summaryrefslogtreecommitdiff
path: root/sys/netinet/ip_ipsp.c
blob: 5e4eb4fea11b0c69a011a313ccbc0312fafff8f0 (plain)
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
/*	$OpenBSD: ip_ipsp.c,v 1.52 1999/08/05 21:58:15 ho Exp $	*/

/*
 * The authors of this code are John Ioannidis (ji@tla.org),
 * Angelos D. Keromytis (kermit@csd.uch.gr), 
 * Niels Provos (provos@physnet.uni-hamburg.de) and
 * Niklas Hallqvist (niklas@appli.se).
 *
 * This code was written by John Ioannidis for BSD/OS in Athens, Greece, 
 * in November 1995.
 *
 * Ported to OpenBSD and NetBSD, with additional transforms, in December 1996,
 * by Angelos D. Keromytis.
 *
 * Additional transforms and features in 1997 and 1998 by Angelos D. Keromytis
 * and Niels Provos.
 *
 * Additional features in 1999 by Angelos D. Keromytis and Niklas Hallqvist.
 *
 * Copyright (c) 1995, 1996, 1997, 1998, 1999 by John Ioannidis,
 * Angelos D. Keromytis and Niels Provos.
 * Copyright (c) 1999 Niklas Hallqvist.
 *
 * Permission to use, copy, and modify this software without fee
 * is hereby granted, provided that this entire notice is included in
 * all copies of any software which is or includes a copy or
 * modification of this software. 
 * You may use this code under the GNU public license if you so wish. Please
 * contribute changes back to the authors under this freer than GPL license
 * so that we may further the use of strong encryption without limitations to
 * all.
 *
 * THIS SOFTWARE IS BEING PROVIDED "AS IS", WITHOUT ANY EXPRESS OR
 * IMPLIED WARRANTY. IN PARTICULAR, NONE OF THE AUTHORS MAKES ANY
 * REPRESENTATION OR WARRANTY OF ANY KIND CONCERNING THE
 * MERCHANTABILITY OF THIS SOFTWARE OR ITS FITNESS FOR ANY PARTICULAR
 * PURPOSE.
 */

/*
 * IPSP Processing
 */

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/domain.h>
#include <sys/protosw.h>
#include <sys/socket.h>
#include <sys/socketvar.h>
#include <sys/errno.h>
#include <sys/time.h>
#include <sys/kernel.h>
#include <sys/proc.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_pcb.h>
#include <netinet/in_var.h>
#include <netinet/ip_var.h>
#include <netinet/ip_icmp.h>

#include <net/raw_cb.h>
#include <net/pfkeyv2.h>

#include <netinet/ip_ipsp.h>
#include <netinet/ip_ah.h>
#include <netinet/ip_esp.h>

#include <dev/rndvar.h>

#ifdef DDB
#include <ddb/db_output.h>
void tdb_hashstats(void);
#endif

#ifdef ENCDEBUG
#define DPRINTF(x)	if (encdebug) printf x
#else
#define DPRINTF(x)
#endif

#ifdef __GNUC__
#define INLINE static __inline
#endif

int		ipsp_kern __P((int, char **, int));
u_int8_t       	get_sa_require  __P((struct inpcb *));
int		check_ipsec_policy  __P((struct inpcb *, u_int32_t));
void		tdb_rehash __P((void));

extern int	ipsec_auth_default_level;
extern int	ipsec_esp_trans_default_level;
extern int	ipsec_esp_network_default_level;

extern int encdebug;
int ipsec_in_use = 0;
u_int32_t kernfs_epoch = 0;

struct expclusterlist_head expclusterlist =
    TAILQ_HEAD_INITIALIZER(expclusterlist);
struct explist_head explist = TAILQ_HEAD_INITIALIZER(explist);

u_int8_t hmac_ipad_buffer[64] = {
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 
    0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36, 0x36 };

u_int8_t hmac_opad_buffer[64] = {
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C,
    0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C, 0x5C };

/*
 * This is the proper place to define the various encapsulation transforms.
 */

struct xformsw xformsw[] = {
    { XF_IP4,	         0,               "IPv4 Simple Encapsulation",
      ipe4_attach,       ipe4_init,       ipe4_zeroize,
      (struct mbuf * (*)(struct mbuf *, struct tdb *))ipe4_input, 
      ipe4_output, },
    { XF_OLD_AH,         XFT_AUTH,	 "Keyed Authentication, RFC 1828/1852",
      ah_old_attach,     ah_old_init,     ah_old_zeroize,
      ah_old_input,      ah_old_output, },
    { XF_OLD_ESP,        XFT_CONF,        "Simple Encryption, RFC 1829/1851",
      esp_old_attach,    esp_old_init,    esp_old_zeroize,
      esp_old_input,     esp_old_output, },
    { XF_NEW_AH,	 XFT_AUTH,	  "HMAC Authentication",
      ah_new_attach,	 ah_new_init,     ah_new_zeroize,
      ah_new_input,	 ah_new_output, },
    { XF_NEW_ESP,	 XFT_CONF|XFT_AUTH,
      "Encryption + Authentication + Replay Protection",
      esp_new_attach,	 esp_new_init,  esp_new_zeroize,
      esp_new_input,	 esp_new_output, },
#ifdef TCP_SIGNATURE
    { XF_TCPSIGNATURE,	 XFT_AUTH, "TCP MD5 Signature Option, RFC 2385",
      tcp_signature_tdb_attach, 	tcp_signature_tdb_init,
      tcp_signature_tdb_zeroize,	tcp_signature_tdb_input,
      tcp_signature_tdb_output, }
#endif /* TCP_SIGNATURE */
};

struct xformsw *xformswNXFORMSW = &xformsw[sizeof(xformsw)/sizeof(xformsw[0])];

unsigned char ipseczeroes[IPSEC_ZEROES_SIZE]; /* zeroes! */ 

#define TDB_HASHSIZE_INIT 32
static struct tdb **tdbh = NULL;
static u_int tdb_hashmask = TDB_HASHSIZE_INIT - 1;
static int tdb_count;

/*
 * Check which transformationes are required
 */

u_int8_t
get_sa_require(struct inpcb *inp)
{
    u_int8_t sareq = 0;
       
    if (inp != NULL)
    {
	sareq |= inp->inp_seclevel[SL_AUTH] >= IPSEC_LEVEL_USE ? 
		 NOTIFY_SATYPE_AUTH : 0;
	sareq |= inp->inp_seclevel[SL_ESP_TRANS] >= IPSEC_LEVEL_USE ?
		 NOTIFY_SATYPE_CONF : 0;
	sareq |= inp->inp_seclevel[SL_ESP_NETWORK] >= IPSEC_LEVEL_USE ?
		 NOTIFY_SATYPE_TUNNEL : 0;
    }
    else
    {
	sareq |= ipsec_auth_default_level >= IPSEC_LEVEL_USE ? 
		 NOTIFY_SATYPE_AUTH : 0;
	sareq |= ipsec_esp_trans_default_level >= IPSEC_LEVEL_USE ? 
		 NOTIFY_SATYPE_CONF : 0;
	sareq |= ipsec_esp_network_default_level >= IPSEC_LEVEL_USE ? 
		 NOTIFY_SATYPE_TUNNEL : 0;
    }
    
    return (sareq);
}

/*
 * Check the socket policy and request a new SA with a key management
 * daemon. Sometime the inp does not contain the destination address
 * in that case use dst.
 */

int
check_ipsec_policy(struct inpcb *inp, u_int32_t daddr)
{
    union sockaddr_union sunion;
    struct socket *so;
    struct route_enc re0, *re = &re0;
    struct sockaddr_encap *dst; 
    u_int8_t sa_require, sa_have;
    int error, i, s;
    struct tdb *tdb = NULL;

    if (inp == NULL || ((so = inp->inp_socket) == 0))
      return (EINVAL);

    /* If IPSEC is not required just use what we got */
    if (!(sa_require = inp->inp_secrequire))
      return 0;

    s = spltdb();
    if (!inp->inp_tdb)
    {
	bzero((caddr_t) re, sizeof(*re));
	dst = (struct sockaddr_encap *) &re->re_dst;
	dst->sen_family = PF_KEY;
	dst->sen_len = SENT_IP4_LEN;
	dst->sen_type = SENT_IP4;
	dst->sen_ip_src = inp->inp_laddr;
	dst->sen_ip_dst.s_addr = inp->inp_faddr.s_addr ? 
	  inp->inp_faddr.s_addr : daddr;
	dst->sen_proto = so->so_proto->pr_protocol;
	switch (dst->sen_proto)
	{
	  case IPPROTO_UDP:
	  case IPPROTO_TCP:
	    dst->sen_sport = inp->inp_lport;
	    dst->sen_dport = inp->inp_fport;
	    break;
	  default:
	    dst->sen_sport = 0;
	    dst->sen_dport = 0;
	}
    
	/* Try to find a flow */
	rtalloc((struct route *) re);
    
	if (re->re_rt != NULL)
	{
	    struct sockaddr_encap *gw;
	    
	    gw = (struct sockaddr_encap *) (re->re_rt->rt_gateway);
	    
	    if (gw->sen_type == SENT_IPSP)
	    {
		bzero(&sunion, sizeof(sunion));
	        sunion.sin.sin_family = AF_INET;
		sunion.sin.sin_len = sizeof(struct sockaddr_in);
		sunion.sin.sin_addr = gw->sen_ipsp_dst;
	      
		tdb = (struct tdb *) gettdb(gw->sen_ipsp_spi, &sunion,
					    gw->sen_ipsp_sproto);
	    }
	    RTFREE(re->re_rt);
	}
    } else
      tdb = inp->inp_tdb;

    if (tdb)
      SPI_CHAIN_ATTRIB(sa_have, tdb_onext, tdb);
    else
      sa_have = 0;
    splx(s);

    /* Check if our requirements are met */
    if (!(sa_require & ~sa_have))
      return 0;

    error = i = 0;

    inp->inp_secresult = SR_WAIT;

    /* If necessary try to notify keymanagement three times */
    while (i < 3)
    {
	/* XXX address */
	DPRINTF(("ipsec: send SA request (%d), remote ip: %s, SA type: %d\n",
		 i + 1, inet_ntoa4(dst->sen_ip_dst), sa_require));

	/* Send notify */
	/* XXX PF_KEYv2 Notify */
	 
	/* 
	 * Wait for the keymanagement daemon to establich a new SA,
	 * even on error check again, perhaps some other process
	 * already established the necessary SA.
	 */
	error = tsleep((caddr_t)inp, PSOCK|PCATCH, "ipsecnotify", 30*hz);
	DPRINTF(("check_ipsec: sleep %d\n", error));

	if (error && error != EWOULDBLOCK)
	  break;
	/* 
	 * A Key Management daemon returned an apropriate SA back
	 * to the kernel, the kernel noted that state in the waiting
	 * socket.
	 */
	if (inp->inp_secresult == SR_SUCCESS)
	  return (0);
	/*
	 * Key Management returned a permanent failure, we do not
	 * need to retry again. XXX - when more than one key
	 * management daemon is available we can not do that.
	 */
	if (inp->inp_secresult == SR_FAILED)
	  break;
	i++;
    }

    return (error ? error : EWOULDBLOCK);
}

/*
 * Add an inpcb to the list of inpcb which reference this tdb directly.
 */

void
tdb_add_inp(struct tdb *tdb, struct inpcb *inp)
{
    int s = spltdb();

    if (inp->inp_tdb)
    {
	if (inp->inp_tdb == tdb)
	{
	    splx(s);
	    return;
	}
	TAILQ_REMOVE(&inp->inp_tdb->tdb_inp, inp, inp_tdb_next);
    }
    inp->inp_tdb = tdb;
    TAILQ_INSERT_TAIL(&tdb->tdb_inp, inp, inp_tdb_next);
    splx(s);

    DPRINTF(("tdb_add_inp: tdb: %p, inp: %p\n", tdb, inp));
}

/*
 * Reserve an SPI; the SA is not valid yet though.  We use SPI_LOCAL_USE as
 * an error return value.  It'll not be a problem that we also use that
 * for demand-keying as that is manually specified.
 */

u_int32_t
reserve_spi(u_int32_t sspi, u_int32_t tspi, union sockaddr_union *src,
	    union sockaddr_union *dst, u_int8_t sproto, int *errval)
{
    struct tdb *tdbp;
    u_int32_t spi;
    int nums;

    /* Don't accept ranges only encompassing reserved SPIs.  */
    if (tspi < sspi || tspi <= SPI_RESERVED_MAX)
    {
	(*errval) = EINVAL;
	return 0;
    }

    /* Limit the range to not include reserved areas.  */
    if (sspi <= SPI_RESERVED_MAX)
      sspi = SPI_RESERVED_MAX + 1;

    if (sspi == tspi)   /* Asking for a specific SPI */
      nums = 1;
    else
      nums = 50;  /* XXX figure out some good value */

    while (nums--)
    {
	if (sspi == tspi)  /* Specific SPI asked */
	  spi = tspi;
	else    /* Range specified */
	{
	    get_random_bytes((void *) &spi, sizeof(spi));
	    spi = sspi + (spi % (tspi - sspi));
	}
	  
	/* Don't allocate reserved SPIs.  */
	if (spi == SPI_LOCAL_USE ||
	    (spi >= SPI_RESERVED_MIN && spi <= SPI_RESERVED_MAX))
	  continue;
	else
	  spi = htonl(spi);

	/* Check whether we're using this SPI already */
	if (gettdb(spi, dst, sproto) != (struct tdb *) NULL)
	  continue;
	
	MALLOC(tdbp, struct tdb *, sizeof(struct tdb), M_TDB, M_WAITOK);
	bzero((caddr_t) tdbp, sizeof(struct tdb));
	
	tdbp->tdb_spi = spi;
	bcopy(&dst->sa, &tdbp->tdb_dst.sa, SA_LEN(&dst->sa));
	bcopy(&src->sa, &tdbp->tdb_src.sa, SA_LEN(&src->sa));
	tdbp->tdb_sproto = sproto;
	tdbp->tdb_flags |= TDBF_INVALID;       /* Mark SA as invalid for now */
	tdbp->tdb_established = time.tv_sec;
	tdbp->tdb_epoch = kernfs_epoch - 1;
	puttdb(tdbp);

	/* XXX Should set up a silent expiration for this */
	
	return spi;
    }

    (*errval) = EEXIST;
    return 0;
}

/*
 * Our hashing function needs to stir things with a non-zero random multiplier
 * so we cannot be DoS-attacked via choosing of the data to hash.
 */
INLINE int
tdb_hash(u_int32_t spi, union sockaddr_union *dst, u_int8_t proto)
{
    static u_int32_t mult1 = 0, mult2 = 0;
    u_int8_t *ptr = (u_int8_t *) dst;
    int i, shift;
    u_int64_t hash;
    int val32 = 0;

    while (mult1 == 0)
	mult1 = arc4random();
    while (mult2 == 0)
	mult2 = arc4random();

    hash = (spi ^ proto) * mult1;
    for (i = 0; i < SA_LEN(&dst->sa); i++)
    {
	val32 = (val32 << 8) | ptr[i];
	if (i % 4 == 3)
	{
	    hash ^= val32 * mult2;
	    val32 = 0;
	}
    }
    if (i % 4 != 0)
	hash ^= val32 * mult2;

    shift = ffs(tdb_hashmask + 1);
    while ((hash & ~tdb_hashmask) != 0)
      hash = (hash >> shift) ^ (hash & tdb_hashmask);

    return hash;
}

/*
 * An IPSP SAID is really the concatenation of the SPI found in the 
 * packet, the destination address of the packet and the IPsec protocol.
 * When we receive an IPSP packet, we need to look up its tunnel descriptor
 * block, based on the SPI in the packet and the destination address (which
 * is really one of our addresses if we received the packet!
 */

struct tdb *
gettdb(u_int32_t spi, union sockaddr_union *dst, u_int8_t proto)
{
    u_int32_t hashval;
    struct tdb *tdbp;
    int s;

    if (tdbh == NULL)
      return (struct tdb *) NULL;

    hashval = tdb_hash(spi, dst, proto);

    s = spltdb();
    for (tdbp = tdbh[hashval]; tdbp != NULL; tdbp = tdbp->tdb_hnext)
      if ((tdbp->tdb_spi == spi) && 
	  !bcmp(&tdbp->tdb_dst, dst, SA_LEN(&dst->sa)) &&
	  (tdbp->tdb_sproto == proto))
	break;
    splx(s);

    return tdbp;
}

#if DDB
void
tdb_hashstats()
{
    int i, cnt, buckets[16];
    struct tdb *tdbp;

    if (tdbh == NULL)
    {
	db_printf("no tdb hash table\n");
	return;
    }

    bzero (buckets, sizeof(buckets));
    for (i = 0; i <= tdb_hashmask; i++)
    {
	cnt = 0;
	for (tdbp = tdbh[i]; cnt < 16 && tdbp != NULL; tdbp = tdbp->tdb_hnext)
	  cnt++;
	buckets[cnt]++;
    }

    db_printf("tdb cnt\t\tbucket cnt\n");
    for (i = 0; i < 16; i++)
      if (buckets[i] > 0)
	db_printf("%d%c\t\t%d\n", i, i == 15 ? "+" : "", buckets[i]);
}
#endif	/* DDB */

/*
 * Caller is responsible for setting at least spltdb().
 */

int
tdb_walk(int (*walker)(struct tdb *, void *), void *arg)
{
    int i, rval = 0;
    struct tdb *tdbp, *next;

    if (tdbh == NULL)
        return ENOENT;

    for (i = 0; i <= tdb_hashmask; i++)
	for (tdbp = tdbh[i]; rval == 0 && tdbp != NULL; tdbp = next)
	{
	    next = tdbp->tdb_hnext;
	    rval = walker(tdbp, (void *)arg);
	}

    return rval;
}

struct flow *
get_flow(void)
{
    struct flow *flow;

    MALLOC(flow, struct flow *, sizeof(struct flow), M_TDB, M_WAITOK);
    bzero(flow, sizeof(struct flow));

    ipsec_in_use++;
    return flow;
}

/*
 * Called at splsoftclock().
 */

void
handle_expirations(void *arg)
{
    struct tdb *tdb;

    for (tdb = TAILQ_FIRST(&explist);
	 tdb && tdb->tdb_timeout <= time.tv_sec;
	 tdb = TAILQ_FIRST(&explist))
    {
	/* Hard expirations first */
	if ((tdb->tdb_flags & TDBF_TIMER) &&
	    (tdb->tdb_exp_timeout <= time.tv_sec))
	{
	    pfkeyv2_expire(tdb, SADB_EXT_LIFETIME_HARD);
	    tdb_delete(tdb, 0, 0);
	    continue;
	}
	else
	  if ((tdb->tdb_flags & TDBF_FIRSTUSE) &&
	      (tdb->tdb_first_use + tdb->tdb_exp_first_use <= time.tv_sec))
	  {
	      pfkeyv2_expire(tdb, SADB_EXT_LIFETIME_HARD);
	      tdb_delete(tdb, 0, 0);
	      continue;
	  }

	/* Soft expirations */
	if ((tdb->tdb_flags & TDBF_SOFT_TIMER) &&
	    (tdb->tdb_soft_timeout <= time.tv_sec))
	{
	    pfkeyv2_expire(tdb, SADB_EXT_LIFETIME_SOFT);
	    tdb->tdb_flags &= ~TDBF_SOFT_TIMER;
	    tdb_expiration(tdb, TDBEXP_EARLY);
	}
	else
	  if ((tdb->tdb_flags & TDBF_SOFT_FIRSTUSE) &&
	      (tdb->tdb_first_use + tdb->tdb_soft_first_use <=
	       time.tv_sec))
	  {
	      pfkeyv2_expire(tdb, SADB_EXT_LIFETIME_SOFT);
	      tdb->tdb_flags &= ~TDBF_SOFT_FIRSTUSE;
	      tdb_expiration(tdb, TDBEXP_EARLY);
	  }
    }

    /* If any tdb is left on the expiration queue, set the timer.  */
    if (tdb)
      timeout(handle_expirations, (void *) NULL, 
	      hz * (tdb->tdb_timeout - time.tv_sec));
}

/*
 * Ensure the tdb is in the right place in the expiration list.
 */

void
tdb_expiration(struct tdb *tdb, int flags)
{
    u_int64_t next_timeout = 0;
    struct tdb *t, *former_expirer, *next_expirer;
    int will_be_first, sole_reason, early;
    int s = spltdb();

    /*
     * If this is the local use SPI, this is an SPD entry, so don't setup any
     * timers.
     */
    if (ntohl(tdb->tdb_spi) == SPI_LOCAL_USE)
    {
	splx(s);
	return;
    }

    /* Find the earliest expiration.  */
    if ((tdb->tdb_flags & TDBF_FIRSTUSE) && tdb->tdb_first_use != 0 &&
	(next_timeout == 0 ||
	 next_timeout > tdb->tdb_first_use + tdb->tdb_exp_first_use))
	next_timeout = tdb->tdb_first_use + tdb->tdb_exp_first_use;
    if ((tdb->tdb_flags & TDBF_SOFT_FIRSTUSE) && tdb->tdb_first_use != 0 &&
	(next_timeout == 0 ||
	 next_timeout > tdb->tdb_first_use + tdb->tdb_soft_first_use))
	next_timeout = tdb->tdb_first_use + tdb->tdb_soft_first_use;
    if ((tdb->tdb_flags & TDBF_TIMER) &&
	(next_timeout == 0 || next_timeout > tdb->tdb_exp_timeout))
	next_timeout = tdb->tdb_exp_timeout;
    if ((tdb->tdb_flags & TDBF_SOFT_TIMER) &&
	(next_timeout == 0 || next_timeout > tdb->tdb_soft_timeout))
	next_timeout = tdb->tdb_soft_timeout;

    /* No change?  */
    if (next_timeout == tdb->tdb_timeout)
    {
      splx(s);
      return;
    }

    /*
     * Find out some useful facts: Will our tdb be first to expire?
     * Was our tdb the sole reason for the old timeout?
     */
    former_expirer = TAILQ_FIRST(&expclusterlist);
    next_expirer = TAILQ_NEXT(tdb, tdb_explink);
    will_be_first = (next_timeout != 0 &&
		     (former_expirer == NULL ||
		      next_timeout < former_expirer->tdb_timeout));
    sole_reason = (tdb == former_expirer &&
		   (next_expirer == NULL ||
		    tdb->tdb_timeout != next_expirer->tdb_timeout));

    /*
     * We need to untimeout if either:
     * - there is an expiration pending and the new timeout is earlier than
     *   what already exists or
     * - the existing first expiration is due to our old timeout value solely
     */
    if ((former_expirer != NULL && will_be_first) || sole_reason)
      untimeout(handle_expirations, (void *) NULL);

    /*
     * We need to timeout if we've been asked to and if either
     * - our tdb has a timeout and no former expiration exist or
     * - the new timeout is earlier than what already exists or
     * - the existing first expiration is due to our old timeout value solely
     *   and another expiration is in the pipe.
     */
    if ((flags & TDBEXP_TIMEOUT) &&
	(will_be_first || (sole_reason && next_expirer != NULL)))
      timeout(handle_expirations, (void *) NULL,
	      hz * ((will_be_first ? next_timeout :
		     next_expirer->tdb_timeout) - time.tv_sec));

    /* Our old position, if any, is not relevant anymore.  */
    if (tdb->tdb_timeout != 0)
    {
        if (tdb->tdb_expnext.tqe_prev != NULL)
	{
	    if (next_expirer && tdb->tdb_timeout == next_expirer->tdb_timeout)
	      TAILQ_INSERT_BEFORE(tdb, next_expirer, tdb_expnext);
	    TAILQ_REMOVE(&expclusterlist, tdb, tdb_expnext);
	    tdb->tdb_expnext.tqe_prev = NULL;
	}
	TAILQ_REMOVE(&explist, tdb, tdb_explink);
    }

    tdb->tdb_timeout = next_timeout;

    if (next_timeout == 0)
    {
      splx(s);
      return;
    }

    /*
     * Search front-to-back if we believe we will end up early, otherwise
     * back-to-front.
     */
    early = will_be_first || (flags & TDBEXP_EARLY);
    for (t = (early ? TAILQ_FIRST(&expclusterlist) :
	      TAILQ_LAST(&expclusterlist, expclusterlist_head));
	 t != NULL && (early ? (t->tdb_timeout <= next_timeout) : 
		       (t->tdb_timeout > next_timeout));
	 t = (early ? TAILQ_NEXT(t, tdb_expnext) :
	      TAILQ_PREV(t, expclusterlist_head, tdb_expnext)))
      ;
    if (t == (early ? TAILQ_FIRST(&expclusterlist) : NULL))
    {
	/* We are to become the first expiration.  */
	TAILQ_INSERT_HEAD(&expclusterlist, tdb, tdb_expnext);
	TAILQ_INSERT_HEAD(&explist, tdb, tdb_explink);
    }
    else
    {
	if (early)
	  t = (t ? TAILQ_PREV(t, expclusterlist_head, tdb_expnext) :
	       TAILQ_LAST(&expclusterlist, expclusterlist_head));
	if (TAILQ_NEXT(t, tdb_expnext))
	  TAILQ_INSERT_BEFORE(TAILQ_NEXT(t, tdb_expnext), tdb, tdb_explink);
	else
	  TAILQ_INSERT_TAIL(&explist, tdb, tdb_explink);
	if (t->tdb_timeout < next_timeout)
	  TAILQ_INSERT_AFTER(&expclusterlist, t, tdb, tdb_expnext);
    }

#ifdef DIAGNOSTIC
    /*
     * Check various invariants.
     */
    if (tdb->tdb_expnext.tqe_prev != NULL)
    {
	t = TAILQ_FIRST(&expclusterlist);
	if (t != tdb && t->tdb_timeout >= tdb->tdb_timeout)
	  panic("tdb_expiration: "
		"expclusterlist first link out of order (%p, %p)",
		tdb, t);
	t = TAILQ_PREV(tdb, expclusterlist_head, tdb_expnext);
	if (t != NULL && t->tdb_timeout >= tdb->tdb_timeout)
	  panic("tdb_expiration: "
		"expclusterlist prev link out of order (%p, %p)",
		tdb, t);
	else if (t == NULL && tdb != TAILQ_FIRST(&expclusterlist))
	  panic("tdb_expiration: "
		"expclusterlist first link out of order (%p, %p)",
		tdb, TAILQ_FIRST(&expclusterlist));
	t = TAILQ_NEXT(tdb, tdb_expnext);
	if (t != NULL && t->tdb_timeout <= tdb->tdb_timeout)
	  panic("tdb_expiration: "
		"expclusterlist next link out of order (%p, %p)",
		tdb, t);
	else if (t == NULL &&
		 tdb != TAILQ_LAST(&expclusterlist, expclusterlist_head))
	  panic("tdb_expiration: "
		"expclusterlist last link out of order (%p, %p)",
		tdb, TAILQ_LAST(&expclusterlist, expclusterlist_head));
	t = TAILQ_LAST(&expclusterlist, expclusterlist_head);
	if (t != tdb && t->tdb_timeout <= tdb->tdb_timeout)
	  panic("tdb_expiration: "
		"expclusterlist last link out of order (%p, %p)",
		tdb, t);
    }
    t = TAILQ_FIRST(&explist);
    if (t != NULL && t->tdb_timeout > tdb->tdb_timeout)
      panic("tdb_expiration: explist first link out of order (%p, %p)", tdb,
	    t);
    t = TAILQ_PREV(tdb, explist_head, tdb_explink);
    if (t != NULL && t->tdb_timeout > tdb->tdb_timeout)
      panic("tdb_expiration: explist prev link out of order (%p, %p)", tdb, t);
    else if (t == NULL && tdb != TAILQ_FIRST(&explist))
      panic("tdb_expiration: explist first link out of order (%p, %p)", tdb,
	    TAILQ_FIRST(&explist));
    t = TAILQ_NEXT(tdb, tdb_explink);
    if (t != NULL && t->tdb_timeout < tdb->tdb_timeout)
      panic("tdb_expiration: explist next link out of order (%p, %p)", tdb, t);
    else if (t == NULL && tdb != TAILQ_LAST(&explist, explist_head))
      panic("tdb_expiration: explist last link out of order (%p, %p)", tdb,
	    TAILQ_LAST(&explist, explist_head));
    t = TAILQ_LAST(&explist, explist_head);
    if (t != tdb && t->tdb_timeout < tdb->tdb_timeout)
      panic("tdb_expiration: explist last link out of order (%p, %p)", tdb, t);
#endif

    splx(s);
}

/*
 * Caller is responsible for setting at least spltdb().
 */

struct flow *
find_flow(union sockaddr_union *src, union sockaddr_union *srcmask,
	  union sockaddr_union *dst, union sockaddr_union *dstmask,
	  u_int8_t proto, struct tdb *tdb)
{
    struct flow *flow;

    for (flow = tdb->tdb_flow; flow; flow = flow->flow_next)
      if (!bcmp(&src->sa, &flow->flow_src.sa, SA_LEN(&src->sa)) &&
	  !bcmp(&dst->sa, &flow->flow_dst.sa, SA_LEN(&dst->sa)) &&
	  !bcmp(&srcmask->sa, &flow->flow_srcmask.sa, SA_LEN(&srcmask->sa)) &&
	  !bcmp(&dstmask->sa, &flow->flow_dstmask.sa, SA_LEN(&dstmask->sa)) &&
	  (proto == flow->flow_proto))
	return flow;

    return (struct flow *) NULL;
}

/*
 * Caller is responsible for setting at least spltdb().
 */

struct flow *
find_global_flow(union sockaddr_union *src, union sockaddr_union *srcmask,
		 union sockaddr_union *dst, union sockaddr_union *dstmask,
		 u_int8_t proto)
{
    struct flow *flow;
    struct tdb *tdb;
    int i;

    if (tdbh == NULL)
      return (struct flow *) NULL;

    for (i = 0; i <= tdb_hashmask; i++)
    {
	for (tdb = tdbh[i]; tdb != NULL; tdb = tdb->tdb_hnext)
	  if ((flow = find_flow(src, srcmask, dst, dstmask, proto, tdb)) !=
	      (struct flow *) NULL)
	    return flow;
    }
    return (struct flow *) NULL;
}

/*
 * Caller is responsible for spltdb().
 */

void
tdb_rehash(void)
{
    struct tdb **new_tdbh, *tdbp, *tdbnp;
    u_int i, old_hashmask = tdb_hashmask;
    u_int32_t hashval;

    tdb_hashmask = (tdb_hashmask << 1) | 1;
    MALLOC(new_tdbh, struct tdb **, sizeof(struct tdb *) * (tdb_hashmask + 1),
	   M_TDB, M_WAITOK);
    bzero(new_tdbh, sizeof(struct tdbh *) * (tdb_hashmask + 1));
    for (i = 0; i <= old_hashmask; i++)
      for (tdbp = tdbh[i]; tdbp != NULL; tdbp = tdbnp)
      {
	  tdbnp = tdbp->tdb_hnext;
      	  hashval = tdb_hash(tdbp->tdb_spi, &tdbp->tdb_dst, tdbp->tdb_sproto);
	  tdbp->tdb_hnext = new_tdbh[hashval];
	  new_tdbh[hashval] = tdbp;
      }
    FREE(tdbh, M_TDB);
    tdbh = new_tdbh;
}

void
puttdb(struct tdb *tdbp)
{
    u_int32_t hashval;
    int s = spltdb();

    if (tdbh == NULL)
    {
	MALLOC(tdbh, struct tdb **, sizeof(struct tdb *) * (tdb_hashmask + 1),
	       M_TDB, M_WAITOK);
	bzero(tdbh, sizeof(struct tdb *) * (tdb_hashmask + 1));
    }

    hashval = tdb_hash(tdbp->tdb_spi, &tdbp->tdb_dst, tdbp->tdb_sproto);
    
    /*
     * Rehash if this tdb would cause a bucket to have more than two items
     * and if the number of tdbs exceed 10% of the bucket count.  This
     * number is arbitratily chosen and is just a measure to not keep rehashing
     * when adding and removing tdbs which happens to always end up in the
     * same bucket, which is not uncommon when doing manual keying.
     */
    if (tdbh[hashval] != NULL && tdbh[hashval]->tdb_hnext != NULL &&
	tdb_count * 10 > tdb_hashmask + 1)
    {
	tdb_rehash();
	hashval = tdb_hash(tdbp->tdb_spi, &tdbp->tdb_dst, tdbp->tdb_sproto);
    }
    tdbp->tdb_hnext = tdbh[hashval];
    tdbh[hashval] = tdbp;
    tdb_count++;
    splx(s);
}

/*
 * Caller is responsible for setting at least spltdb().
 */

void
put_flow(struct flow *flow, struct tdb *tdb)
{
    flow->flow_next = tdb->tdb_flow;
    flow->flow_prev = (struct flow *) NULL;

    tdb->tdb_flow = flow;

    flow->flow_sa = tdb;

    if (flow->flow_next)
      flow->flow_next->flow_prev = flow;
}

/*
 * Caller is responsible for setting at least spltdb().
 */

void
delete_flow(struct flow *flow, struct tdb *tdb)
{
    if (tdb)
    {
	if (tdb->tdb_flow == flow)
	{
	    tdb->tdb_flow = flow->flow_next;
	    if (tdb->tdb_flow)
	      tdb->tdb_flow->flow_prev = (struct flow *) NULL;
	}
	else
	{
	    flow->flow_prev->flow_next = flow->flow_next;
	    if (flow->flow_next)
	      flow->flow_next->flow_prev = flow->flow_prev;
	}
    }

    ipsec_in_use--;
    FREE(flow, M_TDB);
}

void
tdb_delete(struct tdb *tdbp, int delchain, int expflags)
{
    struct tdb *tdbpp;
    struct inpcb *inp;
    u_int32_t hashval = tdbp->tdb_sproto + tdbp->tdb_spi;
    int s;

    if (tdbh == NULL)
      return;

    hashval = tdb_hash(tdbp->tdb_spi, &tdbp->tdb_dst, tdbp->tdb_sproto);

    s = spltdb();
    if (tdbh[hashval] == tdbp)
    {
	tdbpp = tdbp;
	tdbh[hashval] = tdbp->tdb_hnext;
    }
    else
      for (tdbpp = tdbh[hashval]; tdbpp != NULL; tdbpp = tdbpp->tdb_hnext)
	if (tdbpp->tdb_hnext == tdbp)
	{
	    tdbpp->tdb_hnext = tdbp->tdb_hnext;
	    tdbpp = tdbp;
	}

    /*
     * If there was something before us in the chain pointing to us,
     * make it point nowhere
     */
    if ((tdbp->tdb_inext) &&
	(tdbp->tdb_inext->tdb_onext == tdbp))
      tdbp->tdb_inext->tdb_onext = NULL;

    /* 
     * If there was something after us in the chain pointing to us,
     * make it point nowhere
     */
    if ((tdbp->tdb_onext) &&
	(tdbp->tdb_onext->tdb_inext == tdbp))
      tdbp->tdb_onext->tdb_inext = NULL;
    
    tdbpp = tdbp->tdb_onext;
    
    if (tdbp->tdb_xform)
      (*(tdbp->tdb_xform->xf_zeroize))(tdbp);

    while (tdbp->tdb_flow)
    {
        /* Delete the flow and the routing entry that goes with it. */ 
        struct sockaddr_encap encapdst, encapnetmask;

        bzero(&encapdst, sizeof(struct sockaddr_encap));
        bzero(&encapnetmask, sizeof(struct sockaddr_encap));

        encapdst.sen_len = SENT_IP4_LEN;
        encapdst.sen_family = PF_KEY;
        encapdst.sen_type = SENT_IP4;
        encapdst.sen_ip_src = tdbp->tdb_flow->flow_src.sin.sin_addr;
        encapdst.sen_ip_dst = tdbp->tdb_flow->flow_dst.sin.sin_addr;
        encapdst.sen_proto = tdbp->tdb_flow->flow_proto;
        encapdst.sen_sport = tdbp->tdb_flow->flow_src.sin.sin_port;
        encapdst.sen_dport = tdbp->tdb_flow->flow_dst.sin.sin_port;

        encapnetmask.sen_len = SENT_IP4_LEN;
        encapnetmask.sen_family = PF_KEY;
        encapnetmask.sen_type = SENT_IP4;
        encapnetmask.sen_ip_src = tdbp->tdb_flow->flow_srcmask.sin.sin_addr;
        encapnetmask.sen_ip_dst = tdbp->tdb_flow->flow_dstmask.sin.sin_addr;

        if (tdbp->tdb_flow->flow_proto)
        {
            encapnetmask.sen_proto = 0xff;
            if (tdbp->tdb_flow->flow_src.sin.sin_port)
              encapnetmask.sen_sport = 0xffff;
            if (tdbp->tdb_flow->flow_dst.sin.sin_port)
              encapnetmask.sen_dport = 0xffff;
        }

        rtrequest(RTM_DELETE, (struct sockaddr *) &encapdst,
                  (struct sockaddr *) 0,
                  (struct sockaddr *) &encapnetmask,
                  0, (struct rtentry **) 0);

        delete_flow(tdbp->tdb_flow, tdbp);
    }

    /* Cleanup SA-Bindings */
    for (tdbpp = TAILQ_FIRST(&tdbp->tdb_bind_in); tdbpp;
	 tdbpp = TAILQ_FIRST(&tdbp->tdb_bind_in))
    {
        TAILQ_REMOVE(&tdbpp->tdb_bind_in, tdbpp, tdb_bind_in_next);
	tdbpp->tdb_bind_out = NULL;
    }
    /* Cleanup inp references */
    for (inp = TAILQ_FIRST(&tdbp->tdb_inp); inp;
	 inp = TAILQ_FIRST(&tdbp->tdb_inp))
    {
        TAILQ_REMOVE(&tdbp->tdb_inp, inp, inp_tdb_next);
	inp->inp_tdb = NULL;
    }

    if (tdbp->tdb_bind_out)
      TAILQ_REMOVE(&tdbp->tdb_bind_out->tdb_bind_in, tdbp, tdb_bind_in_next);

    /* Remove us from the expiration lists.  */
    if (tdbp->tdb_timeout != 0)
    {
        tdbp->tdb_flags &= ~(TDBF_FIRSTUSE | TDBF_SOFT_FIRSTUSE | TDBF_TIMER |
			     TDBF_SOFT_TIMER);
	tdb_expiration(tdbp, expflags);
    }

    if (tdbp->tdb_srcid)
      FREE(tdbp->tdb_srcid, M_XDATA);

    if (tdbp->tdb_dstid)
      FREE(tdbp->tdb_dstid, M_XDATA);

    FREE(tdbp, M_TDB);
    tdb_count--;

    if (delchain && tdbpp)
      tdb_delete(tdbpp, delchain, expflags);
    splx(s);
}

int
tdb_init(struct tdb *tdbp, u_int16_t alg, struct ipsecinit *ii)
{
    struct xformsw *xsp;

    /* Record establishment time */
    tdbp->tdb_established = time.tv_sec;
    tdbp->tdb_epoch = kernfs_epoch - 1;

    /* Init Incoming SA-Binding Queues */
    TAILQ_INIT(&tdbp->tdb_bind_in);
    TAILQ_INIT(&tdbp->tdb_inp);

    for (xsp = xformsw; xsp < xformswNXFORMSW; xsp++)
      if (xsp->xf_type == alg)
	return (*(xsp->xf_init))(tdbp, xsp, ii);

    DPRINTF(("tdb_init(): no alg %d for spi %08x, addr %s, proto %d\n", 
	     alg, ntohl(tdbp->tdb_spi), ipsp_address(tdbp->tdb_dst),
	     tdbp->tdb_sproto));
    
    return EINVAL;
}

/*
 * Used by kernfs
 */
int
ipsp_kern(int off, char **bufp, int len)
{
    static char buffer[IPSEC_KERNFS_BUFSIZE];
    struct flow *flow;
    struct tdb *tdb, *tdbp;
    int l, i, s;

    if (off == 0)
      kernfs_epoch++;
    
    if (bufp == NULL || tdbh == NULL)
      return 0;

    bzero(buffer, IPSEC_KERNFS_BUFSIZE);

    *bufp = buffer;
    
    for (i = 0; i <= tdb_hashmask; i++)
    {
        s = spltdb();
	for (tdb = tdbh[i]; tdb; tdb = tdb->tdb_hnext)
	  if (tdb->tdb_epoch != kernfs_epoch)
	  {
	      tdb->tdb_epoch = kernfs_epoch;

	      l = sprintf(buffer,
			  "SPI = %08x, Destination = %s, Sproto = %u\n",
			  ntohl(tdb->tdb_spi),
			  ipsp_address(tdb->tdb_dst), tdb->tdb_sproto);

	      l += sprintf(buffer + l, "\tEstablished %d seconds ago\n",
			   time.tv_sec - tdb->tdb_established);

	      l += sprintf(buffer + l, "\tSource = %s",
			   ipsp_address(tdb->tdb_src));

	      if (tdb->tdb_proxy.sa.sa_family)
		l += sprintf(buffer + l, ", Proxy = %s\n",
			     ipsp_address(tdb->tdb_proxy));
	      else
		l += sprintf(buffer + l, "\n");

	      l += sprintf(buffer + l, "\tFlags (%08x) = <", tdb->tdb_flags);

	      if ((tdb->tdb_flags & ~(TDBF_TIMER | TDBF_BYTES |
				      TDBF_ALLOCATIONS | TDBF_FIRSTUSE |
				      TDBF_SOFT_TIMER | TDBF_SOFT_BYTES |
				      TDBF_SOFT_FIRSTUSE |
				      TDBF_SOFT_ALLOCATIONS)) == 0)
		l += sprintf(buffer + l, "none>\n");
	      else
	      {
		  /* We can reuse variable 'i' here, since we're not looping */
		  i = 0;

		  if (tdb->tdb_flags & TDBF_UNIQUE)
		  {
		      if (i)
			l += sprintf(buffer + l, ", ");
		      else
			i = 1;

		      l += sprintf(buffer + l, "unique");
		      i = 1;
		  }

		  if (tdb->tdb_flags & TDBF_INVALID)
		  {
		      if (i)
			l += sprintf(buffer + l, ", ");
		      else
			i = 1;

		      l += sprintf(buffer + l, "invalid");
		  }

		  if (tdb->tdb_flags & TDBF_HALFIV)
		  {
		      if (i)
			l += sprintf(buffer + l, ", ");
		      else
			i = 1;

		      l += sprintf(buffer + l, "halfiv");
		  }

		  if (tdb->tdb_flags & TDBF_PFS)
		  {
		      if (i)
			l += sprintf(buffer + l, ", ");
		      else
			i = 1;

		      l += sprintf(buffer + l, "pfs");
		  }

		  if (tdb->tdb_flags & TDBF_TUNNELING)
		  {
		      if (i)
			l += sprintf(buffer + l, ", ");
		      else
			i = 1;

		      l += sprintf(buffer + l, "tunneling");
		  }

		  l += sprintf(buffer + l, ">\n");
	      }

	      if (tdb->tdb_xform)
		l += sprintf(buffer + l, "\txform = <%s>\n", 
			     tdb->tdb_xform->xf_name);

	      if (tdb->tdb_encalgxform)
		l += sprintf(buffer + l, "\t\tEncryption = <%s>\n",
			     tdb->tdb_encalgxform->name);

	      if (tdb->tdb_authalgxform)
		l += sprintf(buffer + l, "\t\tAuthentication = <%s>\n",
			     tdb->tdb_authalgxform->name);

	      if (tdb->tdb_bind_out)
		l += sprintf(buffer + l,
			     "\tBound SA: SPI = %08x, "
			     "Destination = %s, Sproto = %u\n",
			     ntohl(tdb->tdb_bind_out->tdb_spi),
			     ipsp_address(tdb->tdb_bind_out->tdb_dst),
			     tdb->tdb_bind_out->tdb_sproto);
	      for (i = 0, tdbp = TAILQ_FIRST(&tdb->tdb_bind_in); tdbp;
		   tdbp = TAILQ_NEXT(tdbp, tdb_bind_in_next))
		i++;

	      if (i > 0)
		l += sprintf(buffer + l,
			     "\tReferenced by %d incoming SA%s\n",
			     i, i == 1 ? "" : "s");

	      if (tdb->tdb_onext)
		l += sprintf(buffer + l,
			     "\tNext SA: SPI = %08x, "
			     "Destination = %s, Sproto = %u\n",
			     ntohl(tdb->tdb_onext->tdb_spi),
			     ipsp_address(tdb->tdb_onext->tdb_dst),
			     tdb->tdb_onext->tdb_sproto);

	      if (tdb->tdb_inext)
		l += sprintf(buffer + l,
			     "\tPrevious SA: SPI = %08x, "
			     "Destination = %s, Sproto = %u\n",
			     ntohl(tdb->tdb_inext->tdb_spi),
			     ipsp_address(tdb->tdb_inext->tdb_dst),
			     tdb->tdb_inext->tdb_sproto);

	      for (i = 0, flow = tdb->tdb_flow; flow; flow = flow->flow_next)
		i++;

	      l+= sprintf(buffer + l, "\tCurrently used by %d flows\n", i);

	      l += sprintf(buffer + l, "\t%u flows have used this SA\n",
			   tdb->tdb_cur_allocations);
	    
	      l += sprintf(buffer + l, "\t%qu bytes processed by this SA\n",
			 tdb->tdb_cur_bytes);
	    
	      l += sprintf(buffer + l, "\tExpirations:\n");

	      if (tdb->tdb_flags & TDBF_TIMER)
		l += sprintf(buffer + l,
			     "\t\tHard expiration(1) in %qu seconds\n",
			     tdb->tdb_exp_timeout - time.tv_sec);
	    
	      if (tdb->tdb_flags & TDBF_SOFT_TIMER)
		l += sprintf(buffer + l,
			     "\t\tSoft expiration(1) in %qu seconds\n",
			     tdb->tdb_soft_timeout - time.tv_sec);
	    
	      if (tdb->tdb_flags & TDBF_BYTES)
		l += sprintf(buffer + l,
			     "\t\tHard expiration after %qu bytes\n",
			     tdb->tdb_exp_bytes);
	    
	      if (tdb->tdb_flags & TDBF_SOFT_BYTES)
		l += sprintf(buffer + l,
			     "\t\tSoft expiration after %qu bytes\n",
			     tdb->tdb_soft_bytes);

	      if (tdb->tdb_flags & TDBF_ALLOCATIONS)
		l += sprintf(buffer + l,
			     "\t\tHard expiration after %u flows\n",
			     tdb->tdb_exp_allocations);
	    
	      if (tdb->tdb_flags & TDBF_SOFT_ALLOCATIONS)
		l += sprintf(buffer + l,
			     "\t\tSoft expiration after %u flows\n",
			     tdb->tdb_soft_allocations);

	      if (tdb->tdb_flags & TDBF_FIRSTUSE)
	      {
		  if (tdb->tdb_first_use)
		    l += sprintf(buffer + l,
				 "\t\tHard expiration(2) in %qu seconds\n",
				 (tdb->tdb_first_use +
				  tdb->tdb_exp_first_use) - time.tv_sec);
		  else
		    l += sprintf(buffer + l,
				 "\t\tHard expiration in %qu seconds "
				 "after first use\n",
				 tdb->tdb_exp_first_use);
	      }

	      if (tdb->tdb_flags & TDBF_SOFT_FIRSTUSE)
	      {
		  if (tdb->tdb_first_use)
		    l += sprintf(buffer + l,
				 "\t\tSoft expiration(2) in %qu seconds\n",
				 (tdb->tdb_first_use +
				  tdb->tdb_soft_first_use) - time.tv_sec);
		  else
		    l += sprintf(buffer + l,
				 "\t\tSoft expiration in %qu seconds "
				 "after first use\n",
				 tdb->tdb_soft_first_use);
	      }

	      if (!(tdb->tdb_flags &
		    (TDBF_TIMER | TDBF_SOFT_TIMER | TDBF_BYTES |
		     TDBF_SOFT_ALLOCATIONS | TDBF_ALLOCATIONS |
		     TDBF_SOFT_BYTES | TDBF_FIRSTUSE | TDBF_SOFT_FIRSTUSE)))
		l += sprintf(buffer + l, "\t\t(none)\n");

	      l += sprintf(buffer + l, "\n");

	      splx(s);
	      return l;
	  }
	splx(s);
    }
    return 0;
}

char *
inet_ntoa4(struct in_addr ina)
{
    static char buf[4][4 * sizeof "123"];
    unsigned char *ucp = (unsigned char *) &ina;
    static int i = 1;
 
    i = (i + 1) % 2;
    sprintf(buf[i], "%d.%d.%d.%d", ucp[0] & 0xff, ucp[1] & 0xff,
            ucp[2] & 0xff, ucp[3] & 0xff);
    return (buf[i]);
}

char *
ipsp_address(union sockaddr_union sa)
{
    switch (sa.sa.sa_family)
    {
	case AF_INET:
	    return inet_ntoa4(sa.sin.sin_addr);

#if INET6
	    /* XXX Add AF_INET6 support here */
#endif /* INET6 */

	default:
	    return "(unknown address family)";
    }
}