summaryrefslogtreecommitdiffstats
path: root/source3/modules/vfs_gpfs.c
blob: a8b4e38ff88c00c61285258f92c4f4a67de152a2 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
1022
1023
1024
1025
1026
1027
1028
1029
1030
1031
1032
1033
1034
1035
1036
1037
1038
1039
1040
1041
1042
1043
1044
1045
1046
1047
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
1060
1061
1062
1063
1064
1065
1066
1067
1068
1069
1070
1071
1072
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
1092
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
1103
1104
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
1120
1121
1122
1123
1124
1125
1126
1127
1128
1129
1130
1131
1132
1133
1134
1135
1136
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
1161
1162
1163
1164
1165
1166
1167
1168
1169
1170
1171
1172
1173
1174
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
1216
1217
1218
1219
1220
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
1247
1248
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
1259
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
1310
1311
1312
1313
1314
1315
1316
1317
1318
1319
1320
1321
1322
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
1341
1342
1343
1344
1345
1346
1347
1348
1349
1350
1351
1352
1353
1354
1355
1356
1357
1358
1359
1360
1361
1362
1363
1364
1365
1366
1367
1368
1369
1370
1371
1372
1373
1374
1375
1376
1377
1378
1379
1380
1381
1382
1383
1384
1385
1386
1387
1388
1389
1390
1391
1392
1393
1394
1395
1396
1397
1398
1399
1400
1401
1402
1403
1404
1405
1406
1407
1408
1409
1410
1411
1412
1413
1414
1415
1416
1417
1418
1419
1420
1421
1422
1423
1424
1425
1426
1427
1428
1429
1430
1431
1432
1433
1434
1435
1436
1437
1438
1439
1440
1441
1442
1443
1444
1445
1446
1447
1448
1449
1450
1451
1452
1453
1454
1455
1456
1457
1458
1459
1460
1461
1462
1463
1464
1465
1466
1467
1468
1469
1470
1471
1472
1473
1474
1475
1476
1477
1478
1479
1480
1481
1482
1483
1484
1485
1486
1487
1488
1489
1490
1491
1492
1493
1494
1495
1496
1497
1498
1499
1500
1501
1502
1503
1504
1505
1506
1507
1508
1509
1510
1511
1512
1513
1514
1515
1516
1517
1518
1519
1520
1521
1522
1523
1524
1525
1526
1527
1528
1529
1530
1531
1532
1533
1534
1535
1536
1537
1538
1539
1540
1541
1542
1543
1544
1545
1546
1547
1548
1549
1550
1551
1552
1553
1554
1555
1556
1557
1558
1559
1560
1561
1562
1563
1564
1565
1566
1567
1568
1569
1570
1571
1572
1573
1574
1575
1576
1577
1578
1579
1580
1581
1582
1583
1584
1585
1586
1587
1588
1589
1590
1591
1592
1593
1594
1595
1596
1597
1598
1599
1600
1601
1602
1603
1604
1605
1606
1607
1608
1609
1610
1611
1612
1613
1614
1615
1616
1617
1618
1619
1620
1621
1622
1623
1624
1625
1626
1627
1628
1629
1630
1631
1632
1633
1634
1635
1636
1637
1638
1639
1640
1641
1642
1643
1644
1645
1646
1647
1648
1649
1650
1651
1652
1653
1654
1655
1656
1657
1658
1659
1660
1661
1662
1663
1664
1665
1666
1667
1668
1669
1670
1671
1672
1673
1674
1675
1676
1677
1678
1679
1680
1681
1682
1683
1684
1685
1686
1687
1688
1689
1690
1691
1692
1693
1694
1695
1696
1697
1698
1699
1700
1701
1702
1703
1704
1705
1706
1707
1708
1709
1710
1711
1712
1713
1714
1715
1716
1717
1718
1719
1720
1721
1722
1723
1724
1725
1726
1727
1728
1729
1730
1731
1732
1733
1734
1735
1736
1737
1738
1739
1740
1741
1742
1743
1744
1745
1746
1747
1748
1749
1750
1751
1752
1753
1754
1755
1756
1757
1758
1759
1760
1761
1762
1763
1764
1765
1766
1767
1768
1769
1770
1771
1772
1773
1774
1775
1776
1777
1778
1779
1780
1781
1782
1783
1784
1785
1786
1787
1788
1789
1790
1791
1792
1793
1794
1795
1796
1797
1798
1799
1800
1801
1802
1803
1804
1805
1806
1807
1808
1809
1810
1811
1812
1813
1814
1815
1816
1817
1818
1819
1820
1821
1822
1823
1824
1825
1826
1827
1828
1829
1830
1831
1832
1833
1834
1835
1836
1837
1838
1839
1840
1841
1842
1843
1844
1845
1846
1847
1848
1849
1850
1851
1852
1853
1854
1855
1856
1857
1858
1859
1860
1861
1862
1863
1864
1865
1866
1867
1868
1869
1870
1871
1872
1873
1874
1875
1876
1877
1878
1879
1880
1881
1882
1883
1884
1885
1886
1887
1888
1889
1890
1891
1892
1893
1894
1895
1896
1897
1898
1899
1900
1901
1902
1903
1904
1905
1906
1907
1908
1909
1910
1911
1912
1913
1914
1915
1916
1917
1918
1919
1920
1921
1922
1923
1924
1925
1926
1927
1928
1929
1930
1931
1932
1933
1934
1935
1936
1937
1938
1939
1940
1941
1942
1943
1944
1945
1946
1947
1948
1949
1950
1951
1952
1953
1954
1955
1956
1957
1958
1959
1960
1961
1962
1963
1964
1965
1966
1967
1968
1969
1970
1971
1972
1973
1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987
1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
2015
2016
2017
2018
2019
2020
2021
2022
2023
2024
2025
2026
2027
2028
2029
2030
2031
2032
2033
2034
2035
2036
2037
2038
2039
2040
2041
2042
2043
2044
2045
2046
2047
2048
2049
2050
2051
2052
2053
2054
2055
2056
2057
2058
2059
2060
2061
2062
2063
2064
2065
2066
2067
2068
2069
2070
2071
2072
2073
2074
2075
2076
2077
2078
2079
2080
2081
2082
2083
2084
2085
2086
2087
2088
2089
2090
2091
2092
2093
2094
2095
2096
2097
2098
2099
2100
2101
2102
2103
2104
2105
2106
2107
2108
2109
2110
2111
2112
2113
2114
2115
2116
2117
2118
2119
2120
2121
2122
2123
2124
2125
2126
2127
2128
2129
2130
2131
2132
2133
2134
2135
2136
2137
2138
2139
2140
2141
2142
2143
2144
2145
2146
2147
2148
2149
2150
2151
2152
2153
2154
2155
2156
2157
2158
2159
2160
2161
2162
2163
2164
2165
2166
2167
2168
2169
2170
2171
2172
2173
2174
2175
2176
2177
2178
2179
2180
2181
2182
2183
2184
2185
2186
2187
2188
2189
2190
2191
2192
2193
2194
2195
2196
2197
2198
2199
2200
2201
2202
2203
2204
2205
2206
2207
2208
2209
2210
2211
2212
2213
2214
2215
2216
2217
2218
2219
2220
2221
2222
2223
2224
2225
2226
2227
2228
2229
2230
2231
2232
2233
2234
2235
2236
2237
2238
2239
2240
2241
2242
2243
2244
2245
2246
2247
2248
2249
2250
2251
2252
2253
2254
2255
2256
2257
2258
2259
2260
2261
2262
2263
2264
2265
2266
2267
2268
2269
2270
2271
2272
2273
2274
2275
2276
2277
2278
2279
2280
2281
2282
2283
2284
2285
2286
2287
2288
2289
2290
2291
2292
2293
2294
2295
2296
2297
2298
2299
2300
2301
2302
2303
2304
2305
2306
2307
2308
2309
2310
2311
2312
2313
2314
2315
2316
2317
2318
2319
2320
2321
2322
2323
2324
2325
2326
2327
2328
2329
2330
2331
2332
2333
2334
2335
2336
2337
2338
2339
2340
2341
2342
2343
2344
2345
2346
2347
2348
2349
2350
2351
2352
2353
2354
2355
2356
2357
2358
2359
2360
2361
2362
2363
2364
2365
2366
2367
2368
2369
2370
2371
2372
2373
2374
2375
2376
2377
2378
2379
2380
2381
2382
2383
2384
2385
2386
2387
2388
2389
2390
2391
2392
2393
2394
2395
2396
2397
2398
2399
2400
2401
2402
2403
2404
2405
2406
2407
2408
2409
2410
2411
2412
2413
2414
2415
2416
2417
2418
2419
2420
2421
2422
2423
2424
2425
2426
2427
2428
2429
2430
2431
2432
2433
2434
2435
2436
2437
2438
2439
2440
2441
2442
2443
2444
2445
2446
2447
2448
2449
2450
2451
2452
2453
2454
2455
2456
2457
2458
2459
2460
2461
2462
2463
2464
2465
2466
2467
2468
2469
2470
2471
2472
2473
2474
2475
2476
2477
2478
2479
2480
2481
2482
2483
2484
2485
2486
2487
2488
2489
2490
2491
2492
2493
2494
2495
2496
2497
2498
2499
2500
2501
2502
2503
2504
2505
2506
2507
2508
2509
2510
2511
2512
2513
2514
2515
2516
2517
2518
2519
2520
2521
2522
2523
2524
2525
2526
2527
2528
2529
2530
2531
2532
2533
2534
2535
2536
2537
2538
2539
2540
2541
2542
2543
2544
2545
2546
/*
 *  Unix SMB/CIFS implementation.
 *  Samba VFS module for GPFS filesystem
 *  Copyright (C) Christian Ambach <cambach1@de.ibm.com> 2006
 *  Copyright (C) Christof Schmitt 2015
 *  Major code contributions by Chetan Shringarpure <chetan.sh@in.ibm.com>
 *                           and Gomati Mohanan <gomati.mohanan@in.ibm.com>
 *
 *  This program is free software; you can redistribute it and/or modify
 *  it under the terms of the GNU General Public License as published by
 *  the Free Software Foundation; either version 3 of the License, or
 *  (at your option) any later version.
 *
 *  This program is distributed in the hope that it will be useful,
 *  but WITHOUT ANY WARRANTY; without even the implied warranty of
 *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
 *  GNU General Public License for more details.
 *
 *  You should have received a copy of the GNU General Public License
 *  along with this program; if not, see <http://www.gnu.org/licenses/>.
 */

#include "includes.h"
#include "smbd/smbd.h"
#include "include/smbprofile.h"
#include "modules/non_posix_acls.h"
#include "libcli/security/security.h"
#include "nfs4_acls.h"
#include "system/filesys.h"
#include "auth.h"
#include "lib/util/tevent_unix.h"
#include "lib/util/gpfswrap.h"

#include <gnutls/gnutls.h>
#include <gnutls/crypto.h>
#include "lib/crypto/gnutls_helpers.h"

#undef DBGC_CLASS
#define DBGC_CLASS DBGC_VFS

#ifndef GPFS_GETACL_NATIVE
#define GPFS_GETACL_NATIVE 0x00000004
#endif

struct gpfs_config_data {
	struct smbacl4_vfs_params nfs4_params;
	bool sharemodes;
	bool leases;
	bool hsm;
	bool syncio;
	bool winattr;
	bool ftruncate;
	bool getrealfilename;
	bool dfreequota;
	bool acl;
	bool settimes;
	bool recalls;
	struct {
		bool gpfs_fstat_x;
	} pathref_ok;
};

struct gpfs_fsp_extension {
	bool offline;
};

static inline unsigned int gpfs_acl_flags(gpfs_acl_t *gacl)
{
	if (gacl->acl_level == GPFS_ACL_LEVEL_V4FLAGS) {
		return gacl->v4Level1.acl_flags;
	}
	return 0;
}

static inline gpfs_ace_v4_t *gpfs_ace_ptr(gpfs_acl_t *gacl, unsigned int i)
{
	if (gacl->acl_level == GPFS_ACL_LEVEL_V4FLAGS) {
		return &gacl->v4Level1.ace_v4[i];
	}
	return &gacl->ace_v4[i];
}

static unsigned int vfs_gpfs_access_mask_to_allow(uint32_t access_mask)
{
	unsigned int allow = GPFS_SHARE_NONE;

	if (access_mask & (FILE_WRITE_DATA|FILE_APPEND_DATA)) {
		allow |= GPFS_SHARE_WRITE;
	}
	if (access_mask & (FILE_READ_DATA|FILE_EXECUTE)) {
		allow |= GPFS_SHARE_READ;
	}

	return allow;
}

static unsigned int vfs_gpfs_share_access_to_deny(uint32_t share_access)
{
	unsigned int deny = GPFS_DENY_NONE;

	if (!(share_access & FILE_SHARE_WRITE)) {
		deny |= GPFS_DENY_WRITE;
	}
	if (!(share_access & FILE_SHARE_READ)) {
		deny |= GPFS_DENY_READ;
	}

	/*
	 * GPFS_DENY_DELETE can only be set together with either
	 * GPFS_DENY_WRITE or GPFS_DENY_READ.
	 */
	if ((deny & (GPFS_DENY_WRITE|GPFS_DENY_READ)) &&
	    !(share_access & FILE_SHARE_DELETE)) {
		deny |= GPFS_DENY_DELETE;
	}

	return deny;
}

static int set_gpfs_sharemode(files_struct *fsp, uint32_t access_mask,
			      uint32_t share_access)
{
	unsigned int allow = GPFS_SHARE_NONE;
	unsigned int deny = GPFS_DENY_NONE;
	int result;

	if (access_mask == 0) {
		DBG_DEBUG("Clearing file system share mode.\n");
	} else {
		allow = vfs_gpfs_access_mask_to_allow(access_mask);
		deny = vfs_gpfs_share_access_to_deny(share_access);
	}
	DBG_DEBUG("access_mask=0x%x, allow=0x%x, share_access=0x%x, "
		  "deny=0x%x\n", access_mask, allow, share_access, deny);

	result = gpfswrap_set_share(fsp_get_io_fd(fsp), allow, deny);
	if (result == 0) {
		return 0;
	}

	if (errno == EACCES) {
		DBG_NOTICE("GPFS share mode denied for %s/%s.\n",
			   fsp->conn->connectpath,
			   fsp->fsp_name->base_name);
	} else if (errno == EPERM) {
		DBG_ERR("Samba requested GPFS sharemode for %s/%s, but the "
			"GPFS file system is not configured accordingly. "
			"Configure file system with mmchfs -D nfs4 or "
			"set gpfs:sharemodes=no in Samba.\n",
			fsp->conn->connectpath,
			fsp->fsp_name->base_name);
	} else {
		DBG_ERR("gpfs_set_share failed: %s\n", strerror(errno));
	}

	return result;
}

static int vfs_gpfs_filesystem_sharemode(vfs_handle_struct *handle,
					 files_struct *fsp,
					 uint32_t share_access,
					 uint32_t access_mask)
{

	struct gpfs_config_data *config;
	int ret = 0;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if(!config->sharemodes) {
		return 0;
	}

	/*
	 * A named stream fsp will have the basefile open in the fsp
	 * fd, so lacking a distinct fd for the stream we have to skip
	 * set_gpfs_sharemode for stream.
	 */
	if (fsp_is_alternate_stream(fsp)) {
		DBG_NOTICE("Not requesting GPFS sharemode on stream: %s/%s\n",
			   fsp->conn->connectpath,
			   fsp_str_dbg(fsp));
		return 0;
	}

	ret = set_gpfs_sharemode(fsp, access_mask, share_access);

	return ret;
}

static int vfs_gpfs_close(vfs_handle_struct *handle, files_struct *fsp)
{

	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if (config->sharemodes &&
	    (fsp->fsp_flags.kernel_share_modes_taken))
	{
		/*
		 * Always clear GPFS sharemode in case the actual
		 * close gets deferred due to outstanding POSIX locks
		 * (see fd_close_posix)
		 */
		int ret = gpfswrap_set_share(fsp_get_io_fd(fsp), 0, 0);
		if (ret != 0) {
			DBG_ERR("Clearing GPFS sharemode on close failed for "
				" %s/%s: %s\n",
				fsp->conn->connectpath,
				fsp->fsp_name->base_name,
				strerror(errno));
		}
	}

	return SMB_VFS_NEXT_CLOSE(handle, fsp);
}

#ifdef HAVE_KERNEL_OPLOCKS_LINUX
static int lease_type_to_gpfs(int leasetype)
{
	if (leasetype == F_RDLCK) {
		return GPFS_LEASE_READ;
	}

	if (leasetype == F_WRLCK) {
		return GPFS_LEASE_WRITE;
	}

	return GPFS_LEASE_NONE;
}

static int vfs_gpfs_setlease(vfs_handle_struct *handle,
			     files_struct *fsp,
			     int leasetype)
{
	struct gpfs_config_data *config;
	int ret=0;

	START_PROFILE(syscall_linux_setlease);

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	ret = linux_set_lease_sighandler(fsp_get_io_fd(fsp));
	if (ret == -1) {
		goto failure;
	}

	if (config->leases) {
		int gpfs_lease_type = lease_type_to_gpfs(leasetype);
		int saved_errno = 0;

		/*
		 * Ensure the lease owner is root to allow
		 * correct delivery of lease-break signals.
		 */
		become_root();
		ret = gpfswrap_set_lease(fsp_get_io_fd(fsp), gpfs_lease_type);
		if (ret < 0) {
			saved_errno = errno;
		}
		unbecome_root();

		if (saved_errno != 0) {
			errno = saved_errno;
		}
	}

failure:
	END_PROFILE(syscall_linux_setlease);

	return ret;
}

#else /* HAVE_KERNEL_OPLOCKS_LINUX */

static int vfs_gpfs_setlease(vfs_handle_struct *handle,
				files_struct *fsp,
				int leasetype)
{
	return ENOSYS;
}
#endif /* HAVE_KERNEL_OPLOCKS_LINUX */

static NTSTATUS vfs_gpfs_get_real_filename_at(struct vfs_handle_struct *handle,
					      struct files_struct *dirfsp,
					      const char *name,
					      TALLOC_CTX *mem_ctx,
					      char **found_name)
{
	int result;
	char *full_path = NULL;
	char *to_free = NULL;
	char real_pathname[PATH_MAX+1], tmpbuf[PATH_MAX];
	size_t full_path_len;
	int buflen;
	bool mangled;
	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return NT_STATUS_INTERNAL_ERROR);

	if (!config->getrealfilename) {
		return SMB_VFS_NEXT_GET_REAL_FILENAME_AT(
			handle, dirfsp, name, mem_ctx, found_name);
	}

	mangled = mangle_is_mangled(name, handle->conn->params);
	if (mangled) {
		return SMB_VFS_NEXT_GET_REAL_FILENAME_AT(
			handle, dirfsp, name, mem_ctx, found_name);
	}

	full_path_len = full_path_tos(dirfsp->fsp_name->base_name, name,
				      tmpbuf, sizeof(tmpbuf),
				      &full_path, &to_free);
	if (full_path_len == -1) {
		return NT_STATUS_NO_MEMORY;
	}

	buflen = sizeof(real_pathname) - 1;

	result = gpfswrap_get_realfilename_path(full_path, real_pathname,
						&buflen);

	TALLOC_FREE(to_free);

	if ((result == -1) && (errno == ENOSYS)) {
		return SMB_VFS_NEXT_GET_REAL_FILENAME_AT(
			handle, dirfsp, name, mem_ctx, found_name);
	}

	if (result == -1) {
		DEBUG(10, ("smbd_gpfs_get_realfilename_path returned %s\n",
			   strerror(errno)));
		return map_nt_error_from_unix(errno);
	}

	/*
	 * GPFS does not necessarily null-terminate the returned path
	 * but instead returns the buffer length in buflen.
	 */

	if (buflen < sizeof(real_pathname)) {
		real_pathname[buflen] = '\0';
	} else {
		real_pathname[sizeof(real_pathname)-1] = '\0';
	}

	DBG_DEBUG("%s/%s -> %s\n",
		  fsp_str_dbg(dirfsp),
		  name,
		  real_pathname);

	name = strrchr_m(real_pathname, '/');
	if (name == NULL) {
		return NT_STATUS_OBJECT_NAME_NOT_FOUND;
	}

	*found_name = talloc_strdup(mem_ctx, name+1);
	if (*found_name == NULL) {
		return NT_STATUS_NO_MEMORY;
	}

	return NT_STATUS_OK;
}

static void sd2gpfs_control(uint16_t control, struct gpfs_acl *gacl)
{
	unsigned int gpfs_aclflags = 0;
	control &= SEC_DESC_DACL_PROTECTED | SEC_DESC_SACL_PROTECTED |
		SEC_DESC_DACL_AUTO_INHERITED | SEC_DESC_SACL_AUTO_INHERITED |
		SEC_DESC_DACL_DEFAULTED | SEC_DESC_SACL_DEFAULTED |
		SEC_DESC_DACL_PRESENT | SEC_DESC_SACL_PRESENT;
	gpfs_aclflags = control << 8;
	if (!(control & SEC_DESC_DACL_PRESENT))
		gpfs_aclflags |= ACL4_FLAG_NULL_DACL;
	if (!(control & SEC_DESC_SACL_PRESENT))
		gpfs_aclflags |= ACL4_FLAG_NULL_SACL;
	gacl->acl_level = GPFS_ACL_LEVEL_V4FLAGS;
	gacl->v4Level1.acl_flags = gpfs_aclflags;
}

static uint16_t gpfs2sd_control(unsigned int gpfs_aclflags)
{
	uint16_t control = gpfs_aclflags >> 8;
	control &= SEC_DESC_DACL_PROTECTED | SEC_DESC_SACL_PROTECTED |
		SEC_DESC_DACL_AUTO_INHERITED | SEC_DESC_SACL_AUTO_INHERITED |
		SEC_DESC_DACL_DEFAULTED | SEC_DESC_SACL_DEFAULTED |
		SEC_DESC_DACL_PRESENT | SEC_DESC_SACL_PRESENT;
	control |= SEC_DESC_SELF_RELATIVE;
	return control;
}

static void gpfs_dumpacl(int level, struct gpfs_acl *gacl)
{
	gpfs_aclCount_t i;
	if (gacl==NULL)
	{
		DEBUG(0, ("gpfs acl is NULL\n"));
		return;
	}

	DEBUG(level, ("len: %d, level: %d, version: %d, nace: %d, "
		      "control: %x\n",
		      gacl->acl_len, gacl->acl_level, gacl->acl_version,
		      gacl->acl_nace, gpfs_acl_flags(gacl)));

	for(i=0; i<gacl->acl_nace; i++)
	{
		struct gpfs_ace_v4 *gace = gpfs_ace_ptr(gacl, i);
		DEBUG(level, ("\tace[%d]: type:%d, flags:0x%x, mask:0x%x, "
			      "iflags:0x%x, who:%u\n",
			      i, gace->aceType, gace->aceFlags, gace->aceMask,
			      gace->aceIFlags, gace->aceWho));
	}
}

static int gpfs_getacl_with_capability(struct files_struct *fsp,
				       int flags,
				       void *buf)
{
	int ret, saved_errno;

	set_effective_capability(DAC_OVERRIDE_CAPABILITY);

	ret = gpfswrap_fgetacl(fsp_get_pathref_fd(fsp), flags, buf);
	saved_errno = errno;

	drop_effective_capability(DAC_OVERRIDE_CAPABILITY);

	errno = saved_errno;
	return ret;
}

/*
 * get the ACL from GPFS, allocated on the specified mem_ctx
 * internally retries when initial buffer was too small
 *
 * caller needs to cast result to either
 * raw = yes: struct gpfs_opaque_acl
 * raw = no: struct gpfs_acl
 *
 */
static void *vfs_gpfs_getacl(TALLOC_CTX *mem_ctx,
			 struct files_struct *fsp,
			 const bool raw,
			 const gpfs_aclType_t type)
{
	const char *fname = fsp->fsp_name->base_name;
	void *aclbuf;
	size_t size = 512;
	int ret, flags;
	unsigned int *len;
	size_t struct_size;
	bool use_capability = false;

again:

	aclbuf = talloc_zero_size(mem_ctx, size);
	if (aclbuf == NULL) {
		errno = ENOMEM;
		return NULL;
	}

	if (raw) {
		struct gpfs_opaque_acl *buf = (struct gpfs_opaque_acl *) aclbuf;
		buf->acl_type = type;
		flags = GPFS_GETACL_NATIVE;
		len = (unsigned int *) &(buf->acl_buffer_len);
		struct_size = sizeof(struct gpfs_opaque_acl);
	} else {
		struct gpfs_acl *buf = (struct gpfs_acl *) aclbuf;
		buf->acl_type = type;
		buf->acl_level = GPFS_ACL_LEVEL_V4FLAGS;
		flags = GPFS_GETACL_STRUCT;
		len = &(buf->acl_len);
		/* reserve space for control flags in gpfs 3.5 and beyond */
		struct_size = sizeof(struct gpfs_acl) + sizeof(unsigned int);
	}

	/* set the length of the buffer as input value */
	*len = size;

	if (use_capability) {
		ret = gpfs_getacl_with_capability(fsp, flags, aclbuf);
	} else {
		ret = gpfswrap_fgetacl(fsp_get_pathref_fd(fsp), flags, aclbuf);
		if ((ret != 0) && (errno == EACCES)) {
			DBG_DEBUG("Retry with DAC capability for %s\n", fname);
			use_capability = true;
			ret = gpfs_getacl_with_capability(fsp, flags, aclbuf);
		}
	}

	if ((ret != 0) && (errno == ENOSPC)) {
		/*
		 * get the size needed to accommodate the complete buffer
		 *
		 * the value returned only applies to the ACL blob in the
		 * struct so make sure to also have headroom for the first
		 * struct members by adding room for the complete struct
		 * (might be a few bytes too much then)
		 */
		size = *len + struct_size;
		talloc_free(aclbuf);
		DEBUG(10, ("Increasing ACL buffer size to %zu\n", size));
		goto again;
	}

	if (ret != 0) {
		DEBUG(5, ("smbd_gpfs_getacl failed with %s\n",
			  strerror(errno)));
		talloc_free(aclbuf);
		return NULL;
	}

	return aclbuf;
}

/* Tries to get nfs4 acls and returns SMB ACL allocated.
 * On failure returns 1 if it got non-NFSv4 ACL to prompt 
 * retry with POSIX ACL checks.
 * On failure returns -1 if there is system (GPFS) error, check errno.
 * Returns 0 on success
 */
static int gpfs_get_nfs4_acl(TALLOC_CTX *mem_ctx,
			     struct files_struct *fsp,
			     struct SMB4ACL_T **ppacl)
{
	const char *fname = fsp->fsp_name->base_name;
	gpfs_aclCount_t i;
	struct gpfs_acl *gacl = NULL;
	DEBUG(10, ("gpfs_get_nfs4_acl invoked for %s\n", fname));

	/* Get the ACL */
	gacl = (struct gpfs_acl*) vfs_gpfs_getacl(talloc_tos(), fsp,
						  false, 0);
	if (gacl == NULL) {
		DEBUG(9, ("gpfs_getacl failed for %s with %s\n",
			   fname, strerror(errno)));
		if (errno == ENODATA) {
			/*
			 * GPFS returns ENODATA for snapshot
			 * directories. Retry with POSIX ACLs check.
			 */
			return 1;
		}

		return -1;
	}

	if (gacl->acl_type != GPFS_ACL_TYPE_NFS4) {
		DEBUG(10, ("Got non-nfsv4 acl\n"));
		/* Retry with POSIX ACLs check */
		talloc_free(gacl);
		return 1;
	}

	*ppacl = smb_create_smb4acl(mem_ctx);

	if (gacl->acl_level == GPFS_ACL_LEVEL_V4FLAGS) {
		uint16_t control = gpfs2sd_control(gpfs_acl_flags(gacl));
		smbacl4_set_controlflags(*ppacl, control);
	}

	DEBUG(10, ("len: %d, level: %d, version: %d, nace: %d, control: %x\n",
		   gacl->acl_len, gacl->acl_level, gacl->acl_version,
		   gacl->acl_nace, gpfs_acl_flags(gacl)));

	for (i=0; i<gacl->acl_nace; i++) {
		struct gpfs_ace_v4 *gace = gpfs_ace_ptr(gacl, i);
		SMB_ACE4PROP_T smbace = { 0 };
		DEBUG(10, ("type: %d, iflags: %x, flags: %x, mask: %x, "
			   "who: %d\n", gace->aceType, gace->aceIFlags,
			   gace->aceFlags, gace->aceMask, gace->aceWho));

		if (gace->aceIFlags & ACE4_IFLAG_SPECIAL_ID) {
			smbace.flags |= SMB_ACE4_ID_SPECIAL;
			switch (gace->aceWho) {
			case ACE4_SPECIAL_OWNER:
				smbace.who.special_id = SMB_ACE4_WHO_OWNER;
				break;
			case ACE4_SPECIAL_GROUP:
				smbace.who.special_id = SMB_ACE4_WHO_GROUP;
				break;
			case ACE4_SPECIAL_EVERYONE:
				smbace.who.special_id = SMB_ACE4_WHO_EVERYONE;
				break;
			default:
				DEBUG(8, ("invalid special gpfs id %d "
					  "ignored\n", gace->aceWho));
				continue; /* don't add it */
			}
		} else {
			if (gace->aceFlags & ACE4_FLAG_GROUP_ID)
				smbace.who.gid = gace->aceWho;
			else
				smbace.who.uid = gace->aceWho;
		}

		/* remove redundant deny entries */
		if (i > 0 && gace->aceType == SMB_ACE4_ACCESS_DENIED_ACE_TYPE) {
			struct gpfs_ace_v4 *prev = gpfs_ace_ptr(gacl, i - 1);
			if (prev->aceType == SMB_ACE4_ACCESS_ALLOWED_ACE_TYPE &&
			    prev->aceFlags == gace->aceFlags &&
			    prev->aceIFlags == gace->aceIFlags &&
			    (gace->aceMask & prev->aceMask) == 0 &&
			    gace->aceWho == prev->aceWho) {
				/* it's redundant - skip it */
				continue;
			}
		}

		smbace.aceType = gace->aceType;
		smbace.aceFlags = gace->aceFlags;
		smbace.aceMask = gace->aceMask;
		smb_add_ace4(*ppacl, &smbace);
	}

	talloc_free(gacl);

	return 0;
}

static NTSTATUS gpfsacl_fget_nt_acl(vfs_handle_struct *handle,
	files_struct *fsp, uint32_t security_info,
	TALLOC_CTX *mem_ctx,
	struct security_descriptor **ppdesc)
{
	struct SMB4ACL_T *pacl = NULL;
	int	result;
	struct gpfs_config_data *config;
	TALLOC_CTX *frame = talloc_stackframe();
	NTSTATUS status;

	*ppdesc = NULL;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return NT_STATUS_INTERNAL_ERROR);

	if (!config->acl) {
		status = SMB_VFS_NEXT_FGET_NT_ACL(handle, fsp, security_info,
						  mem_ctx, ppdesc);
		TALLOC_FREE(frame);
		return status;
	}

	result = gpfs_get_nfs4_acl(frame, fsp, &pacl);

	if (result == 0) {
		status = smb_fget_nt_acl_nfs4(fsp, &config->nfs4_params,
					      security_info,
					      mem_ctx, ppdesc, pacl);
		TALLOC_FREE(frame);
		return status;
	}

	if (result > 0) {
		DEBUG(10, ("retrying with posix acl...\n"));
		status = posix_fget_nt_acl(fsp, security_info,
					   mem_ctx, ppdesc);
		TALLOC_FREE(frame);
		return status;
	}

	TALLOC_FREE(frame);

	/* GPFS ACL was not read, something wrong happened, error code is set in errno */
	return map_nt_error_from_unix(errno);
}

static bool vfs_gpfs_nfs4_ace_to_gpfs_ace(SMB_ACE4PROP_T *nfs4_ace,
					  struct gpfs_ace_v4 *gace,
					  uid_t owner_uid)
{
	gace->aceType = nfs4_ace->aceType;
	gace->aceFlags = nfs4_ace->aceFlags;
	gace->aceMask = nfs4_ace->aceMask;

	if (nfs4_ace->flags & SMB_ACE4_ID_SPECIAL) {
		switch(nfs4_ace->who.special_id) {
		case SMB_ACE4_WHO_EVERYONE:
			gace->aceIFlags = ACE4_IFLAG_SPECIAL_ID;
			gace->aceWho = ACE4_SPECIAL_EVERYONE;
			break;
		case SMB_ACE4_WHO_OWNER:
			/*
			 * With GPFS it is not possible to deny ACL or
			 * attribute access to the owner. Setting an
			 * ACL with such an entry is not possible.
			 * Denying ACL or attribute access for the
			 * owner through a named ACL entry can be
			 * stored in an ACL, it is just not effective.
			 *
			 * Map this case to a named entry to allow at
			 * least setting this ACL, which will be
			 * enforced by the smbd permission check. Do
			 * not do this for an inheriting OWNER entry,
			 * as this represents a CREATOR OWNER ACE. The
			 * remaining limitation is that CREATOR OWNER
			 * cannot deny ACL or attribute access.
			 */
			if (!nfs_ace_is_inherit(nfs4_ace) &&
			    nfs4_ace->aceType ==
					SMB_ACE4_ACCESS_DENIED_ACE_TYPE &&
			    nfs4_ace->aceMask & (SMB_ACE4_READ_ATTRIBUTES|
						 SMB_ACE4_WRITE_ATTRIBUTES|
						 SMB_ACE4_READ_ACL|
						 SMB_ACE4_WRITE_ACL)) {
				gace->aceIFlags = 0;
				gace->aceWho = owner_uid;
			} else {
				gace->aceIFlags = ACE4_IFLAG_SPECIAL_ID;
				gace->aceWho = ACE4_SPECIAL_OWNER;
			}
			break;
		case SMB_ACE4_WHO_GROUP:
			gace->aceIFlags = ACE4_IFLAG_SPECIAL_ID;
			gace->aceWho = ACE4_SPECIAL_GROUP;
			break;
		default:
			DBG_WARNING("Unsupported special_id %d\n",
				    nfs4_ace->who.special_id);
			return false;
		}

		return true;
	}

	gace->aceIFlags = 0;
	gace->aceWho = (nfs4_ace->aceFlags & SMB_ACE4_IDENTIFIER_GROUP) ?
		nfs4_ace->who.gid : nfs4_ace->who.uid;

	return true;
}

static struct gpfs_acl *vfs_gpfs_smbacl2gpfsacl(TALLOC_CTX *mem_ctx,
						files_struct *fsp,
						struct SMB4ACL_T *smbacl,
						bool controlflags)
{
	struct gpfs_acl *gacl;
	gpfs_aclLen_t gacl_len;
	struct SMB4ACE_T *smbace;

	gacl_len = offsetof(gpfs_acl_t, ace_v4) + sizeof(unsigned int)
		+ smb_get_naces(smbacl) * sizeof(gpfs_ace_v4_t);

	gacl = (struct gpfs_acl *)TALLOC_SIZE(mem_ctx, gacl_len);
	if (gacl == NULL) {
		DEBUG(0, ("talloc failed\n"));
		errno = ENOMEM;
		return NULL;
	}

	gacl->acl_level = GPFS_ACL_LEVEL_BASE;
	gacl->acl_version = GPFS_ACL_VERSION_NFS4;
	gacl->acl_type = GPFS_ACL_TYPE_NFS4;
	gacl->acl_nace = 0; /* change later... */

	if (controlflags) {
		gacl->acl_level = GPFS_ACL_LEVEL_V4FLAGS;
		sd2gpfs_control(smbacl4_get_controlflags(smbacl), gacl);
	}

	for (smbace=smb_first_ace4(smbacl); smbace!=NULL; smbace = smb_next_ace4(smbace)) {
		struct gpfs_ace_v4 *gace = gpfs_ace_ptr(gacl, gacl->acl_nace);
		SMB_ACE4PROP_T	*aceprop = smb_get_ace4(smbace);
		bool add_ace;

		add_ace = vfs_gpfs_nfs4_ace_to_gpfs_ace(aceprop, gace,
							fsp->fsp_name->st.st_ex_uid);
		if (!add_ace) {
			continue;
		}

		gacl->acl_nace++;
	}
	gacl->acl_len = (char *)gpfs_ace_ptr(gacl, gacl->acl_nace)
		- (char *)gacl;
	return gacl;
}

static bool gpfsacl_process_smbacl(vfs_handle_struct *handle,
				   files_struct *fsp,
				   struct SMB4ACL_T *smbacl)
{
	int ret;
	struct gpfs_acl *gacl;
	TALLOC_CTX *mem_ctx = talloc_tos();

	gacl = vfs_gpfs_smbacl2gpfsacl(mem_ctx, fsp, smbacl, true);
	if (gacl == NULL) { /* out of memory */
		return False;
	}
	ret = gpfswrap_putacl(fsp->fsp_name->base_name,
			      GPFS_PUTACL_STRUCT | GPFS_ACL_SAMBA, gacl);

	if ((ret != 0) && (errno == EINVAL)) {
		DEBUG(10, ("Retry without nfs41 control flags\n"));
		talloc_free(gacl);
		gacl = vfs_gpfs_smbacl2gpfsacl(mem_ctx, fsp, smbacl, false);
		if (gacl == NULL) { /* out of memory */
			return False;
		}
		ret = gpfswrap_putacl(fsp->fsp_name->base_name,
				      GPFS_PUTACL_STRUCT | GPFS_ACL_SAMBA,
				      gacl);
	}

	if (ret != 0) {
		DEBUG(8, ("gpfs_putacl failed with %s\n", strerror(errno)));
		gpfs_dumpacl(8, gacl);
		return False;
	}

	DEBUG(10, ("gpfs_putacl succeeded\n"));
	return True;
}

static NTSTATUS gpfsacl_set_nt_acl_internal(vfs_handle_struct *handle, files_struct *fsp, uint32_t security_info_sent, const struct security_descriptor *psd)
{
	struct gpfs_acl *acl;
	NTSTATUS result = NT_STATUS_ACCESS_DENIED;

	acl = (struct gpfs_acl*) vfs_gpfs_getacl(talloc_tos(),
						 fsp,
						 false, 0);
	if (acl == NULL) {
		return map_nt_error_from_unix(errno);
	}

	if (acl->acl_version == GPFS_ACL_VERSION_NFS4) {
		struct gpfs_config_data *config;

		SMB_VFS_HANDLE_GET_DATA(handle, config,
					struct gpfs_config_data,
					return NT_STATUS_INTERNAL_ERROR);

		result = smb_set_nt_acl_nfs4(handle,
			fsp, &config->nfs4_params, security_info_sent, psd,
			gpfsacl_process_smbacl);
	} else { /* assume POSIX ACL - by default... */
		result = set_nt_acl(fsp, security_info_sent, psd);
	}

	talloc_free(acl);
	return result;
}

static NTSTATUS gpfsacl_fset_nt_acl(vfs_handle_struct *handle, files_struct *fsp, uint32_t security_info_sent, const struct security_descriptor *psd)
{
	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return NT_STATUS_INTERNAL_ERROR);

	if (!config->acl) {
		return SMB_VFS_NEXT_FSET_NT_ACL(handle, fsp, security_info_sent, psd);
	}

	return gpfsacl_set_nt_acl_internal(handle, fsp, security_info_sent, psd);
}

static SMB_ACL_T gpfs2smb_acl(const struct gpfs_acl *pacl, TALLOC_CTX *mem_ctx)
{
	SMB_ACL_T result;
	gpfs_aclCount_t i;

	result = sys_acl_init(mem_ctx);
	if (result == NULL) {
		errno = ENOMEM;
		return NULL;
	}

	result->count = pacl->acl_nace;
	result->acl = talloc_realloc(result, result->acl, struct smb_acl_entry,
				     result->count);
	if (result->acl == NULL) {
		TALLOC_FREE(result);
		errno = ENOMEM;
		return NULL;
	}

	for (i=0; i<pacl->acl_nace; i++) {
		struct smb_acl_entry *ace = &result->acl[i];
		const struct gpfs_ace_v1 *g_ace = &pacl->ace_v1[i];

		DEBUG(10, ("Converting type %d id %lu perm %x\n",
			   (int)g_ace->ace_type, (unsigned long)g_ace->ace_who,
			   (int)g_ace->ace_perm));

		switch (g_ace->ace_type) {
		case GPFS_ACL_USER:
			ace->a_type = SMB_ACL_USER;
			ace->info.user.uid = (uid_t)g_ace->ace_who;
			break;
		case GPFS_ACL_USER_OBJ:
			ace->a_type = SMB_ACL_USER_OBJ;
			break;
		case GPFS_ACL_GROUP:
			ace->a_type = SMB_ACL_GROUP;
			ace->info.group.gid = (gid_t)g_ace->ace_who;
			break;
		case GPFS_ACL_GROUP_OBJ:
 			ace->a_type = SMB_ACL_GROUP_OBJ;
			break;
		case GPFS_ACL_OTHER:
			ace->a_type = SMB_ACL_OTHER;
			break;
		case GPFS_ACL_MASK:
			ace->a_type = SMB_ACL_MASK;
			break;
		default:
			DEBUG(10, ("Got invalid ace_type: %d\n",
				   g_ace->ace_type));
			TALLOC_FREE(result);
			errno = EINVAL;
			return NULL;
		}

		ace->a_perm = 0;
		ace->a_perm |= (g_ace->ace_perm & ACL_PERM_READ) ?
			SMB_ACL_READ : 0;
		ace->a_perm |= (g_ace->ace_perm & ACL_PERM_WRITE) ?
			SMB_ACL_WRITE : 0;
		ace->a_perm |= (g_ace->ace_perm & ACL_PERM_EXECUTE) ?
			SMB_ACL_EXECUTE : 0;

		DEBUGADD(10, ("Converted to %d perm %x\n",
			      ace->a_type, ace->a_perm));
	}

	return result;
}

static SMB_ACL_T gpfsacl_get_posix_acl(struct files_struct *fsp,
				       gpfs_aclType_t type,
				       TALLOC_CTX *mem_ctx)
{
	struct gpfs_acl *pacl;
	SMB_ACL_T result = NULL;

	pacl = vfs_gpfs_getacl(talloc_tos(), fsp, false, type);

	if (pacl == NULL) {
		DBG_DEBUG("vfs_gpfs_getacl failed for %s with %s\n",
			   fsp_str_dbg(fsp), strerror(errno));
		if (errno == 0) {
			errno = EINVAL;
		}
		goto done;
	}

	if (pacl->acl_version != GPFS_ACL_VERSION_POSIX) {
		DEBUG(10, ("Got acl version %d, expected %d\n",
			   pacl->acl_version, GPFS_ACL_VERSION_POSIX));
		errno = EINVAL;
		goto done;
	}

	DEBUG(10, ("len: %d, level: %d, version: %d, nace: %d\n",
		   pacl->acl_len, pacl->acl_level, pacl->acl_version,
		   pacl->acl_nace));

	result = gpfs2smb_acl(pacl, mem_ctx);
	if (result != NULL) {
		errno = 0;
	}

 done:

	if (pacl != NULL) {
		talloc_free(pacl);
	}
	if (errno != 0) {
		TALLOC_FREE(result);
	}
	return result;
}

static SMB_ACL_T gpfsacl_sys_acl_get_fd(vfs_handle_struct *handle,
					files_struct *fsp,
					SMB_ACL_TYPE_T type,
					TALLOC_CTX *mem_ctx)
{
	gpfs_aclType_t gpfs_type;
	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return NULL);

	if (!config->acl) {
		return SMB_VFS_NEXT_SYS_ACL_GET_FD(handle, fsp, type, mem_ctx);
	}

	switch(type) {
	case SMB_ACL_TYPE_ACCESS:
		gpfs_type = GPFS_ACL_TYPE_ACCESS;
		break;
	case SMB_ACL_TYPE_DEFAULT:
		gpfs_type = GPFS_ACL_TYPE_DEFAULT;
		break;
	default:
		DEBUG(0, ("Got invalid type: %d\n", type));
		smb_panic("exiting");
	}
	return gpfsacl_get_posix_acl(fsp, gpfs_type, mem_ctx);
}

static int gpfsacl_sys_acl_blob_get_fd(vfs_handle_struct *handle,
				      files_struct *fsp,
				      TALLOC_CTX *mem_ctx,
				      char **blob_description,
				      DATA_BLOB *blob)
{
	struct gpfs_config_data *config;
	struct gpfs_opaque_acl *acl = NULL;
	DATA_BLOB aclblob;
	int result;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if (!config->acl) {
		return SMB_VFS_NEXT_SYS_ACL_BLOB_GET_FD(handle, fsp, mem_ctx,
							blob_description, blob);
	}

	errno = 0;
	acl = (struct gpfs_opaque_acl *) vfs_gpfs_getacl(mem_ctx,
						fsp,
						true,
						GPFS_ACL_TYPE_NFS4);

	if (errno) {
		DEBUG(5, ("vfs_gpfs_getacl finished with errno %d: %s\n",
					errno, strerror(errno)));

		/* EINVAL means POSIX ACL, bail out on other cases */
		if (errno != EINVAL) {
			return -1;
		}
	}

	if (acl != NULL) {
		/*
		 * file has NFSv4 ACL
		 *
		 * we only need the actual ACL blob here
		 * acl_version will always be NFS4 because we asked
		 * for NFS4
		 * acl_type is only used for POSIX ACLs
		 */
		aclblob.data = (uint8_t*) acl->acl_var_data;
		aclblob.length = acl->acl_buffer_len;

		*blob_description = talloc_strdup(mem_ctx, "gpfs_nfs4_acl");
		if (!*blob_description) {
			talloc_free(acl);
			errno = ENOMEM;
			return -1;
		}

		result = non_posix_sys_acl_blob_get_fd_helper(handle, fsp,
							      aclblob, mem_ctx,
							      blob);

		talloc_free(acl);
		return result;
	}

	/* fall back to POSIX ACL */
	return posix_sys_acl_blob_get_fd(handle, fsp, mem_ctx,
					 blob_description, blob);
}

static struct gpfs_acl *smb2gpfs_acl(const SMB_ACL_T pacl,
				     SMB_ACL_TYPE_T type)
{
	gpfs_aclLen_t len;
	struct gpfs_acl *result;
	int i;

	DEBUG(10, ("smb2gpfs_acl: Got ACL with %d entries\n", pacl->count));

	len = offsetof(gpfs_acl_t, ace_v1) + (pacl->count) *
		sizeof(gpfs_ace_v1_t);

	result = (struct gpfs_acl *)SMB_MALLOC(len);
	if (result == NULL) {
		errno = ENOMEM;
		return result;
	}

	result->acl_len = len;
	result->acl_level = 0;
	result->acl_version = GPFS_ACL_VERSION_POSIX;
	result->acl_type = (type == SMB_ACL_TYPE_DEFAULT) ?
		GPFS_ACL_TYPE_DEFAULT : GPFS_ACL_TYPE_ACCESS;
	result->acl_nace = pacl->count;

	for (i=0; i<pacl->count; i++) {
		const struct smb_acl_entry *ace = &pacl->acl[i];
		struct gpfs_ace_v1 *g_ace = &result->ace_v1[i];

		DEBUG(10, ("Converting type %d perm %x\n",
			   (int)ace->a_type, (int)ace->a_perm));

		g_ace->ace_perm = 0;

		switch(ace->a_type) {
		case SMB_ACL_USER:
			g_ace->ace_type = GPFS_ACL_USER;
			g_ace->ace_who = (gpfs_uid_t)ace->info.user.uid;
			break;
		case SMB_ACL_USER_OBJ:
			g_ace->ace_type = GPFS_ACL_USER_OBJ;
			g_ace->ace_perm |= ACL_PERM_CONTROL;
			g_ace->ace_who = 0;
			break;
		case SMB_ACL_GROUP:
			g_ace->ace_type = GPFS_ACL_GROUP;
			g_ace->ace_who = (gpfs_uid_t)ace->info.group.gid;
			break;
		case SMB_ACL_GROUP_OBJ:
			g_ace->ace_type = GPFS_ACL_GROUP_OBJ;
			g_ace->ace_who = 0;
			break;
		case SMB_ACL_MASK:
			g_ace->ace_type = GPFS_ACL_MASK;
			g_ace->ace_perm = 0x8f;
			g_ace->ace_who = 0;
			break;
		case SMB_ACL_OTHER:
			g_ace->ace_type = GPFS_ACL_OTHER;
			g_ace->ace_who = 0;
			break;
		default:
			DEBUG(10, ("Got invalid ace_type: %d\n", ace->a_type));
			errno = EINVAL;
			SAFE_FREE(result);
			return NULL;
		}

		g_ace->ace_perm |= (ace->a_perm & SMB_ACL_READ) ?
			ACL_PERM_READ : 0;
		g_ace->ace_perm |= (ace->a_perm & SMB_ACL_WRITE) ?
			ACL_PERM_WRITE : 0;
		g_ace->ace_perm |= (ace->a_perm & SMB_ACL_EXECUTE) ?
			ACL_PERM_EXECUTE : 0;

		DEBUGADD(10, ("Converted to %d id %d perm %x\n",
			      g_ace->ace_type, g_ace->ace_who, g_ace->ace_perm));
	}

	return result;
}

static int gpfsacl_sys_acl_set_fd(vfs_handle_struct *handle,
				  files_struct *fsp,
				  SMB_ACL_TYPE_T type,
				  SMB_ACL_T theacl)
{
	struct gpfs_config_data *config;
	struct gpfs_acl *gpfs_acl = NULL;
	int result;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if (!config->acl) {
		return SMB_VFS_NEXT_SYS_ACL_SET_FD(handle, fsp, type, theacl);
	}

	gpfs_acl = smb2gpfs_acl(theacl, type);
	if (gpfs_acl == NULL) {
		return -1;
	}

	/*
	 * This is no longer a handle based call.
	 */
	result = gpfswrap_putacl(fsp->fsp_name->base_name,
				 GPFS_PUTACL_STRUCT|GPFS_ACL_SAMBA,
				 gpfs_acl);
	SAFE_FREE(gpfs_acl);
	return result;
}

static int gpfsacl_sys_acl_delete_def_fd(vfs_handle_struct *handle,
				files_struct *fsp)
{
	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if (!config->acl) {
		return SMB_VFS_NEXT_SYS_ACL_DELETE_DEF_FD(handle, fsp);
	}

	errno = ENOTSUP;
	return -1;
}


/*
 * Assumed: mode bits are shiftable and standard
 * Output: the new aceMask field for an smb nfs4 ace
 */
static uint32_t gpfsacl_mask_filter(uint32_t aceType, uint32_t aceMask, uint32_t rwx)
{
	const uint32_t posix_nfs4map[3] = {
                SMB_ACE4_EXECUTE, /* execute */
		SMB_ACE4_WRITE_DATA | SMB_ACE4_APPEND_DATA, /* write; GPFS specific */
                SMB_ACE4_READ_DATA /* read */
	};
	int     i;
	uint32_t        posix_mask = 0x01;
	uint32_t        posix_bit;
	uint32_t        nfs4_bits;

	for(i=0; i<3; i++) {
		nfs4_bits = posix_nfs4map[i];
		posix_bit = rwx & posix_mask;

		if (aceType==SMB_ACE4_ACCESS_ALLOWED_ACE_TYPE) {
			if (posix_bit)
				aceMask |= nfs4_bits;
			else
				aceMask &= ~nfs4_bits;
		} else {
			/* add deny bits when suitable */
			if (!posix_bit)
				aceMask |= nfs4_bits;
			else
				aceMask &= ~nfs4_bits;
		} /* other ace types are unexpected */

		posix_mask <<= 1;
	}

	return aceMask;
}

static int gpfsacl_emu_chmod(vfs_handle_struct *handle,
			     struct files_struct *fsp,
			     mode_t mode)
{
	struct smb_filename *fname = fsp->fsp_name;
	char *path = fsp->fsp_name->base_name;
	struct SMB4ACL_T *pacl = NULL;
	int     result;
	bool    haveAllowEntry[SMB_ACE4_WHO_EVERYONE + 1] = {False, False, False, False};
	int     i;
	files_struct fake_fsp = { 0 }; /* TODO: rationalize parametrization */
	struct SMB4ACE_T *smbace;
	TALLOC_CTX *frame = talloc_stackframe();

	DEBUG(10, ("gpfsacl_emu_chmod invoked for %s mode %o\n", path, mode));

	result = gpfs_get_nfs4_acl(frame, fsp, &pacl);
	if (result) {
		TALLOC_FREE(frame);
		return result;
	}

	if (mode & ~(S_IRWXU | S_IRWXG | S_IRWXO)) {
		DEBUG(2, ("WARNING: cutting extra mode bits %o on %s\n", mode, path));
	}

	for (smbace=smb_first_ace4(pacl); smbace!=NULL; smbace = smb_next_ace4(smbace)) {
		SMB_ACE4PROP_T  *ace = smb_get_ace4(smbace);
		uint32_t        specid = ace->who.special_id;

		if (ace->flags&SMB_ACE4_ID_SPECIAL &&
		    ace->aceType<=SMB_ACE4_ACCESS_DENIED_ACE_TYPE &&
		    specid <= SMB_ACE4_WHO_EVERYONE) {

			uint32_t newMask;

			if (ace->aceType==SMB_ACE4_ACCESS_ALLOWED_ACE_TYPE)
				haveAllowEntry[specid] = True;

			/* mode >> 6 for @owner, mode >> 3 for @group,
			 * mode >> 0 for @everyone */
			newMask = gpfsacl_mask_filter(ace->aceType, ace->aceMask,
						      mode >> ((SMB_ACE4_WHO_EVERYONE - specid) * 3));
			if (ace->aceMask!=newMask) {
				DEBUG(10, ("ace changed for %s (%o -> %o) id=%d\n",
					   path, ace->aceMask, newMask, specid));
			}
			ace->aceMask = newMask;
		}
	}

	/* make sure we have at least ALLOW entries
	 * for all the 3 special ids (@EVERYONE, @OWNER, @GROUP)
	 * - if necessary
	 */
	for(i = SMB_ACE4_WHO_OWNER; i<=SMB_ACE4_WHO_EVERYONE; i++) {
		SMB_ACE4PROP_T ace = { 0 };

		if (haveAllowEntry[i]==True)
			continue;

		ace.aceType = SMB_ACE4_ACCESS_ALLOWED_ACE_TYPE;
		ace.flags |= SMB_ACE4_ID_SPECIAL;
		ace.who.special_id = i;

		if (i==SMB_ACE4_WHO_GROUP) /* not sure it's necessary... */
			ace.aceFlags |= SMB_ACE4_IDENTIFIER_GROUP;

		ace.aceMask = gpfsacl_mask_filter(ace.aceType, ace.aceMask,
						  mode >> ((SMB_ACE4_WHO_EVERYONE - i) * 3));

		/* don't add unnecessary aces */
		if (!ace.aceMask)
			continue;

		/* we add it to the END - as windows expects allow aces */
		smb_add_ace4(pacl, &ace);
		DEBUG(10, ("Added ALLOW ace for %s, mode=%o, id=%d, aceMask=%x\n",
			   path, mode, i, ace.aceMask));
	}

	/* don't add complementary DENY ACEs here */
	fake_fsp.fsp_name = synthetic_smb_fname(frame,
						path,
						NULL,
						NULL,
						fname->twrp,
						0);
	if (fake_fsp.fsp_name == NULL) {
		errno = ENOMEM;
		TALLOC_FREE(frame);
		return -1;
	}
	/* put the acl */
	if (gpfsacl_process_smbacl(handle, &fake_fsp, pacl) == False) {
		TALLOC_FREE(frame);
		return -1;
	}

	TALLOC_FREE(frame);
	return 0; /* ok for [f]chmod */
}

static int vfs_gpfs_fchmod(vfs_handle_struct *handle, files_struct *fsp, mode_t mode)
{
	SMB_STRUCT_STAT st;
	int rc;

	rc = SMB_VFS_NEXT_FSTAT(handle, fsp, &st);
	if (rc != 0) {
		return -1;
	}

	/* avoid chmod() if possible, to preserve acls */
	if ((st.st_ex_mode & ~S_IFMT) == mode) {
		return 0;
	}

	rc = gpfsacl_emu_chmod(handle, fsp, mode);
	if (rc == 1) {
		return SMB_VFS_NEXT_FCHMOD(handle, fsp, mode);
	}
	return rc;
}

static uint32_t vfs_gpfs_winattrs_to_dosmode(unsigned int winattrs)
{
	uint32_t dosmode = 0;

	if (winattrs & GPFS_WINATTR_ARCHIVE){
		dosmode |= FILE_ATTRIBUTE_ARCHIVE;
	}
	if (winattrs & GPFS_WINATTR_HIDDEN){
		dosmode |= FILE_ATTRIBUTE_HIDDEN;
	}
	if (winattrs & GPFS_WINATTR_SYSTEM){
		dosmode |= FILE_ATTRIBUTE_SYSTEM;
	}
	if (winattrs & GPFS_WINATTR_READONLY){
		dosmode |= FILE_ATTRIBUTE_READONLY;
	}
	if (winattrs & GPFS_WINATTR_SPARSE_FILE) {
		dosmode |= FILE_ATTRIBUTE_SPARSE;
	}
	if (winattrs & GPFS_WINATTR_OFFLINE) {
		dosmode |= FILE_ATTRIBUTE_OFFLINE;
	}

	return dosmode;
}

static unsigned int vfs_gpfs_dosmode_to_winattrs(uint32_t dosmode)
{
	unsigned int winattrs = 0;

	if (dosmode & FILE_ATTRIBUTE_ARCHIVE){
		winattrs |= GPFS_WINATTR_ARCHIVE;
	}
	if (dosmode & FILE_ATTRIBUTE_HIDDEN){
		winattrs |= GPFS_WINATTR_HIDDEN;
	}
	if (dosmode & FILE_ATTRIBUTE_SYSTEM){
		winattrs |= GPFS_WINATTR_SYSTEM;
	}
	if (dosmode & FILE_ATTRIBUTE_READONLY){
		winattrs |= GPFS_WINATTR_READONLY;
	}
	if (dosmode & FILE_ATTRIBUTE_SPARSE) {
		winattrs |= GPFS_WINATTR_SPARSE_FILE;
	}
	if (dosmode & FILE_ATTRIBUTE_OFFLINE) {
		winattrs |= GPFS_WINATTR_OFFLINE;
	}

	return winattrs;
}

static struct timespec gpfs_timestruc64_to_timespec(struct gpfs_timestruc64 g)
{
	return (struct timespec) { .tv_sec = g.tv_sec, .tv_nsec = g.tv_nsec };
}

static NTSTATUS vfs_gpfs_fget_dos_attributes(struct vfs_handle_struct *handle,
					     struct files_struct *fsp,
					     uint32_t *dosmode)
{
	struct gpfs_config_data *config;
	int fd = fsp_get_pathref_fd(fsp);
	struct sys_proc_fd_path_buf buf;
	const char *p = NULL;
	struct gpfs_iattr64 iattr = { };
	unsigned int litemask = 0;
	struct timespec ts;
	int ret;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return NT_STATUS_INTERNAL_ERROR);

	if (!config->winattr) {
		return SMB_VFS_NEXT_FGET_DOS_ATTRIBUTES(handle, fsp, dosmode);
	}

	if (fsp->fsp_flags.is_pathref && !config->pathref_ok.gpfs_fstat_x) {
		if (fsp->fsp_flags.have_proc_fds) {
			p = sys_proc_fd_path(fd, &buf);
		} else {
			p = fsp->fsp_name->base_name;
		}
	}

	if (p != NULL) {
		ret = gpfswrap_stat_x(p, &litemask, &iattr, sizeof(iattr));
	} else {
		ret = gpfswrap_fstat_x(fd, &litemask, &iattr, sizeof(iattr));
	}
	if (ret == -1 && errno == ENOSYS) {
		return SMB_VFS_NEXT_FGET_DOS_ATTRIBUTES(handle, fsp, dosmode);
	}

	if (ret == -1 && errno == EACCES) {
		int saved_errno = 0;

		/*
		 * According to MS-FSA 2.1.5.1.2.1 "Algorithm to Check Access to
		 * an Existing File" FILE_LIST_DIRECTORY on a directory implies
		 * FILE_READ_ATTRIBUTES for directory entries. Being able to
		 * open a file implies FILE_LIST_DIRECTORY.
		 */

		set_effective_capability(DAC_OVERRIDE_CAPABILITY);

		if (p != NULL) {
			ret = gpfswrap_stat_x(p,
					      &litemask,
					      &iattr,
					      sizeof(iattr));
		} else {
			ret = gpfswrap_fstat_x(fd,
					       &litemask,
					       &iattr,
					       sizeof(iattr));
		}
		if (ret == -1) {
			saved_errno = errno;
		}

		drop_effective_capability(DAC_OVERRIDE_CAPABILITY);

		if (saved_errno != 0) {
			errno = saved_errno;
		}
	}

	if (ret == -1) {
		DBG_WARNING("Getting winattrs failed for %s: %s\n",
			    fsp->fsp_name->base_name, strerror(errno));
		return map_nt_error_from_unix(errno);
	}

	ts = gpfs_timestruc64_to_timespec(iattr.ia_createtime);

	*dosmode |= vfs_gpfs_winattrs_to_dosmode(iattr.ia_winflags);
	update_stat_ex_create_time(&fsp->fsp_name->st, ts);

	return NT_STATUS_OK;
}

static NTSTATUS vfs_gpfs_fset_dos_attributes(struct vfs_handle_struct *handle,
					     struct files_struct *fsp,
					     uint32_t dosmode)
{
	struct gpfs_config_data *config;
	struct gpfs_winattr attrs = { };
	int ret;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return NT_STATUS_INTERNAL_ERROR);

	if (!config->winattr) {
		return SMB_VFS_NEXT_FSET_DOS_ATTRIBUTES(handle, fsp, dosmode);
	}

	attrs.winAttrs = vfs_gpfs_dosmode_to_winattrs(dosmode);

	if (!fsp->fsp_flags.is_pathref) {
		ret = gpfswrap_set_winattrs(fsp_get_io_fd(fsp),
					    GPFS_WINATTR_SET_ATTRS, &attrs);
		if (ret == -1) {
			DBG_WARNING("Setting winattrs failed for %s: %s\n",
				    fsp_str_dbg(fsp), strerror(errno));
			return map_nt_error_from_unix(errno);
		}
		return NT_STATUS_OK;
	}

	if (fsp->fsp_flags.have_proc_fds) {
		int fd = fsp_get_pathref_fd(fsp);
		struct sys_proc_fd_path_buf buf;

		ret = gpfswrap_set_winattrs_path(sys_proc_fd_path(fd, &buf),
						 GPFS_WINATTR_SET_ATTRS,
						 &attrs);
		if (ret == -1) {
			DBG_WARNING("Setting winattrs failed for "
				    "[%s][%s]: %s\n",
				    buf.buf,
				    fsp_str_dbg(fsp),
				    strerror(errno));
			return map_nt_error_from_unix(errno);
		}
		return NT_STATUS_OK;
	}

	/*
	 * This is no longer a handle based call.
	 */
	ret = gpfswrap_set_winattrs_path(fsp->fsp_name->base_name,
					 GPFS_WINATTR_SET_ATTRS,
					 &attrs);
	if (ret == -1) {
		DBG_WARNING("Setting winattrs failed for [%s]: %s\n",
			    fsp_str_dbg(fsp), strerror(errno));
		return map_nt_error_from_unix(errno);
	}

	return NT_STATUS_OK;
}

static int timespec_to_gpfs_time(
	struct timespec ts, gpfs_timestruc_t *gt, int idx, int *flags)
{
	if (is_omit_timespec(&ts)) {
		return 0;
	}

	if (ts.tv_sec < 0 || ts.tv_sec > UINT32_MAX) {
		DBG_NOTICE("GPFS uses 32-bit unsigned timestamps "
			   "and cannot handle %jd.\n",
			   (intmax_t)ts.tv_sec);
		errno = ERANGE;
		return -1;
	}

	*flags |= 1 << idx;
	gt[idx].tv_sec = ts.tv_sec;
	gt[idx].tv_nsec = ts.tv_nsec;
	DBG_DEBUG("Setting GPFS time %d, flags 0x%x\n", idx, *flags);

	return 0;
}

static int smbd_gpfs_set_times(struct files_struct *fsp,
			       struct smb_file_time *ft)
{
	gpfs_timestruc_t gpfs_times[4];
	int flags = 0;
	int rc;

	ZERO_ARRAY(gpfs_times);
	rc = timespec_to_gpfs_time(ft->atime, gpfs_times, 0, &flags);
	if (rc != 0) {
		return rc;
	}

	rc = timespec_to_gpfs_time(ft->mtime, gpfs_times, 1, &flags);
	if (rc != 0) {
		return rc;
	}

	/* No good mapping from LastChangeTime to ctime, not storing */
	rc = timespec_to_gpfs_time(ft->create_time, gpfs_times, 3, &flags);
	if (rc != 0) {
		return rc;
	}

	if (!flags) {
		DBG_DEBUG("nothing to do, return to avoid EINVAL\n");
		return 0;
	}

	if (!fsp->fsp_flags.is_pathref) {
		rc = gpfswrap_set_times(fsp_get_io_fd(fsp), flags, gpfs_times);
		if (rc != 0) {
			DBG_WARNING("gpfs_set_times(%s) failed: %s\n",
				    fsp_str_dbg(fsp), strerror(errno));
		}
		return rc;
	}


	if (fsp->fsp_flags.have_proc_fds) {
		int fd = fsp_get_pathref_fd(fsp);
		struct sys_proc_fd_path_buf buf;

		rc = gpfswrap_set_times_path(sys_proc_fd_path(fd, &buf),
					     flags,
					     gpfs_times);
		if (rc != 0) {
			DBG_WARNING("gpfs_set_times_path(%s,%s) failed: %s\n",
				    fsp_str_dbg(fsp),
				    buf.buf,
				    strerror(errno));
		}
		return rc;
	}

	/*
	 * This is no longer a handle based call.
	 */

	rc = gpfswrap_set_times_path(fsp->fsp_name->base_name,
				     flags,
				     gpfs_times);
	if (rc != 0) {
		DBG_WARNING("gpfs_set_times_path(%s) failed: %s\n",
			    fsp_str_dbg(fsp), strerror(errno));
	}
	return rc;
}

static int vfs_gpfs_fntimes(struct vfs_handle_struct *handle,
		files_struct *fsp,
		struct smb_file_time *ft)
{

	struct gpfs_winattr attrs;
	int ret;
	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle,
				config,
				struct gpfs_config_data,
				return -1);

	/* Try to use gpfs_set_times if it is enabled and available */
	if (config->settimes) {
		return smbd_gpfs_set_times(fsp, ft);
	}

	DBG_DEBUG("gpfs_set_times() not available or disabled, "
		  "use ntimes and winattr\n");

	ret = SMB_VFS_NEXT_FNTIMES(handle, fsp, ft);
	if (ret == -1) {
		/* don't complain if access was denied */
		if (errno != EPERM && errno != EACCES) {
			DBG_WARNING("SMB_VFS_NEXT_FNTIMES failed: %s\n",
				    strerror(errno));
		}
		return -1;
	}

	if (is_omit_timespec(&ft->create_time)) {
		DBG_DEBUG("Create Time is NULL\n");
		return 0;
	}

	if (!config->winattr) {
		return 0;
	}

	attrs.winAttrs = 0;
	attrs.creationTime.tv_sec = ft->create_time.tv_sec;
	attrs.creationTime.tv_nsec = ft->create_time.tv_nsec;

	if (!fsp->fsp_flags.is_pathref) {
		ret = gpfswrap_set_winattrs(fsp_get_io_fd(fsp),
					    GPFS_WINATTR_SET_CREATION_TIME,
					    &attrs);
		if (ret == -1 && errno != ENOSYS) {
			DBG_WARNING("Set GPFS ntimes failed %d\n", ret);
			return -1;
		}
		return ret;
	}

	if (fsp->fsp_flags.have_proc_fds) {
		int fd = fsp_get_pathref_fd(fsp);
		struct sys_proc_fd_path_buf buf;

		ret = gpfswrap_set_winattrs_path(
			sys_proc_fd_path(fd, &buf),
			GPFS_WINATTR_SET_CREATION_TIME,
			&attrs);
		if (ret == -1 && errno != ENOSYS) {
			DBG_WARNING("Set GPFS ntimes failed %d\n", ret);
			return -1;
		}
		return ret;
	}

	/*
	 * This is no longer a handle based call.
	 */
	ret = gpfswrap_set_winattrs_path(fsp->fsp_name->base_name,
					 GPFS_WINATTR_SET_CREATION_TIME,
					 &attrs);
	if (ret == -1 && errno != ENOSYS) {
		DBG_WARNING("Set GPFS ntimes failed %d\n", ret);
		return -1;
	}

	return 0;
}

static int vfs_gpfs_fallocate(struct vfs_handle_struct *handle,
			      struct files_struct *fsp, uint32_t mode,
			      off_t offset, off_t len)
{
	if (mode == (VFS_FALLOCATE_FL_PUNCH_HOLE|VFS_FALLOCATE_FL_KEEP_SIZE) &&
	    !fsp->fsp_flags.is_sparse &&
	    lp_strict_allocate(SNUM(fsp->conn))) {
		/*
		 * This is from a ZERO_DATA request on a non-sparse
		 * file. GPFS does not support FL_KEEP_SIZE and thus
		 * cannot fill the whole again in the subsequent
		 * fallocate(FL_KEEP_SIZE). Deny this FL_PUNCH_HOLE
		 * call to not end up with a hole in a non-sparse
		 * file.
		 */
		errno = ENOTSUP;
		return -1;
	}

	return SMB_VFS_NEXT_FALLOCATE(handle, fsp, mode, offset, len);
}

static int vfs_gpfs_ftruncate(vfs_handle_struct *handle, files_struct *fsp,
				off_t len)
{
	int result;
	struct gpfs_config_data *config;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if (!config->ftruncate) {
		return SMB_VFS_NEXT_FTRUNCATE(handle, fsp, len);
	}

	result = gpfswrap_ftruncate(fsp_get_io_fd(fsp), len);
	if ((result == -1) && (errno == ENOSYS)) {
		return SMB_VFS_NEXT_FTRUNCATE(handle, fsp, len);
	}
	return result;
}

static bool vfs_gpfs_is_offline(struct vfs_handle_struct *handle,
				struct files_struct *fsp,
				SMB_STRUCT_STAT *sbuf)
{
	struct gpfs_winattr attrs;
	struct gpfs_config_data *config;
	int ret;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return false);

	if (!config->winattr) {
		return false;
	}

	ret = gpfswrap_get_winattrs(fsp_get_pathref_fd(fsp), &attrs);
	if (ret == -1) {
		return false;
	}

	if ((attrs.winAttrs & GPFS_WINATTR_OFFLINE) != 0) {
		DBG_DEBUG("%s is offline\n", fsp_str_dbg(fsp));
		return true;
	}

	DBG_DEBUG("%s is online\n", fsp_str_dbg(fsp));
	return false;
}

static bool vfs_gpfs_fsp_is_offline(struct vfs_handle_struct *handle,
				    struct files_struct *fsp)
{
	struct gpfs_fsp_extension *ext;

	ext = VFS_FETCH_FSP_EXTENSION(handle, fsp);
	if (ext == NULL) {
		/*
		 * Something bad happened, always ask.
		 */
		return vfs_gpfs_is_offline(handle, fsp,
					   &fsp->fsp_name->st);
	}

	if (ext->offline) {
		/*
		 * As long as it's offline, ask.
		 */
		ext->offline = vfs_gpfs_is_offline(handle, fsp,
						   &fsp->fsp_name->st);
	}

	return ext->offline;
}

static bool vfs_gpfs_aio_force(struct vfs_handle_struct *handle,
			       struct files_struct *fsp)
{
	return vfs_gpfs_fsp_is_offline(handle, fsp);
}

static ssize_t vfs_gpfs_sendfile(vfs_handle_struct *handle, int tofd,
				 files_struct *fsp, const DATA_BLOB *hdr,
				 off_t offset, size_t n)
{
	if (vfs_gpfs_fsp_is_offline(handle, fsp)) {
		errno = ENOSYS;
		return -1;
	}
	return SMB_VFS_NEXT_SENDFILE(handle, tofd, fsp, hdr, offset, n);
}

#ifdef O_PATH
static int vfs_gpfs_check_pathref_fstat_x(struct gpfs_config_data *config,
					  struct connection_struct *conn)
{
	struct gpfs_iattr64 iattr = {0};
	unsigned int litemask = 0;
	int saved_errno;
	int fd;
	int ret;

	fd = open(conn->connectpath, O_PATH);
	if (fd == -1) {
		DBG_ERR("openat() of share with O_PATH failed: %s\n",
			strerror(errno));
		return -1;
	}

	ret = gpfswrap_fstat_x(fd, &litemask, &iattr, sizeof(iattr));
	if (ret == 0) {
		close(fd);
		config->pathref_ok.gpfs_fstat_x = true;
		return 0;
	}

	saved_errno = errno;
	ret = close(fd);
	if (ret != 0) {
		DBG_ERR("close failed: %s\n", strerror(errno));
		return -1;
	}

	if (saved_errno != EBADF) {
		DBG_ERR("gpfswrap_fstat_x() of O_PATH handle failed: %s\n",
			strerror(saved_errno));
		return -1;
	}

	return 0;
}
#endif

static int vfs_gpfs_check_pathref(struct gpfs_config_data *config,
				  struct connection_struct *conn)
{
#ifndef O_PATH
	/*
	 * This code path leaves all struct gpfs_config_data.pathref_ok members
	 * initialized to false.
	 */
	return 0;
#else
	int ret;

	ret = vfs_gpfs_check_pathref_fstat_x(config, conn);
	if (ret != 0) {
		return -1;
	}

	return 0;
#endif
}

static int vfs_gpfs_connect(struct vfs_handle_struct *handle,
			    const char *service, const char *user)
{
	struct gpfs_config_data *config;
	int ret;
	bool check_fstype;

	ret = SMB_VFS_NEXT_CONNECT(handle, service, user);
	if (ret < 0) {
		return ret;
	}

	if (IS_IPC(handle->conn)) {
		return 0;
	}

	ret = gpfswrap_init();
	if (ret < 0) {
		DBG_ERR("Could not load GPFS library.\n");
		return ret;
	}

	ret = gpfswrap_lib_init(0);
	if (ret < 0) {
		DBG_ERR("Could not open GPFS device file: %s\n",
			strerror(errno));
		return ret;
	}

	ret = gpfswrap_register_cifs_export();
	if (ret < 0) {
		DBG_ERR("Failed to register with GPFS: %s\n", strerror(errno));
		return ret;
	}

	config = talloc_zero(handle->conn, struct gpfs_config_data);
	if (!config) {
		DEBUG(0, ("talloc_zero() failed\n"));
		errno = ENOMEM;
		return -1;
	}

	check_fstype = lp_parm_bool(SNUM(handle->conn), "gpfs",
				    "check_fstype", true);

	if (check_fstype) {
		const char *connectpath = handle->conn->connectpath;
		struct statfs buf = { 0 };

		ret = statfs(connectpath, &buf);
		if (ret != 0) {
			DBG_ERR("statfs failed for share %s at path %s: %s\n",
				service, connectpath, strerror(errno));
			TALLOC_FREE(config);
			return ret;
		}

		if (buf.f_type != GPFS_SUPER_MAGIC) {
			DBG_ERR("SMB share %s, path %s not in GPFS file system."
				" statfs magic: 0x%jx\n",
				service,
				connectpath,
				(uintmax_t)buf.f_type);
			errno = EINVAL;
			TALLOC_FREE(config);
			return -1;
		}
	}

	ret = smbacl4_get_vfs_params(handle->conn, &config->nfs4_params);
	if (ret < 0) {
		TALLOC_FREE(config);
		return ret;
	}

	config->sharemodes = lp_parm_bool(SNUM(handle->conn), "gpfs",
					"sharemodes", true);

	config->leases = lp_parm_bool(SNUM(handle->conn), "gpfs",
					"leases", true);

	config->hsm = lp_parm_bool(SNUM(handle->conn), "gpfs",
				   "hsm", false);

	config->syncio = lp_parm_bool(SNUM(handle->conn), "gpfs",
				      "syncio", false);

	config->winattr = lp_parm_bool(SNUM(handle->conn), "gpfs",
				       "winattr", false);

	config->ftruncate = lp_parm_bool(SNUM(handle->conn), "gpfs",
					 "ftruncate", true);

	config->getrealfilename = lp_parm_bool(SNUM(handle->conn), "gpfs",
					       "getrealfilename", true);

	config->dfreequota = lp_parm_bool(SNUM(handle->conn), "gpfs",
					  "dfreequota", false);

	config->acl = lp_parm_bool(SNUM(handle->conn), "gpfs", "acl", true);

	config->settimes = lp_parm_bool(SNUM(handle->conn), "gpfs",
					"settimes", true);
	config->recalls = lp_parm_bool(SNUM(handle->conn), "gpfs",
				       "recalls", true);

	ret = vfs_gpfs_check_pathref(config, handle->conn);
	if (ret != 0) {
		DBG_ERR("vfs_gpfs_check_pathref() on [%s] failed\n",
			handle->conn->connectpath);
		TALLOC_FREE(config);
		return -1;
	}

	SMB_VFS_HANDLE_SET_DATA(handle, config,
				NULL, struct gpfs_config_data,
				return -1);

	if (config->leases) {
		/*
		 * GPFS lease code is based on kernel oplock code
		 * so make sure it is turned on
		 */
		if (!lp_kernel_oplocks(SNUM(handle->conn))) {
			DEBUG(5, ("Enabling kernel oplocks for "
				  "gpfs:leases to work\n"));
			lp_do_parameter(SNUM(handle->conn), "kernel oplocks",
					"true");
		}

		/*
		 * as the kernel does not properly support Level II oplocks
		 * and GPFS leases code is based on kernel infrastructure, we
		 * need to turn off Level II oplocks if gpfs:leases is enabled
		 */
		if (lp_level2_oplocks(SNUM(handle->conn))) {
			DEBUG(5, ("gpfs:leases are enabled, disabling "
				  "Level II oplocks\n"));
			lp_do_parameter(SNUM(handle->conn), "level2 oplocks",
					"false");
		}
	}

	/*
	 * Unless we have an async implementation of get_dos_attributes turn
	 * this off.
	 */
	lp_do_parameter(SNUM(handle->conn), "smbd async dosmode", "false");

	return 0;
}

static int get_gpfs_quota(const char *pathname, int type, int id,
			  struct gpfs_quotaInfo *qi)
{
	int ret;

	ret = gpfswrap_quotactl(pathname, GPFS_QCMD(Q_GETQUOTA, type), id, qi);

	if (ret) {
		if (errno == GPFS_E_NO_QUOTA_INST) {
			DEBUG(10, ("Quotas disabled on GPFS filesystem.\n"));
		} else if (errno != ENOSYS) {
			DEBUG(0, ("Get quota failed, type %d, id, %d, "
				  "errno %d.\n", type, id, errno));
		}

		return ret;
	}

	DEBUG(10, ("quota type %d, id %d, blk u:%lld h:%lld s:%lld gt:%u\n",
		   type, id, qi->blockUsage, qi->blockHardLimit,
		   qi->blockSoftLimit, qi->blockGraceTime));

	return ret;
}

static void vfs_gpfs_disk_free_quota(struct gpfs_quotaInfo qi, time_t cur_time,
				     uint64_t *dfree, uint64_t *dsize)
{
	uint64_t usage, limit;

	/*
	 * The quota reporting is done in units of 1024 byte blocks, but
	 * sys_fsusage uses units of 512 byte blocks, adjust the block number
	 * accordingly. Also filter possibly negative usage counts from gpfs.
	 */
	usage = qi.blockUsage < 0 ? 0 : (uint64_t)qi.blockUsage * 2;
	limit = (uint64_t)qi.blockHardLimit * 2;

	/*
	 * When the grace time for the exceeded soft block quota has been
	 * exceeded, the soft block quota becomes an additional hard limit.
	 */
	if (qi.blockSoftLimit &&
	    qi.blockGraceTime && cur_time > qi.blockGraceTime) {
		/* report disk as full */
		*dfree = 0;
		*dsize = MIN(*dsize, usage);
	}

	if (!qi.blockHardLimit)
		return;

	if (usage >= limit) {
		/* report disk as full */
		*dfree = 0;
		*dsize = MIN(*dsize, usage);

	} else {
		/* limit has not been reached, determine "free space" */
		*dfree = MIN(*dfree, limit - usage);
		*dsize = MIN(*dsize, limit);
	}
}

static uint64_t vfs_gpfs_disk_free(vfs_handle_struct *handle,
				const struct smb_filename *smb_fname,
				uint64_t *bsize,
				uint64_t *dfree,
				uint64_t *dsize)
{
	struct security_unix_token *utok;
	struct gpfs_quotaInfo qi_user = { 0 }, qi_group = { 0 };
	struct gpfs_config_data *config;
	int err;
	time_t cur_time;

	SMB_VFS_HANDLE_GET_DATA(handle, config, struct gpfs_config_data,
				return (uint64_t)-1);
	if (!config->dfreequota) {
		return SMB_VFS_NEXT_DISK_FREE(handle, smb_fname,
					      bsize, dfree, dsize);
	}

	err = sys_fsusage(smb_fname->base_name, dfree, dsize);
	if (err) {
		DEBUG (0, ("Could not get fs usage, errno %d\n", errno));
		return SMB_VFS_NEXT_DISK_FREE(handle, smb_fname,
					      bsize, dfree, dsize);
	}

	/* sys_fsusage returns units of 512 bytes */
	*bsize = 512;

	DEBUG(10, ("fs dfree %llu, dsize %llu\n",
		   (unsigned long long)*dfree, (unsigned long long)*dsize));

	utok = handle->conn->session_info->unix_token;

	err = get_gpfs_quota(smb_fname->base_name,
			GPFS_USRQUOTA, utok->uid, &qi_user);
	if (err) {
		return SMB_VFS_NEXT_DISK_FREE(handle, smb_fname,
					      bsize, dfree, dsize);
	}

	/*
	 * If new files created under this folder get this folder's
	 * GID, then available space is governed by the quota of the
	 * folder's GID, not the primary group of the creating user.
	 */
	if (VALID_STAT(smb_fname->st) &&
	    S_ISDIR(smb_fname->st.st_ex_mode) &&
	    smb_fname->st.st_ex_mode & S_ISGID) {
		become_root();
		err = get_gpfs_quota(smb_fname->base_name, GPFS_GRPQUOTA,
				     smb_fname->st.st_ex_gid, &qi_group);
		unbecome_root();

	} else {
		err = get_gpfs_quota(smb_fname->base_name, GPFS_GRPQUOTA,
				     utok->gid, &qi_group);
	}

	if (err) {
		return SMB_VFS_NEXT_DISK_FREE(handle, smb_fname,
					      bsize, dfree, dsize);
	}

	cur_time = time(NULL);

	/* Adjust free space and size according to quota limits. */
	vfs_gpfs_disk_free_quota(qi_user, cur_time, dfree, dsize);
	vfs_gpfs_disk_free_quota(qi_group, cur_time, dfree, dsize);

	return *dfree / 2;
}

static int vfs_gpfs_get_quota(vfs_handle_struct *handle,
				const struct smb_filename *smb_fname,
				enum SMB_QUOTA_TYPE qtype,
				unid_t id,
				SMB_DISK_QUOTA *dq)
{
	switch(qtype) {
		/*
		 * User/group quota are being used for disk-free
		 * determination, which in this module is done directly
		 * by the disk-free function. It's important that this
		 * module does not return wrong quota values by mistake,
		 * which would modify the correct values set by disk-free.
		 * User/group quota are also being used for processing
		 * NT_TRANSACT_GET_USER_QUOTA in smb1 protocol, which is
		 * currently not supported by this module.
		 */
		case SMB_USER_QUOTA_TYPE:
		case SMB_GROUP_QUOTA_TYPE:
			errno = ENOSYS;
			return -1;
		default:
			return SMB_VFS_NEXT_GET_QUOTA(handle, smb_fname,
					qtype, id, dq);
	}
}

static uint32_t vfs_gpfs_capabilities(struct vfs_handle_struct *handle,
				      enum timestamp_set_resolution *p_ts_res)
{
	struct gpfs_config_data *config;
	uint32_t next;

	next = SMB_VFS_NEXT_FS_CAPABILITIES(handle, p_ts_res);

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return next);

	if (config->hsm) {
		next |= FILE_SUPPORTS_REMOTE_STORAGE;
	}
	return next;
}

static int vfs_gpfs_openat(struct vfs_handle_struct *handle,
			   const struct files_struct *dirfsp,
			   const struct smb_filename *smb_fname,
			   files_struct *fsp,
			   const struct vfs_open_how *_how)
{
	struct vfs_open_how how = *_how;
	struct gpfs_config_data *config = NULL;
	struct gpfs_fsp_extension *ext = NULL;
	int ret;

	SMB_VFS_HANDLE_GET_DATA(handle, config,
				struct gpfs_config_data,
				return -1);

	if (config->hsm && !config->recalls &&
	    !fsp->fsp_flags.is_pathref &&
	    vfs_gpfs_fsp_is_offline(handle, fsp))
	{
		DBG_DEBUG("Refusing access to offline file %s\n",
			  fsp_str_dbg(fsp));
		errno = EACCES;
		return -1;
	}

	if (config->syncio) {
		how.flags |= O_SYNC;
	}

	ext = VFS_ADD_FSP_EXTENSION(handle, fsp, struct gpfs_fsp_extension,
				    NULL);
	if (ext == NULL) {
		errno = ENOMEM;
		return -1;
	}

	/*
	 * Assume the file is offline until gpfs tells us it's online.
	 */
	*ext = (struct gpfs_fsp_extension) { .offline = true };

	ret = SMB_VFS_NEXT_OPENAT(handle, dirfsp, smb_fname, fsp, &how);
	if (ret == -1) {
		VFS_REMOVE_FSP_EXTENSION(handle, fsp);
	}
	return ret;
}

static ssize_t vfs_gpfs_pread(vfs_handle_struct *handle, files_struct *fsp,
			      void *data, size_t n, off_t offset)
{
	ssize_t ret;
	bool was_offline;

	was_offline = vfs_gpfs_fsp_is_offline(handle, fsp);

	ret = SMB_VFS_NEXT_PREAD(handle, fsp, data, n, offset);

	if ((ret != -1) && was_offline) {
		notify_fname(handle->conn, NOTIFY_ACTION_MODIFIED,
			     FILE_NOTIFY_CHANGE_ATTRIBUTES,
			     fsp->fsp_name->base_name);
	}

	return ret;
}

struct vfs_gpfs_pread_state {
	struct files_struct *fsp;
	ssize_t ret;
	bool was_offline;
	struct vfs_aio_state vfs_aio_state;
};

static void vfs_gpfs_pread_done(struct tevent_req *subreq);

static struct tevent_req *vfs_gpfs_pread_send(struct vfs_handle_struct *handle,
					      TALLOC_CTX *mem_ctx,
					      struct tevent_context *ev,
					      struct files_struct *fsp,
					      void *data, size_t n,
					      off_t offset)
{
	struct tevent_req *req, *subreq;
	struct vfs_gpfs_pread_state *state;

	req = tevent_req_create(mem_ctx, &state, struct vfs_gpfs_pread_state);
	if (req == NULL) {
		return NULL;
	}
	state->was_offline = vfs_gpfs_fsp_is_offline(handle, fsp);
	state->fsp = fsp;
	subreq = SMB_VFS_NEXT_PREAD_SEND(state, ev, handle, fsp, data,
					 n, offset);
	if (tevent_req_nomem(subreq, req)) {
		return tevent_req_post(req, ev);
	}
	tevent_req_set_callback(subreq, vfs_gpfs_pread_done, req);
	return req;
}

static void vfs_gpfs_pread_done(struct tevent_req *subreq)
{
	struct tevent_req *req = tevent_req_callback_data(
		subreq, struct tevent_req);
	struct vfs_gpfs_pread_state *state = tevent_req_data(
		req, struct vfs_gpfs_pread_state);

	state->ret = SMB_VFS_PREAD_RECV(subreq, &state->vfs_aio_state);
	TALLOC_FREE(subreq);
	tevent_req_done(req);
}

static ssize_t vfs_gpfs_pread_recv(struct tevent_req *req,
				   struct vfs_aio_state *vfs_aio_state)
{
	struct vfs_gpfs_pread_state *state = tevent_req_data(
		req, struct vfs_gpfs_pread_state);
	struct files_struct *fsp = state->fsp;

	if (tevent_req_is_unix_error(req, &vfs_aio_state->error)) {
		return -1;
	}
	*vfs_aio_state = state->vfs_aio_state;

	if ((state->ret != -1) && state->was_offline) {
		DEBUG(10, ("sending notify\n"));
		notify_fname(fsp->conn, NOTIFY_ACTION_MODIFIED,
			     FILE_NOTIFY_CHANGE_ATTRIBUTES,
			     fsp->fsp_name->base_name);
	}

	return state->ret;
}

static ssize_t vfs_gpfs_pwrite(vfs_handle_struct *handle, files_struct *fsp,
			       const void *data, size_t n, off_t offset)
{
	ssize_t ret;
	bool was_offline;

	was_offline = vfs_gpfs_fsp_is_offline(handle, fsp);

	ret = SMB_VFS_NEXT_PWRITE(handle, fsp, data, n, offset);

	if ((ret != -1) && was_offline) {
		notify_fname(handle->conn, NOTIFY_ACTION_MODIFIED,
			     FILE_NOTIFY_CHANGE_ATTRIBUTES,
			     fsp->fsp_name->base_name);
	}

	return ret;
}

struct vfs_gpfs_pwrite_state {
	struct files_struct *fsp;
	ssize_t ret;
	bool was_offline;
	struct vfs_aio_state vfs_aio_state;
};

static void vfs_gpfs_pwrite_done(struct tevent_req *subreq);

static struct tevent_req *vfs_gpfs_pwrite_send(
	struct vfs_handle_struct *handle,
	TALLOC_CTX *mem_ctx,
	struct tevent_context *ev,
	struct files_struct *fsp,
	const void *data, size_t n,
	off_t offset)
{
	struct tevent_req *req, *subreq;
	struct vfs_gpfs_pwrite_state *state;

	req = tevent_req_create(mem_ctx, &state, struct vfs_gpfs_pwrite_state);
	if (req == NULL) {
		return NULL;
	}
	state->was_offline = vfs_gpfs_fsp_is_offline(handle, fsp);
	state->fsp = fsp;
	subreq = SMB_VFS_NEXT_PWRITE_SEND(state, ev, handle, fsp, data,
					 n, offset);
	if (tevent_req_nomem(subreq, req)) {
		return tevent_req_post(req, ev);
	}
	tevent_req_set_callback(subreq, vfs_gpfs_pwrite_done, req);
	return req;
}

static void vfs_gpfs_pwrite_done(struct tevent_req *subreq)
{
	struct tevent_req *req = tevent_req_callback_data(
		subreq, struct tevent_req);
	struct vfs_gpfs_pwrite_state *state = tevent_req_data(
		req, struct vfs_gpfs_pwrite_state);

	state->ret = SMB_VFS_PWRITE_RECV(subreq, &state->vfs_aio_state);
	TALLOC_FREE(subreq);
	tevent_req_done(req);
}

static ssize_t vfs_gpfs_pwrite_recv(struct tevent_req *req,
				    struct vfs_aio_state *vfs_aio_state)
{
	struct vfs_gpfs_pwrite_state *state = tevent_req_data(
		req, struct vfs_gpfs_pwrite_state);
	struct files_struct *fsp = state->fsp;

	if (tevent_req_is_unix_error(req, &vfs_aio_state->error)) {
		return -1;
	}
	*vfs_aio_state = state->vfs_aio_state;

	if ((state->ret != -1) && state->was_offline) {
		DEBUG(10, ("sending notify\n"));
		notify_fname(fsp->conn, NOTIFY_ACTION_MODIFIED,
			     FILE_NOTIFY_CHANGE_ATTRIBUTES,
			     fsp->fsp_name->base_name);
	}

	return state->ret;
}


static struct vfs_fn_pointers vfs_gpfs_fns = {
	.connect_fn = vfs_gpfs_connect,
	.disk_free_fn = vfs_gpfs_disk_free,
	.get_quota_fn = vfs_gpfs_get_quota,
	.fs_capabilities_fn = vfs_gpfs_capabilities,
	.filesystem_sharemode_fn = vfs_gpfs_filesystem_sharemode,
	.linux_setlease_fn = vfs_gpfs_setlease,
	.get_real_filename_at_fn = vfs_gpfs_get_real_filename_at,
	.get_dos_attributes_send_fn = vfs_not_implemented_get_dos_attributes_send,
	.get_dos_attributes_recv_fn = vfs_not_implemented_get_dos_attributes_recv,
	.fget_dos_attributes_fn = vfs_gpfs_fget_dos_attributes,
	.fset_dos_attributes_fn = vfs_gpfs_fset_dos_attributes,
	.fget_nt_acl_fn = gpfsacl_fget_nt_acl,
	.fset_nt_acl_fn = gpfsacl_fset_nt_acl,
	.sys_acl_get_fd_fn = gpfsacl_sys_acl_get_fd,
	.sys_acl_blob_get_fd_fn = gpfsacl_sys_acl_blob_get_fd,
	.sys_acl_set_fd_fn = gpfsacl_sys_acl_set_fd,
	.sys_acl_delete_def_fd_fn = gpfsacl_sys_acl_delete_def_fd,
	.fchmod_fn = vfs_gpfs_fchmod,
	.close_fn = vfs_gpfs_close,
	.stat_fn = nfs4_acl_stat,
	.fstat_fn = nfs4_acl_fstat,
	.lstat_fn = nfs4_acl_lstat,
	.fstatat_fn = nfs4_acl_fstatat,
	.fntimes_fn = vfs_gpfs_fntimes,
	.aio_force_fn = vfs_gpfs_aio_force,
	.sendfile_fn = vfs_gpfs_sendfile,
	.fallocate_fn = vfs_gpfs_fallocate,
	.openat_fn = vfs_gpfs_openat,
	.pread_fn = vfs_gpfs_pread,
	.pread_send_fn = vfs_gpfs_pread_send,
	.pread_recv_fn = vfs_gpfs_pread_recv,
	.pwrite_fn = vfs_gpfs_pwrite,
	.pwrite_send_fn = vfs_gpfs_pwrite_send,
	.pwrite_recv_fn = vfs_gpfs_pwrite_recv,
	.ftruncate_fn = vfs_gpfs_ftruncate
};

static_decl_vfs;
NTSTATUS vfs_gpfs_init(TALLOC_CTX *ctx)
{
	return smb_register_vfs(SMB_VFS_INTERFACE_VERSION, "gpfs",
				&vfs_gpfs_fns);
}