Training courses

Kernel and Embedded Linux

Bootlin training courses

Embedded Linux, kernel,
Yocto Project, Buildroot, real-time,
graphics, boot time, debugging...

Bootlin logo

Elixir Cross Referencer

   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
/*	$NetBSD: vfs_lookup.c,v 1.212.2.1 2021/06/21 14:50:57 martin Exp $	*/

/*
 * 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_lookup.c	8.10 (Berkeley) 5/27/95
 */

#include <sys/cdefs.h>
__KERNEL_RCSID(0, "$NetBSD: vfs_lookup.c,v 1.212.2.1 2021/06/21 14:50:57 martin Exp $");

#ifdef _KERNEL_OPT
#include "opt_magiclinks.h"
#endif

#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kernel.h>
#include <sys/syslimits.h>
#include <sys/time.h>
#include <sys/namei.h>
#include <sys/vnode.h>
#include <sys/mount.h>
#include <sys/errno.h>
#include <sys/filedesc.h>
#include <sys/hash.h>
#include <sys/proc.h>
#include <sys/syslog.h>
#include <sys/kauth.h>
#include <sys/ktrace.h>
#include <sys/dirent.h>

#ifndef MAGICLINKS
#define MAGICLINKS 0
#endif

int vfs_magiclinks = MAGICLINKS;

__CTASSERT(MAXNAMLEN == NAME_MAX);

/*
 * Substitute replacement text for 'magic' strings in symlinks.
 * Returns 0 if successful, and returns non-zero if an error
 * occurs.  (Currently, the only possible error is running out
 * of temporary pathname space.)
 *
 * Looks for "@<string>" and "@<string>/", where <string> is a
 * recognized 'magic' string.  Replaces the "@<string>" with the
 * appropriate replacement text.  (Note that in some cases the
 * replacement text may have zero length.)
 *
 * This would have been table driven, but the variance in
 * replacement strings (and replacement string lengths) made
 * that impractical.
 */
#define	VNL(x)							\
	(sizeof(x) - 1)

#define	VO	'{'
#define	VC	'}'

#define	MATCH(str)						\
	((termchar == '/' && i + VNL(str) == *len) ||		\
	 (i + VNL(str) < *len &&				\
	  cp[i + VNL(str)] == termchar)) &&			\
	!strncmp((str), &cp[i], VNL(str))

#define	SUBSTITUTE(m, s, sl)					\
	if ((newlen + (sl)) >= MAXPATHLEN)			\
		return 1;					\
	i += VNL(m);						\
	if (termchar != '/')					\
		i++;						\
	(void)memcpy(&tmp[newlen], (s), (sl));			\
	newlen += (sl);						\
	change = 1;						\
	termchar = '/';

static int
symlink_magic(struct proc *p, char *cp, size_t *len)
{
	char *tmp;
	size_t change, i, newlen, slen;
	char termchar = '/';
	char idtmp[11]; /* enough for 32 bit *unsigned* integer */


	tmp = PNBUF_GET();
	for (change = i = newlen = 0; i < *len; ) {
		if (cp[i] != '@') {
			tmp[newlen++] = cp[i++];
			continue;
		}

		i++;

		/* Check for @{var} syntax. */
		if (cp[i] == VO) {
			termchar = VC;
			i++;
		}

		/*
		 * The following checks should be ordered according
		 * to frequency of use.
		 */
		if (MATCH("machine_arch")) {
			slen = VNL(MACHINE_ARCH);
			SUBSTITUTE("machine_arch", MACHINE_ARCH, slen);
		} else if (MATCH("machine")) {
			slen = VNL(MACHINE);
			SUBSTITUTE("machine", MACHINE, slen);
		} else if (MATCH("hostname")) {
			SUBSTITUTE("hostname", hostname, hostnamelen);
		} else if (MATCH("osrelease")) {
			slen = strlen(osrelease);
			SUBSTITUTE("osrelease", osrelease, slen);
		} else if (MATCH("emul")) {
			slen = strlen(p->p_emul->e_name);
			SUBSTITUTE("emul", p->p_emul->e_name, slen);
		} else if (MATCH("kernel_ident")) {
			slen = strlen(kernel_ident);
			SUBSTITUTE("kernel_ident", kernel_ident, slen);
		} else if (MATCH("domainname")) {
			SUBSTITUTE("domainname", domainname, domainnamelen);
		} else if (MATCH("ostype")) {
			slen = strlen(ostype);
			SUBSTITUTE("ostype", ostype, slen);
		} else if (MATCH("uid")) {
			slen = snprintf(idtmp, sizeof(idtmp), "%u",
			    kauth_cred_geteuid(kauth_cred_get()));
			SUBSTITUTE("uid", idtmp, slen);
		} else if (MATCH("ruid")) {
			slen = snprintf(idtmp, sizeof(idtmp), "%u",
			    kauth_cred_getuid(kauth_cred_get()));
			SUBSTITUTE("ruid", idtmp, slen);
		} else if (MATCH("gid")) {
			slen = snprintf(idtmp, sizeof(idtmp), "%u",
			    kauth_cred_getegid(kauth_cred_get()));
			SUBSTITUTE("gid", idtmp, slen);
		} else if (MATCH("rgid")) {
			slen = snprintf(idtmp, sizeof(idtmp), "%u",
			    kauth_cred_getgid(kauth_cred_get()));
			SUBSTITUTE("rgid", idtmp, slen);
		} else {
			tmp[newlen++] = '@';
			if (termchar == VC)
				tmp[newlen++] = VO;
		}
	}

	if (change) {
		(void)memcpy(cp, tmp, newlen);
		*len = newlen;
	}
	PNBUF_PUT(tmp);

	return 0;
}

#undef VNL
#undef VO
#undef VC
#undef MATCH
#undef SUBSTITUTE

////////////////////////////////////////////////////////////

/*
 * Determine the namei hash (for the namecache) for name.
 * If *ep != NULL, hash from name to ep-1.
 * If *ep == NULL, hash from name until the first NUL or '/', and
 * return the location of this termination character in *ep.
 *
 * This function returns an equivalent hash to the MI hash32_strn().
 * The latter isn't used because in the *ep == NULL case, determining
 * the length of the string to the first NUL or `/' and then calling
 * hash32_strn() involves unnecessary double-handling of the data.
 */
uint32_t
namei_hash(const char *name, const char **ep)
{
	uint32_t	hash;

	hash = HASH32_STR_INIT;
	if (*ep != NULL) {
		for (; name < *ep; name++)
			hash = hash * 33 + *(const uint8_t *)name;
	} else {
		for (; *name != '\0' && *name != '/'; name++)
			hash = hash * 33 + *(const uint8_t *)name;
		*ep = name;
	}
	return (hash + (hash >> 5));
}

/*
 * Find the end of the first path component in NAME and return its
 * length.
 */
static size_t
namei_getcomponent(const char *name)
{
	size_t pos;

	pos = 0;
	while (name[pos] != '\0' && name[pos] != '/') {
		pos++;
	}
	return pos;
}

////////////////////////////////////////////////////////////

/*
 * Sealed abstraction for pathnames.
 *
 * System-call-layer level code that is going to call namei should
 * first create a pathbuf and adjust all the bells and whistles on it
 * as needed by context.
 */

struct pathbuf {
	char *pb_path;
	char *pb_pathcopy;
	unsigned pb_pathcopyuses;
};

static struct pathbuf *
pathbuf_create_raw(void)
{
	struct pathbuf *pb;

	pb = kmem_alloc(sizeof(*pb), KM_SLEEP);
	pb->pb_path = PNBUF_GET();
	if (pb->pb_path == NULL) {
		kmem_free(pb, sizeof(*pb));
		return NULL;
	}
	pb->pb_pathcopy = NULL;
	pb->pb_pathcopyuses = 0;
	return pb;
}

void
pathbuf_destroy(struct pathbuf *pb)
{
	KASSERT(pb->pb_pathcopyuses == 0);
	KASSERT(pb->pb_pathcopy == NULL);
	PNBUF_PUT(pb->pb_path);
	kmem_free(pb, sizeof(*pb));
}

struct pathbuf *
pathbuf_assimilate(char *pnbuf)
{
	struct pathbuf *pb;

	pb = kmem_alloc(sizeof(*pb), KM_SLEEP);
	pb->pb_path = pnbuf;
	pb->pb_pathcopy = NULL;
	pb->pb_pathcopyuses = 0;
	return pb;
}

struct pathbuf *
pathbuf_create(const char *path)
{
	struct pathbuf *pb;
	int error;

	pb = pathbuf_create_raw();
	if (pb == NULL) {
		return NULL;
	}
	error = copystr(path, pb->pb_path, PATH_MAX, NULL);
	if (error != 0) {
		KASSERT(!"kernel path too long in pathbuf_create");
		/* make sure it's null-terminated, just in case */
		pb->pb_path[PATH_MAX-1] = '\0';
	}
	return pb;
}

int
pathbuf_copyin(const char *userpath, struct pathbuf **ret)
{
	struct pathbuf *pb;
	int error;

	pb = pathbuf_create_raw();
	if (pb == NULL) {
		return ENOMEM;
	}
	error = copyinstr(userpath, pb->pb_path, PATH_MAX, NULL);
	if (error) {
		pathbuf_destroy(pb);
		return error;
	}
	*ret = pb;
	return 0;
}

/*
 * XXX should not exist:
 *   1. whether a pointer is kernel or user should be statically checkable.
 *   2. copyin should be handled by the upper part of the syscall layer,
 *      not in here.
 */
int
pathbuf_maybe_copyin(const char *path, enum uio_seg seg, struct pathbuf **ret)
{
	if (seg == UIO_USERSPACE) {
		return pathbuf_copyin(path, ret);
	} else {
		*ret = pathbuf_create(path);
		if (*ret == NULL) {
			return ENOMEM;
		}
		return 0;
	}
}

/*
 * Get a copy of the path buffer as it currently exists. If this is
 * called after namei starts the results may be arbitrary.
 */
void
pathbuf_copystring(const struct pathbuf *pb, char *buf, size_t maxlen)
{
	strlcpy(buf, pb->pb_path, maxlen);
}

/*
 * These two functions allow access to a saved copy of the original
 * path string. The first copy should be gotten before namei is
 * called. Each copy that is gotten should be put back.
 */

const char *
pathbuf_stringcopy_get(struct pathbuf *pb)
{
	if (pb->pb_pathcopyuses == 0) {
		pb->pb_pathcopy = PNBUF_GET();
		strcpy(pb->pb_pathcopy, pb->pb_path);
	}
	pb->pb_pathcopyuses++;
	return pb->pb_pathcopy;
}

void
pathbuf_stringcopy_put(struct pathbuf *pb, const char *str)
{
	KASSERT(str == pb->pb_pathcopy);
	KASSERT(pb->pb_pathcopyuses > 0);
	pb->pb_pathcopyuses--;
	if (pb->pb_pathcopyuses == 0) {
		PNBUF_PUT(pb->pb_pathcopy);
		pb->pb_pathcopy = NULL;
	}
}


////////////////////////////////////////////////////////////

/*
 * namei: convert a pathname into a pointer to a (maybe-locked) vnode,
 * and maybe also its parent directory vnode, and assorted other guff.
 * See namei(9) for the interface documentation.
 *
 *
 * The FOLLOW flag is set when symbolic links are to be followed
 * when they occur at the end of the name translation process.
 * Symbolic links are always followed for all other pathname
 * components other than the last.
 *
 * The segflg defines whether the name is to be copied from user
 * space or kernel space.
 *
 * Overall outline of namei:
 *
 *	copy in name
 *	get starting directory
 *	while (!done && !error) {
 *		call lookup to search path.
 *		if symbolic link, massage name in buffer and continue
 *	}
 */

/*
 * Search a pathname.
 * This is a very central and rather complicated routine.
 *
 * The pathname is pointed to by ni_ptr and is of length ni_pathlen.
 * The starting directory is passed in. The pathname is descended
 * until done, or a symbolic link is encountered. The variable ni_more
 * is clear if the path is completed; it is set to one if a symbolic
 * link needing interpretation is encountered.
 *
 * The flag argument is LOOKUP, CREATE, RENAME, or DELETE depending on
 * whether the name is to be looked up, created, renamed, or deleted.
 * When CREATE, RENAME, or DELETE is specified, information usable in
 * creating, renaming, or deleting a directory entry may be calculated.
 * If flag has LOCKPARENT or'ed into it, the parent directory is returned
 * locked.  Otherwise the parent directory is not returned. If the target
 * of the pathname exists and LOCKLEAF is or'ed into the flag the target
 * is returned locked, otherwise it is returned unlocked.  When creating
 * or renaming and LOCKPARENT is specified, the target may not be ".".
 * When deleting and LOCKPARENT is specified, the target may be ".".
 *
 * Overall outline of lookup:
 *
 * dirloop:
 *	identify next component of name at ndp->ni_ptr
 *	handle degenerate case where name is null string
 *	if .. and crossing mount points and on mounted filesys, find parent
 *	call VOP_LOOKUP routine for next component name
 *	    directory vnode returned in ni_dvp, locked.
 *	    component vnode returned in ni_vp (if it exists), locked.
 *	if result vnode is mounted on and crossing mount points,
 *	    find mounted on vnode
 *	if more components of name, do next level at dirloop
 *	return the answer in ni_vp, locked if LOCKLEAF set
 *	    if LOCKPARENT set, return locked parent in ni_dvp
 */


/*
 * Internal state for a namei operation.
 *
 * cnp is always equal to &ndp->ni_cnp.
 */
struct namei_state {
	struct nameidata *ndp;
	struct componentname *cnp;

	int docache;			/* == 0 do not cache last component */
	int rdonly;			/* lookup read-only flag bit */
	int slashes;

	unsigned attempt_retry:1;	/* true if error allows emul retry */
	unsigned root_referenced:1;	/* true if ndp->ni_rootdir and
					     ndp->ni_erootdir were referenced */
};


/*
 * Initialize the namei working state.
 */
static void
namei_init(struct namei_state *state, struct nameidata *ndp)
{

	state->ndp = ndp;
	state->cnp = &ndp->ni_cnd;

	state->docache = 0;
	state->rdonly = 0;
	state->slashes = 0;

	state->root_referenced = 0;

	KASSERTMSG((state->cnp->cn_cred != NULL), "namei: bad cred/proc");
	KASSERTMSG(((state->cnp->cn_nameiop & (~OPMASK)) == 0),
	    "namei: nameiop contaminated with flags: %08"PRIx32,
	    state->cnp->cn_nameiop);
	KASSERTMSG(((state->cnp->cn_flags & OPMASK) == 0),
	    "name: flags contaminated with nameiops: %08"PRIx32,
	    state->cnp->cn_flags);

	/*
	 * The buffer for name translation shall be the one inside the
	 * pathbuf.
	 */
	state->ndp->ni_pnbuf = state->ndp->ni_pathbuf->pb_path;
}

/*
 * Clean up the working namei state, leaving things ready for return
 * from namei.
 */
static void
namei_cleanup(struct namei_state *state)
{
	KASSERT(state->cnp == &state->ndp->ni_cnd);

	if (state->root_referenced) {
		if (state->ndp->ni_rootdir != NULL)
			vrele(state->ndp->ni_rootdir);
		if (state->ndp->ni_erootdir != NULL)
			vrele(state->ndp->ni_erootdir);
	}
}

//////////////////////////////

/*
 * Get the directory context.
 * Initializes the rootdir and erootdir state and returns a reference
 * to the starting dir.
 */
static struct vnode *
namei_getstartdir(struct namei_state *state)
{
	struct nameidata *ndp = state->ndp;
	struct componentname *cnp = state->cnp;
	struct cwdinfo *cwdi;		/* pointer to cwd state */
	struct lwp *self = curlwp;	/* thread doing namei() */
	struct vnode *rootdir, *erootdir, *curdir, *startdir;

	if (state->root_referenced) {
		if (state->ndp->ni_rootdir != NULL)
			vrele(state->ndp->ni_rootdir);
		if (state->ndp->ni_erootdir != NULL)
			vrele(state->ndp->ni_erootdir);
		state->root_referenced = 0;
	}

	cwdi = self->l_proc->p_cwdi;
	rw_enter(&cwdi->cwdi_lock, RW_READER);

	/* root dir */
	if (cwdi->cwdi_rdir == NULL || (cnp->cn_flags & NOCHROOT)) {
		rootdir = rootvnode;
	} else {
		rootdir = cwdi->cwdi_rdir;
	}

	/* emulation root dir, if any */
	if ((cnp->cn_flags & TRYEMULROOT) == 0) {
		/* if we don't want it, don't fetch it */
		erootdir = NULL;
	} else if (cnp->cn_flags & EMULROOTSET) {
		/* explicitly set emulroot; "/../" doesn't override this */
		erootdir = ndp->ni_erootdir;
	} else if (!strncmp(ndp->ni_pnbuf, "/../", 4)) {
		/* explicit reference to real rootdir */
		erootdir = NULL;
	} else {
		/* may be null */
		erootdir = cwdi->cwdi_edir;
	}

	/* current dir */
	curdir = cwdi->cwdi_cdir;

	if (ndp->ni_pnbuf[0] != '/') {
		if (ndp->ni_atdir != NULL) {
			startdir = ndp->ni_atdir;
		} else {
			startdir = curdir;
		}
		erootdir = NULL;
	} else if (cnp->cn_flags & TRYEMULROOT && erootdir != NULL) {
		startdir = erootdir;
	} else {
		startdir = rootdir;
		erootdir = NULL;
	}

	state->ndp->ni_rootdir = rootdir;
	state->ndp->ni_erootdir = erootdir;

	/*
	 * Get a reference to the start dir so we can safely unlock cwdi.
	 *
	 * Must hold references to rootdir and erootdir while we're running.
	 * A multithreaded process may chroot during namei.
	 */
	if (startdir != NULL)
		vref(startdir);
	if (state->ndp->ni_rootdir != NULL)
		vref(state->ndp->ni_rootdir);
	if (state->ndp->ni_erootdir != NULL)
		vref(state->ndp->ni_erootdir);
	state->root_referenced = 1;

	rw_exit(&cwdi->cwdi_lock);
	return startdir;
}

/*
 * Get the directory context for the nfsd case, in parallel to
 * getstartdir. Initializes the rootdir and erootdir state and
 * returns a reference to the passed-in starting dir.
 */
static struct vnode *
namei_getstartdir_for_nfsd(struct namei_state *state)
{
	KASSERT(state->ndp->ni_atdir != NULL);

	/* always use the real root, and never set an emulation root */
	if (rootvnode == NULL) {
		return NULL;
	}
	state->ndp->ni_rootdir = rootvnode;
	state->ndp->ni_erootdir = NULL;

	vref(state->ndp->ni_atdir);
	KASSERT(! state->root_referenced);
	vref(state->ndp->ni_rootdir);
	state->root_referenced = 1;
	return state->ndp->ni_atdir;
}


/*
 * Ktrace the namei operation.
 */
static void
namei_ktrace(struct namei_state *state)
{
	struct nameidata *ndp = state->ndp;
	struct componentname *cnp = state->cnp;
	struct lwp *self = curlwp;	/* thread doing namei() */
	const char *emul_path;

	if (ktrpoint(KTR_NAMEI)) {
		if (ndp->ni_erootdir != NULL) {
			/*
			 * To make any sense, the trace entry need to have the
			 * text of the emulation path prepended.
			 * Usually we can get this from the current process,
			 * but when called from emul_find_interp() it is only
			 * in the exec_package - so we get it passed in ni_next
			 * (this is a hack).
			 */
			if (cnp->cn_flags & EMULROOTSET)
				emul_path = ndp->ni_next;
			else
				emul_path = self->l_proc->p_emul->e_path;
			ktrnamei2(emul_path, strlen(emul_path),
			    ndp->ni_pnbuf, ndp->ni_pathlen);
		} else
			ktrnamei(ndp->ni_pnbuf, ndp->ni_pathlen);
	}
}

/*
 * Start up namei. Find the root dir and cwd, establish the starting
 * directory for lookup, and lock it. Also calls ktrace when
 * appropriate.
 */
static int
namei_start(struct namei_state *state, int isnfsd,
	    struct vnode **startdir_ret)
{
	struct nameidata *ndp = state->ndp;
	struct vnode *startdir;

	/* length includes null terminator (was originally from copyinstr) */
	ndp->ni_pathlen = strlen(ndp->ni_pnbuf) + 1;

	/*
	 * POSIX.1 requirement: "" is not a valid file name.
	 */
	if (ndp->ni_pathlen == 1) {
		ndp->ni_erootdir = NULL;
		return ENOENT;
	}

	ndp->ni_loopcnt = 0;

	/* Get starting directory, set up root, and ktrace. */
	if (isnfsd) {
		startdir = namei_getstartdir_for_nfsd(state);
		/* no ktrace */
	} else {
		startdir = namei_getstartdir(state);
		namei_ktrace(state);
	}

	if (startdir == NULL) {
		return ENOENT;
	}

	/* NDAT may feed us with a non directory namei_getstartdir */
	if (startdir->v_type != VDIR) {
		vrele(startdir);
		return ENOTDIR;
	}

	vn_lock(startdir, LK_EXCLUSIVE | LK_RETRY);

	*startdir_ret = startdir;
	return 0;
}

/*
 * Check for being at a symlink that we're going to follow.
 */
static inline int
namei_atsymlink(struct namei_state *state, struct vnode *foundobj)
{
	return (foundobj->v_type == VLNK) &&
		(state->cnp->cn_flags & (FOLLOW|REQUIREDIR));
}

/*
 * Follow a symlink.
 *
 * Updates searchdir. inhibitmagic causes magic symlinks to not be
 * interpreted; this is used by nfsd.
 *
 * Unlocks foundobj on success (ugh)
 */
static inline int
namei_follow(struct namei_state *state, int inhibitmagic,
	     struct vnode *searchdir, struct vnode *foundobj,
	     struct vnode **newsearchdir_ret)
{
	struct nameidata *ndp = state->ndp;
	struct componentname *cnp = state->cnp;

	struct lwp *self = curlwp;	/* thread doing namei() */
	struct iovec aiov;		/* uio for reading symbolic links */
	struct uio auio;
	char *cp;			/* pointer into pathname argument */
	size_t linklen;
	int error;

	KASSERT(VOP_ISLOCKED(searchdir) == LK_EXCLUSIVE);
	KASSERT(VOP_ISLOCKED(foundobj) == LK_EXCLUSIVE);
	if (ndp->ni_loopcnt++ >= MAXSYMLINKS) {
		return ELOOP;
	}
	if (foundobj->v_mount->mnt_flag & MNT_SYMPERM) {
		error = VOP_ACCESS(foundobj, VEXEC, cnp->cn_cred);
		if (error != 0)
			return error;
	}

	/* FUTURE: fix this to not use a second buffer */
	cp = PNBUF_GET();
	aiov.iov_base = cp;
	aiov.iov_len = MAXPATHLEN;
	auio.uio_iov = &aiov;
	auio.uio_iovcnt = 1;
	auio.uio_offset = 0;
	auio.uio_rw = UIO_READ;
	auio.uio_resid = MAXPATHLEN;
	UIO_SETUP_SYSSPACE(&auio);
	error = VOP_READLINK(foundobj, &auio, cnp->cn_cred);
	if (error) {
		PNBUF_PUT(cp);
		return error;
	}
	linklen = MAXPATHLEN - auio.uio_resid;
	if (linklen == 0) {
		PNBUF_PUT(cp);
		return ENOENT;
	}

	/*
	 * Do symlink substitution, if appropriate, and
	 * check length for potential overflow.
	 *
	 * Inhibit symlink substitution for nfsd.
	 * XXX: This is how it was before; is that a bug or a feature?
	 */
	if ((!inhibitmagic && vfs_magiclinks &&
	     symlink_magic(self->l_proc, cp, &linklen)) ||
	    (linklen + ndp->ni_pathlen >= MAXPATHLEN)) {
		PNBUF_PUT(cp);
		return ENAMETOOLONG;
	}
	if (ndp->ni_pathlen > 1) {
		/* includes a null-terminator */
		memcpy(cp + linklen, ndp->ni_next, ndp->ni_pathlen);
	} else {
		cp[linklen] = '\0';
	}
	ndp->ni_pathlen += linklen;
	memcpy(ndp->ni_pnbuf, cp, ndp->ni_pathlen);
	PNBUF_PUT(cp);

	/* we're now starting from the beginning of the buffer again */
	cnp->cn_nameptr = ndp->ni_pnbuf;

	/* must unlock this before relocking searchdir */
	VOP_UNLOCK(foundobj);

	/*
	 * Check if root directory should replace current directory.
	 */
	if (ndp->ni_pnbuf[0] == '/') {
		vput(searchdir);
		/* Keep absolute symbolic links inside emulation root */
		searchdir = ndp->ni_erootdir;
		if (searchdir == NULL ||
		    (ndp->ni_pnbuf[1] == '.' 
		     && ndp->ni_pnbuf[2] == '.'
		     && ndp->ni_pnbuf[3] == '/')) {
			ndp->ni_erootdir = NULL;
			searchdir = ndp->ni_rootdir;
		}
		vref(searchdir);
		vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
		while (cnp->cn_nameptr[0] == '/') {
			cnp->cn_nameptr++;
			ndp->ni_pathlen--;
		}
	}

	*newsearchdir_ret = searchdir;
	KASSERT(VOP_ISLOCKED(searchdir) == LK_EXCLUSIVE);
	return 0;
}

//////////////////////////////

/*
 * Inspect the leading path component and update the state accordingly.
 */
static int
lookup_parsepath(struct namei_state *state)
{
	const char *cp;			/* pointer into pathname argument */

	struct componentname *cnp = state->cnp;
	struct nameidata *ndp = state->ndp;

	KASSERT(cnp == &ndp->ni_cnd);

	/*
	 * Search a new directory.
	 *
	 * The last component of the filename is left accessible via
	 * cnp->cn_nameptr for callers that need the name. Callers needing
	 * the name set the SAVENAME flag. When done, they assume
	 * responsibility for freeing the pathname buffer.
	 *
	 * At this point, our only vnode state is that the search dir
	 * is held and locked.
	 */
	cnp->cn_consume = 0;
	cnp->cn_namelen = namei_getcomponent(cnp->cn_nameptr);
	cp = cnp->cn_nameptr + cnp->cn_namelen;
	if (cnp->cn_namelen > KERNEL_NAME_MAX) {
		return ENAMETOOLONG;
	}
#ifdef NAMEI_DIAGNOSTIC
	{ char c = *cp;
	*(char *)cp = '\0';
	printf("{%s}: ", cnp->cn_nameptr);
	*(char *)cp = c; }
#endif /* NAMEI_DIAGNOSTIC */
	ndp->ni_pathlen -= cnp->cn_namelen;
	ndp->ni_next = cp;
	/*
	 * If this component is followed by a slash, then move the pointer to
	 * the next component forward, and remember that this component must be
	 * a directory.
	 */
	if (*cp == '/') {
		do {
			cp++;
		} while (*cp == '/');
		state->slashes = cp - ndp->ni_next;
		ndp->ni_pathlen -= state->slashes;
		ndp->ni_next = cp;
		cnp->cn_flags |= REQUIREDIR;
	} else {
		state->slashes = 0;
		cnp->cn_flags &= ~REQUIREDIR;
	}
	/*
	 * We do special processing on the last component, whether or not it's
	 * a directory.  Cache all intervening lookups, but not the final one.
	 */
	if (*cp == '\0') {
		if (state->docache)
			cnp->cn_flags |= MAKEENTRY;
		else
			cnp->cn_flags &= ~MAKEENTRY;
		cnp->cn_flags |= ISLASTCN;
	} else {
		cnp->cn_flags |= MAKEENTRY;
		cnp->cn_flags &= ~ISLASTCN;
	}
	if (cnp->cn_namelen == 2 &&
	    cnp->cn_nameptr[1] == '.' && cnp->cn_nameptr[0] == '.')
		cnp->cn_flags |= ISDOTDOT;
	else
		cnp->cn_flags &= ~ISDOTDOT;

	return 0;
}

/*
 * Call VOP_LOOKUP for a single lookup; return a new search directory
 * (used when crossing mountpoints up or searching union mounts down) and 
 * the found object, which for create operations may be NULL on success.
 *
 * Note that the new search directory may be null, which means the
 * searchdir was unlocked and released. This happens in the common case
 * when crossing a mount point downwards, in order to avoid coupling
 * locks between different file system volumes. Importantly, this can
 * happen even if the call fails. (XXX: this is gross and should be
 * tidied somehow.)
 */
static int
lookup_once(struct namei_state *state,
	    struct vnode *searchdir,
	    struct vnode **newsearchdir_ret,
	    struct vnode **foundobj_ret)
{
	struct vnode *tmpvn;		/* scratch vnode */
	struct vnode *foundobj;		/* result */
	struct mount *mp;		/* mount table entry */
	struct lwp *l = curlwp;
	int error;

	struct componentname *cnp = state->cnp;
	struct nameidata *ndp = state->ndp;

	KASSERT(cnp == &ndp->ni_cnd);
	KASSERT(VOP_ISLOCKED(searchdir) == LK_EXCLUSIVE);
	*newsearchdir_ret = searchdir;

	/*
	 * Handle "..": two special cases.
	 * 1. If at root directory (e.g. after chroot)
	 *    or at absolute root directory
	 *    then ignore it so can't get out.
	 * 1a. If at the root of the emulation filesystem go to the real
	 *    root. So "/../<path>" is always absolute.
	 * 1b. If we have somehow gotten out of a jail, warn
	 *    and also ignore it so we can't get farther out.
	 * 2. If this vnode is the root of a mounted
	 *    filesystem, then replace it with the
	 *    vnode which was mounted on so we take the
	 *    .. in the other file system.
	 */
	if (cnp->cn_flags & ISDOTDOT) {
		struct proc *p = l->l_proc;

		for (;;) {
			if (searchdir == ndp->ni_rootdir ||
			    searchdir == rootvnode) {
				foundobj = searchdir;
				vref(foundobj);
				*foundobj_ret = foundobj;
				error = 0;
				goto done;
			}
			if (ndp->ni_rootdir != rootvnode) {
				int retval;

				VOP_UNLOCK(searchdir);
				retval = vn_isunder(searchdir, ndp->ni_rootdir, l);
				vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
				if (!retval) {
				    /* Oops! We got out of jail! */
				    log(LOG_WARNING,
					"chrooted pid %d uid %d (%s) "
					"detected outside of its chroot\n",
					p->p_pid, kauth_cred_geteuid(l->l_cred),
					p->p_comm);
				    /* Put us at the jail root. */
				    vput(searchdir);
				    searchdir = NULL;
				    foundobj = ndp->ni_rootdir;
				    vref(foundobj);
				    vref(foundobj);
				    vn_lock(foundobj, LK_EXCLUSIVE | LK_RETRY);
				    *newsearchdir_ret = foundobj;
				    *foundobj_ret = foundobj;
				    error = 0;
				    goto done;
				}
			}
			if ((searchdir->v_vflag & VV_ROOT) == 0 ||
			    (cnp->cn_flags & NOCROSSMOUNT))
				break;
			tmpvn = searchdir;
			searchdir = searchdir->v_mount->mnt_vnodecovered;
			vref(searchdir);
			vput(tmpvn);
			vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
			*newsearchdir_ret = searchdir;
		}
	}

	/*
	 * We now have a segment name to search for, and a directory to search.
	 * Our vnode state here is that "searchdir" is held and locked.
	 */
unionlookup:
	foundobj = NULL;
	error = VOP_LOOKUP(searchdir, &foundobj, cnp);

	if (error != 0) {
		KASSERTMSG((foundobj == NULL),
		    "leaf `%s' should be empty but is %p",
		    cnp->cn_nameptr, foundobj);
#ifdef NAMEI_DIAGNOSTIC
		printf("not found\n");
#endif /* NAMEI_DIAGNOSTIC */
		if ((error == ENOENT) &&
		    (searchdir->v_vflag & VV_ROOT) &&
		    (searchdir->v_mount->mnt_flag & MNT_UNION)) {
			tmpvn = searchdir;
			searchdir = searchdir->v_mount->mnt_vnodecovered;
			vref(searchdir);
			vput(tmpvn);
			vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
			*newsearchdir_ret = searchdir;
			goto unionlookup;
		}

		if (error != EJUSTRETURN)
			goto done;

		/*
		 * If this was not the last component, or there were trailing
		 * slashes, and we are not going to create a directory,
		 * then the name must exist.
		 */
		if ((cnp->cn_flags & (REQUIREDIR | CREATEDIR)) == REQUIREDIR) {
			error = ENOENT;
			goto done;
		}

		/*
		 * If creating and at end of pathname, then can consider
		 * allowing file to be created.
		 */
		if (state->rdonly) {
			error = EROFS;
			goto done;
		}

		/*
		 * We return success and a NULL foundobj to indicate
		 * that the entry doesn't currently exist, leaving a
		 * pointer to the (normally, locked) directory vnode
		 * as searchdir.
		 */
		*foundobj_ret = NULL;
		error = 0;
		goto done;
	}
#ifdef NAMEI_DIAGNOSTIC
	printf("found\n");
#endif /* NAMEI_DIAGNOSTIC */

	/*
	 * Take into account any additional components consumed by the
	 * underlying filesystem.  This will include any trailing slashes after
	 * the last component consumed.
	 */
	if (cnp->cn_consume > 0) {
		ndp->ni_pathlen -= cnp->cn_consume - state->slashes;
		ndp->ni_next += cnp->cn_consume - state->slashes;
		cnp->cn_consume = 0;
		if (ndp->ni_next[0] == '\0')
			cnp->cn_flags |= ISLASTCN;
	}

	/*
	 * "searchdir" is locked and held, "foundobj" is held,
	 * they may be the same vnode.
	 */
	if (searchdir != foundobj) {
		if (cnp->cn_flags & ISDOTDOT)
			VOP_UNLOCK(searchdir);
		error = vn_lock(foundobj, LK_EXCLUSIVE);
		if (cnp->cn_flags & ISDOTDOT)
			vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
		if (error != 0) {
			vrele(foundobj);
			goto done;
		}
	}

	/*
	 * Check to see if the vnode has been mounted on;
	 * if so find the root of the mounted file system.
	 */
	KASSERT(searchdir != NULL);
	while (foundobj->v_type == VDIR &&
	       (mp = foundobj->v_mountedhere) != NULL &&
	       (cnp->cn_flags & NOCROSSMOUNT) == 0) {

		KASSERT(searchdir != foundobj);

		error = vfs_busy(mp);
		if (error != 0) {
			vput(foundobj);
			goto done;
		}
		if (searchdir != NULL) {
			VOP_UNLOCK(searchdir);
		}
		vput(foundobj);
		error = VFS_ROOT(mp, &foundobj);
		vfs_unbusy(mp);
		if (error) {
			if (searchdir != NULL) {
				vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
			}
			goto done;
		}
		/*
		 * Avoid locking vnodes from two filesystems because
		 * it's prone to deadlock, e.g. when using puffs.
		 * Also, it isn't a good idea to propagate slowness of
		 * a filesystem up to the root directory. For now,
		 * only handle the common case, where foundobj is
		 * VDIR.
		 *
		 * In this case set searchdir to null to avoid using
		 * it again. It is not correct to set searchdir ==
		 * foundobj here as that will confuse the caller.
		 * (See PR 40740.)
		 */
		if (searchdir == NULL) {
			/* already been here once; do nothing further */
		} else if (foundobj->v_type == VDIR) {
			vrele(searchdir);
			*newsearchdir_ret = searchdir = NULL;
		} else {
			VOP_UNLOCK(foundobj);
			vn_lock(searchdir, LK_EXCLUSIVE | LK_RETRY);
			vn_lock(foundobj, LK_EXCLUSIVE | LK_RETRY);
		}
	}

	*foundobj_ret = foundobj;
	error = 0;
done:
	KASSERT(*newsearchdir_ret == NULL ||
		VOP_ISLOCKED(*newsearchdir_ret) == LK_EXCLUSIVE);
	/*
	 * *foundobj_ret is valid only if error == 0.
	 */
	KASSERT(error != 0 || *foundobj_ret == NULL ||
	    VOP_ISLOCKED(*foundobj_ret) == LK_EXCLUSIVE);
	return error;
}

//////////////////////////////

/*
 * Do a complete path search from a single root directory.
 * (This is called up to twice if TRYEMULROOT is in effect.)
 */
static int
namei_oneroot(struct namei_state *state,
	 int neverfollow, int inhibitmagic, int isnfsd)
{
	struct nameidata *ndp = state->ndp;
	struct componentname *cnp = state->cnp;
	struct vnode *searchdir, *foundobj;
	int error;

	error = namei_start(state, isnfsd, &searchdir);
	if (error) {
		ndp->ni_dvp = NULL;
		ndp->ni_vp = NULL;
		return error;
	}
	KASSERT(searchdir->v_type == VDIR);

	/*
	 * Setup: break out flag bits into variables.
	 */
	state->docache = (cnp->cn_flags & NOCACHE) ^ NOCACHE;
	if (cnp->cn_nameiop == DELETE)
		state->docache = 0;
	state->rdonly = cnp->cn_flags & RDONLY;

	/*
	 * Keep going until we run out of path components.
	 */
	cnp->cn_nameptr = ndp->ni_pnbuf;

	/* drop leading slashes (already used them to choose startdir) */
	while (cnp->cn_nameptr[0] == '/') {
		cnp->cn_nameptr++;
		ndp->ni_pathlen--;
	}
	/* was it just "/"? */
	if (cnp->cn_nameptr[0] == '\0') {
		foundobj = searchdir;
		searchdir = NULL;
		cnp->cn_flags |= ISLASTCN;

		/* bleh */
		goto skiploop;
	}

	for (;;) {
		KASSERT(searchdir != NULL);
		KASSERT(VOP_ISLOCKED(searchdir) == LK_EXCLUSIVE);

		/*
		 * If the directory we're on is unmounted, bail out.
		 * XXX: should this also check if it's unlinked?
		 * XXX: yes it should... but how?
		 */
		if (searchdir->v_mount == NULL) {
			vput(searchdir);
			ndp->ni_dvp = NULL;
			ndp->ni_vp = NULL;
			return (ENOENT);
		}

		/*
		 * Look up the next path component.
		 * (currently, this may consume more than one)
		 */

		/* There should be no slashes here. */
		KASSERT(cnp->cn_nameptr[0] != '/');

		/* and we shouldn't have looped around if we were done */
		KASSERT(cnp->cn_nameptr[0] != '\0');

		error = lookup_parsepath(state);
		if (error) {
			vput(searchdir);
			ndp->ni_dvp = NULL;
			ndp->ni_vp = NULL;
			state->attempt_retry = 1;
			return (error);
		}

		error = lookup_once(state, searchdir, &searchdir, &foundobj);
		if (error) {
			if (searchdir != NULL) {
				vput(searchdir);
			}
			ndp->ni_dvp = NULL;
			ndp->ni_vp = NULL;
			/*
			 * Note that if we're doing TRYEMULROOT we can
			 * retry with the normal root. Where this is
			 * currently set matches previous practice,
			 * but the previous practice didn't make much
			 * sense and somebody should sit down and
			 * figure out which cases should cause retry
			 * and which shouldn't. XXX.
			 */
			state->attempt_retry = 1;
			return (error);
		}

		if (foundobj == NULL) {
			/*
			 * Success with no object returned means we're
			 * creating something and it isn't already
			 * there. Break out of the main loop now so
			 * the code below doesn't have to test for
			 * foundobj == NULL.
			 */
			/* lookup_once can't have dropped the searchdir */
			KASSERT(searchdir != NULL);
			break;
		}

		/*
		 * Check for symbolic link. If we've reached one,
		 * follow it, unless we aren't supposed to. Back up
		 * over any slashes that we skipped, as we will need
		 * them again.
		 */
		if (namei_atsymlink(state, foundobj)) {
			ndp->ni_pathlen += state->slashes;
			ndp->ni_next -= state->slashes;
			if (neverfollow) {
				error = EINVAL;
			} else if (searchdir == NULL) {
				/*
				 * dholland 20160410: lookup_once only
				 * drops searchdir if it crossed a
				 * mount point. Therefore, if we get
				 * here it means we crossed a mount
				 * point to a mounted filesystem whose
				 * root vnode is a symlink. In theory
				 * we could continue at this point by
				 * using the pre-crossing searchdir
				 * (e.g. just take out an extra
				 * reference on it before calling
				 * lookup_once so we still have it),
				 * but this will make an ugly mess and
				 * it should never happen in practice
				 * as only badly broken filesystems
				 * have non-directory root vnodes. (I
				 * have seen this sort of thing with
				 * NFS occasionally but even then it
				 * means something's badly wrong.)
				 */
				error = ENOTDIR;
			} else {
				/*
				 * dholland 20110410: if we're at a
				 * union mount it might make sense to
				 * use the top of the union stack here
				 * rather than the layer we found the
				 * symlink in. (FUTURE)
				 */
				error = namei_follow(state, inhibitmagic,
						     searchdir, foundobj,
						     &searchdir);
			}
			if (error) {
				KASSERT(searchdir != foundobj);
				if (searchdir != NULL) {
					vput(searchdir);
				}
				vput(foundobj);
				ndp->ni_dvp = NULL;
				ndp->ni_vp = NULL;
				return error;
			}
			/* namei_follow unlocks it (ugh) so rele, not put */
			vrele(foundobj);
			foundobj = NULL;

			/*
			 * If we followed a symlink to `/' and there
			 * are no more components after the symlink,
			 * we're done with the loop and what we found
			 * is the searchdir.
			 */
			if (cnp->cn_nameptr[0] == '\0') {
				KASSERT(searchdir != NULL);
				foundobj = searchdir;
				searchdir = NULL;
				cnp->cn_flags |= ISLASTCN;
				break;
			}

			continue;
		}

		/*
		 * Not a symbolic link.
		 *
		 * Check for directory, if the component was
		 * followed by a series of slashes.
		 */
		if ((foundobj->v_type != VDIR) &&
		    (cnp->cn_flags & REQUIREDIR)) {
			KASSERT(foundobj != searchdir);
			if (searchdir) {
				vput(searchdir);
			}
			vput(foundobj);
			ndp->ni_dvp = NULL;
			ndp->ni_vp = NULL;
			state->attempt_retry = 1;
			return ENOTDIR;
		}

		/*
		 * Stop if we've reached the last component.
		 */
		if (cnp->cn_flags & ISLASTCN) {
			break;
		}

		/*
		 * Continue with the next component.
		 */
		cnp->cn_nameptr = ndp->ni_next;
		if (searchdir == foundobj) {
			vrele(searchdir);
		} else if (searchdir != NULL) {
			vput(searchdir);
		}
		searchdir = foundobj;
		foundobj = NULL;
	}

 skiploop:

	if (foundobj != NULL) {
		if (foundobj == ndp->ni_erootdir) {
			/*
			 * We are about to return the emulation root.
			 * This isn't a good idea because code might
			 * repeatedly lookup ".." until the file
			 * matches that returned for "/" and loop
			 * forever.  So convert it to the real root.
			 */
			if (searchdir != NULL) {
				if (searchdir == foundobj)
					vrele(searchdir);
				else
					vput(searchdir);
				searchdir = NULL;
			}
			vput(foundobj);
			foundobj = ndp->ni_rootdir;
			vref(foundobj);
			vn_lock(foundobj, LK_EXCLUSIVE | LK_RETRY);
		}

		/*
		 * If the caller requested the parent node (i.e. it's
		 * a CREATE, DELETE, or RENAME), and we don't have one
		 * (because this is the root directory, or we crossed
		 * a mount point), then we must fail.
		 *
		 * 20210604 dholland when NONEXCLHACK is set (open
		 * with O_CREAT but not O_EXCL) skip this logic. Since
		 * we have a foundobj, open will not be creating, so
		 * it doesn't actually need or use the searchdir, so
		 * it's ok to return it even if it's on a different
		 * volume, and it's also ok to return NULL; by setting
		 * NONEXCLHACK the open code promises to cope with
		 * those cases correctly. (That is, it should do what
		 * it would do anyway, that is, just release the
		 * searchdir, except not crash if it's null.) This is
		 * needed because otherwise opening mountpoints with
		 * O_CREAT but not O_EXCL fails... which is a silly
		 * thing to do but ought to work. (This whole issue
		 * came to light because 3rd party code wanted to open
		 * certain procfs nodes with O_CREAT for some 3rd
		 * party reason, and it failed.)
		 *
		 * Note that NONEXCLHACK is properly a different
		 * nameiop (it is partway between LOOKUP and CREATE)
		 * but it was stuffed in as a flag instead to make the
		 * resulting patch less invasive for pullup. Blah.
		 */
		if (cnp->cn_nameiop != LOOKUP &&
		    (searchdir == NULL ||
		     searchdir->v_mount != foundobj->v_mount) &&
		    (cnp->cn_flags & NONEXCLHACK) == 0) {
			if (searchdir) {
				vput(searchdir);
			}
			vput(foundobj);
			foundobj = NULL;
			ndp->ni_dvp = NULL;
			ndp->ni_vp = NULL;
			state->attempt_retry = 1;

			switch (cnp->cn_nameiop) {
			    case CREATE:
				return EEXIST;
			    case DELETE:
			    case RENAME:
				return EBUSY;
			    default:
				break;
			}
			panic("Invalid nameiop\n");
		}

		/*
		 * Disallow directory write attempts on read-only lookups.
		 * Prefers EEXIST over EROFS for the CREATE case.
		 */
		if (state->rdonly &&
		    (cnp->cn_nameiop == DELETE || cnp->cn_nameiop == RENAME)) {
			if (searchdir) {
				if (foundobj != searchdir) {
					vput(searchdir);
				} else {
					vrele(searchdir);
				}
				searchdir = NULL;
			}
			vput(foundobj);
			foundobj = NULL;
			ndp->ni_dvp = NULL;
			ndp->ni_vp = NULL;
			state->attempt_retry = 1;
			return EROFS;
		}
		if ((cnp->cn_flags & LOCKLEAF) == 0) {
			/*
			 * Note: if LOCKPARENT but not LOCKLEAF is
			 * set, and searchdir == foundobj, this code
			 * necessarily unlocks the parent as well as
			 * the leaf. That is, just because you specify
			 * LOCKPARENT doesn't mean you necessarily get
			 * a locked parent vnode. The code in
			 * vfs_syscalls.c, and possibly elsewhere,
			 * that uses this combination "knows" this, so
			 * it can't be safely changed. Feh. XXX
			 */
			VOP_UNLOCK(foundobj);
		}
	}

	/*
	 * Done.
	 */

	/*
	 * If LOCKPARENT is not set, the parent directory isn't returned.
	 */
	if ((cnp->cn_flags & LOCKPARENT) == 0 && searchdir != NULL) {
		if (searchdir == foundobj) {
			vrele(searchdir);
		} else {
			vput(searchdir);
		}
		searchdir = NULL;
	}

	ndp->ni_dvp = searchdir;
	ndp->ni_vp = foundobj;
	return 0;
}

/*
 * Do namei; wrapper layer that handles TRYEMULROOT.
 */
static int
namei_tryemulroot(struct namei_state *state,
	 int neverfollow, int inhibitmagic, int isnfsd)
{
	int error;

	struct nameidata *ndp = state->ndp;
	struct componentname *cnp = state->cnp;
	const char *savepath = NULL;

	KASSERT(cnp == &ndp->ni_cnd);

	if (cnp->cn_flags & TRYEMULROOT) {
		savepath = pathbuf_stringcopy_get(ndp->ni_pathbuf);
	}

    emul_retry:
	state->attempt_retry = 0;

	error = namei_oneroot(state, neverfollow, inhibitmagic, isnfsd);
	if (error) {
		/*
		 * Once namei has started up, the existence of ni_erootdir
		 * tells us whether we're working from an emulation root.
		 * The TRYEMULROOT flag isn't necessarily authoritative.
		 */
		if (ndp->ni_erootdir != NULL && state->attempt_retry) {
			/* Retry the whole thing using the normal root */
			cnp->cn_flags &= ~TRYEMULROOT;
			state->attempt_retry = 0;

			/* kinda gross */
			strcpy(ndp->ni_pathbuf->pb_path, savepath);
			pathbuf_stringcopy_put(ndp->ni_pathbuf, savepath);
			savepath = NULL;

			goto emul_retry;
		}
	}
	if (savepath != NULL) {
		pathbuf_stringcopy_put(ndp->ni_pathbuf, savepath);
	}
	return error;
}

/*
 * External interface.
 */
int
namei(struct nameidata *ndp)
{
	struct namei_state state;
	int error;

	namei_init(&state, ndp);
	error = namei_tryemulroot(&state,
				  0/*!neverfollow*/, 0/*!inhibitmagic*/,
				  0/*isnfsd*/);
	namei_cleanup(&state);

	if (error) {
		/* make sure no stray refs leak out */
		KASSERT(ndp->ni_dvp == NULL);
		KASSERT(ndp->ni_vp == NULL);
	}

	return error;
}

////////////////////////////////////////////////////////////

/*
 * External interface used by nfsd. This is basically different from
 * namei only in that it has the ability to pass in the "current
 * directory", and uses an extra flag "neverfollow" for which there's
 * no physical flag defined in namei.h. (There used to be a cut&paste
 * copy of about half of namei in nfsd to allow these minor
 * adjustments to exist.)
 *
 * XXX: the namei interface should be adjusted so nfsd can just use
 * ordinary namei().
 */
int
lookup_for_nfsd(struct nameidata *ndp, struct vnode *forcecwd, int neverfollow)
{
	struct namei_state state;
	int error;

	KASSERT(ndp->ni_atdir == NULL);
	ndp->ni_atdir = forcecwd;

	namei_init(&state, ndp);
	error = namei_tryemulroot(&state,
				  neverfollow, 1/*inhibitmagic*/, 1/*isnfsd*/);
	namei_cleanup(&state);

	if (error) {
		/* make sure no stray refs leak out */
		KASSERT(ndp->ni_dvp == NULL);
		KASSERT(ndp->ni_vp == NULL);
	}

	return error;
}

/*
 * A second external interface used by nfsd. This turns out to be a
 * single lookup used by the WebNFS code (ha!) to get "index.html" or
 * equivalent when asked for a directory. It should eventually evolve
 * into some kind of namei_once() call; for the time being it's kind
 * of a mess. XXX.
 *
 * dholland 20110109: I don't think it works, and I don't think it
 * worked before I started hacking and slashing either, and I doubt
 * anyone will ever notice.
 */

/*
 * Internals. This calls lookup_once() after setting up the assorted
 * pieces of state the way they ought to be.
 */
static int
do_lookup_for_nfsd_index(struct namei_state *state)
{
	int error = 0;

	struct componentname *cnp = state->cnp;
	struct nameidata *ndp = state->ndp;
	struct vnode *startdir;
	struct vnode *foundobj;
	const char *cp;			/* pointer into pathname argument */

	KASSERT(cnp == &ndp->ni_cnd);

	startdir = state->ndp->ni_atdir;

	cnp->cn_nameptr = ndp->ni_pnbuf;
	state->docache = 1;
	state->rdonly = cnp->cn_flags & RDONLY;
	ndp->ni_dvp = NULL;

	cnp->cn_consume = 0;
	cnp->cn_namelen = namei_getcomponent(cnp->cn_nameptr);
	cp = cnp->cn_nameptr + cnp->cn_namelen;
	KASSERT(cnp->cn_namelen <= KERNEL_NAME_MAX);
	ndp->ni_pathlen -= cnp->cn_namelen;
	ndp->ni_next = cp;
	state->slashes = 0;
	cnp->cn_flags &= ~REQUIREDIR;
	cnp->cn_flags |= MAKEENTRY|ISLASTCN;

	if (cnp->cn_namelen == 2 &&
	    cnp->cn_nameptr[1] == '.' && cnp->cn_nameptr[0] == '.')
		cnp->cn_flags |= ISDOTDOT;
	else
		cnp->cn_flags &= ~ISDOTDOT;

	/*
	 * Because lookup_once can change the startdir, we need our
	 * own reference to it to avoid consuming the caller's.
	 */
	vref(startdir);
	vn_lock(startdir, LK_EXCLUSIVE | LK_RETRY);
	error = lookup_once(state, startdir, &startdir, &foundobj);
	if (error == 0 && startdir == foundobj) {
		vrele(startdir);
	} else if (startdir != NULL) {
		vput(startdir);
	}
	if (error) {
		goto bad;
	}
	ndp->ni_vp = foundobj;

	if (foundobj == NULL) {
		return 0;
	}

	KASSERT((cnp->cn_flags & LOCKPARENT) == 0);
	if ((cnp->cn_flags & LOCKLEAF) == 0) {
		VOP_UNLOCK(foundobj);
	}
	return (0);

bad:
	ndp->ni_vp = NULL;
	return (error);
}

/*
 * External interface. The partitioning between this function and the
 * above isn't very clear - the above function exists mostly so code
 * that uses "state->" can be shuffled around without having to change
 * it to "state.".
 */
int
lookup_for_nfsd_index(struct nameidata *ndp, struct vnode *startdir)
{
	struct namei_state state;
	int error;

	KASSERT(ndp->ni_atdir == NULL);
	ndp->ni_atdir = startdir;

	/*
	 * Note: the name sent in here (is not|should not be) allowed
	 * to contain a slash.
	 */
	if (strlen(ndp->ni_pathbuf->pb_path) > KERNEL_NAME_MAX) {
		return ENAMETOOLONG;
	}
	if (strchr(ndp->ni_pathbuf->pb_path, '/')) {
		return EINVAL;
	}

	ndp->ni_pathlen = strlen(ndp->ni_pathbuf->pb_path) + 1;
	ndp->ni_pnbuf = NULL;
	ndp->ni_cnd.cn_nameptr = NULL;

	namei_init(&state, ndp);
	error = do_lookup_for_nfsd_index(&state);
	namei_cleanup(&state);

	return error;
}

////////////////////////////////////////////////////////////

/*
 * Reacquire a path name component.
 * dvp is locked on entry and exit.
 * *vpp is locked on exit unless it's NULL.
 */
int
relookup(struct vnode *dvp, struct vnode **vpp, struct componentname *cnp, int dummy)
{
	int rdonly;			/* lookup read-only flag bit */
	int error = 0;
#ifdef DEBUG
	size_t newlen;			/* DEBUG: check name len */
	const char *cp;			/* DEBUG: check name ptr */
#endif /* DEBUG */

	(void)dummy;

	/*
	 * Setup: break out flag bits into variables.
	 */
	rdonly = cnp->cn_flags & RDONLY;

	/*
	 * Search a new directory.
	 *
	 * The cn_hash value is for use by vfs_cache.
	 * The last component of the filename is left accessible via
	 * cnp->cn_nameptr for callers that need the name. Callers needing
	 * the name set the SAVENAME flag. When done, they assume
	 * responsibility for freeing the pathname buffer.
	 */
#ifdef DEBUG
#if 0
	cp = NULL;
	newhash = namei_hash(cnp->cn_nameptr, &cp);
	if ((uint32_t)newhash != (uint32_t)cnp->cn_hash)
		panic("relookup: bad hash");
#endif
	newlen = namei_getcomponent(cnp->cn_nameptr);
	if (cnp->cn_namelen != newlen)
		panic("relookup: bad len");
	cp = cnp->cn_nameptr + cnp->cn_namelen;
	while (*cp == '/')
		cp++;
	if (*cp != 0)
		panic("relookup: not last component");
#endif /* DEBUG */

	/*
	 * Check for degenerate name (e.g. / or "")
	 * which is a way of talking about a directory,
	 * e.g. like "/." or ".".
	 */
	if (cnp->cn_nameptr[0] == '\0')
		panic("relookup: null name");

	if (cnp->cn_flags & ISDOTDOT)
		panic("relookup: lookup on dot-dot");

	/*
	 * We now have a segment name to search for, and a directory to search.
	 */
	*vpp = NULL;
	error = VOP_LOOKUP(dvp, vpp, cnp);
	if ((error) != 0) {
		KASSERTMSG((*vpp == NULL),
		    "leaf `%s' should be empty but is %p",
		    cnp->cn_nameptr, *vpp);
		if (error != EJUSTRETURN)
			goto bad;
	}

	/*
	 * Check for symbolic link
	 */
	KASSERTMSG((*vpp == NULL || (*vpp)->v_type != VLNK ||
		(cnp->cn_flags & FOLLOW) == 0),
	    "relookup: symlink found");

	/*
	 * Check for read-only lookups.
	 */
	if (rdonly && cnp->cn_nameiop != LOOKUP) {
		error = EROFS;
		if (*vpp) {
			vrele(*vpp);
		}
		goto bad;
	}
	/*
	 * Lock result.
	 */
	if (*vpp && *vpp != dvp) {
		error = vn_lock(*vpp, LK_EXCLUSIVE);
		if (error != 0) {
			vrele(*vpp);
			goto bad;
		}
	}
	return (0);

bad:
	*vpp = NULL;
	return (error);
}

/*
 * namei_simple - simple forms of namei.
 *
 * These are wrappers to allow the simple case callers of namei to be
 * left alone while everything else changes under them.
 */

/* Flags */
struct namei_simple_flags_type {
	int dummy;
};
static const struct namei_simple_flags_type ns_nn, ns_nt, ns_fn, ns_ft;
const namei_simple_flags_t NSM_NOFOLLOW_NOEMULROOT = &ns_nn;
const namei_simple_flags_t NSM_NOFOLLOW_TRYEMULROOT = &ns_nt;
const namei_simple_flags_t NSM_FOLLOW_NOEMULROOT = &ns_fn;
const namei_simple_flags_t NSM_FOLLOW_TRYEMULROOT = &ns_ft;

static
int
namei_simple_convert_flags(namei_simple_flags_t sflags)
{
	if (sflags == NSM_NOFOLLOW_NOEMULROOT)
		return NOFOLLOW | 0;
	if (sflags == NSM_NOFOLLOW_TRYEMULROOT)
		return NOFOLLOW | TRYEMULROOT;
	if (sflags == NSM_FOLLOW_NOEMULROOT)
		return FOLLOW | 0;
	if (sflags == NSM_FOLLOW_TRYEMULROOT)
		return FOLLOW | TRYEMULROOT;
	panic("namei_simple_convert_flags: bogus sflags\n");
	return 0;
}

int
namei_simple_kernel(const char *path, namei_simple_flags_t sflags,
	struct vnode **vp_ret)
{
	return nameiat_simple_kernel(NULL, path, sflags, vp_ret);
}

int
nameiat_simple_kernel(struct vnode *dvp, const char *path, 
	namei_simple_flags_t sflags, struct vnode **vp_ret)
{
	struct nameidata nd;
	struct pathbuf *pb;
	int err;

	pb = pathbuf_create(path);
	if (pb == NULL) {
		return ENOMEM;
	}

	NDINIT(&nd,
		LOOKUP,
		namei_simple_convert_flags(sflags),
		pb);

	if (dvp != NULL)
		NDAT(&nd, dvp);

	err = namei(&nd);
	if (err != 0) {
		pathbuf_destroy(pb);
		return err;
	}
	*vp_ret = nd.ni_vp;
	pathbuf_destroy(pb);
	return 0;
}

int
namei_simple_user(const char *path, namei_simple_flags_t sflags,
	struct vnode **vp_ret)
{
	return nameiat_simple_user(NULL, path, sflags, vp_ret);
}

int
nameiat_simple_user(struct vnode *dvp, const char *path,
	namei_simple_flags_t sflags, struct vnode **vp_ret)
{
	struct pathbuf *pb;
	struct nameidata nd;
	int err;

	err = pathbuf_copyin(path, &pb);
	if (err) {
		return err;
	}

	NDINIT(&nd,
		LOOKUP,
		namei_simple_convert_flags(sflags),
		pb);

	if (dvp != NULL)
		NDAT(&nd, dvp);

	err = namei(&nd);
	if (err != 0) {
		pathbuf_destroy(pb);
		return err;
	}
	*vp_ret = nd.ni_vp;
	pathbuf_destroy(pb);
	return 0;
}