aboutsummaryrefslogtreecommitdiff
path: root/sys/kern/kern_descrip.c
blob: 1f422bbf57c046b4c7b8bf586e52624466e2c18a (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
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
/*-
 * SPDX-License-Identifier: BSD-3-Clause
 *
 * Copyright (c) 1982, 1986, 1989, 1991, 1993
 *	The Regents of the University of California.  All rights reserved.
 * (c) UNIX System Laboratories, Inc.
 * All or some portions of this file are derived from material licensed
 * to the University of California by American Telephone and Telegraph
 * Co. or Unix System Laboratories, Inc. and are reproduced herein with
 * the permission of UNIX System Laboratories, Inc.
 *
 * Redistribution and use in source and binary forms, with or without
 * modification, are permitted provided that the following conditions
 * are met:
 * 1. Redistributions of source code must retain the above copyright
 *    notice, this list of conditions and the following disclaimer.
 * 2. Redistributions in binary form must reproduce the above copyright
 *    notice, this list of conditions and the following disclaimer in the
 *    documentation and/or other materials provided with the distribution.
 * 3. Neither the name of the University nor the names of its contributors
 *    may be used to endorse or promote products derived from this software
 *    without specific prior written permission.
 *
 * THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``AS IS'' AND
 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
 * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
 * ARE DISCLAIMED.  IN NO EVENT SHALL THE REGENTS OR CONTRIBUTORS BE LIABLE
 * FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
 * DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
 * OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
 * HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
 * LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
 * OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
 * SUCH DAMAGE.
 *
 *	@(#)kern_descrip.c	8.6 (Berkeley) 4/19/94
 */

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

#include "opt_capsicum.h"
#include "opt_ddb.h"
#include "opt_ktrace.h"

#include <sys/param.h>
#include <sys/systm.h>

#include <sys/capsicum.h>
#include <sys/conf.h>
#include <sys/fcntl.h>
#include <sys/file.h>
#include <sys/filedesc.h>
#include <sys/filio.h>
#include <sys/jail.h>
#include <sys/kernel.h>
#include <sys/limits.h>
#include <sys/lock.h>
#include <sys/malloc.h>
#include <sys/mount.h>
#include <sys/mutex.h>
#include <sys/namei.h>
#include <sys/selinfo.h>
#include <sys/priv.h>
#include <sys/proc.h>
#include <sys/protosw.h>
#include <sys/racct.h>
#include <sys/resourcevar.h>
#include <sys/sbuf.h>
#include <sys/signalvar.h>
#include <sys/kdb.h>
#include <sys/smr.h>
#include <sys/stat.h>
#include <sys/sx.h>
#include <sys/syscallsubr.h>
#include <sys/sysctl.h>
#include <sys/sysproto.h>
#include <sys/unistd.h>
#include <sys/user.h>
#include <sys/vnode.h>
#ifdef KTRACE
#include <sys/ktrace.h>
#endif

#include <net/vnet.h>

#include <security/audit/audit.h>

#include <vm/uma.h>
#include <vm/vm.h>

#include <ddb/ddb.h>

static MALLOC_DEFINE(M_FILEDESC, "filedesc", "Open file descriptor table");
static MALLOC_DEFINE(M_PWD, "pwd", "Descriptor table vnodes");
static MALLOC_DEFINE(M_FILEDESC_TO_LEADER, "filedesc_to_leader",
    "file desc to leader structures");
static MALLOC_DEFINE(M_SIGIO, "sigio", "sigio structures");
MALLOC_DEFINE(M_FILECAPS, "filecaps", "descriptor capabilities");

MALLOC_DECLARE(M_FADVISE);

static __read_mostly uma_zone_t file_zone;
static __read_mostly uma_zone_t filedesc0_zone;
__read_mostly uma_zone_t pwd_zone;
VFS_SMR_DECLARE;

static int	closefp(struct filedesc *fdp, int fd, struct file *fp,
		    struct thread *td, int holdleaders);
static int	fd_first_free(struct filedesc *fdp, int low, int size);
static void	fdgrowtable(struct filedesc *fdp, int nfd);
static void	fdgrowtable_exp(struct filedesc *fdp, int nfd);
static void	fdunused(struct filedesc *fdp, int fd);
static void	fdused(struct filedesc *fdp, int fd);
static int	getmaxfd(struct thread *td);
static u_long	*filecaps_copy_prep(const struct filecaps *src);
static void	filecaps_copy_finish(const struct filecaps *src,
		    struct filecaps *dst, u_long *ioctls);
static u_long 	*filecaps_free_prep(struct filecaps *fcaps);
static void	filecaps_free_finish(u_long *ioctls);

static struct pwd *pwd_alloc(void);

/*
 * Each process has:
 *
 * - An array of open file descriptors (fd_ofiles)
 * - An array of file flags (fd_ofileflags)
 * - A bitmap recording which descriptors are in use (fd_map)
 *
 * A process starts out with NDFILE descriptors.  The value of NDFILE has
 * been selected based the historical limit of 20 open files, and an
 * assumption that the majority of processes, especially short-lived
 * processes like shells, will never need more.
 *
 * If this initial allocation is exhausted, a larger descriptor table and
 * map are allocated dynamically, and the pointers in the process's struct
 * filedesc are updated to point to those.  This is repeated every time
 * the process runs out of file descriptors (provided it hasn't hit its
 * resource limit).
 *
 * Since threads may hold references to individual descriptor table
 * entries, the tables are never freed.  Instead, they are placed on a
 * linked list and freed only when the struct filedesc is released.
 */
#define NDFILE		20
#define NDSLOTSIZE	sizeof(NDSLOTTYPE)
#define	NDENTRIES	(NDSLOTSIZE * __CHAR_BIT)
#define NDSLOT(x)	((x) / NDENTRIES)
#define NDBIT(x)	((NDSLOTTYPE)1 << ((x) % NDENTRIES))
#define	NDSLOTS(x)	(((x) + NDENTRIES - 1) / NDENTRIES)

/*
 * SLIST entry used to keep track of ofiles which must be reclaimed when
 * the process exits.
 */
struct freetable {
	struct fdescenttbl *ft_table;
	SLIST_ENTRY(freetable) ft_next;
};

/*
 * Initial allocation: a filedesc structure + the head of SLIST used to
 * keep track of old ofiles + enough space for NDFILE descriptors.
 */

struct fdescenttbl0 {
	int	fdt_nfiles;
	struct	filedescent fdt_ofiles[NDFILE];
};

struct filedesc0 {
	struct filedesc fd_fd;
	SLIST_HEAD(, freetable) fd_free;
	struct	fdescenttbl0 fd_dfiles;
	NDSLOTTYPE fd_dmap[NDSLOTS(NDFILE)];
};

/*
 * Descriptor management.
 */
static int __exclusive_cache_line openfiles; /* actual number of open files */
struct mtx sigio_lock;		/* mtx to protect pointers to sigio */
void __read_mostly (*mq_fdclose)(struct thread *td, int fd, struct file *fp);

/*
 * If low >= size, just return low. Otherwise find the first zero bit in the
 * given bitmap, starting at low and not exceeding size - 1. Return size if
 * not found.
 */
static int
fd_first_free(struct filedesc *fdp, int low, int size)
{
	NDSLOTTYPE *map = fdp->fd_map;
	NDSLOTTYPE mask;
	int off, maxoff;

	if (low >= size)
		return (low);

	off = NDSLOT(low);
	if (low % NDENTRIES) {
		mask = ~(~(NDSLOTTYPE)0 >> (NDENTRIES - (low % NDENTRIES)));
		if ((mask &= ~map[off]) != 0UL)
			return (off * NDENTRIES + ffsl(mask) - 1);
		++off;
	}
	for (maxoff = NDSLOTS(size); off < maxoff; ++off)
		if (map[off] != ~0UL)
			return (off * NDENTRIES + ffsl(~map[off]) - 1);
	return (size);
}

/*
 * Find the last used fd.
 *
 * Call this variant if fdp can't be modified by anyone else (e.g, during exec).
 * Otherwise use fdlastfile.
 */
int
fdlastfile_single(struct filedesc *fdp)
{
	NDSLOTTYPE *map = fdp->fd_map;
	int off, minoff;

	off = NDSLOT(fdp->fd_nfiles - 1);
	for (minoff = NDSLOT(0); off >= minoff; --off)
		if (map[off] != 0)
			return (off * NDENTRIES + flsl(map[off]) - 1);
	return (-1);
}

int
fdlastfile(struct filedesc *fdp)
{

	FILEDESC_LOCK_ASSERT(fdp);
	return (fdlastfile_single(fdp));
}

static int
fdisused(struct filedesc *fdp, int fd)
{

	KASSERT(fd >= 0 && fd < fdp->fd_nfiles,
	    ("file descriptor %d out of range (0, %d)", fd, fdp->fd_nfiles));

	return ((fdp->fd_map[NDSLOT(fd)] & NDBIT(fd)) != 0);
}

/*
 * Mark a file descriptor as used.
 */
static void
fdused_init(struct filedesc *fdp, int fd)
{

	KASSERT(!fdisused(fdp, fd), ("fd=%d is already used", fd));

	fdp->fd_map[NDSLOT(fd)] |= NDBIT(fd);
}

static void
fdused(struct filedesc *fdp, int fd)
{

	FILEDESC_XLOCK_ASSERT(fdp);

	fdused_init(fdp, fd);
	if (fd == fdp->fd_freefile)
		fdp->fd_freefile++;
}

/*
 * Mark a file descriptor as unused.
 */
static void
fdunused(struct filedesc *fdp, int fd)
{

	FILEDESC_XLOCK_ASSERT(fdp);

	KASSERT(fdisused(fdp, fd), ("fd=%d is already unused", fd));
	KASSERT(fdp->fd_ofiles[fd].fde_file == NULL,
	    ("fd=%d is still in use", fd));

	fdp->fd_map[NDSLOT(fd)] &= ~NDBIT(fd);
	if (fd < fdp->fd_freefile)
		fdp->fd_freefile = fd;
}

/*
 * Free a file descriptor.
 *
 * Avoid some work if fdp is about to be destroyed.
 */
static inline void
fdefree_last(struct filedescent *fde)
{

	filecaps_free(&fde->fde_caps);
}

static inline void
fdfree(struct filedesc *fdp, int fd)
{
	struct filedescent *fde;

	fde = &fdp->fd_ofiles[fd];
#ifdef CAPABILITIES
	seqc_write_begin(&fde->fde_seqc);
#endif
	fde->fde_file = NULL;
#ifdef CAPABILITIES
	seqc_write_end(&fde->fde_seqc);
#endif
	fdefree_last(fde);
	fdunused(fdp, fd);
}

/*
 * System calls on descriptors.
 */
#ifndef _SYS_SYSPROTO_H_
struct getdtablesize_args {
	int	dummy;
};
#endif
/* ARGSUSED */
int
sys_getdtablesize(struct thread *td, struct getdtablesize_args *uap)
{
#ifdef	RACCT
	uint64_t lim;
#endif

	td->td_retval[0] = getmaxfd(td);
#ifdef	RACCT
	PROC_LOCK(td->td_proc);
	lim = racct_get_limit(td->td_proc, RACCT_NOFILE);
	PROC_UNLOCK(td->td_proc);
	if (lim < td->td_retval[0])
		td->td_retval[0] = lim;
#endif
	return (0);
}

/*
 * Duplicate a file descriptor to a particular value.
 *
 * Note: keep in mind that a potential race condition exists when closing
 * descriptors from a shared descriptor table (via rfork).
 */
#ifndef _SYS_SYSPROTO_H_
struct dup2_args {
	u_int	from;
	u_int	to;
};
#endif
/* ARGSUSED */
int
sys_dup2(struct thread *td, struct dup2_args *uap)
{

	return (kern_dup(td, FDDUP_FIXED, 0, (int)uap->from, (int)uap->to));
}

/*
 * Duplicate a file descriptor.
 */
#ifndef _SYS_SYSPROTO_H_
struct dup_args {
	u_int	fd;
};
#endif
/* ARGSUSED */
int
sys_dup(struct thread *td, struct dup_args *uap)
{

	return (kern_dup(td, FDDUP_NORMAL, 0, (int)uap->fd, 0));
}

/*
 * The file control system call.
 */
#ifndef _SYS_SYSPROTO_H_
struct fcntl_args {
	int	fd;
	int	cmd;
	long	arg;
};
#endif
/* ARGSUSED */
int
sys_fcntl(struct thread *td, struct fcntl_args *uap)
{

	return (kern_fcntl_freebsd(td, uap->fd, uap->cmd, uap->arg));
}

int
kern_fcntl_freebsd(struct thread *td, int fd, int cmd, long arg)
{
	struct flock fl;
	struct __oflock ofl;
	intptr_t arg1;
	int error, newcmd;

	error = 0;
	newcmd = cmd;
	switch (cmd) {
	case F_OGETLK:
	case F_OSETLK:
	case F_OSETLKW:
		/*
		 * Convert old flock structure to new.
		 */
		error = copyin((void *)(intptr_t)arg, &ofl, sizeof(ofl));
		fl.l_start = ofl.l_start;
		fl.l_len = ofl.l_len;
		fl.l_pid = ofl.l_pid;
		fl.l_type = ofl.l_type;
		fl.l_whence = ofl.l_whence;
		fl.l_sysid = 0;

		switch (cmd) {
		case F_OGETLK:
			newcmd = F_GETLK;
			break;
		case F_OSETLK:
			newcmd = F_SETLK;
			break;
		case F_OSETLKW:
			newcmd = F_SETLKW;
			break;
		}
		arg1 = (intptr_t)&fl;
		break;
	case F_GETLK:
	case F_SETLK:
	case F_SETLKW:
	case F_SETLK_REMOTE:
		error = copyin((void *)(intptr_t)arg, &fl, sizeof(fl));
		arg1 = (intptr_t)&fl;
		break;
	default:
		arg1 = arg;
		break;
	}
	if (error)
		return (error);
	error = kern_fcntl(td, fd, newcmd, arg1);
	if (error)
		return (error);
	if (cmd == F_OGETLK) {
		ofl.l_start = fl.l_start;
		ofl.l_len = fl.l_len;
		ofl.l_pid = fl.l_pid;
		ofl.l_type = fl.l_type;
		ofl.l_whence = fl.l_whence;
		error = copyout(&ofl, (void *)(intptr_t)arg, sizeof(ofl));
	} else if (cmd == F_GETLK) {
		error = copyout(&fl, (void *)(intptr_t)arg, sizeof(fl));
	}
	return (error);
}

int
kern_fcntl(struct thread *td, int fd, int cmd, intptr_t arg)
{
	struct filedesc *fdp;
	struct flock *flp;
	struct file *fp, *fp2;
	struct filedescent *fde;
	struct proc *p;
	struct vnode *vp;
	struct mount *mp;
	int error, flg, seals, tmp;
	uint64_t bsize;
	off_t foffset;

	error = 0;
	flg = F_POSIX;
	p = td->td_proc;
	fdp = p->p_fd;

	AUDIT_ARG_FD(cmd);
	AUDIT_ARG_CMD(cmd);
	switch (cmd) {
	case F_DUPFD:
		tmp = arg;
		error = kern_dup(td, FDDUP_FCNTL, 0, fd, tmp);
		break;

	case F_DUPFD_CLOEXEC:
		tmp = arg;
		error = kern_dup(td, FDDUP_FCNTL, FDDUP_FLAG_CLOEXEC, fd, tmp);
		break;

	case F_DUP2FD:
		tmp = arg;
		error = kern_dup(td, FDDUP_FIXED, 0, fd, tmp);
		break;

	case F_DUP2FD_CLOEXEC:
		tmp = arg;
		error = kern_dup(td, FDDUP_FIXED, FDDUP_FLAG_CLOEXEC, fd, tmp);
		break;

	case F_GETFD:
		error = EBADF;
		FILEDESC_SLOCK(fdp);
		fde = fdeget_locked(fdp, fd);
		if (fde != NULL) {
			td->td_retval[0] =
			    (fde->fde_flags & UF_EXCLOSE) ? FD_CLOEXEC : 0;
			error = 0;
		}
		FILEDESC_SUNLOCK(fdp);
		break;

	case F_SETFD:
		error = EBADF;
		FILEDESC_XLOCK(fdp);
		fde = fdeget_locked(fdp, fd);
		if (fde != NULL) {
			fde->fde_flags = (fde->fde_flags & ~UF_EXCLOSE) |
			    (arg & FD_CLOEXEC ? UF_EXCLOSE : 0);
			error = 0;
		}
		FILEDESC_XUNLOCK(fdp);
		break;

	case F_GETFL:
		error = fget_fcntl(td, fd, &cap_fcntl_rights, F_GETFL, &fp);
		if (error != 0)
			break;
		td->td_retval[0] = OFLAGS(fp->f_flag);
		fdrop(fp, td);
		break;

	case F_SETFL:
		error = fget_fcntl(td, fd, &cap_fcntl_rights, F_SETFL, &fp);
		if (error != 0)
			break;
		do {
			tmp = flg = fp->f_flag;
			tmp &= ~FCNTLFLAGS;
			tmp |= FFLAGS(arg & ~O_ACCMODE) & FCNTLFLAGS;
		} while(atomic_cmpset_int(&fp->f_flag, flg, tmp) == 0);
		tmp = fp->f_flag & FNONBLOCK;
		error = fo_ioctl(fp, FIONBIO, &tmp, td->td_ucred, td);
		if (error != 0) {
			fdrop(fp, td);
			break;
		}
		tmp = fp->f_flag & FASYNC;
		error = fo_ioctl(fp, FIOASYNC, &tmp, td->td_ucred, td);
		if (error == 0) {
			fdrop(fp, td);
			break;
		}
		atomic_clear_int(&fp->f_flag, FNONBLOCK);
		tmp = 0;
		(void)fo_ioctl(fp, FIONBIO, &tmp, td->td_ucred, td);
		fdrop(fp, td);
		break;

	case F_GETOWN:
		error = fget_fcntl(td, fd, &cap_fcntl_rights, F_GETOWN, &fp);
		if (error != 0)
			break;
		error = fo_ioctl(fp, FIOGETOWN, &tmp, td->td_ucred, td);
		if (error == 0)
			td->td_retval[0] = tmp;
		fdrop(fp, td);
		break;

	case F_SETOWN:
		error = fget_fcntl(td, fd, &cap_fcntl_rights, F_SETOWN, &fp);
		if (error != 0)
			break;
		tmp = arg;
		error = fo_ioctl(fp, FIOSETOWN, &tmp, td->td_ucred, td);
		fdrop(fp, td);
		break;

	case F_SETLK_REMOTE:
		error = priv_check(td, PRIV_NFS_LOCKD);
		if (error != 0)
			return (error);
		flg = F_REMOTE;
		goto do_setlk;

	case F_SETLKW:
		flg |= F_WAIT;
		/* FALLTHROUGH F_SETLK */

	case F_SETLK:
	do_setlk:
		flp = (struct flock *)arg;
		if ((flg & F_REMOTE) != 0 && flp->l_sysid == 0) {
			error = EINVAL;
			break;
		}

		error = fget_unlocked(fdp, fd, &cap_flock_rights, &fp);
		if (error != 0)
			break;
		if (fp->f_type != DTYPE_VNODE) {
			error = EBADF;
			fdrop(fp, td);
			break;
		}

		if (flp->l_whence == SEEK_CUR) {
			foffset = foffset_get(fp);
			if (foffset < 0 ||
			    (flp->l_start > 0 &&
			     foffset > OFF_MAX - flp->l_start)) {
				error = EOVERFLOW;
				fdrop(fp, td);
				break;
			}
			flp->l_start += foffset;
		}

		vp = fp->f_vnode;
		switch (flp->l_type) {
		case F_RDLCK:
			if ((fp->f_flag & FREAD) == 0) {
				error = EBADF;
				break;
			}
			if ((p->p_leader->p_flag & P_ADVLOCK) == 0) {
				PROC_LOCK(p->p_leader);
				p->p_leader->p_flag |= P_ADVLOCK;
				PROC_UNLOCK(p->p_leader);
			}
			error = VOP_ADVLOCK(vp, (caddr_t)p->p_leader, F_SETLK,
			    flp, flg);
			break;
		case F_WRLCK:
			if ((fp->f_flag & FWRITE) == 0) {
				error = EBADF;
				break;
			}
			if ((p->p_leader->p_flag & P_ADVLOCK) == 0) {
				PROC_LOCK(p->p_leader);
				p->p_leader->p_flag |= P_ADVLOCK;
				PROC_UNLOCK(p->p_leader);
			}
			error = VOP_ADVLOCK(vp, (caddr_t)p->p_leader, F_SETLK,
			    flp, flg);
			break;
		case F_UNLCK:
			error = VOP_ADVLOCK(vp, (caddr_t)p->p_leader, F_UNLCK,
			    flp, flg);
			break;
		case F_UNLCKSYS:
			if (flg != F_REMOTE) {
				error = EINVAL;
				break;
			}
			error = VOP_ADVLOCK(vp, (caddr_t)p->p_leader,
			    F_UNLCKSYS, flp, flg);
			break;
		default:
			error = EINVAL;
			break;
		}
		if (error != 0 || flp->l_type == F_UNLCK ||
		    flp->l_type == F_UNLCKSYS) {
			fdrop(fp, td);
			break;
		}

		/*
		 * Check for a race with close.
		 *
		 * The vnode is now advisory locked (or unlocked, but this case
		 * is not really important) as the caller requested.
		 * We had to drop the filedesc lock, so we need to recheck if
		 * the descriptor is still valid, because if it was closed
		 * in the meantime we need to remove advisory lock from the
		 * vnode - close on any descriptor leading to an advisory
		 * locked vnode, removes that lock.
		 * We will return 0 on purpose in that case, as the result of
		 * successful advisory lock might have been externally visible
		 * already. This is fine - effectively we pretend to the caller
		 * that the closing thread was a bit slower and that the
		 * advisory lock succeeded before the close.
		 */
		error = fget_unlocked(fdp, fd, &cap_no_rights, &fp2);
		if (error != 0) {
			fdrop(fp, td);
			break;
		}
		if (fp != fp2) {
			flp->l_whence = SEEK_SET;
			flp->l_start = 0;
			flp->l_len = 0;
			flp->l_type = F_UNLCK;
			(void) VOP_ADVLOCK(vp, (caddr_t)p->p_leader,
			    F_UNLCK, flp, F_POSIX);
		}
		fdrop(fp, td);
		fdrop(fp2, td);
		break;

	case F_GETLK:
		error = fget_unlocked(fdp, fd, &cap_flock_rights, &fp);
		if (error != 0)
			break;
		if (fp->f_type != DTYPE_VNODE) {
			error = EBADF;
			fdrop(fp, td);
			break;
		}
		flp = (struct flock *)arg;
		if (flp->l_type != F_RDLCK && flp->l_type != F_WRLCK &&
		    flp->l_type != F_UNLCK) {
			error = EINVAL;
			fdrop(fp, td);
			break;
		}
		if (flp->l_whence == SEEK_CUR) {
			foffset = foffset_get(fp);
			if ((flp->l_start > 0 &&
			    foffset > OFF_MAX - flp->l_start) ||
			    (flp->l_start < 0 &&
			    foffset < OFF_MIN - flp->l_start)) {
				error = EOVERFLOW;
				fdrop(fp, td);
				break;
			}
			flp->l_start += foffset;
		}
		vp = fp->f_vnode;
		error = VOP_ADVLOCK(vp, (caddr_t)p->p_leader, F_GETLK, flp,
		    F_POSIX);
		fdrop(fp, td);
		break;

	case F_ADD_SEALS:
		error = fget_unlocked(fdp, fd, &cap_no_rights, &fp);
		if (error != 0)
			break;
		error = fo_add_seals(fp, arg);
		fdrop(fp, td);
		break;

	case F_GET_SEALS:
		error = fget_unlocked(fdp, fd, &cap_no_rights, &fp);
		if (error != 0)
			break;
		if (fo_get_seals(fp, &seals) == 0)
			td->td_retval[0] = seals;
		else
			error = EINVAL;
		fdrop(fp, td);
		break;

	case F_RDAHEAD:
		arg = arg ? 128 * 1024: 0;
		/* FALLTHROUGH */
	case F_READAHEAD:
		error = fget_unlocked(fdp, fd, &cap_no_rights, &fp);
		if (error != 0)
			break;
		if (fp->f_type != DTYPE_VNODE) {
			fdrop(fp, td);
			error = EBADF;
			break;
		}
		vp = fp->f_vnode;
		if (vp->v_type != VREG) {
			fdrop(fp, td);
			error = ENOTTY;
			break;
		}

		/*
		 * Exclusive lock synchronizes against f_seqcount reads and
		 * writes in sequential_heuristic().
		 */
		error = vn_lock(vp, LK_EXCLUSIVE);
		if (error != 0) {
			fdrop(fp, td);
			break;
		}
		if (arg >= 0) {
			bsize = fp->f_vnode->v_mount->mnt_stat.f_iosize;
			arg = MIN(arg, INT_MAX - bsize + 1);
			fp->f_seqcount[UIO_READ] = MIN(IO_SEQMAX,
			    (arg + bsize - 1) / bsize);
			atomic_set_int(&fp->f_flag, FRDAHEAD);
		} else {
			atomic_clear_int(&fp->f_flag, FRDAHEAD);
		}
		VOP_UNLOCK(vp);
		fdrop(fp, td);
		break;

	case F_ISUNIONSTACK:
		/*
		 * Check if the vnode is part of a union stack (either the
		 * "union" flag from mount(2) or unionfs).
		 *
		 * Prior to introduction of this op libc's readdir would call
		 * fstatfs(2), in effect unnecessarily copying kilobytes of
		 * data just to check fs name and a mount flag.
		 *
		 * Fixing the code to handle everything in the kernel instead
		 * is a non-trivial endeavor and has low priority, thus this
		 * horrible kludge facilitates the current behavior in a much
		 * cheaper manner until someone(tm) sorts this out.
		 */
		error = fget_unlocked(fdp, fd, &cap_no_rights, &fp);
		if (error != 0)
			break;
		if (fp->f_type != DTYPE_VNODE) {
			fdrop(fp, td);
			error = EBADF;
			break;
		}
		vp = fp->f_vnode;
		/*
		 * Since we don't prevent dooming the vnode even non-null mp
		 * found can become immediately stale. This is tolerable since
		 * mount points are type-stable (providing safe memory access)
		 * and any vfs op on this vnode going forward will return an
		 * error (meaning return value in this case is meaningless).
		 */
		mp = atomic_load_ptr(&vp->v_mount);
		if (__predict_false(mp == NULL)) {
			fdrop(fp, td);
			error = EBADF;
			break;
		}
		td->td_retval[0] = 0;
		if (mp->mnt_kern_flag & MNTK_UNIONFS ||
		    mp->mnt_flag & MNT_UNION)
			td->td_retval[0] = 1;
		fdrop(fp, td);
		break;

	default:
		error = EINVAL;
		break;
	}
	return (error);
}

static int
getmaxfd(struct thread *td)
{

	return (min((int)lim_cur(td, RLIMIT_NOFILE), maxfilesperproc));
}

/*
 * Common code for dup, dup2, fcntl(F_DUPFD) and fcntl(F_DUP2FD).
 */
int
kern_dup(struct thread *td, u_int mode, int flags, int old, int new)
{
	struct filedesc *fdp;
	struct filedescent *oldfde, *newfde;
	struct proc *p;
	struct file *delfp;
	u_long *oioctls, *nioctls;
	int error, maxfd;

	p = td->td_proc;
	fdp = p->p_fd;
	oioctls = NULL;

	MPASS((flags & ~(FDDUP_FLAG_CLOEXEC)) == 0);
	MPASS(mode < FDDUP_LASTMODE);

	AUDIT_ARG_FD(old);
	/* XXXRW: if (flags & FDDUP_FIXED) AUDIT_ARG_FD2(new); */

	/*
	 * Verify we have a valid descriptor to dup from and possibly to
	 * dup to. Unlike dup() and dup2(), fcntl()'s F_DUPFD should
	 * return EINVAL when the new descriptor is out of bounds.
	 */
	if (old < 0)
		return (EBADF);
	if (new < 0)
		return (mode == FDDUP_FCNTL ? EINVAL : EBADF);
	maxfd = getmaxfd(td);
	if (new >= maxfd)
		return (mode == FDDUP_FCNTL ? EINVAL : EBADF);

	error = EBADF;
	FILEDESC_XLOCK(fdp);
	if (fget_locked(fdp, old) == NULL)
		goto unlock;
	if ((mode == FDDUP_FIXED || mode == FDDUP_MUSTREPLACE) && old == new) {
		td->td_retval[0] = new;
		if (flags & FDDUP_FLAG_CLOEXEC)
			fdp->fd_ofiles[new].fde_flags |= UF_EXCLOSE;
		error = 0;
		goto unlock;
	}

	oldfde = &fdp->fd_ofiles[old];
	if (!fhold(oldfde->fde_file))
		goto unlock;

	/*
	 * If the caller specified a file descriptor, make sure the file
	 * table is large enough to hold it, and grab it.  Otherwise, just
	 * allocate a new descriptor the usual way.
	 */
	switch (mode) {
	case FDDUP_NORMAL:
	case FDDUP_FCNTL:
		if ((error = fdalloc(td, new, &new)) != 0) {
			fdrop(oldfde->fde_file, td);
			goto unlock;
		}
		break;
	case FDDUP_MUSTREPLACE:
		/* Target file descriptor must exist. */
		if (fget_locked(fdp, new) == NULL) {
			fdrop(oldfde->fde_file, td);
			goto unlock;
		}
		break;
	case FDDUP_FIXED:
		if (new >= fdp->fd_nfiles) {
			/*
			 * The resource limits are here instead of e.g.
			 * fdalloc(), because the file descriptor table may be
			 * shared between processes, so we can't really use
			 * racct_add()/racct_sub().  Instead of counting the
			 * number of actually allocated descriptors, just put
			 * the limit on the size of the file descriptor table.
			 */
#ifdef RACCT
			if (RACCT_ENABLED()) {
				error = racct_set_unlocked(p, RACCT_NOFILE, new + 1);
				if (error != 0) {
					error = EMFILE;
					fdrop(oldfde->fde_file, td);
					goto unlock;
				}
			}
#endif
			fdgrowtable_exp(fdp, new + 1);
		}
		if (!fdisused(fdp, new))
			fdused(fdp, new);
		break;
	default:
		KASSERT(0, ("%s unsupported mode %d", __func__, mode));
	}

	KASSERT(old != new, ("new fd is same as old"));

	newfde = &fdp->fd_ofiles[new];
	delfp = newfde->fde_file;

	nioctls = filecaps_copy_prep(&oldfde->fde_caps);

	/*
	 * Duplicate the source descriptor.
	 */
#ifdef CAPABILITIES
	seqc_write_begin(&newfde->fde_seqc);
#endif
	oioctls = filecaps_free_prep(&newfde->fde_caps);
	memcpy(newfde, oldfde, fde_change_size);
	filecaps_copy_finish(&oldfde->fde_caps, &newfde->fde_caps,
	    nioctls);
	if ((flags & FDDUP_FLAG_CLOEXEC) != 0)
		newfde->fde_flags = oldfde->fde_flags | UF_EXCLOSE;
	else
		newfde->fde_flags = oldfde->fde_flags & ~UF_EXCLOSE;
#ifdef CAPABILITIES
	seqc_write_end(&newfde->fde_seqc);
#endif
	td->td_retval[0] = new;

	error = 0;

	if (delfp != NULL) {
		(void) closefp(fdp, new, delfp, td, 1);
		FILEDESC_UNLOCK_ASSERT(fdp);
	} else {
unlock:
		FILEDESC_XUNLOCK(fdp);
	}

	filecaps_free_finish(oioctls);
	return (error);
}

/*
 * If sigio is on the list associated with a process or process group,
 * disable signalling from the device, remove sigio from the list and
 * free sigio.
 */
void
funsetown(struct sigio **sigiop)
{
	struct sigio *sigio;

	if (*sigiop == NULL)
		return;
	SIGIO_LOCK();
	sigio = *sigiop;
	if (sigio == NULL) {
		SIGIO_UNLOCK();
		return;
	}
	*(sigio->sio_myref) = NULL;
	if ((sigio)->sio_pgid < 0) {
		struct pgrp *pg = (sigio)->sio_pgrp;
		PGRP_LOCK(pg);
		SLIST_REMOVE(&sigio->sio_pgrp->pg_sigiolst, sigio,
			    sigio, sio_pgsigio);
		PGRP_UNLOCK(pg);
	} else {
		struct proc *p = (sigio)->sio_proc;
		PROC_LOCK(p);
		SLIST_REMOVE(&sigio->sio_proc->p_sigiolst, sigio,
			    sigio, sio_pgsigio);
		PROC_UNLOCK(p);
	}
	SIGIO_UNLOCK();
	crfree(sigio->sio_ucred);
	free(sigio, M_SIGIO);
}

/*
 * Free a list of sigio structures.
 * We only need to lock the SIGIO_LOCK because we have made ourselves
 * inaccessible to callers of fsetown and therefore do not need to lock
 * the proc or pgrp struct for the list manipulation.
 */
void
funsetownlst(struct sigiolst *sigiolst)
{
	struct proc *p;
	struct pgrp *pg;
	struct sigio *sigio;

	sigio = SLIST_FIRST(sigiolst);
	if (sigio == NULL)
		return;
	p = NULL;
	pg = NULL;

	/*
	 * Every entry of the list should belong
	 * to a single proc or pgrp.
	 */
	if (sigio->sio_pgid < 0) {
		pg = sigio->sio_pgrp;
		PGRP_LOCK_ASSERT(pg, MA_NOTOWNED);
	} else /* if (sigio->sio_pgid > 0) */ {
		p = sigio->sio_proc;
		PROC_LOCK_ASSERT(p, MA_NOTOWNED);
	}

	SIGIO_LOCK();
	while ((sigio = SLIST_FIRST(sigiolst)) != NULL) {
		*(sigio->sio_myref) = NULL;
		if (pg != NULL) {
			KASSERT(sigio->sio_pgid < 0,
			    ("Proc sigio in pgrp sigio list"));
			KASSERT(sigio->sio_pgrp == pg,
			    ("Bogus pgrp in sigio list"));
			PGRP_LOCK(pg);
			SLIST_REMOVE(&pg->pg_sigiolst, sigio, sigio,
			    sio_pgsigio);
			PGRP_UNLOCK(pg);
		} else /* if (p != NULL) */ {
			KASSERT(sigio->sio_pgid > 0,
			    ("Pgrp sigio in proc sigio list"));
			KASSERT(sigio->sio_proc == p,
			    ("Bogus proc in sigio list"));
			PROC_LOCK(p);
			SLIST_REMOVE(&p->p_sigiolst, sigio, sigio,
			    sio_pgsigio);
			PROC_UNLOCK(p);
		}
		SIGIO_UNLOCK();
		crfree(sigio->sio_ucred);
		free(sigio, M_SIGIO);
		SIGIO_LOCK();
	}
	SIGIO_UNLOCK();
}

/*
 * This is common code for FIOSETOWN ioctl called by fcntl(fd, F_SETOWN, arg).
 *
 * After permission checking, add a sigio structure to the sigio list for
 * the process or process group.
 */
int
fsetown(pid_t pgid, struct sigio **sigiop)
{
	struct proc *proc;
	struct pgrp *pgrp;
	struct sigio *sigio;
	int ret;

	if (pgid == 0) {
		funsetown(sigiop);
		return (0);
	}

	ret = 0;

	/* Allocate and fill in the new sigio out of locks. */
	sigio = malloc(sizeof(struct sigio), M_SIGIO, M_WAITOK);
	sigio->sio_pgid = pgid;
	sigio->sio_ucred = crhold(curthread->td_ucred);
	sigio->sio_myref = sigiop;

	sx_slock(&proctree_lock);
	if (pgid > 0) {
		proc = pfind(pgid);
		if (proc == NULL) {
			ret = ESRCH;
			goto fail;
		}

		/*
		 * Policy - Don't allow a process to FSETOWN a process
		 * in another session.
		 *
		 * Remove this test to allow maximum flexibility or
		 * restrict FSETOWN to the current process or process
		 * group for maximum safety.
		 */
		PROC_UNLOCK(proc);
		if (proc->p_session != curthread->td_proc->p_session) {
			ret = EPERM;
			goto fail;
		}

		pgrp = NULL;
	} else /* if (pgid < 0) */ {
		pgrp = pgfind(-pgid);
		if (pgrp == NULL) {
			ret = ESRCH;
			goto fail;
		}
		PGRP_UNLOCK(pgrp);

		/*
		 * Policy - Don't allow a process to FSETOWN a process
		 * in another session.
		 *
		 * Remove this test to allow maximum flexibility or
		 * restrict FSETOWN to the current process or process
		 * group for maximum safety.
		 */
		if (pgrp->pg_session != curthread->td_proc->p_session) {
			ret = EPERM;
			goto fail;
		}

		proc = NULL;
	}
	funsetown(sigiop);
	if (pgid > 0) {
		PROC_LOCK(proc);
		/*
		 * Since funsetownlst() is called without the proctree
		 * locked, we need to check for P_WEXIT.
		 * XXX: is ESRCH correct?
		 */
		if ((proc->p_flag & P_WEXIT) != 0) {
			PROC_UNLOCK(proc);
			ret = ESRCH;
			goto fail;
		}
		SLIST_INSERT_HEAD(&proc->p_sigiolst, sigio, sio_pgsigio);
		sigio->sio_proc = proc;
		PROC_UNLOCK(proc);
	} else {
		PGRP_LOCK(pgrp);
		SLIST_INSERT_HEAD(&pgrp->pg_sigiolst, sigio, sio_pgsigio);
		sigio->sio_pgrp = pgrp;
		PGRP_UNLOCK(pgrp);
	}
	sx_sunlock(&proctree_lock);
	SIGIO_LOCK();
	*sigiop = sigio;
	SIGIO_UNLOCK();
	return (0);

fail:
	sx_sunlock(&proctree_lock);
	crfree(sigio->sio_ucred);
	free(sigio, M_SIGIO);
	return (ret);
}

/*
 * This is common code for FIOGETOWN ioctl called by fcntl(fd, F_GETOWN, arg).
 */
pid_t
fgetown(struct sigio **sigiop)
{
	pid_t pgid;

	SIGIO_LOCK();
	pgid = (*sigiop != NULL) ? (*sigiop)->sio_pgid : 0;
	SIGIO_UNLOCK();
	return (pgid);
}

/*
 * Function drops the filedesc lock on return.
 */
static int
closefp(struct filedesc *fdp, int fd, struct file *fp, struct thread *td,
    int holdleaders)
{
	int error;

	FILEDESC_XLOCK_ASSERT(fdp);

	if (holdleaders) {
		if (td->td_proc->p_fdtol != NULL) {
			/*
			 * Ask fdfree() to sleep to ensure that all relevant
			 * process leaders can be traversed in closef().
			 */
			fdp->fd_holdleaderscount++;
		} else {
			holdleaders = 0;
		}
	}

	/*
	 * We now hold the fp reference that used to be owned by the
	 * descriptor array.  We have to unlock the FILEDESC *AFTER*
	 * knote_fdclose to prevent a race of the fd getting opened, a knote
	 * added, and deleteing a knote for the new fd.
	 */
	if (__predict_false(!TAILQ_EMPTY(&fdp->fd_kqlist)))
		knote_fdclose(td, fd);

	/*
	 * We need to notify mqueue if the object is of type mqueue.
	 */
	if (__predict_false(fp->f_type == DTYPE_MQUEUE))
		mq_fdclose(td, fd, fp);
	FILEDESC_XUNLOCK(fdp);

	error = closef(fp, td);
	if (holdleaders) {
		FILEDESC_XLOCK(fdp);
		fdp->fd_holdleaderscount--;
		if (fdp->fd_holdleaderscount == 0 &&
		    fdp->fd_holdleaderswakeup != 0) {
			fdp->fd_holdleaderswakeup = 0;
			wakeup(&fdp->fd_holdleaderscount);
		}
		FILEDESC_XUNLOCK(fdp);
	}
	return (error);
}

/*
 * Close a file descriptor.
 */
#ifndef _SYS_SYSPROTO_H_
struct close_args {
	int     fd;
};
#endif
/* ARGSUSED */
int
sys_close(struct thread *td, struct close_args *uap)
{

	return (kern_close(td, uap->fd));
}

int
kern_close(struct thread *td, int fd)
{
	struct filedesc *fdp;
	struct file *fp;

	fdp = td->td_proc->p_fd;

	AUDIT_SYSCLOSE(td, fd);

	FILEDESC_XLOCK(fdp);
	if ((fp = fget_locked(fdp, fd)) == NULL) {
		FILEDESC_XUNLOCK(fdp);
		return (EBADF);
	}
	fdfree(fdp, fd);

	/* closefp() drops the FILEDESC lock for us. */
	return (closefp(fdp, fd, fp, td, 1));
}

int
kern_close_range(struct thread *td, u_int lowfd, u_int highfd)
{
	struct filedesc *fdp;
	int fd, ret, lastfile;

	ret = 0;
	fdp = td->td_proc->p_fd;
	FILEDESC_SLOCK(fdp);

	/*
	 * Check this prior to clamping; closefrom(3) with only fd 0, 1, and 2
	 * open should not be a usage error.  From a close_range() perspective,
	 * close_range(3, ~0U, 0) in the same scenario should also likely not
	 * be a usage error as all fd above 3 are in-fact already closed.
	 */
	if (highfd < lowfd) {
		ret = EINVAL;
		goto out;
	}

	/*
	 * If lastfile == -1, we're dealing with either a fresh file
	 * table or one in which every fd has been closed.  Just return
	 * successful; there's nothing left to do.
	 */
	lastfile = fdlastfile(fdp);
	if (lastfile == -1)
		goto out;
	/* Clamped to [lowfd, lastfile] */
	highfd = MIN(highfd, lastfile);
	for (fd = lowfd; fd <= highfd; fd++) {
		if (fdp->fd_ofiles[fd].fde_file != NULL) {
			FILEDESC_SUNLOCK(fdp);
			(void)kern_close(td, fd);
			FILEDESC_SLOCK(fdp);
		}
	}
out:
	FILEDESC_SUNLOCK(fdp);
	return (ret);
}

#ifndef _SYS_SYSPROTO_H_
struct close_range_args {
	u_int	lowfd;
	u_int	highfd;
	int	flags;
};
#endif
int
sys_close_range(struct thread *td, struct close_range_args *uap)
{

	/* No flags currently defined */
	if (uap->flags != 0)
		return (EINVAL);
	return (kern_close_range(td, uap->lowfd, uap->highfd));
}

#ifdef COMPAT_FREEBSD12
/*
 * Close open file descriptors.
 */
#ifndef _SYS_SYSPROTO_H_
struct freebsd12_closefrom_args {
	int	lowfd;
};
#endif
/* ARGSUSED */
int
freebsd12_closefrom(struct thread *td, struct freebsd12_closefrom_args *uap)
{
	u_int lowfd;

	AUDIT_ARG_FD(uap->lowfd);

	/*
	 * Treat negative starting file descriptor values identical to
	 * closefrom(0) which closes all files.
	 */
	lowfd = MAX(0, uap->lowfd);
	return (kern_close_range(td, lowfd, ~0U));
}
#endif	/* COMPAT_FREEBSD12 */

#if defined(COMPAT_43)
/*
 * Return status information about a file descriptor.
 */
#ifndef _SYS_SYSPROTO_H_
struct ofstat_args {
	int	fd;
	struct	ostat *sb;
};
#endif
/* ARGSUSED */
int
ofstat(struct thread *td, struct ofstat_args *uap)
{
	struct ostat oub;
	struct stat ub;
	int error;

	error = kern_fstat(td, uap->fd, &ub);
	if (error == 0) {
		cvtstat(&ub, &oub);
		error = copyout(&oub, uap->sb, sizeof(oub));
	}
	return (error);
}
#endif /* COMPAT_43 */

#if defined(COMPAT_FREEBSD11)
int
freebsd11_fstat(struct thread *td, struct freebsd11_fstat_args *uap)
{
	struct stat sb;
	struct freebsd11_stat osb;
	int error;

	error = kern_fstat(td, uap->fd, &sb);
	if (error != 0)
		return (error);
	error = freebsd11_cvtstat(&sb, &osb);
	if (error == 0)
		error = copyout(&osb, uap->sb, sizeof(osb));
	return (error);
}
#endif	/* COMPAT_FREEBSD11 */

/*
 * Return status information about a file descriptor.
 */
#ifndef _SYS_SYSPROTO_H_
struct fstat_args {
	int	fd;
	struct	stat *sb;
};
#endif
/* ARGSUSED */
int
sys_fstat(struct thread *td, struct fstat_args *uap)
{
	struct stat ub;
	int error;

	error = kern_fstat(td, uap->fd, &ub);
	if (error == 0)
		error = copyout(&ub, uap->sb, sizeof(ub));
	return (error);
}

int
kern_fstat(struct thread *td, int fd, struct stat *sbp)
{
	struct file *fp;
	int error;

	AUDIT_ARG_FD(fd);

	error = fget(td, fd, &cap_fstat_rights, &fp);
	if (__predict_false(error != 0))
		return (error);

	AUDIT_ARG_FILE(td->td_proc, fp);

	error = fo_stat(fp, sbp, td->td_ucred, td);
	fdrop(fp, td);
#ifdef __STAT_TIME_T_EXT
	sbp->st_atim_ext = 0;
	sbp->st_mtim_ext = 0;
	sbp->st_ctim_ext = 0;
	sbp->st_btim_ext = 0;
#endif
#ifdef KTRACE
	if (KTRPOINT(td, KTR_STRUCT))
		ktrstat_error(sbp, error);
#endif
	return (error);
}

#if defined(COMPAT_FREEBSD11)
/*
 * Return status information about a file descriptor.
 */
#ifndef _SYS_SYSPROTO_H_
struct freebsd11_nfstat_args {
	int	fd;
	struct	nstat *sb;
};
#endif
/* ARGSUSED */
int
freebsd11_nfstat(struct thread *td, struct freebsd11_nfstat_args *uap)
{
	struct nstat nub;
	struct stat ub;
	int error;

	error = kern_fstat(td, uap->fd, &ub);
	if (error == 0) {
		freebsd11_cvtnstat(&ub, &nub);
		error = copyout(&nub, uap->sb, sizeof(nub));
	}
	return (error);
}
#endif /* COMPAT_FREEBSD11 */

/*
 * Return pathconf information about a file descriptor.
 */
#ifndef _SYS_SYSPROTO_H_
struct fpathconf_args {
	int	fd;
	int	name;
};
#endif
/* ARGSUSED */
int
sys_fpathconf(struct thread *td, struct fpathconf_args *uap)
{
	long value;
	int error;

	error = kern_fpathconf(td, uap->fd, uap->name, &value);
	if (error == 0)
		td->td_retval[0] = value;
	return (error);
}

int
kern_fpathconf(struct thread *td, int fd, int name, long *valuep)
{
	struct file *fp;
	struct vnode *vp;
	int error;

	error = fget(td, fd, &cap_fpathconf_rights, &fp);
	if (error != 0)
		return (error);

	if (name == _PC_ASYNC_IO) {
		*valuep = _POSIX_ASYNCHRONOUS_IO;
		goto out;
	}
	vp = fp->f_vnode;
	if (vp != NULL) {
		vn_lock(vp, LK_SHARED | LK_RETRY);
		error = VOP_PATHCONF(vp, name, valuep);
		VOP_UNLOCK(vp);
	} else if (fp->f_type == DTYPE_PIPE || fp->f_type == DTYPE_SOCKET) {
		if (name != _PC_PIPE_BUF) {
			error = EINVAL;
		} else {
			*valuep = PIPE_BUF;
			error = 0;
		}
	} else {
		error = EOPNOTSUPP;
	}
out:
	fdrop(fp, td);
	return (error);
}

/*
 * Copy filecaps structure allocating memory for ioctls array if needed.
 *
 * The last parameter indicates whether the fdtable is locked. If it is not and
 * ioctls are encountered, copying fails and the caller must lock the table.
 *
 * Note that if the table was not locked, the caller has to check the relevant
 * sequence counter to determine whether the operation was successful.
 */
bool
filecaps_copy(const struct filecaps *src, struct filecaps *dst, bool locked)
{
	size_t size;

	if (src->fc_ioctls != NULL && !locked)
		return (false);
	memcpy(dst, src, sizeof(*src));
	if (src->fc_ioctls == NULL)
		return (true);

	KASSERT(src->fc_nioctls > 0,
	    ("fc_ioctls != NULL, but fc_nioctls=%hd", src->fc_nioctls));

	size = sizeof(src->fc_ioctls[0]) * src->fc_nioctls;
	dst->fc_ioctls = malloc(size, M_FILECAPS, M_WAITOK);
	memcpy(dst->fc_ioctls, src->fc_ioctls, size);
	return (true);
}

static u_long *
filecaps_copy_prep(const struct filecaps *src)
{
	u_long *ioctls;
	size_t size;

	if (__predict_true(src->fc_ioctls == NULL))
		return (NULL);

	KASSERT(src->fc_nioctls > 0,
	    ("fc_ioctls != NULL, but fc_nioctls=%hd", src->fc_nioctls));

	size = sizeof(src->fc_ioctls[0]) * src->fc_nioctls;
	ioctls = malloc(size, M_FILECAPS, M_WAITOK);
	return (ioctls);
}

static void
filecaps_copy_finish(const struct filecaps *src, struct filecaps *dst,
    u_long *ioctls)
{
	size_t size;

	*dst = *src;
	if (__predict_true(src->fc_ioctls == NULL)) {
		MPASS(ioctls == NULL);
		return;
	}

	size = sizeof(src->fc_ioctls[0]) * src->fc_nioctls;
	dst->fc_ioctls = ioctls;
	bcopy(src->fc_ioctls, dst->fc_ioctls, size);
}

/*
 * Move filecaps structure to the new place and clear the old place.
 */
void
filecaps_move(struct filecaps *src, struct filecaps *dst)
{

	*dst = *src;
	bzero(src, sizeof(*src));
}

/*
 * Fill the given filecaps structure with full rights.
 */
static void
filecaps_fill(struct filecaps *fcaps)
{

	CAP_ALL(&fcaps->fc_rights);
	fcaps->fc_ioctls = NULL;
	fcaps->fc_nioctls = -1;
	fcaps->fc_fcntls = CAP_FCNTL_ALL;
}

/*
 * Free memory allocated within filecaps structure.
 */
void
filecaps_free(struct filecaps *fcaps)
{

	free(fcaps->fc_ioctls, M_FILECAPS);
	bzero(fcaps, sizeof(*fcaps));
}

static u_long *
filecaps_free_prep(struct filecaps *fcaps)
{
	u_long *ioctls;

	ioctls = fcaps->fc_ioctls;
	bzero(fcaps, sizeof(*fcaps));
	return (ioctls);
}

static void
filecaps_free_finish(u_long *ioctls)
{

	free(ioctls, M_FILECAPS);
}

/*
 * Validate the given filecaps structure.
 */
static void
filecaps_validate(const struct filecaps *fcaps, const char *func)
{

	KASSERT(cap_rights_is_valid(&fcaps->fc_rights),
	    ("%s: invalid rights", func));
	KASSERT((fcaps->fc_fcntls & ~CAP_FCNTL_ALL) == 0,
	    ("%s: invalid fcntls", func));
	KASSERT(fcaps->fc_fcntls == 0 ||
	    cap_rights_is_set(&fcaps->fc_rights, CAP_FCNTL),
	    ("%s: fcntls without CAP_FCNTL", func));
	KASSERT(fcaps->fc_ioctls != NULL ? fcaps->fc_nioctls > 0 :
	    (fcaps->fc_nioctls == -1 || fcaps->fc_nioctls == 0),
	    ("%s: invalid ioctls", func));
	KASSERT(fcaps->fc_nioctls == 0 ||
	    cap_rights_is_set(&fcaps->fc_rights, CAP_IOCTL),
	    ("%s: ioctls without CAP_IOCTL", func));
}

static void
fdgrowtable_exp(struct filedesc *fdp, int nfd)
{
	int nfd1;

	FILEDESC_XLOCK_ASSERT(fdp);

	nfd1 = fdp->fd_nfiles * 2;
	if (nfd1 < nfd)
		nfd1 = nfd;
	fdgrowtable(fdp, nfd1);
}

/*
 * Grow the file table to accommodate (at least) nfd descriptors.
 */
static void
fdgrowtable(struct filedesc *fdp, int nfd)
{
	struct filedesc0 *fdp0;
	struct freetable *ft;
	struct fdescenttbl *ntable;
	struct fdescenttbl *otable;
	int nnfiles, onfiles;
	NDSLOTTYPE *nmap, *omap;

	KASSERT(fdp->fd_nfiles > 0, ("zero-length file table"));

	/* save old values */
	onfiles = fdp->fd_nfiles;
	otable = fdp->fd_files;
	omap = fdp->fd_map;

	/* compute the size of the new table */
	nnfiles = NDSLOTS(nfd) * NDENTRIES; /* round up */
	if (nnfiles <= onfiles)
		/* the table is already large enough */
		return;

	/*
	 * Allocate a new table.  We need enough space for the number of
	 * entries, file entries themselves and the struct freetable we will use
	 * when we decommission the table and place it on the freelist.
	 * We place the struct freetable in the middle so we don't have
	 * to worry about padding.
	 */
	ntable = malloc(offsetof(struct fdescenttbl, fdt_ofiles) +
	    nnfiles * sizeof(ntable->fdt_ofiles[0]) +
	    sizeof(struct freetable),
	    M_FILEDESC, M_ZERO | M_WAITOK);
	/* copy the old data */
	ntable->fdt_nfiles = nnfiles;
	memcpy(ntable->fdt_ofiles, otable->fdt_ofiles,
	    onfiles * sizeof(ntable->fdt_ofiles[0]));

	/*
	 * Allocate a new map only if the old is not large enough.  It will
	 * grow at a slower rate than the table as it can map more
	 * entries than the table can hold.
	 */
	if (NDSLOTS(nnfiles) > NDSLOTS(onfiles)) {
		nmap = malloc(NDSLOTS(nnfiles) * NDSLOTSIZE, M_FILEDESC,
		    M_ZERO | M_WAITOK);
		/* copy over the old data and update the pointer */
		memcpy(nmap, omap, NDSLOTS(onfiles) * sizeof(*omap));
		fdp->fd_map = nmap;
	}

	/*
	 * Make sure that ntable is correctly initialized before we replace
	 * fd_files poiner. Otherwise fget_unlocked() may see inconsistent
	 * data.
	 */
	atomic_store_rel_ptr((volatile void *)&fdp->fd_files, (uintptr_t)ntable);

	/*
	 * Do not free the old file table, as some threads may still
	 * reference entries within it.  Instead, place it on a freelist
	 * which will be processed when the struct filedesc is released.
	 *
	 * Note that if onfiles == NDFILE, we're dealing with the original
	 * static allocation contained within (struct filedesc0 *)fdp,
	 * which must not be freed.
	 */
	if (onfiles > NDFILE) {
		ft = (struct freetable *)&otable->fdt_ofiles[onfiles];
		fdp0 = (struct filedesc0 *)fdp;
		ft->ft_table = otable;
		SLIST_INSERT_HEAD(&fdp0->fd_free, ft, ft_next);
	}
	/*
	 * The map does not have the same possibility of threads still
	 * holding references to it.  So always free it as long as it
	 * does not reference the original static allocation.
	 */
	if (NDSLOTS(onfiles) > NDSLOTS(NDFILE))
		free(omap, M_FILEDESC);
}

/*
 * Allocate a file descriptor for the process.
 */
int
fdalloc(struct thread *td, int minfd, int *result)
{
	struct proc *p = td->td_proc;
	struct filedesc *fdp = p->p_fd;
	int fd, maxfd, allocfd;
#ifdef RACCT
	int error;
#endif

	FILEDESC_XLOCK_ASSERT(fdp);

	if (fdp->fd_freefile > minfd)
		minfd = fdp->fd_freefile;

	maxfd = getmaxfd(td);

	/*
	 * Search the bitmap for a free descriptor starting at minfd.
	 * If none is found, grow the file table.
	 */
	fd = fd_first_free(fdp, minfd, fdp->fd_nfiles);
	if (__predict_false(fd >= maxfd))
		return (EMFILE);
	if (__predict_false(fd >= fdp->fd_nfiles)) {
		allocfd = min(fd * 2, maxfd);
#ifdef RACCT
		if (RACCT_ENABLED()) {
			error = racct_set_unlocked(p, RACCT_NOFILE, allocfd);
			if (error != 0)
				return (EMFILE);
		}
#endif
		/*
		 * fd is already equal to first free descriptor >= minfd, so
		 * we only need to grow the table and we are done.
		 */
		fdgrowtable_exp(fdp, allocfd);
	}

	/*
	 * Perform some sanity checks, then mark the file descriptor as
	 * used and return it to the caller.
	 */
	KASSERT(fd >= 0 && fd < min(maxfd, fdp->fd_nfiles),
	    ("invalid descriptor %d", fd));
	KASSERT(!fdisused(fdp, fd),
	    ("fd_first_free() returned non-free descriptor"));
	KASSERT(fdp->fd_ofiles[fd].fde_file == NULL,
	    ("file descriptor isn't free"));
	fdused(fdp, fd);
	*result = fd;
	return (0);
}

/*
 * Allocate n file descriptors for the process.
 */
int
fdallocn(struct thread *td, int minfd, int *fds, int n)
{
	struct proc *p = td->td_proc;
	struct filedesc *fdp = p->p_fd;
	int i;

	FILEDESC_XLOCK_ASSERT(fdp);

	for (i = 0; i < n; i++)
		if (fdalloc(td, 0, &fds[i]) != 0)
			break;

	if (i < n) {
		for (i--; i >= 0; i--)
			fdunused(fdp, fds[i]);
		return (EMFILE);
	}

	return (0);
}

/*
 * Create a new open file structure and allocate a file descriptor for the
 * process that refers to it.  We add one reference to the file for the
 * descriptor table and one reference for resultfp. This is to prevent us
 * being preempted and the entry in the descriptor table closed after we
 * release the FILEDESC lock.
 */
int
falloc_caps(struct thread *td, struct file **resultfp, int *resultfd, int flags,
    struct filecaps *fcaps)
{
	struct file *fp;
	int error, fd;

	error = falloc_noinstall(td, &fp);
	if (error)
		return (error);		/* no reference held on error */

	error = finstall(td, fp, &fd, flags, fcaps);
	if (error) {
		fdrop(fp, td);		/* one reference (fp only) */
		return (error);
	}

	if (resultfp != NULL)
		*resultfp = fp;		/* copy out result */
	else
		fdrop(fp, td);		/* release local reference */

	if (resultfd != NULL)
		*resultfd = fd;

	return (0);
}

/*
 * Create a new open file structure without allocating a file descriptor.
 */
int
falloc_noinstall(struct thread *td, struct file **resultfp)
{
	struct file *fp;
	int maxuserfiles = maxfiles - (maxfiles / 20);
	int openfiles_new;
	static struct timeval lastfail;
	static int curfail;

	KASSERT(resultfp != NULL, ("%s: resultfp == NULL", __func__));

	openfiles_new = atomic_fetchadd_int(&openfiles, 1) + 1;
	if ((openfiles_new >= maxuserfiles &&
	    priv_check(td, PRIV_MAXFILES) != 0) ||
	    openfiles_new >= maxfiles) {
		atomic_subtract_int(&openfiles, 1);
		if (ppsratecheck(&lastfail, &curfail, 1)) {
			printf("kern.maxfiles limit exceeded by uid %i, (%s) "
			    "please see tuning(7).\n", td->td_ucred->cr_ruid, td->td_proc->p_comm);
		}
		return (ENFILE);
	}
	fp = uma_zalloc(file_zone, M_WAITOK);
	bzero(fp, sizeof(*fp));
	refcount_init(&fp->f_count, 1);
	fp->f_cred = crhold(td->td_ucred);
	fp->f_ops = &badfileops;
	*resultfp = fp;
	return (0);
}

/*
 * Install a file in a file descriptor table.
 */
void
_finstall(struct filedesc *fdp, struct file *fp, int fd, int flags,
    struct filecaps *fcaps)
{
	struct filedescent *fde;

	MPASS(fp != NULL);
	if (fcaps != NULL)
		filecaps_validate(fcaps, __func__);
	FILEDESC_XLOCK_ASSERT(fdp);

	fde = &fdp->fd_ofiles[fd];
#ifdef CAPABILITIES
	seqc_write_begin(&fde->fde_seqc);
#endif
	fde->fde_file = fp;
	fde->fde_flags = (flags & O_CLOEXEC) != 0 ? UF_EXCLOSE : 0;
	if (fcaps != NULL)
		filecaps_move(fcaps, &fde->fde_caps);
	else
		filecaps_fill(&fde->fde_caps);
#ifdef CAPABILITIES
	seqc_write_end(&fde->fde_seqc);
#endif
}

int
finstall(struct thread *td, struct file *fp, int *fd, int flags,
    struct filecaps *fcaps)
{
	struct filedesc *fdp = td->td_proc->p_fd;
	int error;

	MPASS(fd != NULL);

	if (!fhold(fp))
		return (EBADF);
	FILEDESC_XLOCK(fdp);
	error = fdalloc(td, 0, fd);
	if (__predict_false(error != 0)) {
		FILEDESC_XUNLOCK(fdp);
		fdrop(fp, td);
		return (error);
	}
	_finstall(fdp, fp, *fd, flags, fcaps);
	FILEDESC_XUNLOCK(fdp);
	return (0);
}

/*
 * Build a new filedesc structure from another.
 * Copy the current, root, and jail root vnode references.
 *
 * If fdp is not NULL, return with it shared locked.
 */
struct filedesc *
fdinit(struct filedesc *fdp, bool prepfiles, int *lastfile)
{
	struct filedesc0 *newfdp0;
	struct filedesc *newfdp;
	struct pwd *newpwd;

	if (prepfiles)
		MPASS(lastfile != NULL);
	else
		MPASS(lastfile == NULL);

	newfdp0 = uma_zalloc(filedesc0_zone, M_WAITOK | M_ZERO);
	newfdp = &newfdp0->fd_fd;

	/* Create the file descriptor table. */
	FILEDESC_LOCK_INIT(newfdp);
	refcount_init(&newfdp->fd_refcnt, 1);
	refcount_init(&newfdp->fd_holdcnt, 1);
	newfdp->fd_cmask = CMASK;
	newfdp->fd_map = newfdp0->fd_dmap;
	newfdp->fd_files = (struct fdescenttbl *)&newfdp0->fd_dfiles;
	newfdp->fd_files->fdt_nfiles = NDFILE;

	if (fdp == NULL) {
		newpwd = pwd_alloc();
		smr_serialized_store(&newfdp->fd_pwd, newpwd, true);
		return (newfdp);
	}

	FILEDESC_SLOCK(fdp);
	newpwd = pwd_hold_filedesc(fdp);
	smr_serialized_store(&newfdp->fd_pwd, newpwd, true);
	if (!prepfiles) {
		FILEDESC_SUNLOCK(fdp);
		return (newfdp);
	}

	for (;;) {
		*lastfile = fdlastfile(fdp);
		if (*lastfile < newfdp->fd_nfiles)
			break;
		FILEDESC_SUNLOCK(fdp);
		fdgrowtable(newfdp, *lastfile + 1);
		FILEDESC_SLOCK(fdp);
	}

	return (newfdp);
}

static struct filedesc *
fdhold(struct proc *p)
{
	struct filedesc *fdp;

	PROC_LOCK_ASSERT(p, MA_OWNED);
	fdp = p->p_fd;
	if (fdp != NULL)
		refcount_acquire(&fdp->fd_holdcnt);
	return (fdp);
}

static void
fddrop(struct filedesc *fdp)
{

	if (fdp->fd_holdcnt > 1) {
		if (refcount_release(&fdp->fd_holdcnt) == 0)
			return;
	}

	FILEDESC_LOCK_DESTROY(fdp);
	uma_zfree(filedesc0_zone, fdp);
}

/*
 * Share a filedesc structure.
 */
struct filedesc *
fdshare(struct filedesc *fdp)
{

	refcount_acquire(&fdp->fd_refcnt);
	return (fdp);
}

/*
 * Unshare a filedesc structure, if necessary by making a copy
 */
void
fdunshare(struct thread *td)
{
	struct filedesc *tmp;
	struct proc *p = td->td_proc;

	if (p->p_fd->fd_refcnt == 1)
		return;

	tmp = fdcopy(p->p_fd);
	fdescfree(td);
	p->p_fd = tmp;
}

void
fdinstall_remapped(struct thread *td, struct filedesc *fdp)
{

	fdescfree(td);
	td->td_proc->p_fd = fdp;
}

/*
 * Copy a filedesc structure.  A NULL pointer in returns a NULL reference,
 * this is to ease callers, not catch errors.
 */
struct filedesc *
fdcopy(struct filedesc *fdp)
{
	struct filedesc *newfdp;
	struct filedescent *nfde, *ofde;
	int i, lastfile;

	MPASS(fdp != NULL);

	newfdp = fdinit(fdp, true, &lastfile);
	/* copy all passable descriptors (i.e. not kqueue) */
	newfdp->fd_freefile = -1;
	for (i = 0; i <= lastfile; ++i) {
		ofde = &fdp->fd_ofiles[i];
		if (ofde->fde_file == NULL ||
		    (ofde->fde_file->f_ops->fo_flags & DFLAG_PASSABLE) == 0 ||
		    !fhold(ofde->fde_file)) {
			if (newfdp->fd_freefile == -1)
				newfdp->fd_freefile = i;
			continue;
		}
		nfde = &newfdp->fd_ofiles[i];
		*nfde = *ofde;
		filecaps_copy(&ofde->fde_caps, &nfde->fde_caps, true);
		fdused_init(newfdp, i);
	}
	if (newfdp->fd_freefile == -1)
		newfdp->fd_freefile = i;
	newfdp->fd_cmask = fdp->fd_cmask;
	FILEDESC_SUNLOCK(fdp);
	return (newfdp);
}

/*
 * Copies a filedesc structure, while remapping all file descriptors
 * stored inside using a translation table.
 *
 * File descriptors are copied over to the new file descriptor table,
 * regardless of whether the close-on-exec flag is set.
 */
int
fdcopy_remapped(struct filedesc *fdp, const int *fds, size_t nfds,
    struct filedesc **ret)
{
	struct filedesc *newfdp;
	struct filedescent *nfde, *ofde;
	int error, i, lastfile;

	MPASS(fdp != NULL);

	newfdp = fdinit(fdp, true, &lastfile);
	if (nfds > lastfile + 1) {
		/* New table cannot be larger than the old one. */
		error = E2BIG;
		goto bad;
	}
	/* Copy all passable descriptors (i.e. not kqueue). */
	newfdp->fd_freefile = nfds;
	for (i = 0; i < nfds; ++i) {
		if (fds[i] < 0 || fds[i] > lastfile) {
			/* File descriptor out of bounds. */
			error = EBADF;
			goto bad;
		}
		ofde = &fdp->fd_ofiles[fds[i]];
		if (ofde->fde_file == NULL) {
			/* Unused file descriptor. */
			error = EBADF;
			goto bad;
		}
		if ((ofde->fde_file->f_ops->fo_flags & DFLAG_PASSABLE) == 0) {
			/* File descriptor cannot be passed. */
			error = EINVAL;
			goto bad;
		}
		if (!fhold(nfde->fde_file)) {
			error = EBADF;
			goto bad;
		}
		nfde = &newfdp->fd_ofiles[i];
		*nfde = *ofde;
		filecaps_copy(&ofde->fde_caps, &nfde->fde_caps, true);
		fdused_init(newfdp, i);
	}
	newfdp->fd_cmask = fdp->fd_cmask;
	FILEDESC_SUNLOCK(fdp);
	*ret = newfdp;
	return (0);
bad:
	FILEDESC_SUNLOCK(fdp);
	fdescfree_remapped(newfdp);
	return (error);
}

/*
 * Clear POSIX style locks. This is only used when fdp looses a reference (i.e.
 * one of processes using it exits) and the table used to be shared.
 */
static void
fdclearlocks(struct thread *td)
{
	struct filedesc *fdp;
	struct filedesc_to_leader *fdtol;
	struct flock lf;
	struct file *fp;
	struct proc *p;
	struct vnode *vp;
	int i, lastfile;

	p = td->td_proc;
	fdp = p->p_fd;
	fdtol = p->p_fdtol;
	MPASS(fdtol != NULL);

	FILEDESC_XLOCK(fdp);
	KASSERT(fdtol->fdl_refcount > 0,
	    ("filedesc_to_refcount botch: fdl_refcount=%d",
	    fdtol->fdl_refcount));
	if (fdtol->fdl_refcount == 1 &&
	    (p->p_leader->p_flag & P_ADVLOCK) != 0) {
		lastfile = fdlastfile(fdp);
		for (i = 0; i <= lastfile; i++) {
			fp = fdp->fd_ofiles[i].fde_file;
			if (fp == NULL || fp->f_type != DTYPE_VNODE ||
			    !fhold(fp))
				continue;
			FILEDESC_XUNLOCK(fdp);
			lf.l_whence = SEEK_SET;
			lf.l_start = 0;
			lf.l_len = 0;
			lf.l_type = F_UNLCK;
			vp = fp->f_vnode;
			(void) VOP_ADVLOCK(vp,
			    (caddr_t)p->p_leader, F_UNLCK,
			    &lf, F_POSIX);
			FILEDESC_XLOCK(fdp);
			fdrop(fp, td);
		}
	}
retry:
	if (fdtol->fdl_refcount == 1) {
		if (fdp->fd_holdleaderscount > 0 &&
		    (p->p_leader->p_flag & P_ADVLOCK) != 0) {
			/*
			 * close() or kern_dup() has cleared a reference
			 * in a shared file descriptor table.
			 */
			fdp->fd_holdleaderswakeup = 1;
			sx_sleep(&fdp->fd_holdleaderscount,
			    FILEDESC_LOCK(fdp), PLOCK, "fdlhold", 0);
			goto retry;
		}
		if (fdtol->fdl_holdcount > 0) {
			/*
			 * Ensure that fdtol->fdl_leader remains
			 * valid in closef().
			 */
			fdtol->fdl_wakeup = 1;
			sx_sleep(fdtol, FILEDESC_LOCK(fdp), PLOCK,
			    "fdlhold", 0);
			goto retry;
		}
	}
	fdtol->fdl_refcount--;
	if (fdtol->fdl_refcount == 0 &&
	    fdtol->fdl_holdcount == 0) {
		fdtol->fdl_next->fdl_prev = fdtol->fdl_prev;
		fdtol->fdl_prev->fdl_next = fdtol->fdl_next;
	} else
		fdtol = NULL;
	p->p_fdtol = NULL;
	FILEDESC_XUNLOCK(fdp);
	if (fdtol != NULL)
		free(fdtol, M_FILEDESC_TO_LEADER);
}

/*
 * Release a filedesc structure.
 */
static void
fdescfree_fds(struct thread *td, struct filedesc *fdp, bool needclose)
{
	struct filedesc0 *fdp0;
	struct freetable *ft, *tft;
	struct filedescent *fde;
	struct file *fp;
	int i, lastfile;

	lastfile = fdlastfile_single(fdp);
	for (i = 0; i <= lastfile; i++) {
		fde = &fdp->fd_ofiles[i];
		fp = fde->fde_file;
		if (fp != NULL) {
			fdefree_last(fde);
			if (needclose)
				(void) closef(fp, td);
			else
				fdrop(fp, td);
		}
	}

	if (NDSLOTS(fdp->fd_nfiles) > NDSLOTS(NDFILE))
		free(fdp->fd_map, M_FILEDESC);
	if (fdp->fd_nfiles > NDFILE)
		free(fdp->fd_files, M_FILEDESC);

	fdp0 = (struct filedesc0 *)fdp;
	SLIST_FOREACH_SAFE(ft, &fdp0->fd_free, ft_next, tft)
		free(ft->ft_table, M_FILEDESC);

	fddrop(fdp);
}

void
fdescfree(struct thread *td)
{
	struct proc *p;
	struct filedesc *fdp;
	struct pwd *pwd;

	p = td->td_proc;
	fdp = p->p_fd;
	MPASS(fdp != NULL);

#ifdef RACCT
	if (RACCT_ENABLED())
		racct_set_unlocked(p, RACCT_NOFILE, 0);
#endif

	if (p->p_fdtol != NULL)
		fdclearlocks(td);

	PROC_LOCK(p);
	p->p_fd = NULL;
	PROC_UNLOCK(p);

	if (refcount_release(&fdp->fd_refcnt) == 0)
		return;

	FILEDESC_XLOCK(fdp);
	pwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
	pwd_set(fdp, NULL);
	FILEDESC_XUNLOCK(fdp);

	pwd_drop(pwd);

	fdescfree_fds(td, fdp, 1);
}

void
fdescfree_remapped(struct filedesc *fdp)
{

	pwd_drop(smr_serialized_load(&fdp->fd_pwd, true));
	fdescfree_fds(curthread, fdp, 0);
}

/*
 * For setugid programs, we don't want to people to use that setugidness
 * to generate error messages which write to a file which otherwise would
 * otherwise be off-limits to the process.  We check for filesystems where
 * the vnode can change out from under us after execve (like [lin]procfs).
 *
 * Since fdsetugidsafety calls this only for fd 0, 1 and 2, this check is
 * sufficient.  We also don't check for setugidness since we know we are.
 */
static bool
is_unsafe(struct file *fp)
{
	struct vnode *vp;

	if (fp->f_type != DTYPE_VNODE)
		return (false);

	vp = fp->f_vnode;
	return ((vp->v_vflag & VV_PROCDEP) != 0);
}

/*
 * Make this setguid thing safe, if at all possible.
 */
void
fdsetugidsafety(struct thread *td)
{
	struct filedesc *fdp;
	struct file *fp;
	int i;

	fdp = td->td_proc->p_fd;
	KASSERT(fdp->fd_refcnt == 1, ("the fdtable should not be shared"));
	MPASS(fdp->fd_nfiles >= 3);
	for (i = 0; i <= 2; i++) {
		fp = fdp->fd_ofiles[i].fde_file;
		if (fp != NULL && is_unsafe(fp)) {
			FILEDESC_XLOCK(fdp);
			knote_fdclose(td, i);
			/*
			 * NULL-out descriptor prior to close to avoid
			 * a race while close blocks.
			 */
			fdfree(fdp, i);
			FILEDESC_XUNLOCK(fdp);
			(void) closef(fp, td);
		}
	}
}

/*
 * If a specific file object occupies a specific file descriptor, close the
 * file descriptor entry and drop a reference on the file object.  This is a
 * convenience function to handle a subsequent error in a function that calls
 * falloc() that handles the race that another thread might have closed the
 * file descriptor out from under the thread creating the file object.
 */
void
fdclose(struct thread *td, struct file *fp, int idx)
{
	struct filedesc *fdp = td->td_proc->p_fd;

	FILEDESC_XLOCK(fdp);
	if (fdp->fd_ofiles[idx].fde_file == fp) {
		fdfree(fdp, idx);
		FILEDESC_XUNLOCK(fdp);
		fdrop(fp, td);
	} else
		FILEDESC_XUNLOCK(fdp);
}

/*
 * Close any files on exec?
 */
void
fdcloseexec(struct thread *td)
{
	struct filedesc *fdp;
	struct filedescent *fde;
	struct file *fp;
	int i, lastfile;

	fdp = td->td_proc->p_fd;
	KASSERT(fdp->fd_refcnt == 1, ("the fdtable should not be shared"));
	lastfile = fdlastfile_single(fdp);
	for (i = 0; i <= lastfile; i++) {
		fde = &fdp->fd_ofiles[i];
		fp = fde->fde_file;
		if (fp != NULL && (fp->f_type == DTYPE_MQUEUE ||
		    (fde->fde_flags & UF_EXCLOSE))) {
			FILEDESC_XLOCK(fdp);
			fdfree(fdp, i);
			(void) closefp(fdp, i, fp, td, 0);
			FILEDESC_UNLOCK_ASSERT(fdp);
		}
	}
}

/*
 * It is unsafe for set[ug]id processes to be started with file
 * descriptors 0..2 closed, as these descriptors are given implicit
 * significance in the Standard C library.  fdcheckstd() will create a
 * descriptor referencing /dev/null for each of stdin, stdout, and
 * stderr that is not already open.
 */
int
fdcheckstd(struct thread *td)
{
	struct filedesc *fdp;
	register_t save;
	int i, error, devnull;

	fdp = td->td_proc->p_fd;
	KASSERT(fdp->fd_refcnt == 1, ("the fdtable should not be shared"));
	MPASS(fdp->fd_nfiles >= 3);
	devnull = -1;
	for (i = 0; i <= 2; i++) {
		if (fdp->fd_ofiles[i].fde_file != NULL)
			continue;

		save = td->td_retval[0];
		if (devnull != -1) {
			error = kern_dup(td, FDDUP_FIXED, 0, devnull, i);
		} else {
			error = kern_openat(td, AT_FDCWD, "/dev/null",
			    UIO_SYSSPACE, O_RDWR, 0);
			if (error == 0) {
				devnull = td->td_retval[0];
				KASSERT(devnull == i, ("we didn't get our fd"));
			}
		}
		td->td_retval[0] = save;
		if (error != 0)
			return (error);
	}
	return (0);
}

/*
 * Internal form of close.  Decrement reference count on file structure.
 * Note: td may be NULL when closing a file that was being passed in a
 * message.
 */
int
closef(struct file *fp, struct thread *td)
{
	struct vnode *vp;
	struct flock lf;
	struct filedesc_to_leader *fdtol;
	struct filedesc *fdp;

	/*
	 * POSIX record locking dictates that any close releases ALL
	 * locks owned by this process.  This is handled by setting
	 * a flag in the unlock to free ONLY locks obeying POSIX
	 * semantics, and not to free BSD-style file locks.
	 * If the descriptor was in a message, POSIX-style locks
	 * aren't passed with the descriptor, and the thread pointer
	 * will be NULL.  Callers should be careful only to pass a
	 * NULL thread pointer when there really is no owning
	 * context that might have locks, or the locks will be
	 * leaked.
	 */
	if (fp->f_type == DTYPE_VNODE && td != NULL) {
		vp = fp->f_vnode;
		if ((td->td_proc->p_leader->p_flag & P_ADVLOCK) != 0) {
			lf.l_whence = SEEK_SET;
			lf.l_start = 0;
			lf.l_len = 0;
			lf.l_type = F_UNLCK;
			(void) VOP_ADVLOCK(vp, (caddr_t)td->td_proc->p_leader,
			    F_UNLCK, &lf, F_POSIX);
		}
		fdtol = td->td_proc->p_fdtol;
		if (fdtol != NULL) {
			/*
			 * Handle special case where file descriptor table is
			 * shared between multiple process leaders.
			 */
			fdp = td->td_proc->p_fd;
			FILEDESC_XLOCK(fdp);
			for (fdtol = fdtol->fdl_next;
			    fdtol != td->td_proc->p_fdtol;
			    fdtol = fdtol->fdl_next) {
				if ((fdtol->fdl_leader->p_flag &
				    P_ADVLOCK) == 0)
					continue;
				fdtol->fdl_holdcount++;
				FILEDESC_XUNLOCK(fdp);
				lf.l_whence = SEEK_SET;
				lf.l_start = 0;
				lf.l_len = 0;
				lf.l_type = F_UNLCK;
				vp = fp->f_vnode;
				(void) VOP_ADVLOCK(vp,
				    (caddr_t)fdtol->fdl_leader, F_UNLCK, &lf,
				    F_POSIX);
				FILEDESC_XLOCK(fdp);
				fdtol->fdl_holdcount--;
				if (fdtol->fdl_holdcount == 0 &&
				    fdtol->fdl_wakeup != 0) {
					fdtol->fdl_wakeup = 0;
					wakeup(fdtol);
				}
			}
			FILEDESC_XUNLOCK(fdp);
		}
	}
	return (fdrop(fp, td));
}

/*
 * Initialize the file pointer with the specified properties.
 *
 * The ops are set with release semantics to be certain that the flags, type,
 * and data are visible when ops is.  This is to prevent ops methods from being
 * called with bad data.
 */
void
finit(struct file *fp, u_int flag, short type, void *data, struct fileops *ops)
{
	fp->f_data = data;
	fp->f_flag = flag;
	fp->f_type = type;
	atomic_store_rel_ptr((volatile uintptr_t *)&fp->f_ops, (uintptr_t)ops);
}

int
fget_cap_locked(struct filedesc *fdp, int fd, cap_rights_t *needrightsp,
    struct file **fpp, struct filecaps *havecapsp)
{
	struct filedescent *fde;
	int error;

	FILEDESC_LOCK_ASSERT(fdp);

	fde = fdeget_locked(fdp, fd);
	if (fde == NULL) {
		error = EBADF;
		goto out;
	}

#ifdef CAPABILITIES
	error = cap_check(cap_rights_fde_inline(fde), needrightsp);
	if (error != 0)
		goto out;
#endif

	if (havecapsp != NULL)
		filecaps_copy(&fde->fde_caps, havecapsp, true);

	*fpp = fde->fde_file;

	error = 0;
out:
	return (error);
}

int
fget_cap(struct thread *td, int fd, cap_rights_t *needrightsp,
    struct file **fpp, struct filecaps *havecapsp)
{
	struct filedesc *fdp = td->td_proc->p_fd;
	int error;
#ifndef CAPABILITIES
	error = fget_unlocked(fdp, fd, needrightsp, fpp);
	if (havecapsp != NULL && error == 0)
		filecaps_fill(havecapsp);
#else
	struct file *fp;
	seqc_t seq;

	*fpp = NULL;
	for (;;) {
		error = fget_unlocked_seq(fdp, fd, needrightsp, &fp, &seq);
		if (error != 0)
			return (error);

		if (havecapsp != NULL) {
			if (!filecaps_copy(&fdp->fd_ofiles[fd].fde_caps,
			    havecapsp, false)) {
				fdrop(fp, td);
				goto get_locked;
			}
		}

		if (!fd_modified(fdp, fd, seq))
			break;
		fdrop(fp, td);
	}

	*fpp = fp;
	return (0);

get_locked:
	FILEDESC_SLOCK(fdp);
	error = fget_cap_locked(fdp, fd, needrightsp, fpp, havecapsp);
	if (error == 0 && !fhold(*fpp))
		error = EBADF;
	FILEDESC_SUNLOCK(fdp);
#endif
	return (error);
}

int
fget_unlocked_seq(struct filedesc *fdp, int fd, cap_rights_t *needrightsp,
    struct file **fpp, seqc_t *seqp)
{
#ifdef CAPABILITIES
	const struct filedescent *fde;
#endif
	const struct fdescenttbl *fdt;
	struct file *fp;
#ifdef CAPABILITIES
	seqc_t seq;
	cap_rights_t haverights;
	int error;
#endif

	fdt = fdp->fd_files;
	if (__predict_false((u_int)fd >= fdt->fdt_nfiles))
		return (EBADF);
	/*
	 * Fetch the descriptor locklessly.  We avoid fdrop() races by
	 * never raising a refcount above 0.  To accomplish this we have
	 * to use a cmpset loop rather than an atomic_add.  The descriptor
	 * must be re-verified once we acquire a reference to be certain
	 * that the identity is still correct and we did not lose a race
	 * due to preemption.
	 */
	for (;;) {
#ifdef CAPABILITIES
		seq = seqc_read(fd_seqc(fdt, fd));
		fde = &fdt->fdt_ofiles[fd];
		haverights = *cap_rights_fde_inline(fde);
		fp = fde->fde_file;
		if (!seqc_consistent(fd_seqc(fdt, fd), seq))
			continue;
#else
		fp = fdt->fdt_ofiles[fd].fde_file;
#endif
		if (fp == NULL)
			return (EBADF);
#ifdef CAPABILITIES
		error = cap_check_inline(&haverights, needrightsp);
		if (error != 0)
			return (error);
#endif
		if (__predict_false(!refcount_acquire_if_not_zero(&fp->f_count))) {
			/*
			 * The count was found either saturated or zero.
			 * This re-read is not any more racy than using the
			 * return value from fcmpset.
			 */
			if (fp->f_count != 0)
				return (EBADF);
			/*
			 * Force a reload. Other thread could reallocate the
			 * table before this fd was closed, so it is possible
			 * that there is a stale fp pointer in cached version.
			 */
			fdt = atomic_load_ptr(&fdp->fd_files);
			continue;
		}
		/*
		 * Use an acquire barrier to force re-reading of fdt so it is
		 * refreshed for verification.
		 */
		atomic_thread_fence_acq();
		fdt = fdp->fd_files;
#ifdef	CAPABILITIES
		if (seqc_consistent_nomb(fd_seqc(fdt, fd), seq))
#else
		if (fp == fdt->fdt_ofiles[fd].fde_file)
#endif
			break;
		fdrop(fp, curthread);
	}
	*fpp = fp;
	if (seqp != NULL) {
#ifdef CAPABILITIES
		*seqp = seq;
#endif
	}
	return (0);
}

/*
 * See the comments in fget_unlocked_seq for an explanation of how this works.
 *
 * This is a simplified variant which bails out to the aforementioned routine
 * if anything goes wrong. In practice this only happens when userspace is
 * racing with itself.
 */
int
fget_unlocked(struct filedesc *fdp, int fd, cap_rights_t *needrightsp,
    struct file **fpp)
{
#ifdef CAPABILITIES
	const struct filedescent *fde;
#endif
	const struct fdescenttbl *fdt;
	struct file *fp;
#ifdef CAPABILITIES
	seqc_t seq;
	const cap_rights_t *haverights;
#endif

	fdt = fdp->fd_files;
	if (__predict_false((u_int)fd >= fdt->fdt_nfiles))
		return (EBADF);
#ifdef CAPABILITIES
	seq = seqc_read_any(fd_seqc(fdt, fd));
	if (__predict_false(seqc_in_modify(seq)))
		goto out_fallback;
	fde = &fdt->fdt_ofiles[fd];
	haverights = cap_rights_fde_inline(fde);
	fp = fde->fde_file;
#else
	fp = fdt->fdt_ofiles[fd].fde_file;
#endif
	if (__predict_false(fp == NULL))
		goto out_fallback;
#ifdef CAPABILITIES
	if (__predict_false(cap_check_inline_transient(haverights, needrightsp)))
		goto out_fallback;
#endif
	if (__predict_false(!refcount_acquire_if_not_zero(&fp->f_count)))
		goto out_fallback;

	/*
	 * Use an acquire barrier to force re-reading of fdt so it is
	 * refreshed for verification.
	 */
	atomic_thread_fence_acq();
	fdt = fdp->fd_files;
#ifdef	CAPABILITIES
	if (__predict_false(!seqc_consistent_nomb(fd_seqc(fdt, fd), seq)))
#else
	if (__predict_false(fp != fdt->fdt_ofiles[fd].fde_file))
#endif
		goto out_fdrop;
	*fpp = fp;
	return (0);
out_fdrop:
	fdrop(fp, curthread);
out_fallback:
	return (fget_unlocked_seq(fdp, fd, needrightsp, fpp, NULL));
}

/*
 * Extract the file pointer associated with the specified descriptor for the
 * current user process.
 *
 * If the descriptor doesn't exist or doesn't match 'flags', EBADF is
 * returned.
 *
 * File's rights will be checked against the capability rights mask.
 *
 * If an error occurred the non-zero error is returned and *fpp is set to
 * NULL.  Otherwise *fpp is held and set and zero is returned.  Caller is
 * responsible for fdrop().
 */
static __inline int
_fget(struct thread *td, int fd, struct file **fpp, int flags,
    cap_rights_t *needrightsp)
{
	struct filedesc *fdp;
	struct file *fp;
	int error;

	*fpp = NULL;
	fdp = td->td_proc->p_fd;
	error = fget_unlocked(fdp, fd, needrightsp, &fp);
	if (__predict_false(error != 0))
		return (error);
	if (__predict_false(fp->f_ops == &badfileops)) {
		fdrop(fp, td);
		return (EBADF);
	}

	/*
	 * FREAD and FWRITE failure return EBADF as per POSIX.
	 */
	error = 0;
	switch (flags) {
	case FREAD:
	case FWRITE:
		if ((fp->f_flag & flags) == 0)
			error = EBADF;
		break;
	case FEXEC:
	    	if ((fp->f_flag & (FREAD | FEXEC)) == 0 ||
		    ((fp->f_flag & FWRITE) != 0))
			error = EBADF;
		break;
	case 0:
		break;
	default:
		KASSERT(0, ("wrong flags"));
	}

	if (error != 0) {
		fdrop(fp, td);
		return (error);
	}

	*fpp = fp;
	return (0);
}

int
fget(struct thread *td, int fd, cap_rights_t *rightsp, struct file **fpp)
{

	return (_fget(td, fd, fpp, 0, rightsp));
}

int
fget_mmap(struct thread *td, int fd, cap_rights_t *rightsp, vm_prot_t *maxprotp,
    struct file **fpp)
{
	int error;
#ifndef CAPABILITIES
	error = _fget(td, fd, fpp, 0, rightsp);
	if (maxprotp != NULL)
		*maxprotp = VM_PROT_ALL;
	return (error);
#else
	cap_rights_t fdrights;
	struct filedesc *fdp;
	struct file *fp;
	seqc_t seq;

	*fpp = NULL;
	fdp = td->td_proc->p_fd;
	MPASS(cap_rights_is_set(rightsp, CAP_MMAP));
	for (;;) {
		error = fget_unlocked_seq(fdp, fd, rightsp, &fp, &seq);
		if (__predict_false(error != 0))
			return (error);
		if (__predict_false(fp->f_ops == &badfileops)) {
			fdrop(fp, td);
			return (EBADF);
		}
		if (maxprotp != NULL)
			fdrights = *cap_rights(fdp, fd);
		if (!fd_modified(fdp, fd, seq))
			break;
		fdrop(fp, td);
	}

	/*
	 * If requested, convert capability rights to access flags.
	 */
	if (maxprotp != NULL)
		*maxprotp = cap_rights_to_vmprot(&fdrights);
	*fpp = fp;
	return (0);
#endif
}

int
fget_read(struct thread *td, int fd, cap_rights_t *rightsp, struct file **fpp)
{

	return (_fget(td, fd, fpp, FREAD, rightsp));
}

int
fget_write(struct thread *td, int fd, cap_rights_t *rightsp, struct file **fpp)
{

	return (_fget(td, fd, fpp, FWRITE, rightsp));
}

int
fget_fcntl(struct thread *td, int fd, cap_rights_t *rightsp, int needfcntl,
    struct file **fpp)
{
	struct filedesc *fdp = td->td_proc->p_fd;
#ifndef CAPABILITIES
	return (fget_unlocked(fdp, fd, rightsp, fpp));
#else
	struct file *fp;
	int error;
	seqc_t seq;

	*fpp = NULL;
	MPASS(cap_rights_is_set(rightsp, CAP_FCNTL));
	for (;;) {
		error = fget_unlocked_seq(fdp, fd, rightsp, &fp, &seq);
		if (error != 0)
			return (error);
		error = cap_fcntl_check(fdp, fd, needfcntl);
		if (!fd_modified(fdp, fd, seq))
			break;
		fdrop(fp, td);
	}
	if (error != 0) {
		fdrop(fp, td);
		return (error);
	}
	*fpp = fp;
	return (0);
#endif
}

/*
 * Like fget() but loads the underlying vnode, or returns an error if the
 * descriptor does not represent a vnode.  Note that pipes use vnodes but
 * never have VM objects.  The returned vnode will be vref()'d.
 *
 * XXX: what about the unused flags ?
 */
static __inline int
_fgetvp(struct thread *td, int fd, int flags, cap_rights_t *needrightsp,
    struct vnode **vpp)
{
	struct file *fp;
	int error;

	*vpp = NULL;
	error = _fget(td, fd, &fp, flags, needrightsp);
	if (error != 0)
		return (error);
	if (fp->f_vnode == NULL) {
		error = EINVAL;
	} else {
		*vpp = fp->f_vnode;
		vrefact(*vpp);
	}
	fdrop(fp, td);

	return (error);
}

int
fgetvp(struct thread *td, int fd, cap_rights_t *rightsp, struct vnode **vpp)
{

	return (_fgetvp(td, fd, 0, rightsp, vpp));
}

int
fgetvp_rights(struct thread *td, int fd, cap_rights_t *needrightsp,
    struct filecaps *havecaps, struct vnode **vpp)
{
	struct filecaps caps;
	struct file *fp;
	int error;

	error = fget_cap(td, fd, needrightsp, &fp, &caps);
	if (error != 0)
		return (error);
	if (fp->f_ops == &badfileops) {
		error = EBADF;
		goto out;
	}
	if (fp->f_vnode == NULL) {
		error = EINVAL;
		goto out;
	}

	*havecaps = caps;
	*vpp = fp->f_vnode;
	vrefact(*vpp);
	fdrop(fp, td);

	return (0);
out:
	filecaps_free(&caps);
	fdrop(fp, td);
	return (error);
}

int
fgetvp_read(struct thread *td, int fd, cap_rights_t *rightsp, struct vnode **vpp)
{

	return (_fgetvp(td, fd, FREAD, rightsp, vpp));
}

int
fgetvp_exec(struct thread *td, int fd, cap_rights_t *rightsp, struct vnode **vpp)
{

	return (_fgetvp(td, fd, FEXEC, rightsp, vpp));
}

#ifdef notyet
int
fgetvp_write(struct thread *td, int fd, cap_rights_t *rightsp,
    struct vnode **vpp)
{

	return (_fgetvp(td, fd, FWRITE, rightsp, vpp));
}
#endif

/*
 * Handle the last reference to a file being closed.
 *
 * Without the noinline attribute clang keeps inlining the func thorough this
 * file when fdrop is used.
 */
int __noinline
_fdrop(struct file *fp, struct thread *td)
{
	int error;

	if (fp->f_count != 0)
		panic("fdrop: count %d", fp->f_count);
	error = fo_close(fp, td);
	atomic_subtract_int(&openfiles, 1);
	crfree(fp->f_cred);
	free(fp->f_advice, M_FADVISE);
	uma_zfree(file_zone, fp);

	return (error);
}

/*
 * Apply an advisory lock on a file descriptor.
 *
 * Just attempt to get a record lock of the requested type on the entire file
 * (l_whence = SEEK_SET, l_start = 0, l_len = 0).
 */
#ifndef _SYS_SYSPROTO_H_
struct flock_args {
	int	fd;
	int	how;
};
#endif
/* ARGSUSED */
int
sys_flock(struct thread *td, struct flock_args *uap)
{
	struct file *fp;
	struct vnode *vp;
	struct flock lf;
	int error;

	error = fget(td, uap->fd, &cap_flock_rights, &fp);
	if (error != 0)
		return (error);
	if (fp->f_type != DTYPE_VNODE) {
		fdrop(fp, td);
		return (EOPNOTSUPP);
	}

	vp = fp->f_vnode;
	lf.l_whence = SEEK_SET;
	lf.l_start = 0;
	lf.l_len = 0;
	if (uap->how & LOCK_UN) {
		lf.l_type = F_UNLCK;
		atomic_clear_int(&fp->f_flag, FHASLOCK);
		error = VOP_ADVLOCK(vp, (caddr_t)fp, F_UNLCK, &lf, F_FLOCK);
		goto done2;
	}
	if (uap->how & LOCK_EX)
		lf.l_type = F_WRLCK;
	else if (uap->how & LOCK_SH)
		lf.l_type = F_RDLCK;
	else {
		error = EBADF;
		goto done2;
	}
	atomic_set_int(&fp->f_flag, FHASLOCK);
	error = VOP_ADVLOCK(vp, (caddr_t)fp, F_SETLK, &lf,
	    (uap->how & LOCK_NB) ? F_FLOCK : F_FLOCK | F_WAIT);
done2:
	fdrop(fp, td);
	return (error);
}
/*
 * Duplicate the specified descriptor to a free descriptor.
 */
int
dupfdopen(struct thread *td, struct filedesc *fdp, int dfd, int mode,
    int openerror, int *indxp)
{
	struct filedescent *newfde, *oldfde;
	struct file *fp;
	u_long *ioctls;
	int error, indx;

	KASSERT(openerror == ENODEV || openerror == ENXIO,
	    ("unexpected error %d in %s", openerror, __func__));

	/*
	 * If the to-be-dup'd fd number is greater than the allowed number
	 * of file descriptors, or the fd to be dup'd has already been
	 * closed, then reject.
	 */
	FILEDESC_XLOCK(fdp);
	if ((fp = fget_locked(fdp, dfd)) == NULL) {
		FILEDESC_XUNLOCK(fdp);
		return (EBADF);
	}

	error = fdalloc(td, 0, &indx);
	if (error != 0) {
		FILEDESC_XUNLOCK(fdp);
		return (error);
	}

	/*
	 * There are two cases of interest here.
	 *
	 * For ENODEV simply dup (dfd) to file descriptor (indx) and return.
	 *
	 * For ENXIO steal away the file structure from (dfd) and store it in
	 * (indx).  (dfd) is effectively closed by this operation.
	 */
	switch (openerror) {
	case ENODEV:
		/*
		 * Check that the mode the file is being opened for is a
		 * subset of the mode of the existing descriptor.
		 */
		if (((mode & (FREAD|FWRITE)) | fp->f_flag) != fp->f_flag) {
			fdunused(fdp, indx);
			FILEDESC_XUNLOCK(fdp);
			return (EACCES);
		}
		if (!fhold(fp)) {
			fdunused(fdp, indx);
			FILEDESC_XUNLOCK(fdp);
			return (EBADF);
		}
		newfde = &fdp->fd_ofiles[indx];
		oldfde = &fdp->fd_ofiles[dfd];
		ioctls = filecaps_copy_prep(&oldfde->fde_caps);
#ifdef CAPABILITIES
		seqc_write_begin(&newfde->fde_seqc);
#endif
		memcpy(newfde, oldfde, fde_change_size);
		filecaps_copy_finish(&oldfde->fde_caps, &newfde->fde_caps,
		    ioctls);
#ifdef CAPABILITIES
		seqc_write_end(&newfde->fde_seqc);
#endif
		break;
	case ENXIO:
		/*
		 * Steal away the file pointer from dfd and stuff it into indx.
		 */
		newfde = &fdp->fd_ofiles[indx];
		oldfde = &fdp->fd_ofiles[dfd];
#ifdef CAPABILITIES
		seqc_write_begin(&newfde->fde_seqc);
#endif
		memcpy(newfde, oldfde, fde_change_size);
		oldfde->fde_file = NULL;
		fdunused(fdp, dfd);
#ifdef CAPABILITIES
		seqc_write_end(&newfde->fde_seqc);
#endif
		break;
	}
	FILEDESC_XUNLOCK(fdp);
	*indxp = indx;
	return (0);
}

/*
 * This sysctl determines if we will allow a process to chroot(2) if it
 * has a directory open:
 *	0: disallowed for all processes.
 *	1: allowed for processes that were not already chroot(2)'ed.
 *	2: allowed for all processes.
 */

static int chroot_allow_open_directories = 1;

SYSCTL_INT(_kern, OID_AUTO, chroot_allow_open_directories, CTLFLAG_RW,
    &chroot_allow_open_directories, 0,
    "Allow a process to chroot(2) if it has a directory open");

/*
 * Helper function for raised chroot(2) security function:  Refuse if
 * any filedescriptors are open directories.
 */
static int
chroot_refuse_vdir_fds(struct filedesc *fdp)
{
	struct vnode *vp;
	struct file *fp;
	int fd, lastfile;

	FILEDESC_LOCK_ASSERT(fdp);

	lastfile = fdlastfile(fdp);
	for (fd = 0; fd <= lastfile; fd++) {
		fp = fget_locked(fdp, fd);
		if (fp == NULL)
			continue;
		if (fp->f_type == DTYPE_VNODE) {
			vp = fp->f_vnode;
			if (vp->v_type == VDIR)
				return (EPERM);
		}
	}
	return (0);
}

static void
pwd_fill(struct pwd *oldpwd, struct pwd *newpwd)
{

	if (newpwd->pwd_cdir == NULL && oldpwd->pwd_cdir != NULL) {
		vrefact(oldpwd->pwd_cdir);
		newpwd->pwd_cdir = oldpwd->pwd_cdir;
	}

	if (newpwd->pwd_rdir == NULL && oldpwd->pwd_rdir != NULL) {
		vrefact(oldpwd->pwd_rdir);
		newpwd->pwd_rdir = oldpwd->pwd_rdir;
	}

	if (newpwd->pwd_jdir == NULL && oldpwd->pwd_jdir != NULL) {
		vrefact(oldpwd->pwd_jdir);
		newpwd->pwd_jdir = oldpwd->pwd_jdir;
	}
}

struct pwd *
pwd_hold_filedesc(struct filedesc *fdp)
{
	struct pwd *pwd;

	FILEDESC_LOCK_ASSERT(fdp);
	pwd = FILEDESC_LOCKED_LOAD_PWD(fdp);
	if (pwd != NULL)
		refcount_acquire(&pwd->pwd_refcount);
	return (pwd);
}

struct pwd *
pwd_hold(struct thread *td)
{
	struct filedesc *fdp;
	struct pwd *pwd;

	fdp = td->td_proc->p_fd;

	vfs_smr_enter();
	pwd = vfs_smr_entered_load(&fdp->fd_pwd);
	MPASS(pwd != NULL);
	if (__predict_true(refcount_acquire_if_not_zero(&pwd->pwd_refcount))) {
		vfs_smr_exit();
		return (pwd);
	}
	vfs_smr_exit();
	FILEDESC_SLOCK(fdp);
	pwd = pwd_hold_filedesc(fdp);
	MPASS(pwd != NULL);
	return (pwd);
}

struct pwd *
pwd_get_smr(void)
{
	struct pwd *pwd;

	pwd = vfs_smr_entered_load(&curproc->p_fd->fd_pwd);
	MPASS(pwd != NULL);
	return (pwd);
}

static struct pwd *
pwd_alloc(void)
{
	struct pwd *pwd;

	pwd = uma_zalloc_smr(pwd_zone, M_WAITOK);
	bzero(pwd, sizeof(*pwd));
	refcount_init(&pwd->pwd_refcount, 1);
	return (pwd);
}

void
pwd_drop(struct pwd *pwd)
{

	if (!refcount_release(&pwd->pwd_refcount))
		return;

	if (pwd->pwd_cdir != NULL)
		vrele(pwd->pwd_cdir);
	if (pwd->pwd_rdir != NULL)
		vrele(pwd->pwd_rdir);
	if (pwd->pwd_jdir != NULL)
		vrele(pwd->pwd_jdir);
	uma_zfree_smr(pwd_zone, pwd);
}

/*
* Common routine for kern_chroot() and jail_attach().  The caller is
* responsible for invoking priv_check() and mac_vnode_check_chroot() to
* authorize this operation.
*/
int
pwd_chroot(struct thread *td, struct vnode *vp)
{
	struct filedesc *fdp;
	struct pwd *newpwd, *oldpwd;
	int error;

	fdp = td->td_proc->p_fd;
	newpwd = pwd_alloc();
	FILEDESC_XLOCK(fdp);
	oldpwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
	if (chroot_allow_open_directories == 0 ||
	    (chroot_allow_open_directories == 1 &&
	    oldpwd->pwd_rdir != rootvnode)) {
		error = chroot_refuse_vdir_fds(fdp);
		if (error != 0) {
			FILEDESC_XUNLOCK(fdp);
			pwd_drop(newpwd);
			return (error);
		}
	}

	vrefact(vp);
	newpwd->pwd_rdir = vp;
	if (oldpwd->pwd_jdir == NULL) {
		vrefact(vp);
		newpwd->pwd_jdir = vp;
	}
	pwd_fill(oldpwd, newpwd);
	pwd_set(fdp, newpwd);
	FILEDESC_XUNLOCK(fdp);
	pwd_drop(oldpwd);
	return (0);
}

void
pwd_chdir(struct thread *td, struct vnode *vp)
{
	struct filedesc *fdp;
	struct pwd *newpwd, *oldpwd;

	VNPASS(vp->v_usecount > 0, vp);

	newpwd = pwd_alloc();
	fdp = td->td_proc->p_fd;
	FILEDESC_XLOCK(fdp);
	oldpwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
	newpwd->pwd_cdir = vp;
	pwd_fill(oldpwd, newpwd);
	pwd_set(fdp, newpwd);
	FILEDESC_XUNLOCK(fdp);
	pwd_drop(oldpwd);
}

void
pwd_ensure_dirs(void)
{
	struct filedesc *fdp;
	struct pwd *oldpwd, *newpwd;

	fdp = curproc->p_fd;
	FILEDESC_XLOCK(fdp);
	oldpwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
	if (oldpwd->pwd_cdir != NULL && oldpwd->pwd_rdir != NULL) {
		FILEDESC_XUNLOCK(fdp);
		return;
	}
	FILEDESC_XUNLOCK(fdp);

	newpwd = pwd_alloc();
	FILEDESC_XLOCK(fdp);
	oldpwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
	pwd_fill(oldpwd, newpwd);
	if (newpwd->pwd_cdir == NULL) {
		vrefact(rootvnode);
		newpwd->pwd_cdir = rootvnode;
	}
	if (newpwd->pwd_rdir == NULL) {
		vrefact(rootvnode);
		newpwd->pwd_rdir = rootvnode;
	}
	pwd_set(fdp, newpwd);
	FILEDESC_XUNLOCK(fdp);
	pwd_drop(oldpwd);
}

void
pwd_set_rootvnode(void)
{
	struct filedesc *fdp;
	struct pwd *oldpwd, *newpwd;

	fdp = curproc->p_fd;

	newpwd = pwd_alloc();
	FILEDESC_XLOCK(fdp);
	oldpwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
	vrefact(rootvnode);
	newpwd->pwd_cdir = rootvnode;
	vrefact(rootvnode);
	newpwd->pwd_rdir = rootvnode;
	pwd_fill(oldpwd, newpwd);
	pwd_set(fdp, newpwd);
	FILEDESC_XUNLOCK(fdp);
	pwd_drop(oldpwd);
}

/*
 * Scan all active processes and prisons to see if any of them have a current
 * or root directory of `olddp'. If so, replace them with the new mount point.
 */
void
mountcheckdirs(struct vnode *olddp, struct vnode *newdp)
{
	struct filedesc *fdp;
	struct pwd *newpwd, *oldpwd;
	struct prison *pr;
	struct proc *p;
	int nrele;

	if (vrefcnt(olddp) == 1)
		return;
	nrele = 0;
	newpwd = pwd_alloc();
	sx_slock(&allproc_lock);
	FOREACH_PROC_IN_SYSTEM(p) {
		PROC_LOCK(p);
		fdp = fdhold(p);
		PROC_UNLOCK(p);
		if (fdp == NULL)
			continue;
		FILEDESC_XLOCK(fdp);
		oldpwd = FILEDESC_XLOCKED_LOAD_PWD(fdp);
		if (oldpwd == NULL ||
		    (oldpwd->pwd_cdir != olddp &&
		    oldpwd->pwd_rdir != olddp &&
		    oldpwd->pwd_jdir != olddp)) {
			FILEDESC_XUNLOCK(fdp);
			fddrop(fdp);
			continue;
		}
		if (oldpwd->pwd_cdir == olddp) {
			vrefact(newdp);
			newpwd->pwd_cdir = newdp;
		}
		if (oldpwd->pwd_rdir == olddp) {
			vrefact(newdp);
			newpwd->pwd_rdir = newdp;
		}
		if (oldpwd->pwd_jdir == olddp) {
			vrefact(newdp);
			newpwd->pwd_jdir = newdp;
		}
		pwd_fill(oldpwd, newpwd);
		pwd_set(fdp, newpwd);
		FILEDESC_XUNLOCK(fdp);
		pwd_drop(oldpwd);
		fddrop(fdp);
		newpwd = pwd_alloc();
	}
	sx_sunlock(&allproc_lock);
	pwd_drop(newpwd);
	if (rootvnode == olddp) {
		vrefact(newdp);
		rootvnode = newdp;
		nrele++;
	}
	mtx_lock(&prison0.pr_mtx);
	if (prison0.pr_root == olddp) {
		vrefact(newdp);
		prison0.pr_root = newdp;
		nrele++;
	}
	mtx_unlock(&prison0.pr_mtx);
	sx_slock(&allprison_lock);
	TAILQ_FOREACH(pr, &allprison, pr_list) {
		mtx_lock(&pr->pr_mtx);
		if (pr->pr_root == olddp) {
			vrefact(newdp);
			pr->pr_root = newdp;
			nrele++;
		}
		mtx_unlock(&pr->pr_mtx);
	}
	sx_sunlock(&allprison_lock);
	while (nrele--)
		vrele(olddp);
}

struct filedesc_to_leader *
filedesc_to_leader_alloc(struct filedesc_to_leader *old, struct filedesc *fdp, struct proc *leader)
{
	struct filedesc_to_leader *fdtol;

	fdtol = malloc(sizeof(struct filedesc_to_leader),
	    M_FILEDESC_TO_LEADER, M_WAITOK);
	fdtol->fdl_refcount = 1;
	fdtol->fdl_holdcount = 0;
	fdtol->fdl_wakeup = 0;
	fdtol->fdl_leader = leader;
	if (old != NULL) {
		FILEDESC_XLOCK(fdp);
		fdtol->fdl_next = old->fdl_next;
		fdtol->fdl_prev = old;
		old->fdl_next = fdtol;
		fdtol->fdl_next->fdl_prev = fdtol;
		FILEDESC_XUNLOCK(fdp);
	} else {
		fdtol->fdl_next = fdtol;
		fdtol->fdl_prev = fdtol;
	}
	return (fdtol);
}

static int
sysctl_kern_proc_nfds(SYSCTL_HANDLER_ARGS)
{
	NDSLOTTYPE *map;
	struct filedesc *fdp;
	int count, off, minoff;

	if (*(int *)arg1 != 0)
		return (EINVAL);

	fdp = curproc->p_fd;
	count = 0;
	FILEDESC_SLOCK(fdp);
	map = fdp->fd_map;
	off = NDSLOT(fdp->fd_nfiles - 1);
	for (minoff = NDSLOT(0); off >= minoff; --off)
		count += bitcountl(map[off]);
	FILEDESC_SUNLOCK(fdp);

	return (SYSCTL_OUT(req, &count, sizeof(count)));
}

static SYSCTL_NODE(_kern_proc, KERN_PROC_NFDS, nfds,
    CTLFLAG_RD|CTLFLAG_CAPRD|CTLFLAG_MPSAFE, sysctl_kern_proc_nfds,
    "Number of open file descriptors");

/*
 * Get file structures globally.
 */
static int
sysctl_kern_file(SYSCTL_HANDLER_ARGS)
{
	struct xfile xf;
	struct filedesc *fdp;
	struct file *fp;
	struct proc *p;
	int error, n, lastfile;

	error = sysctl_wire_old_buffer(req, 0);
	if (error != 0)
		return (error);
	if (req->oldptr == NULL) {
		n = 0;
		sx_slock(&allproc_lock);
		FOREACH_PROC_IN_SYSTEM(p) {
			PROC_LOCK(p);
			if (p->p_state == PRS_NEW) {
				PROC_UNLOCK(p);
				continue;
			}
			fdp = fdhold(p);
			PROC_UNLOCK(p);
			if (fdp == NULL)
				continue;
			/* overestimates sparse tables. */
			n += fdp->fd_nfiles;
			fddrop(fdp);
		}
		sx_sunlock(&allproc_lock);
		return (SYSCTL_OUT(req, 0, n * sizeof(xf)));
	}
	error = 0;
	bzero(&xf, sizeof(xf));
	xf.xf_size = sizeof(xf);
	sx_slock(&allproc_lock);
	FOREACH_PROC_IN_SYSTEM(p) {
		PROC_LOCK(p);
		if (p->p_state == PRS_NEW) {
			PROC_UNLOCK(p);
			continue;
		}
		if (p_cansee(req->td, p) != 0) {
			PROC_UNLOCK(p);
			continue;
		}
		xf.xf_pid = p->p_pid;
		xf.xf_uid = p->p_ucred->cr_uid;
		fdp = fdhold(p);
		PROC_UNLOCK(p);
		if (fdp == NULL)
			continue;
		FILEDESC_SLOCK(fdp);
		lastfile = fdlastfile(fdp);
		for (n = 0; fdp->fd_refcnt > 0 && n <= lastfile; ++n) {
			if ((fp = fdp->fd_ofiles[n].fde_file) == NULL)
				continue;
			xf.xf_fd = n;
			xf.xf_file = (uintptr_t)fp;
			xf.xf_data = (uintptr_t)fp->f_data;
			xf.xf_vnode = (uintptr_t)fp->f_vnode;
			xf.xf_type = (uintptr_t)fp->f_type;
			xf.xf_count = fp->f_count;
			xf.xf_msgcount = 0;
			xf.xf_offset = foffset_get(fp);
			xf.xf_flag = fp->f_flag;
			error = SYSCTL_OUT(req, &xf, sizeof(xf));
			if (error)
				break;
		}
		FILEDESC_SUNLOCK(fdp);
		fddrop(fdp);
		if (error)
			break;
	}
	sx_sunlock(&allproc_lock);
	return (error);
}

SYSCTL_PROC(_kern, KERN_FILE, file, CTLTYPE_OPAQUE|CTLFLAG_RD|CTLFLAG_MPSAFE,
    0, 0, sysctl_kern_file, "S,xfile", "Entire file table");

#ifdef KINFO_FILE_SIZE
CTASSERT(sizeof(struct kinfo_file) == KINFO_FILE_SIZE);
#endif

static int
xlate_fflags(int fflags)
{
	static const struct {
		int	fflag;
		int	kf_fflag;
	} fflags_table[] = {
		{ FAPPEND, KF_FLAG_APPEND },
		{ FASYNC, KF_FLAG_ASYNC },
		{ FFSYNC, KF_FLAG_FSYNC },
		{ FHASLOCK, KF_FLAG_HASLOCK },
		{ FNONBLOCK, KF_FLAG_NONBLOCK },
		{ FREAD, KF_FLAG_READ },
		{ FWRITE, KF_FLAG_WRITE },
		{ O_CREAT, KF_FLAG_CREAT },
		{ O_DIRECT, KF_FLAG_DIRECT },
		{ O_EXCL, KF_FLAG_EXCL },
		{ O_EXEC, KF_FLAG_EXEC },
		{ O_EXLOCK, KF_FLAG_EXLOCK },
		{ O_NOFOLLOW, KF_FLAG_NOFOLLOW },
		{ O_SHLOCK, KF_FLAG_SHLOCK },
		{ O_TRUNC, KF_FLAG_TRUNC }
	};
	unsigned int i;
	int kflags;

	kflags = 0;
	for (i = 0; i < nitems(fflags_table); i++)
		if (fflags & fflags_table[i].fflag)
			kflags |=  fflags_table[i].kf_fflag;
	return (kflags);
}

/* Trim unused data from kf_path by truncating the structure size. */
void
pack_kinfo(struct kinfo_file *kif)
{

	kif->kf_structsize = offsetof(struct kinfo_file, kf_path) +
	    strlen(kif->kf_path) + 1;
	kif->kf_structsize = roundup(kif->kf_structsize, sizeof(uint64_t));
}

static void
export_file_to_kinfo(struct file *fp, int fd, cap_rights_t *rightsp,
    struct kinfo_file *kif, struct filedesc *fdp, int flags)
{
	int error;

	bzero(kif, sizeof(*kif));

	/* Set a default type to allow for empty fill_kinfo() methods. */
	kif->kf_type = KF_TYPE_UNKNOWN;
	kif->kf_flags = xlate_fflags(fp->f_flag);
	if (rightsp != NULL)
		kif->kf_cap_rights = *rightsp;
	else
		cap_rights_init_zero(&kif->kf_cap_rights);
	kif->kf_fd = fd;
	kif->kf_ref_count = fp->f_count;
	kif->kf_offset = foffset_get(fp);

	/*
	 * This may drop the filedesc lock, so the 'fp' cannot be
	 * accessed after this call.
	 */
	error = fo_fill_kinfo(fp, kif, fdp);
	if (error == 0)
		kif->kf_status |= KF_ATTR_VALID;
	if ((flags & KERN_FILEDESC_PACK_KINFO) != 0)
		pack_kinfo(kif);
	else
		kif->kf_structsize = roundup2(sizeof(*kif), sizeof(uint64_t));
}

static void
export_vnode_to_kinfo(struct vnode *vp, int fd, int fflags,
    struct kinfo_file *kif, int flags)
{
	int error;

	bzero(kif, sizeof(*kif));

	kif->kf_type = KF_TYPE_VNODE;
	error = vn_fill_kinfo_vnode(vp, kif);
	if (error == 0)
		kif->kf_status |= KF_ATTR_VALID;
	kif->kf_flags = xlate_fflags(fflags);
	cap_rights_init_zero(&kif->kf_cap_rights);
	kif->kf_fd = fd;
	kif->kf_ref_count = -1;
	kif->kf_offset = -1;
	if ((flags & KERN_FILEDESC_PACK_KINFO) != 0)
		pack_kinfo(kif);
	else
		kif->kf_structsize = roundup2(sizeof(*kif), sizeof(uint64_t));
	vrele(vp);
}

struct export_fd_buf {
	struct filedesc		*fdp;
	struct sbuf 		*sb;
	ssize_t			remainder;
	struct kinfo_file	kif;
	int			flags;
};

static int
export_kinfo_to_sb(struct export_fd_buf *efbuf)
{
	struct kinfo_file *kif;

	kif = &efbuf->kif;
	if (efbuf->remainder != -1) {
		if (efbuf->remainder < kif->kf_structsize) {
			/* Terminate export. */
			efbuf->remainder = 0;
			return (0);
		}
		efbuf->remainder -= kif->kf_structsize;
	}
	return (sbuf_bcat(efbuf->sb, kif, kif->kf_structsize) == 0 ? 0 : ENOMEM);
}

static int
export_file_to_sb(struct file *fp, int fd, cap_rights_t *rightsp,
    struct export_fd_buf *efbuf)
{
	int error;

	if (efbuf->remainder == 0)
		return (0);
	export_file_to_kinfo(fp, fd, rightsp, &efbuf->kif, efbuf->fdp,
	    efbuf->flags);
	FILEDESC_SUNLOCK(efbuf->fdp);
	error = export_kinfo_to_sb(efbuf);
	FILEDESC_SLOCK(efbuf->fdp);
	return (error);
}

static int
export_vnode_to_sb(struct vnode *vp, int fd, int fflags,
    struct export_fd_buf *efbuf)
{
	int error;

	if (efbuf->remainder == 0)
		return (0);
	if (efbuf->fdp != NULL)
		FILEDESC_SUNLOCK(efbuf->fdp);
	export_vnode_to_kinfo(vp, fd, fflags, &efbuf->kif, efbuf->flags);
	error = export_kinfo_to_sb(efbuf);
	if (efbuf->fdp != NULL)
		FILEDESC_SLOCK(efbuf->fdp);
	return (error);
}

/*
 * Store a process file descriptor information to sbuf.
 *
 * Takes a locked proc as argument, and returns with the proc unlocked.
 */
int
kern_proc_filedesc_out(struct proc *p,  struct sbuf *sb, ssize_t maxlen,
    int flags)
{
	struct file *fp;
	struct filedesc *fdp;
	struct export_fd_buf *efbuf;
	struct vnode *cttyvp, *textvp, *tracevp;
	struct pwd *pwd;
	int error, i, lastfile;
	cap_rights_t rights;

	PROC_LOCK_ASSERT(p, MA_OWNED);

	/* ktrace vnode */
	tracevp = p->p_tracevp;
	if (tracevp != NULL)
		vrefact(tracevp);
	/* text vnode */
	textvp = p->p_textvp;
	if (textvp != NULL)
		vrefact(textvp);
	/* Controlling tty. */
	cttyvp = NULL;
	if (p->p_pgrp != NULL && p->p_pgrp->pg_session != NULL) {
		cttyvp = p->p_pgrp->pg_session->s_ttyvp;
		if (cttyvp != NULL)
			vrefact(cttyvp);
	}
	fdp = fdhold(p);
	PROC_UNLOCK(p);
	efbuf = malloc(sizeof(*efbuf), M_TEMP, M_WAITOK);
	efbuf->fdp = NULL;
	efbuf->sb = sb;
	efbuf->remainder = maxlen;
	efbuf->flags = flags;
	if (tracevp != NULL)
		export_vnode_to_sb(tracevp, KF_FD_TYPE_TRACE, FREAD | FWRITE,
		    efbuf);
	if (textvp != NULL)
		export_vnode_to_sb(textvp, KF_FD_TYPE_TEXT, FREAD, efbuf);
	if (cttyvp != NULL)
		export_vnode_to_sb(cttyvp, KF_FD_TYPE_CTTY, FREAD | FWRITE,
		    efbuf);
	error = 0;
	if (fdp == NULL)
		goto fail;
	efbuf->fdp = fdp;
	FILEDESC_SLOCK(fdp);
	pwd = pwd_hold_filedesc(fdp);
	if (pwd != NULL) {
		/* working directory */
		if (pwd->pwd_cdir != NULL) {
			vrefact(pwd->pwd_cdir);
			export_vnode_to_sb(pwd->pwd_cdir, KF_FD_TYPE_CWD, FREAD, efbuf);
		}
		/* root directory */
		if (pwd->pwd_rdir != NULL) {
			vrefact(pwd->pwd_rdir);
			export_vnode_to_sb(pwd->pwd_rdir, KF_FD_TYPE_ROOT, FREAD, efbuf);
		}
		/* jail directory */
		if (pwd->pwd_jdir != NULL) {
			vrefact(pwd->pwd_jdir);
			export_vnode_to_sb(pwd->pwd_jdir, KF_FD_TYPE_JAIL, FREAD, efbuf);
		}
		pwd_drop(pwd);
	}
	lastfile = fdlastfile(fdp);
	for (i = 0; fdp->fd_refcnt > 0 && i <= lastfile; i++) {
		if ((fp = fdp->fd_ofiles[i].fde_file) == NULL)
			continue;
#ifdef CAPABILITIES
		rights = *cap_rights(fdp, i);
#else /* !CAPABILITIES */
		rights = cap_no_rights;
#endif
		/*
		 * Create sysctl entry.  It is OK to drop the filedesc
		 * lock inside of export_file_to_sb() as we will
		 * re-validate and re-evaluate its properties when the
		 * loop continues.
		 */
		error = export_file_to_sb(fp, i, &rights, efbuf);
		if (error != 0 || efbuf->remainder == 0)
			break;
	}
	FILEDESC_SUNLOCK(fdp);
	fddrop(fdp);
fail:
	free(efbuf, M_TEMP);
	return (error);
}

#define FILEDESC_SBUF_SIZE	(sizeof(struct kinfo_file) * 5)

/*
 * Get per-process file descriptors for use by procstat(1), et al.
 */
static int
sysctl_kern_proc_filedesc(SYSCTL_HANDLER_ARGS)
{
	struct sbuf sb;
	struct proc *p;
	ssize_t maxlen;
	int error, error2, *name;

	name = (int *)arg1;

	sbuf_new_for_sysctl(&sb, NULL, FILEDESC_SBUF_SIZE, req);
	sbuf_clear_flags(&sb, SBUF_INCLUDENUL);
	error = pget((pid_t)name[0], PGET_CANDEBUG | PGET_NOTWEXIT, &p);
	if (error != 0) {
		sbuf_delete(&sb);
		return (error);
	}
	maxlen = req->oldptr != NULL ? req->oldlen : -1;
	error = kern_proc_filedesc_out(p, &sb, maxlen,
	    KERN_FILEDESC_PACK_KINFO);
	error2 = sbuf_finish(&sb);
	sbuf_delete(&sb);
	return (error != 0 ? error : error2);
}

#ifdef COMPAT_FREEBSD7
#ifdef KINFO_OFILE_SIZE
CTASSERT(sizeof(struct kinfo_ofile) == KINFO_OFILE_SIZE);
#endif

static void
kinfo_to_okinfo(struct kinfo_file *kif, struct kinfo_ofile *okif)
{

	okif->kf_structsize = sizeof(*okif);
	okif->kf_type = kif->kf_type;
	okif->kf_fd = kif->kf_fd;
	okif->kf_ref_count = kif->kf_ref_count;
	okif->kf_flags = kif->kf_flags & (KF_FLAG_READ | KF_FLAG_WRITE |
	    KF_FLAG_APPEND | KF_FLAG_ASYNC | KF_FLAG_FSYNC | KF_FLAG_NONBLOCK |
	    KF_FLAG_DIRECT | KF_FLAG_HASLOCK);
	okif->kf_offset = kif->kf_offset;
	if (kif->kf_type == KF_TYPE_VNODE)
		okif->kf_vnode_type = kif->kf_un.kf_file.kf_file_type;
	else
		okif->kf_vnode_type = KF_VTYPE_VNON;
	strlcpy(okif->kf_path, kif->kf_path, sizeof(okif->kf_path));
	if (kif->kf_type == KF_TYPE_SOCKET) {
		okif->kf_sock_domain = kif->kf_un.kf_sock.kf_sock_domain0;
		okif->kf_sock_type = kif->kf_un.kf_sock.kf_sock_type0;
		okif->kf_sock_protocol = kif->kf_un.kf_sock.kf_sock_protocol0;
		okif->kf_sa_local = kif->kf_un.kf_sock.kf_sa_local;
		okif->kf_sa_peer = kif->kf_un.kf_sock.kf_sa_peer;
	} else {
		okif->kf_sa_local.ss_family = AF_UNSPEC;
		okif->kf_sa_peer.ss_family = AF_UNSPEC;
	}
}

static int
export_vnode_for_osysctl(struct vnode *vp, int type, struct kinfo_file *kif,
    struct kinfo_ofile *okif, struct filedesc *fdp, struct sysctl_req *req)
{
	int error;

	vrefact(vp);
	FILEDESC_SUNLOCK(fdp);
	export_vnode_to_kinfo(vp, type, 0, kif, KERN_FILEDESC_PACK_KINFO);
	kinfo_to_okinfo(kif, okif);
	error = SYSCTL_OUT(req, okif, sizeof(*okif));
	FILEDESC_SLOCK(fdp);
	return (error);
}

/*
 * Get per-process file descriptors for use by procstat(1), et al.
 */
static int
sysctl_kern_proc_ofiledesc(SYSCTL_HANDLER_ARGS)
{
	struct kinfo_ofile *okif;
	struct kinfo_file *kif;
	struct filedesc *fdp;
	struct pwd *pwd;
	int error, i, lastfile, *name;
	struct file *fp;
	struct proc *p;

	name = (int *)arg1;
	error = pget((pid_t)name[0], PGET_CANDEBUG | PGET_NOTWEXIT, &p);
	if (error != 0)
		return (error);
	fdp = fdhold(p);
	PROC_UNLOCK(p);
	if (fdp == NULL)
		return (ENOENT);
	kif = malloc(sizeof(*kif), M_TEMP, M_WAITOK);
	okif = malloc(sizeof(*okif), M_TEMP, M_WAITOK);
	FILEDESC_SLOCK(fdp);
	pwd = pwd_hold_filedesc(fdp);
	if (pwd != NULL) {
		if (pwd->pwd_cdir != NULL)
			export_vnode_for_osysctl(pwd->pwd_cdir, KF_FD_TYPE_CWD, kif,
			    okif, fdp, req);
		if (pwd->pwd_rdir != NULL)
			export_vnode_for_osysctl(pwd->pwd_rdir, KF_FD_TYPE_ROOT, kif,
			    okif, fdp, req);
		if (pwd->pwd_jdir != NULL)
			export_vnode_for_osysctl(pwd->pwd_jdir, KF_FD_TYPE_JAIL, kif,
			    okif, fdp, req);
		pwd_drop(pwd);
	}
	lastfile = fdlastfile(fdp);
	for (i = 0; fdp->fd_refcnt > 0 && i <= lastfile; i++) {
		if ((fp = fdp->fd_ofiles[i].fde_file) == NULL)
			continue;
		export_file_to_kinfo(fp, i, NULL, kif, fdp,
		    KERN_FILEDESC_PACK_KINFO);
		FILEDESC_SUNLOCK(fdp);
		kinfo_to_okinfo(kif, okif);
		error = SYSCTL_OUT(req, okif, sizeof(*okif));
		FILEDESC_SLOCK(fdp);
		if (error)
			break;
	}
	FILEDESC_SUNLOCK(fdp);
	fddrop(fdp);
	free(kif, M_TEMP);
	free(okif, M_TEMP);
	return (0);
}

static SYSCTL_NODE(_kern_proc, KERN_PROC_OFILEDESC, ofiledesc,
    CTLFLAG_RD|CTLFLAG_MPSAFE, sysctl_kern_proc_ofiledesc,
    "Process ofiledesc entries");
#endif	/* COMPAT_FREEBSD7 */

int
vntype_to_kinfo(int vtype)
{
	struct {
		int	vtype;
		int	kf_vtype;
	} vtypes_table[] = {
		{ VBAD, KF_VTYPE_VBAD },
		{ VBLK, KF_VTYPE_VBLK },
		{ VCHR, KF_VTYPE_VCHR },
		{ VDIR, KF_VTYPE_VDIR },
		{ VFIFO, KF_VTYPE_VFIFO },
		{ VLNK, KF_VTYPE_VLNK },
		{ VNON, KF_VTYPE_VNON },
		{ VREG, KF_VTYPE_VREG },
		{ VSOCK, KF_VTYPE_VSOCK }
	};
	unsigned int i;

	/*
	 * Perform vtype translation.
	 */
	for (i = 0; i < nitems(vtypes_table); i++)
		if (vtypes_table[i].vtype == vtype)
			return (vtypes_table[i].kf_vtype);

	return (KF_VTYPE_UNKNOWN);
}

static SYSCTL_NODE(_kern_proc, KERN_PROC_FILEDESC, filedesc,
    CTLFLAG_RD|CTLFLAG_MPSAFE, sysctl_kern_proc_filedesc,
    "Process filedesc entries");

/*
 * Store a process current working directory information to sbuf.
 *
 * Takes a locked proc as argument, and returns with the proc unlocked.
 */
int
kern_proc_cwd_out(struct proc *p,  struct sbuf *sb, ssize_t maxlen)
{
	struct filedesc *fdp;
	struct pwd *pwd;
	struct export_fd_buf *efbuf;
	struct vnode *cdir;
	int error;

	PROC_LOCK_ASSERT(p, MA_OWNED);

	fdp = fdhold(p);
	PROC_UNLOCK(p);
	if (fdp == NULL)
		return (EINVAL);

	efbuf = malloc(sizeof(*efbuf), M_TEMP, M_WAITOK);
	efbuf->fdp = fdp;
	efbuf->sb = sb;
	efbuf->remainder = maxlen;

	FILEDESC_SLOCK(fdp);
	pwd = FILEDESC_LOCKED_LOAD_PWD(fdp);
	cdir = pwd->pwd_cdir;
	if (cdir == NULL) {
		error = EINVAL;
	} else {
		vrefact(cdir);
		error = export_vnode_to_sb(cdir, KF_FD_TYPE_CWD, FREAD, efbuf);
	}
	FILEDESC_SUNLOCK(fdp);
	fddrop(fdp);
	free(efbuf, M_TEMP);
	return (error);
}

/*
 * Get per-process current working directory.
 */
static int
sysctl_kern_proc_cwd(SYSCTL_HANDLER_ARGS)
{
	struct sbuf sb;
	struct proc *p;
	ssize_t maxlen;
	int error, error2, *name;

	name = (int *)arg1;

	sbuf_new_for_sysctl(&sb, NULL, sizeof(struct kinfo_file), req);
	sbuf_clear_flags(&sb, SBUF_INCLUDENUL);
	error = pget((pid_t)name[0], PGET_CANDEBUG | PGET_NOTWEXIT, &p);
	if (error != 0) {
		sbuf_delete(&sb);
		return (error);
	}
	maxlen = req->oldptr != NULL ? req->oldlen : -1;
	error = kern_proc_cwd_out(p, &sb, maxlen);
	error2 = sbuf_finish(&sb);
	sbuf_delete(&sb);
	return (error != 0 ? error : error2);
}

static SYSCTL_NODE(_kern_proc, KERN_PROC_CWD, cwd, CTLFLAG_RD|CTLFLAG_MPSAFE,
    sysctl_kern_proc_cwd, "Process current working directory");

#ifdef DDB
/*
 * For the purposes of debugging, generate a human-readable string for the
 * file type.
 */
static const char *
file_type_to_name(short type)
{

	switch (type) {
	case 0:
		return ("zero");
	case DTYPE_VNODE:
		return ("vnode");
	case DTYPE_SOCKET:
		return ("socket");
	case DTYPE_PIPE:
		return ("pipe");
	case DTYPE_FIFO:
		return ("fifo");
	case DTYPE_KQUEUE:
		return ("kqueue");
	case DTYPE_CRYPTO:
		return ("crypto");
	case DTYPE_MQUEUE:
		return ("mqueue");
	case DTYPE_SHM:
		return ("shm");
	case DTYPE_SEM:
		return ("ksem");
	case DTYPE_PTS:
		return ("pts");
	case DTYPE_DEV:
		return ("dev");
	case DTYPE_PROCDESC:
		return ("proc");
	case DTYPE_LINUXEFD:
		return ("levent");
	case DTYPE_LINUXTFD:
		return ("ltimer");
	default:
		return ("unkn");
	}
}

/*
 * For the purposes of debugging, identify a process (if any, perhaps one of
 * many) that references the passed file in its file descriptor array. Return
 * NULL if none.
 */
static struct proc *
file_to_first_proc(struct file *fp)
{
	struct filedesc *fdp;
	struct proc *p;
	int n;

	FOREACH_PROC_IN_SYSTEM(p) {
		if (p->p_state == PRS_NEW)
			continue;
		fdp = p->p_fd;
		if (fdp == NULL)
			continue;
		for (n = 0; n < fdp->fd_nfiles; n++) {
			if (fp == fdp->fd_ofiles[n].fde_file)
				return (p);
		}
	}
	return (NULL);
}

static void
db_print_file(struct file *fp, int header)
{
#define XPTRWIDTH ((int)howmany(sizeof(void *) * NBBY, 4))
	struct proc *p;

	if (header)
		db_printf("%*s %6s %*s %8s %4s %5s %6s %*s %5s %s\n",
		    XPTRWIDTH, "File", "Type", XPTRWIDTH, "Data", "Flag",
		    "GCFl", "Count", "MCount", XPTRWIDTH, "Vnode", "FPID",
		    "FCmd");
	p = file_to_first_proc(fp);
	db_printf("%*p %6s %*p %08x %04x %5d %6d %*p %5d %s\n", XPTRWIDTH,
	    fp, file_type_to_name(fp->f_type), XPTRWIDTH, fp->f_data,
	    fp->f_flag, 0, fp->f_count, 0, XPTRWIDTH, fp->f_vnode,
	    p != NULL ? p->p_pid : -1, p != NULL ? p->p_comm : "-");

#undef XPTRWIDTH
}

DB_SHOW_COMMAND(file, db_show_file)
{
	struct file *fp;

	if (!have_addr) {
		db_printf("usage: show file <addr>\n");
		return;
	}
	fp = (struct file *)addr;
	db_print_file(fp, 1);
}

DB_SHOW_COMMAND(files, db_show_files)
{
	struct filedesc *fdp;
	struct file *fp;
	struct proc *p;
	int header;
	int n;

	header = 1;
	FOREACH_PROC_IN_SYSTEM(p) {
		if (p->p_state == PRS_NEW)
			continue;
		if ((fdp = p->p_fd) == NULL)
			continue;
		for (n = 0; n < fdp->fd_nfiles; ++n) {
			if ((fp = fdp->fd_ofiles[n].fde_file) == NULL)
				continue;
			db_print_file(fp, header);
			header = 0;
		}
	}
}
#endif

SYSCTL_INT(_kern, KERN_MAXFILESPERPROC, maxfilesperproc, CTLFLAG_RW,
    &maxfilesperproc, 0, "Maximum files allowed open per process");

SYSCTL_INT(_kern, KERN_MAXFILES, maxfiles, CTLFLAG_RW,
    &maxfiles, 0, "Maximum number of files");

SYSCTL_INT(_kern, OID_AUTO, openfiles, CTLFLAG_RD,
    &openfiles, 0, "System-wide number of open files");

/* ARGSUSED*/
static void
filelistinit(void *dummy)
{

	file_zone = uma_zcreate("Files", sizeof(struct file), NULL, NULL,
	    NULL, NULL, UMA_ALIGN_PTR, UMA_ZONE_NOFREE);
	filedesc0_zone = uma_zcreate("filedesc0", sizeof(struct filedesc0),
	    NULL, NULL, NULL, NULL, UMA_ALIGN_PTR, 0);
	pwd_zone = uma_zcreate("PWD", sizeof(struct pwd), NULL, NULL,
	    NULL, NULL, UMA_ALIGN_PTR, UMA_ZONE_SMR);
	/*
	 * XXXMJG this is a temporary hack due to boot ordering issues against
	 * the vnode zone.
	 */
	vfs_smr = uma_zone_get_smr(pwd_zone);
	mtx_init(&sigio_lock, "sigio lock", NULL, MTX_DEF);
}
SYSINIT(select, SI_SUB_LOCK, SI_ORDER_FIRST, filelistinit, NULL);

/*-------------------------------------------------------------------*/

static int
badfo_readwrite(struct file *fp, struct uio *uio, struct ucred *active_cred,
    int flags, struct thread *td)
{

	return (EBADF);
}

static int
badfo_truncate(struct file *fp, off_t length, struct ucred *active_cred,
    struct thread *td)
{

	return (EINVAL);
}

static int
badfo_ioctl(struct file *fp, u_long com, void *data, struct ucred *active_cred,
    struct thread *td)
{

	return (EBADF);
}

static int
badfo_poll(struct file *fp, int events, struct ucred *active_cred,
    struct thread *td)
{

	return (0);
}

static int
badfo_kqfilter(struct file *fp, struct knote *kn)
{

	return (EBADF);
}

static int
badfo_stat(struct file *fp, struct stat *sb, struct ucred *active_cred,
    struct thread *td)
{

	return (EBADF);
}

static int
badfo_close(struct file *fp, struct thread *td)
{

	return (0);
}

static int
badfo_chmod(struct file *fp, mode_t mode, struct ucred *active_cred,
    struct thread *td)
{

	return (EBADF);
}

static int
badfo_chown(struct file *fp, uid_t uid, gid_t gid, struct ucred *active_cred,
    struct thread *td)
{

	return (EBADF);
}

static int
badfo_sendfile(struct file *fp, int sockfd, struct uio *hdr_uio,
    struct uio *trl_uio, off_t offset, size_t nbytes, off_t *sent, int flags,
    struct thread *td)
{

	return (EBADF);
}

static int
badfo_fill_kinfo(struct file *fp, struct kinfo_file *kif, struct filedesc *fdp)
{

	return (0);
}

struct fileops badfileops = {
	.fo_read = badfo_readwrite,
	.fo_write = badfo_readwrite,
	.fo_truncate = badfo_truncate,
	.fo_ioctl = badfo_ioctl,
	.fo_poll = badfo_poll,
	.fo_kqfilter = badfo_kqfilter,
	.fo_stat = badfo_stat,
	.fo_close = badfo_close,
	.fo_chmod = badfo_chmod,
	.fo_chown = badfo_chown,
	.fo_sendfile = badfo_sendfile,
	.fo_fill_kinfo = badfo_fill_kinfo,
};

int
invfo_rdwr(struct file *fp, struct uio *uio, struct ucred *active_cred,
    int flags, struct thread *td)
{

	return (EOPNOTSUPP);
}

int
invfo_truncate(struct file *fp, off_t length, struct ucred *active_cred,
    struct thread *td)
{

	return (EINVAL);
}

int
invfo_ioctl(struct file *fp, u_long com, void *data,
    struct ucred *active_cred, struct thread *td)
{

	return (ENOTTY);
}

int
invfo_poll(struct file *fp, int events, struct ucred *active_cred,
    struct thread *td)
{

	return (poll_no_poll(events));
}

int
invfo_kqfilter(struct file *fp, struct knote *kn)
{

	return (EINVAL);
}

int
invfo_chmod(struct file *fp, mode_t mode, struct ucred *active_cred,
    struct thread *td)
{

	return (EINVAL);
}

int
invfo_chown(struct file *fp, uid_t uid, gid_t gid, struct ucred *active_cred,
    struct thread *td)
{

	return (EINVAL);
}

int
invfo_sendfile(struct file *fp, int sockfd, struct uio *hdr_uio,
    struct uio *trl_uio, off_t offset, size_t nbytes, off_t *sent, int flags,
    struct thread *td)
{

	return (EINVAL);
}

/*-------------------------------------------------------------------*/

/*
 * File Descriptor pseudo-device driver (/dev/fd/).
 *
 * Opening minor device N dup()s the file (if any) connected to file
 * descriptor N belonging to the calling process.  Note that this driver
 * consists of only the ``open()'' routine, because all subsequent
 * references to this file will be direct to the other driver.
 *
 * XXX: we could give this one a cloning event handler if necessary.
 */

/* ARGSUSED */
static int
fdopen(struct cdev *dev, int mode, int type, struct thread *td)
{

	/*
	 * XXX Kludge: set curthread->td_dupfd to contain the value of the
	 * the file descriptor being sought for duplication. The error
	 * return ensures that the vnode for this device will be released
	 * by vn_open. Open will detect this special error and take the
	 * actions in dupfdopen below. Other callers of vn_open or VOP_OPEN
	 * will simply report the error.
	 */
	td->td_dupfd = dev2unit(dev);
	return (ENODEV);
}

static struct cdevsw fildesc_cdevsw = {
	.d_version =	D_VERSION,
	.d_open =	fdopen,
	.d_name =	"FD",
};

static void
fildesc_drvinit(void *unused)
{
	struct cdev *dev;

	dev = make_dev_credf(MAKEDEV_ETERNAL, &fildesc_cdevsw, 0, NULL,
	    UID_ROOT, GID_WHEEL, 0666, "fd/0");
	make_dev_alias(dev, "stdin");
	dev = make_dev_credf(MAKEDEV_ETERNAL, &fildesc_cdevsw, 1, NULL,
	    UID_ROOT, GID_WHEEL, 0666, "fd/1");
	make_dev_alias(dev, "stdout");
	dev = make_dev_credf(MAKEDEV_ETERNAL, &fildesc_cdevsw, 2, NULL,
	    UID_ROOT, GID_WHEEL, 0666, "fd/2");
	make_dev_alias(dev, "stderr");
}

SYSINIT(fildescdev, SI_SUB_DRIVERS, SI_ORDER_MIDDLE, fildesc_drvinit, NULL);