summaryrefslogtreecommitdiff
path: root/lib/netdev.c
blob: c797783782f89349da441dc888ad937ed339bca1 (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
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
/*
 * Copyright (c) 2008, 2009, 2010, 2011, 2012, 2013, 2014, 2016, 2017 Nicira, Inc.
 *
 * Licensed under the Apache License, Version 2.0 (the "License");
 * you may not use this file except in compliance with the License.
 * You may obtain a copy of the License at:
 *
 *     http://www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an "AS IS" BASIS,
 * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include <config.h>
#include "netdev.h"

#include <errno.h>
#include <inttypes.h>
#include <sys/types.h>
#include <netinet/in.h>
#include <stdlib.h>
#include <string.h>
#include <unistd.h>

#ifndef _WIN32
#include <ifaddrs.h>
#include <net/if.h>
#include <sys/ioctl.h>
#endif

#include "cmap.h"
#include "coverage.h"
#include "dpif.h"
#include "dp-packet.h"
#include "openvswitch/dynamic-string.h"
#include "fatal-signal.h"
#include "hash.h"
#include "openvswitch/list.h"
#include "netdev-offload-provider.h"
#include "netdev-provider.h"
#include "netdev-vport.h"
#include "odp-netlink.h"
#include "openflow/openflow.h"
#include "packets.h"
#include "openvswitch/ofp-print.h"
#include "openvswitch/poll-loop.h"
#include "seq.h"
#include "openvswitch/shash.h"
#include "smap.h"
#include "socket-util.h"
#include "sset.h"
#include "svec.h"
#include "openvswitch/vlog.h"
#include "flow.h"
#include "util.h"
#ifdef __linux__
#include "tc.h"
#endif

VLOG_DEFINE_THIS_MODULE(netdev);

COVERAGE_DEFINE(netdev_received);
COVERAGE_DEFINE(netdev_sent);
COVERAGE_DEFINE(netdev_add_router);
COVERAGE_DEFINE(netdev_get_stats);
COVERAGE_DEFINE(netdev_send_prepare_drops);
COVERAGE_DEFINE(netdev_push_header_drops);

struct netdev_saved_flags {
    struct netdev *netdev;
    struct ovs_list node;           /* In struct netdev's saved_flags_list. */
    enum netdev_flags saved_flags;
    enum netdev_flags saved_values;
};

/* Protects 'netdev_shash' and the mutable members of struct netdev. */
static struct ovs_mutex netdev_mutex = OVS_MUTEX_INITIALIZER;

/* All created network devices. */
static struct shash netdev_shash OVS_GUARDED_BY(netdev_mutex)
    = SHASH_INITIALIZER(&netdev_shash);

/* Mutual exclusion of */
static struct ovs_mutex netdev_class_mutex OVS_ACQ_BEFORE(netdev_mutex)
    = OVS_MUTEX_INITIALIZER;

/* Contains 'struct netdev_registered_class'es. */
static struct cmap netdev_classes = CMAP_INITIALIZER;

struct netdev_registered_class {
    struct cmap_node cmap_node; /* In 'netdev_classes', by class->type. */
    const struct netdev_class *class;

    /* Number of references: one for the class itself and one for every
     * instance of the class. */
    struct ovs_refcount refcnt;
};

/* This is set pretty low because we probably won't learn anything from the
 * additional log messages. */
static struct vlog_rate_limit rl = VLOG_RATE_LIMIT_INIT(5, 20);

static void restore_all_flags(void *aux OVS_UNUSED);
void update_device_args(struct netdev *, const struct shash *args);
#ifdef HAVE_AF_XDP
void signal_remove_xdp(struct netdev *netdev);
#endif

int
netdev_n_txq(const struct netdev *netdev)
{
    return netdev->n_txq;
}

int
netdev_n_rxq(const struct netdev *netdev)
{
    return netdev->n_rxq;
}

bool
netdev_is_pmd(const struct netdev *netdev)
{
    return netdev->netdev_class->is_pmd;
}

bool
netdev_has_tunnel_push_pop(const struct netdev *netdev)
{
    return netdev->netdev_class->push_header
           && netdev->netdev_class->pop_header;
}

static void
netdev_initialize(void)
    OVS_EXCLUDED(netdev_mutex)
{
    static struct ovsthread_once once = OVSTHREAD_ONCE_INITIALIZER;

    if (ovsthread_once_start(&once)) {
        fatal_signal_add_hook(restore_all_flags, NULL, NULL, true);

        netdev_vport_patch_register();

#ifdef __linux__
        netdev_register_provider(&netdev_linux_class);
        netdev_register_provider(&netdev_internal_class);
        netdev_register_provider(&netdev_tap_class);
        netdev_vport_tunnel_register();

        netdev_register_flow_api_provider(&netdev_offload_tc);
#ifdef HAVE_AF_XDP
        netdev_register_provider(&netdev_afxdp_class);
        netdev_register_provider(&netdev_afxdp_nonpmd_class);
#endif
#endif
#if defined(__FreeBSD__) || defined(__NetBSD__)
        netdev_register_provider(&netdev_tap_class);
        netdev_register_provider(&netdev_bsd_class);
#endif
#ifdef _WIN32
        netdev_register_provider(&netdev_windows_class);
        netdev_register_provider(&netdev_internal_class);
        netdev_vport_tunnel_register();
#endif
        ovsthread_once_done(&once);
    }
}

/* Performs periodic work needed by all the various kinds of netdevs.
 *
 * If your program opens any netdevs, it must call this function within its
 * main poll loop. */
void
netdev_run(void)
    OVS_EXCLUDED(netdev_mutex)
{
    netdev_initialize();

    struct netdev_registered_class *rc;
    CMAP_FOR_EACH (rc, cmap_node, &netdev_classes) {
        if (rc->class->run) {
            rc->class->run(rc->class);
        }
    }
}

/* Arranges for poll_block() to wake up when netdev_run() needs to be called.
 *
 * If your program opens any netdevs, it must call this function within its
 * main poll loop. */
void
netdev_wait(void)
    OVS_EXCLUDED(netdev_mutex)
{
    netdev_initialize();

    struct netdev_registered_class *rc;
    CMAP_FOR_EACH (rc, cmap_node, &netdev_classes) {
        if (rc->class->wait) {
            rc->class->wait(rc->class);
        }
    }
}

static struct netdev_registered_class *
netdev_lookup_class(const char *type)
{
    struct netdev_registered_class *rc;
    CMAP_FOR_EACH_WITH_HASH (rc, cmap_node, hash_string(type, 0),
                             &netdev_classes) {
        if (!strcmp(type, rc->class->type)) {
            return rc;
        }
    }
    return NULL;
}

/* Initializes and registers a new netdev provider.  After successful
 * registration, new netdevs of that type can be opened using netdev_open(). */
int
netdev_register_provider(const struct netdev_class *new_class)
    OVS_EXCLUDED(netdev_class_mutex, netdev_mutex)
{
    int error;

    ovs_mutex_lock(&netdev_class_mutex);
    if (netdev_lookup_class(new_class->type)) {
        VLOG_WARN("attempted to register duplicate netdev provider: %s",
                   new_class->type);
        error = EEXIST;
    } else {
        error = new_class->init ? new_class->init() : 0;
        if (!error) {
            struct netdev_registered_class *rc;

            rc = xmalloc(sizeof *rc);
            cmap_insert(&netdev_classes, &rc->cmap_node,
                        hash_string(new_class->type, 0));
            rc->class = new_class;
            ovs_refcount_init(&rc->refcnt);
        } else {
            VLOG_ERR("failed to initialize %s network device class: %s",
                     new_class->type, ovs_strerror(error));
        }
    }
    ovs_mutex_unlock(&netdev_class_mutex);

    return error;
}

/* Unregisters a netdev provider.  'type' must have been previously registered
 * and not currently be in use by any netdevs.  After unregistration new
 * netdevs of that type cannot be opened using netdev_open().  (However, the
 * provider may still be accessible from other threads until the next RCU grace
 * period, so the caller must not free or re-register the same netdev_class
 * until that has passed.) */
int
netdev_unregister_provider(const char *type)
    OVS_EXCLUDED(netdev_class_mutex, netdev_mutex)
{
    struct netdev_registered_class *rc;
    int error;

    netdev_initialize();

    ovs_mutex_lock(&netdev_class_mutex);
    rc = netdev_lookup_class(type);
    if (!rc) {
        VLOG_WARN("attempted to unregister a netdev provider that is not "
                  "registered: %s", type);
        error = EAFNOSUPPORT;
    } else if (ovs_refcount_unref(&rc->refcnt) != 1) {
        ovs_refcount_ref(&rc->refcnt);
        VLOG_WARN("attempted to unregister in use netdev provider: %s",
                  type);
        error = EBUSY;
    } else  {
        cmap_remove(&netdev_classes, &rc->cmap_node,
                    hash_string(rc->class->type, 0));
        ovsrcu_postpone(free, rc);
        error = 0;
    }
    ovs_mutex_unlock(&netdev_class_mutex);

    return error;
}

/* Clears 'types' and enumerates the types of all currently registered netdev
 * providers into it.  The caller must first initialize the sset. */
void
netdev_enumerate_types(struct sset *types)
    OVS_EXCLUDED(netdev_mutex)
{
    netdev_initialize();
    sset_clear(types);

    struct netdev_registered_class *rc;
    CMAP_FOR_EACH (rc, cmap_node, &netdev_classes) {
        sset_add(types, rc->class->type);
    }
}

static const char *
netdev_vport_type_from_name(const char *name)
{
    struct netdev_registered_class *rc;
    const char *type;
    CMAP_FOR_EACH (rc, cmap_node, &netdev_classes) {
        const char *dpif_port = netdev_vport_class_get_dpif_port(rc->class);
        if (dpif_port && !strncmp(name, dpif_port, strlen(dpif_port))) {
            type = rc->class->type;
            return type;
        }
    }
    return NULL;
}

/* Check that the network device name is not the same as any of the registered
 * vport providers' dpif_port name (dpif_port is NULL if the vport provider
 * does not define it) or the datapath internal port name (e.g. ovs-system).
 *
 * Returns true if there is a name conflict, false otherwise. */
bool
netdev_is_reserved_name(const char *name)
    OVS_EXCLUDED(netdev_mutex)
{
    netdev_initialize();

    struct netdev_registered_class *rc;
    CMAP_FOR_EACH (rc, cmap_node, &netdev_classes) {
        const char *dpif_port = netdev_vport_class_get_dpif_port(rc->class);
        if (dpif_port && !strncmp(name, dpif_port, strlen(dpif_port))) {
            return true;
        }
    }

    if (!strncmp(name, "ovs-", 4)) {
        struct sset types;
        const char *type;

        sset_init(&types);
        dp_enumerate_types(&types);
        SSET_FOR_EACH (type, &types) {
            if (!strcmp(name+4, type)) {
                sset_destroy(&types);
                return true;
            }
        }
        sset_destroy(&types);
    }

    return false;
}

/* Opens the network device named 'name' (e.g. "eth0") of the specified 'type'
 * (e.g. "system") and returns zero if successful, otherwise a positive errno
 * value.  On success, sets '*netdevp' to the new network device, otherwise to
 * null.
 *
 * Some network devices may need to be configured (with netdev_set_config())
 * before they can be used.
 *
 * Before opening rxqs or sending packets, '*netdevp' may need to be
 * reconfigured (with netdev_is_reconf_required() and netdev_reconfigure()).
 * */
int
netdev_open(const char *name, const char *type, struct netdev **netdevp)
    OVS_EXCLUDED(netdev_mutex)
{
    struct netdev *netdev;
    int error = 0;

    if (!name[0]) {
        /* Reject empty names.  This saves the providers having to do this.  At
         * least one screwed this up: the netdev-linux "tap" implementation
         * passed the name directly to the Linux TUNSETIFF call, which treats
         * an empty string as a request to generate a unique name. */
        return EINVAL;
    }

    netdev_initialize();

    ovs_mutex_lock(&netdev_mutex);
    netdev = shash_find_data(&netdev_shash, name);

    if (netdev && type && type[0]) {
        if (strcmp(type, netdev->netdev_class->type)) {

            if (netdev->auto_classified) {
                /* If this device was first created without a classification
                 * type, for example due to routing or tunneling code, and they
                 * keep a reference, a "classified" call to open will fail.
                 * In this case we remove the classless device, and re-add it
                 * below. We remove the netdev from the shash, and change the
                 * sequence, so owners of the old classless device can
                 * release/cleanup. */
                if (netdev->node) {
                    shash_delete(&netdev_shash, netdev->node);
                    netdev->node = NULL;
                    netdev_change_seq_changed(netdev);
                }

                netdev = NULL;
            } else {
                error = EEXIST;
            }
        } else if (netdev->auto_classified) {
            /* If netdev reopened with type "system", clear auto_classified. */
            netdev->auto_classified = false;
        }
    }

    if (!netdev) {
        struct netdev_registered_class *rc;

        rc = netdev_lookup_class(type && type[0] ? type : "system");
        if (rc && ovs_refcount_try_ref_rcu(&rc->refcnt)) {
            netdev = rc->class->alloc();
            if (netdev) {
                memset(netdev, 0, sizeof *netdev);
                netdev->netdev_class = rc->class;
                netdev->auto_classified = type && type[0] ? false : true;
                netdev->name = xstrdup(name);
                netdev->change_seq = 1;
                netdev->reconfigure_seq = seq_create();
                netdev->last_reconfigure_seq =
                    seq_read(netdev->reconfigure_seq);
                ovsrcu_set(&netdev->flow_api, NULL);
                netdev->hw_info.oor = false;
                atomic_init(&netdev->hw_info.miss_api_supported, false);
                netdev->node = shash_add(&netdev_shash, name, netdev);

                /* By default enable one tx and rx queue per netdev. */
                netdev->n_txq = netdev->netdev_class->send ? 1 : 0;
                netdev->n_rxq = netdev->netdev_class->rxq_alloc ? 1 : 0;

                ovs_list_init(&netdev->saved_flags_list);

                error = rc->class->construct(netdev);
                if (!error) {
                    netdev_change_seq_changed(netdev);
                } else {
                    ovs_refcount_unref(&rc->refcnt);
                    seq_destroy(netdev->reconfigure_seq);
                    free(netdev->name);
                    ovs_assert(ovs_list_is_empty(&netdev->saved_flags_list));
                    shash_delete(&netdev_shash, netdev->node);
                    rc->class->dealloc(netdev);
                }
            } else {
                error = ENOMEM;
            }
        } else {
            VLOG_WARN("could not create netdev %s of unknown type %s",
                      name, type);
            error = EAFNOSUPPORT;
        }
    }

    if (!error) {
        netdev->ref_cnt++;
        *netdevp = netdev;
    } else {
        *netdevp = NULL;
    }
    ovs_mutex_unlock(&netdev_mutex);

    return error;
}

/* Returns a reference to 'netdev_' for the caller to own. Returns null if
 * 'netdev_' is null. */
struct netdev *
netdev_ref(const struct netdev *netdev_)
    OVS_EXCLUDED(netdev_mutex)
{
    struct netdev *netdev = CONST_CAST(struct netdev *, netdev_);

    if (netdev) {
        ovs_mutex_lock(&netdev_mutex);
        ovs_assert(netdev->ref_cnt > 0);
        netdev->ref_cnt++;
        ovs_mutex_unlock(&netdev_mutex);
    }
    return netdev;
}

/* Reconfigures the device 'netdev' with 'args'.  'args' may be empty
 * or NULL if none are needed. */
int
netdev_set_config(struct netdev *netdev, const struct smap *args, char **errp)
    OVS_EXCLUDED(netdev_mutex)
{
    if (netdev->netdev_class->set_config) {
        const struct smap no_args = SMAP_INITIALIZER(&no_args);
        char *verbose_error = NULL;
        int error;

        error = netdev->netdev_class->set_config(netdev,
                                                 args ? args : &no_args,
                                                 &verbose_error);
        if (error) {
            VLOG_WARN_BUF(verbose_error ? NULL : errp,
                          "%s: could not set configuration (%s)",
                          netdev_get_name(netdev), ovs_strerror(error));
            if (verbose_error) {
                if (errp) {
                    *errp = verbose_error;
                } else {
                    free(verbose_error);
                }
            }
        }
        return error;
    } else if (args && !smap_is_empty(args)) {
        VLOG_WARN_BUF(errp, "%s: arguments provided to device that is not configurable",
                      netdev_get_name(netdev));
    }
    return 0;
}

/* Returns the current configuration for 'netdev' in 'args'.  The caller must
 * have already initialized 'args' with smap_init().  Returns 0 on success, in
 * which case 'args' will be filled with 'netdev''s configuration.  On failure
 * returns a positive errno value, in which case 'args' will be empty.
 *
 * The caller owns 'args' and its contents and must eventually free them with
 * smap_destroy(). */
int
netdev_get_config(const struct netdev *netdev, struct smap *args)
    OVS_EXCLUDED(netdev_mutex)
{
    int error;

    smap_clear(args);
    if (netdev->netdev_class->get_config) {
        error = netdev->netdev_class->get_config(netdev, args);
        if (error) {
            smap_clear(args);
        }
    } else {
        error = 0;
    }

    return error;
}

const struct netdev_tunnel_config *
netdev_get_tunnel_config(const struct netdev *netdev)
    OVS_EXCLUDED(netdev_mutex)
{
    if (netdev->netdev_class->get_tunnel_config) {
        return netdev->netdev_class->get_tunnel_config(netdev);
    } else {
        return NULL;
    }
}

/* Returns the id of the numa node the 'netdev' is on.  If the function
 * is not implemented, returns NETDEV_NUMA_UNSPEC. */
int
netdev_get_numa_id(const struct netdev *netdev)
{
    if (netdev->netdev_class->get_numa_id) {
        return netdev->netdev_class->get_numa_id(netdev);
    } else {
        return NETDEV_NUMA_UNSPEC;
    }
}

static void
netdev_unref(struct netdev *dev)
    OVS_RELEASES(netdev_mutex)
{
    ovs_assert(dev->ref_cnt);
    if (!--dev->ref_cnt) {
        const struct netdev_class *class = dev->netdev_class;
        struct netdev_registered_class *rc;

        netdev_uninit_flow_api(dev);

        dev->netdev_class->destruct(dev);

        if (dev->node) {
            shash_delete(&netdev_shash, dev->node);
        }
        free(dev->name);
        seq_destroy(dev->reconfigure_seq);
        dev->netdev_class->dealloc(dev);
        ovs_mutex_unlock(&netdev_mutex);

        rc = netdev_lookup_class(class->type);
        ovs_refcount_unref(&rc->refcnt);
    } else {
        ovs_mutex_unlock(&netdev_mutex);
    }
}

/* Closes and destroys 'netdev'. */
void
netdev_close(struct netdev *netdev)
    OVS_EXCLUDED(netdev_mutex)
{
    if (netdev) {
        ovs_mutex_lock(&netdev_mutex);
        netdev_unref(netdev);
    }
}

/* Removes 'netdev' from the global shash and unrefs 'netdev'.
 *
 * This allows handler and revalidator threads to still retain references
 * to this netdev while the main thread changes interface configuration.
 *
 * This function should only be called by the main thread when closing
 * netdevs during user configuration changes. Otherwise, netdev_close should be
 * used to close netdevs. */
void
netdev_remove(struct netdev *netdev)
{
    if (netdev) {
        ovs_mutex_lock(&netdev_mutex);
        if (netdev->node) {
            shash_delete(&netdev_shash, netdev->node);
            netdev->node = NULL;
            netdev_change_seq_changed(netdev);
        }
        netdev_unref(netdev);
    }
}

/* Parses 'netdev_name_', which is of the form [type@]name into its component
 * pieces.  'name' and 'type' must be freed by the caller. */
void
netdev_parse_name(const char *netdev_name_, char **name, char **type)
{
    char *netdev_name = xstrdup(netdev_name_);
    char *separator;

    separator = strchr(netdev_name, '@');
    if (separator) {
        *separator = '\0';
        *type = netdev_name;
        *name = xstrdup(separator + 1);
    } else {
        *name = netdev_name;
        *type = xstrdup("system");
    }
}

/* Attempts to open a netdev_rxq handle for obtaining packets received on
 * 'netdev'.  On success, returns 0 and stores a nonnull 'netdev_rxq *' into
 * '*rxp'.  On failure, returns a positive errno value and stores NULL into
 * '*rxp'.
 *
 * Some kinds of network devices might not support receiving packets.  This
 * function returns EOPNOTSUPP in that case.*/
int
netdev_rxq_open(struct netdev *netdev, struct netdev_rxq **rxp, int id)
    OVS_EXCLUDED(netdev_mutex)
{
    int error;

    if (netdev->netdev_class->rxq_alloc && id < netdev->n_rxq) {
        struct netdev_rxq *rx = netdev->netdev_class->rxq_alloc();
        if (rx) {
            rx->netdev = netdev;
            rx->queue_id = id;
            error = netdev->netdev_class->rxq_construct(rx);
            if (!error) {
                netdev_ref(netdev);
                *rxp = rx;
                return 0;
            }
            netdev->netdev_class->rxq_dealloc(rx);
        } else {
            error = ENOMEM;
        }
    } else {
        error = EOPNOTSUPP;
    }

    *rxp = NULL;
    return error;
}

/* Closes 'rx'. */
void
netdev_rxq_close(struct netdev_rxq *rx)
    OVS_EXCLUDED(netdev_mutex)
{
    if (rx) {
        struct netdev *netdev = rx->netdev;
        netdev->netdev_class->rxq_destruct(rx);
        netdev->netdev_class->rxq_dealloc(rx);
        netdev_close(netdev);
    }
}

bool netdev_rxq_enabled(struct netdev_rxq *rx)
{
    bool enabled = true;

    if (rx->netdev->netdev_class->rxq_enabled) {
        enabled = rx->netdev->netdev_class->rxq_enabled(rx);
    }
    return enabled;
}

/* Attempts to receive a batch of packets from 'rx'.  'batch' should point to
 * the beginning of an array of NETDEV_MAX_BURST pointers to dp_packet.  If
 * successful, this function stores pointers to up to NETDEV_MAX_BURST
 * dp_packets into the array, transferring ownership of the packets to the
 * caller, stores the number of received packets in 'batch->count', and returns
 * 0.
 *
 * The implementation does not necessarily initialize any non-data members of
 * 'batch'.  That is, the caller must initialize layer pointers and metadata
 * itself, if desired, e.g. with pkt_metadata_init() and miniflow_extract().
 *
 * Returns EAGAIN immediately if no packet is ready to be received or another
 * positive errno value if an error was encountered. */
int
netdev_rxq_recv(struct netdev_rxq *rx, struct dp_packet_batch *batch,
                int *qfill)
{
    int retval;

    retval = rx->netdev->netdev_class->rxq_recv(rx, batch, qfill);
    if (!retval) {
        COVERAGE_INC(netdev_received);
    } else {
        batch->count = 0;
    }
    return retval;
}

/* Arranges for poll_block() to wake up when a packet is ready to be received
 * on 'rx'. */
void
netdev_rxq_wait(struct netdev_rxq *rx)
{
    rx->netdev->netdev_class->rxq_wait(rx);
}

/* Discards any packets ready to be received on 'rx'. */
int
netdev_rxq_drain(struct netdev_rxq *rx)
{
    return (rx->netdev->netdev_class->rxq_drain
            ? rx->netdev->netdev_class->rxq_drain(rx)
            : 0);
}

/* Configures the number of tx queues of 'netdev'. Returns 0 if successful,
 * otherwise a positive errno value.
 *
 * 'n_txq' specifies the exact number of transmission queues to create.
 *
 * The change might not effective immediately.  The caller must check if a
 * reconfiguration is required with netdev_is_reconf_required() and eventually
 * call netdev_reconfigure() before using the new queues.
 *
 * On error, the tx queue configuration is unchanged */
int
netdev_set_tx_multiq(struct netdev *netdev, unsigned int n_txq)
{
    int error;

    error = (netdev->netdev_class->set_tx_multiq
             ? netdev->netdev_class->set_tx_multiq(netdev, MAX(n_txq, 1))
             : EOPNOTSUPP);

    if (error && error != EOPNOTSUPP) {
        VLOG_DBG_RL(&rl, "failed to set tx queue for network device %s:"
                    "%s", netdev_get_name(netdev), ovs_strerror(error));
    }

    return error;
}

enum netdev_pt_mode
netdev_get_pt_mode(const struct netdev *netdev)
{
    return (netdev->netdev_class->get_pt_mode
            ? netdev->netdev_class->get_pt_mode(netdev)
            : NETDEV_PT_LEGACY_L2);
}

/* Check if a 'packet' is compatible with 'netdev_flags'.
 * If a packet is incompatible, return 'false' with the 'errormsg'
 * pointing to a reason. */
static bool
netdev_send_prepare_packet(const uint64_t netdev_flags,
                           struct dp_packet *packet, char **errormsg)
{
    uint64_t l4_mask;

    if (dp_packet_hwol_is_tso(packet)
        && !(netdev_flags & NETDEV_TX_OFFLOAD_TCP_TSO)) {
            /* Fall back to GSO in software. */
            VLOG_ERR_BUF(errormsg, "No TSO support");
            return false;
    }

    l4_mask = dp_packet_hwol_l4_mask(packet);
    if (l4_mask) {
        if (dp_packet_hwol_l4_is_tcp(packet)) {
            if (!(netdev_flags & NETDEV_TX_OFFLOAD_TCP_CKSUM)) {
                /* Fall back to TCP csum in software. */
                VLOG_ERR_BUF(errormsg, "No TCP checksum support");
                return false;
            }
        } else if (dp_packet_hwol_l4_is_udp(packet)) {
            if (!(netdev_flags & NETDEV_TX_OFFLOAD_UDP_CKSUM)) {
                /* Fall back to UDP csum in software. */
                VLOG_ERR_BUF(errormsg, "No UDP checksum support");
                return false;
            }
        } else if (dp_packet_hwol_l4_is_sctp(packet)) {
            if (!(netdev_flags & NETDEV_TX_OFFLOAD_SCTP_CKSUM)) {
                /* Fall back to SCTP csum in software. */
                VLOG_ERR_BUF(errormsg, "No SCTP checksum support");
                return false;
            }
        } else {
            VLOG_ERR_BUF(errormsg, "No L4 checksum support: mask: %"PRIu64,
                         l4_mask);
            return false;
        }
    }

    return true;
}

/* Check if each packet in 'batch' is compatible with 'netdev' features,
 * otherwise either fall back to software implementation or drop it. */
static void
netdev_send_prepare_batch(const struct netdev *netdev,
                          struct dp_packet_batch *batch)
{
    struct dp_packet *packet;
    size_t i, size = dp_packet_batch_size(batch);

    DP_PACKET_BATCH_REFILL_FOR_EACH (i, size, packet, batch) {
        char *errormsg = NULL;

        if (netdev_send_prepare_packet(netdev->ol_flags, packet, &errormsg)) {
            dp_packet_batch_refill(batch, packet, i);
        } else {
            dp_packet_delete(packet);
            COVERAGE_INC(netdev_send_prepare_drops);
            VLOG_WARN_RL(&rl, "%s: Packet dropped: %s",
                         netdev_get_name(netdev), errormsg);
            free(errormsg);
        }
    }
}

/* Sends 'batch' on 'netdev'.  Returns 0 if successful (for every packet),
 * otherwise a positive errno value.  Returns EAGAIN without blocking if
 * at least one the packets cannot be queued immediately.  Returns EMSGSIZE
 * if a partial packet was transmitted or if a packet is too big or too small
 * to transmit on the device.
 *
 * The caller must make sure that 'netdev' supports sending by making sure that
 * 'netdev_n_txq(netdev)' returns >= 1.
 *
 * If the function returns a non-zero value, some of the packets might have
 * been sent anyway.
 *
 * The caller transfers ownership of all the packets to the network device,
 * regardless of success.
 *
 * If 'concurrent_txq' is true, the caller may perform concurrent calls
 * to netdev_send() with the same 'qid'. The netdev provider is responsible
 * for making sure that these concurrent calls do not create a race condition
 * by using locking or other synchronization if required.
 *
 * The network device is expected to maintain one or more packet
 * transmission queues, so that the caller does not ordinarily have to
 * do additional queuing of packets.  'qid' specifies the queue to use
 * and can be ignored if the implementation does not support multiple
 * queues. */
int
netdev_send(struct netdev *netdev, int qid, struct dp_packet_batch *batch,
            bool concurrent_txq)
{
    int error;

    netdev_send_prepare_batch(netdev, batch);
    if (OVS_UNLIKELY(dp_packet_batch_is_empty(batch))) {
        return 0;
    }

    error = netdev->netdev_class->send(netdev, qid, batch, concurrent_txq);
    if (!error) {
        COVERAGE_INC(netdev_sent);
    }
    return error;
}

/* Pop tunnel header, build tunnel metadata and resize 'batch->packets'
 * for further processing.
 *
 * The caller must make sure that 'netdev' support this operation by checking
 * that netdev_has_tunnel_push_pop() returns true. */
void
netdev_pop_header(struct netdev *netdev, struct dp_packet_batch *batch)
{
    struct dp_packet *packet;
    size_t i, size = dp_packet_batch_size(batch);

    DP_PACKET_BATCH_REFILL_FOR_EACH (i, size, packet, batch) {
        packet = netdev->netdev_class->pop_header(packet);
        if (packet) {
            /* Reset the offload flags if present, to avoid wrong
             * interpretation in the further packet processing when
             * recirculated.*/
            dp_packet_reset_offload(packet);
            pkt_metadata_init_conn(&packet->md);
            dp_packet_batch_refill(batch, packet, i);
        }
    }
}

void
netdev_init_tnl_build_header_params(struct netdev_tnl_build_header_params *params,
                                    const struct flow *tnl_flow,
                                    const struct in6_addr *src,
                                    struct eth_addr dmac,
                                    struct eth_addr smac)
{
    params->flow = tnl_flow;
    params->dmac = dmac;
    params->smac = smac;
    params->s_ip = src;
    params->is_ipv6 = !IN6_IS_ADDR_V4MAPPED(src);
}

int netdev_build_header(const struct netdev *netdev,
                        struct ovs_action_push_tnl *data,
                        const struct netdev_tnl_build_header_params *params)
{
    if (netdev->netdev_class->build_header) {
        return netdev->netdev_class->build_header(netdev, data, params);
    }
    return EOPNOTSUPP;
}

/* Push tunnel header (reading from tunnel metadata) and resize
 * 'batch->packets' for further processing.
 *
 * The caller must make sure that 'netdev' support this operation by checking
 * that netdev_has_tunnel_push_pop() returns true. */
int
netdev_push_header(const struct netdev *netdev,
                   struct dp_packet_batch *batch,
                   const struct ovs_action_push_tnl *data)
{
    struct dp_packet *packet;
    size_t i, size = dp_packet_batch_size(batch);

    DP_PACKET_BATCH_REFILL_FOR_EACH (i, size, packet, batch) {
        if (OVS_UNLIKELY(dp_packet_hwol_is_tso(packet)
                         || dp_packet_hwol_l4_mask(packet))) {
            COVERAGE_INC(netdev_push_header_drops);
            dp_packet_delete(packet);
            VLOG_WARN_RL(&rl, "%s: Tunneling packets with HW offload flags is "
                         "not supported: packet dropped",
                         netdev_get_name(netdev));
        } else {
            netdev->netdev_class->push_header(netdev, packet, data);
            pkt_metadata_init(&packet->md, data->out_port);
            dp_packet_batch_refill(batch, packet, i);
        }
    }

    return 0;
}

/* Registers with the poll loop to wake up from the next call to poll_block()
 * when the packet transmission queue has sufficient room to transmit a packet
 * with netdev_send().
 *
 * The network device is expected to maintain one or more packet
 * transmission queues, so that the caller does not ordinarily have to
 * do additional queuing of packets.  'qid' specifies the queue to use
 * and can be ignored if the implementation does not support multiple
 * queues. */
void
netdev_send_wait(struct netdev *netdev, int qid)
{
    if (netdev->netdev_class->send_wait) {
        netdev->netdev_class->send_wait(netdev, qid);
    }
}

/* Attempts to set 'netdev''s MAC address to 'mac'.  Returns 0 if successful,
 * otherwise a positive errno value. */
int
netdev_set_etheraddr(struct netdev *netdev, const struct eth_addr mac)
{
    return netdev->netdev_class->set_etheraddr(netdev, mac);
}

/* Retrieves 'netdev''s MAC address.  If successful, returns 0 and copies the
 * the MAC address into 'mac'.  On failure, returns a positive errno value and
 * clears 'mac' to all-zeros. */
int
netdev_get_etheraddr(const struct netdev *netdev, struct eth_addr *mac)
{
    int error;

    error = netdev->netdev_class->get_etheraddr(netdev, mac);
    if (error) {
        memset(mac, 0, sizeof *mac);
    }
    return error;
}

/* Returns the name of the network device that 'netdev' represents,
 * e.g. "eth0".  The caller must not modify or free the returned string. */
const char *
netdev_get_name(const struct netdev *netdev)
{
    return netdev->name;
}

/* Retrieves the MTU of 'netdev'.  The MTU is the maximum size of transmitted
 * (and received) packets, in bytes, not including the hardware header; thus,
 * this is typically 1500 bytes for Ethernet devices.
 *
 * If successful, returns 0 and stores the MTU size in '*mtup'.  Returns
 * EOPNOTSUPP if 'netdev' does not have an MTU (as e.g. some tunnels do not).
 * On other failure, returns a positive errno value.  On failure, sets '*mtup'
 * to 0. */
int
netdev_get_mtu(const struct netdev *netdev, int *mtup)
{
    const struct netdev_class *class = netdev->netdev_class;
    int error;

    error = class->get_mtu ? class->get_mtu(netdev, mtup) : EOPNOTSUPP;
    if (error) {
        *mtup = 0;
        if (error != EOPNOTSUPP) {
            VLOG_DBG_RL(&rl, "failed to retrieve MTU for network device %s: "
                         "%s", netdev_get_name(netdev), ovs_strerror(error));
        }
    }
    return error;
}

/* Sets the MTU of 'netdev'.  The MTU is the maximum size of transmitted
 * (and received) packets, in bytes.
 *
 * If successful, returns 0.  Returns EOPNOTSUPP if 'netdev' does not have an
 * MTU (as e.g. some tunnels do not).  On other failure, returns a positive
 * errno value. */
int
netdev_set_mtu(struct netdev *netdev, int mtu)
{
    const struct netdev_class *class = netdev->netdev_class;
    int error;

    error = class->set_mtu ? class->set_mtu(netdev, mtu) : EOPNOTSUPP;
    if (error && error != EOPNOTSUPP) {
        VLOG_WARN_RL(&rl, "failed to set MTU for network device %s: %s",
                     netdev_get_name(netdev), ovs_strerror(error));
    }

    return error;
}

/* If 'user_config' is true, the user wants to control 'netdev''s MTU and we
 * should not override it.  If 'user_config' is false, we may adjust
 * 'netdev''s MTU (e.g., if 'netdev' is internal). */
void
netdev_mtu_user_config(struct netdev *netdev, bool user_config)
{
    if (netdev->mtu_user_config != user_config) {
        netdev_change_seq_changed(netdev);
        netdev->mtu_user_config = user_config;
    }
}

/* Returns 'true' if the user explicitly specified an MTU value for 'netdev'.
 * Otherwise, returns 'false', in which case we are allowed to adjust the
 * device MTU. */
bool
netdev_mtu_is_user_config(struct netdev *netdev)
{
    return netdev->mtu_user_config;
}

/* Returns the ifindex of 'netdev', if successful, as a positive number.  On
 * failure, returns a negative errno value.
 *
 * The desired semantics of the ifindex value are a combination of those
 * specified by POSIX for if_nametoindex() and by SNMP for ifIndex.  An ifindex
 * value should be unique within a host and remain stable at least until
 * reboot.  SNMP says an ifindex "ranges between 1 and the value of ifNumber"
 * but many systems do not follow this rule anyhow.
 *
 * Some network devices may not implement support for this function.  In such
 * cases this function will always return -EOPNOTSUPP.
 */
int
netdev_get_ifindex(const struct netdev *netdev)
{
    int (*get_ifindex)(const struct netdev *);

    get_ifindex = netdev->netdev_class->get_ifindex;

    return get_ifindex ? get_ifindex(netdev) : -EOPNOTSUPP;
}

/* Stores the features supported by 'netdev' into each of '*current',
 * '*advertised', '*supported', and '*peer' that are non-null.  Each value is a
 * bitmap of "enum ofp_port_features" bits, in host byte order.  Returns 0 if
 * successful, otherwise a positive errno value.  On failure, all of the
 * passed-in values are set to 0.
 *
 * Some network devices may not implement support for this function.  In such
 * cases this function will always return EOPNOTSUPP. */
int
netdev_get_features(const struct netdev *netdev,
                    enum netdev_features *current,
                    enum netdev_features *advertised,
                    enum netdev_features *supported,
                    enum netdev_features *peer)
{
    int (*get_features)(const struct netdev *netdev,
                        enum netdev_features *current,
                        enum netdev_features *advertised,
                        enum netdev_features *supported,
                        enum netdev_features *peer);
    enum netdev_features dummy[4];
    int error;

    if (!current) {
        current = &dummy[0];
    }
    if (!advertised) {
        advertised = &dummy[1];
    }
    if (!supported) {
        supported = &dummy[2];
    }
    if (!peer) {
        peer = &dummy[3];
    }

    get_features = netdev->netdev_class->get_features;
    error = get_features
                    ? get_features(netdev, current, advertised, supported,
                                   peer)
                    : EOPNOTSUPP;
    if (error) {
        *current = *advertised = *supported = *peer = 0;
    }
    return error;
}

/* Returns the maximum speed of a network connection that has the NETDEV_F_*
 * bits in 'features', in bits per second.  If no bits that indicate a speed
 * are set in 'features', returns 'default_bps'. */
uint64_t
netdev_features_to_bps(enum netdev_features features,
                       uint64_t default_bps)
{
    enum {
        F_1000000MB = NETDEV_F_1TB_FD,
        F_100000MB = NETDEV_F_100GB_FD,
        F_40000MB = NETDEV_F_40GB_FD,
        F_10000MB = NETDEV_F_10GB_FD,
        F_1000MB = NETDEV_F_1GB_HD | NETDEV_F_1GB_FD,
        F_100MB = NETDEV_F_100MB_HD | NETDEV_F_100MB_FD,
        F_10MB = NETDEV_F_10MB_HD | NETDEV_F_10MB_FD
    };

    return (  features & F_1000000MB ? UINT64_C(1000000000000)
            : features & F_100000MB  ? UINT64_C(100000000000)
            : features & F_40000MB   ? UINT64_C(40000000000)
            : features & F_10000MB   ? UINT64_C(10000000000)
            : features & F_1000MB    ? UINT64_C(1000000000)
            : features & F_100MB     ? UINT64_C(100000000)
            : features & F_10MB      ? UINT64_C(10000000)
                                     : default_bps);
}

/* Returns true if any of the NETDEV_F_* bits that indicate a full-duplex link
 * are set in 'features', otherwise false. */
bool
netdev_features_is_full_duplex(enum netdev_features features)
{
    return (features & (NETDEV_F_10MB_FD | NETDEV_F_100MB_FD | NETDEV_F_1GB_FD
                        | NETDEV_F_10GB_FD | NETDEV_F_40GB_FD
                        | NETDEV_F_100GB_FD | NETDEV_F_1TB_FD)) != 0;
}

/* Set the features advertised by 'netdev' to 'advertise'.  Returns 0 if
 * successful, otherwise a positive errno value. */
int
netdev_set_advertisements(struct netdev *netdev,
                          enum netdev_features advertise)
{
    return (netdev->netdev_class->set_advertisements
            ? netdev->netdev_class->set_advertisements(
                    netdev, advertise)
            : EOPNOTSUPP);
}

static const char *
netdev_feature_to_name(uint32_t bit)
{
    enum netdev_features f = bit;

    switch (f) {
    case NETDEV_F_10MB_HD:    return "10MB-HD";
    case NETDEV_F_10MB_FD:    return "10MB-FD";
    case NETDEV_F_100MB_HD:   return "100MB-HD";
    case NETDEV_F_100MB_FD:   return "100MB-FD";
    case NETDEV_F_1GB_HD:     return "1GB-HD";
    case NETDEV_F_1GB_FD:     return "1GB-FD";
    case NETDEV_F_10GB_FD:    return "10GB-FD";
    case NETDEV_F_40GB_FD:    return "40GB-FD";
    case NETDEV_F_100GB_FD:   return "100GB-FD";
    case NETDEV_F_1TB_FD:     return "1TB-FD";
    case NETDEV_F_OTHER:      return "OTHER";
    case NETDEV_F_COPPER:     return "COPPER";
    case NETDEV_F_FIBER:      return "FIBER";
    case NETDEV_F_AUTONEG:    return "AUTO_NEG";
    case NETDEV_F_PAUSE:      return "AUTO_PAUSE";
    case NETDEV_F_PAUSE_ASYM: return "AUTO_PAUSE_ASYM";
    }

    return NULL;
}

void
netdev_features_format(struct ds *s, enum netdev_features features)
{
    ofp_print_bit_names(s, features, netdev_feature_to_name, ' ');
    ds_put_char(s, '\n');
}

/* Assigns 'addr' as 'netdev''s IPv4 address and 'mask' as its netmask.  If
 * 'addr' is INADDR_ANY, 'netdev''s IPv4 address is cleared.  Returns a
 * positive errno value. */
int
netdev_set_in4(struct netdev *netdev, struct in_addr addr, struct in_addr mask)
{
    return (netdev->netdev_class->set_in4
            ? netdev->netdev_class->set_in4(netdev, addr, mask)
            : EOPNOTSUPP);
}

static int
netdev_get_addresses_by_name(const char *device_name,
                             struct in6_addr **addrsp, int *n_addrsp)
{
    struct netdev *netdev;
    int error = netdev_open(device_name, NULL, &netdev);
    if (error) {
        *addrsp = NULL;
        *n_addrsp = 0;
        return error;
    }

    struct in6_addr *masks;
    error = netdev_get_addr_list(netdev, addrsp, &masks, n_addrsp);
    netdev_close(netdev);
    free(masks);
    return error;
}

/* Obtains an IPv4 address from 'device_name' and save the address in '*in4'.
 * Returns 0 if successful, otherwise a positive errno value. */
int
netdev_get_in4_by_name(const char *device_name, struct in_addr *in4)
{
    struct in6_addr *addrs;
    int n;
    int error = netdev_get_addresses_by_name(device_name, &addrs, &n);

    in4->s_addr = 0;
    if (!error) {
        error = ENOENT;
        for (int i = 0; i < n; i++) {
            if (IN6_IS_ADDR_V4MAPPED(&addrs[i])) {
                in4->s_addr = in6_addr_get_mapped_ipv4(&addrs[i]);
                error = 0;
                break;
            }
        }
    }
    free(addrs);

    return error;
}

/* Obtains an IPv4 or IPv6 address from 'device_name' and save the address in
 * '*in6', representing IPv4 addresses as v6-mapped.  Returns 0 if successful,
 * otherwise a positive errno value. */
int
netdev_get_ip_by_name(const char *device_name, struct in6_addr *in6)
{
    struct in6_addr *addrs;
    int n;
    int error = netdev_get_addresses_by_name(device_name, &addrs, &n);

    *in6 = in6addr_any;
    if (!error) {
        error = ENOENT;
        for (int i = 0; i < n; i++) {
            if (!in6_is_lla(&addrs[i])) {
                *in6 = addrs[i];
                error = 0;
                break;
            }
        }
    }
    free(addrs);

    return error;
}

/* Adds 'router' as a default IP gateway for the TCP/IP stack that corresponds
 * to 'netdev'. */
int
netdev_add_router(struct netdev *netdev, struct in_addr router)
{
    COVERAGE_INC(netdev_add_router);
    return (netdev->netdev_class->add_router
            ? netdev->netdev_class->add_router(netdev, router)
            : EOPNOTSUPP);
}

/* Looks up the next hop for 'host' for the TCP/IP stack that corresponds to
 * 'netdev'.  If a route cannot not be determined, sets '*next_hop' to 0,
 * '*netdev_name' to null, and returns a positive errno value.  Otherwise, if a
 * next hop is found, stores the next hop gateway's address (0 if 'host' is on
 * a directly connected network) in '*next_hop' and a copy of the name of the
 * device to reach 'host' in '*netdev_name', and returns 0.  The caller is
 * responsible for freeing '*netdev_name' (by calling free()). */
int
netdev_get_next_hop(const struct netdev *netdev,
                    const struct in_addr *host, struct in_addr *next_hop,
                    char **netdev_name)
{
    int error = (netdev->netdev_class->get_next_hop
                 ? netdev->netdev_class->get_next_hop(
                        host, next_hop, netdev_name)
                 : EOPNOTSUPP);
    if (error) {
        next_hop->s_addr = 0;
        *netdev_name = NULL;
    }
    return error;
}

/* Populates 'smap' with status information.
 *
 * Populates 'smap' with 'netdev' specific status information.  This
 * information may be used to populate the status column of the Interface table
 * as defined in ovs-vswitchd.conf.db(5). */
int
netdev_get_status(const struct netdev *netdev, struct smap *smap)
{
    return (netdev->netdev_class->get_status
            ? netdev->netdev_class->get_status(netdev, smap)
            : EOPNOTSUPP);
}

/* Returns all assigned IP address to  'netdev' and returns 0.
 * API allocates array of address and masks and set it to
 * '*addr' and '*mask'.
 * Otherwise, returns a positive errno value and sets '*addr', '*mask
 * and '*n_addr' to NULL.
 *
 * The following error values have well-defined meanings:
 *
 *   - EADDRNOTAVAIL: 'netdev' has no assigned IPv6 address.
 *
 *   - EOPNOTSUPP: No IPv6 network stack attached to 'netdev'.
 *
 * 'addr' may be null, in which case the address itself is not reported. */
int
netdev_get_addr_list(const struct netdev *netdev, struct in6_addr **addr,
                     struct in6_addr **mask, int *n_addr)
{
    int error;

    error = (netdev->netdev_class->get_addr_list
             ? netdev->netdev_class->get_addr_list(netdev, addr, mask, n_addr): EOPNOTSUPP);
    if (error && addr) {
        *addr = NULL;
        *mask = NULL;
        *n_addr = 0;
    }

    return error;
}

/* On 'netdev', turns off the flags in 'off' and then turns on the flags in
 * 'on'.  Returns 0 if successful, otherwise a positive errno value. */
static int
do_update_flags(struct netdev *netdev, enum netdev_flags off,
                enum netdev_flags on, enum netdev_flags *old_flagsp,
                struct netdev_saved_flags **sfp)
    OVS_EXCLUDED(netdev_mutex)
{
    struct netdev_saved_flags *sf = NULL;
    enum netdev_flags old_flags;
    int error;

    error = netdev->netdev_class->update_flags(netdev, off & ~on, on,
                                               &old_flags);
    if (error) {
        VLOG_WARN_RL(&rl, "failed to %s flags for network device %s: %s",
                     off || on ? "set" : "get", netdev_get_name(netdev),
                     ovs_strerror(error));
        old_flags = 0;
    } else if ((off || on) && sfp) {
        enum netdev_flags new_flags = (old_flags & ~off) | on;
        enum netdev_flags changed_flags = old_flags ^ new_flags;
        if (changed_flags) {
            ovs_mutex_lock(&netdev_mutex);
            *sfp = sf = xmalloc(sizeof *sf);
            sf->netdev = netdev;
            ovs_list_push_front(&netdev->saved_flags_list, &sf->node);
            sf->saved_flags = changed_flags;
            sf->saved_values = changed_flags & new_flags;

            netdev->ref_cnt++;
            ovs_mutex_unlock(&netdev_mutex);
        }
    }

    if (old_flagsp) {
        *old_flagsp = old_flags;
    }
    if (sfp) {
        *sfp = sf;
    }

    return error;
}

/* Obtains the current flags for 'netdev' and stores them into '*flagsp'.
 * Returns 0 if successful, otherwise a positive errno value.  On failure,
 * stores 0 into '*flagsp'. */
int
netdev_get_flags(const struct netdev *netdev_, enum netdev_flags *flagsp)
{
    struct netdev *netdev = CONST_CAST(struct netdev *, netdev_);
    return do_update_flags(netdev, 0, 0, flagsp, NULL);
}

/* Sets the flags for 'netdev' to 'flags'.
 * Returns 0 if successful, otherwise a positive errno value. */
int
netdev_set_flags(struct netdev *netdev, enum netdev_flags flags,
                 struct netdev_saved_flags **sfp)
{
    return do_update_flags(netdev, -1, flags, NULL, sfp);
}

/* Turns on the specified 'flags' on 'netdev':
 *
 *    - On success, returns 0.  If 'sfp' is nonnull, sets '*sfp' to a newly
 *      allocated 'struct netdev_saved_flags *' that may be passed to
 *      netdev_restore_flags() to restore the original values of 'flags' on
 *      'netdev' (this will happen automatically at program termination if
 *      netdev_restore_flags() is never called) , or to NULL if no flags were
 *      actually changed.
 *
 *    - On failure, returns a positive errno value.  If 'sfp' is nonnull, sets
 *      '*sfp' to NULL. */
int
netdev_turn_flags_on(struct netdev *netdev, enum netdev_flags flags,
                     struct netdev_saved_flags **sfp)
{
    return do_update_flags(netdev, 0, flags, NULL, sfp);
}

/* Turns off the specified 'flags' on 'netdev'.  See netdev_turn_flags_on() for
 * details of the interface. */
int
netdev_turn_flags_off(struct netdev *netdev, enum netdev_flags flags,
                      struct netdev_saved_flags **sfp)
{
    return do_update_flags(netdev, flags, 0, NULL, sfp);
}

/* Restores the flags that were saved in 'sf', and destroys 'sf'.
 * Does nothing if 'sf' is NULL. */
void
netdev_restore_flags(struct netdev_saved_flags *sf)
    OVS_EXCLUDED(netdev_mutex)
{
    if (sf) {
        struct netdev *netdev = sf->netdev;
        enum netdev_flags old_flags;

        netdev->netdev_class->update_flags(netdev,
                                           sf->saved_flags & sf->saved_values,
                                           sf->saved_flags & ~sf->saved_values,
                                           &old_flags);

        ovs_mutex_lock(&netdev_mutex);
        ovs_list_remove(&sf->node);
        free(sf);
        netdev_unref(netdev);
    }
}

/* Looks up the ARP table entry for 'ip' on 'netdev'.  If one exists and can be
 * successfully retrieved, it stores the corresponding MAC address in 'mac' and
 * returns 0.  Otherwise, it returns a positive errno value; in particular,
 * ENXIO indicates that there is no ARP table entry for 'ip' on 'netdev'. */
int
netdev_arp_lookup(const struct netdev *netdev,
                  ovs_be32 ip, struct eth_addr *mac)
{
    int error = (netdev->netdev_class->arp_lookup
                 ? netdev->netdev_class->arp_lookup(netdev, ip, mac)
                 : EOPNOTSUPP);
    if (error) {
        *mac = eth_addr_zero;
    }
    return error;
}

/* Returns true if carrier is active (link light is on) on 'netdev'. */
bool
netdev_get_carrier(const struct netdev *netdev)
{
    int error;
    enum netdev_flags flags;
    bool carrier;

    netdev_get_flags(netdev, &flags);
    if (!(flags & NETDEV_UP)) {
        return false;
    }

    if (!netdev->netdev_class->get_carrier) {
        return true;
    }

    error = netdev->netdev_class->get_carrier(netdev, &carrier);
    if (error) {
        VLOG_DBG("%s: failed to get network device carrier status, assuming "
                 "down: %s", netdev_get_name(netdev), ovs_strerror(error));
        carrier = false;
    }

    return carrier;
}

/* Returns the number of times 'netdev''s carrier has changed. */
long long int
netdev_get_carrier_resets(const struct netdev *netdev)
{
    return (netdev->netdev_class->get_carrier_resets
            ? netdev->netdev_class->get_carrier_resets(netdev)
            : 0);
}

/* Attempts to force netdev_get_carrier() to poll 'netdev''s MII registers for
 * link status instead of checking 'netdev''s carrier.  'netdev''s MII
 * registers will be polled once ever 'interval' milliseconds.  If 'netdev'
 * does not support MII, another method may be used as a fallback.  If
 * 'interval' is less than or equal to zero, reverts netdev_get_carrier() to
 * its normal behavior.
 *
 * Returns 0 if successful, otherwise a positive errno value. */
int
netdev_set_miimon_interval(struct netdev *netdev, long long int interval)
{
    return (netdev->netdev_class->set_miimon_interval
            ? netdev->netdev_class->set_miimon_interval(netdev, interval)
            : EOPNOTSUPP);
}

/* Retrieves current device stats for 'netdev'. */
int
netdev_get_stats(const struct netdev *netdev, struct netdev_stats *stats)
{
    int error;

    /* Statistics are initialized before passing it to particular device
     * implementation so all values are filtered out by default. */
    memset(stats, 0xFF, sizeof *stats);

    COVERAGE_INC(netdev_get_stats);
    error = (netdev->netdev_class->get_stats
             ? netdev->netdev_class->get_stats(netdev, stats)
             : EOPNOTSUPP);
    if (error) {
        /* In case of error all statistics are filtered out */
        memset(stats, 0xff, sizeof *stats);
    }
    return error;
}

/* Retrieves current device custom stats for 'netdev'. */
int
netdev_get_custom_stats(const struct netdev *netdev,
                        struct netdev_custom_stats *custom_stats)
{
    int error;
    memset(custom_stats, 0, sizeof *custom_stats);
    error = (netdev->netdev_class->get_custom_stats
             ? netdev->netdev_class->get_custom_stats(netdev, custom_stats)
             : EOPNOTSUPP);

    return error;
}

/* Attempts to set input rate limiting (policing) policy, such that:
 * - up to 'kbits_rate' kbps of traffic is accepted, with a maximum
 *   accumulative burst size of 'kbits' kb; and
 * - up to 'kpkts' kpps of traffic is accepted, with a maximum
 *   accumulative burst size of 'kpkts' kilo packets.
 */
int
netdev_set_policing(struct netdev *netdev, uint32_t kbits_rate,
                    uint32_t kbits_burst, uint32_t kpkts_rate,
                    uint32_t kpkts_burst)
{
    return (netdev->netdev_class->set_policing
            ? netdev->netdev_class->set_policing(netdev,
                    kbits_rate, kbits_burst, kpkts_rate, kpkts_burst)
            : EOPNOTSUPP);
}

/* Adds to 'types' all of the forms of QoS supported by 'netdev', or leaves it
 * empty if 'netdev' does not support QoS.  Any names added to 'types' should
 * be documented as valid for the "type" column in the "QoS" table in
 * vswitchd/vswitch.xml (which is built as ovs-vswitchd.conf.db(8)).
 *
 * Every network device supports disabling QoS with a type of "", but this type
 * will not be added to 'types'.
 *
 * The caller must initialize 'types' (e.g. with sset_init()) before calling
 * this function.  The caller is responsible for destroying 'types' (e.g. with
 * sset_destroy()) when it is no longer needed.
 *
 * Returns 0 if successful, otherwise a positive errno value. */
int
netdev_get_qos_types(const struct netdev *netdev, struct sset *types)
{
    const struct netdev_class *class = netdev->netdev_class;
    return (class->get_qos_types
            ? class->get_qos_types(netdev, types)
            : 0);
}

/* Queries 'netdev' for its capabilities regarding the specified 'type' of QoS,
 * which should be "" or one of the types returned by netdev_get_qos_types()
 * for 'netdev'.  Returns 0 if successful, otherwise a positive errno value.
 * On success, initializes 'caps' with the QoS capabilities; on failure, clears
 * 'caps' to all zeros. */
int
netdev_get_qos_capabilities(const struct netdev *netdev, const char *type,
                            struct netdev_qos_capabilities *caps)
{
    const struct netdev_class *class = netdev->netdev_class;

    if (*type) {
        int retval = (class->get_qos_capabilities
                      ? class->get_qos_capabilities(netdev, type, caps)
                      : EOPNOTSUPP);
        if (retval) {
            memset(caps, 0, sizeof *caps);
        }
        return retval;
    } else {
        /* Every netdev supports turning off QoS. */
        memset(caps, 0, sizeof *caps);
        return 0;
    }
}

/* Obtains the number of queues supported by 'netdev' for the specified 'type'
 * of QoS.  Returns 0 if successful, otherwise a positive errno value.  Stores
 * the number of queues (zero on failure) in '*n_queuesp'.
 *
 * This is just a simple wrapper around netdev_get_qos_capabilities(). */
int
netdev_get_n_queues(const struct netdev *netdev,
                    const char *type, unsigned int *n_queuesp)
{
    struct netdev_qos_capabilities caps;
    int retval;

    retval = netdev_get_qos_capabilities(netdev, type, &caps);
    *n_queuesp = caps.n_queues;
    return retval;
}

/* Queries 'netdev' about its currently configured form of QoS.  If successful,
 * stores the name of the current form of QoS into '*typep', stores any details
 * of configuration as string key-value pairs in 'details', and returns 0.  On
 * failure, sets '*typep' to NULL and returns a positive errno value.
 *
 * A '*typep' of "" indicates that QoS is currently disabled on 'netdev'.
 *
 * The caller must initialize 'details' as an empty smap (e.g. with
 * smap_init()) before calling this function.  The caller must free 'details'
 * when it is no longer needed (e.g. with smap_destroy()).
 *
 * The caller must not modify or free '*typep'.
 *
 * '*typep' will be one of the types returned by netdev_get_qos_types() for
 * 'netdev'.  The contents of 'details' should be documented as valid for
 * '*typep' in the "other_config" column in the "QoS" table in
 * vswitchd/vswitch.xml (which is built as ovs-vswitchd.conf.db(8)). */
int
netdev_get_qos(const struct netdev *netdev,
               const char **typep, struct smap *details)
{
    const struct netdev_class *class = netdev->netdev_class;
    int retval;

    if (class->get_qos) {
        retval = class->get_qos(netdev, typep, details);
        if (retval) {
            *typep = NULL;
            smap_clear(details);
        }
        return retval;
    } else {
        /* 'netdev' doesn't support QoS, so report that QoS is disabled. */
        *typep = "";
        return 0;
    }
}

/* Attempts to reconfigure QoS on 'netdev', changing the form of QoS to 'type'
 * with details of configuration from 'details'.  Returns 0 if successful,
 * otherwise a positive errno value.  On error, the previous QoS configuration
 * is retained.
 *
 * When this function changes the type of QoS (not just 'details'), this also
 * resets all queue configuration for 'netdev' to their defaults (which depend
 * on the specific type of QoS).  Otherwise, the queue configuration for
 * 'netdev' is unchanged.
 *
 * 'type' should be "" (to disable QoS) or one of the types returned by
 * netdev_get_qos_types() for 'netdev'.  The contents of 'details' should be
 * documented as valid for the given 'type' in the "other_config" column in the
 * "QoS" table in vswitchd/vswitch.xml (which is built as
 * ovs-vswitchd.conf.db(8)).
 *
 * NULL may be specified for 'details' if there are no configuration
 * details. */
int
netdev_set_qos(struct netdev *netdev,
               const char *type, const struct smap *details)
{
    const struct netdev_class *class = netdev->netdev_class;

    if (!type) {
        type = "";
    }

    if (class->set_qos) {
        if (!details) {
            static const struct smap empty = SMAP_INITIALIZER(&empty);
            details = &empty;
        }
        return class->set_qos(netdev, type, details);
    } else {
        return *type ? EOPNOTSUPP : 0;
    }
}

/* Queries 'netdev' for information about the queue numbered 'queue_id'.  If
 * successful, adds that information as string key-value pairs to 'details'.
 * Returns 0 if successful, otherwise a positive errno value.
 *
 * 'queue_id' must be less than the number of queues supported by 'netdev' for
 * the current form of QoS (e.g. as returned by netdev_get_n_queues(netdev)).
 *
 * The returned contents of 'details' should be documented as valid for the
 * given 'type' in the "other_config" column in the "Queue" table in
 * vswitchd/vswitch.xml (which is built as ovs-vswitchd.conf.db(8)).
 *
 * The caller must initialize 'details' (e.g. with smap_init()) before calling
 * this function.  The caller must free 'details' when it is no longer needed
 * (e.g. with smap_destroy()). */
int
netdev_get_queue(const struct netdev *netdev,
                 unsigned int queue_id, struct smap *details)
{
    const struct netdev_class *class = netdev->netdev_class;
    int retval;

    retval = (class->get_queue
              ? class->get_queue(netdev, queue_id, details)
              : EOPNOTSUPP);
    if (retval) {
        smap_clear(details);
    }
    return retval;
}

/* Configures the queue numbered 'queue_id' on 'netdev' with the key-value
 * string pairs in 'details'.  The contents of 'details' should be documented
 * as valid for the given 'type' in the "other_config" column in the "Queue"
 * table in vswitchd/vswitch.xml (which is built as ovs-vswitchd.conf.db(8)).
 * Returns 0 if successful, otherwise a positive errno value.  On failure, the
 * given queue's configuration should be unmodified.
 *
 * 'queue_id' must be less than the number of queues supported by 'netdev' for
 * the current form of QoS (e.g. as returned by netdev_get_n_queues(netdev)).
 *
 * This function does not modify 'details', and the caller retains ownership of
 * it. */
int
netdev_set_queue(struct netdev *netdev,
                 unsigned int queue_id, const struct smap *details)
{
    const struct netdev_class *class = netdev->netdev_class;
    return (class->set_queue
            ? class->set_queue(netdev, queue_id, details)
            : EOPNOTSUPP);
}

/* Attempts to delete the queue numbered 'queue_id' from 'netdev'.  Some kinds
 * of QoS may have a fixed set of queues, in which case attempts to delete them
 * will fail with EOPNOTSUPP.
 *
 * Returns 0 if successful, otherwise a positive errno value.  On failure, the
 * given queue will be unmodified.
 *
 * 'queue_id' must be less than the number of queues supported by 'netdev' for
 * the current form of QoS (e.g. as returned by
 * netdev_get_n_queues(netdev)). */
int
netdev_delete_queue(struct netdev *netdev, unsigned int queue_id)
{
    const struct netdev_class *class = netdev->netdev_class;
    return (class->delete_queue
            ? class->delete_queue(netdev, queue_id)
            : EOPNOTSUPP);
}

/* Obtains statistics about 'queue_id' on 'netdev'.  On success, returns 0 and
 * fills 'stats' with the queue's statistics; individual members of 'stats' may
 * be set to all-1-bits if the statistic is unavailable.  On failure, returns a
 * positive errno value and fills 'stats' with values indicating unsupported
 * statistics. */
int
netdev_get_queue_stats(const struct netdev *netdev, unsigned int queue_id,
                       struct netdev_queue_stats *stats)
{
    const struct netdev_class *class = netdev->netdev_class;
    int retval;

    retval = (class->get_queue_stats
              ? class->get_queue_stats(netdev, queue_id, stats)
              : EOPNOTSUPP);
    if (retval) {
        stats->tx_bytes = UINT64_MAX;
        stats->tx_packets = UINT64_MAX;
        stats->tx_errors = UINT64_MAX;
        stats->created = LLONG_MIN;
    }
    return retval;
}

/* Initializes 'dump' to begin dumping the queues in a netdev.
 *
 * This function provides no status indication.  An error status for the entire
 * dump operation is provided when it is completed by calling
 * netdev_queue_dump_done().
 */
void
netdev_queue_dump_start(struct netdev_queue_dump *dump,
                        const struct netdev *netdev)
{
    dump->netdev = netdev_ref(netdev);
    if (netdev->netdev_class->queue_dump_start) {
        dump->error = netdev->netdev_class->queue_dump_start(netdev,
                                                             &dump->state);
    } else {
        dump->error = EOPNOTSUPP;
    }
}

/* Attempts to retrieve another queue from 'dump', which must have been
 * initialized with netdev_queue_dump_start().  On success, stores a new queue
 * ID into '*queue_id', fills 'details' with configuration details for the
 * queue, and returns true.  On failure, returns false.
 *
 * Queues are not necessarily dumped in increasing order of queue ID (or any
 * other predictable order).
 *
 * Failure might indicate an actual error or merely that the last queue has
 * been dumped.  An error status for the entire dump operation is provided when
 * it is completed by calling netdev_queue_dump_done().
 *
 * The returned contents of 'details' should be documented as valid for the
 * given 'type' in the "other_config" column in the "Queue" table in
 * vswitchd/vswitch.xml (which is built as ovs-vswitchd.conf.db(8)).
 *
 * The caller must initialize 'details' (e.g. with smap_init()) before calling
 * this function.  This function will clear and replace its contents.  The
 * caller must free 'details' when it is no longer needed (e.g. with
 * smap_destroy()). */
bool
netdev_queue_dump_next(struct netdev_queue_dump *dump,
                       unsigned int *queue_id, struct smap *details)
{
    smap_clear(details);

    const struct netdev *netdev = dump->netdev;
    if (dump->error) {
        return false;
    }

    dump->error = netdev->netdev_class->queue_dump_next(netdev, dump->state,
                                                        queue_id, details);

    if (dump->error) {
        netdev->netdev_class->queue_dump_done(netdev, dump->state);
        return false;
    }
    return true;
}

/* Completes queue table dump operation 'dump', which must have been
 * initialized with netdev_queue_dump_start().  Returns 0 if the dump operation
 * was error-free, otherwise a positive errno value describing the problem. */
int
netdev_queue_dump_done(struct netdev_queue_dump *dump)
{
    const struct netdev *netdev = dump->netdev;
    if (!dump->error && netdev->netdev_class->queue_dump_done) {
        dump->error = netdev->netdev_class->queue_dump_done(netdev,
                                                            dump->state);
    }
    netdev_close(dump->netdev);
    return dump->error == EOF ? 0 : dump->error;
}

/* Iterates over all of 'netdev''s queues, calling 'cb' with the queue's ID,
 * its statistics, and the 'aux' specified by the caller.  The order of
 * iteration is unspecified, but (when successful) each queue is visited
 * exactly once.
 *
 * Calling this function may be more efficient than calling
 * netdev_get_queue_stats() for every queue.
 *
 * 'cb' must not modify or free the statistics passed in.
 *
 * Returns 0 if successful, otherwise a positive errno value.  On error, some
 * configured queues may not have been included in the iteration. */
int
netdev_dump_queue_stats(const struct netdev *netdev,
                        netdev_dump_queue_stats_cb *cb, void *aux)
{
    const struct netdev_class *class = netdev->netdev_class;
    return (class->dump_queue_stats
            ? class->dump_queue_stats(netdev, cb, aux)
            : EOPNOTSUPP);
}


/* Returns the class type of 'netdev'.
 *
 * The caller must not free the returned value. */
const char *
netdev_get_type(const struct netdev *netdev)
{
    return netdev->netdev_class->type;
}

/* Returns the class associated with 'netdev'. */
const struct netdev_class *
netdev_get_class(const struct netdev *netdev)
{
    return netdev->netdev_class;
}

/* Set the type of 'dpif' this 'netdev' belongs to. */
void
netdev_set_dpif_type(struct netdev *netdev, const char *type)
{
    netdev->dpif_type = type;
}

/* Returns the type of 'dpif' this 'netdev' belongs to.
 *
 * The caller must not free the returned value. */
const char *
netdev_get_dpif_type(const struct netdev *netdev)
{
    return netdev->dpif_type;
}

/* Returns the netdev with 'name' or NULL if there is none.
 *
 * The caller must free the returned netdev with netdev_close(). */
struct netdev *
netdev_from_name(const char *name)
    OVS_EXCLUDED(netdev_mutex)
{
    struct netdev *netdev;

    ovs_mutex_lock(&netdev_mutex);
    netdev = shash_find_data(&netdev_shash, name);
    if (netdev) {
        netdev->ref_cnt++;
    }
    ovs_mutex_unlock(&netdev_mutex);

    return netdev;
}

/* Fills 'device_list' with devices that match 'netdev_class'.
 *
 * The caller is responsible for initializing and destroying 'device_list' and
 * must close each device on the list. */
void
netdev_get_devices(const struct netdev_class *netdev_class,
                   struct shash *device_list)
    OVS_EXCLUDED(netdev_mutex)
{
    struct shash_node *node;

    ovs_mutex_lock(&netdev_mutex);
    SHASH_FOR_EACH (node, &netdev_shash) {
        struct netdev *dev = node->data;

        if (dev->netdev_class == netdev_class) {
            dev->ref_cnt++;
            shash_add(device_list, node->name, node->data);
        }
    }
    ovs_mutex_unlock(&netdev_mutex);
}

/* Extracts pointers to all 'netdev-vports' into an array 'vports'
 * and returns it.  Stores the size of the array into '*size'.
 *
 * The caller is responsible for freeing 'vports' and must close
 * each 'netdev-vport' in the list. */
struct netdev **
netdev_get_vports(size_t *size)
    OVS_EXCLUDED(netdev_mutex)
{
    struct netdev **vports;
    struct shash_node *node;
    size_t n = 0;

    if (!size) {
        return NULL;
    }

    /* Explicitly allocates big enough chunk of memory. */
    ovs_mutex_lock(&netdev_mutex);
    vports = xmalloc(shash_count(&netdev_shash) * sizeof *vports);
    SHASH_FOR_EACH (node, &netdev_shash) {
        struct netdev *dev = node->data;

        if (netdev_vport_is_vport_class(dev->netdev_class)) {
            dev->ref_cnt++;
            vports[n] = dev;
            n++;
        }
    }
    ovs_mutex_unlock(&netdev_mutex);
    *size = n;

    return vports;
}

const char *
netdev_get_type_from_name(const char *name)
{
    struct netdev *dev;
    const char *type;
    type = netdev_vport_type_from_name(name);
    if (type == NULL) {
        dev = netdev_from_name(name);
        type = dev ? netdev_get_type(dev) : NULL;
        netdev_close(dev);
    }
    return type;
}

struct netdev *
netdev_rxq_get_netdev(const struct netdev_rxq *rx)
{
    ovs_assert(rx->netdev->ref_cnt > 0);
    return rx->netdev;
}

const char *
netdev_rxq_get_name(const struct netdev_rxq *rx)
{
    return netdev_get_name(netdev_rxq_get_netdev(rx));
}

int
netdev_rxq_get_queue_id(const struct netdev_rxq *rx)
{
    return rx->queue_id;
}

static void
restore_all_flags(void *aux OVS_UNUSED)
{
    struct shash_node *node;

    SHASH_FOR_EACH (node, &netdev_shash) {
        struct netdev *netdev = node->data;
        const struct netdev_saved_flags *sf;
        enum netdev_flags saved_values;
        enum netdev_flags saved_flags;

        saved_values = saved_flags = 0;
        LIST_FOR_EACH (sf, node, &netdev->saved_flags_list) {
            saved_flags |= sf->saved_flags;
            saved_values &= ~sf->saved_flags;
            saved_values |= sf->saved_flags & sf->saved_values;
        }
        if (saved_flags) {
            enum netdev_flags old_flags;

            netdev->netdev_class->update_flags(netdev,
                                               saved_flags & saved_values,
                                               saved_flags & ~saved_values,
                                               &old_flags);
        }
#ifdef HAVE_AF_XDP
        if (netdev->netdev_class == &netdev_afxdp_class) {
            signal_remove_xdp(netdev);
        }
#endif
    }
}

uint64_t
netdev_get_change_seq(const struct netdev *netdev)
{
    uint64_t change_seq;

    atomic_read_explicit(&CONST_CAST(struct netdev *, netdev)->change_seq,
                        &change_seq, memory_order_acquire);

    return change_seq;
}

#ifndef _WIN32
/* This implementation is shared by Linux and BSD. */

static struct ifaddrs *if_addr_list;
static struct ovs_mutex if_addr_list_lock = OVS_MUTEX_INITIALIZER;

void
netdev_get_addrs_list_flush(void)
{
    ovs_mutex_lock(&if_addr_list_lock);
    if (if_addr_list) {
        freeifaddrs(if_addr_list);
        if_addr_list = NULL;
    }
    ovs_mutex_unlock(&if_addr_list_lock);
}

int
netdev_get_addrs(const char dev[], struct in6_addr **paddr,
                 struct in6_addr **pmask, int *n_in)
{
    struct in6_addr *addr_array, *mask_array;
    const struct ifaddrs *ifa;
    int cnt = 0, i = 0;
    int retries = 3;

    ovs_mutex_lock(&if_addr_list_lock);
    if (!if_addr_list) {
        int err;

retry:
        err = getifaddrs(&if_addr_list);
        if (err) {
            ovs_mutex_unlock(&if_addr_list_lock);
            return -err;
        }
        retries--;
    }

    for (ifa = if_addr_list; ifa; ifa = ifa->ifa_next) {
        if (!ifa->ifa_name) {
            if (retries) {
                /* Older versions of glibc have a bug on race condition with
                 * address addition which may cause one of the returned
                 * ifa_name values to be NULL. In such case, we know that we've
                 * got an inconsistent dump. Retry but beware of an endless
                 * loop. From glibc 2.28 and beyond, this workaround is not
                 * needed and should be eventually removed. */
                freeifaddrs(if_addr_list);
                goto retry;
            } else {
                VLOG_WARN("Proceeding with an inconsistent dump of "
                          "interfaces from the kernel. Some may be missing");
            }
        }
        if (ifa->ifa_addr && ifa->ifa_name && ifa->ifa_netmask) {
            int family;

            family = ifa->ifa_addr->sa_family;
            if (family == AF_INET || family == AF_INET6) {
                if (!strncmp(ifa->ifa_name, dev, IFNAMSIZ)) {
                    cnt++;
                }
            }
        }
    }

    if (!cnt) {
        ovs_mutex_unlock(&if_addr_list_lock);
        return EADDRNOTAVAIL;
    }
    addr_array = xzalloc(sizeof *addr_array * cnt);
    mask_array = xzalloc(sizeof *mask_array * cnt);
    for (ifa = if_addr_list; ifa; ifa = ifa->ifa_next) {
        if (ifa->ifa_name
            && ifa->ifa_addr
            && ifa->ifa_netmask
            && !strncmp(ifa->ifa_name, dev, IFNAMSIZ)
            && sa_is_ip(ifa->ifa_addr)) {
            addr_array[i] = sa_get_address(ifa->ifa_addr);
            mask_array[i] = sa_get_address(ifa->ifa_netmask);
            i++;
        }
    }
    ovs_mutex_unlock(&if_addr_list_lock);
    if (paddr) {
        *n_in = cnt;
        *paddr = addr_array;
        *pmask = mask_array;
    } else {
        free(addr_array);
        free(mask_array);
    }
    return 0;
}
#endif

void
netdev_wait_reconf_required(struct netdev *netdev)
{
    seq_wait(netdev->reconfigure_seq, netdev->last_reconfigure_seq);
}

bool
netdev_is_reconf_required(struct netdev *netdev)
{
    return seq_read(netdev->reconfigure_seq) != netdev->last_reconfigure_seq;
}

/* Give a chance to 'netdev' to reconfigure some of its parameters.
 *
 * If a module uses netdev_send() and netdev_rxq_recv(), it must call this
 * function when netdev_is_reconf_required() returns true.
 *
 * Return 0 if successful, otherwise a positive errno value.  If the
 * reconfiguration fails the netdev will not be able to send or receive
 * packets.
 *
 * When this function is called, no call to netdev_rxq_recv() or netdev_send()
 * must be issued. */
int
netdev_reconfigure(struct netdev *netdev)
{
    const struct netdev_class *class = netdev->netdev_class;

    netdev->last_reconfigure_seq = seq_read(netdev->reconfigure_seq);

    return (class->reconfigure
            ? class->reconfigure(netdev)
            : EOPNOTSUPP);
}

void
netdev_free_custom_stats_counters(struct netdev_custom_stats *custom_stats)
{
    if (custom_stats) {
        if (custom_stats->counters) {
            free(custom_stats->counters);
            custom_stats->counters = NULL;
            custom_stats->size = 0;
        }
    }
}