-
Notifications
You must be signed in to change notification settings - Fork 24
/
tcp_input.c
4840 lines (4445 loc) · 124 KB
/
tcp_input.c
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
/* $NetBSD: tcp_input.c,v 1.314 2011/05/25 23:20:57 gdt Exp $ */
/*
* Copyright (C) 1995, 1996, 1997, and 1998 WIDE Project.
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the project nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE PROJECT AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE PROJECT OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*/
/*
* @(#)COPYRIGHT 1.1 (NRL) 17 January 1995
*
* NRL grants permission for redistribution and use in source and binary
* forms, with or without modification, of the software and documentation
* created at NRL provided that the following conditions are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgements:
* This product includes software developed by the University of
* California, Berkeley and its contributors.
* This product includes software developed at the Information
* Technology Division, US Naval Research Laboratory.
* 4. Neither the name of the NRL nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THE SOFTWARE PROVIDED BY NRL IS PROVIDED BY NRL AND CONTRIBUTORS ``AS
* IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A
* PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL NRL OR
* CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL,
* EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO,
* PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR
* PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF
* LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING
* NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
* The views and conclusions contained in the software and documentation
* are those of the authors and should not be interpreted as representing
* official policies, either expressed or implied, of the US Naval
* Research Laboratory (NRL).
*/
/*-
* Copyright (c) 1997, 1998, 1999, 2001, 2005, 2006,
* 2011 The NetBSD Foundation, Inc.
* All rights reserved.
*
* This code is derived from software contributed to The NetBSD Foundation
* by Coyote Point Systems, Inc.
* This code is derived from software contributed to The NetBSD Foundation
* by Jason R. Thorpe and Kevin M. Lahey of the Numerical Aerospace Simulation
* Facility, NASA Ames Research Center.
* This code is derived from software contributed to The NetBSD Foundation
* by Charles M. Hannum.
* This code is derived from software contributed to The NetBSD Foundation
* by Rui Paulo.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
* ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
* TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
* PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
* BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
* CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
* SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
* INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
* CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
* ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
* POSSIBILITY OF SUCH DAMAGE.
*/
/*
* Copyright (c) 1982, 1986, 1988, 1990, 1993, 1994, 1995
* The Regents of the University of California. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. Neither the name of the University nor the names of its contributors
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* @(#)tcp_input.c 8.12 (Berkeley) 5/24/95
*/
/*
* TODO list for SYN cache stuff:
*
* Find room for a "state" field, which is needed to keep a
* compressed state for TIME_WAIT TCBs. It's been noted already
* that this is fairly important for very high-volume web and
* mail servers, which use a large number of short-lived
* connections.
*/
#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: tcp_input.c,v 1.314 2011/05/25 23:20:57 gdt Exp $");
#include "opt_inet.h"
#include "opt_ipsec.h"
#include "opt_inet_csum.h"
#include "opt_tcp_debug.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/errno.h>
#include <sys/syslog.h>
#include <sys/pool.h>
#include <sys/domain.h>
#include <sys/kernel.h>
#ifdef TCP_SIGNATURE
#include <sys/md5.h>
#endif
#include <sys/lwp.h> /* for lwp0 */
#include <net/if.h>
#include <net/route.h>
#include <net/if_types.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/in_pcb.h>
#include <netinet/in_var.h>
#include <netinet/ip_var.h>
#include <netinet/in_offload.h>
#ifdef INET6
#ifndef INET
#include <netinet/in.h>
#endif
#include <netinet/ip6.h>
#include <netinet6/ip6_var.h>
#include <netinet6/in6_pcb.h>
#include <netinet6/ip6_var.h>
#include <netinet6/in6_var.h>
#include <netinet/icmp6.h>
#include <netinet6/nd6.h>
#ifdef TCP_SIGNATURE
#include <netinet6/scope6_var.h>
#endif
#endif
#ifndef INET6
/* always need ip6.h for IP6_EXTHDR_GET */
#include <netinet/ip6.h>
#endif
#include <netinet/tcp.h>
#include <netinet/tcp_fsm.h>
#include <netinet/tcp_seq.h>
#include <netinet/tcp_timer.h>
#include <netinet/tcp_var.h>
#include <netinet/tcp_private.h>
#include <netinet/tcpip.h>
#include <netinet/tcp_congctl.h>
#include <netinet/tcp_debug.h>
#include <machine/stdarg.h>
#ifdef IPSEC
#include <netinet6/ipsec.h>
#include <netinet6/ipsec_private.h>
#include <netkey/key.h>
#endif /*IPSEC*/
#ifdef INET6
#include "faith.h"
#if defined(NFAITH) && NFAITH > 0
#include <net/if_faith.h>
#endif
#endif /* IPSEC */
#ifdef FAST_IPSEC
#include <netipsec/ipsec.h>
#include <netipsec/ipsec_var.h>
#include <netipsec/ipsec_private.h>
#include <netipsec/key.h>
#ifdef INET6
#include <netipsec/ipsec6.h>
#endif
#endif /* FAST_IPSEC*/
#include <netinet/tcp_vtw.h>
int tcprexmtthresh = 3;
int tcp_log_refused;
int tcp_do_autorcvbuf = 1;
int tcp_autorcvbuf_inc = 16 * 1024;
int tcp_autorcvbuf_max = 256 * 1024;
int tcp_msl = (TCPTV_MSL / PR_SLOWHZ);
static int tcp_rst_ppslim_count = 0;
static struct timeval tcp_rst_ppslim_last;
static int tcp_ackdrop_ppslim_count = 0;
static struct timeval tcp_ackdrop_ppslim_last;
#define TCP_PAWS_IDLE (24U * 24 * 60 * 60 * PR_SLOWHZ)
/* for modulo comparisons of timestamps */
#define TSTMP_LT(a,b) ((int)((a)-(b)) < 0)
#define TSTMP_GEQ(a,b) ((int)((a)-(b)) >= 0)
/*
* Neighbor Discovery, Neighbor Unreachability Detection Upper layer hint.
*/
#ifdef INET6
static inline void
nd6_hint(struct tcpcb *tp)
{
struct rtentry *rt;
if (tp != NULL && tp->t_in6pcb != NULL && tp->t_family == AF_INET6 &&
(rt = rtcache_validate(&tp->t_in6pcb->in6p_route)) != NULL)
nd6_nud_hint(rt, NULL, 0);
}
#else
static inline void
nd6_hint(struct tcpcb *tp)
{
}
#endif
/*
* Compute ACK transmission behavior. Delay the ACK unless
* we have already delayed an ACK (must send an ACK every two segments).
* We also ACK immediately if we received a PUSH and the ACK-on-PUSH
* option is enabled.
*/
static void
tcp_setup_ack(struct tcpcb *tp, const struct tcphdr *th)
{
if (tp->t_flags & TF_DELACK ||
(tcp_ack_on_push && th->th_flags & TH_PUSH))
tp->t_flags |= TF_ACKNOW;
else
TCP_SET_DELACK(tp);
}
static void
icmp_check(struct tcpcb *tp, const struct tcphdr *th, int acked)
{
/*
* If we had a pending ICMP message that refers to data that have
* just been acknowledged, disregard the recorded ICMP message.
*/
if ((tp->t_flags & TF_PMTUD_PEND) &&
SEQ_GT(th->th_ack, tp->t_pmtud_th_seq))
tp->t_flags &= ~TF_PMTUD_PEND;
/*
* Keep track of the largest chunk of data
* acknowledged since last PMTU update
*/
if (tp->t_pmtud_mss_acked < acked)
tp->t_pmtud_mss_acked = acked;
}
/*
* Convert TCP protocol fields to host order for easier processing.
*/
static void
tcp_fields_to_host(struct tcphdr *th)
{
NTOHL(th->th_seq);
NTOHL(th->th_ack);
NTOHS(th->th_win);
NTOHS(th->th_urp);
}
/*
* ... and reverse the above.
*/
static void
tcp_fields_to_net(struct tcphdr *th)
{
HTONL(th->th_seq);
HTONL(th->th_ack);
HTONS(th->th_win);
HTONS(th->th_urp);
}
#ifdef TCP_CSUM_COUNTERS
#include <sys/device.h>
#if defined(INET)
extern struct evcnt tcp_hwcsum_ok;
extern struct evcnt tcp_hwcsum_bad;
extern struct evcnt tcp_hwcsum_data;
extern struct evcnt tcp_swcsum;
#endif /* defined(INET) */
#if defined(INET6)
extern struct evcnt tcp6_hwcsum_ok;
extern struct evcnt tcp6_hwcsum_bad;
extern struct evcnt tcp6_hwcsum_data;
extern struct evcnt tcp6_swcsum;
#endif /* defined(INET6) */
#define TCP_CSUM_COUNTER_INCR(ev) (ev)->ev_count++
#else
#define TCP_CSUM_COUNTER_INCR(ev) /* nothing */
#endif /* TCP_CSUM_COUNTERS */
#ifdef TCP_REASS_COUNTERS
#include <sys/device.h>
extern struct evcnt tcp_reass_;
extern struct evcnt tcp_reass_empty;
extern struct evcnt tcp_reass_iteration[8];
extern struct evcnt tcp_reass_prependfirst;
extern struct evcnt tcp_reass_prepend;
extern struct evcnt tcp_reass_insert;
extern struct evcnt tcp_reass_inserttail;
extern struct evcnt tcp_reass_append;
extern struct evcnt tcp_reass_appendtail;
extern struct evcnt tcp_reass_overlaptail;
extern struct evcnt tcp_reass_overlapfront;
extern struct evcnt tcp_reass_segdup;
extern struct evcnt tcp_reass_fragdup;
#define TCP_REASS_COUNTER_INCR(ev) (ev)->ev_count++
#else
#define TCP_REASS_COUNTER_INCR(ev) /* nothing */
#endif /* TCP_REASS_COUNTERS */
static int tcp_reass(struct tcpcb *, const struct tcphdr *, struct mbuf *,
int *);
static int tcp_dooptions(struct tcpcb *, const u_char *, int,
struct tcphdr *, struct mbuf *, int, struct tcp_opt_info *);
#ifdef INET
static void tcp4_log_refused(const struct ip *, const struct tcphdr *);
#endif
#ifdef INET6
static void tcp6_log_refused(const struct ip6_hdr *, const struct tcphdr *);
#endif
#define TRAVERSE(x) while ((x)->m_next) (x) = (x)->m_next
#if defined(MBUFTRACE)
struct mowner tcp_reass_mowner = MOWNER_INIT("tcp", "reass");
#endif /* defined(MBUFTRACE) */
static struct pool tcpipqent_pool;
void
tcpipqent_init(void)
{
pool_init(&tcpipqent_pool, sizeof(struct ipqent), 0, 0, 0, "tcpipqepl",
NULL, IPL_VM);
}
struct ipqent *
tcpipqent_alloc(void)
{
struct ipqent *ipqe;
int s;
s = splvm();
ipqe = pool_get(&tcpipqent_pool, PR_NOWAIT);
splx(s);
return ipqe;
}
void
tcpipqent_free(struct ipqent *ipqe)
{
int s;
s = splvm();
pool_put(&tcpipqent_pool, ipqe);
splx(s);
}
static int
tcp_reass(struct tcpcb *tp, const struct tcphdr *th, struct mbuf *m, int *tlen)
{
struct ipqent *p, *q, *nq, *tiqe = NULL;
struct socket *so = NULL;
int pkt_flags;
tcp_seq pkt_seq;
unsigned pkt_len;
u_long rcvpartdupbyte = 0;
u_long rcvoobyte;
#ifdef TCP_REASS_COUNTERS
u_int count = 0;
#endif
uint64_t *tcps;
if (tp->t_inpcb)
so = tp->t_inpcb->inp_socket;
#ifdef INET6
else if (tp->t_in6pcb)
so = tp->t_in6pcb->in6p_socket;
#endif
TCP_REASS_LOCK_CHECK(tp);
/*
* Call with th==0 after become established to
* force pre-ESTABLISHED data up to user socket.
*/
if (th == 0)
goto present;
m_claimm(m, &tcp_reass_mowner);
rcvoobyte = *tlen;
/*
* Copy these to local variables because the tcpiphdr
* gets munged while we are collapsing mbufs.
*/
pkt_seq = th->th_seq;
pkt_len = *tlen;
pkt_flags = th->th_flags;
TCP_REASS_COUNTER_INCR(&tcp_reass_);
if ((p = TAILQ_LAST(&tp->segq, ipqehead)) != NULL) {
/*
* When we miss a packet, the vast majority of time we get
* packets that follow it in order. So optimize for that.
*/
if (pkt_seq == p->ipqe_seq + p->ipqe_len) {
p->ipqe_len += pkt_len;
p->ipqe_flags |= pkt_flags;
m_cat(p->ipre_mlast, m);
TRAVERSE(p->ipre_mlast);
m = NULL;
tiqe = p;
TAILQ_REMOVE(&tp->timeq, p, ipqe_timeq);
TCP_REASS_COUNTER_INCR(&tcp_reass_appendtail);
goto skip_replacement;
}
/*
* While we're here, if the pkt is completely beyond
* anything we have, just insert it at the tail.
*/
if (SEQ_GT(pkt_seq, p->ipqe_seq + p->ipqe_len)) {
TCP_REASS_COUNTER_INCR(&tcp_reass_inserttail);
goto insert_it;
}
}
q = TAILQ_FIRST(&tp->segq);
if (q != NULL) {
/*
* If this segment immediately precedes the first out-of-order
* block, simply slap the segment in front of it and (mostly)
* skip the complicated logic.
*/
if (pkt_seq + pkt_len == q->ipqe_seq) {
q->ipqe_seq = pkt_seq;
q->ipqe_len += pkt_len;
q->ipqe_flags |= pkt_flags;
m_cat(m, q->ipqe_m);
q->ipqe_m = m;
q->ipre_mlast = m; /* last mbuf may have changed */
TRAVERSE(q->ipre_mlast);
tiqe = q;
TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
TCP_REASS_COUNTER_INCR(&tcp_reass_prependfirst);
goto skip_replacement;
}
} else {
TCP_REASS_COUNTER_INCR(&tcp_reass_empty);
}
/*
* Find a segment which begins after this one does.
*/
for (p = NULL; q != NULL; q = nq) {
nq = TAILQ_NEXT(q, ipqe_q);
#ifdef TCP_REASS_COUNTERS
count++;
#endif
/*
* If the received segment is just right after this
* fragment, merge the two together and then check
* for further overlaps.
*/
if (q->ipqe_seq + q->ipqe_len == pkt_seq) {
#ifdef TCPREASS_DEBUG
printf("tcp_reass[%p]: concat %u:%u(%u) to %u:%u(%u)\n",
tp, pkt_seq, pkt_seq + pkt_len, pkt_len,
q->ipqe_seq, q->ipqe_seq + q->ipqe_len, q->ipqe_len);
#endif
pkt_len += q->ipqe_len;
pkt_flags |= q->ipqe_flags;
pkt_seq = q->ipqe_seq;
m_cat(q->ipre_mlast, m);
TRAVERSE(q->ipre_mlast);
m = q->ipqe_m;
TCP_REASS_COUNTER_INCR(&tcp_reass_append);
goto free_ipqe;
}
/*
* If the received segment is completely past this
* fragment, we need to go the next fragment.
*/
if (SEQ_LT(q->ipqe_seq + q->ipqe_len, pkt_seq)) {
p = q;
continue;
}
/*
* If the fragment is past the received segment,
* it (or any following) can't be concatenated.
*/
if (SEQ_GT(q->ipqe_seq, pkt_seq + pkt_len)) {
TCP_REASS_COUNTER_INCR(&tcp_reass_insert);
break;
}
/*
* We've received all the data in this segment before.
* mark it as a duplicate and return.
*/
if (SEQ_LEQ(q->ipqe_seq, pkt_seq) &&
SEQ_GEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) {
tcps = TCP_STAT_GETREF();
tcps[TCP_STAT_RCVDUPPACK]++;
tcps[TCP_STAT_RCVDUPBYTE] += pkt_len;
TCP_STAT_PUTREF();
tcp_new_dsack(tp, pkt_seq, pkt_len);
m_freem(m);
if (tiqe != NULL) {
tcpipqent_free(tiqe);
}
TCP_REASS_COUNTER_INCR(&tcp_reass_segdup);
goto out;
}
/*
* Received segment completely overlaps this fragment
* so we drop the fragment (this keeps the temporal
* ordering of segments correct).
*/
if (SEQ_GEQ(q->ipqe_seq, pkt_seq) &&
SEQ_LEQ(q->ipqe_seq + q->ipqe_len, pkt_seq + pkt_len)) {
rcvpartdupbyte += q->ipqe_len;
m_freem(q->ipqe_m);
TCP_REASS_COUNTER_INCR(&tcp_reass_fragdup);
goto free_ipqe;
}
/*
* RX'ed segment extends past the end of the
* fragment. Drop the overlapping bytes. Then
* merge the fragment and segment then treat as
* a longer received packet.
*/
if (SEQ_LT(q->ipqe_seq, pkt_seq) &&
SEQ_GT(q->ipqe_seq + q->ipqe_len, pkt_seq)) {
int overlap = q->ipqe_seq + q->ipqe_len - pkt_seq;
#ifdef TCPREASS_DEBUG
printf("tcp_reass[%p]: trim starting %d bytes of %u:%u(%u)\n",
tp, overlap,
pkt_seq, pkt_seq + pkt_len, pkt_len);
#endif
m_adj(m, overlap);
rcvpartdupbyte += overlap;
m_cat(q->ipre_mlast, m);
TRAVERSE(q->ipre_mlast);
m = q->ipqe_m;
pkt_seq = q->ipqe_seq;
pkt_len += q->ipqe_len - overlap;
rcvoobyte -= overlap;
TCP_REASS_COUNTER_INCR(&tcp_reass_overlaptail);
goto free_ipqe;
}
/*
* RX'ed segment extends past the front of the
* fragment. Drop the overlapping bytes on the
* received packet. The packet will then be
* contatentated with this fragment a bit later.
*/
if (SEQ_GT(q->ipqe_seq, pkt_seq) &&
SEQ_LT(q->ipqe_seq, pkt_seq + pkt_len)) {
int overlap = pkt_seq + pkt_len - q->ipqe_seq;
#ifdef TCPREASS_DEBUG
printf("tcp_reass[%p]: trim trailing %d bytes of %u:%u(%u)\n",
tp, overlap,
pkt_seq, pkt_seq + pkt_len, pkt_len);
#endif
m_adj(m, -overlap);
pkt_len -= overlap;
rcvpartdupbyte += overlap;
TCP_REASS_COUNTER_INCR(&tcp_reass_overlapfront);
rcvoobyte -= overlap;
}
/*
* If the received segment immediates precedes this
* fragment then tack the fragment onto this segment
* and reinsert the data.
*/
if (q->ipqe_seq == pkt_seq + pkt_len) {
#ifdef TCPREASS_DEBUG
printf("tcp_reass[%p]: append %u:%u(%u) to %u:%u(%u)\n",
tp, q->ipqe_seq, q->ipqe_seq + q->ipqe_len, q->ipqe_len,
pkt_seq, pkt_seq + pkt_len, pkt_len);
#endif
pkt_len += q->ipqe_len;
pkt_flags |= q->ipqe_flags;
m_cat(m, q->ipqe_m);
TAILQ_REMOVE(&tp->segq, q, ipqe_q);
TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
tp->t_segqlen--;
KASSERT(tp->t_segqlen >= 0);
KASSERT(tp->t_segqlen != 0 ||
(TAILQ_EMPTY(&tp->segq) &&
TAILQ_EMPTY(&tp->timeq)));
if (tiqe == NULL) {
tiqe = q;
} else {
tcpipqent_free(q);
}
TCP_REASS_COUNTER_INCR(&tcp_reass_prepend);
break;
}
/*
* If the fragment is before the segment, remember it.
* When this loop is terminated, p will contain the
* pointer to fragment that is right before the received
* segment.
*/
if (SEQ_LEQ(q->ipqe_seq, pkt_seq))
p = q;
continue;
/*
* This is a common operation. It also will allow
* to save doing a malloc/free in most instances.
*/
free_ipqe:
TAILQ_REMOVE(&tp->segq, q, ipqe_q);
TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
tp->t_segqlen--;
KASSERT(tp->t_segqlen >= 0);
KASSERT(tp->t_segqlen != 0 ||
(TAILQ_EMPTY(&tp->segq) && TAILQ_EMPTY(&tp->timeq)));
if (tiqe == NULL) {
tiqe = q;
} else {
tcpipqent_free(q);
}
}
#ifdef TCP_REASS_COUNTERS
if (count > 7)
TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[0]);
else if (count > 0)
TCP_REASS_COUNTER_INCR(&tcp_reass_iteration[count]);
#endif
insert_it:
/*
* Allocate a new queue entry since the received segment did not
* collapse onto any other out-of-order block; thus we are allocating
* a new block. If it had collapsed, tiqe would not be NULL and
* we would be reusing it.
* XXX If we can't, just drop the packet. XXX
*/
if (tiqe == NULL) {
tiqe = tcpipqent_alloc();
if (tiqe == NULL) {
TCP_STATINC(TCP_STAT_RCVMEMDROP);
m_freem(m);
goto out;
}
}
/*
* Update the counters.
*/
tcps = TCP_STAT_GETREF();
tcps[TCP_STAT_RCVOOPACK]++;
tcps[TCP_STAT_RCVOOBYTE] += rcvoobyte;
if (rcvpartdupbyte) {
tcps[TCP_STAT_RCVPARTDUPPACK]++;
tcps[TCP_STAT_RCVPARTDUPBYTE] += rcvpartdupbyte;
}
TCP_STAT_PUTREF();
/*
* Insert the new fragment queue entry into both queues.
*/
tiqe->ipqe_m = m;
tiqe->ipre_mlast = m;
tiqe->ipqe_seq = pkt_seq;
tiqe->ipqe_len = pkt_len;
tiqe->ipqe_flags = pkt_flags;
if (p == NULL) {
TAILQ_INSERT_HEAD(&tp->segq, tiqe, ipqe_q);
#ifdef TCPREASS_DEBUG
if (tiqe->ipqe_seq != tp->rcv_nxt)
printf("tcp_reass[%p]: insert %u:%u(%u) at front\n",
tp, pkt_seq, pkt_seq + pkt_len, pkt_len);
#endif
} else {
TAILQ_INSERT_AFTER(&tp->segq, p, tiqe, ipqe_q);
#ifdef TCPREASS_DEBUG
printf("tcp_reass[%p]: insert %u:%u(%u) after %u:%u(%u)\n",
tp, pkt_seq, pkt_seq + pkt_len, pkt_len,
p->ipqe_seq, p->ipqe_seq + p->ipqe_len, p->ipqe_len);
#endif
}
tp->t_segqlen++;
skip_replacement:
TAILQ_INSERT_HEAD(&tp->timeq, tiqe, ipqe_timeq);
present:
/*
* Present data to user, advancing rcv_nxt through
* completed sequence space.
*/
if (TCPS_HAVEESTABLISHED(tp->t_state) == 0)
goto out;
q = TAILQ_FIRST(&tp->segq);
if (q == NULL || q->ipqe_seq != tp->rcv_nxt)
goto out;
if (tp->t_state == TCPS_SYN_RECEIVED && q->ipqe_len)
goto out;
tp->rcv_nxt += q->ipqe_len;
pkt_flags = q->ipqe_flags & TH_FIN;
nd6_hint(tp);
TAILQ_REMOVE(&tp->segq, q, ipqe_q);
TAILQ_REMOVE(&tp->timeq, q, ipqe_timeq);
tp->t_segqlen--;
KASSERT(tp->t_segqlen >= 0);
KASSERT(tp->t_segqlen != 0 ||
(TAILQ_EMPTY(&tp->segq) && TAILQ_EMPTY(&tp->timeq)));
if (so->so_state & SS_CANTRCVMORE)
m_freem(q->ipqe_m);
else
sbappendstream(&so->so_rcv, q->ipqe_m);
tcpipqent_free(q);
TCP_REASS_UNLOCK(tp);
sorwakeup(so);
return (pkt_flags);
out:
TCP_REASS_UNLOCK(tp);
return (0);
}
#ifdef INET6
int
tcp6_input(struct mbuf **mp, int *offp, int proto)
{
struct mbuf *m = *mp;
/*
* draft-itojun-ipv6-tcp-to-anycast
* better place to put this in?
*/
if (m->m_flags & M_ANYCAST6) {
struct ip6_hdr *ip6;
if (m->m_len < sizeof(struct ip6_hdr)) {
if ((m = m_pullup(m, sizeof(struct ip6_hdr))) == NULL) {
TCP_STATINC(TCP_STAT_RCVSHORT);
return IPPROTO_DONE;
}
}
ip6 = mtod(m, struct ip6_hdr *);
icmp6_error(m, ICMP6_DST_UNREACH, ICMP6_DST_UNREACH_ADDR,
(char *)&ip6->ip6_dst - (char *)ip6);
return IPPROTO_DONE;
}
tcp_input(m, *offp, proto);
return IPPROTO_DONE;
}
#endif
#ifdef INET
static void
tcp4_log_refused(const struct ip *ip, const struct tcphdr *th)
{
char src[4*sizeof "123"];
char dst[4*sizeof "123"];
if (ip) {
strlcpy(src, inet_ntoa(ip->ip_src), sizeof(src));
strlcpy(dst, inet_ntoa(ip->ip_dst), sizeof(dst));
}
else {
strlcpy(src, "(unknown)", sizeof(src));
strlcpy(dst, "(unknown)", sizeof(dst));
}
log(LOG_INFO,
"Connection attempt to TCP %s:%d from %s:%d\n",
dst, ntohs(th->th_dport),
src, ntohs(th->th_sport));
}
#endif
#ifdef INET6
static void
tcp6_log_refused(const struct ip6_hdr *ip6, const struct tcphdr *th)
{
char src[INET6_ADDRSTRLEN];
char dst[INET6_ADDRSTRLEN];
if (ip6) {
strlcpy(src, ip6_sprintf(&ip6->ip6_src), sizeof(src));
strlcpy(dst, ip6_sprintf(&ip6->ip6_dst), sizeof(dst));
}
else {
strlcpy(src, "(unknown v6)", sizeof(src));
strlcpy(dst, "(unknown v6)", sizeof(dst));
}
log(LOG_INFO,
"Connection attempt to TCP [%s]:%d from [%s]:%d\n",
dst, ntohs(th->th_dport),
src, ntohs(th->th_sport));
}
#endif
/*
* Checksum extended TCP header and data.
*/
int
tcp_input_checksum(int af, struct mbuf *m, const struct tcphdr *th,
int toff, int off, int tlen)
{
/*
* XXX it's better to record and check if this mbuf is
* already checked.
*/
switch (af) {
#ifdef INET
case AF_INET:
switch (m->m_pkthdr.csum_flags &
((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_TCPv4) |
M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) {
case M_CSUM_TCPv4|M_CSUM_TCP_UDP_BAD:
TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_bad);
goto badcsum;
case M_CSUM_TCPv4|M_CSUM_DATA: {
u_int32_t hw_csum = m->m_pkthdr.csum_data;
TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_data);
if (m->m_pkthdr.csum_flags & M_CSUM_NO_PSEUDOHDR) {
const struct ip *ip =
mtod(m, const struct ip *);
hw_csum = in_cksum_phdr(ip->ip_src.s_addr,
ip->ip_dst.s_addr,
htons(hw_csum + tlen + off + IPPROTO_TCP));
}
if ((hw_csum ^ 0xffff) != 0)
goto badcsum;
break;
}
case M_CSUM_TCPv4:
/* Checksum was okay. */
TCP_CSUM_COUNTER_INCR(&tcp_hwcsum_ok);
break;
default:
/*
* Must compute it ourselves. Maybe skip checksum
* on loopback interfaces.
*/
if (__predict_true(!(m->m_pkthdr.rcvif->if_flags &
IFF_LOOPBACK) ||
tcp_do_loopback_cksum)) {
TCP_CSUM_COUNTER_INCR(&tcp_swcsum);
if (in4_cksum(m, IPPROTO_TCP, toff,
tlen + off) != 0)
goto badcsum;
}
break;
}
break;
#endif /* INET4 */
#ifdef INET6
case AF_INET6:
switch (m->m_pkthdr.csum_flags &
((m->m_pkthdr.rcvif->if_csum_flags_rx & M_CSUM_TCPv6) |
M_CSUM_TCP_UDP_BAD | M_CSUM_DATA)) {
case M_CSUM_TCPv6|M_CSUM_TCP_UDP_BAD:
TCP_CSUM_COUNTER_INCR(&tcp6_hwcsum_bad);
goto badcsum;
#if 0 /* notyet */
case M_CSUM_TCPv6|M_CSUM_DATA:
#endif
case M_CSUM_TCPv6:
/* Checksum was okay. */
TCP_CSUM_COUNTER_INCR(&tcp6_hwcsum_ok);
break;
default:
/*
* Must compute it ourselves. Maybe skip checksum
* on loopback interfaces.
*/
if (__predict_true((m->m_flags & M_LOOP) == 0 ||
tcp_do_loopback_cksum)) {
TCP_CSUM_COUNTER_INCR(&tcp6_swcsum);
if (in6_cksum(m, IPPROTO_TCP, toff,
tlen + off) != 0)
goto badcsum;
}
}
break;
#endif /* INET6 */
}
return 0;
badcsum:
TCP_STATINC(TCP_STAT_RCVBADSUM);
return -1;
}
/* When a packet arrives addressed to a vestigial tcpbp, we
* nevertheless have to respond to it per the spec.