_snd, acked); 2450 tp->t_lastoff -= acked; 2451 tp->snd_wnd -= acked; 2452 ourfinisacked = 0; 2453 } 2454 sowwakeup(so); 2455 2456 icmp_check(tp, th, acked); 2457 2458 tp->snd_una = th->th_ack; 2459 if (SEQ_GT(tp->snd_una, tp->snd_fack)) 2460 tp->snd_fack = tp->snd_una; 2461 if (SEQ_LT(tp->snd_nxt, tp->snd_una)) 2462 tp->snd_nxt = tp->snd_una; 2463 if (SEQ_LT(tp->snd_high, tp->snd_una)) 2464 tp->snd_high = tp->snd_una; 2465 2466 switch (tp->t_state) { 2467 2468 /* 2469 * In FIN_WAIT_1 STATE in addition to the processing 2470 * for the ESTABLISHED state if our FIN is now acknowledged 2471 * then enter FIN_WAIT_2. 2472 */ 2473 case TCPS_FIN_WAIT_1: 2474 if (ourfinisacked) { 2475 /* 2476 * If we can't receive any more 2477 * data, then closing user can proceed. 2478 * Starting the timer is contrary to the 2479 * specification, but if we don't get a FIN 2480 * we'll hang forever. 2481 */ 2482 if (so->so_state & SS_CANTRCVMORE) { 2483 soisdisconnected(so); 2484 if (tp->t_maxidle > 0) 2485 TCP_TIMER_ARM(tp, TCPT_2MSL, 2486 tp->t_maxidle); 2487 } 2488 tp->t_state = TCPS_FIN_WAIT_2; 2489 } 2490 break; 2491 2492 /* 2493 * In CLOSING STATE in addition to the processing for 2494 * the ESTABLISHED state if the ACK acknowledges our FIN 2495 * then enter the TIME-WAIT state, otherwise ignore 2496 * the segment. 2497 */ 2498 case TCPS_CLOSING: 2499 if (ourfinisacked) { 2500 tp->t_state = TCPS_TIME_WAIT; 2501 tcp_canceltimers(tp); 2502 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2503 soisdisconnected(so); 2504 } 2505 break; 2506 2507 /* 2508 * In LAST_ACK, we may still be waiting for data to drain 2509 * and/or to be acked, as well as for the ack of our FIN. 2510 * If our FIN is now acknowledged, delete the TCB, 2511 * enter the closed state and return. 2512 */ 2513 case TCPS_LAST_ACK: 2514 if (ourfinisacked) { 2515 tp = tcp_close(tp); 2516 goto drop; 2517 } 2518 break; 2519 2520 /* 2521 * In TIME_WAIT state the only thing that should arrive 2522 * is a retransmission of the remote FIN. Acknowledge 2523 * it and restart the finack timer. 2524 */ 2525 case TCPS_TIME_WAIT: 2526 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2527 goto dropafterack; 2528 } 2529 } 2530 2531 step6: 2532 /* 2533 * Update window information. 2534 * Don't look at window if no ACK: TAC's send garbage on first SYN. 2535 */ 2536 if ((tiflags & TH_ACK) && (SEQ_LT(tp->snd_wl1, th->th_seq) || 2537 (tp->snd_wl1 == th->th_seq && (SEQ_LT(tp->snd_wl2, th->th_ack) || 2538 (tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd))))) { 2539 /* keep track of pure window updates */ 2540 if (tlen == 0 && 2541 tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd) 2542 TCP_STATINC(TCP_STAT_RCVWINUPD); 2543 tp->snd_wnd = tiwin; 2544 tp->snd_wl1 = th->th_seq; 2545 tp->snd_wl2 = th->th_ack; 2546 if (tp->snd_wnd > tp->max_sndwnd) 2547 tp->max_sndwnd = tp->snd_wnd; 2548 needoutput = 1; 2549 } 2550 2551 /* 2552 * Process segments with URG. 2553 */ 2554 if ((tiflags & TH_URG) && th->th_urp && 2555 TCPS_HAVERCVDFIN(tp->t_state) == 0) { 2556 /* 2557 * This is a kludge, but if we receive and accept 2558 * random urgent pointers, we'll crash in 2559 * soreceive. It's hard to imagine someone 2560 * actually wanting to send this much urgent data. 2561 */ 2562 if (th->th_urp + so->so_rcv.sb_cc > sb_max) { 2563 th->th_urp = 0; /* XXX */ 2564 tiflags &= ~TH_URG; /* XXX */ 2565 goto dodata; /* XXX */ 2566 } 2567 /* 2568 * If this segment advances the known urgent pointer, 2569 * then mark the data stream. This should not happen 2570 * in CLOSE_WAIT, CLOSING, LAST_ACK or TIME_WAIT STATES since 2571 * a FIN has been received from the remote side. 2572 * In these states we ignore the URG. 2573 * 2574 * According to RFC961 (Assigned Protocols), 2575 * the urgent pointer points to the last octet 2576 * of urgent data. We continue, however, 2577 * to consider it to indicate the first octet 2578 * of data past the urgent section as the original 2579 * spec states (in one of two places). 2580 */ 2581 if (SEQ_GT(th->th_seq+th->th_urp, tp->rcv_up)) { 2582 tp->rcv_up = th->th_seq + th->th_urp; 2583 so->so_oobmark = so->so_rcv.sb_cc + 2584 (tp->rcv_up - tp->rcv_nxt) - 1; 2585 if (so->so_oobmark == 0) 2586 so->so_state |= SS_RCVATMARK; 2587 sohasoutofband(so); 2588 tp->t_oobflags &= ~(TCPOOB_HAVEDATA | TCPOOB_HADDATA); 2589 } 2590 /* 2591 * Remove out of band data so doesn't get presented to user. 2592 * This can happen independent of advancing the URG pointer, 2593 * but if two URG's are pending at once, some out-of-band 2594 * data may creep in... ick. 2595 */ 2596 if (th->th_urp <= (u_int16_t) tlen 2597 #ifdef SO_OOBINLINE 2598 && (so->so_options & SO_OOBINLINE) == 0 2599 #endif 2600 ) 2601 tcp_pulloutofband(so, th, m, hdroptlen); 2602 } else 2603 /* 2604 * If no out of band data is expected, 2605 * pull receive urgent pointer along 2606 * with the receive window. 2607 */ 2608 if (SEQ_GT(tp->rcv_nxt, tp->rcv_up)) 2609 tp->rcv_up = tp->rcv_nxt; 2610 dodata: /* XXX */ 2611 2612 /* 2613 * Process the segment text, merging it into the TCP sequencing queue, 2614 * and arranging for acknowledgement of receipt if necessary. 2615 * This process logically involves adjusting tp->rcv_wnd as data 2616 * is presented to the user (this happens in tcp_usrreq.c, 2617 * case PRU_RCVD). If a FIN has already been received on this 2618 * connection then we just ignore the text. 2619 */ 2620 if ((tlen || (tiflags & TH_FIN)) && 2621 TCPS_HAVERCVDFIN(tp->t_state) == 0) { 2622 /* 2623 * Insert segment ti into reassembly queue of tcp with 2624 * control block tp. Return TH_FIN if reassembly now includes 2625 * a segment with FIN. The macro form does the common case 2626 * inline (segment is the next to be received on an 2627 * established connection, and the queue is empty), 2628 * avoiding linkage into and removal from the queue and 2629 * repetition of various conversions. 2630 * Set DELACK for segments received in order, but ack 2631 * immediately when segments are out of order 2632 * (so fast retransmit can work). 2633 */ 2634 /* NOTE: this was TCP_REASS() macro, but used only once */ 2635 TCP_REASS_LOCK(tp); 2636 if (th->th_seq == tp->rcv_nxt && 2637 TAILQ_FIRST(&tp->segq) == NULL && 2638 tp->t_state == TCPS_ESTABLISHED) { 2639 tcp_setup_ack(tp, th); 2640 tp->rcv_nxt += tlen; 2641 tiflags = th->th_flags & TH_FIN; 2642 tcps = TCP_STAT_GETREF(); 2643 tcps[TCP_STAT_RCVPACK]++; 2644 tcps[TCP_STAT_RCVBYTE] += tlen; 2645 TCP_STAT_PUTREF(); 2646 nd6_hint(tp); 2647 if (so->so_state & SS_CANTRCVMORE) 2648 m_freem(m); 2649 else { 2650 m_adj(m, hdroptlen); 2651 sbappendstream(&(so)->so_rcv, m); 2652 } 2653 TCP_REASS_UNLOCK(tp); 2654 sorwakeup(so); 2655 } else { 2656 m_adj(m, hdroptlen); 2657 tiflags = tcp_reass(tp, th, m, &tlen); 2658 tp->t_flags |= TF_ACKNOW; 2659 TCP_REASS_UNLOCK(tp); 2660 } 2661 2662 /* 2663 * Note the amount of data that peer has sent into 2664 * our window, in order to estimate the sender's 2665 * buffer size. 2666 */ 2667 len = so->so_rcv.sb_hiwat - (tp->rcv_adv - tp->rcv_nxt); 2668 } else { 2669 m_freem(m); 2670 m = NULL; 2671 tiflags &= ~TH_FIN; 2672 } 2673 2674 /* 2675 * If FIN is received ACK the FIN and let the user know 2676 * that the connection is closing. Ignore a FIN received before 2677 * the connection is fully established. 2678 */ 2679 if ((tiflags & TH_FIN) && TCPS_HAVEESTABLISHED(tp->t_state)) { 2680 if (TCPS_HAVERCVDFIN(tp->t_state) == 0) { 2681 socantrcvmore(so); 2682 tp->t_flags |= TF_ACKNOW; 2683 tp->rcv_nxt++; 2684 } 2685 switch (tp->t_state) { 2686 2687 /* 2688 * In ESTABLISHED STATE enter the CLOSE_WAIT state. 2689 */ 2690 case TCPS_ESTABLISHED: 2691 tp->t_state = TCPS_CLOSE_WAIT; 2692 break; 2693 2694 /* 2695 * If still in FIN_WAIT_1 STATE FIN has not been acked so 2696 * enter the CLOSING state. 2697 */ 2698 case TCPS_FIN_WAIT_1: 2699 tp->t_state = TCPS_CLOSING; 2700 break; 2701 2702 /* 2703 * In FIN_WAIT_2 state enter the TIME_WAIT state, 2704 * starting the time-wait timer, turning off the other 2705 * standard timers. 2706 */ 2707 case TCPS_FIN_WAIT_2: 2708 tp->t_state = TCPS_TIME_WAIT; 2709 tcp_canceltimers(tp); 2710 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2711 soisdisconnected(so); 2712 break; 2713 2714 /* 2715 * In TIME_WAIT state restart the 2 MSL time_wait timer. 2716 */ 2717 case TCPS_TIME_WAIT: 2718 TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * TCPTV_MSL); 2719 break; 2720 } 2721 } 2722 #ifdef TCP_DEBUG 2723 if (so->so_options & SO_DEBUG) 2724 tcp_trace(TA_INPUT, ostate, tp, tcp_saveti, 0); 2725 #endif 2726 2727 /* 2728 * Return any desired output. 2729 */ 2730 if (needoutput || (tp->t_flags & TF_ACKNOW)) { 2731 (void) tcp_output(tp); 2732 } 2733 if (tcp_saveti) 2734 m_freem(tcp_saveti); 2735 return; 2736 2737 badsyn: 2738 /* 2739 * Received a bad SYN. Increment counters and dropwithreset. 2740 */ 2741 TCP_STATINC(TCP_STAT_BADSYN); 2742 tp = NULL; 2743 goto dropwithreset; 2744 2745 dropafterack: 2746 /* 2747 * Generate an ACK dropping incoming segment if it occupies 2748 * sequence space, where the ACK reflects our state. 2749 */ 2750 if (tiflags & TH_RST) 2751 goto drop; 2752 goto dropafterack2; 2753 2754 dropafterack_ratelim: 2755 /* 2756 * We may want to rate-limit ACKs against SYN/RST attack. 2757 */ 2758 if (ppsratecheck(&tcp_ackdrop_ppslim_last, &tcp_ackdrop_ppslim_count, 2759 tcp_ackdrop_ppslim) == 0) { 2760 /* XXX stat */ 2761 goto drop; 2762 } 2763 /* ...fall into dropafterack2... */ 2764 2765 dropafterack2: 2766 m_freem(m); 2767 tp->t_flags |= TF_ACKNOW; 2768 (void) tcp_output(tp); 2769 if (tcp_saveti) 2770 m_freem(tcp_saveti); 2771 return; 2772 2773 dropwithreset_ratelim: 2774 /* 2775 * We may want to rate-limit RSTs in certain situations, 2776 * particularly if we are sending an RST in response to 2777 * an attempt to connect to or otherwise communicate with 2778 * a port for which we have no socket. 2779 */ 2780 if (ppsratecheck(&tcp_rst_ppslim_last, &tcp_rst_ppslim_count, 2781 tcp_rst_ppslim) == 0) { 2782 /* XXX stat */ 2783 goto drop; 2784 } 2785 /* ...fall into dropwithreset... */ 2786 2787 dropwithreset: 2788 /* 2789 * Generate a RST, dropping incoming segment. 2790 * Make ACK acceptable to originator of segment. 2791 */ 2792 if (tiflags & TH_RST) 2793 goto drop; 2794 2795 switch (af) { 2796 #ifdef INET6 2797 case AF_INET6: 2798 /* For following calls to tcp_respond */ 2799 if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) 2800 goto drop; 2801 break; 2802 #endif /* INET6 */ 2803 case AF_INET: 2804 if (IN_MULTICAST(ip->ip_dst.s_addr) || 2805 in_broadcast(ip->ip_dst, m->m_pkthdr.rcvif)) 2806 goto drop; 2807 } 2808 2809 if (tiflags & TH_ACK) 2810 (void)tcp_respond(tp, m, m, th, (tcp_seq)0, th->th_ack, TH_RST); 2811 else { 2812 if (tiflags & TH_SYN) 2813 tlen++; 2814 (void)tcp_respond(tp, m, m, th, th->th_seq + tlen, (tcp_seq)0, 2815 TH_RST|TH_ACK); 2816 } 2817 if (tcp_saveti) 2818 m_freem(tcp_saveti); 2819 return; 2820 2821 badcsum: 2822 drop: 2823 /* 2824 * Drop space held by incoming segment and return. 2825 */ 2826 if (tp) { 2827 if (tp->t_inpcb) 2828 so = tp->t_inpcb->inp_socket; 2829 #ifdef INET6 2830 else if (tp->t_in6pcb) 2831 so = tp->t_in6pcb->in6p_socket; 2832 #endif 2833 else 2834 so = NULL; 2835 #ifdef TCP_DEBUG 2836 if (so && (so->so_options & SO_DEBUG) != 0) 2837 tcp_trace(TA_DROP, ostate, tp, tcp_saveti, 0); 2838 #endif 2839 } 2840 if (tcp_saveti) 2841 m_freem(tcp_saveti); 2842 m_freem(m); 2843 return; 2844 } 2845 2846 #ifdef TCP_SIGNATURE 2847 int 2848 tcp_signature_apply(void *fstate, void *data, u_int len) 2849 { 2850 2851 MD5Update(fstate, (u_char *)data, len); 2852 return (0); 2853 } 2854 2855 struct secasvar * 2856 tcp_signature_getsav(struct mbuf *m, struct tcphdr *th) 2857 { 2858 struct secasvar *sav; 2859 #ifdef FAST_IPSEC 2860 union sockaddr_union dst; 2861 #endif 2862 struct ip *ip; 2863 struct ip6_hdr *ip6; 2864 2865 ip = mtod(m, struct ip *); 2866 switch (ip->ip_v) { 2867 case 4: 2868 ip = mtod(m, struct ip *); 2869 ip6 = NULL; 2870 break; 2871 case 6: 2872 ip = NULL; 2873 ip6 = mtod(m, struct ip6_hdr *); 2874 break; 2875 default: 2876 return (NULL); 2877 } 2878 2879 #ifdef FAST_IPSEC 2880 /* Extract the destination from the IP header in the mbuf. */ 2881 memset(&dst, 0, sizeof(union sockaddr_union)); 2882 if (ip !=NULL) { 2883 dst.sa.sa_len = sizeof(struct sockaddr_in); 2884 dst.sa.sa_family = AF_INET; 2885 dst.sin.sin_addr = ip->ip_dst; 2886 } else { 2887 dst.sa.sa_len = sizeof(struct sockaddr_in6); 2888 dst.sa.sa_family = AF_INET6; 2889 dst.sin6.sin6_addr = ip6->ip6_dst; 2890 } 2891 2892 /* 2893 * Look up an SADB entry which matches the address of the peer. 2894 */ 2895 sav = KEY_ALLOCSA(&dst, IPPROTO_TCP, htonl(TCP_SIG_SPI)); 2896 #else 2897 if (ip) 2898 sav = key_allocsa(AF_INET, (void *)&ip->ip_src, 2899 (void *)&ip->ip_dst, IPPROTO_TCP, 2900 htonl(TCP_SIG_SPI), 0, 0); 2901 else 2902 sav = key_allocsa(AF_INET6, (void *)&ip6->ip6_src, 2903 (void *)&ip6->ip6_dst, IPPROTO_TCP, 2904 htonl(TCP_SIG_SPI), 0, 0); 2905 #endif 2906 2907 return (sav); /* freesav must be performed by caller */ 2908 } 2909 2910 int 2911 tcp_signature(struct mbuf *m, struct tcphdr *th, int thoff, 2912 struct secasvar *sav, char *sig) 2913 { 2914 MD5_CTX ctx; 2915 struct ip *ip; 2916 struct ipovly *ipovly; 2917 struct ip6_hdr *ip6; 2918 struct ippseudo ippseudo; 2919 struct ip6_hdr_pseudo ip6pseudo; 2920 struct tcphdr th0; 2921 int l, tcphdrlen; 2922 2923 if (sav == NULL) 2924 return (-1); 2925 2926 tcphdrlen = th->th_off * 4; 2927 2928 switch (mtod(m, struct ip *)->ip_v) { 2929 case 4: 2930 ip = mtod(m, struct ip *); 2931 ip6 = NULL; 2932 break; 2933 case 6: 2934 ip = NULL; 2935 ip6 = mtod(m, struct ip6_hdr *); 2936 break; 2937 default: 2938 return (-1); 2939 } 2940 2941 MD5Init(&ctx); 2942 2943 if (ip) { 2944 memset(&ippseudo, 0, sizeof(ippseudo)); 2945 ipovly = (struct ipovly *)ip; 2946 ippseudo.ippseudo_src = ipovly->ih_src; 2947 ippseudo.ippseudo_dst = ipovly->ih_dst; 2948 ippseudo.ippseudo_pad = 0; 2949 ippseudo.ippseudo_p = IPPROTO_TCP; 2950 ippseudo.ippseudo_len = htons(m->m_pkthdr.len - thoff); 2951 MD5Update(&ctx, (char *)&ippseudo, sizeof(ippseudo)); 2952 } else { 2953 memset(&ip6pseudo, 0, sizeof(ip6pseudo)); 2954 ip6pseudo.ip6ph_src = ip6->ip6_src; 2955 in6_clearscope(&ip6pseudo.ip6ph_src); 2956 ip6pseudo.ip6ph_dst = ip6->ip6_dst; 2957 in6_clearscope(&ip6pseudo.ip6ph_dst); 2958 ip6pseudo.ip6ph_len = htons(m->m_pkthdr.len - thoff); 2959 ip6pseudo.ip6ph_nxt = IPPROTO_TCP; 2960 MD5Update(&ctx, (char *)&ip6pseudo, sizeof(ip6pseudo)); 2961 } 2962 2963 th0 = *th; 2964 th0.th_sum = 0; 2965 MD5Update(&ctx, (char *)&th0, sizeof(th0)); 2966 2967 l = m->m_pkthdr.len - thoff - tcphdrlen; 2968 if (l > 0) 2969 m_apply(m, thoff + tcphdrlen, 2970 m->m_pkthdr.len - thoff - tcphdrlen, 2971 tcp_signature_apply, &ctx); 2972 2973 MD5Update(&ctx, _KEYBUF(sav->key_auth), _KEYLEN(sav->key_auth)); 2974 MD5Final(sig, &ctx); 2975 2976 return (0); 2977 } 2978 #endif 2979 2980 static int 2981 tcp_dooptions(struct tcpcb *tp, const u_char *cp, int cnt, 2982 struct tcphdr *th, 2983 struct mbuf *m, int toff, struct tcp_opt_info *oi) 2984 { 2985 u_int16_t mss; 2986 int opt, optlen = 0; 2987 #ifdef TCP_SIGNATURE 2988 void *sigp = NULL; 2989 char sigbuf[TCP_SIGLEN]; 2990 struct secasvar *sav = NULL; 2991 #endif 2992 2993 for (; cp && cnt > 0; cnt -= optlen, cp += optlen) { 2994 opt = cp[0]; 2995 if (opt == TCPOPT_EOL) 2996 break; 2997 if (opt == TCPOPT_NOP) 2998 optlen = 1; 2999 else { 3000 if (cnt < 2) 3001 break; 3002 optlen = cp[1]; 3003 if (optlen < 2 || optlen > cnt) 3004 break; 3005 } 3006 switch (opt) { 3007 3008 default: 3009 continue; 3010 3011 case TCPOPT_MAXSEG: 3012 if (optlen != TCPOLEN_MAXSEG) 3013 continue; 3014 if (!(th->th_flags & TH_SYN)) 3015 continue; 3016 if (TCPS_HAVERCVDSYN(tp->t_state)) 3017 continue; 3018 bcopy(cp + 2, &mss, sizeof(mss)); 3019 oi->maxseg = ntohs(mss); 3020 break; 3021 3022 case TCPOPT_WINDOW: 3023 if (optlen != TCPOLEN_WINDOW) 3024 continue; 3025 if (!(th->th_flags & TH_SYN)) 3026 continue; 3027 if (TCPS_HAVERCVDSYN(tp->t_state)) 3028 continue; 3029 tp->t_flags |= TF_RCVD_SCALE; 3030 tp->requested_s_scale = cp[2]; 3031 if (tp->requested_s_scale > TCP_MAX_WINSHIFT) { 3032 #if 0 /*XXX*/ 3033 char *p; 3034 3035 if (ip) 3036 p = ntohl(ip->ip_src); 3037 #ifdef INET6 3038 else if (ip6) 3039 p = ip6_sprintf(&ip6->ip6_src); 3040 #endif 3041 else 3042 p = "(unknown)"; 3043 log(LOG_ERR, "TCP: invalid wscale %d from %s, " 3044 "assuming %d\n", 3045 tp->requested_s_scale, p, 3046 TCP_MAX_WINSHIFT); 3047 #else 3048 log(LOG_ERR, "TCP: invalid wscale %d, " 3049 "assuming %d\n", 3050 tp->requested_s_scale, 3051 TCP_MAX_WINSHIFT); 3052 #endif 3053 tp->requested_s_scale = TCP_MAX_WINSHIFT; 3054 } 3055 break; 3056 3057 case TCPOPT_TIMESTAMP: 3058 if (optlen != TCPOLEN_TIMESTAMP) 3059 continue; 3060 oi->ts_present = 1; 3061 bcopy(cp + 2, &oi->ts_val, sizeof(oi->ts_val)); 3062 NTOHL(oi->ts_val); 3063 bcopy(cp + 6, &oi->ts_ecr, sizeof(oi->ts_ecr)); 3064 NTOHL(oi->ts_ecr); 3065 3066 if (!(th->th_flags & TH_SYN)) 3067 continue; 3068 if (TCPS_HAVERCVDSYN(tp->t_state)) 3069 continue; 3070 /* 3071 * A timestamp received in a SYN makes 3072 * it ok to send timestamp requests and replies. 3073 */ 3074 tp->t_flags |= TF_RCVD_TSTMP; 3075 tp->ts_recent = oi->ts_val; 3076 tp->ts_recent_age = tcp_now; 3077 break; 3078 3079 case TCPOPT_SACK_PERMITTED: 3080 if (optlen != TCPOLEN_SACK_PERMITTED) 3081 continue; 3082 if (!(th->th_flags & TH_SYN)) 3083 continue; 3084 if (TCPS_HAVERCVDSYN(tp->t_state)) 3085 continue; 3086 if (tcp_do_sack) { 3087 tp->t_flags |= TF_SACK_PERMIT; 3088 tp->t_flags |= TF_WILL_SACK; 3089 } 3090 break; 3091 3092 case TCPOPT_SACK: 3093 tcp_sack_option(tp, th, cp, optlen); 3094 break; 3095 #ifdef TCP_SIGNATURE 3096 case TCPOPT_SIGNATURE: 3097 if (optlen != TCPOLEN_SIGNATURE) 3098 continue; 3099 if (sigp && memcmp(sigp, cp + 2, TCP_SIGLEN)) 3100 return (-1); 3101 3102 sigp = sigbuf; 3103 memcpy(sigbuf, cp + 2, TCP_SIGLEN); 3104 tp->t_flags |= TF_SIGNATURE; 3105 break; 3106 #endif 3107 } 3108 } 3109 3110 #ifdef TCP_SIGNATURE 3111 if (tp->t_flags & TF_SIGNATURE) { 3112 3113 sav = tcp_signature_getsav(m, th); 3114 3115 if (sav == NULL && tp->t_state == TCPS_LISTEN) 3116 return (-1); 3117 } 3118 3119 if ((sigp ? TF_SIGNATURE : 0) ^ (tp->t_flags & TF_SIGNATURE)) { 3120 if (sav == NULL) 3121 return (-1); 3122 #ifdef FAST_IPSEC 3123 KEY_FREESAV(&sav); 3124 #else 3125 key_freesav(sav); 3126 #endif 3127 return (-1); 3128 } 3129 3130 if (sigp) { 3131 char sig[TCP_SIGLEN]; 3132 3133 tcp_fields_to_net(th); 3134 if (tcp_signature(m, th, toff, sav, sig) < 0) { 3135 tcp_fields_to_host(th); 3136 if (sav == NULL) 3137 return (-1); 3138 #ifdef FAST_IPSEC 3139 KEY_FREESAV(&sav); 3140 #else 3141 key_freesav(sav); 3142 #endif 3143 return (-1); 3144 } 3145 tcp_fields_to_host(th); 3146 3147 if (memcmp(sig, sigp, TCP_SIGLEN)) { 3148 TCP_STATINC(TCP_STAT_BADSIG); 3149 if (sav == NULL) 3150 return (-1); 3151 #ifdef FAST_IPSEC 3152 KEY_FREESAV(&sav); 3153 #else 3154 key_freesav(sav); 3155 #endif 3156 return (-1); 3157 } else 3158 TCP_STATINC(TCP_STAT_GOODSIG); 3159 3160 key_sa_recordxfer(sav, m); 3161 #ifdef FAST_IPSEC 3162 KEY_FREESAV(&sav); 3163 #else 3164 key_freesav(sav); 3165 #endif 3166 } 3167 #endif 3168 3169 return (0); 3170 } 3171 3172 /* 3173 * Pull out of band byte out of a segment so 3174 * it doesn't appear in the user's data queue. 3175 * It is still reflected in the segment length for 3176 * sequencing purposes. 3177 */ 3178 void 3179 tcp_pulloutofband(struct socket *so, struct tcphdr *th, 3180 struct mbuf *m, int off) 3181 { 3182 int cnt = off + th->th_urp - 1; 3183 3184 while (cnt >= 0) { 3185 if (m->m_len > cnt) { 3186 char *cp = mtod(m, char *) + cnt; 3187 struct tcpcb *tp = sototcpcb(so); 3188 3189 tp->t_iobc = *cp; 3190 tp->t_oobflags |= TCPOOB_HAVEDATA; 3191 bcopy(cp+1, cp, (unsigned)(m->m_len - cnt - 1)); 3192 m->m_len--; 3193 return; 3194 } 3195 cnt -= m->m_len; 3196 m = m->m_next; 3197 if (m == 0) 3198 break; 3199 } 3200 panic("tcp_pulloutofband"); 3201 } 3202 3203 /* 3204 * Collect new round-trip time estimate 3205 * and update averages and current timeout. 3206 */ 3207 void 3208 tcp_xmit_timer(struct tcpcb *tp, uint32_t rtt) 3209 { 3210 int32_t delta; 3211 3212 TCP_STATINC(TCP_STAT_RTTUPDATED); 3213 if (tp->t_srtt != 0) { 3214 /* 3215 * srtt is stored as fixed point with 3 bits after the 3216 * binary point (i.e., scaled by 8). The following magic 3217 * is equivalent to the smoothing algorithm in rfc793 with 3218 * an alpha of .875 (srtt = rtt/8 + srtt*7/8 in fixed 3219 * point). Adjust rtt to origin 0. 3220 */ 3221 delta = (rtt << 2) - (tp->t_srtt >> TCP_RTT_SHIFT); 3222 if ((tp->t_srtt += delta) <= 0) 3223 tp->t_srtt = 1 << 2; 3224 /* 3225 * We accumulate a smoothed rtt variance (actually, a 3226 * smoothed mean difference), then set the retransmit 3227 * timer to smoothed rtt + 4 times the smoothed variance. 3228 * rttvar is stored as fixed point with 2 bits after the 3229 * binary point (scaled by 4). The following is 3230 * equivalent to rfc793 smoothing with an alpha of .75 3231 * (rttvar = rttvar*3/4 + |delta| / 4). This replaces 3232 * rfc793's wired-in beta. 3233 */ 3234 if (delta < 0) 3235 delta = -delta; 3236 delta -= (tp->t_rttvar >> TCP_RTTVAR_SHIFT); 3237 if ((tp->t_rttvar += delta) <= 0) 3238 tp->t_rttvar = 1 << 2; 3239 } else { 3240 /* 3241 * No rtt measurement yet - use the unsmoothed rtt. 3242 * Set the variance to half the rtt (so our first 3243 * retransmit happens at 3*rtt). 3244 */ 3245 tp->t_srtt = rtt << (TCP_RTT_SHIFT + 2); 3246 tp->t_rttvar = rtt << (TCP_RTTVAR_SHIFT + 2 - 1); 3247 } 3248 tp->t_rtttime = 0; 3249 tp->t_rxtshift = 0; 3250 3251 /* 3252 * the retransmit should happen at rtt + 4 * rttvar. 3253 * Because of the way we do the smoothing, srtt and rttvar 3254 * will each average +1/2 tick of bias. When we compute 3255 * the retransmit timer, we want 1/2 tick of rounding and 3256 * 1 extra tick because of +-1/2 tick uncertainty in the 3257 * firing of the timer. The bias will give us exactly the 3258 * 1.5 tick we need. But, because the bias is 3259 * statistical, we have to test that we don't drop below 3260 * the minimum feasible timer (which is 2 ticks). 3261 */ 3262 TCPT_RANGESET(tp->t_rxtcur, TCP_REXMTVAL(tp), 3263 max(tp->t_rttmin, rtt + 2), TCPTV_REXMTMAX); 3264 3265 /* 3266 * We received an ack for a packet that wasn't retransmitted; 3267 * it is probably safe to discard any error indications we've 3268 * received recently. This isn't quite right, but close enough 3269 * for now (a route might have failed after we sent a segment, 3270 * and the return path might not be symmetrical). 3271 */ 3272 tp->t_softerror = 0; 3273 } 3274 3275 3276 /* 3277 * TCP compressed state engine. Currently used to hold compressed 3278 * state for SYN_RECEIVED. 3279 */ 3280 3281 u_long syn_cache_count; 3282 u_int32_t syn_hash1, syn_hash2; 3283 3284 #define SYN_HASH(sa, sp, dp) \ 3285 ((((sa)->s_addr^syn_hash1)*(((((u_int32_t)(dp))<<16) + \ 3286 ((u_int32_t)(sp)))^syn_hash2))) 3287 #ifndef INET6 3288 #define SYN_HASHALL(hash, src, dst) \ 3289 do { \ 3290 hash = SYN_HASH(&((const struct sockaddr_in *)(src))->sin_addr, \ 3291 ((const struct sockaddr_in *)(src))->sin_port, \ 3292 ((const struct sockaddr_in *)(dst))->sin_port); \ 3293 } while (/*CONSTCOND*/ 0) 3294 #else 3295 #define SYN_HASH6(sa, sp, dp) \ 3296 ((((sa)->s6_addr32[0] ^ (sa)->s6_addr32[3] ^ syn_hash1) * \ 3297 (((((u_int32_t)(dp))<<16) + ((u_int32_t)(sp)))^syn_hash2)) \ 3298 & 0x7fffffff) 3299 3300 #define SYN_HASHALL(hash, src, dst) \ 3301 do { \ 3302 switch ((src)->sa_family) { \ 3303 case AF_INET: \ 3304 hash = SYN_HASH(&((const struct sockaddr_in *)(src))->sin_addr, \ 3305 ((const struct sockaddr_in *)(src))->sin_port, \ 3306 ((const struct sockaddr_in *)(dst))->sin_port); \ 3307 break; \ 3308 case AF_INET6: \ 3309 hash = SYN_HASH6(&((const struct sockaddr_in6 *)(src))->sin6_addr, \ 3310 ((const struct sockaddr_in6 *)(src))->sin6_port, \ 3311 ((const struct sockaddr_in6 *)(dst))->sin6_port); \ 3312 break; \ 3313 default: \ 3314 hash = 0; \ 3315 } \ 3316 } while (/*CONSTCOND*/0) 3317 #endif /* INET6 */ 3318 3319 static struct pool syn_cache_pool; 3320 3321 /* 3322 * We don't estimate RTT with SYNs, so each packet starts with the default 3323 * RTT and each timer step has a fixed timeout value. 3324 */ 3325 #define SYN_CACHE_TIMER_ARM(sc) \ 3326 do { \ 3327 TCPT_RANGESET((sc)->sc_rxtcur, \ 3328 TCPTV_SRTTDFLT * tcp_backoff[(sc)->sc_rxtshift], TCPTV_MIN, \ 3329 TCPTV_REXMTMAX); \ 3330 callout_reset(&(sc)->sc_timer, \ 3331 (sc)->sc_rxtcur * (hz / PR_SLOWHZ), syn_cache_timer, (sc)); \ 3332 } while (/*CONSTCOND*/0) 3333 3334 #define SYN_CACHE_TIMESTAMP(sc) (tcp_now - (sc)->sc_timebase) 3335 3336 static inline void 3337 syn_cache_rm(struct syn_cache *sc) 3338 { 3339 TAILQ_REMOVE(&tcp_syn_cache[sc->sc_bucketidx].sch_bucket, 3340 sc, sc_bucketq); 3341 sc->sc_tp = NULL; 3342 LIST_REMOVE(sc, sc_tpq); 3343 tcp_syn_cache[sc->sc_bucketidx].sch_length--; 3344 callout_stop(&sc->sc_timer); 3345 syn_cache_count--; 3346 } 3347 3348 static inline void 3349 syn_cache_put(struct syn_cache *sc) 3350 { 3351 if (sc->sc_ipopts) 3352 (void) m_free(sc->sc_ipopts); 3353 rtcache_free(&sc->sc_route); 3354 if (callout_invoking(&sc->sc_timer)) 3355 sc->sc_flags |= SCF_DEAD; 3356 else { 3357 callout_destroy(&sc->sc_timer); 3358 pool_put(&syn_cache_pool, sc); 3359 } 3360 } 3361 3362 void 3363 syn_cache_init(void) 3364 { 3365 int i; 3366 3367 pool_init(&syn_cache_pool, sizeof(struct syn_cache), 0, 0, 0, 3368 "synpl", NULL, IPL_SOFTNET); 3369 3370 /* Initialize the hash buckets. */ 3371 for (i = 0; i < tcp_syn_cache_size; i++) 3372 TAILQ_INIT(&tcp_syn_cache[i].sch_bucket); 3373 } 3374 3375 void 3376 syn_cache_insert(struct syn_cache *sc, struct tcpcb *tp) 3377 { 3378 struct syn_cache_head *scp; 3379 struct syn_cache *sc2; 3380 int s; 3381 3382 /* 3383 * If there are no entries in the hash table, reinitialize 3384 * the hash secrets. 3385 */ 3386 if (syn_cache_count == 0) { 3387 syn_hash1 = arc4random(); 3388 syn_hash2 = arc4random(); 3389 } 3390 3391 SYN_HASHALL(sc->sc_hash, &sc->sc_src.sa, &sc->sc_dst.sa); 3392 sc->sc_bucketidx = sc->sc_hash % tcp_syn_cache_size; 3393 scp = &tcp_syn_cache[sc->sc_bucketidx]; 3394 3395 /* 3396 * Make sure that we don't overflow the per-bucket 3397 * limit or the total cache size limit. 3398 */ 3399 s = splsoftnet(); 3400 if (scp->sch_length >= tcp_syn_bucket_limit) { 3401 TCP_STATINC(TCP_STAT_SC_BUCKETOVERFLOW); 3402 /* 3403 * The bucket is full. Toss the oldest element in the 3404 * bucket. This will be the first entry in the bucket. 3405 */ 3406 sc2 = TAILQ_FIRST(&scp->sch_bucket); 3407 #ifdef DIAGNOSTIC 3408 /* 3409 * This should never happen; we should always find an 3410 * entry in our bucket. 3411 */ 3412 if (sc2 == NULL) 3413 panic("syn_cache_insert: bucketoverflow: impossible"); 3414 #endif 3415 syn_cache_rm(sc2); 3416 syn_cache_put(sc2); /* calls pool_put but see spl above */ 3417 } else if (syn_cache_count >= tcp_syn_cache_limit) { 3418 struct syn_cache_head *scp2, *sce; 3419 3420 TCP_STATINC(TCP_STAT_SC_OVERFLOWED); 3421 /* 3422 * The cache is full. Toss the oldest entry in the 3423 * first non-empty bucket we can find. 3424 * 3425 * XXX We would really like to toss the oldest 3426 * entry in the cache, but we hope that this 3427 * condition doesn't happen very often. 3428 */ 3429 scp2 = scp; 3430 if (TAILQ_EMPTY(&scp2->sch_bucket)) { 3431 sce = &tcp_syn_cache[tcp_syn_cache_size]; 3432 for (++scp2; scp2 != scp; scp2++) { 3433 if (scp2 >= sce) 3434 scp2 = &tcp_syn_cache[0]; 3435 if (! TAILQ_EMPTY(&scp2->sch_bucket)) 3436 break; 3437 } 3438 #ifdef DIAGNOSTIC 3439 /* 3440 * This should never happen; we should always find a 3441 * non-empty bucket. 3442 */ 3443 if (scp2 == scp) 3444 panic("syn_cache_insert: cacheoverflow: " 3445 "impossible"); 3446 #endif 3447 } 3448 sc2 = TAILQ_FIRST(&scp2->sch_bucket); 3449 syn_cache_rm(sc2); 3450 syn_cache_put(sc2); /* calls pool_put but see spl above */ 3451 } 3452 3453 /* 3454 * Initialize the entry's timer. 3455 */ 3456 sc->sc_rxttot = 0; 3457 sc->sc_rxtshift = 0; 3458 SYN_CACHE_TIMER_ARM(sc); 3459 3460 /* Link it from tcpcb entry */ 3461 LIST_INSERT_HEAD(&tp->t_sc, sc, sc_tpq); 3462 3463 /* Put it into the bucket. */ 3464 TAILQ_INSERT_TAIL(&scp->sch_bucket, sc, sc_bucketq); 3465 scp->sch_length++; 3466 syn_cache_count++; 3467 3468 TCP_STATINC(TCP_STAT_SC_ADDED); 3469 splx(s); 3470 } 3471 3472 /* 3473 * Walk the timer queues, looking for SYN,ACKs that need to be retransmitted. 3474 * If we have retransmitted an entry the maximum number of times, expire 3475 * that entry. 3476 */ 3477 void 3478 syn_cache_timer(void *arg) 3479 { 3480 struct syn_cache *sc = arg; 3481 3482 mutex_enter(softnet_lock); 3483 KERNEL_LOCK(1, NULL); 3484 callout_ack(&sc->sc_timer); 3485 3486 if (__predict_false(sc->sc_flags & SCF_DEAD)) { 3487 TCP_STATINC(TCP_STAT_SC_DELAYED_FREE); 3488 callout_destroy(&sc->sc_timer); 3489 pool_put(&syn_cache_pool, sc); 3490 KERNEL_UNLOCK_ONE(NULL); 3491 mutex_exit(softnet_lock); 3492 return; 3493 } 3494 3495 if (__predict_false(sc->sc_rxtshift == TCP_MAXRXTSHIFT)) { 3496 /* Drop it -- too many retransmissions. */ 3497 goto dropit; 3498 } 3499 3500 /* 3501 * Compute the total amount of time this entry has 3502 * been on a queue. If this entry has been on longer 3503 * than the keep alive timer would allow, expire it. 3504 */ 3505 sc->sc_rxttot += sc->sc_rxtcur; 3506 if (sc->sc_rxttot >= tcp_keepinit) 3507 goto dropit; 3508 3509 TCP_STATINC(TCP_STAT_SC_RETRANSMITTED); 3510 (void) syn_cache_respond(sc, NULL); 3511 3512 /* Advance the timer back-off. */ 3513 sc->sc_rxtshift++; 3514 SYN_CACHE_TIMER_ARM(sc); 3515 3516 KERNEL_UNLOCK_ONE(NULL); 3517 mutex_exit(softnet_lock); 3518 return; 3519 3520 dropit: 3521 TCP_STATINC(TCP_STAT_SC_TIMED_OUT); 3522 syn_cache_rm(sc); 3523 syn_cache_put(sc); /* calls pool_put but see spl above */ 3524 KERNEL_UNLOCK_ONE(NULL); 3525 mutex_exit(softnet_lock); 3526 } 3527 3528 /* 3529 * Remove syn cache created by the specified tcb entry, 3530 * because this does not make sense to keep them 3531 * (if there's no tcb entry, syn cache entry will never be used) 3532 */ 3533 void 3534 syn_cache_cleanup(struct tcpcb *tp) 3535 { 3536 struct syn_cache *sc, *nsc; 3537 int s; 3538 3539 s = splsoftnet(); 3540 3541 for (sc = LIST_FIRST(&tp->t_sc); sc != NULL; sc = nsc) { 3542 nsc = LIST_NEXT(sc, sc_tpq); 3543 3544 #ifdef DIAGNOSTIC 3545 if (sc->sc_tp != tp) 3546 panic("invalid sc_tp in syn_cache_cleanup"); 3547 #endif 3548 syn_cache_rm(sc); 3549 syn_cache_put(sc); /* calls pool_put but see spl above */ 3550 } 3551 /* just for safety */ 3552 LIST_INIT(&tp->t_sc); 3553 3554 splx(s); 3555 } 3556 3557 /* 3558 * Find an entry in the syn cache. 3559 */ 3560 struct syn_cache * 3561 syn_cache_lookup(const struct sockaddr *src, const struct sockaddr *dst, 3562 struct syn_cache_head **headp) 3563 { 3564 struct syn_cache *sc; 3565 struct syn_cache_head *scp; 3566 u_int32_t hash; 3567 int s; 3568 3569 SYN_HASHALL(hash, src, dst); 3570 3571 scp = &tcp_syn_cache[hash % tcp_syn_cache_size]; 3572 *headp = scp; 3573 s = splsoftnet(); 3574 for (sc = TAILQ_FIRST(&scp->sch_bucket); sc != NULL; 3575 sc = TAILQ_NEXT(sc, sc_bucketq)) { 3576 if (sc->sc_hash != hash) 3577 continue; 3578 if (!memcmp(&sc->sc_src, src, src->sa_len) && 3579 !memcmp(&sc->sc_dst, dst, dst->sa_len)) { 3580 splx(s); 3581 return (sc); 3582 } 3583 } 3584 splx(s); 3585 return (NULL); 3586 } 3587 3588 /* 3589 * This function gets called when we receive an ACK for a 3590 * socket in the LISTEN state. We look up the connection 3591 * in the syn cache, and if its there, we pull it out of 3592 * the cache and turn it into a full-blown connection in 3593 * the SYN-RECEIVED state. 3594 * 3595 * The return values may not be immediately obvious, and their effects 3596 * can be subtle, so here they are: 3597 * 3598 * NULL SYN was not found in cache; caller should drop the 3599 * packet and send an RST. 3600 * 3601 * -1 We were unable to create the new connection, and are 3602 * aborting it. An ACK,RST is being sent to the peer 3603 * (unless we got screwey sequence numbners; see below), 3604 * because the 3-way handshake has been completed. Caller 3605 * should not free the mbuf, since we may be using it. If 3606 * we are not, we will free it. 3607 * 3608 * Otherwise, the return value is a pointer to the new socket 3609 * associated with the connection. 3610 */ 3611 struct socket * 3612 syn_cache_get(struct sockaddr *src, struct sockaddr *dst, 3613 struct tcphdr *th, unsigned int hlen, unsigned int tlen, 3614 struct socket *so, struct mbuf *m) 3615 { 3616 struct syn_cache *sc; 3617 struct syn_cache_head *scp; 3618 struct inpcb *inp = NULL; 3619 #ifdef INET6 3620 struct in6pcb *in6p = NULL; 3621 #endif 3622 struct tcpcb *tp = 0; 3623 struct mbuf *am; 3624 int s; 3625 struct socket *oso; 3626 3627 s = splsoftnet(); 3628 if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) { 3629 splx(s); 3630 return (NULL); 3631 } 3632 3633 /* 3634 * Verify the sequence and ack numbers. Try getting the correct 3635 * response again. 3636 */ 3637 if ((th->th_ack != sc->sc_iss + 1) || 3638 SEQ_LEQ(th->th_seq, sc->sc_irs) || 3639 SEQ_GT(th->th_seq, sc->sc_irs + 1 + sc->sc_win)) { 3640 (void) syn_cache_respond(sc, m); 3641 splx(s); 3642 return ((struct socket *)(-1)); 3643 } 3644 3645 /* Remove this cache entry */ 3646 syn_cache_rm(sc); 3647 splx(s); 3648 3649 /* 3650 * Ok, create the full blown connection, and set things up 3651 * as they would have been set up if we had created the 3652 * connection when the SYN arrived. If we can't create 3653 * the connection, abort it. 3654 */ 3655 /* 3656 * inp still has the OLD in_pcb stuff, set the 3657 * v6-related flags on the new guy, too. This is 3658 * done particularly for the case where an AF_INET6 3659 * socket is bound only to a port, and a v4 connection 3660 * comes in on that port. 3661 * we also copy the flowinfo from the original pcb 3662 * to the new one. 3663 */ 3664 oso = so; 3665 so = sonewconn(so, SS_ISCONNECTED); 3666 if (so == NULL) 3667 goto resetandabort; 3668 3669 switch (so->so_proto->pr_domain->dom_family) { 3670 #ifdef INET 3671 case AF_INET: 3672 inp = sotoinpcb(so); 3673 break; 3674 #endif 3675 #ifdef INET6 3676 case AF_INET6: 3677 in6p = sotoin6pcb(so); 3678 break; 3679 #endif 3680 } 3681 switch (src->sa_family) { 3682 #ifdef INET 3683 case AF_INET: 3684 if (inp) { 3685 inp->inp_laddr = ((struct sockaddr_in *)dst)->sin_addr; 3686 inp->inp_lport = ((struct sockaddr_in *)dst)->sin_port; 3687 inp->inp_options = ip_srcroute(); 3688 in_pcbstate(inp, INP_BOUND); 3689 if (inp->inp_options == NULL) { 3690 inp->inp_options = sc->sc_ipopts; 3691 sc->sc_ipopts = NULL; 3692 } 3693 } 3694 #ifdef INET6 3695 else if (in6p) { 3696 /* IPv4 packet to AF_INET6 socket */ 3697 memset(&in6p->in6p_laddr, 0, sizeof(in6p->in6p_laddr)); 3698 in6p->in6p_laddr.s6_addr16[5] = htons(0xffff); 3699 bcopy(&((struct sockaddr_in *)dst)->sin_addr, 3700 &in6p->in6p_laddr.s6_addr32[3], 3701 sizeof(((struct sockaddr_in *)dst)->sin_addr)); 3702 in6p->in6p_lport = ((struct sockaddr_in *)dst)->sin_port; 3703 in6totcpcb(in6p)->t_family = AF_INET; 3704 if (sotoin6pcb(oso)->in6p_flags & IN6P_IPV6_V6ONLY) 3705 in6p->in6p_flags |= IN6P_IPV6_V6ONLY; 3706 else 3707 in6p->in6p_flags &= ~IN6P_IPV6_V6ONLY; 3708 in6_pcbstate(in6p, IN6P_BOUND); 3709 } 3710 #endif 3711 break; 3712 #endif 3713 #ifdef INET6 3714 case AF_INET6: 3715 if (in6p) { 3716 in6p->in6p_laddr = ((struct sockaddr_in6 *)dst)->sin6_addr; 3717 in6p->in6p_lport = ((struct sockaddr_in6 *)dst)->sin6_port; 3718 in6_pcbstate(in6p, IN6P_BOUND); 3719 } 3720 break; 3721 #endif 3722 } 3723 #ifdef INET6 3724 if (in6p && in6totcpcb(in6p)->t_family == AF_INET6 && sotoinpcb(oso)) { 3725 struct in6pcb *oin6p = sotoin6pcb(oso); 3726 /* inherit socket options from the listening socket */ 3727 in6p->in6p_flags |= (oin6p->in6p_flags & IN6P_CONTROLOPTS); 3728 if (in6p->in6p_flags & IN6P_CONTROLOPTS) { 3729 m_freem(in6p->in6p_options); 3730 in6p->in6p_options = 0; 3731 } 3732 ip6_savecontrol(in6p, &in6p->in6p_options, 3733 mtod(m, struct ip6_hdr *), m); 3734 } 3735 #endif 3736 3737 #if defined(IPSEC) || defined(FAST_IPSEC) 3738 /* 3739 * we make a copy of policy, instead of sharing the policy, 3740 * for better behavior in terms of SA lookup and dead SA removal. 3741 */ 3742 if (inp) { 3743 /* copy old policy into new socket's */ 3744 if (ipsec_copy_pcbpolicy(sotoinpcb(oso)->inp_sp, inp->inp_sp)) 3745 printf("tcp_input: could not copy policy\n"); 3746 } 3747 #ifdef INET6 3748 else if (in6p) { 3749 /* copy old policy into new socket's */ 3750 if (ipsec_copy_pcbpolicy(sotoin6pcb(oso)->in6p_sp, 3751 in6p->in6p_sp)) 3752 printf("tcp_input: could not copy policy\n"); 3753 } 3754 #endif 3755 #endif 3756 3757 /* 3758 * Give the new socket our cached route reference. 3759 */ 3760 if (inp) { 3761 rtcache_copy(&inp->inp_route, &sc->sc_route); 3762 rtcache_free(&sc->sc_route); 3763 } 3764 #ifdef INET6 3765 else { 3766 rtcache_copy(&in6p->in6p_route, &sc->sc_route); 3767 rtcache_free(&sc->sc_route); 3768 } 3769 #endif 3770 3771 am = m_get(M_DONTWAIT, MT_SONAME); /* XXX */ 3772 if (am == NULL) 3773 goto resetandabort; 3774 MCLAIM(am, &tcp_mowner); 3775 am->m_len = src->sa_len; 3776 bcopy(src, mtod(am, void *), src->sa_len); 3777 if (inp) { 3778 if (in_pcbconnect(inp, am, &lwp0)) { 3779 (void) m_free(am); 3780 goto resetandabort; 3781 } 3782 } 3783 #ifdef INET6 3784 else if (in6p) { 3785 if (src->sa_family == AF_INET) { 3786 /* IPv4 packet to AF_INET6 socket */ 3787 struct sockaddr_in6 *sin6; 3788 sin6 = mtod(am, struct sockaddr_in6 *); 3789 am->m_len = sizeof(*sin6); 3790 memset(sin6, 0, sizeof(*sin6)); 3791 sin6->sin6_family = AF_INET6; 3792 sin6->sin6_len = sizeof(*sin6); 3793 sin6->sin6_port = ((struct sockaddr_in *)src)->sin_port; 3794 sin6->sin6_addr.s6_addr16[5] = htons(0xffff); 3795 bcopy(&((struct sockaddr_in *)src)->sin_addr, 3796 &sin6->sin6_addr.s6_addr32[3], 3797 sizeof(sin6->sin6_addr.s6_addr32[3])); 3798 } 3799 if (in6_pcbconnect(in6p, am, NULL)) { 3800 (void) m_free(am); 3801 goto resetandabort; 3802 } 3803 } 3804 #endif 3805 else { 3806 (void) m_free(am); 3807 goto resetandabort; 3808 } 3809 (void) m_free(am); 3810 3811 if (inp) 3812 tp = intotcpcb(inp); 3813 #ifdef INET6 3814 else if (in6p) 3815 tp = in6totcpcb(in6p); 3816 #endif 3817 else 3818 tp = NULL; 3819 tp->t_flags = sototcpcb(oso)->t_flags & TF_NODELAY; 3820 if (sc->sc_request_r_scale != 15) { 3821 tp->requested_s_scale = sc->sc_requested_s_scale; 3822 tp->request_r_scale = sc->sc_request_r_scale; 3823 tp->snd_scale = sc->sc_requested_s_scale; 3824 tp->rcv_scale = sc->sc_request_r_scale; 3825 tp->t_flags |= TF_REQ_SCALE|TF_RCVD_SCALE; 3826 } 3827 if (sc->sc_flags & SCF_TIMESTAMP) 3828 tp->t_flags |= TF_REQ_TSTMP|TF_RCVD_TSTMP; 3829 tp->ts_timebase = sc->sc_timebase; 3830 3831 tp->t_template = tcp_template(tp); 3832 if (tp->t_template == 0) { 3833 tp = tcp_drop(tp, ENOBUFS); /* destroys socket */ 3834 so = NULL; 3835 m_freem(m); 3836 goto abort; 3837 } 3838 3839 tp->iss = sc->sc_iss; 3840 tp->irs = sc->sc_irs; 3841 tcp_sendseqinit(tp); 3842 tcp_rcvseqinit(tp); 3843 tp->t_state = TCPS_SYN_RECEIVED; 3844 TCP_TIMER_ARM(tp, TCPT_KEEP, tp->t_keepinit); 3845 TCP_STATINC(TCP_STAT_ACCEPTS); 3846 3847 if ((sc->sc_flags & SCF_SACK_PERMIT) && tcp_do_sack) 3848 tp->t_flags |= TF_WILL_SACK; 3849 3850 if ((sc->sc_flags & SCF_ECN_PERMIT) && tcp_do_ecn) 3851 tp->t_flags |= TF_ECN_PERMIT; 3852 3853 #ifdef TCP_SIGNATURE 3854 if (sc->sc_flags & SCF_SIGNATURE) 3855 tp->t_flags |= TF_SIGNATURE; 3856 #endif 3857 3858 /* Initialize tp->t_ourmss before we deal with the peer's! */ 3859 tp->t_ourmss = sc->sc_ourmaxseg; 3860 tcp_mss_from_peer(tp, sc->sc_peermaxseg); 3861 3862 /* 3863 * Initialize the initial congestion window. If we 3864 * had to retransmit the SYN,ACK, we must initialize cwnd 3865 * to 1 segment (i.e. the Loss Window). 3866 */ 3867 if (sc->sc_rxtshift) 3868 tp->snd_cwnd = tp->t_peermss; 3869 else { 3870 int ss = tcp_init_win; 3871 #ifdef INET 3872 if (inp != NULL && in_localaddr(inp->inp_faddr)) 3873 ss = tcp_init_win_local; 3874 #endif 3875 #ifdef INET6 3876 if (in6p != NULL && in6_localaddr(&in6p->in6p_faddr)) 3877 ss = tcp_init_win_local; 3878 #endif 3879 tp->snd_cwnd = TCP_INITIAL_WINDOW(ss, tp->t_peermss); 3880 } 3881 3882 tcp_rmx_rtt(tp); 3883 tp->snd_wl1 = sc->sc_irs; 3884 tp->rcv_up = sc->sc_irs + 1; 3885 3886 /* 3887 * This is what whould have happened in tcp_output() when 3888 * the SYN,ACK was sent. 3889 */ 3890 tp->snd_up = tp->snd_una; 3891 tp->snd_max = tp->snd_nxt = tp->iss+1; 3892 TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur); 3893 if (sc->sc_win > 0 && SEQ_GT(tp->rcv_nxt + sc->sc_win, tp->rcv_adv)) 3894 tp->rcv_adv = tp->rcv_nxt + sc->sc_win; 3895 tp->last_ack_sent = tp->rcv_nxt; 3896 tp->t_partialacks = -1; 3897 tp->t_dupacks = 0; 3898 3899 TCP_STATINC(TCP_STAT_SC_COMPLETED); 3900 s = splsoftnet(); 3901 syn_cache_put(sc); 3902 splx(s); 3903 return (so); 3904 3905 resetandabort: 3906 (void)tcp_respond(NULL, m, m, th, (tcp_seq)0, th->th_ack, TH_RST); 3907 abort: 3908 if (so != NULL) { 3909 (void) soqremque(so, 1); 3910 (void) soabort(so); 3911 mutex_enter(softnet_lock); 3912 } 3913 s = splsoftnet(); 3914 syn_cache_put(sc); 3915 splx(s); 3916 TCP_STATINC(TCP_STAT_SC_ABORTED); 3917 return ((struct socket *)(-1)); 3918 } 3919 3920 /* 3921 * This function is called when we get a RST for a 3922 * non-existent connection, so that we can see if the 3923 * connection is in the syn cache. If it is, zap it. 3924 */ 3925 3926 void 3927 syn_cache_reset(struct sockaddr *src, struct sockaddr *dst, struct tcphdr *th) 3928 { 3929 struct syn_cache *sc; 3930 struct syn_cache_head *scp; 3931 int s = splsoftnet(); 3932 3933 if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) { 3934 splx(s); 3935 return; 3936 } 3937 if (SEQ_LT(th->th_seq, sc->sc_irs) || 3938 SEQ_GT(th->th_seq, sc->sc_irs+1)) { 3939 splx(s); 3940 return; 3941 } 3942 syn_cache_rm(sc); 3943 TCP_STATINC(TCP_STAT_SC_RESET); 3944 syn_cache_put(sc); /* calls pool_put but see spl above */ 3945 splx(s); 3946 } 3947 3948 void 3949 syn_cache_unreach(const struct sockaddr *src, const struct sockaddr *dst, 3950 struct tcphdr *th) 3951 { 3952 struct syn_cache *sc; 3953 struct syn_cache_head *scp; 3954 int s; 3955 3956 s = splsoftnet(); 3957 if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) { 3958 splx(s); 3959 return; 3960 } 3961 /* If the sequence number != sc_iss, then it's a bogus ICMP msg */ 3962 if (ntohl (th->th_seq) != sc->sc_iss) { 3963 splx(s); 3964 return; 3965 } 3966 3967 /* 3968 * If we've retransmitted 3 times and this is our second error, 3969 * we remove the entry. Otherwise, we allow it to continue on. 3970 * This prevents us from incorrectly nuking an entry during a 3971 * spurious network outage. 3972 * 3973 * See tcp_notify(). 3974 */ 3975 if ((sc->sc_flags & SCF_UNREACH) == 0 || sc->sc_rxtshift < 3) { 3976 sc->sc_flags |= SCF_UNREACH; 3977 splx(s); 3978 return; 3979 } 3980 3981 syn_cache_rm(sc); 3982 TCP_STATINC(TCP_STAT_SC_UNREACH); 3983 syn_cache_put(sc); /* calls pool_put but see spl above */ 3984 splx(s); 3985 } 3986 3987 /* 3988 * Given a LISTEN socket and an inbound SYN request, add 3989 * this to the syn cache, and send back a segment: 3990 * <SEQ=ISS><ACK=RCV_NXT><CTL=SYN,ACK> 3991 * to the source. 3992 * 3993 * IMPORTANT NOTE: We do _NOT_ ACK data that might accompany the SYN. 3994 * Doing so would require that we hold onto the data and deliver it 3995 * to the application. However, if we are the target of a SYN-flood 3996 * DoS attack, an attacker could send data which would eventually 3997 * consume all available buffer space if it were ACKed. By not ACKing 3998 * the data, we avoid this DoS scenario. 3999 */ 4000 4001 int 4002 syn_cache_add(struct sockaddr *src, struct sockaddr *dst, struct tcphdr *th, 4003 unsigned int hlen, struct socket *so, struct mbuf *m, u_char *optp, 4004 int optlen, struct tcp_opt_info *oi) 4005 { 4006 struct tcpcb tb, *tp; 4007 long win; 4008 struct syn_cache *sc; 4009 struct syn_cache_head *scp; 4010 struct mbuf *ipopts; 4011 struct tcp_opt_info opti; 4012 int s; 4013 4014 tp = sototcpcb(so); 4015 4016 memset(&opti, 0, sizeof(opti)); 4017 4018 /* 4019 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN 4020 * 4021 * Note this check is performed in tcp_input() very early on. 4022 */ 4023 4024 /* 4025 * Initialize some local state. 4026 */ 4027 win = sbspace(&so->so_rcv); 4028 if (win > TCP_MAXWIN) 4029 win = TCP_MAXWIN; 4030 4031 switch (src->sa_family) { 4032 #ifdef INET 4033 case AF_INET: 4034 /* 4035 * Remember the IP options, if any. 4036 */ 4037 ipopts = ip_srcroute(); 4038 break; 4039 #endif 4040 default: 4041 ipopts = NULL; 4042 } 4043 4044 #ifdef TCP_SIGNATURE 4045 if (optp || (tp->t_flags & TF_SIGNATURE)) 4046 #else 4047 if (optp) 4048 #endif 4049 { 4050 tb.t_flags = tcp_do_rfc1323 ? (TF_REQ_SCALE|TF_REQ_TSTMP) : 0; 4051 #ifdef TCP_SIGNATURE 4052 tb.t_flags |= (tp->t_flags & TF_SIGNATURE); 4053 #endif 4054 tb.t_state = TCPS_LISTEN; 4055 if (tcp_dooptions(&tb, optp, optlen, th, m, m->m_pkthdr.len - 4056 sizeof(struct tcphdr) - optlen - hlen, oi) < 0) 4057 return (0); 4058 } else 4059 tb.t_flags = 0; 4060 4061 /* 4062 * See if we already have an entry for this connection. 4063 * If we do, resend the SYN,ACK. We do not count this 4064 * as a retransmission (XXX though maybe we should). 4065 */ 4066 if ((sc = syn_cache_lookup(src, dst, &scp)) != NULL) { 4067 TCP_STATINC(TCP_STAT_SC_DUPESYN); 4068 if (ipopts) { 4069 /* 4070 * If we were remembering a previous source route, 4071 * forget it and use the new one we've been given. 4072 */ 4073 if (sc->sc_ipopts) 4074 (void) m_free(sc->sc_ipopts); 4075 sc->sc_ipopts = ipopts; 4076 } 4077 sc->sc_timestamp = tb.ts_recent; 4078 if (syn_cache_respond(sc, m) == 0) { 4079 uint64_t *tcps = TCP_STAT_GETREF(); 4080 tcps[TCP_STAT_SNDACKS]++; 4081 tcps[TCP_STAT_SNDTOTAL]++; 4082 TCP_STAT_PUTREF(); 4083 } 4084 return (1); 4085 } 4086 4087 s = splsoftnet(); 4088 sc = pool_get(&syn_cache_pool, PR_NOWAIT); 4089 splx(s); 4090 if (sc == NULL) { 4091 if (ipopts) 4092 (void) m_free(ipopts); 4093 return (0); 4094 } 4095 4096 /* 4097 * Fill in the cache, and put the necessary IP and TCP 4098 * options into the reply. 4099 */ 4100 memset(sc, 0, sizeof(struct syn_cache)); 4101 callout_init(&sc->sc_timer, CALLOUT_MPSAFE); 4102 bcopy(src, &sc->sc_src, src->sa_len); 4103 bcopy(dst, &sc->sc_dst, dst->sa_len); 4104 sc->sc_flags = 0; 4105 sc->sc_ipopts = ipopts; 4106 sc->sc_irs = th->th_seq; 4107 switch (src->sa_family) { 4108 #ifdef INET 4109 case AF_INET: 4110 { 4111 struct sockaddr_in *srcin = (void *) src; 4112 struct sockaddr_in *dstin = (void *) dst; 4113 4114 sc->sc_iss = tcp_new_iss1(&dstin->sin_addr, 4115 &srcin->sin_addr, dstin->sin_port, 4116 srcin->sin_port, sizeof(dstin->sin_addr), 0); 4117 break; 4118 } 4119 #endif /* INET */ 4120 #ifdef INET6 4121 case AF_INET6: 4122 { 4123 struct sockaddr_in6 *srcin6 = (void *) src; 4124 struct sockaddr_in6 *dstin6 = (void *) dst; 4125 4126 sc->sc_iss = tcp_new_iss1(&dstin6->sin6_addr, 4127 &srcin6->sin6_addr, dstin6->sin6_port, 4128 srcin6->sin6_port, sizeof(dstin6->sin6_addr), 0); 4129 break; 4130 } 4131 #endif /* INET6 */ 4132 } 4133 sc->sc_peermaxseg = oi->maxseg; 4134 sc->sc_ourmaxseg = tcp_mss_to_advertise(m->m_flags & M_PKTHDR ? 4135 m->m_pkthdr.rcvif : NULL, 4136 sc->sc_src.sa.sa_family); 4137 sc->sc_win = win; 4138 sc->sc_timebase = tcp_now - 1; /* see tcp_newtcpcb() */ 4139 sc->sc_timestamp = tb.ts_recent; 4140 if ((tb.t_flags & (TF_REQ_TSTMP|TF_RCVD_TSTMP)) == 4141 (TF_REQ_TSTMP|TF_RCVD_TSTMP)) 4142 sc->sc_flags |= SCF_TIMESTAMP; 4143 if ((tb.t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) == 4144 (TF_RCVD_SCALE|TF_REQ_SCALE)) { 4145 sc->sc_requested_s_scale = tb.requested_s_scale; 4146 sc->sc_request_r_scale = 0; 4147 /* 4148 * Pick the smallest possible scaling factor that 4149 * will still allow us to scale up to sb_max. 4150 * 4151 * We do this because there are broken firewalls that 4152 * will corrupt the window scale option, leading to 4153 * the other endpoint believing that our advertised 4154 * window is unscaled. At scale factors larger than 4155 * 5 the unscaled window will drop below 1500 bytes, 4156 * leading to serious problems when traversing these 4157 * broken firewalls. 4158 * 4159 * With the default sbmax of 256K, a scale factor 4160 * of 3 will be chosen by this algorithm. Those who 4161 * choose a larger sbmax should watch out 4162 * for the compatiblity problems mentioned above. 4163 * 4164 * RFC1323: The Window field in a SYN (i.e., a <SYN> 4165 * or <SYN,ACK>) segment itself is never scaled. 4166 */ 4167 while (sc->sc_request_r_scale < TCP_MAX_WINSHIFT && 4168 (TCP_MAXWIN << sc->sc_request_r_scale) < sb_max) 4169 sc->sc_request_r_scale++; 4170 } else { 4171 sc->sc_requested_s_scale = 15; 4172 sc->sc_request_r_scale = 15; 4173 } 4174 if ((tb.t_flags & TF_SACK_PERMIT) && tcp_do_sack) 4175 sc->sc_flags |= SCF_SACK_PERMIT; 4176 4177 /* 4178 * ECN setup packet recieved. 4179 */ 4180 if ((th->th_flags & (TH_ECE|TH_CWR)) && tcp_do_ecn) 4181 sc->sc_flags |= SCF_ECN_PERMIT; 4182 4183 #ifdef TCP_SIGNATURE 4184 if (tb.t_flags & TF_SIGNATURE) 4185 sc->sc_flags |= SCF_SIGNATURE; 4186 #endif 4187 sc->sc_tp = tp; 4188 if (syn_cache_respond(sc, m) == 0) { 4189 uint64_t *tcps = TCP_STAT_GETREF(); 4190 tcps[TCP_STAT_SNDACKS]++; 4191 tcps[TCP_STAT_SNDTOTAL]++; 4192 TCP_STAT_PUTREF(); 4193 syn_cache_insert(sc, tp); 4194 } else { 4195 s = splsoftnet(); 4196 syn_cache_put(sc); 4197 splx(s); 4198 TCP_STATINC(TCP_STAT_SC_DROPPED); 4199 } 4200 return (1); 4201 } 4202 4203 int 4204 syn_cache_respond(struct syn_cache *sc, struct mbuf *m) 4205 { 4206 #ifdef INET6 4207 struct rtentry *rt; 4208 #endif 4209 struct route *ro; 4210 u_int8_t *optp; 4211 int optlen, error; 4212 u_int16_t tlen; 4213 struct ip *ip = NULL; 4214 #ifdef INET6 4215 struct ip6_hdr *ip6 = NULL; 4216 #endif 4217 struct tcpcb *tp = NULL; 4218 struct tcphdr *th; 4219 u_int hlen; 4220 struct socket *so; 4221 4222 ro = &sc->sc_route; 4223 switch (sc->sc_src.sa.sa_family) { 4224 case AF_INET: 4225 hlen = sizeof(struct ip); 4226 break; 4227 #ifdef INET6 4228 case AF_INET6: 4229 hlen = sizeof(struct ip6_hdr); 4230 break; 4231 #endif 4232 default: 4233 if (m) 4234 m_freem(m); 4235 return (EAFNOSUPPORT); 4236 } 4237 4238 /* Compute the size of the TCP options. */ 4239 optlen = 4 + (sc->sc_request_r_scale != 15 ? 4 : 0) + 4240 ((sc->sc_flags & SCF_SACK_PERMIT) ? (TCPOLEN_SACK_PERMITTED + 2) : 0) + 4241 #ifdef TCP_SIGNATURE 4242 ((sc->sc_flags & SCF_SIGNATURE) ? (TCPOLEN_SIGNATURE + 2) : 0) + 4243 #endif 4244 ((sc->sc_flags & SCF_TIMESTAMP) ? TCPOLEN_TSTAMP_APPA : 0); 4245 4246 tlen = hlen + sizeof(struct tcphdr) + optlen; 4247 4248 /* 4249 * Create the IP+TCP header from scratch. 4250 */ 4251 if (m) 4252 m_freem(m); 4253 #ifdef DIAGNOSTIC 4254 if (max_linkhdr + tlen > MCLBYTES) 4255 return (ENOBUFS); 4256 #endif 4257 MGETHDR(m, M_DONTWAIT, MT_DATA); 4258 if (m && tlen > MHLEN) { 4259 MCLGET(m, M_DONTWAIT); 4260 if ((m->m_flags & M_EXT) == 0) { 4261 m_freem(m); 4262 m = NULL; 4263 } 4264 } 4265 if (m == NULL) 4266 return (ENOBUFS); 4267 MCLAIM(m, &tcp_tx_mowner); 4268 4269 /* Fixup the mbuf. */ 4270 m->m_data += max_linkhdr; 4271 m->m_len = m->m_pkthdr.len = tlen; 4272 if (sc->sc_tp) { 4273 tp = sc->sc_tp; 4274 if (tp->t_inpcb) 4275 so = tp->t_inpcb->inp_socket; 4276 #ifdef INET6 4277 else if (tp->t_in6pcb) 4278 so = tp->t_in6pcb->in6p_socket; 4279 #endif 4280 else 4281 so = NULL; 4282 } else 4283 so = NULL; 4284 m->m_pkthdr.rcvif = NULL; 4285 memset(mtod(m, u_char *), 0, tlen); 4286 4287 switch (sc->sc_src.sa.sa_family) { 4288 case AF_INET: 4289 ip = mtod(m, struct ip *); 4290 ip->ip_v = 4; 4291 ip->ip_dst = sc->sc_src.sin.sin_addr; 4292 ip->ip_src = sc->sc_dst.sin.sin_addr; 4293 ip->ip_p = IPPROTO_TCP; 4294 th = (struct tcphdr *)(ip + 1); 4295 th->th_dport = sc->sc_src.sin.sin_port; 4296 th->th_sport = sc->sc_dst.sin.sin_port; 4297 break; 4298 #ifdef INET6 4299 case AF_INET6: 4300 ip6 = mtod(m, struct ip6_hdr *); 4301 ip6->ip6_vfc = IPV6_VERSION; 4302 ip6->ip6_dst = sc->sc_src.sin6.sin6_addr; 4303 ip6->ip6_src = sc->sc_dst.sin6.sin6_addr; 4304 ip6->ip6_nxt = IPPROTO_TCP; 4305 /* ip6_plen will be updated in ip6_output() */ 4306 th = (struct tcphdr *)(ip6 + 1); 4307 th->th_dport = sc->sc_src.sin6.sin6_port; 4308 th->th_sport = sc->sc_dst.sin6.sin6_port; 4309 break; 4310 #endif 4311 default: 4312 th = NULL; 4313 } 4314 4315 th->th_seq = htonl(sc->sc_iss); 4316 th->th_ack = htonl(sc->sc_irs + 1); 4317 th->th_off = (sizeof(struct tcphdr) + optlen) >> 2; 4318 th->th_flags = TH_SYN|TH_ACK; 4319 th->th_win = htons(sc->sc_win); 4320 /* th_sum already 0 */ 4321 /* th_urp already 0 */ 4322 4323 /* Tack on the TCP options. */ 4324 optp = (u_int8_t *)(th + 1); 4325 *optp++ = TCPOPT_MAXSEG; 4326 *optp++ = 4; 4327 *optp++ = (sc->sc_ourmaxseg >> 8) & 0xff; 4328 *optp++ = sc->sc_ourmaxseg & 0xff; 4329 4330 if (sc->sc_request_r_scale != 15) { 4331 *((u_int32_t *)optp) = htonl(TCPOPT_NOP << 24 | 4332 TCPOPT_WINDOW << 16 | TCPOLEN_WINDOW << 8 | 4333 sc->sc_request_r_scale); 4334 optp += 4; 4335 } 4336 4337 if (sc->sc_flags & SCF_TIMESTAMP) { 4338 u_int32_t *lp = (u_int32_t *)(optp); 4339 /* Form timestamp option as shown in appendix A of RFC 1323. */ 4340 *lp++ = htonl(TCPOPT_TSTAMP_HDR); 4341 *lp++ = htonl(SYN_CACHE_TIMESTAMP(sc)); 4342 *lp = htonl(sc->sc_timestamp); 4343 optp += TCPOLEN_TSTAMP_APPA; 4344 } 4345 4346 if (sc->sc_flags & SCF_SACK_PERMIT) { 4347 u_int8_t *p = optp; 4348 4349 /* Let the peer know that we will SACK. */ 4350 p[0] = TCPOPT_SACK_PERMITTED; 4351 p[1] = 2; 4352 p[2] = TCPOPT_NOP; 4353 p[3] = TCPOPT_NOP; 4354 optp += 4; 4355 } 4356 4357 /* 4358 * Send ECN SYN-ACK setup packet. 4359 * Routes can be asymetric, so, even if we receive a packet 4360 * with ECE and CWR set, we must not assume no one will block 4361 * the ECE packet we are about to send. 4362 */ 4363 if ((sc->sc_flags & SCF_ECN_PERMIT) && tp && 4364 SEQ_GEQ(tp->snd_nxt, tp->snd_max)) { 4365 th->th_flags |= TH_ECE; 4366 TCP_STATINC(TCP_STAT_ECN_SHS); 4367 4368 /* 4369 * draft-ietf-tcpm-ecnsyn-00.txt 4370 * 4371 * "[...] a TCP node MAY respond to an ECN-setup 4372 * SYN packet by setting ECT in the responding 4373 * ECN-setup SYN/ACK packet, indicating to routers 4374 * that the SYN/ACK packet is ECN-Capable. 4375 * This allows a congested router along the path 4376 * to mark the packet instead of dropping the 4377 * packet as an indication of congestion." 4378 * 4379 * "[...] There can be a great benefit in setting 4380 * an ECN-capable codepoint in SYN/ACK packets [...] 4381 * Congestion is most likely to occur in 4382 * the server-to-client direction. As a result, 4383 * setting an ECN-capable codepoint in SYN/ACK 4384 * packets can reduce the occurence of three-second 4385 * retransmit timeouts resulting from the drop 4386 * of SYN/ACK packets." 4387 * 4388 * Page 4 and 6, January 2006. 4389 */ 4390 4391 switch (sc->sc_src.sa.sa_family) { 4392 #ifdef INET 4393 case AF_INET: 4394 ip->ip_tos |= IPTOS_ECN_ECT0; 4395 break; 4396 #endif 4397 #ifdef INET6 4398 case AF_INET6: 4399 ip6->ip6_flow |= htonl(IPTOS_ECN_ECT0 << 20); 4400 break; 4401 #endif 4402 } 4403 TCP_STATINC(TCP_STAT_ECN_ECT); 4404 } 4405 4406 #ifdef TCP_SIGNATURE 4407 if (sc->sc_flags & SCF_SIGNATURE) { 4408 struct secasvar *sav; 4409 u_int8_t *sigp; 4410 4411 sav = tcp_signature_getsav(m, th); 4412 4413 if (sav == NULL) { 4414 if (m) 4415 m_freem(m); 4416 return (EPERM); 4417 } 4418 4419 *optp++ = TCPOPT_SIGNATURE; 4420 *optp++ = TCPOLEN_SIGNATURE; 4421 sigp = optp; 4422 memset(optp, 0, TCP_SIGLEN); 4423 optp += TCP_SIGLEN; 4424 *optp++ = TCPOPT_NOP; 4425 *optp++ = TCPOPT_EOL; 4426 4427 (void)tcp_signature(m, th, hlen, sav, sigp); 4428 4429 key_sa_recordxfer(sav, m); 4430 #ifdef FAST_IPSEC 4431 KEY_FREESAV(&sav); 4432 #else 4433 key_freesav(sav); 4434 #endif 4435 } 4436 #endif 4437 4438 /* Compute the packet's checksum. */ 4439 switch (sc->sc_src.sa.sa_family) { 4440 case AF_INET: 4441 ip->ip_len = htons(tlen - hlen); 4442 th->th_sum = 0; 4443 th->th_sum = in4_cksum(m, IPPROTO_TCP, hlen, tlen - hlen); 4444 break; 4445 #ifdef INET6 4446 case AF_INET6: 4447 ip6->ip6_plen = htons(tlen - hlen); 4448 th->th_sum = 0; 4449 th->th_sum = in6_cksum(m, IPPROTO_TCP, hlen, tlen - hlen); 4450 break; 4451 #endif 4452 } 4453 4454 /* 4455 * Fill in some straggling IP bits. Note the stack expects 4456 * ip_len to be in host order, for convenience. 4457 */ 4458 switch (sc->sc_src.sa.sa_family) { 4459 #ifdef INET 4460 case AF_INET: 4461 ip->ip_len = htons(tlen); 4462 ip->ip_ttl = ip_defttl; 4463 /* XXX tos? */ 4464 break; 4465 #endif 4466 #ifdef INET6 4467 case AF_INET6: 4468 ip6->ip6_vfc &= ~IPV6_VERSION_MASK; 4469 ip6->ip6_vfc |= IPV6_VERSION; 4470 ip6->ip6_plen = htons(tlen - hlen); 4471 /* ip6_hlim will be initialized afterwards */ 4472 /* XXX flowlabel? */ 4473 break; 4474 #endif 4475 } 4476 4477 /* XXX use IPsec policy on listening socket, on SYN ACK */ 4478 tp = sc->sc_tp; 4479 4480 switch (sc->sc_src.sa.sa_family) { 4481 #ifdef INET 4482 case AF_INET: 4483 error = ip_output(m, sc->sc_ipopts, ro, 4484 (ip_mtudisc ? IP_MTUDISC : 0), 4485 (struct ip_moptions *)NULL, so); 4486 break; 4487 #endif 4488 #ifdef INET6 4489 case AF_INET6: 4490 ip6->ip6_hlim = in6_selecthlim(NULL, 4491 (rt = rtcache_validate(ro)) != NULL ? rt->rt_ifp 4492 : NULL); 4493 4494 error = ip6_output(m, NULL /*XXX*/, ro, 0, NULL, so, NULL); 4495 break; 4496 #endif 4497 default: 4498 error = EAFNOSUPPORT; 4499 break; 4500 } 4501 return (error); 4502 } 4503
Indexes created Mon Sep 22 05:09:51 GMT 2025