summaryrefslogtreecommitdiffstats
path: root/freebsd/sys/netinet6/nd6.c
blob: ead446201ad9042470e3d83889e9677f9ecf55a6 (plain) (blame)
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
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
#include <machine/rtems-bsd-kernel-space.h>

/*-
 * 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.
 *
 *	$KAME: nd6.c,v 1.144 2001/05/24 07:44:00 itojun Exp $
 */

#include <sys/cdefs.h>
__FBSDID("$FreeBSD$");

#include <rtems/bsd/local/opt_inet.h>
#include <rtems/bsd/local/opt_inet6.h>

#include <rtems/bsd/sys/param.h>
#include <sys/systm.h>
#include <sys/callout.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/sockio.h>
#include <rtems/bsd/sys/time.h>
#include <sys/kernel.h>
#include <sys/protosw.h>
#include <rtems/bsd/sys/errno.h>
#include <sys/syslog.h>
#include <rtems/bsd/sys/lock.h>
#include <sys/rwlock.h>
#include <sys/queue.h>
#include <sys/sysctl.h>

#include <net/if.h>
#include <net/if_arc.h>
#include <net/if_dl.h>
#include <net/if_types.h>
#include <net/iso88025.h>
#include <net/fddi.h>
#include <net/route.h>
#include <net/vnet.h>

#include <netinet/in.h>
#include <net/if_llatbl.h>
#define	L3_ADDR_SIN6(le)	((struct sockaddr_in6 *) L3_ADDR(le))
#include <netinet/if_ether.h>
#include <netinet6/in6_var.h>
#include <netinet/ip6.h>
#include <netinet6/ip6_var.h>
#include <netinet6/scope6_var.h>
#include <netinet6/nd6.h>
#include <netinet/icmp6.h>

#include <sys/limits.h>

#include <security/mac/mac_framework.h>

#define ND6_SLOWTIMER_INTERVAL (60 * 60) /* 1 hour */
#define ND6_RECALC_REACHTM_INTERVAL (60 * 120) /* 2 hours */

#define SIN6(s) ((struct sockaddr_in6 *)s)

/* timer values */
VNET_DEFINE(int, nd6_prune)	= 1;	/* walk list every 1 seconds */
VNET_DEFINE(int, nd6_delay)	= 5;	/* delay first probe time 5 second */
VNET_DEFINE(int, nd6_umaxtries)	= 3;	/* maximum unicast query */
VNET_DEFINE(int, nd6_mmaxtries)	= 3;	/* maximum multicast query */
VNET_DEFINE(int, nd6_useloopback) = 1;	/* use loopback interface for
					 * local traffic */
VNET_DEFINE(int, nd6_gctimer)	= (60 * 60 * 24); /* 1 day: garbage
					 * collection timer */

/* preventing too many loops in ND option parsing */
static VNET_DEFINE(int, nd6_maxndopt) = 10; /* max # of ND options allowed */

VNET_DEFINE(int, nd6_maxnudhint) = 0;	/* max # of subsequent upper
					 * layer hints */
static VNET_DEFINE(int, nd6_maxqueuelen) = 1; /* max pkts cached in unresolved
					 * ND entries */
#define	V_nd6_maxndopt			VNET(nd6_maxndopt)
#define	V_nd6_maxqueuelen		VNET(nd6_maxqueuelen)

#ifdef ND6_DEBUG
VNET_DEFINE(int, nd6_debug) = 1;
#else
VNET_DEFINE(int, nd6_debug) = 0;
#endif

/* for debugging? */
#if 0
static int nd6_inuse, nd6_allocated;
#endif

VNET_DEFINE(struct nd_drhead, nd_defrouter);
VNET_DEFINE(struct nd_prhead, nd_prefix);

VNET_DEFINE(int, nd6_recalc_reachtm_interval) = ND6_RECALC_REACHTM_INTERVAL;
#define	V_nd6_recalc_reachtm_interval	VNET(nd6_recalc_reachtm_interval)

static struct sockaddr_in6 all1_sa;

static int nd6_is_new_addr_neighbor __P((struct sockaddr_in6 *,
	struct ifnet *));
static void nd6_setmtu0(struct ifnet *, struct nd_ifinfo *);
static void nd6_slowtimo(void *);
static int regen_tmpaddr(struct in6_ifaddr *);
static struct llentry *nd6_free(struct llentry *, int);
static void nd6_llinfo_timer(void *);
static void clear_llinfo_pqueue(struct llentry *);

static VNET_DEFINE(struct callout, nd6_slowtimo_ch);
#define	V_nd6_slowtimo_ch		VNET(nd6_slowtimo_ch)

VNET_DEFINE(struct callout, nd6_timer_ch);

void
nd6_init(void)
{
	int i;

	LIST_INIT(&V_nd_prefix);

	all1_sa.sin6_family = AF_INET6;
	all1_sa.sin6_len = sizeof(struct sockaddr_in6);
	for (i = 0; i < sizeof(all1_sa.sin6_addr); i++)
		all1_sa.sin6_addr.s6_addr[i] = 0xff;

	/* initialization of the default router list */
	TAILQ_INIT(&V_nd_defrouter);

	/* start timer */
	callout_init(&V_nd6_slowtimo_ch, 0);
	callout_reset(&V_nd6_slowtimo_ch, ND6_SLOWTIMER_INTERVAL * hz,
	    nd6_slowtimo, curvnet);
}

#ifdef VIMAGE
void
nd6_destroy()
{

	callout_drain(&V_nd6_slowtimo_ch);
	callout_drain(&V_nd6_timer_ch);
}
#endif

struct nd_ifinfo *
nd6_ifattach(struct ifnet *ifp)
{
	struct nd_ifinfo *nd;

	nd = (struct nd_ifinfo *)malloc(sizeof(*nd), M_IP6NDP, M_WAITOK);
	bzero(nd, sizeof(*nd));

	nd->initialized = 1;

	nd->chlim = IPV6_DEFHLIM;
	nd->basereachable = REACHABLE_TIME;
	nd->reachable = ND_COMPUTE_RTIME(nd->basereachable);
	nd->retrans = RETRANS_TIMER;
	/*
	 * Note that the default value of ip6_accept_rtadv is 0, which means
	 * we won't accept RAs by default even if we set ND6_IFF_ACCEPT_RTADV
	 * here.
	 */
	nd->flags = (ND6_IFF_PERFORMNUD | ND6_IFF_ACCEPT_RTADV);

	/* XXX: we cannot call nd6_setmtu since ifp is not fully initialized */
	nd6_setmtu0(ifp, nd);

	return nd;
}

void
nd6_ifdetach(struct nd_ifinfo *nd)
{

	free(nd, M_IP6NDP);
}

/*
 * Reset ND level link MTU. This function is called when the physical MTU
 * changes, which means we might have to adjust the ND level MTU.
 */
void
nd6_setmtu(struct ifnet *ifp)
{

	nd6_setmtu0(ifp, ND_IFINFO(ifp));
}

/* XXX todo: do not maintain copy of ifp->if_mtu in ndi->maxmtu */
void
nd6_setmtu0(struct ifnet *ifp, struct nd_ifinfo *ndi)
{
	u_int32_t omaxmtu;

	omaxmtu = ndi->maxmtu;

	switch (ifp->if_type) {
	case IFT_ARCNET:
		ndi->maxmtu = MIN(ARC_PHDS_MAXMTU, ifp->if_mtu); /* RFC2497 */
		break;
	case IFT_FDDI:
		ndi->maxmtu = MIN(FDDIIPMTU, ifp->if_mtu); /* RFC2467 */
		break;
	case IFT_ISO88025:
		 ndi->maxmtu = MIN(ISO88025_MAX_MTU, ifp->if_mtu);
		 break;
	default:
		ndi->maxmtu = ifp->if_mtu;
		break;
	}

	/*
	 * Decreasing the interface MTU under IPV6 minimum MTU may cause
	 * undesirable situation.  We thus notify the operator of the change
	 * explicitly.  The check for omaxmtu is necessary to restrict the
	 * log to the case of changing the MTU, not initializing it.
	 */
	if (omaxmtu >= IPV6_MMTU && ndi->maxmtu < IPV6_MMTU) {
		log(LOG_NOTICE, "nd6_setmtu0: "
		    "new link MTU on %s (%lu) is too small for IPv6\n",
		    if_name(ifp), (unsigned long)ndi->maxmtu);
	}

	if (ndi->maxmtu > V_in6_maxmtu)
		in6_setmaxmtu(); /* check all interfaces just in case */

}

void
nd6_option_init(void *opt, int icmp6len, union nd_opts *ndopts)
{

	bzero(ndopts, sizeof(*ndopts));
	ndopts->nd_opts_search = (struct nd_opt_hdr *)opt;
	ndopts->nd_opts_last
		= (struct nd_opt_hdr *)(((u_char *)opt) + icmp6len);

	if (icmp6len == 0) {
		ndopts->nd_opts_done = 1;
		ndopts->nd_opts_search = NULL;
	}
}

/*
 * Take one ND option.
 */
struct nd_opt_hdr *
nd6_option(union nd_opts *ndopts)
{
	struct nd_opt_hdr *nd_opt;
	int olen;

	if (ndopts == NULL)
		panic("ndopts == NULL in nd6_option");
	if (ndopts->nd_opts_last == NULL)
		panic("uninitialized ndopts in nd6_option");
	if (ndopts->nd_opts_search == NULL)
		return NULL;
	if (ndopts->nd_opts_done)
		return NULL;

	nd_opt = ndopts->nd_opts_search;

	/* make sure nd_opt_len is inside the buffer */
	if ((caddr_t)&nd_opt->nd_opt_len >= (caddr_t)ndopts->nd_opts_last) {
		bzero(ndopts, sizeof(*ndopts));
		return NULL;
	}

	olen = nd_opt->nd_opt_len << 3;
	if (olen == 0) {
		/*
		 * Message validation requires that all included
		 * options have a length that is greater than zero.
		 */
		bzero(ndopts, sizeof(*ndopts));
		return NULL;
	}

	ndopts->nd_opts_search = (struct nd_opt_hdr *)((caddr_t)nd_opt + olen);
	if (ndopts->nd_opts_search > ndopts->nd_opts_last) {
		/* option overruns the end of buffer, invalid */
		bzero(ndopts, sizeof(*ndopts));
		return NULL;
	} else if (ndopts->nd_opts_search == ndopts->nd_opts_last) {
		/* reached the end of options chain */
		ndopts->nd_opts_done = 1;
		ndopts->nd_opts_search = NULL;
	}
	return nd_opt;
}

/*
 * Parse multiple ND options.
 * This function is much easier to use, for ND routines that do not need
 * multiple options of the same type.
 */
int
nd6_options(union nd_opts *ndopts)
{
	struct nd_opt_hdr *nd_opt;
	int i = 0;

	if (ndopts == NULL)
		panic("ndopts == NULL in nd6_options");
	if (ndopts->nd_opts_last == NULL)
		panic("uninitialized ndopts in nd6_options");
	if (ndopts->nd_opts_search == NULL)
		return 0;

	while (1) {
		nd_opt = nd6_option(ndopts);
		if (nd_opt == NULL && ndopts->nd_opts_last == NULL) {
			/*
			 * Message validation requires that all included
			 * options have a length that is greater than zero.
			 */
			ICMP6STAT_INC(icp6s_nd_badopt);
			bzero(ndopts, sizeof(*ndopts));
			return -1;
		}

		if (nd_opt == NULL)
			goto skip1;

		switch (nd_opt->nd_opt_type) {
		case ND_OPT_SOURCE_LINKADDR:
		case ND_OPT_TARGET_LINKADDR:
		case ND_OPT_MTU:
		case ND_OPT_REDIRECTED_HEADER:
			if (ndopts->nd_opt_array[nd_opt->nd_opt_type]) {
				nd6log((LOG_INFO,
				    "duplicated ND6 option found (type=%d)\n",
				    nd_opt->nd_opt_type));
				/* XXX bark? */
			} else {
				ndopts->nd_opt_array[nd_opt->nd_opt_type]
					= nd_opt;
			}
			break;
		case ND_OPT_PREFIX_INFORMATION:
			if (ndopts->nd_opt_array[nd_opt->nd_opt_type] == 0) {
				ndopts->nd_opt_array[nd_opt->nd_opt_type]
					= nd_opt;
			}
			ndopts->nd_opts_pi_end =
				(struct nd_opt_prefix_info *)nd_opt;
			break;
		/* What about ND_OPT_ROUTE_INFO? RFC 4191 */
		case ND_OPT_RDNSS:	/* RFC 6106 */
		case ND_OPT_DNSSL:	/* RFC 6106 */
			/*
			 * Silently ignore options we know and do not care about
			 * in the kernel.
			 */
			break;
		default:
			/*
			 * Unknown options must be silently ignored,
			 * to accomodate future extension to the protocol.
			 */
			nd6log((LOG_DEBUG,
			    "nd6_options: unsupported option %d - "
			    "option ignored\n", nd_opt->nd_opt_type));
		}

skip1:
		i++;
		if (i > V_nd6_maxndopt) {
			ICMP6STAT_INC(icp6s_nd_toomanyopt);
			nd6log((LOG_INFO, "too many loop in nd opt\n"));
			break;
		}

		if (ndopts->nd_opts_done)
			break;
	}

	return 0;
}

/*
 * ND6 timer routine to handle ND6 entries
 */
void
nd6_llinfo_settimer_locked(struct llentry *ln, long tick)
{
	int canceled;

	LLE_WLOCK_ASSERT(ln);

	if (tick < 0) {
		ln->la_expire = 0;
		ln->ln_ntick = 0;
		canceled = callout_stop(&ln->ln_timer_ch);
	} else {
		ln->la_expire = time_second + tick / hz;
		LLE_ADDREF(ln);
		if (tick > INT_MAX) {
			ln->ln_ntick = tick - INT_MAX;
			canceled = callout_reset(&ln->ln_timer_ch, INT_MAX,
			    nd6_llinfo_timer, ln);
		} else {
			ln->ln_ntick = 0;
			canceled = callout_reset(&ln->ln_timer_ch, tick,
			    nd6_llinfo_timer, ln);
		}
	}
	if (canceled)
		LLE_REMREF(ln);
}

void
nd6_llinfo_settimer(struct llentry *ln, long tick)
{

	LLE_WLOCK(ln);
	nd6_llinfo_settimer_locked(ln, tick);
	LLE_WUNLOCK(ln);
}

static void
nd6_llinfo_timer(void *arg)
{
	struct llentry *ln;
	struct in6_addr *dst;
	struct ifnet *ifp;
	struct nd_ifinfo *ndi = NULL;

	KASSERT(arg != NULL, ("%s: arg NULL", __func__));
	ln = (struct llentry *)arg;
	LLE_WLOCK_ASSERT(ln);
	ifp = ln->lle_tbl->llt_ifp;

	CURVNET_SET(ifp->if_vnet);

	if (ln->ln_ntick > 0) {
		if (ln->ln_ntick > INT_MAX) {
			ln->ln_ntick -= INT_MAX;
			nd6_llinfo_settimer_locked(ln, INT_MAX);
		} else {
			ln->ln_ntick = 0;
			nd6_llinfo_settimer_locked(ln, ln->ln_ntick);
		}
		goto done;
	}

	ndi = ND_IFINFO(ifp);
	dst = &L3_ADDR_SIN6(ln)->sin6_addr;
	if (ln->la_flags & LLE_STATIC) {
		goto done;
	}

	if (ln->la_flags & LLE_DELETED) {
		(void)nd6_free(ln, 0);
		ln = NULL;
		goto done;
	}

	switch (ln->ln_state) {
	case ND6_LLINFO_INCOMPLETE:
		if (ln->la_asked < V_nd6_mmaxtries) {
			ln->la_asked++;
			nd6_llinfo_settimer_locked(ln, (long)ndi->retrans * hz / 1000);
			LLE_WUNLOCK(ln);
			nd6_ns_output(ifp, NULL, dst, ln, 0);
			LLE_WLOCK(ln);
		} else {
			struct mbuf *m = ln->la_hold;
			if (m) {
				struct mbuf *m0;

				/*
				 * assuming every packet in la_hold has the
				 * same IP header.  Send error after unlock.
				 */
				m0 = m->m_nextpkt;
				m->m_nextpkt = NULL;
				ln->la_hold = m0;
				clear_llinfo_pqueue(ln);
			}
			(void)nd6_free(ln, 0);
			ln = NULL;
			if (m != NULL)
				icmp6_error2(m, ICMP6_DST_UNREACH,
				    ICMP6_DST_UNREACH_ADDR, 0, ifp);
		}
		break;
	case ND6_LLINFO_REACHABLE:
		if (!ND6_LLINFO_PERMANENT(ln)) {
			ln->ln_state = ND6_LLINFO_STALE;
			nd6_llinfo_settimer_locked(ln, (long)V_nd6_gctimer * hz);
		}
		break;

	case ND6_LLINFO_STALE:
		/* Garbage Collection(RFC 2461 5.3) */
		if (!ND6_LLINFO_PERMANENT(ln)) {
			(void)nd6_free(ln, 1);
			ln = NULL;
		}
		break;

	case ND6_LLINFO_DELAY:
		if (ndi && (ndi->flags & ND6_IFF_PERFORMNUD) != 0) {
			/* We need NUD */
			ln->la_asked = 1;
			ln->ln_state = ND6_LLINFO_PROBE;
			nd6_llinfo_settimer_locked(ln, (long)ndi->retrans * hz / 1000);
			LLE_WUNLOCK(ln);
			nd6_ns_output(ifp, dst, dst, ln, 0);
			LLE_WLOCK(ln);
		} else {
			ln->ln_state = ND6_LLINFO_STALE; /* XXX */
			nd6_llinfo_settimer_locked(ln, (long)V_nd6_gctimer * hz);
		}
		break;
	case ND6_LLINFO_PROBE:
		if (ln->la_asked < V_nd6_umaxtries) {
			ln->la_asked++;
			nd6_llinfo_settimer_locked(ln, (long)ndi->retrans * hz / 1000);
			LLE_WUNLOCK(ln);
			nd6_ns_output(ifp, dst, dst, ln, 0);
			LLE_WLOCK(ln);
		} else {
			(void)nd6_free(ln, 0);
			ln = NULL;
		}
		break;
	default:
		panic("%s: paths in a dark night can be confusing: %d",
		    __func__, ln->ln_state);
	}
done:
	if (ln != NULL)
		LLE_FREE_LOCKED(ln);
	CURVNET_RESTORE();
}


/*
 * ND6 timer routine to expire default route list and prefix list
 */
void
nd6_timer(void *arg)
{
	CURVNET_SET((struct vnet *) arg);
	int s;
	struct nd_defrouter *dr, *ndr;
	struct nd_prefix *pr, *npr;
	struct in6_ifaddr *ia6, *nia6;

	callout_reset(&V_nd6_timer_ch, V_nd6_prune * hz,
	    nd6_timer, curvnet);

	/* expire default router list */
	s = splnet();
	TAILQ_FOREACH_SAFE(dr, &V_nd_defrouter, dr_entry, ndr) {
		if (dr->expire && dr->expire < time_second)
			defrtrlist_del(dr);
	}

	/*
	 * expire interface addresses.
	 * in the past the loop was inside prefix expiry processing.
	 * However, from a stricter speci-confrmance standpoint, we should
	 * rather separate address lifetimes and prefix lifetimes.
	 *
	 * XXXRW: in6_ifaddrhead locking.
	 */
  addrloop:
	TAILQ_FOREACH_SAFE(ia6, &V_in6_ifaddrhead, ia_link, nia6) {
		/* check address lifetime */
		if (IFA6_IS_INVALID(ia6)) {
			int regen = 0;

			/*
			 * If the expiring address is temporary, try
			 * regenerating a new one.  This would be useful when
			 * we suspended a laptop PC, then turned it on after a
			 * period that could invalidate all temporary
			 * addresses.  Although we may have to restart the
			 * loop (see below), it must be after purging the
			 * address.  Otherwise, we'd see an infinite loop of
			 * regeneration.
			 */
			if (V_ip6_use_tempaddr &&
			    (ia6->ia6_flags & IN6_IFF_TEMPORARY) != 0) {
				if (regen_tmpaddr(ia6) == 0)
					regen = 1;
			}

			in6_purgeaddr(&ia6->ia_ifa);

			if (regen)
				goto addrloop; /* XXX: see below */
		} else if (IFA6_IS_DEPRECATED(ia6)) {
			int oldflags = ia6->ia6_flags;

			ia6->ia6_flags |= IN6_IFF_DEPRECATED;

			/*
			 * If a temporary address has just become deprecated,
			 * regenerate a new one if possible.
			 */
			if (V_ip6_use_tempaddr &&
			    (ia6->ia6_flags & IN6_IFF_TEMPORARY) != 0 &&
			    (oldflags & IN6_IFF_DEPRECATED) == 0) {

				if (regen_tmpaddr(ia6) == 0) {
					/*
					 * A new temporary address is
					 * generated.
					 * XXX: this means the address chain
					 * has changed while we are still in
					 * the loop.  Although the change
					 * would not cause disaster (because
					 * it's not a deletion, but an
					 * addition,) we'd rather restart the
					 * loop just for safety.  Or does this
					 * significantly reduce performance??
					 */
					goto addrloop;
				}
			}
		} else {
			/*
			 * A new RA might have made a deprecated address
			 * preferred.
			 */
			ia6->ia6_flags &= ~IN6_IFF_DEPRECATED;
		}
	}

	/* expire prefix list */
	LIST_FOREACH_SAFE(pr, &V_nd_prefix, ndpr_entry, npr) {
		/*
		 * check prefix lifetime.
		 * since pltime is just for autoconf, pltime processing for
		 * prefix is not necessary.
		 */
		if (pr->ndpr_vltime != ND6_INFINITE_LIFETIME &&
		    time_second - pr->ndpr_lastupdate > pr->ndpr_vltime) {

			/*
			 * address expiration and prefix expiration are
			 * separate.  NEVER perform in6_purgeaddr here.
			 */
			prelist_remove(pr);
		}
	}
	splx(s);
	CURVNET_RESTORE();
}

/*
 * ia6 - deprecated/invalidated temporary address
 */
static int
regen_tmpaddr(struct in6_ifaddr *ia6)
{
	struct ifaddr *ifa;
	struct ifnet *ifp;
	struct in6_ifaddr *public_ifa6 = NULL;

	ifp = ia6->ia_ifa.ifa_ifp;
	IF_ADDR_RLOCK(ifp);
	TAILQ_FOREACH(ifa, &ifp->if_addrhead, ifa_link) {
		struct in6_ifaddr *it6;

		if (ifa->ifa_addr->sa_family != AF_INET6)
			continue;

		it6 = (struct in6_ifaddr *)ifa;

		/* ignore no autoconf addresses. */
		if ((it6->ia6_flags & IN6_IFF_AUTOCONF) == 0)
			continue;

		/* ignore autoconf addresses with different prefixes. */
		if (it6->ia6_ndpr == NULL || it6->ia6_ndpr != ia6->ia6_ndpr)
			continue;

		/*
		 * Now we are looking at an autoconf address with the same
		 * prefix as ours.  If the address is temporary and is still
		 * preferred, do not create another one.  It would be rare, but
		 * could happen, for example, when we resume a laptop PC after
		 * a long period.
		 */
		if ((it6->ia6_flags & IN6_IFF_TEMPORARY) != 0 &&
		    !IFA6_IS_DEPRECATED(it6)) {
			public_ifa6 = NULL;
			break;
		}

		/*
		 * This is a public autoconf address that has the same prefix
		 * as ours.  If it is preferred, keep it.  We can't break the
		 * loop here, because there may be a still-preferred temporary
		 * address with the prefix.
		 */
		if (!IFA6_IS_DEPRECATED(it6))
		    public_ifa6 = it6;

		if (public_ifa6 != NULL)
			ifa_ref(&public_ifa6->ia_ifa);
	}
	IF_ADDR_RUNLOCK(ifp);

	if (public_ifa6 != NULL) {
		int e;

		if ((e = in6_tmpifadd(public_ifa6, 0, 0)) != 0) {
			ifa_free(&public_ifa6->ia_ifa);
			log(LOG_NOTICE, "regen_tmpaddr: failed to create a new"
			    " tmp addr,errno=%d\n", e);
			return (-1);
		}
		ifa_free(&public_ifa6->ia_ifa);
		return (0);
	}

	return (-1);
}

/*
 * Nuke neighbor cache/prefix/default router management table, right before
 * ifp goes away.
 */
void
nd6_purge(struct ifnet *ifp)
{
	struct nd_defrouter *dr, *ndr;
	struct nd_prefix *pr, *npr;

	/*
	 * Nuke default router list entries toward ifp.
	 * We defer removal of default router list entries that is installed
	 * in the routing table, in order to keep additional side effects as
	 * small as possible.
	 */
	TAILQ_FOREACH_SAFE(dr, &V_nd_defrouter, dr_entry, ndr) {
		if (dr->installed)
			continue;

		if (dr->ifp == ifp)
			defrtrlist_del(dr);
	}

	TAILQ_FOREACH_SAFE(dr, &V_nd_defrouter, dr_entry, ndr) {
		if (!dr->installed)
			continue;

		if (dr->ifp == ifp)
			defrtrlist_del(dr);
	}

	/* Nuke prefix list entries toward ifp */
	LIST_FOREACH_SAFE(pr, &V_nd_prefix, ndpr_entry, npr) {
		if (pr->ndpr_ifp == ifp) {
			/*
			 * Because if_detach() does *not* release prefixes
			 * while purging addresses the reference count will
			 * still be above zero. We therefore reset it to
			 * make sure that the prefix really gets purged.
			 */
			pr->ndpr_refcnt = 0;

			/*
			 * Previously, pr->ndpr_addr is removed as well,
			 * but I strongly believe we don't have to do it.
			 * nd6_purge() is only called from in6_ifdetach(),
			 * which removes all the associated interface addresses
			 * by itself.
			 * (jinmei@kame.net 20010129)
			 */
			prelist_remove(pr);
		}
	}

	/* cancel default outgoing interface setting */
	if (V_nd6_defifindex == ifp->if_index)
		nd6_setdefaultiface(0);

	if (!V_ip6_forwarding && V_ip6_accept_rtadv) { /* XXX: too restrictive? */
		/* refresh default router list
		 *
		 * 
		 */
		defrouter_select();

	}

	/* XXXXX
	 * We do not nuke the neighbor cache entries here any more
	 * because the neighbor cache is kept in if_afdata[AF_INET6].
	 * nd6_purge() is invoked by in6_ifdetach() which is called
	 * from if_detach() where everything gets purged. So let
	 * in6_domifdetach() do the actual L2 table purging work.
	 */
}

/* 
 * the caller acquires and releases the lock on the lltbls
 * Returns the llentry locked
 */
struct llentry *
nd6_lookup(struct in6_addr *addr6, int flags, struct ifnet *ifp)
{
	struct sockaddr_in6 sin6;
	struct llentry *ln;
	int llflags;
	
	bzero(&sin6, sizeof(sin6));
	sin6.sin6_len = sizeof(struct sockaddr_in6);
	sin6.sin6_family = AF_INET6;
	sin6.sin6_addr = *addr6;

	IF_AFDATA_LOCK_ASSERT(ifp);

	llflags = 0;
	if (flags & ND6_CREATE)
	    llflags |= LLE_CREATE;
	if (flags & ND6_EXCLUSIVE)
	    llflags |= LLE_EXCLUSIVE;	
	
	ln = lla_lookup(LLTABLE6(ifp), llflags, (struct sockaddr *)&sin6);
	if ((ln != NULL) && (llflags & LLE_CREATE))
		ln->ln_state = ND6_LLINFO_NOSTATE;
	
	return (ln);
}

/*
 * Test whether a given IPv6 address is a neighbor or not, ignoring
 * the actual neighbor cache.  The neighbor cache is ignored in order
 * to not reenter the routing code from within itself.
 */
static int
nd6_is_new_addr_neighbor(struct sockaddr_in6 *addr, struct ifnet *ifp)
{
	struct nd_prefix *pr;
	struct ifaddr *dstaddr;

	/*
	 * A link-local address is always a neighbor.
	 * XXX: a link does not necessarily specify a single interface.
	 */
	if (IN6_IS_ADDR_LINKLOCAL(&addr->sin6_addr)) {
		struct sockaddr_in6 sin6_copy;
		u_int32_t zone;

		/*
		 * We need sin6_copy since sa6_recoverscope() may modify the
		 * content (XXX).
		 */
		sin6_copy = *addr;
		if (sa6_recoverscope(&sin6_copy))
			return (0); /* XXX: should be impossible */
		if (in6_setscope(&sin6_copy.sin6_addr, ifp, &zone))
			return (0);
		if (sin6_copy.sin6_scope_id == zone)
			return (1);
		else
			return (0);
	}

	/*
	 * If the address matches one of our addresses,
	 * it should be a neighbor.
	 * If the address matches one of our on-link prefixes, it should be a
	 * neighbor.
	 */
	LIST_FOREACH(pr, &V_nd_prefix, ndpr_entry) {
		if (pr->ndpr_ifp != ifp)
			continue;

		if (!(pr->ndpr_stateflags & NDPRF_ONLINK)) {
			struct rtentry *rt;

			/* Always use the default FIB here. */
			rt = in6_rtalloc1((struct sockaddr *)&pr->ndpr_prefix,
			    0, 0, RT_DEFAULT_FIB);
			if (rt == NULL)
				continue;
			/*
			 * This is the case where multiple interfaces
			 * have the same prefix, but only one is installed 
			 * into the routing table and that prefix entry
			 * is not the one being examined here. In the case
			 * where RADIX_MPATH is enabled, multiple route
			 * entries (of the same rt_key value) will be 
			 * installed because the interface addresses all
			 * differ.
			 */
			if (!IN6_ARE_ADDR_EQUAL(&pr->ndpr_prefix.sin6_addr,
			       &((struct sockaddr_in6 *)rt_key(rt))->sin6_addr)) {
				RTFREE_LOCKED(rt);
				continue;
			}
			RTFREE_LOCKED(rt);
		}

		if (IN6_ARE_MASKED_ADDR_EQUAL(&pr->ndpr_prefix.sin6_addr,
		    &addr->sin6_addr, &pr->ndpr_mask))
			return (1);
	}

	/*
	 * If the address is assigned on the node of the other side of
	 * a p2p interface, the address should be a neighbor.
	 */
	dstaddr = ifa_ifwithdstaddr((struct sockaddr *)addr);
	if (dstaddr != NULL) {
		if (dstaddr->ifa_ifp == ifp) {
			ifa_free(dstaddr);
			return (1);
		}
		ifa_free(dstaddr);
	}

	/*
	 * If the default router list is empty, all addresses are regarded
	 * as on-link, and thus, as a neighbor.
	 * XXX: we restrict the condition to hosts, because routers usually do
	 * not have the "default router list".
	 */
	if (!V_ip6_forwarding && TAILQ_EMPTY(&V_nd_defrouter) &&
	    V_nd6_defifindex == ifp->if_index) {
		return (1);
	}

	return (0);
}


/*
 * Detect if a given IPv6 address identifies a neighbor on a given link.
 * XXX: should take care of the destination of a p2p link?
 */
int
nd6_is_addr_neighbor(struct sockaddr_in6 *addr, struct ifnet *ifp)
{
	struct llentry *lle;
	int rc = 0;

	IF_AFDATA_UNLOCK_ASSERT(ifp);
	if (nd6_is_new_addr_neighbor(addr, ifp))
		return (1);

	/*
	 * Even if the address matches none of our addresses, it might be
	 * in the neighbor cache.
	 */
	IF_AFDATA_RLOCK(ifp);
	if ((lle = nd6_lookup(&addr->sin6_addr, 0, ifp)) != NULL) {
		LLE_RUNLOCK(lle);
		rc = 1;
	}
	IF_AFDATA_RUNLOCK(ifp);
	return (rc);
}

/*
 * Free an nd6 llinfo entry.
 * Since the function would cause significant changes in the kernel, DO NOT
 * make it global, unless you have a strong reason for the change, and are sure
 * that the change is safe.
 */
static struct llentry *
nd6_free(struct llentry *ln, int gc)
{
        struct llentry *next;
	struct nd_defrouter *dr;
	struct ifnet *ifp;

	LLE_WLOCK_ASSERT(ln);

	/*
	 * we used to have pfctlinput(PRC_HOSTDEAD) here.
	 * even though it is not harmful, it was not really necessary.
	 */

	/* cancel timer */
	nd6_llinfo_settimer_locked(ln, -1);

	ifp = ln->lle_tbl->llt_ifp;

	if (!V_ip6_forwarding) {

		dr = defrouter_lookup(&L3_ADDR_SIN6(ln)->sin6_addr, ifp);

		if (dr != NULL && dr->expire &&
		    ln->ln_state == ND6_LLINFO_STALE && gc) {
			/*
			 * If the reason for the deletion is just garbage
			 * collection, and the neighbor is an active default
			 * router, do not delete it.  Instead, reset the GC
			 * timer using the router's lifetime.
			 * Simply deleting the entry would affect default
			 * router selection, which is not necessarily a good
			 * thing, especially when we're using router preference
			 * values.
			 * XXX: the check for ln_state would be redundant,
			 *      but we intentionally keep it just in case.
			 */
			if (dr->expire > time_second)
				nd6_llinfo_settimer_locked(ln,
				    (dr->expire - time_second) * hz);
			else
				nd6_llinfo_settimer_locked(ln,
				    (long)V_nd6_gctimer * hz);

			next = LIST_NEXT(ln, lle_next);
			LLE_REMREF(ln);
			LLE_WUNLOCK(ln);
			return (next);
		}

		if (dr) {
			/*
			 * Unreachablity of a router might affect the default
			 * router selection and on-link detection of advertised
			 * prefixes.
			 */

			/*
			 * Temporarily fake the state to choose a new default
			 * router and to perform on-link determination of
			 * prefixes correctly.
			 * Below the state will be set correctly,
			 * or the entry itself will be deleted.
			 */
			ln->ln_state = ND6_LLINFO_INCOMPLETE;
		}

		if (ln->ln_router || dr) {

			/*
			 * We need to unlock to avoid a LOR with rt6_flush() with the
			 * rnh and for the calls to pfxlist_onlink_check() and
			 * defrouter_select() in the block further down for calls
			 * into nd6_lookup().  We still hold a ref.
			 */
			LLE_WUNLOCK(ln);

			/*
			 * rt6_flush must be called whether or not the neighbor
			 * is in the Default Router List.
			 * See a corresponding comment in nd6_na_input().
			 */
			rt6_flush(&L3_ADDR_SIN6(ln)->sin6_addr, ifp);
		}

		if (dr) {
			/*
			 * Since defrouter_select() does not affect the
			 * on-link determination and MIP6 needs the check
			 * before the default router selection, we perform
			 * the check now.
			 */
			pfxlist_onlink_check();

			/*
			 * Refresh default router list.
			 */
			defrouter_select();
		}

		if (ln->ln_router || dr)
			LLE_WLOCK(ln);
	}

	/*
	 * Before deleting the entry, remember the next entry as the
	 * return value.  We need this because pfxlist_onlink_check() above
	 * might have freed other entries (particularly the old next entry) as
	 * a side effect (XXX).
	 */
	next = LIST_NEXT(ln, lle_next);

	/*
	 * Save to unlock. We still hold an extra reference and will not
	 * free(9) in llentry_free() if someone else holds one as well.
	 */
	LLE_WUNLOCK(ln);
	IF_AFDATA_LOCK(ifp);
	LLE_WLOCK(ln);
	LLE_REMREF(ln);
	llentry_free(ln);
	IF_AFDATA_UNLOCK(ifp);

	return (next);
}

/*
 * Upper-layer reachability hint for Neighbor Unreachability Detection.
 *
 * XXX cost-effective methods?
 */
void
nd6_nud_hint(struct rtentry *rt, struct in6_addr *dst6, int force)
{
	struct llentry *ln;
	struct ifnet *ifp;

	if ((dst6 == NULL) || (rt == NULL))
		return;

	ifp = rt->rt_ifp;
	IF_AFDATA_LOCK(ifp);
	ln = nd6_lookup(dst6, ND6_EXCLUSIVE, NULL);
	IF_AFDATA_UNLOCK(ifp);
	if (ln == NULL)
		return;

	if (ln->ln_state < ND6_LLINFO_REACHABLE)
		goto done;

	/*
	 * if we get upper-layer reachability confirmation many times,
	 * it is possible we have false information.
	 */
	if (!force) {
		ln->ln_byhint++;
		if (ln->ln_byhint > V_nd6_maxnudhint) {
			goto done;
		}
	}

 	ln->ln_state = ND6_LLINFO_REACHABLE;
	if (!ND6_LLINFO_PERMANENT(ln)) {
		nd6_llinfo_settimer_locked(ln,
		    (long)ND_IFINFO(rt->rt_ifp)->reachable * hz);
	}
done:
	LLE_WUNLOCK(ln);
}


/*
 * Rejuvenate this function for routing operations related
 * processing.
 */
void
nd6_rtrequest(int req, struct rtentry *rt, struct rt_addrinfo *info)
{
	struct sockaddr_in6 *gateway = (struct sockaddr_in6 *)rt->rt_gateway;
	struct nd_defrouter *dr;
	struct ifnet *ifp = rt->rt_ifp;

	RT_LOCK_ASSERT(rt);

	switch (req) {
	case RTM_ADD:
		break;

	case RTM_DELETE:
		if (!ifp)
			return;
		/*
		 * Only indirect routes are interesting.
		 */
		if ((rt->rt_flags & RTF_GATEWAY) == 0)
			return;
		/*
		 * check for default route
		 */
		if (IN6_ARE_ADDR_EQUAL(&in6addr_any, 
				       &SIN6(rt_key(rt))->sin6_addr)) {

			dr = defrouter_lookup(&gateway->sin6_addr, ifp);
			if (dr != NULL)
				dr->installed = 0;
		}
		break;
	}
}


int
nd6_ioctl(u_long cmd, caddr_t data, struct ifnet *ifp)
{
	struct in6_drlist *drl = (struct in6_drlist *)data;
	struct in6_oprlist *oprl = (struct in6_oprlist *)data;
	struct in6_ndireq *ndi = (struct in6_ndireq *)data;
	struct in6_nbrinfo *nbi = (struct in6_nbrinfo *)data;
	struct in6_ndifreq *ndif = (struct in6_ndifreq *)data;
	struct nd_defrouter *dr;
	struct nd_prefix *pr;
	int i = 0, error = 0;
	int s;

	switch (cmd) {
	case SIOCGDRLST_IN6:
		/*
		 * obsolete API, use sysctl under net.inet6.icmp6
		 */
		bzero(drl, sizeof(*drl));
		s = splnet();
		TAILQ_FOREACH(dr, &V_nd_defrouter, dr_entry) {
			if (i >= DRLSTSIZ)
				break;
			drl->defrouter[i].rtaddr = dr->rtaddr;
			in6_clearscope(&drl->defrouter[i].rtaddr);

			drl->defrouter[i].flags = dr->flags;
			drl->defrouter[i].rtlifetime = dr->rtlifetime;
			drl->defrouter[i].expire = dr->expire;
			drl->defrouter[i].if_index = dr->ifp->if_index;
			i++;
		}
		splx(s);
		break;
	case SIOCGPRLST_IN6:
		/*
		 * obsolete API, use sysctl under net.inet6.icmp6
		 *
		 * XXX the structure in6_prlist was changed in backward-
		 * incompatible manner.  in6_oprlist is used for SIOCGPRLST_IN6,
		 * in6_prlist is used for nd6_sysctl() - fill_prlist().
		 */
		/*
		 * XXX meaning of fields, especialy "raflags", is very
		 * differnet between RA prefix list and RR/static prefix list.
		 * how about separating ioctls into two?
		 */
		bzero(oprl, sizeof(*oprl));
		s = splnet();
		LIST_FOREACH(pr, &V_nd_prefix, ndpr_entry) {
			struct nd_pfxrouter *pfr;
			int j;

			if (i >= PRLSTSIZ)
				break;
			oprl->prefix[i].prefix = pr->ndpr_prefix.sin6_addr;
			oprl->prefix[i].raflags = pr->ndpr_raf;
			oprl->prefix[i].prefixlen = pr->ndpr_plen;
			oprl->prefix[i].vltime = pr->ndpr_vltime;
			oprl->prefix[i].pltime = pr->ndpr_pltime;
			oprl->prefix[i].if_index = pr->ndpr_ifp->if_index;
			if (pr->ndpr_vltime == ND6_INFINITE_LIFETIME)
				oprl->prefix[i].expire = 0;
			else {
				time_t maxexpire;

				/* XXX: we assume time_t is signed. */
				maxexpire = (-1) &
				    ~((time_t)1 <<
				    ((sizeof(maxexpire) * 8) - 1));
				if (pr->ndpr_vltime <
				    maxexpire - pr->ndpr_lastupdate) {
					oprl->prefix[i].expire =
					    pr->ndpr_lastupdate +
					    pr->ndpr_vltime;
				} else
					oprl->prefix[i].expire = maxexpire;
			}

			j = 0;
			LIST_FOREACH(pfr, &pr->ndpr_advrtrs, pfr_entry) {
				if (j < DRLSTSIZ) {
#define RTRADDR oprl->prefix[i].advrtr[j]
					RTRADDR = pfr->router->rtaddr;
					in6_clearscope(&RTRADDR);
#undef RTRADDR
				}
				j++;
			}
			oprl->prefix[i].advrtrs = j;
			oprl->prefix[i].origin = PR_ORIG_RA;

			i++;
		}
		splx(s);

		break;
	case OSIOCGIFINFO_IN6:
#define ND	ndi->ndi
		/* XXX: old ndp(8) assumes a positive value for linkmtu. */
		bzero(&ND, sizeof(ND));
		ND.linkmtu = IN6_LINKMTU(ifp);
		ND.maxmtu = ND_IFINFO(ifp)->maxmtu;
		ND.basereachable = ND_IFINFO(ifp)->basereachable;
		ND.reachable = ND_IFINFO(ifp)->reachable;
		ND.retrans = ND_IFINFO(ifp)->retrans;
		ND.flags = ND_IFINFO(ifp)->flags;
		ND.recalctm = ND_IFINFO(ifp)->recalctm;
		ND.chlim = ND_IFINFO(ifp)->chlim;
		break;
	case SIOCGIFINFO_IN6:
		ND = *ND_IFINFO(ifp);
		break;
	case SIOCSIFINFO_IN6:
		/*
		 * used to change host variables from userland.
		 * intented for a use on router to reflect RA configurations.
		 */
		/* 0 means 'unspecified' */
		if (ND.linkmtu != 0) {
			if (ND.linkmtu < IPV6_MMTU ||
			    ND.linkmtu > IN6_LINKMTU(ifp)) {
				error = EINVAL;
				break;
			}
			ND_IFINFO(ifp)->linkmtu = ND.linkmtu;
		}

		if (ND.basereachable != 0) {
			int obasereachable = ND_IFINFO(ifp)->basereachable;

			ND_IFINFO(ifp)->basereachable = ND.basereachable;
			if (ND.basereachable != obasereachable)
				ND_IFINFO(ifp)->reachable =
				    ND_COMPUTE_RTIME(ND.basereachable);
		}
		if (ND.retrans != 0)
			ND_IFINFO(ifp)->retrans = ND.retrans;
		if (ND.chlim != 0)
			ND_IFINFO(ifp)->chlim = ND.chlim;
		/* FALLTHROUGH */
	case SIOCSIFINFO_FLAGS:
		ND_IFINFO(ifp)->flags = ND.flags;
		break;
#undef ND
	case SIOCSNDFLUSH_IN6:	/* XXX: the ioctl name is confusing... */
		/* sync kernel routing table with the default router list */
		defrouter_reset();
		defrouter_select();
		break;
	case SIOCSPFXFLUSH_IN6:
	{
		/* flush all the prefix advertised by routers */
		struct nd_prefix *pr, *next;

		s = splnet();
		LIST_FOREACH_SAFE(pr, &V_nd_prefix, ndpr_entry, next) {
			struct in6_ifaddr *ia, *ia_next;

			if (IN6_IS_ADDR_LINKLOCAL(&pr->ndpr_prefix.sin6_addr))
				continue; /* XXX */

			/* do we really have to remove addresses as well? */
			/* XXXRW: in6_ifaddrhead locking. */
			TAILQ_FOREACH_SAFE(ia, &V_in6_ifaddrhead, ia_link,
			    ia_next) {
				if ((ia->ia6_flags & IN6_IFF_AUTOCONF) == 0)
					continue;

				if (ia->ia6_ndpr == pr)
					in6_purgeaddr(&ia->ia_ifa);
			}
			prelist_remove(pr);
		}
		splx(s);
		break;
	}
	case SIOCSRTRFLUSH_IN6:
	{
		/* flush all the default routers */
		struct nd_defrouter *dr, *next;

		s = splnet();
		defrouter_reset();
		TAILQ_FOREACH_SAFE(dr, &V_nd_defrouter, dr_entry, next) {
			defrtrlist_del(dr);
		}
		defrouter_select();
		splx(s);
		break;
	}
	case SIOCGNBRINFO_IN6:
	{
		struct llentry *ln;
		struct in6_addr nb_addr = nbi->addr; /* make local for safety */

		if ((error = in6_setscope(&nb_addr, ifp, NULL)) != 0)
			return (error);

		IF_AFDATA_RLOCK(ifp);
		ln = nd6_lookup(&nb_addr, 0, ifp);
		IF_AFDATA_RUNLOCK(ifp);

		if (ln == NULL) {
			error = EINVAL;
			break;
		}
		nbi->state = ln->ln_state;
		nbi->asked = ln->la_asked;
		nbi->isrouter = ln->ln_router;
		nbi->expire = ln->la_expire;
		LLE_RUNLOCK(ln);
		break;
	}
	case SIOCGDEFIFACE_IN6:	/* XXX: should be implemented as a sysctl? */
		ndif->ifindex = V_nd6_defifindex;
		break;
	case SIOCSDEFIFACE_IN6:	/* XXX: should be implemented as a sysctl? */
		return (nd6_setdefaultiface(ndif->ifindex));
	}
	return (error);
}

/*
 * Create neighbor cache entry and cache link-layer address,
 * on reception of inbound ND6 packets.  (RS/RA/NS/redirect)
 *
 * type - ICMP6 type
 * code - type dependent information
 *
 * XXXXX
 *  The caller of this function already acquired the ndp 
 *  cache table lock because the cache entry is returned.
 */
struct llentry *
nd6_cache_lladdr(struct ifnet *ifp, struct in6_addr *from, char *lladdr,
    int lladdrlen, int type, int code)
{
	struct llentry *ln = NULL;
	int is_newentry;
	int do_update;
	int olladdr;
	int llchange;
	int flags;
	int newstate = 0;
	uint16_t router = 0;
	struct sockaddr_in6 sin6;
	struct mbuf *chain = NULL;
	int static_route = 0;

	IF_AFDATA_UNLOCK_ASSERT(ifp);

	if (ifp == NULL)
		panic("ifp == NULL in nd6_cache_lladdr");
	if (from == NULL)
		panic("from == NULL in nd6_cache_lladdr");

	/* nothing must be updated for unspecified address */
	if (IN6_IS_ADDR_UNSPECIFIED(from))
		return NULL;

	/*
	 * Validation about ifp->if_addrlen and lladdrlen must be done in
	 * the caller.
	 *
	 * XXX If the link does not have link-layer adderss, what should
	 * we do? (ifp->if_addrlen == 0)
	 * Spec says nothing in sections for RA, RS and NA.  There's small
	 * description on it in NS section (RFC 2461 7.2.3).
	 */
	flags = lladdr ? ND6_EXCLUSIVE : 0;
	IF_AFDATA_LOCK(ifp);
	ln = nd6_lookup(from, flags, ifp);

	if (ln == NULL) {
		flags |= ND6_EXCLUSIVE;
		ln = nd6_lookup(from, flags | ND6_CREATE, ifp);
		IF_AFDATA_UNLOCK(ifp);
		is_newentry = 1;
	} else {
		IF_AFDATA_UNLOCK(ifp);		
		/* do nothing if static ndp is set */
		if (ln->la_flags & LLE_STATIC) {
			static_route = 1;
			goto done;
		}
		is_newentry = 0;
	}
	if (ln == NULL)
		return (NULL);

	olladdr = (ln->la_flags & LLE_VALID) ? 1 : 0;
	if (olladdr && lladdr) {
		llchange = bcmp(lladdr, &ln->ll_addr,
		    ifp->if_addrlen);
	} else
		llchange = 0;

	/*
	 * newentry olladdr  lladdr  llchange	(*=record)
	 *	0	n	n	--	(1)
	 *	0	y	n	--	(2)
	 *	0	n	y	--	(3) * STALE
	 *	0	y	y	n	(4) *
	 *	0	y	y	y	(5) * STALE
	 *	1	--	n	--	(6)   NOSTATE(= PASSIVE)
	 *	1	--	y	--	(7) * STALE
	 */

	if (lladdr) {		/* (3-5) and (7) */
		/*
		 * Record source link-layer address
		 * XXX is it dependent to ifp->if_type?
		 */
		bcopy(lladdr, &ln->ll_addr, ifp->if_addrlen);
		ln->la_flags |= LLE_VALID;
	}

	if (!is_newentry) {
		if ((!olladdr && lladdr != NULL) ||	/* (3) */
		    (olladdr && lladdr != NULL && llchange)) {	/* (5) */
			do_update = 1;
			newstate = ND6_LLINFO_STALE;
		} else					/* (1-2,4) */
			do_update = 0;
	} else {
		do_update = 1;
		if (lladdr == NULL)			/* (6) */
			newstate = ND6_LLINFO_NOSTATE;
		else					/* (7) */
			newstate = ND6_LLINFO_STALE;
	}

	if (do_update) {
		/*
		 * Update the state of the neighbor cache.
		 */
		ln->ln_state = newstate;

		if (ln->ln_state == ND6_LLINFO_STALE) {
			/*
			 * XXX: since nd6_output() below will cause
			 * state tansition to DELAY and reset the timer,
			 * we must set the timer now, although it is actually
			 * meaningless.
			 */
			nd6_llinfo_settimer_locked(ln, (long)V_nd6_gctimer * hz);

			if (ln->la_hold) {
				struct mbuf *m_hold, *m_hold_next;

				/*
				 * reset the la_hold in advance, to explicitly
				 * prevent a la_hold lookup in nd6_output()
				 * (wouldn't happen, though...)
				 */
				for (m_hold = ln->la_hold, ln->la_hold = NULL;
				    m_hold; m_hold = m_hold_next) {
					m_hold_next = m_hold->m_nextpkt;
					m_hold->m_nextpkt = NULL;

					/*
					 * we assume ifp is not a p2p here, so
					 * just set the 2nd argument as the
					 * 1st one.
					 */
					nd6_output_lle(ifp, ifp, m_hold, L3_ADDR_SIN6(ln), NULL, ln, &chain);
				}
				/*
				 * If we have mbufs in the chain we need to do
				 * deferred transmit. Copy the address from the
				 * llentry before dropping the lock down below.
				 */
				if (chain != NULL)
					memcpy(&sin6, L3_ADDR_SIN6(ln), sizeof(sin6));
			}
		} else if (ln->ln_state == ND6_LLINFO_INCOMPLETE) {
			/* probe right away */
			nd6_llinfo_settimer_locked((void *)ln, 0);
		}
	}

	/*
	 * ICMP6 type dependent behavior.
	 *
	 * NS: clear IsRouter if new entry
	 * RS: clear IsRouter
	 * RA: set IsRouter if there's lladdr
	 * redir: clear IsRouter if new entry
	 *
	 * RA case, (1):
	 * The spec says that we must set IsRouter in the following cases:
	 * - If lladdr exist, set IsRouter.  This means (1-5).
	 * - If it is old entry (!newentry), set IsRouter.  This means (7).
	 * So, based on the spec, in (1-5) and (7) cases we must set IsRouter.
	 * A quetion arises for (1) case.  (1) case has no lladdr in the
	 * neighbor cache, this is similar to (6).
	 * This case is rare but we figured that we MUST NOT set IsRouter.
	 *
	 * newentry olladdr  lladdr  llchange	    NS  RS  RA	redir
	 *							D R
	 *	0	n	n	--	(1)	c   ?     s
	 *	0	y	n	--	(2)	c   s     s
	 *	0	n	y	--	(3)	c   s     s
	 *	0	y	y	n	(4)	c   s     s
	 *	0	y	y	y	(5)	c   s     s
	 *	1	--	n	--	(6) c	c	c s
	 *	1	--	y	--	(7) c	c   s	c s
	 *
	 *					(c=clear s=set)
	 */
	switch (type & 0xff) {
	case ND_NEIGHBOR_SOLICIT:
		/*
		 * New entry must have is_router flag cleared.
		 */
		if (is_newentry)	/* (6-7) */
			ln->ln_router = 0;
		break;
	case ND_REDIRECT:
		/*
		 * If the icmp is a redirect to a better router, always set the
		 * is_router flag.  Otherwise, if the entry is newly created,
		 * clear the flag.  [RFC 2461, sec 8.3]
		 */
		if (code == ND_REDIRECT_ROUTER)
			ln->ln_router = 1;
		else if (is_newentry) /* (6-7) */
			ln->ln_router = 0;
		break;
	case ND_ROUTER_SOLICIT:
		/*
		 * is_router flag must always be cleared.
		 */
		ln->ln_router = 0;
		break;
	case ND_ROUTER_ADVERT:
		/*
		 * Mark an entry with lladdr as a router.
		 */
		if ((!is_newentry && (olladdr || lladdr)) ||	/* (2-5) */
		    (is_newentry && lladdr)) {			/* (7) */
			ln->ln_router = 1;
		}
		break;
	}

	if (ln != NULL) {
		static_route = (ln->la_flags & LLE_STATIC);
		router = ln->ln_router;

		if (flags & ND6_EXCLUSIVE)
			LLE_WUNLOCK(ln);
		else
			LLE_RUNLOCK(ln);
		if (static_route)
			ln = NULL;
	}
	if (chain)
		nd6_output_flush(ifp, ifp, chain, &sin6, NULL);
	
	/*
	 * When the link-layer address of a router changes, select the
	 * best router again.  In particular, when the neighbor entry is newly
	 * created, it might affect the selection policy.
	 * Question: can we restrict the first condition to the "is_newentry"
	 * case?
	 * XXX: when we hear an RA from a new router with the link-layer
	 * address option, defrouter_select() is called twice, since
	 * defrtrlist_update called the function as well.  However, I believe
	 * we can compromise the overhead, since it only happens the first
	 * time.
	 * XXX: although defrouter_select() should not have a bad effect
	 * for those are not autoconfigured hosts, we explicitly avoid such
	 * cases for safety.
	 */
	if (do_update && router && !V_ip6_forwarding && V_ip6_accept_rtadv) {
		/*
		 * guaranteed recursion
		 */
		defrouter_select();
	}
	
	return (ln);
done:	
	if (ln != NULL) {
		if (flags & ND6_EXCLUSIVE)
			LLE_WUNLOCK(ln);
		else
			LLE_RUNLOCK(ln);
		if (static_route)
			ln = NULL;
	}
	return (ln);
}

static void
nd6_slowtimo(void *arg)
{
	CURVNET_SET((struct vnet *) arg);
	struct nd_ifinfo *nd6if;
	struct ifnet *ifp;

	callout_reset(&V_nd6_slowtimo_ch, ND6_SLOWTIMER_INTERVAL * hz,
	    nd6_slowtimo, curvnet);
	IFNET_RLOCK_NOSLEEP();
	TAILQ_FOREACH(ifp, &V_ifnet, if_list) {
		nd6if = ND_IFINFO(ifp);
		if (nd6if->basereachable && /* already initialized */
		    (nd6if->recalctm -= ND6_SLOWTIMER_INTERVAL) <= 0) {
			/*
			 * Since reachable time rarely changes by router
			 * advertisements, we SHOULD insure that a new random
			 * value gets recomputed at least once every few hours.
			 * (RFC 2461, 6.3.4)
			 */
			nd6if->recalctm = V_nd6_recalc_reachtm_interval;
			nd6if->reachable = ND_COMPUTE_RTIME(nd6if->basereachable);
		}
	}
	IFNET_RUNLOCK_NOSLEEP();
	CURVNET_RESTORE();
}

int
nd6_output(struct ifnet *ifp, struct ifnet *origifp, struct mbuf *m0,
    struct sockaddr_in6 *dst, struct rtentry *rt0)
{

	return (nd6_output_lle(ifp, origifp, m0, dst, rt0, NULL, NULL));
}


/*
 * Note that I'm not enforcing any global serialization
 * lle state or asked changes here as the logic is too
 * complicated to avoid having to always acquire an exclusive
 * lock
 * KMM
 *
 */
#define senderr(e) { error = (e); goto bad;}

int
nd6_output_lle(struct ifnet *ifp, struct ifnet *origifp, struct mbuf *m0,
    struct sockaddr_in6 *dst, struct rtentry *rt0, struct llentry *lle,
	struct mbuf **chain)
{
	struct mbuf *m = m0;
	struct llentry *ln = lle;
	int error = 0;
	int flags = 0;

#ifdef INVARIANTS
	if (lle != NULL) {
		
		LLE_WLOCK_ASSERT(lle);

		KASSERT(chain != NULL, (" lle locked but no mbuf chain pointer passed"));
	}
#endif
	if (IN6_IS_ADDR_MULTICAST(&dst->sin6_addr))
		goto sendpkt;

	if (nd6_need_cache(ifp) == 0)
		goto sendpkt;

	/*
	 * next hop determination.  This routine is derived from ether_output.
	 */

	/*
	 * Address resolution or Neighbor Unreachability Detection
	 * for the next hop.
	 * At this point, the destination of the packet must be a unicast
	 * or an anycast address(i.e. not a multicast).
	 */

	flags = ((m != NULL) || (lle != NULL)) ? LLE_EXCLUSIVE : 0;
	if (ln == NULL) {
	retry:
		IF_AFDATA_LOCK(ifp);
		ln = lla_lookup(LLTABLE6(ifp), flags, (struct sockaddr *)dst);
		IF_AFDATA_UNLOCK(ifp);
		if ((ln == NULL) && nd6_is_addr_neighbor(dst, ifp))  {
			/*
			 * Since nd6_is_addr_neighbor() internally calls nd6_lookup(),
			 * the condition below is not very efficient.  But we believe
			 * it is tolerable, because this should be a rare case.
			 */
			flags = ND6_CREATE | (m ? ND6_EXCLUSIVE : 0);
			IF_AFDATA_LOCK(ifp);
			ln = nd6_lookup(&dst->sin6_addr, flags, ifp);
			IF_AFDATA_UNLOCK(ifp);
		}
	} 
	if (ln == NULL) {
		if ((ifp->if_flags & IFF_POINTOPOINT) == 0 &&
		    !(ND_IFINFO(ifp)->flags & ND6_IFF_PERFORMNUD)) {
			char ip6buf[INET6_ADDRSTRLEN];
			log(LOG_DEBUG,
			    "nd6_output: can't allocate llinfo for %s "
			    "(ln=%p)\n",
			    ip6_sprintf(ip6buf, &dst->sin6_addr), ln);
			senderr(EIO);	/* XXX: good error? */
		}
		goto sendpkt;	/* send anyway */
	}

	/* We don't have to do link-layer address resolution on a p2p link. */
	if ((ifp->if_flags & IFF_POINTOPOINT) != 0 &&
	    ln->ln_state < ND6_LLINFO_REACHABLE) {
		if ((flags & LLE_EXCLUSIVE) == 0) {
			flags |= LLE_EXCLUSIVE;
			goto retry;
		}
		ln->ln_state = ND6_LLINFO_STALE;
		nd6_llinfo_settimer_locked(ln, (long)V_nd6_gctimer * hz);
	}

	/*
	 * The first time we send a packet to a neighbor whose entry is
	 * STALE, we have to change the state to DELAY and a sets a timer to
	 * expire in DELAY_FIRST_PROBE_TIME seconds to ensure do
	 * neighbor unreachability detection on expiration.
	 * (RFC 2461 7.3.3)
	 */
	if (ln->ln_state == ND6_LLINFO_STALE) {
		if ((flags & LLE_EXCLUSIVE) == 0) {
			flags |= LLE_EXCLUSIVE;
			LLE_RUNLOCK(ln);
			goto retry;
		}
		ln->la_asked = 0;
		ln->ln_state = ND6_LLINFO_DELAY;
		nd6_llinfo_settimer_locked(ln, (long)V_nd6_delay * hz);
	}

	/*
	 * If the neighbor cache entry has a state other than INCOMPLETE
	 * (i.e. its link-layer address is already resolved), just
	 * send the packet.
	 */
	if (ln->ln_state > ND6_LLINFO_INCOMPLETE)
		goto sendpkt;

	/*
	 * There is a neighbor cache entry, but no ethernet address
	 * response yet.  Append this latest packet to the end of the
	 * packet queue in the mbuf, unless the number of the packet
	 * does not exceed nd6_maxqueuelen.  When it exceeds nd6_maxqueuelen,
	 * the oldest packet in the queue will be removed.
	 */
	if (ln->ln_state == ND6_LLINFO_NOSTATE)
		ln->ln_state = ND6_LLINFO_INCOMPLETE;

	if ((flags & LLE_EXCLUSIVE) == 0) {
		flags |= LLE_EXCLUSIVE;
		LLE_RUNLOCK(ln);
		goto retry;
	}

	LLE_WLOCK_ASSERT(ln);

	if (ln->la_hold) {
		struct mbuf *m_hold;
		int i;
		
		i = 0;
		for (m_hold = ln->la_hold; m_hold; m_hold = m_hold->m_nextpkt) {
			i++;
			if (m_hold->m_nextpkt == NULL) {
				m_hold->m_nextpkt = m;
				break;
			}
		}
		while (i >= V_nd6_maxqueuelen) {
			m_hold = ln->la_hold;
			ln->la_hold = ln->la_hold->m_nextpkt;
			m_freem(m_hold);
			i--;
		}
	} else {
		ln->la_hold = m;
	}

	/*
	 * If there has been no NS for the neighbor after entering the
	 * INCOMPLETE state, send the first solicitation.
	 */
	if (!ND6_LLINFO_PERMANENT(ln) && ln->la_asked == 0) {
		ln->la_asked++;
		
		nd6_llinfo_settimer_locked(ln,
		    (long)ND_IFINFO(ifp)->retrans * hz / 1000);
		LLE_WUNLOCK(ln);
		nd6_ns_output(ifp, NULL, &dst->sin6_addr, ln, 0);
		if (lle != NULL && ln == lle)
			LLE_WLOCK(lle);

	} else if (lle == NULL || ln != lle) {
		/*
		 * We did the lookup (no lle arg) so we
		 * need to do the unlock here.
		 */
		LLE_WUNLOCK(ln);
	}

	return (0);

  sendpkt:
	/* discard the packet if IPv6 operation is disabled on the interface */
	if ((ND_IFINFO(ifp)->flags & ND6_IFF_IFDISABLED)) {
		error = ENETDOWN; /* better error? */
		goto bad;
	}
	/*
	 * ln is valid and the caller did not pass in 
	 * an llentry
	 */
	if ((ln != NULL) && (lle == NULL)) {
		if (flags & LLE_EXCLUSIVE)
			LLE_WUNLOCK(ln);
		else
			LLE_RUNLOCK(ln);
	}

#ifdef MAC
	mac_netinet6_nd6_send(ifp, m);
#endif
	/*
	 * We were passed in a pointer to an lle with the lock held 
	 * this means that we can't call if_output as we will
	 * recurse on the lle lock - so what we do is we create
	 * a list of mbufs to send and transmit them in the caller
	 * after the lock is dropped
	 */
	if (lle != NULL) {
		if (*chain == NULL)
			*chain = m;
		else {
			struct mbuf *mb;

			/*
			 * append mbuf to end of deferred chain
			 */
			mb = *chain;
			while (mb->m_nextpkt != NULL)
				mb = mb->m_nextpkt;
			mb->m_nextpkt = m;
		}
		return (error);
	}
	if ((ifp->if_flags & IFF_LOOPBACK) != 0) {
		return ((*ifp->if_output)(origifp, m, (struct sockaddr *)dst,
		    NULL));
	}
	error = (*ifp->if_output)(ifp, m, (struct sockaddr *)dst, NULL);
	return (error);

  bad:
	/*
	 * ln is valid and the caller did not pass in 
	 * an llentry
	 */
	if ((ln != NULL) && (lle == NULL)) {
		if (flags & LLE_EXCLUSIVE)
			LLE_WUNLOCK(ln);
		else
			LLE_RUNLOCK(ln);
	}
	if (m)
		m_freem(m);
	return (error);
}
#undef senderr


int
nd6_output_flush(struct ifnet *ifp, struct ifnet *origifp, struct mbuf *chain,
    struct sockaddr_in6 *dst, struct route *ro)
{
	struct mbuf *m, *m_head;
	struct ifnet *outifp;
	int error = 0;

	m_head = chain;
	if ((ifp->if_flags & IFF_LOOPBACK) != 0)
		outifp = origifp;
	else
		outifp = ifp;
	
	while (m_head) {
		m = m_head;
		m_head = m_head->m_nextpkt;
		error = (*ifp->if_output)(ifp, m, (struct sockaddr *)dst, ro);			       
	}

	/*
	 * XXX
	 * note that intermediate errors are blindly ignored - but this is 
	 * the same convention as used with nd6_output when called by
	 * nd6_cache_lladdr
	 */
	return (error);
}	


int
nd6_need_cache(struct ifnet *ifp)
{
	/*
	 * XXX: we currently do not make neighbor cache on any interface
	 * other than ARCnet, Ethernet, FDDI and GIF.
	 *
	 * RFC2893 says:
	 * - unidirectional tunnels needs no ND
	 */
	switch (ifp->if_type) {
	case IFT_ARCNET:
	case IFT_ETHER:
	case IFT_FDDI:
	case IFT_IEEE1394:
#ifdef IFT_L2VLAN
	case IFT_L2VLAN:
#endif
#ifdef IFT_IEEE80211
	case IFT_IEEE80211:
#endif
#ifdef IFT_CARP
	case IFT_CARP:
#endif
	case IFT_GIF:		/* XXX need more cases? */
	case IFT_PPP:
	case IFT_TUNNEL:
	case IFT_BRIDGE:
	case IFT_PROPVIRTUAL:
		return (1);
	default:
		return (0);
	}
}

/*
 * the callers of this function need to be re-worked to drop
 * the lle lock, drop here for now
 */
int
nd6_storelladdr(struct ifnet *ifp, struct mbuf *m,
    struct sockaddr *dst, u_char *desten, struct llentry **lle)
{
	struct llentry *ln;

	*lle = NULL;
	IF_AFDATA_UNLOCK_ASSERT(ifp);
	if (m->m_flags & M_MCAST) {
		int i;

		switch (ifp->if_type) {
		case IFT_ETHER:
		case IFT_FDDI:
#ifdef IFT_L2VLAN
		case IFT_L2VLAN:
#endif
#ifdef IFT_IEEE80211
		case IFT_IEEE80211:
#endif
		case IFT_BRIDGE:
		case IFT_ISO88025:
			ETHER_MAP_IPV6_MULTICAST(&SIN6(dst)->sin6_addr,
						 desten);
			return (0);
		case IFT_IEEE1394:
			/*
			 * netbsd can use if_broadcastaddr, but we don't do so
			 * to reduce # of ifdef.
			 */
			for (i = 0; i < ifp->if_addrlen; i++)
				desten[i] = ~0;
			return (0);
		case IFT_ARCNET:
			*desten = 0;
			return (0);
		default:
			m_freem(m);
			return (EAFNOSUPPORT);
		}
	}


	/*
	 * the entry should have been created in nd6_store_lladdr
	 */
	IF_AFDATA_RLOCK(ifp);
	ln = lla_lookup(LLTABLE6(ifp), 0, dst);
	IF_AFDATA_RUNLOCK(ifp);
	if ((ln == NULL) || !(ln->la_flags & LLE_VALID)) {
		if (ln != NULL)
			LLE_RUNLOCK(ln);
		/* this could happen, if we could not allocate memory */
		m_freem(m);
		return (1);
	}

	bcopy(&ln->ll_addr, desten, ifp->if_addrlen);
	*lle = ln;
	LLE_RUNLOCK(ln);
	/*
	 * A *small* use after free race exists here
	 */
	return (0);
}

static void 
clear_llinfo_pqueue(struct llentry *ln)
{
	struct mbuf *m_hold, *m_hold_next;

	for (m_hold = ln->la_hold; m_hold; m_hold = m_hold_next) {
		m_hold_next = m_hold->m_nextpkt;
		m_hold->m_nextpkt = NULL;
		m_freem(m_hold);
	}

	ln->la_hold = NULL;
	return;
}

static int nd6_sysctl_drlist(SYSCTL_HANDLER_ARGS);
static int nd6_sysctl_prlist(SYSCTL_HANDLER_ARGS);
#ifdef SYSCTL_DECL
SYSCTL_DECL(_net_inet6_icmp6);
#endif
SYSCTL_NODE(_net_inet6_icmp6, ICMPV6CTL_ND6_DRLIST, nd6_drlist,
	CTLFLAG_RD, nd6_sysctl_drlist, "");
SYSCTL_NODE(_net_inet6_icmp6, ICMPV6CTL_ND6_PRLIST, nd6_prlist,
	CTLFLAG_RD, nd6_sysctl_prlist, "");
SYSCTL_VNET_INT(_net_inet6_icmp6, ICMPV6CTL_ND6_MAXQLEN, nd6_maxqueuelen,
	CTLFLAG_RW, &VNET_NAME(nd6_maxqueuelen), 1, "");

static int
nd6_sysctl_drlist(SYSCTL_HANDLER_ARGS)
{
	struct in6_defrouter d;
	struct nd_defrouter *dr;
	int error;

	if (req->newptr)
		return (EPERM);

	bzero(&d, sizeof(d));
	d.rtaddr.sin6_family = AF_INET6;
	d.rtaddr.sin6_len = sizeof(d.rtaddr);

	/*
	 * XXX locking
	 */
	TAILQ_FOREACH(dr, &V_nd_defrouter, dr_entry) {
		d.rtaddr.sin6_addr = dr->rtaddr;
		error = sa6_recoverscope(&d.rtaddr);
		if (error != 0)
			return (error);
		d.flags = dr->flags;
		d.rtlifetime = dr->rtlifetime;
		d.expire = dr->expire;
		d.if_index = dr->ifp->if_index;
		error = SYSCTL_OUT(req, &d, sizeof(d));
		if (error != 0)
			return (error);
	}
	return (0);
}

static int
nd6_sysctl_prlist(SYSCTL_HANDLER_ARGS)
{
	struct in6_prefix p;
	struct sockaddr_in6 s6;
	struct nd_prefix *pr;
	struct nd_pfxrouter *pfr;
	time_t maxexpire;
	int error;
	char ip6buf[INET6_ADDRSTRLEN];

	if (req->newptr)
		return (EPERM);

	bzero(&p, sizeof(p));
	p.origin = PR_ORIG_RA;
	bzero(&s6, sizeof(s6));
	s6.sin6_family = AF_INET6;
	s6.sin6_len = sizeof(s6);

	/*
	 * XXX locking
	 */
	LIST_FOREACH(pr, &V_nd_prefix, ndpr_entry) {
		p.prefix = pr->ndpr_prefix;
		if (sa6_recoverscope(&p.prefix)) {
			log(LOG_ERR, "scope error in prefix list (%s)\n",
			    ip6_sprintf(ip6buf, &p.prefix.sin6_addr));
			/* XXX: press on... */
		}
		p.raflags = pr->ndpr_raf;
		p.prefixlen = pr->ndpr_plen;
		p.vltime = pr->ndpr_vltime;
		p.pltime = pr->ndpr_pltime;
		p.if_index = pr->ndpr_ifp->if_index;
		if (pr->ndpr_vltime == ND6_INFINITE_LIFETIME)
			p.expire = 0;
		else {
			/* XXX: we assume time_t is signed. */
			maxexpire = (-1) &
			    ~((time_t)1 << ((sizeof(maxexpire) * 8) - 1));
			if (pr->ndpr_vltime < maxexpire - pr->ndpr_lastupdate)
				p.expire = pr->ndpr_lastupdate +
				    pr->ndpr_vltime;
			else
				p.expire = maxexpire;
		}
		p.refcnt = pr->ndpr_refcnt;
		p.flags = pr->ndpr_stateflags;
		p.advrtrs = 0;
		LIST_FOREACH(pfr, &pr->ndpr_advrtrs, pfr_entry)
			p.advrtrs++;
		error = SYSCTL_OUT(req, &p, sizeof(p));
		if (error != 0)
			return (error);
		LIST_FOREACH(pfr, &pr->ndpr_advrtrs, pfr_entry) {
			s6.sin6_addr = pfr->router->rtaddr;
			if (sa6_recoverscope(&s6))
				log(LOG_ERR,
				    "scope error in prefix list (%s)\n",
				    ip6_sprintf(ip6buf, &pfr->router->rtaddr));
			error = SYSCTL_OUT(req, &s6, sizeof(s6));
			if (error != 0)
				return (error);
		}
	}
	return (0);
}