summaryrefslogtreecommitdiff
path: root/usr/src/uts/common/fs/zfs/zfs_vfsops.c
blob: 288dc93e3c415fb8322d0965bfc4d1402a6eab51 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
/*
 * CDDL HEADER START
 *
 * The contents of this file are subject to the terms of the
 * Common Development and Distribution License (the "License").
 * You may not use this file except in compliance with the License.
 *
 * You can obtain a copy of the license at usr/src/OPENSOLARIS.LICENSE
 * or http://www.opensolaris.org/os/licensing.
 * See the License for the specific language governing permissions
 * and limitations under the License.
 *
 * When distributing Covered Code, include this CDDL HEADER in each
 * file and include the License file at usr/src/OPENSOLARIS.LICENSE.
 * If applicable, add the following below this CDDL HEADER, with the
 * fields enclosed by brackets "[]" replaced with your own identifying
 * information: Portions Copyright [yyyy] [name of copyright owner]
 *
 * CDDL HEADER END
 */
/*
 * Copyright (c) 2005, 2010, Oracle and/or its affiliates. All rights reserved.
 * Copyright (c) 2012, 2015 by Delphix. All rights reserved.
 * Copyright (c) 2014 Integros [integros.com]
 * Copyright 2016 Nexenta Systems, Inc. All rights reserved.
 * Copyright 2019 Joyent, Inc.
 * Copyright 2020 Joshua M. Clulow <josh@sysmgr.org>
 * Copyright 2020 OmniOS Community Edition (OmniOSce) Association.
 * Copyright 2022 Oxide Computer Company
 */

/* Portions Copyright 2010 Robert Milkowski */

#include <sys/types.h>
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/sysmacros.h>
#include <sys/kmem.h>
#include <sys/pathname.h>
#include <sys/vnode.h>
#include <sys/vfs.h>
#include <sys/vfs_opreg.h>
#include <sys/mntent.h>
#include <sys/mount.h>
#include <sys/cmn_err.h>
#include "fs/fs_subr.h"
#include <sys/zfs_znode.h>
#include <sys/zfs_dir.h>
#include <sys/zil.h>
#include <sys/fs/zfs.h>
#include <sys/dmu.h>
#include <sys/dsl_prop.h>
#include <sys/dsl_dataset.h>
#include <sys/dsl_deleg.h>
#include <sys/spa.h>
#include <sys/zap.h>
#include <sys/sa.h>
#include <sys/sa_impl.h>
#include <sys/varargs.h>
#include <sys/policy.h>
#include <sys/atomic.h>
#include <sys/mkdev.h>
#include <sys/modctl.h>
#include <sys/refstr.h>
#include <sys/zfs_ioctl.h>
#include <sys/zfs_ctldir.h>
#include <sys/zfs_fuid.h>
#include <sys/bootconf.h>
#include <sys/ddi.h>
#include <sys/sunddi.h>
#include <sys/dnlc.h>
#include <sys/dmu_objset.h>
#include <sys/spa_boot.h>
#include <sys/vdev_impl.h>
#include "zfs_comutil.h"

int zfsfstype;
vfsops_t *zfs_vfsops = NULL;
static major_t zfs_major;
static minor_t zfs_minor;
static kmutex_t	zfs_dev_mtx;

extern int sys_shutdown;

static int zfs_mount(vfs_t *vfsp, vnode_t *mvp, struct mounta *uap, cred_t *cr);
static int zfs_umount(vfs_t *vfsp, int fflag, cred_t *cr);
static int zfs_mountroot(vfs_t *vfsp, enum whymountroot);
static int zfs_root(vfs_t *vfsp, vnode_t **vpp);
static int zfs_statvfs(vfs_t *vfsp, struct statvfs64 *statp);
static int zfs_vget(vfs_t *vfsp, vnode_t **vpp, fid_t *fidp);
static void zfs_freevfs(vfs_t *vfsp);

static const fs_operation_def_t zfs_vfsops_template[] = {
	VFSNAME_MOUNT,		{ .vfs_mount = zfs_mount },
	VFSNAME_MOUNTROOT,	{ .vfs_mountroot = zfs_mountroot },
	VFSNAME_UNMOUNT,	{ .vfs_unmount = zfs_umount },
	VFSNAME_ROOT,		{ .vfs_root = zfs_root },
	VFSNAME_STATVFS,	{ .vfs_statvfs = zfs_statvfs },
	VFSNAME_SYNC,		{ .vfs_sync = zfs_sync },
	VFSNAME_VGET,		{ .vfs_vget = zfs_vget },
	VFSNAME_FREEVFS,	{ .vfs_freevfs = zfs_freevfs },
	NULL,			NULL
};

/*
 * We need to keep a count of active fs's.
 * This is necessary to prevent our module
 * from being unloaded after a umount -f
 */
static uint32_t	zfs_active_fs_count = 0;

static char *noatime_cancel[] = { MNTOPT_ATIME, NULL };
static char *atime_cancel[] = { MNTOPT_NOATIME, NULL };
static char *noxattr_cancel[] = { MNTOPT_XATTR, NULL };
static char *xattr_cancel[] = { MNTOPT_NOXATTR, NULL };

/*
 * MO_DEFAULT is not used since the default value is determined
 * by the equivalent property.
 */
static mntopt_t mntopts[] = {
	{ MNTOPT_NOXATTR, noxattr_cancel, NULL, 0, NULL },
	{ MNTOPT_XATTR, xattr_cancel, NULL, 0, NULL },
	{ MNTOPT_NOATIME, noatime_cancel, NULL, 0, NULL },
	{ MNTOPT_ATIME, atime_cancel, NULL, 0, NULL }
};

static mntopts_t zfs_mntopts = {
	sizeof (mntopts) / sizeof (mntopt_t),
	mntopts
};

/*ARGSUSED*/
int
zfs_sync(vfs_t *vfsp, short flag, cred_t *cr)
{
	/*
	 * Data integrity is job one.  We don't want a compromised kernel
	 * writing to the storage pool, so we never sync during panic.
	 */
	if (panicstr)
		return (0);

	/*
	 * SYNC_ATTR is used by fsflush() to force old filesystems like UFS
	 * to sync metadata, which they would otherwise cache indefinitely.
	 * Semantically, the only requirement is that the sync be initiated.
	 * The DMU syncs out txgs frequently, so there's nothing to do.
	 */
	if (flag & SYNC_ATTR)
		return (0);

	if (vfsp != NULL) {
		/*
		 * Sync a specific filesystem.
		 */
		zfsvfs_t *zfsvfs = vfsp->vfs_data;
		dsl_pool_t *dp;

		ZFS_ENTER(zfsvfs);
		dp = dmu_objset_pool(zfsvfs->z_os);

		/*
		 * If the system is shutting down, then skip any
		 * filesystems which may exist on a suspended pool.
		 */
		if (sys_shutdown && spa_suspended(dp->dp_spa)) {
			ZFS_EXIT(zfsvfs);
			return (0);
		}

		if (zfsvfs->z_log != NULL)
			zil_commit(zfsvfs->z_log, 0);

		ZFS_EXIT(zfsvfs);
	} else {
		/*
		 * Sync all ZFS filesystems.  This is what happens when you
		 * run sync(8).  Unlike other filesystems, ZFS honors the
		 * request by waiting for all pools to commit all dirty data.
		 */
		spa_sync_allpools();
	}

	return (0);
}

static int
zfs_create_unique_device(dev_t *dev)
{
	major_t new_major;

	do {
		ASSERT3U(zfs_minor, <=, MAXMIN32);
		minor_t start = zfs_minor;
		do {
			mutex_enter(&zfs_dev_mtx);
			if (zfs_minor >= MAXMIN32) {
				/*
				 * If we're still using the real major
				 * keep out of /dev/zfs and /dev/zvol minor
				 * number space.  If we're using a getudev()'ed
				 * major number, we can use all of its minors.
				 */
				if (zfs_major == ddi_name_to_major(ZFS_DRIVER))
					zfs_minor = ZFS_MIN_MINOR;
				else
					zfs_minor = 0;
			} else {
				zfs_minor++;
			}
			*dev = makedevice(zfs_major, zfs_minor);
			mutex_exit(&zfs_dev_mtx);
		} while (vfs_devismounted(*dev) && zfs_minor != start);
		if (zfs_minor == start) {
			/*
			 * We are using all ~262,000 minor numbers for the
			 * current major number.  Create a new major number.
			 */
			if ((new_major = getudev()) == (major_t)-1) {
				cmn_err(CE_WARN,
				    "zfs_mount: Can't get unique major "
				    "device number.");
				return (-1);
			}
			mutex_enter(&zfs_dev_mtx);
			zfs_major = new_major;
			zfs_minor = 0;

			mutex_exit(&zfs_dev_mtx);
		} else {
			break;
		}
		/* CONSTANTCONDITION */
	} while (1);

	return (0);
}

static void
atime_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	if (newval == TRUE) {
		zfsvfs->z_atime = TRUE;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NOATIME);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_ATIME, NULL, 0);
	} else {
		zfsvfs->z_atime = FALSE;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_ATIME);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NOATIME, NULL, 0);
	}
}

static void
xattr_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	if (newval == TRUE) {
		/* XXX locking on vfs_flag? */
		zfsvfs->z_vfs->vfs_flag |= VFS_XATTR;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NOXATTR);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_XATTR, NULL, 0);
	} else {
		/* XXX locking on vfs_flag? */
		zfsvfs->z_vfs->vfs_flag &= ~VFS_XATTR;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_XATTR);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NOXATTR, NULL, 0);
	}
}

static void
blksz_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;
	ASSERT3U(newval, <=, spa_maxblocksize(dmu_objset_spa(zfsvfs->z_os)));
	ASSERT3U(newval, >=, SPA_MINBLOCKSIZE);
	ASSERT(ISP2(newval));

	zfsvfs->z_max_blksz = newval;
	zfsvfs->z_vfs->vfs_bsize = newval;
}

static void
readonly_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	if (newval) {
		/* XXX locking on vfs_flag? */
		zfsvfs->z_vfs->vfs_flag |= VFS_RDONLY;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_RW);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_RO, NULL, 0);
	} else {
		/* XXX locking on vfs_flag? */
		zfsvfs->z_vfs->vfs_flag &= ~VFS_RDONLY;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_RO);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_RW, NULL, 0);
	}
}

static void
devices_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	if (newval == FALSE) {
		zfsvfs->z_vfs->vfs_flag |= VFS_NODEVICES;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_DEVICES);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NODEVICES, NULL, 0);
	} else {
		zfsvfs->z_vfs->vfs_flag &= ~VFS_NODEVICES;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NODEVICES);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_DEVICES, NULL, 0);
	}
}

static void
setuid_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	if (newval == FALSE) {
		zfsvfs->z_vfs->vfs_flag |= VFS_NOSETUID;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_SETUID);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NOSETUID, NULL, 0);
	} else {
		zfsvfs->z_vfs->vfs_flag &= ~VFS_NOSETUID;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NOSETUID);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_SETUID, NULL, 0);
	}
}

static void
exec_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	if (newval == FALSE) {
		zfsvfs->z_vfs->vfs_flag |= VFS_NOEXEC;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_EXEC);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NOEXEC, NULL, 0);
	} else {
		zfsvfs->z_vfs->vfs_flag &= ~VFS_NOEXEC;
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NOEXEC);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_EXEC, NULL, 0);
	}
}

/*
 * The nbmand mount option can be changed at mount time.
 * We can't allow it to be toggled on live file systems or incorrect
 * behavior may be seen from cifs clients
 *
 * This property isn't registered via dsl_prop_register(), but this callback
 * will be called when a file system is first mounted
 */
static void
nbmand_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;
	if (newval == FALSE) {
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NBMAND);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NONBMAND, NULL, 0);
	} else {
		vfs_clearmntopt(zfsvfs->z_vfs, MNTOPT_NONBMAND);
		vfs_setmntopt(zfsvfs->z_vfs, MNTOPT_NBMAND, NULL, 0);
	}
}

static void
snapdir_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	zfsvfs->z_show_ctldir = newval;
}

static void
vscan_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	zfsvfs->z_vscan = newval;
}

static void
acl_mode_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	zfsvfs->z_acl_mode = newval;
}

static void
acl_inherit_changed_cb(void *arg, uint64_t newval)
{
	zfsvfs_t *zfsvfs = arg;

	zfsvfs->z_acl_inherit = newval;
}

static int
zfs_register_callbacks(vfs_t *vfsp)
{
	struct dsl_dataset *ds = NULL;
	objset_t *os = NULL;
	zfsvfs_t *zfsvfs = NULL;
	uint64_t nbmand;
	boolean_t readonly = B_FALSE;
	boolean_t do_readonly = B_FALSE;
	boolean_t setuid = B_FALSE;
	boolean_t do_setuid = B_FALSE;
	boolean_t exec = B_FALSE;
	boolean_t do_exec = B_FALSE;
	boolean_t devices = B_FALSE;
	boolean_t do_devices = B_FALSE;
	boolean_t xattr = B_FALSE;
	boolean_t do_xattr = B_FALSE;
	boolean_t atime = B_FALSE;
	boolean_t do_atime = B_FALSE;
	int error = 0;

	ASSERT(vfsp);
	zfsvfs = vfsp->vfs_data;
	ASSERT(zfsvfs);
	os = zfsvfs->z_os;

	/*
	 * The act of registering our callbacks will destroy any mount
	 * options we may have.  In order to enable temporary overrides
	 * of mount options, we stash away the current values and
	 * restore them after we register the callbacks.
	 */
	if (vfs_optionisset(vfsp, MNTOPT_RO, NULL) ||
	    !spa_writeable(dmu_objset_spa(os))) {
		readonly = B_TRUE;
		do_readonly = B_TRUE;
	} else if (vfs_optionisset(vfsp, MNTOPT_RW, NULL)) {
		readonly = B_FALSE;
		do_readonly = B_TRUE;
	}
	if (vfs_optionisset(vfsp, MNTOPT_NOSUID, NULL)) {
		devices = B_FALSE;
		setuid = B_FALSE;
		do_devices = B_TRUE;
		do_setuid = B_TRUE;
	} else {
		if (vfs_optionisset(vfsp, MNTOPT_NODEVICES, NULL)) {
			devices = B_FALSE;
			do_devices = B_TRUE;
		} else if (vfs_optionisset(vfsp, MNTOPT_DEVICES, NULL)) {
			devices = B_TRUE;
			do_devices = B_TRUE;
		}

		if (vfs_optionisset(vfsp, MNTOPT_NOSETUID, NULL)) {
			setuid = B_FALSE;
			do_setuid = B_TRUE;
		} else if (vfs_optionisset(vfsp, MNTOPT_SETUID, NULL)) {
			setuid = B_TRUE;
			do_setuid = B_TRUE;
		}
	}
	if (vfs_optionisset(vfsp, MNTOPT_NOEXEC, NULL)) {
		exec = B_FALSE;
		do_exec = B_TRUE;
	} else if (vfs_optionisset(vfsp, MNTOPT_EXEC, NULL)) {
		exec = B_TRUE;
		do_exec = B_TRUE;
	}
	if (vfs_optionisset(vfsp, MNTOPT_NOXATTR, NULL)) {
		xattr = B_FALSE;
		do_xattr = B_TRUE;
	} else if (vfs_optionisset(vfsp, MNTOPT_XATTR, NULL)) {
		xattr = B_TRUE;
		do_xattr = B_TRUE;
	}
	if (vfs_optionisset(vfsp, MNTOPT_NOATIME, NULL)) {
		atime = B_FALSE;
		do_atime = B_TRUE;
	} else if (vfs_optionisset(vfsp, MNTOPT_ATIME, NULL)) {
		atime = B_TRUE;
		do_atime = B_TRUE;
	}

	/*
	 * nbmand is a special property.  It can only be changed at
	 * mount time.
	 *
	 * This is weird, but it is documented to only be changeable
	 * at mount time.
	 */
	if (vfs_optionisset(vfsp, MNTOPT_NONBMAND, NULL)) {
		nbmand = B_FALSE;
	} else if (vfs_optionisset(vfsp, MNTOPT_NBMAND, NULL)) {
		nbmand = B_TRUE;
	} else {
		char osname[ZFS_MAX_DATASET_NAME_LEN];

		dmu_objset_name(os, osname);
		if (error = dsl_prop_get_integer(osname, "nbmand", &nbmand,
		    NULL)) {
			return (error);
		}
	}

	/*
	 * Register property callbacks.
	 *
	 * It would probably be fine to just check for i/o error from
	 * the first prop_register(), but I guess I like to go
	 * overboard...
	 */
	ds = dmu_objset_ds(os);
	dsl_pool_config_enter(dmu_objset_pool(os), FTAG);
	error = dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_ATIME), atime_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_XATTR), xattr_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_RECORDSIZE), blksz_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_READONLY), readonly_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_DEVICES), devices_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_SETUID), setuid_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_EXEC), exec_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_SNAPDIR), snapdir_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_ACLMODE), acl_mode_changed_cb, zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_ACLINHERIT), acl_inherit_changed_cb,
	    zfsvfs);
	error = error ? error : dsl_prop_register(ds,
	    zfs_prop_to_name(ZFS_PROP_VSCAN), vscan_changed_cb, zfsvfs);
	dsl_pool_config_exit(dmu_objset_pool(os), FTAG);
	if (error)
		goto unregister;

	/*
	 * Invoke our callbacks to restore temporary mount options.
	 */
	if (do_readonly)
		readonly_changed_cb(zfsvfs, readonly);
	if (do_setuid)
		setuid_changed_cb(zfsvfs, setuid);
	if (do_exec)
		exec_changed_cb(zfsvfs, exec);
	if (do_devices)
		devices_changed_cb(zfsvfs, devices);
	if (do_xattr)
		xattr_changed_cb(zfsvfs, xattr);
	if (do_atime)
		atime_changed_cb(zfsvfs, atime);

	nbmand_changed_cb(zfsvfs, nbmand);

	return (0);

unregister:
	dsl_prop_unregister_all(ds, zfsvfs);
	return (error);
}

static int
zfs_space_delta_cb(dmu_object_type_t bonustype, void *data,
    uint64_t *userp, uint64_t *groupp, uint64_t *projectp)
{
	sa_hdr_phys_t sa;
	sa_hdr_phys_t *sap = data;
	uint64_t flags;
	int hdrsize;
	boolean_t swap = B_FALSE;

	/*
	 * Is it a valid type of object to track?
	 */
	if (bonustype != DMU_OT_ZNODE && bonustype != DMU_OT_SA)
		return (SET_ERROR(ENOENT));

	/*
	 * If we have a NULL data pointer
	 * then assume the id's aren't changing and
	 * return EEXIST to the dmu to let it know to
	 * use the same ids
	 */
	if (data == NULL)
		return (SET_ERROR(EEXIST));

	if (bonustype == DMU_OT_ZNODE) {
		znode_phys_t *znp = data;
		*userp = znp->zp_uid;
		*groupp = znp->zp_gid;
		*projectp = ZFS_DEFAULT_PROJID;
		return (0);
	}

	if (sap->sa_magic == 0) {
		/*
		 * This should only happen for newly created files
		 * that haven't had the znode data filled in yet.
		 */
		*userp = 0;
		*groupp = 0;
		*projectp = ZFS_DEFAULT_PROJID;
		return (0);
	}

	sa = *sap;
	if (sa.sa_magic == BSWAP_32(SA_MAGIC)) {
		sa.sa_magic = SA_MAGIC;
		sa.sa_layout_info = BSWAP_16(sa.sa_layout_info);
		swap = B_TRUE;
	} else {
		VERIFY3U(sa.sa_magic, ==, SA_MAGIC);
	}

	hdrsize = sa_hdrsize(&sa);
	VERIFY3U(hdrsize, >=, sizeof (sa_hdr_phys_t));

	*userp = *((uint64_t *)((uintptr_t)data + hdrsize + SA_UID_OFFSET));
	*groupp = *((uint64_t *)((uintptr_t)data + hdrsize + SA_GID_OFFSET));
	flags = *((uint64_t *)((uintptr_t)data + hdrsize + SA_FLAGS_OFFSET));
	if (swap)
		flags = BSWAP_64(flags);

	if (flags & ZFS_PROJID)
		*projectp = *((uint64_t *)((uintptr_t)data + hdrsize +
		    SA_PROJID_OFFSET));
	else
		*projectp = ZFS_DEFAULT_PROJID;

	if (swap) {
		*userp = BSWAP_64(*userp);
		*groupp = BSWAP_64(*groupp);
		*projectp = BSWAP_64(*projectp);
	}
	return (0);
}

static void
fuidstr_to_sid(zfsvfs_t *zfsvfs, const char *fuidstr,
    char *domainbuf, int buflen, uid_t *ridp)
{
	uint64_t fuid;
	const char *domain;

	fuid = zfs_strtonum(fuidstr, NULL);

	domain = zfs_fuid_find_by_idx(zfsvfs, FUID_INDEX(fuid));
	if (domain)
		(void) strlcpy(domainbuf, domain, buflen);
	else
		domainbuf[0] = '\0';
	*ridp = FUID_RID(fuid);
}

static uint64_t
zfs_userquota_prop_to_obj(zfsvfs_t *zfsvfs, zfs_userquota_prop_t type)
{
	switch (type) {
	case ZFS_PROP_USERUSED:
	case ZFS_PROP_USEROBJUSED:
		return (DMU_USERUSED_OBJECT);
	case ZFS_PROP_GROUPUSED:
	case ZFS_PROP_GROUPOBJUSED:
		return (DMU_GROUPUSED_OBJECT);
	case ZFS_PROP_PROJECTUSED:
	case ZFS_PROP_PROJECTOBJUSED:
		return (DMU_PROJECTUSED_OBJECT);
	case ZFS_PROP_USERQUOTA:
		return (zfsvfs->z_userquota_obj);
	case ZFS_PROP_GROUPQUOTA:
		return (zfsvfs->z_groupquota_obj);
	case ZFS_PROP_USEROBJQUOTA:
		return (zfsvfs->z_userobjquota_obj);
	case ZFS_PROP_GROUPOBJQUOTA:
		return (zfsvfs->z_groupobjquota_obj);
	case ZFS_PROP_PROJECTQUOTA:
		return (zfsvfs->z_projectquota_obj);
	case ZFS_PROP_PROJECTOBJQUOTA:
		return (zfsvfs->z_projectobjquota_obj);
	default:
		return (ZFS_NO_OBJECT);
	}
}

int
zfs_userspace_many(zfsvfs_t *zfsvfs, zfs_userquota_prop_t type,
    uint64_t *cookiep, void *vbuf, uint64_t *bufsizep)
{
	int error;
	zap_cursor_t zc;
	zap_attribute_t za;
	zfs_useracct_t *buf = vbuf;
	uint64_t obj;
	int offset = 0;

	if (!dmu_objset_userspace_present(zfsvfs->z_os))
		return (SET_ERROR(ENOTSUP));

	if ((type == ZFS_PROP_PROJECTQUOTA || type == ZFS_PROP_PROJECTUSED ||
	    type == ZFS_PROP_PROJECTOBJQUOTA ||
	    type == ZFS_PROP_PROJECTOBJUSED) &&
	    !dmu_objset_projectquota_present(zfsvfs->z_os))
		return (SET_ERROR(ENOTSUP));

	if ((type == ZFS_PROP_USEROBJUSED || type == ZFS_PROP_GROUPOBJUSED ||
	    type == ZFS_PROP_USEROBJQUOTA || type == ZFS_PROP_GROUPOBJQUOTA ||
	    type == ZFS_PROP_PROJECTOBJUSED ||
	    type == ZFS_PROP_PROJECTOBJQUOTA) &&
	    !dmu_objset_userobjspace_present(zfsvfs->z_os))
		return (SET_ERROR(ENOTSUP));

	obj = zfs_userquota_prop_to_obj(zfsvfs, type);
	if (obj == ZFS_NO_OBJECT) {
		*bufsizep = 0;
		return (0);
	}

	if (type == ZFS_PROP_USEROBJUSED || type == ZFS_PROP_GROUPOBJUSED ||
	    type == ZFS_PROP_PROJECTOBJUSED)
		offset = DMU_OBJACCT_PREFIX_LEN;

	for (zap_cursor_init_serialized(&zc, zfsvfs->z_os, obj, *cookiep);
	    (error = zap_cursor_retrieve(&zc, &za)) == 0;
	    zap_cursor_advance(&zc)) {
		if ((uintptr_t)buf - (uintptr_t)vbuf + sizeof (zfs_useracct_t) >
		    *bufsizep)
			break;

		/*
		 * skip object quota (with zap name prefix DMU_OBJACCT_PREFIX)
		 * when dealing with block quota and vice versa.
		 */
		if ((offset > 0) != (strncmp(za.za_name, DMU_OBJACCT_PREFIX,
		    DMU_OBJACCT_PREFIX_LEN) == 0))
			continue;

		fuidstr_to_sid(zfsvfs, za.za_name + offset,
		    buf->zu_domain, sizeof (buf->zu_domain), &buf->zu_rid);

		buf->zu_space = za.za_first_integer;
		buf++;
	}
	if (error == ENOENT)
		error = 0;

	ASSERT3U((uintptr_t)buf - (uintptr_t)vbuf, <=, *bufsizep);
	*bufsizep = (uintptr_t)buf - (uintptr_t)vbuf;
	*cookiep = zap_cursor_serialize(&zc);
	zap_cursor_fini(&zc);
	return (error);
}

/*
 * buf must be big enough (eg, 32 bytes)
 */
static int
id_to_fuidstr(zfsvfs_t *zfsvfs, const char *domain, uid_t rid,
    char *buf, boolean_t addok)
{
	uint64_t fuid;
	int domainid = 0;

	if (domain && domain[0]) {
		domainid = zfs_fuid_find_by_domain(zfsvfs, domain, NULL, addok);
		if (domainid == -1)
			return (SET_ERROR(ENOENT));
	}
	fuid = FUID_ENCODE(domainid, rid);
	(void) sprintf(buf, "%llx", (longlong_t)fuid);
	return (0);
}

int
zfs_userspace_one(zfsvfs_t *zfsvfs, zfs_userquota_prop_t type,
    const char *domain, uint64_t rid, uint64_t *valp)
{
	char buf[20 + DMU_OBJACCT_PREFIX_LEN];
	int offset = 0;
	int err;
	uint64_t obj;

	*valp = 0;

	if (!dmu_objset_userspace_present(zfsvfs->z_os))
		return (SET_ERROR(ENOTSUP));

	if ((type == ZFS_PROP_USEROBJUSED || type == ZFS_PROP_GROUPOBJUSED ||
	    type == ZFS_PROP_USEROBJQUOTA || type == ZFS_PROP_GROUPOBJQUOTA ||
	    type == ZFS_PROP_PROJECTOBJUSED ||
	    type == ZFS_PROP_PROJECTOBJQUOTA) &&
	    !dmu_objset_userobjspace_present(zfsvfs->z_os))
		return (SET_ERROR(ENOTSUP));

	if (type == ZFS_PROP_PROJECTQUOTA || type == ZFS_PROP_PROJECTUSED ||
	    type == ZFS_PROP_PROJECTOBJQUOTA ||
	    type == ZFS_PROP_PROJECTOBJUSED) {
		if (!dmu_objset_projectquota_present(zfsvfs->z_os))
			return (SET_ERROR(ENOTSUP));
		if (!zpl_is_valid_projid(rid))
			return (SET_ERROR(EINVAL));
	}

	obj = zfs_userquota_prop_to_obj(zfsvfs, type);
	if (obj == ZFS_NO_OBJECT)
		return (0);

	if (type == ZFS_PROP_USEROBJUSED || type == ZFS_PROP_GROUPOBJUSED ||
	    type == ZFS_PROP_PROJECTOBJUSED) {
		strncpy(buf, DMU_OBJACCT_PREFIX, DMU_OBJACCT_PREFIX_LEN);
		offset = DMU_OBJACCT_PREFIX_LEN;
	}

	err = id_to_fuidstr(zfsvfs, domain, rid, buf + offset, B_FALSE);
	if (err)
		return (err);

	err = zap_lookup(zfsvfs->z_os, obj, buf, 8, 1, valp);
	if (err == ENOENT)
		err = 0;
	return (err);
}

int
zfs_set_userquota(zfsvfs_t *zfsvfs, zfs_userquota_prop_t type,
    const char *domain, uint64_t rid, uint64_t quota)
{
	char buf[32];
	int err;
	dmu_tx_t *tx;
	uint64_t *objp;
	boolean_t fuid_dirtied;

	if (zfsvfs->z_version < ZPL_VERSION_USERSPACE)
		return (SET_ERROR(ENOTSUP));

	switch (type) {
	case ZFS_PROP_USERQUOTA:
		objp = &zfsvfs->z_userquota_obj;
		break;
	case ZFS_PROP_GROUPQUOTA:
		objp = &zfsvfs->z_groupquota_obj;
		break;
	case ZFS_PROP_USEROBJQUOTA:
		objp = &zfsvfs->z_userobjquota_obj;
		break;
	case ZFS_PROP_GROUPOBJQUOTA:
		objp = &zfsvfs->z_groupobjquota_obj;
		break;
	case ZFS_PROP_PROJECTQUOTA:
		if (!dmu_objset_projectquota_enabled(zfsvfs->z_os))
			return (SET_ERROR(ENOTSUP));
		if (!zpl_is_valid_projid(rid))
			return (SET_ERROR(EINVAL));

		objp = &zfsvfs->z_projectquota_obj;
		break;
	case ZFS_PROP_PROJECTOBJQUOTA:
		if (!dmu_objset_projectquota_enabled(zfsvfs->z_os))
			return (SET_ERROR(ENOTSUP));
		if (!zpl_is_valid_projid(rid))
			return (SET_ERROR(EINVAL));

		objp = &zfsvfs->z_projectobjquota_obj;
		break;
	default:
		return (SET_ERROR(EINVAL));
	}

	err = id_to_fuidstr(zfsvfs, domain, rid, buf, B_TRUE);
	if (err)
		return (err);
	fuid_dirtied = zfsvfs->z_fuid_dirty;

	tx = dmu_tx_create(zfsvfs->z_os);
	dmu_tx_hold_zap(tx, *objp ? *objp : DMU_NEW_OBJECT, B_TRUE, NULL);
	if (*objp == 0) {
		dmu_tx_hold_zap(tx, MASTER_NODE_OBJ, B_TRUE,
		    zfs_userquota_prop_prefixes[type]);
	}
	if (fuid_dirtied)
		zfs_fuid_txhold(zfsvfs, tx);
	err = dmu_tx_assign(tx, TXG_WAIT);
	if (err) {
		dmu_tx_abort(tx);
		return (err);
	}

	mutex_enter(&zfsvfs->z_lock);
	if (*objp == 0) {
		*objp = zap_create(zfsvfs->z_os, DMU_OT_USERGROUP_QUOTA,
		    DMU_OT_NONE, 0, tx);
		VERIFY(0 == zap_add(zfsvfs->z_os, MASTER_NODE_OBJ,
		    zfs_userquota_prop_prefixes[type], 8, 1, objp, tx));
	}
	mutex_exit(&zfsvfs->z_lock);

	if (quota == 0) {
		err = zap_remove(zfsvfs->z_os, *objp, buf, tx);
		if (err == ENOENT)
			err = 0;
	} else {
		err = zap_update(zfsvfs->z_os, *objp, buf, 8, 1, &quota, tx);
	}
	ASSERT(err == 0);
	if (fuid_dirtied)
		zfs_fuid_sync(zfsvfs, tx);
	dmu_tx_commit(tx);
	return (err);
}

boolean_t
zfs_id_overobjquota(zfsvfs_t *zfsvfs, uint64_t usedobj, uint64_t id)
{
	char buf[20 + DMU_OBJACCT_PREFIX_LEN];
	uint64_t used, quota, quotaobj;
	int err;

	if (!dmu_objset_userobjspace_present(zfsvfs->z_os)) {
		if (dmu_objset_userobjspace_upgradable(zfsvfs->z_os)) {
			dsl_pool_config_enter(
			    dmu_objset_pool(zfsvfs->z_os), FTAG);
			dmu_objset_id_quota_upgrade(zfsvfs->z_os);
			dsl_pool_config_exit(
			    dmu_objset_pool(zfsvfs->z_os), FTAG);
		}
		return (B_FALSE);
	}

	if (usedobj == DMU_PROJECTUSED_OBJECT) {
		if (!dmu_objset_projectquota_present(zfsvfs->z_os)) {
			if (dmu_objset_projectquota_upgradable(zfsvfs->z_os)) {
				dsl_pool_config_enter(
				    dmu_objset_pool(zfsvfs->z_os), FTAG);
				dmu_objset_id_quota_upgrade(zfsvfs->z_os);
				dsl_pool_config_exit(
				    dmu_objset_pool(zfsvfs->z_os), FTAG);
			}
			return (B_FALSE);
		}
		quotaobj = zfsvfs->z_projectobjquota_obj;
	} else if (usedobj == DMU_USERUSED_OBJECT) {
		quotaobj = zfsvfs->z_userobjquota_obj;
	} else if (usedobj == DMU_GROUPUSED_OBJECT) {
		quotaobj = zfsvfs->z_groupobjquota_obj;
	} else {
		return (B_FALSE);
	}
	if (quotaobj == 0 || zfsvfs->z_replay)
		return (B_FALSE);

	(void) sprintf(buf, "%llx", (longlong_t)id);
	err = zap_lookup(zfsvfs->z_os, quotaobj, buf, 8, 1, &quota);
	if (err != 0)
		return (B_FALSE);

	(void) sprintf(buf, DMU_OBJACCT_PREFIX "%llx", (longlong_t)id);
	err = zap_lookup(zfsvfs->z_os, usedobj, buf, 8, 1, &used);
	if (err != 0)
		return (B_FALSE);
	return (used >= quota);
}

boolean_t
zfs_id_overblockquota(zfsvfs_t *zfsvfs, uint64_t usedobj, uint64_t id)
{
	char buf[20];
	uint64_t used, quota, quotaobj;
	int err;

	if (usedobj == DMU_PROJECTUSED_OBJECT) {
		if (!dmu_objset_projectquota_present(zfsvfs->z_os)) {
			if (dmu_objset_projectquota_upgradable(zfsvfs->z_os)) {
				dsl_pool_config_enter(
				    dmu_objset_pool(zfsvfs->z_os), FTAG);
				dmu_objset_id_quota_upgrade(zfsvfs->z_os);
				dsl_pool_config_exit(
				    dmu_objset_pool(zfsvfs->z_os), FTAG);
			}
			return (B_FALSE);
		}
		quotaobj = zfsvfs->z_projectquota_obj;
	} else if (usedobj == DMU_USERUSED_OBJECT) {
		quotaobj = zfsvfs->z_userquota_obj;
	} else if (usedobj == DMU_GROUPUSED_OBJECT) {
		quotaobj = zfsvfs->z_groupquota_obj;
	} else {
		return (B_FALSE);
	}
	if (quotaobj == 0 || zfsvfs->z_replay)
		return (B_FALSE);

	(void) sprintf(buf, "%llx", (longlong_t)id);
	err = zap_lookup(zfsvfs->z_os, quotaobj, buf, 8, 1, &quota);
	if (err != 0)
		return (B_FALSE);

	err = zap_lookup(zfsvfs->z_os, usedobj, buf, 8, 1, &used);
	if (err != 0)
		return (B_FALSE);
	return (used >= quota);
}

boolean_t
zfs_id_overquota(zfsvfs_t *zfsvfs, uint64_t usedobj, uint64_t id)
{
	return (zfs_id_overblockquota(zfsvfs, usedobj, id) ||
	    zfs_id_overobjquota(zfsvfs, usedobj, id));
}

/*
 * Associate this zfsvfs with the given objset, which must be owned.
 * This will cache a bunch of on-disk state from the objset in the
 * zfsvfs.
 */
static int
zfsvfs_init(zfsvfs_t *zfsvfs, objset_t *os)
{
	int error;
	uint64_t val;

	zfsvfs->z_max_blksz = SPA_OLD_MAXBLOCKSIZE;
	zfsvfs->z_show_ctldir = ZFS_SNAPDIR_VISIBLE;
	zfsvfs->z_os = os;

	error = zfs_get_zplprop(os, ZFS_PROP_VERSION, &zfsvfs->z_version);
	if (error != 0)
		return (error);
	if (zfsvfs->z_version >
	    zfs_zpl_version_map(spa_version(dmu_objset_spa(os)))) {
		(void) printf("Can't mount a version %lld file system "
		    "on a version %lld pool\n. Pool must be upgraded to mount "
		    "this file system.", (u_longlong_t)zfsvfs->z_version,
		    (u_longlong_t)spa_version(dmu_objset_spa(os)));
		return (SET_ERROR(ENOTSUP));
	}
	error = zfs_get_zplprop(os, ZFS_PROP_NORMALIZE, &val);
	if (error != 0)
		return (error);
	zfsvfs->z_norm = (int)val;

	error = zfs_get_zplprop(os, ZFS_PROP_UTF8ONLY, &val);
	if (error != 0)
		return (error);
	zfsvfs->z_utf8 = (val != 0);

	error = zfs_get_zplprop(os, ZFS_PROP_CASE, &val);
	if (error != 0)
		return (error);
	zfsvfs->z_case = (uint_t)val;

	/*
	 * Fold case on file systems that are always or sometimes case
	 * insensitive.
	 */
	if (zfsvfs->z_case == ZFS_CASE_INSENSITIVE ||
	    zfsvfs->z_case == ZFS_CASE_MIXED)
		zfsvfs->z_norm |= U8_TEXTPREP_TOUPPER;

	zfsvfs->z_use_fuids = USE_FUIDS(zfsvfs->z_version, zfsvfs->z_os);
	zfsvfs->z_use_sa = USE_SA(zfsvfs->z_version, zfsvfs->z_os);

	uint64_t sa_obj = 0;
	if (zfsvfs->z_use_sa) {
		/* should either have both of these objects or none */
		error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_SA_ATTRS, 8, 1,
		    &sa_obj);
		if (error != 0)
			return (error);
	}

	error = sa_setup(os, sa_obj, zfs_attr_table, ZPL_END,
	    &zfsvfs->z_attr_table);
	if (error != 0)
		return (error);

	if (zfsvfs->z_version >= ZPL_VERSION_SA)
		sa_register_update_callback(os, zfs_sa_upgrade);

	error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_ROOT_OBJ, 8, 1,
	    &zfsvfs->z_root);
	if (error != 0)
		return (error);
	ASSERT(zfsvfs->z_root != 0);

	error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_UNLINKED_SET, 8, 1,
	    &zfsvfs->z_unlinkedobj);
	if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ,
	    zfs_userquota_prop_prefixes[ZFS_PROP_USERQUOTA],
	    8, 1, &zfsvfs->z_userquota_obj);
	if (error == ENOENT)
		zfsvfs->z_userquota_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ,
	    zfs_userquota_prop_prefixes[ZFS_PROP_GROUPQUOTA],
	    8, 1, &zfsvfs->z_groupquota_obj);
	if (error == ENOENT)
		zfsvfs->z_groupquota_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ,
	    zfs_userquota_prop_prefixes[ZFS_PROP_PROJECTQUOTA],
	    8, 1, &zfsvfs->z_projectquota_obj);
	if (error == ENOENT)
		zfsvfs->z_projectquota_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ,
	    zfs_userquota_prop_prefixes[ZFS_PROP_USEROBJQUOTA],
	    8, 1, &zfsvfs->z_userobjquota_obj);
	if (error == ENOENT)
		zfsvfs->z_userobjquota_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ,
	    zfs_userquota_prop_prefixes[ZFS_PROP_GROUPOBJQUOTA],
	    8, 1, &zfsvfs->z_groupobjquota_obj);
	if (error == ENOENT)
		zfsvfs->z_groupobjquota_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ,
	    zfs_userquota_prop_prefixes[ZFS_PROP_PROJECTOBJQUOTA],
	    8, 1, &zfsvfs->z_projectobjquota_obj);
	if (error == ENOENT)
		zfsvfs->z_projectobjquota_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_FUID_TABLES, 8, 1,
	    &zfsvfs->z_fuid_obj);
	if (error == ENOENT)
		zfsvfs->z_fuid_obj = 0;
	else if (error != 0)
		return (error);

	error = zap_lookup(os, MASTER_NODE_OBJ, ZFS_SHARES_DIR, 8, 1,
	    &zfsvfs->z_shares_dir);
	if (error == ENOENT)
		zfsvfs->z_shares_dir = 0;
	else if (error != 0)
		return (error);

	return (0);
}

int
zfsvfs_create(const char *osname, boolean_t readonly, zfsvfs_t **zfvp)
{
	objset_t *os;
	zfsvfs_t *zfsvfs;
	int error;
	boolean_t ro = (readonly || (strchr(osname, '@') != NULL));

	zfsvfs = kmem_zalloc(sizeof (zfsvfs_t), KM_SLEEP);

	error = dmu_objset_own(osname, DMU_OST_ZFS, ro, B_TRUE, zfsvfs, &os);
	if (error != 0) {
		kmem_free(zfsvfs, sizeof (zfsvfs_t));
		return (error);
	}

	error = zfsvfs_create_impl(zfvp, zfsvfs, os);
	if (error != 0) {
		dmu_objset_disown(os, B_TRUE, zfsvfs);
	}
	return (error);
}


int
zfsvfs_create_impl(zfsvfs_t **zfvp, zfsvfs_t *zfsvfs, objset_t *os)
{
	int error;

	zfsvfs->z_vfs = NULL;
	zfsvfs->z_parent = zfsvfs;

	mutex_init(&zfsvfs->z_znodes_lock, NULL, MUTEX_DEFAULT, NULL);
	mutex_init(&zfsvfs->z_lock, NULL, MUTEX_DEFAULT, NULL);
	list_create(&zfsvfs->z_all_znodes, sizeof (znode_t),
	    offsetof(znode_t, z_link_node));
	rrm_init(&zfsvfs->z_teardown_lock, B_FALSE);
	rw_init(&zfsvfs->z_teardown_inactive_lock, NULL, RW_DEFAULT, NULL);
	rw_init(&zfsvfs->z_fuid_lock, NULL, RW_DEFAULT, NULL);
	for (int i = 0; i != ZFS_OBJ_MTX_SZ; i++)
		mutex_init(&zfsvfs->z_hold_mtx[i], NULL, MUTEX_DEFAULT, NULL);

	error = zfsvfs_init(zfsvfs, os);
	if (error != 0) {
		*zfvp = NULL;
		kmem_free(zfsvfs, sizeof (zfsvfs_t));
		return (error);
	}

	zfsvfs->z_drain_task = TASKQID_INVALID;
	zfsvfs->z_draining = B_FALSE;
	zfsvfs->z_drain_cancel = B_TRUE;

	*zfvp = zfsvfs;
	return (0);
}

static int
zfsvfs_setup(zfsvfs_t *zfsvfs, boolean_t mounting)
{
	int error;

	error = zfs_register_callbacks(zfsvfs->z_vfs);
	if (error)
		return (error);

	zfsvfs->z_log = zil_open(zfsvfs->z_os, zfs_get_data);

	/*
	 * If we are not mounting (ie: online recv), then we don't
	 * have to worry about replaying the log as we blocked all
	 * operations out since we closed the ZIL.
	 */
	if (mounting) {
		boolean_t readonly;

		/*
		 * During replay we remove the read only flag to
		 * allow replays to succeed.
		 */
		readonly = zfsvfs->z_vfs->vfs_flag & VFS_RDONLY;
		if (readonly != 0) {
			zfsvfs->z_vfs->vfs_flag &= ~VFS_RDONLY;
		} else {
			zfs_unlinked_drain(zfsvfs);
		}

		/*
		 * Parse and replay the intent log.
		 *
		 * Because of ziltest, this must be done after
		 * zfs_unlinked_drain().  (Further note: ziltest
		 * doesn't use readonly mounts, where
		 * zfs_unlinked_drain() isn't called.)  This is because
		 * ziltest causes spa_sync() to think it's committed,
		 * but actually it is not, so the intent log contains
		 * many txg's worth of changes.
		 *
		 * In particular, if object N is in the unlinked set in
		 * the last txg to actually sync, then it could be
		 * actually freed in a later txg and then reallocated
		 * in a yet later txg.  This would write a "create
		 * object N" record to the intent log.  Normally, this
		 * would be fine because the spa_sync() would have
		 * written out the fact that object N is free, before
		 * we could write the "create object N" intent log
		 * record.
		 *
		 * But when we are in ziltest mode, we advance the "open
		 * txg" without actually spa_sync()-ing the changes to
		 * disk.  So we would see that object N is still
		 * allocated and in the unlinked set, and there is an
		 * intent log record saying to allocate it.
		 */
		if (spa_writeable(dmu_objset_spa(zfsvfs->z_os))) {
			if (zil_replay_disable) {
				zil_destroy(zfsvfs->z_log, B_FALSE);
			} else {
				zfsvfs->z_replay = B_TRUE;
				zil_replay(zfsvfs->z_os, zfsvfs,
				    zfs_replay_vector);
				zfsvfs->z_replay = B_FALSE;
			}
		}

		/* restore readonly bit */
		if (readonly != 0)
			zfsvfs->z_vfs->vfs_flag |= VFS_RDONLY;
	}

	/*
	 * Set the objset user_ptr to track its zfsvfs.
	 */
	mutex_enter(&zfsvfs->z_os->os_user_ptr_lock);
	dmu_objset_set_user(zfsvfs->z_os, zfsvfs);
	mutex_exit(&zfsvfs->z_os->os_user_ptr_lock);

	return (0);
}

void
zfsvfs_free(zfsvfs_t *zfsvfs)
{
	int i;
	extern krwlock_t zfsvfs_lock; /* in zfs_znode.c */

	/*
	 * This is a barrier to prevent the filesystem from going away in
	 * zfs_znode_move() until we can safely ensure that the filesystem is
	 * not unmounted. We consider the filesystem valid before the barrier
	 * and invalid after the barrier.
	 */
	rw_enter(&zfsvfs_lock, RW_READER);
	rw_exit(&zfsvfs_lock);

	zfs_fuid_destroy(zfsvfs);

	mutex_destroy(&zfsvfs->z_znodes_lock);
	mutex_destroy(&zfsvfs->z_lock);
	list_destroy(&zfsvfs->z_all_znodes);
	rrm_destroy(&zfsvfs->z_teardown_lock);
	rw_destroy(&zfsvfs->z_teardown_inactive_lock);
	rw_destroy(&zfsvfs->z_fuid_lock);
	for (i = 0; i != ZFS_OBJ_MTX_SZ; i++)
		mutex_destroy(&zfsvfs->z_hold_mtx[i]);
	kmem_free(zfsvfs, sizeof (zfsvfs_t));
}

static void
zfs_set_fuid_feature(zfsvfs_t *zfsvfs)
{
	zfsvfs->z_use_fuids = USE_FUIDS(zfsvfs->z_version, zfsvfs->z_os);
	if (zfsvfs->z_vfs) {
		if (zfsvfs->z_use_fuids) {
			vfs_set_feature(zfsvfs->z_vfs, VFSFT_XVATTR);
			vfs_set_feature(zfsvfs->z_vfs, VFSFT_SYSATTR_VIEWS);
			vfs_set_feature(zfsvfs->z_vfs, VFSFT_ACEMASKONACCESS);
			vfs_set_feature(zfsvfs->z_vfs, VFSFT_ACLONCREATE);
			vfs_set_feature(zfsvfs->z_vfs, VFSFT_ACCESS_FILTER);
			vfs_set_feature(zfsvfs->z_vfs, VFSFT_REPARSE);
		} else {
			vfs_clear_feature(zfsvfs->z_vfs, VFSFT_XVATTR);
			vfs_clear_feature(zfsvfs->z_vfs, VFSFT_SYSATTR_VIEWS);
			vfs_clear_feature(zfsvfs->z_vfs, VFSFT_ACEMASKONACCESS);
			vfs_clear_feature(zfsvfs->z_vfs, VFSFT_ACLONCREATE);
			vfs_clear_feature(zfsvfs->z_vfs, VFSFT_ACCESS_FILTER);
			vfs_clear_feature(zfsvfs->z_vfs, VFSFT_REPARSE);
		}
	}
	zfsvfs->z_use_sa = USE_SA(zfsvfs->z_version, zfsvfs->z_os);
}

static int
zfs_domount(vfs_t *vfsp, char *osname)
{
	dev_t mount_dev;
	uint64_t recordsize, fsid_guid;
	int error = 0;
	zfsvfs_t *zfsvfs;
	boolean_t readonly = vfsp->vfs_flag & VFS_RDONLY ? B_TRUE : B_FALSE;

	ASSERT(vfsp);
	ASSERT(osname);

	error = zfsvfs_create(osname, readonly, &zfsvfs);
	if (error)
		return (error);
	zfsvfs->z_vfs = vfsp;

	/* Initialize the generic filesystem structure. */
	vfsp->vfs_bcount = 0;
	vfsp->vfs_data = NULL;

	if (zfs_create_unique_device(&mount_dev) == -1) {
		error = SET_ERROR(ENODEV);
		goto out;
	}
	ASSERT(vfs_devismounted(mount_dev) == 0);

	if (error = dsl_prop_get_integer(osname, "recordsize", &recordsize,
	    NULL))
		goto out;

	vfsp->vfs_dev = mount_dev;
	vfsp->vfs_fstype = zfsfstype;
	vfsp->vfs_bsize = recordsize;
	vfsp->vfs_flag |= VFS_NOTRUNC;
	vfsp->vfs_data = zfsvfs;

	/*
	 * The fsid is 64 bits, composed of an 8-bit fs type, which
	 * separates our fsid from any other filesystem types, and a
	 * 56-bit objset unique ID.  The objset unique ID is unique to
	 * all objsets open on this system, provided by unique_create().
	 * The 8-bit fs type must be put in the low bits of fsid[1]
	 * because that's where other Solaris filesystems put it.
	 */
	fsid_guid = dmu_objset_fsid_guid(zfsvfs->z_os);
	ASSERT((fsid_guid & ~((1ULL<<56)-1)) == 0);
	vfsp->vfs_fsid.val[0] = fsid_guid;
	vfsp->vfs_fsid.val[1] = ((fsid_guid>>32) << 8) |
	    zfsfstype & 0xFF;

	/*
	 * Set features for file system.
	 */
	zfs_set_fuid_feature(zfsvfs);
	if (zfsvfs->z_case == ZFS_CASE_INSENSITIVE) {
		vfs_set_feature(vfsp, VFSFT_DIRENTFLAGS);
		vfs_set_feature(vfsp, VFSFT_CASEINSENSITIVE);
		vfs_set_feature(vfsp, VFSFT_NOCASESENSITIVE);
	} else if (zfsvfs->z_case == ZFS_CASE_MIXED) {
		vfs_set_feature(vfsp, VFSFT_DIRENTFLAGS);
		vfs_set_feature(vfsp, VFSFT_CASEINSENSITIVE);
	}
	vfs_set_feature(vfsp, VFSFT_ZEROCOPY_SUPPORTED);

	if (dmu_objset_is_snapshot(zfsvfs->z_os)) {
		uint64_t pval;

		atime_changed_cb(zfsvfs, B_FALSE);
		readonly_changed_cb(zfsvfs, B_TRUE);
		if (error = dsl_prop_get_integer(osname, "xattr", &pval, NULL))
			goto out;
		xattr_changed_cb(zfsvfs, pval);
		zfsvfs->z_issnap = B_TRUE;
		zfsvfs->z_os->os_sync = ZFS_SYNC_DISABLED;

		mutex_enter(&zfsvfs->z_os->os_user_ptr_lock);
		dmu_objset_set_user(zfsvfs->z_os, zfsvfs);
		mutex_exit(&zfsvfs->z_os->os_user_ptr_lock);
	} else {
		error = zfsvfs_setup(zfsvfs, B_TRUE);
	}

	/* cache the root vnode for this mount */
	znode_t *rootzp;
	if (error = zfs_zget(zfsvfs, zfsvfs->z_root, &rootzp)) {
		goto out;
	}
	zfsvfs->z_rootdir = ZTOV(rootzp);

	if (!zfsvfs->z_issnap)
		zfsctl_create(zfsvfs);
out:
	if (error) {
		dmu_objset_disown(zfsvfs->z_os, B_TRUE, zfsvfs);
		zfsvfs_free(zfsvfs);
	} else {
		atomic_inc_32(&zfs_active_fs_count);
	}

	return (error);
}

void
zfs_unregister_callbacks(zfsvfs_t *zfsvfs)
{
	objset_t *os = zfsvfs->z_os;

	if (!dmu_objset_is_snapshot(os))
		dsl_prop_unregister_all(dmu_objset_ds(os), zfsvfs);
}

/*
 * Convert a decimal digit string to a uint64_t integer.
 */
static int
str_to_uint64(char *str, uint64_t *objnum)
{
	uint64_t num = 0;

	while (*str) {
		if (*str < '0' || *str > '9')
			return (SET_ERROR(EINVAL));

		num = num*10 + *str++ - '0';
	}

	*objnum = num;
	return (0);
}

/*
 * The boot path passed from the boot loader is in the form of
 * "rootpool-name/root-filesystem-object-number'. Convert this
 * string to a dataset name: "rootpool-name/root-filesystem-name".
 */
static int
zfs_parse_bootfs(char *bpath, char *outpath)
{
	char *slashp;
	uint64_t objnum;
	int error;

	if (*bpath == 0 || *bpath == '/')
		return (SET_ERROR(EINVAL));

	(void) strcpy(outpath, bpath);

	slashp = strchr(bpath, '/');

	/* if no '/', just return the pool name */
	if (slashp == NULL) {
		return (0);
	}

	/* if not a number, just return the root dataset name */
	if (str_to_uint64(slashp+1, &objnum)) {
		return (0);
	}

	*slashp = '\0';
	error = dsl_dsobj_to_dsname(bpath, objnum, outpath);
	*slashp = '/';

	return (error);
}

/*
 * Check that the hex label string is appropriate for the dataset being
 * mounted into the global_zone proper.
 *
 * Return an error if the hex label string is not default or
 * admin_low/admin_high.  For admin_low labels, the corresponding
 * dataset must be readonly.
 */
int
zfs_check_global_label(const char *dsname, const char *hexsl)
{
	if (strcasecmp(hexsl, ZFS_MLSLABEL_DEFAULT) == 0)
		return (0);
	if (strcasecmp(hexsl, ADMIN_HIGH) == 0)
		return (0);
	if (strcasecmp(hexsl, ADMIN_LOW) == 0) {
		/* must be readonly */
		uint64_t rdonly;

		if (dsl_prop_get_integer(dsname,
		    zfs_prop_to_name(ZFS_PROP_READONLY), &rdonly, NULL))
			return (SET_ERROR(EACCES));
		return (rdonly ? 0 : EACCES);
	}
	return (SET_ERROR(EACCES));
}

static int
zfs_statfs_project(zfsvfs_t *zfsvfs, znode_t *zp, struct statvfs64 *statp,
    uint32_t bshift)
{
	char buf[20 + DMU_OBJACCT_PREFIX_LEN];
	uint64_t offset = DMU_OBJACCT_PREFIX_LEN;
	uint64_t quota;
	uint64_t used;
	int err;

	strlcpy(buf, DMU_OBJACCT_PREFIX, DMU_OBJACCT_PREFIX_LEN + 1);
	err = id_to_fuidstr(zfsvfs, NULL, zp->z_projid, buf + offset, B_FALSE);
	if (err)
		return (err);

	if (zfsvfs->z_projectquota_obj == 0)
		goto objs;

	err = zap_lookup(zfsvfs->z_os, zfsvfs->z_projectquota_obj,
	    buf + offset, 8, 1, &quota);
	if (err == ENOENT)
		goto objs;
	else if (err)
		return (err);

	err = zap_lookup(zfsvfs->z_os, DMU_PROJECTUSED_OBJECT,
	    buf + offset, 8, 1, &used);
	if (unlikely(err == ENOENT)) {
		uint32_t blksize;
		u_longlong_t nblocks;

		/*
		 * Quota accounting is async, so it is possible race case.
		 * There is at least one object with the given project ID.
		 */
		sa_object_size(zp->z_sa_hdl, &blksize, &nblocks);
		if (unlikely(zp->z_blksz == 0))
			blksize = zfsvfs->z_max_blksz;

		used = blksize * nblocks;
	} else if (err) {
		return (err);
	}

	statp->f_blocks = quota >> bshift;
	statp->f_bfree = (quota > used) ? ((quota - used) >> bshift) : 0;
	statp->f_bavail = statp->f_bfree;

objs:
	if (zfsvfs->z_projectobjquota_obj == 0)
		return (0);

	err = zap_lookup(zfsvfs->z_os, zfsvfs->z_projectobjquota_obj,
	    buf + offset, 8, 1, &quota);
	if (err == ENOENT)
		return (0);
	else if (err)
		return (err);

	err = zap_lookup(zfsvfs->z_os, DMU_PROJECTUSED_OBJECT,
	    buf, 8, 1, &used);
	if (unlikely(err == ENOENT)) {
		/*
		 * Quota accounting is async, so it is possible race case.
		 * There is at least one object with the given project ID.
		 */
		used = 1;
	} else if (err) {
		return (err);
	}

	statp->f_files = quota;
	statp->f_ffree = (quota > used) ? (quota - used) : 0;

	return (0);
}

/*
 * Determine whether the mount is allowed according to MAC check.
 * by comparing (where appropriate) label of the dataset against
 * the label of the zone being mounted into.  If the dataset has
 * no label, create one.
 *
 * Returns 0 if access allowed, error otherwise (e.g. EACCES)
 */
static int
zfs_mount_label_policy(vfs_t *vfsp, char *osname)
{
	int		error, retv;
	zone_t		*mntzone = NULL;
	ts_label_t	*mnt_tsl;
	bslabel_t	*mnt_sl;
	bslabel_t	ds_sl;
	char		ds_hexsl[MAXNAMELEN];

	retv = EACCES;				/* assume the worst */

	/*
	 * Start by getting the dataset label if it exists.
	 */
	error = dsl_prop_get(osname, zfs_prop_to_name(ZFS_PROP_MLSLABEL),
	    1, sizeof (ds_hexsl), &ds_hexsl, NULL);
	if (error)
		return (SET_ERROR(EACCES));

	/*
	 * If labeling is NOT enabled, then disallow the mount of datasets
	 * which have a non-default label already.  No other label checks
	 * are needed.
	 */
	if (!is_system_labeled()) {
		if (strcasecmp(ds_hexsl, ZFS_MLSLABEL_DEFAULT) == 0)
			return (0);
		return (SET_ERROR(EACCES));
	}

	/*
	 * Get the label of the mountpoint.  If mounting into the global
	 * zone (i.e. mountpoint is not within an active zone and the
	 * zoned property is off), the label must be default or
	 * admin_low/admin_high only; no other checks are needed.
	 */
	mntzone = zone_find_by_any_path(refstr_value(vfsp->vfs_mntpt), B_FALSE);
	if (mntzone->zone_id == GLOBAL_ZONEID) {
		uint64_t zoned;

		zone_rele(mntzone);

		if (dsl_prop_get_integer(osname,
		    zfs_prop_to_name(ZFS_PROP_ZONED), &zoned, NULL))
			return (SET_ERROR(EACCES));
		if (!zoned)
			return (zfs_check_global_label(osname, ds_hexsl));
		else
			/*
			 * This is the case of a zone dataset being mounted
			 * initially, before the zone has been fully created;
			 * allow this mount into global zone.
			 */
			return (0);
	}

	mnt_tsl = mntzone->zone_slabel;
	ASSERT(mnt_tsl != NULL);
	label_hold(mnt_tsl);
	mnt_sl = label2bslabel(mnt_tsl);

	if (strcasecmp(ds_hexsl, ZFS_MLSLABEL_DEFAULT) == 0) {
		/*
		 * The dataset doesn't have a real label, so fabricate one.
		 */
		char *str = NULL;

		if (l_to_str_internal(mnt_sl, &str) == 0 &&
		    dsl_prop_set_string(osname,
		    zfs_prop_to_name(ZFS_PROP_MLSLABEL),
		    ZPROP_SRC_LOCAL, str) == 0)
			retv = 0;
		if (str != NULL)
			kmem_free(str, strlen(str) + 1);
	} else if (hexstr_to_label(ds_hexsl, &ds_sl) == 0) {
		/*
		 * Now compare labels to complete the MAC check.  If the
		 * labels are equal then allow access.  If the mountpoint
		 * label dominates the dataset label, allow readonly access.
		 * Otherwise, access is denied.
		 */
		if (blequal(mnt_sl, &ds_sl))
			retv = 0;
		else if (bldominates(mnt_sl, &ds_sl)) {
			vfs_setmntopt(vfsp, MNTOPT_RO, NULL, 0);
			retv = 0;
		}
	}

	label_rele(mnt_tsl);
	zone_rele(mntzone);
	return (retv);
}

/*
 * Load a string-valued boot property and attempt to convert it to a 64-bit
 * unsigned integer.  If the value is not present, or the conversion fails,
 * return the provided default value.
 */
static uint64_t
spa_get_bootprop_uint64(const char *name, uint64_t defval)
{
	char *propval;
	u_longlong_t r;
	int e;

	if ((propval = spa_get_bootprop(name)) == NULL) {
		/*
		 * The property does not exist.
		 */
		return (defval);
	}

	e = ddi_strtoull(propval, NULL, 10, &r);

	spa_free_bootprop(propval);

	/*
	 * If the conversion succeeded, return the value.  If there was any
	 * kind of failure, just return the default value.
	 */
	return (e == 0 ? r : defval);
}

static int
zfs_mountroot(vfs_t *vfsp, enum whymountroot why)
{
	int error = 0;
	static int zfsrootdone = 0;
	zfsvfs_t *zfsvfs = NULL;
	znode_t *zp = NULL;
	vnode_t *vp = NULL;
	char *zfs_bootfs;
	char *zfs_devid;
	char *zfs_rootdisk_path;
	uint64_t zfs_bootpool;
	uint64_t zfs_bootvdev;

	ASSERT(vfsp);

	/*
	 * The filesystem that we mount as root is defined in the
	 * boot property "zfs-bootfs" with a format of
	 * "poolname/root-dataset-objnum".
	 */
	if (why == ROOT_INIT) {
		if (zfsrootdone++)
			return (SET_ERROR(EBUSY));

		/*
		 * the process of doing a spa_load will require the
		 * clock to be set before we could (for example) do
		 * something better by looking at the timestamp on
		 * an uberblock, so just set it to -1.
		 */
		clkset(-1);

		if ((zfs_bootfs = spa_get_bootprop("zfs-bootfs")) == NULL) {
			cmn_err(CE_NOTE, "spa_get_bootfs: can not get "
			    "bootfs name");
			return (SET_ERROR(EINVAL));
		}
		zfs_devid = spa_get_bootprop("diskdevid");

		/*
		 * The boot loader may also provide us with the GUID for both
		 * the pool and the nominated boot vdev.  A GUID value of 0 is
		 * explicitly invalid (see "spa_change_guid()"), so we use this
		 * as a sentinel value when no GUID is present.
		 */
		zfs_bootpool = spa_get_bootprop_uint64("zfs-bootpool", 0);
		zfs_bootvdev = spa_get_bootprop_uint64("zfs-bootvdev", 0);

		/*
		 * If we have been given a root disk override path, we want to
		 * ignore device paths from the pool configuration and use only
		 * the specific path we were given in the boot properties.
		 */
		zfs_rootdisk_path = spa_get_bootprop("zfs-rootdisk-path");

		/*
		 * Initialise the early boot device rescan mechanism.  A scan
		 * will not actually be performed unless we need to do so in
		 * order to find the correct /devices path for a relocated
		 * device.
		 */
		vdev_disk_preroot_init(zfs_rootdisk_path);

		error = spa_import_rootpool(rootfs.bo_name, zfs_devid,
		    zfs_bootpool, zfs_bootvdev);

		spa_free_bootprop(zfs_devid);

		if (error != 0) {
			spa_free_bootprop(zfs_bootfs);
			spa_free_bootprop(zfs_rootdisk_path);
			vdev_disk_preroot_fini();
			cmn_err(CE_NOTE, "spa_import_rootpool: error %d",
			    error);
			return (error);
		}

		if (error = zfs_parse_bootfs(zfs_bootfs, rootfs.bo_name)) {
			spa_free_bootprop(zfs_bootfs);
			spa_free_bootprop(zfs_rootdisk_path);
			vdev_disk_preroot_fini();
			cmn_err(CE_NOTE, "zfs_parse_bootfs: error %d",
			    error);
			return (error);
		}

		spa_free_bootprop(zfs_bootfs);
		spa_free_bootprop(zfs_rootdisk_path);

		if ((error = vfs_lock(vfsp)) != 0) {
			vdev_disk_preroot_fini();
			return (error);
		}

		if (error = zfs_domount(vfsp, rootfs.bo_name)) {
			cmn_err(CE_NOTE, "zfs_domount: error %d", error);
			goto out;
		}

		/* zfs_domount has already cached the root vnode for us */
		zfsvfs = (zfsvfs_t *)vfsp->vfs_data;
		ASSERT(zfsvfs);
		ASSERT(zfsvfs->z_rootdir);

		vp = zfsvfs->z_rootdir;
		mutex_enter(&vp->v_lock);
		vp->v_flag |= VROOT;
		mutex_exit(&vp->v_lock);

		/*
		 * Leave rootvp held.  The root file system is never unmounted.
		 */
		VN_HOLD(vp);
		rootvp = vp;

		vfs_add((struct vnode *)0, vfsp,
		    (vfsp->vfs_flag & VFS_RDONLY) ? MS_RDONLY : 0);
out:
		vdev_disk_preroot_fini();
		vfs_unlock(vfsp);
		return (error);
	} else if (why == ROOT_REMOUNT) {
		readonly_changed_cb(vfsp->vfs_data, B_FALSE);
		vfsp->vfs_flag |= VFS_REMOUNT;

		/* refresh mount options */
		zfs_unregister_callbacks(vfsp->vfs_data);
		return (zfs_register_callbacks(vfsp));

	} else if (why == ROOT_UNMOUNT) {
		zfs_unregister_callbacks((zfsvfs_t *)vfsp->vfs_data);
		(void) zfs_sync(vfsp, 0, 0);
		return (0);
	}

	/*
	 * if "why" is equal to anything else other than ROOT_INIT,
	 * ROOT_REMOUNT, or ROOT_UNMOUNT, we do not support it.
	 */
	return (SET_ERROR(ENOTSUP));
}

/*ARGSUSED*/
static int
zfs_mount(vfs_t *vfsp, vnode_t *mvp, struct mounta *uap, cred_t *cr)
{
	char		*osname;
	pathname_t	spn;
	int		error = 0;
	uio_seg_t	fromspace = (uap->flags & MS_SYSSPACE) ?
	    UIO_SYSSPACE : UIO_USERSPACE;
	int		canwrite;

	if (mvp->v_type != VDIR)
		return (SET_ERROR(ENOTDIR));

	mutex_enter(&mvp->v_lock);
	if ((uap->flags & MS_REMOUNT) == 0 &&
	    (uap->flags & MS_OVERLAY) == 0 &&
	    (mvp->v_count != 1 || (mvp->v_flag & VROOT))) {
		mutex_exit(&mvp->v_lock);
		return (SET_ERROR(EBUSY));
	}
	mutex_exit(&mvp->v_lock);

	/*
	 * ZFS does not support passing unparsed data in via MS_DATA.
	 * Users should use the MS_OPTIONSTR interface; this means
	 * that all option parsing is already done and the options struct
	 * can be interrogated.
	 */
	if ((uap->flags & MS_DATA) && uap->datalen > 0)
		return (SET_ERROR(EINVAL));

	/*
	 * Get the objset name (the "special" mount argument).
	 */
	if (error = pn_get(uap->spec, fromspace, &spn))
		return (error);

	osname = spn.pn_path;

	/*
	 * Check for mount privilege?
	 *
	 * If we don't have privilege then see if
	 * we have local permission to allow it
	 */
	error = secpolicy_fs_mount(cr, mvp, vfsp);
	if (error) {
		if (dsl_deleg_access(osname, ZFS_DELEG_PERM_MOUNT, cr) == 0) {
			vattr_t		vattr;

			/*
			 * Make sure user is the owner of the mount point
			 * or has sufficient privileges.
			 */

			vattr.va_mask = AT_UID;

			if (VOP_GETATTR(mvp, &vattr, 0, cr, NULL)) {
				goto out;
			}

			if (secpolicy_vnode_owner(cr, vattr.va_uid) != 0 &&
			    VOP_ACCESS(mvp, VWRITE, 0, cr, NULL) != 0) {
				goto out;
			}
			secpolicy_fs_mount_clearopts(cr, vfsp);
		} else {
			goto out;
		}
	}

	/*
	 * Refuse to mount a filesystem if we are in a local zone and the
	 * dataset is not visible.
	 */
	if (!INGLOBALZONE(curproc) &&
	    (!zone_dataset_visible(osname, &canwrite) || !canwrite)) {
		error = SET_ERROR(EPERM);
		goto out;
	}

	error = zfs_mount_label_policy(vfsp, osname);
	if (error)
		goto out;

	/*
	 * When doing a remount, we simply refresh our temporary properties
	 * according to those options set in the current VFS options.
	 */
	if (uap->flags & MS_REMOUNT) {
		/* refresh mount options */
		zfs_unregister_callbacks(vfsp->vfs_data);
		error = zfs_register_callbacks(vfsp);
		goto out;
	}

	error = zfs_domount(vfsp, osname);

	/*
	 * Add an extra VFS_HOLD on our parent vfs so that it can't
	 * disappear due to a forced unmount.
	 */
	if (error == 0 && ((zfsvfs_t *)vfsp->vfs_data)->z_issnap)
		VFS_HOLD(mvp->v_vfsp);

out:
	pn_free(&spn);
	return (error);
}

static int
zfs_statvfs(vfs_t *vfsp, struct statvfs64 *statp)
{
	zfsvfs_t *zfsvfs = vfsp->vfs_data;
	dev32_t d32;
	uint64_t refdbytes, availbytes, usedobjs, availobjs;
	int err = 0;

	ZFS_ENTER(zfsvfs);

	dmu_objset_space(zfsvfs->z_os,
	    &refdbytes, &availbytes, &usedobjs, &availobjs);

	/*
	 * The underlying storage pool actually uses multiple block sizes.
	 * We report the fragsize as the smallest block size we support,
	 * and we report our blocksize as the filesystem's maximum blocksize.
	 */
	statp->f_frsize = 1UL << SPA_MINBLOCKSHIFT;
	statp->f_bsize = zfsvfs->z_max_blksz;

	/*
	 * The following report "total" blocks of various kinds in the
	 * file system, but reported in terms of f_frsize - the
	 * "fragment" size.
	 */

	statp->f_blocks = (refdbytes + availbytes) >> SPA_MINBLOCKSHIFT;
	statp->f_bfree = availbytes >> SPA_MINBLOCKSHIFT;
	statp->f_bavail = statp->f_bfree; /* no root reservation */

	/*
	 * statvfs() should really be called statufs(), because it assumes
	 * static metadata.  ZFS doesn't preallocate files, so the best
	 * we can do is report the max that could possibly fit in f_files,
	 * and that minus the number actually used in f_ffree.
	 * For f_ffree, report the smaller of the number of object available
	 * and the number of blocks (each object will take at least a block).
	 */
	statp->f_ffree = MIN(availobjs, statp->f_bfree);
	statp->f_favail = statp->f_ffree;	/* no "root reservation" */
	statp->f_files = statp->f_ffree + usedobjs;

	(void) cmpldev(&d32, vfsp->vfs_dev);
	statp->f_fsid = d32;

	/*
	 * We're a zfs filesystem.
	 */
	(void) strcpy(statp->f_basetype, vfssw[vfsp->vfs_fstype].vsw_name);

	statp->f_flag = vf_to_stf(vfsp->vfs_flag);

	statp->f_namemax = MAXNAMELEN - 1;

	/*
	 * We have all of 32 characters to stuff a string here.
	 * Is there anything useful we could/should provide?
	 */
	bzero(statp->f_fstr, sizeof (statp->f_fstr));

	if (dmu_objset_projectquota_enabled(zfsvfs->z_os) &&
	    dmu_objset_projectquota_present(zfsvfs->z_os)) {
		znode_t *zp;

		/*
		 * In ZoL, zfs_statvfs is passed a Linux dentry (directory
		 * entry), instead of a vfsp. The ZoL code uses the dentry
		 * to get the znode from the dentry's inode. This represents
		 * whatever filename was passed to the user-level statvfs
		 * syscall.
		 *
		 * We're using the VFS root znode here, so this represents a
		 * potential difference from ZoL.
		 */
		if (zfs_zget(zfsvfs, zfsvfs->z_root, &zp) == 0) {
			uint32_t bshift = ddi_fls(statp->f_bsize) - 1;

			if (zp->z_pflags & ZFS_PROJINHERIT && zp->z_projid &&
			    zpl_is_valid_projid(zp->z_projid))
				err = zfs_statfs_project(zfsvfs, zp, statp,
				    bshift);
			VN_RELE(ZTOV(zp));
		}
	}

	ZFS_EXIT(zfsvfs);
	return (err);
}

static int
zfs_root(vfs_t *vfsp, vnode_t **vpp)
{
	zfsvfs_t *zfsvfs = vfsp->vfs_data;
	struct vnode *vp;
	int error;

	ZFS_ENTER(zfsvfs);

	vp = zfsvfs->z_rootdir;
	if (vp != NULL) {
		VN_HOLD(vp);
		error = 0;
	} else {
		/* forced unmount */
		error = EIO;
	}
	*vpp = vp;

	ZFS_EXIT(zfsvfs);
	return (error);

}

/*
 * Teardown the zfsvfs::z_os.
 *
 * Note, if 'unmounting' is FALSE, we return with the 'z_teardown_lock'
 * and 'z_teardown_inactive_lock' held.
 */
static int
zfsvfs_teardown(zfsvfs_t *zfsvfs, boolean_t unmounting)
{
	znode_t	*zp;

	zfs_unlinked_drain_stop_wait(zfsvfs);

	rrm_enter(&zfsvfs->z_teardown_lock, RW_WRITER, FTAG);

	if (!unmounting) {
		/*
		 * We purge the parent filesystem's vfsp as the parent
		 * filesystem and all of its snapshots have their vnode's
		 * v_vfsp set to the parent's filesystem's vfsp.  Note,
		 * 'z_parent' is self referential for non-snapshots.
		 */
		(void) dnlc_purge_vfsp(zfsvfs->z_parent->z_vfs, 0);
	}

	/*
	 * Close the zil. NB: Can't close the zil while zfs_inactive
	 * threads are blocked as zil_close can call zfs_inactive.
	 */
	if (zfsvfs->z_log) {
		zil_close(zfsvfs->z_log);
		zfsvfs->z_log = NULL;
	}

	rw_enter(&zfsvfs->z_teardown_inactive_lock, RW_WRITER);

	/*
	 * If we are not unmounting (ie: online recv) and someone already
	 * unmounted this file system while we were doing the switcheroo,
	 * or a reopen of z_os failed then just bail out now.
	 */
	if (!unmounting && (zfsvfs->z_unmounted || zfsvfs->z_os == NULL)) {
		rw_exit(&zfsvfs->z_teardown_inactive_lock);
		rrm_exit(&zfsvfs->z_teardown_lock, FTAG);
		return (SET_ERROR(EIO));
	}

	/*
	 * At this point there are no vops active, and any new vops will
	 * fail with EIO since we have z_teardown_lock for writer (only
	 * relavent for forced unmount).
	 *
	 * Release all holds on dbufs.
	 */
	mutex_enter(&zfsvfs->z_znodes_lock);
	for (zp = list_head(&zfsvfs->z_all_znodes); zp != NULL;
	    zp = list_next(&zfsvfs->z_all_znodes, zp))
		if (zp->z_sa_hdl) {
			ASSERT(ZTOV(zp)->v_count > 0);
			zfs_znode_dmu_fini(zp);
		}
	mutex_exit(&zfsvfs->z_znodes_lock);

	/*
	 * If we are unmounting, set the unmounted flag and let new vops
	 * unblock.  zfs_inactive will have the unmounted behavior, and all
	 * other vops will fail with EIO.
	 */
	if (unmounting) {
		/*
		 * Clear the cached root vnode now that we are unmounted.
		 * Its release must be performed outside the teardown locks to
		 * avoid recursive lock entry via zfs_inactive().
		 */
		vnode_t *vp = zfsvfs->z_rootdir;
		zfsvfs->z_rootdir = NULL;

		zfsvfs->z_unmounted = B_TRUE;
		rw_exit(&zfsvfs->z_teardown_inactive_lock);
		rrm_exit(&zfsvfs->z_teardown_lock, FTAG);

		/* Drop the cached root vp now that it is safe */
		VN_RELE(vp);
	}

	/*
	 * z_os will be NULL if there was an error in attempting to reopen
	 * zfsvfs, so just return as the properties had already been
	 * unregistered and cached data had been evicted before.
	 */
	if (zfsvfs->z_os == NULL)
		return (0);

	/*
	 * Unregister properties.
	 */
	zfs_unregister_callbacks(zfsvfs);

	/*
	 * Evict cached data
	 */
	if (dsl_dataset_is_dirty(dmu_objset_ds(zfsvfs->z_os)) &&
	    !(zfsvfs->z_vfs->vfs_flag & VFS_RDONLY))
		txg_wait_synced(dmu_objset_pool(zfsvfs->z_os), 0);
	dmu_objset_evict_dbufs(zfsvfs->z_os);

	return (0);
}

/*ARGSUSED*/
static int
zfs_umount(vfs_t *vfsp, int fflag, cred_t *cr)
{
	zfsvfs_t *zfsvfs = vfsp->vfs_data;
	objset_t *os;
	int ret;

	ret = secpolicy_fs_unmount(cr, vfsp);
	if (ret) {
		if (dsl_deleg_access((char *)refstr_value(vfsp->vfs_resource),
		    ZFS_DELEG_PERM_MOUNT, cr))
			return (ret);
	}

	/*
	 * We purge the parent filesystem's vfsp as the parent filesystem
	 * and all of its snapshots have their vnode's v_vfsp set to the
	 * parent's filesystem's vfsp.  Note, 'z_parent' is self
	 * referential for non-snapshots.
	 */
	(void) dnlc_purge_vfsp(zfsvfs->z_parent->z_vfs, 0);

	/*
	 * Unmount any snapshots mounted under .zfs before unmounting the
	 * dataset itself.
	 */
	if (zfsvfs->z_ctldir != NULL &&
	    (ret = zfsctl_umount_snapshots(vfsp, fflag, cr)) != 0) {
		return (ret);
	}

	if (!(fflag & MS_FORCE)) {
		/*
		 * Check the number of active vnodes in the file system.
		 * Our count is maintained in the vfs structure, but the
		 * number is off by 1 to indicate a hold on the vfs
		 * structure itself.
		 */
		boolean_t draining;
		uint_t thresh = 1;
		vnode_t *ctlvp, *rvp;

		/*
		 * The cached vnode for the root directory of the mount also
		 * maintains a hold on the vfs structure.
		 */
		rvp = zfsvfs->z_rootdir;
		thresh++;

		/*
		 * The '.zfs' directory maintains a reference of its own, and
		 * any active references underneath are reflected in the vnode
		 * count. Allow one additional reference for it.
		 */
		ctlvp = zfsvfs->z_ctldir;
		if (ctlvp != NULL) {
			thresh++;
		}

		/*
		 * If it's running, the asynchronous unlinked drain task needs
		 * to be stopped before the number of active vnodes can be
		 * reliably checked.
		 */
		draining = zfsvfs->z_draining;
		if (draining)
			zfs_unlinked_drain_stop_wait(zfsvfs);

		if (vfsp->vfs_count > thresh || rvp->v_count > 1 ||
		    (ctlvp != NULL && ctlvp->v_count > 1)) {
			if (draining) {
				/* If it was draining, restart the task */
				zfs_unlinked_drain(zfsvfs);
			}
			return (SET_ERROR(EBUSY));
		}
	}

	vfsp->vfs_flag |= VFS_UNMOUNTED;

	VERIFY(zfsvfs_teardown(zfsvfs, B_TRUE) == 0);
	os = zfsvfs->z_os;

	/*
	 * z_os will be NULL if there was an error in
	 * attempting to reopen zfsvfs.
	 */
	if (os != NULL) {
		/*
		 * Unset the objset user_ptr.
		 */
		mutex_enter(&os->os_user_ptr_lock);
		dmu_objset_set_user(os, NULL);
		mutex_exit(&os->os_user_ptr_lock);

		/*
		 * Finally release the objset
		 */
		dmu_objset_disown(os, B_TRUE, zfsvfs);
	}

	/*
	 * We can now safely destroy the '.zfs' directory node.
	 */
	if (zfsvfs->z_ctldir != NULL)
		zfsctl_destroy(zfsvfs);

	return (0);
}

static int
zfs_vget(vfs_t *vfsp, vnode_t **vpp, fid_t *fidp)
{
	zfsvfs_t	*zfsvfs = vfsp->vfs_data;
	znode_t		*zp;
	uint64_t	object = 0;
	uint64_t	fid_gen = 0;
	uint64_t	gen_mask;
	uint64_t	zp_gen;
	int		i, err;

	*vpp = NULL;

	ZFS_ENTER(zfsvfs);

	if (fidp->fid_len == LONG_FID_LEN) {
		zfid_long_t	*zlfid = (zfid_long_t *)fidp;
		uint64_t	objsetid = 0;
		uint64_t	setgen = 0;

		for (i = 0; i < sizeof (zlfid->zf_setid); i++)
			objsetid |= ((uint64_t)zlfid->zf_setid[i]) << (8 * i);

		for (i = 0; i < sizeof (zlfid->zf_setgen); i++)
			setgen |= ((uint64_t)zlfid->zf_setgen[i]) << (8 * i);

		ZFS_EXIT(zfsvfs);

		err = zfsctl_lookup_objset(vfsp, objsetid, &zfsvfs);
		if (err)
			return (SET_ERROR(EINVAL));
		ZFS_ENTER(zfsvfs);
	}

	if (fidp->fid_len == SHORT_FID_LEN || fidp->fid_len == LONG_FID_LEN) {
		zfid_short_t	*zfid = (zfid_short_t *)fidp;

		for (i = 0; i < sizeof (zfid->zf_object); i++)
			object |= ((uint64_t)zfid->zf_object[i]) << (8 * i);

		for (i = 0; i < sizeof (zfid->zf_gen); i++)
			fid_gen |= ((uint64_t)zfid->zf_gen[i]) << (8 * i);
	} else {
		ZFS_EXIT(zfsvfs);
		return (SET_ERROR(EINVAL));
	}

	/* A zero fid_gen means we are in the .zfs control directories */
	if (fid_gen == 0 &&
	    (object == ZFSCTL_INO_ROOT || object == ZFSCTL_INO_SNAPDIR)) {
		*vpp = zfsvfs->z_ctldir;
		ASSERT(*vpp != NULL);
		if (object == ZFSCTL_INO_SNAPDIR) {
			VERIFY(zfsctl_root_lookup(*vpp, "snapshot", vpp, NULL,
			    0, NULL, NULL, NULL, NULL, NULL) == 0);
		} else {
			VN_HOLD(*vpp);
		}
		ZFS_EXIT(zfsvfs);
		return (0);
	}

	gen_mask = -1ULL >> (64 - 8 * i);

	dprintf("getting %llu [%u mask %llx]\n", object, fid_gen, gen_mask);
	if (err = zfs_zget(zfsvfs, object, &zp)) {
		ZFS_EXIT(zfsvfs);
		return (err);
	}
	(void) sa_lookup(zp->z_sa_hdl, SA_ZPL_GEN(zfsvfs), &zp_gen,
	    sizeof (uint64_t));
	zp_gen = zp_gen & gen_mask;
	if (zp_gen == 0)
		zp_gen = 1;
	if (zp->z_unlinked || zp_gen != fid_gen) {
		dprintf("znode gen (%u) != fid gen (%u)\n", zp_gen, fid_gen);
		VN_RELE(ZTOV(zp));
		ZFS_EXIT(zfsvfs);
		return (SET_ERROR(EINVAL));
	}

	*vpp = ZTOV(zp);
	ZFS_EXIT(zfsvfs);
	return (0);
}

/*
 * Block out VOPs and close zfsvfs_t::z_os
 *
 * Note, if successful, then we return with the 'z_teardown_lock' and
 * 'z_teardown_inactive_lock' write held.  We leave ownership of the underlying
 * dataset and objset intact so that they can be atomically handed off during
 * a subsequent rollback or recv operation and the resume thereafter.
 */
int
zfs_suspend_fs(zfsvfs_t *zfsvfs)
{
	int error;

	if ((error = zfsvfs_teardown(zfsvfs, B_FALSE)) != 0)
		return (error);

	return (0);
}

/*
 * Rebuild SA and release VOPs.  Note that ownership of the underlying dataset
 * is an invariant across any of the operations that can be performed while the
 * filesystem was suspended.  Whether it succeeded or failed, the preconditions
 * are the same: the relevant objset and associated dataset are owned by
 * zfsvfs, held, and long held on entry.
 */
int
zfs_resume_fs(zfsvfs_t *zfsvfs, dsl_dataset_t *ds)
{
	int err;
	znode_t *zp;

	ASSERT(RRM_WRITE_HELD(&zfsvfs->z_teardown_lock));
	ASSERT(RW_WRITE_HELD(&zfsvfs->z_teardown_inactive_lock));

	/*
	 * We already own this, so just update the objset_t, as the one we
	 * had before may have been evicted.
	 */
	objset_t *os;
	VERIFY3P(ds->ds_owner, ==, zfsvfs);
	VERIFY(dsl_dataset_long_held(ds));
	VERIFY0(dmu_objset_from_ds(ds, &os));

	err = zfsvfs_init(zfsvfs, os);
	if (err != 0)
		goto bail;

	VERIFY(zfsvfs_setup(zfsvfs, B_FALSE) == 0);

	zfs_set_fuid_feature(zfsvfs);

	/*
	 * Attempt to re-establish all the active znodes with
	 * their dbufs.  If a zfs_rezget() fails, then we'll let
	 * any potential callers discover that via ZFS_ENTER_VERIFY_VP
	 * when they try to use their znode.
	 */
	mutex_enter(&zfsvfs->z_znodes_lock);
	for (zp = list_head(&zfsvfs->z_all_znodes); zp;
	    zp = list_next(&zfsvfs->z_all_znodes, zp)) {
		(void) zfs_rezget(zp);
	}
	mutex_exit(&zfsvfs->z_znodes_lock);

	if (((zfsvfs->z_vfs->vfs_flag & VFS_RDONLY) == 0) &&
	    !zfsvfs->z_unmounted) {
		/*
		 * zfs_suspend_fs() could have interrupted freeing
		 * of dnodes. We need to restart this freeing so
		 * that we don't "leak" the space.
		 */
		zfs_unlinked_drain(zfsvfs);
	}

bail:
	/* release the VOPs */
	rw_exit(&zfsvfs->z_teardown_inactive_lock);
	rrm_exit(&zfsvfs->z_teardown_lock, FTAG);

	if (err) {
		/*
		 * Since we couldn't setup the sa framework, try to force
		 * unmount this file system.
		 */
		if (vn_vfswlock(zfsvfs->z_vfs->vfs_vnodecovered) == 0)
			(void) dounmount(zfsvfs->z_vfs, MS_FORCE, CRED());
	}
	return (err);
}

static void
zfs_freevfs(vfs_t *vfsp)
{
	zfsvfs_t *zfsvfs = vfsp->vfs_data;

	/*
	 * If this is a snapshot, we have an extra VFS_HOLD on our parent
	 * from zfs_mount().  Release it here.  If we came through
	 * zfs_mountroot() instead, we didn't grab an extra hold, so
	 * skip the VFS_RELE for rootvfs.
	 */
	if (zfsvfs->z_issnap && (vfsp != rootvfs))
		VFS_RELE(zfsvfs->z_parent->z_vfs);

	zfsvfs_free(zfsvfs);

	atomic_dec_32(&zfs_active_fs_count);
}

/*
 * VFS_INIT() initialization.  Note that there is no VFS_FINI(),
 * so we can't safely do any non-idempotent initialization here.
 * Leave that to zfs_init() and zfs_fini(), which are called
 * from the module's _init() and _fini() entry points.
 */
/*ARGSUSED*/
static int
zfs_vfsinit(int fstype, char *name)
{
	int error;

	zfsfstype = fstype;

	/*
	 * Setup vfsops and vnodeops tables.
	 */
	error = vfs_setfsops(fstype, zfs_vfsops_template, &zfs_vfsops);
	if (error != 0) {
		cmn_err(CE_WARN, "zfs: bad vfs ops template");
	}

	error = zfs_create_op_tables();
	if (error) {
		zfs_remove_op_tables();
		cmn_err(CE_WARN, "zfs: bad vnode ops template");
		(void) vfs_freevfsops_by_type(zfsfstype);
		return (error);
	}

	mutex_init(&zfs_dev_mtx, NULL, MUTEX_DEFAULT, NULL);

	/*
	 * Unique major number for all zfs mounts.
	 * If we run out of 32-bit minors, we'll getudev() another major.
	 */
	zfs_major = ddi_name_to_major(ZFS_DRIVER);
	zfs_minor = ZFS_MIN_MINOR;

	return (0);
}

void
zfs_init(void)
{
	/*
	 * Initialize .zfs directory structures
	 */
	zfsctl_init();

	/*
	 * Initialize znode cache, vnode ops, etc...
	 */
	zfs_znode_init();

	dmu_objset_register_type(DMU_OST_ZFS, zfs_space_delta_cb);
}

void
zfs_fini(void)
{
	zfsctl_fini();
	zfs_znode_fini();
}

int
zfs_busy(void)
{
	return (zfs_active_fs_count != 0);
}

int
zfs_set_version(zfsvfs_t *zfsvfs, uint64_t newvers)
{
	int error;
	objset_t *os = zfsvfs->z_os;
	dmu_tx_t *tx;

	if (newvers < ZPL_VERSION_INITIAL || newvers > ZPL_VERSION)
		return (SET_ERROR(EINVAL));

	if (newvers < zfsvfs->z_version)
		return (SET_ERROR(EINVAL));

	if (zfs_spa_version_map(newvers) >
	    spa_version(dmu_objset_spa(zfsvfs->z_os)))
		return (SET_ERROR(ENOTSUP));

	tx = dmu_tx_create(os);
	dmu_tx_hold_zap(tx, MASTER_NODE_OBJ, B_FALSE, ZPL_VERSION_STR);
	if (newvers >= ZPL_VERSION_SA && !zfsvfs->z_use_sa) {
		dmu_tx_hold_zap(tx, MASTER_NODE_OBJ, B_TRUE,
		    ZFS_SA_ATTRS);
		dmu_tx_hold_zap(tx, DMU_NEW_OBJECT, FALSE, NULL);
	}
	error = dmu_tx_assign(tx, TXG_WAIT);
	if (error) {
		dmu_tx_abort(tx);
		return (error);
	}

	error = zap_update(os, MASTER_NODE_OBJ, ZPL_VERSION_STR,
	    8, 1, &newvers, tx);

	if (error) {
		dmu_tx_commit(tx);
		return (error);
	}

	if (newvers >= ZPL_VERSION_SA && !zfsvfs->z_use_sa) {
		uint64_t sa_obj;

		ASSERT3U(spa_version(dmu_objset_spa(zfsvfs->z_os)), >=,
		    SPA_VERSION_SA);
		sa_obj = zap_create(os, DMU_OT_SA_MASTER_NODE,
		    DMU_OT_NONE, 0, tx);

		error = zap_add(os, MASTER_NODE_OBJ,
		    ZFS_SA_ATTRS, 8, 1, &sa_obj, tx);
		ASSERT0(error);

		VERIFY(0 == sa_set_sa_object(os, sa_obj));
		sa_register_update_callback(os, zfs_sa_upgrade);
	}

	spa_history_log_internal_ds(dmu_objset_ds(os), "upgrade", tx,
	    "from %llu to %llu", zfsvfs->z_version, newvers);

	dmu_tx_commit(tx);

	zfsvfs->z_version = newvers;
	os->os_version = newvers;

	zfs_set_fuid_feature(zfsvfs);

	return (0);
}

/*
 * Read a property stored within the master node.
 */
int
zfs_get_zplprop(objset_t *os, zfs_prop_t prop, uint64_t *value)
{
	uint64_t *cached_copy = NULL;

	/*
	 * Figure out where in the objset_t the cached copy would live, if it
	 * is available for the requested property.
	 */
	if (os != NULL) {
		switch (prop) {
		case ZFS_PROP_VERSION:
			cached_copy = &os->os_version;
			break;
		case ZFS_PROP_NORMALIZE:
			cached_copy = &os->os_normalization;
			break;
		case ZFS_PROP_UTF8ONLY:
			cached_copy = &os->os_utf8only;
			break;
		case ZFS_PROP_CASE:
			cached_copy = &os->os_casesensitivity;
			break;
		default:
			break;
		}
	}
	if (cached_copy != NULL && *cached_copy != OBJSET_PROP_UNINITIALIZED) {
		*value = *cached_copy;
		return (0);
	}

	/*
	 * If the property wasn't cached, look up the file system's value for
	 * the property. For the version property, we look up a slightly
	 * different string.
	 */
	const char *pname;
	int error = ENOENT;
	if (prop == ZFS_PROP_VERSION) {
		pname = ZPL_VERSION_STR;
	} else {
		pname = zfs_prop_to_name(prop);
	}

	if (os != NULL) {
		ASSERT3U(os->os_phys->os_type, ==, DMU_OST_ZFS);
		error = zap_lookup(os, MASTER_NODE_OBJ, pname, 8, 1, value);
	}

	if (error == ENOENT) {
		/* No value set, use the default value */
		switch (prop) {
		case ZFS_PROP_VERSION:
			*value = ZPL_VERSION;
			break;
		case ZFS_PROP_NORMALIZE:
		case ZFS_PROP_UTF8ONLY:
			*value = 0;
			break;
		case ZFS_PROP_CASE:
			*value = ZFS_CASE_SENSITIVE;
			break;
		default:
			return (error);
		}
		error = 0;
	}

	/*
	 * If one of the methods for getting the property value above worked,
	 * copy it into the objset_t's cache.
	 */
	if (error == 0 && cached_copy != NULL) {
		*cached_copy = *value;
	}

	return (error);
}

/*
 * Return true if the coresponding vfs's unmounted flag is set.
 * Otherwise return false.
 * If this function returns true we know VFS unmount has been initiated.
 */
boolean_t
zfs_get_vfs_flag_unmounted(objset_t *os)
{
	zfsvfs_t *zfvp;
	boolean_t unmounted = B_FALSE;

	ASSERT(dmu_objset_type(os) == DMU_OST_ZFS);

	mutex_enter(&os->os_user_ptr_lock);
	zfvp = dmu_objset_get_user(os);
	if (zfvp != NULL && zfvp->z_vfs != NULL &&
	    (zfvp->z_vfs->vfs_flag & VFS_UNMOUNTED))
		unmounted = B_TRUE;
	mutex_exit(&os->os_user_ptr_lock);

	return (unmounted);
}

static vfsdef_t vfw = {
	VFSDEF_VERSION,
	MNTTYPE_ZFS,
	zfs_vfsinit,
	VSW_HASPROTO|VSW_CANRWRO|VSW_CANREMOUNT|VSW_VOLATILEDEV|VSW_STATS|
	    VSW_XID|VSW_ZMOUNT,
	&zfs_mntopts
};

struct modlfs zfs_modlfs = {
	&mod_fsops, "ZFS filesystem version " SPA_VERSION_STRING, &vfw
};