summaryrefslogtreecommitdiffstats
path: root/net/decnet/dn_route.c
blob: ef94ca56d7bda0de8b46d03b65894c479a207968 (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
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
/*
 * DECnet       An implementation of the DECnet protocol suite for the LINUX
 *              operating system.  DECnet is implemented using the  BSD Socket
 *              interface as the means of communication with the user level.
 *
 *              DECnet Routing Functions (Endnode and Router)
 *
 * Authors:     Steve Whitehouse <SteveW@ACM.org>
 *              Eduardo Marcelo Serrat <emserrat@geocities.com>
 *
 * Changes:
 *              Steve Whitehouse : Fixes to allow "intra-ethernet" and
 *                                 "return-to-sender" bits on outgoing
 *                                 packets.
 *		Steve Whitehouse : Timeouts for cached routes.
 *              Steve Whitehouse : Use dst cache for input routes too.
 *              Steve Whitehouse : Fixed error values in dn_send_skb.
 *              Steve Whitehouse : Rework routing functions to better fit
 *                                 DECnet routing design
 *              Alexey Kuznetsov : New SMP locking
 *              Steve Whitehouse : More SMP locking changes & dn_cache_dump()
 *              Steve Whitehouse : Prerouting NF hook, now really is prerouting.
 *				   Fixed possible skb leak in rtnetlink funcs.
 *              Steve Whitehouse : Dave Miller's dynamic hash table sizing and
 *                                 Alexey Kuznetsov's finer grained locking
 *                                 from ipv4/route.c.
 *              Steve Whitehouse : Routing is now starting to look like a
 *                                 sensible set of code now, mainly due to
 *                                 my copying the IPv4 routing code. The
 *                                 hooks here are modified and will continue
 *                                 to evolve for a while.
 *              Steve Whitehouse : Real SMP at last :-) Also new netfilter
 *                                 stuff. Look out raw sockets your days
 *                                 are numbered!
 *              Steve Whitehouse : Added return-to-sender functions. Added
 *                                 backlog congestion level return codes.
 *		Steve Whitehouse : Fixed bug where routes were set up with
 *                                 no ref count on net devices.
 *              Steve Whitehouse : RCU for the route cache
 *              Steve Whitehouse : Preparations for the flow cache
 *              Steve Whitehouse : Prepare for nonlinear skbs
 */

/******************************************************************************
    (c) 1995-1998 E.M. Serrat		emserrat@geocities.com

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.
*******************************************************************************/

#include <linux/errno.h>
#include <linux/types.h>
#include <linux/socket.h>
#include <linux/in.h>
#include <linux/kernel.h>
#include <linux/sockios.h>
#include <linux/net.h>
#include <linux/netdevice.h>
#include <linux/inet.h>
#include <linux/route.h>
#include <linux/in_route.h>
#include <net/sock.h>
#include <linux/mm.h>
#include <linux/proc_fs.h>
#include <linux/seq_file.h>
#include <linux/init.h>
#include <linux/rtnetlink.h>
#include <linux/string.h>
#include <linux/netfilter_decnet.h>
#include <linux/rcupdate.h>
#include <linux/times.h>
#include <asm/errno.h>
#include <net/neighbour.h>
#include <net/dst.h>
#include <net/flow.h>
#include <net/fib_rules.h>
#include <net/dn.h>
#include <net/dn_dev.h>
#include <net/dn_nsp.h>
#include <net/dn_route.h>
#include <net/dn_neigh.h>
#include <net/dn_fib.h>

struct dn_rt_hash_bucket
{
	struct dn_route *chain;
	spinlock_t lock;
} __attribute__((__aligned__(8)));

extern struct neigh_table dn_neigh_table;


static unsigned char dn_hiord_addr[6] = {0xAA,0x00,0x04,0x00,0x00,0x00};

static const int dn_rt_min_delay = 2 * HZ;
static const int dn_rt_max_delay = 10 * HZ;
static const int dn_rt_mtu_expires = 10 * 60 * HZ;

static unsigned long dn_rt_deadline;

static int dn_dst_gc(void);
static struct dst_entry *dn_dst_check(struct dst_entry *, __u32);
static struct dst_entry *dn_dst_negative_advice(struct dst_entry *);
static void dn_dst_link_failure(struct sk_buff *);
static void dn_dst_update_pmtu(struct dst_entry *dst, u32 mtu);
static int dn_route_input(struct sk_buff *);
static void dn_run_flush(unsigned long dummy);

static struct dn_rt_hash_bucket *dn_rt_hash_table;
static unsigned dn_rt_hash_mask;

static struct timer_list dn_route_timer;
static DEFINE_TIMER(dn_rt_flush_timer, dn_run_flush, 0, 0);
int decnet_dst_gc_interval = 2;

static struct dst_ops dn_dst_ops = {
	.family =		PF_DECnet,
	.protocol =		__constant_htons(ETH_P_DNA_RT),
	.gc_thresh =		128,
	.gc =			dn_dst_gc,
	.check =		dn_dst_check,
	.negative_advice =	dn_dst_negative_advice,
	.link_failure =		dn_dst_link_failure,
	.update_pmtu =		dn_dst_update_pmtu,
	.entry_size =		sizeof(struct dn_route),
	.entries =		ATOMIC_INIT(0),
};

static __inline__ unsigned dn_hash(__le16 src, __le16 dst)
{
	__u16 tmp = (__u16 __force)(src ^ dst);
	tmp ^= (tmp >> 3);
	tmp ^= (tmp >> 5);
	tmp ^= (tmp >> 10);
	return dn_rt_hash_mask & (unsigned)tmp;
}

static inline void dnrt_free(struct dn_route *rt)
{
	call_rcu_bh(&rt->u.dst.rcu_head, dst_rcu_free);
}

static inline void dnrt_drop(struct dn_route *rt)
{
	dst_release(&rt->u.dst);
	call_rcu_bh(&rt->u.dst.rcu_head, dst_rcu_free);
}

static void dn_dst_check_expire(unsigned long dummy)
{
	int i;
	struct dn_route *rt, **rtp;
	unsigned long now = jiffies;
	unsigned long expire = 120 * HZ;

	for(i = 0; i <= dn_rt_hash_mask; i++) {
		rtp = &dn_rt_hash_table[i].chain;

		spin_lock(&dn_rt_hash_table[i].lock);
		while((rt=*rtp) != NULL) {
			if (atomic_read(&rt->u.dst.__refcnt) ||
					(now - rt->u.dst.lastuse) < expire) {
				rtp = &rt->u.dst.dn_next;
				continue;
			}
			*rtp = rt->u.dst.dn_next;
			rt->u.dst.dn_next = NULL;
			dnrt_free(rt);
		}
		spin_unlock(&dn_rt_hash_table[i].lock);

		if ((jiffies - now) > 0)
			break;
	}

	mod_timer(&dn_route_timer, now + decnet_dst_gc_interval * HZ);
}

static int dn_dst_gc(void)
{
	struct dn_route *rt, **rtp;
	int i;
	unsigned long now = jiffies;
	unsigned long expire = 10 * HZ;

	for(i = 0; i <= dn_rt_hash_mask; i++) {

		spin_lock_bh(&dn_rt_hash_table[i].lock);
		rtp = &dn_rt_hash_table[i].chain;

		while((rt=*rtp) != NULL) {
			if (atomic_read(&rt->u.dst.__refcnt) ||
					(now - rt->u.dst.lastuse) < expire) {
				rtp = &rt->u.dst.dn_next;
				continue;
			}
			*rtp = rt->u.dst.dn_next;
			rt->u.dst.dn_next = NULL;
			dnrt_drop(rt);
			break;
		}
		spin_unlock_bh(&dn_rt_hash_table[i].lock);
	}

	return 0;
}

/*
 * The decnet standards don't impose a particular minimum mtu, what they
 * do insist on is that the routing layer accepts a datagram of at least
 * 230 bytes long. Here we have to subtract the routing header length from
 * 230 to get the minimum acceptable mtu. If there is no neighbour, then we
 * assume the worst and use a long header size.
 *
 * We update both the mtu and the advertised mss (i.e. the segment size we
 * advertise to the other end).
 */
static void dn_dst_update_pmtu(struct dst_entry *dst, u32 mtu)
{
	u32 min_mtu = 230;
	struct dn_dev *dn = dst->neighbour ?
			    (struct dn_dev *)dst->neighbour->dev->dn_ptr : NULL;

	if (dn && dn->use_long == 0)
		min_mtu -= 6;
	else
		min_mtu -= 21;

	if (dst->metrics[RTAX_MTU-1] > mtu && mtu >= min_mtu) {
		if (!(dst_metric_locked(dst, RTAX_MTU))) {
			dst->metrics[RTAX_MTU-1] = mtu;
			dst_set_expires(dst, dn_rt_mtu_expires);
		}
		if (!(dst_metric_locked(dst, RTAX_ADVMSS))) {
			u32 mss = mtu - DN_MAX_NSP_DATA_HEADER;
			if (dst->metrics[RTAX_ADVMSS-1] > mss)
				dst->metrics[RTAX_ADVMSS-1] = mss;
		}
	}
}

/*
 * When a route has been marked obsolete. (e.g. routing cache flush)
 */
static struct dst_entry *dn_dst_check(struct dst_entry *dst, __u32 cookie)
{
	return NULL;
}

static struct dst_entry *dn_dst_negative_advice(struct dst_entry *dst)
{
	dst_release(dst);
	return NULL;
}

static void dn_dst_link_failure(struct sk_buff *skb)
{
	return;
}

static inline int compare_keys(struct flowi *fl1, struct flowi *fl2)
{
	return ((fl1->nl_u.dn_u.daddr ^ fl2->nl_u.dn_u.daddr) |
		(fl1->nl_u.dn_u.saddr ^ fl2->nl_u.dn_u.saddr) |
		(fl1->mark ^ fl2->mark) |
		(fl1->nl_u.dn_u.scope ^ fl2->nl_u.dn_u.scope) |
		(fl1->oif ^ fl2->oif) |
		(fl1->iif ^ fl2->iif)) == 0;
}

static int dn_insert_route(struct dn_route *rt, unsigned hash, struct dn_route **rp)
{
	struct dn_route *rth, **rthp;
	unsigned long now = jiffies;

	rthp = &dn_rt_hash_table[hash].chain;

	spin_lock_bh(&dn_rt_hash_table[hash].lock);
	while((rth = *rthp) != NULL) {
		if (compare_keys(&rth->fl, &rt->fl)) {
			/* Put it first */
			*rthp = rth->u.dst.dn_next;
			rcu_assign_pointer(rth->u.dst.dn_next,
					   dn_rt_hash_table[hash].chain);
			rcu_assign_pointer(dn_rt_hash_table[hash].chain, rth);

			rth->u.dst.__use++;
			dst_hold(&rth->u.dst);
			rth->u.dst.lastuse = now;
			spin_unlock_bh(&dn_rt_hash_table[hash].lock);

			dnrt_drop(rt);
			*rp = rth;
			return 0;
		}
		rthp = &rth->u.dst.dn_next;
	}

	rcu_assign_pointer(rt->u.dst.dn_next, dn_rt_hash_table[hash].chain);
	rcu_assign_pointer(dn_rt_hash_table[hash].chain, rt);

	dst_hold(&rt->u.dst);
	rt->u.dst.__use++;
	rt->u.dst.lastuse = now;
	spin_unlock_bh(&dn_rt_hash_table[hash].lock);
	*rp = rt;
	return 0;
}

void dn_run_flush(unsigned long dummy)
{
	int i;
	struct dn_route *rt, *next;

	for(i = 0; i < dn_rt_hash_mask; i++) {
		spin_lock_bh(&dn_rt_hash_table[i].lock);

		if ((rt = xchg(&dn_rt_hash_table[i].chain, NULL)) == NULL)
			goto nothing_to_declare;

		for(; rt; rt=next) {
			next = rt->u.dst.dn_next;
			rt->u.dst.dn_next = NULL;
			dst_free((struct dst_entry *)rt);
		}

nothing_to_declare:
		spin_unlock_bh(&dn_rt_hash_table[i].lock);
	}
}

static DEFINE_SPINLOCK(dn_rt_flush_lock);

void dn_rt_cache_flush(int delay)
{
	unsigned long now = jiffies;
	int user_mode = !in_interrupt();

	if (delay < 0)
		delay = dn_rt_min_delay;

	spin_lock_bh(&dn_rt_flush_lock);

	if (del_timer(&dn_rt_flush_timer) && delay > 0 && dn_rt_deadline) {
		long tmo = (long)(dn_rt_deadline - now);

		if (user_mode && tmo < dn_rt_max_delay - dn_rt_min_delay)
			tmo = 0;

		if (delay > tmo)
			delay = tmo;
	}

	if (delay <= 0) {
		spin_unlock_bh(&dn_rt_flush_lock);
		dn_run_flush(0);
		return;
	}

	if (dn_rt_deadline == 0)
		dn_rt_deadline = now + dn_rt_max_delay;

	dn_rt_flush_timer.expires = now + delay;
	add_timer(&dn_rt_flush_timer);
	spin_unlock_bh(&dn_rt_flush_lock);
}

/**
 * dn_return_short - Return a short packet to its sender
 * @skb: The packet to return
 *
 */
static int dn_return_short(struct sk_buff *skb)
{
	struct dn_skb_cb *cb;
	unsigned char *ptr;
	__le16 *src;
	__le16 *dst;
	__le16 tmp;

	/* Add back headers */
	skb_push(skb, skb->data - skb->nh.raw);

	if ((skb = skb_unshare(skb, GFP_ATOMIC)) == NULL)
		return NET_RX_DROP;

	cb = DN_SKB_CB(skb);
	/* Skip packet length and point to flags */
	ptr = skb->data + 2;
	*ptr++ = (cb->rt_flags & ~DN_RT_F_RQR) | DN_RT_F_RTS;

	dst = (__le16 *)ptr;
	ptr += 2;
	src = (__le16 *)ptr;
	ptr += 2;
	*ptr = 0; /* Zero hop count */

	/* Swap source and destination */
	tmp  = *src;
	*src = *dst;
	*dst = tmp;

	skb->pkt_type = PACKET_OUTGOING;
	dn_rt_finish_output(skb, NULL, NULL);
	return NET_RX_SUCCESS;
}

/**
 * dn_return_long - Return a long packet to its sender
 * @skb: The long format packet to return
 *
 */
static int dn_return_long(struct sk_buff *skb)
{
	struct dn_skb_cb *cb;
	unsigned char *ptr;
	unsigned char *src_addr, *dst_addr;
	unsigned char tmp[ETH_ALEN];

	/* Add back all headers */
	skb_push(skb, skb->data - skb->nh.raw);

	if ((skb = skb_unshare(skb, GFP_ATOMIC)) == NULL)
		return NET_RX_DROP;

	cb = DN_SKB_CB(skb);
	/* Ignore packet length and point to flags */
	ptr = skb->data + 2;

	/* Skip padding */
	if (*ptr & DN_RT_F_PF) {
		char padlen = (*ptr & ~DN_RT_F_PF);
		ptr += padlen;
	}

	*ptr++ = (cb->rt_flags & ~DN_RT_F_RQR) | DN_RT_F_RTS;
	ptr += 2;
	dst_addr = ptr;
	ptr += 8;
	src_addr = ptr;
	ptr += 6;
	*ptr = 0; /* Zero hop count */

	/* Swap source and destination */
	memcpy(tmp, src_addr, ETH_ALEN);
	memcpy(src_addr, dst_addr, ETH_ALEN);
	memcpy(dst_addr, tmp, ETH_ALEN);

	skb->pkt_type = PACKET_OUTGOING;
	dn_rt_finish_output(skb, dst_addr, src_addr);
	return NET_RX_SUCCESS;
}

/**
 * dn_route_rx_packet - Try and find a route for an incoming packet
 * @skb: The packet to find a route for
 *
 * Returns: result of input function if route is found, error code otherwise
 */
static int dn_route_rx_packet(struct sk_buff *skb)
{
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	int err;

	if ((err = dn_route_input(skb)) == 0)
		return dst_input(skb);

	if (decnet_debug_level & 4) {
		char *devname = skb->dev ? skb->dev->name : "???";
		struct dn_skb_cb *cb = DN_SKB_CB(skb);
		printk(KERN_DEBUG
			"DECnet: dn_route_rx_packet: rt_flags=0x%02x dev=%s len=%d src=0x%04hx dst=0x%04hx err=%d type=%d\n",
			(int)cb->rt_flags, devname, skb->len,
			dn_ntohs(cb->src), dn_ntohs(cb->dst),
			err, skb->pkt_type);
	}

	if ((skb->pkt_type == PACKET_HOST) && (cb->rt_flags & DN_RT_F_RQR)) {
		switch(cb->rt_flags & DN_RT_PKT_MSK) {
			case DN_RT_PKT_SHORT:
				return dn_return_short(skb);
			case DN_RT_PKT_LONG:
				return dn_return_long(skb);
		}
	}

	kfree_skb(skb);
	return NET_RX_DROP;
}

static int dn_route_rx_long(struct sk_buff *skb)
{
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	unsigned char *ptr = skb->data;

	if (!pskb_may_pull(skb, 21)) /* 20 for long header, 1 for shortest nsp */
		goto drop_it;

	skb_pull(skb, 20);
	skb->h.raw = skb->data;

	/* Destination info */
	ptr += 2;
	cb->dst = dn_eth2dn(ptr);
	if (memcmp(ptr, dn_hiord_addr, 4) != 0)
		goto drop_it;
	ptr += 6;


	/* Source info */
	ptr += 2;
	cb->src = dn_eth2dn(ptr);
	if (memcmp(ptr, dn_hiord_addr, 4) != 0)
		goto drop_it;
	ptr += 6;
	/* Other junk */
	ptr++;
	cb->hops = *ptr++; /* Visit Count */

	return NF_HOOK(PF_DECnet, NF_DN_PRE_ROUTING, skb, skb->dev, NULL, dn_route_rx_packet);

drop_it:
	kfree_skb(skb);
	return NET_RX_DROP;
}



static int dn_route_rx_short(struct sk_buff *skb)
{
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	unsigned char *ptr = skb->data;

	if (!pskb_may_pull(skb, 6)) /* 5 for short header + 1 for shortest nsp */
		goto drop_it;

	skb_pull(skb, 5);
	skb->h.raw = skb->data;

	cb->dst = *(__le16 *)ptr;
	ptr += 2;
	cb->src = *(__le16 *)ptr;
	ptr += 2;
	cb->hops = *ptr & 0x3f;

	return NF_HOOK(PF_DECnet, NF_DN_PRE_ROUTING, skb, skb->dev, NULL, dn_route_rx_packet);

drop_it:
	kfree_skb(skb);
	return NET_RX_DROP;
}

static int dn_route_discard(struct sk_buff *skb)
{
	/*
	 * I know we drop the packet here, but thats considered success in
	 * this case
	 */
	kfree_skb(skb);
	return NET_RX_SUCCESS;
}

static int dn_route_ptp_hello(struct sk_buff *skb)
{
	dn_dev_hello(skb);
	dn_neigh_pointopoint_hello(skb);
	return NET_RX_SUCCESS;
}

int dn_route_rcv(struct sk_buff *skb, struct net_device *dev, struct packet_type *pt, struct net_device *orig_dev)
{
	struct dn_skb_cb *cb;
	unsigned char flags = 0;
	__u16 len = dn_ntohs(*(__le16 *)skb->data);
	struct dn_dev *dn = (struct dn_dev *)dev->dn_ptr;
	unsigned char padlen = 0;

	if (dn == NULL)
		goto dump_it;

	if ((skb = skb_share_check(skb, GFP_ATOMIC)) == NULL)
		goto out;

	if (!pskb_may_pull(skb, 3))
		goto dump_it;

	skb_pull(skb, 2);

	if (len > skb->len)
		goto dump_it;

	skb_trim(skb, len);

	flags = *skb->data;

	cb = DN_SKB_CB(skb);
	cb->stamp = jiffies;
	cb->iif = dev->ifindex;

	/*
	 * If we have padding, remove it.
	 */
	if (flags & DN_RT_F_PF) {
		padlen = flags & ~DN_RT_F_PF;
		if (!pskb_may_pull(skb, padlen + 1))
			goto dump_it;
		skb_pull(skb, padlen);
		flags = *skb->data;
	}

	skb->nh.raw = skb->data;

	/*
	 * Weed out future version DECnet
	 */
	if (flags & DN_RT_F_VER)
		goto dump_it;

	cb->rt_flags = flags;

	if (decnet_debug_level & 1)
		printk(KERN_DEBUG
			"dn_route_rcv: got 0x%02x from %s [%d %d %d]\n",
			(int)flags, (dev) ? dev->name : "???", len, skb->len,
			padlen);

	if (flags & DN_RT_PKT_CNTL) {
		if (unlikely(skb_linearize(skb)))
			goto dump_it;

		switch(flags & DN_RT_CNTL_MSK) {
			case DN_RT_PKT_INIT:
				dn_dev_init_pkt(skb);
				break;
			case DN_RT_PKT_VERI:
				dn_dev_veri_pkt(skb);
				break;
		}

		if (dn->parms.state != DN_DEV_S_RU)
			goto dump_it;

		switch(flags & DN_RT_CNTL_MSK) {
			case DN_RT_PKT_HELO:
				return NF_HOOK(PF_DECnet, NF_DN_HELLO, skb, skb->dev, NULL, dn_route_ptp_hello);

			case DN_RT_PKT_L1RT:
			case DN_RT_PKT_L2RT:
				return NF_HOOK(PF_DECnet, NF_DN_ROUTE, skb, skb->dev, NULL, dn_route_discard);
			case DN_RT_PKT_ERTH:
				return NF_HOOK(PF_DECnet, NF_DN_HELLO, skb, skb->dev, NULL, dn_neigh_router_hello);

			case DN_RT_PKT_EEDH:
				return NF_HOOK(PF_DECnet, NF_DN_HELLO, skb, skb->dev, NULL, dn_neigh_endnode_hello);
		}
	} else {
		if (dn->parms.state != DN_DEV_S_RU)
			goto dump_it;

		skb_pull(skb, 1); /* Pull flags */

		switch(flags & DN_RT_PKT_MSK) {
			case DN_RT_PKT_LONG:
				return dn_route_rx_long(skb);
			case DN_RT_PKT_SHORT:
				return dn_route_rx_short(skb);
		}
	}

dump_it:
	kfree_skb(skb);
out:
	return NET_RX_DROP;
}

static int dn_output(struct sk_buff *skb)
{
	struct dst_entry *dst = skb->dst;
	struct dn_route *rt = (struct dn_route *)dst;
	struct net_device *dev = dst->dev;
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	struct neighbour *neigh;

	int err = -EINVAL;

	if ((neigh = dst->neighbour) == NULL)
		goto error;

	skb->dev = dev;

	cb->src = rt->rt_saddr;
	cb->dst = rt->rt_daddr;

	/*
	 * Always set the Intra-Ethernet bit on all outgoing packets
	 * originated on this node. Only valid flag from upper layers
	 * is return-to-sender-requested. Set hop count to 0 too.
	 */
	cb->rt_flags &= ~DN_RT_F_RQR;
	cb->rt_flags |= DN_RT_F_IE;
	cb->hops = 0;

	return NF_HOOK(PF_DECnet, NF_DN_LOCAL_OUT, skb, NULL, dev, neigh->output);

error:
	if (net_ratelimit())
		printk(KERN_DEBUG "dn_output: This should not happen\n");

	kfree_skb(skb);

	return err;
}

static int dn_forward(struct sk_buff *skb)
{
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	struct dst_entry *dst = skb->dst;
	struct dn_dev *dn_db = dst->dev->dn_ptr;
	struct dn_route *rt;
	struct neighbour *neigh = dst->neighbour;
	int header_len;
#ifdef CONFIG_NETFILTER
	struct net_device *dev = skb->dev;
#endif

	if (skb->pkt_type != PACKET_HOST)
		goto drop;

	/* Ensure that we have enough space for headers */
	rt = (struct dn_route *)skb->dst;
	header_len = dn_db->use_long ? 21 : 6;
	if (skb_cow(skb, LL_RESERVED_SPACE(rt->u.dst.dev)+header_len))
		goto drop;

	/*
	 * Hop count exceeded.
	 */
	if (++cb->hops > 30)
		goto drop;

	skb->dev = rt->u.dst.dev;

	/*
	 * If packet goes out same interface it came in on, then set
	 * the Intra-Ethernet bit. This has no effect for short
	 * packets, so we don't need to test for them here.
	 */
	cb->rt_flags &= ~DN_RT_F_IE;
	if (rt->rt_flags & RTCF_DOREDIRECT)
		cb->rt_flags |= DN_RT_F_IE;

	return NF_HOOK(PF_DECnet, NF_DN_FORWARD, skb, dev, skb->dev, neigh->output);

drop:
	kfree_skb(skb);
	return NET_RX_DROP;
}

/*
 * Drop packet. This is used for endnodes and for
 * when we should not be forwarding packets from
 * this dest.
 */
static int dn_blackhole(struct sk_buff *skb)
{
	kfree_skb(skb);
	return NET_RX_DROP;
}

/*
 * Used to catch bugs. This should never normally get
 * called.
 */
static int dn_rt_bug(struct sk_buff *skb)
{
	if (net_ratelimit()) {
		struct dn_skb_cb *cb = DN_SKB_CB(skb);

		printk(KERN_DEBUG "dn_rt_bug: skb from:%04x to:%04x\n",
				dn_ntohs(cb->src), dn_ntohs(cb->dst));
	}

	kfree_skb(skb);

	return NET_RX_BAD;
}

static int dn_rt_set_next_hop(struct dn_route *rt, struct dn_fib_res *res)
{
	struct dn_fib_info *fi = res->fi;
	struct net_device *dev = rt->u.dst.dev;
	struct neighbour *n;
	unsigned mss;

	if (fi) {
		if (DN_FIB_RES_GW(*res) &&
		    DN_FIB_RES_NH(*res).nh_scope == RT_SCOPE_LINK)
			rt->rt_gateway = DN_FIB_RES_GW(*res);
		memcpy(rt->u.dst.metrics, fi->fib_metrics,
		       sizeof(rt->u.dst.metrics));
	}
	rt->rt_type = res->type;

	if (dev != NULL && rt->u.dst.neighbour == NULL) {
		n = __neigh_lookup_errno(&dn_neigh_table, &rt->rt_gateway, dev);
		if (IS_ERR(n))
			return PTR_ERR(n);
		rt->u.dst.neighbour = n;
	}

	if (rt->u.dst.metrics[RTAX_MTU-1] == 0 ||
	    rt->u.dst.metrics[RTAX_MTU-1] > rt->u.dst.dev->mtu)
		rt->u.dst.metrics[RTAX_MTU-1] = rt->u.dst.dev->mtu;
	mss = dn_mss_from_pmtu(dev, dst_mtu(&rt->u.dst));
	if (rt->u.dst.metrics[RTAX_ADVMSS-1] == 0 ||
	    rt->u.dst.metrics[RTAX_ADVMSS-1] > mss)
		rt->u.dst.metrics[RTAX_ADVMSS-1] = mss;
	return 0;
}

static inline int dn_match_addr(__le16 addr1, __le16 addr2)
{
	__u16 tmp = dn_ntohs(addr1) ^ dn_ntohs(addr2);
	int match = 16;
	while(tmp) {
		tmp >>= 1;
		match--;
	}
	return match;
}

static __le16 dnet_select_source(const struct net_device *dev, __le16 daddr, int scope)
{
	__le16 saddr = 0;
	struct dn_dev *dn_db = dev->dn_ptr;
	struct dn_ifaddr *ifa;
	int best_match = 0;
	int ret;

	read_lock(&dev_base_lock);
	for(ifa = dn_db->ifa_list; ifa; ifa = ifa->ifa_next) {
		if (ifa->ifa_scope > scope)
			continue;
		if (!daddr) {
			saddr = ifa->ifa_local;
			break;
		}
		ret = dn_match_addr(daddr, ifa->ifa_local);
		if (ret > best_match)
			saddr = ifa->ifa_local;
		if (best_match == 0)
			saddr = ifa->ifa_local;
	}
	read_unlock(&dev_base_lock);

	return saddr;
}

static inline __le16 __dn_fib_res_prefsrc(struct dn_fib_res *res)
{
	return dnet_select_source(DN_FIB_RES_DEV(*res), DN_FIB_RES_GW(*res), res->scope);
}

static inline __le16 dn_fib_rules_map_destination(__le16 daddr, struct dn_fib_res *res)
{
	__le16 mask = dnet_make_mask(res->prefixlen);
	return (daddr&~mask)|res->fi->fib_nh->nh_gw;
}

static int dn_route_output_slow(struct dst_entry **pprt, const struct flowi *oldflp, int try_hard)
{
	struct flowi fl = { .nl_u = { .dn_u =
				      { .daddr = oldflp->fld_dst,
					.saddr = oldflp->fld_src,
					.scope = RT_SCOPE_UNIVERSE,
				     } },
			    .mark = oldflp->mark,
			    .iif = loopback_dev.ifindex,
			    .oif = oldflp->oif };
	struct dn_route *rt = NULL;
	struct net_device *dev_out = NULL;
	struct neighbour *neigh = NULL;
	unsigned hash;
	unsigned flags = 0;
	struct dn_fib_res res = { .fi = NULL, .type = RTN_UNICAST };
	int err;
	int free_res = 0;
	__le16 gateway = 0;

	if (decnet_debug_level & 16)
		printk(KERN_DEBUG
		       "dn_route_output_slow: dst=%04x src=%04x mark=%d"
		       " iif=%d oif=%d\n", dn_ntohs(oldflp->fld_dst),
		       dn_ntohs(oldflp->fld_src),
		       oldflp->mark, loopback_dev.ifindex, oldflp->oif);

	/* If we have an output interface, verify its a DECnet device */
	if (oldflp->oif) {
		dev_out = dev_get_by_index(oldflp->oif);
		err = -ENODEV;
		if (dev_out && dev_out->dn_ptr == NULL) {
			dev_put(dev_out);
			dev_out = NULL;
		}
		if (dev_out == NULL)
			goto out;
	}

	/* If we have a source address, verify that its a local address */
	if (oldflp->fld_src) {
		err = -EADDRNOTAVAIL;

		if (dev_out) {
			if (dn_dev_islocal(dev_out, oldflp->fld_src))
				goto source_ok;
			dev_put(dev_out);
			goto out;
		}
		read_lock(&dev_base_lock);
		for(dev_out = dev_base; dev_out; dev_out = dev_out->next) {
			if (!dev_out->dn_ptr)
				continue;
			if (!dn_dev_islocal(dev_out, oldflp->fld_src))
				continue;
			if ((dev_out->flags & IFF_LOOPBACK) &&
			    oldflp->fld_dst &&
			    !dn_dev_islocal(dev_out, oldflp->fld_dst))
				continue;
			break;
		}
		read_unlock(&dev_base_lock);
		if (dev_out == NULL)
			goto out;
		dev_hold(dev_out);
source_ok:
		;
	}

	/* No destination? Assume its local */
	if (!fl.fld_dst) {
		fl.fld_dst = fl.fld_src;

		err = -EADDRNOTAVAIL;
		if (dev_out)
			dev_put(dev_out);
		dev_out = &loopback_dev;
		dev_hold(dev_out);
		if (!fl.fld_dst) {
			fl.fld_dst =
			fl.fld_src = dnet_select_source(dev_out, 0,
						       RT_SCOPE_HOST);
			if (!fl.fld_dst)
				goto out;
		}
		fl.oif = loopback_dev.ifindex;
		res.type = RTN_LOCAL;
		goto make_route;
	}

	if (decnet_debug_level & 16)
		printk(KERN_DEBUG
		       "dn_route_output_slow: initial checks complete."
		       " dst=%o4x src=%04x oif=%d try_hard=%d\n",
		       dn_ntohs(fl.fld_dst), dn_ntohs(fl.fld_src),
		       fl.oif, try_hard);

	/*
	 * N.B. If the kernel is compiled without router support then
	 * dn_fib_lookup() will evaluate to non-zero so this if () block
	 * will always be executed.
	 */
	err = -ESRCH;
	if (try_hard || (err = dn_fib_lookup(&fl, &res)) != 0) {
		struct dn_dev *dn_db;
		if (err != -ESRCH)
			goto out;
		/*
		 * Here the fallback is basically the standard algorithm for
		 * routing in endnodes which is described in the DECnet routing
		 * docs
		 *
		 * If we are not trying hard, look in neighbour cache.
		 * The result is tested to ensure that if a specific output
		 * device/source address was requested, then we honour that
		 * here
		 */
		if (!try_hard) {
			neigh = neigh_lookup_nodev(&dn_neigh_table, &fl.fld_dst);
			if (neigh) {
				if ((oldflp->oif &&
				    (neigh->dev->ifindex != oldflp->oif)) ||
				    (oldflp->fld_src &&
				    (!dn_dev_islocal(neigh->dev,
						      oldflp->fld_src)))) {
					neigh_release(neigh);
					neigh = NULL;
				} else {
					if (dev_out)
						dev_put(dev_out);
					if (dn_dev_islocal(neigh->dev, fl.fld_dst)) {
						dev_out = &loopback_dev;
						res.type = RTN_LOCAL;
					} else {
						dev_out = neigh->dev;
					}
					dev_hold(dev_out);
					goto select_source;
				}
			}
		}

		/* Not there? Perhaps its a local address */
		if (dev_out == NULL)
			dev_out = dn_dev_get_default();
		err = -ENODEV;
		if (dev_out == NULL)
			goto out;
		dn_db = dev_out->dn_ptr;
		/* Possible improvement - check all devices for local addr */
		if (dn_dev_islocal(dev_out, fl.fld_dst)) {
			dev_put(dev_out);
			dev_out = &loopback_dev;
			dev_hold(dev_out);
			res.type = RTN_LOCAL;
			goto select_source;
		}
		/* Not local either.... try sending it to the default router */
		neigh = neigh_clone(dn_db->router);
		BUG_ON(neigh && neigh->dev != dev_out);

		/* Ok then, we assume its directly connected and move on */
select_source:
		if (neigh)
			gateway = ((struct dn_neigh *)neigh)->addr;
		if (gateway == 0)
			gateway = fl.fld_dst;
		if (fl.fld_src == 0) {
			fl.fld_src = dnet_select_source(dev_out, gateway,
							 res.type == RTN_LOCAL ?
							 RT_SCOPE_HOST :
							 RT_SCOPE_LINK);
			if (fl.fld_src == 0 && res.type != RTN_LOCAL)
				goto e_addr;
		}
		fl.oif = dev_out->ifindex;
		goto make_route;
	}
	free_res = 1;

	if (res.type == RTN_NAT)
		goto e_inval;

	if (res.type == RTN_LOCAL) {
		if (!fl.fld_src)
			fl.fld_src = fl.fld_dst;
		if (dev_out)
			dev_put(dev_out);
		dev_out = &loopback_dev;
		dev_hold(dev_out);
		fl.oif = dev_out->ifindex;
		if (res.fi)
			dn_fib_info_put(res.fi);
		res.fi = NULL;
		goto make_route;
	}

	if (res.fi->fib_nhs > 1 && fl.oif == 0)
		dn_fib_select_multipath(&fl, &res);

	/*
	 * We could add some logic to deal with default routes here and
	 * get rid of some of the special casing above.
	 */

	if (!fl.fld_src)
		fl.fld_src = DN_FIB_RES_PREFSRC(res);

	if (dev_out)
		dev_put(dev_out);
	dev_out = DN_FIB_RES_DEV(res);
	dev_hold(dev_out);
	fl.oif = dev_out->ifindex;
	gateway = DN_FIB_RES_GW(res);

make_route:
	if (dev_out->flags & IFF_LOOPBACK)
		flags |= RTCF_LOCAL;

	rt = dst_alloc(&dn_dst_ops);
	if (rt == NULL)
		goto e_nobufs;

	atomic_set(&rt->u.dst.__refcnt, 1);
	rt->u.dst.flags   = DST_HOST;

	rt->fl.fld_src    = oldflp->fld_src;
	rt->fl.fld_dst    = oldflp->fld_dst;
	rt->fl.oif        = oldflp->oif;
	rt->fl.iif        = 0;
	rt->fl.mark       = oldflp->mark;

	rt->rt_saddr      = fl.fld_src;
	rt->rt_daddr      = fl.fld_dst;
	rt->rt_gateway    = gateway ? gateway : fl.fld_dst;
	rt->rt_local_src  = fl.fld_src;

	rt->rt_dst_map    = fl.fld_dst;
	rt->rt_src_map    = fl.fld_src;

	rt->u.dst.dev = dev_out;
	dev_hold(dev_out);
	rt->u.dst.neighbour = neigh;
	neigh = NULL;

	rt->u.dst.lastuse = jiffies;
	rt->u.dst.output  = dn_output;
	rt->u.dst.input   = dn_rt_bug;
	rt->rt_flags      = flags;
	if (flags & RTCF_LOCAL)
		rt->u.dst.input = dn_nsp_rx;

	err = dn_rt_set_next_hop(rt, &res);
	if (err)
		goto e_neighbour;

	hash = dn_hash(rt->fl.fld_src, rt->fl.fld_dst);
	dn_insert_route(rt, hash, (struct dn_route **)pprt);

done:
	if (neigh)
		neigh_release(neigh);
	if (free_res)
		dn_fib_res_put(&res);
	if (dev_out)
		dev_put(dev_out);
out:
	return err;

e_addr:
	err = -EADDRNOTAVAIL;
	goto done;
e_inval:
	err = -EINVAL;
	goto done;
e_nobufs:
	err = -ENOBUFS;
	goto done;
e_neighbour:
	dst_free(&rt->u.dst);
	goto e_nobufs;
}


/*
 * N.B. The flags may be moved into the flowi at some future stage.
 */
static int __dn_route_output_key(struct dst_entry **pprt, const struct flowi *flp, int flags)
{
	unsigned hash = dn_hash(flp->fld_src, flp->fld_dst);
	struct dn_route *rt = NULL;

	if (!(flags & MSG_TRYHARD)) {
		rcu_read_lock_bh();
		for(rt = rcu_dereference(dn_rt_hash_table[hash].chain); rt;
			rt = rcu_dereference(rt->u.dst.dn_next)) {
			if ((flp->fld_dst == rt->fl.fld_dst) &&
			    (flp->fld_src == rt->fl.fld_src) &&
			    (flp->mark == rt->fl.mark) &&
			    (rt->fl.iif == 0) &&
			    (rt->fl.oif == flp->oif)) {
				rt->u.dst.lastuse = jiffies;
				dst_hold(&rt->u.dst);
				rt->u.dst.__use++;
				rcu_read_unlock_bh();
				*pprt = &rt->u.dst;
				return 0;
			}
		}
		rcu_read_unlock_bh();
	}

	return dn_route_output_slow(pprt, flp, flags);
}

static int dn_route_output_key(struct dst_entry **pprt, struct flowi *flp, int flags)
{
	int err;

	err = __dn_route_output_key(pprt, flp, flags);
	if (err == 0 && flp->proto) {
		err = xfrm_lookup(pprt, flp, NULL, 0);
	}
	return err;
}

int dn_route_output_sock(struct dst_entry **pprt, struct flowi *fl, struct sock *sk, int flags)
{
	int err;

	err = __dn_route_output_key(pprt, fl, flags & MSG_TRYHARD);
	if (err == 0 && fl->proto) {
		err = xfrm_lookup(pprt, fl, sk, !(flags & MSG_DONTWAIT));
	}
	return err;
}

static int dn_route_input_slow(struct sk_buff *skb)
{
	struct dn_route *rt = NULL;
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	struct net_device *in_dev = skb->dev;
	struct net_device *out_dev = NULL;
	struct dn_dev *dn_db;
	struct neighbour *neigh = NULL;
	unsigned hash;
	int flags = 0;
	__le16 gateway = 0;
	__le16 local_src = 0;
	struct flowi fl = { .nl_u = { .dn_u =
				     { .daddr = cb->dst,
				       .saddr = cb->src,
				       .scope = RT_SCOPE_UNIVERSE,
				    } },
			    .mark = skb->mark,
			    .iif = skb->dev->ifindex };
	struct dn_fib_res res = { .fi = NULL, .type = RTN_UNREACHABLE };
	int err = -EINVAL;
	int free_res = 0;

	dev_hold(in_dev);

	if ((dn_db = in_dev->dn_ptr) == NULL)
		goto out;

	/* Zero source addresses are not allowed */
	if (fl.fld_src == 0)
		goto out;

	/*
	 * In this case we've just received a packet from a source
	 * outside ourselves pretending to come from us. We don't
	 * allow it any further to prevent routing loops, spoofing and
	 * other nasties. Loopback packets already have the dst attached
	 * so this only affects packets which have originated elsewhere.
	 */
	err  = -ENOTUNIQ;
	if (dn_dev_islocal(in_dev, cb->src))
		goto out;

	err = dn_fib_lookup(&fl, &res);
	if (err) {
		if (err != -ESRCH)
			goto out;
		/*
		 * Is the destination us ?
		 */
		if (!dn_dev_islocal(in_dev, cb->dst))
			goto e_inval;

		res.type = RTN_LOCAL;
	} else {
		__le16 src_map = fl.fld_src;
		free_res = 1;

		out_dev = DN_FIB_RES_DEV(res);
		if (out_dev == NULL) {
			if (net_ratelimit())
				printk(KERN_CRIT "Bug in dn_route_input_slow() "
						 "No output device\n");
			goto e_inval;
		}
		dev_hold(out_dev);

		if (res.r)
			src_map = fl.fld_src; /* no NAT support for now */

		gateway = DN_FIB_RES_GW(res);
		if (res.type == RTN_NAT) {
			fl.fld_dst = dn_fib_rules_map_destination(fl.fld_dst, &res);
			dn_fib_res_put(&res);
			free_res = 0;
			if (dn_fib_lookup(&fl, &res))
				goto e_inval;
			free_res = 1;
			if (res.type != RTN_UNICAST)
				goto e_inval;
			flags |= RTCF_DNAT;
			gateway = fl.fld_dst;
		}
		fl.fld_src = src_map;
	}

	switch(res.type) {
	case RTN_UNICAST:
		/*
		 * Forwarding check here, we only check for forwarding
		 * being turned off, if you want to only forward intra
		 * area, its up to you to set the routing tables up
		 * correctly.
		 */
		if (dn_db->parms.forwarding == 0)
			goto e_inval;

		if (res.fi->fib_nhs > 1 && fl.oif == 0)
			dn_fib_select_multipath(&fl, &res);

		/*
		 * Check for out_dev == in_dev. We use the RTCF_DOREDIRECT
		 * flag as a hint to set the intra-ethernet bit when
		 * forwarding. If we've got NAT in operation, we don't do
		 * this optimisation.
		 */
		if (out_dev == in_dev && !(flags & RTCF_NAT))
			flags |= RTCF_DOREDIRECT;

		local_src = DN_FIB_RES_PREFSRC(res);

	case RTN_BLACKHOLE:
	case RTN_UNREACHABLE:
		break;
	case RTN_LOCAL:
		flags |= RTCF_LOCAL;
		fl.fld_src = cb->dst;
		fl.fld_dst = cb->src;

		/* Routing tables gave us a gateway */
		if (gateway)
			goto make_route;

		/* Packet was intra-ethernet, so we know its on-link */
		if (cb->rt_flags & DN_RT_F_IE) {
			gateway = cb->src;
			flags |= RTCF_DIRECTSRC;
			goto make_route;
		}

		/* Use the default router if there is one */
		neigh = neigh_clone(dn_db->router);
		if (neigh) {
			gateway = ((struct dn_neigh *)neigh)->addr;
			goto make_route;
		}

		/* Close eyes and pray */
		gateway = cb->src;
		flags |= RTCF_DIRECTSRC;
		goto make_route;
	default:
		goto e_inval;
	}

make_route:
	rt = dst_alloc(&dn_dst_ops);
	if (rt == NULL)
		goto e_nobufs;

	rt->rt_saddr      = fl.fld_src;
	rt->rt_daddr      = fl.fld_dst;
	rt->rt_gateway    = fl.fld_dst;
	if (gateway)
		rt->rt_gateway = gateway;
	rt->rt_local_src  = local_src ? local_src : rt->rt_saddr;

	rt->rt_dst_map    = fl.fld_dst;
	rt->rt_src_map    = fl.fld_src;

	rt->fl.fld_src    = cb->src;
	rt->fl.fld_dst    = cb->dst;
	rt->fl.oif        = 0;
	rt->fl.iif        = in_dev->ifindex;
	rt->fl.mark       = fl.mark;

	rt->u.dst.flags = DST_HOST;
	rt->u.dst.neighbour = neigh;
	rt->u.dst.dev = out_dev;
	rt->u.dst.lastuse = jiffies;
	rt->u.dst.output = dn_rt_bug;
	switch(res.type) {
		case RTN_UNICAST:
			rt->u.dst.input = dn_forward;
			break;
		case RTN_LOCAL:
			rt->u.dst.output = dn_output;
			rt->u.dst.input = dn_nsp_rx;
			rt->u.dst.dev = in_dev;
			flags |= RTCF_LOCAL;
			break;
		default:
		case RTN_UNREACHABLE:
		case RTN_BLACKHOLE:
			rt->u.dst.input = dn_blackhole;
	}
	rt->rt_flags = flags;
	if (rt->u.dst.dev)
		dev_hold(rt->u.dst.dev);

	err = dn_rt_set_next_hop(rt, &res);
	if (err)
		goto e_neighbour;

	hash = dn_hash(rt->fl.fld_src, rt->fl.fld_dst);
	dn_insert_route(rt, hash, (struct dn_route **)&skb->dst);

done:
	if (neigh)
		neigh_release(neigh);
	if (free_res)
		dn_fib_res_put(&res);
	dev_put(in_dev);
	if (out_dev)
		dev_put(out_dev);
out:
	return err;

e_inval:
	err = -EINVAL;
	goto done;

e_nobufs:
	err = -ENOBUFS;
	goto done;

e_neighbour:
	dst_free(&rt->u.dst);
	goto done;
}

int dn_route_input(struct sk_buff *skb)
{
	struct dn_route *rt;
	struct dn_skb_cb *cb = DN_SKB_CB(skb);
	unsigned hash = dn_hash(cb->src, cb->dst);

	if (skb->dst)
		return 0;

	rcu_read_lock();
	for(rt = rcu_dereference(dn_rt_hash_table[hash].chain); rt != NULL;
	    rt = rcu_dereference(rt->u.dst.dn_next)) {
		if ((rt->fl.fld_src == cb->src) &&
		    (rt->fl.fld_dst == cb->dst) &&
		    (rt->fl.oif == 0) &&
		    (rt->fl.mark == skb->mark) &&
		    (rt->fl.iif == cb->iif)) {
			rt->u.dst.lastuse = jiffies;
			dst_hold(&rt->u.dst);
			rt->u.dst.__use++;
			rcu_read_unlock();
			skb->dst = (struct dst_entry *)rt;
			return 0;
		}
	}
	rcu_read_unlock();

	return dn_route_input_slow(skb);
}

static int dn_rt_fill_info(struct sk_buff *skb, u32 pid, u32 seq,
			   int event, int nowait, unsigned int flags)
{
	struct dn_route *rt = (struct dn_route *)skb->dst;
	struct rtmsg *r;
	struct nlmsghdr *nlh;
	unsigned char *b = skb->tail;
	long expires;

	nlh = NLMSG_NEW(skb, pid, seq, event, sizeof(*r), flags);
	r = NLMSG_DATA(nlh);
	r->rtm_family = AF_DECnet;
	r->rtm_dst_len = 16;
	r->rtm_src_len = 0;
	r->rtm_tos = 0;
	r->rtm_table = RT_TABLE_MAIN;
	RTA_PUT_U32(skb, RTA_TABLE, RT_TABLE_MAIN);
	r->rtm_type = rt->rt_type;
	r->rtm_flags = (rt->rt_flags & ~0xFFFF) | RTM_F_CLONED;
	r->rtm_scope = RT_SCOPE_UNIVERSE;
	r->rtm_protocol = RTPROT_UNSPEC;
	if (rt->rt_flags & RTCF_NOTIFY)
		r->rtm_flags |= RTM_F_NOTIFY;
	RTA_PUT(skb, RTA_DST, 2, &rt->rt_daddr);
	if (rt->fl.fld_src) {
		r->rtm_src_len = 16;
		RTA_PUT(skb, RTA_SRC, 2, &rt->fl.fld_src);
	}
	if (rt->u.dst.dev)
		RTA_PUT(skb, RTA_OIF, sizeof(int), &rt->u.dst.dev->ifindex);
	/*
	 * Note to self - change this if input routes reverse direction when
	 * they deal only with inputs and not with replies like they do
	 * currently.
	 */
	RTA_PUT(skb, RTA_PREFSRC, 2, &rt->rt_local_src);
	if (rt->rt_daddr != rt->rt_gateway)
		RTA_PUT(skb, RTA_GATEWAY, 2, &rt->rt_gateway);
	if (rtnetlink_put_metrics(skb, rt->u.dst.metrics) < 0)
		goto rtattr_failure;
	expires = rt->u.dst.expires ? rt->u.dst.expires - jiffies : 0;
	if (rtnl_put_cacheinfo(skb, &rt->u.dst, 0, 0, 0, expires,
			       rt->u.dst.error) < 0)
		goto rtattr_failure;
	if (rt->fl.iif)
		RTA_PUT(skb, RTA_IIF, sizeof(int), &rt->fl.iif);

	nlh->nlmsg_len = skb->tail - b;
	return skb->len;

nlmsg_failure:
rtattr_failure:
	skb_trim(skb, b - skb->data);
	return -1;
}

/*
 * This is called by both endnodes and routers now.
 */
int dn_cache_getroute(struct sk_buff *in_skb, struct nlmsghdr *nlh, void *arg)
{
	struct rtattr **rta = arg;
	struct rtmsg *rtm = NLMSG_DATA(nlh);
	struct dn_route *rt = NULL;
	struct dn_skb_cb *cb;
	int err;
	struct sk_buff *skb;
	struct flowi fl;

	memset(&fl, 0, sizeof(fl));
	fl.proto = DNPROTO_NSP;

	skb = alloc_skb(NLMSG_GOODSIZE, GFP_KERNEL);
	if (skb == NULL)
		return -ENOBUFS;
	skb_reset_mac_header(skb);
	cb = DN_SKB_CB(skb);

	if (rta[RTA_SRC-1])
		memcpy(&fl.fld_src, RTA_DATA(rta[RTA_SRC-1]), 2);
	if (rta[RTA_DST-1])
		memcpy(&fl.fld_dst, RTA_DATA(rta[RTA_DST-1]), 2);
	if (rta[RTA_IIF-1])
		memcpy(&fl.iif, RTA_DATA(rta[RTA_IIF-1]), sizeof(int));

	if (fl.iif) {
		struct net_device *dev;
		if ((dev = dev_get_by_index(fl.iif)) == NULL) {
			kfree_skb(skb);
			return -ENODEV;
		}
		if (!dev->dn_ptr) {
			dev_put(dev);
			kfree_skb(skb);
			return -ENODEV;
		}
		skb->protocol = __constant_htons(ETH_P_DNA_RT);
		skb->dev = dev;
		cb->src = fl.fld_src;
		cb->dst = fl.fld_dst;
		local_bh_disable();
		err = dn_route_input(skb);
		local_bh_enable();
		memset(cb, 0, sizeof(struct dn_skb_cb));
		rt = (struct dn_route *)skb->dst;
		if (!err && -rt->u.dst.error)
			err = rt->u.dst.error;
	} else {
		int oif = 0;
		if (rta[RTA_OIF - 1])
			memcpy(&oif, RTA_DATA(rta[RTA_OIF - 1]), sizeof(int));
		fl.oif = oif;
		err = dn_route_output_key((struct dst_entry **)&rt, &fl, 0);
	}

	if (skb->dev)
		dev_put(skb->dev);
	skb->dev = NULL;
	if (err)
		goto out_free;
	skb->dst = &rt->u.dst;
	if (rtm->rtm_flags & RTM_F_NOTIFY)
		rt->rt_flags |= RTCF_NOTIFY;

	err = dn_rt_fill_info(skb, NETLINK_CB(in_skb).pid, nlh->nlmsg_seq, RTM_NEWROUTE, 0, 0);

	if (err == 0)
		goto out_free;
	if (err < 0) {
		err = -EMSGSIZE;
		goto out_free;
	}

	return rtnl_unicast(skb, NETLINK_CB(in_skb).pid);

out_free:
	kfree_skb(skb);
	return err;
}

/*
 * For routers, this is called from dn_fib_dump, but for endnodes its
 * called directly from the rtnetlink dispatch table.
 */
int dn_cache_dump(struct sk_buff *skb, struct netlink_callback *cb)
{
	struct dn_route *rt;
	int h, s_h;
	int idx, s_idx;

	if (NLMSG_PAYLOAD(cb->nlh, 0) < sizeof(struct rtmsg))
		return -EINVAL;
	if (!(((struct rtmsg *)NLMSG_DATA(cb->nlh))->rtm_flags&RTM_F_CLONED))
		return 0;

	s_h = cb->args[0];
	s_idx = idx = cb->args[1];
	for(h = 0; h <= dn_rt_hash_mask; h++) {
		if (h < s_h)
			continue;
		if (h > s_h)
			s_idx = 0;
		rcu_read_lock_bh();
		for(rt = rcu_dereference(dn_rt_hash_table[h].chain), idx = 0;
			rt;
			rt = rcu_dereference(rt->u.dst.dn_next), idx++) {
			if (idx < s_idx)
				continue;
			skb->dst = dst_clone(&rt->u.dst);
			if (dn_rt_fill_info(skb, NETLINK_CB(cb->skb).pid,
					cb->nlh->nlmsg_seq, RTM_NEWROUTE,
					1, NLM_F_MULTI) <= 0) {
				dst_release(xchg(&skb->dst, NULL));
				rcu_read_unlock_bh();
				goto done;
			}
			dst_release(xchg(&skb->dst, NULL));
		}
		rcu_read_unlock_bh();
	}

done:
	cb->args[0] = h;
	cb->args[1] = idx;
	return skb->len;
}

#ifdef CONFIG_PROC_FS
struct dn_rt_cache_iter_state {
	int bucket;
};

static struct dn_route *dn_rt_cache_get_first(struct seq_file *seq)
{
	struct dn_route *rt = NULL;
	struct dn_rt_cache_iter_state *s = seq->private;

	for(s->bucket = dn_rt_hash_mask; s->bucket >= 0; --s->bucket) {
		rcu_read_lock_bh();
		rt = dn_rt_hash_table[s->bucket].chain;
		if (rt)
			break;
		rcu_read_unlock_bh();
	}
	return rt;
}

static struct dn_route *dn_rt_cache_get_next(struct seq_file *seq, struct dn_route *rt)
{
	struct dn_rt_cache_iter_state *s = rcu_dereference(seq->private);

	rt = rt->u.dst.dn_next;
	while(!rt) {
		rcu_read_unlock_bh();
		if (--s->bucket < 0)
			break;
		rcu_read_lock_bh();
		rt = dn_rt_hash_table[s->bucket].chain;
	}
	return rt;
}

static void *dn_rt_cache_seq_start(struct seq_file *seq, loff_t *pos)
{
	struct dn_route *rt = dn_rt_cache_get_first(seq);

	if (rt) {
		while(*pos && (rt = dn_rt_cache_get_next(seq, rt)))
			--*pos;
	}
	return *pos ? NULL : rt;
}

static void *dn_rt_cache_seq_next(struct seq_file *seq, void *v, loff_t *pos)
{
	struct dn_route *rt = dn_rt_cache_get_next(seq, v);
	++*pos;
	return rt;
}

static void dn_rt_cache_seq_stop(struct seq_file *seq, void *v)
{
	if (v)
		rcu_read_unlock_bh();
}

static int dn_rt_cache_seq_show(struct seq_file *seq, void *v)
{
	struct dn_route *rt = v;
	char buf1[DN_ASCBUF_LEN], buf2[DN_ASCBUF_LEN];

	seq_printf(seq, "%-8s %-7s %-7s %04d %04d %04d\n",
			rt->u.dst.dev ? rt->u.dst.dev->name : "*",
			dn_addr2asc(dn_ntohs(rt->rt_daddr), buf1),
			dn_addr2asc(dn_ntohs(rt->rt_saddr), buf2),
			atomic_read(&rt->u.dst.__refcnt),
			rt->u.dst.__use,
			(int) dst_metric(&rt->u.dst, RTAX_RTT));
	return 0;
}

static struct seq_operations dn_rt_cache_seq_ops = {
	.start	= dn_rt_cache_seq_start,
	.next	= dn_rt_cache_seq_next,
	.stop	= dn_rt_cache_seq_stop,
	.show	= dn_rt_cache_seq_show,
};

static int dn_rt_cache_seq_open(struct inode *inode, struct file *file)
{
	struct seq_file *seq;
	int rc = -ENOMEM;
	struct dn_rt_cache_iter_state *s = kmalloc(sizeof(*s), GFP_KERNEL);

	if (!s)
		goto out;
	rc = seq_open(file, &dn_rt_cache_seq_ops);
	if (rc)
		goto out_kfree;
	seq		= file->private_data;
	seq->private	= s;
	memset(s, 0, sizeof(*s));
out:
	return rc;
out_kfree:
	kfree(s);
	goto out;
}

static const struct file_operations dn_rt_cache_seq_fops = {
	.owner	 = THIS_MODULE,
	.open	 = dn_rt_cache_seq_open,
	.read	 = seq_read,
	.llseek	 = seq_lseek,
	.release = seq_release_private,
};

#endif /* CONFIG_PROC_FS */

void __init dn_route_init(void)
{
	int i, goal, order;

	dn_dst_ops.kmem_cachep =
		kmem_cache_create("dn_dst_cache", sizeof(struct dn_route), 0,
				  SLAB_HWCACHE_ALIGN|SLAB_PANIC, NULL, NULL);
	init_timer(&dn_route_timer);
	dn_route_timer.function = dn_dst_check_expire;
	dn_route_timer.expires = jiffies + decnet_dst_gc_interval * HZ;
	add_timer(&dn_route_timer);

	goal = num_physpages >> (26 - PAGE_SHIFT);

	for(order = 0; (1UL << order) < goal; order++)
		/* NOTHING */;

	/*
	 * Only want 1024 entries max, since the table is very, very unlikely
	 * to be larger than that.
	 */
	while(order && ((((1UL << order) * PAGE_SIZE) /
				sizeof(struct dn_rt_hash_bucket)) >= 2048))
		order--;

	do {
		dn_rt_hash_mask = (1UL << order) * PAGE_SIZE /
			sizeof(struct dn_rt_hash_bucket);
		while(dn_rt_hash_mask & (dn_rt_hash_mask - 1))
			dn_rt_hash_mask--;
		dn_rt_hash_table = (struct dn_rt_hash_bucket *)
			__get_free_pages(GFP_ATOMIC, order);
	} while (dn_rt_hash_table == NULL && --order > 0);

	if (!dn_rt_hash_table)
		panic("Failed to allocate DECnet route cache hash table\n");

	printk(KERN_INFO
		"DECnet: Routing cache hash table of %u buckets, %ldKbytes\n",
		dn_rt_hash_mask,
		(long)(dn_rt_hash_mask*sizeof(struct dn_rt_hash_bucket))/1024);

	dn_rt_hash_mask--;
	for(i = 0; i <= dn_rt_hash_mask; i++) {
		spin_lock_init(&dn_rt_hash_table[i].lock);
		dn_rt_hash_table[i].chain = NULL;
	}

	dn_dst_ops.gc_thresh = (dn_rt_hash_mask + 1);

	proc_net_fops_create("decnet_cache", S_IRUGO, &dn_rt_cache_seq_fops);
}

void __exit dn_route_cleanup(void)
{
	del_timer(&dn_route_timer);
	dn_run_flush(0);

	proc_net_remove("decnet_cache");
}