summaryrefslogtreecommitdiffstats
path: root/sys/net/route.c
blob: de677bce95bef0c51812ade70a41f1ea9170727a (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
/*-
 * Copyright (c) 1980, 1986, 1991, 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.
 * 4. 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.
 *
 *	@(#)route.c	8.3.1.1 (Berkeley) 2/23/95
 * $FreeBSD$
 */
/************************************************************************
 * Note: In this file a 'fib' is a "forwarding information base"	*
 * Which is the new name for an in kernel routing (next hop) table.	*
 ***********************************************************************/

#include "opt_inet.h"
#include "opt_route.h"
#include "opt_mrouting.h"
#include "opt_mpath.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/syslog.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/sysctl.h>
#include <sys/syslog.h>
#include <sys/sysproto.h>
#include <sys/proc.h>
#include <sys/domain.h>
#include <sys/kernel.h>
#include <sys/vimage.h>

#include <net/if.h>
#include <net/if_dl.h>
#include <net/route.h>

#ifdef RADIX_MPATH
#include <net/radix_mpath.h>
#endif

#include <netinet/in.h>
#include <netinet/ip_mroute.h>

#include <vm/uma.h>

u_int rt_numfibs = RT_NUMFIBS;
SYSCTL_INT(_net, OID_AUTO, fibs, CTLFLAG_RD, &rt_numfibs, 0, "");
/*
 * Allow the boot code to allow LESS than RT_MAXFIBS to be used.
 * We can't do more because storage is statically allocated for now.
 * (for compatibility reasons.. this will change).
 */
TUNABLE_INT("net.fibs", &rt_numfibs);

/*
 * By default add routes to all fibs for new interfaces.
 * Once this is set to 0 then only allocate routes on interface
 * changes for the FIB of the caller when adding a new set of addresses
 * to an interface.  XXX this is a shotgun aproach to a problem that needs
 * a more fine grained solution.. that will come.
 */
u_int rt_add_addr_allfibs = 1;
SYSCTL_INT(_net, OID_AUTO, add_addr_allfibs, CTLFLAG_RW,
    &rt_add_addr_allfibs, 0, "");
TUNABLE_INT("net.add_addr_allfibs", &rt_add_addr_allfibs);

VNET_DEFINE(struct radix_node_head *, rt_tables);
static VNET_DEFINE(uma_zone_t, rtzone);		/* Routing table UMA zone. */
VNET_DEFINE(int, rttrash);		/* routes not in table but not freed */
VNET_DEFINE(struct rtstat, rtstat);

#define	V_rt_tables	VNET(rt_tables)
#define	V_rtzone	VNET(rtzone)
#define	V_rttrash	VNET(rttrash)
#define	V_rtstat	VNET(rtstat)

static void rt_maskedcopy(struct sockaddr *,
	    struct sockaddr *, struct sockaddr *);
static int vnet_route_iattach(const void *);
#ifdef VIMAGE
static int vnet_route_idetach(const void *);

static const vnet_modinfo_t vnet_rtable_modinfo = {
	.vmi_id		= VNET_MOD_RTABLE,
	.vmi_name	= "rtable",
	.vmi_iattach	= vnet_route_iattach,
	.vmi_idetach	= vnet_route_idetach
};
#endif

/* compare two sockaddr structures */
#define	sa_equal(a1, a2) (bcmp((a1), (a2), (a1)->sa_len) == 0)

/*
 * Convert a 'struct radix_node *' to a 'struct rtentry *'.
 * The operation can be done safely (in this code) because a
 * 'struct rtentry' starts with two 'struct radix_node''s, the first
 * one representing leaf nodes in the routing tree, which is
 * what the code in radix.c passes us as a 'struct radix_node'.
 *
 * But because there are a lot of assumptions in this conversion,
 * do not cast explicitly, but always use the macro below.
 */
#define RNTORT(p)	((struct rtentry *)(p))

#if 0
/* default fib for tunnels to use */
u_int tunnel_fib = 0;
SYSCTL_INT(_net, OID_AUTO, tunnelfib, CTLFLAG_RD, &tunnel_fib, 0, "");
#endif

/*
 * handler for net.my_fibnum
 */
static int
sysctl_my_fibnum(SYSCTL_HANDLER_ARGS)
{
        int fibnum;
        int error;
 
        fibnum = curthread->td_proc->p_fibnum;
        error = sysctl_handle_int(oidp, &fibnum, 0, req);
        return (error);
}

SYSCTL_PROC(_net, OID_AUTO, my_fibnum, CTLTYPE_INT|CTLFLAG_RD,
            NULL, 0, &sysctl_my_fibnum, "I", "default FIB of caller");

static __inline struct radix_node_head **
rt_tables_get_rnh_ptr(int table, int fam)
{
	struct radix_node_head **rnh;

	KASSERT(table >= 0 && table < rt_numfibs, ("%s: table out of bounds.",
	    __func__));
	KASSERT(fam >= 0 && fam < (AF_MAX+1), ("%s: fam out of bounds.",
	    __func__));

	/* rnh is [fib=0][af=0]. */
	rnh = (struct radix_node_head **)V_rt_tables;
	/* Get the offset to the requested table and fam. */
	rnh += table * (AF_MAX+1) + fam;

	return (rnh);
}

struct radix_node_head *
rt_tables_get_rnh(int table, int fam)
{

	return (*rt_tables_get_rnh_ptr(table, fam));
}

static void
route_init(void)
{

	/* whack the tunable ints into  line. */
	if (rt_numfibs > RT_MAXFIBS)
		rt_numfibs = RT_MAXFIBS;
	if (rt_numfibs == 0)
		rt_numfibs = 1;
	rn_init();	/* initialize all zeroes, all ones, mask table */

#ifdef VIMAGE
	vnet_mod_register(&vnet_rtable_modinfo);
#else
	vnet_route_iattach(NULL);
#endif
}

static int
vnet_route_iattach(const void *unused __unused)
{
	struct domain *dom;
	struct radix_node_head **rnh;
	int table;
	int fam;

	V_rt_tables = malloc(rt_numfibs * (AF_MAX+1) *
	    sizeof(struct radix_node_head *), M_RTABLE, M_WAITOK|M_ZERO);

	V_rtzone = uma_zcreate("rtentry", sizeof(struct rtentry), NULL, NULL,
	    NULL, NULL, UMA_ALIGN_PTR, 0);
	for (dom = domains; dom; dom = dom->dom_next) {
		if (dom->dom_rtattach)  {
			for  (table = 0; table < rt_numfibs; table++) {
				if ( (fam = dom->dom_family) == AF_INET ||
				    table == 0) {
 			        	/* for now only AF_INET has > 1 table */
					/* XXX MRT 
					 * rtattach will be also called
					 * from vfs_export.c but the
					 * offset will be 0
					 * (only for AF_INET and AF_INET6
					 * which don't need it anyhow)
					 */
					rnh = rt_tables_get_rnh_ptr(table, fam);
					if (rnh == NULL)
						panic("%s: rnh NULL", __func__);
					dom->dom_rtattach((void **)rnh,
				    	    dom->dom_rtoffset);
				} else {
					break;
				}
			}
		}
	}

	return (0);
}

#ifdef VIMAGE
static int
vnet_route_idetach(const void *unused __unused)
{
	int table;
	int fam;
	struct domain *dom;
	struct radix_node_head **rnh;

	for (dom = domains; dom; dom = dom->dom_next) {
		if (dom->dom_rtdetach) {
			for (table = 0; table < rt_numfibs; table++) {
				if ( (fam = dom->dom_family) == AF_INET ||
				    table == 0) {
					/* For now only AF_INET has > 1 tbl. */
					rnh = rt_tables_get_rnh_ptr(table, fam);
					if (rnh == NULL)
						panic("%s: rnh NULL", __func__);
					dom->dom_rtdetach((void **)rnh,
					    dom->dom_rtoffset);
				} else {
					break;
				}
			}
		}
	}
	return (0);
}
#endif

#ifndef _SYS_SYSPROTO_H_
struct setfib_args {
	int     fibnum;
};
#endif
int
setfib(struct thread *td, struct setfib_args *uap)
{
	if (uap->fibnum < 0 || uap->fibnum >= rt_numfibs)
		return EINVAL;
	td->td_proc->p_fibnum = uap->fibnum;
	return (0);
}

/*
 * Packet routing routines.
 */
void
rtalloc(struct route *ro)
{
	rtalloc_ign_fib(ro, 0UL, 0);
}

void
rtalloc_fib(struct route *ro, u_int fibnum)
{
	rtalloc_ign_fib(ro, 0UL, fibnum);
}

void
rtalloc_ign(struct route *ro, u_long ignore)
{
	struct rtentry *rt;

	if ((rt = ro->ro_rt) != NULL) {
		if (rt->rt_ifp != NULL && rt->rt_flags & RTF_UP)
			return;
		RTFREE(rt);
		ro->ro_rt = NULL;
	}
	ro->ro_rt = rtalloc1_fib(&ro->ro_dst, 1, ignore, 0);
	if (ro->ro_rt)
		RT_UNLOCK(ro->ro_rt);
}

void
rtalloc_ign_fib(struct route *ro, u_long ignore, u_int fibnum)
{
	struct rtentry *rt;

	if ((rt = ro->ro_rt) != NULL) {
		if (rt->rt_ifp != NULL && rt->rt_flags & RTF_UP)
			return;
		RTFREE(rt);
		ro->ro_rt = NULL;
	}
	ro->ro_rt = rtalloc1_fib(&ro->ro_dst, 1, ignore, fibnum);
	if (ro->ro_rt)
		RT_UNLOCK(ro->ro_rt);
}

/*
 * Look up the route that matches the address given
 * Or, at least try.. Create a cloned route if needed.
 *
 * The returned route, if any, is locked.
 */
struct rtentry *
rtalloc1(struct sockaddr *dst, int report, u_long ignflags)
{
	return (rtalloc1_fib(dst, report, ignflags, 0));
}

struct rtentry *
rtalloc1_fib(struct sockaddr *dst, int report, u_long ignflags,
		    u_int fibnum)
{
	struct radix_node_head *rnh;
	struct rtentry *rt;
	struct radix_node *rn;
	struct rtentry *newrt;
	struct rt_addrinfo info;
	int err = 0, msgtype = RTM_MISS;
	int needlock;

	KASSERT((fibnum < rt_numfibs), ("rtalloc1_fib: bad fibnum"));
	if (dst->sa_family != AF_INET)	/* Only INET supports > 1 fib now */
		fibnum = 0;
	rnh = rt_tables_get_rnh(fibnum, dst->sa_family);
	newrt = NULL;
	/*
	 * Look up the address in the table for that Address Family
	 */
	if (rnh == NULL) {
		V_rtstat.rts_unreach++;
		goto miss;
	}
	needlock = !(ignflags & RTF_RNH_LOCKED);
	if (needlock)
		RADIX_NODE_HEAD_RLOCK(rnh);
#ifdef INVARIANTS	
	else
		RADIX_NODE_HEAD_LOCK_ASSERT(rnh);
#endif
	rn = rnh->rnh_matchaddr(dst, rnh);
	if (rn && ((rn->rn_flags & RNF_ROOT) == 0)) {
		newrt = rt = RNTORT(rn);
		RT_LOCK(newrt);
		RT_ADDREF(newrt);
		if (needlock)
			RADIX_NODE_HEAD_RUNLOCK(rnh);
		goto done;

	} else if (needlock)
		RADIX_NODE_HEAD_RUNLOCK(rnh);
	
	/*
	 * Either we hit the root or couldn't find any match,
	 * Which basically means
	 * "caint get there frm here"
	 */
	V_rtstat.rts_unreach++;
miss:
	if (report) {
		/*
		 * If required, report the failure to the supervising
		 * Authorities.
		 * For a delete, this is not an error. (report == 0)
		 */
		bzero(&info, sizeof(info));
		info.rti_info[RTAX_DST] = dst;
		rt_missmsg(msgtype, &info, 0, err);
	}	
done:
	if (newrt)
		RT_LOCK_ASSERT(newrt);
	return (newrt);
}

/*
 * Remove a reference count from an rtentry.
 * If the count gets low enough, take it out of the routing table
 */
void
rtfree(struct rtentry *rt)
{
	struct radix_node_head *rnh;

	KASSERT(rt != NULL,("%s: NULL rt", __func__));
	rnh = rt_tables_get_rnh(rt->rt_fibnum, rt_key(rt)->sa_family);
	KASSERT(rnh != NULL,("%s: NULL rnh", __func__));

	RT_LOCK_ASSERT(rt);

	/*
	 * The callers should use RTFREE_LOCKED() or RTFREE(), so
	 * we should come here exactly with the last reference.
	 */
	RT_REMREF(rt);
	if (rt->rt_refcnt > 0) {
		log(LOG_DEBUG, "%s: %p has %d refs\n", __func__, rt, rt->rt_refcnt);
		goto done;
	}

	/*
	 * On last reference give the "close method" a chance
	 * to cleanup private state.  This also permits (for
	 * IPv4 and IPv6) a chance to decide if the routing table
	 * entry should be purged immediately or at a later time.
	 * When an immediate purge is to happen the close routine
	 * typically calls rtexpunge which clears the RTF_UP flag
	 * on the entry so that the code below reclaims the storage.
	 */
	if (rt->rt_refcnt == 0 && rnh->rnh_close)
		rnh->rnh_close((struct radix_node *)rt, rnh);

	/*
	 * If we are no longer "up" (and ref == 0)
	 * then we can free the resources associated
	 * with the route.
	 */
	if ((rt->rt_flags & RTF_UP) == 0) {
		if (rt->rt_nodes->rn_flags & (RNF_ACTIVE | RNF_ROOT))
			panic("rtfree 2");
		/*
		 * the rtentry must have been removed from the routing table
		 * so it is represented in rttrash.. remove that now.
		 */
		V_rttrash--;
#ifdef	DIAGNOSTIC
		if (rt->rt_refcnt < 0) {
			printf("rtfree: %p not freed (neg refs)\n", rt);
			goto done;
		}
#endif
		/*
		 * release references on items we hold them on..
		 * e.g other routes and ifaddrs.
		 */
		if (rt->rt_ifa)
			ifa_free(rt->rt_ifa);
		/*
		 * The key is separatly alloc'd so free it (see rt_setgate()).
		 * This also frees the gateway, as they are always malloc'd
		 * together.
		 */
		Free(rt_key(rt));

		/*
		 * and the rtentry itself of course
		 */
		RT_LOCK_DESTROY(rt);
		uma_zfree(V_rtzone, rt);
		return;
	}
done:
	RT_UNLOCK(rt);
}


/*
 * Force a routing table entry to the specified
 * destination to go through the given gateway.
 * Normally called as a result of a routing redirect
 * message from the network layer.
 */
void
rtredirect(struct sockaddr *dst,
	struct sockaddr *gateway,
	struct sockaddr *netmask,
	int flags,
	struct sockaddr *src)
{
	rtredirect_fib(dst, gateway, netmask, flags, src, 0);
}

void
rtredirect_fib(struct sockaddr *dst,
	struct sockaddr *gateway,
	struct sockaddr *netmask,
	int flags,
	struct sockaddr *src,
	u_int fibnum)
{
	struct rtentry *rt, *rt0 = NULL;
	int error = 0;
	short *stat = NULL;
	struct rt_addrinfo info;
	struct ifaddr *ifa;
	struct radix_node_head *rnh;

	ifa = NULL;
	rnh = rt_tables_get_rnh(fibnum, dst->sa_family);
	if (rnh == NULL) {
		error = EAFNOSUPPORT;
		goto out;
	}

	/* verify the gateway is directly reachable */
	if ((ifa = ifa_ifwithnet(gateway)) == NULL) {
		error = ENETUNREACH;
		goto out;
	}
	rt = rtalloc1_fib(dst, 0, 0UL, fibnum);	/* NB: rt is locked */
	/*
	 * If the redirect isn't from our current router for this dst,
	 * it's either old or wrong.  If it redirects us to ourselves,
	 * we have a routing loop, perhaps as a result of an interface
	 * going down recently.
	 */
	if (!(flags & RTF_DONE) && rt &&
	     (!sa_equal(src, rt->rt_gateway) || rt->rt_ifa != ifa))
		error = EINVAL;
	else if (ifa_ifwithaddr_check(gateway))
		error = EHOSTUNREACH;
	if (error)
		goto done;
	/*
	 * Create a new entry if we just got back a wildcard entry
	 * or the the lookup failed.  This is necessary for hosts
	 * which use routing redirects generated by smart gateways
	 * to dynamically build the routing tables.
	 */
	if (rt == NULL || (rt_mask(rt) && rt_mask(rt)->sa_len < 2))
		goto create;
	/*
	 * Don't listen to the redirect if it's
	 * for a route to an interface.
	 */
	if (rt->rt_flags & RTF_GATEWAY) {
		if (((rt->rt_flags & RTF_HOST) == 0) && (flags & RTF_HOST)) {
			/*
			 * Changing from route to net => route to host.
			 * Create new route, rather than smashing route to net.
			 */
		create:
			rt0 = rt;
			rt = NULL;
		
			flags |=  RTF_GATEWAY | RTF_DYNAMIC;
			bzero((caddr_t)&info, sizeof(info));
			info.rti_info[RTAX_DST] = dst;
			info.rti_info[RTAX_GATEWAY] = gateway;
			info.rti_info[RTAX_NETMASK] = netmask;
			info.rti_ifa = ifa;
			info.rti_flags = flags;
			if (rt0 != NULL)
				RT_UNLOCK(rt0);	/* drop lock to avoid LOR with RNH */
			error = rtrequest1_fib(RTM_ADD, &info, &rt, fibnum);
			if (rt != NULL) {
				RT_LOCK(rt);
				if (rt0 != NULL)
					EVENTHANDLER_INVOKE(route_redirect_event, rt0, rt, dst);
				flags = rt->rt_flags;
			}
			if (rt0 != NULL)
				RTFREE(rt0);
			
			stat = &V_rtstat.rts_dynamic;
		} else {
			struct rtentry *gwrt;

			/*
			 * Smash the current notion of the gateway to
			 * this destination.  Should check about netmask!!!
			 */
			rt->rt_flags |= RTF_MODIFIED;
			flags |= RTF_MODIFIED;
			stat = &V_rtstat.rts_newgateway;
			/*
			 * add the key and gateway (in one malloc'd chunk).
			 */
			RT_UNLOCK(rt);
			RADIX_NODE_HEAD_LOCK(rnh);
			RT_LOCK(rt);
			rt_setgate(rt, rt_key(rt), gateway);
			gwrt = rtalloc1(gateway, 1, RTF_RNH_LOCKED);
			RADIX_NODE_HEAD_UNLOCK(rnh);
			EVENTHANDLER_INVOKE(route_redirect_event, rt, gwrt, dst);
			RTFREE_LOCKED(gwrt);
		}
	} else
		error = EHOSTUNREACH;
done:
	if (rt)
		RTFREE_LOCKED(rt);
out:
	if (error)
		V_rtstat.rts_badredirect++;
	else if (stat != NULL)
		(*stat)++;
	bzero((caddr_t)&info, sizeof(info));
	info.rti_info[RTAX_DST] = dst;
	info.rti_info[RTAX_GATEWAY] = gateway;
	info.rti_info[RTAX_NETMASK] = netmask;
	info.rti_info[RTAX_AUTHOR] = src;
	rt_missmsg(RTM_REDIRECT, &info, flags, error);
	if (ifa != NULL)
		ifa_free(ifa);
}

int
rtioctl(u_long req, caddr_t data)
{
	return (rtioctl_fib(req, data, 0));
}

/*
 * Routing table ioctl interface.
 */
int
rtioctl_fib(u_long req, caddr_t data, u_int fibnum)
{

	/*
	 * If more ioctl commands are added here, make sure the proper
	 * super-user checks are being performed because it is possible for
	 * prison-root to make it this far if raw sockets have been enabled
	 * in jails.
	 */
#ifdef INET
	/* Multicast goop, grrr... */
	return mrt_ioctl ? mrt_ioctl(req, data, fibnum) : EOPNOTSUPP;
#else /* INET */
	return ENXIO;
#endif /* INET */
}

/*
 * For both ifa_ifwithroute() routines, 'ifa' is returned referenced.
 */
struct ifaddr *
ifa_ifwithroute(int flags, struct sockaddr *dst, struct sockaddr *gateway)
{
	return (ifa_ifwithroute_fib(flags, dst, gateway, 0));
}

struct ifaddr *
ifa_ifwithroute_fib(int flags, struct sockaddr *dst, struct sockaddr *gateway,
				u_int fibnum)
{
	register struct ifaddr *ifa;
	int not_found = 0;

	if ((flags & RTF_GATEWAY) == 0) {
		/*
		 * If we are adding a route to an interface,
		 * and the interface is a pt to pt link
		 * we should search for the destination
		 * as our clue to the interface.  Otherwise
		 * we can use the local address.
		 */
		ifa = NULL;
		if (flags & RTF_HOST)
			ifa = ifa_ifwithdstaddr(dst);
		if (ifa == NULL)
			ifa = ifa_ifwithaddr(gateway);
	} else {
		/*
		 * If we are adding a route to a remote net
		 * or host, the gateway may still be on the
		 * other end of a pt to pt link.
		 */
		ifa = ifa_ifwithdstaddr(gateway);
	}
	if (ifa == NULL)
		ifa = ifa_ifwithnet(gateway);
	if (ifa == NULL) {
		struct rtentry *rt = rtalloc1_fib(gateway, 0, RTF_RNH_LOCKED, fibnum);
		if (rt == NULL)
			return (NULL);
		/*
		 * dismiss a gateway that is reachable only
		 * through the default router
		 */
		switch (gateway->sa_family) {
		case AF_INET:
			if (satosin(rt_key(rt))->sin_addr.s_addr == INADDR_ANY)
				not_found = 1;
			break;
		case AF_INET6:
			if (IN6_IS_ADDR_UNSPECIFIED(&satosin6(rt_key(rt))->sin6_addr))
				not_found = 1;
			break;
		default:
			break;
		}
		if (!not_found && rt->rt_ifa != NULL) {
			ifa = rt->rt_ifa;
			ifa_ref(ifa);
		}
		RT_REMREF(rt);
		RT_UNLOCK(rt);
		if (not_found || ifa == NULL)
			return (NULL);
	}
	if (ifa->ifa_addr->sa_family != dst->sa_family) {
		struct ifaddr *oifa = ifa;
		ifa = ifaof_ifpforaddr(dst, ifa->ifa_ifp);
		if (ifa == NULL)
			ifa = oifa;
		else
			ifa_free(oifa);
	}
	return (ifa);
}

/*
 * Do appropriate manipulations of a routing tree given
 * all the bits of info needed
 */
int
rtrequest(int req,
	struct sockaddr *dst,
	struct sockaddr *gateway,
	struct sockaddr *netmask,
	int flags,
	struct rtentry **ret_nrt)
{
	return (rtrequest_fib(req, dst, gateway, netmask, flags, ret_nrt, 0));
}

int
rtrequest_fib(int req,
	struct sockaddr *dst,
	struct sockaddr *gateway,
	struct sockaddr *netmask,
	int flags,
	struct rtentry **ret_nrt,
	u_int fibnum)
{
	struct rt_addrinfo info;

	if (dst->sa_len == 0)
		return(EINVAL);

	bzero((caddr_t)&info, sizeof(info));
	info.rti_flags = flags;
	info.rti_info[RTAX_DST] = dst;
	info.rti_info[RTAX_GATEWAY] = gateway;
	info.rti_info[RTAX_NETMASK] = netmask;
	return rtrequest1_fib(req, &info, ret_nrt, fibnum);
}

/*
 * These (questionable) definitions of apparent local variables apply
 * to the next two functions.  XXXXXX!!!
 */
#define	dst	info->rti_info[RTAX_DST]
#define	gateway	info->rti_info[RTAX_GATEWAY]
#define	netmask	info->rti_info[RTAX_NETMASK]
#define	ifaaddr	info->rti_info[RTAX_IFA]
#define	ifpaddr	info->rti_info[RTAX_IFP]
#define	flags	info->rti_flags

int
rt_getifa(struct rt_addrinfo *info)
{
	return (rt_getifa_fib(info, 0));
}

/*
 * Look up rt_addrinfo for a specific fib.  Note that if rti_ifa is defined,
 * it will be referenced so the caller must free it.
 */
int
rt_getifa_fib(struct rt_addrinfo *info, u_int fibnum)
{
	struct ifaddr *ifa;
	int error = 0;

	/*
	 * ifp may be specified by sockaddr_dl
	 * when protocol address is ambiguous.
	 */
	if (info->rti_ifp == NULL && ifpaddr != NULL &&
	    ifpaddr->sa_family == AF_LINK &&
	    (ifa = ifa_ifwithnet(ifpaddr)) != NULL) {
		info->rti_ifp = ifa->ifa_ifp;
		ifa_free(ifa);
	}
	if (info->rti_ifa == NULL && ifaaddr != NULL)
		info->rti_ifa = ifa_ifwithaddr(ifaaddr);
	if (info->rti_ifa == NULL) {
		struct sockaddr *sa;

		sa = ifaaddr != NULL ? ifaaddr :
		    (gateway != NULL ? gateway : dst);
		if (sa != NULL && info->rti_ifp != NULL)
			info->rti_ifa = ifaof_ifpforaddr(sa, info->rti_ifp);
		else if (dst != NULL && gateway != NULL)
			info->rti_ifa = ifa_ifwithroute_fib(flags, dst, gateway,
							fibnum);
		else if (sa != NULL)
			info->rti_ifa = ifa_ifwithroute_fib(flags, sa, sa,
							fibnum);
	}
	if ((ifa = info->rti_ifa) != NULL) {
		if (info->rti_ifp == NULL)
			info->rti_ifp = ifa->ifa_ifp;
	} else
		error = ENETUNREACH;
	return (error);
}

/*
 * Expunges references to a route that's about to be reclaimed.
 * The route must be locked.
 */
int
rtexpunge(struct rtentry *rt)
{
	struct radix_node *rn;
	struct radix_node_head *rnh;
	struct ifaddr *ifa;
	int error = 0;

	/*
	 * Find the correct routing tree to use for this Address Family
	 */
	rnh = rt_tables_get_rnh(rt->rt_fibnum, rt_key(rt)->sa_family);
	RT_LOCK_ASSERT(rt);
	if (rnh == NULL)
		return (EAFNOSUPPORT);
	RADIX_NODE_HEAD_LOCK_ASSERT(rnh);
#if 0
	/*
	 * We cannot assume anything about the reference count
	 * because protocols call us in many situations; often
	 * before unwinding references to the table entry.
	 */
	KASSERT(rt->rt_refcnt <= 1, ("bogus refcnt %ld", rt->rt_refcnt));
#endif
	/*
	 * Remove the item from the tree; it should be there,
	 * but when callers invoke us blindly it may not (sigh).
	 */
	rn = rnh->rnh_deladdr(rt_key(rt), rt_mask(rt), rnh);
	if (rn == NULL) {
		error = ESRCH;
		goto bad;
	}
	KASSERT((rn->rn_flags & (RNF_ACTIVE | RNF_ROOT)) == 0,
		("unexpected flags 0x%x", rn->rn_flags));
	KASSERT(rt == RNTORT(rn),
		("lookup mismatch, rt %p rn %p", rt, rn));

	rt->rt_flags &= ~RTF_UP;

	/*
	 * Give the protocol a chance to keep things in sync.
	 */
	if ((ifa = rt->rt_ifa) && ifa->ifa_rtrequest) {
		struct rt_addrinfo info;

		bzero((caddr_t)&info, sizeof(info));
		info.rti_flags = rt->rt_flags;
		info.rti_info[RTAX_DST] = rt_key(rt);
		info.rti_info[RTAX_GATEWAY] = rt->rt_gateway;
		info.rti_info[RTAX_NETMASK] = rt_mask(rt);
		ifa->ifa_rtrequest(RTM_DELETE, rt, &info);
	}

	/*
	 * one more rtentry floating around that is not
	 * linked to the routing table.
	 */
	V_rttrash++;
bad:
	return (error);
}

#ifdef RADIX_MPATH
static int
rn_mpath_update(int req, struct rt_addrinfo *info,
    struct radix_node_head *rnh, struct rtentry **ret_nrt)
{
	/*
	 * if we got multipath routes, we require users to specify
	 * a matching RTAX_GATEWAY.
	 */
	struct rtentry *rt, *rto = NULL;
	register struct radix_node *rn;
	int error = 0;

	rn = rnh->rnh_matchaddr(dst, rnh);
	if (rn == NULL)
		return (ESRCH);
	rto = rt = RNTORT(rn);
	rt = rt_mpath_matchgate(rt, gateway);
	if (rt == NULL)
		return (ESRCH);
	/*
	 * this is the first entry in the chain
	 */
	if (rto == rt) {
		rn = rn_mpath_next((struct radix_node *)rt);
		/*
		 * there is another entry, now it's active
		 */
		if (rn) {
			rto = RNTORT(rn);
			RT_LOCK(rto);
			rto->rt_flags |= RTF_UP;
			RT_UNLOCK(rto);
		} else if (rt->rt_flags & RTF_GATEWAY) {
			/*
			 * For gateway routes, we need to 
			 * make sure that we we are deleting
			 * the correct gateway. 
			 * rt_mpath_matchgate() does not 
			 * check the case when there is only
			 * one route in the chain.  
			 */
			if (gateway &&
			    (rt->rt_gateway->sa_len != gateway->sa_len ||
				memcmp(rt->rt_gateway, gateway, gateway->sa_len)))
				error = ESRCH;
			else {
				/*
				 * remove from tree before returning it
				 * to the caller
				 */
				rn = rnh->rnh_deladdr(dst, netmask, rnh);
				KASSERT(rt == RNTORT(rn), ("radix node disappeared"));
				goto gwdelete;
			}
			
		}
		/*
		 * use the normal delete code to remove
		 * the first entry
		 */
		if (req != RTM_DELETE) 
			goto nondelete;

		error = ENOENT;
		goto done;
	}
		
	/*
	 * if the entry is 2nd and on up
	 */
	if ((req == RTM_DELETE) && !rt_mpath_deldup(rto, rt))
		panic ("rtrequest1: rt_mpath_deldup");
gwdelete:
	RT_LOCK(rt);
	RT_ADDREF(rt);
	if (req == RTM_DELETE) {
		rt->rt_flags &= ~RTF_UP;
		/*
		 * One more rtentry floating around that is not
		 * linked to the routing table. rttrash will be decremented
		 * when RTFREE(rt) is eventually called.
		 */
		V_rttrash++;
	}
	
nondelete:
	if (req != RTM_DELETE)
		panic("unrecognized request %d", req);
	

	/*
	 * If the caller wants it, then it can have it,
	 * but it's up to it to free the rtentry as we won't be
	 * doing it.
	 */
	if (ret_nrt) {
		*ret_nrt = rt;
		RT_UNLOCK(rt);
	} else
		RTFREE_LOCKED(rt);
done:
	return (error);
}
#endif

int
rtrequest1_fib(int req, struct rt_addrinfo *info, struct rtentry **ret_nrt,
				u_int fibnum)
{
	int error = 0, needlock = 0;
	register struct rtentry *rt;
	register struct radix_node *rn;
	register struct radix_node_head *rnh;
	struct ifaddr *ifa;
	struct sockaddr *ndst;
#define senderr(x) { error = x ; goto bad; }

	KASSERT((fibnum < rt_numfibs), ("rtrequest1_fib: bad fibnum"));
	if (dst->sa_family != AF_INET)	/* Only INET supports > 1 fib now */
		fibnum = 0;
	/*
	 * Find the correct routing tree to use for this Address Family
	 */
	rnh = rt_tables_get_rnh(fibnum, dst->sa_family);
	if (rnh == NULL)
		return (EAFNOSUPPORT);
	needlock = ((flags & RTF_RNH_LOCKED) == 0);
	flags &= ~RTF_RNH_LOCKED;
	if (needlock)
		RADIX_NODE_HEAD_LOCK(rnh);
	else
		RADIX_NODE_HEAD_LOCK_ASSERT(rnh);
	/*
	 * If we are adding a host route then we don't want to put
	 * a netmask in the tree, nor do we want to clone it.
	 */
	if (flags & RTF_HOST)
		netmask = NULL;

	switch (req) {
	case RTM_DELETE:
#ifdef RADIX_MPATH
		if (rn_mpath_capable(rnh)) {
			error = rn_mpath_update(req, info, rnh, ret_nrt);
			/*
			 * "bad" holds true for the success case
			 * as well
			 */
			if (error != ENOENT)
				goto bad;
		}
#endif
		/*
		 * Remove the item from the tree and return it.
		 * Complain if it is not there and do no more processing.
		 */
		rn = rnh->rnh_deladdr(dst, netmask, rnh);
		if (rn == NULL)
			senderr(ESRCH);
		if (rn->rn_flags & (RNF_ACTIVE | RNF_ROOT))
			panic ("rtrequest delete");
		rt = RNTORT(rn);
		RT_LOCK(rt);
		RT_ADDREF(rt);
		rt->rt_flags &= ~RTF_UP;

		/*
		 * give the protocol a chance to keep things in sync.
		 */
		if ((ifa = rt->rt_ifa) && ifa->ifa_rtrequest)
			ifa->ifa_rtrequest(RTM_DELETE, rt, info);

		/*
		 * One more rtentry floating around that is not
		 * linked to the routing table. rttrash will be decremented
		 * when RTFREE(rt) is eventually called.
		 */
		V_rttrash++;

		/*
		 * If the caller wants it, then it can have it,
		 * but it's up to it to free the rtentry as we won't be
		 * doing it.
		 */
		if (ret_nrt) {
			*ret_nrt = rt;
			RT_UNLOCK(rt);
		} else
			RTFREE_LOCKED(rt);
		break;
	case RTM_RESOLVE:
		/*
		 * resolve was only used for route cloning
		 * here for compat
		 */
		break;
	case RTM_ADD:
		if ((flags & RTF_GATEWAY) && !gateway)
			senderr(EINVAL);
		if (dst && gateway && (dst->sa_family != gateway->sa_family) && 
		    (gateway->sa_family != AF_UNSPEC) && (gateway->sa_family != AF_LINK))
			senderr(EINVAL);

		if (info->rti_ifa == NULL) {
			error = rt_getifa_fib(info, fibnum);
			if (error)
				senderr(error);
		} else
			ifa_ref(info->rti_ifa);
		ifa = info->rti_ifa;
		rt = uma_zalloc(V_rtzone, M_NOWAIT | M_ZERO);
		if (rt == NULL) {
			if (ifa != NULL)
				ifa_free(ifa);
			senderr(ENOBUFS);
		}
		RT_LOCK_INIT(rt);
		rt->rt_flags = RTF_UP | flags;
		rt->rt_fibnum = fibnum;
		/*
		 * Add the gateway. Possibly re-malloc-ing the storage for it
		 * 
		 */
		RT_LOCK(rt);
		if ((error = rt_setgate(rt, dst, gateway)) != 0) {
			RT_LOCK_DESTROY(rt);
			if (ifa != NULL)
				ifa_free(ifa);
			uma_zfree(V_rtzone, rt);
			senderr(error);
		}

		/*
		 * point to the (possibly newly malloc'd) dest address.
		 */
		ndst = (struct sockaddr *)rt_key(rt);

		/*
		 * make sure it contains the value we want (masked if needed).
		 */
		if (netmask) {
			rt_maskedcopy(dst, ndst, netmask);
		} else
			bcopy(dst, ndst, dst->sa_len);

		/*
		 * We use the ifa reference returned by rt_getifa_fib().
		 * This moved from below so that rnh->rnh_addaddr() can
		 * examine the ifa and  ifa->ifa_ifp if it so desires.
		 */
		rt->rt_ifa = ifa;
		rt->rt_ifp = ifa->ifa_ifp;
		rt->rt_rmx.rmx_weight = 1;

#ifdef RADIX_MPATH
		/* do not permit exactly the same dst/mask/gw pair */
		if (rn_mpath_capable(rnh) &&
			rt_mpath_conflict(rnh, rt, netmask)) {
			if (rt->rt_ifa) {
				ifa_free(rt->rt_ifa);
			}
			Free(rt_key(rt));
			RT_LOCK_DESTROY(rt);
			uma_zfree(V_rtzone, rt);
			senderr(EEXIST);
		}
#endif

		/* XXX mtu manipulation will be done in rnh_addaddr -- itojun */
		rn = rnh->rnh_addaddr(ndst, netmask, rnh, rt->rt_nodes);
		/*
		 * If it still failed to go into the tree,
		 * then un-make it (this should be a function)
		 */
		if (rn == NULL) {
			if (rt->rt_ifa)
				ifa_free(rt->rt_ifa);
			Free(rt_key(rt));
			RT_LOCK_DESTROY(rt);
			uma_zfree(V_rtzone, rt);
			senderr(EEXIST);
		}

		/*
		 * If this protocol has something to add to this then
		 * allow it to do that as well.
		 */
		if (ifa->ifa_rtrequest)
			ifa->ifa_rtrequest(req, rt, info);

		/*
		 * actually return a resultant rtentry and
		 * give the caller a single reference.
		 */
		if (ret_nrt) {
			*ret_nrt = rt;
			RT_ADDREF(rt);
		}
		RT_UNLOCK(rt);
		break;
	default:
		error = EOPNOTSUPP;
	}
bad:
	if (needlock)
		RADIX_NODE_HEAD_UNLOCK(rnh);
	return (error);
#undef senderr
}

#undef dst
#undef gateway
#undef netmask
#undef ifaaddr
#undef ifpaddr
#undef flags

int
rt_setgate(struct rtentry *rt, struct sockaddr *dst, struct sockaddr *gate)
{
	/* XXX dst may be overwritten, can we move this to below */
	int dlen = SA_SIZE(dst), glen = SA_SIZE(gate);
#ifdef INVARIANTS
	struct radix_node_head *rnh;

	rnh = rt_tables_get_rnh(rt->rt_fibnum, dst->sa_family);
#endif

	RT_LOCK_ASSERT(rt);
	RADIX_NODE_HEAD_LOCK_ASSERT(rnh);
	
	/*
	 * Prepare to store the gateway in rt->rt_gateway.
	 * Both dst and gateway are stored one after the other in the same
	 * malloc'd chunk. If we have room, we can reuse the old buffer,
	 * rt_gateway already points to the right place.
	 * Otherwise, malloc a new block and update the 'dst' address.
	 */
	if (rt->rt_gateway == NULL || glen > SA_SIZE(rt->rt_gateway)) {
		caddr_t new;

		R_Malloc(new, caddr_t, dlen + glen);
		if (new == NULL)
			return ENOBUFS;
		/*
		 * XXX note, we copy from *dst and not *rt_key(rt) because
		 * rt_setgate() can be called to initialize a newly
		 * allocated route entry, in which case rt_key(rt) == NULL
		 * (and also rt->rt_gateway == NULL).
		 * Free()/free() handle a NULL argument just fine.
		 */
		bcopy(dst, new, dlen);
		Free(rt_key(rt));	/* free old block, if any */
		rt_key(rt) = (struct sockaddr *)new;
		rt->rt_gateway = (struct sockaddr *)(new + dlen);
	}

	/*
	 * Copy the new gateway value into the memory chunk.
	 */
	bcopy(gate, rt->rt_gateway, glen);

	return (0);
}

static void
rt_maskedcopy(struct sockaddr *src, struct sockaddr *dst, struct sockaddr *netmask)
{
	register u_char *cp1 = (u_char *)src;
	register u_char *cp2 = (u_char *)dst;
	register u_char *cp3 = (u_char *)netmask;
	u_char *cplim = cp2 + *cp3;
	u_char *cplim2 = cp2 + *cp1;

	*cp2++ = *cp1++; *cp2++ = *cp1++; /* copies sa_len & sa_family */
	cp3 += 2;
	if (cplim > cplim2)
		cplim = cplim2;
	while (cp2 < cplim)
		*cp2++ = *cp1++ & *cp3++;
	if (cp2 < cplim2)
		bzero((caddr_t)cp2, (unsigned)(cplim2 - cp2));
}

/*
 * Set up a routing table entry, normally
 * for an interface.
 */
#define _SOCKADDR_TMPSIZE 128 /* Not too big.. kernel stack size is limited */
static inline  int
rtinit1(struct ifaddr *ifa, int cmd, int flags, int fibnum)
{
	struct sockaddr *dst;
	struct sockaddr *netmask;
	struct rtentry *rt = NULL;
	struct rt_addrinfo info;
	int error = 0;
	int startfib, endfib;
	char tempbuf[_SOCKADDR_TMPSIZE];
	int didwork = 0;
	int a_failure = 0;
	static struct sockaddr_dl null_sdl = {sizeof(null_sdl), AF_LINK};

	if (flags & RTF_HOST) {
		dst = ifa->ifa_dstaddr;
		netmask = NULL;
	} else {
		dst = ifa->ifa_addr;
		netmask = ifa->ifa_netmask;
	}
	if ( dst->sa_family != AF_INET)
		fibnum = 0;
	if (fibnum == -1) {
		if (rt_add_addr_allfibs == 0 && cmd == (int)RTM_ADD) {
			startfib = endfib = curthread->td_proc->p_fibnum;
		} else {
			startfib = 0;
			endfib = rt_numfibs - 1;
		}
	} else {
		KASSERT((fibnum < rt_numfibs), ("rtinit1: bad fibnum"));
		startfib = fibnum;
		endfib = fibnum;
	}
	if (dst->sa_len == 0)
		return(EINVAL);

	/*
	 * If it's a delete, check that if it exists,
	 * it's on the correct interface or we might scrub
	 * a route to another ifa which would
	 * be confusing at best and possibly worse.
	 */
	if (cmd == RTM_DELETE) {
		/*
		 * It's a delete, so it should already exist..
		 * If it's a net, mask off the host bits
		 * (Assuming we have a mask)
		 * XXX this is kinda inet specific..
		 */
		if (netmask != NULL) {
			rt_maskedcopy(dst, (struct sockaddr *)tempbuf, netmask);
			dst = (struct sockaddr *)tempbuf;
		}
	}
	/*
	 * Now go through all the requested tables (fibs) and do the
	 * requested action. Realistically, this will either be fib 0
	 * for protocols that don't do multiple tables or all the
	 * tables for those that do. XXX For this version only AF_INET.
	 * When that changes code should be refactored to protocol
	 * independent parts and protocol dependent parts.
	 */
	for ( fibnum = startfib; fibnum <= endfib; fibnum++) {
		if (cmd == RTM_DELETE) {
			struct radix_node_head *rnh;
			struct radix_node *rn;
			/*
			 * Look up an rtentry that is in the routing tree and
			 * contains the correct info.
			 */
			rnh = rt_tables_get_rnh(fibnum, dst->sa_family);
			if (rnh == NULL)
				/* this table doesn't exist but others might */
				continue;
			RADIX_NODE_HEAD_LOCK(rnh);
#ifdef RADIX_MPATH
			if (rn_mpath_capable(rnh)) {

				rn = rnh->rnh_matchaddr(dst, rnh);
				if (rn == NULL) 
					error = ESRCH;
				else {
					rt = RNTORT(rn);
					/*
					 * for interface route the
					 * rt->rt_gateway is sockaddr_intf
					 * for cloning ARP entries, so
					 * rt_mpath_matchgate must use the
					 * interface address
					 */
					rt = rt_mpath_matchgate(rt,
					    ifa->ifa_addr);
					if (!rt) 
						error = ESRCH;
				}
			}
			else
#endif
			rn = rnh->rnh_lookup(dst, netmask, rnh);
			error = (rn == NULL ||
			    (rn->rn_flags & RNF_ROOT) ||
			    RNTORT(rn)->rt_ifa != ifa ||
			    !sa_equal((struct sockaddr *)rn->rn_key, dst));
			RADIX_NODE_HEAD_UNLOCK(rnh);
			if (error) {
				/* this is only an error if bad on ALL tables */
				continue;
			}
		}
		/*
		 * Do the actual request
		 */
		bzero((caddr_t)&info, sizeof(info));
		info.rti_ifa = ifa;
		info.rti_flags = flags | ifa->ifa_flags;
		info.rti_info[RTAX_DST] = dst;
		/* 
		 * doing this for compatibility reasons
		 */
		if (cmd == RTM_ADD)
			info.rti_info[RTAX_GATEWAY] =
			    (struct sockaddr *)&null_sdl;
		else
			info.rti_info[RTAX_GATEWAY] = ifa->ifa_addr;
		info.rti_info[RTAX_NETMASK] = netmask;
		error = rtrequest1_fib(cmd, &info, &rt, fibnum);
		if (error == 0 && rt != NULL) {
			/*
			 * notify any listening routing agents of the change
			 */
			RT_LOCK(rt);
#ifdef RADIX_MPATH
			/*
			 * in case address alias finds the first address
			 * e.g. ifconfig bge0 192.103.54.246/24
			 * e.g. ifconfig bge0 192.103.54.247/24
			 * the address set in the route is 192.103.54.246
			 * so we need to replace it with 192.103.54.247
			 */
			if (memcmp(rt->rt_ifa->ifa_addr,
			    ifa->ifa_addr, ifa->ifa_addr->sa_len)) {
				ifa_free(rt->rt_ifa);
				ifa_ref(ifa);
				rt->rt_ifp = ifa->ifa_ifp;
				rt->rt_ifa = ifa;
			}
#endif
			/* 
			 * doing this for compatibility reasons
			 */
			if (cmd == RTM_ADD) {
			    ((struct sockaddr_dl *)rt->rt_gateway)->sdl_type  =
				rt->rt_ifp->if_type;
			    ((struct sockaddr_dl *)rt->rt_gateway)->sdl_index =
				rt->rt_ifp->if_index;
			}
			rt_newaddrmsg(cmd, ifa, error, rt);
			if (cmd == RTM_DELETE) {
				/*
				 * If we are deleting, and we found an entry,
				 * then it's been removed from the tree..
				 * now throw it away.
				 */
				RTFREE_LOCKED(rt);
			} else {
				if (cmd == RTM_ADD) {
					/*
					 * We just wanted to add it..
					 * we don't actually need a reference.
					 */
					RT_REMREF(rt);
				}
				RT_UNLOCK(rt);
			}
			didwork = 1;
		}
		if (error)
			a_failure = error;
	}
	if (cmd == RTM_DELETE) {
		if (didwork) {
			error = 0;
		} else {
			/* we only give an error if it wasn't in any table */
			error = ((flags & RTF_HOST) ?
			    EHOSTUNREACH : ENETUNREACH);
		}
	} else {
		if (a_failure) {
			/* return an error if any of them failed */
			error = a_failure;
		}
	}
	return (error);
}

/* special one for inet internal use. may not use. */
int
rtinit_fib(struct ifaddr *ifa, int cmd, int flags)
{
	return (rtinit1(ifa, cmd, flags, -1));
}

/*
 * Set up a routing table entry, normally
 * for an interface.
 */
int
rtinit(struct ifaddr *ifa, int cmd, int flags)
{
	struct sockaddr *dst;
	int fib = 0;

	if (flags & RTF_HOST) {
		dst = ifa->ifa_dstaddr;
	} else {
		dst = ifa->ifa_addr;
	}

	if (dst->sa_family == AF_INET)
		fib = -1;
	return (rtinit1(ifa, cmd, flags, fib));
}

/* This must be before ip6_init2(), which is now SI_ORDER_MIDDLE */
SYSINIT(route, SI_SUB_PROTO_DOMAIN, SI_ORDER_THIRD, route_init, 0);
OpenPOWER on IntegriCloud