summaryrefslogtreecommitdiff
path: root/sys/kern/vfs_vnops.c
blob: 3bbb3b9fdcee008fe9b6c689586bd4c4e4925b47 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
/*	$NetBSD: vfs_vnops.c,v 1.241 2023/04/22 13:53:02 riastradh Exp $	*/

/*-
 * Copyright (c) 2009 The NetBSD Foundation, Inc.
 * All rights reserved.
 *
 * This code is derived from software contributed to The NetBSD Foundation
 * by Andrew Doran.
 *
 * 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.
 *
 * THIS SOFTWARE IS PROVIDED BY THE NETBSD FOUNDATION, INC. 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 FOUNDATION 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.
 */

/*
 * Copyright (c) 1982, 1986, 1989, 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.
 *
 *	@(#)vfs_vnops.c	8.14 (Berkeley) 6/15/95
 */

#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: vfs_vnops.c,v 1.241 2023/04/22 13:53:02 riastradh Exp $");

#include "veriexec.h"

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/file.h>
#include <sys/stat.h>
#include <sys/buf.h>
#include <sys/proc.h>
#include <sys/mount.h>
#include <sys/namei.h>
#include <sys/vnode_impl.h>
#include <sys/ioctl.h>
#include <sys/tty.h>
#include <sys/poll.h>
#include <sys/kauth.h>
#include <sys/syslog.h>
#include <sys/fstrans.h>
#include <sys/atomic.h>
#include <sys/filedesc.h>
#include <sys/wapbl.h>
#include <sys/mman.h>

#include <miscfs/specfs/specdev.h>
#include <miscfs/fifofs/fifo.h>

#include <uvm/uvm_extern.h>
#include <uvm/uvm_readahead.h>
#include <uvm/uvm_device.h>

#ifdef UNION
#include <fs/union/union.h>
#endif

#ifndef COMPAT_ZERODEV
#define COMPAT_ZERODEV(dev)	(0)
#endif

int (*vn_union_readdir_hook)(struct vnode **, struct file *, struct lwp *);

#include <sys/verified_exec.h>

static int vn_read(file_t *fp, off_t *offset, struct uio *uio,
    kauth_cred_t cred, int flags);
static int vn_write(file_t *fp, off_t *offset, struct uio *uio,
    kauth_cred_t cred, int flags);
static int vn_closefile(file_t *fp);
static int vn_poll(file_t *fp, int events);
static int vn_fcntl(file_t *fp, u_int com, void *data);
static int vn_statfile(file_t *fp, struct stat *sb);
static int vn_ioctl(file_t *fp, u_long com, void *data);
static int vn_mmap(struct file *, off_t *, size_t, int, int *, int *,
    struct uvm_object **, int *);
static int vn_seek(struct file *, off_t, int, off_t *, int);
static int vn_advlock(struct file *, void *, int, struct flock *, int);
static int vn_fpathconf(struct file *, int, register_t *);
static int vn_posix_fadvise(struct file *, off_t, off_t, int);

const struct fileops vnops = {
	.fo_name = "vn",
	.fo_read = vn_read,
	.fo_write = vn_write,
	.fo_ioctl = vn_ioctl,
	.fo_fcntl = vn_fcntl,
	.fo_poll = vn_poll,
	.fo_stat = vn_statfile,
	.fo_close = vn_closefile,
	.fo_kqfilter = vn_kqfilter,
	.fo_restart = fnullop_restart,
	.fo_mmap = vn_mmap,
	.fo_seek = vn_seek,
	.fo_advlock = vn_advlock,
	.fo_fpathconf = vn_fpathconf,
	.fo_posix_fadvise = vn_posix_fadvise,
};

/*
 * Common code for vnode open operations.
 * Check permissions, and call the VOP_OPEN or VOP_CREATE routine.
 *
 * at_dvp is the directory for openat(), if any.
 * pb is the path.
 * nmode is additional namei flags, restricted to TRYEMULROOT and NOCHROOT.
 * fmode is the open flags, converted from O_* to F*
 * cmode is the creation file permissions.
 *
 * XXX shouldn't cmode be mode_t?
 *
 * On success produces either a locked vnode in *ret_vp, or NULL in
 * *ret_vp and a file descriptor number in *ret_fd.
 *
 * The caller may pass NULL for ret_fd (and ret_domove), in which case
 * EOPNOTSUPP will be produced in the cases that would otherwise return
 * a file descriptor.
 *
 * Note that callers that want no-follow behavior should pass
 * O_NOFOLLOW in fmode. Neither FOLLOW nor NOFOLLOW in nmode is
 * honored.
 */
int
vn_open(struct vnode *at_dvp, struct pathbuf *pb,
	int nmode, int fmode, int cmode,
	struct vnode **ret_vp, bool *ret_domove, int *ret_fd)
{
	struct nameidata nd;
	struct vnode *vp = NULL;
	struct lwp *l = curlwp;
	kauth_cred_t cred = l->l_cred;
	struct vattr va;
	int error;
	const char *pathstring;

	KASSERT((nmode & (TRYEMULROOT | NOCHROOT)) == nmode);

	KASSERT(ret_vp != NULL);
	KASSERT((ret_domove == NULL) == (ret_fd == NULL));

	if ((fmode & (O_CREAT | O_DIRECTORY)) == (O_CREAT | O_DIRECTORY))
		return EINVAL;

	NDINIT(&nd, LOOKUP, nmode, pb);
	if (at_dvp != NULL)
		NDAT(&nd, at_dvp);

	nd.ni_cnd.cn_flags &= TRYEMULROOT | NOCHROOT;

	if (fmode & O_CREAT) {
		nd.ni_cnd.cn_nameiop = CREATE;
		nd.ni_cnd.cn_flags |= LOCKPARENT | LOCKLEAF;
		if ((fmode & O_EXCL) == 0 &&
		    ((fmode & O_NOFOLLOW) == 0))
			nd.ni_cnd.cn_flags |= FOLLOW;
		if ((fmode & O_EXCL) == 0)
			nd.ni_cnd.cn_flags |= NONEXCLHACK;
	} else {
		nd.ni_cnd.cn_nameiop = LOOKUP;
		nd.ni_cnd.cn_flags |= LOCKLEAF;
		if ((fmode & O_NOFOLLOW) == 0)
			nd.ni_cnd.cn_flags |= FOLLOW;
	}

	pathstring = pathbuf_stringcopy_get(nd.ni_pathbuf);
	if (pathstring == NULL) {
		return ENOMEM;
	}

	/*
	 * When this "interface" was exposed to do_open() it used
	 * to initialize l_dupfd to -newfd-1 (thus passing in the
	 * new file handle number to use)... but nothing in the
	 * kernel uses that value. So just send 0.
	 */
	l->l_dupfd = 0;

	error = namei(&nd);
	if (error)
		goto out;

	vp = nd.ni_vp;

#if NVERIEXEC > 0
	error = veriexec_openchk(l, nd.ni_vp, pathstring, fmode);
	if (error) {
		/* We have to release the locks ourselves */
		/*
		 * 20210604 dholland passing NONEXCLHACK means we can
		 * get ni_dvp == NULL back if ni_vp exists, and we should
		 * treat that like the non-O_CREAT case.
		 */
		if ((fmode & O_CREAT) != 0 && nd.ni_dvp != NULL) {
			if (vp == NULL) {
				vput(nd.ni_dvp);
			} else {
				VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
				if (nd.ni_dvp == nd.ni_vp)
					vrele(nd.ni_dvp);
				else
					vput(nd.ni_dvp);
				nd.ni_dvp = NULL;
				vput(vp);
				vp = NULL;
			}
		} else {
			vput(vp);
			vp = NULL;
		}
		goto out;
	}
#endif /* NVERIEXEC > 0 */

	/*
	 * 20210604 dholland ditto
	 */
	if ((fmode & O_CREAT) != 0 && nd.ni_dvp != NULL) {
		if (nd.ni_vp == NULL) {
			vattr_null(&va);
			va.va_type = VREG;
			va.va_mode = cmode;
			if (fmode & O_EXCL)
				 va.va_vaflags |= VA_EXCLUSIVE;
			error = VOP_CREATE(nd.ni_dvp, &nd.ni_vp,
					   &nd.ni_cnd, &va);
			if (error) {
				vput(nd.ni_dvp);
				goto out;
			}
			fmode &= ~O_TRUNC;
			vp = nd.ni_vp;
			vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
			vput(nd.ni_dvp);
		} else {
			VOP_ABORTOP(nd.ni_dvp, &nd.ni_cnd);
			if (nd.ni_dvp == nd.ni_vp)
				vrele(nd.ni_dvp);
			else
				vput(nd.ni_dvp);
			nd.ni_dvp = NULL;
			vp = nd.ni_vp;
			if (fmode & O_EXCL) {
				error = EEXIST;
				goto bad;
			}
			fmode &= ~O_CREAT;
		}
	} else if ((fmode & O_CREAT) != 0) {
		/*
		 * 20210606 dholland passing NONEXCLHACK means this
		 * case exists; it is the same as the following one
		 * but also needs to do things in the second (exists)
		 * half of the following block. (Besides handle
		 * ni_dvp, anyway.)
		 */
		vp = nd.ni_vp;
		KASSERT((fmode & O_EXCL) == 0);
		fmode &= ~O_CREAT;
	} else {
		vp = nd.ni_vp;
	}
	if (vp->v_type == VSOCK) {
		error = EOPNOTSUPP;
		goto bad;
	}
	if (nd.ni_vp->v_type == VLNK) {
		error = EFTYPE;
		goto bad;
	}

	if ((fmode & O_CREAT) == 0) {
		error = vn_openchk(vp, cred, fmode);
		if (error != 0)
			goto bad;
	}

	if (fmode & O_TRUNC) {
		vattr_null(&va);
		va.va_size = 0;
		error = VOP_SETATTR(vp, &va, cred);
		if (error != 0)
			goto bad;
	}
	if ((error = VOP_OPEN(vp, fmode, cred)) != 0)
		goto bad;
	if (fmode & FWRITE) {
		mutex_enter(vp->v_interlock);
		vp->v_writecount++;
		mutex_exit(vp->v_interlock);
	}

bad:
	if (error) {
		vput(vp);
		vp = NULL;
	}
out:
	pathbuf_stringcopy_put(nd.ni_pathbuf, pathstring);

	switch (error) {
	case EDUPFD:
	case EMOVEFD:
		/* if the caller isn't prepared to handle fds, fail for them */
		if (ret_fd == NULL) {
			error = EOPNOTSUPP;
			break;
		}
		*ret_vp = NULL;
		*ret_domove = error == EMOVEFD;
		*ret_fd = l->l_dupfd;
		error = 0;
		break;
	case 0:
		KASSERT(VOP_ISLOCKED(vp) == LK_EXCLUSIVE);
		*ret_vp = vp;
		break;
	}
	l->l_dupfd = 0;
	return error;
}

/*
 * Check for write permissions on the specified vnode.
 * Prototype text segments cannot be written.
 */
int
vn_writechk(struct vnode *vp)
{

	/*
	 * If the vnode is in use as a process's text,
	 * we can't allow writing.
	 */
	if (vp->v_iflag & VI_TEXT)
		return ETXTBSY;
	return 0;
}

int
vn_openchk(struct vnode *vp, kauth_cred_t cred, int fflags)
{
	int permbits = 0;
	int error;

	if (vp->v_type == VNON || vp->v_type == VBAD)
		return ENXIO;

	if ((fflags & O_DIRECTORY) != 0 && vp->v_type != VDIR)
		return ENOTDIR;

	if ((fflags & O_REGULAR) != 0 && vp->v_type != VREG)
		return EFTYPE;

	if ((fflags & FREAD) != 0) {
		permbits = VREAD;
	}
	if ((fflags & FEXEC) != 0) {
		permbits |= VEXEC;
	}
	if ((fflags & (FWRITE | O_TRUNC)) != 0) {
		permbits |= VWRITE;
		if (vp->v_type == VDIR) {
			error = EISDIR;
			goto bad;
		}
		error = vn_writechk(vp);
		if (error != 0)
			goto bad;
	}
	error = VOP_ACCESS(vp, permbits, cred);
bad:
	return error;
}

/*
 * Mark a vnode as having executable mappings.
 */
void
vn_markexec(struct vnode *vp)
{

	if ((vp->v_iflag & VI_EXECMAP) != 0) {
		/* Safe unlocked, as long as caller holds a reference. */
		return;
	}

	rw_enter(vp->v_uobj.vmobjlock, RW_WRITER);
	mutex_enter(vp->v_interlock);
	if ((vp->v_iflag & VI_EXECMAP) == 0) {
		cpu_count(CPU_COUNT_EXECPAGES, vp->v_uobj.uo_npages);
		vp->v_iflag |= VI_EXECMAP;
	}
	mutex_exit(vp->v_interlock);
	rw_exit(vp->v_uobj.vmobjlock);
}

/*
 * Mark a vnode as being the text of a process.
 * Fail if the vnode is currently writable.
 */
int
vn_marktext(struct vnode *vp)
{

	if ((vp->v_iflag & (VI_TEXT|VI_EXECMAP)) == (VI_TEXT|VI_EXECMAP)) {
		/* Safe unlocked, as long as caller holds a reference. */
		return 0;
	}

	rw_enter(vp->v_uobj.vmobjlock, RW_WRITER);
	mutex_enter(vp->v_interlock);
	if (vp->v_writecount != 0) {
		KASSERT((vp->v_iflag & VI_TEXT) == 0);
		mutex_exit(vp->v_interlock);
		rw_exit(vp->v_uobj.vmobjlock);
		return ETXTBSY;
	}
	if ((vp->v_iflag & VI_EXECMAP) == 0) {
		cpu_count(CPU_COUNT_EXECPAGES, vp->v_uobj.uo_npages);
	}
	vp->v_iflag |= (VI_TEXT | VI_EXECMAP);
	mutex_exit(vp->v_interlock);
	rw_exit(vp->v_uobj.vmobjlock);
	return 0;
}

/*
 * Vnode close call
 *
 * Note: takes an unlocked vnode, while VOP_CLOSE takes a locked node.
 */
int
vn_close(struct vnode *vp, int flags, kauth_cred_t cred)
{
	int error;

	vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	if (flags & FWRITE) {
		mutex_enter(vp->v_interlock);
		KASSERT(vp->v_writecount > 0);
		vp->v_writecount--;
		mutex_exit(vp->v_interlock);
	}
	error = VOP_CLOSE(vp, flags, cred);
	vput(vp);
	return error;
}

static int
enforce_rlimit_fsize(struct vnode *vp, struct uio *uio, int ioflag)
{
	struct lwp *l = curlwp;
	off_t testoff;

	if (uio->uio_rw != UIO_WRITE || vp->v_type != VREG)
		return 0;

	KASSERT(VOP_ISLOCKED(vp) == LK_EXCLUSIVE);
	if (ioflag & IO_APPEND)
		testoff = vp->v_size;
	else
		testoff = uio->uio_offset;

	if (testoff + uio->uio_resid >
	    l->l_proc->p_rlimit[RLIMIT_FSIZE].rlim_cur) {
		mutex_enter(&proc_lock);
		psignal(l->l_proc, SIGXFSZ);
		mutex_exit(&proc_lock);
		return EFBIG;
	}

	return 0;
}

/*
 * Package up an I/O request on a vnode into a uio and do it.
 */
int
vn_rdwr(enum uio_rw rw, struct vnode *vp, void *base, int len, off_t offset,
    enum uio_seg segflg, int ioflg, kauth_cred_t cred, size_t *aresid,
    struct lwp *l)
{
	struct uio auio;
	struct iovec aiov;
	int error;

	if ((ioflg & IO_NODELOCKED) == 0) {
		if (rw == UIO_READ) {
			vn_lock(vp, LK_SHARED | LK_RETRY);
		} else /* UIO_WRITE */ {
			vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
		}
	}
	auio.uio_iov = &aiov;
	auio.uio_iovcnt = 1;
	aiov.iov_base = base;
	aiov.iov_len = len;
	auio.uio_resid = len;
	auio.uio_offset = offset;
	auio.uio_rw = rw;
	if (segflg == UIO_SYSSPACE) {
		UIO_SETUP_SYSSPACE(&auio);
	} else {
		auio.uio_vmspace = l->l_proc->p_vmspace;
	}

	if ((error = enforce_rlimit_fsize(vp, &auio, ioflg)) != 0)
		goto out;

	if (rw == UIO_READ) {
		error = VOP_READ(vp, &auio, ioflg, cred);
	} else {
		error = VOP_WRITE(vp, &auio, ioflg, cred);
	}

	if (aresid)
		*aresid = auio.uio_resid;
	else
		if (auio.uio_resid && error == 0)
			error = EIO;

 out:
	if ((ioflg & IO_NODELOCKED) == 0) {
		VOP_UNLOCK(vp);
	}
	return error;
}

int
vn_readdir(file_t *fp, char *bf, int segflg, u_int count, int *done,
    struct lwp *l, off_t **cookies, int *ncookies)
{
	struct vnode *vp = fp->f_vnode;
	struct iovec aiov;
	struct uio auio;
	int error, eofflag;

	/* Limit the size on any kernel buffers used by VOP_READDIR */
	count = uimin(MAXBSIZE, count);

unionread:
	if (vp->v_type != VDIR)
		return EINVAL;
	aiov.iov_base = bf;
	aiov.iov_len = count;
	auio.uio_iov = &aiov;
	auio.uio_iovcnt = 1;
	auio.uio_rw = UIO_READ;
	if (segflg == UIO_SYSSPACE) {
		UIO_SETUP_SYSSPACE(&auio);
	} else {
		KASSERT(l == curlwp);
		auio.uio_vmspace = l->l_proc->p_vmspace;
	}
	auio.uio_resid = count;
	vn_lock(vp, LK_SHARED | LK_RETRY);
	mutex_enter(&fp->f_lock);
	auio.uio_offset = fp->f_offset;
	mutex_exit(&fp->f_lock);
	error = VOP_READDIR(vp, &auio, fp->f_cred, &eofflag, cookies,
	    ncookies);
	mutex_enter(&fp->f_lock);
	fp->f_offset = auio.uio_offset;
	mutex_exit(&fp->f_lock);
	VOP_UNLOCK(vp);
	if (error)
		return error;

	if (count == auio.uio_resid && vn_union_readdir_hook) {
		struct vnode *ovp = vp;

		error = (*vn_union_readdir_hook)(&vp, fp, l);
		if (error)
			return error;
		if (vp != ovp)
			goto unionread;
	}

	if (count == auio.uio_resid && (vp->v_vflag & VV_ROOT) &&
	    (vp->v_mount->mnt_flag & MNT_UNION)) {
		struct vnode *tvp = vp;
		vp = vp->v_mount->mnt_vnodecovered;
		vref(vp);
		mutex_enter(&fp->f_lock);
		fp->f_vnode = vp;
		fp->f_offset = 0;
		mutex_exit(&fp->f_lock);
		vrele(tvp);
		goto unionread;
	}
	*done = count - auio.uio_resid;
	return error;
}

/*
 * File table vnode read routine.
 */
static int
vn_read(file_t *fp, off_t *offset, struct uio *uio, kauth_cred_t cred,
    int flags)
{
	struct vnode *vp = fp->f_vnode;
	int error, ioflag, fflag;
	size_t count;

	ioflag = IO_ADV_ENCODE(fp->f_advice);
	fflag = fp->f_flag;
	if (fflag & FNONBLOCK)
		ioflag |= IO_NDELAY;
	if ((fflag & (FFSYNC | FRSYNC)) == (FFSYNC | FRSYNC))
		ioflag |= IO_SYNC;
	if (fflag & FALTIO)
		ioflag |= IO_ALTSEMANTICS;
	if (fflag & FDIRECT)
		ioflag |= IO_DIRECT;
	if (offset == &fp->f_offset && (flags & FOF_UPDATE_OFFSET) != 0)
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	else
		vn_lock(vp, LK_SHARED | LK_RETRY);
	if (__predict_false(vp->v_type == VDIR) &&
	    offset == &fp->f_offset && (flags & FOF_UPDATE_OFFSET) == 0)
		mutex_enter(&fp->f_lock);
	uio->uio_offset = *offset;
	if (__predict_false(vp->v_type == VDIR) &&
	    offset == &fp->f_offset && (flags & FOF_UPDATE_OFFSET) == 0)
		mutex_enter(&fp->f_lock);
	count = uio->uio_resid;
	error = VOP_READ(vp, uio, ioflag, cred);
	if (flags & FOF_UPDATE_OFFSET)
		*offset += count - uio->uio_resid;
	VOP_UNLOCK(vp);
	return error;
}

/*
 * File table vnode write routine.
 */
static int
vn_write(file_t *fp, off_t *offset, struct uio *uio, kauth_cred_t cred,
    int flags)
{
	struct vnode *vp = fp->f_vnode;
	int error, ioflag, fflag;
	size_t count;

	ioflag = IO_ADV_ENCODE(fp->f_advice) | IO_UNIT;
	fflag = fp->f_flag;
	if (vp->v_type == VREG && (fflag & O_APPEND))
		ioflag |= IO_APPEND;
	if (fflag & FNONBLOCK)
		ioflag |= IO_NDELAY;
	if (fflag & FFSYNC ||
	    (vp->v_mount && (vp->v_mount->mnt_flag & MNT_SYNCHRONOUS)))
		ioflag |= IO_SYNC;
	else if (fflag & FDSYNC)
		ioflag |= IO_DSYNC;
	if (fflag & FALTIO)
		ioflag |= IO_ALTSEMANTICS;
	if (fflag & FDIRECT)
		ioflag |= IO_DIRECT;
	vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	uio->uio_offset = *offset;
	count = uio->uio_resid;

	if ((error = enforce_rlimit_fsize(vp, uio, ioflag)) != 0)
		goto out;

	error = VOP_WRITE(vp, uio, ioflag, cred);

	if (flags & FOF_UPDATE_OFFSET) {
		if (ioflag & IO_APPEND) {
			/*
			 * SUSv3 describes behaviour for count = 0 as following:
			 * "Before any action ... is taken, and if nbyte is zero
			 * and the file is a regular file, the write() function
			 * ... in the absence of errors ... shall return zero
			 * and have no other results."
			 */ 
			if (count)
				*offset = uio->uio_offset;
		} else
			*offset += count - uio->uio_resid;
	}

 out:
	VOP_UNLOCK(vp);
	return error;
}

/*
 * File table vnode stat routine.
 */
static int
vn_statfile(file_t *fp, struct stat *sb)
{
	struct vnode *vp = fp->f_vnode;
	int error;

	vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	error = vn_stat(vp, sb);
	VOP_UNLOCK(vp);
	return error;
}

int
vn_stat(struct vnode *vp, struct stat *sb)
{
	struct vattr va;
	int error;
	mode_t mode;

	memset(&va, 0, sizeof(va));
	error = VOP_GETATTR(vp, &va, kauth_cred_get());
	if (error)
		return error;
	/*
	 * Copy from vattr table
	 */
	memset(sb, 0, sizeof(*sb));
	sb->st_dev = va.va_fsid;
	sb->st_ino = va.va_fileid;
	mode = va.va_mode;
	switch (vp->v_type) {
	case VREG:
		mode |= S_IFREG;
		break;
	case VDIR:
		mode |= S_IFDIR;
		break;
	case VBLK:
		mode |= S_IFBLK;
		break;
	case VCHR:
		mode |= S_IFCHR;
		break;
	case VLNK:
		mode |= S_IFLNK;
		break;
	case VSOCK:
		mode |= S_IFSOCK;
		break;
	case VFIFO:
		mode |= S_IFIFO;
		break;
	default:
		return EBADF;
	}
	sb->st_mode = mode;
	sb->st_nlink = va.va_nlink;
	sb->st_uid = va.va_uid;
	sb->st_gid = va.va_gid;
	sb->st_rdev = va.va_rdev;
	sb->st_size = va.va_size;
	sb->st_atimespec = va.va_atime;
	sb->st_mtimespec = va.va_mtime;
	sb->st_ctimespec = va.va_ctime;
	sb->st_birthtimespec = va.va_birthtime;
	sb->st_blksize = va.va_blocksize;
	sb->st_flags = va.va_flags;
	sb->st_gen = 0;
	sb->st_blocks = va.va_bytes / S_BLKSIZE;
	return 0;
}

/*
 * File table vnode fcntl routine.
 */
static int
vn_fcntl(file_t *fp, u_int com, void *data)
{
	struct vnode *vp = fp->f_vnode;
	int error;

	error = VOP_FCNTL(vp, com, data, fp->f_flag, kauth_cred_get());
	return error;
}

/*
 * File table vnode ioctl routine.
 */
static int
vn_ioctl(file_t *fp, u_long com, void *data)
{
	struct vnode *vp = fp->f_vnode, *ovp;
	struct vattr vattr;
	int error;

	switch (vp->v_type) {

	case VREG:
	case VDIR:
		if (com == FIONREAD) {
			vn_lock(vp, LK_SHARED | LK_RETRY);
			error = VOP_GETATTR(vp, &vattr, kauth_cred_get());
			if (error == 0) {
				if (vp->v_type == VDIR)
					mutex_enter(&fp->f_lock);
				*(int *)data = vattr.va_size - fp->f_offset;
				if (vp->v_type == VDIR)
					mutex_exit(&fp->f_lock);
			}
			VOP_UNLOCK(vp);
			if (error)
				return error;
			return 0;
		}
		if ((com == FIONWRITE) || (com == FIONSPACE)) {
			/*
			 * Files don't have send queues, so there never
			 * are any bytes in them, nor is there any
			 * open space in them.
			 */
			*(int *)data = 0;
			return 0;
		}
		if (com == FIOGETBMAP) {
			daddr_t *block;

			if (*(daddr_t *)data < 0)
				return EINVAL;
			block = (daddr_t *)data;
			vn_lock(vp, LK_SHARED | LK_RETRY);
			error = VOP_BMAP(vp, *block, NULL, block, NULL);
			VOP_UNLOCK(vp);
			return error;
		}
		if (com == OFIOGETBMAP) {
			daddr_t ibn, obn;

			if (*(int32_t *)data < 0)
				return EINVAL;
			ibn = (daddr_t)*(int32_t *)data;
			vn_lock(vp, LK_SHARED | LK_RETRY);
			error = VOP_BMAP(vp, ibn, NULL, &obn, NULL);
			VOP_UNLOCK(vp);
			*(int32_t *)data = (int32_t)obn;
			return error;
		}
		if (com == FIONBIO || com == FIOASYNC)	/* XXX */
			return 0;			/* XXX */
		/* FALLTHROUGH */
	case VFIFO:
	case VCHR:
	case VBLK:
		error = VOP_IOCTL(vp, com, data, fp->f_flag,
		    kauth_cred_get());
		if (error == 0 && com == TIOCSCTTY) {
			vref(vp);
			mutex_enter(&proc_lock);
			ovp = curproc->p_session->s_ttyvp;
			curproc->p_session->s_ttyvp = vp;
			mutex_exit(&proc_lock);
			if (ovp != NULL)
				vrele(ovp);
		}
		return error;

	default:
		return EPASSTHROUGH;
	}
}

/*
 * File table vnode poll routine.
 */
static int
vn_poll(file_t *fp, int events)
{

	return VOP_POLL(fp->f_vnode, events);
}

/*
 * File table vnode kqfilter routine.
 */
int
vn_kqfilter(file_t *fp, struct knote *kn)
{

	return VOP_KQFILTER(fp->f_vnode, kn);
}

static int
vn_mmap(struct file *fp, off_t *offp, size_t size, int prot, int *flagsp,
    int *advicep, struct uvm_object **uobjp, int *maxprotp)
{
	struct uvm_object *uobj;
	struct vnode *vp;
	struct vattr va;
	struct lwp *l;
	vm_prot_t maxprot;
	off_t off;
	int error, flags;
	bool needwritemap;

	l = curlwp;

	off = *offp;
	flags = *flagsp;
	maxprot = VM_PROT_EXECUTE;

	KASSERT(size > 0);

	vp = fp->f_vnode;
	if (vp->v_type != VREG && vp->v_type != VCHR &&
	    vp->v_type != VBLK) {
		/* only REG/CHR/BLK support mmap */
		return ENODEV;
	}
	if (vp->v_type != VCHR && off < 0) {
		return EINVAL;
	}
#if SIZE_MAX > UINT32_MAX	/* XXX -Wtype-limits */
	if (vp->v_type != VCHR && size > __type_max(off_t)) {
		return EOVERFLOW;
	}
#endif
	if (vp->v_type != VCHR && off > __type_max(off_t) - size) {
		/* no offset wrapping */
		return EOVERFLOW;
	}

	/* special case: catch SunOS style /dev/zero */
	if (vp->v_type == VCHR &&
	    (vp->v_rdev == zerodev || COMPAT_ZERODEV(vp->v_rdev))) {
		*uobjp = NULL;
		*maxprotp = VM_PROT_ALL;
		return 0;
	}

	/*
	 * Old programs may not select a specific sharing type, so
	 * default to an appropriate one.
	 *
	 * XXX: how does MAP_ANON fit in the picture?
	 */
	if ((flags & (MAP_SHARED|MAP_PRIVATE)) == 0) {
#if defined(DEBUG)
		struct proc *p = l->l_proc;
		printf("WARNING: defaulted mmap() share type to "
		       "%s (pid %d command %s)\n", vp->v_type == VCHR ?
		       "MAP_SHARED" : "MAP_PRIVATE", p->p_pid,
		       p->p_comm);
#endif
		if (vp->v_type == VCHR)
			flags |= MAP_SHARED;	/* for a device */
		else
			flags |= MAP_PRIVATE;	/* for a file */
	}

	/*
	 * MAP_PRIVATE device mappings don't make sense (and aren't
	 * supported anyway).  However, some programs rely on this,
	 * so just change it to MAP_SHARED.
	 */
	if (vp->v_type == VCHR && (flags & MAP_PRIVATE) != 0) {
		flags = (flags & ~MAP_PRIVATE) | MAP_SHARED;
	}

	/*
	 * now check protection
	 */

	/* check read access */
	if (fp->f_flag & FREAD)
		maxprot |= VM_PROT_READ;
	else if (prot & PROT_READ) {
		return EACCES;
	}

	/* check write access, shared case first */
	if (flags & MAP_SHARED) {
		/*
		 * if the file is writable, only add PROT_WRITE to
		 * maxprot if the file is not immutable, append-only.
		 * otherwise, if we have asked for PROT_WRITE, return
		 * EPERM.
		 */
		if (fp->f_flag & FWRITE) {
			vn_lock(vp, LK_SHARED | LK_RETRY);
			error = VOP_GETATTR(vp, &va, l->l_cred);
			VOP_UNLOCK(vp);
			if (error) {
				return error;
			}
			if ((va.va_flags &
			     (SF_SNAPSHOT|IMMUTABLE|APPEND)) == 0)
				maxprot |= VM_PROT_WRITE;
			else if (prot & PROT_WRITE) {
				return EPERM;
			}
		} else if (prot & PROT_WRITE) {
			return EACCES;
		}
	} else {
		/* MAP_PRIVATE mappings can always write to */
		maxprot |= VM_PROT_WRITE;
	}

	/*
	 * Don't allow mmap for EXEC if the file system
	 * is mounted NOEXEC.
	 */
	if ((prot & PROT_EXEC) != 0 &&
	    (vp->v_mount->mnt_flag & MNT_NOEXEC) != 0) {
		return EACCES;
	}

	if (vp->v_type != VCHR) {
		error = VOP_MMAP(vp, prot, curlwp->l_cred);
		if (error) {
			return error;
		}
		vref(vp);
		uobj = &vp->v_uobj;

		/*
		 * If the vnode is being mapped with PROT_EXEC,
		 * then mark it as text.
		 */
		if (prot & PROT_EXEC) {
			vn_markexec(vp);
		}
	} else {
		int i = maxprot;

		/*
		 * XXX Some devices don't like to be mapped with
		 * XXX PROT_EXEC or PROT_WRITE, but we don't really
		 * XXX have a better way of handling this, right now
		 */
		do {
			uobj = udv_attach(vp->v_rdev,
					  (flags & MAP_SHARED) ? i :
					  (i & ~VM_PROT_WRITE), off, size);
			i--;
		} while ((uobj == NULL) && (i > 0));
		if (uobj == NULL) {
			return EINVAL;
		}
		*advicep = UVM_ADV_RANDOM;
	}

	/*
	 * Set vnode flags to indicate the new kinds of mapping.
	 * We take the vnode lock in exclusive mode here to serialize
	 * with direct I/O.
	 *
	 * Safe to check for these flag values without a lock, as
	 * long as a reference to the vnode is held.
	 */
	needwritemap = (vp->v_iflag & VI_WRMAP) == 0 &&
		(flags & MAP_SHARED) != 0 &&
		(maxprot & VM_PROT_WRITE) != 0;
	if ((vp->v_vflag & VV_MAPPED) == 0 || needwritemap) {
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
		vp->v_vflag |= VV_MAPPED;
		if (needwritemap) {
			rw_enter(vp->v_uobj.vmobjlock, RW_WRITER);
			mutex_enter(vp->v_interlock);
			vp->v_iflag |= VI_WRMAP;
			mutex_exit(vp->v_interlock);
			rw_exit(vp->v_uobj.vmobjlock);
		}
		VOP_UNLOCK(vp);
	}

#if NVERIEXEC > 0

	/*
	 * Check if the file can be executed indirectly.
	 *
	 * XXX: This gives false warnings about "Incorrect access type"
	 * XXX: if the mapping is not executable. Harmless, but will be
	 * XXX: fixed as part of other changes.
	 */
	if (veriexec_verify(l, vp, "(mmap)", VERIEXEC_INDIRECT,
			    NULL)) {

		/*
		 * Don't allow executable mappings if we can't
		 * indirectly execute the file.
		 */
		if (prot & VM_PROT_EXECUTE) {
			return EPERM;
		}

		/*
		 * Strip the executable bit from 'maxprot' to make sure
		 * it can't be made executable later.
		 */
		maxprot &= ~VM_PROT_EXECUTE;
	}
#endif /* NVERIEXEC > 0 */

	*uobjp = uobj;
	*maxprotp = maxprot;
	*flagsp = flags;

	return 0;
}

static int
vn_seek(struct file *fp, off_t delta, int whence, off_t *newoffp,
    int flags)
{
	const off_t OFF_MIN = __type_min(off_t);
	const off_t OFF_MAX = __type_max(off_t);
	kauth_cred_t cred = fp->f_cred;
	off_t oldoff, newoff;
	struct vnode *vp = fp->f_vnode;
	struct vattr vattr;
	int error;

	if (vp->v_type == VFIFO)
		return ESPIPE;

	if (flags & FOF_UPDATE_OFFSET)
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	else
		vn_lock(vp, LK_SHARED | LK_RETRY);

	/* Compute the old and new offsets.  */
	if (vp->v_type == VDIR && (flags & FOF_UPDATE_OFFSET) == 0)
		mutex_enter(&fp->f_lock);
	oldoff = fp->f_offset;
	if (vp->v_type == VDIR && (flags & FOF_UPDATE_OFFSET) == 0)
		mutex_exit(&fp->f_lock);
	switch (whence) {
	case SEEK_CUR:
		if (delta > 0) {
			if (oldoff > 0 && delta > OFF_MAX - oldoff) {
				newoff = OFF_MAX;
				break;
			}
		} else {
			if (oldoff < 0 && delta < OFF_MIN - oldoff) {
				newoff = OFF_MIN;
				break;
			}
		}
		newoff = oldoff + delta;
		break;
	case SEEK_END:
		error = VOP_GETATTR(vp, &vattr, cred);
		if (error)
			goto out;
		if (vattr.va_size > OFF_MAX ||
		    delta > OFF_MAX - (off_t)vattr.va_size) {
			newoff = OFF_MAX;
			break;
		}
		newoff = delta + vattr.va_size;
		break;
	case SEEK_SET:
		newoff = delta;
		break;
	default:
		error = EINVAL;
		goto out;
	}

	/* Pass the proposed change to the file system to audit.  */
	error = VOP_SEEK(vp, oldoff, newoff, cred);
	if (error)
		goto out;

	/* Success!  */
	if (newoffp)
		*newoffp = newoff;
	if (flags & FOF_UPDATE_OFFSET)
		fp->f_offset = newoff;
	error = 0;

out:	VOP_UNLOCK(vp);
	return error;
}

static int
vn_advlock(struct file *fp, void *id, int op, struct flock *fl,
    int flags)
{
	struct vnode *const vp = fp->f_vnode;

	if (fl->l_whence == SEEK_CUR) {
		vn_lock(vp, LK_SHARED | LK_RETRY);
		fl->l_start += fp->f_offset;
		VOP_UNLOCK(vp);
	}

	return VOP_ADVLOCK(vp, id, op, fl, flags);
}

static int
vn_fpathconf(struct file *fp, int name, register_t *retval)
{
	struct vnode *const vp = fp->f_vnode;
	int error;

	vn_lock(vp, LK_SHARED | LK_RETRY);
	error = VOP_PATHCONF(vp, name, retval);
	VOP_UNLOCK(vp);

	return error;
}

static int
vn_posix_fadvise(struct file *fp, off_t offset, off_t len, int advice)
{
	const off_t OFF_MAX = __type_max(off_t);
	struct vnode *vp = fp->f_vnode;
	off_t endoffset;
	int error;

	if (offset < 0) {
		return EINVAL;
	}
	if (len == 0) {
		endoffset = OFF_MAX;
	} else if (len > 0 && (OFF_MAX - offset) >= len) {
		endoffset = offset + len;
	} else {
		return EINVAL;
	}

	CTASSERT(POSIX_FADV_NORMAL == UVM_ADV_NORMAL);
	CTASSERT(POSIX_FADV_RANDOM == UVM_ADV_RANDOM);
	CTASSERT(POSIX_FADV_SEQUENTIAL == UVM_ADV_SEQUENTIAL);

	switch (advice) {
	case POSIX_FADV_WILLNEED:
	case POSIX_FADV_DONTNEED:
		if (vp->v_type != VREG && vp->v_type != VBLK)
			return 0;
		break;
	}

	switch (advice) {
	case POSIX_FADV_NORMAL:
	case POSIX_FADV_RANDOM:
	case POSIX_FADV_SEQUENTIAL:
		/*
		 * We ignore offset and size.  Must lock the file to
		 * do this, as f_advice is sub-word sized.
		 */
		mutex_enter(&fp->f_lock);
		fp->f_advice = (u_char)advice;
		mutex_exit(&fp->f_lock);
		error = 0;
		break;

	case POSIX_FADV_WILLNEED:
		error = uvm_readahead(&vp->v_uobj, offset, endoffset - offset);
		break;

	case POSIX_FADV_DONTNEED:
		/*
		 * Align the region to page boundaries as VOP_PUTPAGES expects
		 * by shrinking it.  We shrink instead of expand because we
		 * do not want to deactivate cache outside of the requested
		 * region.  It means that if the specified region is smaller
		 * than PAGE_SIZE, we do nothing.
		 */
		if (offset <= trunc_page(OFF_MAX) &&
		    round_page(offset) < trunc_page(endoffset)) {
			rw_enter(vp->v_uobj.vmobjlock, RW_WRITER);
			error = VOP_PUTPAGES(vp,
			    round_page(offset), trunc_page(endoffset),
			    PGO_DEACTIVATE | PGO_CLEANIT);
		} else {
			error = 0;
		}
		break;

	case POSIX_FADV_NOREUSE:
		/* Not implemented yet. */
		error = 0;
		break;
	default:
		error = EINVAL;
		break;
	}

	return error;
}

/*
 * Check that the vnode is still valid, and if so
 * acquire requested lock.
 */
int
vn_lock(struct vnode *vp, int flags)
{
	struct lwp *l;
	int error;

	KASSERT(vrefcnt(vp) > 0);
	KASSERT((flags & ~(LK_SHARED|LK_EXCLUSIVE|LK_NOWAIT|LK_RETRY|
	    LK_UPGRADE|LK_DOWNGRADE)) == 0);
	KASSERT((flags & LK_NOWAIT) != 0 || !mutex_owned(vp->v_interlock));

#ifdef DIAGNOSTIC
	if (wapbl_vphaswapbl(vp))
		WAPBL_JUNLOCK_ASSERT(wapbl_vptomp(vp));
#endif

	/* Get a more useful report for lockstat. */
	l = curlwp;
	KASSERT(l->l_rwcallsite == 0);
	l->l_rwcallsite = (uintptr_t)__builtin_return_address(0);	

	error = VOP_LOCK(vp, flags);

	l->l_rwcallsite = 0;

	switch (flags & (LK_RETRY | LK_NOWAIT)) {
	case 0:
		KASSERT(error == 0 || error == ENOENT);
		break;
	case LK_RETRY:
		KASSERT(error == 0);
		break;
	case LK_NOWAIT:
		KASSERT(error == 0 || error == EBUSY || error == ENOENT);
		break;
	case LK_RETRY | LK_NOWAIT:
		KASSERT(error == 0 || error == EBUSY);
		break;
	}

	return error;
}

/*
 * File table vnode close routine.
 */
static int
vn_closefile(file_t *fp)
{

	return vn_close(fp->f_vnode, fp->f_flag, fp->f_cred);
}

/*
 * Simplified in-kernel wrapper calls for extended attribute access.
 * Both calls pass in a NULL credential, authorizing a "kernel" access.
 * Set IO_NODELOCKED in ioflg if the vnode is already locked.
 */
int
vn_extattr_get(struct vnode *vp, int ioflg, int attrnamespace,
    const char *attrname, size_t *buflen, void *bf, struct lwp *l)
{
	struct uio auio;
	struct iovec aiov;
	int error;

	aiov.iov_len = *buflen;
	aiov.iov_base = bf;

	auio.uio_iov = &aiov;
	auio.uio_iovcnt = 1;
	auio.uio_rw = UIO_READ;
	auio.uio_offset = 0;
	auio.uio_resid = *buflen;
	UIO_SETUP_SYSSPACE(&auio);

	if ((ioflg & IO_NODELOCKED) == 0)
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);

	error = VOP_GETEXTATTR(vp, attrnamespace, attrname, &auio, NULL,
	    NOCRED);

	if ((ioflg & IO_NODELOCKED) == 0)
		VOP_UNLOCK(vp);

	if (error == 0)
		*buflen = *buflen - auio.uio_resid;

	return error;
}

/*
 * XXX Failure mode if partially written?
 */
int
vn_extattr_set(struct vnode *vp, int ioflg, int attrnamespace,
    const char *attrname, size_t buflen, const void *bf, struct lwp *l)
{
	struct uio auio;
	struct iovec aiov;
	int error;

	aiov.iov_len = buflen;
	aiov.iov_base = __UNCONST(bf);		/* XXXUNCONST kills const */

	auio.uio_iov = &aiov;
	auio.uio_iovcnt = 1;
	auio.uio_rw = UIO_WRITE;
	auio.uio_offset = 0;
	auio.uio_resid = buflen;
	UIO_SETUP_SYSSPACE(&auio);

	if ((ioflg & IO_NODELOCKED) == 0) {
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	}

	error = VOP_SETEXTATTR(vp, attrnamespace, attrname, &auio, NOCRED);

	if ((ioflg & IO_NODELOCKED) == 0) {
		VOP_UNLOCK(vp);
	}

	return error;
}

int
vn_extattr_rm(struct vnode *vp, int ioflg, int attrnamespace,
    const char *attrname, struct lwp *l)
{
	int error;

	if ((ioflg & IO_NODELOCKED) == 0) {
		vn_lock(vp, LK_EXCLUSIVE | LK_RETRY);
	}

	error = VOP_DELETEEXTATTR(vp, attrnamespace, attrname, NOCRED);
	if (error == EOPNOTSUPP)
		error = VOP_SETEXTATTR(vp, attrnamespace, attrname, NULL,
		    NOCRED);

	if ((ioflg & IO_NODELOCKED) == 0) {
		VOP_UNLOCK(vp);
	}

	return error;
}

int
vn_fifo_bypass(void *v)
{
	struct vop_generic_args *ap = v;

	return VOCALL(fifo_vnodeop_p, ap->a_desc->vdesc_offset, v);
}

/*
 * Open block device by device number
 */
int
vn_bdev_open(dev_t dev, struct vnode **vpp, struct lwp *l)
{
	int     error;

	if ((error = bdevvp(dev, vpp)) != 0)
		return error;

	vn_lock(*vpp, LK_EXCLUSIVE | LK_RETRY);
	if ((error = VOP_OPEN(*vpp, FREAD | FWRITE, l->l_cred)) != 0) {
		vput(*vpp);
		return error;
	}
	mutex_enter((*vpp)->v_interlock);
	(*vpp)->v_writecount++;
	mutex_exit((*vpp)->v_interlock);
	VOP_UNLOCK(*vpp);

	return 0;
}

/*
 * Lookup the provided name in the filesystem.  If the file exists,
 * is a valid block device, and isn't being used by anyone else,
 * set *vpp to the file's vnode.
 */
int
vn_bdev_openpath(struct pathbuf *pb, struct vnode **vpp, struct lwp *l)
{
	struct vnode *vp;
	dev_t dev;
	enum vtype vt;
	int     error;

	error = vn_open(NULL, pb, 0, FREAD | FWRITE, 0, &vp, NULL, NULL);
	if (error != 0)
		return error;

	dev = vp->v_rdev;
	vt = vp->v_type;

	VOP_UNLOCK(vp);
	(void) vn_close(vp, FREAD | FWRITE, l->l_cred);

	if (vt != VBLK)
		return ENOTBLK;

	return vn_bdev_open(dev, vpp, l);
}

static long
vn_knote_to_interest(const struct knote *kn)
{
	switch (kn->kn_filter) {
	case EVFILT_READ:
		/*
		 * Writing to the file or changing its attributes can
		 * set the file size, which impacts the readability
		 * filter.
		 *
		 * (No need to set NOTE_EXTEND here; it's only ever
		 * send with other hints; see vnode_if.c.)
		 */
		return NOTE_WRITE | NOTE_ATTRIB;

	case EVFILT_VNODE:
		return kn->kn_sfflags;

	case EVFILT_WRITE:
	default:
		return 0;
	}
}

void
vn_knote_attach(struct vnode *vp, struct knote *kn)
{
	struct vnode_klist *vk = vp->v_klist;
	long interest = 0;

	/*
	 * In the case of layered / stacked file systems, knotes
	 * should only ever be associated with the base vnode.
	 */
	KASSERT(kn->kn_hook == vp);
	KASSERT(vp->v_klist == &VNODE_TO_VIMPL(vp)->vi_klist);

	/*
	 * We maintain a bitmask of the kevents that there is interest in,
	 * to minimize the impact of having watchers.  It's silly to have
	 * to traverse vn_klist every time a read or write happens simply
	 * because there is someone interested in knowing when the file
	 * is deleted, for example.
	 */

	mutex_enter(vp->v_interlock);
	SLIST_INSERT_HEAD(&vk->vk_klist, kn, kn_selnext);
	SLIST_FOREACH(kn, &vk->vk_klist, kn_selnext) {
		interest |= vn_knote_to_interest(kn);
	}
	vk->vk_interest = interest;
	mutex_exit(vp->v_interlock);
}

void
vn_knote_detach(struct vnode *vp, struct knote *kn)
{
	struct vnode_klist *vk = vp->v_klist;
	long interest = 0;

	/* See above. */
	KASSERT(kn->kn_hook == vp);
	KASSERT(vp->v_klist == &VNODE_TO_VIMPL(vp)->vi_klist);

	/*
	 * We special case removing the head of the list, because:
	 *
	 * 1. It's extremely likely that we're detaching the only
	 *    knote.
	 *
	 * 2. We're already traversing the whole list, so we don't
	 *    want to use the generic SLIST_REMOVE() which would
	 *    traverse it *again*.
	 */

	mutex_enter(vp->v_interlock);
	if (__predict_true(kn == SLIST_FIRST(&vk->vk_klist))) {
		SLIST_REMOVE_HEAD(&vk->vk_klist, kn_selnext);
		SLIST_FOREACH(kn, &vk->vk_klist, kn_selnext) {
			interest |= vn_knote_to_interest(kn);
		}
		vk->vk_interest = interest;
	} else {
		struct knote *thiskn, *nextkn, *prevkn = NULL;

		SLIST_FOREACH_SAFE(thiskn, &vk->vk_klist, kn_selnext, nextkn) {
			if (thiskn == kn) {
				KASSERT(kn != NULL);
				KASSERT(prevkn != NULL);
				SLIST_REMOVE_AFTER(prevkn, kn_selnext);
				kn = NULL;
			} else {
				interest |= vn_knote_to_interest(thiskn);
				prevkn = thiskn;
			}
		}
		vk->vk_interest = interest;
	}
	mutex_exit(vp->v_interlock);
}