Loading...
   1
   2
   3
   4
   5
   6
   7
   8
   9
  10
  11
  12
  13
  14
  15
  16
  17
  18
  19
  20
  21
  22
  23
  24
  25
  26
  27
  28
  29
  30
  31
  32
  33
  34
  35
  36
  37
  38
  39
  40
  41
  42
  43
  44
  45
  46
  47
  48
  49
  50
  51
  52
  53
  54
  55
  56
  57
  58
  59
  60
  61
  62
  63
  64
  65
  66
  67
  68
  69
  70
  71
  72
  73
  74
  75
  76
  77
  78
  79
  80
  81
  82
  83
  84
  85
  86
  87
  88
  89
  90
  91
  92
  93
  94
  95
  96
  97
  98
  99
 100
 101
 102
 103
 104
 105
 106
 107
 108
 109
 110
 111
 112
 113
 114
 115
 116
 117
 118
 119
 120
 121
 122
 123
 124
 125
 126
 127
 128
 129
 130
 131
 132
 133
 134
 135
 136
 137
 138
 139
 140
 141
 142
 143
 144
 145
 146
 147
 148
 149
 150
 151
 152
 153
 154
 155
 156
 157
 158
 159
 160
 161
 162
 163
 164
 165
 166
 167
 168
 169
 170
 171
 172
 173
 174
 175
 176
 177
 178
 179
 180
 181
 182
 183
 184
 185
 186
 187
 188
 189
 190
 191
 192
 193
 194
 195
 196
 197
 198
 199
 200
 201
 202
 203
 204
 205
 206
 207
 208
 209
 210
 211
 212
 213
 214
 215
 216
 217
 218
 219
 220
 221
 222
 223
 224
 225
 226
 227
 228
 229
 230
 231
 232
 233
 234
 235
 236
 237
 238
 239
 240
 241
 242
 243
 244
 245
 246
 247
 248
 249
 250
 251
 252
 253
 254
 255
 256
 257
 258
 259
 260
 261
 262
 263
 264
 265
 266
 267
 268
 269
 270
 271
 272
 273
 274
 275
 276
 277
 278
 279
 280
 281
 282
 283
 284
 285
 286
 287
 288
 289
 290
 291
 292
 293
 294
 295
 296
 297
 298
 299
 300
 301
 302
 303
 304
 305
 306
 307
 308
 309
 310
 311
 312
 313
 314
 315
 316
 317
 318
 319
 320
 321
 322
 323
 324
 325
 326
 327
 328
 329
 330
 331
 332
 333
 334
 335
 336
 337
 338
 339
 340
 341
 342
 343
 344
 345
 346
 347
 348
 349
 350
 351
 352
 353
 354
 355
 356
 357
 358
 359
 360
 361
 362
 363
 364
 365
 366
 367
 368
 369
 370
 371
 372
 373
 374
 375
 376
 377
 378
 379
 380
 381
 382
 383
 384
 385
 386
 387
 388
 389
 390
 391
 392
 393
 394
 395
 396
 397
 398
 399
 400
 401
 402
 403
 404
 405
 406
 407
 408
 409
 410
 411
 412
 413
 414
 415
 416
 417
 418
 419
 420
 421
 422
 423
 424
 425
 426
 427
 428
 429
 430
 431
 432
 433
 434
 435
 436
 437
 438
 439
 440
 441
 442
 443
 444
 445
 446
 447
 448
 449
 450
 451
 452
 453
 454
 455
 456
 457
 458
 459
 460
 461
 462
 463
 464
 465
 466
 467
 468
 469
 470
 471
 472
 473
 474
 475
 476
 477
 478
 479
 480
 481
 482
 483
 484
 485
 486
 487
 488
 489
 490
 491
 492
 493
 494
 495
 496
 497
 498
 499
 500
 501
 502
 503
 504
 505
 506
 507
 508
 509
 510
 511
 512
 513
 514
 515
 516
 517
 518
 519
 520
 521
 522
 523
 524
 525
 526
 527
 528
 529
 530
 531
 532
 533
 534
 535
 536
 537
 538
 539
 540
 541
 542
 543
 544
 545
 546
 547
 548
 549
 550
 551
 552
 553
 554
 555
 556
 557
 558
 559
 560
 561
 562
 563
 564
 565
 566
 567
 568
 569
 570
 571
 572
 573
 574
 575
 576
 577
 578
 579
 580
 581
 582
 583
 584
 585
 586
 587
 588
 589
 590
 591
 592
 593
 594
 595
 596
 597
 598
 599
 600
 601
 602
 603
 604
 605
 606
 607
 608
 609
 610
 611
 612
 613
 614
 615
 616
 617
 618
 619
 620
 621
 622
 623
 624
 625
 626
 627
 628
 629
 630
 631
 632
 633
 634
 635
 636
 637
 638
 639
 640
 641
 642
 643
 644
 645
 646
 647
 648
 649
 650
 651
 652
 653
 654
 655
 656
 657
 658
 659
 660
 661
 662
 663
 664
 665
 666
 667
 668
 669
 670
 671
 672
 673
 674
 675
 676
 677
 678
 679
 680
 681
 682
 683
 684
 685
 686
 687
 688
 689
 690
 691
 692
 693
 694
 695
 696
 697
 698
 699
 700
 701
 702
 703
 704
 705
 706
 707
 708
 709
 710
 711
 712
 713
 714
 715
 716
 717
 718
 719
 720
 721
 722
 723
 724
 725
 726
 727
 728
 729
 730
 731
 732
 733
 734
 735
 736
 737
 738
 739
 740
 741
 742
 743
 744
 745
 746
 747
 748
 749
 750
 751
 752
 753
 754
 755
 756
 757
 758
 759
 760
 761
 762
 763
 764
 765
 766
 767
 768
 769
 770
 771
 772
 773
 774
 775
 776
 777
 778
 779
 780
 781
 782
 783
 784
 785
 786
 787
 788
 789
 790
 791
 792
 793
 794
 795
 796
 797
 798
 799
 800
 801
 802
 803
 804
 805
 806
 807
 808
 809
 810
 811
 812
 813
 814
 815
 816
 817
 818
 819
 820
 821
 822
 823
 824
 825
 826
 827
 828
 829
 830
 831
 832
 833
 834
 835
 836
 837
 838
 839
 840
 841
 842
 843
 844
 845
 846
 847
 848
 849
 850
 851
 852
 853
 854
 855
 856
 857
 858
 859
 860
 861
 862
 863
 864
 865
 866
 867
 868
 869
 870
 871
 872
 873
 874
 875
 876
 877
 878
 879
 880
 881
 882
 883
 884
 885
 886
 887
 888
 889
 890
 891
 892
 893
 894
 895
 896
 897
 898
 899
 900
 901
 902
 903
 904
 905
 906
 907
 908
 909
 910
 911
 912
 913
 914
 915
 916
 917
 918
 919
 920
 921
 922
 923
 924
 925
 926
 927
 928
 929
 930
 931
 932
 933
 934
 935
 936
 937
 938
 939
 940
 941
 942
 943
 944
 945
 946
 947
 948
 949
 950
 951
 952
 953
 954
 955
 956
 957
 958
 959
 960
 961
 962
 963
 964
 965
 966
 967
 968
 969
 970
 971
 972
 973
 974
 975
 976
 977
 978
 979
 980
 981
 982
 983
 984
 985
 986
 987
 988
 989
 990
 991
 992
 993
 994
 995
 996
 997
 998
 999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
2547
2548
2549
2550
2551
2552
2553
2554
2555
2556
2557
2558
2559
2560
2561
2562
2563
2564
2565
2566
2567
2568
2569
2570
2571
2572
2573
2574
2575
2576
2577
2578
2579
2580
2581
2582
2583
2584
2585
2586
2587
2588
2589
2590
2591
2592
2593
2594
2595
2596
2597
2598
2599
2600
2601
2602
2603
2604
2605
2606
2607
2608
2609
2610
2611
2612
2613
2614
2615
2616
2617
2618
2619
2620
2621
2622
2623
2624
2625
2626
2627
2628
2629
2630
2631
2632
2633
2634
2635
2636
2637
2638
2639
2640
2641
2642
2643
2644
2645
2646
2647
2648
2649
2650
2651
2652
2653
2654
2655
2656
2657
2658
2659
2660
2661
2662
2663
2664
2665
2666
2667
2668
2669
2670
2671
2672
2673
2674
2675
2676
2677
2678
2679
2680
2681
2682
2683
2684
2685
2686
2687
2688
2689
2690
2691
2692
2693
2694
2695
2696
2697
2698
2699
2700
2701
2702
2703
2704
2705
2706
2707
2708
2709
2710
2711
2712
2713
2714
2715
2716
2717
2718
2719
2720
2721
2722
2723
2724
2725
2726
2727
2728
2729
2730
2731
2732
2733
2734
2735
2736
2737
2738
2739
2740
2741
2742
2743
2744
2745
2746
2747
2748
2749
2750
2751
2752
2753
2754
2755
2756
2757
2758
2759
2760
2761
2762
2763
2764
2765
2766
2767
2768
2769
2770
2771
2772
2773
2774
2775
2776
2777
2778
2779
2780
2781
2782
2783
2784
2785
2786
2787
2788
2789
2790
2791
2792
2793
2794
2795
2796
2797
2798
2799
2800
2801
2802
2803
2804
2805
2806
2807
2808
2809
2810
2811
2812
2813
2814
2815
2816
2817
2818
2819
2820
2821
2822
2823
2824
2825
2826
2827
2828
2829
2830
2831
2832
2833
2834
2835
2836
2837
2838
2839
2840
2841
2842
2843
2844
2845
2846
2847
2848
2849
2850
2851
2852
2853
2854
2855
2856
2857
2858
2859
2860
2861
2862
2863
2864
2865
2866
2867
2868
2869
2870
2871
2872
2873
2874
2875
2876
2877
2878
2879
2880
2881
2882
2883
2884
2885
2886
2887
2888
2889
2890
2891
2892
2893
2894
2895
2896
2897
2898
2899
2900
2901
2902
2903
2904
2905
2906
2907
2908
2909
2910
2911
2912
2913
2914
2915
2916
2917
2918
2919
2920
2921
2922
2923
2924
2925
2926
2927
2928
2929
2930
2931
2932
2933
2934
2935
2936
2937
2938
2939
2940
2941
2942
2943
2944
2945
2946
2947
2948
2949
2950
2951
2952
2953
2954
2955
2956
2957
2958
2959
2960
2961
2962
2963
2964
2965
2966
2967
2968
2969
2970
2971
2972
2973
2974
2975
2976
2977
2978
2979
2980
2981
2982
2983
2984
2985
2986
2987
2988
2989
2990
2991
2992
2993
2994
2995
2996
2997
2998
2999
3000
3001
3002
3003
3004
3005
3006
3007
3008
3009
3010
3011
3012
3013
3014
3015
3016
3017
3018
3019
3020
3021
3022
3023
3024
3025
3026
3027
3028
3029
3030
3031
3032
3033
3034
3035
3036
3037
3038
3039
3040
3041
3042
3043
3044
3045
3046
3047
3048
3049
3050
3051
3052
3053
3054
3055
3056
3057
3058
3059
3060
3061
3062
3063
3064
3065
3066
3067
3068
3069
3070
3071
3072
3073
3074
3075
3076
3077
3078
3079
3080
3081
3082
3083
3084
3085
3086
3087
3088
3089
3090
3091
3092
3093
3094
3095
3096
3097
3098
3099
3100
3101
3102
3103
3104
3105
3106
3107
3108
3109
3110
3111
3112
3113
3114
3115
3116
3117
3118
3119
3120
3121
3122
3123
3124
3125
3126
3127
3128
3129
3130
3131
3132
3133
3134
3135
3136
3137
3138
3139
3140
3141
3142
3143
3144
3145
3146
3147
3148
3149
3150
3151
3152
3153
3154
3155
3156
3157
3158
3159
3160
3161
3162
3163
3164
3165
3166
3167
3168
3169
3170
3171
3172
3173
3174
3175
3176
3177
3178
3179
3180
3181
3182
3183
3184
3185
3186
3187
3188
3189
3190
3191
3192
3193
3194
3195
3196
3197
3198
3199
3200
3201
3202
3203
3204
3205
3206
3207
3208
3209
3210
3211
3212
3213
3214
3215
3216
3217
3218
3219
3220
3221
3222
3223
3224
3225
3226
3227
3228
3229
3230
3231
3232
3233
3234
3235
3236
3237
3238
3239
3240
3241
3242
3243
3244
3245
3246
3247
3248
3249
3250
3251
3252
3253
3254
3255
3256
3257
3258
3259
3260
3261
3262
3263
3264
3265
3266
3267
3268
3269
3270
3271
3272
3273
3274
3275
3276
3277
3278
3279
3280
3281
3282
3283
3284
3285
3286
3287
3288
3289
3290
3291
3292
3293
3294
3295
3296
3297
3298
3299
3300
3301
3302
3303
3304
3305
3306
3307
3308
3309
3310
3311
3312
3313
3314
3315
3316
3317
3318
3319
3320
3321
3322
3323
3324
3325
3326
3327
3328
3329
3330
3331
3332
3333
3334
3335
3336
3337
3338
3339
3340
3341
3342
3343
3344
3345
3346
3347
3348
3349
3350
3351
3352
3353
3354
3355
3356
3357
3358
3359
3360
3361
3362
3363
3364
3365
3366
3367
3368
3369
3370
3371
3372
3373
3374
3375
3376
3377
3378
3379
3380
3381
3382
3383
3384
3385
3386
3387
3388
3389
3390
3391
3392
3393
3394
3395
3396
3397
3398
3399
3400
3401
3402
3403
3404
3405
3406
3407
3408
3409
3410
3411
3412
3413
3414
3415
3416
3417
3418
3419
3420
3421
3422
3423
3424
3425
3426
3427
3428
3429
3430
3431
3432
3433
3434
3435
3436
3437
3438
3439
3440
3441
3442
3443
3444
3445
3446
3447
3448
3449
3450
3451
3452
3453
3454
3455
3456
3457
3458
3459
3460
3461
3462
3463
3464
3465
3466
3467
3468
3469
3470
3471
3472
3473
3474
3475
3476
3477
3478
3479
3480
3481
3482
3483
3484
3485
3486
3487
3488
3489
3490
3491
3492
3493
3494
3495
3496
3497
3498
3499
3500
3501
3502
3503
3504
3505
3506
3507
3508
3509
3510
3511
3512
3513
3514
3515
3516
3517
3518
3519
3520
3521
3522
3523
3524
3525
3526
3527
3528
3529
3530
3531
3532
3533
3534
3535
3536
3537
3538
3539
3540
3541
3542
3543
3544
3545
3546
3547
3548
3549
3550
3551
3552
3553
3554
3555
3556
3557
3558
3559
3560
3561
3562
3563
3564
3565
3566
3567
3568
3569
3570
3571
3572
3573
3574
3575
3576
3577
3578
3579
3580
3581
3582
3583
3584
3585
3586
3587
3588
3589
3590
3591
3592
3593
3594
3595
3596
3597
3598
3599
3600
3601
3602
3603
3604
3605
3606
3607
3608
3609
3610
3611
3612
3613
3614
3615
3616
3617
3618
3619
3620
3621
3622
3623
3624
3625
3626
3627
3628
3629
3630
3631
3632
3633
3634
3635
3636
3637
3638
3639
3640
3641
3642
3643
3644
3645
3646
3647
3648
3649
3650
3651
3652
3653
3654
3655
3656
3657
3658
3659
3660
3661
3662
3663
3664
3665
3666
3667
3668
3669
3670
3671
3672
3673
3674
3675
3676
3677
3678
3679
3680
3681
3682
3683
3684
3685
3686
3687
3688
3689
3690
3691
3692
3693
3694
3695
3696
3697
3698
3699
3700
3701
3702
3703
3704
3705
3706
3707
3708
3709
3710
3711
3712
3713
3714
3715
3716
3717
3718
3719
3720
3721
3722
3723
3724
3725
3726
3727
3728
3729
3730
3731
3732
3733
3734
3735
3736
3737
3738
3739
3740
3741
3742
3743
3744
3745
3746
3747
3748
3749
3750
3751
3752
3753
3754
3755
3756
3757
3758
3759
3760
3761
3762
3763
3764
3765
3766
3767
3768
3769
3770
3771
3772
3773
3774
3775
3776
3777
3778
3779
3780
3781
3782
3783
3784
3785
3786
3787
3788
3789
3790
3791
3792
3793
3794
3795
3796
3797
3798
3799
3800
3801
3802
3803
3804
3805
3806
3807
3808
3809
3810
3811
3812
3813
3814
3815
3816
3817
3818
3819
3820
3821
3822
3823
3824
3825
3826
3827
3828
3829
3830
3831
3832
3833
3834
3835
3836
3837
3838
3839
3840
3841
3842
3843
3844
3845
3846
3847
3848
3849
3850
3851
3852
3853
3854
3855
3856
3857
3858
3859
3860
3861
3862
3863
3864
3865
3866
3867
3868
3869
3870
3871
3872
3873
3874
3875
3876
3877
3878
3879
3880
3881
3882
3883
3884
3885
3886
3887
3888
3889
3890
3891
3892
3893
3894
3895
3896
3897
3898
3899
3900
3901
3902
3903
3904
3905
3906
3907
3908
3909
3910
3911
3912
3913
3914
3915
3916
3917
3918
3919
3920
3921
3922
3923
3924
3925
3926
3927
3928
3929
3930
3931
3932
3933
3934
3935
3936
3937
3938
3939
3940
3941
3942
3943
3944
3945
3946
3947
3948
3949
3950
3951
3952
3953
3954
3955
3956
3957
3958
3959
3960
3961
3962
3963
3964
3965
3966
3967
3968
3969
3970
3971
3972
3973
3974
3975
3976
3977
3978
3979
3980
3981
3982
3983
3984
3985
3986
3987
3988
3989
3990
3991
3992
3993
3994
3995
3996
3997
3998
3999
4000
4001
4002
4003
4004
4005
4006
4007
4008
4009
4010
4011
4012
4013
4014
4015
4016
4017
4018
4019
4020
4021
4022
4023
4024
4025
4026
4027
4028
4029
4030
4031
4032
4033
4034
4035
4036
4037
4038
4039
4040
4041
4042
4043
4044
4045
4046
4047
4048
4049
4050
4051
4052
4053
4054
4055
4056
4057
4058
4059
4060
4061
4062
4063
4064
4065
4066
4067
4068
4069
4070
4071
4072
4073
4074
4075
4076
4077
4078
4079
4080
4081
4082
4083
4084
4085
4086
4087
4088
4089
4090
4091
4092
4093
4094
4095
4096
4097
4098
4099
4100
4101
4102
4103
4104
4105
4106
4107
4108
4109
4110
4111
4112
4113
4114
4115
4116
4117
4118
4119
4120
4121
4122
4123
4124
4125
4126
4127
4128
4129
4130
4131
4132
4133
4134
4135
4136
4137
4138
4139
4140
4141
4142
4143
4144
4145
4146
4147
4148
4149
4150
4151
4152
4153
4154
4155
4156
4157
4158
4159
4160
4161
4162
4163
4164
4165
4166
4167
4168
4169
4170
4171
4172
4173
4174
4175
4176
4177
4178
4179
4180
4181
4182
4183
4184
4185
4186
4187
4188
4189
4190
4191
4192
4193
4194
4195
4196
4197
4198
4199
4200
4201
4202
4203
4204
4205
4206
4207
4208
4209
4210
4211
4212
4213
4214
4215
4216
4217
4218
4219
4220
4221
4222
4223
4224
4225
4226
4227
4228
4229
4230
4231
4232
4233
4234
4235
4236
4237
4238
4239
4240
4241
4242
4243
4244
4245
4246
4247
4248
4249
4250
4251
4252
4253
4254
4255
4256
4257
4258
4259
4260
4261
4262
4263
4264
4265
4266
4267
4268
4269
4270
4271
4272
4273
4274
4275
4276
4277
4278
4279
4280
4281
4282
4283
4284
4285
4286
4287
4288
4289
4290
4291
4292
4293
4294
4295
4296
4297
4298
4299
4300
4301
4302
4303
4304
4305
4306
4307
4308
4309
4310
4311
4312
4313
4314
4315
4316
4317
4318
4319
4320
4321
4322
4323
4324
4325
4326
4327
4328
4329
4330
4331
4332
4333
4334
4335
4336
4337
4338
4339
4340
4341
4342
4343
4344
4345
4346
4347
4348
4349
4350
4351
4352
4353
4354
4355
4356
4357
4358
4359
4360
4361
4362
4363
4364
4365
4366
4367
4368
4369
4370
4371
4372
4373
4374
4375
4376
4377
4378
4379
4380
4381
4382
4383
4384
4385
4386
4387
4388
4389
4390
4391
4392
4393
4394
4395
4396
4397
4398
4399
4400
4401
4402
4403
4404
4405
4406
4407
4408
4409
4410
4411
4412
4413
4414
4415
4416
4417
4418
4419
4420
4421
4422
4423
4424
4425
4426
4427
4428
4429
4430
4431
4432
4433
4434
4435
4436
4437
4438
4439
4440
4441
4442
4443
4444
4445
4446
4447
4448
4449
4450
4451
4452
4453
4454
4455
4456
4457
4458
4459
4460
4461
4462
4463
4464
4465
4466
4467
4468
4469
4470
4471
4472
4473
4474
4475
4476
4477
4478
4479
4480
4481
4482
4483
4484
4485
4486
4487
4488
4489
4490
4491
4492
4493
4494
4495
4496
4497
4498
4499
4500
4501
4502
4503
4504
4505
4506
4507
4508
4509
4510
4511
4512
4513
4514
4515
4516
4517
4518
4519
4520
4521
4522
4523
4524
4525
4526
4527
4528
4529
4530
4531
4532
4533
4534
4535
4536
4537
4538
4539
4540
4541
4542
4543
4544
4545
4546
4547
4548
4549
4550
4551
4552
4553
4554
4555
4556
4557
4558
4559
4560
4561
4562
4563
4564
4565
4566
4567
4568
4569
4570
4571
4572
4573
4574
4575
4576
4577
4578
4579
4580
4581
4582
4583
4584
4585
4586
4587
4588
4589
4590
4591
4592
4593
4594
4595
4596
4597
4598
4599
4600
4601
4602
4603
4604
4605
4606
4607
4608
4609
4610
4611
4612
4613
4614
4615
4616
4617
4618
4619
4620
4621
4622
4623
4624
4625
4626
4627
4628
4629
4630
4631
4632
4633
4634
4635
4636
4637
4638
4639
4640
4641
4642
4643
4644
4645
4646
4647
4648
4649
4650
4651
4652
4653
4654
4655
4656
4657
4658
4659
4660
4661
4662
4663
4664
4665
4666
4667
4668
4669
4670
4671
4672
4673
4674
4675
4676
4677
4678
4679
4680
4681
4682
4683
4684
4685
4686
4687
4688
4689
4690
4691
4692
4693
4694
4695
4696
4697
4698
4699
4700
4701
4702
4703
4704
4705
4706
4707
4708
4709
4710
4711
4712
4713
4714
4715
4716
4717
4718
4719
4720
4721
4722
4723
4724
4725
4726
4727
4728
4729
4730
4731
4732
4733
4734
4735
4736
4737
4738
4739
4740
4741
4742
4743
4744
4745
4746
4747
4748
4749
4750
4751
4752
4753
4754
4755
4756
4757
4758
4759
4760
4761
4762
4763
4764
4765
4766
4767
4768
4769
4770
4771
4772
4773
4774
4775
4776
4777
4778
4779
4780
4781
4782
4783
4784
4785
4786
4787
4788
4789
4790
4791
4792
4793
4794
4795
4796
4797
4798
4799
4800
4801
4802
4803
4804
4805
4806
4807
4808
4809
4810
4811
4812
4813
4814
4815
4816
4817
4818
4819
4820
4821
4822
4823
4824
4825
4826
4827
4828
4829
4830
4831
4832
4833
4834
4835
4836
4837
4838
4839
4840
4841
4842
4843
4844
4845
4846
4847
4848
4849
4850
4851
4852
4853
4854
4855
4856
4857
4858
4859
4860
4861
4862
4863
4864
4865
4866
4867
4868
4869
4870
4871
4872
4873
4874
4875
4876
4877
4878
4879
4880
4881
4882
4883
4884
4885
4886
4887
4888
4889
4890
4891
4892
4893
4894
4895
4896
4897
4898
4899
4900
4901
4902
4903
4904
4905
4906
4907
4908
4909
4910
4911
4912
4913
4914
4915
4916
4917
4918
4919
4920
4921
4922
4923
4924
4925
4926
4927
4928
4929
4930
4931
4932
4933
4934
4935
4936
4937
4938
4939
4940
4941
4942
4943
4944
4945
4946
4947
4948
4949
4950
4951
4952
4953
4954
4955
4956
4957
4958
4959
4960
4961
4962
4963
4964
4965
4966
4967
4968
4969
4970
4971
4972
4973
4974
4975
4976
4977
4978
4979
4980
4981
4982
4983
4984
4985
4986
4987
4988
4989
4990
4991
4992
4993
4994
4995
4996
4997
4998
4999
5000
5001
5002
5003
5004
5005
5006
5007
5008
5009
5010
5011
5012
5013
5014
5015
5016
5017
5018
5019
5020
5021
5022
5023
5024
5025
5026
5027
5028
5029
5030
5031
5032
5033
5034
5035
5036
5037
5038
5039
5040
5041
5042
5043
5044
5045
5046
5047
5048
5049
5050
5051
5052
5053
5054
5055
5056
5057
5058
5059
5060
5061
5062
5063
5064
5065
5066
5067
5068
5069
5070
5071
5072
5073
5074
5075
5076
5077
5078
5079
5080
5081
5082
5083
5084
5085
5086
5087
5088
5089
5090
5091
5092
5093
5094
5095
5096
5097
5098
5099
5100
5101
5102
5103
5104
5105
5106
5107
5108
5109
5110
5111
5112
5113
5114
5115
5116
5117
5118
5119
5120
5121
5122
5123
5124
5125
5126
5127
5128
5129
5130
5131
5132
5133
5134
5135
5136
5137
5138
5139
5140
5141
5142
5143
5144
5145
5146
5147
5148
5149
5150
5151
5152
5153
5154
5155
5156
5157
5158
5159
5160
5161
5162
5163
5164
5165
5166
5167
5168
5169
5170
5171
5172
5173
5174
5175
5176
5177
5178
5179
5180
5181
5182
5183
5184
5185
5186
5187
5188
5189
5190
5191
5192
5193
5194
5195
5196
5197
5198
5199
5200
5201
5202
5203
5204
5205
5206
5207
5208
5209
5210
5211
5212
5213
5214
5215
5216
5217
5218
5219
5220
5221
5222
5223
5224
5225
5226
5227
5228
5229
5230
5231
5232
5233
5234
5235
5236
5237
5238
5239
5240
5241
5242
5243
5244
5245
5246
5247
5248
5249
5250
5251
5252
5253
5254
5255
5256
5257
5258
5259
5260
5261
5262
5263
5264
5265
5266
5267
5268
5269
5270
5271
5272
5273
5274
5275
5276
5277
5278
5279
5280
5281
5282
5283
5284
5285
5286
5287
5288
5289
5290
5291
5292
5293
5294
5295
5296
5297
5298
5299
5300
5301
5302
5303
5304
5305
5306
5307
5308
5309
5310
5311
5312
5313
5314
5315
5316
5317
5318
5319
5320
5321
5322
5323
5324
5325
5326
5327
5328
5329
5330
5331
5332
5333
5334
5335
5336
5337
5338
5339
5340
5341
5342
5343
5344
5345
5346
5347
5348
5349
5350
5351
5352
5353
5354
5355
5356
5357
5358
5359
5360
5361
5362
5363
5364
5365
5366
5367
5368
5369
5370
5371
5372
5373
5374
5375
5376
5377
5378
5379
5380
5381
5382
5383
5384
5385
5386
5387
5388
5389
5390
5391
5392
5393
5394
5395
5396
5397
5398
5399
5400
5401
5402
5403
5404
5405
5406
5407
5408
5409
5410
5411
5412
5413
5414
5415
5416
5417
5418
5419
5420
5421
5422
5423
5424
5425
5426
5427
5428
5429
5430
5431
5432
5433
5434
5435
5436
5437
5438
5439
5440
5441
/*
 * Copyright (c) 1999-2009 Apple Inc. All rights reserved.
 *
 * @APPLE_OSREFERENCE_LICENSE_HEADER_START@
 * 
 * This file contains Original Code and/or Modifications of Original Code
 * as defined in and that are subject to the Apple Public Source License
 * Version 2.0 (the 'License'). You may not use this file except in
 * compliance with the License. The rights granted to you under the License
 * may not be used to create, or enable the creation or redistribution of,
 * unlawful or unlicensed copies of an Apple operating system, or to
 * circumvent, violate, or enable the circumvention or violation of, any
 * terms of an Apple operating system software license agreement.
 * 
 * Please obtain a copy of the License at
 * http://www.opensource.apple.com/apsl/ and read it before using this file.
 * 
 * The Original Code and all software distributed under the License are
 * distributed on an 'AS IS' basis, WITHOUT WARRANTY OF ANY KIND, EITHER
 * EXPRESS OR IMPLIED, AND APPLE HEREBY DISCLAIMS ALL SUCH WARRANTIES,
 * INCLUDING WITHOUT LIMITATION, ANY WARRANTIES OF MERCHANTABILITY,
 * FITNESS FOR A PARTICULAR PURPOSE, QUIET ENJOYMENT OR NON-INFRINGEMENT.
 * Please see the License for the specific language governing rights and
 * limitations under the License.
 * 
 * @APPLE_OSREFERENCE_LICENSE_HEADER_END@
 */
/*
 * Copyright (c) 1991, 1993, 1994
 *	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. All advertising materials mentioning features or use of this software
 *    must display the following acknowledgement:
 *	This product includes software developed by the University of
 *	California, Berkeley and its contributors.
 * 4. 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.
 *
 *      hfs_vfsops.c
 *  derived from	@(#)ufs_vfsops.c	8.8 (Berkeley) 5/20/95
 *
 *      (c) Copyright 1997-2002 Apple Computer, Inc. All rights reserved.
 *
 *      hfs_vfsops.c -- VFS layer for loadable HFS file system.
 *
 */
#include <sys/param.h>
#include <sys/systm.h>
#include <sys/kauth.h>

#include <sys/ubc.h>
#include <sys/ubc_internal.h>
#include <sys/vnode_internal.h>
#include <sys/mount_internal.h>
#include <sys/sysctl.h>
#include <sys/malloc.h>
#include <sys/stat.h>
#include <sys/quota.h>
#include <sys/disk.h>
#include <sys/paths.h>
#include <sys/utfconv.h>
#include <sys/kdebug.h>
#include <sys/fslog.h>

#include <kern/locks.h>

#include <vfs/vfs_journal.h>

#include <miscfs/specfs/specdev.h>
#include <hfs/hfs_mount.h>

#include <libkern/crypto/md5.h>
#include <uuid/uuid.h>

#include "hfs.h"
#include "hfs_catalog.h"
#include "hfs_cnode.h"
#include "hfs_dbg.h"
#include "hfs_endian.h"
#include "hfs_hotfiles.h"
#include "hfs_quota.h"

#include "hfscommon/headers/FileMgrInternal.h"
#include "hfscommon/headers/BTreesInternal.h"

#if	HFS_DIAGNOSTIC
int hfs_dbg_all = 0;
int hfs_dbg_err = 0;
#endif


lck_grp_attr_t *  hfs_group_attr;
lck_attr_t *  hfs_lock_attr;
lck_grp_t *  hfs_mutex_group;
lck_grp_t *  hfs_rwlock_group;

extern struct vnodeopv_desc hfs_vnodeop_opv_desc;
extern struct vnodeopv_desc hfs_std_vnodeop_opv_desc;

/* not static so we can re-use in hfs_readwrite.c for build_path calls */
int hfs_vfs_vget(struct mount *mp, ino64_t ino, struct vnode **vpp, vfs_context_t context);

static int hfs_changefs(struct mount *mp, struct hfs_mount_args *args);
static int hfs_fhtovp(struct mount *mp, int fhlen, unsigned char *fhp, struct vnode **vpp, vfs_context_t context);
static int hfs_flushfiles(struct mount *, int, struct proc *);
static int hfs_flushMDB(struct hfsmount *hfsmp, int waitfor, int altflush);
static int hfs_getmountpoint(struct vnode *vp, struct hfsmount **hfsmpp);
static int hfs_init(struct vfsconf *vfsp);
static int hfs_mount(struct mount *mp, vnode_t  devvp, user_addr_t data, vfs_context_t context);
static int hfs_mountfs(struct vnode *devvp, struct mount *mp, struct hfs_mount_args *args, int journal_replay_only, vfs_context_t context);
static int hfs_reload(struct mount *mp);
static int hfs_vfs_root(struct mount *mp, struct vnode **vpp, vfs_context_t context);
static int hfs_quotactl(struct mount *, int, uid_t, caddr_t, vfs_context_t context);
static int hfs_start(struct mount *mp, int flags, vfs_context_t context);
static int hfs_statfs(struct mount *mp, register struct vfsstatfs *sbp, vfs_context_t context);
static int hfs_sync(struct mount *mp, int waitfor, vfs_context_t context);
static int hfs_sysctl(int *name, u_int namelen, user_addr_t oldp, size_t *oldlenp, 
                      user_addr_t newp, size_t newlen, vfs_context_t context);
static int hfs_unmount(struct mount *mp, int mntflags, vfs_context_t context);
static int hfs_vptofh(struct vnode *vp, int *fhlenp, unsigned char *fhp, vfs_context_t context);

static int hfs_reclaimspace(struct hfsmount *hfsmp, u_int32_t startblk, u_int32_t reclaimblks, vfs_context_t context);
static int hfs_overlapped_overflow_extents(struct hfsmount *hfsmp, u_int32_t startblk,
                                           u_int32_t catblks, u_int32_t fileID, int rsrcfork);
static int hfs_journal_replay(vnode_t devvp, vfs_context_t context);


/*
 * Called by vfs_mountroot when mounting HFS Plus as root.
 */

__private_extern__
int
hfs_mountroot(mount_t mp, vnode_t rvp, vfs_context_t context)
{
	struct hfsmount *hfsmp;
	ExtendedVCB *vcb;
	struct vfsstatfs *vfsp;
	int error;

	if ((error = hfs_mountfs(rvp, mp, NULL, 0, context)))
		return (error);

	/* Init hfsmp */
	hfsmp = VFSTOHFS(mp);

	hfsmp->hfs_uid = UNKNOWNUID;
	hfsmp->hfs_gid = UNKNOWNGID;
	hfsmp->hfs_dir_mask = (S_IRWXU | S_IRGRP|S_IXGRP | S_IROTH|S_IXOTH); /* 0755 */
	hfsmp->hfs_file_mask = (S_IRWXU | S_IRGRP|S_IXGRP | S_IROTH|S_IXOTH); /* 0755 */

	/* Establish the free block reserve. */
	vcb = HFSTOVCB(hfsmp);
	vcb->reserveBlocks = ((u_int64_t)vcb->totalBlocks * HFS_MINFREE) / 100;
	vcb->reserveBlocks = MIN(vcb->reserveBlocks, HFS_MAXRESERVE / vcb->blockSize);

	vfsp = vfs_statfs(mp);
	(void)hfs_statfs(mp, vfsp, NULL);

	return (0);
}


/*
 * VFS Operations.
 *
 * mount system call
 */

static int
hfs_mount(struct mount *mp, vnode_t devvp, user_addr_t data, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	struct hfsmount *hfsmp = NULL;
	struct hfs_mount_args args;
	int retval = E_NONE;
	u_int32_t cmdflags;

	if ((retval = copyin(data, (caddr_t)&args, sizeof(args)))) {
		return (retval);
	}
	cmdflags = (u_int32_t)vfs_flags(mp) & MNT_CMDFLAGS;
	if (cmdflags & MNT_UPDATE) {
		hfsmp = VFSTOHFS(mp);

		/* Reload incore data after an fsck. */
		if (cmdflags & MNT_RELOAD) {
			if (vfs_isrdonly(mp))
				return hfs_reload(mp);
			else
				return (EINVAL);
		}

		/* Change to a read-only file system. */
		if (((hfsmp->hfs_flags & HFS_READ_ONLY) == 0) &&
		    vfs_isrdonly(mp)) {
			int flags;

			/* Set flag to indicate that a downgrade to read-only
			 * is in progress and therefore block any further 
			 * modifications to the file system.
			 */
			hfs_global_exclusive_lock_acquire(hfsmp);
			hfsmp->hfs_flags |= HFS_RDONLY_DOWNGRADE;
			hfsmp->hfs_downgrading_proc = current_thread();
			hfs_global_exclusive_lock_release(hfsmp);

			/* use VFS_SYNC to push out System (btree) files */
			retval = VFS_SYNC(mp, MNT_WAIT, context);
			if (retval && ((cmdflags & MNT_FORCE) == 0)) {
				hfsmp->hfs_flags &= ~HFS_RDONLY_DOWNGRADE;
				hfsmp->hfs_downgrading_proc = NULL;
				goto out;
			}
		
			flags = WRITECLOSE;
			if (cmdflags & MNT_FORCE)
				flags |= FORCECLOSE;
				
			if ((retval = hfs_flushfiles(mp, flags, p))) {
				hfsmp->hfs_flags &= ~HFS_RDONLY_DOWNGRADE;
				hfsmp->hfs_downgrading_proc = NULL;
				goto out;
			}

			/* mark the volume cleanly unmounted */
			hfsmp->vcbAtrb |= kHFSVolumeUnmountedMask;
			retval = hfs_flushvolumeheader(hfsmp, MNT_WAIT, 0);
			hfsmp->hfs_flags |= HFS_READ_ONLY;

			/* also get the volume bitmap blocks */
			if (!retval) {
				if (vnode_mount(hfsmp->hfs_devvp) == mp) {
					retval = hfs_fsync(hfsmp->hfs_devvp, MNT_WAIT, 0, p);
				} else {
					vnode_get(hfsmp->hfs_devvp);
					retval = VNOP_FSYNC(hfsmp->hfs_devvp, MNT_WAIT, context);
					vnode_put(hfsmp->hfs_devvp);
				}
			}
			if (retval) {
				hfsmp->hfs_flags &= ~HFS_RDONLY_DOWNGRADE;
				hfsmp->hfs_downgrading_proc = NULL;
				hfsmp->hfs_flags &= ~HFS_READ_ONLY;
				goto out;
			}
			if (hfsmp->jnl) {
			    hfs_global_exclusive_lock_acquire(hfsmp);

			    journal_close(hfsmp->jnl);
			    hfsmp->jnl = NULL;

			    // Note: we explicitly don't want to shutdown
			    //       access to the jvp because we may need
			    //       it later if we go back to being read-write.

			    hfs_global_exclusive_lock_release(hfsmp);
			}

			hfsmp->hfs_downgrading_proc = NULL;
		}

		/* Change to a writable file system. */
		if (vfs_iswriteupgrade(mp)) {

			/*
			 * On inconsistent disks, do not allow read-write mount
			 * unless it is the boot volume being mounted.
			 */
			if (!(vfs_flags(mp) & MNT_ROOTFS) &&
					(hfsmp->vcbAtrb & kHFSVolumeInconsistentMask)) {
				retval = EINVAL;
				goto out;
			}

			// If the journal was shut-down previously because we were
			// asked to be read-only, let's start it back up again now
			
			if (   (HFSTOVCB(hfsmp)->vcbAtrb & kHFSVolumeJournaledMask)
			    && hfsmp->jnl == NULL
			    && hfsmp->jvp != NULL) {
			    int jflags;

			    if (hfsmp->hfs_flags & HFS_NEED_JNL_RESET) {
					jflags = JOURNAL_RESET;
			    } else {
					jflags = 0;
			    }
			    
			    hfs_global_exclusive_lock_acquire(hfsmp);

			    hfsmp->jnl = journal_open(hfsmp->jvp,
						      (hfsmp->jnl_start * HFSTOVCB(hfsmp)->blockSize) + (off_t)HFSTOVCB(hfsmp)->hfsPlusIOPosOffset,
						      hfsmp->jnl_size,
						      hfsmp->hfs_devvp,
						      hfsmp->hfs_logical_block_size,
						      jflags,
						      0,
						      hfs_sync_metadata, hfsmp->hfs_mp);

			    hfs_global_exclusive_lock_release(hfsmp);

			    if (hfsmp->jnl == NULL) {
				retval = EINVAL;
				goto out;
			    } else {
				hfsmp->hfs_flags &= ~HFS_NEED_JNL_RESET;
			    }

			}

			/* See if we need to erase unused Catalog nodes due to <rdar://problem/6947811>. */
			retval = hfs_erase_unused_nodes(hfsmp);
			if (retval != E_NONE)
				goto out;
			
			/* Only clear HFS_READ_ONLY after a successful write */
			hfsmp->hfs_flags &= ~HFS_READ_ONLY;

			/* If this mount point was downgraded from read-write 
			 * to read-only, clear that information as we are now 
			 * moving back to read-write.
			 */
			hfsmp->hfs_flags &= ~HFS_RDONLY_DOWNGRADE;
			hfsmp->hfs_downgrading_proc = NULL;

			/* mark the volume dirty (clear clean unmount bit) */
			hfsmp->vcbAtrb &= ~kHFSVolumeUnmountedMask;

			retval = hfs_flushvolumeheader(hfsmp, MNT_WAIT, 0);
			if (retval != E_NONE)
				goto out;

			if (!(hfsmp->hfs_flags & (HFS_READ_ONLY | HFS_STANDARD))) {
				/* Setup private/hidden directories for hardlinks. */
				hfs_privatedir_init(hfsmp, FILE_HARDLINKS);
				hfs_privatedir_init(hfsmp, DIR_HARDLINKS);

				hfs_remove_orphans(hfsmp);

				/*
				 * Allow hot file clustering if conditions allow.
				 */
				if (hfsmp->hfs_flags & HFS_METADATA_ZONE) {
					(void) hfs_recording_init(hfsmp);
				}
				/* Force ACLs on HFS+ file systems. */
				if (vfs_extendedsecurity(HFSTOVFS(hfsmp)) == 0) {
					vfs_setextendedsecurity(HFSTOVFS(hfsmp));
				}
			}
		}

		/* Update file system parameters. */
		retval = hfs_changefs(mp, &args);

	} else /* not an update request */ {

		/* Set the mount flag to indicate that we support volfs  */
		vfs_setflags(mp, (u_int64_t)((unsigned int)MNT_DOVOLFS));

		retval = hfs_mountfs(devvp, mp, &args, 0, context);
	}
out:
	if (retval == 0) {
		(void)hfs_statfs(mp, vfs_statfs(mp), context);
	}
	return (retval);
}


struct hfs_changefs_cargs {
	struct hfsmount *hfsmp;
        int		namefix;
        int		permfix;
        int		permswitch;
};

static int
hfs_changefs_callback(struct vnode *vp, void *cargs)
{
	ExtendedVCB *vcb;
	struct cnode *cp;
	struct cat_desc cndesc;
	struct cat_attr cnattr;
	struct hfs_changefs_cargs *args;
	int lockflags;
	int error;

	args = (struct hfs_changefs_cargs *)cargs;

	cp = VTOC(vp);
	vcb = HFSTOVCB(args->hfsmp);

	lockflags = hfs_systemfile_lock(args->hfsmp, SFL_CATALOG, HFS_SHARED_LOCK);
	error = cat_lookup(args->hfsmp, &cp->c_desc, 0, &cndesc, &cnattr, NULL, NULL);
	hfs_systemfile_unlock(args->hfsmp, lockflags);
	if (error) {
	        /*
		 * If we couldn't find this guy skip to the next one
		 */
	        if (args->namefix)
		        cache_purge(vp);

		return (VNODE_RETURNED);
	}
	/*
	 * Get the real uid/gid and perm mask from disk.
	 */
	if (args->permswitch || args->permfix) {
	        cp->c_uid = cnattr.ca_uid;
		cp->c_gid = cnattr.ca_gid;
		cp->c_mode = cnattr.ca_mode;
	}
	/*
	 * If we're switching name converters then...
	 *   Remove the existing entry from the namei cache.
	 *   Update name to one based on new encoder.
	 */
	if (args->namefix) {
	        cache_purge(vp);
		replace_desc(cp, &cndesc);

		if (cndesc.cd_cnid == kHFSRootFolderID) {
		        strlcpy((char *)vcb->vcbVN, (const char *)cp->c_desc.cd_nameptr, NAME_MAX+1);
			cp->c_desc.cd_encoding = args->hfsmp->hfs_encoding;
		}
	} else {
	        cat_releasedesc(&cndesc);
	}
	return (VNODE_RETURNED);
}

/* Change fs mount parameters */
static int
hfs_changefs(struct mount *mp, struct hfs_mount_args *args)
{
	int retval = 0;
	int namefix, permfix, permswitch;
	struct hfsmount *hfsmp;
	ExtendedVCB *vcb;
	hfs_to_unicode_func_t	get_unicode_func;
	unicode_to_hfs_func_t	get_hfsname_func;
	u_int32_t old_encoding = 0;
	struct hfs_changefs_cargs cargs;
	u_int32_t mount_flags;

	hfsmp = VFSTOHFS(mp);
	vcb = HFSTOVCB(hfsmp);
	mount_flags = (unsigned int)vfs_flags(mp);

	hfsmp->hfs_flags |= HFS_IN_CHANGEFS;
	
	permswitch = (((hfsmp->hfs_flags & HFS_UNKNOWN_PERMS) &&
	               ((mount_flags & MNT_UNKNOWNPERMISSIONS) == 0)) ||
	              (((hfsmp->hfs_flags & HFS_UNKNOWN_PERMS) == 0) &&
	               (mount_flags & MNT_UNKNOWNPERMISSIONS)));

	/* The root filesystem must operate with actual permissions: */
	if (permswitch && (mount_flags & MNT_ROOTFS) && (mount_flags & MNT_UNKNOWNPERMISSIONS)) {
		vfs_clearflags(mp, (u_int64_t)((unsigned int)MNT_UNKNOWNPERMISSIONS));	/* Just say "No". */
		retval = EINVAL;
		goto exit;
	}
	if (mount_flags & MNT_UNKNOWNPERMISSIONS)
		hfsmp->hfs_flags |= HFS_UNKNOWN_PERMS;
	else
		hfsmp->hfs_flags &= ~HFS_UNKNOWN_PERMS;

	namefix = permfix = 0;

	/*
	 * Tracking of hot files requires up-to-date access times.  So if
	 * access time updates are disabled, we must also disable hot files.
	 */
	if (mount_flags & MNT_NOATIME) {
		(void) hfs_recording_suspend(hfsmp);
	}
	
	/* Change the timezone (Note: this affects all hfs volumes and hfs+ volume create dates) */
	if (args->hfs_timezone.tz_minuteswest != VNOVAL) {
		gTimeZone = args->hfs_timezone;
	}

	/* Change the default uid, gid and/or mask */
	if ((args->hfs_uid != (uid_t)VNOVAL) && (hfsmp->hfs_uid != args->hfs_uid)) {
		hfsmp->hfs_uid = args->hfs_uid;
		if (vcb->vcbSigWord == kHFSPlusSigWord)
			++permfix;
	}
	if ((args->hfs_gid != (gid_t)VNOVAL) && (hfsmp->hfs_gid != args->hfs_gid)) {
		hfsmp->hfs_gid = args->hfs_gid;
		if (vcb->vcbSigWord == kHFSPlusSigWord)
			++permfix;
	}
	if (args->hfs_mask != (mode_t)VNOVAL) {
		if (hfsmp->hfs_dir_mask != (args->hfs_mask & ALLPERMS)) {
			hfsmp->hfs_dir_mask = args->hfs_mask & ALLPERMS;
			hfsmp->hfs_file_mask = args->hfs_mask & ALLPERMS;
			if ((args->flags != VNOVAL) && (args->flags & HFSFSMNT_NOXONFILES))
				hfsmp->hfs_file_mask = (args->hfs_mask & DEFFILEMODE);
			if (vcb->vcbSigWord == kHFSPlusSigWord)
				++permfix;
		}
	}
	
	/* Change the hfs encoding value (hfs only) */
	if ((vcb->vcbSigWord == kHFSSigWord)	&&
	    (args->hfs_encoding != (u_int32_t)VNOVAL)              &&
	    (hfsmp->hfs_encoding != args->hfs_encoding)) {

		retval = hfs_getconverter(args->hfs_encoding, &get_unicode_func, &get_hfsname_func);
		if (retval)
			goto exit;

		/*
		 * Connect the new hfs_get_unicode converter but leave
		 * the old hfs_get_hfsname converter in place so that
		 * we can lookup existing vnodes to get their correctly
		 * encoded names.
		 *
		 * When we're all finished, we can then connect the new
		 * hfs_get_hfsname converter and release our interest
		 * in the old converters.
		 */
		hfsmp->hfs_get_unicode = get_unicode_func;
		old_encoding = hfsmp->hfs_encoding;
		hfsmp->hfs_encoding = args->hfs_encoding;
		++namefix;
	}

	if (!(namefix || permfix || permswitch))
		goto exit;

	/* XXX 3762912 hack to support HFS filesystem 'owner' */
	if (permfix)
		vfs_setowner(mp,
		    hfsmp->hfs_uid == UNKNOWNUID ? KAUTH_UID_NONE : hfsmp->hfs_uid,
		    hfsmp->hfs_gid == UNKNOWNGID ? KAUTH_GID_NONE : hfsmp->hfs_gid);
	
	/*
	 * For each active vnode fix things that changed
	 *
	 * Note that we can visit a vnode more than once
	 * and we can race with fsync.
	 *
	 * hfs_changefs_callback will be called for each vnode
	 * hung off of this mount point
	 *
	 * The vnode will be properly referenced and unreferenced 
	 * around the callback
	 */
	cargs.hfsmp = hfsmp;
	cargs.namefix = namefix;
	cargs.permfix = permfix;
	cargs.permswitch = permswitch;

	vnode_iterate(mp, 0, hfs_changefs_callback, (void *)&cargs);

	/*
	 * If we're switching name converters we can now
	 * connect the new hfs_get_hfsname converter and
	 * release our interest in the old converters.
	 */
	if (namefix) {
		hfsmp->hfs_get_hfsname = get_hfsname_func;
		vcb->volumeNameEncodingHint = args->hfs_encoding;
		(void) hfs_relconverter(old_encoding);
	}
exit:
	hfsmp->hfs_flags &= ~HFS_IN_CHANGEFS;
	return (retval);
}


struct hfs_reload_cargs {
	struct hfsmount *hfsmp;
        int		error;
};

static int
hfs_reload_callback(struct vnode *vp, void *cargs)
{
	struct cnode *cp;
	struct hfs_reload_cargs *args;
	int lockflags;

	args = (struct hfs_reload_cargs *)cargs;
	/*
	 * flush all the buffers associated with this node
	 */
	(void) buf_invalidateblks(vp, 0, 0, 0);

	cp = VTOC(vp);
	/* 
	 * Remove any directory hints
	 */
	if (vnode_isdir(vp))
	        hfs_reldirhints(cp, 0);

	/*
	 * Re-read cnode data for all active vnodes (non-metadata files).
	 */
	if (!vnode_issystem(vp) && !VNODE_IS_RSRC(vp)) {
	        struct cat_fork *datafork;
		struct cat_desc desc;

		datafork = cp->c_datafork ? &cp->c_datafork->ff_data : NULL;

		/* lookup by fileID since name could have changed */
		lockflags = hfs_systemfile_lock(args->hfsmp, SFL_CATALOG, HFS_SHARED_LOCK);
		args->error = cat_idlookup(args->hfsmp, cp->c_fileid, 0, &desc, &cp->c_attr, datafork);
		hfs_systemfile_unlock(args->hfsmp, lockflags);
		if (args->error) {
		        return (VNODE_RETURNED_DONE);
		}

		/* update cnode's catalog descriptor */
		(void) replace_desc(cp, &desc);
	}
	return (VNODE_RETURNED);
}

/*
 * Reload all incore data for a filesystem (used after running fsck on
 * the root filesystem and finding things to fix). The filesystem must
 * be mounted read-only.
 *
 * Things to do to update the mount:
 *	invalidate all cached meta-data.
 *	invalidate all inactive vnodes.
 *	invalidate all cached file data.
 *	re-read volume header from disk.
 *	re-load meta-file info (extents, file size).
 *	re-load B-tree header data.
 *	re-read cnode data for all active vnodes.
 */
static int
hfs_reload(struct mount *mountp)
{
	register struct vnode *devvp;
	struct buf *bp;
	int error, i;
	struct hfsmount *hfsmp;
	struct HFSPlusVolumeHeader *vhp;
	ExtendedVCB *vcb;
	struct filefork *forkp;
    	struct cat_desc cndesc;
	struct hfs_reload_cargs args;
	daddr64_t priIDSector;

    	hfsmp = VFSTOHFS(mountp);
	vcb = HFSTOVCB(hfsmp);

	if (vcb->vcbSigWord == kHFSSigWord)
		return (EINVAL);	/* rooting from HFS is not supported! */

	/*
	 * Invalidate all cached meta-data.
	 */
	devvp = hfsmp->hfs_devvp;
	if (buf_invalidateblks(devvp, 0, 0, 0))
		panic("hfs_reload: dirty1");

	args.hfsmp = hfsmp;
	args.error = 0;
	/*
	 * hfs_reload_callback will be called for each vnode
	 * hung off of this mount point that can't be recycled...
	 * vnode_iterate will recycle those that it can (the VNODE_RELOAD option)
	 * the vnode will be in an 'unbusy' state (VNODE_WAIT) and 
	 * properly referenced and unreferenced around the callback
	 */
	vnode_iterate(mountp, VNODE_RELOAD | VNODE_WAIT, hfs_reload_callback, (void *)&args);

	if (args.error)
	        return (args.error);

	/*
	 * Re-read VolumeHeader from disk.
	 */
	priIDSector = (daddr64_t)((vcb->hfsPlusIOPosOffset / hfsmp->hfs_logical_block_size) + 
			HFS_PRI_SECTOR(hfsmp->hfs_logical_block_size));

	error = (int)buf_meta_bread(hfsmp->hfs_devvp,
			HFS_PHYSBLK_ROUNDDOWN(priIDSector, hfsmp->hfs_log_per_phys),
			hfsmp->hfs_physical_block_size, NOCRED, &bp);
	if (error) {
        	if (bp != NULL)
        		buf_brelse(bp);
		return (error);
	}

	vhp = (HFSPlusVolumeHeader *) (buf_dataptr(bp) + HFS_PRI_OFFSET(hfsmp->hfs_physical_block_size));

	/* Do a quick sanity check */
	if ((SWAP_BE16(vhp->signature) != kHFSPlusSigWord &&
	     SWAP_BE16(vhp->signature) != kHFSXSigWord) ||
	    (SWAP_BE16(vhp->version) != kHFSPlusVersion &&
	     SWAP_BE16(vhp->version) != kHFSXVersion) ||
	    SWAP_BE32(vhp->blockSize) != vcb->blockSize) {
		buf_brelse(bp);
		return (EIO);
	}

	vcb->vcbLsMod		= to_bsd_time(SWAP_BE32(vhp->modifyDate));
	vcb->vcbAtrb		= SWAP_BE32 (vhp->attributes);
	vcb->vcbJinfoBlock  = SWAP_BE32(vhp->journalInfoBlock);
	vcb->vcbClpSiz		= SWAP_BE32 (vhp->rsrcClumpSize);
	vcb->vcbNxtCNID		= SWAP_BE32 (vhp->nextCatalogID);
	vcb->vcbVolBkUp		= to_bsd_time(SWAP_BE32(vhp->backupDate));
	vcb->vcbWrCnt		= SWAP_BE32 (vhp->writeCount);
	vcb->vcbFilCnt		= SWAP_BE32 (vhp->fileCount);
	vcb->vcbDirCnt		= SWAP_BE32 (vhp->folderCount);
	HFS_UPDATE_NEXT_ALLOCATION(vcb, SWAP_BE32 (vhp->nextAllocation));
	vcb->totalBlocks	= SWAP_BE32 (vhp->totalBlocks);
	vcb->freeBlocks		= SWAP_BE32 (vhp->freeBlocks);
	vcb->encodingsBitmap	= SWAP_BE64 (vhp->encodingsBitmap);
	bcopy(vhp->finderInfo, vcb->vcbFndrInfo, sizeof(vhp->finderInfo));    
	vcb->localCreateDate	= SWAP_BE32 (vhp->createDate); /* hfs+ create date is in local time */ 

	/*
	 * Re-load meta-file vnode data (extent info, file size, etc).
	 */
	forkp = VTOF((struct vnode *)vcb->extentsRefNum);
	for (i = 0; i < kHFSPlusExtentDensity; i++) {
		forkp->ff_extents[i].startBlock =
			SWAP_BE32 (vhp->extentsFile.extents[i].startBlock);
		forkp->ff_extents[i].blockCount =
			SWAP_BE32 (vhp->extentsFile.extents[i].blockCount);
	}
	forkp->ff_size      = SWAP_BE64 (vhp->extentsFile.logicalSize);
	forkp->ff_blocks    = SWAP_BE32 (vhp->extentsFile.totalBlocks);
	forkp->ff_clumpsize = SWAP_BE32 (vhp->extentsFile.clumpSize);


	forkp = VTOF((struct vnode *)vcb->catalogRefNum);
	for (i = 0; i < kHFSPlusExtentDensity; i++) {
		forkp->ff_extents[i].startBlock	=
			SWAP_BE32 (vhp->catalogFile.extents[i].startBlock);
		forkp->ff_extents[i].blockCount	=
			SWAP_BE32 (vhp->catalogFile.extents[i].blockCount);
	}
	forkp->ff_size      = SWAP_BE64 (vhp->catalogFile.logicalSize);
	forkp->ff_blocks    = SWAP_BE32 (vhp->catalogFile.totalBlocks);
	forkp->ff_clumpsize = SWAP_BE32 (vhp->catalogFile.clumpSize);

	if (hfsmp->hfs_attribute_vp) {
		forkp = VTOF(hfsmp->hfs_attribute_vp);
		for (i = 0; i < kHFSPlusExtentDensity; i++) {
			forkp->ff_extents[i].startBlock	=
				SWAP_BE32 (vhp->attributesFile.extents[i].startBlock);
			forkp->ff_extents[i].blockCount	=
				SWAP_BE32 (vhp->attributesFile.extents[i].blockCount);
		}
		forkp->ff_size      = SWAP_BE64 (vhp->attributesFile.logicalSize);
		forkp->ff_blocks    = SWAP_BE32 (vhp->attributesFile.totalBlocks);
		forkp->ff_clumpsize = SWAP_BE32 (vhp->attributesFile.clumpSize);
	}

	forkp = VTOF((struct vnode *)vcb->allocationsRefNum);
	for (i = 0; i < kHFSPlusExtentDensity; i++) {
		forkp->ff_extents[i].startBlock	=
			SWAP_BE32 (vhp->allocationFile.extents[i].startBlock);
		forkp->ff_extents[i].blockCount	=
			SWAP_BE32 (vhp->allocationFile.extents[i].blockCount);
	}
	forkp->ff_size      = SWAP_BE64 (vhp->allocationFile.logicalSize);
	forkp->ff_blocks    = SWAP_BE32 (vhp->allocationFile.totalBlocks);
	forkp->ff_clumpsize = SWAP_BE32 (vhp->allocationFile.clumpSize);

	buf_brelse(bp);
	vhp = NULL;

	/*
	 * Re-load B-tree header data
	 */
	forkp = VTOF((struct vnode *)vcb->extentsRefNum);
	if ( (error = MacToVFSError( BTReloadData((FCB*)forkp) )) )
		return (error);

	forkp = VTOF((struct vnode *)vcb->catalogRefNum);
	if ( (error = MacToVFSError( BTReloadData((FCB*)forkp) )) )
		return (error);

	if (hfsmp->hfs_attribute_vp) {
		forkp = VTOF(hfsmp->hfs_attribute_vp);
		if ( (error = MacToVFSError( BTReloadData((FCB*)forkp) )) )
			return (error);
	}

	/* Reload the volume name */
	if ((error = cat_idlookup(hfsmp, kHFSRootFolderID, 0, &cndesc, NULL, NULL)))
		return (error);
	vcb->volumeNameEncodingHint = cndesc.cd_encoding;
	bcopy(cndesc.cd_nameptr, vcb->vcbVN, min(255, cndesc.cd_namelen));
	cat_releasedesc(&cndesc);

	/* Re-establish private/hidden directories. */
	hfs_privatedir_init(hfsmp, FILE_HARDLINKS);
	hfs_privatedir_init(hfsmp, DIR_HARDLINKS);

	/* In case any volume information changed to trigger a notification */
	hfs_generate_volume_notifications(hfsmp);
    
	return (0);
}



static void
hfs_syncer(void *arg0, void *unused)
{
#pragma unused(unused)

    struct hfsmount *hfsmp = arg0;
    clock_sec_t secs;
    clock_usec_t usecs;
    uint32_t delay = HFS_META_DELAY;
    uint64_t now;
    static int no_max=1;

    clock_get_calendar_microtime(&secs, &usecs);
    now = ((uint64_t)secs * 1000000ULL) + (uint64_t)usecs;

    //
    // If the amount of pending writes is more than our limit, wait
    // for 2/3 of it to drain and then flush the journal. 
    //
    if (hfsmp->hfs_mp->mnt_pending_write_size > hfsmp->hfs_max_pending_io) {
	    int counter=0;
	    uint64_t pending_io, start, rate;
	    
	    no_max = 0;

	    hfs_start_transaction(hfsmp);   // so we hold off any new i/o's

	    pending_io = hfsmp->hfs_mp->mnt_pending_write_size;
	    
	    clock_get_calendar_microtime(&secs, &usecs);
	    start = ((uint64_t)secs * 1000000ULL) + (uint64_t)usecs;

	    while(hfsmp->hfs_mp->mnt_pending_write_size > (pending_io/3) && counter++ < 500) {
		    tsleep((caddr_t)hfsmp, PRIBIO, "hfs-wait-for-io-to-drain", 10);
	    }

	    if (counter >= 500) {
		    printf("hfs: timed out waiting for io to drain (%lld)\n", (int64_t)hfsmp->hfs_mp->mnt_pending_write_size);
	    }

	    if (hfsmp->jnl) {
		    journal_flush(hfsmp->jnl);
	    } else {
		    hfs_sync(hfsmp->hfs_mp, MNT_WAIT, vfs_context_kernel());
	    }

	    clock_get_calendar_microtime(&secs, &usecs);
	    now = ((uint64_t)secs * 1000000ULL) + (uint64_t)usecs;
	    hfsmp->hfs_last_sync_time = now;
	    rate = ((pending_io * 1000000ULL) / (now - start));     // yields bytes per second

	    hfs_end_transaction(hfsmp);
	    
	    //
	    // If a reasonable amount of time elapsed then check the
	    // i/o rate.  If it's taking less than 1 second or more
	    // than 2 seconds, adjust hfs_max_pending_io so that we
	    // will allow about 1.5 seconds of i/o to queue up.
	    //
	    if ((now - start) >= 300000) {
		    uint64_t scale = (pending_io * 100) / rate;
		    
		    if (scale < 100 || scale > 200) {
			    // set it so that it should take about 1.5 seconds to drain
			    hfsmp->hfs_max_pending_io = (rate * 150ULL) / 100ULL;
		    }
	    }
  
    } else if (   ((now - hfsmp->hfs_last_sync_time) >= 5000000ULL)
	       || (((now - hfsmp->hfs_last_sync_time) >= 100000LL)
		   && ((now - hfsmp->hfs_last_sync_request_time) >= 100000LL)
		   && (hfsmp->hfs_active_threads == 0)
		   && (hfsmp->hfs_global_lock_nesting == 0))) {

	    //
	    // Flush the journal if more than 5 seconds elapsed since
	    // the last sync OR we have not sync'ed recently and the
	    // last sync request time was more than 100 milliseconds
	    // ago and no one is in the middle of a transaction right
	    // now.  Else we defer the sync and reschedule it.
	    //
	    if (hfsmp->jnl) {
		    lck_rw_lock_shared(&hfsmp->hfs_global_lock);

		    journal_flush(hfsmp->jnl);

		    lck_rw_unlock_shared(&hfsmp->hfs_global_lock);
	    } else {
		    hfs_sync(hfsmp->hfs_mp, MNT_WAIT, vfs_context_kernel());
	    }

	    clock_get_calendar_microtime(&secs, &usecs);
	    now = ((uint64_t)secs * 1000000ULL) + (uint64_t)usecs;
	    hfsmp->hfs_last_sync_time = now;
	    
    } else if (hfsmp->hfs_active_threads == 0) {
	    uint64_t deadline;

	    clock_interval_to_deadline(delay, HFS_MILLISEC_SCALE, &deadline);
	    thread_call_enter_delayed(hfsmp->hfs_syncer, deadline);

	    // note: we intentionally return early here and do not
	    // decrement the sync_scheduled and sync_incomplete
	    // variables because we rescheduled the timer.

	    return;
    }
	    
    //
    // NOTE: we decrement these *after* we're done the journal_flush() since
    // it can take a significant amount of time and so we don't want more
    // callbacks scheduled until we're done this one.
    //
    OSDecrementAtomic((volatile SInt32 *)&hfsmp->hfs_sync_scheduled);
    OSDecrementAtomic((volatile SInt32 *)&hfsmp->hfs_sync_incomplete);
    wakeup((caddr_t)&hfsmp->hfs_sync_incomplete);
}


extern int IOBSDIsMediaEjectable( const char *cdev_name );

/*
 * Common code for mount and mountroot
 */
static int
hfs_mountfs(struct vnode *devvp, struct mount *mp, struct hfs_mount_args *args,
            int journal_replay_only, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	int retval = E_NONE;
	struct hfsmount	*hfsmp = NULL;
	struct buf *bp;
	dev_t dev;
	HFSMasterDirectoryBlock *mdbp = NULL;
	int ronly;
#if QUOTA
	int i;
#endif
	int mntwrapper;
	kauth_cred_t cred;
	u_int64_t disksize;
	daddr64_t log_blkcnt;
	u_int32_t log_blksize;
	u_int32_t phys_blksize;
	u_int32_t minblksize;
	u_int32_t iswritable;
	daddr64_t mdb_offset;
	int isvirtual = 0;
	int isroot = 0;

	if (args == NULL) {
		/* only hfs_mountroot passes us NULL as the 'args' argument */
		isroot = 1;
	}

	ronly = vfs_isrdonly(mp);
	dev = vnode_specrdev(devvp);
	cred = p ? vfs_context_ucred(context) : NOCRED;
	mntwrapper = 0;

	bp = NULL;
	hfsmp = NULL;
	mdbp = NULL;
	minblksize = kHFSBlockSize;

	/* Advisory locking should be handled at the VFS layer */
	vfs_setlocklocal(mp);

	/* Get the logical block size (treated as physical block size everywhere) */
	if (VNOP_IOCTL(devvp, DKIOCGETBLOCKSIZE, (caddr_t)&log_blksize, 0, context)) {
		retval = ENXIO;
		goto error_exit;
	}
	if (log_blksize == 0 || log_blksize > 1024*1024*1024) {
		printf("hfs: logical block size 0x%x looks bad.  Not mounting.\n", log_blksize);
		retval = ENXIO;
		goto error_exit;
	}
	
	/* Get the physical block size. */
	retval = VNOP_IOCTL(devvp, DKIOCGETPHYSICALBLOCKSIZE, (caddr_t)&phys_blksize, 0, context);
	if (retval) {
		if ((retval != ENOTSUP) && (retval != ENOTTY)) {
			retval = ENXIO;
			goto error_exit;
		}
		/* If device does not support this ioctl, assume that physical 
		 * block size is same as logical block size 
		 */
		phys_blksize = log_blksize;
	}
	if (phys_blksize == 0 || phys_blksize > 1024*1024*1024) {
		printf("hfs: physical block size 0x%x looks bad.  Not mounting.\n", phys_blksize);
		retval = ENXIO;
		goto error_exit;
	}

	/* Switch to 512 byte sectors (temporarily) */
	if (log_blksize > 512) {
		u_int32_t size512 = 512;

		if (VNOP_IOCTL(devvp, DKIOCSETBLOCKSIZE, (caddr_t)&size512, FWRITE, context)) {
			retval = ENXIO;
			goto error_exit;
		}
	}
	/* Get the number of 512 byte physical blocks. */
	if (VNOP_IOCTL(devvp, DKIOCGETBLOCKCOUNT, (caddr_t)&log_blkcnt, 0, context)) {
		/* resetting block size may fail if getting block count did */
		(void)VNOP_IOCTL(devvp, DKIOCSETBLOCKSIZE, (caddr_t)&log_blksize, FWRITE, context);

		retval = ENXIO;
		goto error_exit;
	}
	/* Compute an accurate disk size (i.e. within 512 bytes) */
	disksize = (u_int64_t)log_blkcnt * (u_int64_t)512;

	/*
	 * On Tiger it is not necessary to switch the device 
	 * block size to be 4k if there are more than 31-bits
	 * worth of blocks but to insure compatibility with
	 * pre-Tiger systems we have to do it.
	 *
	 * If the device size is not a multiple of 4K (8 * 512), then
	 * switching the logical block size isn't going to help because
	 * we will be unable to write the alternate volume header.
	 * In this case, just leave the logical block size unchanged.
	 */
	if (log_blkcnt > 0x000000007fffffff && (log_blkcnt & 7) == 0) {
		minblksize = log_blksize = 4096;
		if (phys_blksize < log_blksize)
			phys_blksize = log_blksize;
	}
	
	/*
	 * The cluster layer is not currently prepared to deal with a logical
	 * block size larger than the system's page size.  (It can handle
	 * blocks per page, but not multiple pages per block.)  So limit the
	 * logical block size to the page size.
	 */
	if (log_blksize > PAGE_SIZE)
		log_blksize = PAGE_SIZE;
	
	/* Now switch to our preferred physical block size. */
	if (log_blksize > 512) {
		if (VNOP_IOCTL(devvp, DKIOCSETBLOCKSIZE, (caddr_t)&log_blksize, FWRITE, context)) {
			retval = ENXIO;
			goto error_exit;
		}
		/* Get the count of physical blocks. */
		if (VNOP_IOCTL(devvp, DKIOCGETBLOCKCOUNT, (caddr_t)&log_blkcnt, 0, context)) {
			retval = ENXIO;
			goto error_exit;
		}
	}
	/*
	 * At this point:
	 *   minblksize is the minimum physical block size
	 *   log_blksize has our preferred physical block size
	 *   log_blkcnt has the total number of physical blocks
	 */

	mdb_offset = (daddr64_t)HFS_PRI_SECTOR(log_blksize);
	if ((retval = (int)buf_meta_bread(devvp, 
				HFS_PHYSBLK_ROUNDDOWN(mdb_offset, (phys_blksize/log_blksize)), 
				phys_blksize, cred, &bp))) {
		goto error_exit;
	}
	MALLOC(mdbp, HFSMasterDirectoryBlock *, kMDBSize, M_TEMP, M_WAITOK);
	if (mdbp == NULL) {
		retval = ENOMEM;
		goto error_exit;
	}
	bcopy((char *)buf_dataptr(bp) + HFS_PRI_OFFSET(phys_blksize), mdbp, kMDBSize);
	buf_brelse(bp);
	bp = NULL;

	MALLOC(hfsmp, struct hfsmount *, sizeof(struct hfsmount), M_HFSMNT, M_WAITOK);
	if (hfsmp == NULL) {
		retval = ENOMEM;
		goto error_exit;
	}
	bzero(hfsmp, sizeof(struct hfsmount));
	
	hfs_chashinit_finish(hfsmp);

	/*
	 *  Init the volume information structure
	 */
	
	lck_mtx_init(&hfsmp->hfs_mutex, hfs_mutex_group, hfs_lock_attr);
	lck_mtx_init(&hfsmp->hfc_mutex, hfs_mutex_group, hfs_lock_attr);
	lck_rw_init(&hfsmp->hfs_global_lock, hfs_rwlock_group, hfs_lock_attr);
	lck_rw_init(&hfsmp->hfs_insync, hfs_rwlock_group, hfs_lock_attr);

	vfs_setfsprivate(mp, hfsmp);
	hfsmp->hfs_mp = mp;			/* Make VFSTOHFS work */
	hfsmp->hfs_raw_dev = vnode_specrdev(devvp);
	hfsmp->hfs_devvp = devvp;
	vnode_ref(devvp);  /* Hold a ref on the device, dropped when hfsmp is freed. */
	hfsmp->hfs_logical_block_size = log_blksize;
	hfsmp->hfs_logical_block_count = log_blkcnt;
	hfsmp->hfs_physical_block_size = phys_blksize;
	hfsmp->hfs_log_per_phys = (phys_blksize / log_blksize);
	hfsmp->hfs_flags |= HFS_WRITEABLE_MEDIA;
	if (ronly)
		hfsmp->hfs_flags |= HFS_READ_ONLY;
	if (((unsigned int)vfs_flags(mp)) & MNT_UNKNOWNPERMISSIONS)
		hfsmp->hfs_flags |= HFS_UNKNOWN_PERMS;

#if QUOTA
	for (i = 0; i < MAXQUOTAS; i++)
		dqfileinit(&hfsmp->hfs_qfiles[i]);
#endif

	if (args) {
		hfsmp->hfs_uid = (args->hfs_uid == (uid_t)VNOVAL) ? UNKNOWNUID : args->hfs_uid;
		if (hfsmp->hfs_uid == 0xfffffffd) hfsmp->hfs_uid = UNKNOWNUID;
		hfsmp->hfs_gid = (args->hfs_gid == (gid_t)VNOVAL) ? UNKNOWNGID : args->hfs_gid;
		if (hfsmp->hfs_gid == 0xfffffffd) hfsmp->hfs_gid = UNKNOWNGID;
		vfs_setowner(mp, hfsmp->hfs_uid, hfsmp->hfs_gid);				/* tell the VFS */
		if (args->hfs_mask != (mode_t)VNOVAL) {
			hfsmp->hfs_dir_mask = args->hfs_mask & ALLPERMS;
			if (args->flags & HFSFSMNT_NOXONFILES) {
				hfsmp->hfs_file_mask = (args->hfs_mask & DEFFILEMODE);
			} else {
				hfsmp->hfs_file_mask = args->hfs_mask & ALLPERMS;
			}
		} else {
			hfsmp->hfs_dir_mask = UNKNOWNPERMISSIONS & ALLPERMS;		/* 0777: rwx---rwx */
			hfsmp->hfs_file_mask = UNKNOWNPERMISSIONS & DEFFILEMODE;	/* 0666: no --x by default? */
		}
		if ((args->flags != (int)VNOVAL) && (args->flags & HFSFSMNT_WRAPPER))
			mntwrapper = 1;
	} else {
		/* Even w/o explicit mount arguments, MNT_UNKNOWNPERMISSIONS requires setting up uid, gid, and mask: */
		if (((unsigned int)vfs_flags(mp)) & MNT_UNKNOWNPERMISSIONS) {
			hfsmp->hfs_uid = UNKNOWNUID;
			hfsmp->hfs_gid = UNKNOWNGID;
			vfs_setowner(mp, hfsmp->hfs_uid, hfsmp->hfs_gid);			/* tell the VFS */
			hfsmp->hfs_dir_mask = UNKNOWNPERMISSIONS & ALLPERMS;		/* 0777: rwx---rwx */
			hfsmp->hfs_file_mask = UNKNOWNPERMISSIONS & DEFFILEMODE;	/* 0666: no --x by default? */
		}
	}

	/* Find out if disk media is writable. */
	if (VNOP_IOCTL(devvp, DKIOCISWRITABLE, (caddr_t)&iswritable, 0, context) == 0) {
		if (iswritable)
			hfsmp->hfs_flags |= HFS_WRITEABLE_MEDIA;
		else
			hfsmp->hfs_flags &= ~HFS_WRITEABLE_MEDIA;
	}

	// record the current time at which we're mounting this volume
	struct timeval tv;
	microtime(&tv);
	hfsmp->hfs_mount_time = tv.tv_sec;

	/* Mount a standard HFS disk */
	if ((SWAP_BE16(mdbp->drSigWord) == kHFSSigWord) &&
	    (mntwrapper || (SWAP_BE16(mdbp->drEmbedSigWord) != kHFSPlusSigWord))) {

		/* On 10.6 and beyond, non read-only mounts for HFS standard vols get rejected */
		if (vfs_isrdwr(mp)) {
			retval = EROFS;
			goto error_exit;
		}
		/* Treat it as if it's read-only and not writeable */
		hfsmp->hfs_flags |= HFS_READ_ONLY;
		hfsmp->hfs_flags &= ~HFS_WRITEABLE_MEDIA;

	   	/* If only journal replay is requested, exit immediately */
		if (journal_replay_only) {
			retval = 0;
			goto error_exit;
		}

	        if ((vfs_flags(mp) & MNT_ROOTFS)) {	
			retval = EINVAL;  /* Cannot root from HFS standard disks */
			goto error_exit;
		}
		/* HFS disks can only use 512 byte physical blocks */
		if (log_blksize > kHFSBlockSize) {
			log_blksize = kHFSBlockSize;
			if (VNOP_IOCTL(devvp, DKIOCSETBLOCKSIZE, (caddr_t)&log_blksize, FWRITE, context)) {
				retval = ENXIO;
				goto error_exit;
			}
			if (VNOP_IOCTL(devvp, DKIOCGETBLOCKCOUNT, (caddr_t)&log_blkcnt, 0, context)) {
				retval = ENXIO;
				goto error_exit;
			}
			hfsmp->hfs_logical_block_size = log_blksize;
			hfsmp->hfs_logical_block_count = log_blkcnt;
			hfsmp->hfs_physical_block_size = log_blksize;
			hfsmp->hfs_log_per_phys = 1;
		}
		if (args) {
			hfsmp->hfs_encoding = args->hfs_encoding;
			HFSTOVCB(hfsmp)->volumeNameEncodingHint = args->hfs_encoding;

			/* establish the timezone */
			gTimeZone = args->hfs_timezone;
		}

		retval = hfs_getconverter(hfsmp->hfs_encoding, &hfsmp->hfs_get_unicode,
					&hfsmp->hfs_get_hfsname);
		if (retval)
			goto error_exit;

		retval = hfs_MountHFSVolume(hfsmp, mdbp, p);
		if (retval)
			(void) hfs_relconverter(hfsmp->hfs_encoding);

	} else /* Mount an HFS Plus disk */ {
		HFSPlusVolumeHeader *vhp;
		off_t embeddedOffset;
		int   jnl_disable = 0;
	
		/* Get the embedded Volume Header */
		if (SWAP_BE16(mdbp->drEmbedSigWord) == kHFSPlusSigWord) {
			embeddedOffset = SWAP_BE16(mdbp->drAlBlSt) * kHFSBlockSize;
			embeddedOffset += (u_int64_t)SWAP_BE16(mdbp->drEmbedExtent.startBlock) *
			                  (u_int64_t)SWAP_BE32(mdbp->drAlBlkSiz);

			/*
			 * If the embedded volume doesn't start on a block
			 * boundary, then switch the device to a 512-byte
			 * block size so everything will line up on a block
			 * boundary.
			 */
			if ((embeddedOffset % log_blksize) != 0) {
				printf("hfs_mountfs: embedded volume offset not"
				    " a multiple of physical block size (%d);"
				    " switching to 512\n", log_blksize);
				log_blksize = 512;
				if (VNOP_IOCTL(devvp, DKIOCSETBLOCKSIZE,
				    (caddr_t)&log_blksize, FWRITE, context)) {
					retval = ENXIO;
					goto error_exit;
				}
				if (VNOP_IOCTL(devvp, DKIOCGETBLOCKCOUNT,
				    (caddr_t)&log_blkcnt, 0, context)) {
					retval = ENXIO;
					goto error_exit;
				}
				/* Note: relative block count adjustment */
				hfsmp->hfs_logical_block_count *=
				    hfsmp->hfs_logical_block_size / log_blksize;
				
				/* Update logical /physical block size */
				hfsmp->hfs_logical_block_size = log_blksize;
				hfsmp->hfs_physical_block_size = log_blksize;
				phys_blksize = log_blksize;
				hfsmp->hfs_log_per_phys = 1;
			}

			disksize = (u_int64_t)SWAP_BE16(mdbp->drEmbedExtent.blockCount) *
			           (u_int64_t)SWAP_BE32(mdbp->drAlBlkSiz);

			hfsmp->hfs_logical_block_count = disksize / log_blksize;
	
			mdb_offset = (daddr64_t)((embeddedOffset / log_blksize) + HFS_PRI_SECTOR(log_blksize));
			retval = (int)buf_meta_bread(devvp, HFS_PHYSBLK_ROUNDDOWN(mdb_offset, hfsmp->hfs_log_per_phys),
					phys_blksize, cred, &bp);
			if (retval)
				goto error_exit;
			bcopy((char *)buf_dataptr(bp) + HFS_PRI_OFFSET(phys_blksize), mdbp, 512);
			buf_brelse(bp);
			bp = NULL;
			vhp = (HFSPlusVolumeHeader*) mdbp;

		} else /* pure HFS+ */ {
			embeddedOffset = 0;
			vhp = (HFSPlusVolumeHeader*) mdbp;
		}

		/*
		 * On inconsistent disks, do not allow read-write mount
		 * unless it is the boot volume being mounted.  We also
		 * always want to replay the journal if the journal_replay_only
		 * flag is set because that will (most likely) get the
		 * disk into a consistent state before fsck_hfs starts
		 * looking at it.
		 */
		if (  !(vfs_flags(mp) & MNT_ROOTFS)
		   && (SWAP_BE32(vhp->attributes) & kHFSVolumeInconsistentMask)
		   && !journal_replay_only
		   && !(hfsmp->hfs_flags & HFS_READ_ONLY)) {
			retval = EINVAL;
			goto error_exit;
		}


		// XXXdbg
		//
		hfsmp->jnl = NULL;
		hfsmp->jvp = NULL;
		if (args != NULL && (args->flags & HFSFSMNT_EXTENDED_ARGS) && 
		    args->journal_disable) {
		    jnl_disable = 1;
		}
				
		//
		// We only initialize the journal here if the last person
		// to mount this volume was journaling aware.  Otherwise
		// we delay journal initialization until later at the end
		// of hfs_MountHFSPlusVolume() because the last person who
		// mounted it could have messed things up behind our back
		// (so we need to go find the .journal file, make sure it's
		// the right size, re-sync up if it was moved, etc).
		//
		if (   (SWAP_BE32(vhp->lastMountedVersion) == kHFSJMountVersion)
			&& (SWAP_BE32(vhp->attributes) & kHFSVolumeJournaledMask)
			&& !jnl_disable) {
			
			// if we're able to init the journal, mark the mount
			// point as journaled.
			//
			if ((retval = hfs_early_journal_init(hfsmp, vhp, args, embeddedOffset, mdb_offset, mdbp, cred)) == 0) {
				vfs_setflags(mp, (u_int64_t)((unsigned int)MNT_JOURNALED));
			} else {
				if (retval == EROFS) {
					// EROFS is a special error code that means the volume has an external
					// journal which we couldn't find.  in that case we do not want to
					// rewrite the volume header - we'll just refuse to mount the volume.
					retval = EINVAL;
					goto error_exit;
				}

				// if the journal failed to open, then set the lastMountedVersion
				// to be "FSK!" which fsck_hfs will see and force the fsck instead
				// of just bailing out because the volume is journaled.
				if (!ronly) {
				    HFSPlusVolumeHeader *jvhp;

				    hfsmp->hfs_flags |= HFS_NEED_JNL_RESET;
				    
				    if (mdb_offset == 0) {
					mdb_offset = (daddr64_t)((embeddedOffset / log_blksize) + HFS_PRI_SECTOR(log_blksize));
				    }

				    bp = NULL;
				    retval = (int)buf_meta_bread(devvp, 
						    HFS_PHYSBLK_ROUNDDOWN(mdb_offset, hfsmp->hfs_log_per_phys), 
						    phys_blksize, cred, &bp);
				    if (retval == 0) {
					jvhp = (HFSPlusVolumeHeader *)(buf_dataptr(bp) + HFS_PRI_OFFSET(phys_blksize));
					    
					if (SWAP_BE16(jvhp->signature) == kHFSPlusSigWord || SWAP_BE16(jvhp->signature) == kHFSXSigWord) {
						printf ("hfs(1): Journal replay fail.  Writing lastMountVersion as FSK!\n");
					    jvhp->lastMountedVersion = SWAP_BE32(kFSKMountVersion);
					    buf_bwrite(bp);
					} else {
					    buf_brelse(bp);
					}
					bp = NULL;
				    } else if (bp) {
					buf_brelse(bp);
					// clear this so the error exit path won't try to use it
					bp = NULL;
				    }
				}

				// if this isn't the root device just bail out.
				// If it is the root device we just continue on
				// in the hopes that fsck_hfs will be able to
				// fix any damage that exists on the volume.
				if ( !(vfs_flags(mp) & MNT_ROOTFS)) {
				    retval = EINVAL;
				    goto error_exit;
				}
			}
		}
		// XXXdbg
	
		/* Either the journal is replayed successfully, or there 
		 * was nothing to replay, or no journal exists.  In any case,
		 * return success.
		 */
		if (journal_replay_only) {
			retval = 0;
			goto error_exit;
		}

		(void) hfs_getconverter(0, &hfsmp->hfs_get_unicode, &hfsmp->hfs_get_hfsname);

		retval = hfs_MountHFSPlusVolume(hfsmp, vhp, embeddedOffset, disksize, p, args, cred);
		/*
		 * If the backend didn't like our physical blocksize
		 * then retry with physical blocksize of 512.
		 */
		if ((retval == ENXIO) && (log_blksize > 512) && (log_blksize != minblksize)) {
			printf("hfs_mountfs: could not use physical block size "
				"(%d) switching to 512\n", log_blksize);
			log_blksize = 512;
			if (VNOP_IOCTL(devvp, DKIOCSETBLOCKSIZE, (caddr_t)&log_blksize, FWRITE, context)) {
				retval = ENXIO;
				goto error_exit;
			}
			if (VNOP_IOCTL(devvp, DKIOCGETBLOCKCOUNT, (caddr_t)&log_blkcnt, 0, context)) {
				retval = ENXIO;
				goto error_exit;
			}
			devvp->v_specsize = log_blksize;
			/* Note: relative block count adjustment (in case this is an embedded volume). */
    			hfsmp->hfs_logical_block_count *= hfsmp->hfs_logical_block_size / log_blksize;
     			hfsmp->hfs_logical_block_size = log_blksize;
     			hfsmp->hfs_log_per_phys = hfsmp->hfs_physical_block_size / log_blksize;
 
			if (hfsmp->jnl && hfsmp->jvp == devvp) {
			    // close and re-open this with the new block size
			    journal_close(hfsmp->jnl);
			    hfsmp->jnl = NULL;
			    if (hfs_early_journal_init(hfsmp, vhp, args, embeddedOffset, mdb_offset, mdbp, cred) == 0) {
					vfs_setflags(mp, (u_int64_t)((unsigned int)MNT_JOURNALED));
				} else {
					// if the journal failed to open, then set the lastMountedVersion
					// to be "FSK!" which fsck_hfs will see and force the fsck instead
					// of just bailing out because the volume is journaled.
					if (!ronly) {
				    	HFSPlusVolumeHeader *jvhp;

				    	hfsmp->hfs_flags |= HFS_NEED_JNL_RESET;
				    
				    	if (mdb_offset == 0) {
							mdb_offset = (daddr64_t)((embeddedOffset / log_blksize) + HFS_PRI_SECTOR(log_blksize));
				    	}

				   	 	bp = NULL;
				    	retval = (int)buf_meta_bread(devvp, HFS_PHYSBLK_ROUNDDOWN(mdb_offset, hfsmp->hfs_log_per_phys), 
							phys_blksize, cred, &bp);
				    	if (retval == 0) {
							jvhp = (HFSPlusVolumeHeader *)(buf_dataptr(bp) + HFS_PRI_OFFSET(phys_blksize));
					    
							if (SWAP_BE16(jvhp->signature) == kHFSPlusSigWord || SWAP_BE16(jvhp->signature) == kHFSXSigWord) {
								printf ("hfs(2): Journal replay fail.  Writing lastMountVersion as FSK!\n");
					    		jvhp->lastMountedVersion = SWAP_BE32(kFSKMountVersion);
					    		buf_bwrite(bp);
							} else {
					    		buf_brelse(bp);
							}
							bp = NULL;
				    	} else if (bp) {
							buf_brelse(bp);
							// clear this so the error exit path won't try to use it
							bp = NULL;
				    	}
					}

					// if this isn't the root device just bail out.
					// If it is the root device we just continue on
					// in the hopes that fsck_hfs will be able to
					// fix any damage that exists on the volume.
					if ( !(vfs_flags(mp) & MNT_ROOTFS)) {
				    	retval = EINVAL;
				    	goto error_exit;
					}
				}
			}

			/* Try again with a smaller block size... */
			retval = hfs_MountHFSPlusVolume(hfsmp, vhp, embeddedOffset, disksize, p, args, cred);
		}
		if (retval)
			(void) hfs_relconverter(0);
	}

	// save off a snapshot of the mtime from the previous mount
	// (for matador).
	hfsmp->hfs_last_mounted_mtime = hfsmp->hfs_mtime;

	if ( retval ) {
		goto error_exit;
	}

	mp->mnt_vfsstat.f_fsid.val[0] = (long)dev;
	mp->mnt_vfsstat.f_fsid.val[1] = vfs_typenum(mp);
	vfs_setmaxsymlen(mp, 0);

	mp->mnt_vtable->vfc_vfsflags |= VFC_VFSNATIVEXATTR;
#if NAMEDSTREAMS
	mp->mnt_kern_flag |= MNTK_NAMED_STREAMS;
#endif
	if (!(hfsmp->hfs_flags & HFS_STANDARD)) {
		/* Tell VFS that we support directory hard links. */
		mp->mnt_vtable->vfc_vfsflags |= VFC_VFSDIRLINKS;
	} else {
		/* HFS standard doesn't support extended readdir! */
		mp->mnt_vtable->vfc_vfsflags &= ~VFC_VFSREADDIR_EXTENDED;
	}

	if (args) {
		/*
		 * Set the free space warning levels for a non-root volume:
		 *
		 * Set the "danger" limit to 1% of the volume size or 100MB, whichever
		 * is less.  Set the "warning" limit to 2% of the volume size or 150MB,
		 * whichever is less.  And last, set the "desired" freespace level to
		 * to 3% of the volume size or 200MB, whichever is less.
		 */
		hfsmp->hfs_freespace_notify_dangerlimit =
			MIN(HFS_VERYLOWDISKTRIGGERLEVEL / HFSTOVCB(hfsmp)->blockSize,
				(HFSTOVCB(hfsmp)->totalBlocks / 100) * HFS_VERYLOWDISKTRIGGERFRACTION);
		hfsmp->hfs_freespace_notify_warninglimit =
			MIN(HFS_LOWDISKTRIGGERLEVEL / HFSTOVCB(hfsmp)->blockSize,
				(HFSTOVCB(hfsmp)->totalBlocks / 100) * HFS_LOWDISKTRIGGERFRACTION);
		hfsmp->hfs_freespace_notify_desiredlevel =
			MIN(HFS_LOWDISKSHUTOFFLEVEL / HFSTOVCB(hfsmp)->blockSize,
				(HFSTOVCB(hfsmp)->totalBlocks / 100) * HFS_LOWDISKSHUTOFFFRACTION);
	} else {
		/*
		 * Set the free space warning levels for the root volume:
		 *
		 * Set the "danger" limit to 5% of the volume size or 125MB, whichever
		 * is less.  Set the "warning" limit to 10% of the volume size or 250MB,
		 * whichever is less.  And last, set the "desired" freespace level to
		 * to 11% of the volume size or 375MB, whichever is less.
		 */
		hfsmp->hfs_freespace_notify_dangerlimit =
			MIN(HFS_ROOTVERYLOWDISKTRIGGERLEVEL / HFSTOVCB(hfsmp)->blockSize,
				(HFSTOVCB(hfsmp)->totalBlocks / 100) * HFS_ROOTVERYLOWDISKTRIGGERFRACTION);
		hfsmp->hfs_freespace_notify_warninglimit =
			MIN(HFS_ROOTLOWDISKTRIGGERLEVEL / HFSTOVCB(hfsmp)->blockSize,
				(HFSTOVCB(hfsmp)->totalBlocks / 100) * HFS_ROOTLOWDISKTRIGGERFRACTION);
		hfsmp->hfs_freespace_notify_desiredlevel =
			MIN(HFS_ROOTLOWDISKSHUTOFFLEVEL / HFSTOVCB(hfsmp)->blockSize,
				(HFSTOVCB(hfsmp)->totalBlocks / 100) * HFS_ROOTLOWDISKSHUTOFFFRACTION);
	};
	
	/* Check if the file system exists on virtual device, like disk image */
	if (VNOP_IOCTL(devvp, DKIOCISVIRTUAL, (caddr_t)&isvirtual, 0, context) == 0) {
		if (isvirtual) {
			hfsmp->hfs_flags |= HFS_VIRTUAL_DEVICE;
		}
	}

	/* do not allow ejectability checks on the root device */
	if (isroot == 0) {
		if ((hfsmp->hfs_flags & HFS_VIRTUAL_DEVICE) == 0 && 
				IOBSDIsMediaEjectable(mp->mnt_vfsstat.f_mntfromname)) {
			hfsmp->hfs_max_pending_io = 4096*1024;   // a reasonable value to start with.
			hfsmp->hfs_syncer = thread_call_allocate(hfs_syncer, hfsmp);
			if (hfsmp->hfs_syncer == NULL) {
				printf("hfs: failed to allocate syncer thread callback for %s (%s)\n",
						mp->mnt_vfsstat.f_mntfromname, mp->mnt_vfsstat.f_mntonname);
			}
		}
	}

	/*
	 * Start looking for free space to drop below this level and generate a
	 * warning immediately if needed:
	 */
	hfsmp->hfs_notification_conditions = 0;
	hfs_generate_volume_notifications(hfsmp);

	if (ronly == 0) {
		(void) hfs_flushvolumeheader(hfsmp, MNT_WAIT, 0);
	}
	FREE(mdbp, M_TEMP);
	return (0);

error_exit:
	if (bp)
		buf_brelse(bp);
	if (mdbp)
		FREE(mdbp, M_TEMP);

	if (hfsmp && hfsmp->jvp && hfsmp->jvp != hfsmp->hfs_devvp) {
		vnode_clearmountedon(hfsmp->jvp);
		(void)VNOP_CLOSE(hfsmp->jvp, ronly ? FREAD : FREAD|FWRITE, vfs_context_kernel());
		hfsmp->jvp = NULL;
	}
	if (hfsmp) {
		if (hfsmp->hfs_devvp) {
			vnode_rele(hfsmp->hfs_devvp);
		}
		hfs_delete_chash(hfsmp);

		FREE(hfsmp, M_HFSMNT);
		vfs_setfsprivate(mp, NULL);
	}
        return (retval);
}


/*
 * Make a filesystem operational.
 * Nothing to do at the moment.
 */
/* ARGSUSED */
static int
hfs_start(__unused struct mount *mp, __unused int flags, __unused vfs_context_t context)
{
	return (0);
}


/*
 * unmount system call
 */
static int
hfs_unmount(struct mount *mp, int mntflags, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	struct hfsmount *hfsmp = VFSTOHFS(mp);
	int retval = E_NONE;
	int flags;
	int force;
	int started_tr = 0;

	flags = 0;
	force = 0;
	if (mntflags & MNT_FORCE) {
		flags |= FORCECLOSE;
		force = 1;
	}

	if ((retval = hfs_flushfiles(mp, flags, p)) && !force)
 		return (retval);

	if (hfsmp->hfs_flags & HFS_METADATA_ZONE)
		(void) hfs_recording_suspend(hfsmp);

	/*
	 * Cancel any pending timers for this volume.  Then wait for any timers
	 * which have fired, but whose callbacks have not yet completed.
	 */
	if (hfsmp->hfs_syncer)
	{
		struct timespec ts = {0, 100000000};	/* 0.1 seconds */
		
		/*
		 * Cancel any timers that have been scheduled, but have not
		 * fired yet.  NOTE: The kernel considers a timer complete as
		 * soon as it starts your callback, so the kernel does not
		 * keep track of the number of callbacks in progress.
		 */
		if (thread_call_cancel(hfsmp->hfs_syncer))
			OSDecrementAtomic((volatile SInt32 *)&hfsmp->hfs_sync_incomplete);
		thread_call_free(hfsmp->hfs_syncer);
		hfsmp->hfs_syncer = NULL;
		
		/*
		 * This waits for all of the callbacks that were entered before
		 * we did thread_call_cancel above, but have not completed yet.
		 */
		while(hfsmp->hfs_sync_incomplete > 0)
		{
			msleep((caddr_t)&hfsmp->hfs_sync_incomplete, NULL, PWAIT, "hfs_unmount", &ts);
		}
		
		if (hfsmp->hfs_sync_incomplete < 0)
			panic("hfs_unmount: pm_sync_incomplete underflow!\n");
	}
	
	/*
	 * Flush out the b-trees, volume bitmap and Volume Header
	 */
	if ((hfsmp->hfs_flags & HFS_READ_ONLY) == 0) {
		retval = hfs_start_transaction(hfsmp);
		if (retval == 0) {
		    started_tr = 1;
		} else if (!force) {
		    goto err_exit;
		}

		if (hfsmp->hfs_startup_vp) {
			(void) hfs_lock(VTOC(hfsmp->hfs_startup_vp), HFS_EXCLUSIVE_LOCK);
			retval = hfs_fsync(hfsmp->hfs_startup_vp, MNT_WAIT, 0, p);
			hfs_unlock(VTOC(hfsmp->hfs_startup_vp));
			if (retval && !force)
				goto err_exit;
		}

		if (hfsmp->hfs_attribute_vp) {
			(void) hfs_lock(VTOC(hfsmp->hfs_attribute_vp), HFS_EXCLUSIVE_LOCK);
			retval = hfs_fsync(hfsmp->hfs_attribute_vp, MNT_WAIT, 0, p);
			hfs_unlock(VTOC(hfsmp->hfs_attribute_vp));
			if (retval && !force)
				goto err_exit;
		}

		(void) hfs_lock(VTOC(hfsmp->hfs_catalog_vp), HFS_EXCLUSIVE_LOCK);
		retval = hfs_fsync(hfsmp->hfs_catalog_vp, MNT_WAIT, 0, p);
		hfs_unlock(VTOC(hfsmp->hfs_catalog_vp));
		if (retval && !force)
			goto err_exit;
		
		(void) hfs_lock(VTOC(hfsmp->hfs_extents_vp), HFS_EXCLUSIVE_LOCK);
		retval = hfs_fsync(hfsmp->hfs_extents_vp, MNT_WAIT, 0, p);
		hfs_unlock(VTOC(hfsmp->hfs_extents_vp));
		if (retval && !force)
			goto err_exit;
			
		if (hfsmp->hfs_allocation_vp) {
			(void) hfs_lock(VTOC(hfsmp->hfs_allocation_vp), HFS_EXCLUSIVE_LOCK);
			retval = hfs_fsync(hfsmp->hfs_allocation_vp, MNT_WAIT, 0, p);
			hfs_unlock(VTOC(hfsmp->hfs_allocation_vp));
			if (retval && !force)
				goto err_exit;
		}

		if (hfsmp->hfc_filevp && vnode_issystem(hfsmp->hfc_filevp)) {
			retval = hfs_fsync(hfsmp->hfc_filevp, MNT_WAIT, 0, p);
			if (retval && !force)
				goto err_exit;
		}

		/* If runtime corruption was detected, indicate that the volume
		 * was not unmounted cleanly.
		 */
		if (hfsmp->vcbAtrb & kHFSVolumeInconsistentMask) {
			HFSTOVCB(hfsmp)->vcbAtrb &= ~kHFSVolumeUnmountedMask;
		} else {
			HFSTOVCB(hfsmp)->vcbAtrb |= kHFSVolumeUnmountedMask;
		}

		if (hfsmp->hfs_flags & HFS_HAS_SPARSE_DEVICE) {
			int i;
			u_int32_t min_start = hfsmp->totalBlocks;

			// set the nextAllocation pointer to the smallest free block number
			// we've seen so on the next mount we won't rescan unnecessarily
			for(i=0; i < (int)hfsmp->vcbFreeExtCnt; i++) {
				if (hfsmp->vcbFreeExt[i].startBlock < min_start) {
					min_start = hfsmp->vcbFreeExt[i].startBlock;
				}
			}
			if (min_start < hfsmp->nextAllocation) {
				hfsmp->nextAllocation = min_start;
			}
		}


		retval = hfs_flushvolumeheader(hfsmp, MNT_WAIT, 0);
		if (retval) {
			HFSTOVCB(hfsmp)->vcbAtrb &= ~kHFSVolumeUnmountedMask;
			if (!force)
				goto err_exit;	/* could not flush everything */
		}

		if (started_tr) {
		    hfs_end_transaction(hfsmp);
		    started_tr = 0;
		}
	}

	if (hfsmp->jnl) {
		hfs_journal_flush(hfsmp);
	}
	
	/*
	 *	Invalidate our caches and release metadata vnodes
	 */
	(void) hfsUnmount(hfsmp, p);

	/*
	 * Last chance to dump unreferenced system files.
	 */
	(void) vflush(mp, NULLVP, FORCECLOSE);

	if (HFSTOVCB(hfsmp)->vcbSigWord == kHFSSigWord)
		(void) hfs_relconverter(hfsmp->hfs_encoding);

	// XXXdbg
	if (hfsmp->jnl) {
	    journal_close(hfsmp->jnl);
	    hfsmp->jnl = NULL;
	}

	VNOP_FSYNC(hfsmp->hfs_devvp, MNT_WAIT, context);

	if (hfsmp->jvp && hfsmp->jvp != hfsmp->hfs_devvp) {
	    vnode_clearmountedon(hfsmp->jvp);
	    retval = VNOP_CLOSE(hfsmp->jvp,
	                       hfsmp->hfs_flags & HFS_READ_ONLY ? FREAD : FREAD|FWRITE,
			       vfs_context_kernel());
	    vnode_put(hfsmp->jvp);
	    hfsmp->jvp = NULL;
	}
	// XXXdbg

#ifdef HFS_SPARSE_DEV
	/* Drop our reference on the backing fs (if any). */
	if ((hfsmp->hfs_flags & HFS_HAS_SPARSE_DEVICE) && hfsmp->hfs_backingfs_rootvp) {
		struct vnode * tmpvp;

		hfsmp->hfs_flags &= ~HFS_HAS_SPARSE_DEVICE;
		tmpvp = hfsmp->hfs_backingfs_rootvp;
		hfsmp->hfs_backingfs_rootvp = NULLVP;
		vnode_rele(tmpvp);
	}
#endif /* HFS_SPARSE_DEV */
	lck_mtx_destroy(&hfsmp->hfc_mutex, hfs_mutex_group);
	vnode_rele(hfsmp->hfs_devvp);

	hfs_delete_chash(hfsmp);
	FREE(hfsmp, M_HFSMNT);

	return (0);

  err_exit:
	if (started_tr) {
		hfs_end_transaction(hfsmp);
	}
	return retval;
}


/*
 * Return the root of a filesystem.
 */
static int
hfs_vfs_root(struct mount *mp, struct vnode **vpp, __unused vfs_context_t context)
{
	return hfs_vget(VFSTOHFS(mp), (cnid_t)kHFSRootFolderID, vpp, 1);
}


/*
 * Do operations associated with quotas
 */
#if !QUOTA
static int
hfs_quotactl(__unused struct mount *mp, __unused int cmds, __unused uid_t uid, __unused caddr_t datap, __unused vfs_context_t context)
{
	return (ENOTSUP);
}
#else
static int
hfs_quotactl(struct mount *mp, int cmds, uid_t uid, caddr_t datap, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	int cmd, type, error;

	if (uid == ~0U)
		uid = vfs_context_ucred(context)->cr_ruid;
	cmd = cmds >> SUBCMDSHIFT;

	switch (cmd) {
	case Q_SYNC:
	case Q_QUOTASTAT:
		break;
	case Q_GETQUOTA:
		if (uid == vfs_context_ucred(context)->cr_ruid)
			break;
		/* fall through */
	default:
		if ( (error = vfs_context_suser(context)) )
			return (error);
	}

	type = cmds & SUBCMDMASK;
	if ((u_int)type >= MAXQUOTAS)
		return (EINVAL);
	if (vfs_busy(mp, LK_NOWAIT))
		return (0);

	switch (cmd) {

	case Q_QUOTAON:
		error = hfs_quotaon(p, mp, type, datap);
		break;

	case Q_QUOTAOFF:
		error = hfs_quotaoff(p, mp, type);
		break;

	case Q_SETQUOTA:
		error = hfs_setquota(mp, uid, type, datap);
		break;

	case Q_SETUSE:
		error = hfs_setuse(mp, uid, type, datap);
		break;

	case Q_GETQUOTA:
		error = hfs_getquota(mp, uid, type, datap);
		break;

	case Q_SYNC:
		error = hfs_qsync(mp);
		break;

	case Q_QUOTASTAT:
		error = hfs_quotastat(mp, type, datap);
		break;

	default:
		error = EINVAL;
		break;
	}
	vfs_unbusy(mp);

	return (error);
}
#endif /* QUOTA */

/* Subtype is composite of bits */
#define HFS_SUBTYPE_JOURNALED      0x01
#define HFS_SUBTYPE_CASESENSITIVE  0x02
/* bits 2 - 6 reserved */
#define HFS_SUBTYPE_STANDARDHFS    0x80

/*
 * Get file system statistics.
 */
static int
hfs_statfs(struct mount *mp, register struct vfsstatfs *sbp, __unused vfs_context_t context)
{
	ExtendedVCB *vcb = VFSTOVCB(mp);
	struct hfsmount *hfsmp = VFSTOHFS(mp);
	u_int32_t freeCNIDs;
	u_int16_t subtype = 0;

	freeCNIDs = (u_int32_t)0xFFFFFFFF - (u_int32_t)vcb->vcbNxtCNID;

	sbp->f_bsize = (u_int32_t)vcb->blockSize;
	sbp->f_iosize = (size_t)cluster_max_io_size(mp, 0);
	sbp->f_blocks = (u_int64_t)((u_int32_t)vcb->totalBlocks);
	sbp->f_bfree = (u_int64_t)((u_int32_t )hfs_freeblks(hfsmp, 0));
	sbp->f_bavail = (u_int64_t)((u_int32_t )hfs_freeblks(hfsmp, 1));
	sbp->f_files = (u_int64_t)((u_int32_t )(vcb->totalBlocks - 2));  /* max files is constrained by total blocks */
	sbp->f_ffree = (u_int64_t)((u_int32_t )(MIN(freeCNIDs, sbp->f_bavail)));

	/*
	 * Subtypes (flavors) for HFS
	 *   0:   Mac OS Extended
	 *   1:   Mac OS Extended (Journaled) 
	 *   2:   Mac OS Extended (Case Sensitive) 
	 *   3:   Mac OS Extended (Case Sensitive, Journaled) 
	 *   4 - 127:   Reserved
	 * 128:   Mac OS Standard
	 * 
	 */
	if (hfsmp->hfs_flags & HFS_STANDARD) {
		subtype = HFS_SUBTYPE_STANDARDHFS;
	} else /* HFS Plus */ {
		if (hfsmp->jnl)
			subtype |= HFS_SUBTYPE_JOURNALED;
		if (hfsmp->hfs_flags & HFS_CASE_SENSITIVE)
			subtype |= HFS_SUBTYPE_CASESENSITIVE;
	}
	sbp->f_fssubtype = subtype;

	return (0);
}


//
// XXXdbg -- this is a callback to be used by the journal to
//           get meta data blocks flushed out to disk.
//
// XXXdbg -- be smarter and don't flush *every* block on each
//           call.  try to only flush some so we don't wind up
//           being too synchronous.
//
__private_extern__
void
hfs_sync_metadata(void *arg)
{
	struct mount *mp = (struct mount *)arg;
	struct hfsmount *hfsmp;
	ExtendedVCB *vcb;
	buf_t	bp;
	int  retval;
	daddr64_t priIDSector;
	hfsmp = VFSTOHFS(mp);
	vcb = HFSTOVCB(hfsmp);

	// now make sure the super block is flushed
	priIDSector = (daddr64_t)((vcb->hfsPlusIOPosOffset / hfsmp->hfs_logical_block_size) +
				  HFS_PRI_SECTOR(hfsmp->hfs_logical_block_size));

	retval = (int)buf_meta_bread(hfsmp->hfs_devvp, 
			HFS_PHYSBLK_ROUNDDOWN(priIDSector, hfsmp->hfs_log_per_phys),
			hfsmp->hfs_physical_block_size, NOCRED, &bp);
	if ((retval != 0 ) && (retval != ENXIO)) {
		printf("hfs_sync_metadata: can't read volume header at %d! (retval 0x%x)\n",
		       (int)priIDSector, retval);
	}

	if (retval == 0 && ((buf_flags(bp) & (B_DELWRI | B_LOCKED)) == B_DELWRI)) {
	    buf_bwrite(bp);
	} else if (bp) {
	    buf_brelse(bp);
	}

	// the alternate super block...
	// XXXdbg - we probably don't need to do this each and every time.
	//          hfs_btreeio.c:FlushAlternate() should flag when it was
	//          written...
	if (hfsmp->hfs_alt_id_sector) {
		retval = (int)buf_meta_bread(hfsmp->hfs_devvp, 
				HFS_PHYSBLK_ROUNDDOWN(hfsmp->hfs_alt_id_sector, hfsmp->hfs_log_per_phys),
				hfsmp->hfs_physical_block_size, NOCRED, &bp);
		if (retval == 0 && ((buf_flags(bp) & (B_DELWRI | B_LOCKED)) == B_DELWRI)) {
		    buf_bwrite(bp);
		} else if (bp) {
		    buf_brelse(bp);
		}
	}
}


struct hfs_sync_cargs {
        kauth_cred_t cred;
        struct proc  *p;
        int    waitfor;
        int    error;
};


static int
hfs_sync_callback(struct vnode *vp, void *cargs)
{
	struct cnode *cp;
	struct hfs_sync_cargs *args;
	int error;

	args = (struct hfs_sync_cargs *)cargs;

	if (hfs_lock(VTOC(vp), HFS_EXCLUSIVE_LOCK) != 0) {
		return (VNODE_RETURNED);
	}
	cp = VTOC(vp);

	if ((cp->c_flag & C_MODIFIED) ||
	    (cp->c_touch_acctime | cp->c_touch_chgtime | cp->c_touch_modtime) ||
	    vnode_hasdirtyblks(vp)) {
	        error = hfs_fsync(vp, args->waitfor, 0, args->p);

		if (error)
		        args->error = error;
	}
	hfs_unlock(cp);
	return (VNODE_RETURNED);
}



/*
 * Go through the disk queues to initiate sandbagged IO;
 * go through the inodes to write those that have been modified;
 * initiate the writing of the super block if it has been modified.
 *
 * Note: we are always called with the filesystem marked `MPBUSY'.
 */
static int
hfs_sync(struct mount *mp, int waitfor, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	struct cnode *cp;
	struct hfsmount *hfsmp;
	ExtendedVCB *vcb;
	struct vnode *meta_vp[4];
	int i;
	int error, allerror = 0;
	struct hfs_sync_cargs args;

	hfsmp = VFSTOHFS(mp);

	/*
	 * hfs_changefs might be manipulating vnodes so back off
	 */
	if (hfsmp->hfs_flags & HFS_IN_CHANGEFS)
		return (0);

	if (hfsmp->hfs_flags & HFS_READ_ONLY)
		return (EROFS);

	/* skip over frozen volumes */
	if (!lck_rw_try_lock_shared(&hfsmp->hfs_insync))
		return 0;

	args.cred = kauth_cred_get();
	args.waitfor = waitfor;
	args.p = p;
	args.error = 0;
	/*
	 * hfs_sync_callback will be called for each vnode
	 * hung off of this mount point... the vnode will be
	 * properly referenced and unreferenced around the callback
	 */
	vnode_iterate(mp, 0, hfs_sync_callback, (void *)&args);

	if (args.error)
	        allerror = args.error;

	vcb = HFSTOVCB(hfsmp);

	meta_vp[0] = vcb->extentsRefNum;
	meta_vp[1] = vcb->catalogRefNum;
	meta_vp[2] = vcb->allocationsRefNum;  /* This is NULL for standard HFS */
	meta_vp[3] = hfsmp->hfs_attribute_vp; /* Optional file */

	/* Now sync our three metadata files */
	for (i = 0; i < 4; ++i) {
		struct vnode *btvp;

		btvp = meta_vp[i];;
		if ((btvp==0) || (vnode_mount(btvp) != mp))
			continue;

		/* XXX use hfs_systemfile_lock instead ? */
		(void) hfs_lock(VTOC(btvp), HFS_EXCLUSIVE_LOCK);
		cp = VTOC(btvp);

		if (((cp->c_flag &  C_MODIFIED) == 0) &&
		    (cp->c_touch_acctime == 0) &&
		    (cp->c_touch_chgtime == 0) &&
		    (cp->c_touch_modtime == 0) &&
		    vnode_hasdirtyblks(btvp) == 0) {
			hfs_unlock(VTOC(btvp));
			continue;
		}
		error = vnode_get(btvp);
		if (error) {
			hfs_unlock(VTOC(btvp));
			continue;
		}
		if ((error = hfs_fsync(btvp, waitfor, 0, p)))
			allerror = error;

		hfs_unlock(cp);
		vnode_put(btvp);
	};

	/*
	 * Force stale file system control information to be flushed.
	 */
	if (vcb->vcbSigWord == kHFSSigWord) {
		if ((error = VNOP_FSYNC(hfsmp->hfs_devvp, waitfor, context))) {
			allerror = error;
		}
	}
#if QUOTA
	hfs_qsync(mp);
#endif /* QUOTA */

	hfs_hotfilesync(hfsmp, vfs_context_kernel());

	/*
	 * Write back modified superblock.
	 */
	if (IsVCBDirty(vcb)) {
		error = hfs_flushvolumeheader(hfsmp, waitfor, 0);
		if (error)
			allerror = error;
	}

	if (hfsmp->jnl) {
	    hfs_journal_flush(hfsmp);
	}

	{
		clock_sec_t secs;
		clock_usec_t usecs;
		uint64_t now;

		clock_get_calendar_microtime(&secs, &usecs);
		now = ((uint64_t)secs * 1000000ULL) + (uint64_t)usecs;
		hfsmp->hfs_last_sync_time = now;
	}

	lck_rw_unlock_shared(&hfsmp->hfs_insync);	
	return (allerror);
}


/*
 * File handle to vnode
 *
 * Have to be really careful about stale file handles:
 * - check that the cnode id is valid
 * - call hfs_vget() to get the locked cnode
 * - check for an unallocated cnode (i_mode == 0)
 * - check that the given client host has export rights and return
 *   those rights via. exflagsp and credanonp
 */
static int
hfs_fhtovp(struct mount *mp, int fhlen, unsigned char *fhp, struct vnode **vpp, __unused vfs_context_t context)
{
	struct hfsfid *hfsfhp;
	struct vnode *nvp;
	int result;

	*vpp = NULL;
	hfsfhp = (struct hfsfid *)fhp;

	if (fhlen < (int)sizeof(struct hfsfid))
		return (EINVAL);

	result = hfs_vget(VFSTOHFS(mp), ntohl(hfsfhp->hfsfid_cnid), &nvp, 0);
	if (result) {
		if (result == ENOENT)
			result = ESTALE;
		return result;
	}

	/* 
	 * We used to use the create time as the gen id of the file handle,
	 * but it is not static enough because it can change at any point 
	 * via system calls.  We still don't have another volume ID or other
	 * unique identifier to use for a generation ID across reboots that
	 * persists until the file is removed.  Using only the CNID exposes
	 * us to the potential wrap-around case, but as of 2/2008, it would take
	 * over 2 months to wrap around if the machine did nothing but allocate
	 * CNIDs.  Using some kind of wrap counter would only be effective if
	 * each file had the wrap counter associated with it.  For now, 
	 * we use only the CNID to identify the file as it's good enough.
	 */	 

	*vpp = nvp;

	hfs_unlock(VTOC(nvp));
	return (0);
}


/*
 * Vnode pointer to File handle
 */
/* ARGSUSED */
static int
hfs_vptofh(struct vnode *vp, int *fhlenp, unsigned char *fhp, __unused vfs_context_t context)
{
	struct cnode *cp;
	struct hfsfid *hfsfhp;

	if (ISHFS(VTOVCB(vp)))
		return (ENOTSUP);	/* hfs standard is not exportable */

	if (*fhlenp < (int)sizeof(struct hfsfid))
		return (EOVERFLOW);

	cp = VTOC(vp);
	hfsfhp = (struct hfsfid *)fhp;
	/* only the CNID is used to identify the file now */
	hfsfhp->hfsfid_cnid = htonl(cp->c_fileid);
	hfsfhp->hfsfid_gen = htonl(cp->c_fileid);
	*fhlenp = sizeof(struct hfsfid);
	
	return (0);
}


/*
 * Initial HFS filesystems, done only once.
 */
static int
hfs_init(__unused struct vfsconf *vfsp)
{
	static int done = 0;

	if (done)
		return (0);
	done = 1;
	hfs_chashinit();
	hfs_converterinit();

	BTReserveSetup();
	
	
	hfs_lock_attr    = lck_attr_alloc_init();
	hfs_group_attr   = lck_grp_attr_alloc_init();
	hfs_mutex_group  = lck_grp_alloc_init("hfs-mutex", hfs_group_attr);
	hfs_rwlock_group = lck_grp_alloc_init("hfs-rwlock", hfs_group_attr);
	
#if HFS_COMPRESSION
    decmpfs_init();
#endif

	return (0);
}

static int
hfs_getmountpoint(struct vnode *vp, struct hfsmount **hfsmpp)
{
	struct hfsmount * hfsmp;
	char fstypename[MFSNAMELEN];

	if (vp == NULL)
		return (EINVAL);
	
	if (!vnode_isvroot(vp))
		return (EINVAL);

	vnode_vfsname(vp, fstypename);
	if (strncmp(fstypename, "hfs", sizeof(fstypename)) != 0)
		return (EINVAL);

	hfsmp = VTOHFS(vp);

	if (HFSTOVCB(hfsmp)->vcbSigWord == kHFSSigWord)
		return (EINVAL);

	*hfsmpp = hfsmp;

	return (0);
}

// XXXdbg
#include <sys/filedesc.h>

/*
 * HFS filesystem related variables.
 */
static int
hfs_sysctl(int *name, __unused u_int namelen, user_addr_t oldp, size_t *oldlenp, 
			user_addr_t newp, size_t newlen, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	int error;
	struct hfsmount *hfsmp;

	/* all sysctl names at this level are terminal */

	if (name[0] == HFS_ENCODINGBIAS) {
		int bias;

		bias = hfs_getencodingbias();
		error = sysctl_int(oldp, oldlenp, newp, newlen, &bias);
		if (error == 0 && newp)
			hfs_setencodingbias(bias);
		return (error);

	} else if (name[0] == HFS_EXTEND_FS) {
        u_int64_t  newsize;
		vnode_t vp = vfs_context_cwd(context);

		if (newp == USER_ADDR_NULL || vp == NULLVP)
			return (EINVAL);
		if ((error = hfs_getmountpoint(vp, &hfsmp)))
			return (error);
		error = sysctl_quad(oldp, oldlenp, newp, newlen, (quad_t *)&newsize);
		if (error)
			return (error);
	
		error = hfs_extendfs(hfsmp, newsize, context);		
		return (error);

	} else if (name[0] == HFS_ENCODINGHINT) {
		size_t bufsize;
		size_t bytes;
		u_int32_t hint;
		u_int16_t *unicode_name = NULL;
		char *filename = NULL;

		if ((newlen <= 0) || (newlen > MAXPATHLEN)) 
			return (EINVAL);

		bufsize = MAX(newlen * 3, MAXPATHLEN);
		MALLOC(filename, char *, newlen, M_TEMP, M_WAITOK);
		if (filename == NULL) {
			error = ENOMEM;
			goto encodinghint_exit;
		}
		MALLOC(unicode_name, u_int16_t *, bufsize, M_TEMP, M_WAITOK);
		if (filename == NULL) {
			error = ENOMEM;
			goto encodinghint_exit;
		}

		error = copyin(newp, (caddr_t)filename, newlen);
		if (error == 0) {
			error = utf8_decodestr((u_int8_t *)filename, newlen - 1, unicode_name,
			                       &bytes, bufsize, 0, UTF_DECOMPOSED);
			if (error == 0) {
				hint = hfs_pickencoding(unicode_name, bytes / 2);
				error = sysctl_int(oldp, oldlenp, USER_ADDR_NULL, 0, (int32_t *)&hint);
			}
		}

encodinghint_exit:
		if (unicode_name)
			FREE(unicode_name, M_TEMP);
		if (filename)
			FREE(filename, M_TEMP);
		return (error);

	} else if (name[0] == HFS_ENABLE_JOURNALING) {
		// make the file system journaled...
		vnode_t vp = vfs_context_cwd(context);
		vnode_t jvp;
		ExtendedVCB *vcb;
		struct cat_attr jnl_attr, jinfo_attr;
		struct cat_fork jnl_fork, jinfo_fork;
		void *jnl = NULL;
		int lockflags;

		/* Only root can enable journaling */
		if (!is_suser()) {
			return (EPERM);
		}
		if (vp == NULLVP)
		        return EINVAL;

		hfsmp = VTOHFS(vp);
		if (hfsmp->hfs_flags & HFS_READ_ONLY) {
			return EROFS;
		}
		if (HFSTOVCB(hfsmp)->vcbSigWord == kHFSSigWord) {
			printf("hfs: can't make a plain hfs volume journaled.\n");
			return EINVAL;
		}

		if (hfsmp->jnl) {
		    printf("hfs: volume @ mp %p is already journaled!\n", vnode_mount(vp));
		    return EAGAIN;
		}

		vcb = HFSTOVCB(hfsmp);
		lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG | SFL_EXTENTS, HFS_EXCLUSIVE_LOCK);
		if (BTHasContiguousNodes(VTOF(vcb->catalogRefNum)) == 0 ||
			BTHasContiguousNodes(VTOF(vcb->extentsRefNum)) == 0) {

			printf("hfs: volume has a btree w/non-contiguous nodes.  can not enable journaling.\n");
			hfs_systemfile_unlock(hfsmp, lockflags);
			return EINVAL;
		}
		hfs_systemfile_unlock(hfsmp, lockflags);

		// make sure these both exist!
		if (   GetFileInfo(vcb, kHFSRootFolderID, ".journal_info_block", &jinfo_attr, &jinfo_fork) == 0
			|| GetFileInfo(vcb, kHFSRootFolderID, ".journal", &jnl_attr, &jnl_fork) == 0) {

			return EINVAL;
		}

		hfs_sync(hfsmp->hfs_mp, MNT_WAIT, context);

		printf("hfs: Initializing the journal (joffset 0x%llx sz 0x%llx)...\n",
			   (off_t)name[2], (off_t)name[3]);

		//
		// XXXdbg - note that currently (Sept, 08) hfs_util does not support
		//          enabling the journal on a separate device so it is safe
		//          to just copy hfs_devvp here.  If hfs_util gets the ability
		//          to dynamically enable the journal on a separate device then
		//          we will have to do the same thing as hfs_early_journal_init()
		//          to locate and open the journal device.
		//
		jvp = hfsmp->hfs_devvp;
		jnl = journal_create(jvp,
							 (off_t)name[2] * (off_t)HFSTOVCB(hfsmp)->blockSize
							 + HFSTOVCB(hfsmp)->hfsPlusIOPosOffset,
							 (off_t)((unsigned)name[3]),
							 hfsmp->hfs_devvp,
							 hfsmp->hfs_logical_block_size,
							 0,
							 0,
							 hfs_sync_metadata, hfsmp->hfs_mp);

		if (jnl == NULL) {
			printf("hfs: FAILED to create the journal!\n");
			if (jvp && jvp != hfsmp->hfs_devvp) {
				vnode_clearmountedon(jvp);
				VNOP_CLOSE(jvp, hfsmp->hfs_flags & HFS_READ_ONLY ? FREAD : FREAD|FWRITE, vfs_context_kernel());
			}
			jvp = NULL;

			return EINVAL;
		} 

		hfs_global_exclusive_lock_acquire(hfsmp);
		
		/*
		 * Flush all dirty metadata buffers.
		 */
		buf_flushdirtyblks(hfsmp->hfs_devvp, MNT_WAIT, 0, "hfs_sysctl");
		buf_flushdirtyblks(hfsmp->hfs_extents_vp, MNT_WAIT, 0, "hfs_sysctl");
		buf_flushdirtyblks(hfsmp->hfs_catalog_vp, MNT_WAIT, 0, "hfs_sysctl");
		buf_flushdirtyblks(hfsmp->hfs_allocation_vp, MNT_WAIT, 0, "hfs_sysctl");
		if (hfsmp->hfs_attribute_vp)
			buf_flushdirtyblks(hfsmp->hfs_attribute_vp, MNT_WAIT, 0, "hfs_sysctl");

		HFSTOVCB(hfsmp)->vcbJinfoBlock = name[1];
		HFSTOVCB(hfsmp)->vcbAtrb |= kHFSVolumeJournaledMask;
		hfsmp->jvp = jvp;
		hfsmp->jnl = jnl;

		// save this off for the hack-y check in hfs_remove()
		hfsmp->jnl_start        = (u_int32_t)name[2];
		hfsmp->jnl_size         = (off_t)((unsigned)name[3]);
		hfsmp->hfs_jnlinfoblkid = jinfo_attr.ca_fileid;
		hfsmp->hfs_jnlfileid    = jnl_attr.ca_fileid;

		vfs_setflags(hfsmp->hfs_mp, (u_int64_t)((unsigned int)MNT_JOURNALED));

		hfs_global_exclusive_lock_release(hfsmp);
		hfs_flushvolumeheader(hfsmp, MNT_WAIT, 1);

		{
			fsid_t fsid;
		
			fsid.val[0] = (int32_t)hfsmp->hfs_raw_dev;
			fsid.val[1] = (int32_t)vfs_typenum(HFSTOVFS(hfsmp));
			vfs_event_signal(&fsid, VQ_UPDATE, (intptr_t)NULL);
		}
		return 0;
	} else if (name[0] == HFS_DISABLE_JOURNALING) {
		// clear the journaling bit 
		vnode_t vp = vfs_context_cwd(context);
		
		/* Only root can disable journaling */
		if (!is_suser()) {
			return (EPERM);
		}
		if (vp == NULLVP)
		        return EINVAL;

		hfsmp = VTOHFS(vp);

		/* 
		 * Disabling journaling is disallowed on volumes with directory hard links
		 * because we have not tested the relevant code path.
		 */  
		if (hfsmp->hfs_private_attr[DIR_HARDLINKS].ca_entries != 0){
			printf("hfs: cannot disable journaling on volumes with directory hardlinks\n");
			return EPERM;
		}

		printf("hfs: disabling journaling for mount @ %p\n", vnode_mount(vp));

		hfs_global_exclusive_lock_acquire(hfsmp);

		// Lights out for you buddy!
		journal_close(hfsmp->jnl);
		hfsmp->jnl = NULL;

		if (hfsmp->jvp && hfsmp->jvp != hfsmp->hfs_devvp) {
			vnode_clearmountedon(hfsmp->jvp);
			VNOP_CLOSE(hfsmp->jvp, hfsmp->hfs_flags & HFS_READ_ONLY ? FREAD : FREAD|FWRITE, vfs_context_kernel());
			vnode_put(hfsmp->jvp);
		}
		hfsmp->jvp = NULL;
		vfs_clearflags(hfsmp->hfs_mp, (u_int64_t)((unsigned int)MNT_JOURNALED));
		hfsmp->jnl_start        = 0;
		hfsmp->hfs_jnlinfoblkid = 0;
		hfsmp->hfs_jnlfileid    = 0;
		
		HFSTOVCB(hfsmp)->vcbAtrb &= ~kHFSVolumeJournaledMask;
		
		hfs_global_exclusive_lock_release(hfsmp);
		hfs_flushvolumeheader(hfsmp, MNT_WAIT, 1);

		{
			fsid_t fsid;
		
			fsid.val[0] = (int32_t)hfsmp->hfs_raw_dev;
			fsid.val[1] = (int32_t)vfs_typenum(HFSTOVFS(hfsmp));
			vfs_event_signal(&fsid, VQ_UPDATE, (intptr_t)NULL);
		}
		return 0;
	} else if (name[0] == HFS_GET_JOURNAL_INFO) {
		vnode_t vp = vfs_context_cwd(context);
		off_t jnl_start, jnl_size;

		if (vp == NULLVP)
		        return EINVAL;

		/* 64-bit processes won't work with this sysctl -- can't fit a pointer into an int! */
		if (proc_is64bit(current_proc()))
			return EINVAL;

		hfsmp = VTOHFS(vp);
	    if (hfsmp->jnl == NULL) {
			jnl_start = 0;
			jnl_size  = 0;
	    } else {
			jnl_start = (off_t)(hfsmp->jnl_start * HFSTOVCB(hfsmp)->blockSize) + (off_t)HFSTOVCB(hfsmp)->hfsPlusIOPosOffset;
			jnl_size  = (off_t)hfsmp->jnl_size;
	    }

	    if ((error = copyout((caddr_t)&jnl_start, CAST_USER_ADDR_T(name[1]), sizeof(off_t))) != 0) {
			return error;
		}
	    if ((error = copyout((caddr_t)&jnl_size, CAST_USER_ADDR_T(name[2]), sizeof(off_t))) != 0) {
			return error;
		}

		return 0;
	} else if (name[0] == HFS_SET_PKG_EXTENSIONS) {

	    return set_package_extensions_table((user_addr_t)((unsigned)name[1]), name[2], name[3]);
	    
	} else if (name[0] == VFS_CTL_QUERY) {
    	struct sysctl_req *req;
    	union union_vfsidctl vc;
    	struct mount *mp;
 	    struct vfsquery vq;
	
		req = CAST_DOWN(struct sysctl_req *, oldp);	/* we're new style vfs sysctl. */
        
        error = SYSCTL_IN(req, &vc, proc_is64bit(p)? sizeof(vc.vc64):sizeof(vc.vc32));
		if (error) return (error);

		mp = vfs_getvfs(&vc.vc32.vc_fsid); /* works for 32 and 64 */
        if (mp == NULL) return (ENOENT);
        
		hfsmp = VFSTOHFS(mp);
		bzero(&vq, sizeof(vq));
		vq.vq_flags = hfsmp->hfs_notification_conditions;
		return SYSCTL_OUT(req, &vq, sizeof(vq));;
	} else if (name[0] == HFS_REPLAY_JOURNAL) {
		vnode_t devvp = NULL;
		int device_fd;
		if (namelen != 2) {
			return (EINVAL);
		}
		device_fd = name[1];
		error = file_vnode(device_fd, &devvp);
		if (error) {
			return error;
		}
		error = vnode_getwithref(devvp);
		if (error) {
			file_drop(device_fd);
			return error;
		}
		error = hfs_journal_replay(devvp, context);
		file_drop(device_fd);
		vnode_put(devvp);
		return error;
	}

	return (ENOTSUP);
}

/* 
 * hfs_vfs_vget is not static since it is used in hfs_readwrite.c to support
 * the build_path ioctl.  We use it to leverage the code below that updates
 * the origin list cache if necessary
 */

int
hfs_vfs_vget(struct mount *mp, ino64_t ino, struct vnode **vpp, __unused vfs_context_t context)
{
	int error;
	int lockflags;
	struct hfsmount *hfsmp;

	hfsmp = VFSTOHFS(mp);

	error = hfs_vget(hfsmp, (cnid_t)ino, vpp, 1);
	if (error)
		return (error);

	/*
	 * ADLs may need to have their origin state updated
	 * since build_path needs a valid parent.  The same is true
	 * for hardlinked files as well.  There isn't a race window here
	 * in re-acquiring the cnode lock since we aren't pulling any data 
	 * out of the cnode; instead, we're going to the catalog.
	 */
	if ((VTOC(*vpp)->c_flag & C_HARDLINK) &&
	    (hfs_lock(VTOC(*vpp), HFS_EXCLUSIVE_LOCK) == 0)) {
		cnode_t *cp = VTOC(*vpp);
		struct cat_desc cdesc;
		
		if (!hfs_haslinkorigin(cp)) {
			lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG, HFS_SHARED_LOCK);
			error = cat_findname(hfsmp, (cnid_t)ino, &cdesc);
			hfs_systemfile_unlock(hfsmp, lockflags);
			if (error == 0) {
				if ((cdesc.cd_parentcnid != hfsmp->hfs_private_desc[DIR_HARDLINKS].cd_cnid) &&
					(cdesc.cd_parentcnid != hfsmp->hfs_private_desc[FILE_HARDLINKS].cd_cnid)) {
					hfs_savelinkorigin(cp, cdesc.cd_parentcnid);
				}
				cat_releasedesc(&cdesc);
			}
		}
		hfs_unlock(cp);
	}
	return (0);
}


/*
 * Look up an HFS object by ID.
 *
 * The object is returned with an iocount reference and the cnode locked.
 *
 * If the object is a file then it will represent the data fork.
 */
__private_extern__
int
hfs_vget(struct hfsmount *hfsmp, cnid_t cnid, struct vnode **vpp, int skiplock)
{
	struct vnode *vp = NULLVP;
	struct cat_desc cndesc;
	struct cat_attr cnattr;
	struct cat_fork cnfork;
	u_int32_t linkref = 0;
	int error;
	
	/* Check for cnids that should't be exported. */
	if ((cnid < kHFSFirstUserCatalogNodeID) &&
	    (cnid != kHFSRootFolderID && cnid != kHFSRootParentID)) {
		return (ENOENT);
	}
	/* Don't export our private directories. */
	if (cnid == hfsmp->hfs_private_desc[FILE_HARDLINKS].cd_cnid ||
	    cnid == hfsmp->hfs_private_desc[DIR_HARDLINKS].cd_cnid) {
		return (ENOENT);
	}
	/*
	 * Check the hash first
	 */
	vp = hfs_chash_getvnode(hfsmp, cnid, 0, skiplock);
	if (vp) {
		*vpp = vp;
		return(0);
	}

	bzero(&cndesc, sizeof(cndesc));
	bzero(&cnattr, sizeof(cnattr));
	bzero(&cnfork, sizeof(cnfork));

	/*
	 * Not in hash, lookup in catalog
	 */
	if (cnid == kHFSRootParentID) {
		static char hfs_rootname[] = "/";

		cndesc.cd_nameptr = (const u_int8_t *)&hfs_rootname[0];
		cndesc.cd_namelen = 1;
		cndesc.cd_parentcnid = kHFSRootParentID;
		cndesc.cd_cnid = kHFSRootFolderID;
		cndesc.cd_flags = CD_ISDIR;

		cnattr.ca_fileid = kHFSRootFolderID;
		cnattr.ca_linkcount = 1;
		cnattr.ca_entries = 1;
		cnattr.ca_dircount = 1;
		cnattr.ca_mode = (S_IFDIR | S_IRWXU | S_IRWXG | S_IRWXO);
	} else {
		int lockflags;
		cnid_t pid;
		const char *nameptr;

		lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG, HFS_SHARED_LOCK);
		error = cat_idlookup(hfsmp, cnid, 0, &cndesc, &cnattr, &cnfork);
		hfs_systemfile_unlock(hfsmp, lockflags);

		if (error) {
			*vpp = NULL;
			return (error);
		}

		/*
		 * Check for a raw hardlink inode and save its linkref.
		 */
		pid = cndesc.cd_parentcnid;
		nameptr = (const char *)cndesc.cd_nameptr;

		if ((pid == hfsmp->hfs_private_desc[FILE_HARDLINKS].cd_cnid) &&
		    (bcmp(nameptr, HFS_INODE_PREFIX, HFS_INODE_PREFIX_LEN) == 0)) {
			linkref = strtoul(&nameptr[HFS_INODE_PREFIX_LEN], NULL, 10);

		} else if ((pid == hfsmp->hfs_private_desc[DIR_HARDLINKS].cd_cnid) &&
		           (bcmp(nameptr, HFS_DIRINODE_PREFIX, HFS_DIRINODE_PREFIX_LEN) == 0)) {
			linkref = strtoul(&nameptr[HFS_DIRINODE_PREFIX_LEN], NULL, 10);

		} else if ((pid == hfsmp->hfs_private_desc[FILE_HARDLINKS].cd_cnid) &&
		           (bcmp(nameptr, HFS_DELETE_PREFIX, HFS_DELETE_PREFIX_LEN) == 0)) {
			*vpp = NULL;
			cat_releasedesc(&cndesc);
			return (ENOENT);  /* open unlinked file */
		}
	}

	/*
	 * Finish initializing cnode descriptor for hardlinks.
	 *
	 * We need a valid name and parent for reverse lookups.
	 */
	if (linkref) {
		cnid_t nextlinkid;
		cnid_t prevlinkid;
		struct cat_desc linkdesc;
		int lockflags;

		cnattr.ca_linkref = linkref;

		/*
		 * Pick up the first link in the chain and get a descriptor for it.
		 * This allows blind volfs paths to work for hardlinks.
		 */
		if ((hfs_lookuplink(hfsmp, linkref, &prevlinkid,  &nextlinkid) == 0) &&
		    (nextlinkid != 0)) {
			lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG, HFS_SHARED_LOCK);
			error = cat_findname(hfsmp, nextlinkid, &linkdesc);
			hfs_systemfile_unlock(hfsmp, lockflags);
			if (error == 0) {
				cat_releasedesc(&cndesc);
				bcopy(&linkdesc, &cndesc, sizeof(linkdesc));
			}
		}	
	}

	if (linkref) {
		error = hfs_getnewvnode(hfsmp, NULL, NULL, &cndesc, 0, &cnattr, &cnfork, &vp);
		if (error == 0) {
			VTOC(vp)->c_flag |= C_HARDLINK;
			vnode_setmultipath(vp);
		}
	} else {
		struct componentname cn;

		/* Supply hfs_getnewvnode with a component name. */
		MALLOC_ZONE(cn.cn_pnbuf, caddr_t, MAXPATHLEN, M_NAMEI, M_WAITOK);
		cn.cn_nameiop = LOOKUP;
		cn.cn_flags = ISLASTCN | HASBUF;
		cn.cn_context = NULL;
		cn.cn_pnlen = MAXPATHLEN;
		cn.cn_nameptr = cn.cn_pnbuf;
		cn.cn_namelen = cndesc.cd_namelen;
		cn.cn_hash = 0;
		cn.cn_consume = 0;
		bcopy(cndesc.cd_nameptr, cn.cn_nameptr, cndesc.cd_namelen + 1);
	
		error = hfs_getnewvnode(hfsmp, NULLVP, &cn, &cndesc, 0, &cnattr, &cnfork, &vp);

		if (error == 0 && (VTOC(vp)->c_flag & C_HARDLINK)) {
			hfs_savelinkorigin(VTOC(vp), cndesc.cd_parentcnid);
		}
		FREE_ZONE(cn.cn_pnbuf, cn.cn_pnlen, M_NAMEI);
	}
	cat_releasedesc(&cndesc);

	*vpp = vp;
	if (vp && skiplock) {
		hfs_unlock(VTOC(vp));
	}
	return (error);
}


/*
 * Flush out all the files in a filesystem.
 */
static int
#if QUOTA
hfs_flushfiles(struct mount *mp, int flags, struct proc *p)
#else
hfs_flushfiles(struct mount *mp, int flags, __unused struct proc *p)
#endif /* QUOTA */
{
	struct hfsmount *hfsmp;
	struct vnode *skipvp = NULLVP;
	int error;
#if QUOTA
	int quotafilecnt;
	int i;
#endif

	hfsmp = VFSTOHFS(mp);

#if QUOTA
	/*
	 * The open quota files have an indirect reference on
	 * the root directory vnode.  We must account for this
	 * extra reference when doing the intial vflush.
	 */
	quotafilecnt = 0;
	if (((unsigned int)vfs_flags(mp)) & MNT_QUOTA) {

		/* Find out how many quota files we have open. */
		for (i = 0; i < MAXQUOTAS; i++) {
			if (hfsmp->hfs_qfiles[i].qf_vp != NULLVP)
				++quotafilecnt;
		}

		/* Obtain the root vnode so we can skip over it. */
		skipvp = hfs_chash_getvnode(hfsmp, kHFSRootFolderID, 0, 0);
	}
#endif /* QUOTA */

	error = vflush(mp, skipvp, SKIPSYSTEM | SKIPSWAP | flags);
	if (error != 0)
		return(error);

	error = vflush(mp, skipvp, SKIPSYSTEM | flags);

#if QUOTA
	if (((unsigned int)vfs_flags(mp)) & MNT_QUOTA) {
		if (skipvp) {
			/*
			 * See if there are additional references on the
			 * root vp besides the ones obtained from the open
			 * quota files and the hfs_chash_getvnode call above.
			 */
			if ((error == 0) &&
			    (vnode_isinuse(skipvp,  quotafilecnt))) {
				error = EBUSY;  /* root directory is still open */
			}
			hfs_unlock(VTOC(skipvp));
			vnode_put(skipvp);
		}
		if (error && (flags & FORCECLOSE) == 0)
			return (error);

		for (i = 0; i < MAXQUOTAS; i++) {
			if (hfsmp->hfs_qfiles[i].qf_vp == NULLVP)
				continue;
			hfs_quotaoff(p, mp, i);
		}
		error = vflush(mp, NULLVP, SKIPSYSTEM | flags);
	}
#endif /* QUOTA */

	return (error);
}

/*
 * Update volume encoding bitmap (HFS Plus only)
 */
__private_extern__
void
hfs_setencodingbits(struct hfsmount *hfsmp, u_int32_t encoding)
{
#define  kIndexMacUkrainian	48  /* MacUkrainian encoding is 152 */
#define  kIndexMacFarsi		49  /* MacFarsi encoding is 140 */

	u_int32_t	index;

	switch (encoding) {
	case kTextEncodingMacUkrainian:
		index = kIndexMacUkrainian;
		break;
	case kTextEncodingMacFarsi:
		index = kIndexMacFarsi;
		break;
	default:
		index = encoding;
		break;
	}

	if (index < 64 && (hfsmp->encodingsBitmap & (u_int64_t)(1ULL << index)) == 0) {
		HFS_MOUNT_LOCK(hfsmp, TRUE)
		hfsmp->encodingsBitmap |= (u_int64_t)(1ULL << index);
		MarkVCBDirty(hfsmp);
		HFS_MOUNT_UNLOCK(hfsmp, TRUE);
	}
}

/*
 * Update volume stats
 *
 * On journal volumes this will cause a volume header flush
 */
__private_extern__
int
hfs_volupdate(struct hfsmount *hfsmp, enum volop op, int inroot)
{
	struct timeval tv;

	microtime(&tv);

	lck_mtx_lock(&hfsmp->hfs_mutex);

	MarkVCBDirty(hfsmp);
	hfsmp->hfs_mtime = tv.tv_sec;

	switch (op) {
	case VOL_UPDATE:
		break;
	case VOL_MKDIR:
		if (hfsmp->hfs_dircount != 0xFFFFFFFF)
			++hfsmp->hfs_dircount;
		if (inroot && hfsmp->vcbNmRtDirs != 0xFFFF)
			++hfsmp->vcbNmRtDirs;
		break;
	case VOL_RMDIR:
		if (hfsmp->hfs_dircount != 0)
			--hfsmp->hfs_dircount;
		if (inroot && hfsmp->vcbNmRtDirs != 0xFFFF)
			--hfsmp->vcbNmRtDirs;
		break;
	case VOL_MKFILE:
		if (hfsmp->hfs_filecount != 0xFFFFFFFF)
			++hfsmp->hfs_filecount;
		if (inroot && hfsmp->vcbNmFls != 0xFFFF)
			++hfsmp->vcbNmFls;
		break;
	case VOL_RMFILE:
		if (hfsmp->hfs_filecount != 0)
			--hfsmp->hfs_filecount;
		if (inroot && hfsmp->vcbNmFls != 0xFFFF)
			--hfsmp->vcbNmFls;
		break;
	}

	lck_mtx_unlock(&hfsmp->hfs_mutex);

	if (hfsmp->jnl) {
		hfs_flushvolumeheader(hfsmp, 0, 0);
	}

	return (0);
}


static int
hfs_flushMDB(struct hfsmount *hfsmp, int waitfor, int altflush)
{
	ExtendedVCB *vcb = HFSTOVCB(hfsmp);
	struct filefork *fp;
	HFSMasterDirectoryBlock	*mdb;
	struct buf *bp = NULL;
	int retval;
	int sectorsize;
	ByteCount namelen;

	sectorsize = hfsmp->hfs_logical_block_size;
	retval = (int)buf_bread(hfsmp->hfs_devvp, (daddr64_t)HFS_PRI_SECTOR(sectorsize), sectorsize, NOCRED, &bp);
	if (retval) {
		if (bp)
			buf_brelse(bp);
		return retval;
	}

	lck_mtx_lock(&hfsmp->hfs_mutex);

	mdb = (HFSMasterDirectoryBlock *)(buf_dataptr(bp) + HFS_PRI_OFFSET(sectorsize));
    
	mdb->drCrDate	= SWAP_BE32 (UTCToLocal(to_hfs_time(vcb->vcbCrDate)));
	mdb->drLsMod	= SWAP_BE32 (UTCToLocal(to_hfs_time(vcb->vcbLsMod)));
	mdb->drAtrb	= SWAP_BE16 (vcb->vcbAtrb);
	mdb->drNmFls	= SWAP_BE16 (vcb->vcbNmFls);
	mdb->drAllocPtr	= SWAP_BE16 (vcb->nextAllocation);
	mdb->drClpSiz	= SWAP_BE32 (vcb->vcbClpSiz);
	mdb->drNxtCNID	= SWAP_BE32 (vcb->vcbNxtCNID);
	mdb->drFreeBks	= SWAP_BE16 (vcb->freeBlocks);

	namelen = strlen((char *)vcb->vcbVN);
	retval = utf8_to_hfs(vcb, namelen, vcb->vcbVN, mdb->drVN);
	/* Retry with MacRoman in case that's how it was exported. */
	if (retval)
		retval = utf8_to_mac_roman(namelen, vcb->vcbVN, mdb->drVN);
	
	mdb->drVolBkUp	= SWAP_BE32 (UTCToLocal(to_hfs_time(vcb->vcbVolBkUp)));
	mdb->drWrCnt	= SWAP_BE32 (vcb->vcbWrCnt);
	mdb->drNmRtDirs	= SWAP_BE16 (vcb->vcbNmRtDirs);
	mdb->drFilCnt	= SWAP_BE32 (vcb->vcbFilCnt);
	mdb->drDirCnt	= SWAP_BE32 (vcb->vcbDirCnt);
	
	bcopy(vcb->vcbFndrInfo, mdb->drFndrInfo, sizeof(mdb->drFndrInfo));

	fp = VTOF(vcb->extentsRefNum);
	mdb->drXTExtRec[0].startBlock = SWAP_BE16 (fp->ff_extents[0].startBlock);
	mdb->drXTExtRec[0].blockCount = SWAP_BE16 (fp->ff_extents[0].blockCount);
	mdb->drXTExtRec[1].startBlock = SWAP_BE16 (fp->ff_extents[1].startBlock);
	mdb->drXTExtRec[1].blockCount = SWAP_BE16 (fp->ff_extents[1].blockCount);
	mdb->drXTExtRec[2].startBlock = SWAP_BE16 (fp->ff_extents[2].startBlock);
	mdb->drXTExtRec[2].blockCount = SWAP_BE16 (fp->ff_extents[2].blockCount);
	mdb->drXTFlSize	= SWAP_BE32 (fp->ff_blocks * vcb->blockSize);
	mdb->drXTClpSiz	= SWAP_BE32 (fp->ff_clumpsize);
	FTOC(fp)->c_flag &= ~C_MODIFIED;
	
	fp = VTOF(vcb->catalogRefNum);
	mdb->drCTExtRec[0].startBlock = SWAP_BE16 (fp->ff_extents[0].startBlock);
	mdb->drCTExtRec[0].blockCount = SWAP_BE16 (fp->ff_extents[0].blockCount);
	mdb->drCTExtRec[1].startBlock = SWAP_BE16 (fp->ff_extents[1].startBlock);
	mdb->drCTExtRec[1].blockCount = SWAP_BE16 (fp->ff_extents[1].blockCount);
	mdb->drCTExtRec[2].startBlock = SWAP_BE16 (fp->ff_extents[2].startBlock);
	mdb->drCTExtRec[2].blockCount = SWAP_BE16 (fp->ff_extents[2].blockCount);
	mdb->drCTFlSize	= SWAP_BE32 (fp->ff_blocks * vcb->blockSize);
	mdb->drCTClpSiz	= SWAP_BE32 (fp->ff_clumpsize);
	FTOC(fp)->c_flag &= ~C_MODIFIED;

	MarkVCBClean( vcb );

	lck_mtx_unlock(&hfsmp->hfs_mutex);

	/* If requested, flush out the alternate MDB */
	if (altflush) {
		struct buf *alt_bp = NULL;

		if (buf_meta_bread(hfsmp->hfs_devvp, hfsmp->hfs_alt_id_sector, sectorsize, NOCRED, &alt_bp) == 0) {
			bcopy(mdb, (char *)buf_dataptr(alt_bp) + HFS_ALT_OFFSET(sectorsize), kMDBSize);

			(void) VNOP_BWRITE(alt_bp);
		} else if (alt_bp)
			buf_brelse(alt_bp);
	}

	if (waitfor != MNT_WAIT)
		buf_bawrite(bp);
	else 
		retval = VNOP_BWRITE(bp);

	return (retval);
}

/*
 *  Flush any dirty in-memory mount data to the on-disk
 *  volume header.
 *
 *  Note: the on-disk volume signature is intentionally
 *  not flushed since the on-disk "H+" and "HX" signatures
 *  are always stored in-memory as "H+".
 */
__private_extern__
int
hfs_flushvolumeheader(struct hfsmount *hfsmp, int waitfor, int altflush)
{
	ExtendedVCB *vcb = HFSTOVCB(hfsmp);
	struct filefork *fp;
	HFSPlusVolumeHeader *volumeHeader, *altVH;
	int retval;
	struct buf *bp, *alt_bp;
	int i;
	daddr64_t priIDSector;
	int critical;
	u_int16_t  signature;
	u_int16_t  hfsversion;

	if (hfsmp->hfs_flags & HFS_READ_ONLY) {
		return(0);
	}
	if (hfsmp->hfs_flags & HFS_STANDARD) {
		return hfs_flushMDB(hfsmp, waitfor, altflush);
	}
	critical = altflush;
	priIDSector = (daddr64_t)((vcb->hfsPlusIOPosOffset / hfsmp->hfs_logical_block_size) +
				  HFS_PRI_SECTOR(hfsmp->hfs_logical_block_size));

	if (hfs_start_transaction(hfsmp) != 0) {
	    return EINVAL;
	}

	bp = NULL;
	alt_bp = NULL;

	retval = (int)buf_meta_bread(hfsmp->hfs_devvp, 
			HFS_PHYSBLK_ROUNDDOWN(priIDSector, hfsmp->hfs_log_per_phys),
			hfsmp->hfs_physical_block_size, NOCRED, &bp);
	if (retval) {
		printf("hfs: err %d reading VH blk (%s)\n", retval, vcb->vcbVN);
		goto err_exit;
	}

	volumeHeader = (HFSPlusVolumeHeader *)((char *)buf_dataptr(bp) + 
			HFS_PRI_OFFSET(hfsmp->hfs_physical_block_size));

	/*
	 * Sanity check what we just read.  If it's bad, try the alternate
	 * instead.
	 */
	signature = SWAP_BE16 (volumeHeader->signature);
	hfsversion   = SWAP_BE16 (volumeHeader->version);
	if ((signature != kHFSPlusSigWord && signature != kHFSXSigWord) ||
	    (hfsversion < kHFSPlusVersion) || (hfsversion > 100) ||
	    (SWAP_BE32 (volumeHeader->blockSize) != vcb->blockSize)) {
		printf("hfs: corrupt VH on %s, sig 0x%04x, ver %d, blksize %d%s\n",
		      vcb->vcbVN, signature, hfsversion,
		      SWAP_BE32 (volumeHeader->blockSize),
		      hfsmp->hfs_alt_id_sector ? "; trying alternate" : "");
		hfs_mark_volume_inconsistent(hfsmp);
		
		if (hfsmp->hfs_alt_id_sector) {
			retval = buf_meta_bread(hfsmp->hfs_devvp, 
			    HFS_PHYSBLK_ROUNDDOWN(hfsmp->hfs_alt_id_sector, hfsmp->hfs_log_per_phys),
			    hfsmp->hfs_physical_block_size, NOCRED, &alt_bp);
			if (retval) {
				printf("hfs: err %d reading alternate VH (%s)\n", retval, vcb->vcbVN);
				goto err_exit;
			}
			
			altVH = (HFSPlusVolumeHeader *)((char *)buf_dataptr(alt_bp) + 
				HFS_ALT_OFFSET(hfsmp->hfs_physical_block_size));
			signature = SWAP_BE16(altVH->signature);
			hfsversion = SWAP_BE16(altVH->version);
			
			if ((signature != kHFSPlusSigWord && signature != kHFSXSigWord) ||
			    (hfsversion < kHFSPlusVersion) || (kHFSPlusVersion > 100) ||
			    (SWAP_BE32(altVH->blockSize) != vcb->blockSize)) {
				printf("hfs: corrupt alternate VH on %s, sig 0x%04x, ver %d, blksize %d\n",
				    vcb->vcbVN, signature, hfsversion,
				    SWAP_BE32(altVH->blockSize));
				retval = EIO;
				goto err_exit;
			}
			
			/* The alternate is plausible, so use it. */
			bcopy(altVH, volumeHeader, kMDBSize);
			buf_brelse(alt_bp);
			alt_bp = NULL;
		} else {
			/* No alternate VH, nothing more we can do. */
			retval = EIO;
			goto err_exit;
		}
	}

	if (hfsmp->jnl) {
		journal_modify_block_start(hfsmp->jnl, bp);
	}

	/*
	 * For embedded HFS+ volumes, update create date if it changed
	 * (ie from a setattrlist call)
	 */
	if ((vcb->hfsPlusIOPosOffset != 0) &&
	    (SWAP_BE32 (volumeHeader->createDate) != vcb->localCreateDate)) {
		struct buf *bp2;
		HFSMasterDirectoryBlock	*mdb;

		retval = (int)buf_meta_bread(hfsmp->hfs_devvp, 
				HFS_PHYSBLK_ROUNDDOWN(HFS_PRI_SECTOR(hfsmp->hfs_logical_block_size), hfsmp->hfs_log_per_phys),
				hfsmp->hfs_physical_block_size, NOCRED, &bp2);
		if (retval) {
			if (bp2)
				buf_brelse(bp2);
			retval = 0;
		} else {
			mdb = (HFSMasterDirectoryBlock *)(buf_dataptr(bp2) +
				HFS_PRI_OFFSET(hfsmp->hfs_physical_block_size));

			if ( SWAP_BE32 (mdb->drCrDate) != vcb->localCreateDate )
			  {
				if (hfsmp->jnl) {
				    journal_modify_block_start(hfsmp->jnl, bp2);
				}

				mdb->drCrDate = SWAP_BE32 (vcb->localCreateDate);	/* pick up the new create date */

				if (hfsmp->jnl) {
					journal_modify_block_end(hfsmp->jnl, bp2, NULL, NULL);
				} else {
					(void) VNOP_BWRITE(bp2);		/* write out the changes */
				}
			  }
			else
			  {
				buf_brelse(bp2);						/* just release it */
			  }
		  }	
	}

	lck_mtx_lock(&hfsmp->hfs_mutex);

	/* Note: only update the lower 16 bits worth of attributes */
	volumeHeader->attributes       = SWAP_BE32 (vcb->vcbAtrb);
	volumeHeader->journalInfoBlock = SWAP_BE32 (vcb->vcbJinfoBlock);
	if (hfsmp->jnl) {
		volumeHeader->lastMountedVersion = SWAP_BE32 (kHFSJMountVersion);
	} else {
		volumeHeader->lastMountedVersion = SWAP_BE32 (kHFSPlusMountVersion);
	}
	volumeHeader->createDate	= SWAP_BE32 (vcb->localCreateDate);  /* volume create date is in local time */
	volumeHeader->modifyDate	= SWAP_BE32 (to_hfs_time(vcb->vcbLsMod));
	volumeHeader->backupDate	= SWAP_BE32 (to_hfs_time(vcb->vcbVolBkUp));
	volumeHeader->fileCount		= SWAP_BE32 (vcb->vcbFilCnt);
	volumeHeader->folderCount	= SWAP_BE32 (vcb->vcbDirCnt);
	volumeHeader->totalBlocks	= SWAP_BE32 (vcb->totalBlocks);
	volumeHeader->freeBlocks	= SWAP_BE32 (vcb->freeBlocks);
	volumeHeader->nextAllocation	= SWAP_BE32 (vcb->nextAllocation);
	volumeHeader->rsrcClumpSize	= SWAP_BE32 (vcb->vcbClpSiz);
	volumeHeader->dataClumpSize	= SWAP_BE32 (vcb->vcbClpSiz);
	volumeHeader->nextCatalogID	= SWAP_BE32 (vcb->vcbNxtCNID);
	volumeHeader->writeCount	= SWAP_BE32 (vcb->vcbWrCnt);
	volumeHeader->encodingsBitmap	= SWAP_BE64 (vcb->encodingsBitmap);

	if (bcmp(vcb->vcbFndrInfo, volumeHeader->finderInfo, sizeof(volumeHeader->finderInfo)) != 0) {
		bcopy(vcb->vcbFndrInfo, volumeHeader->finderInfo, sizeof(volumeHeader->finderInfo));
		critical = 1;
	}

	/*
	 * System files are only dirty when altflush is set.
	 */
	if (altflush == 0) {
		goto done;
	}

	/* Sync Extents over-flow file meta data */
	fp = VTOF(vcb->extentsRefNum);
	if (FTOC(fp)->c_flag & C_MODIFIED) {
		for (i = 0; i < kHFSPlusExtentDensity; i++) {
			volumeHeader->extentsFile.extents[i].startBlock	=
				SWAP_BE32 (fp->ff_extents[i].startBlock);
			volumeHeader->extentsFile.extents[i].blockCount	=
				SWAP_BE32 (fp->ff_extents[i].blockCount);
		}
		volumeHeader->extentsFile.logicalSize = SWAP_BE64 (fp->ff_size);
		volumeHeader->extentsFile.totalBlocks = SWAP_BE32 (fp->ff_blocks);
		volumeHeader->extentsFile.clumpSize   = SWAP_BE32 (fp->ff_clumpsize);
		FTOC(fp)->c_flag &= ~C_MODIFIED;
	}

	/* Sync Catalog file meta data */
	fp = VTOF(vcb->catalogRefNum);
	if (FTOC(fp)->c_flag & C_MODIFIED) {
		for (i = 0; i < kHFSPlusExtentDensity; i++) {
			volumeHeader->catalogFile.extents[i].startBlock	=
				SWAP_BE32 (fp->ff_extents[i].startBlock);
			volumeHeader->catalogFile.extents[i].blockCount	=
				SWAP_BE32 (fp->ff_extents[i].blockCount);
		}
		volumeHeader->catalogFile.logicalSize = SWAP_BE64 (fp->ff_size);
		volumeHeader->catalogFile.totalBlocks = SWAP_BE32 (fp->ff_blocks);
		volumeHeader->catalogFile.clumpSize   = SWAP_BE32 (fp->ff_clumpsize);
		FTOC(fp)->c_flag &= ~C_MODIFIED;
	}

	/* Sync Allocation file meta data */
	fp = VTOF(vcb->allocationsRefNum);
	if (FTOC(fp)->c_flag & C_MODIFIED) {
		for (i = 0; i < kHFSPlusExtentDensity; i++) {
			volumeHeader->allocationFile.extents[i].startBlock =
				SWAP_BE32 (fp->ff_extents[i].startBlock);
			volumeHeader->allocationFile.extents[i].blockCount =
				SWAP_BE32 (fp->ff_extents[i].blockCount);
		}
		volumeHeader->allocationFile.logicalSize = SWAP_BE64 (fp->ff_size);
		volumeHeader->allocationFile.totalBlocks = SWAP_BE32 (fp->ff_blocks);
		volumeHeader->allocationFile.clumpSize   = SWAP_BE32 (fp->ff_clumpsize);
		FTOC(fp)->c_flag &= ~C_MODIFIED;
	}

	/* Sync Attribute file meta data */
	if (hfsmp->hfs_attribute_vp) {
		fp = VTOF(hfsmp->hfs_attribute_vp);
		for (i = 0; i < kHFSPlusExtentDensity; i++) {
			volumeHeader->attributesFile.extents[i].startBlock =
				SWAP_BE32 (fp->ff_extents[i].startBlock);
			volumeHeader->attributesFile.extents[i].blockCount =
				SWAP_BE32 (fp->ff_extents[i].blockCount);
		}
		FTOC(fp)->c_flag &= ~C_MODIFIED;
		volumeHeader->attributesFile.logicalSize = SWAP_BE64 (fp->ff_size);
		volumeHeader->attributesFile.totalBlocks = SWAP_BE32 (fp->ff_blocks);
		volumeHeader->attributesFile.clumpSize   = SWAP_BE32 (fp->ff_clumpsize);
	}

	/* Sync Startup file meta data */
	if (hfsmp->hfs_startup_vp) {
		fp = VTOF(hfsmp->hfs_startup_vp);
		if (FTOC(fp)->c_flag & C_MODIFIED) {
			for (i = 0; i < kHFSPlusExtentDensity; i++) {
				volumeHeader->startupFile.extents[i].startBlock =
					SWAP_BE32 (fp->ff_extents[i].startBlock);
				volumeHeader->startupFile.extents[i].blockCount =
					SWAP_BE32 (fp->ff_extents[i].blockCount);
			}
			volumeHeader->startupFile.logicalSize = SWAP_BE64 (fp->ff_size);
			volumeHeader->startupFile.totalBlocks = SWAP_BE32 (fp->ff_blocks);
			volumeHeader->startupFile.clumpSize   = SWAP_BE32 (fp->ff_clumpsize);
			FTOC(fp)->c_flag &= ~C_MODIFIED;
		}
	}

done:
	MarkVCBClean(hfsmp);
	lck_mtx_unlock(&hfsmp->hfs_mutex);

	/* If requested, flush out the alternate volume header */
	if (altflush && hfsmp->hfs_alt_id_sector) {
		if (buf_meta_bread(hfsmp->hfs_devvp, 
				HFS_PHYSBLK_ROUNDDOWN(hfsmp->hfs_alt_id_sector, hfsmp->hfs_log_per_phys),
				hfsmp->hfs_physical_block_size, NOCRED, &alt_bp) == 0) {
			if (hfsmp->jnl) {
				journal_modify_block_start(hfsmp->jnl, alt_bp);
			}

			bcopy(volumeHeader, (char *)buf_dataptr(alt_bp) + 
					HFS_ALT_OFFSET(hfsmp->hfs_physical_block_size), 
					kMDBSize);

			if (hfsmp->jnl) {
				journal_modify_block_end(hfsmp->jnl, alt_bp, NULL, NULL);
			} else {
				(void) VNOP_BWRITE(alt_bp);
			}
		} else if (alt_bp)
			buf_brelse(alt_bp);
	}

	if (hfsmp->jnl) {
		journal_modify_block_end(hfsmp->jnl, bp, NULL, NULL);
	} else {
		if (waitfor != MNT_WAIT)
			buf_bawrite(bp);
		else {
		    retval = VNOP_BWRITE(bp);
		    /* When critical data changes, flush the device cache */
		    if (critical && (retval == 0)) {
			(void) VNOP_IOCTL(hfsmp->hfs_devvp, DKIOCSYNCHRONIZECACHE,
					 NULL, FWRITE, NULL);
		    }
		}
	}
	hfs_end_transaction(hfsmp);
 
	return (retval);

err_exit:
	if (alt_bp)
		buf_brelse(alt_bp);
	if (bp)
		buf_brelse(bp);
	hfs_end_transaction(hfsmp);
	return retval;
}


/*
 * Extend a file system.
 */
__private_extern__
int
hfs_extendfs(struct hfsmount *hfsmp, u_int64_t newsize, vfs_context_t context)
{
	struct proc *p = vfs_context_proc(context);
	kauth_cred_t cred = vfs_context_ucred(context);
	struct  vnode *vp;
	struct  vnode *devvp;
	struct  buf *bp;
	struct  filefork *fp = NULL;
	ExtendedVCB  *vcb;
	struct  cat_fork forkdata;
	u_int64_t  oldsize;
	u_int64_t  newblkcnt;
	u_int64_t  prev_phys_block_count;
	u_int32_t  addblks;
	u_int64_t  sectorcnt;
	u_int32_t  sectorsize;
	u_int32_t  phys_sectorsize;
	daddr64_t  prev_alt_sector;
	daddr_t	   bitmapblks;
	int  lockflags;
	int  error;
	int64_t oldBitmapSize;
	Boolean  usedExtendFileC = false;
	
	devvp = hfsmp->hfs_devvp;
	vcb = HFSTOVCB(hfsmp);

	/*
	 * - HFS Plus file systems only. 
	 * - Journaling must be enabled.
	 * - No embedded volumes.
	 */
	if ((vcb->vcbSigWord == kHFSSigWord) ||
	     (hfsmp->jnl == NULL) ||
	     (vcb->hfsPlusIOPosOffset != 0)) {
		return (EPERM);
	}
	/*
	 * If extending file system by non-root, then verify
	 * ownership and check permissions.
	 */
	if (suser(cred, NULL)) {
		error = hfs_vget(hfsmp, kHFSRootFolderID, &vp, 0);

		if (error)
			return (error);
		error = hfs_owner_rights(hfsmp, VTOC(vp)->c_uid, cred, p, 0);
		if (error == 0) {
			error = hfs_write_access(vp, cred, p, false);
		}
		hfs_unlock(VTOC(vp));
		vnode_put(vp);
		if (error)
			return (error);

		error = vnode_authorize(devvp, NULL, KAUTH_VNODE_READ_DATA | KAUTH_VNODE_WRITE_DATA, context);
		if (error)
			return (error);
	}
	if (VNOP_IOCTL(devvp, DKIOCGETBLOCKSIZE, (caddr_t)&sectorsize, 0, context)) {
		return (ENXIO);
	}
	if (sectorsize != hfsmp->hfs_logical_block_size) {
		return (ENXIO);
	}
	if (VNOP_IOCTL(devvp, DKIOCGETBLOCKCOUNT, (caddr_t)&sectorcnt, 0, context)) {
		return (ENXIO);
	}
	if ((sectorsize * sectorcnt) < newsize) {
		printf("hfs_extendfs: not enough space on device\n");
		return (ENOSPC);
	}
	error = VNOP_IOCTL(devvp, DKIOCGETPHYSICALBLOCKSIZE, (caddr_t)&phys_sectorsize, 0, context);
	if (error) {
		if ((error != ENOTSUP) && (error != ENOTTY)) {
			return (ENXIO);
		}
		/* If ioctl is not supported, force physical and logical sector size to be same */
		phys_sectorsize = sectorsize;
	}
	oldsize = (u_int64_t)hfsmp->totalBlocks * (u_int64_t)hfsmp->blockSize;

	/*
	 * Validate new size.
	 */
	if ((newsize <= oldsize) || (newsize % sectorsize) || (newsize % phys_sectorsize)) {
		printf("hfs_extendfs: invalid size\n");
		return (EINVAL);
	}
	newblkcnt = newsize / vcb->blockSize;
	if (newblkcnt > (u_int64_t)0xFFFFFFFF)
		return (EOVERFLOW);

	addblks = newblkcnt - vcb->totalBlocks;

	printf("hfs_extendfs: growing %s by %d blocks\n", vcb->vcbVN, addblks);
	/*
	 * Enclose changes inside a transaction.
	 */
	if (hfs_start_transaction(hfsmp) != 0) {
		return (EINVAL);
	}

	/*
	 * Note: we take the attributes lock in case we have an attribute data vnode
	 * which needs to change size.
	 */
	lockflags = hfs_systemfile_lock(hfsmp, SFL_ATTRIBUTE | SFL_EXTENTS | SFL_BITMAP, HFS_EXCLUSIVE_LOCK);
	vp = vcb->allocationsRefNum;
	fp = VTOF(vp);
	bcopy(&fp->ff_data, &forkdata, sizeof(forkdata));

	/*
	 * Calculate additional space required (if any) by allocation bitmap.
	 */
	oldBitmapSize = fp->ff_size;
	bitmapblks = roundup((newblkcnt+7) / 8, vcb->vcbVBMIOSize) / vcb->blockSize;
	if (bitmapblks > (daddr_t)fp->ff_blocks)
		bitmapblks -= fp->ff_blocks;
	else
		bitmapblks = 0;

	if (bitmapblks > 0) {
		daddr64_t blkno;
		daddr_t blkcnt;
		off_t bytesAdded;

		/*
		 * Get the bitmap's current size (in allocation blocks) so we know
		 * where to start zero filling once the new space is added.  We've
		 * got to do this before the bitmap is grown.
		 */
		blkno  = (daddr64_t)fp->ff_blocks;

		/*
		 * Try to grow the allocation file in the normal way, using allocation
		 * blocks already existing in the file system.  This way, we might be
		 * able to grow the bitmap contiguously, or at least in the metadata
		 * zone.
		 */
		error = ExtendFileC(vcb, fp, bitmapblks * vcb->blockSize, 0,
				kEFAllMask | kEFNoClumpMask | kEFReserveMask | kEFMetadataMask,
				&bytesAdded);

		if (error == 0) {
			usedExtendFileC = true;
		} else {
			/*
			 * If the above allocation failed, fall back to allocating the new
			 * extent of the bitmap from the space we're going to add.  Since those
			 * blocks don't yet belong to the file system, we have to update the
			 * extent list directly, and manually adjust the file size.
			 */
			bytesAdded = 0;
			error = AddFileExtent(vcb, fp, vcb->totalBlocks, bitmapblks);
			if (error) {
				printf("hfs_extendfs: error %d adding extents\n", error);
				goto out;
			}
			fp->ff_blocks += bitmapblks;
			VTOC(vp)->c_blocks = fp->ff_blocks;
			VTOC(vp)->c_flag |= C_MODIFIED;
		}
		
		/*
		 * Update the allocation file's size to include the newly allocated
		 * blocks.  Note that ExtendFileC doesn't do this, which is why this
		 * statement is outside the above "if" statement.
		 */
		fp->ff_size += (u_int64_t)bitmapblks * (u_int64_t)vcb->blockSize;
		
		/*
		 * Zero out the new bitmap blocks.
		 */
		{
	
			bp = NULL;
			blkcnt = bitmapblks;
			while (blkcnt > 0) {
				error = (int)buf_meta_bread(vp, blkno, vcb->blockSize, NOCRED, &bp);
				if (error) {
					if (bp) {
						buf_brelse(bp);
					}
					break;
				}
				bzero((char *)buf_dataptr(bp), vcb->blockSize);
				buf_markaged(bp);
				error = (int)buf_bwrite(bp);
				if (error)
					break;
				--blkcnt;
				++blkno;
			}
		}
		if (error) {
			printf("hfs_extendfs: error %d  clearing blocks\n", error);
			goto out;
		}
		/*
		 * Mark the new bitmap space as allocated.
		 *
		 * Note that ExtendFileC will have marked any blocks it allocated, so
		 * this is only needed if we used AddFileExtent.  Also note that this
		 * has to come *after* the zero filling of new blocks in the case where
		 * we used AddFileExtent (since the part of the bitmap we're touching
		 * is in those newly allocated blocks).
		 */
		if (!usedExtendFileC) {
			error = BlockMarkAllocated(vcb, vcb->totalBlocks, bitmapblks);
			if (error) {
				printf("hfs_extendfs: error %d setting bitmap\n", error);
				goto out;
			}
			vcb->freeBlocks -= bitmapblks;
		}
	}
	/*
	 * Mark the new alternate VH as allocated.
	 */
	if (vcb->blockSize == 512)
		error = BlockMarkAllocated(vcb, vcb->totalBlocks + addblks - 2, 2);
	else
		error = BlockMarkAllocated(vcb, vcb->totalBlocks + addblks - 1, 1);
	if (error) {
		printf("hfs_extendfs: error %d setting bitmap (VH)\n", error);
		goto out;
	}
	/*
	 * Mark the old alternate VH as free.
	 */
	if (vcb->blockSize == 512)
		(void) BlockMarkFree(vcb, vcb->totalBlocks - 2, 2);
	else 
		(void) BlockMarkFree(vcb, vcb->totalBlocks - 1, 1);
	/*
	 * Adjust file system variables for new space.
	 */
	prev_phys_block_count = hfsmp->hfs_logical_block_count;
	prev_alt_sector = hfsmp->hfs_alt_id_sector;

	vcb->totalBlocks += addblks;
	vcb->freeBlocks += addblks;
	hfsmp->hfs_logical_block_count = newsize / sectorsize;
	hfsmp->hfs_alt_id_sector = (hfsmp->hfsPlusIOPosOffset / sectorsize) +
	                          HFS_ALT_SECTOR(sectorsize, hfsmp->hfs_logical_block_count);
	MarkVCBDirty(vcb);
	error = hfs_flushvolumeheader(hfsmp, MNT_WAIT, HFS_ALTFLUSH);
	if (error) {
		printf("hfs_extendfs: couldn't flush volume headers (%d)", error);
		/*
		 * Restore to old state.
		 */
		if (usedExtendFileC) {
			(void) TruncateFileC(vcb, fp, oldBitmapSize, false);
		} else {
			fp->ff_blocks -= bitmapblks;
			fp->ff_size -= (u_int64_t)bitmapblks * (u_int64_t)vcb->blockSize;
			/*
			 * No need to mark the excess blocks free since those bitmap blocks
			 * are no longer part of the bitmap.  But we do need to undo the
			 * effect of the "vcb->freeBlocks -= bitmapblks" above.
			 */
			vcb->freeBlocks += bitmapblks;
		}
		vcb->totalBlocks -= addblks;
		vcb->freeBlocks -= addblks;
		hfsmp->hfs_logical_block_count = prev_phys_block_count;
		hfsmp->hfs_alt_id_sector = prev_alt_sector;
		MarkVCBDirty(vcb);
		if (vcb->blockSize == 512)
			(void) BlockMarkAllocated(vcb, vcb->totalBlocks - 2, 2);
		else
			(void) BlockMarkAllocated(vcb, vcb->totalBlocks - 1, 1);
		goto out;
	}
	/*
	 * Invalidate the old alternate volume header.
	 */
	bp = NULL;
	if (prev_alt_sector) {
		if (buf_meta_bread(hfsmp->hfs_devvp, 
				HFS_PHYSBLK_ROUNDDOWN(prev_alt_sector, hfsmp->hfs_log_per_phys),
				hfsmp->hfs_physical_block_size, NOCRED, &bp) == 0) {
			journal_modify_block_start(hfsmp->jnl, bp);
	
			bzero((char *)buf_dataptr(bp) + HFS_ALT_OFFSET(hfsmp->hfs_physical_block_size), kMDBSize);
	
			journal_modify_block_end(hfsmp->jnl, bp, NULL, NULL);
		} else if (bp) {
			buf_brelse(bp);
		}
	}
	
	/*
	 * TODO: Adjust the size of the metadata zone based on new volume size?
	 */
	 
	/*
	 * Adjust the size of hfsmp->hfs_attrdata_vp
	 */
	if (hfsmp->hfs_attrdata_vp) {
		struct cnode *attr_cp;
		struct filefork *attr_fp;
		
		if (vnode_get(hfsmp->hfs_attrdata_vp) == 0) {
			attr_cp = VTOC(hfsmp->hfs_attrdata_vp);
			attr_fp = VTOF(hfsmp->hfs_attrdata_vp);
			
			attr_cp->c_blocks = newblkcnt;
			attr_fp->ff_blocks = newblkcnt;
			attr_fp->ff_extents[0].blockCount = newblkcnt;
			attr_fp->ff_size = (off_t) newblkcnt * hfsmp->blockSize;
			ubc_setsize(hfsmp->hfs_attrdata_vp, attr_fp->ff_size);
			vnode_put(hfsmp->hfs_attrdata_vp);
		}
	}

out:
	if (error && fp) {
		/* Restore allocation fork. */
		bcopy(&forkdata, &fp->ff_data, sizeof(forkdata));
		VTOC(vp)->c_blocks = fp->ff_blocks;

	}
	/*
	   Regardless of whether or not the totalblocks actually increased,
	   we should reset the allocLimit field. If it changed, it will
	   get updated; if not, it will remain the same.
	*/
	hfsmp->allocLimit = vcb->totalBlocks;
	hfs_systemfile_unlock(hfsmp, lockflags);
	hfs_end_transaction(hfsmp);

	return (error);
}

#define HFS_MIN_SIZE  (32LL * 1024LL * 1024LL)

/*
 * Truncate a file system (while still mounted).
 */
__private_extern__
int
hfs_truncatefs(struct hfsmount *hfsmp, u_int64_t newsize, vfs_context_t context)
{
	struct  buf *bp = NULL;
	u_int64_t oldsize;
	u_int32_t newblkcnt;
	u_int32_t reclaimblks = 0;
	int lockflags = 0;
	int transaction_begun = 0;
	int error;

	lck_mtx_lock(&hfsmp->hfs_mutex);
	if (hfsmp->hfs_flags & HFS_RESIZE_IN_PROGRESS) {
		lck_mtx_unlock(&hfsmp->hfs_mutex);
		return (EALREADY);
	}
	hfsmp->hfs_flags |= HFS_RESIZE_IN_PROGRESS;
	hfsmp->hfs_resize_filesmoved = 0;
	hfsmp->hfs_resize_totalfiles = 0;
	lck_mtx_unlock(&hfsmp->hfs_mutex);

	/*
	 * - Journaled HFS Plus volumes only.
	 * - No embedded volumes.
	 */
	if ((hfsmp->jnl == NULL) ||
	    (hfsmp->hfsPlusIOPosOffset != 0)) {
		error = EPERM;
		goto out;
	}
	oldsize = (u_int64_t)hfsmp->totalBlocks * (u_int64_t)hfsmp->blockSize;
	newblkcnt = newsize / hfsmp->blockSize;
	reclaimblks = hfsmp->totalBlocks - newblkcnt;

	/* Make sure new size is valid. */
	if ((newsize < HFS_MIN_SIZE) ||
	    (newsize >= oldsize) ||
	    (newsize % hfsmp->hfs_logical_block_size) ||
	    (newsize % hfsmp->hfs_physical_block_size)) {
		printf ("hfs_truncatefs: invalid size\n");
		error = EINVAL;
		goto out;
	}
	/* Make sure there's enough space to work with. */
	if (reclaimblks >= hfs_freeblks(hfsmp, 1)) {
		printf("hfs_truncatefs: insufficient space (need %u blocks; have %u blocks)\n", reclaimblks, hfs_freeblks(hfsmp, 1));
		error = ENOSPC;
		goto out;
	}
	
	/* Start with a clean journal. */
	hfs_journal_flush(hfsmp);
	
	if (hfs_start_transaction(hfsmp) != 0) {
		error = EINVAL;
		goto out;
	}
	transaction_begun = 1;

	/*
	 * Prevent new allocations from using the part we're trying to truncate.
	 *
	 * NOTE: allocLimit is set to the allocation block number where the new
	 * alternate volume header will be.  That way there will be no files to
	 * interfere with allocating the new alternate volume header, and no files
	 * in the allocation blocks beyond (i.e. the blocks we're trying to
	 * truncate away.
	 */
	lck_mtx_lock(&hfsmp->hfs_mutex);
	if (hfsmp->blockSize == 512) 
		hfsmp->allocLimit = newblkcnt - 2;
	else
		hfsmp->allocLimit = newblkcnt - 1;
	hfsmp->freeBlocks -= reclaimblks;
	lck_mtx_unlock(&hfsmp->hfs_mutex);
	
	/*
	 * Look for files that have blocks at or beyond the location of the
	 * new alternate volume header.
	 */
	if (hfs_isallocated(hfsmp, hfsmp->allocLimit, reclaimblks)) {
		/*
		 * hfs_reclaimspace will use separate transactions when
		 * relocating files (so we don't overwhelm the journal).
		 */
		hfs_end_transaction(hfsmp);
		transaction_begun = 0;

		/* Attempt to reclaim some space. */ 
		if (hfs_reclaimspace(hfsmp, hfsmp->allocLimit, reclaimblks, context) != 0) {
			printf("hfs_truncatefs: couldn't reclaim space on %s\n", hfsmp->vcbVN);
			error = ENOSPC;
			goto out;
		}
		if (hfs_start_transaction(hfsmp) != 0) {
			error = EINVAL;
			goto out;
		}
		transaction_begun = 1;
		
		/* Check if we're clear now. */
		if (hfs_isallocated(hfsmp, hfsmp->allocLimit, reclaimblks)) {
			printf("hfs_truncatefs: didn't reclaim enough space on %s\n", hfsmp->vcbVN);
			error = EAGAIN;  /* tell client to try again */
			goto out;
		}
	}
	
	/*
	 * Note: we take the attributes lock in case we have an attribute data vnode
	 * which needs to change size.
	 */
	lockflags = hfs_systemfile_lock(hfsmp, SFL_ATTRIBUTE | SFL_EXTENTS | SFL_BITMAP, HFS_EXCLUSIVE_LOCK);

	/*
	 * Mark the old alternate volume header as free. 
	 * We don't bother shrinking allocation bitmap file.
	 */
	if (hfsmp->blockSize == 512) 
		(void) BlockMarkFree(hfsmp, hfsmp->totalBlocks - 2, 2);
	else 
		(void) BlockMarkFree(hfsmp, hfsmp->totalBlocks - 1, 1);

	/*
	 * Allocate last 1KB for alternate volume header.
	 */
	error = BlockMarkAllocated(hfsmp, hfsmp->allocLimit, (hfsmp->blockSize == 512) ? 2 : 1);
	if (error) {
		printf("hfs_truncatefs: Error %d allocating new alternate volume header\n", error);
		goto out;
	}

	/*
	 * Invalidate the existing alternate volume header.
	 *
	 * Don't include this in a transaction (don't call journal_modify_block)
	 * since this block will be outside of the truncated file system!
	 */
	if (hfsmp->hfs_alt_id_sector) {
		if (buf_meta_bread(hfsmp->hfs_devvp, 
				HFS_PHYSBLK_ROUNDDOWN(hfsmp->hfs_alt_id_sector, hfsmp->hfs_log_per_phys),
				hfsmp->hfs_physical_block_size, NOCRED, &bp) == 0) {
	
			bzero((void*)((char *)buf_dataptr(bp) + HFS_ALT_OFFSET(hfsmp->hfs_physical_block_size)), kMDBSize);
			(void) VNOP_BWRITE(bp);
		} else if (bp) {
			buf_brelse(bp);
		}
		bp = NULL;
	}

	/* Log successful shrinking. */
	printf("hfs_truncatefs: shrank \"%s\" to %d blocks (was %d blocks)\n",
	       hfsmp->vcbVN, newblkcnt, hfsmp->totalBlocks);

	/*
	 * Adjust file system variables and flush them to disk.
	 */
	hfsmp->totalBlocks = newblkcnt;
	hfsmp->hfs_logical_block_count = newsize / hfsmp->hfs_logical_block_size;
	hfsmp->hfs_alt_id_sector = HFS_ALT_SECTOR(hfsmp->hfs_logical_block_size, hfsmp->hfs_logical_block_count);
	MarkVCBDirty(hfsmp);
	error = hfs_flushvolumeheader(hfsmp, MNT_WAIT, HFS_ALTFLUSH);
	if (error)
		panic("hfs_truncatefs: unexpected error flushing volume header (%d)\n", error);
	
	/*
	 * TODO: Adjust the size of the metadata zone based on new volume size?
	 */
	 
	/*
	 * Adjust the size of hfsmp->hfs_attrdata_vp
	 */
	if (hfsmp->hfs_attrdata_vp) {
		struct cnode *cp;
		struct filefork *fp;
		
		if (vnode_get(hfsmp->hfs_attrdata_vp) == 0) {
			cp = VTOC(hfsmp->hfs_attrdata_vp);
			fp = VTOF(hfsmp->hfs_attrdata_vp);
			
			cp->c_blocks = newblkcnt;
			fp->ff_blocks = newblkcnt;
			fp->ff_extents[0].blockCount = newblkcnt;
			fp->ff_size = (off_t) newblkcnt * hfsmp->blockSize;
			ubc_setsize(hfsmp->hfs_attrdata_vp, fp->ff_size);
			vnode_put(hfsmp->hfs_attrdata_vp);
		}
	}
	
out:
	if (error)
		hfsmp->freeBlocks += reclaimblks;
	
	lck_mtx_lock(&hfsmp->hfs_mutex);
	hfsmp->allocLimit = hfsmp->totalBlocks;
	if (hfsmp->nextAllocation >= hfsmp->allocLimit)
		hfsmp->nextAllocation = hfsmp->hfs_metazone_end + 1;
	hfsmp->hfs_flags &= ~HFS_RESIZE_IN_PROGRESS;
	lck_mtx_unlock(&hfsmp->hfs_mutex);
	
	if (lockflags) {
		hfs_systemfile_unlock(hfsmp, lockflags);
	}
	if (transaction_begun) {
		hfs_end_transaction(hfsmp);
		hfs_journal_flush(hfsmp);
	}

	return (error);
}


/*
 * Invalidate the physical block numbers associated with buffer cache blocks
 * in the given extent of the given vnode.
 */
struct hfs_inval_blk_no {
	daddr64_t sectorStart;
	daddr64_t sectorCount;
};
static int
hfs_invalidate_block_numbers_callback(buf_t bp, void *args_in)
{
	daddr64_t blkno;
	struct hfs_inval_blk_no *args;
	
	blkno = buf_blkno(bp);
	args = args_in;
	
	if (blkno >= args->sectorStart && blkno < args->sectorStart+args->sectorCount)
		buf_setblkno(bp, buf_lblkno(bp));

	return BUF_RETURNED;
}
static void
hfs_invalidate_sectors(struct vnode *vp, daddr64_t sectorStart, daddr64_t sectorCount)
{
	struct hfs_inval_blk_no args;
	args.sectorStart = sectorStart;
	args.sectorCount = sectorCount;
	
	buf_iterate(vp, hfs_invalidate_block_numbers_callback, BUF_SCAN_DIRTY|BUF_SCAN_CLEAN, &args);
}


/*
 * Copy the contents of an extent to a new location.  Also invalidates the
 * physical block number of any buffer cache block in the copied extent
 * (so that if the block is written, it will go through VNOP_BLOCKMAP to
 * determine the new physical block number).
 */
static int
hfs_copy_extent(
	struct hfsmount *hfsmp,
	struct vnode *vp,		/* The file whose extent is being copied. */
	u_int32_t oldStart,		/* The start of the source extent. */
	u_int32_t newStart,		/* The start of the destination extent. */
	u_int32_t blockCount,	/* The number of allocation blocks to copy. */
	vfs_context_t context)
{
	int err = 0;
	size_t bufferSize;
	void *buffer = NULL;
	struct vfsioattr ioattr;
	buf_t bp = NULL;
	off_t resid;
	size_t ioSize;
	u_int32_t ioSizeSectors;	/* Device sectors in this I/O */
	daddr64_t srcSector, destSector;
	u_int32_t sectorsPerBlock = hfsmp->blockSize / hfsmp->hfs_logical_block_size;

	/*
	 * Sanity check that we have locked the vnode of the file we're copying.
	 *
	 * But since hfs_systemfile_lock() doesn't actually take the lock on
	 * the allocation file if a journal is active, ignore the check if the
	 * file being copied is the allocation file.
	 */
	struct cnode *cp = VTOC(vp);
	if (cp != hfsmp->hfs_allocation_cp && cp->c_lockowner != current_thread())
		panic("hfs_copy_extent: vp=%p (cp=%p) not owned?\n", vp, cp);

	/*
	 * Wait for any in-progress writes to this vnode to complete, so that we'll
	 * be copying consistent bits.  (Otherwise, it's possible that an async
	 * write will complete to the old extent after we read from it.  That
	 * could lead to corruption.)
	 */
	err = vnode_waitforwrites(vp, 0, 0, 0, "hfs_copy_extent");
	if (err) {
		printf("hfs_copy_extent: Error %d from vnode_waitforwrites\n", err);
		return err;
	}
	
	/*
	 * Determine the I/O size to use
	 *
	 * NOTE: Many external drives will result in an ioSize of 128KB.
	 * TODO: Should we use a larger buffer, doing several consecutive
	 * reads, then several consecutive writes?
	 */
	vfs_ioattr(hfsmp->hfs_mp, &ioattr);
	bufferSize = MIN(ioattr.io_maxreadcnt, ioattr.io_maxwritecnt);
	if (kmem_alloc(kernel_map, (vm_offset_t*) &buffer, bufferSize))
		return ENOMEM;

	/* Get a buffer for doing the I/O */
	bp = buf_alloc(hfsmp->hfs_devvp);
	buf_setdataptr(bp, (uintptr_t)buffer);
	
	resid = (off_t) blockCount * (off_t) hfsmp->blockSize;
	srcSector = (daddr64_t) oldStart * hfsmp->blockSize / hfsmp->hfs_logical_block_size;
	destSector = (daddr64_t) newStart * hfsmp->blockSize / hfsmp->hfs_logical_block_size;
	while (resid > 0) {
		ioSize = MIN(bufferSize, (size_t) resid);
		ioSizeSectors = ioSize / hfsmp->hfs_logical_block_size;
		
		/* Prepare the buffer for reading */
		buf_reset(bp, B_READ);
		buf_setsize(bp, ioSize);
		buf_setcount(bp, ioSize);
		buf_setblkno(bp, srcSector);
		buf_setlblkno(bp, srcSector);
		
		/* Do the read */
		err = VNOP_STRATEGY(bp);
		if (!err)
			err = buf_biowait(bp);
		if (err) {
			printf("hfs_copy_extent: Error %d from VNOP_STRATEGY (read)\n", err);
			break;
		}
		
		/* Prepare the buffer for writing */
		buf_reset(bp, B_WRITE);
		buf_setsize(bp, ioSize);
		buf_setcount(bp, ioSize);
		buf_setblkno(bp, destSector);
		buf_setlblkno(bp, destSector);
		if (journal_uses_fua(hfsmp->jnl))
			buf_markfua(bp);
			
		/* Do the write */
		vnode_startwrite(hfsmp->hfs_devvp);
		err = VNOP_STRATEGY(bp);
		if (!err)
			err = buf_biowait(bp);
		if (err) {
			printf("hfs_copy_extent: Error %d from VNOP_STRATEGY (write)\n", err);
			break;
		}
		
		resid -= ioSize;
		srcSector += ioSizeSectors;
		destSector += ioSizeSectors;
	}
	if (bp)
		buf_free(bp);
	if (buffer)
		kmem_free(kernel_map, (vm_offset_t)buffer, bufferSize);

	/* Make sure all writes have been flushed to disk. */
	if (!journal_uses_fua(hfsmp->jnl)) {
		err = VNOP_IOCTL(hfsmp->hfs_devvp, DKIOCSYNCHRONIZECACHE, NULL, FWRITE, context);
		if (err) {
			printf("hfs_copy_extent: DKIOCSYNCHRONIZECACHE failed (%d)\n", err);
			err = 0;	/* Don't fail the copy. */
		}
	}

	if (!err)
		hfs_invalidate_sectors(vp, (daddr64_t)oldStart*sectorsPerBlock, (daddr64_t)blockCount*sectorsPerBlock);

	return err;
}


/*
 * Reclaim space at the end of a volume, used by a given system file.
 *
 * This routine attempts to move any extent which contains allocation blocks
 * at or after "startblk."  A separate transaction is used to do the move.
 * The contents of any moved extents are read and written via the volume's
 * device vnode -- NOT via "vp."  During the move, moved blocks which are part
 * of a transaction have their physical block numbers invalidated so they will
 * eventually be written to their new locations.
 *
 * This routine can be used to move overflow extents for the allocation file.
 *
 * Inputs:
 *    hfsmp       The volume being resized.
 *    startblk    Blocks >= this allocation block need to be moved.
 *    locks       Which locks need to be taken for the given system file.
 *    vp          The vnode for the system file.
 *
 * Outputs:
 *    moved       Set to true if any extents were moved.
 */
static int
hfs_relocate_callback(__unused HFSPlusExtentKey *key, HFSPlusExtentRecord *record, HFSPlusExtentRecord *state)
{
	bcopy(state, record, sizeof(HFSPlusExtentRecord));
	return 0;
}
static int
hfs_reclaim_sys_file(struct hfsmount *hfsmp, struct vnode *vp, u_long startblk, int locks, Boolean *moved, vfs_context_t context)
{
	int error;
	int lockflags;
	int i;
	u_long datablks;
	u_long block;
	u_int32_t oldStartBlock;
	u_int32_t newStartBlock;
	u_int32_t blockCount;
	struct filefork *fp;

	/* If there is no vnode for this file, then there's nothing to do. */	
	if (vp == NULL)
		return 0;

	/* printf("hfs_reclaim_sys_file: %.*s\n", VTOC(vp)->c_desc.cd_namelen, VTOC(vp)->c_desc.cd_nameptr); */
	
	/* We always need the allocation bitmap and extents B-tree */
	locks |= SFL_BITMAP | SFL_EXTENTS;
	
	error = hfs_start_transaction(hfsmp);
	if (error) {
		printf("hfs_reclaim_sys_file: hfs_start_transaction returned %d\n", error);
		return error;
	}
	lockflags = hfs_systemfile_lock(hfsmp, locks, HFS_EXCLUSIVE_LOCK);
	fp = VTOF(vp);
	datablks = 0;

	/* Relocate non-overflow extents */
	for (i = 0; i < kHFSPlusExtentDensity; ++i) {
		if (fp->ff_extents[i].blockCount == 0)
			break;
		oldStartBlock = fp->ff_extents[i].startBlock;
		blockCount = fp->ff_extents[i].blockCount;
		datablks += blockCount;
		block = oldStartBlock + blockCount;
		if (block > startblk) {
			error = BlockAllocate(hfsmp, 1, blockCount, blockCount, true, true, &newStartBlock, &blockCount);
			if (error) {
				printf("hfs_reclaim_sys_file: BlockAllocate returned %d\n", error);
				goto fail;
			}
			if (blockCount != fp->ff_extents[i].blockCount) {
				printf("hfs_reclaim_sys_file: new blockCount=%u, original blockCount=%u", blockCount, fp->ff_extents[i].blockCount);
				goto free_fail;
			}
			error = hfs_copy_extent(hfsmp, vp, oldStartBlock, newStartBlock, blockCount, context);
			if (error) {
				printf("hfs_reclaim_sys_file: hfs_copy_extent returned %d\n", error);
				goto free_fail;
			}
			fp->ff_extents[i].startBlock = newStartBlock;
			VTOC(vp)->c_flag |= C_MODIFIED;
			*moved = true;
			error = BlockDeallocate(hfsmp, oldStartBlock, blockCount);
			if (error) {
				/* TODO: Mark volume inconsistent? */
				printf("hfs_reclaim_sys_file: BlockDeallocate returned %d\n", error);
				goto fail;
			}
			error = hfs_flushvolumeheader(hfsmp, MNT_WAIT, HFS_ALTFLUSH);
			if (error) {
				/* TODO: Mark volume inconsistent? */
				printf("hfs_reclaim_sys_file: hfs_flushvolumeheader returned %d\n", error);
				goto fail;
			}
		}
	}

	/* Relocate overflow extents (if any) */
	if (i == kHFSPlusExtentDensity && fp->ff_blocks > datablks) {
		struct BTreeIterator *iterator = NULL;
		struct FSBufferDescriptor btdata;
		HFSPlusExtentRecord record;
		HFSPlusExtentKey *key;
		FCB *fcb;
		u_int32_t fileID;
		u_int8_t forktype;

		forktype = VNODE_IS_RSRC(vp) ? 0xFF : 0;
		fileID = VTOC(vp)->c_cnid;
		if (kmem_alloc(kernel_map, (vm_offset_t*) &iterator, sizeof(*iterator))) {
			printf("hfs_reclaim_sys_file: kmem_alloc failed!\n");
			error = ENOMEM;
			goto fail;
		}

		bzero(iterator, sizeof(*iterator));
		key = (HFSPlusExtentKey *) &iterator->key;
		key->keyLength = kHFSPlusExtentKeyMaximumLength;
		key->forkType = forktype;
		key->fileID = fileID;
		key->startBlock = datablks;
	
		btdata.bufferAddress = &record;
		btdata.itemSize = sizeof(record);
		btdata.itemCount = 1;
	
		fcb = VTOF(hfsmp->hfs_extents_vp);

		error = BTSearchRecord(fcb, iterator, &btdata, NULL, iterator);
		while (error == 0) {
			/* Stop when we encounter a different file or fork. */
			if ((key->fileID != fileID) ||
				(key->forkType != forktype)) {
				break;
			}
			/* 
			 * Check if the file overlaps target space.
			 */
			for (i = 0; i < kHFSPlusExtentDensity; ++i) {
				if (record[i].blockCount == 0) {
					goto overflow_done;
				}
				oldStartBlock = record[i].startBlock;
				blockCount = record[i].blockCount;
				block = oldStartBlock + blockCount;
				if (block > startblk) {
					error = BlockAllocate(hfsmp, 1, blockCount, blockCount, true, true, &newStartBlock, &blockCount);
					if (error) {
						printf("hfs_reclaim_sys_file: BlockAllocate returned %d\n", error);
						goto overflow_done;
					}
					if (blockCount != record[i].blockCount) {
						printf("hfs_reclaim_sys_file: new blockCount=%u, original blockCount=%u", blockCount, fp->ff_extents[i].blockCount);
						kmem_free(kernel_map, (vm_offset_t)iterator, sizeof(*iterator));
						goto free_fail;
					}
					error = hfs_copy_extent(hfsmp, vp, oldStartBlock, newStartBlock, blockCount, context);
					if (error) {
						printf("hfs_reclaim_sys_file: hfs_copy_extent returned %d\n", error);
						kmem_free(kernel_map, (vm_offset_t)iterator, sizeof(*iterator));
						goto free_fail;
					}
					record[i].startBlock = newStartBlock;
					VTOC(vp)->c_flag |= C_MODIFIED;
					*moved = true;
					/*
					 * NOTE: To support relocating overflow extents of the
					 * allocation file, we must update the BTree record BEFORE
					 * deallocating the old extent so that BlockDeallocate will
					 * use the extent's new location to calculate physical block
					 * numbers.  (This is for the case where the old extent's
					 * bitmap bits actually reside in the extent being moved.)
					 */
					error = BTUpdateRecord(fcb, iterator, (IterateCallBackProcPtr) hfs_relocate_callback, &record);
					if (error) {
						/* TODO: Mark volume inconsistent? */
						printf("hfs_reclaim_sys_file: BTUpdateRecord returned %d\n", error);
						goto overflow_done;
					}
					error = BlockDeallocate(hfsmp, oldStartBlock, blockCount);
					if (error) {
						/* TODO: Mark volume inconsistent? */
						printf("hfs_reclaim_sys_file: BlockDeallocate returned %d\n", error);
						goto overflow_done;
					}
				}
			}
			/* Look for more records. */
			error = BTIterateRecord(fcb, kBTreeNextRecord, iterator, &btdata, NULL);
			if (error == btNotFound) {
				error = 0;
				break;
			}
		}
overflow_done:
		kmem_free(kernel_map, (vm_offset_t)iterator, sizeof(*iterator));
		if (error) {
			goto fail;
		}
	}
	
	hfs_systemfile_unlock(hfsmp, lockflags);
	error = hfs_end_transaction(hfsmp);
	if (error) {
		printf("hfs_reclaim_sys_file: hfs_end_transaction returned %d\n", error);
	}

	return error;

free_fail:
	(void) BlockDeallocate(hfsmp, newStartBlock, blockCount);
fail:
	(void) hfs_systemfile_unlock(hfsmp, lockflags);
	(void) hfs_end_transaction(hfsmp);
	return error;
}


/*
 * This journal_relocate callback updates the journal info block to point
 * at the new journal location.  This write must NOT be done using the
 * transaction.  We must write the block immediately.  We must also force
 * it to get to the media so that the new journal location will be seen by
 * the replay code before we can safely let journaled blocks be written
 * to their normal locations.
 *
 * The tests for journal_uses_fua below are mildly hacky.  Since the journal
 * and the file system are both on the same device, I'm leveraging what
 * the journal has decided about FUA.
 */
struct hfs_journal_relocate_args {
	struct hfsmount *hfsmp;
	vfs_context_t context;
	u_int32_t newStartBlock;
};

static errno_t
hfs_journal_relocate_callback(void *_args)
{
	int error;
	struct hfs_journal_relocate_args *args = _args;
	struct hfsmount *hfsmp = args->hfsmp;
	buf_t bp;
	JournalInfoBlock *jibp;

	error = buf_meta_bread(hfsmp->hfs_devvp,
		hfsmp->vcbJinfoBlock * (hfsmp->blockSize/hfsmp->hfs_logical_block_size),
		hfsmp->blockSize, vfs_context_ucred(args->context), &bp);
	if (error) {
		printf("hfs_reclaim_journal_file: failed to read JIB (%d)\n", error);
		return error;
	}
	jibp = (JournalInfoBlock*) buf_dataptr(bp);
	jibp->offset = SWAP_BE64((u_int64_t)args->newStartBlock * hfsmp->blockSize);
	jibp->size = SWAP_BE64(hfsmp->jnl_size);
	if (journal_uses_fua(hfsmp->jnl))
		buf_markfua(bp);
	error = buf_bwrite(bp);
	if (error) {
		printf("hfs_reclaim_journal_file: failed to write JIB (%d)\n", error);
		return error;
	}
	if (!journal_uses_fua(hfsmp->jnl)) {
		error = VNOP_IOCTL(hfsmp->hfs_devvp, DKIOCSYNCHRONIZECACHE, NULL, FWRITE, args->context);
		if (error) {
			printf("hfs_reclaim_journal_file: DKIOCSYNCHRONIZECACHE failed (%d)\n", error);
			error = 0;		/* Don't fail the operation. */
		}
	}

	return error;
}


static int
hfs_reclaim_journal_file(struct hfsmount *hfsmp, vfs_context_t context)
{
	int error;
	int lockflags;
	u_int32_t newStartBlock;
	u_int32_t oldBlockCount;
	u_int32_t newBlockCount;
	struct cat_desc journal_desc;
	struct cat_attr journal_attr;
	struct cat_fork journal_fork;
	struct hfs_journal_relocate_args callback_args;

	error = hfs_start_transaction(hfsmp);
	if (error) {
		printf("hfs_reclaim_journal_file: hfs_start_transaction returned %d\n", error);
		return error;
	}
	lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG | SFL_BITMAP, HFS_EXCLUSIVE_LOCK);
	
	oldBlockCount = hfsmp->jnl_size / hfsmp->blockSize;
	
	/* TODO: Allow the journal to change size based on the new volume size. */
	error = BlockAllocate(hfsmp, 1, oldBlockCount, oldBlockCount, true, true, &newStartBlock, &newBlockCount);
	if (error) {
		printf("hfs_reclaim_journal_file: BlockAllocate returned %d\n", error);
		goto fail;
	}
	if (newBlockCount != oldBlockCount) {
		printf("hfs_reclaim_journal_file: newBlockCount != oldBlockCount (%u, %u)\n", newBlockCount, oldBlockCount);
		goto free_fail;
	}
	
	error = BlockDeallocate(hfsmp, hfsmp->jnl_start, oldBlockCount);
	if (error) {
		printf("hfs_reclaim_journal_file: BlockDeallocate returned %d\n", error);
		goto free_fail;
	}

	/* Update the catalog record for .journal */
	error = cat_idlookup(hfsmp, hfsmp->hfs_jnlfileid, 1, &journal_desc, &journal_attr, &journal_fork);
	if (error) {
		printf("hfs_reclaim_journal_file: cat_idlookup returned %d\n", error);
		goto free_fail;
	}
	journal_fork.cf_size = newBlockCount * hfsmp->blockSize;
	journal_fork.cf_extents[0].startBlock = newStartBlock;
	journal_fork.cf_extents[0].blockCount = newBlockCount;
	journal_fork.cf_blocks = newBlockCount;
	error = cat_update(hfsmp, &journal_desc, &journal_attr, &journal_fork, NULL);
	cat_releasedesc(&journal_desc);  /* all done with cat descriptor */
	if (error) {
		printf("hfs_reclaim_journal_file: cat_update returned %d\n", error);
		goto free_fail;
	}
	callback_args.hfsmp = hfsmp;
	callback_args.context = context;
	callback_args.newStartBlock = newStartBlock;
	
	error = journal_relocate(hfsmp->jnl, (off_t)newStartBlock*hfsmp->blockSize,
		(off_t)newBlockCount*hfsmp->blockSize, 0,
		hfs_journal_relocate_callback, &callback_args);
	if (error) {
		/* NOTE: journal_relocate will mark the journal invalid. */
		printf("hfs_reclaim_journal_file: journal_relocate returned %d\n", error);
		goto fail;
	}
	hfsmp->jnl_start = newStartBlock;
	hfsmp->jnl_size = (off_t)newBlockCount * hfsmp->blockSize;

	hfs_systemfile_unlock(hfsmp, lockflags);
	error = hfs_end_transaction(hfsmp);
	if (error) {
		printf("hfs_reclaim_journal_file: hfs_end_transaction returned %d\n", error);
	}
	
	return error;

free_fail:
	(void) BlockDeallocate(hfsmp, newStartBlock, newBlockCount);
fail:
	hfs_systemfile_unlock(hfsmp, lockflags);
	(void) hfs_end_transaction(hfsmp);
	return error;
}


/*
 * Move the journal info block to a new location.  We have to make sure the
 * new copy of the journal info block gets to the media first, then change
 * the field in the volume header and the catalog record.
 */
static int
hfs_reclaim_journal_info_block(struct hfsmount *hfsmp, vfs_context_t context)
{
	int error;
	int lockflags;
	u_int32_t newBlock;
	u_int32_t blockCount;
	struct cat_desc jib_desc;
	struct cat_attr jib_attr;
	struct cat_fork jib_fork;
	buf_t old_bp, new_bp;
	
	error = hfs_start_transaction(hfsmp);
	if (error) {
		printf("hfs_reclaim_journal_info_block: hfs_start_transaction returned %d\n", error);
		return error;
	}
	lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG | SFL_BITMAP, HFS_EXCLUSIVE_LOCK);
	
	error = BlockAllocate(hfsmp, 1, 1, 1, true, true, &newBlock, &blockCount);
	if (error) {
		printf("hfs_reclaim_journal_info_block: BlockAllocate returned %d\n", error);
		goto fail;
	}
	if (blockCount != 1) {
		printf("hfs_reclaim_journal_info_block: blockCount != 1 (%u)\n", blockCount);
		goto free_fail;
	}
	error = BlockDeallocate(hfsmp, hfsmp->vcbJinfoBlock, 1);
	if (error) {
		printf("hfs_reclaim_journal_info_block: BlockDeallocate returned %d\n", error);
		goto free_fail;
	}
	
	/* Copy the old journal info block content to the new location */
	error = buf_meta_bread(hfsmp->hfs_devvp,
		hfsmp->vcbJinfoBlock * (hfsmp->blockSize/hfsmp->hfs_logical_block_size),
		hfsmp->blockSize, vfs_context_ucred(context), &old_bp);
	if (error) {
		printf("hfs_reclaim_journal_info_block: failed to read JIB (%d)\n", error);
		goto free_fail;
	}
	new_bp = buf_getblk(hfsmp->hfs_devvp,
		newBlock * (hfsmp->blockSize/hfsmp->hfs_logical_block_size),
		hfsmp->blockSize, 0, 0, BLK_META);
	bcopy((char*)buf_dataptr(old_bp), (char*)buf_dataptr(new_bp), hfsmp->blockSize);
	buf_brelse(old_bp);
	if (journal_uses_fua(hfsmp->jnl))
		buf_markfua(new_bp);
	error = buf_bwrite(new_bp);
	if (error) {
		printf("hfs_reclaim_journal_info_block: failed to write new JIB (%d)\n", error);
		goto free_fail;
	}
	if (!journal_uses_fua(hfsmp->jnl)) {
		error = VNOP_IOCTL(hfsmp->hfs_devvp, DKIOCSYNCHRONIZECACHE, NULL, FWRITE, context);
		if (error) {
			printf("hfs_reclaim_journal_info_block: DKIOCSYNCHRONIZECACHE failed (%d)\n", error);
			/* Don't fail the operation. */
		}
	}
	
	/* Update the catalog record for .journal_info_block */
	error = cat_idlookup(hfsmp, hfsmp->hfs_jnlinfoblkid, 1, &jib_desc, &jib_attr, &jib_fork);
	if (error) {
		printf("hfs_reclaim_journal_file: cat_idlookup returned %d\n", error);
		goto fail;
	}
	jib_fork.cf_size = hfsmp->blockSize;
	jib_fork.cf_extents[0].startBlock = newBlock;
	jib_fork.cf_extents[0].blockCount = 1;
	jib_fork.cf_blocks = 1;
	error = cat_update(hfsmp, &jib_desc, &jib_attr, &jib_fork, NULL);
	cat_releasedesc(&jib_desc);  /* all done with cat descriptor */
	if (error) {
		printf("hfs_reclaim_journal_info_block: cat_update returned %d\n", error);
		goto fail;
	}
	
	/* Update the pointer to the journal info block in the volume header. */
	hfsmp->vcbJinfoBlock = newBlock;
	error = hfs_flushvolumeheader(hfsmp, MNT_WAIT, HFS_ALTFLUSH);
	if (error) {
		printf("hfs_reclaim_journal_info_block: hfs_flushvolumeheader returned %d\n", error);
		goto fail;
	}
	hfs_systemfile_unlock(hfsmp, lockflags);
	error = hfs_end_transaction(hfsmp);
	if (error) {
		printf("hfs_reclaim_journal_info_block: hfs_end_transaction returned %d\n", error);
	}
	error = hfs_journal_flush(hfsmp);
	if (error) {
		printf("hfs_reclaim_journal_info_block: journal_flush returned %d\n", error);
	}
	return error;

free_fail:
	(void) BlockDeallocate(hfsmp, newBlock, blockCount);
fail:
	hfs_systemfile_unlock(hfsmp, lockflags);
	(void) hfs_end_transaction(hfsmp);
	return error;
}


/*
 * Reclaim space at the end of a file system.
 */
static int
hfs_reclaimspace(struct hfsmount *hfsmp, u_int32_t startblk, u_int32_t reclaimblks, vfs_context_t context)
{
	struct vnode *vp = NULL;
	FCB *fcb;
	struct BTreeIterator * iterator = NULL;
	struct FSBufferDescriptor btdata;
	struct HFSPlusCatalogFile filerec;
	u_int32_t  saved_next_allocation;
	cnid_t * cnidbufp;
	size_t cnidbufsize;
	int filecnt = 0;
	int maxfilecnt;
	u_int32_t block;
	u_int32_t datablks;
	u_int32_t rsrcblks;
	u_int32_t blkstomove = 0;
	int lockflags;
	int i;
	int error;
	int lastprogress = 0;
	Boolean system_file_moved = false;

	/* Relocate extents of the Allocation file if they're in the way. */
	error = hfs_reclaim_sys_file(hfsmp, hfsmp->hfs_allocation_vp, startblk, SFL_BITMAP, &system_file_moved, context);
	if (error) {
		printf("hfs_reclaimspace: reclaim allocation file returned %d\n", error);
		return error;
	}
	/* Relocate extents of the Extents B-tree if they're in the way. */
	error = hfs_reclaim_sys_file(hfsmp, hfsmp->hfs_extents_vp, startblk, SFL_EXTENTS, &system_file_moved, context);
	if (error) {
		printf("hfs_reclaimspace: reclaim extents b-tree returned %d\n", error);
		return error;
	}
	/* Relocate extents of the Catalog B-tree if they're in the way. */
	error = hfs_reclaim_sys_file(hfsmp, hfsmp->hfs_catalog_vp, startblk, SFL_CATALOG, &system_file_moved, context);
	if (error) {
		printf("hfs_reclaimspace: reclaim catalog b-tree returned %d\n", error);
		return error;
	}
	/* Relocate extents of the Attributes B-tree if they're in the way. */
	error = hfs_reclaim_sys_file(hfsmp, hfsmp->hfs_attribute_vp, startblk, SFL_ATTRIBUTE, &system_file_moved, context);
	if (error) {
		printf("hfs_reclaimspace: reclaim attribute b-tree returned %d\n", error);
		return error;
	}
	/* Relocate extents of the Startup File if there is one and they're in the way. */
	error = hfs_reclaim_sys_file(hfsmp, hfsmp->hfs_startup_vp, startblk, SFL_STARTUP, &system_file_moved, context);
	if (error) {
		printf("hfs_reclaimspace: reclaim startup file returned %d\n", error);
		return error;
	}
	
	/*
	 * We need to make sure the alternate volume header gets flushed if we moved
	 * any extents in the volume header.  But we need to do that before
	 * shrinking the size of the volume, or else the journal code will panic
	 * with an invalid (too large) block number.
	 *
	 * Note that system_file_moved will be set if ANY extent was moved, even
	 * if it was just an overflow extent.  In this case, the journal_flush isn't
	 * strictly required, but shouldn't hurt.
	 */
	if (system_file_moved)
		hfs_journal_flush(hfsmp);

	if (hfsmp->jnl_start + (hfsmp->jnl_size / hfsmp->blockSize) > startblk) {
		error = hfs_reclaim_journal_file(hfsmp, context);
		if (error) {
			printf("hfs_reclaimspace: hfs_reclaim_journal_file failed (%d)\n", error);
			return error;
		}
	}
	
	if (hfsmp->vcbJinfoBlock >= startblk) {
		error = hfs_reclaim_journal_info_block(hfsmp, context);
		if (error) {
			printf("hfs_reclaimspace: hfs_reclaim_journal_info_block failed (%d)\n", error);
			return error;
		}
	}
	
	/* For now move a maximum of 250,000 files. */
	maxfilecnt = MIN(hfsmp->hfs_filecount, 250000);
	maxfilecnt = MIN((u_int32_t)maxfilecnt, reclaimblks);
	cnidbufsize = maxfilecnt * sizeof(cnid_t);
	if (kmem_alloc(kernel_map, (vm_offset_t *)&cnidbufp, cnidbufsize)) {
		return (ENOMEM);
	}	
	if (kmem_alloc(kernel_map, (vm_offset_t *)&iterator, sizeof(*iterator))) {
		kmem_free(kernel_map, (vm_offset_t)cnidbufp, cnidbufsize);
		return (ENOMEM);
	}	

	saved_next_allocation = hfsmp->nextAllocation;
	HFS_UPDATE_NEXT_ALLOCATION(hfsmp, hfsmp->hfs_metazone_start);

	fcb = VTOF(hfsmp->hfs_catalog_vp);
	bzero(iterator, sizeof(*iterator));

	btdata.bufferAddress = &filerec;
	btdata.itemSize = sizeof(filerec);
	btdata.itemCount = 1;

	/* Keep the Catalog and extents files locked during iteration. */
	lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG | SFL_EXTENTS, HFS_SHARED_LOCK);

	error = BTIterateRecord(fcb, kBTreeFirstRecord, iterator, NULL, NULL);
	if (error) {
		goto end_iteration;
	}
	/*
	 * Iterate over all the catalog records looking for files
	 * that overlap into the space we're trying to free up.
	 */
	for (filecnt = 0; filecnt < maxfilecnt; ) {
		error = BTIterateRecord(fcb, kBTreeNextRecord, iterator, &btdata, NULL);
		if (error) {
			if (error == fsBTRecordNotFoundErr || error == fsBTEndOfIterationErr) {
				error = 0;				
			}
			break;
		}
		if (filerec.recordType != kHFSPlusFileRecord) {
			continue;
		}
		datablks = rsrcblks = 0;
		/* 
		 * Check if either fork overlaps target space.
		 */
		for (i = 0; i < kHFSPlusExtentDensity; ++i) {
			if (filerec.dataFork.extents[i].blockCount != 0) {
				datablks += filerec.dataFork.extents[i].blockCount;
				block = filerec.dataFork.extents[i].startBlock +
						filerec.dataFork.extents[i].blockCount;
				if (block >= startblk) {
					if ((filerec.fileID == hfsmp->hfs_jnlfileid) ||
						(filerec.fileID == hfsmp->hfs_jnlinfoblkid)) {
						printf("hfs_reclaimspace: cannot move active journal\n");
						error = EPERM;
						goto end_iteration;
					}
					cnidbufp[filecnt++] = filerec.fileID;
					blkstomove += filerec.dataFork.totalBlocks;
					break;
				}
			}
			if (filerec.resourceFork.extents[i].blockCount != 0) {
				rsrcblks += filerec.resourceFork.extents[i].blockCount;
				block = filerec.resourceFork.extents[i].startBlock +
						filerec.resourceFork.extents[i].blockCount;
				if (block >= startblk) {
					cnidbufp[filecnt++] = filerec.fileID;
					blkstomove += filerec.resourceFork.totalBlocks;
					break;
				}
			}
		}
		/*
		 * Check for any overflow extents that overlap.
		 */
		if (i == kHFSPlusExtentDensity) {
			if (filerec.dataFork.totalBlocks > datablks) {
				if (hfs_overlapped_overflow_extents(hfsmp, startblk, datablks, filerec.fileID, 0)) {
					cnidbufp[filecnt++] = filerec.fileID;
					blkstomove += filerec.dataFork.totalBlocks;
				}
			} else if (filerec.resourceFork.totalBlocks > rsrcblks) {
				if (hfs_overlapped_overflow_extents(hfsmp, startblk, rsrcblks, filerec.fileID, 1)) {
					cnidbufp[filecnt++] = filerec.fileID;
					blkstomove += filerec.resourceFork.totalBlocks;
				}
			}
		}
	}

end_iteration:
	if (filecnt == 0 && !system_file_moved) {
		printf("hfs_reclaimspace: no files moved\n");
		error = ENOSPC;
	}
	/* All done with catalog. */
	hfs_systemfile_unlock(hfsmp, lockflags);
	if (error || filecnt == 0)
		goto out;

	/*
	 * Double check space requirements to make sure
	 * there is enough space to relocate any files
	 * that reside in the reclaim area.
	 *
	 *                                          Blocks To Move --------------
	 *                                                            |    |    |
	 *                                                            V    V    V
	 * ------------------------------------------------------------------------
	 * |                                                        | /   ///  // |
	 * |                                                        | /   ///  // |
	 * |                                                        | /   ///  // |
	 * ------------------------------------------------------------------------
	 *
	 * <------------------- New Total Blocks ------------------><-- Reclaim -->
	 *
	 * <------------------------ Original Total Blocks ----------------------->
	 *
	 */
	if (blkstomove >= hfs_freeblks(hfsmp, 1)) {
		printf("hfs_truncatefs: insufficient space (need %u blocks; have %u blocks)\n", blkstomove, hfs_freeblks(hfsmp, 1));
		error = ENOSPC;
		goto out;
	}
	hfsmp->hfs_resize_filesmoved = 0;
	hfsmp->hfs_resize_totalfiles = filecnt;
	
	/* Now move any files that are in the way. */
	for (i = 0; i < filecnt; ++i) {
		struct vnode * rvp;
        struct cnode * cp;

		if (hfs_vget(hfsmp, cnidbufp[i], &vp, 0) != 0)
			continue;

        /* Relocating directory hard links is not supported, so we
         * punt (see radar 6217026). */
        cp = VTOC(vp);
        if ((cp->c_flag & C_HARDLINK) && vnode_isdir(vp)) {
            printf("hfs_reclaimspace: unable to relocate directory hard link %d\n", cp->c_cnid);
            error = EINVAL;
            goto out;
        }

		/* Relocate any data fork blocks. */
		if (VTOF(vp) && VTOF(vp)->ff_blocks > 0) {
			error = hfs_relocate(vp, hfsmp->hfs_metazone_end + 1, kauth_cred_get(), current_proc());
		}
		if (error) 
			break;

		/* Relocate any resource fork blocks. */
		if ((cp->c_blocks - (VTOF(vp) ? VTOF((vp))->ff_blocks : 0)) > 0) {
			error = hfs_vgetrsrc(hfsmp, vp, &rvp, TRUE);
			if (error)
				break;
			error = hfs_relocate(rvp, hfsmp->hfs_metazone_end + 1, kauth_cred_get(), current_proc());
			VTOC(rvp)->c_flag |= C_NEED_RVNODE_PUT;
			if (error)
				break;
		}
		hfs_unlock(cp);
		vnode_put(vp);
		vp = NULL;

		++hfsmp->hfs_resize_filesmoved;

		/* Report intermediate progress. */
		if (filecnt > 100) {
			int progress;

			progress = (i * 100) / filecnt;
			if (progress > (lastprogress + 9)) {
				printf("hfs_reclaimspace: %d%% done...\n", progress);
				lastprogress = progress;
			}
		}
	}
	if (vp) {
		hfs_unlock(VTOC(vp));
		vnode_put(vp);
		vp = NULL;
	}
	if (hfsmp->hfs_resize_filesmoved != 0) {
		printf("hfs_reclaimspace: relocated %d files on \"%s\"\n",
		       (int)hfsmp->hfs_resize_filesmoved, hfsmp->vcbVN);
	}
out:
	kmem_free(kernel_map, (vm_offset_t)iterator, sizeof(*iterator));
	kmem_free(kernel_map, (vm_offset_t)cnidbufp, cnidbufsize);

	/*
	 * Restore the roving allocation pointer on errors.
	 * (but only if we didn't move any files)
	 */
	if (error && hfsmp->hfs_resize_filesmoved == 0) {
		HFS_UPDATE_NEXT_ALLOCATION(hfsmp, saved_next_allocation);
	}
	return (error);
}


/*
 * Check if there are any overflow extents that overlap.
 */
static int
hfs_overlapped_overflow_extents(struct hfsmount *hfsmp, u_int32_t startblk, u_int32_t catblks, u_int32_t fileID, int rsrcfork)
{
	struct BTreeIterator * iterator = NULL;
	struct FSBufferDescriptor btdata;
	HFSPlusExtentRecord extrec;
	HFSPlusExtentKey *extkeyptr;
	FCB *fcb;
	u_int32_t block;
	u_int8_t forktype;
	int overlapped = 0;
	int i;
	int error;

	forktype = rsrcfork ? 0xFF : 0;
	if (kmem_alloc(kernel_map, (vm_offset_t *)&iterator, sizeof(*iterator))) {
		return (0);
	}	
	bzero(iterator, sizeof(*iterator));
	extkeyptr = (HFSPlusExtentKey *)&iterator->key;
	extkeyptr->keyLength = kHFSPlusExtentKeyMaximumLength;
	extkeyptr->forkType = forktype;
	extkeyptr->fileID = fileID;
	extkeyptr->startBlock = catblks;

	btdata.bufferAddress = &extrec;
	btdata.itemSize = sizeof(extrec);
	btdata.itemCount = 1;
	
	fcb = VTOF(hfsmp->hfs_extents_vp);

	error = BTSearchRecord(fcb, iterator, &btdata, NULL, iterator);
	while (error == 0) {
		/* Stop when we encounter a different file. */
		if ((extkeyptr->fileID != fileID) ||
		    (extkeyptr->forkType != forktype)) {
			break;
		}
		/* 
		 * Check if the file overlaps target space.
		 */
		for (i = 0; i < kHFSPlusExtentDensity; ++i) {
			if (extrec[i].blockCount == 0) {
				break;
			}
			block = extrec[i].startBlock + extrec[i].blockCount;
			if (block >= startblk) {
				overlapped = 1;
				break;
			}
		}
		/* Look for more records. */
		error = BTIterateRecord(fcb, kBTreeNextRecord, iterator, &btdata, NULL);
	}

	kmem_free(kernel_map, (vm_offset_t)iterator, sizeof(*iterator));
	return (overlapped);
}


/*
 * Calculate the progress of a file system resize operation.
 */
__private_extern__
int
hfs_resize_progress(struct hfsmount *hfsmp, u_int32_t *progress)
{
	if ((hfsmp->hfs_flags & HFS_RESIZE_IN_PROGRESS) == 0) {
		return (ENXIO);
	}

	if (hfsmp->hfs_resize_totalfiles > 0)
		*progress = (hfsmp->hfs_resize_filesmoved * 100) / hfsmp->hfs_resize_totalfiles;
	else
		*progress = 0;

	return (0);
}


/*
 * Creates a UUID from a unique "name" in the HFS UUID Name space.
 * See version 3 UUID.
 */
static void
hfs_getvoluuid(struct hfsmount *hfsmp, uuid_t result)
{
	MD5_CTX  md5c;
	uint8_t  rawUUID[8];

	((uint32_t *)rawUUID)[0] = hfsmp->vcbFndrInfo[6];
	((uint32_t *)rawUUID)[1] = hfsmp->vcbFndrInfo[7];

	MD5Init( &md5c );
	MD5Update( &md5c, HFS_UUID_NAMESPACE_ID, sizeof( uuid_t ) );
	MD5Update( &md5c, rawUUID, sizeof (rawUUID) );
	MD5Final( result, &md5c );

	result[6] = 0x30 | ( result[6] & 0x0F );
	result[8] = 0x80 | ( result[8] & 0x3F );
}

/*
 * Get file system attributes.
 */
static int
hfs_vfs_getattr(struct mount *mp, struct vfs_attr *fsap, __unused vfs_context_t context)
{
#define HFS_ATTR_CMN_VALIDMASK (ATTR_CMN_VALIDMASK & ~(ATTR_CMN_NAMEDATTRCOUNT | ATTR_CMN_NAMEDATTRLIST))
#define HFS_ATTR_FILE_VALIDMASK (ATTR_FILE_VALIDMASK & ~(ATTR_FILE_FILETYPE | ATTR_FILE_FORKCOUNT | ATTR_FILE_FORKLIST))

	ExtendedVCB *vcb = VFSTOVCB(mp);
	struct hfsmount *hfsmp = VFSTOHFS(mp);
	u_int32_t freeCNIDs;

	freeCNIDs = (u_int32_t)0xFFFFFFFF - (u_int32_t)hfsmp->vcbNxtCNID;

	VFSATTR_RETURN(fsap, f_objcount, (u_int64_t)hfsmp->vcbFilCnt + (u_int64_t)hfsmp->vcbDirCnt);
	VFSATTR_RETURN(fsap, f_filecount, (u_int64_t)hfsmp->vcbFilCnt);
	VFSATTR_RETURN(fsap, f_dircount, (u_int64_t)hfsmp->vcbDirCnt);
	VFSATTR_RETURN(fsap, f_maxobjcount, (u_int64_t)0xFFFFFFFF);
	VFSATTR_RETURN(fsap, f_iosize, (size_t)cluster_max_io_size(mp, 0));
	VFSATTR_RETURN(fsap, f_blocks, (u_int64_t)hfsmp->totalBlocks);
	VFSATTR_RETURN(fsap, f_bfree, (u_int64_t)hfs_freeblks(hfsmp, 0));
	VFSATTR_RETURN(fsap, f_bavail, (u_int64_t)hfs_freeblks(hfsmp, 1));
	VFSATTR_RETURN(fsap, f_bsize, (u_int32_t)vcb->blockSize);
	/* XXX needs clarification */
	VFSATTR_RETURN(fsap, f_bused, hfsmp->totalBlocks - hfs_freeblks(hfsmp, 1));
	/* Maximum files is constrained by total blocks. */
	VFSATTR_RETURN(fsap, f_files, (u_int64_t)(hfsmp->totalBlocks - 2));
	VFSATTR_RETURN(fsap, f_ffree, MIN((u_int64_t)freeCNIDs, (u_int64_t)hfs_freeblks(hfsmp, 1)));

	fsap->f_fsid.val[0] = hfsmp->hfs_raw_dev;
	fsap->f_fsid.val[1] = vfs_typenum(mp);
	VFSATTR_SET_SUPPORTED(fsap, f_fsid);

	VFSATTR_RETURN(fsap, f_signature, vcb->vcbSigWord);
	VFSATTR_RETURN(fsap, f_carbon_fsid, 0);

	if (VFSATTR_IS_ACTIVE(fsap, f_capabilities)) {
		vol_capabilities_attr_t *cap;
	
		cap = &fsap->f_capabilities;

		if (hfsmp->hfs_flags & HFS_STANDARD) {
			cap->capabilities[VOL_CAPABILITIES_FORMAT] =
				VOL_CAP_FMT_PERSISTENTOBJECTIDS |
				VOL_CAP_FMT_CASE_PRESERVING |
				VOL_CAP_FMT_FAST_STATFS |
				VOL_CAP_FMT_HIDDEN_FILES |
				VOL_CAP_FMT_PATH_FROM_ID;
		} else {
			cap->capabilities[VOL_CAPABILITIES_FORMAT] =
				VOL_CAP_FMT_PERSISTENTOBJECTIDS |
				VOL_CAP_FMT_SYMBOLICLINKS |
				VOL_CAP_FMT_HARDLINKS |
				VOL_CAP_FMT_JOURNAL |
				VOL_CAP_FMT_ZERO_RUNS |
				(hfsmp->jnl ? VOL_CAP_FMT_JOURNAL_ACTIVE : 0) |
				(hfsmp->hfs_flags & HFS_CASE_SENSITIVE ? VOL_CAP_FMT_CASE_SENSITIVE : 0) |
				VOL_CAP_FMT_CASE_PRESERVING |
				VOL_CAP_FMT_FAST_STATFS | 
				VOL_CAP_FMT_2TB_FILESIZE |
				VOL_CAP_FMT_HIDDEN_FILES |
#if HFS_COMPRESSION
				VOL_CAP_FMT_PATH_FROM_ID |
				VOL_CAP_FMT_DECMPFS_COMPRESSION;
#else
				VOL_CAP_FMT_PATH_FROM_ID;
#endif
		}
		cap->capabilities[VOL_CAPABILITIES_INTERFACES] =
			VOL_CAP_INT_SEARCHFS |
			VOL_CAP_INT_ATTRLIST |
			VOL_CAP_INT_NFSEXPORT |
			VOL_CAP_INT_READDIRATTR |
			VOL_CAP_INT_EXCHANGEDATA |
			VOL_CAP_INT_ALLOCATE |
			VOL_CAP_INT_VOL_RENAME |
			VOL_CAP_INT_ADVLOCK |
			VOL_CAP_INT_FLOCK |
#if NAMEDSTREAMS
			VOL_CAP_INT_EXTENDED_ATTR |
			VOL_CAP_INT_NAMEDSTREAMS;
#else
			VOL_CAP_INT_EXTENDED_ATTR;
#endif
		cap->capabilities[VOL_CAPABILITIES_RESERVED1] = 0;
		cap->capabilities[VOL_CAPABILITIES_RESERVED2] = 0;

		cap->valid[VOL_CAPABILITIES_FORMAT] =
			VOL_CAP_FMT_PERSISTENTOBJECTIDS |
			VOL_CAP_FMT_SYMBOLICLINKS |
			VOL_CAP_FMT_HARDLINKS |
			VOL_CAP_FMT_JOURNAL |
			VOL_CAP_FMT_JOURNAL_ACTIVE |
			VOL_CAP_FMT_NO_ROOT_TIMES |
			VOL_CAP_FMT_SPARSE_FILES |
			VOL_CAP_FMT_ZERO_RUNS |
			VOL_CAP_FMT_CASE_SENSITIVE |
			VOL_CAP_FMT_CASE_PRESERVING |
			VOL_CAP_FMT_FAST_STATFS |
			VOL_CAP_FMT_2TB_FILESIZE |
			VOL_CAP_FMT_OPENDENYMODES |
			VOL_CAP_FMT_HIDDEN_FILES |
#if HFS_COMPRESSION
			VOL_CAP_FMT_PATH_FROM_ID |
			VOL_CAP_FMT_DECMPFS_COMPRESSION;
#else
			VOL_CAP_FMT_PATH_FROM_ID;
#endif
		cap->valid[VOL_CAPABILITIES_INTERFACES] =
			VOL_CAP_INT_SEARCHFS |
			VOL_CAP_INT_ATTRLIST |
			VOL_CAP_INT_NFSEXPORT |
			VOL_CAP_INT_READDIRATTR |
			VOL_CAP_INT_EXCHANGEDATA |
			VOL_CAP_INT_COPYFILE |
			VOL_CAP_INT_ALLOCATE |
			VOL_CAP_INT_VOL_RENAME |
			VOL_CAP_INT_ADVLOCK |
			VOL_CAP_INT_FLOCK |
			VOL_CAP_INT_MANLOCK |
#if NAMEDSTREAMS
			VOL_CAP_INT_EXTENDED_ATTR |
			VOL_CAP_INT_NAMEDSTREAMS;
#else
			VOL_CAP_INT_EXTENDED_ATTR;
#endif
		cap->valid[VOL_CAPABILITIES_RESERVED1] = 0;
		cap->valid[VOL_CAPABILITIES_RESERVED2] = 0;
		VFSATTR_SET_SUPPORTED(fsap, f_capabilities);
	}
	if (VFSATTR_IS_ACTIVE(fsap, f_attributes)) {
		vol_attributes_attr_t *attrp = &fsap->f_attributes;

        	attrp->validattr.commonattr = HFS_ATTR_CMN_VALIDMASK;
        	attrp->validattr.volattr = ATTR_VOL_VALIDMASK & ~ATTR_VOL_INFO;
        	attrp->validattr.dirattr = ATTR_DIR_VALIDMASK;
        	attrp->validattr.fileattr = HFS_ATTR_FILE_VALIDMASK;
        	attrp->validattr.forkattr = 0;

        	attrp->nativeattr.commonattr = HFS_ATTR_CMN_VALIDMASK;
        	attrp->nativeattr.volattr = ATTR_VOL_VALIDMASK & ~ATTR_VOL_INFO;
        	attrp->nativeattr.dirattr = ATTR_DIR_VALIDMASK;
        	attrp->nativeattr.fileattr = HFS_ATTR_FILE_VALIDMASK;
        	attrp->nativeattr.forkattr = 0;
		VFSATTR_SET_SUPPORTED(fsap, f_attributes);
	}	
	fsap->f_create_time.tv_sec = hfsmp->vcbCrDate;
	fsap->f_create_time.tv_nsec = 0;
	VFSATTR_SET_SUPPORTED(fsap, f_create_time);
	fsap->f_modify_time.tv_sec = hfsmp->vcbLsMod;
	fsap->f_modify_time.tv_nsec = 0;
	VFSATTR_SET_SUPPORTED(fsap, f_modify_time);

	fsap->f_backup_time.tv_sec = hfsmp->vcbVolBkUp;
	fsap->f_backup_time.tv_nsec = 0;
	VFSATTR_SET_SUPPORTED(fsap, f_backup_time);
	if (VFSATTR_IS_ACTIVE(fsap, f_fssubtype)) {
		u_int16_t subtype = 0;

		/*
		 * Subtypes (flavors) for HFS
		 *   0:   Mac OS Extended
		 *   1:   Mac OS Extended (Journaled) 
		 *   2:   Mac OS Extended (Case Sensitive) 
		 *   3:   Mac OS Extended (Case Sensitive, Journaled) 
		 *   4 - 127:   Reserved
		 * 128:   Mac OS Standard
		 * 
		 */
		if (hfsmp->hfs_flags & HFS_STANDARD) {
			subtype = HFS_SUBTYPE_STANDARDHFS;
		} else /* HFS Plus */ {
			if (hfsmp->jnl)
				subtype |= HFS_SUBTYPE_JOURNALED;
			if (hfsmp->hfs_flags & HFS_CASE_SENSITIVE)
				subtype |= HFS_SUBTYPE_CASESENSITIVE;
		}
		fsap->f_fssubtype = subtype;
		VFSATTR_SET_SUPPORTED(fsap, f_fssubtype);
	}

	if (VFSATTR_IS_ACTIVE(fsap, f_vol_name)) {
		strlcpy(fsap->f_vol_name, (char *) hfsmp->vcbVN, MAXPATHLEN);
		VFSATTR_SET_SUPPORTED(fsap, f_vol_name);
	}
	if (VFSATTR_IS_ACTIVE(fsap, f_uuid)) {
		hfs_getvoluuid(hfsmp, fsap->f_uuid);
		VFSATTR_SET_SUPPORTED(fsap, f_uuid);
	}
	return (0);
}

/*
 * Perform a volume rename.  Requires the FS' root vp.
 */
static int
hfs_rename_volume(struct vnode *vp, const char *name, proc_t p)
{
	ExtendedVCB *vcb = VTOVCB(vp);
	struct cnode *cp = VTOC(vp);
	struct hfsmount *hfsmp = VTOHFS(vp);
	struct cat_desc to_desc;
	struct cat_desc todir_desc;
	struct cat_desc new_desc;
	cat_cookie_t cookie;
	int lockflags;
	int error = 0;

	/*
	 * Ignore attempts to rename a volume to a zero-length name.
	 */
	if (name[0] == 0)
		return(0);

	bzero(&to_desc, sizeof(to_desc));
	bzero(&todir_desc, sizeof(todir_desc));
	bzero(&new_desc, sizeof(new_desc));
	bzero(&cookie, sizeof(cookie));

	todir_desc.cd_parentcnid = kHFSRootParentID;
	todir_desc.cd_cnid = kHFSRootFolderID;
	todir_desc.cd_flags = CD_ISDIR;

	to_desc.cd_nameptr = (const u_int8_t *)name;
	to_desc.cd_namelen = strlen(name);
	to_desc.cd_parentcnid = kHFSRootParentID;
	to_desc.cd_cnid = cp->c_cnid;
	to_desc.cd_flags = CD_ISDIR;

	if ((error = hfs_lock(cp, HFS_EXCLUSIVE_LOCK)) == 0) {
		if ((error = hfs_start_transaction(hfsmp)) == 0) {
			if ((error = cat_preflight(hfsmp, CAT_RENAME, &cookie, p)) == 0) {
				lockflags = hfs_systemfile_lock(hfsmp, SFL_CATALOG, HFS_EXCLUSIVE_LOCK);

				error = cat_rename(hfsmp, &cp->c_desc, &todir_desc, &to_desc, &new_desc);

				/*
				 * If successful, update the name in the VCB, ensure it's terminated.
				 */
				if (!error) {
					strlcpy((char *)vcb->vcbVN, name, sizeof(vcb->vcbVN));
				}

				hfs_systemfile_unlock(hfsmp, lockflags);
				cat_postflight(hfsmp, &cookie, p);
			
				if (error)
					MarkVCBDirty(vcb);
				(void) hfs_flushvolumeheader(hfsmp, MNT_WAIT, 0);
			}
			hfs_end_transaction(hfsmp);
		}			
		if (!error) {
			/* Release old allocated name buffer */
			if (cp->c_desc.cd_flags & CD_HASBUF) {
				const char *tmp_name = (const char *)cp->c_desc.cd_nameptr;
		
				cp->c_desc.cd_nameptr = 0;
				cp->c_desc.cd_namelen = 0;
				cp->c_desc.cd_flags &= ~CD_HASBUF;
				vfs_removename(tmp_name);
			}			
			/* Update cnode's catalog descriptor */
			replace_desc(cp, &new_desc);
			vcb->volumeNameEncodingHint = new_desc.cd_encoding;
			cp->c_touch_chgtime = TRUE;
		}

		hfs_unlock(cp);
	}
	
	return(error);
}

/*
 * Get file system attributes.
 */
static int
hfs_vfs_setattr(struct mount *mp, struct vfs_attr *fsap, __unused vfs_context_t context)
{
	kauth_cred_t cred = vfs_context_ucred(context);
	int error = 0;

	/*
	 * Must be superuser or owner of filesystem to change volume attributes
	 */
	if (!kauth_cred_issuser(cred) && (kauth_cred_getuid(cred) != vfs_statfs(mp)->f_owner))
		return(EACCES);

	if (VFSATTR_IS_ACTIVE(fsap, f_vol_name)) {
		vnode_t root_vp;
		
		error = hfs_vfs_root(mp, &root_vp, context);
		if (error)
			goto out;

		error = hfs_rename_volume(root_vp, fsap->f_vol_name, vfs_context_proc(context));
		(void) vnode_put(root_vp);
		if (error)
			goto out;

		VFSATTR_SET_SUPPORTED(fsap, f_vol_name);
	}

out:
	return error;
}

/* If a runtime corruption is detected, set the volume inconsistent 
 * bit in the volume attributes.  The volume inconsistent bit is a persistent
 * bit which represents that the volume is corrupt and needs repair.  
 * The volume inconsistent bit can be set from the kernel when it detects
 * runtime corruption or from file system repair utilities like fsck_hfs when
 * a repair operation fails.  The bit should be cleared only from file system 
 * verify/repair utility like fsck_hfs when a verify/repair succeeds.
 */
void hfs_mark_volume_inconsistent(struct hfsmount *hfsmp)
{
	HFS_MOUNT_LOCK(hfsmp, TRUE);	
	if ((hfsmp->vcbAtrb & kHFSVolumeInconsistentMask) == 0) {
		hfsmp->vcbAtrb |= kHFSVolumeInconsistentMask;
		MarkVCBDirty(hfsmp);
	}
	if ((hfsmp->hfs_flags & HFS_READ_ONLY)==0) {	
		/* Log information to ASL log */
		fslog_fs_corrupt(hfsmp->hfs_mp);
		printf("hfs: Runtime corruption detected on %s, fsck will be forced on next mount.\n", hfsmp->vcbVN);
	}
	HFS_MOUNT_UNLOCK(hfsmp, TRUE);
}

/* Replay the journal on the device node provided.  Returns zero if 
 * journal replay succeeded or no journal was supposed to be replayed.
 */
static int hfs_journal_replay(vnode_t devvp, vfs_context_t context)
{
	int retval = 0;
	struct mount *mp = NULL;
	struct hfs_mount_args *args = NULL;

	/* Replay allowed only on raw devices */
	if (!vnode_ischr(devvp)) {
		retval = EINVAL;
		goto out;
	}

	/* Create dummy mount structures */
	MALLOC(mp, struct mount *, sizeof(struct mount), M_TEMP, M_WAITOK);
	if (mp == NULL) {
		retval = ENOMEM;
		goto out;
	}
	bzero(mp, sizeof(struct mount));
	mount_lock_init(mp);

	MALLOC(args, struct hfs_mount_args *, sizeof(struct hfs_mount_args), M_TEMP, M_WAITOK);
	if (args == NULL) {
		retval = ENOMEM;
		goto out;
	}
	bzero(args, sizeof(struct hfs_mount_args));

	retval = hfs_mountfs(devvp, mp, args, 1, context);
	buf_flushdirtyblks(devvp, MNT_WAIT, 0, "hfs_journal_replay");

out:
	if (mp) {
		mount_lock_destroy(mp);
		FREE(mp, M_TEMP);
	}
	if (args) {
		FREE(args, M_TEMP);
	}
	return retval;
}

/*
 * hfs vfs operations.
 */
struct vfsops hfs_vfsops = {
	hfs_mount,
	hfs_start,
	hfs_unmount,
	hfs_vfs_root,
	hfs_quotactl,
	hfs_vfs_getattr, 	/* was hfs_statfs */
	hfs_sync,
	hfs_vfs_vget,
	hfs_fhtovp,
	hfs_vptofh,
	hfs_init,
	hfs_sysctl,
	hfs_vfs_setattr,
	{NULL}
};