aboutsummaryrefslogtreecommitdiff
path: root/sys/net/if_vlan.c
blob: 0f2ded3f604010a5cf6398f1dfeecbf1a04597ac (plain) (blame)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
/*-
 * Copyright 1998 Massachusetts Institute of Technology
 * Copyright 2012 ADARA Networks, Inc.
 * Copyright 2017 Dell EMC Isilon
 *
 * Portions of this software were developed by Robert N. M. Watson under
 * contract to ADARA Networks, Inc.
 *
 * Permission to use, copy, modify, and distribute this software and
 * its documentation for any purpose and without fee is hereby
 * granted, provided that both the above copyright notice and this
 * permission notice appear in all copies, that both the above
 * copyright notice and this permission notice appear in all
 * supporting documentation, and that the name of M.I.T. not be used
 * in advertising or publicity pertaining to distribution of the
 * software without specific, written prior permission.  M.I.T. makes
 * no representations about the suitability of this software for any
 * purpose.  It is provided "as is" without express or implied
 * warranty.
 *
 * THIS SOFTWARE IS PROVIDED BY M.I.T. ``AS IS''.  M.I.T. DISCLAIMS
 * ALL EXPRESS OR IMPLIED WARRANTIES WITH REGARD TO THIS SOFTWARE,
 * INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
 * MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. IN NO EVENT
 * SHALL M.I.T. BE LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL,
 * SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT
 * LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF
 * USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
 * ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
 * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
 * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 */

/*
 * if_vlan.c - pseudo-device driver for IEEE 802.1Q virtual LANs.
 * This is sort of sneaky in the implementation, since
 * we need to pretend to be enough of an Ethernet implementation
 * to make arp work.  The way we do this is by telling everyone
 * that we are an Ethernet, and then catch the packets that
 * ether_output() sends to us via if_transmit(), rewrite them for
 * use by the real outgoing interface, and ask it to send them.
 */

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

#include "opt_inet.h"
#include "opt_inet6.h"
#include "opt_kern_tls.h"
#include "opt_vlan.h"
#include "opt_ratelimit.h"

#include <sys/param.h>
#include <sys/eventhandler.h>
#include <sys/kernel.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mbuf.h>
#include <sys/module.h>
#include <sys/rmlock.h>
#include <sys/priv.h>
#include <sys/queue.h>
#include <sys/socket.h>
#include <sys/sockio.h>
#include <sys/sysctl.h>
#include <sys/systm.h>
#include <sys/sx.h>
#include <sys/taskqueue.h>

#include <net/bpf.h>
#include <net/ethernet.h>
#include <net/if.h>
#include <net/if_var.h>
#include <net/if_private.h>
#include <net/if_clone.h>
#include <net/if_dl.h>
#include <net/if_types.h>
#include <net/if_vlan_var.h>
#include <net/route.h>
#include <net/vnet.h>

#ifdef INET
#include <netinet/in.h>
#include <netinet/if_ether.h>
#endif

#define	VLAN_DEF_HWIDTH	4
#define	VLAN_IFFLAGS	(IFF_BROADCAST | IFF_MULTICAST)

#define	UP_AND_RUNNING(ifp) \
    ((ifp)->if_flags & IFF_UP && (ifp)->if_drv_flags & IFF_DRV_RUNNING)

CK_SLIST_HEAD(ifvlanhead, ifvlan);

struct ifvlantrunk {
	struct	ifnet   *parent;	/* parent interface of this trunk */
	struct	mtx	lock;
#ifdef VLAN_ARRAY
#define	VLAN_ARRAY_SIZE	(EVL_VLID_MASK + 1)
	struct	ifvlan	*vlans[VLAN_ARRAY_SIZE]; /* static table */
#else
	struct	ifvlanhead *hash;	/* dynamic hash-list table */
	uint16_t	hmask;
	uint16_t	hwidth;
#endif
	int		refcnt;
};

#if defined(KERN_TLS) || defined(RATELIMIT)
struct vlan_snd_tag {
	struct m_snd_tag com;
	struct m_snd_tag *tag;
};

static inline struct vlan_snd_tag *
mst_to_vst(struct m_snd_tag *mst)
{

	return (__containerof(mst, struct vlan_snd_tag, com));
}
#endif

/*
 * This macro provides a facility to iterate over every vlan on a trunk with
 * the assumption that none will be added/removed during iteration.
 */
#ifdef VLAN_ARRAY
#define VLAN_FOREACH(_ifv, _trunk) \
	size_t _i; \
	for (_i = 0; _i < VLAN_ARRAY_SIZE; _i++) \
		if (((_ifv) = (_trunk)->vlans[_i]) != NULL)
#else /* VLAN_ARRAY */
#define VLAN_FOREACH(_ifv, _trunk) \
	struct ifvlan *_next; \
	size_t _i; \
	for (_i = 0; _i < (1 << (_trunk)->hwidth); _i++) \
		CK_SLIST_FOREACH_SAFE((_ifv), &(_trunk)->hash[_i], ifv_list, _next)
#endif /* VLAN_ARRAY */

/*
 * This macro provides a facility to iterate over every vlan on a trunk while
 * also modifying the number of vlans on the trunk. The iteration continues
 * until some condition is met or there are no more vlans on the trunk.
 */
#ifdef VLAN_ARRAY
/* The VLAN_ARRAY case is simple -- just a for loop using the condition. */
#define VLAN_FOREACH_UNTIL_SAFE(_ifv, _trunk, _cond) \
	size_t _i; \
	for (_i = 0; !(_cond) && _i < VLAN_ARRAY_SIZE; _i++) \
		if (((_ifv) = (_trunk)->vlans[_i]))
#else /* VLAN_ARRAY */
/*
 * The hash table case is more complicated. We allow for the hash table to be
 * modified (i.e. vlans removed) while we are iterating over it. To allow for
 * this we must restart the iteration every time we "touch" something during
 * the iteration, since removal will resize the hash table and invalidate our
 * current position. If acting on the touched element causes the trunk to be
 * emptied, then iteration also stops.
 */
#define VLAN_FOREACH_UNTIL_SAFE(_ifv, _trunk, _cond) \
	size_t _i; \
	bool _touch = false; \
	for (_i = 0; \
	    !(_cond) && _i < (1 << (_trunk)->hwidth); \
	    _i = (_touch && ((_trunk) != NULL) ? 0 : _i + 1), _touch = false) \
		if (((_ifv) = CK_SLIST_FIRST(&(_trunk)->hash[_i])) != NULL && \
		    (_touch = true))
#endif /* VLAN_ARRAY */

struct vlan_mc_entry {
	struct sockaddr_dl		mc_addr;
	CK_SLIST_ENTRY(vlan_mc_entry)	mc_entries;
	struct epoch_context		mc_epoch_ctx;
};

struct ifvlan {
	struct	ifvlantrunk *ifv_trunk;
	struct	ifnet *ifv_ifp;
#define	TRUNK(ifv)	((ifv)->ifv_trunk)
#define	PARENT(ifv)	(TRUNK(ifv)->parent)
	void	*ifv_cookie;
	int	ifv_pflags;	/* special flags we have set on parent */
	int	ifv_capenable;
	int	ifv_encaplen;	/* encapsulation length */
	int	ifv_mtufudge;	/* MTU fudged by this much */
	int	ifv_mintu;	/* min transmission unit */
	struct  ether_8021q_tag ifv_qtag;
#define ifv_proto	ifv_qtag.proto
#define ifv_vid		ifv_qtag.vid
#define ifv_pcp		ifv_qtag.pcp
	struct task lladdr_task;
	CK_SLIST_HEAD(, vlan_mc_entry) vlan_mc_listhead;
#ifndef VLAN_ARRAY
	CK_SLIST_ENTRY(ifvlan) ifv_list;
#endif
};

/* Special flags we should propagate to parent. */
static struct {
	int flag;
	int (*func)(struct ifnet *, int);
} vlan_pflags[] = {
	{IFF_PROMISC, ifpromisc},
	{IFF_ALLMULTI, if_allmulti},
	{0, NULL}
};

VNET_DECLARE(int, vlan_mtag_pcp);
#define	V_vlan_mtag_pcp	VNET(vlan_mtag_pcp)

static const char vlanname[] = "vlan";
static MALLOC_DEFINE(M_VLAN, vlanname, "802.1Q Virtual LAN Interface");

static eventhandler_tag ifdetach_tag;
static eventhandler_tag iflladdr_tag;
static eventhandler_tag ifevent_tag;

/*
 * if_vlan uses two module-level synchronizations primitives to allow concurrent
 * modification of vlan interfaces and (mostly) allow for vlans to be destroyed
 * while they are being used for tx/rx. To accomplish this in a way that has
 * acceptable performance and cooperation with other parts of the network stack
 * there is a non-sleepable epoch(9) and an sx(9).
 *
 * The performance-sensitive paths that warrant using the epoch(9) are
 * vlan_transmit and vlan_input. Both have to check for the vlan interface's
 * existence using if_vlantrunk, and being in the network tx/rx paths the use
 * of an epoch(9) gives a measureable improvement in performance.
 *
 * The reason for having an sx(9) is mostly because there are still areas that
 * must be sleepable and also have safe concurrent access to a vlan interface.
 * Since the sx(9) exists, it is used by default in most paths unless sleeping
 * is not permitted, or if it is not clear whether sleeping is permitted.
 *
 */
#define _VLAN_SX_ID ifv_sx

static struct sx _VLAN_SX_ID;

#define VLAN_LOCKING_INIT() \
	sx_init_flags(&_VLAN_SX_ID, "vlan_sx", SX_RECURSE)

#define VLAN_LOCKING_DESTROY() \
	sx_destroy(&_VLAN_SX_ID)

#define	VLAN_SLOCK()			sx_slock(&_VLAN_SX_ID)
#define	VLAN_SUNLOCK()			sx_sunlock(&_VLAN_SX_ID)
#define	VLAN_XLOCK()			sx_xlock(&_VLAN_SX_ID)
#define	VLAN_XUNLOCK()			sx_xunlock(&_VLAN_SX_ID)
#define	VLAN_SLOCK_ASSERT()		sx_assert(&_VLAN_SX_ID, SA_SLOCKED)
#define	VLAN_XLOCK_ASSERT()		sx_assert(&_VLAN_SX_ID, SA_XLOCKED)
#define	VLAN_SXLOCK_ASSERT()		sx_assert(&_VLAN_SX_ID, SA_LOCKED)

/*
 * We also have a per-trunk mutex that should be acquired when changing
 * its state.
 */
#define	TRUNK_LOCK_INIT(trunk)		mtx_init(&(trunk)->lock, vlanname, NULL, MTX_DEF)
#define	TRUNK_LOCK_DESTROY(trunk)	mtx_destroy(&(trunk)->lock)
#define	TRUNK_WLOCK(trunk)		mtx_lock(&(trunk)->lock)
#define	TRUNK_WUNLOCK(trunk)		mtx_unlock(&(trunk)->lock)
#define	TRUNK_WLOCK_ASSERT(trunk)	mtx_assert(&(trunk)->lock, MA_OWNED);

/*
 * The VLAN_ARRAY substitutes the dynamic hash with a static array
 * with 4096 entries. In theory this can give a boost in processing,
 * however in practice it does not. Probably this is because the array
 * is too big to fit into CPU cache.
 */
#ifndef VLAN_ARRAY
static	void vlan_inithash(struct ifvlantrunk *trunk);
static	void vlan_freehash(struct ifvlantrunk *trunk);
static	int vlan_inshash(struct ifvlantrunk *trunk, struct ifvlan *ifv);
static	int vlan_remhash(struct ifvlantrunk *trunk, struct ifvlan *ifv);
static	void vlan_growhash(struct ifvlantrunk *trunk, int howmuch);
static __inline struct ifvlan * vlan_gethash(struct ifvlantrunk *trunk,
	uint16_t vid);
#endif
static	void trunk_destroy(struct ifvlantrunk *trunk);

static	void vlan_init(void *foo);
static	void vlan_input(struct ifnet *ifp, struct mbuf *m);
static	int vlan_ioctl(struct ifnet *ifp, u_long cmd, caddr_t addr);
#if defined(KERN_TLS) || defined(RATELIMIT)
static	int vlan_snd_tag_alloc(struct ifnet *,
    union if_snd_tag_alloc_params *, struct m_snd_tag **);
static	int vlan_snd_tag_modify(struct m_snd_tag *,
    union if_snd_tag_modify_params *);
static	int vlan_snd_tag_query(struct m_snd_tag *,
    union if_snd_tag_query_params *);
static	void vlan_snd_tag_free(struct m_snd_tag *);
static struct m_snd_tag *vlan_next_snd_tag(struct m_snd_tag *);
static void vlan_ratelimit_query(struct ifnet *,
    struct if_ratelimit_query_results *);
#endif
static	void vlan_qflush(struct ifnet *ifp);
static	int vlan_setflag(struct ifnet *ifp, int flag, int status,
    int (*func)(struct ifnet *, int));
static	int vlan_setflags(struct ifnet *ifp, int status);
static	int vlan_setmulti(struct ifnet *ifp);
static	int vlan_transmit(struct ifnet *ifp, struct mbuf *m);
#ifdef ALTQ
static void vlan_altq_start(struct ifnet *ifp);
static	int vlan_altq_transmit(struct ifnet *ifp, struct mbuf *m);
#endif
static	int vlan_output(struct ifnet *ifp, struct mbuf *m,
    const struct sockaddr *dst, struct route *ro);
static	void vlan_unconfig(struct ifnet *ifp);
static	void vlan_unconfig_locked(struct ifnet *ifp, int departing);
static	int vlan_config(struct ifvlan *ifv, struct ifnet *p, uint16_t tag,
	uint16_t proto);
static	void vlan_link_state(struct ifnet *ifp);
static	void vlan_capabilities(struct ifvlan *ifv);
static	void vlan_trunk_capabilities(struct ifnet *ifp);

static	struct ifnet *vlan_clone_match_ethervid(const char *, int *);
static	int vlan_clone_match(struct if_clone *, const char *);
static	int vlan_clone_create(struct if_clone *, char *, size_t,
    struct ifc_data *, struct ifnet **);
static	int vlan_clone_destroy(struct if_clone *, struct ifnet *, uint32_t);

static	void vlan_ifdetach(void *arg, struct ifnet *ifp);
static  void vlan_iflladdr(void *arg, struct ifnet *ifp);
static  void vlan_ifevent(void *arg, struct ifnet *ifp, int event);

static  void vlan_lladdr_fn(void *arg, int pending);

static struct if_clone *vlan_cloner;

#ifdef VIMAGE
VNET_DEFINE_STATIC(struct if_clone *, vlan_cloner);
#define	V_vlan_cloner	VNET(vlan_cloner)
#endif

#ifdef RATELIMIT
static const struct if_snd_tag_sw vlan_snd_tag_ul_sw = {
	.snd_tag_modify = vlan_snd_tag_modify,
	.snd_tag_query = vlan_snd_tag_query,
	.snd_tag_free = vlan_snd_tag_free,
	.next_snd_tag = vlan_next_snd_tag,
	.type = IF_SND_TAG_TYPE_UNLIMITED
};

static const struct if_snd_tag_sw vlan_snd_tag_rl_sw = {
	.snd_tag_modify = vlan_snd_tag_modify,
	.snd_tag_query = vlan_snd_tag_query,
	.snd_tag_free = vlan_snd_tag_free,
	.next_snd_tag = vlan_next_snd_tag,
	.type = IF_SND_TAG_TYPE_RATE_LIMIT
};
#endif

#ifdef KERN_TLS
static const struct if_snd_tag_sw vlan_snd_tag_tls_sw = {
	.snd_tag_modify = vlan_snd_tag_modify,
	.snd_tag_query = vlan_snd_tag_query,
	.snd_tag_free = vlan_snd_tag_free,
	.next_snd_tag = vlan_next_snd_tag,
	.type = IF_SND_TAG_TYPE_TLS
};

#ifdef RATELIMIT
static const struct if_snd_tag_sw vlan_snd_tag_tls_rl_sw = {
	.snd_tag_modify = vlan_snd_tag_modify,
	.snd_tag_query = vlan_snd_tag_query,
	.snd_tag_free = vlan_snd_tag_free,
	.next_snd_tag = vlan_next_snd_tag,
	.type = IF_SND_TAG_TYPE_TLS_RATE_LIMIT
};
#endif
#endif

static void
vlan_mc_free(struct epoch_context *ctx)
{
	struct vlan_mc_entry *mc = __containerof(ctx, struct vlan_mc_entry, mc_epoch_ctx);
	free(mc, M_VLAN);
}

#ifndef VLAN_ARRAY
#define HASH(n, m)	((((n) >> 8) ^ ((n) >> 4) ^ (n)) & (m))

static void
vlan_inithash(struct ifvlantrunk *trunk)
{
	int i, n;

	/*
	 * The trunk must not be locked here since we call malloc(M_WAITOK).
	 * It is OK in case this function is called before the trunk struct
	 * gets hooked up and becomes visible from other threads.
	 */

	KASSERT(trunk->hwidth == 0 && trunk->hash == NULL,
	    ("%s: hash already initialized", __func__));

	trunk->hwidth = VLAN_DEF_HWIDTH;
	n = 1 << trunk->hwidth;
	trunk->hmask = n - 1;
	trunk->hash = malloc(sizeof(struct ifvlanhead) * n, M_VLAN, M_WAITOK);
	for (i = 0; i < n; i++)
		CK_SLIST_INIT(&trunk->hash[i]);
}

static void
vlan_freehash(struct ifvlantrunk *trunk)
{
#ifdef INVARIANTS
	int i;

	KASSERT(trunk->hwidth > 0, ("%s: hwidth not positive", __func__));
	for (i = 0; i < (1 << trunk->hwidth); i++)
		KASSERT(CK_SLIST_EMPTY(&trunk->hash[i]),
		    ("%s: hash table not empty", __func__));
#endif
	free(trunk->hash, M_VLAN);
	trunk->hash = NULL;
	trunk->hwidth = trunk->hmask = 0;
}

static int
vlan_inshash(struct ifvlantrunk *trunk, struct ifvlan *ifv)
{
	int i, b;
	struct ifvlan *ifv2;

	VLAN_XLOCK_ASSERT();
	KASSERT(trunk->hwidth > 0, ("%s: hwidth not positive", __func__));

	b = 1 << trunk->hwidth;
	i = HASH(ifv->ifv_vid, trunk->hmask);
	CK_SLIST_FOREACH(ifv2, &trunk->hash[i], ifv_list)
		if (ifv->ifv_vid == ifv2->ifv_vid)
			return (EEXIST);

	/*
	 * Grow the hash when the number of vlans exceeds half of the number of
	 * hash buckets squared. This will make the average linked-list length
	 * buckets/2.
	 */
	if (trunk->refcnt > (b * b) / 2) {
		vlan_growhash(trunk, 1);
		i = HASH(ifv->ifv_vid, trunk->hmask);
	}
	CK_SLIST_INSERT_HEAD(&trunk->hash[i], ifv, ifv_list);
	trunk->refcnt++;

	return (0);
}

static int
vlan_remhash(struct ifvlantrunk *trunk, struct ifvlan *ifv)
{
	int i, b;
	struct ifvlan *ifv2;

	VLAN_XLOCK_ASSERT();
	KASSERT(trunk->hwidth > 0, ("%s: hwidth not positive", __func__));

	b = 1 << (trunk->hwidth - 1);
	i = HASH(ifv->ifv_vid, trunk->hmask);
	CK_SLIST_FOREACH(ifv2, &trunk->hash[i], ifv_list)
		if (ifv2 == ifv) {
			trunk->refcnt--;
			CK_SLIST_REMOVE(&trunk->hash[i], ifv2, ifvlan, ifv_list);
			if (trunk->refcnt < (b * b) / 2)
				vlan_growhash(trunk, -1);
			return (0);
		}

	panic("%s: vlan not found\n", __func__);
	return (ENOENT); /*NOTREACHED*/
}

/*
 * Grow the hash larger or smaller if memory permits.
 */
static void
vlan_growhash(struct ifvlantrunk *trunk, int howmuch)
{
	struct ifvlan *ifv;
	struct ifvlanhead *hash2;
	int hwidth2, i, j, n, n2;

	VLAN_XLOCK_ASSERT();
	KASSERT(trunk->hwidth > 0, ("%s: hwidth not positive", __func__));

	if (howmuch == 0) {
		/* Harmless yet obvious coding error */
		printf("%s: howmuch is 0\n", __func__);
		return;
	}

	hwidth2 = trunk->hwidth + howmuch;
	n = 1 << trunk->hwidth;
	n2 = 1 << hwidth2;
	/* Do not shrink the table below the default */
	if (hwidth2 < VLAN_DEF_HWIDTH)
		return;

	hash2 = malloc(sizeof(struct ifvlanhead) * n2, M_VLAN, M_WAITOK);
	if (hash2 == NULL) {
		printf("%s: out of memory -- hash size not changed\n",
		    __func__);
		return;		/* We can live with the old hash table */
	}
	for (j = 0; j < n2; j++)
		CK_SLIST_INIT(&hash2[j]);
	for (i = 0; i < n; i++)
		while ((ifv = CK_SLIST_FIRST(&trunk->hash[i])) != NULL) {
			CK_SLIST_REMOVE(&trunk->hash[i], ifv, ifvlan, ifv_list);
			j = HASH(ifv->ifv_vid, n2 - 1);
			CK_SLIST_INSERT_HEAD(&hash2[j], ifv, ifv_list);
		}
	NET_EPOCH_WAIT();
	free(trunk->hash, M_VLAN);
	trunk->hash = hash2;
	trunk->hwidth = hwidth2;
	trunk->hmask = n2 - 1;

	if (bootverbose)
		if_printf(trunk->parent,
		    "VLAN hash table resized from %d to %d buckets\n", n, n2);
}

static __inline struct ifvlan *
vlan_gethash(struct ifvlantrunk *trunk, uint16_t vid)
{
	struct ifvlan *ifv;

	NET_EPOCH_ASSERT();

	CK_SLIST_FOREACH(ifv, &trunk->hash[HASH(vid, trunk->hmask)], ifv_list)
		if (ifv->ifv_vid == vid)
			return (ifv);
	return (NULL);
}

#if 0
/* Debugging code to view the hashtables. */
static void
vlan_dumphash(struct ifvlantrunk *trunk)
{
	int i;
	struct ifvlan *ifv;

	for (i = 0; i < (1 << trunk->hwidth); i++) {
		printf("%d: ", i);
		CK_SLIST_FOREACH(ifv, &trunk->hash[i], ifv_list)
			printf("%s ", ifv->ifv_ifp->if_xname);
		printf("\n");
	}
}
#endif /* 0 */
#else

static __inline struct ifvlan *
vlan_gethash(struct ifvlantrunk *trunk, uint16_t vid)
{

	return trunk->vlans[vid];
}

static __inline int
vlan_inshash(struct ifvlantrunk *trunk, struct ifvlan *ifv)
{

	if (trunk->vlans[ifv->ifv_vid] != NULL)
		return EEXIST;
	trunk->vlans[ifv->ifv_vid] = ifv;
	trunk->refcnt++;

	return (0);
}

static __inline int
vlan_remhash(struct ifvlantrunk *trunk, struct ifvlan *ifv)
{

	trunk->vlans[ifv->ifv_vid] = NULL;
	trunk->refcnt--;

	return (0);
}

static __inline void
vlan_freehash(struct ifvlantrunk *trunk)
{
}

static __inline void
vlan_inithash(struct ifvlantrunk *trunk)
{
}

#endif /* !VLAN_ARRAY */

static void
trunk_destroy(struct ifvlantrunk *trunk)
{
	VLAN_XLOCK_ASSERT();

	vlan_freehash(trunk);
	trunk->parent->if_vlantrunk = NULL;
	TRUNK_LOCK_DESTROY(trunk);
	if_rele(trunk->parent);
	free(trunk, M_VLAN);
}

/*
 * Program our multicast filter. What we're actually doing is
 * programming the multicast filter of the parent. This has the
 * side effect of causing the parent interface to receive multicast
 * traffic that it doesn't really want, which ends up being discarded
 * later by the upper protocol layers. Unfortunately, there's no way
 * to avoid this: there really is only one physical interface.
 */
static int
vlan_setmulti(struct ifnet *ifp)
{
	struct ifnet		*ifp_p;
	struct ifmultiaddr	*ifma;
	struct ifvlan		*sc;
	struct vlan_mc_entry	*mc;
	int			error;

	VLAN_XLOCK_ASSERT();

	/* Find the parent. */
	sc = ifp->if_softc;
	ifp_p = PARENT(sc);

	CURVNET_SET_QUIET(ifp_p->if_vnet);

	/* First, remove any existing filter entries. */
	while ((mc = CK_SLIST_FIRST(&sc->vlan_mc_listhead)) != NULL) {
		CK_SLIST_REMOVE_HEAD(&sc->vlan_mc_listhead, mc_entries);
		(void)if_delmulti(ifp_p, (struct sockaddr *)&mc->mc_addr);
		NET_EPOCH_CALL(vlan_mc_free, &mc->mc_epoch_ctx);
	}

	/* Now program new ones. */
	IF_ADDR_WLOCK(ifp);
	CK_STAILQ_FOREACH(ifma, &ifp->if_multiaddrs, ifma_link) {
		if (ifma->ifma_addr->sa_family != AF_LINK)
			continue;
		mc = malloc(sizeof(struct vlan_mc_entry), M_VLAN, M_NOWAIT);
		if (mc == NULL) {
			IF_ADDR_WUNLOCK(ifp);
			CURVNET_RESTORE();
			return (ENOMEM);
		}
		bcopy(ifma->ifma_addr, &mc->mc_addr, ifma->ifma_addr->sa_len);
		mc->mc_addr.sdl_index = ifp_p->if_index;
		CK_SLIST_INSERT_HEAD(&sc->vlan_mc_listhead, mc, mc_entries);
	}
	IF_ADDR_WUNLOCK(ifp);
	CK_SLIST_FOREACH (mc, &sc->vlan_mc_listhead, mc_entries) {
		error = if_addmulti(ifp_p, (struct sockaddr *)&mc->mc_addr,
		    NULL);
		if (error) {
			CURVNET_RESTORE();
			return (error);
		}
	}

	CURVNET_RESTORE();
	return (0);
}

/*
 * A handler for interface ifnet events.
 */
static void
vlan_ifevent(void *arg __unused, struct ifnet *ifp, int event)
{
	struct epoch_tracker et;
	struct ifvlan *ifv;
	struct ifvlantrunk *trunk;

	if (event != IFNET_EVENT_UPDATE_BAUDRATE)
		return;

	NET_EPOCH_ENTER(et);
	trunk = ifp->if_vlantrunk;
	if (trunk == NULL) {
		NET_EPOCH_EXIT(et);
		return;
	}

	TRUNK_WLOCK(trunk);
	VLAN_FOREACH(ifv, trunk) {
		ifv->ifv_ifp->if_baudrate = ifp->if_baudrate;
	}
	TRUNK_WUNLOCK(trunk);
	NET_EPOCH_EXIT(et);
}

/*
 * A handler for parent interface link layer address changes.
 * If the parent interface link layer address is changed we
 * should also change it on all children vlans.
 */
static void
vlan_iflladdr(void *arg __unused, struct ifnet *ifp)
{
	struct epoch_tracker et;
	struct ifvlan *ifv;
	struct ifnet *ifv_ifp;
	struct ifvlantrunk *trunk;
	struct sockaddr_dl *sdl;

	/* Need the epoch since this is run on taskqueue_swi. */
	NET_EPOCH_ENTER(et);
	trunk = ifp->if_vlantrunk;
	if (trunk == NULL) {
		NET_EPOCH_EXIT(et);
		return;
	}

	/*
	 * OK, it's a trunk.  Loop over and change all vlan's lladdrs on it.
	 * We need an exclusive lock here to prevent concurrent SIOCSIFLLADDR
	 * ioctl calls on the parent garbling the lladdr of the child vlan.
	 */
	TRUNK_WLOCK(trunk);
	VLAN_FOREACH(ifv, trunk) {
		/*
		 * Copy new new lladdr into the ifv_ifp, enqueue a task
		 * to actually call if_setlladdr. if_setlladdr needs to
		 * be deferred to a taskqueue because it will call into
		 * the if_vlan ioctl path and try to acquire the global
		 * lock.
		 */
		ifv_ifp = ifv->ifv_ifp;
		bcopy(IF_LLADDR(ifp), IF_LLADDR(ifv_ifp),
		    ifp->if_addrlen);
		sdl = (struct sockaddr_dl *)ifv_ifp->if_addr->ifa_addr;
		sdl->sdl_alen = ifp->if_addrlen;
		taskqueue_enqueue(taskqueue_thread, &ifv->lladdr_task);
	}
	TRUNK_WUNLOCK(trunk);
	NET_EPOCH_EXIT(et);
}

/*
 * A handler for network interface departure events.
 * Track departure of trunks here so that we don't access invalid
 * pointers or whatever if a trunk is ripped from under us, e.g.,
 * by ejecting its hot-plug card.  However, if an ifnet is simply
 * being renamed, then there's no need to tear down the state.
 */
static void
vlan_ifdetach(void *arg __unused, struct ifnet *ifp)
{
	struct ifvlan *ifv;
	struct ifvlantrunk *trunk;

	/* If the ifnet is just being renamed, don't do anything. */
	if (ifp->if_flags & IFF_RENAMING)
		return;
	VLAN_XLOCK();
	trunk = ifp->if_vlantrunk;
	if (trunk == NULL) {
		VLAN_XUNLOCK();
		return;
	}

	/*
	 * OK, it's a trunk.  Loop over and detach all vlan's on it.
	 * Check trunk pointer after each vlan_unconfig() as it will
	 * free it and set to NULL after the last vlan was detached.
	 */
	VLAN_FOREACH_UNTIL_SAFE(ifv, ifp->if_vlantrunk,
	    ifp->if_vlantrunk == NULL)
		vlan_unconfig_locked(ifv->ifv_ifp, 1);

	/* Trunk should have been destroyed in vlan_unconfig(). */
	KASSERT(ifp->if_vlantrunk == NULL, ("%s: purge failed", __func__));
	VLAN_XUNLOCK();
}

/*
 * Return the trunk device for a virtual interface.
 */
static struct ifnet  *
vlan_trunkdev(struct ifnet *ifp)
{
	struct ifvlan *ifv;

	NET_EPOCH_ASSERT();

	if (ifp->if_type != IFT_L2VLAN)
		return (NULL);

	ifv = ifp->if_softc;
	ifp = NULL;
	if (ifv->ifv_trunk)
		ifp = PARENT(ifv);
	return (ifp);
}

/*
 * Return the 12-bit VLAN VID for this interface, for use by external
 * components such as Infiniband.
 *
 * XXXRW: Note that the function name here is historical; it should be named
 * vlan_vid().
 */
static int
vlan_tag(struct ifnet *ifp, uint16_t *vidp)
{
	struct ifvlan *ifv;

	if (ifp->if_type != IFT_L2VLAN)
		return (EINVAL);
	ifv = ifp->if_softc;
	*vidp = ifv->ifv_vid;
	return (0);
}

static int
vlan_pcp(struct ifnet *ifp, uint16_t *pcpp)
{
	struct ifvlan *ifv;

	if (ifp->if_type != IFT_L2VLAN)
		return (EINVAL);
	ifv = ifp->if_softc;
	*pcpp = ifv->ifv_pcp;
	return (0);
}

/*
 * Return a driver specific cookie for this interface.  Synchronization
 * with setcookie must be provided by the driver.
 */
static void *
vlan_cookie(struct ifnet *ifp)
{
	struct ifvlan *ifv;

	if (ifp->if_type != IFT_L2VLAN)
		return (NULL);
	ifv = ifp->if_softc;
	return (ifv->ifv_cookie);
}

/*
 * Store a cookie in our softc that drivers can use to store driver
 * private per-instance data in.
 */
static int
vlan_setcookie(struct ifnet *ifp, void *cookie)
{
	struct ifvlan *ifv;

	if (ifp->if_type != IFT_L2VLAN)
		return (EINVAL);
	ifv = ifp->if_softc;
	ifv->ifv_cookie = cookie;
	return (0);
}

/*
 * Return the vlan device present at the specific VID.
 */
static struct ifnet *
vlan_devat(struct ifnet *ifp, uint16_t vid)
{
	struct ifvlantrunk *trunk;
	struct ifvlan *ifv;

	NET_EPOCH_ASSERT();

	trunk = ifp->if_vlantrunk;
	if (trunk == NULL)
		return (NULL);
	ifp = NULL;
	ifv = vlan_gethash(trunk, vid);
	if (ifv)
		ifp = ifv->ifv_ifp;
	return (ifp);
}

/*
 * VLAN support can be loaded as a module.  The only place in the
 * system that's intimately aware of this is ether_input.  We hook
 * into this code through vlan_input_p which is defined there and
 * set here.  No one else in the system should be aware of this so
 * we use an explicit reference here.
 */
extern	void (*vlan_input_p)(struct ifnet *, struct mbuf *);

/* For if_link_state_change() eyes only... */
extern	void (*vlan_link_state_p)(struct ifnet *);

static struct if_clone_addreq vlan_addreq = {
	.match_f = vlan_clone_match,
	.create_f = vlan_clone_create,
	.destroy_f = vlan_clone_destroy,
};

static int
vlan_modevent(module_t mod, int type, void *data)
{

	switch (type) {
	case MOD_LOAD:
		ifdetach_tag = EVENTHANDLER_REGISTER(ifnet_departure_event,
		    vlan_ifdetach, NULL, EVENTHANDLER_PRI_ANY);
		if (ifdetach_tag == NULL)
			return (ENOMEM);
		iflladdr_tag = EVENTHANDLER_REGISTER(iflladdr_event,
		    vlan_iflladdr, NULL, EVENTHANDLER_PRI_ANY);
		if (iflladdr_tag == NULL)
			return (ENOMEM);
		ifevent_tag = EVENTHANDLER_REGISTER(ifnet_event,
		    vlan_ifevent, NULL, EVENTHANDLER_PRI_ANY);
		if (ifevent_tag == NULL)
			return (ENOMEM);
		VLAN_LOCKING_INIT();
		vlan_input_p = vlan_input;
		vlan_link_state_p = vlan_link_state;
		vlan_trunk_cap_p = vlan_trunk_capabilities;
		vlan_trunkdev_p = vlan_trunkdev;
		vlan_cookie_p = vlan_cookie;
		vlan_setcookie_p = vlan_setcookie;
		vlan_tag_p = vlan_tag;
		vlan_pcp_p = vlan_pcp;
		vlan_devat_p = vlan_devat;
#ifndef VIMAGE
		vlan_cloner = ifc_attach_cloner(vlanname, &vlan_addreq);
#endif
		if (bootverbose)
			printf("vlan: initialized, using "
#ifdef VLAN_ARRAY
			       "full-size arrays"
#else
			       "hash tables with chaining"
#endif

			       "\n");
		break;
	case MOD_UNLOAD:
#ifndef VIMAGE
		ifc_detach_cloner(vlan_cloner);
#endif
		EVENTHANDLER_DEREGISTER(ifnet_departure_event, ifdetach_tag);
		EVENTHANDLER_DEREGISTER(iflladdr_event, iflladdr_tag);
		EVENTHANDLER_DEREGISTER(ifnet_event, ifevent_tag);
		vlan_input_p = NULL;
		vlan_link_state_p = NULL;
		vlan_trunk_cap_p = NULL;
		vlan_trunkdev_p = NULL;
		vlan_tag_p = NULL;
		vlan_cookie_p = NULL;
		vlan_setcookie_p = NULL;
		vlan_devat_p = NULL;
		VLAN_LOCKING_DESTROY();
		if (bootverbose)
			printf("vlan: unloaded\n");
		break;
	default:
		return (EOPNOTSUPP);
	}
	return (0);
}

static moduledata_t vlan_mod = {
	"if_vlan",
	vlan_modevent,
	0
};

DECLARE_MODULE(if_vlan, vlan_mod, SI_SUB_PSEUDO, SI_ORDER_ANY);
MODULE_VERSION(if_vlan, 3);

#ifdef VIMAGE
static void
vnet_vlan_init(const void *unused __unused)
{
	vlan_cloner = ifc_attach_cloner(vlanname, &vlan_addreq);
	V_vlan_cloner = vlan_cloner;
}
VNET_SYSINIT(vnet_vlan_init, SI_SUB_PROTO_IFATTACHDOMAIN, SI_ORDER_ANY,
    vnet_vlan_init, NULL);

static void
vnet_vlan_uninit(const void *unused __unused)
{

	ifc_detach_cloner(V_vlan_cloner);
}
VNET_SYSUNINIT(vnet_vlan_uninit, SI_SUB_INIT_IF, SI_ORDER_ANY,
    vnet_vlan_uninit, NULL);
#endif

/*
 * Check for <etherif>.<vlan>[.<vlan> ...] style interface names.
 */
static struct ifnet *
vlan_clone_match_ethervid(const char *name, int *vidp)
{
	char ifname[IFNAMSIZ];
	char *cp;
	struct ifnet *ifp;
	int vid;

	strlcpy(ifname, name, IFNAMSIZ);
	if ((cp = strrchr(ifname, '.')) == NULL)
		return (NULL);
	*cp = '\0';
	if ((ifp = ifunit_ref(ifname)) == NULL)
		return (NULL);
	/* Parse VID. */
	if (*++cp == '\0') {
		if_rele(ifp);
		return (NULL);
	}
	vid = 0;
	for(; *cp >= '0' && *cp <= '9'; cp++)
		vid = (vid * 10) + (*cp - '0');
	if (*cp != '\0') {
		if_rele(ifp);
		return (NULL);
	}
	if (vidp != NULL)
		*vidp = vid;

	return (ifp);
}

static int
vlan_clone_match(struct if_clone *ifc, const char *name)
{
	struct ifnet *ifp;
	const char *cp;

	ifp = vlan_clone_match_ethervid(name, NULL);
	if (ifp != NULL) {
		if_rele(ifp);
		return (1);
	}

	if (strncmp(vlanname, name, strlen(vlanname)) != 0)
		return (0);
	for (cp = name + 4; *cp != '\0'; cp++) {
		if (*cp < '0' || *cp > '9')
			return (0);
	}

	return (1);
}

static int
vlan_clone_create(struct if_clone *ifc, char *name, size_t len,
    struct ifc_data *ifd, struct ifnet **ifpp)
{
	char *dp;
	bool wildcard = false;
	bool subinterface = false;
	int unit;
	int error;
	int vid = 0;
	uint16_t proto = ETHERTYPE_VLAN;
	struct ifvlan *ifv;
	struct ifnet *ifp;
	struct ifnet *p = NULL;
	struct ifaddr *ifa;
	struct sockaddr_dl *sdl;
	struct vlanreq vlr;
	static const u_char eaddr[ETHER_ADDR_LEN];	/* 00:00:00:00:00:00 */


	/*
	 * There are three ways to specify the cloned device:
	 * o pass a parameter block with the clone request.
	 * o specify parameters in the text of the clone device name
	 * o specify no parameters and get an unattached device that
	 *   must be configured separately.
	 * The first technique is preferred; the latter two are supported
	 * for backwards compatibility.
	 *
	 * XXXRW: Note historic use of the word "tag" here.  New ioctls may be
	 * called for.
	 */

	if (ifd->params != NULL) {
		error = ifc_copyin(ifd, &vlr, sizeof(vlr));
		if (error)
			return error;
		vid = vlr.vlr_tag;
		proto = vlr.vlr_proto;

#ifdef COMPAT_FREEBSD12
		if (proto == 0)
			proto = ETHERTYPE_VLAN;
#endif
		p = ifunit_ref(vlr.vlr_parent);
		if (p == NULL)
			return (ENXIO);
	}

	if ((error = ifc_name2unit(name, &unit)) == 0) {

		/*
		 * vlanX interface. Set wildcard to true if the unit number
		 * is not fixed (-1)
		 */
		wildcard = (unit < 0);
	} else {
		struct ifnet *p_tmp = vlan_clone_match_ethervid(name, &vid);
		if (p_tmp != NULL) {
			error = 0;
			subinterface = true;
			unit = IF_DUNIT_NONE;
			wildcard = false;
			if (p != NULL) {
				if_rele(p_tmp);
				if (p != p_tmp)
					error = EINVAL;
			} else
				p = p_tmp;
		} else
			error = ENXIO;
	}

	if (error != 0) {
		if (p != NULL)
			if_rele(p);
		return (error);
	}

	if (!subinterface) {
		/* vlanX interface, mark X as busy or allocate new unit # */
		error = ifc_alloc_unit(ifc, &unit);
		if (error != 0) {
			if (p != NULL)
				if_rele(p);
			return (error);
		}
	}

	/* In the wildcard case, we need to update the name. */
	if (wildcard) {
		for (dp = name; *dp != '\0'; dp++);
		if (snprintf(dp, len - (dp-name), "%d", unit) >
		    len - (dp-name) - 1) {
			panic("%s: interface name too long", __func__);
		}
	}

	ifv = malloc(sizeof(struct ifvlan), M_VLAN, M_WAITOK | M_ZERO);
	ifp = ifv->ifv_ifp = if_alloc(IFT_ETHER);
	if (ifp == NULL) {
		if (!subinterface)
			ifc_free_unit(ifc, unit);
		free(ifv, M_VLAN);
		if (p != NULL)
			if_rele(p);
		return (ENOSPC);
	}
	CK_SLIST_INIT(&ifv->vlan_mc_listhead);
	ifp->if_softc = ifv;
	/*
	 * Set the name manually rather than using if_initname because
	 * we don't conform to the default naming convention for interfaces.
	 */
	strlcpy(ifp->if_xname, name, IFNAMSIZ);
	ifp->if_dname = vlanname;
	ifp->if_dunit = unit;

	ifp->if_init = vlan_init;
#ifdef ALTQ
	ifp->if_start = vlan_altq_start;
	ifp->if_transmit = vlan_altq_transmit;
	IFQ_SET_MAXLEN(&ifp->if_snd, ifqmaxlen);
	ifp->if_snd.ifq_drv_maxlen = 0;
	IFQ_SET_READY(&ifp->if_snd);
#else
	ifp->if_transmit = vlan_transmit;
#endif
	ifp->if_qflush = vlan_qflush;
	ifp->if_ioctl = vlan_ioctl;
#if defined(KERN_TLS) || defined(RATELIMIT)
	ifp->if_snd_tag_alloc = vlan_snd_tag_alloc;
	ifp->if_ratelimit_query = vlan_ratelimit_query;
#endif
	ifp->if_flags = VLAN_IFFLAGS;
	ether_ifattach(ifp, eaddr);
	/* Now undo some of the damage... */
	ifp->if_baudrate = 0;
	ifp->if_type = IFT_L2VLAN;
	ifp->if_hdrlen = ETHER_VLAN_ENCAP_LEN;
	ifa = ifp->if_addr;
	sdl = (struct sockaddr_dl *)ifa->ifa_addr;
	sdl->sdl_type = IFT_L2VLAN;

	if (p != NULL) {
		error = vlan_config(ifv, p, vid, proto);
		if_rele(p);
		if (error != 0) {
			/*
			 * Since we've partially failed, we need to back
			 * out all the way, otherwise userland could get
			 * confused.  Thus, we destroy the interface.
			 */
			ether_ifdetach(ifp);
			vlan_unconfig(ifp);
			if_free(ifp);
			if (!subinterface)
				ifc_free_unit(ifc, unit);
			free(ifv, M_VLAN);

			return (error);
		}
	}
	*ifpp = ifp;

	return (0);
}

static int
vlan_clone_destroy(struct if_clone *ifc, struct ifnet *ifp, uint32_t flags)
{
	struct ifvlan *ifv = ifp->if_softc;
	int unit = ifp->if_dunit;

	if (ifp->if_vlantrunk)
		return (EBUSY);

#ifdef ALTQ
	IFQ_PURGE(&ifp->if_snd);
#endif
	ether_ifdetach(ifp);	/* first, remove it from system-wide lists */
	vlan_unconfig(ifp);	/* now it can be unconfigured and freed */
	/*
	 * We should have the only reference to the ifv now, so we can now
	 * drain any remaining lladdr task before freeing the ifnet and the
	 * ifvlan.
	 */
	taskqueue_drain(taskqueue_thread, &ifv->lladdr_task);
	NET_EPOCH_WAIT();
	if_free(ifp);
	free(ifv, M_VLAN);
	if (unit != IF_DUNIT_NONE)
		ifc_free_unit(ifc, unit);

	return (0);
}

/*
 * The ifp->if_init entry point for vlan(4) is a no-op.
 */
static void
vlan_init(void *foo __unused)
{
}

/*
 * The if_transmit method for vlan(4) interface.
 */
static int
vlan_transmit(struct ifnet *ifp, struct mbuf *m)
{
	struct ifvlan *ifv;
	struct ifnet *p;
	int error, len, mcast;

	NET_EPOCH_ASSERT();

	ifv = ifp->if_softc;
	if (TRUNK(ifv) == NULL) {
		if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
		m_freem(m);
		return (ENETDOWN);
	}
	p = PARENT(ifv);
	len = m->m_pkthdr.len;
	mcast = (m->m_flags & (M_MCAST | M_BCAST)) ? 1 : 0;

	BPF_MTAP(ifp, m);

#if defined(KERN_TLS) || defined(RATELIMIT)
	if (m->m_pkthdr.csum_flags & CSUM_SND_TAG) {
		struct vlan_snd_tag *vst;
		struct m_snd_tag *mst;

		MPASS(m->m_pkthdr.snd_tag->ifp == ifp);
		mst = m->m_pkthdr.snd_tag;
		vst = mst_to_vst(mst);
		if (vst->tag->ifp != p) {
			if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
			m_freem(m);
			return (EAGAIN);
		}

		m->m_pkthdr.snd_tag = m_snd_tag_ref(vst->tag);
		m_snd_tag_rele(mst);
	}
#endif

	/*
	 * Do not run parent's if_transmit() if the parent is not up,
	 * or parent's driver will cause a system crash.
	 */
	if (!UP_AND_RUNNING(p)) {
		if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
		m_freem(m);
		return (ENETDOWN);
	}

	if (!ether_8021q_frame(&m, ifp, p, &ifv->ifv_qtag)) {
		if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
		return (0);
	}

	/*
	 * Send it, precisely as ether_output() would have.
	 */
	error = (p->if_transmit)(p, m);
	if (error == 0) {
		if_inc_counter(ifp, IFCOUNTER_OPACKETS, 1);
		if_inc_counter(ifp, IFCOUNTER_OBYTES, len);
		if_inc_counter(ifp, IFCOUNTER_OMCASTS, mcast);
	} else
		if_inc_counter(ifp, IFCOUNTER_OERRORS, 1);
	return (error);
}

static int
vlan_output(struct ifnet *ifp, struct mbuf *m, const struct sockaddr *dst,
    struct route *ro)
{
	struct ifvlan *ifv;
	struct ifnet *p;

	NET_EPOCH_ASSERT();

	/*
	 * Find the first non-VLAN parent interface.
	 */
	ifv = ifp->if_softc;
	do {
		if (TRUNK(ifv) == NULL) {
			m_freem(m);
			return (ENETDOWN);
		}
		p = PARENT(ifv);
		ifv = p->if_softc;
	} while (p->if_type == IFT_L2VLAN);

	return p->if_output(ifp, m, dst, ro);
}

#ifdef ALTQ
static void
vlan_altq_start(if_t ifp)
{
	struct ifaltq *ifq = &ifp->if_snd;
	struct mbuf *m;

	IFQ_LOCK(ifq);
	IFQ_DEQUEUE_NOLOCK(ifq, m);
	while (m != NULL) {
		vlan_transmit(ifp, m);
		IFQ_DEQUEUE_NOLOCK(ifq, m);
	}
	IFQ_UNLOCK(ifq);
}

static int
vlan_altq_transmit(if_t ifp, struct mbuf *m)
{
	int err;

	if (ALTQ_IS_ENABLED(&ifp->if_snd)) {
		IFQ_ENQUEUE(&ifp->if_snd, m, err);
		if (err == 0)
			vlan_altq_start(ifp);
	} else
		err = vlan_transmit(ifp, m);

	return (err);
}
#endif	/* ALTQ */

/*
 * The ifp->if_qflush entry point for vlan(4) is a no-op.
 */
static void
vlan_qflush(struct ifnet *ifp __unused)
{
}

static void
vlan_input(struct ifnet *ifp, struct mbuf *m)
{
	struct ifvlantrunk *trunk;
	struct ifvlan *ifv;
	struct m_tag *mtag;
	uint16_t vid, tag;

	NET_EPOCH_ASSERT();

	trunk = ifp->if_vlantrunk;
	if (trunk == NULL) {
		m_freem(m);
		return;
	}

	if (m->m_flags & M_VLANTAG) {
		/*
		 * Packet is tagged, but m contains a normal
		 * Ethernet frame; the tag is stored out-of-band.
		 */
		tag = m->m_pkthdr.ether_vtag;
		m->m_flags &= ~M_VLANTAG;
	} else {
		struct ether_vlan_header *evl;

		/*
		 * Packet is tagged in-band as specified by 802.1q.
		 */
		switch (ifp->if_type) {
		case IFT_ETHER:
			if (m->m_len < sizeof(*evl) &&
			    (m = m_pullup(m, sizeof(*evl))) == NULL) {
				if_printf(ifp, "cannot pullup VLAN header\n");
				return;
			}
			evl = mtod(m, struct ether_vlan_header *);
			tag = ntohs(evl->evl_tag);

			/*
			 * Remove the 802.1q header by copying the Ethernet
			 * addresses over it and adjusting the beginning of
			 * the data in the mbuf.  The encapsulated Ethernet
			 * type field is already in place.
			 */
			bcopy((char *)evl, (char *)evl + ETHER_VLAN_ENCAP_LEN,
			      ETHER_HDR_LEN - ETHER_TYPE_LEN);
			m_adj(m, ETHER_VLAN_ENCAP_LEN);
			break;

		default:
#ifdef INVARIANTS
			panic("%s: %s has unsupported if_type %u",
			      __func__, ifp->if_xname, ifp->if_type);
#endif
			if_inc_counter(ifp, IFCOUNTER_NOPROTO, 1);
			m_freem(m);
			return;
		}
	}

	vid = EVL_VLANOFTAG(tag);

	ifv = vlan_gethash(trunk, vid);
	if (ifv == NULL || !UP_AND_RUNNING(ifv->ifv_ifp)) {
		if_inc_counter(ifp, IFCOUNTER_NOPROTO, 1);
		m_freem(m);
		return;
	}

	if (V_vlan_mtag_pcp) {
		/*
		 * While uncommon, it is possible that we will find a 802.1q
		 * packet encapsulated inside another packet that also had an
		 * 802.1q header.  For example, ethernet tunneled over IPSEC
		 * arriving over ethernet.  In that case, we replace the
		 * existing 802.1q PCP m_tag value.
		 */
		mtag = m_tag_locate(m, MTAG_8021Q, MTAG_8021Q_PCP_IN, NULL);
		if (mtag == NULL) {
			mtag = m_tag_alloc(MTAG_8021Q, MTAG_8021Q_PCP_IN,
			    sizeof(uint8_t), M_NOWAIT);
			if (mtag == NULL) {
				if_inc_counter(ifp, IFCOUNTER_IERRORS, 1);
				m_freem(m);
				return;
			}
			m_tag_prepend(m, mtag);
		}
		*(uint8_t *)(mtag + 1) = EVL_PRIOFTAG(tag);
	}

	m->m_pkthdr.rcvif = ifv->ifv_ifp;
	if_inc_counter(ifv->ifv_ifp, IFCOUNTER_IPACKETS, 1);

	/* Pass it back through the parent's input routine. */
	(*ifv->ifv_ifp->if_input)(ifv->ifv_ifp, m);
}

static void
vlan_lladdr_fn(void *arg, int pending __unused)
{
	struct ifvlan *ifv;
	struct ifnet *ifp;

	ifv = (struct ifvlan *)arg;
	ifp = ifv->ifv_ifp;

	CURVNET_SET(ifp->if_vnet);

	/* The ifv_ifp already has the lladdr copied in. */
	if_setlladdr(ifp, IF_LLADDR(ifp), ifp->if_addrlen);

	CURVNET_RESTORE();
}

static int
vlan_config(struct ifvlan *ifv, struct ifnet *p, uint16_t vid,
	uint16_t proto)
{
	struct epoch_tracker et;
	struct ifvlantrunk *trunk;
	struct ifnet *ifp;
	int error = 0;

	/*
	 * We can handle non-ethernet hardware types as long as
	 * they handle the tagging and headers themselves.
	 */
	if (p->if_type != IFT_ETHER &&
	    p->if_type != IFT_L2VLAN &&
	    (p->if_capenable & IFCAP_VLAN_HWTAGGING) == 0)
		return (EPROTONOSUPPORT);
	if ((p->if_flags & VLAN_IFFLAGS) != VLAN_IFFLAGS)
		return (EPROTONOSUPPORT);
	/*
	 * Don't let the caller set up a VLAN VID with
	 * anything except VLID bits.
	 * VID numbers 0x0 and 0xFFF are reserved.
	 */
	if (vid == 0 || vid == 0xFFF || (vid & ~EVL_VLID_MASK))
		return (EINVAL);
	if (ifv->ifv_trunk) {
		trunk = ifv->ifv_trunk;
		if (trunk->parent != p)
			return (EBUSY);

		VLAN_XLOCK();

		ifv->ifv_proto = proto;

		if (ifv->ifv_vid != vid) {
			/* Re-hash */
			vlan_remhash(trunk, ifv);
			ifv->ifv_vid = vid;
			error = vlan_inshash(trunk, ifv);
		}
		/* Will unlock */
		goto done;
	}

	VLAN_XLOCK();
	if (p->if_vlantrunk == NULL) {
		trunk = malloc(sizeof(struct ifvlantrunk),
		    M_VLAN, M_WAITOK | M_ZERO);
		vlan_inithash(trunk);
		TRUNK_LOCK_INIT(trunk);
		TRUNK_WLOCK(trunk);
		p->if_vlantrunk = trunk;
		trunk->parent = p;
		if_ref(trunk->parent);
		TRUNK_WUNLOCK(trunk);
	} else {
		trunk = p->if_vlantrunk;
	}

	ifv->ifv_vid = vid;	/* must set this before vlan_inshash() */
	ifv->ifv_pcp = 0;       /* Default: best effort delivery. */
	error = vlan_inshash(trunk, ifv);
	if (error)
		goto done;
	ifv->ifv_proto = proto;
	ifv->ifv_encaplen = ETHER_VLAN_ENCAP_LEN;
	ifv->ifv_mintu = ETHERMIN;
	ifv->ifv_pflags = 0;
	ifv->ifv_capenable = -1;

	/*
	 * If the parent supports the VLAN_MTU capability,
	 * i.e. can Tx/Rx larger than ETHER_MAX_LEN frames,
	 * use it.
	 */
	if (p->if_capenable & IFCAP_VLAN_MTU) {
		/*
		 * No need to fudge the MTU since the parent can
		 * handle extended frames.
		 */
		ifv->ifv_mtufudge = 0;
	} else {
		/*
		 * Fudge the MTU by the encapsulation size.  This
		 * makes us incompatible with strictly compliant
		 * 802.1Q implementations, but allows us to use
		 * the feature with other NetBSD implementations,
		 * which might still be useful.
		 */
		ifv->ifv_mtufudge = ifv->ifv_encaplen;
	}

	ifv->ifv_trunk = trunk;
	ifp = ifv->ifv_ifp;
	/*
	 * Initialize fields from our parent.  This duplicates some
	 * work with ether_ifattach() but allows for non-ethernet
	 * interfaces to also work.
	 */
	ifp->if_mtu = p->if_mtu - ifv->ifv_mtufudge;
	ifp->if_baudrate = p->if_baudrate;
	ifp->if_input = p->if_input;
	ifp->if_resolvemulti = p->if_resolvemulti;
	ifp->if_addrlen = p->if_addrlen;
	ifp->if_broadcastaddr = p->if_broadcastaddr;
	ifp->if_pcp = ifv->ifv_pcp;

	/*
	 * We wrap the parent's if_output using vlan_output to ensure that it
	 * can't become stale.
	 */
	ifp->if_output = vlan_output;

	/*
	 * Copy only a selected subset of flags from the parent.
	 * Other flags are none of our business.
	 */
#define VLAN_COPY_FLAGS (IFF_SIMPLEX)
	ifp->if_flags &= ~VLAN_COPY_FLAGS;
	ifp->if_flags |= p->if_flags & VLAN_COPY_FLAGS;
#undef VLAN_COPY_FLAGS

	ifp->if_link_state = p->if_link_state;

	NET_EPOCH_ENTER(et);
	vlan_capabilities(ifv);
	NET_EPOCH_EXIT(et);

	/*
	 * Set up our interface address to reflect the underlying
	 * physical interface's.
	 */
	TASK_INIT(&ifv->lladdr_task, 0, vlan_lladdr_fn, ifv);
	((struct sockaddr_dl *)ifp->if_addr->ifa_addr)->sdl_alen =
	    p->if_addrlen;

	/*
	 * Do not schedule link address update if it was the same
	 * as previous parent's. This helps avoid updating for each
	 * associated llentry.
	 */
	if (memcmp(IF_LLADDR(p), IF_LLADDR(ifp), p->if_addrlen) != 0) {
		bcopy(IF_LLADDR(p), IF_LLADDR(ifp), p->if_addrlen);
		taskqueue_enqueue(taskqueue_thread, &ifv->lladdr_task);
	}

	/* We are ready for operation now. */
	ifp->if_drv_flags |= IFF_DRV_RUNNING;

	/* Update flags on the parent, if necessary. */
	vlan_setflags(ifp, 1);

	/*
	 * Configure multicast addresses that may already be
	 * joined on the vlan device.
	 */
	(void)vlan_setmulti(ifp);

done:
	if (error == 0)
		EVENTHANDLER_INVOKE(vlan_config, p, ifv->ifv_vid);
	VLAN_XUNLOCK();

	return (error);
}

static void
vlan_unconfig(struct ifnet *ifp)
{

	VLAN_XLOCK();
	vlan_unconfig_locked(ifp, 0);
	VLAN_XUNLOCK();
}

static void
vlan_unconfig_locked(struct ifnet *ifp, int departing)
{
	struct ifvlantrunk *trunk;
	struct vlan_mc_entry *mc;
	struct ifvlan *ifv;
	struct ifnet  *parent;
	int error;

	VLAN_XLOCK_ASSERT();

	ifv = ifp->if_softc;
	trunk = ifv->ifv_trunk;
	parent = NULL;

	if (trunk != NULL) {
		parent = trunk->parent;

		/*
		 * Since the interface is being unconfigured, we need to
		 * empty the list of multicast groups that we may have joined
		 * while we were alive from the parent's list.
		 */
		while ((mc = CK_SLIST_FIRST(&ifv->vlan_mc_listhead)) != NULL) {
			/*
			 * If the parent interface is being detached,
			 * all its multicast addresses have already
			 * been removed.  Warn about errors if
			 * if_delmulti() does fail, but don't abort as
			 * all callers expect vlan destruction to
			 * succeed.
			 */
			if (!departing) {
				error = if_delmulti(parent,
				    (struct sockaddr *)&mc->mc_addr);
				if (error)
					if_printf(ifp,
		    "Failed to delete multicast address from parent: %d\n",
					    error);
			}
			CK_SLIST_REMOVE_HEAD(&ifv->vlan_mc_listhead, mc_entries);
			NET_EPOCH_CALL(vlan_mc_free, &mc->mc_epoch_ctx);
		}

		vlan_setflags(ifp, 0); /* clear special flags on parent */

		vlan_remhash(trunk, ifv);
		ifv->ifv_trunk = NULL;

		/*
		 * Check if we were the last.
		 */
		if (trunk->refcnt == 0) {
			parent->if_vlantrunk = NULL;
			NET_EPOCH_WAIT();
			trunk_destroy(trunk);
		}
	}

	/* Disconnect from parent. */
	if (ifv->ifv_pflags)
		if_printf(ifp, "%s: ifv_pflags unclean\n", __func__);
	ifp->if_mtu = ETHERMTU;
	ifp->if_link_state = LINK_STATE_UNKNOWN;
	ifp->if_drv_flags &= ~IFF_DRV_RUNNING;

	/*
	 * Only dispatch an event if vlan was
	 * attached, otherwise there is nothing
	 * to cleanup anyway.
	 */
	if (parent != NULL)
		EVENTHANDLER_INVOKE(vlan_unconfig, parent, ifv->ifv_vid);
}

/* Handle a reference counted flag that should be set on the parent as well */
static int
vlan_setflag(struct ifnet *ifp, int flag, int status,
	     int (*func)(struct ifnet *, int))
{
	struct ifvlan *ifv;
	int error;

	VLAN_SXLOCK_ASSERT();

	ifv = ifp->if_softc;
	status = status ? (ifp->if_flags & flag) : 0;
	/* Now "status" contains the flag value or 0 */

	/*
	 * See if recorded parent's status is different from what
	 * we want it to be.  If it is, flip it.  We record parent's
	 * status in ifv_pflags so that we won't clear parent's flag
	 * we haven't set.  In fact, we don't clear or set parent's
	 * flags directly, but get or release references to them.
	 * That's why we can be sure that recorded flags still are
	 * in accord with actual parent's flags.
	 */
	if (status != (ifv->ifv_pflags & flag)) {
		error = (*func)(PARENT(ifv), status);
		if (error)
			return (error);
		ifv->ifv_pflags &= ~flag;
		ifv->ifv_pflags |= status;
	}
	return (0);
}

/*
 * Handle IFF_* flags that require certain changes on the parent:
 * if "status" is true, update parent's flags respective to our if_flags;
 * if "status" is false, forcedly clear the flags set on parent.
 */
static int
vlan_setflags(struct ifnet *ifp, int status)
{
	int error, i;

	for (i = 0; vlan_pflags[i].flag; i++) {
		error = vlan_setflag(ifp, vlan_pflags[i].flag,
				     status, vlan_pflags[i].func);
		if (error)
			return (error);
	}
	return (0);
}

/* Inform all vlans that their parent has changed link state */
static void
vlan_link_state(struct ifnet *ifp)
{
	struct epoch_tracker et;
	struct ifvlantrunk *trunk;
	struct ifvlan *ifv;

	NET_EPOCH_ENTER(et);
	trunk = ifp->if_vlantrunk;
	if (trunk == NULL) {
		NET_EPOCH_EXIT(et);
		return;
	}

	TRUNK_WLOCK(trunk);
	VLAN_FOREACH(ifv, trunk) {
		ifv->ifv_ifp->if_baudrate = trunk->parent->if_baudrate;
		if_link_state_change(ifv->ifv_ifp,
		    trunk->parent->if_link_state);
	}
	TRUNK_WUNLOCK(trunk);
	NET_EPOCH_EXIT(et);
}

static void
vlan_capabilities(struct ifvlan *ifv)
{
	struct ifnet *p;
	struct ifnet *ifp;
	struct ifnet_hw_tsomax hw_tsomax;
	int cap = 0, ena = 0, mena;
	u_long hwa = 0;

	NET_EPOCH_ASSERT();
	VLAN_SXLOCK_ASSERT();

	p = PARENT(ifv);
	ifp = ifv->ifv_ifp;

	/* Mask parent interface enabled capabilities disabled by user. */
	mena = p->if_capenable & ifv->ifv_capenable;

	/*
	 * If the parent interface can do checksum offloading
	 * on VLANs, then propagate its hardware-assisted
	 * checksumming flags. Also assert that checksum
	 * offloading requires hardware VLAN tagging.
	 */
	if (p->if_capabilities & IFCAP_VLAN_HWCSUM)
		cap |= p->if_capabilities & (IFCAP_HWCSUM | IFCAP_HWCSUM_IPV6);
	if (p->if_capenable & IFCAP_VLAN_HWCSUM &&
	    p->if_capenable & IFCAP_VLAN_HWTAGGING) {
		ena |= mena & (IFCAP_HWCSUM | IFCAP_HWCSUM_IPV6);
		if (ena & IFCAP_TXCSUM)
			hwa |= p->if_hwassist & (CSUM_IP | CSUM_TCP |
			    CSUM_UDP | CSUM_SCTP);
		if (ena & IFCAP_TXCSUM_IPV6)
			hwa |= p->if_hwassist & (CSUM_TCP_IPV6 |
			    CSUM_UDP_IPV6 | CSUM_SCTP_IPV6);
	}

	/*
	 * If the parent interface can do TSO on VLANs then
	 * propagate the hardware-assisted flag. TSO on VLANs
	 * does not necessarily require hardware VLAN tagging.
	 */
	memset(&hw_tsomax, 0, sizeof(hw_tsomax));
	if_hw_tsomax_common(p, &hw_tsomax);
	if_hw_tsomax_update(ifp, &hw_tsomax);
	if (p->if_capabilities & IFCAP_VLAN_HWTSO)
		cap |= p->if_capabilities & IFCAP_TSO;
	if (p->if_capenable & IFCAP_VLAN_HWTSO) {
		ena |= mena & IFCAP_TSO;
		if (ena & IFCAP_TSO)
			hwa |= p->if_hwassist & CSUM_TSO;
	}

	/*
	 * If the parent interface can do LRO and checksum offloading on
	 * VLANs, then guess it may do LRO on VLANs.  False positive here
	 * cost nothing, while false negative may lead to some confusions.
	 */
	if (p->if_capabilities & IFCAP_VLAN_HWCSUM)
		cap |= p->if_capabilities & IFCAP_LRO;
	if (p->if_capenable & IFCAP_VLAN_HWCSUM)
		ena |= p->if_capenable & IFCAP_LRO;

	/*
	 * If the parent interface can offload TCP connections over VLANs then
	 * propagate its TOE capability to the VLAN interface.
	 *
	 * All TOE drivers in the tree today can deal with VLANs.  If this
	 * changes then IFCAP_VLAN_TOE should be promoted to a full capability
	 * with its own bit.
	 */
#define	IFCAP_VLAN_TOE IFCAP_TOE
	if (p->if_capabilities & IFCAP_VLAN_TOE)
		cap |= p->if_capabilities & IFCAP_TOE;
	if (p->if_capenable & IFCAP_VLAN_TOE) {
		SETTOEDEV(ifp, TOEDEV(p));
		ena |= mena & IFCAP_TOE;
	}

	/*
	 * If the parent interface supports dynamic link state, so does the
	 * VLAN interface.
	 */
	cap |= (p->if_capabilities & IFCAP_LINKSTATE);
	ena |= (mena & IFCAP_LINKSTATE);

#ifdef RATELIMIT
	/*
	 * If the parent interface supports ratelimiting, so does the
	 * VLAN interface.
	 */
	cap |= (p->if_capabilities & IFCAP_TXRTLMT);
	ena |= (mena & IFCAP_TXRTLMT);
#endif

	/*
	 * If the parent interface supports unmapped mbufs, so does
	 * the VLAN interface.  Note that this should be fine even for
	 * interfaces that don't support hardware tagging as headers
	 * are prepended in normal mbufs to unmapped mbufs holding
	 * payload data.
	 */
	cap |= (p->if_capabilities & IFCAP_MEXTPG);
	ena |= (mena & IFCAP_MEXTPG);

	/*
	 * If the parent interface can offload encryption and segmentation
	 * of TLS records over TCP, propagate it's capability to the VLAN
	 * interface.
	 *
	 * All TLS drivers in the tree today can deal with VLANs.  If
	 * this ever changes, then a new IFCAP_VLAN_TXTLS can be
	 * defined.
	 */
	if (p->if_capabilities & (IFCAP_TXTLS | IFCAP_TXTLS_RTLMT))
		cap |= p->if_capabilities & (IFCAP_TXTLS | IFCAP_TXTLS_RTLMT);
	if (p->if_capenable & (IFCAP_TXTLS | IFCAP_TXTLS_RTLMT))
		ena |= mena & (IFCAP_TXTLS | IFCAP_TXTLS_RTLMT);

	ifp->if_capabilities = cap;
	ifp->if_capenable = ena;
	ifp->if_hwassist = hwa;
}

static void
vlan_trunk_capabilities(struct ifnet *ifp)
{
	struct epoch_tracker et;
	struct ifvlantrunk *trunk;
	struct ifvlan *ifv;

	VLAN_SLOCK();
	trunk = ifp->if_vlantrunk;
	if (trunk == NULL) {
		VLAN_SUNLOCK();
		return;
	}
	NET_EPOCH_ENTER(et);
	VLAN_FOREACH(ifv, trunk)
		vlan_capabilities(ifv);
	NET_EPOCH_EXIT(et);
	VLAN_SUNLOCK();
}

static int
vlan_ioctl(struct ifnet *ifp, u_long cmd, caddr_t data)
{
	struct ifnet *p;
	struct ifreq *ifr;
#ifdef INET
	struct ifaddr *ifa;
#endif
	struct ifvlan *ifv;
	struct ifvlantrunk *trunk;
	struct vlanreq vlr;
	int error = 0, oldmtu;

	ifr = (struct ifreq *)data;
#ifdef INET
	ifa = (struct ifaddr *) data;
#endif
	ifv = ifp->if_softc;

	switch (cmd) {
	case SIOCSIFADDR:
		ifp->if_flags |= IFF_UP;
#ifdef INET
		if (ifa->ifa_addr->sa_family == AF_INET)
			arp_ifinit(ifp, ifa);
#endif
		break;
	case SIOCGIFADDR:
		bcopy(IF_LLADDR(ifp), &ifr->ifr_addr.sa_data[0],
		    ifp->if_addrlen);
		break;
	case SIOCGIFMEDIA:
		VLAN_SLOCK();
		if (TRUNK(ifv) != NULL) {
			p = PARENT(ifv);
			if_ref(p);
			error = (*p->if_ioctl)(p, SIOCGIFMEDIA, data);
			if_rele(p);
			/* Limit the result to the parent's current config. */
			if (error == 0) {
				struct ifmediareq *ifmr;

				ifmr = (struct ifmediareq *)data;
				if (ifmr->ifm_count >= 1 && ifmr->ifm_ulist) {
					ifmr->ifm_count = 1;
					error = copyout(&ifmr->ifm_current,
						ifmr->ifm_ulist,
						sizeof(int));
				}
			}
		} else {
			error = EINVAL;
		}
		VLAN_SUNLOCK();
		break;

	case SIOCSIFMEDIA:
		error = EINVAL;
		break;

	case SIOCSIFMTU:
		/*
		 * Set the interface MTU.
		 */
		VLAN_SLOCK();
		trunk = TRUNK(ifv);
		if (trunk != NULL) {
			TRUNK_WLOCK(trunk);
			if (ifr->ifr_mtu >
			     (PARENT(ifv)->if_mtu - ifv->ifv_mtufudge) ||
			    ifr->ifr_mtu <
			     (ifv->ifv_mintu - ifv->ifv_mtufudge))
				error = EINVAL;
			else
				ifp->if_mtu = ifr->ifr_mtu;
			TRUNK_WUNLOCK(trunk);
		} else
			error = EINVAL;
		VLAN_SUNLOCK();
		break;

	case SIOCSETVLAN:
#ifdef VIMAGE
		/*
		 * XXXRW/XXXBZ: The goal in these checks is to allow a VLAN
		 * interface to be delegated to a jail without allowing the
		 * jail to change what underlying interface/VID it is
		 * associated with.  We are not entirely convinced that this
		 * is the right way to accomplish that policy goal.
		 */
		if (ifp->if_vnet != ifp->if_home_vnet) {
			error = EPERM;
			break;
		}
#endif
		error = copyin(ifr_data_get_ptr(ifr), &vlr, sizeof(vlr));
		if (error)
			break;
		if (vlr.vlr_parent[0] == '\0') {
			vlan_unconfig(ifp);
			break;
		}
		p = ifunit_ref(vlr.vlr_parent);
		if (p == NULL) {
			error = ENOENT;
			break;
		}
#ifdef COMPAT_FREEBSD12
		if (vlr.vlr_proto == 0)
			vlr.vlr_proto = ETHERTYPE_VLAN;
#endif
		oldmtu = ifp->if_mtu;
		error = vlan_config(ifv, p, vlr.vlr_tag, vlr.vlr_proto);
		if_rele(p);

		/*
		 * VLAN MTU may change during addition of the vlandev.
		 * If it did, do network layer specific procedure.
		 */
		if (ifp->if_mtu != oldmtu)
			if_notifymtu(ifp);
		break;

	case SIOCGETVLAN:
#ifdef VIMAGE
		if (ifp->if_vnet != ifp->if_home_vnet) {
			error = EPERM;
			break;
		}
#endif
		bzero(&vlr, sizeof(vlr));
		VLAN_SLOCK();
		if (TRUNK(ifv) != NULL) {
			strlcpy(vlr.vlr_parent, PARENT(ifv)->if_xname,
			    sizeof(vlr.vlr_parent));
			vlr.vlr_tag = ifv->ifv_vid;
			vlr.vlr_proto = ifv->ifv_proto;
		}
		VLAN_SUNLOCK();
		error = copyout(&vlr, ifr_data_get_ptr(ifr), sizeof(vlr));
		break;

	case SIOCSIFFLAGS:
		/*
		 * We should propagate selected flags to the parent,
		 * e.g., promiscuous mode.
		 */
		VLAN_XLOCK();
		if (TRUNK(ifv) != NULL)
			error = vlan_setflags(ifp, 1);
		VLAN_XUNLOCK();
		break;

	case SIOCADDMULTI:
	case SIOCDELMULTI:
		/*
		 * If we don't have a parent, just remember the membership for
		 * when we do.
		 *
		 * XXX We need the rmlock here to avoid sleeping while
		 * holding in6_multi_mtx.
		 */
		VLAN_XLOCK();
		trunk = TRUNK(ifv);
		if (trunk != NULL)
			error = vlan_setmulti(ifp);
		VLAN_XUNLOCK();

		break;
	case SIOCGVLANPCP:
#ifdef VIMAGE
		if (ifp->if_vnet != ifp->if_home_vnet) {
			error = EPERM;
			break;
		}
#endif
		ifr->ifr_vlan_pcp = ifv->ifv_pcp;
		break;

	case SIOCSVLANPCP:
#ifdef VIMAGE
		if (ifp->if_vnet != ifp->if_home_vnet) {
			error = EPERM;
			break;
		}
#endif
		error = priv_check(curthread, PRIV_NET_SETVLANPCP);
		if (error)
			break;
		if (ifr->ifr_vlan_pcp > VLAN_PCP_MAX) {
			error = EINVAL;
			break;
		}
		ifv->ifv_pcp = ifr->ifr_vlan_pcp;
		ifp->if_pcp = ifv->ifv_pcp;
		/* broadcast event about PCP change */
		EVENTHANDLER_INVOKE(ifnet_event, ifp, IFNET_EVENT_PCP);
		break;

	case SIOCSIFCAP:
		VLAN_SLOCK();
		ifv->ifv_capenable = ifr->ifr_reqcap;
		trunk = TRUNK(ifv);
		if (trunk != NULL) {
			struct epoch_tracker et;

			NET_EPOCH_ENTER(et);
			vlan_capabilities(ifv);
			NET_EPOCH_EXIT(et);
		}
		VLAN_SUNLOCK();
		break;

	default:
		error = EINVAL;
		break;
	}

	return (error);
}

#if defined(KERN_TLS) || defined(RATELIMIT)
static int
vlan_snd_tag_alloc(struct ifnet *ifp,
    union if_snd_tag_alloc_params *params,
    struct m_snd_tag **ppmt)
{
	struct epoch_tracker et;
	const struct if_snd_tag_sw *sw;
	struct vlan_snd_tag *vst;
	struct ifvlan *ifv;
	struct ifnet *parent;
	struct m_snd_tag *mst;
	int error;

	NET_EPOCH_ENTER(et);
	ifv = ifp->if_softc;

	switch (params->hdr.type) {
#ifdef RATELIMIT
	case IF_SND_TAG_TYPE_UNLIMITED:
		sw = &vlan_snd_tag_ul_sw;
		break;
	case IF_SND_TAG_TYPE_RATE_LIMIT:
		sw = &vlan_snd_tag_rl_sw;
		break;
#endif
#ifdef KERN_TLS
	case IF_SND_TAG_TYPE_TLS:
		sw = &vlan_snd_tag_tls_sw;
		break;
	case IF_SND_TAG_TYPE_TLS_RX:
		sw = NULL;
		if (params->tls_rx.vlan_id != 0)
			goto failure;
		params->tls_rx.vlan_id = ifv->ifv_vid;
		break;
#ifdef RATELIMIT
	case IF_SND_TAG_TYPE_TLS_RATE_LIMIT:
		sw = &vlan_snd_tag_tls_rl_sw;
		break;
#endif
#endif
	default:
		goto failure;
	}

	if (ifv->ifv_trunk != NULL)
		parent = PARENT(ifv);
	else
		parent = NULL;
	if (parent == NULL)
		goto failure;
	if_ref(parent);
	NET_EPOCH_EXIT(et);

	if (sw != NULL) {
		vst = malloc(sizeof(*vst), M_VLAN, M_NOWAIT);
		if (vst == NULL) {
			if_rele(parent);
			return (ENOMEM);
		}
	} else
		vst = NULL;

	error = m_snd_tag_alloc(parent, params, &mst);
	if_rele(parent);
	if (error) {
		free(vst, M_VLAN);
		return (error);
	}

	if (sw != NULL) {
		m_snd_tag_init(&vst->com, ifp, sw);
		vst->tag = mst;

		*ppmt = &vst->com;
	} else
		*ppmt = mst;

	return (0);
failure:
	NET_EPOCH_EXIT(et);
	return (EOPNOTSUPP);
}

static struct m_snd_tag *
vlan_next_snd_tag(struct m_snd_tag *mst)
{
	struct vlan_snd_tag *vst;

	vst = mst_to_vst(mst);
	return (vst->tag);
}

static int
vlan_snd_tag_modify(struct m_snd_tag *mst,
    union if_snd_tag_modify_params *params)
{
	struct vlan_snd_tag *vst;

	vst = mst_to_vst(mst);
	return (vst->tag->sw->snd_tag_modify(vst->tag, params));
}

static int
vlan_snd_tag_query(struct m_snd_tag *mst,
    union if_snd_tag_query_params *params)
{
	struct vlan_snd_tag *vst;

	vst = mst_to_vst(mst);
	return (vst->tag->sw->snd_tag_query(vst->tag, params));
}

static void
vlan_snd_tag_free(struct m_snd_tag *mst)
{
	struct vlan_snd_tag *vst;

	vst = mst_to_vst(mst);
	m_snd_tag_rele(vst->tag);
	free(vst, M_VLAN);
}

static void
vlan_ratelimit_query(struct ifnet *ifp __unused, struct if_ratelimit_query_results *q)
{
	/*
	 * For vlan, we have an indirect
	 * interface. The caller needs to
	 * get a ratelimit tag on the actual
	 * interface the flow will go on.
	 */
	q->rate_table = NULL;
	q->flags = RT_IS_INDIRECT;
	q->max_flows = 0;
	q->number_of_rates = 0;
}

#endif