summaryrefslogtreecommitdiff
path: root/sys/netinet/if_arp.c
blob: 8f1cfe4fefd080807812c74681bbc191f14ef3d1 (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
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
/*	$NetBSD: if_arp.c,v 1.155 2014/02/25 18:30:12 pooka Exp $	*/

/*-
 * Copyright (c) 1998, 2000, 2008 The NetBSD Foundation, Inc.
 * All rights reserved.
 *
 * This code is derived from software contributed to The NetBSD Foundation
 * by Public Access Networks Corporation ("Panix").  It was developed under
 * contract to Panix by Eric Haszlakiewicz and Thor Lancelot Simon.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 *
 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. AND CONTRIBUTORS
 * ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED
 * TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
 * PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE FOUNDATION OR CONTRIBUTORS
 * BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
 * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
 * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
 * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
 * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
 * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
 * POSSIBILITY OF SUCH DAMAGE.
 */

/*
 * Copyright (c) 1982, 1986, 1988, 1993
 *	The Regents of the University of California.  All rights reserved.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)if_ether.c	8.2 (Berkeley) 9/26/94
 */

/*
 * Ethernet address resolution protocol.
 * TODO:
 *	add "inuse/lock" bit (or ref. count) along with valid bit
 */

#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: if_arp.c,v 1.155 2014/02/25 18:30:12 pooka Exp $");

#include "opt_ddb.h"
#include "opt_inet.h"

#ifdef INET

#include "bridge.h"

#include <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/time.h>
#include <sys/timetc.h>
#include <sys/kernel.h>
#include <sys/errno.h>
#include <sys/ioctl.h>
#include <sys/syslog.h>
#include <sys/proc.h>
#include <sys/protosw.h>
#include <sys/domain.h>
#include <sys/sysctl.h>
#include <sys/socketvar.h>
#include <sys/percpu.h>

#include <net/ethertypes.h>
#include <net/if.h>
#include <net/if_dl.h>
#include <net/if_token.h>
#include <net/if_types.h>
#include <net/if_ether.h>
#include <net/route.h>
#include <net/net_stats.h>

#include <netinet/in.h>
#include <netinet/in_systm.h>
#include <netinet/in_var.h>
#include <netinet/ip.h>
#include <netinet/if_inarp.h>

#include "arcnet.h"
#if NARCNET > 0
#include <net/if_arc.h>
#endif
#include "fddi.h"
#if NFDDI > 0
#include <net/if_fddi.h>
#endif
#include "token.h"
#include "carp.h"
#if NCARP > 0
#include <netinet/ip_carp.h>
#endif

#define SIN(s) ((struct sockaddr_in *)s)
#define SRP(s) ((struct sockaddr_inarp *)s)

/*
 * ARP trailer negotiation.  Trailer protocol is not IP specific,
 * but ARP request/response use IP addresses.
 */
#define ETHERTYPE_IPTRAILERS ETHERTYPE_TRAIL

/* timer values */
int	arpt_prune = (5*60*1);	/* walk list every 5 minutes */
int	arpt_keep = (20*60);	/* once resolved, good for 20 more minutes */
int	arpt_down = 20;		/* once declared down, don't send for 20 secs */
int	arpt_refresh = (5*60);	/* time left before refreshing */
#define	rt_expire rt_rmx.rmx_expire
#define	rt_pksent rt_rmx.rmx_pksent

static	struct sockaddr *arp_setgate(struct rtentry *, struct sockaddr *,
	    const struct sockaddr *);
static	void arptfree(struct llinfo_arp *);
static	void arptimer(void *);
static	struct llinfo_arp *arplookup1(struct mbuf *, const struct in_addr *,
				      int, int, struct rtentry *);
static	struct llinfo_arp *arplookup(struct mbuf *, const struct in_addr *,
					  int, int);
static	void in_arpinput(struct mbuf *);
static	void arp_drainstub(void);

LIST_HEAD(, llinfo_arp) llinfo_arp;
struct	ifqueue arpintrq = {
	.ifq_head = NULL,
	.ifq_tail = NULL,
	.ifq_len = 0,
	.ifq_maxlen = 50,
	.ifq_drops = 0,
};
int	arp_inuse, arp_allocated;
int	arp_maxtries = 5;
int	useloopback = 1;	/* use loopback interface for local traffic */
int	arpinit_done = 0;

static percpu_t *arpstat_percpu;

#define	ARP_STAT_GETREF()	_NET_STAT_GETREF(arpstat_percpu)
#define	ARP_STAT_PUTREF()	_NET_STAT_PUTREF(arpstat_percpu)

#define	ARP_STATINC(x)		_NET_STATINC(arpstat_percpu, x)
#define	ARP_STATADD(x, v)	_NET_STATADD(arpstat_percpu, x, v)

struct	callout arptimer_ch;

/* revarp state */
struct	in_addr myip, srv_ip;
int	myip_initialized = 0;
int	revarp_in_progress = 0;
struct	ifnet *myip_ifp = NULL;

#ifdef DDB
static void db_print_sa(const struct sockaddr *);
static void db_print_ifa(struct ifaddr *);
static void db_print_llinfo(void *);
static int db_show_rtentry(struct rtentry *, void *);
#endif

static int arp_drainwanted;

static int log_movements = 1;
static int log_permanent_modify = 1;
static int log_wrong_iface = 1;

/*
 * this should be elsewhere.
 */

static char *
lla_snprintf(u_int8_t *, int);

static char *
lla_snprintf(u_int8_t *adrp, int len)
{
#define NUMBUFS 3
	static char buf[NUMBUFS][16*3];
	static int bnum = 0;

	int i;
	char *p;

	p = buf[bnum];

	*p++ = hexdigits[(*adrp)>>4];
	*p++ = hexdigits[(*adrp++)&0xf];

	for (i=1; i<len && i<16; i++) {
		*p++ = ':';
		*p++ = hexdigits[(*adrp)>>4];
		*p++ = hexdigits[(*adrp++)&0xf];
	}

	*p = 0;
	p = buf[bnum];
	bnum = (bnum + 1) % NUMBUFS;
	return p;
}

DOMAIN_DEFINE(arpdomain);	/* forward declare and add to link set */

static void
arp_fasttimo(void)
{
	if (arp_drainwanted) {
		arp_drain();
		arp_drainwanted = 0;
	}
}

const struct protosw arpsw[] = {
	{ .pr_type = 0,
	  .pr_domain = &arpdomain,
	  .pr_protocol = 0,
	  .pr_flags = 0,
	  .pr_input = 0,
	  .pr_output = 0,
	  .pr_ctlinput = 0,
	  .pr_ctloutput = 0,
	  .pr_usrreq =  0,
	  .pr_init = arp_init,
	  .pr_fasttimo = arp_fasttimo,
	  .pr_slowtimo = 0,
	  .pr_drain = arp_drainstub,
	}
};


struct domain arpdomain = {
	.dom_family = PF_ARP,
	.dom_name = "arp",
	.dom_protosw = arpsw,
	.dom_protoswNPROTOSW = &arpsw[__arraycount(arpsw)],
};

/*
 * ARP table locking.
 *
 * to prevent lossage vs. the arp_drain routine (which may be called at
 * any time, including in a device driver context), we do two things:
 *
 * 1) manipulation of la->la_hold is done at splnet() (for all of
 * about two instructions).
 *
 * 2) manipulation of the arp table's linked list is done under the
 * protection of the ARP_LOCK; if arp_drain() or arptimer is called
 * while the arp table is locked, we punt and try again later.
 */

static int	arp_locked;
static inline int arp_lock_try(int);
static inline void arp_unlock(void);

static inline int
arp_lock_try(int recurse)
{
	int s;

	/*
	 * Use splvm() -- we're blocking things that would cause
	 * mbuf allocation.
	 */
	s = splvm();
	if (!recurse && arp_locked) {
		splx(s);
		return 0;
	}
	arp_locked++;
	splx(s);
	return 1;
}

static inline void
arp_unlock(void)
{
	int s;

	s = splvm();
	arp_locked--;
	splx(s);
}

#ifdef DIAGNOSTIC
#define	ARP_LOCK(recurse)						\
do {									\
	if (arp_lock_try(recurse) == 0) {				\
		printf("%s:%d: arp already locked\n", __FILE__, __LINE__); \
		panic("arp_lock");					\
	}								\
} while (/*CONSTCOND*/ 0)
#define	ARP_LOCK_CHECK()						\
do {									\
	if (arp_locked == 0) {						\
		printf("%s:%d: arp lock not held\n", __FILE__, __LINE__); \
		panic("arp lock check");				\
	}								\
} while (/*CONSTCOND*/ 0)
#else
#define	ARP_LOCK(x)		(void) arp_lock_try(x)
#define	ARP_LOCK_CHECK()	/* nothing */
#endif

#define	ARP_UNLOCK()		arp_unlock()

static void sysctl_net_inet_arp_setup(struct sysctllog **);

void
arp_init(void)
{

	sysctl_net_inet_arp_setup(NULL);
	arpstat_percpu = percpu_alloc(sizeof(uint64_t) * ARP_NSTATS);
}

static void
arp_drainstub(void)
{
	arp_drainwanted = 1;
}

/*
 * ARP protocol drain routine.  Called when memory is in short supply.
 * Called at splvm();  don't acquire softnet_lock as can be called from
 * hardware interrupt handlers.
 */
void
arp_drain(void)
{
	struct llinfo_arp *la, *nla;
	int count = 0;
	struct mbuf *mold;

	KERNEL_LOCK(1, NULL);

	if (arp_lock_try(0) == 0) {
		KERNEL_UNLOCK_ONE(NULL);
		return;
	}

	for (la = LIST_FIRST(&llinfo_arp); la != NULL; la = nla) {
		nla = LIST_NEXT(la, la_list);

		mold = la->la_hold;
		la->la_hold = 0;

		if (mold) {
			m_freem(mold);
			count++;
		}
	}
	ARP_UNLOCK();
	ARP_STATADD(ARP_STAT_DFRDROPPED, count);
	KERNEL_UNLOCK_ONE(NULL);
}


/*
 * Timeout routine.  Age arp_tab entries periodically.
 */
/* ARGSUSED */
static void
arptimer(void *arg)
{
	struct llinfo_arp *la, *nla;

	mutex_enter(softnet_lock);
	KERNEL_LOCK(1, NULL);

	if (arp_lock_try(0) == 0) {
		/* get it later.. */
		KERNEL_UNLOCK_ONE(NULL);
		mutex_exit(softnet_lock);
		return;
	}

	callout_reset(&arptimer_ch, arpt_prune * hz, arptimer, NULL);
	for (la = LIST_FIRST(&llinfo_arp); la != NULL; la = nla) {
		struct rtentry *rt = la->la_rt;

		nla = LIST_NEXT(la, la_list);
		if (rt->rt_expire == 0)
			continue;
		if ((rt->rt_expire - time_second) < arpt_refresh &&
		    rt->rt_pksent > (time_second - arpt_keep)) {
			/*
			 * If the entry has been used during since last
			 * refresh, try to renew it before deleting.
			 */
			arprequest(rt->rt_ifp,
			    &satocsin(rt->rt_ifa->ifa_addr)->sin_addr,
			    &satocsin(rt_getkey(rt))->sin_addr,
			    CLLADDR(rt->rt_ifp->if_sadl));
		} else if (rt->rt_expire <= time_second)
			arptfree(la); /* timer has expired; clear */
	}

	ARP_UNLOCK();

	KERNEL_UNLOCK_ONE(NULL);
	mutex_exit(softnet_lock);
}

/*
 * We set the gateway for RTF_CLONING routes to a "prototype"
 * link-layer sockaddr whose interface type (if_type) and interface
 * index (if_index) fields are prepared.
 */
static struct sockaddr *
arp_setgate(struct rtentry *rt, struct sockaddr *gate,
    const struct sockaddr *netmask)
{
	const struct ifnet *ifp = rt->rt_ifp;
	uint8_t namelen = strlen(ifp->if_xname);
	uint8_t addrlen = ifp->if_addrlen;

	/*
	 * XXX: If this is a manually added route to interface
	 * such as older version of routed or gated might provide,
	 * restore cloning bit.
	 */
	if ((rt->rt_flags & RTF_HOST) == 0 && netmask != NULL &&
	    satocsin(netmask)->sin_addr.s_addr != 0xffffffff)
		rt->rt_flags |= RTF_CLONING;
	if (rt->rt_flags & RTF_CLONING) {
		union {
			struct sockaddr sa;
			struct sockaddr_storage ss;
			struct sockaddr_dl sdl;
		} u;
		/*
		 * Case 1: This route should come from a route to iface.
		 */
		sockaddr_dl_init(&u.sdl, sizeof(u.ss),
		    ifp->if_index, ifp->if_type, NULL, namelen, NULL, addrlen);
		rt_setgate(rt, &u.sa);
		gate = rt->rt_gateway;
	}
	return gate;
}

/*
 * Parallel to llc_rtrequest.
 */
void
arp_rtrequest(int req, struct rtentry *rt, const struct rt_addrinfo *info)
{
	struct sockaddr *gate = rt->rt_gateway;
	struct llinfo_arp *la = (struct llinfo_arp *)rt->rt_llinfo;
	size_t allocsize;
	struct mbuf *mold;
	int s;
	struct in_ifaddr *ia;
	struct ifaddr *ifa;
	struct ifnet *ifp = rt->rt_ifp;

	if (!arpinit_done) {
		arpinit_done = 1;
		/*
		 * We generate expiration times from time_second
		 * so avoid accidentally creating permanent routes.
		 */
		if (time_second == 0) {
			struct timespec ts;
			ts.tv_sec = 1;
			ts.tv_nsec = 0;
			tc_setclock(&ts);
		}
		callout_init(&arptimer_ch, CALLOUT_MPSAFE);
		callout_reset(&arptimer_ch, hz, arptimer, NULL);
	}

	if (req == RTM_LLINFO_UPD) {
		struct in_addr *in;

		if ((ifa = info->rti_ifa) == NULL)
			return;

		in = &ifatoia(ifa)->ia_addr.sin_addr;

		arprequest(ifa->ifa_ifp, in, in,
		    CLLADDR(ifa->ifa_ifp->if_sadl));
		return;
	}

	if ((rt->rt_flags & RTF_GATEWAY) != 0) {
		if (req != RTM_ADD)
			return;

		/*
		 * linklayers with particular link MTU limitation.
		 */
		switch(ifp->if_type) {
#if NFDDI > 0
		case IFT_FDDI:
			if (ifp->if_mtu > FDDIIPMTU)
				rt->rt_rmx.rmx_mtu = FDDIIPMTU;
			break;
#endif
#if NARC > 0
		case IFT_ARCNET:
		    {
			int arcipifmtu;

			if (ifp->if_flags & IFF_LINK0)
				arcipifmtu = arc_ipmtu;
			else
				arcipifmtu = ARCMTU;
			if (ifp->if_mtu > arcipifmtu)
				rt->rt_rmx.rmx_mtu = arcipifmtu;
			break;
		    }
#endif
		}
		return;
	}

	ARP_LOCK(1);		/* we may already be locked here. */

	switch (req) {
	case RTM_SETGATE:
		gate = arp_setgate(rt, gate, info->rti_info[RTAX_NETMASK]);
		break;
	case RTM_ADD:
		gate = arp_setgate(rt, gate, info->rti_info[RTAX_NETMASK]);
		if (rt->rt_flags & RTF_CLONING) {
			/*
			 * Give this route an expiration time, even though
			 * it's a "permanent" route, so that routes cloned
			 * from it do not need their expiration time set.
			 */
			rt->rt_expire = time_second;
			/*
			 * linklayers with particular link MTU limitation.
			 */
			switch (ifp->if_type) {
#if NFDDI > 0
			case IFT_FDDI:
				if ((rt->rt_rmx.rmx_locks & RTV_MTU) == 0 &&
				    (rt->rt_rmx.rmx_mtu > FDDIIPMTU ||
				     (rt->rt_rmx.rmx_mtu == 0 &&
				      ifp->if_mtu > FDDIIPMTU)))
					rt->rt_rmx.rmx_mtu = FDDIIPMTU;
				break;
#endif
#if NARC > 0
			case IFT_ARCNET:
			    {
				int arcipifmtu;
				if (ifp->if_flags & IFF_LINK0)
					arcipifmtu = arc_ipmtu;
				else
					arcipifmtu = ARCMTU;

				if ((rt->rt_rmx.rmx_locks & RTV_MTU) == 0 &&
				    (rt->rt_rmx.rmx_mtu > arcipifmtu ||
				     (rt->rt_rmx.rmx_mtu == 0 &&
				      ifp->if_mtu > arcipifmtu)))
					rt->rt_rmx.rmx_mtu = arcipifmtu;
				break;
			    }
#endif
			}
			break;
		}
		/* Announce a new entry if requested. */
		if (rt->rt_flags & RTF_ANNOUNCE) {
			arprequest(ifp,
			    &satocsin(rt_getkey(rt))->sin_addr,
			    &satocsin(rt_getkey(rt))->sin_addr,
			    CLLADDR(satocsdl(gate)));
		}
		/*FALLTHROUGH*/
	case RTM_RESOLVE:
		if (gate->sa_family != AF_LINK ||
		    gate->sa_len < sockaddr_dl_measure(0, ifp->if_addrlen)) {
			log(LOG_DEBUG, "arp_rtrequest: bad gateway value\n");
			break;
		}
		satosdl(gate)->sdl_type = ifp->if_type;
		satosdl(gate)->sdl_index = ifp->if_index;
		if (la != NULL)
			break; /* This happens on a route change */
		/*
		 * Case 2:  This route may come from cloning, or a manual route
		 * add with a LL address.
		 */
		switch (ifp->if_type) {
#if NTOKEN > 0
		case IFT_ISO88025:
			allocsize = sizeof(*la) + sizeof(struct token_rif);
			break;
#endif /* NTOKEN > 0 */
		default:
			allocsize = sizeof(*la);
		}
		R_Malloc(la, struct llinfo_arp *, allocsize);
		rt->rt_llinfo = (void *)la;
		if (la == NULL) {
			log(LOG_DEBUG, "arp_rtrequest: malloc failed\n");
			break;
		}
		arp_inuse++, arp_allocated++;
		memset(la, 0, allocsize);
		la->la_rt = rt;
		rt->rt_flags |= RTF_LLINFO;
		LIST_INSERT_HEAD(&llinfo_arp, la, la_list);

		INADDR_TO_IA(satocsin(rt_getkey(rt))->sin_addr, ia);
		while (ia && ia->ia_ifp != ifp)
			NEXT_IA_WITH_SAME_ADDR(ia);
		if (ia) {
			/*
			 * This test used to be
			 *	if (lo0ifp->if_flags & IFF_UP)
			 * It allowed local traffic to be forced through
			 * the hardware by configuring the loopback down.
			 * However, it causes problems during network
			 * configuration for boards that can't receive
			 * packets they send.  It is now necessary to clear
			 * "useloopback" and remove the route to force
			 * traffic out to the hardware.
			 *
			 * In 4.4BSD, the above "if" statement checked
			 * rt->rt_ifa against rt_getkey(rt).  It was changed
			 * to the current form so that we can provide a
			 * better support for multiple IPv4 addresses on a
			 * interface.
			 */
			rt->rt_expire = 0;
			if (sockaddr_dl_init(satosdl(gate), gate->sa_len,
			    ifp->if_index, ifp->if_type, NULL, 0,
			    CLLADDR(ifp->if_sadl), ifp->if_addrlen) == NULL) {
				panic("%s(%s): sockaddr_dl_init cannot fail",
				    __func__, ifp->if_xname);
			}
			if (useloopback)
				ifp = rt->rt_ifp = lo0ifp;
			/*
			 * make sure to set rt->rt_ifa to the interface
			 * address we are using, otherwise we will have trouble
			 * with source address selection.
			 */
			ifa = &ia->ia_ifa;
			if (ifa != rt->rt_ifa)
				rt_replace_ifa(rt, ifa);
		}
		break;

	case RTM_DELETE:
		if (la == NULL)
			break;
		arp_inuse--;
		LIST_REMOVE(la, la_list);
		rt->rt_llinfo = NULL;
		rt->rt_flags &= ~RTF_LLINFO;

		s = splnet();
		mold = la->la_hold;
		la->la_hold = 0;
		splx(s);

		if (mold)
			m_freem(mold);

		Free((void *)la);
	}
	ARP_UNLOCK();
}

/*
 * Broadcast an ARP request. Caller specifies:
 *	- arp header source ip address
 *	- arp header target ip address
 *	- arp header source ethernet address
 */
void
arprequest(struct ifnet *ifp,
    const struct in_addr *sip, const struct in_addr *tip,
    const u_int8_t *enaddr)
{
	struct mbuf *m;
	struct arphdr *ah;
	struct sockaddr sa;
	uint64_t *arps;

	if ((m = m_gethdr(M_DONTWAIT, MT_DATA)) == NULL)
		return;
	MCLAIM(m, &arpdomain.dom_mowner);
	switch (ifp->if_type) {
	case IFT_IEEE1394:
		m->m_len = sizeof(*ah) + 2 * sizeof(struct in_addr) +
		    ifp->if_addrlen;
		break;
	default:
		m->m_len = sizeof(*ah) + 2 * sizeof(struct in_addr) +
		    2 * ifp->if_addrlen;
		break;
	}
	m->m_pkthdr.len = m->m_len;
	MH_ALIGN(m, m->m_len);
	ah = mtod(m, struct arphdr *);
	memset(ah, 0, m->m_len);
	switch (ifp->if_type) {
	case IFT_IEEE1394:	/* RFC2734 */
		/* fill it now for ar_tpa computation */
		ah->ar_hrd = htons(ARPHRD_IEEE1394);
		break;
	default:
		/* ifp->if_output will fill ar_hrd */
		break;
	}
	ah->ar_pro = htons(ETHERTYPE_IP);
	ah->ar_hln = ifp->if_addrlen;		/* hardware address length */
	ah->ar_pln = sizeof(struct in_addr);	/* protocol address length */
	ah->ar_op = htons(ARPOP_REQUEST);
	memcpy(ar_sha(ah), enaddr, ah->ar_hln);
	memcpy(ar_spa(ah), sip, ah->ar_pln);
	memcpy(ar_tpa(ah), tip, ah->ar_pln);
	sa.sa_family = AF_ARP;
	sa.sa_len = 2;
	m->m_flags |= M_BCAST;
	arps = ARP_STAT_GETREF();
	arps[ARP_STAT_SNDTOTAL]++;
	arps[ARP_STAT_SENDREQUEST]++;
	ARP_STAT_PUTREF();
	(*ifp->if_output)(ifp, m, &sa, NULL);
}

/*
 * Resolve an IP address into an ethernet address.  If success,
 * desten is filled in.  If there is no entry in arptab,
 * set one up and broadcast a request for the IP address.
 * Hold onto this mbuf and resend it once the address
 * is finally resolved.  A return value of 1 indicates
 * that desten has been filled in and the packet should be sent
 * normally; a 0 return indicates that the packet has been
 * taken over here, either now or for later transmission.
 */
int
arpresolve(struct ifnet *ifp, struct rtentry *rt, struct mbuf *m,
    const struct sockaddr *dst, u_char *desten)
{
	struct llinfo_arp *la;
	const struct sockaddr_dl *sdl;
	struct mbuf *mold;
	int s;

	if ((la = arplookup1(m, &satocsin(dst)->sin_addr, 1, 0, rt)) != NULL)
		rt = la->la_rt;

	if (la == NULL || rt == NULL) {
		ARP_STATINC(ARP_STAT_ALLOCFAIL);
		log(LOG_DEBUG,
		    "arpresolve: can't allocate llinfo on %s for %s\n",
		    ifp->if_xname, in_fmtaddr(satocsin(dst)->sin_addr));
		m_freem(m);
		return 0;
	}
	sdl = satocsdl(rt->rt_gateway);
	/*
	 * Check the address family and length is valid, the address
	 * is resolved; otherwise, try to resolve.
	 */
	if ((rt->rt_expire == 0 || rt->rt_expire > time_second) &&
	    sdl->sdl_family == AF_LINK && sdl->sdl_alen != 0) {
		memcpy(desten, CLLADDR(sdl),
		    min(sdl->sdl_alen, ifp->if_addrlen));
		rt->rt_pksent = time_second; /* Time for last pkt sent */
		return 1;
	}
	/*
	 * There is an arptab entry, but no ethernet address
	 * response yet.  Replace the held mbuf with this
	 * latest one.
	 */

	ARP_STATINC(ARP_STAT_DFRTOTAL);
	s = splnet();
	mold = la->la_hold;
	la->la_hold = m;
	splx(s);

	if (mold) {
		ARP_STATINC(ARP_STAT_DFRDROPPED);
		m_freem(mold);
	}

	/*
	 * Re-send the ARP request when appropriate.
	 */
#ifdef	DIAGNOSTIC
	if (rt->rt_expire == 0) {
		/* This should never happen. (Should it? -gwr) */
		printf("arpresolve: unresolved and rt_expire == 0\n");
		/* Set expiration time to now (expired). */
		rt->rt_expire = time_second;
	}
#endif
	if (rt->rt_expire) {
		rt->rt_flags &= ~RTF_REJECT;
		if (la->la_asked == 0 || rt->rt_expire != time_second) {
			rt->rt_expire = time_second;
			if (la->la_asked++ < arp_maxtries) {
				arprequest(ifp,
				    &satocsin(rt->rt_ifa->ifa_addr)->sin_addr,
				    &satocsin(dst)->sin_addr,
#if NCARP > 0
				    (rt->rt_ifp->if_type == IFT_CARP) ?
				    CLLADDR(rt->rt_ifp->if_sadl):
#endif
				    CLLADDR(ifp->if_sadl));
			} else {
				rt->rt_flags |= RTF_REJECT;
				rt->rt_expire += arpt_down;
				la->la_asked = 0;
			}
		}
	}
	return 0;
}

/*
 * Common length and type checks are done here,
 * then the protocol-specific routine is called.
 */
void
arpintr(void)
{
	struct mbuf *m;
	struct arphdr *ar;
	int s;
	int arplen;

	mutex_enter(softnet_lock);
	KERNEL_LOCK(1, NULL);
	while (arpintrq.ifq_head) {
		s = splnet();
		IF_DEQUEUE(&arpintrq, m);
		splx(s);
		if (m == 0 || (m->m_flags & M_PKTHDR) == 0)
			panic("arpintr");

		MCLAIM(m, &arpdomain.dom_mowner);
		ARP_STATINC(ARP_STAT_RCVTOTAL);

		/*
		 * First, make sure we have at least struct arphdr.
		 */
		if (m->m_len < sizeof(struct arphdr) ||
		    (ar = mtod(m, struct arphdr *)) == NULL)
			goto badlen;

		switch (m->m_pkthdr.rcvif->if_type) {
		case IFT_IEEE1394:
			arplen = sizeof(struct arphdr) +
			    ar->ar_hln + 2 * ar->ar_pln;
			break;
		default:
			arplen = sizeof(struct arphdr) +
			    2 * ar->ar_hln + 2 * ar->ar_pln;
			break;
		}

		if (/* XXX ntohs(ar->ar_hrd) == ARPHRD_ETHER && */
		    m->m_len >= arplen)
			switch (ntohs(ar->ar_pro)) {
			case ETHERTYPE_IP:
			case ETHERTYPE_IPTRAILERS:
				in_arpinput(m);
				continue;
			default:
				ARP_STATINC(ARP_STAT_RCVBADPROTO);
			}
		else {
badlen:
			ARP_STATINC(ARP_STAT_RCVBADLEN);
		}
		m_freem(m);
	}
	KERNEL_UNLOCK_ONE(NULL);
	mutex_exit(softnet_lock);
}

/*
 * ARP for Internet protocols on 10 Mb/s Ethernet.
 * Algorithm is that given in RFC 826.
 * In addition, a sanity check is performed on the sender
 * protocol address, to catch impersonators.
 * We no longer handle negotiations for use of trailer protocol:
 * Formerly, ARP replied for protocol type ETHERTYPE_TRAIL sent
 * along with IP replies if we wanted trailers sent to us,
 * and also sent them in response to IP replies.
 * This allowed either end to announce the desire to receive
 * trailer packets.
 * We no longer reply to requests for ETHERTYPE_TRAIL protocol either,
 * but formerly didn't normally send requests.
 */
static void
in_arpinput(struct mbuf *m)
{
	struct arphdr *ah;
	struct ifnet *ifp = m->m_pkthdr.rcvif;
	struct llinfo_arp *la = NULL;
	struct rtentry  *rt;
	struct in_ifaddr *ia;
#if NBRIDGE > 0
	struct in_ifaddr *bridge_ia = NULL;
#endif
#if NCARP > 0
	u_int32_t count = 0, index = 0;
#endif
	struct sockaddr_dl *sdl;
	struct sockaddr sa;
	struct in_addr isaddr, itaddr, myaddr;
	int op;
	struct mbuf *mold;
	void *tha;
	int s;
	uint64_t *arps;

	if (__predict_false(m_makewritable(&m, 0, m->m_pkthdr.len, M_DONTWAIT)))
		goto out;
	ah = mtod(m, struct arphdr *);
	op = ntohs(ah->ar_op);

	/*
	 * Fix up ah->ar_hrd if necessary, before using ar_tha() or
	 * ar_tpa().
	 */
	switch (ifp->if_type) {
	case IFT_IEEE1394:
		if (ntohs(ah->ar_hrd) == ARPHRD_IEEE1394)
			;
		else {
			/* XXX this is to make sure we compute ar_tha right */
			/* XXX check ar_hrd more strictly? */
			ah->ar_hrd = htons(ARPHRD_IEEE1394);
		}
		break;
	default:
		/* XXX check ar_hrd? */
		break;
	}

	memcpy(&isaddr, ar_spa(ah), sizeof (isaddr));
	memcpy(&itaddr, ar_tpa(ah), sizeof (itaddr));

	if (m->m_flags & (M_BCAST|M_MCAST))
		ARP_STATINC(ARP_STAT_RCVMCAST);

	/*
	 * If the target IP address is zero, ignore the packet.
	 * This prevents the code below from tring to answer
	 * when we are using IP address zero (booting).
	 */
	if (in_nullhost(itaddr)) {
		ARP_STATINC(ARP_STAT_RCVZEROTPA);
		goto out;
	}


	/*
	 * Search for a matching interface address
	 * or any address on the interface to use
	 * as a dummy address in the rest of this function
	 */
	
	INADDR_TO_IA(itaddr, ia);
	while (ia != NULL) {
#if NCARP > 0
		if (ia->ia_ifp->if_type == IFT_CARP &&
		    ((ia->ia_ifp->if_flags & (IFF_UP|IFF_RUNNING)) ==
		    (IFF_UP|IFF_RUNNING))) {
			index++;
			if (ia->ia_ifp == m->m_pkthdr.rcvif &&
			    carp_iamatch(ia, ar_sha(ah),
			    &count, index)) {
				break;
				}
		} else
#endif
			    if (ia->ia_ifp == m->m_pkthdr.rcvif)
				break;
#if NBRIDGE > 0
		/*
		 * If the interface we received the packet on
		 * is part of a bridge, check to see if we need
		 * to "bridge" the packet to ourselves at this
		 * layer.  Note we still prefer a perfect match,
		 * but allow this weaker match if necessary.
		 */
		if (m->m_pkthdr.rcvif->if_bridge != NULL &&
		    m->m_pkthdr.rcvif->if_bridge == ia->ia_ifp->if_bridge)
			bridge_ia = ia;
#endif /* NBRIDGE > 0 */

		NEXT_IA_WITH_SAME_ADDR(ia);
	}

#if NBRIDGE > 0
	if (ia == NULL && bridge_ia != NULL) {
		ia = bridge_ia;
		ifp = bridge_ia->ia_ifp;
	}
#endif

	if (ia == NULL) {
		INADDR_TO_IA(isaddr, ia);
		while ((ia != NULL) && ia->ia_ifp != m->m_pkthdr.rcvif)
			NEXT_IA_WITH_SAME_ADDR(ia);

		if (ia == NULL) {
			IFP_TO_IA(ifp, ia);
			if (ia == NULL) {
				ARP_STATINC(ARP_STAT_RCVNOINT);
				goto out;
			}
		}
	}

	myaddr = ia->ia_addr.sin_addr;

	/* XXX checks for bridge case? */
	if (!memcmp(ar_sha(ah), CLLADDR(ifp->if_sadl), ifp->if_addrlen)) {
		ARP_STATINC(ARP_STAT_RCVLOCALSHA);
		goto out;	/* it's from me, ignore it. */
	}

	/* XXX checks for bridge case? */
	if (!memcmp(ar_sha(ah), ifp->if_broadcastaddr, ifp->if_addrlen)) {
		ARP_STATINC(ARP_STAT_RCVBCASTSHA);
		log(LOG_ERR,
		    "%s: arp: link address is broadcast for IP address %s!\n",
		    ifp->if_xname, in_fmtaddr(isaddr));
		goto out;
	}

	/*
	 * If the source IP address is zero, this is an RFC 5227 ARP probe
	 */
	if (in_nullhost(isaddr)) {
		ARP_STATINC(ARP_STAT_RCVZEROSPA);
		goto reply;
	}

	if (in_hosteq(isaddr, myaddr)) {
		ARP_STATINC(ARP_STAT_RCVLOCALSPA);
		log(LOG_ERR,
		   "duplicate IP address %s sent from link address %s\n",
		   in_fmtaddr(isaddr), lla_snprintf(ar_sha(ah), ah->ar_hln));
		itaddr = myaddr;
		goto reply;
	}
	la = arplookup(m, &isaddr, in_hosteq(itaddr, myaddr), 0);
	if (la != NULL && (rt = la->la_rt) && (sdl = satosdl(rt->rt_gateway))) {
		if (sdl->sdl_alen &&
		    memcmp(ar_sha(ah), CLLADDR(sdl), sdl->sdl_alen)) {
			if (rt->rt_flags & RTF_STATIC) {
				ARP_STATINC(ARP_STAT_RCVOVERPERM);
				if (!log_permanent_modify)
					goto out;
				log(LOG_INFO,
				    "%s tried to overwrite permanent arp info"
				    " for %s\n",
				    lla_snprintf(ar_sha(ah), ah->ar_hln),
				    in_fmtaddr(isaddr));
				goto out;
			} else if (rt->rt_ifp != ifp) {
				ARP_STATINC(ARP_STAT_RCVOVERINT);
				if (!log_wrong_iface)
					goto out;
				log(LOG_INFO,
				    "%s on %s tried to overwrite "
				    "arp info for %s on %s\n",
				    lla_snprintf(ar_sha(ah), ah->ar_hln),
				    ifp->if_xname, in_fmtaddr(isaddr),
				    rt->rt_ifp->if_xname);
				    goto out;
			} else {
				ARP_STATINC(ARP_STAT_RCVOVER);
				if (log_movements)
					log(LOG_INFO, "arp info overwritten "
					    "for %s by %s\n",
					    in_fmtaddr(isaddr),
					    lla_snprintf(ar_sha(ah),
					    ah->ar_hln));
			}
		}
		/*
		 * sanity check for the address length.
		 * XXX this does not work for protocols with variable address
		 * length. -is
		 */
		if (sdl->sdl_alen &&
		    sdl->sdl_alen != ah->ar_hln) {
			ARP_STATINC(ARP_STAT_RCVLENCHG);
			log(LOG_WARNING,
			    "arp from %s: new addr len %d, was %d\n",
			    in_fmtaddr(isaddr), ah->ar_hln, sdl->sdl_alen);
		}
		if (ifp->if_addrlen != ah->ar_hln) {
			ARP_STATINC(ARP_STAT_RCVBADLEN);
			log(LOG_WARNING,
			    "arp from %s: addr len: new %d, i/f %d (ignored)\n",
			    in_fmtaddr(isaddr), ah->ar_hln,
			    ifp->if_addrlen);
			goto reply;
		}
#if NTOKEN > 0
		/*
		 * XXX uses m_data and assumes the complete answer including
		 * XXX token-ring headers is in the same buf
		 */
		if (ifp->if_type == IFT_ISO88025) {
			struct token_header *trh;

			trh = (struct token_header *)M_TRHSTART(m);
			if (trh->token_shost[0] & TOKEN_RI_PRESENT) {
				struct token_rif	*rif;
				size_t	riflen;

				rif = TOKEN_RIF(trh);
				riflen = (ntohs(rif->tr_rcf) &
				    TOKEN_RCF_LEN_MASK) >> 8;

				if (riflen > 2 &&
				    riflen < sizeof(struct token_rif) &&
				    (riflen & 1) == 0) {
					rif->tr_rcf ^= htons(TOKEN_RCF_DIRECTION);
					rif->tr_rcf &= htons(~TOKEN_RCF_BROADCAST_MASK);
					memcpy(TOKEN_RIF(la), rif, riflen);
				}
			}
		}
#endif /* NTOKEN > 0 */
		(void)sockaddr_dl_setaddr(sdl, sdl->sdl_len, ar_sha(ah),
		    ah->ar_hln);
		if (rt->rt_expire)
			rt->rt_expire = time_second + arpt_keep;
		rt->rt_flags &= ~RTF_REJECT;
		la->la_asked = 0;

		s = splnet();
		mold = la->la_hold;
		la->la_hold = 0;
		splx(s);

		if (mold) {
			ARP_STATINC(ARP_STAT_DFRSENT);
			(*ifp->if_output)(ifp, mold, rt_getkey(rt), rt);
		}
	}
reply:
	if (op != ARPOP_REQUEST) {
		if (op == ARPOP_REPLY)
			ARP_STATINC(ARP_STAT_RCVREPLY);
	out:
		m_freem(m);
		return;
	}
	ARP_STATINC(ARP_STAT_RCVREQUEST);
	if (in_hosteq(itaddr, myaddr)) {
		/* I am the target */
		tha = ar_tha(ah);
		if (tha)
			memcpy(tha, ar_sha(ah), ah->ar_hln);
		memcpy(ar_sha(ah), CLLADDR(ifp->if_sadl), ah->ar_hln);
	} else {
		la = arplookup(m, &itaddr, 0, SIN_PROXY);
		if (la == NULL)
			goto out;
		rt = la->la_rt;
		if (rt->rt_ifp->if_type == IFT_CARP &&
		    m->m_pkthdr.rcvif->if_type != IFT_CARP)
			goto out;
		tha = ar_tha(ah);
		if (tha)
			memcpy(tha, ar_sha(ah), ah->ar_hln);
		sdl = satosdl(rt->rt_gateway);
		memcpy(ar_sha(ah), CLLADDR(sdl), ah->ar_hln);
	}

	memcpy(ar_tpa(ah), ar_spa(ah), ah->ar_pln);
	memcpy(ar_spa(ah), &itaddr, ah->ar_pln);
	ah->ar_op = htons(ARPOP_REPLY);
	ah->ar_pro = htons(ETHERTYPE_IP); /* let's be sure! */
	switch (ifp->if_type) {
	case IFT_IEEE1394:
		/*
		 * ieee1394 arp reply is broadcast
		 */
		m->m_flags &= ~M_MCAST;
		m->m_flags |= M_BCAST;
		m->m_len = sizeof(*ah) + (2 * ah->ar_pln) + ah->ar_hln;
		break;

	default:
		m->m_flags &= ~(M_BCAST|M_MCAST); /* never reply by broadcast */
		m->m_len = sizeof(*ah) + (2 * ah->ar_pln) + (2 * ah->ar_hln);
		break;
	}
	m->m_pkthdr.len = m->m_len;
	sa.sa_family = AF_ARP;
	sa.sa_len = 2;
	arps = ARP_STAT_GETREF();
	arps[ARP_STAT_SNDTOTAL]++;
	arps[ARP_STAT_SNDREPLY]++;
	ARP_STAT_PUTREF();
	(*ifp->if_output)(ifp, m, &sa, NULL);
	return;
}

/*
 * Free an arp entry.
 */
static void arptfree(struct llinfo_arp *la)
{
	struct rtentry *rt = la->la_rt;
	struct sockaddr_dl *sdl;

	ARP_LOCK_CHECK();

	if (rt == NULL)
		panic("arptfree");
	if (rt->rt_refcnt > 0 && (sdl = satosdl(rt->rt_gateway)) &&
	    sdl->sdl_family == AF_LINK) {
		sdl->sdl_alen = 0;
		la->la_asked = 0;
		rt->rt_flags &= ~RTF_REJECT;
		return;
	}
	rtrequest(RTM_DELETE, rt_getkey(rt), NULL, rt_mask(rt), 0, NULL);
}

static struct llinfo_arp *
arplookup(struct mbuf *m, const struct in_addr *addr, int create, int proxy)
{
	return arplookup1(m, addr, create, proxy, NULL);
}

/*
 * Lookup or enter a new address in arptab.
 */
static struct llinfo_arp *
arplookup1(struct mbuf *m, const struct in_addr *addr, int create, int proxy,
    struct rtentry *rt0)
{
	struct arphdr *ah;
	struct ifnet *ifp = m->m_pkthdr.rcvif;
	struct rtentry *rt;
	struct sockaddr_inarp sin;
	const char *why = NULL;

	ah = mtod(m, struct arphdr *);
	if (rt0 == NULL) {
		memset(&sin, 0, sizeof(sin));
		sin.sin_len = sizeof(sin);
		sin.sin_family = AF_INET;
		sin.sin_addr = *addr;
		sin.sin_other = proxy ? SIN_PROXY : 0;
		rt = rtalloc1(sintosa(&sin), create);
		if (rt == NULL)
			return NULL;
		rt->rt_refcnt--;
	} else
		rt = rt0;

#define	IS_LLINFO(__rt)							  \
	(((__rt)->rt_flags & (RTF_GATEWAY | RTF_LLINFO)) == RTF_LLINFO && \
	 (__rt)->rt_gateway->sa_family == AF_LINK)


	if (IS_LLINFO(rt))
		return (struct llinfo_arp *)rt->rt_llinfo;

	if (create) {
		if (rt->rt_flags & RTF_GATEWAY)
			why = "host is not on local network";
		else if ((rt->rt_flags & RTF_LLINFO) == 0) {
			ARP_STATINC(ARP_STAT_ALLOCFAIL);
			why = "could not allocate llinfo";
		} else
			why = "gateway route is not ours";
		log(LOG_DEBUG, "arplookup: unable to enter address"
		    " for %s@%s on %s (%s)\n",
		    in_fmtaddr(*addr), lla_snprintf(ar_sha(ah), ah->ar_hln),
		    (ifp) ? ifp->if_xname : "null", why);
		if (rt->rt_refcnt <= 0 && (rt->rt_flags & RTF_CLONED) != 0) {
			rtrequest(RTM_DELETE, rt_getkey(rt),
		    	    rt->rt_gateway, rt_mask(rt), rt->rt_flags, NULL);
		}
	}
	return NULL;
}

int
arpioctl(u_long cmd, void *data)
{

	return EOPNOTSUPP;
}

void
arp_ifinit(struct ifnet *ifp, struct ifaddr *ifa)
{
	struct in_addr *ip;

	/*
	 * Warn the user if another station has this IP address,
	 * but only if the interface IP address is not zero.
	 */
	ip = &IA_SIN(ifa)->sin_addr;
	if (!in_nullhost(*ip))
		arprequest(ifp, ip, ip, CLLADDR(ifp->if_sadl));

	ifa->ifa_rtrequest = arp_rtrequest;
	ifa->ifa_flags |= RTF_CLONING;
}

/*
 * Called from 10 Mb/s Ethernet interrupt handlers
 * when ether packet type ETHERTYPE_REVARP
 * is received.  Common length and type checks are done here,
 * then the protocol-specific routine is called.
 */
void
revarpinput(struct mbuf *m)
{
	struct arphdr *ar;

	if (m->m_len < sizeof(struct arphdr))
		goto out;
	ar = mtod(m, struct arphdr *);
#if 0 /* XXX I don't think we need this... and it will prevent other LL */
	if (ntohs(ar->ar_hrd) != ARPHRD_ETHER)
		goto out;
#endif
	if (m->m_len < sizeof(struct arphdr) + 2 * (ar->ar_hln + ar->ar_pln))
		goto out;
	switch (ntohs(ar->ar_pro)) {
	case ETHERTYPE_IP:
	case ETHERTYPE_IPTRAILERS:
		in_revarpinput(m);
		return;

	default:
		break;
	}
out:
	m_freem(m);
}

/*
 * RARP for Internet protocols on 10 Mb/s Ethernet.
 * Algorithm is that given in RFC 903.
 * We are only using for bootstrap purposes to get an ip address for one of
 * our interfaces.  Thus we support no user-interface.
 *
 * Since the contents of the RARP reply are specific to the interface that
 * sent the request, this code must ensure that they are properly associated.
 *
 * Note: also supports ARP via RARP packets, per the RFC.
 */
void
in_revarpinput(struct mbuf *m)
{
	struct ifnet *ifp;
	struct arphdr *ah;
	void *tha;
	int op;

	ah = mtod(m, struct arphdr *);
	op = ntohs(ah->ar_op);

	switch (m->m_pkthdr.rcvif->if_type) {
	case IFT_IEEE1394:
		/* ARP without target hardware address is not supported */
		goto out;
	default:
		break;
	}

	switch (op) {
	case ARPOP_REQUEST:
	case ARPOP_REPLY:	/* per RFC */
		in_arpinput(m);
		return;
	case ARPOP_REVREPLY:
		break;
	case ARPOP_REVREQUEST:	/* handled by rarpd(8) */
	default:
		goto out;
	}
	if (!revarp_in_progress)
		goto out;
	ifp = m->m_pkthdr.rcvif;
	if (ifp != myip_ifp) /* !same interface */
		goto out;
	if (myip_initialized)
		goto wake;
	tha = ar_tha(ah);
	if (tha == NULL)
		goto out;
	if (memcmp(tha, CLLADDR(ifp->if_sadl), ifp->if_sadl->sdl_alen))
		goto out;
	memcpy(&srv_ip, ar_spa(ah), sizeof(srv_ip));
	memcpy(&myip, ar_tpa(ah), sizeof(myip));
	myip_initialized = 1;
wake:	/* Do wakeup every time in case it was missed. */
	wakeup((void *)&myip);

out:
	m_freem(m);
}

/*
 * Send a RARP request for the ip address of the specified interface.
 * The request should be RFC 903-compliant.
 */
void
revarprequest(struct ifnet *ifp)
{
	struct sockaddr sa;
	struct mbuf *m;
	struct arphdr *ah;
	void *tha;

	if ((m = m_gethdr(M_DONTWAIT, MT_DATA)) == NULL)
		return;
	MCLAIM(m, &arpdomain.dom_mowner);
	m->m_len = sizeof(*ah) + 2*sizeof(struct in_addr) +
	    2*ifp->if_addrlen;
	m->m_pkthdr.len = m->m_len;
	MH_ALIGN(m, m->m_len);
	ah = mtod(m, struct arphdr *);
	memset(ah, 0, m->m_len);
	ah->ar_pro = htons(ETHERTYPE_IP);
	ah->ar_hln = ifp->if_addrlen;		/* hardware address length */
	ah->ar_pln = sizeof(struct in_addr);	/* protocol address length */
	ah->ar_op = htons(ARPOP_REVREQUEST);

	memcpy(ar_sha(ah), CLLADDR(ifp->if_sadl), ah->ar_hln);
	tha = ar_tha(ah);
	if (tha == NULL)
		return;
	memcpy(tha, CLLADDR(ifp->if_sadl), ah->ar_hln);

	sa.sa_family = AF_ARP;
	sa.sa_len = 2;
	m->m_flags |= M_BCAST;
	(*ifp->if_output)(ifp, m, &sa, NULL);

}

/*
 * RARP for the ip address of the specified interface, but also
 * save the ip address of the server that sent the answer.
 * Timeout if no response is received.
 */
int
revarpwhoarewe(struct ifnet *ifp, struct in_addr *serv_in,
    struct in_addr *clnt_in)
{
	int result, count = 20;

	myip_initialized = 0;
	myip_ifp = ifp;

	revarp_in_progress = 1;
	while (count--) {
		revarprequest(ifp);
		result = tsleep((void *)&myip, PSOCK, "revarp", hz/2);
		if (result != EWOULDBLOCK)
			break;
	}
	revarp_in_progress = 0;

	if (!myip_initialized)
		return ENETUNREACH;

	memcpy(serv_in, &srv_ip, sizeof(*serv_in));
	memcpy(clnt_in, &myip, sizeof(*clnt_in));
	return 0;
}



#ifdef DDB

#include <machine/db_machdep.h>
#include <ddb/db_interface.h>
#include <ddb/db_output.h>

static void
db_print_sa(const struct sockaddr *sa)
{
	int len;
	const u_char *p;

	if (sa == NULL) {
		db_printf("[NULL]");
		return;
	}

	p = (const u_char *)sa;
	len = sa->sa_len;
	db_printf("[");
	while (len > 0) {
		db_printf("%d", *p);
		p++; len--;
		if (len) db_printf(",");
	}
	db_printf("]\n");
}

static void
db_print_ifa(struct ifaddr *ifa)
{
	if (ifa == NULL)
		return;
	db_printf("  ifa_addr=");
	db_print_sa(ifa->ifa_addr);
	db_printf("  ifa_dsta=");
	db_print_sa(ifa->ifa_dstaddr);
	db_printf("  ifa_mask=");
	db_print_sa(ifa->ifa_netmask);
	db_printf("  flags=0x%x,refcnt=%d,metric=%d\n",
			  ifa->ifa_flags,
			  ifa->ifa_refcnt,
			  ifa->ifa_metric);
}

static void
db_print_llinfo(void *li)
{
	struct llinfo_arp *la;

	if (li == NULL)
		return;
	la = (struct llinfo_arp *)li;
	db_printf("  la_rt=%p la_hold=%p, la_asked=0x%lx\n",
			  la->la_rt, la->la_hold, la->la_asked);
}

/*
 * Function to pass to rt_walktree().
 * Return non-zero error to abort walk.
 */
static int
db_show_rtentry(struct rtentry *rt, void *w)
{
	db_printf("rtentry=%p", rt);

	db_printf(" flags=0x%x refcnt=%d use=%"PRId64" expire=%"PRId64"\n",
			  rt->rt_flags, rt->rt_refcnt,
			  rt->rt_use, (uint64_t)rt->rt_expire);

	db_printf(" key="); db_print_sa(rt_getkey(rt));
	db_printf(" mask="); db_print_sa(rt_mask(rt));
	db_printf(" gw="); db_print_sa(rt->rt_gateway);

	db_printf(" ifp=%p ", rt->rt_ifp);
	if (rt->rt_ifp)
		db_printf("(%s)", rt->rt_ifp->if_xname);
	else
		db_printf("(NULL)");

	db_printf(" ifa=%p\n", rt->rt_ifa);
	db_print_ifa(rt->rt_ifa);

	db_printf(" gwroute=%p llinfo=%p\n",
			  rt->rt_gwroute, rt->rt_llinfo);
	db_print_llinfo(rt->rt_llinfo);

	return 0;
}

/*
 * Function to print all the route trees.
 * Use this from ddb:  "show arptab"
 */
void
db_show_arptab(db_expr_t addr, bool have_addr,
    db_expr_t count, const char *modif)
{
	rt_walktree(AF_INET, db_show_rtentry, NULL);
}
#endif

static int
sysctl_net_inet_arp_stats(SYSCTLFN_ARGS)
{

	return NETSTAT_SYSCTL(arpstat_percpu, ARP_NSTATS);
}

static void
sysctl_net_inet_arp_setup(struct sysctllog **clog)
{
	const struct sysctlnode *node;

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT,
			CTLTYPE_NODE, "inet", NULL,
			NULL, 0, NULL, 0,
			CTL_NET, PF_INET, CTL_EOL);
	sysctl_createv(clog, 0, NULL, &node,
			CTLFLAG_PERMANENT,
			CTLTYPE_NODE, "arp",
			SYSCTL_DESCR("Address Resolution Protocol"),
			NULL, 0, NULL, 0,
			CTL_NET, PF_INET, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "prune",
			SYSCTL_DESCR("ARP cache pruning interval in seconds"),
			NULL, 0, &arpt_prune, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "keep",
			SYSCTL_DESCR("Valid ARP entry lifetime in seconds"),
			NULL, 0, &arpt_keep, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "down",
			SYSCTL_DESCR("Failed ARP entry lifetime in seconds"),
			NULL, 0, &arpt_down, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "refresh",
			SYSCTL_DESCR("ARP entry refresh interval"),
			NULL, 0, &arpt_refresh, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);
	
	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT,
			CTLTYPE_STRUCT, "stats",
			SYSCTL_DESCR("ARP statistics"),
			sysctl_net_inet_arp_stats, 0, NULL, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "log_movements",
			SYSCTL_DESCR("log ARP replies from MACs different than"
			    " the one in the cache"),
			NULL, 0, &log_movements, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "log_permanent_modify",
			SYSCTL_DESCR("log ARP replies from MACs different than"
			    " the one in the permanent arp entry"),
			NULL, 0, &log_permanent_modify, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);

	sysctl_createv(clog, 0, NULL, NULL,
			CTLFLAG_PERMANENT|CTLFLAG_READWRITE,
			CTLTYPE_INT, "log_wrong_iface",
			SYSCTL_DESCR("log ARP packets arriving on the wrong"
			    " interface"),
			NULL, 0, &log_wrong_iface, 0,
			CTL_NET,PF_INET, node->sysctl_num, CTL_CREATE, CTL_EOL);
}

#endif /* INET */