summaryrefslogtreecommitdiffstats
path: root/sys/net/route.c
blob: adb28a0393e09023be7853bf57bfd413cfb5edeb (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
/*-
 * 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
 * $FreeBSD$
 */

#include "opt_inet.h"
#include "opt_mrouting.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/socket.h>
#include <sys/domain.h>
#include <sys/kernel.h>

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

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

#include <vm/uma.h>

static struct rtstat rtstat;
struct radix_node_head *rt_tables[AF_MAX+1];

static int	rttrash;		/* routes not in table but not freed */

static void rt_maskedcopy(struct sockaddr *,
	    struct sockaddr *, struct sockaddr *);
static void rtable_init(void **);

/* 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))

static void
rtable_init(void **table)
{
	struct domain *dom;
	for (dom = domains; dom; dom = dom->dom_next)
		if (dom->dom_rtattach)
			dom->dom_rtattach(&table[dom->dom_family],
			    dom->dom_rtoffset);
}

static uma_zone_t rtzone;		/* Routing table UMA zone. */

static void
route_init(void)
{
	rtzone = uma_zcreate("rtentry", sizeof(struct rtentry), NULL, NULL,
	    NULL, NULL, UMA_ALIGN_PTR, 0);
	rn_init();	/* initialize all zeroes, all ones, mask table */
	rtable_init((void **)rt_tables);
}

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

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(&ro->ro_dst, 1, ignore);
	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)
{
	struct radix_node_head *rnh = rt_tables[dst->sa_family];
	struct rtentry *rt;
	struct radix_node *rn;
	struct rtentry *newrt;
	struct rt_addrinfo info;
	u_long nflags;
	int err = 0, msgtype = RTM_MISS;

	newrt = NULL;
	bzero(&info, sizeof(info));
	/*
	 * Look up the address in the table for that Address Family
	 */
	if (rnh == NULL) {
		rtstat.rts_unreach++;
		goto miss2;
	}
	RADIX_NODE_HEAD_LOCK(rnh);
	if ((rn = rnh->rnh_matchaddr(dst, rnh)) &&
	    (rn->rn_flags & RNF_ROOT) == 0) {
		/*
		 * If we find it and it's not the root node, then
		 * get a refernce on the rtentry associated.
		 */
		newrt = rt = RNTORT(rn);
		nflags = rt->rt_flags & ~ignflags;
		if (report && (nflags & RTF_CLONING)) {
			/*
			 * We are apparently adding (report = 0 in delete).
			 * If it requires that it be cloned, do so.
			 * (This implies it wasn't a HOST route.)
			 */
			err = rtrequest(RTM_RESOLVE, dst, NULL,
					      NULL, 0, &newrt);
			if (err) {
				/*
				 * If the cloning didn't succeed, maybe
				 * what we have will do. Return that.
				 */
				newrt = rt;		/* existing route */
				RT_LOCK(newrt);
				RT_ADDREF(newrt);
				goto miss;
			}
			KASSERT(newrt, ("no route and no error"));
			RT_LOCK(newrt);
			if (newrt->rt_flags & RTF_XRESOLVE) {
				/*
				 * If the new route specifies it be
				 * externally resolved, then go do that.
				 */
				msgtype = RTM_RESOLVE;
				goto miss;
			}
			/* Inform listeners of the new route. */
			info.rti_info[RTAX_DST] = rt_key(newrt);
			info.rti_info[RTAX_NETMASK] = rt_mask(newrt);
			info.rti_info[RTAX_GATEWAY] = newrt->rt_gateway;
			if (newrt->rt_ifp != NULL) {
				info.rti_info[RTAX_IFP] =
				    newrt->rt_ifp->if_addr->ifa_addr;
				info.rti_info[RTAX_IFA] = newrt->rt_ifa->ifa_addr;
			}
			rt_missmsg(RTM_ADD, &info, newrt->rt_flags, 0);
		} else {
			KASSERT(rt == newrt, ("locking wrong route"));
			RT_LOCK(newrt);
			RT_ADDREF(newrt);
		}
		RADIX_NODE_HEAD_UNLOCK(rnh);
	} else {
		/*
		 * Either we hit the root or couldn't find any match,
		 * Which basically means
		 * "caint get there frm here"
		 */
		rtstat.rts_unreach++;
	miss:
		RADIX_NODE_HEAD_UNLOCK(rnh);
	miss2:	if (report) {
			/*
			 * If required, report the failure to the supervising
			 * Authorities.
			 * For a delete, this is not an error. (report == 0)
			 */
			info.rti_info[RTAX_DST] = dst;
			rt_missmsg(msgtype, &info, 0, err);
		}
	}
	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;

	/* XXX the NULL checks are probably useless */
	if (rt == NULL)
		panic("rtfree: NULL rt");
	rnh = rt_tables[rt_key(rt)->sa_family];
	if (rnh == NULL)
		panic("rtfree: NULL rnh");

	RT_LOCK_ASSERT(rt);

	/*
	 * decrement the reference count by one and if it reaches 0,
	 * and there is a close function defined, call the close function
	 */
	RT_REMREF(rt);
	if (rt->rt_refcnt > 0)
		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.
		 */
		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)
			IFAFREE(rt->rt_ifa);
		rt->rt_parent = NULL;		/* NB: no refcnt on parent */

		/*
		 * 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(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)
{
	struct rtentry *rt;
	int error = 0;
	short *stat = NULL;
	struct rt_addrinfo info;
	struct ifaddr *ifa;

	/* verify the gateway is directly reachable */
	if ((ifa = ifa_ifwithnet(gateway)) == NULL) {
		error = ENETUNREACH;
		goto out;
	}
	rt = rtalloc1(dst, 0, 0UL);	/* 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(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:
			if (rt)
				rtfree(rt);
			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;
			rt = NULL;
			error = rtrequest1(RTM_ADD, &info, &rt);
			if (rt != NULL) {
				RT_LOCK(rt);
				flags = rt->rt_flags;
			}
			stat = &rtstat.rts_dynamic;
		} else {
			/*
			 * Smash the current notion of the gateway to
			 * this destination.  Should check about netmask!!!
			 */
			rt->rt_flags |= RTF_MODIFIED;
			flags |= RTF_MODIFIED;
			stat = &rtstat.rts_newgateway;
			/*
			 * add the key and gateway (in one malloc'd chunk).
			 */
			rt_setgate(rt, rt_key(rt), gateway);
		}
	} else
		error = EHOSTUNREACH;
done:
	if (rt)
		rtfree(rt);
out:
	if (error)
		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);
}

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

	/*
	 * 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) : EOPNOTSUPP;
#else /* INET */
	return ENXIO;
#endif /* INET */
}

struct ifaddr *
ifa_ifwithroute(int flags, struct sockaddr *dst, struct sockaddr *gateway)
{
	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(gateway, 0, 0UL);
		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;
		}
		RT_REMREF(rt);
		RT_UNLOCK(rt);
		if (not_found)
			return (NULL);
		if ((ifa = rt->rt_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;
	}
	return (ifa);
}

static walktree_f_t rt_fixdelete;
static walktree_f_t rt_fixchange;

struct rtfc_arg {
	struct rtentry *rt0;
	struct radix_node_head *rnh;
};

/*
 * 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)
{
	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(req, &info, ret_nrt);
}

/*
 * 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)
{
	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;
	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(flags, dst, gateway);
		else if (sa != NULL)
			info->rti_ifa = ifa_ifwithroute(flags, sa, sa);
	}
	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;

	RT_LOCK_ASSERT(rt);
#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
	/*
	 * Find the correct routing tree to use for this Address Family
	 */
	rnh = rt_tables[rt_key(rt)->sa_family];
	if (rnh == NULL)
		return (EAFNOSUPPORT);

	RADIX_NODE_HEAD_LOCK(rnh);

	/*
	 * 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;

	/*
	 * Now search what's left of the subtree for any cloned
	 * routes which might have been formed from this node.
	 */
	if ((rt->rt_flags & RTF_CLONING) && rt_mask(rt))
		rnh->rnh_walktree_from(rnh, rt_key(rt), rt_mask(rt),
				       rt_fixdelete, rt);

	/*
	 * Remove any external references we may have.
	 * This might result in another rtentry being freed if
	 * we held its last reference.
	 */
	if (rt->rt_gwroute) {
		RTFREE(rt->rt_gwroute);
		rt->rt_gwroute = NULL;
	}

	/*
	 * 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.
	 */
	rttrash++;
bad:
	RADIX_NODE_HEAD_UNLOCK(rnh);
	return (error);
}

int
rtrequest1(int req, struct rt_addrinfo *info, struct rtentry **ret_nrt)
{
	int error = 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; }

	/*
	 * Find the correct routing tree to use for this Address Family
	 */
	rnh = rt_tables[dst->sa_family];
	if (rnh == NULL)
		return (EAFNOSUPPORT);
	RADIX_NODE_HEAD_LOCK(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;
		flags &= ~RTF_CLONING;
	}
	switch (req) {
	case RTM_DELETE:
		/*
		 * 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;

		/*
		 * Now search what's left of the subtree for any cloned
		 * routes which might have been formed from this node.
		 */
		if ((rt->rt_flags & RTF_CLONING) &&
		    rt_mask(rt)) {
			rnh->rnh_walktree_from(rnh, dst, rt_mask(rt),
					       rt_fixdelete, rt);
		}

		/*
		 * Remove any external references we may have.
		 * This might result in another rtentry being freed if
		 * we held its last reference.
		 */
		if (rt->rt_gwroute) {
			RTFREE(rt->rt_gwroute);
			rt->rt_gwroute = NULL;
		}

		/*
		 * 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.
		 */
		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:
		if (ret_nrt == NULL || (rt = *ret_nrt) == NULL)
			senderr(EINVAL);
		ifa = rt->rt_ifa;
		/* XXX locking? */
		flags = rt->rt_flags &
		    ~(RTF_CLONING | RTF_STATIC);
		flags |= RTF_WASCLONED;
		gateway = rt->rt_gateway;
		if ((netmask = rt->rt_genmask) == NULL)
			flags |= RTF_HOST;
		goto makeroute;

	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(info)))
			senderr(error);
		ifa = info->rti_ifa;

	makeroute:
		rt = uma_zalloc(rtzone, M_NOWAIT | M_ZERO);
		if (rt == NULL)
			senderr(ENOBUFS);
		RT_LOCK_INIT(rt);
		rt->rt_flags = RTF_UP | flags;
		/*
		 * Add the gateway. Possibly re-malloc-ing the storage for it
		 * also add the rt_gwroute if possible.
		 */
		RT_LOCK(rt);
		if ((error = rt_setgate(rt, dst, gateway)) != 0) {
			RT_LOCK_DESTROY(rt);
			uma_zfree(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);

		/*
		 * Note that we now have a reference to the ifa.
		 * This moved from below so that rnh->rnh_addaddr() can
		 * examine the ifa and  ifa->ifa_ifp if it so desires.
		 */
		IFAREF(ifa);
		rt->rt_ifa = ifa;
		rt->rt_ifp = ifa->ifa_ifp;

		/* XXX mtu manipulation will be done in rnh_addaddr -- itojun */
		rn = rnh->rnh_addaddr(ndst, netmask, rnh, rt->rt_nodes);
		if (rn == NULL) {
			struct rtentry *rt2;
			/*
			 * Uh-oh, we already have one of these in the tree.
			 * We do a special hack: if the route that's already
			 * there was generated by the cloning mechanism
			 * then we just blow it away and retry the insertion
			 * of the new one.
			 */
			rt2 = rtalloc1(dst, 0, 0);
			if (rt2 && rt2->rt_parent) {
				rtexpunge(rt2);
				RT_UNLOCK(rt2);
				rn = rnh->rnh_addaddr(ndst, netmask,
						      rnh, rt->rt_nodes);
			} else if (rt2) {
				/* undo the extra ref we got */
				RTFREE_LOCKED(rt2);
			}
		}

		/*
		 * If it still failed to go into the tree,
		 * then un-make it (this should be a function)
		 */
		if (rn == NULL) {
			if (rt->rt_gwroute)
				RTFREE(rt->rt_gwroute);
			if (rt->rt_ifa)
				IFAFREE(rt->rt_ifa);
			Free(rt_key(rt));
			RT_LOCK_DESTROY(rt);
			uma_zfree(rtzone, rt);
			senderr(EEXIST);
		}

		rt->rt_parent = NULL;

		/*
		 * If we got here from RESOLVE, then we are cloning
		 * so clone the rest, and note that we
		 * are a clone (and increment the parent's references)
		 */
		if (req == RTM_RESOLVE) {
			KASSERT(ret_nrt && *ret_nrt,
				("no route to clone from"));
			rt->rt_rmx = (*ret_nrt)->rt_rmx; /* copy metrics */
			rt->rt_rmx.rmx_pksent = 0; /* reset packet counter */
			if ((*ret_nrt)->rt_flags & RTF_CLONING) {
				/*
				 * NB: We do not bump the refcnt on the parent
				 * entry under the assumption that it will
				 * remain so long as we do.  This is
				 * important when deleting the parent route
				 * as this operation requires traversing
				 * the tree to delete all clones and futzing
				 * with refcnts requires us to double-lock
				 * parent through this back reference.
				 */
				rt->rt_parent = *ret_nrt;
			}
		}

		/*
		 * 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);

		/*
		 * We repeat the same procedure from rt_setgate() here because
		 * it doesn't fire when we call it there because the node
		 * hasn't been added to the tree yet.
		 */
		if (req == RTM_ADD &&
		    !(rt->rt_flags & RTF_HOST) && rt_mask(rt) != NULL) {
			struct rtfc_arg arg;
			arg.rnh = rnh;
			arg.rt0 = rt;
			rnh->rnh_walktree_from(rnh, rt_key(rt), rt_mask(rt),
					       rt_fixchange, &arg);
		}

		/*
		 * 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:
	RADIX_NODE_HEAD_UNLOCK(rnh);
	return (error);
#undef senderr
}

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

/*
 * Called from rtrequest(RTM_DELETE, ...) to fix up the route's ``family''
 * (i.e., the routes related to it by the operation of cloning).  This
 * routine is iterated over all potential former-child-routes by way of
 * rnh->rnh_walktree_from() above, and those that actually are children of
 * the late parent (passed in as VP here) are themselves deleted.
 */
static int
rt_fixdelete(struct radix_node *rn, void *vp)
{
	struct rtentry *rt = RNTORT(rn);
	struct rtentry *rt0 = vp;

	if (rt->rt_parent == rt0 &&
	    !(rt->rt_flags & (RTF_PINNED | RTF_CLONING))) {
		return rtrequest(RTM_DELETE, rt_key(rt), NULL, rt_mask(rt),
				 rt->rt_flags, NULL);
	}
	return 0;
}

/*
 * This routine is called from rt_setgate() to do the analogous thing for
 * adds and changes.  There is the added complication in this case of a
 * middle insert; i.e., insertion of a new network route between an older
 * network route and (cloned) host routes.  For this reason, a simple check
 * of rt->rt_parent is insufficient; each candidate route must be tested
 * against the (mask, value) of the new route (passed as before in vp)
 * to see if the new route matches it.
 *
 * XXX - it may be possible to do fixdelete() for changes and reserve this
 * routine just for adds.  I'm not sure why I thought it was necessary to do
 * changes this way.
 */

static int
rt_fixchange(struct radix_node *rn, void *vp)
{
	struct rtentry *rt = RNTORT(rn);
	struct rtfc_arg *ap = vp;
	struct rtentry *rt0 = ap->rt0;
	struct radix_node_head *rnh = ap->rnh;
	u_char *xk1, *xm1, *xk2, *xmp;
	int i, len, mlen;

	/* make sure we have a parent, and route is not pinned or cloning */
	if (!rt->rt_parent ||
	    (rt->rt_flags & (RTF_PINNED | RTF_CLONING)))
		return 0;

	if (rt->rt_parent == rt0)	/* parent match */
		goto delete_rt;
	/*
	 * There probably is a function somewhere which does this...
	 * if not, there should be.
	 */
	len = imin(rt_key(rt0)->sa_len, rt_key(rt)->sa_len);

	xk1 = (u_char *)rt_key(rt0);
	xm1 = (u_char *)rt_mask(rt0);
	xk2 = (u_char *)rt_key(rt);

	/* avoid applying a less specific route */
	xmp = (u_char *)rt_mask(rt->rt_parent);
	mlen = rt_key(rt->rt_parent)->sa_len;
	if (mlen > rt_key(rt0)->sa_len)		/* less specific route */
		return 0;
	for (i = rnh->rnh_treetop->rn_offset; i < mlen; i++)
		if ((xmp[i] & ~(xmp[i] ^ xm1[i])) != xmp[i])
			return 0;	/* less specific route */

	for (i = rnh->rnh_treetop->rn_offset; i < len; i++)
		if ((xk2[i] & xm1[i]) != xk1[i])
			return 0;	/* no match */

	/*
	 * OK, this node is a clone, and matches the node currently being
	 * changed/added under the node's mask.  So, get rid of it.
	 */
delete_rt:
	return rtrequest(RTM_DELETE, rt_key(rt), NULL,
			 rt_mask(rt), rt->rt_flags, NULL);
}

int
rt_setgate(struct rtentry *rt, struct sockaddr *dst, struct sockaddr *gate)
{
	/* XXX dst may be overwritten, can we move this to below */
	struct radix_node_head *rnh = rt_tables[dst->sa_family];
	int dlen = SA_SIZE(dst), glen = SA_SIZE(gate);

	RT_LOCK_ASSERT(rt);

	/*
	 * A host route with the destination equal to the gateway
	 * will interfere with keeping LLINFO in the routing
	 * table, so disallow it.
	 */
	if (((rt->rt_flags & (RTF_HOST|RTF_GATEWAY|RTF_LLINFO)) ==
					(RTF_HOST|RTF_GATEWAY)) &&
	    dst->sa_len == gate->sa_len &&
	    bcmp(dst, gate, dst->sa_len) == 0) {
		/*
		 * The route might already exist if this is an RTM_CHANGE
		 * or a routing redirect, so try to delete it.
		 */
		if (rt_key(rt))
			rtexpunge(rt);
		return EADDRNOTAVAIL;
	}

	/*
	 * Cloning loop avoidance in case of bad configuration.
	 */
	if (rt->rt_flags & RTF_GATEWAY) {
		struct rtentry *gwrt;

		RT_UNLOCK(rt);		/* XXX workaround LOR */
		gwrt = rtalloc1(gate, 1, 0);
		if (gwrt == rt) {
			RT_LOCK_ASSERT(rt);
			RT_REMREF(rt);
			return (EADDRINUSE); /* failure */
		}
		RT_LOCK(rt);
		/*
		 * If there is already a gwroute, then drop it. If we
		 * are asked to replace route with itself, then do
		 * not leak its refcounter.
		 */
		if (rt->rt_gwroute != NULL) {
			if (rt->rt_gwroute == gwrt) {
				RT_REMREF(rt->rt_gwroute);
			} else
				RTFREE(rt->rt_gwroute);
		}

		if ((rt->rt_gwroute = gwrt) != NULL)
			RT_UNLOCK(rt->rt_gwroute);
	}

	/*
	 * 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);

	/*
	 * This isn't going to do anything useful for host routes, so
	 * don't bother.  Also make sure we have a reasonable mask
	 * (we don't yet have one during adds).
	 */
	if (!(rt->rt_flags & RTF_HOST) && rt_mask(rt) != 0) {
		struct rtfc_arg arg;

		arg.rnh = rnh;
		arg.rt0 = rt;
		RT_UNLOCK(rt);		/* XXX workaround LOR */
		RADIX_NODE_HEAD_LOCK(rnh);
		RT_LOCK(rt);
		rnh->rnh_walktree_from(rnh, rt_key(rt), rt_mask(rt),
				       rt_fixchange, &arg);
		RADIX_NODE_HEAD_UNLOCK(rnh);
	}

	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.
 */
int
rtinit(struct ifaddr *ifa, int cmd, int flags)
{
	struct sockaddr *dst;
	struct sockaddr *netmask;
	struct mbuf *m = NULL;
	struct rtentry *rt = NULL;
	struct rt_addrinfo info;
	int error;

	if (flags & RTF_HOST) {
		dst = ifa->ifa_dstaddr;
		netmask = NULL;
	} else {
		dst = ifa->ifa_addr;
		netmask = ifa->ifa_netmask;
	}
	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) {
		struct sockaddr *deldst;
		struct radix_node_head *rnh;
		struct radix_node *rn;

		/*
		 * It's a delete, so it should already exist..
		 * If it's a net, mask off the host bits
		 * (Assuming we have a mask)
		 */
		if (netmask != NULL) {
			m = m_get(M_DONTWAIT, MT_SONAME);
			if (m == NULL)
				return(ENOBUFS);
			deldst = mtod(m, struct sockaddr *);
			rt_maskedcopy(dst, deldst, netmask);
			dst = deldst;
		}
		/*
		 * Look up an rtentry that is in the routing tree and
		 * contains the correct info.
		 */
		if ((rnh = rt_tables[dst->sa_family]) == NULL)
			goto bad;
		RADIX_NODE_HEAD_LOCK(rnh);
		error = ((rn = rnh->rnh_lookup(dst, netmask, rnh)) == 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) {
bad:
			if (m)
				(void) m_free(m);
			return (flags & RTF_HOST ? EHOSTUNREACH : ENETUNREACH);
		}
	}
	/*
	 * 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;
	info.rti_info[RTAX_GATEWAY] = ifa->ifa_addr;
	info.rti_info[RTAX_NETMASK] = netmask;
	error = rtrequest1(cmd, &info, &rt);
	if (error == 0 && rt != NULL) {
		/*
		 * notify any listening routing agents of the change
		 */
		RT_LOCK(rt);
		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);
		}
	}
	if (m)
		(void) m_free(m);
	return (error);
}

/*
 * rt_check() is invoked on each layer 2 output path, prior to
 * encapsulating outbound packets.
 *
 * The function is mostly used to find a routing entry for the gateway,
 * which in some protocol families could also point to the link-level
 * address for the gateway itself (the side effect of revalidating the
 * route to the destination is rather pointless at this stage, we did it
 * already a moment before in the pr_output() routine to locate the ifp
 * and gateway to use).
 *
 * When we remove the layer-3 to layer-2 mapping tables from the
 * routing table, this function can be removed.
 *
 * === On input ===
 *   *dst is the address of the NEXT HOP (which coincides with the
 *	final destination if directly reachable);
 *   *lrt0 points to the cached route to the final destination;
 *   *lrt is not meaningful;
 *
 * === Operation ===
 * If the route is marked down try to find a new route.  If the route
 * to the gateway is gone, try to setup a new route.  Otherwise,
 * if the route is marked for packets to be rejected, enforce that.
 *
 * === On return ===
 *   *dst is unchanged;
 *   *lrt0 points to the (possibly new) route to the final destination
 *   *lrt points to the route to the next hop
 *
 * Their values are meaningful ONLY if no error is returned.
 */
int
rt_check(struct rtentry **lrt, struct rtentry **lrt0, struct sockaddr *dst)
{
#define senderr(x) { error = x ; goto bad; }
	struct rtentry *rt;
	struct rtentry *rt0;
	int error;

	KASSERT(*lrt0 != NULL, ("rt_check"));
	rt = rt0 = *lrt0;

	/* NB: the locking here is tortuous... */
	RT_LOCK(rt);
	if ((rt->rt_flags & RTF_UP) == 0) {
		RT_UNLOCK(rt);
		rt = rtalloc1(dst, 1, 0UL);
		if (rt != NULL) {
			RT_REMREF(rt);
			/* XXX what about if change? */
		} else
			senderr(EHOSTUNREACH);
		rt0 = rt;
	}
	/* XXX BSD/OS checks dst->sa_family != AF_NS */
	if (rt->rt_flags & RTF_GATEWAY) {
		if (rt->rt_gwroute == NULL)
			goto lookup;
		rt = rt->rt_gwroute;
		RT_LOCK(rt);		/* NB: gwroute */
		if ((rt->rt_flags & RTF_UP) == 0) {
			rtfree(rt);	/* unlock gwroute */
			rt = rt0;
		lookup:
			RT_UNLOCK(rt0);
			rt = rtalloc1(rt->rt_gateway, 1, 0UL);
			if (rt == rt0) {
				rt0->rt_gwroute = NULL;
				RT_REMREF(rt0);
				RT_UNLOCK(rt0);
				senderr(ENETUNREACH);
			}
			RT_LOCK(rt0);
			rt0->rt_gwroute = rt;
			if (rt == NULL) {
				RT_UNLOCK(rt0);
				senderr(EHOSTUNREACH);
			}
		}
		RT_UNLOCK(rt0);
	}
	/* XXX why are we inspecting rmx_expire? */
	error = (rt->rt_flags & RTF_REJECT) &&
		(rt->rt_rmx.rmx_expire == 0 ||
			time_uptime < rt->rt_rmx.rmx_expire);
	if (error) {
		RT_UNLOCK(rt);
		senderr(rt == rt0 ? EHOSTDOWN : EHOSTUNREACH);
	}

	*lrt = rt;
	*lrt0 = rt0;
	return (0);
bad:
	/* NB: lrt and lrt0 should not be interpreted if error is non-zero */
	return (error);
#undef senderr
}

/* 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