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
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
|
/*-
* Copyright (c) 2001 McAfee, Inc.
* Copyright (c) 2006 Andre Oppermann, Internet Business Solutions AG
* All rights reserved.
*
* This software was developed for the FreeBSD Project by Jonathan Lemon
* and McAfee Research, the Security Research Division of McAfee, Inc. under
* DARPA/SPAWAR contract N66001-01-C-8035 ("CBOSS"), as part of the
* DARPA CHATS research program.
*
* 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 AUTHOR 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 AUTHOR 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.
*
* $FreeBSD$
*/
#include "opt_inet.h"
#include "opt_inet6.h"
#include "opt_ipsec.h"
#include "opt_mac.h"
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/sysctl.h>
#include <sys/lock.h>
#include <sys/mutex.h>
#include <sys/malloc.h>
#include <sys/mac.h>
#include <sys/mbuf.h>
#include <sys/md5.h>
#include <sys/proc.h> /* for proc0 declaration */
#include <sys/random.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <vm/uma.h>
#include <net/if.h>
#include <net/route.h>
#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/ip.h>
#include <netinet/in_var.h>
#include <netinet/in_pcb.h>
#include <netinet/ip_var.h>
#include <netinet/ip_options.h>
#ifdef INET6
#include <netinet/ip6.h>
#include <netinet/icmp6.h>
#include <netinet6/nd6.h>
#include <netinet6/ip6_var.h>
#include <netinet6/in6_pcb.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>
#ifdef INET6
#include <netinet6/tcp6_var.h>
#endif
#ifdef IPSEC
#include <netinet6/ipsec.h>
#ifdef INET6
#include <netinet6/ipsec6.h>
#endif
#endif /*IPSEC*/
#ifdef FAST_IPSEC
#include <netipsec/ipsec.h>
#ifdef INET6
#include <netipsec/ipsec6.h>
#endif
#include <netipsec/key.h>
#endif /*FAST_IPSEC*/
#include <machine/in_cksum.h>
static int tcp_syncookies = 1;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, syncookies, CTLFLAG_RW,
&tcp_syncookies, 0,
"Use TCP SYN cookies if the syncache overflows");
static int tcp_syncookiesonly = 0;
SYSCTL_INT(_net_inet_tcp, OID_AUTO, syncookies_only, CTLFLAG_RW,
&tcp_syncookiesonly, 0,
"Use only TCP SYN cookies");
#define SYNCOOKIE_SECRET_SIZE 8 /* dwords */
#define SYNCOOKIE_LIFETIME 16 /* seconds */
struct syncache {
TAILQ_ENTRY(syncache) sc_hash;
struct in_conninfo sc_inc; /* addresses */
u_long sc_rxttime; /* retransmit time */
u_int16_t sc_rxmits; /* retransmit counter */
u_int32_t sc_tsreflect; /* timestamp to reflect */
u_int32_t sc_ts; /* our timestamp to send */
u_int32_t sc_tsoff; /* ts offset w/ syncookies */
u_int32_t sc_flowlabel; /* IPv6 flowlabel */
tcp_seq sc_irs; /* seq from peer */
tcp_seq sc_iss; /* our ISS */
struct mbuf *sc_ipopts; /* source route */
u_int16_t sc_peer_mss; /* peer's MSS */
u_int16_t sc_wnd; /* advertised window */
u_int8_t sc_ip_ttl; /* IPv4 TTL */
u_int8_t sc_ip_tos; /* IPv4 TOS */
u_int8_t sc_requested_s_scale:4,
sc_requested_r_scale:4;
u_int8_t sc_flags;
#define SCF_NOOPT 0x01 /* no TCP options */
#define SCF_WINSCALE 0x02 /* negotiated window scaling */
#define SCF_TIMESTAMP 0x04 /* negotiated timestamps */
/* MSS is implicit */
#define SCF_UNREACH 0x10 /* icmp unreachable received */
#define SCF_SIGNATURE 0x20 /* send MD5 digests */
#define SCF_SACK 0x80 /* send SACK option */
};
struct syncache_head {
struct mtx sch_mtx;
TAILQ_HEAD(sch_head, syncache) sch_bucket;
struct callout sch_timer;
int sch_nextc;
u_int sch_length;
u_int sch_oddeven;
u_int32_t sch_secbits_odd[SYNCOOKIE_SECRET_SIZE];
u_int32_t sch_secbits_even[SYNCOOKIE_SECRET_SIZE];
u_int sch_reseed; /* time_uptime, seconds */
};
static void syncache_drop(struct syncache *, struct syncache_head *);
static void syncache_free(struct syncache *);
static void syncache_insert(struct syncache *, struct syncache_head *);
struct syncache *syncache_lookup(struct in_conninfo *, struct syncache_head **);
static int syncache_respond(struct syncache *, struct mbuf *);
static struct socket *syncache_socket(struct syncache *, struct socket *,
struct mbuf *m);
static void syncache_timer(void *);
static void syncookie_generate(struct syncache_head *, struct syncache *,
u_int32_t *);
static struct syncache
*syncookie_lookup(struct in_conninfo *, struct syncache_head *,
struct syncache *, struct tcpopt *, struct tcphdr *,
struct socket *);
/*
* Transmit the SYN,ACK fewer times than TCP_MAXRXTSHIFT specifies.
* 3 retransmits corresponds to a timeout of (1 + 2 + 4 + 8 == 15) seconds,
* the odds are that the user has given up attempting to connect by then.
*/
#define SYNCACHE_MAXREXMTS 3
/* Arbitrary values */
#define TCP_SYNCACHE_HASHSIZE 512
#define TCP_SYNCACHE_BUCKETLIMIT 30
struct tcp_syncache {
struct syncache_head *hashbase;
uma_zone_t zone;
u_int hashsize;
u_int hashmask;
u_int bucket_limit;
u_int cache_count; /* XXX: unprotected */
u_int cache_limit;
u_int rexmt_limit;
u_int hash_secret;
};
static struct tcp_syncache tcp_syncache;
SYSCTL_NODE(_net_inet_tcp, OID_AUTO, syncache, CTLFLAG_RW, 0, "TCP SYN cache");
SYSCTL_INT(_net_inet_tcp_syncache, OID_AUTO, bucketlimit, CTLFLAG_RDTUN,
&tcp_syncache.bucket_limit, 0, "Per-bucket hash limit for syncache");
SYSCTL_INT(_net_inet_tcp_syncache, OID_AUTO, cachelimit, CTLFLAG_RDTUN,
&tcp_syncache.cache_limit, 0, "Overall entry limit for syncache");
SYSCTL_INT(_net_inet_tcp_syncache, OID_AUTO, count, CTLFLAG_RD,
&tcp_syncache.cache_count, 0, "Current number of entries in syncache");
SYSCTL_INT(_net_inet_tcp_syncache, OID_AUTO, hashsize, CTLFLAG_RDTUN,
&tcp_syncache.hashsize, 0, "Size of TCP syncache hashtable");
SYSCTL_INT(_net_inet_tcp_syncache, OID_AUTO, rexmtlimit, CTLFLAG_RW,
&tcp_syncache.rexmt_limit, 0, "Limit on SYN/ACK retransmissions");
static MALLOC_DEFINE(M_SYNCACHE, "syncache", "TCP syncache");
#define SYNCACHE_HASH(inc, mask) \
((tcp_syncache.hash_secret ^ \
(inc)->inc_faddr.s_addr ^ \
((inc)->inc_faddr.s_addr >> 16) ^ \
(inc)->inc_fport ^ (inc)->inc_lport) & mask)
#define SYNCACHE_HASH6(inc, mask) \
((tcp_syncache.hash_secret ^ \
(inc)->inc6_faddr.s6_addr32[0] ^ \
(inc)->inc6_faddr.s6_addr32[3] ^ \
(inc)->inc_fport ^ (inc)->inc_lport) & mask)
#define ENDPTS_EQ(a, b) ( \
(a)->ie_fport == (b)->ie_fport && \
(a)->ie_lport == (b)->ie_lport && \
(a)->ie_faddr.s_addr == (b)->ie_faddr.s_addr && \
(a)->ie_laddr.s_addr == (b)->ie_laddr.s_addr \
)
#define ENDPTS6_EQ(a, b) (memcmp(a, b, sizeof(*a)) == 0)
#define SYNCACHE_TIMEOUT(sc, sch, co) do { \
(sc)->sc_rxmits++; \
(sc)->sc_rxttime = ticks + \
TCPTV_RTOBASE * tcp_backoff[(sc)->sc_rxmits - 1]; \
if ((sch)->sch_nextc > (sc)->sc_rxttime) \
(sch)->sch_nextc = (sc)->sc_rxttime; \
if (!TAILQ_EMPTY(&(sch)->sch_bucket) && !(co)) \
callout_reset(&(sch)->sch_timer, \
(sch)->sch_nextc - ticks, \
syncache_timer, (void *)(sch)); \
} while (0)
#define SCH_LOCK(sch) mtx_lock(&(sch)->sch_mtx)
#define SCH_UNLOCK(sch) mtx_unlock(&(sch)->sch_mtx)
#define SCH_LOCK_ASSERT(sch) mtx_assert(&(sch)->sch_mtx, MA_OWNED)
/*
* Requires the syncache entry to be already removed from the bucket list.
*/
static void
syncache_free(struct syncache *sc)
{
if (sc->sc_ipopts)
(void) m_free(sc->sc_ipopts);
uma_zfree(tcp_syncache.zone, sc);
}
void
syncache_init(void)
{
int i;
tcp_syncache.cache_count = 0;
tcp_syncache.hashsize = TCP_SYNCACHE_HASHSIZE;
tcp_syncache.bucket_limit = TCP_SYNCACHE_BUCKETLIMIT;
tcp_syncache.rexmt_limit = SYNCACHE_MAXREXMTS;
tcp_syncache.hash_secret = arc4random();
TUNABLE_INT_FETCH("net.inet.tcp.syncache.hashsize",
&tcp_syncache.hashsize);
TUNABLE_INT_FETCH("net.inet.tcp.syncache.bucketlimit",
&tcp_syncache.bucket_limit);
if (!powerof2(tcp_syncache.hashsize) || tcp_syncache.hashsize == 0) {
printf("WARNING: syncache hash size is not a power of 2.\n");
tcp_syncache.hashsize = TCP_SYNCACHE_HASHSIZE;
}
tcp_syncache.hashmask = tcp_syncache.hashsize - 1;
/* Set limits. */
tcp_syncache.cache_limit =
tcp_syncache.hashsize * tcp_syncache.bucket_limit;
TUNABLE_INT_FETCH("net.inet.tcp.syncache.cachelimit",
&tcp_syncache.cache_limit);
/* Allocate the hash table. */
MALLOC(tcp_syncache.hashbase, struct syncache_head *,
tcp_syncache.hashsize * sizeof(struct syncache_head),
M_SYNCACHE, M_WAITOK | M_ZERO);
/* Initialize the hash buckets. */
for (i = 0; i < tcp_syncache.hashsize; i++) {
TAILQ_INIT(&tcp_syncache.hashbase[i].sch_bucket);
mtx_init(&tcp_syncache.hashbase[i].sch_mtx, "tcp_sc_head",
NULL, MTX_DEF);
callout_init_mtx(&tcp_syncache.hashbase[i].sch_timer,
&tcp_syncache.hashbase[i].sch_mtx, 0);
tcp_syncache.hashbase[i].sch_length = 0;
}
/* Create the syncache entry zone. */
tcp_syncache.zone = uma_zcreate("syncache", sizeof(struct syncache),
NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
uma_zone_set_max(tcp_syncache.zone, tcp_syncache.cache_limit);
}
/*
* Inserts a syncache entry into the specified bucket row.
* Locks and unlocks the syncache_head autonomously.
*/
static void
syncache_insert(struct syncache *sc, struct syncache_head *sch)
{
struct syncache *sc2;
SCH_LOCK(sch);
/*
* Make sure that we don't overflow the per-bucket limit.
* If the bucket is full, toss the oldest element.
*/
if (sch->sch_length >= tcp_syncache.bucket_limit) {
KASSERT(!TAILQ_EMPTY(&sch->sch_bucket),
("sch->sch_length incorrect"));
sc2 = TAILQ_LAST(&sch->sch_bucket, sch_head);
syncache_drop(sc2, sch);
tcpstat.tcps_sc_bucketoverflow++;
}
/* Put it into the bucket. */
TAILQ_INSERT_HEAD(&sch->sch_bucket, sc, sc_hash);
sch->sch_length++;
/* Reinitialize the bucket row's timer. */
SYNCACHE_TIMEOUT(sc, sch, 1);
SCH_UNLOCK(sch);
tcp_syncache.cache_count++;
tcpstat.tcps_sc_added++;
}
/*
* Remove and free entry from syncache bucket row.
* Expects locked syncache head.
*/
static void
syncache_drop(struct syncache *sc, struct syncache_head *sch)
{
SCH_LOCK_ASSERT(sch);
TAILQ_REMOVE(&sch->sch_bucket, sc, sc_hash);
sch->sch_length--;
syncache_free(sc);
tcp_syncache.cache_count--;
}
/*
* Walk the timer queues, looking for SYN,ACKs that need to be retransmitted.
* If we have retransmitted an entry the maximum number of times, expire it.
* One separate timer for each bucket row.
*/
static void
syncache_timer(void *xsch)
{
struct syncache_head *sch = (struct syncache_head *)xsch;
struct syncache *sc, *nsc;
int tick = ticks;
/* NB: syncache_head has already been locked by the callout. */
SCH_LOCK_ASSERT(sch);
TAILQ_FOREACH_SAFE(sc, &sch->sch_bucket, sc_hash, nsc) {
/*
* We do not check if the listen socket still exists
* and accept the case where the listen socket may be
* gone by the time we resend the SYN/ACK. We do
* not expect this to happens often. If it does,
* then the RST will be sent by the time the remote
* host does the SYN/ACK->ACK.
*/
if (sc->sc_rxttime >= tick) {
if (sc->sc_rxttime < sch->sch_nextc)
sch->sch_nextc = sc->sc_rxttime;
continue;
}
if (sc->sc_rxmits > tcp_syncache.rexmt_limit) {
syncache_drop(sc, sch);
tcpstat.tcps_sc_stale++;
continue;
}
(void) syncache_respond(sc, NULL);
tcpstat.tcps_sc_retransmitted++;
SYNCACHE_TIMEOUT(sc, sch, 0);
}
if (!TAILQ_EMPTY(&(sch)->sch_bucket))
callout_reset(&(sch)->sch_timer, (sch)->sch_nextc - tick,
syncache_timer, (void *)(sch));
}
/*
* Find an entry in the syncache.
* Returns always with locked syncache_head plus a matching entry or NULL.
*/
struct syncache *
syncache_lookup(struct in_conninfo *inc, struct syncache_head **schp)
{
struct syncache *sc;
struct syncache_head *sch;
#ifdef INET6
if (inc->inc_isipv6) {
sch = &tcp_syncache.hashbase[
SYNCACHE_HASH6(inc, tcp_syncache.hashmask)];
*schp = sch;
SCH_LOCK(sch);
/* Circle through bucket row to find matching entry. */
TAILQ_FOREACH(sc, &sch->sch_bucket, sc_hash) {
if (ENDPTS6_EQ(&inc->inc_ie, &sc->sc_inc.inc_ie))
return (sc);
}
} else
#endif
{
sch = &tcp_syncache.hashbase[
SYNCACHE_HASH(inc, tcp_syncache.hashmask)];
*schp = sch;
SCH_LOCK(sch);
/* Circle through bucket row to find matching entry. */
TAILQ_FOREACH(sc, &sch->sch_bucket, sc_hash) {
#ifdef INET6
if (sc->sc_inc.inc_isipv6)
continue;
#endif
if (ENDPTS_EQ(&inc->inc_ie, &sc->sc_inc.inc_ie))
return (sc);
}
}
SCH_LOCK_ASSERT(*schp);
return (NULL); /* always returns with locked sch */
}
/*
* This function is called when we get a RST for a
* non-existent connection, so that we can see if the
* connection is in the syn cache. If it is, zap it.
*/
void
syncache_chkrst(struct in_conninfo *inc, struct tcphdr *th)
{
struct syncache *sc;
struct syncache_head *sch;
sc = syncache_lookup(inc, &sch); /* returns locked sch */
SCH_LOCK_ASSERT(sch);
if (sc == NULL)
goto done;
/*
* If the RST bit is set, check the sequence number to see
* if this is a valid reset segment.
* RFC 793 page 37:
* In all states except SYN-SENT, all reset (RST) segments
* are validated by checking their SEQ-fields. A reset is
* valid if its sequence number is in the window.
*
* The sequence number in the reset segment is normally an
* echo of our outgoing acknowlegement numbers, but some hosts
* send a reset with the sequence number at the rightmost edge
* of our receive window, and we have to handle this case.
*/
if (SEQ_GEQ(th->th_seq, sc->sc_irs) &&
SEQ_LEQ(th->th_seq, sc->sc_irs + sc->sc_wnd)) {
syncache_drop(sc, sch);
tcpstat.tcps_sc_reset++;
}
done:
SCH_UNLOCK(sch);
}
void
syncache_badack(struct in_conninfo *inc)
{
struct syncache *sc;
struct syncache_head *sch;
sc = syncache_lookup(inc, &sch); /* returns locked sch */
SCH_LOCK_ASSERT(sch);
if (sc != NULL) {
syncache_drop(sc, sch);
tcpstat.tcps_sc_badack++;
}
SCH_UNLOCK(sch);
}
void
syncache_unreach(struct in_conninfo *inc, struct tcphdr *th)
{
struct syncache *sc;
struct syncache_head *sch;
sc = syncache_lookup(inc, &sch); /* returns locked sch */
SCH_LOCK_ASSERT(sch);
if (sc == NULL)
goto done;
/* If the sequence number != sc_iss, then it's a bogus ICMP msg */
if (ntohl(th->th_seq) != sc->sc_iss)
goto done;
/*
* If we've rertransmitted 3 times and this is our second error,
* we remove the entry. Otherwise, we allow it to continue on.
* This prevents us from incorrectly nuking an entry during a
* spurious network outage.
*
* See tcp_notify().
*/
if ((sc->sc_flags & SCF_UNREACH) == 0 || sc->sc_rxmits < 3 + 1) {
sc->sc_flags |= SCF_UNREACH;
goto done;
}
syncache_drop(sc, sch);
tcpstat.tcps_sc_unreach++;
done:
SCH_UNLOCK(sch);
}
/*
* Build a new TCP socket structure from a syncache entry.
*/
static struct socket *
syncache_socket(struct syncache *sc, struct socket *lso, struct mbuf *m)
{
struct inpcb *inp = NULL;
struct socket *so;
struct tcpcb *tp;
NET_ASSERT_GIANT();
INP_INFO_WLOCK_ASSERT(&tcbinfo);
/*
* Ok, create the full blown connection, and set things up
* as they would have been set up if we had created the
* connection when the SYN arrived. If we can't create
* the connection, abort it.
*/
so = sonewconn(lso, SS_ISCONNECTED);
if (so == NULL) {
/*
* Drop the connection; we will send a RST if the peer
* retransmits the ACK,
*/
tcpstat.tcps_listendrop++;
goto abort2;
}
#ifdef MAC
SOCK_LOCK(so);
mac_set_socket_peer_from_mbuf(m, so);
SOCK_UNLOCK(so);
#endif
inp = sotoinpcb(so);
INP_LOCK(inp);
/* Insert new socket into PCB hash list. */
inp->inp_inc.inc_isipv6 = sc->sc_inc.inc_isipv6;
#ifdef INET6
if (sc->sc_inc.inc_isipv6) {
inp->in6p_laddr = sc->sc_inc.inc6_laddr;
} else {
inp->inp_vflag &= ~INP_IPV6;
inp->inp_vflag |= INP_IPV4;
#endif
inp->inp_laddr = sc->sc_inc.inc_laddr;
#ifdef INET6
}
#endif
inp->inp_lport = sc->sc_inc.inc_lport;
if (in_pcbinshash(inp) != 0) {
/*
* Undo the assignments above if we failed to
* put the PCB on the hash lists.
*/
#ifdef INET6
if (sc->sc_inc.inc_isipv6)
inp->in6p_laddr = in6addr_any;
else
#endif
inp->inp_laddr.s_addr = INADDR_ANY;
inp->inp_lport = 0;
goto abort;
}
#ifdef IPSEC
/* Copy old policy into new socket's. */
if (ipsec_copy_pcbpolicy(sotoinpcb(lso)->inp_sp, inp->inp_sp))
printf("syncache_socket: could not copy policy\n");
#endif
#ifdef FAST_IPSEC
/* Copy old policy into new socket's. */
if (ipsec_copy_policy(sotoinpcb(lso)->inp_sp, inp->inp_sp))
printf("syncache_socket: could not copy policy\n");
#endif
#ifdef INET6
if (sc->sc_inc.inc_isipv6) {
struct inpcb *oinp = sotoinpcb(lso);
struct in6_addr laddr6;
struct sockaddr_in6 sin6;
/*
* Inherit socket options from the listening socket.
* Note that in6p_inputopts are not (and should not be)
* copied, since it stores previously received options and is
* used to detect if each new option is different than the
* previous one and hence should be passed to a user.
* If we copied in6p_inputopts, a user would not be able to
* receive options just after calling the accept system call.
*/
inp->inp_flags |= oinp->inp_flags & INP_CONTROLOPTS;
if (oinp->in6p_outputopts)
inp->in6p_outputopts =
ip6_copypktopts(oinp->in6p_outputopts, M_NOWAIT);
sin6.sin6_family = AF_INET6;
sin6.sin6_len = sizeof(sin6);
sin6.sin6_addr = sc->sc_inc.inc6_faddr;
sin6.sin6_port = sc->sc_inc.inc_fport;
sin6.sin6_flowinfo = sin6.sin6_scope_id = 0;
laddr6 = inp->in6p_laddr;
if (IN6_IS_ADDR_UNSPECIFIED(&inp->in6p_laddr))
inp->in6p_laddr = sc->sc_inc.inc6_laddr;
if (in6_pcbconnect(inp, (struct sockaddr *)&sin6,
thread0.td_ucred)) {
inp->in6p_laddr = laddr6;
goto abort;
}
/* Override flowlabel from in6_pcbconnect. */
inp->in6p_flowinfo &= ~IPV6_FLOWLABEL_MASK;
inp->in6p_flowinfo |= sc->sc_flowlabel;
} else
#endif
{
struct in_addr laddr;
struct sockaddr_in sin;
inp->inp_options = ip_srcroute(m);
if (inp->inp_options == NULL) {
inp->inp_options = sc->sc_ipopts;
sc->sc_ipopts = NULL;
}
sin.sin_family = AF_INET;
sin.sin_len = sizeof(sin);
sin.sin_addr = sc->sc_inc.inc_faddr;
sin.sin_port = sc->sc_inc.inc_fport;
bzero((caddr_t)sin.sin_zero, sizeof(sin.sin_zero));
laddr = inp->inp_laddr;
if (inp->inp_laddr.s_addr == INADDR_ANY)
inp->inp_laddr = sc->sc_inc.inc_laddr;
if (in_pcbconnect(inp, (struct sockaddr *)&sin,
thread0.td_ucred)) {
inp->inp_laddr = laddr;
goto abort;
}
}
tp = intotcpcb(inp);
tp->t_state = TCPS_SYN_RECEIVED;
tp->iss = sc->sc_iss;
tp->irs = sc->sc_irs;
tcp_rcvseqinit(tp);
tcp_sendseqinit(tp);
tp->snd_wl1 = sc->sc_irs;
tp->rcv_up = sc->sc_irs + 1;
tp->rcv_wnd = sc->sc_wnd;
tp->rcv_adv += tp->rcv_wnd;
tp->t_flags = sototcpcb(lso)->t_flags & (TF_NOPUSH|TF_NODELAY);
if (sc->sc_flags & SCF_NOOPT)
tp->t_flags |= TF_NOOPT;
else {
if (sc->sc_flags & SCF_WINSCALE) {
tp->t_flags |= TF_REQ_SCALE|TF_RCVD_SCALE;
tp->snd_scale = sc->sc_requested_s_scale;
tp->request_r_scale = sc->sc_requested_r_scale;
}
if (sc->sc_flags & SCF_TIMESTAMP) {
tp->t_flags |= TF_REQ_TSTMP|TF_RCVD_TSTMP;
tp->ts_recent = sc->sc_tsreflect;
tp->ts_recent_age = ticks;
tp->ts_offset = sc->sc_tsoff;
}
#ifdef TCP_SIGNATURE
if (sc->sc_flags & SCF_SIGNATURE)
tp->t_flags |= TF_SIGNATURE;
#endif
if (sc->sc_flags & SCF_SACK) {
tp->sack_enable = 1;
tp->t_flags |= TF_SACK_PERMIT;
}
}
/*
* Set up MSS and get cached values from tcp_hostcache.
* This might overwrite some of the defaults we just set.
*/
tcp_mss(tp, sc->sc_peer_mss);
/*
* If the SYN,ACK was retransmitted, reset cwnd to 1 segment.
*/
if (sc->sc_rxmits > 1)
tp->snd_cwnd = tp->t_maxseg;
callout_reset(tp->tt_keep, tcp_keepinit, tcp_timer_keep, tp);
INP_UNLOCK(inp);
tcpstat.tcps_accepts++;
return (so);
abort:
INP_UNLOCK(inp);
abort2:
if (so != NULL)
soabort(so);
return (NULL);
}
/*
* This function gets called when we receive an ACK for a
* socket in the LISTEN state. We look up the connection
* in the syncache, and if its there, we pull it out of
* the cache and turn it into a full-blown connection in
* the SYN-RECEIVED state.
*/
int
syncache_expand(struct in_conninfo *inc, struct tcpopt *to, struct tcphdr *th,
struct socket **lsop, struct mbuf *m)
{
struct syncache *sc;
struct syncache_head *sch;
struct socket *so;
struct syncache scs;
/*
* Global TCP locks are held because we manipulate the PCB lists
* and create a new socket.
*/
INP_INFO_WLOCK_ASSERT(&tcbinfo);
sc = syncache_lookup(inc, &sch); /* returns locked sch */
SCH_LOCK_ASSERT(sch);
if (sc == NULL) {
/*
* There is no syncache entry, so see if this ACK is
* a returning syncookie. To do this, first:
* A. See if this socket has had a syncache entry dropped in
* the past. We don't want to accept a bogus syncookie
* if we've never received a SYN.
* B. check that the syncookie is valid. If it is, then
* cobble up a fake syncache entry, and return.
*/
if (!tcp_syncookies) {
SCH_UNLOCK(sch);
goto failed;
}
bzero(&scs, sizeof(scs));
sc = syncookie_lookup(inc, sch, &scs, to, th, *lsop);
SCH_UNLOCK(sch);
if (sc == NULL)
goto failed;
tcpstat.tcps_sc_recvcookie++;
} else {
/* Pull out the entry to unlock the bucket row. */
TAILQ_REMOVE(&sch->sch_bucket, sc, sc_hash);
sch->sch_length--;
tcp_syncache.cache_count--;
SCH_UNLOCK(sch);
}
/*
* If seg contains an ACK, but not for our SYN/ACK, send a RST.
*/
if (th->th_ack != sc->sc_iss + 1)
goto failed;
so = syncache_socket(sc, *lsop, m);
if (so == NULL) {
#if 0
resetandabort:
/* XXXjlemon check this - is this correct? */
(void) tcp_respond(NULL, m, m, th,
th->th_seq + tlen, (tcp_seq)0, TH_RST|TH_ACK);
#endif
m_freem(m); /* XXX: only needed for above */
tcpstat.tcps_sc_aborted++;
if (sc != &scs) {
syncache_insert(sc, sch); /* try again later */
sc = NULL;
}
goto failed;
} else
tcpstat.tcps_sc_completed++;
*lsop = so;
if (sc != &scs)
syncache_free(sc);
return (1);
failed:
if (sc != NULL && sc != &scs)
syncache_free(sc);
return (0);
}
/*
* Given a LISTEN socket and an inbound SYN request, add
* this to the syn cache, and send back a segment:
* <SEQ=ISS><ACK=RCV_NXT><CTL=SYN,ACK>
* to the source.
*
* IMPORTANT NOTE: We do _NOT_ ACK data that might accompany the SYN.
* Doing so would require that we hold onto the data and deliver it
* to the application. However, if we are the target of a SYN-flood
* DoS attack, an attacker could send data which would eventually
* consume all available buffer space if it were ACKed. By not ACKing
* the data, we avoid this DoS scenario.
*/
int
syncache_add(struct in_conninfo *inc, struct tcpopt *to, struct tcphdr *th,
struct inpcb *inp, struct socket **lsop, struct mbuf *m)
{
struct tcpcb *tp;
struct socket *so;
struct syncache *sc = NULL;
struct syncache_head *sch;
struct mbuf *ipopts = NULL;
u_int32_t flowtmp;
int win, sb_hiwat, ip_ttl, ip_tos, noopt;
#ifdef INET6
int autoflowlabel = 0;
#endif
struct syncache scs;
INP_INFO_WLOCK_ASSERT(&tcbinfo);
INP_LOCK_ASSERT(inp); /* listen socket */
/*
* Combine all so/tp operations very early to drop the INP lock as
* soon as possible.
*/
so = *lsop;
tp = sototcpcb(so);
#ifdef INET6
if (inc->inc_isipv6 &&
(inp->in6p_flags & IN6P_AUTOFLOWLABEL))
autoflowlabel = 1;
#endif
ip_ttl = inp->inp_ip_ttl;
ip_tos = inp->inp_ip_tos;
win = sbspace(&so->so_rcv);
sb_hiwat = so->so_rcv.sb_hiwat;
noopt = (tp->t_flags & TF_NOOPT);
so = NULL;
tp = NULL;
INP_UNLOCK(inp);
INP_INFO_WUNLOCK(&tcbinfo);
/*
* Remember the IP options, if any.
*/
#ifdef INET6
if (!inc->inc_isipv6)
#endif
ipopts = ip_srcroute(m);
/*
* See if we already have an entry for this connection.
* If we do, resend the SYN,ACK, and reset the retransmit timer.
*
* XXX: should the syncache be re-initialized with the contents
* of the new SYN here (which may have different options?)
*/
sc = syncache_lookup(inc, &sch); /* returns locked entry */
SCH_LOCK_ASSERT(sch);
if (sc != NULL) {
tcpstat.tcps_sc_dupsyn++;
if (ipopts) {
/*
* If we were remembering a previous source route,
* forget it and use the new one we've been given.
*/
if (sc->sc_ipopts)
(void) m_free(sc->sc_ipopts);
sc->sc_ipopts = ipopts;
}
/*
* Update timestamp if present.
*/
if (sc->sc_flags & SCF_TIMESTAMP)
sc->sc_tsreflect = to->to_tsval;
if (syncache_respond(sc, m) == 0) {
SYNCACHE_TIMEOUT(sc, sch, 1);
tcpstat.tcps_sndacks++;
tcpstat.tcps_sndtotal++;
}
SCH_UNLOCK(sch);
goto done;
}
sc = uma_zalloc(tcp_syncache.zone, M_NOWAIT | M_ZERO);
if (sc == NULL) {
/*
* The zone allocator couldn't provide more entries.
* Treat this as if the cache was full; drop the oldest
* entry and insert the new one.
*/
tcpstat.tcps_sc_zonefail++;
sc = TAILQ_LAST(&sch->sch_bucket, sch_head);
syncache_drop(sc, sch);
sc = uma_zalloc(tcp_syncache.zone, M_NOWAIT | M_ZERO);
if (sc == NULL) {
if (tcp_syncookies) {
bzero(&scs, sizeof(scs));
sc = &scs;
} else {
SCH_UNLOCK(sch);
if (ipopts)
(void) m_free(ipopts);
goto done;
}
}
}
/*
* Fill in the syncache values.
*/
sc->sc_ipopts = ipopts;
bcopy(inc, &sc->sc_inc, sizeof(struct in_conninfo));
#ifdef INET6
if (!inc->inc_isipv6)
#endif
{
sc->sc_ip_tos = ip_tos;
sc->sc_ip_ttl = ip_ttl;
}
sc->sc_irs = th->th_seq;
sc->sc_iss = arc4random();
sc->sc_flags = 0;
sc->sc_flowlabel = 0;
/*
* Initial receive window: clip sbspace to [0 .. TCP_MAXWIN].
* win was derived from socket earlier in the function.
*/
win = imax(win, 0);
win = imin(win, TCP_MAXWIN);
sc->sc_wnd = win;
if (tcp_do_rfc1323) {
/*
* A timestamp received in a SYN makes
* it ok to send timestamp requests and replies.
*/
if (to->to_flags & TOF_TS) {
sc->sc_tsreflect = to->to_tsval;
sc->sc_flags |= SCF_TIMESTAMP;
}
if (to->to_flags & TOF_SCALE) {
int wscale = 0;
/* Compute proper scaling value from buffer space */
while (wscale < TCP_MAX_WINSHIFT &&
(TCP_MAXWIN << wscale) < sb_hiwat)
wscale++;
sc->sc_requested_r_scale = wscale;
sc->sc_requested_s_scale = to->to_requested_s_scale;
sc->sc_flags |= SCF_WINSCALE;
}
}
#ifdef TCP_SIGNATURE
/*
* If listening socket requested TCP digests, and received SYN
* contains the option, flag this in the syncache so that
* syncache_respond() will do the right thing with the SYN+ACK.
* XXX: Currently we always record the option by default and will
* attempt to use it in syncache_respond().
*/
if (to->to_flags & TOF_SIGNATURE)
sc->sc_flags |= SCF_SIGNATURE;
#endif
if (to->to_flags & TOF_SACK)
sc->sc_flags |= SCF_SACK;
if (to->to_flags & TOF_MSS)
sc->sc_peer_mss = to->to_mss; /* peer mss may be zero */
if (noopt)
sc->sc_flags |= SCF_NOOPT;
if (tcp_syncookies) {
syncookie_generate(sch, sc, &flowtmp);
#ifdef INET6
if (autoflowlabel)
sc->sc_flowlabel = flowtmp;
#endif
} else {
#ifdef INET6
if (autoflowlabel)
sc->sc_flowlabel =
(htonl(ip6_randomflowlabel()) & IPV6_FLOWLABEL_MASK);
#endif
}
SCH_UNLOCK(sch);
/*
* Do a standard 3-way handshake.
*/
if (syncache_respond(sc, m) == 0) {
if (tcp_syncookies && tcp_syncookiesonly && sc != &scs)
syncache_free(sc);
else if (sc != &scs)
syncache_insert(sc, sch); /* locks and unlocks sch */
tcpstat.tcps_sndacks++;
tcpstat.tcps_sndtotal++;
} else {
syncache_free(sc);
tcpstat.tcps_sc_dropped++;
}
done:
*lsop = NULL;
return (1);
}
static int
syncache_respond(struct syncache *sc, struct mbuf *m)
{
struct ip *ip = NULL;
struct tcphdr *th;
int optlen, error;
u_int16_t tlen, hlen, mssopt;
u_int8_t *optp;
#ifdef INET6
struct ip6_hdr *ip6 = NULL;
#endif
#ifdef MAC
struct inpcb *inp = NULL;
#endif
hlen =
#ifdef INET6
(sc->sc_inc.inc_isipv6) ? sizeof(struct ip6_hdr) :
#endif
sizeof(struct ip);
/* Determine MSS we advertize to other end of connection. */
mssopt = tcp_mssopt(&sc->sc_inc);
if (sc->sc_peer_mss)
mssopt = max( min(sc->sc_peer_mss, mssopt), tcp_minmss);
/* Compute the size of the TCP options. */
if (sc->sc_flags & SCF_NOOPT) {
optlen = 0;
} else {
optlen = TCPOLEN_MAXSEG +
((sc->sc_flags & SCF_WINSCALE) ? 4 : 0) +
((sc->sc_flags & SCF_TIMESTAMP) ? TCPOLEN_TSTAMP_APPA : 0);
#ifdef TCP_SIGNATURE
if (sc->sc_flags & SCF_SIGNATURE)
optlen += TCPOLEN_SIGNATURE;
#endif
if (sc->sc_flags & SCF_SACK)
optlen += TCPOLEN_SACK_PERMITTED;
optlen = roundup2(optlen, 4);
}
tlen = hlen + sizeof(struct tcphdr) + optlen;
/*
* XXX: Assume that the entire packet will fit in a header mbuf.
*/
KASSERT(max_linkhdr + tlen <= MHLEN, ("syncache: mbuf too small"));
/* Create the IP+TCP header from scratch. */
if (m)
m_freem(m);
m = m_gethdr(M_DONTWAIT, MT_DATA);
if (m == NULL)
return (ENOBUFS);
m->m_data += max_linkhdr;
m->m_len = tlen;
m->m_pkthdr.len = tlen;
m->m_pkthdr.rcvif = NULL;
#ifdef MAC
/*
* For MAC look up the inpcb to get access to the label information.
* We don't store the inpcb pointer in struct syncache to make locking
* less complicated and to save locking operations. However for MAC
* this gives a slight overhead as we have to do a full pcblookup here.
*/
INP_INFO_RLOCK(&tcbinfo);
if (inp == NULL) {
#ifdef INET6 /* && MAC */
if (sc->sc_inc.inc_isipv6)
inp = in6_pcblookup_hash(&tcbinfo,
&sc->sc_inc.inc6_faddr, sc->sc_inc.inc_fport,
&sc->sc_inc.inc6_laddr, sc->sc_inc.inc_lport,
1, NULL);
else
#endif /* INET6 */
inp = in_pcblookup_hash(&tcbinfo,
sc->sc_inc.inc_faddr, sc->sc_inc.inc_fport,
sc->sc_inc.inc_laddr, sc->sc_inc.inc_lport,
1, NULL);
if (inp == NULL) {
m_freem(m);
INP_INFO_RUNLOCK(&tcbinfo);
return (ESHUTDOWN);
}
}
INP_LOCK(inp);
if (!inp->inp_socket->so_options & SO_ACCEPTCONN) {
m_freem(m);
INP_UNLOCK(inp);
INP_INFO_RUNLOCK(&tcbinfo);
return (ESHUTDOWN);
}
mac_create_mbuf_from_inpcb(inp, m);
INP_UNLOCK(inp);
INP_INFO_RUNLOCK(&tcbinfo);
#endif /* MAC */
#ifdef INET6
if (sc->sc_inc.inc_isipv6) {
ip6 = mtod(m, struct ip6_hdr *);
ip6->ip6_vfc = IPV6_VERSION;
ip6->ip6_nxt = IPPROTO_TCP;
ip6->ip6_src = sc->sc_inc.inc6_laddr;
ip6->ip6_dst = sc->sc_inc.inc6_faddr;
ip6->ip6_plen = htons(tlen - hlen);
/* ip6_hlim is set after checksum */
ip6->ip6_flow &= ~IPV6_FLOWLABEL_MASK;
ip6->ip6_flow |= sc->sc_flowlabel;
th = (struct tcphdr *)(ip6 + 1);
} else
#endif
{
ip = mtod(m, struct ip *);
ip->ip_v = IPVERSION;
ip->ip_hl = sizeof(struct ip) >> 2;
ip->ip_len = tlen;
ip->ip_id = 0;
ip->ip_off = 0;
ip->ip_sum = 0;
ip->ip_p = IPPROTO_TCP;
ip->ip_src = sc->sc_inc.inc_laddr;
ip->ip_dst = sc->sc_inc.inc_faddr;
ip->ip_ttl = sc->sc_ip_ttl;
ip->ip_tos = sc->sc_ip_tos;
/*
* See if we should do MTU discovery. Route lookups are
* expensive, so we will only unset the DF bit if:
*
* 1) path_mtu_discovery is disabled
* 2) the SCF_UNREACH flag has been set
*/
if (path_mtu_discovery && ((sc->sc_flags & SCF_UNREACH) == 0))
ip->ip_off |= IP_DF;
th = (struct tcphdr *)(ip + 1);
}
th->th_sport = sc->sc_inc.inc_lport;
th->th_dport = sc->sc_inc.inc_fport;
th->th_seq = htonl(sc->sc_iss);
th->th_ack = htonl(sc->sc_irs + 1);
th->th_off = (sizeof(struct tcphdr) + optlen) >> 2;
th->th_x2 = 0;
th->th_flags = TH_SYN|TH_ACK;
th->th_win = htons(sc->sc_wnd);
th->th_urp = 0;
/* Tack on the TCP options. */
if (optlen != 0) {
optp = (u_int8_t *)(th + 1);
*optp++ = TCPOPT_MAXSEG;
*optp++ = TCPOLEN_MAXSEG;
*optp++ = (mssopt >> 8) & 0xff;
*optp++ = mssopt & 0xff;
if (sc->sc_flags & SCF_WINSCALE) {
*((u_int32_t *)optp) = htonl(TCPOPT_NOP << 24 |
TCPOPT_WINDOW << 16 | TCPOLEN_WINDOW << 8 |
sc->sc_requested_r_scale);
optp += 4;
}
if (sc->sc_flags & SCF_TIMESTAMP) {
u_int32_t *lp = (u_int32_t *)(optp);
/* Form timestamp option per appendix A of RFC 1323. */
*lp++ = htonl(TCPOPT_TSTAMP_HDR);
if (sc->sc_ts)
*lp++ = htonl(sc->sc_ts);
else
*lp++ = htonl(ticks);
*lp = htonl(sc->sc_tsreflect);
optp += TCPOLEN_TSTAMP_APPA;
}
#ifdef TCP_SIGNATURE
/*
* Handle TCP-MD5 passive opener response.
*/
if (sc->sc_flags & SCF_SIGNATURE) {
u_int8_t *bp = optp;
int i;
*bp++ = TCPOPT_SIGNATURE;
*bp++ = TCPOLEN_SIGNATURE;
for (i = 0; i < TCP_SIGLEN; i++)
*bp++ = 0;
tcp_signature_compute(m, sizeof(struct ip), 0, optlen,
optp + 2, IPSEC_DIR_OUTBOUND);
optp += TCPOLEN_SIGNATURE;
}
#endif /* TCP_SIGNATURE */
if (sc->sc_flags & SCF_SACK) {
*optp++ = TCPOPT_SACK_PERMITTED;
*optp++ = TCPOLEN_SACK_PERMITTED;
}
{
/* Pad TCP options to a 4 byte boundary */
int padlen = optlen - (optp - (u_int8_t *)(th + 1));
while (padlen-- > 0)
*optp++ = TCPOPT_EOL;
}
}
#ifdef INET6
if (sc->sc_inc.inc_isipv6) {
th->th_sum = 0;
th->th_sum = in6_cksum(m, IPPROTO_TCP, hlen, tlen - hlen);
ip6->ip6_hlim = in6_selecthlim(NULL, NULL);
error = ip6_output(m, NULL, NULL, 0, NULL, NULL, NULL);
} else
#endif
{
th->th_sum = in_pseudo(ip->ip_src.s_addr, ip->ip_dst.s_addr,
htons(tlen - hlen + IPPROTO_TCP));
m->m_pkthdr.csum_flags = CSUM_TCP;
m->m_pkthdr.csum_data = offsetof(struct tcphdr, th_sum);
error = ip_output(m, sc->sc_ipopts, NULL, 0, NULL, NULL);
}
return (error);
}
/*
* The purpose of SYN cookies is to avoid keeping track of all SYN's we
* receive and to be able to handle SYN floods from bogus source addresses
* (where we will never receive any reply). SYN floods try to exhaust all
* our memory and available slots in the SYN cache table to cause a denial
* of service to legitimate users of the local host.
*
* The idea of SYN cookies is to encode and include all necessary information
* about the connection setup state within the SYN-ACK we send back and thus
* to get along without keeping any local state until the ACK to the SYN-ACK
* arrives (if ever). Everything we need to know should be available from
* the information we encoded in the SYN-ACK.
*
* More information about the theory behind SYN cookies and its first
* discussion and specification can be found at:
* http://cr.yp.to/syncookies.html (overview)
* http://cr.yp.to/syncookies/archive (gory details)
*
* This implementation extends the orginal idea and first implementation
* of FreeBSD by using not only the initial sequence number field to store
* information but also the timestamp field if present. This way we can
* keep track of the entire state we need to know to recreate the session in
* its original form. Almost all TCP speakers implement RFC1323 timestamps
* these days. For those that do not we still have to live with the known
* shortcomings of the ISN only SYN cookies.
*
* Cookie layers:
*
* Initial sequence number we send:
* 31|................................|0
* DDDDDDDDDDDDDDDDDDDDDDDDDMMMRRRP
* D = MD5 Digest (first dword)
* M = MSS index
* R = Rotation of secret
* P = Odd or Even secret
*
* The MD5 Digest is computed with over following parameters:
* a) randomly rotated secret
* b) struct in_conninfo containing the remote/local ip/port (IPv4&IPv6)
* c) the received initial sequence number from remote host
* d) the rotation offset and odd/even bit
*
* Timestamp we send:
* 31|................................|0
* DDDDDDDDDDDDDDDDDDDDDDSSSSRRRRA5
* D = MD5 Digest (third dword) (only as filler)
* S = Requested send window scale
* R = Requested receive window scale
* A = SACK allowed
* 5 = TCP-MD5 enabled (not implemented yet)
* XORed with MD5 Digest (forth dword)
*
* The timestamp isn't cryptographically secure and doesn't need to be.
* The double use of the MD5 digest dwords ties it to a specific remote/
* local host/port, remote initial sequence number and our local time
* limited secret. A received timestamp is reverted (XORed) and then
* the contained MD5 dword is compared to the computed one to ensure the
* timestamp belongs to the SYN-ACK we sent. The other parameters may
* have been tampered with but this isn't different from supplying bogus
* values in the SYN in the first place.
*
* Some problems with SYN cookies remain however:
* Consider the problem of a recreated (and retransmitted) cookie. If the
* original SYN was accepted, the connection is established. The second
* SYN is inflight, and if it arrives with an ISN that falls within the
* receive window, the connection is killed.
*
* Notes:
* A heuristic to determine when to accept syn cookies is not necessary.
* An ACK flood would cause the syncookie verification to be attempted,
* but a SYN flood causes syncookies to be generated. Both are of equal
* cost, so there's no point in trying to optimize the ACK flood case.
* Also, if you don't process certain ACKs for some reason, then all someone
* would have to do is launch a SYN and ACK flood at the same time, which
* would stop cookie verification and defeat the entire purpose of syncookies.
*/
static int tcp_sc_msstab[] = { 0, 256, 468, 536, 996, 1452, 1460, 8960 };
static void
syncookie_generate(struct syncache_head *sch, struct syncache *sc,
u_int32_t *flowlabel)
{
MD5_CTX ctx;
u_int32_t md5_buffer[MD5_DIGEST_LENGTH / sizeof(u_int32_t)];
u_int32_t data;
u_int32_t *secbits;
u_int off, pmss, mss;
int i;
SCH_LOCK_ASSERT(sch);
/* Which of the two secrets to use. */
secbits = sch->sch_oddeven ?
sch->sch_secbits_odd : sch->sch_secbits_even;
/* Reseed secret if too old. */
if (sch->sch_reseed < time_uptime) {
sch->sch_oddeven = sch->sch_oddeven ? 0 : 1; /* toggle */
secbits = sch->sch_oddeven ?
sch->sch_secbits_odd : sch->sch_secbits_even;
for (i = 0; i < SYNCOOKIE_SECRET_SIZE; i++)
secbits[i] = arc4random();
sch->sch_reseed = time_uptime + SYNCOOKIE_LIFETIME;
}
/* Secret rotation offset. */
off = sc->sc_iss & 0x7; /* iss was randomized before */
/* Maximum segment size calculation. */
pmss = max( min(sc->sc_peer_mss, tcp_mssopt(&sc->sc_inc)), tcp_minmss);
for (mss = sizeof(tcp_sc_msstab) / sizeof(int) - 1; mss > 0; mss--)
if (tcp_sc_msstab[mss] <= pmss)
break;
/* Fold parameters and MD5 digest into the ISN we will send. */
data = sch->sch_oddeven;/* odd or even secret, 1 bit */
data |= off << 1; /* secret offset, derived from iss, 3 bits */
data |= mss << 4; /* mss, 3 bits */
MD5Init(&ctx);
MD5Update(&ctx, ((u_int8_t *)secbits) + off,
SYNCOOKIE_SECRET_SIZE * sizeof(*secbits) - off);
MD5Update(&ctx, secbits, off);
MD5Update(&ctx, &sc->sc_inc, sizeof(sc->sc_inc));
MD5Update(&ctx, &sc->sc_irs, sizeof(sc->sc_irs));
MD5Update(&ctx, &data, sizeof(data));
MD5Final((u_int8_t *)&md5_buffer, &ctx);
data |= (md5_buffer[0] << 7);
sc->sc_iss = data;
#ifdef INET6
*flowlabel = md5_buffer[1] & IPV6_FLOWLABEL_MASK;
#endif
/* Additional parameters are stored in the timestamp if present. */
if (sc->sc_flags & SCF_TIMESTAMP) {
data = ((sc->sc_flags & SCF_SIGNATURE) ? 1 : 0); /* TCP-MD5, 1 bit */
data |= ((sc->sc_flags & SCF_SACK) ? 1 : 0) << 1; /* SACK, 1 bit */
data |= sc->sc_requested_s_scale << 2; /* SWIN scale, 4 bits */
data |= sc->sc_requested_r_scale << 6; /* RWIN scale, 4 bits */
data |= md5_buffer[2] << 10; /* more digest bits */
data ^= md5_buffer[3];
sc->sc_ts = data;
sc->sc_tsoff = data - ticks; /* after XOR */
} else
sc->sc_ts = 0;
return;
}
static struct syncache *
syncookie_lookup(struct in_conninfo *inc, struct syncache_head *sch,
struct syncache *sc, struct tcpopt *to, struct tcphdr *th,
struct socket *so)
{
MD5_CTX ctx;
u_int32_t md5_buffer[MD5_DIGEST_LENGTH / sizeof(u_int32_t)];
u_int32_t data = 0;
u_int32_t *secbits;
tcp_seq ack, seq;
int off, mss, wnd, flags;
SCH_LOCK_ASSERT(sch);
/*
* Pull information out of SYN-ACK/ACK and
* revert sequence number advances.
*/
ack = th->th_ack - 1;
seq = th->th_seq - 1;
off = (ack >> 1) & 0x7;
mss = (ack >> 4) & 0x7;
flags = ack & 0x7f;
/* Which of the two secrets to use. */
secbits = (flags & 0x1) ? sch->sch_secbits_odd : sch->sch_secbits_even;
/*
* The secret wasn't updated for the lifetime of a syncookie,
* so this SYN-ACK/ACK is either too old (replay) or totally bogus.
*/
if (sch->sch_reseed < time_uptime) {
return (NULL);
}
/* Recompute the digest so we can compare it. */
MD5Init(&ctx);
MD5Update(&ctx, ((u_int8_t *)secbits) + off,
SYNCOOKIE_SECRET_SIZE * sizeof(*secbits) - off);
MD5Update(&ctx, secbits, off);
MD5Update(&ctx, inc, sizeof(*inc));
MD5Update(&ctx, &seq, sizeof(seq));
MD5Update(&ctx, &flags, sizeof(flags));
MD5Final((u_int8_t *)&md5_buffer, &ctx);
/* Does the digest part of or ACK'ed ISS match? */
if ((ack & (~0x7f)) != (md5_buffer[0] << 7))
return (NULL);
/* Does the digest part of our reflected timestamp match? */
if (to->to_flags & TOF_TS) {
data = md5_buffer[3] ^ to->to_tsecr;
if ((data & (~0x3ff)) != (md5_buffer[2] << 10))
return (NULL);
}
/* Fill in the syncache values. */
bcopy(inc, &sc->sc_inc, sizeof(struct in_conninfo));
sc->sc_ipopts = NULL;
sc->sc_irs = seq;
sc->sc_iss = ack;
#ifdef INET6
if (inc->inc_isipv6) {
if (sotoinpcb(so)->in6p_flags & IN6P_AUTOFLOWLABEL)
sc->sc_flowlabel = md5_buffer[1] & IPV6_FLOWLABEL_MASK;
} else
#endif
{
sc->sc_ip_ttl = sotoinpcb(so)->inp_ip_ttl;
sc->sc_ip_tos = sotoinpcb(so)->inp_ip_tos;
}
/* Additional parameters that were encoded in the timestamp. */
if (data) {
sc->sc_flags |= SCF_TIMESTAMP;
sc->sc_tsreflect = to->to_tsval;
sc->sc_tsoff = to->to_tsecr - ticks;
sc->sc_flags |= (data & 0x1) ? SCF_SIGNATURE : 0;
sc->sc_flags |= ((data >> 1) & 0x1) ? SCF_SACK : 0;
sc->sc_requested_s_scale = min((data >> 2) & 0xf,
TCP_MAX_WINSHIFT);
sc->sc_requested_r_scale = min((data >> 6) & 0xf,
TCP_MAX_WINSHIFT);
if (sc->sc_requested_s_scale || sc->sc_requested_r_scale)
sc->sc_flags |= SCF_WINSCALE;
} else
sc->sc_flags |= SCF_NOOPT;
wnd = sbspace(&so->so_rcv);
wnd = imax(wnd, 0);
wnd = imin(wnd, TCP_MAXWIN);
sc->sc_wnd = wnd;
sc->sc_rxmits = 0;
sc->sc_peer_mss = tcp_sc_msstab[mss];
return (sc);
}
|