xref: /netbsd-src/sys/netinet/tcp_input.c (revision cef8759bd76c1b621f8eab8faa6f208faabc2e15)
1 /*	$NetBSD: tcp_input.c,v 1.418 2020/07/06 18:49:12 christos 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,
74  * 2011 The NetBSD Foundation, Inc.
75  * All rights reserved.
76  *
77  * This code is derived from software contributed to The NetBSD Foundation
78  * by Coyote Point Systems, Inc.
79  * This code is derived from software contributed to The NetBSD Foundation
80  * by Jason R. Thorpe and Kevin M. Lahey of the Numerical Aerospace Simulation
81  * Facility, NASA Ames Research Center.
82  * This code is derived from software contributed to The NetBSD Foundation
83  * by Charles M. Hannum.
84  * This code is derived from software contributed to The NetBSD Foundation
85  * by Rui Paulo.
86  *
87  * Redistribution and use in source and binary forms, with or without
88  * modification, are permitted provided that the following conditions
89  * are met:
90  * 1. Redistributions of source code must retain the above copyright
91  *    notice, this list of conditions and the following disclaimer.
92  * 2. Redistributions in binary form must reproduce the above copyright
93  *    notice, this list of conditions and the following disclaimer in the
94  *    documentation and/or other materials provided with the distribution.
95  *
96  * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
97  * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
98  * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
99  * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
100  * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
101  * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
102  * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
103  * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
104  * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
105  * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
106  * POSSIBILITY OF SUCH DAMAGE.
107  */
108 
109 /*
110  * Copyright (c) 1982, 1986, 1988, 1990, 1993, 1994, 1995
111  *	The Regents of the University of California.  All rights reserved.
112  *
113  * Redistribution and use in source and binary forms, with or without
114  * modification, are permitted provided that the following conditions
115  * are met:
116  * 1. Redistributions of source code must retain the above copyright
117  *    notice, this list of conditions and the following disclaimer.
118  * 2. Redistributions in binary form must reproduce the above copyright
119  *    notice, this list of conditions and the following disclaimer in the
120  *    documentation and/or other materials provided with the distribution.
121  * 3. Neither the name of the University nor the names of its contributors
122  *    may be used to endorse or promote products derived from this software
123  *    without specific prior written permission.
124  *
125  * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
126  * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
127  * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
128  * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
129  * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
130  * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
131  * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
132  * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
133  * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
134  * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
135  * SUCH DAMAGE.
136  *
137  *	@(#)tcp_input.c	8.12 (Berkeley) 5/24/95
138  */
139 
140 /*
141  *	TODO list for SYN cache stuff:
142  *
143  *	Find room for a "state" field, which is needed to keep a
144  *	compressed state for TIME_WAIT TCBs.  It's been noted already
145  *	that this is fairly important for very high-volume web and
146  *	mail servers, which use a large number of short-lived
147  *	connections.
148  */
149 
150 #include <sys/cdefs.h>
151 __KERNEL_RCSID(0, "$NetBSD: tcp_input.c,v 1.418 2020/07/06 18:49:12 christos Exp $");
152 
153 #ifdef _KERNEL_OPT
154 #include "opt_inet.h"
155 #include "opt_ipsec.h"
156 #include "opt_inet_csum.h"
157 #include "opt_tcp_debug.h"
158 #endif
159 
160 #include <sys/param.h>
161 #include <sys/systm.h>
162 #include <sys/malloc.h>
163 #include <sys/mbuf.h>
164 #include <sys/protosw.h>
165 #include <sys/socket.h>
166 #include <sys/socketvar.h>
167 #include <sys/errno.h>
168 #include <sys/syslog.h>
169 #include <sys/pool.h>
170 #include <sys/domain.h>
171 #include <sys/kernel.h>
172 #ifdef TCP_SIGNATURE
173 #include <sys/md5.h>
174 #endif
175 #include <sys/lwp.h> /* for lwp0 */
176 #include <sys/cprng.h>
177 
178 #include <net/if.h>
179 #include <net/if_types.h>
180 
181 #include <netinet/in.h>
182 #include <netinet/in_systm.h>
183 #include <netinet/ip.h>
184 #include <netinet/in_pcb.h>
185 #include <netinet/in_var.h>
186 #include <netinet/ip_var.h>
187 #include <netinet/in_offload.h>
188 
189 #ifdef INET6
190 #include <netinet/ip6.h>
191 #include <netinet6/ip6_var.h>
192 #include <netinet6/in6_pcb.h>
193 #include <netinet6/ip6_var.h>
194 #include <netinet6/in6_var.h>
195 #include <netinet/icmp6.h>
196 #include <netinet6/nd6.h>
197 #ifdef TCP_SIGNATURE
198 #include <netinet6/scope6_var.h>
199 #endif
200 #endif
201 
202 #ifndef INET6
203 #include <netinet/ip6.h>
204 #endif
205 
206 #include <netinet/tcp.h>
207 #include <netinet/tcp_fsm.h>
208 #include <netinet/tcp_seq.h>
209 #include <netinet/tcp_timer.h>
210 #include <netinet/tcp_var.h>
211 #include <netinet/tcp_private.h>
212 #include <netinet/tcp_congctl.h>
213 #include <netinet/tcp_debug.h>
214 
215 #ifdef INET6
216 #include "faith.h"
217 #if defined(NFAITH) && NFAITH > 0
218 #include <net/if_faith.h>
219 #endif
220 #endif
221 
222 #ifdef IPSEC
223 #include <netipsec/ipsec.h>
224 #include <netipsec/key.h>
225 #ifdef INET6
226 #include <netipsec/ipsec6.h>
227 #endif
228 #endif	/* IPSEC*/
229 
230 #include <netinet/tcp_vtw.h>
231 
232 int	tcprexmtthresh = 3;
233 int	tcp_log_refused;
234 
235 int	tcp_do_autorcvbuf = 1;
236 int	tcp_autorcvbuf_inc = 16 * 1024;
237 int	tcp_autorcvbuf_max = 256 * 1024;
238 int	tcp_msl = (TCPTV_MSL / PR_SLOWHZ);
239 
240 static int tcp_rst_ppslim_count = 0;
241 static struct timeval tcp_rst_ppslim_last;
242 static int tcp_ackdrop_ppslim_count = 0;
243 static struct timeval tcp_ackdrop_ppslim_last;
244 
245 static void syn_cache_timer(void *);
246 
247 #define TCP_PAWS_IDLE	(24U * 24 * 60 * 60 * PR_SLOWHZ)
248 
249 /* for modulo comparisons of timestamps */
250 #define TSTMP_LT(a,b)	((int)((a)-(b)) < 0)
251 #define TSTMP_GEQ(a,b)	((int)((a)-(b)) >= 0)
252 
253 /*
254  * Neighbor Discovery, Neighbor Unreachability Detection Upper layer hint.
255  */
256 #ifdef INET6
257 static inline void
258 nd6_hint(struct tcpcb *tp)
259 {
260 	struct rtentry *rt = NULL;
261 
262 	if (tp != NULL && tp->t_in6pcb != NULL && tp->t_family == AF_INET6 &&
263 	    (rt = rtcache_validate(&tp->t_in6pcb->in6p_route)) != NULL) {
264 		nd6_nud_hint(rt);
265 		rtcache_unref(rt, &tp->t_in6pcb->in6p_route);
266 	}
267 }
268 #else
269 static inline void
270 nd6_hint(struct tcpcb *tp)
271 {
272 }
273 #endif
274 
275 /*
276  * Compute ACK transmission behavior.  Delay the ACK unless
277  * we have already delayed an ACK (must send an ACK every two segments).
278  * We also ACK immediately if we received a PUSH and the ACK-on-PUSH
279  * option is enabled.
280  */
281 static void
282 tcp_setup_ack(struct tcpcb *tp, const struct tcphdr *th)
283 {
284 
285 	if (tp->t_flags & TF_DELACK ||
286 	    (tcp_ack_on_push && th->th_flags & TH_PUSH))
287 		tp->t_flags |= TF_ACKNOW;
288 	else
289 		TCP_SET_DELACK(tp);
290 }
291 
292 static void
293 icmp_check(struct tcpcb *tp, const struct tcphdr *th, int acked)
294 {
295 
296 	/*
297 	 * If we had a pending ICMP message that refers to data that have
298 	 * just been acknowledged, disregard the recorded ICMP message.
299 	 */
300 	if ((tp->t_flags & TF_PMTUD_PEND) &&
301 	    SEQ_GT(th->th_ack, tp->t_pmtud_th_seq))
302 		tp->t_flags &= ~TF_PMTUD_PEND;
303 
304 	/*
305 	 * Keep track of the largest chunk of data
306 	 * acknowledged since last PMTU update
307 	 */
308 	if (tp->t_pmtud_mss_acked < acked)
309 		tp->t_pmtud_mss_acked = acked;
310 }
311 
312 /*
313  * Convert TCP protocol fields to host order for easier processing.
314  */
315 static void
316 tcp_fields_to_host(struct tcphdr *th)
317 {
318 
319 	NTOHL(th->th_seq);
320 	NTOHL(th->th_ack);
321 	NTOHS(th->th_win);
322 	NTOHS(th->th_urp);
323 }
324 
325 /*
326  * ... and reverse the above.
327  */
328 static void
329 tcp_fields_to_net(struct tcphdr *th)
330 {
331 
332 	HTONL(th->th_seq);
333 	HTONL(th->th_ack);
334 	HTONS(th->th_win);
335 	HTONS(th->th_urp);
336 }
337 
338 static void
339 tcp_urp_drop(struct tcphdr *th, int todrop, int *tiflags)
340 {
341 	if (th->th_urp > todrop) {
342 		th->th_urp -= todrop;
343 	} else {
344 		*tiflags &= ~TH_URG;
345 		th->th_urp = 0;
346 	}
347 }
348 
349 #ifdef TCP_CSUM_COUNTERS
350 #include <sys/device.h>
351 
352 extern struct evcnt tcp_hwcsum_ok;
353 extern struct evcnt tcp_hwcsum_bad;
354 extern struct evcnt tcp_hwcsum_data;
355 extern struct evcnt tcp_swcsum;
356 #if defined(INET6)
357 extern struct evcnt tcp6_hwcsum_ok;
358 extern struct evcnt tcp6_hwcsum_bad;
359 extern struct evcnt tcp6_hwcsum_data;
360 extern struct evcnt tcp6_swcsum;
361 #endif /* defined(INET6) */
362 
363 #define	TCP_CSUM_COUNTER_INCR(ev)	(ev)->ev_count++
364 
365 #else
366 
367 #define	TCP_CSUM_COUNTER_INCR(ev)	/* nothing */
368 
369 #endif /* TCP_CSUM_COUNTERS */
370 
371 #ifdef TCP_REASS_COUNTERS
372 #include <sys/device.h>
373 
374 extern struct evcnt tcp_reass_;
375 extern struct evcnt tcp_reass_empty;
376 extern struct evcnt tcp_reass_iteration[8];
377 extern struct evcnt tcp_reass_prependfirst;
378 extern struct evcnt tcp_reass_prepend;
379 extern struct evcnt tcp_reass_insert;
380 extern struct evcnt tcp_reass_inserttail;
381 extern struct evcnt tcp_reass_append;
382 extern struct evcnt tcp_reass_appendtail;
383 extern struct evcnt tcp_reass_overlaptail;
384 extern struct evcnt tcp_reass_overlapfront;
385 extern struct evcnt tcp_reass_segdup;
386 extern struct evcnt tcp_reass_fragdup;
387 
388 #define	TCP_REASS_COUNTER_INCR(ev)	(ev)->ev_count++
389 
390 #else
391 
392 #define	TCP_REASS_COUNTER_INCR(ev)	/* nothing */
393 
394 #endif /* TCP_REASS_COUNTERS */
395 
396 static int tcp_reass(struct tcpcb *, const struct tcphdr *, struct mbuf *,
397     int);
398 static int tcp_dooptions(struct tcpcb *, const u_char *, int,
399     struct tcphdr *, struct mbuf *, int, struct tcp_opt_info *);
400 
401 static void tcp4_log_refused(const struct ip *, const struct tcphdr *);
402 #ifdef INET6
403 static void tcp6_log_refused(const struct ip6_hdr *, const struct tcphdr *);
404 #endif
405 
406 #if defined(MBUFTRACE)
407 struct mowner tcp_reass_mowner = MOWNER_INIT("tcp", "reass");
408 #endif /* defined(MBUFTRACE) */
409 
410 static struct pool tcpipqent_pool;
411 
412 void
413 tcpipqent_init(void)
414 {
415 
416 	pool_init(&tcpipqent_pool, sizeof(struct ipqent), 0, 0, 0, "tcpipqepl",
417 	    NULL, IPL_VM);
418 }
419 
420 struct ipqent *
421 tcpipqent_alloc(void)
422 {
423 	struct ipqent *ipqe;
424 	int s;
425 
426 	s = splvm();
427 	ipqe = pool_get(&tcpipqent_pool, PR_NOWAIT);
428 	splx(s);
429 
430 	return ipqe;
431 }
432 
433 void
434 tcpipqent_free(struct ipqent *ipqe)
435 {
436 	int s;
437 
438 	s = splvm();
439 	pool_put(&tcpipqent_pool, ipqe);
440 	splx(s);
441 }
442 
443 /*
444  * Insert segment ti into reassembly queue of tcp with
445  * control block tp.  Return TH_FIN if reassembly now includes
446  * a segment with FIN.
447  */
448 static int
449 tcp_reass(struct tcpcb *tp, const struct tcphdr *th, struct mbuf *m, int tlen)
450 {
451 	struct ipqent *p, *q, *nq, *tiqe = NULL;
452 	struct socket *so = NULL;
453 	int pkt_flags;
454 	tcp_seq pkt_seq;
455 	unsigned pkt_len;
456 	u_long rcvpartdupbyte = 0;
457 	u_long rcvoobyte;
458 #ifdef TCP_REASS_COUNTERS
459 	u_int count = 0;
460 #endif
461 	uint64_t *tcps;
462 
463 	if (tp->t_inpcb)
464 		so = tp->t_inpcb->inp_socket;
465 #ifdef INET6
466 	else if (tp->t_in6pcb)
467 		so = tp->t_in6pcb->in6p_socket;
468 #endif
469 
470 	TCP_REASS_LOCK_CHECK(tp);
471 
472 	/*
473 	 * Call with th==NULL after become established to
474 	 * force pre-ESTABLISHED data up to user socket.
475 	 */
476 	if (th == NULL)
477 		goto present;
478 
479 	m_claimm(m, &tcp_reass_mowner);
480 
481 	rcvoobyte = tlen;
482 	/*
483 	 * Copy these to local variables because the TCP header gets munged
484 	 * while we are collapsing mbufs.
485 	 */
486 	pkt_seq = th->th_seq;
487 	pkt_len = tlen;
488 	pkt_flags = th->th_flags;
489 
490 	TCP_REASS_COUNTER_INCR(&tcp_reass_);
491 
492 	if ((p = TAILQ_LAST(&tp->segq, ipqehead)) != NULL) {
493 		/*
494 		 * When we miss a packet, the vast majority of time we get
495 		 * packets that follow it in order.  So optimize for that.
496 		 */
497 		if (pkt_seq == p->ipqe_seq + p->ipqe_len) {
498 			p->ipqe_len += pkt_len;
499 			p->ipqe_flags |= pkt_flags;
500 			m_cat(p->ipqe_m, m);
501 			m = NULL;
502 			tiqe = p;
503 			TAILQ_REMOVE(&tp->timeq, p, ipqe_timeq);
504 			TCP_REASS_COUNTER_INCR(&tcp_reass_appendtail);
505 			goto skip_replacement;
506 		}
507 		/*
508 		 * While we're here, if the pkt is completely beyond
509 		 * anything we have, just insert it at the tail.
510 		 */
511 		if (SEQ_GT(pkt_seq, p->ipqe_seq + p->ipqe_len)) {
512 			TCP_REASS_COUNTER_INCR(&tcp_reass_inserttail);
513 			goto insert_it;
514 		}
515 	}
516 
517 	q = TAILQ_FIRST(&tp->segq);
518 
519 	if (q != NULL) {
520 		/*
521 		 * If this segment immediately precedes the first out-of-order
522 		 * block, simply slap the segment in front of it and (mostly)
523 		 * skip the complicated logic.
524 		 */
525 		if (pkt_seq + pkt_len == q->ipqe_seq) {
526 			q->ipqe_seq = pkt_seq;
527 			q->ipqe_len += pkt_len;
528 			q->ipqe_flags |= pkt_flags;
529 			m_cat(m, q->ipqe_m);
530 			q->ipqe_m = m;
531 			tiqe = q;
532 			TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
533 			TCP_REASS_COUNTER_INCR(&tcp_reass_prependfirst);
534 			goto skip_replacement;
535 		}
536 	} else {
537 		TCP_REASS_COUNTER_INCR(&tcp_reass_empty);
538 	}
539 
540 	/*
541 	 * Find a segment which begins after this one does.
542 	 */
543 	for (p = NULL; q != NULL; q = nq) {
544 		nq = TAILQ_NEXT(q, ipqe_q);
545 #ifdef TCP_REASS_COUNTERS
546 		count++;
547 #endif
548 
549 		/*
550 		 * If the received segment is just right after this
551 		 * fragment, merge the two together and then check
552 		 * for further overlaps.
553 		 */
554 		if (q->ipqe_seq + q->ipqe_len == pkt_seq) {
555 			pkt_len += q->ipqe_len;
556 			pkt_flags |= q->ipqe_flags;
557 			pkt_seq = q->ipqe_seq;
558 			m_cat(q->ipqe_m, m);
559 			m = q->ipqe_m;
560 			TCP_REASS_COUNTER_INCR(&tcp_reass_append);
561 			goto free_ipqe;
562 		}
563 
564 		/*
565 		 * If the received segment is completely past this
566 		 * fragment, we need to go to the next fragment.
567 		 */
568 		if (SEQ_LT(q->ipqe_seq + q->ipqe_len, pkt_seq)) {
569 			p = q;
570 			continue;
571 		}
572 
573 		/*
574 		 * If the fragment is past the received segment,
575 		 * it (or any following) can't be concatenated.
576 		 */
577 		if (SEQ_GT(q->ipqe_seq, pkt_seq + pkt_len)) {
578 			TCP_REASS_COUNTER_INCR(&tcp_reass_insert);
579 			break;
580 		}
581 
582 		/*
583 		 * We've received all the data in this segment before.
584 		 * Mark it as a duplicate and return.
585 		 */
586 		if (SEQ_LEQ(q->ipqe_seq, pkt_seq) &&
587 		    SEQ_GEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) {
588 			tcps = TCP_STAT_GETREF();
589 			tcps[TCP_STAT_RCVDUPPACK]++;
590 			tcps[TCP_STAT_RCVDUPBYTE] += pkt_len;
591 			TCP_STAT_PUTREF();
592 			tcp_new_dsack(tp, pkt_seq, pkt_len);
593 			m_freem(m);
594 			if (tiqe != NULL) {
595 				tcpipqent_free(tiqe);
596 			}
597 			TCP_REASS_COUNTER_INCR(&tcp_reass_segdup);
598 			goto out;
599 		}
600 
601 		/*
602 		 * Received segment completely overlaps this fragment
603 		 * so we drop the fragment (this keeps the temporal
604 		 * ordering of segments correct).
605 		 */
606 		if (SEQ_GEQ(q->ipqe_seq, pkt_seq) &&
607 		    SEQ_LEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) {
608 			rcvpartdupbyte += q->ipqe_len;
609 			m_freem(q->ipqe_m);
610 			TCP_REASS_COUNTER_INCR(&tcp_reass_fragdup);
611 			goto free_ipqe;
612 		}
613 
614 		/*
615 		 * Received segment extends past the end of the fragment.
616 		 * Drop the overlapping bytes, merge the fragment and
617 		 * segment, and treat as a longer received packet.
618 		 */
619 		if (SEQ_LT(q->ipqe_seq, pkt_seq) &&
620 		    SEQ_GT(q->ipqe_seq + q->ipqe_len, pkt_seq))  {
621 			int overlap = q->ipqe_seq + q->ipqe_len - pkt_seq;
622 			m_adj(m, overlap);
623 			rcvpartdupbyte += overlap;
624 			m_cat(q->ipqe_m, m);
625 			m = q->ipqe_m;
626 			pkt_seq = q->ipqe_seq;
627 			pkt_len += q->ipqe_len - overlap;
628 			rcvoobyte -= overlap;
629 			TCP_REASS_COUNTER_INCR(&tcp_reass_overlaptail);
630 			goto free_ipqe;
631 		}
632 
633 		/*
634 		 * Received segment extends past the front of the fragment.
635 		 * Drop the overlapping bytes on the received packet. The
636 		 * packet will then be concatenated with this fragment a
637 		 * bit later.
638 		 */
639 		if (SEQ_GT(q->ipqe_seq, pkt_seq) &&
640 		    SEQ_LT(q->ipqe_seq, pkt_seq + pkt_len))  {
641 			int overlap = pkt_seq + pkt_len - q->ipqe_seq;
642 			m_adj(m, -overlap);
643 			pkt_len -= overlap;
644 			rcvpartdupbyte += overlap;
645 			TCP_REASS_COUNTER_INCR(&tcp_reass_overlapfront);
646 			rcvoobyte -= overlap;
647 		}
648 
649 		/*
650 		 * If the received segment immediately precedes this
651 		 * fragment then tack the fragment onto this segment
652 		 * and reinsert the data.
653 		 */
654 		if (q->ipqe_seq == pkt_seq + pkt_len) {
655 			pkt_len += q->ipqe_len;
656 			pkt_flags |= q->ipqe_flags;
657 			m_cat(m, q->ipqe_m);
658 			TAILQ_REMOVE(&tp->segq, q, ipqe_q);
659 			TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
660 			tp->t_segqlen--;
661 			KASSERT(tp->t_segqlen >= 0);
662 			KASSERT(tp->t_segqlen != 0 ||
663 			    (TAILQ_EMPTY(&tp->segq) &&
664 			    TAILQ_EMPTY(&tp->timeq)));
665 			if (tiqe == NULL) {
666 				tiqe = q;
667 			} else {
668 				tcpipqent_free(q);
669 			}
670 			TCP_REASS_COUNTER_INCR(&tcp_reass_prepend);
671 			break;
672 		}
673 
674 		/*
675 		 * If the fragment is before the segment, remember it.
676 		 * When this loop is terminated, p will contain the
677 		 * pointer to the fragment that is right before the
678 		 * received segment.
679 		 */
680 		if (SEQ_LEQ(q->ipqe_seq, pkt_seq))
681 			p = q;
682 
683 		continue;
684 
685 		/*
686 		 * This is a common operation.  It also will allow
687 		 * to save doing a malloc/free in most instances.
688 		 */
689 	  free_ipqe:
690 		TAILQ_REMOVE(&tp->segq, q, ipqe_q);
691 		TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
692 		tp->t_segqlen--;
693 		KASSERT(tp->t_segqlen >= 0);
694 		KASSERT(tp->t_segqlen != 0 ||
695 		    (TAILQ_EMPTY(&tp->segq) && TAILQ_EMPTY(&tp->timeq)));
696 		if (tiqe == NULL) {
697 			tiqe = q;
698 		} else {
699 			tcpipqent_free(q);
700 		}
701 	}
702 
703 #ifdef TCP_REASS_COUNTERS
704 	if (count > 7)
705 		TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[0]);
706 	else if (count > 0)
707 		TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[count]);
708 #endif
709 
710 insert_it:
711 	/*
712 	 * Allocate a new queue entry (block) since the received segment
713 	 * did not collapse onto any other out-of-order block. If it had
714 	 * collapsed, tiqe would not be NULL and we would be reusing it.
715 	 *
716 	 * If the allocation fails, drop the packet.
717 	 */
718 	if (tiqe == NULL) {
719 		tiqe = tcpipqent_alloc();
720 		if (tiqe == NULL) {
721 			TCP_STATINC(TCP_STAT_RCVMEMDROP);
722 			m_freem(m);
723 			goto out;
724 		}
725 	}
726 
727 	/*
728 	 * Update the counters.
729 	 */
730 	tp->t_rcvoopack++;
731 	tcps = TCP_STAT_GETREF();
732 	tcps[TCP_STAT_RCVOOPACK]++;
733 	tcps[TCP_STAT_RCVOOBYTE] += rcvoobyte;
734 	if (rcvpartdupbyte) {
735 	    tcps[TCP_STAT_RCVPARTDUPPACK]++;
736 	    tcps[TCP_STAT_RCVPARTDUPBYTE] += rcvpartdupbyte;
737 	}
738 	TCP_STAT_PUTREF();
739 
740 	/*
741 	 * Insert the new fragment queue entry into both queues.
742 	 */
743 	tiqe->ipqe_m = m;
744 	tiqe->ipqe_seq = pkt_seq;
745 	tiqe->ipqe_len = pkt_len;
746 	tiqe->ipqe_flags = pkt_flags;
747 	if (p == NULL) {
748 		TAILQ_INSERT_HEAD(&tp->segq, tiqe, ipqe_q);
749 	} else {
750 		TAILQ_INSERT_AFTER(&tp->segq, p, tiqe, ipqe_q);
751 	}
752 	tp->t_segqlen++;
753 
754 skip_replacement:
755 	TAILQ_INSERT_HEAD(&tp->timeq, tiqe, ipqe_timeq);
756 
757 present:
758 	/*
759 	 * Present data to user, advancing rcv_nxt through
760 	 * completed sequence space.
761 	 */
762 	if (TCPS_HAVEESTABLISHED(tp->t_state) == 0)
763 		goto out;
764 	q = TAILQ_FIRST(&tp->segq);
765 	if (q == NULL || q->ipqe_seq != tp->rcv_nxt)
766 		goto out;
767 	if (tp->t_state == TCPS_SYN_RECEIVED && q->ipqe_len)
768 		goto out;
769 
770 	tp->rcv_nxt += q->ipqe_len;
771 	pkt_flags = q->ipqe_flags & TH_FIN;
772 	nd6_hint(tp);
773 
774 	TAILQ_REMOVE(&tp->segq, q, ipqe_q);
775 	TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
776 	tp->t_segqlen--;
777 	KASSERT(tp->t_segqlen >= 0);
778 	KASSERT(tp->t_segqlen != 0 ||
779 	    (TAILQ_EMPTY(&tp->segq) && TAILQ_EMPTY(&tp->timeq)));
780 	if (so->so_state & SS_CANTRCVMORE)
781 		m_freem(q->ipqe_m);
782 	else
783 		sbappendstream(&so->so_rcv, q->ipqe_m);
784 	tcpipqent_free(q);
785 	TCP_REASS_UNLOCK(tp);
786 	sorwakeup(so);
787 	return pkt_flags;
788 
789 out:
790 	TCP_REASS_UNLOCK(tp);
791 	return 0;
792 }
793 
794 #ifdef INET6
795 int
796 tcp6_input(struct mbuf **mp, int *offp, int proto)
797 {
798 	struct mbuf *m = *mp;
799 
800 	/*
801 	 * draft-itojun-ipv6-tcp-to-anycast
802 	 * better place to put this in?
803 	 */
804 	if (m->m_flags & M_ANYCAST6) {
805 		struct ip6_hdr *ip6;
806 		if (m->m_len < sizeof(struct ip6_hdr)) {
807 			if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) {
808 				TCP_STATINC(TCP_STAT_RCVSHORT);
809 				return IPPROTO_DONE;
810 			}
811 		}
812 		ip6 = mtod(m, struct ip6_hdr *);
813 		icmp6_error(m, ICMP6_DST_UNREACH, ICMP6_DST_UNREACH_ADDR,
814 		    (char *)&ip6->ip6_dst - (char *)ip6);
815 		return IPPROTO_DONE;
816 	}
817 
818 	tcp_input(m, *offp, proto);
819 	return IPPROTO_DONE;
820 }
821 #endif
822 
823 static void
824 tcp4_log_refused(const struct ip *ip, const struct tcphdr *th)
825 {
826 	char src[INET_ADDRSTRLEN];
827 	char dst[INET_ADDRSTRLEN];
828 
829 	if (ip) {
830 		in_print(src, sizeof(src), &ip->ip_src);
831 		in_print(dst, sizeof(dst), &ip->ip_dst);
832 	} else {
833 		strlcpy(src, "(unknown)", sizeof(src));
834 		strlcpy(dst, "(unknown)", sizeof(dst));
835 	}
836 	log(LOG_INFO,
837 	    "Connection attempt to TCP %s:%d from %s:%d\n",
838 	    dst, ntohs(th->th_dport),
839 	    src, ntohs(th->th_sport));
840 }
841 
842 #ifdef INET6
843 static void
844 tcp6_log_refused(const struct ip6_hdr *ip6, const struct tcphdr *th)
845 {
846 	char src[INET6_ADDRSTRLEN];
847 	char dst[INET6_ADDRSTRLEN];
848 
849 	if (ip6) {
850 		in6_print(src, sizeof(src), &ip6->ip6_src);
851 		in6_print(dst, sizeof(dst), &ip6->ip6_dst);
852 	} else {
853 		strlcpy(src, "(unknown v6)", sizeof(src));
854 		strlcpy(dst, "(unknown v6)", sizeof(dst));
855 	}
856 	log(LOG_INFO,
857 	    "Connection attempt to TCP [%s]:%d from [%s]:%d\n",
858 	    dst, ntohs(th->th_dport),
859 	    src, ntohs(th->th_sport));
860 }
861 #endif
862 
863 /*
864  * Checksum extended TCP header and data.
865  */
866 int
867 tcp_input_checksum(int af, struct mbuf *m, const struct tcphdr *th,
868     int toff, int off, int tlen)
869 {
870 	struct ifnet *rcvif;
871 	int s;
872 
873 	/*
874 	 * XXX it's better to record and check if this mbuf is
875 	 * already checked.
876 	 */
877 
878 	rcvif = m_get_rcvif(m, &s);
879 	if (__predict_false(rcvif == NULL))
880 		goto badcsum; /* XXX */
881 
882 	switch (af) {
883 	case AF_INET:
884 		switch (m->m_pkthdr.csum_flags &
885 			((rcvif->if_csum_flags_rx & M_CSUM_TCPv4) |
886 			 M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) {
887 		case M_CSUM_TCPv4|M_CSUM_TCP_UDP_BAD:
888 			TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_bad);
889 			goto badcsum;
890 
891 		case M_CSUM_TCPv4|M_CSUM_DATA: {
892 			u_int32_t hw_csum = m->m_pkthdr.csum_data;
893 
894 			TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_data);
895 			if (m->m_pkthdr.csum_flags & M_CSUM_NO_PSEUDOHDR) {
896 				const struct ip *ip =
897 				    mtod(m, const struct ip *);
898 
899 				hw_csum = in_cksum_phdr(ip->ip_src.s_addr,
900 				    ip->ip_dst.s_addr,
901 				    htons(hw_csum + tlen + off + IPPROTO_TCP));
902 			}
903 			if ((hw_csum ^ 0xffff) != 0)
904 				goto badcsum;
905 			break;
906 		}
907 
908 		case M_CSUM_TCPv4:
909 			/* Checksum was okay. */
910 			TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_ok);
911 			break;
912 
913 		default:
914 			/*
915 			 * Must compute it ourselves.  Maybe skip checksum
916 			 * on loopback interfaces.
917 			 */
918 			if (__predict_true(!(rcvif->if_flags & IFF_LOOPBACK) ||
919 					   tcp_do_loopback_cksum)) {
920 				TCP_CSUM_COUNTER_INCR(&tcp_swcsum);
921 				if (in4_cksum(m, IPPROTO_TCP, toff,
922 					      tlen + off) != 0)
923 					goto badcsum;
924 			}
925 			break;
926 		}
927 		break;
928 
929 #ifdef INET6
930 	case AF_INET6:
931 		switch (m->m_pkthdr.csum_flags &
932 			((rcvif->if_csum_flags_rx & M_CSUM_TCPv6) |
933 			 M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) {
934 		case M_CSUM_TCPv6|M_CSUM_TCP_UDP_BAD:
935 			TCP_CSUM_COUNTER_INCR(&tcp6_hwcsum_bad);
936 			goto badcsum;
937 
938 #if 0 /* notyet */
939 		case M_CSUM_TCPv6|M_CSUM_DATA:
940 #endif
941 
942 		case M_CSUM_TCPv6:
943 			/* Checksum was okay. */
944 			TCP_CSUM_COUNTER_INCR(&tcp6_hwcsum_ok);
945 			break;
946 
947 		default:
948 			/*
949 			 * Must compute it ourselves.  Maybe skip checksum
950 			 * on loopback interfaces.
951 			 */
952 			if (__predict_true((m->m_flags & M_LOOP) == 0 ||
953 			    tcp_do_loopback_cksum)) {
954 				TCP_CSUM_COUNTER_INCR(&tcp6_swcsum);
955 				if (in6_cksum(m, IPPROTO_TCP, toff,
956 				    tlen + off) != 0)
957 					goto badcsum;
958 			}
959 		}
960 		break;
961 #endif /* INET6 */
962 	}
963 	m_put_rcvif(rcvif, &s);
964 
965 	return 0;
966 
967 badcsum:
968 	m_put_rcvif(rcvif, &s);
969 	TCP_STATINC(TCP_STAT_RCVBADSUM);
970 	return -1;
971 }
972 
973 /*
974  * When a packet arrives addressed to a vestigial tcpbp, we
975  * nevertheless have to respond to it per the spec.
976  *
977  * This code is duplicated from the one in tcp_input().
978  */
979 static void tcp_vtw_input(struct tcphdr *th, vestigial_inpcb_t *vp,
980     struct mbuf *m, int tlen)
981 {
982 	int tiflags;
983 	int todrop;
984 	uint32_t t_flags = 0;
985 	uint64_t *tcps;
986 
987 	tiflags = th->th_flags;
988 	todrop  = vp->rcv_nxt - th->th_seq;
989 
990 	if (todrop > 0) {
991 		if (tiflags & TH_SYN) {
992 			tiflags &= ~TH_SYN;
993 			th->th_seq++;
994 			tcp_urp_drop(th, 1, &tiflags);
995 			todrop--;
996 		}
997 		if (todrop > tlen ||
998 		    (todrop == tlen && (tiflags & TH_FIN) == 0)) {
999 			/*
1000 			 * Any valid FIN or RST must be to the left of the
1001 			 * window.  At this point the FIN or RST must be a
1002 			 * duplicate or out of sequence; drop it.
1003 			 */
1004 			if (tiflags & TH_RST)
1005 				goto drop;
1006 			tiflags &= ~(TH_FIN|TH_RST);
1007 
1008 			/*
1009 			 * Send an ACK to resynchronize and drop any data.
1010 			 * But keep on processing for RST or ACK.
1011 			 */
1012 			t_flags |= TF_ACKNOW;
1013 			todrop = tlen;
1014 			tcps = TCP_STAT_GETREF();
1015 			tcps[TCP_STAT_RCVDUPPACK] += 1;
1016 			tcps[TCP_STAT_RCVDUPBYTE] += todrop;
1017 			TCP_STAT_PUTREF();
1018 		} else if ((tiflags & TH_RST) &&
1019 		    th->th_seq != vp->rcv_nxt) {
1020 			/*
1021 			 * Test for reset before adjusting the sequence
1022 			 * number for overlapping data.
1023 			 */
1024 			goto dropafterack_ratelim;
1025 		} else {
1026 			tcps = TCP_STAT_GETREF();
1027 			tcps[TCP_STAT_RCVPARTDUPPACK] += 1;
1028 			tcps[TCP_STAT_RCVPARTDUPBYTE] += todrop;
1029 			TCP_STAT_PUTREF();
1030 		}
1031 
1032 //		tcp_new_dsack(tp, th->th_seq, todrop);
1033 //		hdroptlen += todrop;	/*drop from head afterwards*/
1034 
1035 		th->th_seq += todrop;
1036 		tlen -= todrop;
1037 		tcp_urp_drop(th, todrop, &tiflags);
1038 	}
1039 
1040 	/*
1041 	 * If new data are received on a connection after the
1042 	 * user processes are gone, then RST the other end.
1043 	 */
1044 	if (tlen) {
1045 		TCP_STATINC(TCP_STAT_RCVAFTERCLOSE);
1046 		goto dropwithreset;
1047 	}
1048 
1049 	/*
1050 	 * If segment ends after window, drop trailing data
1051 	 * (and PUSH and FIN); if nothing left, just ACK.
1052 	 */
1053 	todrop = (th->th_seq + tlen) - (vp->rcv_nxt + vp->rcv_wnd);
1054 
1055 	if (todrop > 0) {
1056 		TCP_STATINC(TCP_STAT_RCVPACKAFTERWIN);
1057 		if (todrop >= tlen) {
1058 			/*
1059 			 * The segment actually starts after the window.
1060 			 * th->th_seq + tlen - vp->rcv_nxt - vp->rcv_wnd >= tlen
1061 			 * th->th_seq - vp->rcv_nxt - vp->rcv_wnd >= 0
1062 			 * th->th_seq >= vp->rcv_nxt + vp->rcv_wnd
1063 			 */
1064 			TCP_STATADD(TCP_STAT_RCVBYTEAFTERWIN, tlen);
1065 
1066 			/*
1067 			 * If a new connection request is received
1068 			 * while in TIME_WAIT, drop the old connection
1069 			 * and start over if the sequence numbers
1070 			 * are above the previous ones.
1071 			 */
1072 			if ((tiflags & TH_SYN) &&
1073 			    SEQ_GT(th->th_seq, vp->rcv_nxt)) {
1074 				/*
1075 				 * We only support this in the !NOFDREF case, which
1076 				 * is to say: not here.
1077 				 */
1078 				goto dropwithreset;
1079 			}
1080 
1081 			/*
1082 			 * If window is closed can only take segments at
1083 			 * window edge, and have to drop data and PUSH from
1084 			 * incoming segments.  Continue processing, but
1085 			 * remember to ack.  Otherwise, drop segment
1086 			 * and (if not RST) ack.
1087 			 */
1088 			if (vp->rcv_wnd == 0 && th->th_seq == vp->rcv_nxt) {
1089 				t_flags |= TF_ACKNOW;
1090 				TCP_STATINC(TCP_STAT_RCVWINPROBE);
1091 			} else {
1092 				goto dropafterack;
1093 			}
1094 		} else {
1095 			TCP_STATADD(TCP_STAT_RCVBYTEAFTERWIN, todrop);
1096 		}
1097 		m_adj(m, -todrop);
1098 		tlen -= todrop;
1099 		tiflags &= ~(TH_PUSH|TH_FIN);
1100 	}
1101 
1102 	if (tiflags & TH_RST) {
1103 		if (th->th_seq != vp->rcv_nxt)
1104 			goto dropafterack_ratelim;
1105 
1106 		vtw_del(vp->ctl, vp->vtw);
1107 		goto drop;
1108 	}
1109 
1110 	/*
1111 	 * If the ACK bit is off we drop the segment and return.
1112 	 */
1113 	if ((tiflags & TH_ACK) == 0) {
1114 		if (t_flags & TF_ACKNOW)
1115 			goto dropafterack;
1116 		goto drop;
1117 	}
1118 
1119 	/*
1120 	 * In TIME_WAIT state the only thing that should arrive
1121 	 * is a retransmission of the remote FIN.  Acknowledge
1122 	 * it and restart the finack timer.
1123 	 */
1124 	vtw_restart(vp);
1125 	goto dropafterack;
1126 
1127 dropafterack:
1128 	/*
1129 	 * Generate an ACK dropping incoming segment if it occupies
1130 	 * sequence space, where the ACK reflects our state.
1131 	 */
1132 	if (tiflags & TH_RST)
1133 		goto drop;
1134 	goto dropafterack2;
1135 
1136 dropafterack_ratelim:
1137 	/*
1138 	 * We may want to rate-limit ACKs against SYN/RST attack.
1139 	 */
1140 	if (ppsratecheck(&tcp_ackdrop_ppslim_last, &tcp_ackdrop_ppslim_count,
1141 	    tcp_ackdrop_ppslim) == 0) {
1142 		/* XXX stat */
1143 		goto drop;
1144 	}
1145 	/* ...fall into dropafterack2... */
1146 
1147 dropafterack2:
1148 	(void)tcp_respond(0, m, m, th, th->th_seq + tlen, th->th_ack, TH_ACK);
1149 	return;
1150 
1151 dropwithreset:
1152 	/*
1153 	 * Generate a RST, dropping incoming segment.
1154 	 * Make ACK acceptable to originator of segment.
1155 	 */
1156 	if (tiflags & TH_RST)
1157 		goto drop;
1158 
1159 	if (tiflags & TH_ACK) {
1160 		tcp_respond(0, m, m, th, (tcp_seq)0, th->th_ack, TH_RST);
1161 	} else {
1162 		if (tiflags & TH_SYN)
1163 			++tlen;
1164 		(void)tcp_respond(0, m, m, th, th->th_seq + tlen, (tcp_seq)0,
1165 		    TH_RST|TH_ACK);
1166 	}
1167 	return;
1168 drop:
1169 	m_freem(m);
1170 }
1171 
1172 /*
1173  * TCP input routine, follows pages 65-76 of RFC 793 very closely.
1174  */
1175 void
1176 tcp_input(struct mbuf *m, int off, int proto)
1177 {
1178 	struct tcphdr *th;
1179 	struct ip *ip;
1180 	struct inpcb *inp;
1181 #ifdef INET6
1182 	struct ip6_hdr *ip6;
1183 	struct in6pcb *in6p;
1184 #endif
1185 	u_int8_t *optp = NULL;
1186 	int optlen = 0;
1187 	int len, tlen, hdroptlen = 0;
1188 	struct tcpcb *tp = NULL;
1189 	int tiflags;
1190 	struct socket *so = NULL;
1191 	int todrop, acked, ourfinisacked, needoutput = 0;
1192 	bool dupseg;
1193 #ifdef TCP_DEBUG
1194 	short ostate = 0;
1195 #endif
1196 	u_long tiwin;
1197 	struct tcp_opt_info opti;
1198 	int thlen, iphlen;
1199 	int af;		/* af on the wire */
1200 	struct mbuf *tcp_saveti = NULL;
1201 	uint32_t ts_rtt;
1202 	uint8_t iptos;
1203 	uint64_t *tcps;
1204 	vestigial_inpcb_t vestige;
1205 
1206 	vestige.valid = 0;
1207 
1208 	MCLAIM(m, &tcp_rx_mowner);
1209 
1210 	TCP_STATINC(TCP_STAT_RCVTOTAL);
1211 
1212 	memset(&opti, 0, sizeof(opti));
1213 	opti.ts_present = 0;
1214 	opti.maxseg = 0;
1215 
1216 	/*
1217 	 * RFC1122 4.2.3.10, p. 104: discard bcast/mcast SYN.
1218 	 *
1219 	 * TCP is, by definition, unicast, so we reject all
1220 	 * multicast outright.
1221 	 *
1222 	 * Note, there are additional src/dst address checks in
1223 	 * the AF-specific code below.
1224 	 */
1225 	if (m->m_flags & (M_BCAST|M_MCAST)) {
1226 		/* XXX stat */
1227 		goto drop;
1228 	}
1229 #ifdef INET6
1230 	if (m->m_flags & M_ANYCAST6) {
1231 		/* XXX stat */
1232 		goto drop;
1233 	}
1234 #endif
1235 
1236 	M_REGION_GET(th, struct tcphdr *, m, off, sizeof(struct tcphdr));
1237 	if (th == NULL) {
1238 		TCP_STATINC(TCP_STAT_RCVSHORT);
1239 		return;
1240 	}
1241 
1242 	/*
1243 	 * Enforce alignment requirements that are violated in
1244 	 * some cases, see kern/50766 for details.
1245 	 */
1246 	if (TCP_HDR_ALIGNED_P(th) == 0) {
1247 		m = m_copyup(m, off + sizeof(struct tcphdr), 0);
1248 		if (m == NULL) {
1249 			TCP_STATINC(TCP_STAT_RCVSHORT);
1250 			return;
1251 		}
1252 		th = (struct tcphdr *)(mtod(m, char *) + off);
1253 	}
1254 	KASSERT(TCP_HDR_ALIGNED_P(th));
1255 
1256 	/*
1257 	 * Get IP and TCP header.
1258 	 * Note: IP leaves IP header in first mbuf.
1259 	 */
1260 	ip = mtod(m, struct ip *);
1261 #ifdef INET6
1262 	ip6 = mtod(m, struct ip6_hdr *);
1263 #endif
1264 	switch (ip->ip_v) {
1265 	case 4:
1266 		af = AF_INET;
1267 		iphlen = sizeof(struct ip);
1268 
1269 		if (IN_MULTICAST(ip->ip_dst.s_addr) ||
1270 		    in_broadcast(ip->ip_dst, m_get_rcvif_NOMPSAFE(m)))
1271 			goto drop;
1272 
1273 		/* We do the checksum after PCB lookup... */
1274 		len = ntohs(ip->ip_len);
1275 		tlen = len - off;
1276 		iptos = ip->ip_tos;
1277 		break;
1278 #ifdef INET6
1279 	case 6:
1280 		iphlen = sizeof(struct ip6_hdr);
1281 		af = AF_INET6;
1282 
1283 		/*
1284 		 * Be proactive about unspecified IPv6 address in source.
1285 		 * As we use all-zero to indicate unbounded/unconnected pcb,
1286 		 * unspecified IPv6 address can be used to confuse us.
1287 		 *
1288 		 * Note that packets with unspecified IPv6 destination is
1289 		 * already dropped in ip6_input.
1290 		 */
1291 		if (IN6_IS_ADDR_UNSPECIFIED(&ip6->ip6_src)) {
1292 			/* XXX stat */
1293 			goto drop;
1294 		}
1295 
1296 		/*
1297 		 * Make sure destination address is not multicast.
1298 		 * Source address checked in ip6_input().
1299 		 */
1300 		if (IN6_IS_ADDR_MULTICAST(&ip6->ip6_dst)) {
1301 			/* XXX stat */
1302 			goto drop;
1303 		}
1304 
1305 		/* We do the checksum after PCB lookup... */
1306 		len = m->m_pkthdr.len;
1307 		tlen = len - off;
1308 		iptos = (ntohl(ip6->ip6_flow) >> 20) & 0xff;
1309 		break;
1310 #endif
1311 	default:
1312 		m_freem(m);
1313 		return;
1314 	}
1315 
1316 
1317 	/*
1318 	 * Check that TCP offset makes sense, pull out TCP options and
1319 	 * adjust length.
1320 	 */
1321 	thlen = th->th_off << 2;
1322 	if (thlen < sizeof(struct tcphdr) || thlen > tlen) {
1323 		TCP_STATINC(TCP_STAT_RCVBADOFF);
1324 		goto drop;
1325 	}
1326 	tlen -= thlen;
1327 
1328 	if (thlen > sizeof(struct tcphdr)) {
1329 		M_REGION_GET(th, struct tcphdr *, m, off, thlen);
1330 		if (th == NULL) {
1331 			TCP_STATINC(TCP_STAT_RCVSHORT);
1332 			return;
1333 		}
1334 		KASSERT(TCP_HDR_ALIGNED_P(th));
1335 		optlen = thlen - sizeof(struct tcphdr);
1336 		optp = ((u_int8_t *)th) + sizeof(struct tcphdr);
1337 
1338 		/*
1339 		 * Do quick retrieval of timestamp options.
1340 		 *
1341 		 * If timestamp is the only option and it's formatted as
1342 		 * recommended in RFC 1323 appendix A, we quickly get the
1343 		 * values now and don't bother calling tcp_dooptions(),
1344 		 * etc.
1345 		 */
1346 		if ((optlen == TCPOLEN_TSTAMP_APPA ||
1347 		     (optlen > TCPOLEN_TSTAMP_APPA &&
1348 		      optp[TCPOLEN_TSTAMP_APPA] == TCPOPT_EOL)) &&
1349 		    be32dec(optp) == TCPOPT_TSTAMP_HDR &&
1350 		    (th->th_flags & TH_SYN) == 0) {
1351 			opti.ts_present = 1;
1352 			opti.ts_val = be32dec(optp + 4);
1353 			opti.ts_ecr = be32dec(optp + 8);
1354 			optp = NULL;	/* we've parsed the options */
1355 		}
1356 	}
1357 	tiflags = th->th_flags;
1358 
1359 	/*
1360 	 * Checksum extended TCP header and data
1361 	 */
1362 	if (tcp_input_checksum(af, m, th, off, thlen, tlen))
1363 		goto badcsum;
1364 
1365 	/*
1366 	 * Locate pcb for segment.
1367 	 */
1368 findpcb:
1369 	inp = NULL;
1370 #ifdef INET6
1371 	in6p = NULL;
1372 #endif
1373 	switch (af) {
1374 	case AF_INET:
1375 		inp = in_pcblookup_connect(&tcbtable, ip->ip_src, th->th_sport,
1376 		    ip->ip_dst, th->th_dport, &vestige);
1377 		if (inp == NULL && !vestige.valid) {
1378 			TCP_STATINC(TCP_STAT_PCBHASHMISS);
1379 			inp = in_pcblookup_bind(&tcbtable, ip->ip_dst,
1380 			    th->th_dport);
1381 		}
1382 #ifdef INET6
1383 		if (inp == NULL && !vestige.valid) {
1384 			struct in6_addr s, d;
1385 
1386 			/* mapped addr case */
1387 			in6_in_2_v4mapin6(&ip->ip_src, &s);
1388 			in6_in_2_v4mapin6(&ip->ip_dst, &d);
1389 			in6p = in6_pcblookup_connect(&tcbtable, &s,
1390 			    th->th_sport, &d, th->th_dport, 0, &vestige);
1391 			if (in6p == 0 && !vestige.valid) {
1392 				TCP_STATINC(TCP_STAT_PCBHASHMISS);
1393 				in6p = in6_pcblookup_bind(&tcbtable, &d,
1394 				    th->th_dport, 0);
1395 			}
1396 		}
1397 #endif
1398 #ifndef INET6
1399 		if (inp == NULL && !vestige.valid)
1400 #else
1401 		if (inp == NULL && in6p == NULL && !vestige.valid)
1402 #endif
1403 		{
1404 			TCP_STATINC(TCP_STAT_NOPORT);
1405 			if (tcp_log_refused &&
1406 			    (tiflags & (TH_RST|TH_ACK|TH_SYN)) == TH_SYN) {
1407 				tcp4_log_refused(ip, th);
1408 			}
1409 			tcp_fields_to_host(th);
1410 			goto dropwithreset_ratelim;
1411 		}
1412 #if defined(IPSEC)
1413 		if (ipsec_used) {
1414 			if (inp && ipsec_in_reject(m, inp)) {
1415 				goto drop;
1416 			}
1417 #ifdef INET6
1418 			else if (in6p && ipsec_in_reject(m, in6p)) {
1419 				goto drop;
1420 			}
1421 #endif
1422 		}
1423 #endif /*IPSEC*/
1424 		break;
1425 #ifdef INET6
1426 	case AF_INET6:
1427 	    {
1428 		int faith;
1429 
1430 #if defined(NFAITH) && NFAITH > 0
1431 		faith = faithprefix(&ip6->ip6_dst);
1432 #else
1433 		faith = 0;
1434 #endif
1435 		in6p = in6_pcblookup_connect(&tcbtable, &ip6->ip6_src,
1436 		    th->th_sport, &ip6->ip6_dst, th->th_dport, faith, &vestige);
1437 		if (!in6p && !vestige.valid) {
1438 			TCP_STATINC(TCP_STAT_PCBHASHMISS);
1439 			in6p = in6_pcblookup_bind(&tcbtable, &ip6->ip6_dst,
1440 			    th->th_dport, faith);
1441 		}
1442 		if (!in6p && !vestige.valid) {
1443 			TCP_STATINC(TCP_STAT_NOPORT);
1444 			if (tcp_log_refused &&
1445 			    (tiflags & (TH_RST|TH_ACK|TH_SYN)) == TH_SYN) {
1446 				tcp6_log_refused(ip6, th);
1447 			}
1448 			tcp_fields_to_host(th);
1449 			goto dropwithreset_ratelim;
1450 		}
1451 #if defined(IPSEC)
1452 		if (ipsec_used && in6p && ipsec_in_reject(m, in6p)) {
1453 			goto drop;
1454 		}
1455 #endif
1456 		break;
1457 	    }
1458 #endif
1459 	}
1460 
1461 	tcp_fields_to_host(th);
1462 
1463 	/*
1464 	 * If the state is CLOSED (i.e., TCB does not exist) then
1465 	 * all data in the incoming segment is discarded.
1466 	 * If the TCB exists but is in CLOSED state, it is embryonic,
1467 	 * but should either do a listen or a connect soon.
1468 	 */
1469 	tp = NULL;
1470 	so = NULL;
1471 	if (inp) {
1472 		/* Check the minimum TTL for socket. */
1473 		if (ip->ip_ttl < inp->inp_ip_minttl)
1474 			goto drop;
1475 
1476 		tp = intotcpcb(inp);
1477 		so = inp->inp_socket;
1478 	}
1479 #ifdef INET6
1480 	else if (in6p) {
1481 		tp = in6totcpcb(in6p);
1482 		so = in6p->in6p_socket;
1483 	}
1484 #endif
1485 	else if (vestige.valid) {
1486 		/* We do not support the resurrection of vtw tcpcps. */
1487 		tcp_vtw_input(th, &vestige, m, tlen);
1488 		m = NULL;
1489 		goto drop;
1490 	}
1491 
1492 	if (tp == NULL)
1493 		goto dropwithreset_ratelim;
1494 	if (tp->t_state == TCPS_CLOSED)
1495 		goto drop;
1496 
1497 	KASSERT(so->so_lock == softnet_lock);
1498 	KASSERT(solocked(so));
1499 
1500 	/* Unscale the window into a 32-bit value. */
1501 	if ((tiflags & TH_SYN) == 0)
1502 		tiwin = th->th_win << tp->snd_scale;
1503 	else
1504 		tiwin = th->th_win;
1505 
1506 #ifdef INET6
1507 	/* save packet options if user wanted */
1508 	if (in6p && (in6p->in6p_flags & IN6P_CONTROLOPTS)) {
1509 		if (in6p->in6p_options) {
1510 			m_freem(in6p->in6p_options);
1511 			in6p->in6p_options = NULL;
1512 		}
1513 		ip6_savecontrol(in6p, &in6p->in6p_options, ip6, m);
1514 	}
1515 #endif
1516 
1517 	if (so->so_options & SO_DEBUG) {
1518 #ifdef TCP_DEBUG
1519 		ostate = tp->t_state;
1520 #endif
1521 
1522 		tcp_saveti = NULL;
1523 		if (iphlen + sizeof(struct tcphdr) > MHLEN)
1524 			goto nosave;
1525 
1526 		if (m->m_len > iphlen && (m->m_flags & M_EXT) == 0) {
1527 			tcp_saveti = m_copym(m, 0, iphlen, M_DONTWAIT);
1528 			if (tcp_saveti == NULL)
1529 				goto nosave;
1530 		} else {
1531 			MGETHDR(tcp_saveti, M_DONTWAIT, MT_HEADER);
1532 			if (tcp_saveti == NULL)
1533 				goto nosave;
1534 			MCLAIM(m, &tcp_mowner);
1535 			tcp_saveti->m_len = iphlen;
1536 			m_copydata(m, 0, iphlen,
1537 			    mtod(tcp_saveti, void *));
1538 		}
1539 
1540 		if (M_TRAILINGSPACE(tcp_saveti) < sizeof(struct tcphdr)) {
1541 			m_freem(tcp_saveti);
1542 			tcp_saveti = NULL;
1543 		} else {
1544 			tcp_saveti->m_len += sizeof(struct tcphdr);
1545 			memcpy(mtod(tcp_saveti, char *) + iphlen, th,
1546 			    sizeof(struct tcphdr));
1547 		}
1548 nosave:;
1549 	}
1550 
1551 	if (so->so_options & SO_ACCEPTCONN) {
1552 		union syn_cache_sa src;
1553 		union syn_cache_sa dst;
1554 
1555 		KASSERT(tp->t_state == TCPS_LISTEN);
1556 
1557 		memset(&src, 0, sizeof(src));
1558 		memset(&dst, 0, sizeof(dst));
1559 		switch (af) {
1560 		case AF_INET:
1561 			src.sin.sin_len = sizeof(struct sockaddr_in);
1562 			src.sin.sin_family = AF_INET;
1563 			src.sin.sin_addr = ip->ip_src;
1564 			src.sin.sin_port = th->th_sport;
1565 
1566 			dst.sin.sin_len = sizeof(struct sockaddr_in);
1567 			dst.sin.sin_family = AF_INET;
1568 			dst.sin.sin_addr = ip->ip_dst;
1569 			dst.sin.sin_port = th->th_dport;
1570 			break;
1571 #ifdef INET6
1572 		case AF_INET6:
1573 			src.sin6.sin6_len = sizeof(struct sockaddr_in6);
1574 			src.sin6.sin6_family = AF_INET6;
1575 			src.sin6.sin6_addr = ip6->ip6_src;
1576 			src.sin6.sin6_port = th->th_sport;
1577 
1578 			dst.sin6.sin6_len = sizeof(struct sockaddr_in6);
1579 			dst.sin6.sin6_family = AF_INET6;
1580 			dst.sin6.sin6_addr = ip6->ip6_dst;
1581 			dst.sin6.sin6_port = th->th_dport;
1582 			break;
1583 #endif
1584 		}
1585 
1586 		if ((tiflags & (TH_RST|TH_ACK|TH_SYN)) != TH_SYN) {
1587 			if (tiflags & TH_RST) {
1588 				syn_cache_reset(&src.sa, &dst.sa, th);
1589 			} else if ((tiflags & (TH_ACK|TH_SYN)) ==
1590 			    (TH_ACK|TH_SYN)) {
1591 				/*
1592 				 * Received a SYN,ACK. This should never
1593 				 * happen while we are in LISTEN. Send an RST.
1594 				 */
1595 				goto badsyn;
1596 			} else if (tiflags & TH_ACK) {
1597 				so = syn_cache_get(&src.sa, &dst.sa, th, so, m);
1598 				if (so == NULL) {
1599 					/*
1600 					 * We don't have a SYN for this ACK;
1601 					 * send an RST.
1602 					 */
1603 					goto badsyn;
1604 				} else if (so == (struct socket *)(-1)) {
1605 					/*
1606 					 * We were unable to create the
1607 					 * connection. If the 3-way handshake
1608 					 * was completed, and RST has been
1609 					 * sent to the peer. Since the mbuf
1610 					 * might be in use for the reply, do
1611 					 * not free it.
1612 					 */
1613 					m = NULL;
1614 				} else {
1615 					/*
1616 					 * We have created a full-blown
1617 					 * connection.
1618 					 */
1619 					tp = NULL;
1620 					inp = NULL;
1621 #ifdef INET6
1622 					in6p = NULL;
1623 #endif
1624 					switch (so->so_proto->pr_domain->dom_family) {
1625 					case AF_INET:
1626 						inp = sotoinpcb(so);
1627 						tp = intotcpcb(inp);
1628 						break;
1629 #ifdef INET6
1630 					case AF_INET6:
1631 						in6p = sotoin6pcb(so);
1632 						tp = in6totcpcb(in6p);
1633 						break;
1634 #endif
1635 					}
1636 					if (tp == NULL)
1637 						goto badsyn;	/*XXX*/
1638 					tiwin <<= tp->snd_scale;
1639 					goto after_listen;
1640 				}
1641 			} else {
1642 				/*
1643 				 * None of RST, SYN or ACK was set.
1644 				 * This is an invalid packet for a
1645 				 * TCB in LISTEN state.  Send a RST.
1646 				 */
1647 				goto badsyn;
1648 			}
1649 		} else {
1650 			/*
1651 			 * Received a SYN.
1652 			 */
1653 
1654 #ifdef INET6
1655 			/*
1656 			 * If deprecated address is forbidden, we do
1657 			 * not accept SYN to deprecated interface
1658 			 * address to prevent any new inbound
1659 			 * connection from getting established.
1660 			 * When we do not accept SYN, we send a TCP
1661 			 * RST, with deprecated source address (instead
1662 			 * of dropping it).  We compromise it as it is
1663 			 * much better for peer to send a RST, and
1664 			 * RST will be the final packet for the
1665 			 * exchange.
1666 			 *
1667 			 * If we do not forbid deprecated addresses, we
1668 			 * accept the SYN packet.  RFC2462 does not
1669 			 * suggest dropping SYN in this case.
1670 			 * If we decipher RFC2462 5.5.4, it says like
1671 			 * this:
1672 			 * 1. use of deprecated addr with existing
1673 			 *    communication is okay - "SHOULD continue
1674 			 *    to be used"
1675 			 * 2. use of it with new communication:
1676 			 *   (2a) "SHOULD NOT be used if alternate
1677 			 *        address with sufficient scope is
1678 			 *        available"
1679 			 *   (2b) nothing mentioned otherwise.
1680 			 * Here we fall into (2b) case as we have no
1681 			 * choice in our source address selection - we
1682 			 * must obey the peer.
1683 			 *
1684 			 * The wording in RFC2462 is confusing, and
1685 			 * there are multiple description text for
1686 			 * deprecated address handling - worse, they
1687 			 * are not exactly the same.  I believe 5.5.4
1688 			 * is the best one, so we follow 5.5.4.
1689 			 */
1690 			if (af == AF_INET6 && !ip6_use_deprecated) {
1691 				struct in6_ifaddr *ia6;
1692 				int s;
1693 				struct ifnet *rcvif = m_get_rcvif(m, &s);
1694 				if (rcvif == NULL)
1695 					goto dropwithreset; /* XXX */
1696 				if ((ia6 = in6ifa_ifpwithaddr(rcvif,
1697 				    &ip6->ip6_dst)) &&
1698 				    (ia6->ia6_flags & IN6_IFF_DEPRECATED)) {
1699 					tp = NULL;
1700 					m_put_rcvif(rcvif, &s);
1701 					goto dropwithreset;
1702 				}
1703 				m_put_rcvif(rcvif, &s);
1704 			}
1705 #endif
1706 
1707 			/*
1708 			 * LISTEN socket received a SYN from itself? This
1709 			 * can't possibly be valid; drop the packet.
1710 			 */
1711 			if (th->th_sport == th->th_dport) {
1712 				int eq = 0;
1713 
1714 				switch (af) {
1715 				case AF_INET:
1716 					eq = in_hosteq(ip->ip_src, ip->ip_dst);
1717 					break;
1718 #ifdef INET6
1719 				case AF_INET6:
1720 					eq = IN6_ARE_ADDR_EQUAL(&ip6->ip6_src,
1721 					    &ip6->ip6_dst);
1722 					break;
1723 #endif
1724 				}
1725 				if (eq) {
1726 					TCP_STATINC(TCP_STAT_BADSYN);
1727 					goto drop;
1728 				}
1729 			}
1730 
1731 			/*
1732 			 * SYN looks ok; create compressed TCP
1733 			 * state for it.
1734 			 */
1735 			if (so->so_qlen <= so->so_qlimit &&
1736 			    syn_cache_add(&src.sa, &dst.sa, th, off,
1737 			    so, m, optp, optlen, &opti))
1738 				m = NULL;
1739 		}
1740 
1741 		goto drop;
1742 	}
1743 
1744 after_listen:
1745 	/*
1746 	 * From here on, we're dealing with !LISTEN.
1747 	 */
1748 	KASSERT(tp->t_state != TCPS_LISTEN);
1749 
1750 	/*
1751 	 * Segment received on connection.
1752 	 * Reset idle time and keep-alive timer.
1753 	 */
1754 	tp->t_rcvtime = tcp_now;
1755 	if (TCPS_HAVEESTABLISHED(tp->t_state))
1756 		TCP_TIMER_ARM(tp, TCPT_KEEP, tp->t_keepidle);
1757 
1758 	/*
1759 	 * Process options.
1760 	 */
1761 #ifdef TCP_SIGNATURE
1762 	if (optp || (tp->t_flags & TF_SIGNATURE))
1763 #else
1764 	if (optp)
1765 #endif
1766 		if (tcp_dooptions(tp, optp, optlen, th, m, off, &opti) < 0)
1767 			goto drop;
1768 
1769 	if (TCP_SACK_ENABLED(tp)) {
1770 		tcp_del_sackholes(tp, th);
1771 	}
1772 
1773 	if (TCP_ECN_ALLOWED(tp)) {
1774 		if (tiflags & TH_CWR) {
1775 			tp->t_flags &= ~TF_ECN_SND_ECE;
1776 		}
1777 		switch (iptos & IPTOS_ECN_MASK) {
1778 		case IPTOS_ECN_CE:
1779 			tp->t_flags |= TF_ECN_SND_ECE;
1780 			TCP_STATINC(TCP_STAT_ECN_CE);
1781 			break;
1782 		case IPTOS_ECN_ECT0:
1783 			TCP_STATINC(TCP_STAT_ECN_ECT);
1784 			break;
1785 		case IPTOS_ECN_ECT1:
1786 			/* XXX: ignore for now -- rpaulo */
1787 			break;
1788 		}
1789 		/*
1790 		 * Congestion experienced.
1791 		 * Ignore if we are already trying to recover.
1792 		 */
1793 		if ((tiflags & TH_ECE) && SEQ_GEQ(tp->snd_una, tp->snd_recover))
1794 			tp->t_congctl->cong_exp(tp);
1795 	}
1796 
1797 	if (opti.ts_present && opti.ts_ecr) {
1798 		/*
1799 		 * Calculate the RTT from the returned time stamp and the
1800 		 * connection's time base.  If the time stamp is later than
1801 		 * the current time, or is extremely old, fall back to non-1323
1802 		 * RTT calculation.  Since ts_rtt is unsigned, we can test both
1803 		 * at the same time.
1804 		 *
1805 		 * Note that ts_rtt is in units of slow ticks (500
1806 		 * ms).  Since most earthbound RTTs are < 500 ms,
1807 		 * observed values will have large quantization noise.
1808 		 * Our smoothed RTT is then the fraction of observed
1809 		 * samples that are 1 tick instead of 0 (times 500
1810 		 * ms).
1811 		 *
1812 		 * ts_rtt is increased by 1 to denote a valid sample,
1813 		 * with 0 indicating an invalid measurement.  This
1814 		 * extra 1 must be removed when ts_rtt is used, or
1815 		 * else an erroneous extra 500 ms will result.
1816 		 */
1817 		ts_rtt = TCP_TIMESTAMP(tp) - opti.ts_ecr + 1;
1818 		if (ts_rtt > TCP_PAWS_IDLE)
1819 			ts_rtt = 0;
1820 	} else {
1821 		ts_rtt = 0;
1822 	}
1823 
1824 	/*
1825 	 * Fast path: check for the two common cases of a uni-directional
1826 	 * data transfer. If:
1827 	 *    o We are in the ESTABLISHED state, and
1828 	 *    o The packet has no control flags, and
1829 	 *    o The packet is in-sequence, and
1830 	 *    o The window didn't change, and
1831 	 *    o We are not retransmitting
1832 	 * It's a candidate.
1833 	 *
1834 	 * If the length (tlen) is zero and the ack moved forward, we're
1835 	 * the sender side of the transfer. Just free the data acked and
1836 	 * wake any higher level process that was blocked waiting for
1837 	 * space.
1838 	 *
1839 	 * If the length is non-zero and the ack didn't move, we're the
1840 	 * receiver side. If we're getting packets in-order (the reassembly
1841 	 * queue is empty), add the data to the socket buffer and note
1842 	 * that we need a delayed ack.
1843 	 */
1844 	if (tp->t_state == TCPS_ESTABLISHED &&
1845 	    (tiflags & (TH_SYN|TH_FIN|TH_RST|TH_URG|TH_ECE|TH_CWR|TH_ACK))
1846 	        == TH_ACK &&
1847 	    (!opti.ts_present || TSTMP_GEQ(opti.ts_val, tp->ts_recent)) &&
1848 	    th->th_seq == tp->rcv_nxt &&
1849 	    tiwin && tiwin == tp->snd_wnd &&
1850 	    tp->snd_nxt == tp->snd_max) {
1851 
1852 		/*
1853 		 * If last ACK falls within this segment's sequence numbers,
1854 		 * record the timestamp.
1855 		 * NOTE that the test is modified according to the latest
1856 		 * proposal of the tcplw@cray.com list (Braden 1993/04/26).
1857 		 *
1858 		 * note that we already know
1859 		 *	TSTMP_GEQ(opti.ts_val, tp->ts_recent)
1860 		 */
1861 		if (opti.ts_present && SEQ_LEQ(th->th_seq, tp->last_ack_sent)) {
1862 			tp->ts_recent_age = tcp_now;
1863 			tp->ts_recent = opti.ts_val;
1864 		}
1865 
1866 		if (tlen == 0) {
1867 			/* Ack prediction. */
1868 			if (SEQ_GT(th->th_ack, tp->snd_una) &&
1869 			    SEQ_LEQ(th->th_ack, tp->snd_max) &&
1870 			    tp->snd_cwnd >= tp->snd_wnd &&
1871 			    tp->t_partialacks < 0) {
1872 				/*
1873 				 * this is a pure ack for outstanding data.
1874 				 */
1875 				if (ts_rtt)
1876 					tcp_xmit_timer(tp, ts_rtt - 1);
1877 				else if (tp->t_rtttime &&
1878 				    SEQ_GT(th->th_ack, tp->t_rtseq))
1879 					tcp_xmit_timer(tp,
1880 					  tcp_now - tp->t_rtttime);
1881 				acked = th->th_ack - tp->snd_una;
1882 				tcps = TCP_STAT_GETREF();
1883 				tcps[TCP_STAT_PREDACK]++;
1884 				tcps[TCP_STAT_RCVACKPACK]++;
1885 				tcps[TCP_STAT_RCVACKBYTE] += acked;
1886 				TCP_STAT_PUTREF();
1887 				nd6_hint(tp);
1888 
1889 				if (acked > (tp->t_lastoff - tp->t_inoff))
1890 					tp->t_lastm = NULL;
1891 				sbdrop(&so->so_snd, acked);
1892 				tp->t_lastoff -= acked;
1893 
1894 				icmp_check(tp, th, acked);
1895 
1896 				tp->snd_una = th->th_ack;
1897 				tp->snd_fack = tp->snd_una;
1898 				if (SEQ_LT(tp->snd_high, tp->snd_una))
1899 					tp->snd_high = tp->snd_una;
1900 				m_freem(m);
1901 
1902 				/*
1903 				 * If all outstanding data are acked, stop
1904 				 * retransmit timer, otherwise restart timer
1905 				 * using current (possibly backed-off) value.
1906 				 * If process is waiting for space,
1907 				 * wakeup/selnotify/signal.  If data
1908 				 * are ready to send, let tcp_output
1909 				 * decide between more output or persist.
1910 				 */
1911 				if (tp->snd_una == tp->snd_max)
1912 					TCP_TIMER_DISARM(tp, TCPT_REXMT);
1913 				else if (TCP_TIMER_ISARMED(tp,
1914 				    TCPT_PERSIST) == 0)
1915 					TCP_TIMER_ARM(tp, TCPT_REXMT,
1916 					    tp->t_rxtcur);
1917 
1918 				sowwakeup(so);
1919 				if (so->so_snd.sb_cc) {
1920 					KERNEL_LOCK(1, NULL);
1921 					(void)tcp_output(tp);
1922 					KERNEL_UNLOCK_ONE(NULL);
1923 				}
1924 				if (tcp_saveti)
1925 					m_freem(tcp_saveti);
1926 				return;
1927 			}
1928 		} else if (th->th_ack == tp->snd_una &&
1929 		    TAILQ_FIRST(&tp->segq) == NULL &&
1930 		    tlen <= sbspace(&so->so_rcv)) {
1931 			int newsize = 0;
1932 
1933 			/*
1934 			 * this is a pure, in-sequence data packet
1935 			 * with nothing on the reassembly queue and
1936 			 * we have enough buffer space to take it.
1937 			 */
1938 			tp->rcv_nxt += tlen;
1939 			tcps = TCP_STAT_GETREF();
1940 			tcps[TCP_STAT_PREDDAT]++;
1941 			tcps[TCP_STAT_RCVPACK]++;
1942 			tcps[TCP_STAT_RCVBYTE] += tlen;
1943 			TCP_STAT_PUTREF();
1944 			nd6_hint(tp);
1945 
1946 		/*
1947 		 * Automatic sizing enables the performance of large buffers
1948 		 * and most of the efficiency of small ones by only allocating
1949 		 * space when it is needed.
1950 		 *
1951 		 * On the receive side the socket buffer memory is only rarely
1952 		 * used to any significant extent.  This allows us to be much
1953 		 * more aggressive in scaling the receive socket buffer.  For
1954 		 * the case that the buffer space is actually used to a large
1955 		 * extent and we run out of kernel memory we can simply drop
1956 		 * the new segments; TCP on the sender will just retransmit it
1957 		 * later.  Setting the buffer size too big may only consume too
1958 		 * much kernel memory if the application doesn't read() from
1959 		 * the socket or packet loss or reordering makes use of the
1960 		 * reassembly queue.
1961 		 *
1962 		 * The criteria to step up the receive buffer one notch are:
1963 		 *  1. the number of bytes received during the time it takes
1964 		 *     one timestamp to be reflected back to us (the RTT);
1965 		 *  2. received bytes per RTT is within seven eighth of the
1966 		 *     current socket buffer size;
1967 		 *  3. receive buffer size has not hit maximal automatic size;
1968 		 *
1969 		 * This algorithm does one step per RTT at most and only if
1970 		 * we receive a bulk stream w/o packet losses or reorderings.
1971 		 * Shrinking the buffer during idle times is not necessary as
1972 		 * it doesn't consume any memory when idle.
1973 		 *
1974 		 * TODO: Only step up if the application is actually serving
1975 		 * the buffer to better manage the socket buffer resources.
1976 		 */
1977 			if (tcp_do_autorcvbuf &&
1978 			    opti.ts_ecr &&
1979 			    (so->so_rcv.sb_flags & SB_AUTOSIZE)) {
1980 				if (opti.ts_ecr > tp->rfbuf_ts &&
1981 				    opti.ts_ecr - tp->rfbuf_ts < PR_SLOWHZ) {
1982 					if (tp->rfbuf_cnt >
1983 					    (so->so_rcv.sb_hiwat / 8 * 7) &&
1984 					    so->so_rcv.sb_hiwat <
1985 					    tcp_autorcvbuf_max) {
1986 						newsize =
1987 						    uimin(so->so_rcv.sb_hiwat +
1988 						    tcp_autorcvbuf_inc,
1989 						    tcp_autorcvbuf_max);
1990 					}
1991 					/* Start over with next RTT. */
1992 					tp->rfbuf_ts = 0;
1993 					tp->rfbuf_cnt = 0;
1994 				} else
1995 					tp->rfbuf_cnt += tlen;	/* add up */
1996 			}
1997 
1998 			/*
1999 			 * Drop TCP, IP headers and TCP options then add data
2000 			 * to socket buffer.
2001 			 */
2002 			if (so->so_state & SS_CANTRCVMORE) {
2003 				m_freem(m);
2004 			} else {
2005 				/*
2006 				 * Set new socket buffer size.
2007 				 * Give up when limit is reached.
2008 				 */
2009 				if (newsize)
2010 					if (!sbreserve(&so->so_rcv,
2011 					    newsize, so))
2012 						so->so_rcv.sb_flags &= ~SB_AUTOSIZE;
2013 				m_adj(m, off + thlen);
2014 				sbappendstream(&so->so_rcv, m);
2015 			}
2016 			sorwakeup(so);
2017 			tcp_setup_ack(tp, th);
2018 			if (tp->t_flags & TF_ACKNOW) {
2019 				KERNEL_LOCK(1, NULL);
2020 				(void)tcp_output(tp);
2021 				KERNEL_UNLOCK_ONE(NULL);
2022 			}
2023 			if (tcp_saveti)
2024 				m_freem(tcp_saveti);
2025 			return;
2026 		}
2027 	}
2028 
2029 	/*
2030 	 * Compute mbuf offset to TCP data segment.
2031 	 */
2032 	hdroptlen = off + thlen;
2033 
2034 	/*
2035 	 * Calculate amount of space in receive window. Receive window is
2036 	 * amount of space in rcv queue, but not less than advertised
2037 	 * window.
2038 	 */
2039 	{
2040 		int win;
2041 		win = sbspace(&so->so_rcv);
2042 		if (win < 0)
2043 			win = 0;
2044 		tp->rcv_wnd = imax(win, (int)(tp->rcv_adv - tp->rcv_nxt));
2045 	}
2046 
2047 	/* Reset receive buffer auto scaling when not in bulk receive mode. */
2048 	tp->rfbuf_ts = 0;
2049 	tp->rfbuf_cnt = 0;
2050 
2051 	switch (tp->t_state) {
2052 	/*
2053 	 * If the state is SYN_SENT:
2054 	 *	if seg contains an ACK, but not for our SYN, drop the input.
2055 	 *	if seg contains a RST, then drop the connection.
2056 	 *	if seg does not contain SYN, then drop it.
2057 	 * Otherwise this is an acceptable SYN segment
2058 	 *	initialize tp->rcv_nxt and tp->irs
2059 	 *	if seg contains ack then advance tp->snd_una
2060 	 *	if seg contains a ECE and ECN support is enabled, the stream
2061 	 *	    is ECN capable.
2062 	 *	if SYN has been acked change to ESTABLISHED else SYN_RCVD state
2063 	 *	arrange for segment to be acked (eventually)
2064 	 *	continue processing rest of data/controls, beginning with URG
2065 	 */
2066 	case TCPS_SYN_SENT:
2067 		if ((tiflags & TH_ACK) &&
2068 		    (SEQ_LEQ(th->th_ack, tp->iss) ||
2069 		     SEQ_GT(th->th_ack, tp->snd_max)))
2070 			goto dropwithreset;
2071 		if (tiflags & TH_RST) {
2072 			if (tiflags & TH_ACK)
2073 				tp = tcp_drop(tp, ECONNREFUSED);
2074 			goto drop;
2075 		}
2076 		if ((tiflags & TH_SYN) == 0)
2077 			goto drop;
2078 		if (tiflags & TH_ACK) {
2079 			tp->snd_una = th->th_ack;
2080 			if (SEQ_LT(tp->snd_nxt, tp->snd_una))
2081 				tp->snd_nxt = tp->snd_una;
2082 			if (SEQ_LT(tp->snd_high, tp->snd_una))
2083 				tp->snd_high = tp->snd_una;
2084 			TCP_TIMER_DISARM(tp, TCPT_REXMT);
2085 
2086 			if ((tiflags & TH_ECE) && tcp_do_ecn) {
2087 				tp->t_flags |= TF_ECN_PERMIT;
2088 				TCP_STATINC(TCP_STAT_ECN_SHS);
2089 			}
2090 		}
2091 		tp->irs = th->th_seq;
2092 		tcp_rcvseqinit(tp);
2093 		tp->t_flags |= TF_ACKNOW;
2094 		tcp_mss_from_peer(tp, opti.maxseg);
2095 
2096 		/*
2097 		 * Initialize the initial congestion window.  If we
2098 		 * had to retransmit the SYN, we must initialize cwnd
2099 		 * to 1 segment (i.e. the Loss Window).
2100 		 */
2101 		if (tp->t_flags & TF_SYN_REXMT)
2102 			tp->snd_cwnd = tp->t_peermss;
2103 		else {
2104 			int ss = tcp_init_win;
2105 			if (inp != NULL && in_localaddr(inp->inp_faddr))
2106 				ss = tcp_init_win_local;
2107 #ifdef INET6
2108 			if (in6p != NULL && in6_localaddr(&in6p->in6p_faddr))
2109 				ss = tcp_init_win_local;
2110 #endif
2111 			tp->snd_cwnd = TCP_INITIAL_WINDOW(ss, tp->t_peermss);
2112 		}
2113 
2114 		tcp_rmx_rtt(tp);
2115 		if (tiflags & TH_ACK) {
2116 			TCP_STATINC(TCP_STAT_CONNECTS);
2117 			/*
2118 			 * move tcp_established before soisconnected
2119 			 * because upcall handler can drive tcp_output
2120 			 * functionality.
2121 			 * XXX we might call soisconnected at the end of
2122 			 * all processing
2123 			 */
2124 			tcp_established(tp);
2125 			soisconnected(so);
2126 			/* Do window scaling on this connection? */
2127 			if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) ==
2128 			    (TF_RCVD_SCALE|TF_REQ_SCALE)) {
2129 				tp->snd_scale = tp->requested_s_scale;
2130 				tp->rcv_scale = tp->request_r_scale;
2131 			}
2132 			TCP_REASS_LOCK(tp);
2133 			(void)tcp_reass(tp, NULL, NULL, tlen);
2134 			/*
2135 			 * if we didn't have to retransmit the SYN,
2136 			 * use its rtt as our initial srtt & rtt var.
2137 			 */
2138 			if (tp->t_rtttime)
2139 				tcp_xmit_timer(tp, tcp_now - tp->t_rtttime);
2140 		} else {
2141 			tp->t_state = TCPS_SYN_RECEIVED;
2142 		}
2143 
2144 		/*
2145 		 * Advance th->th_seq to correspond to first data byte.
2146 		 * If data, trim to stay within window,
2147 		 * dropping FIN if necessary.
2148 		 */
2149 		th->th_seq++;
2150 		if (tlen > tp->rcv_wnd) {
2151 			todrop = tlen - tp->rcv_wnd;
2152 			m_adj(m, -todrop);
2153 			tlen = tp->rcv_wnd;
2154 			tiflags &= ~TH_FIN;
2155 			tcps = TCP_STAT_GETREF();
2156 			tcps[TCP_STAT_RCVPACKAFTERWIN]++;
2157 			tcps[TCP_STAT_RCVBYTEAFTERWIN] += todrop;
2158 			TCP_STAT_PUTREF();
2159 		}
2160 		tp->snd_wl1 = th->th_seq - 1;
2161 		tp->rcv_up = th->th_seq;
2162 		goto step6;
2163 
2164 	/*
2165 	 * If the state is SYN_RECEIVED:
2166 	 *	If seg contains an ACK, but not for our SYN, drop the input
2167 	 *	and generate an RST.  See page 36, rfc793
2168 	 */
2169 	case TCPS_SYN_RECEIVED:
2170 		if ((tiflags & TH_ACK) &&
2171 		    (SEQ_LEQ(th->th_ack, tp->iss) ||
2172 		     SEQ_GT(th->th_ack, tp->snd_max)))
2173 			goto dropwithreset;
2174 		break;
2175 	}
2176 
2177 	/*
2178 	 * From here on, we're dealing with !LISTEN and !SYN_SENT.
2179 	 */
2180 	KASSERT(tp->t_state != TCPS_LISTEN &&
2181 	    tp->t_state != TCPS_SYN_SENT);
2182 
2183 	/*
2184 	 * RFC1323 PAWS: if we have a timestamp reply on this segment and
2185 	 * it's less than ts_recent, drop it.
2186 	 */
2187 	if (opti.ts_present && (tiflags & TH_RST) == 0 && tp->ts_recent &&
2188 	    TSTMP_LT(opti.ts_val, tp->ts_recent)) {
2189 		/* Check to see if ts_recent is over 24 days old.  */
2190 		if (tcp_now - tp->ts_recent_age > TCP_PAWS_IDLE) {
2191 			/*
2192 			 * Invalidate ts_recent.  If this segment updates
2193 			 * ts_recent, the age will be reset later and ts_recent
2194 			 * will get a valid value.  If it does not, setting
2195 			 * ts_recent to zero will at least satisfy the
2196 			 * requirement that zero be placed in the timestamp
2197 			 * echo reply when ts_recent isn't valid.  The
2198 			 * age isn't reset until we get a valid ts_recent
2199 			 * because we don't want out-of-order segments to be
2200 			 * dropped when ts_recent is old.
2201 			 */
2202 			tp->ts_recent = 0;
2203 		} else {
2204 			tcps = TCP_STAT_GETREF();
2205 			tcps[TCP_STAT_RCVDUPPACK]++;
2206 			tcps[TCP_STAT_RCVDUPBYTE] += tlen;
2207 			tcps[TCP_STAT_PAWSDROP]++;
2208 			TCP_STAT_PUTREF();
2209 			tcp_new_dsack(tp, th->th_seq, tlen);
2210 			goto dropafterack;
2211 		}
2212 	}
2213 
2214 	/*
2215 	 * Check that at least some bytes of the segment are within the
2216 	 * receive window. If segment begins before rcv_nxt, drop leading
2217 	 * data (and SYN); if nothing left, just ack.
2218 	 */
2219 	todrop = tp->rcv_nxt - th->th_seq;
2220 	dupseg = false;
2221 	if (todrop > 0) {
2222 		if (tiflags & TH_SYN) {
2223 			tiflags &= ~TH_SYN;
2224 			th->th_seq++;
2225 			tcp_urp_drop(th, 1, &tiflags);
2226 			todrop--;
2227 		}
2228 		if (todrop > tlen ||
2229 		    (todrop == tlen && (tiflags & TH_FIN) == 0)) {
2230 			/*
2231 			 * Any valid FIN or RST must be to the left of the
2232 			 * window.  At this point the FIN or RST must be a
2233 			 * duplicate or out of sequence; drop it.
2234 			 */
2235 			if (tiflags & TH_RST)
2236 				goto drop;
2237 			tiflags &= ~(TH_FIN|TH_RST);
2238 
2239 			/*
2240 			 * Send an ACK to resynchronize and drop any data.
2241 			 * But keep on processing for RST or ACK.
2242 			 */
2243 			tp->t_flags |= TF_ACKNOW;
2244 			todrop = tlen;
2245 			dupseg = true;
2246 			tcps = TCP_STAT_GETREF();
2247 			tcps[TCP_STAT_RCVDUPPACK]++;
2248 			tcps[TCP_STAT_RCVDUPBYTE] += todrop;
2249 			TCP_STAT_PUTREF();
2250 		} else if ((tiflags & TH_RST) && th->th_seq != tp->rcv_nxt) {
2251 			/*
2252 			 * Test for reset before adjusting the sequence
2253 			 * number for overlapping data.
2254 			 */
2255 			goto dropafterack_ratelim;
2256 		} else {
2257 			tcps = TCP_STAT_GETREF();
2258 			tcps[TCP_STAT_RCVPARTDUPPACK]++;
2259 			tcps[TCP_STAT_RCVPARTDUPBYTE] += todrop;
2260 			TCP_STAT_PUTREF();
2261 		}
2262 		tcp_new_dsack(tp, th->th_seq, todrop);
2263 		hdroptlen += todrop;	/* drop from head afterwards (m_adj) */
2264 		th->th_seq += todrop;
2265 		tlen -= todrop;
2266 		tcp_urp_drop(th, todrop, &tiflags);
2267 	}
2268 
2269 	/*
2270 	 * If new data is received on a connection after the user processes
2271 	 * are gone, then RST the other end.
2272 	 */
2273 	if ((so->so_state & SS_NOFDREF) &&
2274 	    tp->t_state > TCPS_CLOSE_WAIT && tlen) {
2275 		tp = tcp_close(tp);
2276 		TCP_STATINC(TCP_STAT_RCVAFTERCLOSE);
2277 		goto dropwithreset;
2278 	}
2279 
2280 	/*
2281 	 * If the segment ends after the window, drop trailing data (and
2282 	 * PUSH and FIN); if nothing left, just ACK.
2283 	 */
2284 	todrop = (th->th_seq + tlen) - (tp->rcv_nxt + tp->rcv_wnd);
2285 	if (todrop > 0) {
2286 		TCP_STATINC(TCP_STAT_RCVPACKAFTERWIN);
2287 		if (todrop >= tlen) {
2288 			/*
2289 			 * The segment actually starts after the window.
2290 			 * th->th_seq + tlen - tp->rcv_nxt - tp->rcv_wnd >= tlen
2291 			 * th->th_seq - tp->rcv_nxt - tp->rcv_wnd >= 0
2292 			 * th->th_seq >= tp->rcv_nxt + tp->rcv_wnd
2293 			 */
2294 			TCP_STATADD(TCP_STAT_RCVBYTEAFTERWIN, tlen);
2295 
2296 			/*
2297 			 * If a new connection request is received while in
2298 			 * TIME_WAIT, drop the old connection and start over
2299 			 * if the sequence numbers are above the previous
2300 			 * ones.
2301 			 *
2302 			 * NOTE: We need to put the header fields back into
2303 			 * network order.
2304 			 */
2305 			if ((tiflags & TH_SYN) &&
2306 			    tp->t_state == TCPS_TIME_WAIT &&
2307 			    SEQ_GT(th->th_seq, tp->rcv_nxt)) {
2308 				tp = tcp_close(tp);
2309 				tcp_fields_to_net(th);
2310 				m_freem(tcp_saveti);
2311 				tcp_saveti = NULL;
2312 				goto findpcb;
2313 			}
2314 
2315 			/*
2316 			 * If window is closed can only take segments at
2317 			 * window edge, and have to drop data and PUSH from
2318 			 * incoming segments.  Continue processing, but
2319 			 * remember to ack.  Otherwise, drop segment
2320 			 * and (if not RST) ack.
2321 			 */
2322 			if (tp->rcv_wnd == 0 && th->th_seq == tp->rcv_nxt) {
2323 				KASSERT(todrop == tlen);
2324 				tp->t_flags |= TF_ACKNOW;
2325 				TCP_STATINC(TCP_STAT_RCVWINPROBE);
2326 			} else {
2327 				goto dropafterack;
2328 			}
2329 		} else {
2330 			TCP_STATADD(TCP_STAT_RCVBYTEAFTERWIN, todrop);
2331 		}
2332 		m_adj(m, -todrop);
2333 		tlen -= todrop;
2334 		tiflags &= ~(TH_PUSH|TH_FIN);
2335 	}
2336 
2337 	/*
2338 	 * If last ACK falls within this segment's sequence numbers,
2339 	 *  record the timestamp.
2340 	 * NOTE:
2341 	 * 1) That the test incorporates suggestions from the latest
2342 	 *    proposal of the tcplw@cray.com list (Braden 1993/04/26).
2343 	 * 2) That updating only on newer timestamps interferes with
2344 	 *    our earlier PAWS tests, so this check should be solely
2345 	 *    predicated on the sequence space of this segment.
2346 	 * 3) That we modify the segment boundary check to be
2347 	 *        Last.ACK.Sent <= SEG.SEQ + SEG.Len
2348 	 *    instead of RFC1323's
2349 	 *        Last.ACK.Sent < SEG.SEQ + SEG.Len,
2350 	 *    This modified check allows us to overcome RFC1323's
2351 	 *    limitations as described in Stevens TCP/IP Illustrated
2352 	 *    Vol. 2 p.869. In such cases, we can still calculate the
2353 	 *    RTT correctly when RCV.NXT == Last.ACK.Sent.
2354 	 */
2355 	if (opti.ts_present &&
2356 	    SEQ_LEQ(th->th_seq, tp->last_ack_sent) &&
2357 	    SEQ_LEQ(tp->last_ack_sent, th->th_seq + tlen +
2358 	         ((tiflags & (TH_SYN|TH_FIN)) != 0))) {
2359 		tp->ts_recent_age = tcp_now;
2360 		tp->ts_recent = opti.ts_val;
2361 	}
2362 
2363 	/*
2364 	 * If the RST bit is set examine the state:
2365 	 *    RECEIVED state:
2366 	 *        If passive open, return to LISTEN state.
2367 	 *        If active open, inform user that connection was refused.
2368 	 *    ESTABLISHED, FIN_WAIT_1, FIN_WAIT2, CLOSE_WAIT states:
2369 	 *        Inform user that connection was reset, and close tcb.
2370 	 *    CLOSING, LAST_ACK, TIME_WAIT states:
2371 	 *        Close the tcb.
2372 	 */
2373 	if (tiflags & TH_RST) {
2374 		if (th->th_seq != tp->rcv_nxt)
2375 			goto dropafterack_ratelim;
2376 
2377 		switch (tp->t_state) {
2378 		case TCPS_SYN_RECEIVED:
2379 			so->so_error = ECONNREFUSED;
2380 			goto close;
2381 
2382 		case TCPS_ESTABLISHED:
2383 		case TCPS_FIN_WAIT_1:
2384 		case TCPS_FIN_WAIT_2:
2385 		case TCPS_CLOSE_WAIT:
2386 			so->so_error = ECONNRESET;
2387 		close:
2388 			tp->t_state = TCPS_CLOSED;
2389 			TCP_STATINC(TCP_STAT_DROPS);
2390 			tp = tcp_close(tp);
2391 			goto drop;
2392 
2393 		case TCPS_CLOSING:
2394 		case TCPS_LAST_ACK:
2395 		case TCPS_TIME_WAIT:
2396 			tp = tcp_close(tp);
2397 			goto drop;
2398 		}
2399 	}
2400 
2401 	/*
2402 	 * Since we've covered the SYN-SENT and SYN-RECEIVED states above
2403 	 * we must be in a synchronized state.  RFC793 states (under Reset
2404 	 * Generation) that any unacceptable segment (an out-of-order SYN
2405 	 * qualifies) received in a synchronized state must elicit only an
2406 	 * empty acknowledgment segment ... and the connection remains in
2407 	 * the same state.
2408 	 */
2409 	if (tiflags & TH_SYN) {
2410 		if (tp->rcv_nxt == th->th_seq) {
2411 			tcp_respond(tp, m, m, th, (tcp_seq)0, th->th_ack - 1,
2412 			    TH_ACK);
2413 			if (tcp_saveti)
2414 				m_freem(tcp_saveti);
2415 			return;
2416 		}
2417 
2418 		goto dropafterack_ratelim;
2419 	}
2420 
2421 	/*
2422 	 * If the ACK bit is off we drop the segment and return.
2423 	 */
2424 	if ((tiflags & TH_ACK) == 0) {
2425 		if (tp->t_flags & TF_ACKNOW)
2426 			goto dropafterack;
2427 		goto drop;
2428 	}
2429 
2430 	/*
2431 	 * From here on, we're doing ACK processing.
2432 	 */
2433 
2434 	switch (tp->t_state) {
2435 	/*
2436 	 * In SYN_RECEIVED state if the ack ACKs our SYN then enter
2437 	 * ESTABLISHED state and continue processing, otherwise
2438 	 * send an RST.
2439 	 */
2440 	case TCPS_SYN_RECEIVED:
2441 		if (SEQ_GT(tp->snd_una, th->th_ack) ||
2442 		    SEQ_GT(th->th_ack, tp->snd_max))
2443 			goto dropwithreset;
2444 		TCP_STATINC(TCP_STAT_CONNECTS);
2445 		soisconnected(so);
2446 		tcp_established(tp);
2447 		/* Do window scaling? */
2448 		if ((tp->t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) ==
2449 		    (TF_RCVD_SCALE|TF_REQ_SCALE)) {
2450 			tp->snd_scale = tp->requested_s_scale;
2451 			tp->rcv_scale = tp->request_r_scale;
2452 		}
2453 		TCP_REASS_LOCK(tp);
2454 		(void)tcp_reass(tp, NULL, NULL, tlen);
2455 		tp->snd_wl1 = th->th_seq - 1;
2456 		/* FALLTHROUGH */
2457 
2458 	/*
2459 	 * In ESTABLISHED state: drop duplicate ACKs; ACK out of range
2460 	 * ACKs.  If the ack is in the range
2461 	 *	tp->snd_una < th->th_ack <= tp->snd_max
2462 	 * then advance tp->snd_una to th->th_ack and drop
2463 	 * data from the retransmission queue.  If this ACK reflects
2464 	 * more up to date window information we update our window information.
2465 	 */
2466 	case TCPS_ESTABLISHED:
2467 	case TCPS_FIN_WAIT_1:
2468 	case TCPS_FIN_WAIT_2:
2469 	case TCPS_CLOSE_WAIT:
2470 	case TCPS_CLOSING:
2471 	case TCPS_LAST_ACK:
2472 	case TCPS_TIME_WAIT:
2473 		if (SEQ_LEQ(th->th_ack, tp->snd_una)) {
2474 			if (tlen == 0 && !dupseg && tiwin == tp->snd_wnd) {
2475 				TCP_STATINC(TCP_STAT_RCVDUPACK);
2476 				/*
2477 				 * If we have outstanding data (other than
2478 				 * a window probe), this is a completely
2479 				 * duplicate ack (ie, window info didn't
2480 				 * change), the ack is the biggest we've
2481 				 * seen and we've seen exactly our rexmt
2482 				 * threshhold of them, assume a packet
2483 				 * has been dropped and retransmit it.
2484 				 * Kludge snd_nxt & the congestion
2485 				 * window so we send only this one
2486 				 * packet.
2487 				 */
2488 				if (TCP_TIMER_ISARMED(tp, TCPT_REXMT) == 0 ||
2489 				    th->th_ack != tp->snd_una)
2490 					tp->t_dupacks = 0;
2491 				else if (tp->t_partialacks < 0 &&
2492 				    (++tp->t_dupacks == tcprexmtthresh ||
2493 				     TCP_FACK_FASTRECOV(tp))) {
2494 					/*
2495 					 * Do the fast retransmit, and adjust
2496 					 * congestion control paramenters.
2497 					 */
2498 					if (tp->t_congctl->fast_retransmit(tp, th)) {
2499 						/* False fast retransmit */
2500 						break;
2501 					}
2502 					goto drop;
2503 				} else if (tp->t_dupacks > tcprexmtthresh) {
2504 					tp->snd_cwnd += tp->t_segsz;
2505 					KERNEL_LOCK(1, NULL);
2506 					(void)tcp_output(tp);
2507 					KERNEL_UNLOCK_ONE(NULL);
2508 					goto drop;
2509 				}
2510 			} else {
2511 				/*
2512 				 * If the ack appears to be very old, only
2513 				 * allow data that is in-sequence.  This
2514 				 * makes it somewhat more difficult to insert
2515 				 * forged data by guessing sequence numbers.
2516 				 * Sent an ack to try to update the send
2517 				 * sequence number on the other side.
2518 				 */
2519 				if (tlen && th->th_seq != tp->rcv_nxt &&
2520 				    SEQ_LT(th->th_ack,
2521 				    tp->snd_una - tp->max_sndwnd))
2522 					goto dropafterack;
2523 			}
2524 			break;
2525 		}
2526 		/*
2527 		 * If the congestion window was inflated to account
2528 		 * for the other side's cached packets, retract it.
2529 		 */
2530 		tp->t_congctl->fast_retransmit_newack(tp, th);
2531 
2532 		if (SEQ_GT(th->th_ack, tp->snd_max)) {
2533 			TCP_STATINC(TCP_STAT_RCVACKTOOMUCH);
2534 			goto dropafterack;
2535 		}
2536 		acked = th->th_ack - tp->snd_una;
2537 		tcps = TCP_STAT_GETREF();
2538 		tcps[TCP_STAT_RCVACKPACK]++;
2539 		tcps[TCP_STAT_RCVACKBYTE] += acked;
2540 		TCP_STAT_PUTREF();
2541 
2542 		/*
2543 		 * If we have a timestamp reply, update smoothed
2544 		 * round trip time.  If no timestamp is present but
2545 		 * transmit timer is running and timed sequence
2546 		 * number was acked, update smoothed round trip time.
2547 		 * Since we now have an rtt measurement, cancel the
2548 		 * timer backoff (cf., Phil Karn's retransmit alg.).
2549 		 * Recompute the initial retransmit timer.
2550 		 */
2551 		if (ts_rtt)
2552 			tcp_xmit_timer(tp, ts_rtt - 1);
2553 		else if (tp->t_rtttime && SEQ_GT(th->th_ack, tp->t_rtseq))
2554 			tcp_xmit_timer(tp, tcp_now - tp->t_rtttime);
2555 
2556 		/*
2557 		 * If all outstanding data is acked, stop retransmit
2558 		 * timer and remember to restart (more output or persist).
2559 		 * If there is more data to be acked, restart retransmit
2560 		 * timer, using current (possibly backed-off) value.
2561 		 */
2562 		if (th->th_ack == tp->snd_max) {
2563 			TCP_TIMER_DISARM(tp, TCPT_REXMT);
2564 			needoutput = 1;
2565 		} else if (TCP_TIMER_ISARMED(tp, TCPT_PERSIST) == 0)
2566 			TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur);
2567 
2568 		/*
2569 		 * New data has been acked, adjust the congestion window.
2570 		 */
2571 		tp->t_congctl->newack(tp, th);
2572 
2573 		nd6_hint(tp);
2574 		if (acked > so->so_snd.sb_cc) {
2575 			tp->snd_wnd -= so->so_snd.sb_cc;
2576 			sbdrop(&so->so_snd, (int)so->so_snd.sb_cc);
2577 			ourfinisacked = 1;
2578 		} else {
2579 			if (acked > (tp->t_lastoff - tp->t_inoff))
2580 				tp->t_lastm = NULL;
2581 			sbdrop(&so->so_snd, acked);
2582 			tp->t_lastoff -= acked;
2583 			if (tp->snd_wnd > acked)
2584 				tp->snd_wnd -= acked;
2585 			else
2586 				tp->snd_wnd = 0;
2587 			ourfinisacked = 0;
2588 		}
2589 		sowwakeup(so);
2590 
2591 		icmp_check(tp, th, acked);
2592 
2593 		tp->snd_una = th->th_ack;
2594 		if (SEQ_GT(tp->snd_una, tp->snd_fack))
2595 			tp->snd_fack = tp->snd_una;
2596 		if (SEQ_LT(tp->snd_nxt, tp->snd_una))
2597 			tp->snd_nxt = tp->snd_una;
2598 		if (SEQ_LT(tp->snd_high, tp->snd_una))
2599 			tp->snd_high = tp->snd_una;
2600 
2601 		switch (tp->t_state) {
2602 
2603 		/*
2604 		 * In FIN_WAIT_1 STATE in addition to the processing
2605 		 * for the ESTABLISHED state if our FIN is now acknowledged
2606 		 * then enter FIN_WAIT_2.
2607 		 */
2608 		case TCPS_FIN_WAIT_1:
2609 			if (ourfinisacked) {
2610 				/*
2611 				 * If we can't receive any more
2612 				 * data, then closing user can proceed.
2613 				 * Starting the timer is contrary to the
2614 				 * specification, but if we don't get a FIN
2615 				 * we'll hang forever.
2616 				 */
2617 				if (so->so_state & SS_CANTRCVMORE) {
2618 					soisdisconnected(so);
2619 					if (tp->t_maxidle > 0)
2620 						TCP_TIMER_ARM(tp, TCPT_2MSL,
2621 						    tp->t_maxidle);
2622 				}
2623 				tp->t_state = TCPS_FIN_WAIT_2;
2624 			}
2625 			break;
2626 
2627 	 	/*
2628 		 * In CLOSING STATE in addition to the processing for
2629 		 * the ESTABLISHED state if the ACK acknowledges our FIN
2630 		 * then enter the TIME-WAIT state, otherwise ignore
2631 		 * the segment.
2632 		 */
2633 		case TCPS_CLOSING:
2634 			if (ourfinisacked) {
2635 				tp->t_state = TCPS_TIME_WAIT;
2636 				tcp_canceltimers(tp);
2637 				TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * tp->t_msl);
2638 				soisdisconnected(so);
2639 			}
2640 			break;
2641 
2642 		/*
2643 		 * In LAST_ACK, we may still be waiting for data to drain
2644 		 * and/or to be acked, as well as for the ack of our FIN.
2645 		 * If our FIN is now acknowledged, delete the TCB,
2646 		 * enter the closed state and return.
2647 		 */
2648 		case TCPS_LAST_ACK:
2649 			if (ourfinisacked) {
2650 				tp = tcp_close(tp);
2651 				goto drop;
2652 			}
2653 			break;
2654 
2655 		/*
2656 		 * In TIME_WAIT state the only thing that should arrive
2657 		 * is a retransmission of the remote FIN.  Acknowledge
2658 		 * it and restart the finack timer.
2659 		 */
2660 		case TCPS_TIME_WAIT:
2661 			TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * tp->t_msl);
2662 			goto dropafterack;
2663 		}
2664 	}
2665 
2666 step6:
2667 	/*
2668 	 * Update window information.
2669 	 * Don't look at window if no ACK: TAC's send garbage on first SYN.
2670 	 */
2671 	if ((tiflags & TH_ACK) && (SEQ_LT(tp->snd_wl1, th->th_seq) ||
2672 	    (tp->snd_wl1 == th->th_seq && (SEQ_LT(tp->snd_wl2, th->th_ack) ||
2673 	    (tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd))))) {
2674 		/* keep track of pure window updates */
2675 		if (tlen == 0 &&
2676 		    tp->snd_wl2 == th->th_ack && tiwin > tp->snd_wnd)
2677 			TCP_STATINC(TCP_STAT_RCVWINUPD);
2678 		tp->snd_wnd = tiwin;
2679 		tp->snd_wl1 = th->th_seq;
2680 		tp->snd_wl2 = th->th_ack;
2681 		if (tp->snd_wnd > tp->max_sndwnd)
2682 			tp->max_sndwnd = tp->snd_wnd;
2683 		needoutput = 1;
2684 	}
2685 
2686 	/*
2687 	 * Process segments with URG.
2688 	 */
2689 	if ((tiflags & TH_URG) && th->th_urp &&
2690 	    TCPS_HAVERCVDFIN(tp->t_state) == 0) {
2691 		/*
2692 		 * This is a kludge, but if we receive and accept
2693 		 * random urgent pointers, we'll crash in
2694 		 * soreceive.  It's hard to imagine someone
2695 		 * actually wanting to send this much urgent data.
2696 		 */
2697 		if (th->th_urp + so->so_rcv.sb_cc > sb_max) {
2698 			th->th_urp = 0;			/* XXX */
2699 			tiflags &= ~TH_URG;		/* XXX */
2700 			goto dodata;			/* XXX */
2701 		}
2702 
2703 		/*
2704 		 * If this segment advances the known urgent pointer,
2705 		 * then mark the data stream.  This should not happen
2706 		 * in CLOSE_WAIT, CLOSING, LAST_ACK or TIME_WAIT STATES since
2707 		 * a FIN has been received from the remote side.
2708 		 * In these states we ignore the URG.
2709 		 *
2710 		 * According to RFC961 (Assigned Protocols),
2711 		 * the urgent pointer points to the last octet
2712 		 * of urgent data.  We continue, however,
2713 		 * to consider it to indicate the first octet
2714 		 * of data past the urgent section as the original
2715 		 * spec states (in one of two places).
2716 		 */
2717 		if (SEQ_GT(th->th_seq+th->th_urp, tp->rcv_up)) {
2718 			tp->rcv_up = th->th_seq + th->th_urp;
2719 			so->so_oobmark = so->so_rcv.sb_cc +
2720 			    (tp->rcv_up - tp->rcv_nxt) - 1;
2721 			if (so->so_oobmark == 0)
2722 				so->so_state |= SS_RCVATMARK;
2723 			sohasoutofband(so);
2724 			tp->t_oobflags &= ~(TCPOOB_HAVEDATA | TCPOOB_HADDATA);
2725 		}
2726 
2727 		/*
2728 		 * Remove out of band data so doesn't get presented to user.
2729 		 * This can happen independent of advancing the URG pointer,
2730 		 * but if two URG's are pending at once, some out-of-band
2731 		 * data may creep in... ick.
2732 		 */
2733 		if (th->th_urp <= (u_int16_t)tlen &&
2734 		    (so->so_options & SO_OOBINLINE) == 0)
2735 			tcp_pulloutofband(so, th, m, hdroptlen);
2736 	} else {
2737 		/*
2738 		 * If no out of band data is expected,
2739 		 * pull receive urgent pointer along
2740 		 * with the receive window.
2741 		 */
2742 		if (SEQ_GT(tp->rcv_nxt, tp->rcv_up))
2743 			tp->rcv_up = tp->rcv_nxt;
2744 	}
2745 dodata:
2746 
2747 	/*
2748 	 * Process the segment text, merging it into the TCP sequencing queue,
2749 	 * and arranging for acknowledgement of receipt if necessary.
2750 	 * This process logically involves adjusting tp->rcv_wnd as data
2751 	 * is presented to the user (this happens in tcp_usrreq.c,
2752 	 * tcp_rcvd()).  If a FIN has already been received on this
2753 	 * connection then we just ignore the text.
2754 	 */
2755 	if ((tlen || (tiflags & TH_FIN)) &&
2756 	    TCPS_HAVERCVDFIN(tp->t_state) == 0) {
2757 		/*
2758 		 * Handle the common case:
2759 		 *  o Segment is the next to be received, and
2760 		 *  o The queue is empty, and
2761 		 *  o The connection is established
2762 		 * In this case, we avoid calling tcp_reass.
2763 		 *
2764 		 * tcp_setup_ack: set DELACK for segments received in order,
2765 		 * but ack immediately when segments are out of order (so that
2766 		 * fast retransmit can work).
2767 		 */
2768 		TCP_REASS_LOCK(tp);
2769 		if (th->th_seq == tp->rcv_nxt &&
2770 		    TAILQ_FIRST(&tp->segq) == NULL &&
2771 		    tp->t_state == TCPS_ESTABLISHED) {
2772 			tcp_setup_ack(tp, th);
2773 			tp->rcv_nxt += tlen;
2774 			tiflags = th->th_flags & TH_FIN;
2775 			tcps = TCP_STAT_GETREF();
2776 			tcps[TCP_STAT_RCVPACK]++;
2777 			tcps[TCP_STAT_RCVBYTE] += tlen;
2778 			TCP_STAT_PUTREF();
2779 			nd6_hint(tp);
2780 			if (so->so_state & SS_CANTRCVMORE) {
2781 				m_freem(m);
2782 			} else {
2783 				m_adj(m, hdroptlen);
2784 				sbappendstream(&(so)->so_rcv, m);
2785 			}
2786 			TCP_REASS_UNLOCK(tp);
2787 			sorwakeup(so);
2788 		} else {
2789 			m_adj(m, hdroptlen);
2790 			tiflags = tcp_reass(tp, th, m, tlen);
2791 			tp->t_flags |= TF_ACKNOW;
2792 		}
2793 
2794 		/*
2795 		 * Note the amount of data that peer has sent into
2796 		 * our window, in order to estimate the sender's
2797 		 * buffer size.
2798 		 */
2799 		len = so->so_rcv.sb_hiwat - (tp->rcv_adv - tp->rcv_nxt);
2800 	} else {
2801 		m_freem(m);
2802 		m = NULL;
2803 		tiflags &= ~TH_FIN;
2804 	}
2805 
2806 	/*
2807 	 * If FIN is received ACK the FIN and let the user know
2808 	 * that the connection is closing.  Ignore a FIN received before
2809 	 * the connection is fully established.
2810 	 */
2811 	if ((tiflags & TH_FIN) && TCPS_HAVEESTABLISHED(tp->t_state)) {
2812 		if (TCPS_HAVERCVDFIN(tp->t_state) == 0) {
2813 			socantrcvmore(so);
2814 			tp->t_flags |= TF_ACKNOW;
2815 			tp->rcv_nxt++;
2816 		}
2817 		switch (tp->t_state) {
2818 
2819 	 	/*
2820 		 * In ESTABLISHED STATE enter the CLOSE_WAIT state.
2821 		 */
2822 		case TCPS_ESTABLISHED:
2823 			tp->t_state = TCPS_CLOSE_WAIT;
2824 			break;
2825 
2826 	 	/*
2827 		 * If still in FIN_WAIT_1 STATE FIN has not been acked so
2828 		 * enter the CLOSING state.
2829 		 */
2830 		case TCPS_FIN_WAIT_1:
2831 			tp->t_state = TCPS_CLOSING;
2832 			break;
2833 
2834 	 	/*
2835 		 * In FIN_WAIT_2 state enter the TIME_WAIT state,
2836 		 * starting the time-wait timer, turning off the other
2837 		 * standard timers.
2838 		 */
2839 		case TCPS_FIN_WAIT_2:
2840 			tp->t_state = TCPS_TIME_WAIT;
2841 			tcp_canceltimers(tp);
2842 			TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * tp->t_msl);
2843 			soisdisconnected(so);
2844 			break;
2845 
2846 		/*
2847 		 * In TIME_WAIT state restart the 2 MSL time_wait timer.
2848 		 */
2849 		case TCPS_TIME_WAIT:
2850 			TCP_TIMER_ARM(tp, TCPT_2MSL, 2 * tp->t_msl);
2851 			break;
2852 		}
2853 	}
2854 #ifdef TCP_DEBUG
2855 	if (so->so_options & SO_DEBUG)
2856 		tcp_trace(TA_INPUT, ostate, tp, tcp_saveti, 0);
2857 #endif
2858 
2859 	/*
2860 	 * Return any desired output.
2861 	 */
2862 	if (needoutput || (tp->t_flags & TF_ACKNOW)) {
2863 		KERNEL_LOCK(1, NULL);
2864 		(void)tcp_output(tp);
2865 		KERNEL_UNLOCK_ONE(NULL);
2866 	}
2867 	if (tcp_saveti)
2868 		m_freem(tcp_saveti);
2869 
2870 	if (tp->t_state == TCPS_TIME_WAIT
2871 	    && (so->so_state & SS_NOFDREF)
2872 	    && (tp->t_inpcb || af != AF_INET)
2873 	    && (tp->t_in6pcb || af != AF_INET6)
2874 	    && ((af == AF_INET ? tcp4_vtw_enable : tcp6_vtw_enable) & 1) != 0
2875 	    && TAILQ_EMPTY(&tp->segq)
2876 	    && vtw_add(af, tp)) {
2877 		;
2878 	}
2879 	return;
2880 
2881 badsyn:
2882 	/*
2883 	 * Received a bad SYN.  Increment counters and dropwithreset.
2884 	 */
2885 	TCP_STATINC(TCP_STAT_BADSYN);
2886 	tp = NULL;
2887 	goto dropwithreset;
2888 
2889 dropafterack:
2890 	/*
2891 	 * Generate an ACK dropping incoming segment if it occupies
2892 	 * sequence space, where the ACK reflects our state.
2893 	 */
2894 	if (tiflags & TH_RST)
2895 		goto drop;
2896 	goto dropafterack2;
2897 
2898 dropafterack_ratelim:
2899 	/*
2900 	 * We may want to rate-limit ACKs against SYN/RST attack.
2901 	 */
2902 	if (ppsratecheck(&tcp_ackdrop_ppslim_last, &tcp_ackdrop_ppslim_count,
2903 	    tcp_ackdrop_ppslim) == 0) {
2904 		/* XXX stat */
2905 		goto drop;
2906 	}
2907 
2908 dropafterack2:
2909 	m_freem(m);
2910 	tp->t_flags |= TF_ACKNOW;
2911 	KERNEL_LOCK(1, NULL);
2912 	(void)tcp_output(tp);
2913 	KERNEL_UNLOCK_ONE(NULL);
2914 	if (tcp_saveti)
2915 		m_freem(tcp_saveti);
2916 	return;
2917 
2918 dropwithreset_ratelim:
2919 	/*
2920 	 * We may want to rate-limit RSTs in certain situations,
2921 	 * particularly if we are sending an RST in response to
2922 	 * an attempt to connect to or otherwise communicate with
2923 	 * a port for which we have no socket.
2924 	 */
2925 	if (ppsratecheck(&tcp_rst_ppslim_last, &tcp_rst_ppslim_count,
2926 	    tcp_rst_ppslim) == 0) {
2927 		/* XXX stat */
2928 		goto drop;
2929 	}
2930 
2931 dropwithreset:
2932 	/*
2933 	 * Generate a RST, dropping incoming segment.
2934 	 * Make ACK acceptable to originator of segment.
2935 	 */
2936 	if (tiflags & TH_RST)
2937 		goto drop;
2938 	if (tiflags & TH_ACK) {
2939 		(void)tcp_respond(tp, m, m, th, (tcp_seq)0, th->th_ack, TH_RST);
2940 	} else {
2941 		if (tiflags & TH_SYN)
2942 			tlen++;
2943 		(void)tcp_respond(tp, m, m, th, th->th_seq + tlen, (tcp_seq)0,
2944 		    TH_RST|TH_ACK);
2945 	}
2946 	if (tcp_saveti)
2947 		m_freem(tcp_saveti);
2948 	return;
2949 
2950 badcsum:
2951 drop:
2952 	/*
2953 	 * Drop space held by incoming segment and return.
2954 	 */
2955 	if (tp) {
2956 		if (tp->t_inpcb)
2957 			so = tp->t_inpcb->inp_socket;
2958 #ifdef INET6
2959 		else if (tp->t_in6pcb)
2960 			so = tp->t_in6pcb->in6p_socket;
2961 #endif
2962 		else
2963 			so = NULL;
2964 #ifdef TCP_DEBUG
2965 		if (so && (so->so_options & SO_DEBUG) != 0)
2966 			tcp_trace(TA_DROP, ostate, tp, tcp_saveti, 0);
2967 #endif
2968 	}
2969 	if (tcp_saveti)
2970 		m_freem(tcp_saveti);
2971 	m_freem(m);
2972 	return;
2973 }
2974 
2975 #ifdef TCP_SIGNATURE
2976 int
2977 tcp_signature_apply(void *fstate, void *data, u_int len)
2978 {
2979 
2980 	MD5Update(fstate, (u_char *)data, len);
2981 	return (0);
2982 }
2983 
2984 struct secasvar *
2985 tcp_signature_getsav(struct mbuf *m)
2986 {
2987 	struct ip *ip;
2988 	struct ip6_hdr *ip6;
2989 
2990 	ip = mtod(m, struct ip *);
2991 	switch (ip->ip_v) {
2992 	case 4:
2993 		ip = mtod(m, struct ip *);
2994 		ip6 = NULL;
2995 		break;
2996 	case 6:
2997 		ip = NULL;
2998 		ip6 = mtod(m, struct ip6_hdr *);
2999 		break;
3000 	default:
3001 		return (NULL);
3002 	}
3003 
3004 #ifdef IPSEC
3005 	union sockaddr_union dst;
3006 
3007 	/* Extract the destination from the IP header in the mbuf. */
3008 	memset(&dst, 0, sizeof(union sockaddr_union));
3009 	if (ip != NULL) {
3010 		dst.sa.sa_len = sizeof(struct sockaddr_in);
3011 		dst.sa.sa_family = AF_INET;
3012 		dst.sin.sin_addr = ip->ip_dst;
3013 	} else {
3014 		dst.sa.sa_len = sizeof(struct sockaddr_in6);
3015 		dst.sa.sa_family = AF_INET6;
3016 		dst.sin6.sin6_addr = ip6->ip6_dst;
3017 	}
3018 
3019 	/*
3020 	 * Look up an SADB entry which matches the address of the peer.
3021 	 */
3022 	return KEY_LOOKUP_SA(&dst, IPPROTO_TCP, htonl(TCP_SIG_SPI), 0, 0);
3023 #else
3024 	return NULL;
3025 #endif
3026 }
3027 
3028 int
3029 tcp_signature(struct mbuf *m, struct tcphdr *th, int thoff,
3030     struct secasvar *sav, char *sig)
3031 {
3032 	MD5_CTX ctx;
3033 	struct ip *ip;
3034 	struct ipovly *ipovly;
3035 #ifdef INET6
3036 	struct ip6_hdr *ip6;
3037 	struct ip6_hdr_pseudo ip6pseudo;
3038 #endif
3039 	struct ippseudo ippseudo;
3040 	struct tcphdr th0;
3041 	int l, tcphdrlen;
3042 
3043 	if (sav == NULL)
3044 		return (-1);
3045 
3046 	tcphdrlen = th->th_off * 4;
3047 
3048 	switch (mtod(m, struct ip *)->ip_v) {
3049 	case 4:
3050 		MD5Init(&ctx);
3051 		ip = mtod(m, struct ip *);
3052 		memset(&ippseudo, 0, sizeof(ippseudo));
3053 		ipovly = (struct ipovly *)ip;
3054 		ippseudo.ippseudo_src = ipovly->ih_src;
3055 		ippseudo.ippseudo_dst = ipovly->ih_dst;
3056 		ippseudo.ippseudo_pad = 0;
3057 		ippseudo.ippseudo_p = IPPROTO_TCP;
3058 		ippseudo.ippseudo_len = htons(m->m_pkthdr.len - thoff);
3059 		MD5Update(&ctx, (char *)&ippseudo, sizeof(ippseudo));
3060 		break;
3061 #if INET6
3062 	case 6:
3063 		MD5Init(&ctx);
3064 		ip6 = mtod(m, struct ip6_hdr *);
3065 		memset(&ip6pseudo, 0, sizeof(ip6pseudo));
3066 		ip6pseudo.ip6ph_src = ip6->ip6_src;
3067 		in6_clearscope(&ip6pseudo.ip6ph_src);
3068 		ip6pseudo.ip6ph_dst = ip6->ip6_dst;
3069 		in6_clearscope(&ip6pseudo.ip6ph_dst);
3070 		ip6pseudo.ip6ph_len = htons(m->m_pkthdr.len - thoff);
3071 		ip6pseudo.ip6ph_nxt = IPPROTO_TCP;
3072 		MD5Update(&ctx, (char *)&ip6pseudo, sizeof(ip6pseudo));
3073 		break;
3074 #endif
3075 	default:
3076 		return (-1);
3077 	}
3078 
3079 	th0 = *th;
3080 	th0.th_sum = 0;
3081 	MD5Update(&ctx, (char *)&th0, sizeof(th0));
3082 
3083 	l = m->m_pkthdr.len - thoff - tcphdrlen;
3084 	if (l > 0)
3085 		m_apply(m, thoff + tcphdrlen,
3086 		    m->m_pkthdr.len - thoff - tcphdrlen,
3087 		    tcp_signature_apply, &ctx);
3088 
3089 	MD5Update(&ctx, _KEYBUF(sav->key_auth), _KEYLEN(sav->key_auth));
3090 	MD5Final(sig, &ctx);
3091 
3092 	return (0);
3093 }
3094 #endif
3095 
3096 /*
3097  * Parse and process tcp options.
3098  *
3099  * Returns -1 if this segment should be dropped.  (eg. wrong signature)
3100  * Otherwise returns 0.
3101  */
3102 static int
3103 tcp_dooptions(struct tcpcb *tp, const u_char *cp, int cnt, struct tcphdr *th,
3104     struct mbuf *m, int toff, struct tcp_opt_info *oi)
3105 {
3106 	u_int16_t mss;
3107 	int opt, optlen = 0;
3108 #ifdef TCP_SIGNATURE
3109 	void *sigp = NULL;
3110 	char sigbuf[TCP_SIGLEN];
3111 	struct secasvar *sav = NULL;
3112 #endif
3113 
3114 	for (; cp && cnt > 0; cnt -= optlen, cp += optlen) {
3115 		opt = cp[0];
3116 		if (opt == TCPOPT_EOL)
3117 			break;
3118 		if (opt == TCPOPT_NOP)
3119 			optlen = 1;
3120 		else {
3121 			if (cnt < 2)
3122 				break;
3123 			optlen = cp[1];
3124 			if (optlen < 2 || optlen > cnt)
3125 				break;
3126 		}
3127 		switch (opt) {
3128 
3129 		default:
3130 			continue;
3131 
3132 		case TCPOPT_MAXSEG:
3133 			if (optlen != TCPOLEN_MAXSEG)
3134 				continue;
3135 			if (!(th->th_flags & TH_SYN))
3136 				continue;
3137 			if (TCPS_HAVERCVDSYN(tp->t_state))
3138 				continue;
3139 			memcpy(&mss, cp + 2, sizeof(mss));
3140 			oi->maxseg = ntohs(mss);
3141 			break;
3142 
3143 		case TCPOPT_WINDOW:
3144 			if (optlen != TCPOLEN_WINDOW)
3145 				continue;
3146 			if (!(th->th_flags & TH_SYN))
3147 				continue;
3148 			if (TCPS_HAVERCVDSYN(tp->t_state))
3149 				continue;
3150 			tp->t_flags |= TF_RCVD_SCALE;
3151 			tp->requested_s_scale = cp[2];
3152 			if (tp->requested_s_scale > TCP_MAX_WINSHIFT) {
3153 				char buf[INET6_ADDRSTRLEN];
3154 				struct ip *ip = mtod(m, struct ip *);
3155 #ifdef INET6
3156 				struct ip6_hdr *ip6 = mtod(m, struct ip6_hdr *);
3157 #endif
3158 
3159 				switch (ip->ip_v) {
3160 				case 4:
3161 					in_print(buf, sizeof(buf),
3162 					    &ip->ip_src);
3163 					break;
3164 #ifdef INET6
3165 				case 6:
3166 					in6_print(buf, sizeof(buf),
3167 					    &ip6->ip6_src);
3168 					break;
3169 #endif
3170 				default:
3171 					strlcpy(buf, "(unknown)", sizeof(buf));
3172 					break;
3173 				}
3174 
3175 				log(LOG_ERR, "TCP: invalid wscale %d from %s, "
3176 				    "assuming %d\n",
3177 				    tp->requested_s_scale, buf,
3178 				    TCP_MAX_WINSHIFT);
3179 				tp->requested_s_scale = TCP_MAX_WINSHIFT;
3180 			}
3181 			break;
3182 
3183 		case TCPOPT_TIMESTAMP:
3184 			if (optlen != TCPOLEN_TIMESTAMP)
3185 				continue;
3186 			oi->ts_present = 1;
3187 			memcpy(&oi->ts_val, cp + 2, sizeof(oi->ts_val));
3188 			NTOHL(oi->ts_val);
3189 			memcpy(&oi->ts_ecr, cp + 6, sizeof(oi->ts_ecr));
3190 			NTOHL(oi->ts_ecr);
3191 
3192 			if (!(th->th_flags & TH_SYN))
3193 				continue;
3194 			if (TCPS_HAVERCVDSYN(tp->t_state))
3195 				continue;
3196 			/*
3197 			 * A timestamp received in a SYN makes
3198 			 * it ok to send timestamp requests and replies.
3199 			 */
3200 			tp->t_flags |= TF_RCVD_TSTMP;
3201 			tp->ts_recent = oi->ts_val;
3202 			tp->ts_recent_age = tcp_now;
3203                         break;
3204 
3205 		case TCPOPT_SACK_PERMITTED:
3206 			if (optlen != TCPOLEN_SACK_PERMITTED)
3207 				continue;
3208 			if (!(th->th_flags & TH_SYN))
3209 				continue;
3210 			if (TCPS_HAVERCVDSYN(tp->t_state))
3211 				continue;
3212 			if (tcp_do_sack) {
3213 				tp->t_flags |= TF_SACK_PERMIT;
3214 				tp->t_flags |= TF_WILL_SACK;
3215 			}
3216 			break;
3217 
3218 		case TCPOPT_SACK:
3219 			tcp_sack_option(tp, th, cp, optlen);
3220 			break;
3221 #ifdef TCP_SIGNATURE
3222 		case TCPOPT_SIGNATURE:
3223 			if (optlen != TCPOLEN_SIGNATURE)
3224 				continue;
3225 			if (sigp &&
3226 			    !consttime_memequal(sigp, cp + 2, TCP_SIGLEN))
3227 				return (-1);
3228 
3229 			sigp = sigbuf;
3230 			memcpy(sigbuf, cp + 2, TCP_SIGLEN);
3231 			tp->t_flags |= TF_SIGNATURE;
3232 			break;
3233 #endif
3234 		}
3235 	}
3236 
3237 #ifndef TCP_SIGNATURE
3238 	return 0;
3239 #else
3240 	if (tp->t_flags & TF_SIGNATURE) {
3241 		sav = tcp_signature_getsav(m);
3242 		if (sav == NULL && tp->t_state == TCPS_LISTEN)
3243 			return (-1);
3244 	}
3245 
3246 	if ((sigp ? TF_SIGNATURE : 0) ^ (tp->t_flags & TF_SIGNATURE))
3247 		goto out;
3248 
3249 	if (sigp) {
3250 		char sig[TCP_SIGLEN];
3251 
3252 		tcp_fields_to_net(th);
3253 		if (tcp_signature(m, th, toff, sav, sig) < 0) {
3254 			tcp_fields_to_host(th);
3255 			goto out;
3256 		}
3257 		tcp_fields_to_host(th);
3258 
3259 		if (!consttime_memequal(sig, sigp, TCP_SIGLEN)) {
3260 			TCP_STATINC(TCP_STAT_BADSIG);
3261 			goto out;
3262 		} else
3263 			TCP_STATINC(TCP_STAT_GOODSIG);
3264 
3265 		key_sa_recordxfer(sav, m);
3266 		KEY_SA_UNREF(&sav);
3267 	}
3268 	return 0;
3269 out:
3270 	if (sav != NULL)
3271 		KEY_SA_UNREF(&sav);
3272 	return -1;
3273 #endif
3274 }
3275 
3276 /*
3277  * Pull out of band byte out of a segment so
3278  * it doesn't appear in the user's data queue.
3279  * It is still reflected in the segment length for
3280  * sequencing purposes.
3281  */
3282 void
3283 tcp_pulloutofband(struct socket *so, struct tcphdr *th,
3284     struct mbuf *m, int off)
3285 {
3286 	int cnt = off + th->th_urp - 1;
3287 
3288 	while (cnt >= 0) {
3289 		if (m->m_len > cnt) {
3290 			char *cp = mtod(m, char *) + cnt;
3291 			struct tcpcb *tp = sototcpcb(so);
3292 
3293 			tp->t_iobc = *cp;
3294 			tp->t_oobflags |= TCPOOB_HAVEDATA;
3295 			memmove(cp, cp + 1, (unsigned)(m->m_len - cnt - 1));
3296 			m->m_len--;
3297 			return;
3298 		}
3299 		cnt -= m->m_len;
3300 		m = m->m_next;
3301 		if (m == NULL)
3302 			break;
3303 	}
3304 	panic("tcp_pulloutofband");
3305 }
3306 
3307 /*
3308  * Collect new round-trip time estimate
3309  * and update averages and current timeout.
3310  *
3311  * rtt is in units of slow ticks (typically 500 ms) -- essentially the
3312  * difference of two timestamps.
3313  */
3314 void
3315 tcp_xmit_timer(struct tcpcb *tp, uint32_t rtt)
3316 {
3317 	int32_t delta;
3318 
3319 	TCP_STATINC(TCP_STAT_RTTUPDATED);
3320 	if (tp->t_srtt != 0) {
3321 		/*
3322 		 * Compute the amount to add to srtt for smoothing,
3323 		 * *alpha, or 2^(-TCP_RTT_SHIFT).  Because
3324 		 * srtt is stored in 1/32 slow ticks, we conceptually
3325 		 * shift left 5 bits, subtract srtt to get the
3326 		 * diference, and then shift right by TCP_RTT_SHIFT
3327 		 * (3) to obtain 1/8 of the difference.
3328 		 */
3329 		delta = (rtt << 2) - (tp->t_srtt >> TCP_RTT_SHIFT);
3330 		/*
3331 		 * This can never happen, because delta's lowest
3332 		 * possible value is 1/8 of t_srtt.  But if it does,
3333 		 * set srtt to some reasonable value, here chosen
3334 		 * as 1/8 tick.
3335 		 */
3336 		if ((tp->t_srtt += delta) <= 0)
3337 			tp->t_srtt = 1 << 2;
3338 		/*
3339 		 * RFC2988 requires that rttvar be updated first.
3340 		 * This code is compliant because "delta" is the old
3341 		 * srtt minus the new observation (scaled).
3342 		 *
3343 		 * RFC2988 says:
3344 		 *   rttvar = (1-beta) * rttvar + beta * |srtt-observed|
3345 		 *
3346 		 * delta is in units of 1/32 ticks, and has then been
3347 		 * divided by 8.  This is equivalent to being in 1/16s
3348 		 * units and divided by 4.  Subtract from it 1/4 of
3349 		 * the existing rttvar to form the (signed) amount to
3350 		 * adjust.
3351 		 */
3352 		if (delta < 0)
3353 			delta = -delta;
3354 		delta -= (tp->t_rttvar >> TCP_RTTVAR_SHIFT);
3355 		/*
3356 		 * As with srtt, this should never happen.  There is
3357 		 * no support in RFC2988 for this operation.  But 1/4s
3358 		 * as rttvar when faced with something arguably wrong
3359 		 * is ok.
3360 		 */
3361 		if ((tp->t_rttvar += delta) <= 0)
3362 			tp->t_rttvar = 1 << 2;
3363 
3364 		/*
3365 		 * If srtt exceeds .01 second, ensure we use the 'remote' MSL
3366 		 * Problem is: it doesn't work.  Disabled by defaulting
3367 		 * tcp_rttlocal to 0; see corresponding code in
3368 		 * tcp_subr that selects local vs remote in a different way.
3369 		 *
3370 		 * The static branch prediction hint here should be removed
3371 		 * when the rtt estimator is fixed and the rtt_enable code
3372 		 * is turned back on.
3373 		 */
3374 		if (__predict_false(tcp_rttlocal) && tcp_msl_enable
3375 		    && tp->t_srtt > tcp_msl_remote_threshold
3376 		    && tp->t_msl  < tcp_msl_remote) {
3377 			tp->t_msl = MIN(tcp_msl_remote, TCP_MAXMSL);
3378 		}
3379 	} else {
3380 		/*
3381 		 * This is the first measurement.  Per RFC2988, 2.2,
3382 		 * set rtt=R and srtt=R/2.
3383 		 * For srtt, storage representation is 1/32 ticks,
3384 		 * so shift left by 5.
3385 		 * For rttvar, storage representation is 1/16 ticks,
3386 		 * So shift left by 4, but then right by 1 to halve.
3387 		 */
3388 		tp->t_srtt = rtt << (TCP_RTT_SHIFT + 2);
3389 		tp->t_rttvar = rtt << (TCP_RTTVAR_SHIFT + 2 - 1);
3390 	}
3391 	tp->t_rtttime = 0;
3392 	tp->t_rxtshift = 0;
3393 
3394 	/*
3395 	 * the retransmit should happen at rtt + 4 * rttvar.
3396 	 * Because of the way we do the smoothing, srtt and rttvar
3397 	 * will each average +1/2 tick of bias.  When we compute
3398 	 * the retransmit timer, we want 1/2 tick of rounding and
3399 	 * 1 extra tick because of +-1/2 tick uncertainty in the
3400 	 * firing of the timer.  The bias will give us exactly the
3401 	 * 1.5 tick we need.  But, because the bias is
3402 	 * statistical, we have to test that we don't drop below
3403 	 * the minimum feasible timer (which is 2 ticks).
3404 	 */
3405 	TCPT_RANGESET(tp->t_rxtcur, TCP_REXMTVAL(tp),
3406 	    uimax(tp->t_rttmin, rtt + 2), TCPTV_REXMTMAX);
3407 
3408 	/*
3409 	 * We received an ack for a packet that wasn't retransmitted;
3410 	 * it is probably safe to discard any error indications we've
3411 	 * received recently.  This isn't quite right, but close enough
3412 	 * for now (a route might have failed after we sent a segment,
3413 	 * and the return path might not be symmetrical).
3414 	 */
3415 	tp->t_softerror = 0;
3416 }
3417 
3418 
3419 /*
3420  * TCP compressed state engine.  Currently used to hold compressed
3421  * state for SYN_RECEIVED.
3422  */
3423 
3424 u_long	syn_cache_count;
3425 u_int32_t syn_hash1, syn_hash2;
3426 
3427 #define SYN_HASH(sa, sp, dp) \
3428 	((((sa)->s_addr^syn_hash1)*(((((u_int32_t)(dp))<<16) + \
3429 				     ((u_int32_t)(sp)))^syn_hash2)))
3430 #ifndef INET6
3431 #define	SYN_HASHALL(hash, src, dst) \
3432 do {									\
3433 	hash = SYN_HASH(&((const struct sockaddr_in *)(src))->sin_addr,	\
3434 		((const struct sockaddr_in *)(src))->sin_port,		\
3435 		((const struct sockaddr_in *)(dst))->sin_port);		\
3436 } while (/*CONSTCOND*/ 0)
3437 #else
3438 #define SYN_HASH6(sa, sp, dp) \
3439 	((((sa)->s6_addr32[0] ^ (sa)->s6_addr32[3] ^ syn_hash1) * \
3440 	  (((((u_int32_t)(dp))<<16) + ((u_int32_t)(sp)))^syn_hash2)) \
3441 	 & 0x7fffffff)
3442 
3443 #define SYN_HASHALL(hash, src, dst) \
3444 do {									\
3445 	switch ((src)->sa_family) {					\
3446 	case AF_INET:							\
3447 		hash = SYN_HASH(&((const struct sockaddr_in *)(src))->sin_addr, \
3448 			((const struct sockaddr_in *)(src))->sin_port,	\
3449 			((const struct sockaddr_in *)(dst))->sin_port);	\
3450 		break;							\
3451 	case AF_INET6:							\
3452 		hash = SYN_HASH6(&((const struct sockaddr_in6 *)(src))->sin6_addr, \
3453 			((const struct sockaddr_in6 *)(src))->sin6_port,	\
3454 			((const struct sockaddr_in6 *)(dst))->sin6_port);	\
3455 		break;							\
3456 	default:							\
3457 		hash = 0;						\
3458 	}								\
3459 } while (/*CONSTCOND*/0)
3460 #endif /* INET6 */
3461 
3462 static struct pool syn_cache_pool;
3463 
3464 /*
3465  * We don't estimate RTT with SYNs, so each packet starts with the default
3466  * RTT and each timer step has a fixed timeout value.
3467  */
3468 static inline void
3469 syn_cache_timer_arm(struct syn_cache *sc)
3470 {
3471 
3472 	TCPT_RANGESET(sc->sc_rxtcur,
3473 	    TCPTV_SRTTDFLT * tcp_backoff[sc->sc_rxtshift], TCPTV_MIN,
3474 	    TCPTV_REXMTMAX);
3475 	callout_reset(&sc->sc_timer,
3476 	    sc->sc_rxtcur * (hz / PR_SLOWHZ), syn_cache_timer, sc);
3477 }
3478 
3479 #define	SYN_CACHE_TIMESTAMP(sc)	(tcp_now - (sc)->sc_timebase)
3480 
3481 static inline void
3482 syn_cache_rm(struct syn_cache *sc)
3483 {
3484 	TAILQ_REMOVE(&tcp_syn_cache[sc->sc_bucketidx].sch_bucket,
3485 	    sc, sc_bucketq);
3486 	sc->sc_tp = NULL;
3487 	LIST_REMOVE(sc, sc_tpq);
3488 	tcp_syn_cache[sc->sc_bucketidx].sch_length--;
3489 	callout_stop(&sc->sc_timer);
3490 	syn_cache_count--;
3491 }
3492 
3493 static inline void
3494 syn_cache_put(struct syn_cache *sc)
3495 {
3496 	if (sc->sc_ipopts)
3497 		(void) m_free(sc->sc_ipopts);
3498 	rtcache_free(&sc->sc_route);
3499 	sc->sc_flags |= SCF_DEAD;
3500 	if (!callout_invoking(&sc->sc_timer))
3501 		callout_schedule(&(sc)->sc_timer, 1);
3502 }
3503 
3504 void
3505 syn_cache_init(void)
3506 {
3507 	int i;
3508 
3509 	pool_init(&syn_cache_pool, sizeof(struct syn_cache), 0, 0, 0,
3510 	    "synpl", NULL, IPL_SOFTNET);
3511 
3512 	/* Initialize the hash buckets. */
3513 	for (i = 0; i < tcp_syn_cache_size; i++)
3514 		TAILQ_INIT(&tcp_syn_cache[i].sch_bucket);
3515 }
3516 
3517 void
3518 syn_cache_insert(struct syn_cache *sc, struct tcpcb *tp)
3519 {
3520 	struct syn_cache_head *scp;
3521 	struct syn_cache *sc2;
3522 	int s;
3523 
3524 	/*
3525 	 * If there are no entries in the hash table, reinitialize
3526 	 * the hash secrets.
3527 	 */
3528 	if (syn_cache_count == 0) {
3529 		syn_hash1 = cprng_fast32();
3530 		syn_hash2 = cprng_fast32();
3531 	}
3532 
3533 	SYN_HASHALL(sc->sc_hash, &sc->sc_src.sa, &sc->sc_dst.sa);
3534 	sc->sc_bucketidx = sc->sc_hash % tcp_syn_cache_size;
3535 	scp = &tcp_syn_cache[sc->sc_bucketidx];
3536 
3537 	/*
3538 	 * Make sure that we don't overflow the per-bucket
3539 	 * limit or the total cache size limit.
3540 	 */
3541 	s = splsoftnet();
3542 	if (scp->sch_length >= tcp_syn_bucket_limit) {
3543 		TCP_STATINC(TCP_STAT_SC_BUCKETOVERFLOW);
3544 		/*
3545 		 * The bucket is full.  Toss the oldest element in the
3546 		 * bucket.  This will be the first entry in the bucket.
3547 		 */
3548 		sc2 = TAILQ_FIRST(&scp->sch_bucket);
3549 #ifdef DIAGNOSTIC
3550 		/*
3551 		 * This should never happen; we should always find an
3552 		 * entry in our bucket.
3553 		 */
3554 		if (sc2 == NULL)
3555 			panic("syn_cache_insert: bucketoverflow: impossible");
3556 #endif
3557 		syn_cache_rm(sc2);
3558 		syn_cache_put(sc2);	/* calls pool_put but see spl above */
3559 	} else if (syn_cache_count >= tcp_syn_cache_limit) {
3560 		struct syn_cache_head *scp2, *sce;
3561 
3562 		TCP_STATINC(TCP_STAT_SC_OVERFLOWED);
3563 		/*
3564 		 * The cache is full.  Toss the oldest entry in the
3565 		 * first non-empty bucket we can find.
3566 		 *
3567 		 * XXX We would really like to toss the oldest
3568 		 * entry in the cache, but we hope that this
3569 		 * condition doesn't happen very often.
3570 		 */
3571 		scp2 = scp;
3572 		if (TAILQ_EMPTY(&scp2->sch_bucket)) {
3573 			sce = &tcp_syn_cache[tcp_syn_cache_size];
3574 			for (++scp2; scp2 != scp; scp2++) {
3575 				if (scp2 >= sce)
3576 					scp2 = &tcp_syn_cache[0];
3577 				if (! TAILQ_EMPTY(&scp2->sch_bucket))
3578 					break;
3579 			}
3580 #ifdef DIAGNOSTIC
3581 			/*
3582 			 * This should never happen; we should always find a
3583 			 * non-empty bucket.
3584 			 */
3585 			if (scp2 == scp)
3586 				panic("syn_cache_insert: cacheoverflow: "
3587 				    "impossible");
3588 #endif
3589 		}
3590 		sc2 = TAILQ_FIRST(&scp2->sch_bucket);
3591 		syn_cache_rm(sc2);
3592 		syn_cache_put(sc2);	/* calls pool_put but see spl above */
3593 	}
3594 
3595 	/*
3596 	 * Initialize the entry's timer.
3597 	 */
3598 	sc->sc_rxttot = 0;
3599 	sc->sc_rxtshift = 0;
3600 	syn_cache_timer_arm(sc);
3601 
3602 	/* Link it from tcpcb entry */
3603 	LIST_INSERT_HEAD(&tp->t_sc, sc, sc_tpq);
3604 
3605 	/* Put it into the bucket. */
3606 	TAILQ_INSERT_TAIL(&scp->sch_bucket, sc, sc_bucketq);
3607 	scp->sch_length++;
3608 	syn_cache_count++;
3609 
3610 	TCP_STATINC(TCP_STAT_SC_ADDED);
3611 	splx(s);
3612 }
3613 
3614 /*
3615  * Walk the timer queues, looking for SYN,ACKs that need to be retransmitted.
3616  * If we have retransmitted an entry the maximum number of times, expire
3617  * that entry.
3618  */
3619 static void
3620 syn_cache_timer(void *arg)
3621 {
3622 	struct syn_cache *sc = arg;
3623 
3624 	mutex_enter(softnet_lock);
3625 	KERNEL_LOCK(1, NULL);
3626 
3627 	callout_ack(&sc->sc_timer);
3628 
3629 	if (__predict_false(sc->sc_flags & SCF_DEAD)) {
3630 		TCP_STATINC(TCP_STAT_SC_DELAYED_FREE);
3631 		goto free;
3632 	}
3633 
3634 	if (__predict_false(sc->sc_rxtshift == TCP_MAXRXTSHIFT)) {
3635 		/* Drop it -- too many retransmissions. */
3636 		goto dropit;
3637 	}
3638 
3639 	/*
3640 	 * Compute the total amount of time this entry has
3641 	 * been on a queue.  If this entry has been on longer
3642 	 * than the keep alive timer would allow, expire it.
3643 	 */
3644 	sc->sc_rxttot += sc->sc_rxtcur;
3645 	if (sc->sc_rxttot >= MIN(tcp_keepinit, TCP_TIMER_MAXTICKS))
3646 		goto dropit;
3647 
3648 	TCP_STATINC(TCP_STAT_SC_RETRANSMITTED);
3649 	(void)syn_cache_respond(sc);
3650 
3651 	/* Advance the timer back-off. */
3652 	sc->sc_rxtshift++;
3653 	syn_cache_timer_arm(sc);
3654 
3655 	goto out;
3656 
3657  dropit:
3658 	TCP_STATINC(TCP_STAT_SC_TIMED_OUT);
3659 	syn_cache_rm(sc);
3660 	if (sc->sc_ipopts)
3661 		(void) m_free(sc->sc_ipopts);
3662 	rtcache_free(&sc->sc_route);
3663 
3664  free:
3665 	callout_destroy(&sc->sc_timer);
3666 	pool_put(&syn_cache_pool, sc);
3667 
3668  out:
3669 	KERNEL_UNLOCK_ONE(NULL);
3670 	mutex_exit(softnet_lock);
3671 }
3672 
3673 /*
3674  * Remove syn cache created by the specified tcb entry,
3675  * because this does not make sense to keep them
3676  * (if there's no tcb entry, syn cache entry will never be used)
3677  */
3678 void
3679 syn_cache_cleanup(struct tcpcb *tp)
3680 {
3681 	struct syn_cache *sc, *nsc;
3682 	int s;
3683 
3684 	s = splsoftnet();
3685 
3686 	for (sc = LIST_FIRST(&tp->t_sc); sc != NULL; sc = nsc) {
3687 		nsc = LIST_NEXT(sc, sc_tpq);
3688 
3689 #ifdef DIAGNOSTIC
3690 		if (sc->sc_tp != tp)
3691 			panic("invalid sc_tp in syn_cache_cleanup");
3692 #endif
3693 		syn_cache_rm(sc);
3694 		syn_cache_put(sc);	/* calls pool_put but see spl above */
3695 	}
3696 	/* just for safety */
3697 	LIST_INIT(&tp->t_sc);
3698 
3699 	splx(s);
3700 }
3701 
3702 /*
3703  * Find an entry in the syn cache.
3704  */
3705 struct syn_cache *
3706 syn_cache_lookup(const struct sockaddr *src, const struct sockaddr *dst,
3707     struct syn_cache_head **headp)
3708 {
3709 	struct syn_cache *sc;
3710 	struct syn_cache_head *scp;
3711 	u_int32_t hash;
3712 	int s;
3713 
3714 	SYN_HASHALL(hash, src, dst);
3715 
3716 	scp = &tcp_syn_cache[hash % tcp_syn_cache_size];
3717 	*headp = scp;
3718 	s = splsoftnet();
3719 	for (sc = TAILQ_FIRST(&scp->sch_bucket); sc != NULL;
3720 	     sc = TAILQ_NEXT(sc, sc_bucketq)) {
3721 		if (sc->sc_hash != hash)
3722 			continue;
3723 		if (!memcmp(&sc->sc_src, src, src->sa_len) &&
3724 		    !memcmp(&sc->sc_dst, dst, dst->sa_len)) {
3725 			splx(s);
3726 			return (sc);
3727 		}
3728 	}
3729 	splx(s);
3730 	return (NULL);
3731 }
3732 
3733 /*
3734  * This function gets called when we receive an ACK for a socket in the
3735  * LISTEN state. We look up the connection in the syn cache, and if it's
3736  * there, we pull it out of the cache and turn it into a full-blown
3737  * connection in the SYN-RECEIVED state.
3738  *
3739  * The return values may not be immediately obvious, and their effects
3740  * can be subtle, so here they are:
3741  *
3742  *	NULL	SYN was not found in cache; caller should drop the
3743  *		packet and send an RST.
3744  *
3745  *	-1	We were unable to create the new connection, and are
3746  *		aborting it.  An ACK,RST is being sent to the peer
3747  *		(unless we got screwey sequence numbers; see below),
3748  *		because the 3-way handshake has been completed.  Caller
3749  *		should not free the mbuf, since we may be using it.  If
3750  *		we are not, we will free it.
3751  *
3752  *	Otherwise, the return value is a pointer to the new socket
3753  *	associated with the connection.
3754  */
3755 struct socket *
3756 syn_cache_get(struct sockaddr *src, struct sockaddr *dst,
3757     struct tcphdr *th, struct socket *so, struct mbuf *m)
3758 {
3759 	struct syn_cache *sc;
3760 	struct syn_cache_head *scp;
3761 	struct inpcb *inp = NULL;
3762 #ifdef INET6
3763 	struct in6pcb *in6p = NULL;
3764 #endif
3765 	struct tcpcb *tp;
3766 	int s;
3767 	struct socket *oso;
3768 
3769 	s = splsoftnet();
3770 	if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) {
3771 		splx(s);
3772 		return NULL;
3773 	}
3774 
3775 	/*
3776 	 * Verify the sequence and ack numbers.  Try getting the correct
3777 	 * response again.
3778 	 */
3779 	if ((th->th_ack != sc->sc_iss + 1) ||
3780 	    SEQ_LEQ(th->th_seq, sc->sc_irs) ||
3781 	    SEQ_GT(th->th_seq, sc->sc_irs + 1 + sc->sc_win)) {
3782 		m_freem(m);
3783 		(void)syn_cache_respond(sc);
3784 		splx(s);
3785 		return ((struct socket *)(-1));
3786 	}
3787 
3788 	/* Remove this cache entry */
3789 	syn_cache_rm(sc);
3790 	splx(s);
3791 
3792 	/*
3793 	 * Ok, create the full blown connection, and set things up
3794 	 * as they would have been set up if we had created the
3795 	 * connection when the SYN arrived.  If we can't create
3796 	 * the connection, abort it.
3797 	 */
3798 	/*
3799 	 * inp still has the OLD in_pcb stuff, set the
3800 	 * v6-related flags on the new guy, too.   This is
3801 	 * done particularly for the case where an AF_INET6
3802 	 * socket is bound only to a port, and a v4 connection
3803 	 * comes in on that port.
3804 	 * we also copy the flowinfo from the original pcb
3805 	 * to the new one.
3806 	 */
3807 	oso = so;
3808 	so = sonewconn(so, true);
3809 	if (so == NULL)
3810 		goto resetandabort;
3811 
3812 	switch (so->so_proto->pr_domain->dom_family) {
3813 	case AF_INET:
3814 		inp = sotoinpcb(so);
3815 		break;
3816 #ifdef INET6
3817 	case AF_INET6:
3818 		in6p = sotoin6pcb(so);
3819 		break;
3820 #endif
3821 	}
3822 
3823 	switch (src->sa_family) {
3824 	case AF_INET:
3825 		if (inp) {
3826 			inp->inp_laddr = ((struct sockaddr_in *)dst)->sin_addr;
3827 			inp->inp_lport = ((struct sockaddr_in *)dst)->sin_port;
3828 			inp->inp_options = ip_srcroute(m);
3829 			in_pcbstate(inp, INP_BOUND);
3830 			if (inp->inp_options == NULL) {
3831 				inp->inp_options = sc->sc_ipopts;
3832 				sc->sc_ipopts = NULL;
3833 			}
3834 		}
3835 #ifdef INET6
3836 		else if (in6p) {
3837 			/* IPv4 packet to AF_INET6 socket */
3838 			memset(&in6p->in6p_laddr, 0, sizeof(in6p->in6p_laddr));
3839 			in6p->in6p_laddr.s6_addr16[5] = htons(0xffff);
3840 			bcopy(&((struct sockaddr_in *)dst)->sin_addr,
3841 				&in6p->in6p_laddr.s6_addr32[3],
3842 				sizeof(((struct sockaddr_in *)dst)->sin_addr));
3843 			in6p->in6p_lport = ((struct sockaddr_in *)dst)->sin_port;
3844 			in6totcpcb(in6p)->t_family = AF_INET;
3845 			if (sotoin6pcb(oso)->in6p_flags & IN6P_IPV6_V6ONLY)
3846 				in6p->in6p_flags |= IN6P_IPV6_V6ONLY;
3847 			else
3848 				in6p->in6p_flags &= ~IN6P_IPV6_V6ONLY;
3849 			in6_pcbstate(in6p, IN6P_BOUND);
3850 		}
3851 #endif
3852 		break;
3853 #ifdef INET6
3854 	case AF_INET6:
3855 		if (in6p) {
3856 			in6p->in6p_laddr = ((struct sockaddr_in6 *)dst)->sin6_addr;
3857 			in6p->in6p_lport = ((struct sockaddr_in6 *)dst)->sin6_port;
3858 			in6_pcbstate(in6p, IN6P_BOUND);
3859 		}
3860 		break;
3861 #endif
3862 	}
3863 
3864 #ifdef INET6
3865 	if (in6p && in6totcpcb(in6p)->t_family == AF_INET6 && sotoinpcb(oso)) {
3866 		struct in6pcb *oin6p = sotoin6pcb(oso);
3867 		/* inherit socket options from the listening socket */
3868 		in6p->in6p_flags |= (oin6p->in6p_flags & IN6P_CONTROLOPTS);
3869 		if (in6p->in6p_flags & IN6P_CONTROLOPTS) {
3870 			m_freem(in6p->in6p_options);
3871 			in6p->in6p_options = NULL;
3872 		}
3873 		ip6_savecontrol(in6p, &in6p->in6p_options,
3874 		    mtod(m, struct ip6_hdr *), m);
3875 	}
3876 #endif
3877 
3878 	/*
3879 	 * Give the new socket our cached route reference.
3880 	 */
3881 	if (inp) {
3882 		rtcache_copy(&inp->inp_route, &sc->sc_route);
3883 		rtcache_free(&sc->sc_route);
3884 	}
3885 #ifdef INET6
3886 	else {
3887 		rtcache_copy(&in6p->in6p_route, &sc->sc_route);
3888 		rtcache_free(&sc->sc_route);
3889 	}
3890 #endif
3891 
3892 	if (inp) {
3893 		struct sockaddr_in sin;
3894 		memcpy(&sin, src, src->sa_len);
3895 		if (in_pcbconnect(inp, &sin, &lwp0)) {
3896 			goto resetandabort;
3897 		}
3898 	}
3899 #ifdef INET6
3900 	else if (in6p) {
3901 		struct sockaddr_in6 sin6;
3902 		memcpy(&sin6, src, src->sa_len);
3903 		if (src->sa_family == AF_INET) {
3904 			/* IPv4 packet to AF_INET6 socket */
3905 			in6_sin_2_v4mapsin6((struct sockaddr_in *)src, &sin6);
3906 		}
3907 		if (in6_pcbconnect(in6p, &sin6, NULL)) {
3908 			goto resetandabort;
3909 		}
3910 	}
3911 #endif
3912 	else {
3913 		goto resetandabort;
3914 	}
3915 
3916 	if (inp)
3917 		tp = intotcpcb(inp);
3918 #ifdef INET6
3919 	else if (in6p)
3920 		tp = in6totcpcb(in6p);
3921 #endif
3922 	else
3923 		tp = NULL;
3924 
3925 	tp->t_flags = sototcpcb(oso)->t_flags & TF_NODELAY;
3926 	if (sc->sc_request_r_scale != 15) {
3927 		tp->requested_s_scale = sc->sc_requested_s_scale;
3928 		tp->request_r_scale = sc->sc_request_r_scale;
3929 		tp->snd_scale = sc->sc_requested_s_scale;
3930 		tp->rcv_scale = sc->sc_request_r_scale;
3931 		tp->t_flags |= TF_REQ_SCALE|TF_RCVD_SCALE;
3932 	}
3933 	if (sc->sc_flags & SCF_TIMESTAMP)
3934 		tp->t_flags |= TF_REQ_TSTMP|TF_RCVD_TSTMP;
3935 	tp->ts_timebase = sc->sc_timebase;
3936 
3937 	tp->t_template = tcp_template(tp);
3938 	if (tp->t_template == 0) {
3939 		tp = tcp_drop(tp, ENOBUFS);	/* destroys socket */
3940 		so = NULL;
3941 		m_freem(m);
3942 		goto abort;
3943 	}
3944 
3945 	tp->iss = sc->sc_iss;
3946 	tp->irs = sc->sc_irs;
3947 	tcp_sendseqinit(tp);
3948 	tcp_rcvseqinit(tp);
3949 	tp->t_state = TCPS_SYN_RECEIVED;
3950 	TCP_TIMER_ARM(tp, TCPT_KEEP, tp->t_keepinit);
3951 	TCP_STATINC(TCP_STAT_ACCEPTS);
3952 
3953 	if ((sc->sc_flags & SCF_SACK_PERMIT) && tcp_do_sack)
3954 		tp->t_flags |= TF_WILL_SACK;
3955 
3956 	if ((sc->sc_flags & SCF_ECN_PERMIT) && tcp_do_ecn)
3957 		tp->t_flags |= TF_ECN_PERMIT;
3958 
3959 #ifdef TCP_SIGNATURE
3960 	if (sc->sc_flags & SCF_SIGNATURE)
3961 		tp->t_flags |= TF_SIGNATURE;
3962 #endif
3963 
3964 	/* Initialize tp->t_ourmss before we deal with the peer's! */
3965 	tp->t_ourmss = sc->sc_ourmaxseg;
3966 	tcp_mss_from_peer(tp, sc->sc_peermaxseg);
3967 
3968 	/*
3969 	 * Initialize the initial congestion window.  If we
3970 	 * had to retransmit the SYN,ACK, we must initialize cwnd
3971 	 * to 1 segment (i.e. the Loss Window).
3972 	 */
3973 	if (sc->sc_rxtshift)
3974 		tp->snd_cwnd = tp->t_peermss;
3975 	else {
3976 		int ss = tcp_init_win;
3977 		if (inp != NULL && in_localaddr(inp->inp_faddr))
3978 			ss = tcp_init_win_local;
3979 #ifdef INET6
3980 		if (in6p != NULL && in6_localaddr(&in6p->in6p_faddr))
3981 			ss = tcp_init_win_local;
3982 #endif
3983 		tp->snd_cwnd = TCP_INITIAL_WINDOW(ss, tp->t_peermss);
3984 	}
3985 
3986 	tcp_rmx_rtt(tp);
3987 	tp->snd_wl1 = sc->sc_irs;
3988 	tp->rcv_up = sc->sc_irs + 1;
3989 
3990 	/*
3991 	 * This is what whould have happened in tcp_output() when
3992 	 * the SYN,ACK was sent.
3993 	 */
3994 	tp->snd_up = tp->snd_una;
3995 	tp->snd_max = tp->snd_nxt = tp->iss+1;
3996 	TCP_TIMER_ARM(tp, TCPT_REXMT, tp->t_rxtcur);
3997 	if (sc->sc_win > 0 && SEQ_GT(tp->rcv_nxt + sc->sc_win, tp->rcv_adv))
3998 		tp->rcv_adv = tp->rcv_nxt + sc->sc_win;
3999 	tp->last_ack_sent = tp->rcv_nxt;
4000 	tp->t_partialacks = -1;
4001 	tp->t_dupacks = 0;
4002 
4003 	TCP_STATINC(TCP_STAT_SC_COMPLETED);
4004 	s = splsoftnet();
4005 	syn_cache_put(sc);
4006 	splx(s);
4007 	return so;
4008 
4009 resetandabort:
4010 	(void)tcp_respond(NULL, m, m, th, (tcp_seq)0, th->th_ack, TH_RST);
4011 abort:
4012 	if (so != NULL) {
4013 		(void) soqremque(so, 1);
4014 		(void) soabort(so);
4015 		mutex_enter(softnet_lock);
4016 	}
4017 	s = splsoftnet();
4018 	syn_cache_put(sc);
4019 	splx(s);
4020 	TCP_STATINC(TCP_STAT_SC_ABORTED);
4021 	return ((struct socket *)(-1));
4022 }
4023 
4024 /*
4025  * This function is called when we get a RST for a
4026  * non-existent connection, so that we can see if the
4027  * connection is in the syn cache.  If it is, zap it.
4028  */
4029 
4030 void
4031 syn_cache_reset(struct sockaddr *src, struct sockaddr *dst, struct tcphdr *th)
4032 {
4033 	struct syn_cache *sc;
4034 	struct syn_cache_head *scp;
4035 	int s = splsoftnet();
4036 
4037 	if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) {
4038 		splx(s);
4039 		return;
4040 	}
4041 	if (SEQ_LT(th->th_seq, sc->sc_irs) ||
4042 	    SEQ_GT(th->th_seq, sc->sc_irs+1)) {
4043 		splx(s);
4044 		return;
4045 	}
4046 	syn_cache_rm(sc);
4047 	TCP_STATINC(TCP_STAT_SC_RESET);
4048 	syn_cache_put(sc);	/* calls pool_put but see spl above */
4049 	splx(s);
4050 }
4051 
4052 void
4053 syn_cache_unreach(const struct sockaddr *src, const struct sockaddr *dst,
4054     struct tcphdr *th)
4055 {
4056 	struct syn_cache *sc;
4057 	struct syn_cache_head *scp;
4058 	int s;
4059 
4060 	s = splsoftnet();
4061 	if ((sc = syn_cache_lookup(src, dst, &scp)) == NULL) {
4062 		splx(s);
4063 		return;
4064 	}
4065 	/* If the sequence number != sc_iss, then it's a bogus ICMP msg */
4066 	if (ntohl(th->th_seq) != sc->sc_iss) {
4067 		splx(s);
4068 		return;
4069 	}
4070 
4071 	/*
4072 	 * If we've retransmitted 3 times and this is our second error,
4073 	 * we remove the entry.  Otherwise, we allow it to continue on.
4074 	 * This prevents us from incorrectly nuking an entry during a
4075 	 * spurious network outage.
4076 	 *
4077 	 * See tcp_notify().
4078 	 */
4079 	if ((sc->sc_flags & SCF_UNREACH) == 0 || sc->sc_rxtshift < 3) {
4080 		sc->sc_flags |= SCF_UNREACH;
4081 		splx(s);
4082 		return;
4083 	}
4084 
4085 	syn_cache_rm(sc);
4086 	TCP_STATINC(TCP_STAT_SC_UNREACH);
4087 	syn_cache_put(sc);	/* calls pool_put but see spl above */
4088 	splx(s);
4089 }
4090 
4091 /*
4092  * Given a LISTEN socket and an inbound SYN request, add this to the syn
4093  * cache, and send back a segment:
4094  *	<SEQ=ISS><ACK=RCV_NXT><CTL=SYN,ACK>
4095  * to the source.
4096  *
4097  * IMPORTANT NOTE: We do _NOT_ ACK data that might accompany the SYN.
4098  * Doing so would require that we hold onto the data and deliver it
4099  * to the application.  However, if we are the target of a SYN-flood
4100  * DoS attack, an attacker could send data which would eventually
4101  * consume all available buffer space if it were ACKed.  By not ACKing
4102  * the data, we avoid this DoS scenario.
4103  */
4104 int
4105 syn_cache_add(struct sockaddr *src, struct sockaddr *dst, struct tcphdr *th,
4106     unsigned int toff, struct socket *so, struct mbuf *m, u_char *optp,
4107     int optlen, struct tcp_opt_info *oi)
4108 {
4109 	struct tcpcb tb, *tp;
4110 	long win;
4111 	struct syn_cache *sc;
4112 	struct syn_cache_head *scp;
4113 	struct mbuf *ipopts;
4114 	int s;
4115 
4116 	tp = sototcpcb(so);
4117 
4118 	/*
4119 	 * Initialize some local state.
4120 	 */
4121 	win = sbspace(&so->so_rcv);
4122 	if (win > TCP_MAXWIN)
4123 		win = TCP_MAXWIN;
4124 
4125 #ifdef TCP_SIGNATURE
4126 	if (optp || (tp->t_flags & TF_SIGNATURE))
4127 #else
4128 	if (optp)
4129 #endif
4130 	{
4131 		tb.t_flags = tcp_do_rfc1323 ? (TF_REQ_SCALE|TF_REQ_TSTMP) : 0;
4132 #ifdef TCP_SIGNATURE
4133 		tb.t_flags |= (tp->t_flags & TF_SIGNATURE);
4134 #endif
4135 		tb.t_state = TCPS_LISTEN;
4136 		if (tcp_dooptions(&tb, optp, optlen, th, m, toff, oi) < 0)
4137 			return 0;
4138 	} else
4139 		tb.t_flags = 0;
4140 
4141 	switch (src->sa_family) {
4142 	case AF_INET:
4143 		/* Remember the IP options, if any. */
4144 		ipopts = ip_srcroute(m);
4145 		break;
4146 	default:
4147 		ipopts = NULL;
4148 	}
4149 
4150 	/*
4151 	 * See if we already have an entry for this connection.
4152 	 * If we do, resend the SYN,ACK.  We do not count this
4153 	 * as a retransmission (XXX though maybe we should).
4154 	 */
4155 	if ((sc = syn_cache_lookup(src, dst, &scp)) != NULL) {
4156 		TCP_STATINC(TCP_STAT_SC_DUPESYN);
4157 		if (ipopts) {
4158 			/*
4159 			 * If we were remembering a previous source route,
4160 			 * forget it and use the new one we've been given.
4161 			 */
4162 			if (sc->sc_ipopts)
4163 				(void)m_free(sc->sc_ipopts);
4164 			sc->sc_ipopts = ipopts;
4165 		}
4166 		sc->sc_timestamp = tb.ts_recent;
4167 		m_freem(m);
4168 		if (syn_cache_respond(sc) == 0) {
4169 			uint64_t *tcps = TCP_STAT_GETREF();
4170 			tcps[TCP_STAT_SNDACKS]++;
4171 			tcps[TCP_STAT_SNDTOTAL]++;
4172 			TCP_STAT_PUTREF();
4173 		}
4174 		return 1;
4175 	}
4176 
4177 	s = splsoftnet();
4178 	sc = pool_get(&syn_cache_pool, PR_NOWAIT);
4179 	splx(s);
4180 	if (sc == NULL) {
4181 		if (ipopts)
4182 			(void)m_free(ipopts);
4183 		return 0;
4184 	}
4185 
4186 	/*
4187 	 * Fill in the cache, and put the necessary IP and TCP
4188 	 * options into the reply.
4189 	 */
4190 	memset(sc, 0, sizeof(struct syn_cache));
4191 	callout_init(&sc->sc_timer, CALLOUT_MPSAFE);
4192 	memcpy(&sc->sc_src, src, src->sa_len);
4193 	memcpy(&sc->sc_dst, dst, dst->sa_len);
4194 	sc->sc_flags = 0;
4195 	sc->sc_ipopts = ipopts;
4196 	sc->sc_irs = th->th_seq;
4197 	switch (src->sa_family) {
4198 	case AF_INET:
4199 	    {
4200 		struct sockaddr_in *srcin = (void *)src;
4201 		struct sockaddr_in *dstin = (void *)dst;
4202 
4203 		sc->sc_iss = tcp_new_iss1(&dstin->sin_addr,
4204 		    &srcin->sin_addr, dstin->sin_port,
4205 		    srcin->sin_port, sizeof(dstin->sin_addr), 0);
4206 		break;
4207 	    }
4208 #ifdef INET6
4209 	case AF_INET6:
4210 	    {
4211 		struct sockaddr_in6 *srcin6 = (void *)src;
4212 		struct sockaddr_in6 *dstin6 = (void *)dst;
4213 
4214 		sc->sc_iss = tcp_new_iss1(&dstin6->sin6_addr,
4215 		    &srcin6->sin6_addr, dstin6->sin6_port,
4216 		    srcin6->sin6_port, sizeof(dstin6->sin6_addr), 0);
4217 		break;
4218 	    }
4219 #endif
4220 	}
4221 	sc->sc_peermaxseg = oi->maxseg;
4222 	sc->sc_ourmaxseg = tcp_mss_to_advertise(m->m_flags & M_PKTHDR ?
4223 	    m_get_rcvif_NOMPSAFE(m) : NULL, sc->sc_src.sa.sa_family);
4224 	sc->sc_win = win;
4225 	sc->sc_timebase = tcp_now - 1;	/* see tcp_newtcpcb() */
4226 	sc->sc_timestamp = tb.ts_recent;
4227 	if ((tb.t_flags & (TF_REQ_TSTMP|TF_RCVD_TSTMP)) ==
4228 	    (TF_REQ_TSTMP|TF_RCVD_TSTMP))
4229 		sc->sc_flags |= SCF_TIMESTAMP;
4230 	if ((tb.t_flags & (TF_RCVD_SCALE|TF_REQ_SCALE)) ==
4231 	    (TF_RCVD_SCALE|TF_REQ_SCALE)) {
4232 		sc->sc_requested_s_scale = tb.requested_s_scale;
4233 		sc->sc_request_r_scale = 0;
4234 		/*
4235 		 * Pick the smallest possible scaling factor that
4236 		 * will still allow us to scale up to sb_max.
4237 		 *
4238 		 * We do this because there are broken firewalls that
4239 		 * will corrupt the window scale option, leading to
4240 		 * the other endpoint believing that our advertised
4241 		 * window is unscaled.  At scale factors larger than
4242 		 * 5 the unscaled window will drop below 1500 bytes,
4243 		 * leading to serious problems when traversing these
4244 		 * broken firewalls.
4245 		 *
4246 		 * With the default sbmax of 256K, a scale factor
4247 		 * of 3 will be chosen by this algorithm.  Those who
4248 		 * choose a larger sbmax should watch out
4249 		 * for the compatiblity problems mentioned above.
4250 		 *
4251 		 * RFC1323: The Window field in a SYN (i.e., a <SYN>
4252 		 * or <SYN,ACK>) segment itself is never scaled.
4253 		 */
4254 		while (sc->sc_request_r_scale < TCP_MAX_WINSHIFT &&
4255 		    (TCP_MAXWIN << sc->sc_request_r_scale) < sb_max)
4256 			sc->sc_request_r_scale++;
4257 	} else {
4258 		sc->sc_requested_s_scale = 15;
4259 		sc->sc_request_r_scale = 15;
4260 	}
4261 	if ((tb.t_flags & TF_SACK_PERMIT) && tcp_do_sack)
4262 		sc->sc_flags |= SCF_SACK_PERMIT;
4263 
4264 	/*
4265 	 * ECN setup packet received.
4266 	 */
4267 	if ((th->th_flags & (TH_ECE|TH_CWR)) && tcp_do_ecn)
4268 		sc->sc_flags |= SCF_ECN_PERMIT;
4269 
4270 #ifdef TCP_SIGNATURE
4271 	if (tb.t_flags & TF_SIGNATURE)
4272 		sc->sc_flags |= SCF_SIGNATURE;
4273 #endif
4274 	sc->sc_tp = tp;
4275 	m_freem(m);
4276 	if (syn_cache_respond(sc) == 0) {
4277 		uint64_t *tcps = TCP_STAT_GETREF();
4278 		tcps[TCP_STAT_SNDACKS]++;
4279 		tcps[TCP_STAT_SNDTOTAL]++;
4280 		TCP_STAT_PUTREF();
4281 		syn_cache_insert(sc, tp);
4282 	} else {
4283 		s = splsoftnet();
4284 		/*
4285 		 * syn_cache_put() will try to schedule the timer, so
4286 		 * we need to initialize it
4287 		 */
4288 		syn_cache_timer_arm(sc);
4289 		syn_cache_put(sc);
4290 		splx(s);
4291 		TCP_STATINC(TCP_STAT_SC_DROPPED);
4292 	}
4293 	return 1;
4294 }
4295 
4296 /*
4297  * syn_cache_respond: (re)send SYN+ACK.
4298  *
4299  * Returns 0 on success.
4300  */
4301 
4302 int
4303 syn_cache_respond(struct syn_cache *sc)
4304 {
4305 #ifdef INET6
4306 	struct rtentry *rt = NULL;
4307 #endif
4308 	struct route *ro;
4309 	u_int8_t *optp;
4310 	int optlen, error;
4311 	u_int16_t tlen;
4312 	struct ip *ip = NULL;
4313 #ifdef INET6
4314 	struct ip6_hdr *ip6 = NULL;
4315 #endif
4316 	struct tcpcb *tp;
4317 	struct tcphdr *th;
4318 	struct mbuf *m;
4319 	u_int hlen;
4320 #ifdef TCP_SIGNATURE
4321 	struct secasvar *sav = NULL;
4322 	u_int8_t *sigp = NULL;
4323 #endif
4324 
4325 	ro = &sc->sc_route;
4326 	switch (sc->sc_src.sa.sa_family) {
4327 	case AF_INET:
4328 		hlen = sizeof(struct ip);
4329 		break;
4330 #ifdef INET6
4331 	case AF_INET6:
4332 		hlen = sizeof(struct ip6_hdr);
4333 		break;
4334 #endif
4335 	default:
4336 		return EAFNOSUPPORT;
4337 	}
4338 
4339 	/* Worst case scanario, since we don't know the option size yet. */
4340 	tlen = hlen + sizeof(struct tcphdr) + MAX_TCPOPTLEN;
4341 	KASSERT(max_linkhdr + tlen <= MCLBYTES);
4342 
4343 	/*
4344 	 * Create the IP+TCP header from scratch.
4345 	 */
4346 	MGETHDR(m, M_DONTWAIT, MT_DATA);
4347 	if (m && (max_linkhdr + tlen) > MHLEN) {
4348 		MCLGET(m, M_DONTWAIT);
4349 		if ((m->m_flags & M_EXT) == 0) {
4350 			m_freem(m);
4351 			m = NULL;
4352 		}
4353 	}
4354 	if (m == NULL)
4355 		return ENOBUFS;
4356 	MCLAIM(m, &tcp_tx_mowner);
4357 
4358 	tp = sc->sc_tp;
4359 
4360 	/* Fixup the mbuf. */
4361 	m->m_data += max_linkhdr;
4362 	m_reset_rcvif(m);
4363 	memset(mtod(m, void *), 0, tlen);
4364 
4365 	switch (sc->sc_src.sa.sa_family) {
4366 	case AF_INET:
4367 		ip = mtod(m, struct ip *);
4368 		ip->ip_v = 4;
4369 		ip->ip_dst = sc->sc_src.sin.sin_addr;
4370 		ip->ip_src = sc->sc_dst.sin.sin_addr;
4371 		ip->ip_p = IPPROTO_TCP;
4372 		th = (struct tcphdr *)(ip + 1);
4373 		th->th_dport = sc->sc_src.sin.sin_port;
4374 		th->th_sport = sc->sc_dst.sin.sin_port;
4375 		break;
4376 #ifdef INET6
4377 	case AF_INET6:
4378 		ip6 = mtod(m, struct ip6_hdr *);
4379 		ip6->ip6_vfc = IPV6_VERSION;
4380 		ip6->ip6_dst = sc->sc_src.sin6.sin6_addr;
4381 		ip6->ip6_src = sc->sc_dst.sin6.sin6_addr;
4382 		ip6->ip6_nxt = IPPROTO_TCP;
4383 		/* ip6_plen will be updated in ip6_output() */
4384 		th = (struct tcphdr *)(ip6 + 1);
4385 		th->th_dport = sc->sc_src.sin6.sin6_port;
4386 		th->th_sport = sc->sc_dst.sin6.sin6_port;
4387 		break;
4388 #endif
4389 	default:
4390 		panic("%s: impossible (1)", __func__);
4391 	}
4392 
4393 	th->th_seq = htonl(sc->sc_iss);
4394 	th->th_ack = htonl(sc->sc_irs + 1);
4395 	th->th_flags = TH_SYN|TH_ACK;
4396 	th->th_win = htons(sc->sc_win);
4397 	/* th_x2, th_sum, th_urp already 0 from memset */
4398 
4399 	/* Tack on the TCP options. */
4400 	optp = (u_int8_t *)(th + 1);
4401 	optlen = 0;
4402 	*optp++ = TCPOPT_MAXSEG;
4403 	*optp++ = TCPOLEN_MAXSEG;
4404 	*optp++ = (sc->sc_ourmaxseg >> 8) & 0xff;
4405 	*optp++ = sc->sc_ourmaxseg & 0xff;
4406 	optlen += TCPOLEN_MAXSEG;
4407 
4408 	if (sc->sc_request_r_scale != 15) {
4409 		*((u_int32_t *)optp) = htonl(TCPOPT_NOP << 24 |
4410 		    TCPOPT_WINDOW << 16 | TCPOLEN_WINDOW << 8 |
4411 		    sc->sc_request_r_scale);
4412 		optp += TCPOLEN_WINDOW + TCPOLEN_NOP;
4413 		optlen += TCPOLEN_WINDOW + TCPOLEN_NOP;
4414 	}
4415 
4416 	if (sc->sc_flags & SCF_SACK_PERMIT) {
4417 		/* Let the peer know that we will SACK. */
4418 		*optp++ = TCPOPT_SACK_PERMITTED;
4419 		*optp++ = TCPOLEN_SACK_PERMITTED;
4420 		optlen += TCPOLEN_SACK_PERMITTED;
4421 	}
4422 
4423 	if (sc->sc_flags & SCF_TIMESTAMP) {
4424 		while (optlen % 4 != 2) {
4425 			optlen += TCPOLEN_NOP;
4426 			*optp++ = TCPOPT_NOP;
4427 		}
4428 		*optp++ = TCPOPT_TIMESTAMP;
4429 		*optp++ = TCPOLEN_TIMESTAMP;
4430 		u_int32_t *lp = (u_int32_t *)(optp);
4431 		/* Form timestamp option as shown in appendix A of RFC 1323. */
4432 		*lp++ = htonl(SYN_CACHE_TIMESTAMP(sc));
4433 		*lp   = htonl(sc->sc_timestamp);
4434 		optp += TCPOLEN_TIMESTAMP - 2;
4435 		optlen += TCPOLEN_TIMESTAMP;
4436 	}
4437 
4438 #ifdef TCP_SIGNATURE
4439 	if (sc->sc_flags & SCF_SIGNATURE) {
4440 		sav = tcp_signature_getsav(m);
4441 		if (sav == NULL) {
4442 			m_freem(m);
4443 			return EPERM;
4444 		}
4445 
4446 		*optp++ = TCPOPT_SIGNATURE;
4447 		*optp++ = TCPOLEN_SIGNATURE;
4448 		sigp = optp;
4449 		memset(optp, 0, TCP_SIGLEN);
4450 		optp += TCP_SIGLEN;
4451 		optlen += TCPOLEN_SIGNATURE;
4452 	}
4453 #endif
4454 
4455 	/*
4456 	 * Terminate and pad TCP options to a 4 byte boundary.
4457 	 *
4458 	 * According to RFC793: "The content of the header beyond the
4459 	 * End-of-Option option must be header padding (i.e., zero)."
4460 	 * And later: "The padding is composed of zeros."
4461 	 */
4462 	if (optlen % 4) {
4463 		optlen += TCPOLEN_EOL;
4464 		*optp++ = TCPOPT_EOL;
4465 	}
4466 	while (optlen % 4) {
4467 		optlen += TCPOLEN_PAD;
4468 		*optp++ = TCPOPT_PAD;
4469 	}
4470 
4471 	/* Compute the actual values now that we've added the options. */
4472 	tlen = hlen + sizeof(struct tcphdr) + optlen;
4473 	m->m_len = m->m_pkthdr.len = tlen;
4474 	th->th_off = (sizeof(struct tcphdr) + optlen) >> 2;
4475 
4476 #ifdef TCP_SIGNATURE
4477 	if (sav) {
4478 		(void)tcp_signature(m, th, hlen, sav, sigp);
4479 		key_sa_recordxfer(sav, m);
4480 		KEY_SA_UNREF(&sav);
4481 	}
4482 #endif
4483 
4484 	/*
4485 	 * Send ECN SYN-ACK setup packet.
4486 	 * Routes can be asymetric, so, even if we receive a packet
4487 	 * with ECE and CWR set, we must not assume no one will block
4488 	 * the ECE packet we are about to send.
4489 	 */
4490 	if ((sc->sc_flags & SCF_ECN_PERMIT) && tp &&
4491 	    SEQ_GEQ(tp->snd_nxt, tp->snd_max)) {
4492 		th->th_flags |= TH_ECE;
4493 		TCP_STATINC(TCP_STAT_ECN_SHS);
4494 
4495 		/*
4496 		 * draft-ietf-tcpm-ecnsyn-00.txt
4497 		 *
4498 		 * "[...] a TCP node MAY respond to an ECN-setup
4499 		 * SYN packet by setting ECT in the responding
4500 		 * ECN-setup SYN/ACK packet, indicating to routers
4501 		 * that the SYN/ACK packet is ECN-Capable.
4502 		 * This allows a congested router along the path
4503 		 * to mark the packet instead of dropping the
4504 		 * packet as an indication of congestion."
4505 		 *
4506 		 * "[...] There can be a great benefit in setting
4507 		 * an ECN-capable codepoint in SYN/ACK packets [...]
4508 		 * Congestion is  most likely to occur in
4509 		 * the server-to-client direction.  As a result,
4510 		 * setting an ECN-capable codepoint in SYN/ACK
4511 		 * packets can reduce the occurence of three-second
4512 		 * retransmit timeouts resulting from the drop
4513 		 * of SYN/ACK packets."
4514 		 *
4515 		 * Page 4 and 6, January 2006.
4516 		 */
4517 
4518 		switch (sc->sc_src.sa.sa_family) {
4519 		case AF_INET:
4520 			ip->ip_tos |= IPTOS_ECN_ECT0;
4521 			break;
4522 #ifdef INET6
4523 		case AF_INET6:
4524 			ip6->ip6_flow |= htonl(IPTOS_ECN_ECT0 << 20);
4525 			break;
4526 #endif
4527 		}
4528 		TCP_STATINC(TCP_STAT_ECN_ECT);
4529 	}
4530 
4531 
4532 	/*
4533 	 * Compute the packet's checksum.
4534 	 *
4535 	 * Fill in some straggling IP bits.  Note the stack expects
4536 	 * ip_len to be in host order, for convenience.
4537 	 */
4538 	switch (sc->sc_src.sa.sa_family) {
4539 	case AF_INET:
4540 		ip->ip_len = htons(tlen - hlen);
4541 		th->th_sum = 0;
4542 		th->th_sum = in4_cksum(m, IPPROTO_TCP, hlen, tlen - hlen);
4543 		ip->ip_len = htons(tlen);
4544 		ip->ip_ttl = ip_defttl;
4545 		/* XXX tos? */
4546 		break;
4547 #ifdef INET6
4548 	case AF_INET6:
4549 		ip6->ip6_plen = htons(tlen - hlen);
4550 		th->th_sum = 0;
4551 		th->th_sum = in6_cksum(m, IPPROTO_TCP, hlen, tlen - hlen);
4552 		ip6->ip6_vfc &= ~IPV6_VERSION_MASK;
4553 		ip6->ip6_vfc |= IPV6_VERSION;
4554 		ip6->ip6_plen = htons(tlen - hlen);
4555 		/* ip6_hlim will be initialized afterwards */
4556 		/* XXX flowlabel? */
4557 		break;
4558 #endif
4559 	}
4560 
4561 	/* XXX use IPsec policy on listening socket, on SYN ACK */
4562 	tp = sc->sc_tp;
4563 
4564 	switch (sc->sc_src.sa.sa_family) {
4565 	case AF_INET:
4566 		error = ip_output(m, sc->sc_ipopts, ro,
4567 		    (ip_mtudisc ? IP_MTUDISC : 0),
4568 		    NULL, tp ? tp->t_inpcb : NULL);
4569 		break;
4570 #ifdef INET6
4571 	case AF_INET6:
4572 		ip6->ip6_hlim = in6_selecthlim(NULL,
4573 		    (rt = rtcache_validate(ro)) != NULL ? rt->rt_ifp : NULL);
4574 		rtcache_unref(rt, ro);
4575 
4576 		error = ip6_output(m, NULL /*XXX*/, ro, 0, NULL,
4577 		    tp ? tp->t_in6pcb : NULL, NULL);
4578 		break;
4579 #endif
4580 	default:
4581 		panic("%s: impossible (2)", __func__);
4582 	}
4583 
4584 	return error;
4585 }
4586