aboutsummaryrefslogtreecommitdiffstats
path: root/fs/ocfs2/file.c
blob: a80f31776d94beacc2383b53fe04dd7b1808077b (plain) (blame)
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
/* -*- mode: c; c-basic-offset: 8; -*-
 * vim: noexpandtab sw=8 ts=8 sts=0:
 *
 * file.c
 *
 * File open, close, extend, truncate
 *
 * Copyright (C) 2002, 2004 Oracle.  All rights reserved.
 *
 * 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 2 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, write to the
 * Free Software Foundation, Inc., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#include <linux/capability.h>
#include <linux/fs.h>
#include <linux/types.h>
#include <linux/slab.h>
#include <linux/highmem.h>
#include <linux/pagemap.h>
#include <linux/uio.h>
#include <linux/sched.h>
#include <linux/splice.h>
#include <linux/mount.h>
#include <linux/writeback.h>

#define MLOG_MASK_PREFIX ML_INODE
#include <cluster/masklog.h>

#include "ocfs2.h"

#include "alloc.h"
#include "aops.h"
#include "dir.h"
#include "dlmglue.h"
#include "extent_map.h"
#include "file.h"
#include "sysfile.h"
#include "inode.h"
#include "ioctl.h"
#include "journal.h"
#include "mmap.h"
#include "suballoc.h"
#include "super.h"

#include "buffer_head_io.h"

static int ocfs2_sync_inode(struct inode *inode)
{
	filemap_fdatawrite(inode->i_mapping);
	return sync_mapping_buffers(inode->i_mapping);
}

static int ocfs2_file_open(struct inode *inode, struct file *file)
{
	int status;
	int mode = file->f_flags;
	struct ocfs2_inode_info *oi = OCFS2_I(inode);

	mlog_entry("(0x%p, 0x%p, '%.*s')\n", inode, file,
		   file->f_path.dentry->d_name.len, file->f_path.dentry->d_name.name);

	spin_lock(&oi->ip_lock);

	/* Check that the inode hasn't been wiped from disk by another
	 * node. If it hasn't then we're safe as long as we hold the
	 * spin lock until our increment of open count. */
	if (OCFS2_I(inode)->ip_flags & OCFS2_INODE_DELETED) {
		spin_unlock(&oi->ip_lock);

		status = -ENOENT;
		goto leave;
	}

	if (mode & O_DIRECT)
		oi->ip_flags |= OCFS2_INODE_OPEN_DIRECT;

	oi->ip_open_count++;
	spin_unlock(&oi->ip_lock);
	status = 0;
leave:
	mlog_exit(status);
	return status;
}

static int ocfs2_file_release(struct inode *inode, struct file *file)
{
	struct ocfs2_inode_info *oi = OCFS2_I(inode);

	mlog_entry("(0x%p, 0x%p, '%.*s')\n", inode, file,
		       file->f_path.dentry->d_name.len,
		       file->f_path.dentry->d_name.name);

	spin_lock(&oi->ip_lock);
	if (!--oi->ip_open_count)
		oi->ip_flags &= ~OCFS2_INODE_OPEN_DIRECT;
	spin_unlock(&oi->ip_lock);

	mlog_exit(0);

	return 0;
}

static int ocfs2_sync_file(struct file *file,
			   struct dentry *dentry,
			   int datasync)
{
	int err = 0;
	journal_t *journal;
	struct inode *inode = dentry->d_inode;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	mlog_entry("(0x%p, 0x%p, %d, '%.*s')\n", file, dentry, datasync,
		   dentry->d_name.len, dentry->d_name.name);

	err = ocfs2_sync_inode(dentry->d_inode);
	if (err)
		goto bail;

	journal = osb->journal->j_journal;
	err = journal_force_commit(journal);

bail:
	mlog_exit(err);

	return (err < 0) ? -EIO : 0;
}

int ocfs2_should_update_atime(struct inode *inode,
			      struct vfsmount *vfsmnt)
{
	struct timespec now;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	if (ocfs2_is_hard_readonly(osb) || ocfs2_is_soft_readonly(osb))
		return 0;

	if ((inode->i_flags & S_NOATIME) ||
	    ((inode->i_sb->s_flags & MS_NODIRATIME) && S_ISDIR(inode->i_mode)))
		return 0;

	/*
	 * We can be called with no vfsmnt structure - NFSD will
	 * sometimes do this.
	 *
	 * Note that our action here is different than touch_atime() -
	 * if we can't tell whether this is a noatime mount, then we
	 * don't know whether to trust the value of s_atime_quantum.
	 */
	if (vfsmnt == NULL)
		return 0;

	if ((vfsmnt->mnt_flags & MNT_NOATIME) ||
	    ((vfsmnt->mnt_flags & MNT_NODIRATIME) && S_ISDIR(inode->i_mode)))
		return 0;

	if (vfsmnt->mnt_flags & MNT_RELATIME) {
		if ((timespec_compare(&inode->i_atime, &inode->i_mtime) <= 0) ||
		    (timespec_compare(&inode->i_atime, &inode->i_ctime) <= 0))
			return 1;

		return 0;
	}

	now = CURRENT_TIME;
	if ((now.tv_sec - inode->i_atime.tv_sec <= osb->s_atime_quantum))
		return 0;
	else
		return 1;
}

int ocfs2_update_inode_atime(struct inode *inode,
			     struct buffer_head *bh)
{
	int ret;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	handle_t *handle;

	mlog_entry_void();

	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
	if (handle == NULL) {
		ret = -ENOMEM;
		mlog_errno(ret);
		goto out;
	}

	inode->i_atime = CURRENT_TIME;
	ret = ocfs2_mark_inode_dirty(handle, inode, bh);
	if (ret < 0)
		mlog_errno(ret);

	ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
out:
	mlog_exit(ret);
	return ret;
}

static int ocfs2_set_inode_size(handle_t *handle,
				struct inode *inode,
				struct buffer_head *fe_bh,
				u64 new_i_size)
{
	int status;

	mlog_entry_void();
	i_size_write(inode, new_i_size);
	inode->i_blocks = ocfs2_inode_sector_count(inode);
	inode->i_ctime = inode->i_mtime = CURRENT_TIME;

	status = ocfs2_mark_inode_dirty(handle, inode, fe_bh);
	if (status < 0) {
		mlog_errno(status);
		goto bail;
	}

bail:
	mlog_exit(status);
	return status;
}

static int ocfs2_simple_size_update(struct inode *inode,
				    struct buffer_head *di_bh,
				    u64 new_i_size)
{
	int ret;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	handle_t *handle = NULL;

	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
	if (handle == NULL) {
		ret = -ENOMEM;
		mlog_errno(ret);
		goto out;
	}

	ret = ocfs2_set_inode_size(handle, inode, di_bh,
				   new_i_size);
	if (ret < 0)
		mlog_errno(ret);

	ocfs2_commit_trans(osb, handle);
out:
	return ret;
}

static int ocfs2_orphan_for_truncate(struct ocfs2_super *osb,
				     struct inode *inode,
				     struct buffer_head *fe_bh,
				     u64 new_i_size)
{
	int status;
	handle_t *handle;
	struct ocfs2_dinode *di;

	mlog_entry_void();

	/* TODO: This needs to actually orphan the inode in this
	 * transaction. */

	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
	if (IS_ERR(handle)) {
		status = PTR_ERR(handle);
		mlog_errno(status);
		goto out;
	}

	status = ocfs2_journal_access(handle, inode, fe_bh,
				      OCFS2_JOURNAL_ACCESS_WRITE);
	if (status < 0) {
		mlog_errno(status);
		goto out_commit;
	}

	/*
	 * Do this before setting i_size.
	 */
	status = ocfs2_zero_tail_for_truncate(inode, handle, new_i_size);
	if (status) {
		mlog_errno(status);
		goto out_commit;
	}

	i_size_write(inode, new_i_size);
	inode->i_blocks = ocfs2_align_bytes_to_sectors(new_i_size);
	inode->i_ctime = inode->i_mtime = CURRENT_TIME;

	di = (struct ocfs2_dinode *) fe_bh->b_data;
	di->i_size = cpu_to_le64(new_i_size);
	di->i_ctime = di->i_mtime = cpu_to_le64(inode->i_ctime.tv_sec);
	di->i_ctime_nsec = di->i_mtime_nsec = cpu_to_le32(inode->i_ctime.tv_nsec);

	status = ocfs2_journal_dirty(handle, fe_bh);
	if (status < 0)
		mlog_errno(status);

out_commit:
	ocfs2_commit_trans(osb, handle);
out:

	mlog_exit(status);
	return status;
}

static int ocfs2_truncate_file(struct inode *inode,
			       struct buffer_head *di_bh,
			       u64 new_i_size)
{
	int status = 0;
	struct ocfs2_dinode *fe = NULL;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	struct ocfs2_truncate_context *tc = NULL;

	mlog_entry("(inode = %llu, new_i_size = %llu\n",
		   (unsigned long long)OCFS2_I(inode)->ip_blkno,
		   (unsigned long long)new_i_size);

	fe = (struct ocfs2_dinode *) di_bh->b_data;
	if (!OCFS2_IS_VALID_DINODE(fe)) {
		OCFS2_RO_ON_INVALID_DINODE(inode->i_sb, fe);
		status = -EIO;
		goto bail;
	}

	mlog_bug_on_msg(le64_to_cpu(fe->i_size) != i_size_read(inode),
			"Inode %llu, inode i_size = %lld != di "
			"i_size = %llu, i_flags = 0x%x\n",
			(unsigned long long)OCFS2_I(inode)->ip_blkno,
			i_size_read(inode),
			(unsigned long long)le64_to_cpu(fe->i_size),
			le32_to_cpu(fe->i_flags));

	if (new_i_size > le64_to_cpu(fe->i_size)) {
		mlog(0, "asked to truncate file with size (%llu) to size (%llu)!\n",
		     (unsigned long long)le64_to_cpu(fe->i_size),
		     (unsigned long long)new_i_size);
		status = -EINVAL;
		mlog_errno(status);
		goto bail;
	}

	mlog(0, "inode %llu, i_size = %llu, new_i_size = %llu\n",
	     (unsigned long long)le64_to_cpu(fe->i_blkno),
	     (unsigned long long)le64_to_cpu(fe->i_size),
	     (unsigned long long)new_i_size);

	/* lets handle the simple truncate cases before doing any more
	 * cluster locking. */
	if (new_i_size == le64_to_cpu(fe->i_size))
		goto bail;

	down_write(&OCFS2_I(inode)->ip_alloc_sem);

	/* This forces other nodes to sync and drop their pages. Do
	 * this even if we have a truncate without allocation change -
	 * ocfs2 cluster sizes can be much greater than page size, so
	 * we have to truncate them anyway.  */
	status = ocfs2_data_lock(inode, 1);
	if (status < 0) {
		up_write(&OCFS2_I(inode)->ip_alloc_sem);

		mlog_errno(status);
		goto bail;
	}

	unmap_mapping_range(inode->i_mapping, new_i_size + PAGE_SIZE - 1, 0, 1);
	truncate_inode_pages(inode->i_mapping, new_i_size);

	/* alright, we're going to need to do a full blown alloc size
	 * change. Orphan the inode so that recovery can complete the
	 * truncate if necessary. This does the task of marking
	 * i_size. */
	status = ocfs2_orphan_for_truncate(osb, inode, di_bh, new_i_size);
	if (status < 0) {
		mlog_errno(status);
		goto bail_unlock_data;
	}

	status = ocfs2_prepare_truncate(osb, inode, di_bh, &tc);
	if (status < 0) {
		mlog_errno(status);
		goto bail_unlock_data;
	}

	status = ocfs2_commit_truncate(osb, inode, di_bh, tc);
	if (status < 0) {
		mlog_errno(status);
		goto bail_unlock_data;
	}

	/* TODO: orphan dir cleanup here. */
bail_unlock_data:
	ocfs2_data_unlock(inode, 1);

	up_write(&OCFS2_I(inode)->ip_alloc_sem);

bail:

	mlog_exit(status);
	return status;
}

/*
 * extend allocation only here.
 * we'll update all the disk stuff, and oip->alloc_size
 *
 * expect stuff to be locked, a transaction started and enough data /
 * metadata reservations in the contexts.
 *
 * Will return -EAGAIN, and a reason if a restart is needed.
 * If passed in, *reason will always be set, even in error.
 */
int ocfs2_do_extend_allocation(struct ocfs2_super *osb,
			       struct inode *inode,
			       u32 *logical_offset,
			       u32 clusters_to_add,
			       struct buffer_head *fe_bh,
			       handle_t *handle,
			       struct ocfs2_alloc_context *data_ac,
			       struct ocfs2_alloc_context *meta_ac,
			       enum ocfs2_alloc_restarted *reason_ret)
{
	int status = 0;
	int free_extents;
	struct ocfs2_dinode *fe = (struct ocfs2_dinode *) fe_bh->b_data;
	enum ocfs2_alloc_restarted reason = RESTART_NONE;
	u32 bit_off, num_bits;
	u64 block;

	BUG_ON(!clusters_to_add);

	free_extents = ocfs2_num_free_extents(osb, inode, fe);
	if (free_extents < 0) {
		status = free_extents;
		mlog_errno(status);
		goto leave;
	}

	/* there are two cases which could cause us to EAGAIN in the
	 * we-need-more-metadata case:
	 * 1) we haven't reserved *any*
	 * 2) we are so fragmented, we've needed to add metadata too
	 *    many times. */
	if (!free_extents && !meta_ac) {
		mlog(0, "we haven't reserved any metadata!\n");
		status = -EAGAIN;
		reason = RESTART_META;
		goto leave;
	} else if ((!free_extents)
		   && (ocfs2_alloc_context_bits_left(meta_ac)
		       < ocfs2_extend_meta_needed(fe))) {
		mlog(0, "filesystem is really fragmented...\n");
		status = -EAGAIN;
		reason = RESTART_META;
		goto leave;
	}

	status = ocfs2_claim_clusters(osb, handle, data_ac, 1,
				      &bit_off, &num_bits);
	if (status < 0) {
		if (status != -ENOSPC)
			mlog_errno(status);
		goto leave;
	}

	BUG_ON(num_bits > clusters_to_add);

	/* reserve our write early -- insert_extent may update the inode */
	status = ocfs2_journal_access(handle, inode, fe_bh,
				      OCFS2_JOURNAL_ACCESS_WRITE);
	if (status < 0) {
		mlog_errno(status);
		goto leave;
	}

	block = ocfs2_clusters_to_blocks(osb->sb, bit_off);
	mlog(0, "Allocating %u clusters at block %u for inode %llu\n",
	     num_bits, bit_off, (unsigned long long)OCFS2_I(inode)->ip_blkno);
	status = ocfs2_insert_extent(osb, handle, inode, fe_bh,
				     *logical_offset, block, num_bits,
				     meta_ac);
	if (status < 0) {
		mlog_errno(status);
		goto leave;
	}

	status = ocfs2_journal_dirty(handle, fe_bh);
	if (status < 0) {
		mlog_errno(status);
		goto leave;
	}

	clusters_to_add -= num_bits;
	*logical_offset += num_bits;

	if (clusters_to_add) {
		mlog(0, "need to alloc once more, clusters = %u, wanted = "
		     "%u\n", fe->i_clusters, clusters_to_add);
		status = -EAGAIN;
		reason = RESTART_TRANS;
	}

leave:
	mlog_exit(status);
	if (reason_ret)
		*reason_ret = reason;
	return status;
}

/*
 * For a given allocation, determine which allocators will need to be
 * accessed, and lock them, reserving the appropriate number of bits.
 *
 * Called from ocfs2_extend_allocation() for file systems which don't
 * support holes, and from ocfs2_write() for file systems which
 * understand sparse inodes.
 */
int ocfs2_lock_allocators(struct inode *inode, struct ocfs2_dinode *di,
			  u32 clusters_to_add,
			  struct ocfs2_alloc_context **data_ac,
			  struct ocfs2_alloc_context **meta_ac)
{
	int ret, num_free_extents;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	*meta_ac = NULL;
	*data_ac = NULL;

	mlog(0, "extend inode %llu, i_size = %lld, di->i_clusters = %u, "
	     "clusters_to_add = %u\n",
	     (unsigned long long)OCFS2_I(inode)->ip_blkno, i_size_read(inode),
	     le32_to_cpu(di->i_clusters), clusters_to_add);

	num_free_extents = ocfs2_num_free_extents(osb, inode, di);
	if (num_free_extents < 0) {
		ret = num_free_extents;
		mlog_errno(ret);
		goto out;
	}

	/*
	 * Sparse allocation file systems need to be more conservative
	 * with reserving room for expansion - the actual allocation
	 * happens while we've got a journal handle open so re-taking
	 * a cluster lock (because we ran out of room for another
	 * extent) will violate ordering rules.
	 *
	 * Most of the time we'll only be seeing this 1 cluster at a time
	 * anyway.
	 */
	if (!num_free_extents ||
	    (ocfs2_sparse_alloc(osb) && num_free_extents < clusters_to_add)) {
		ret = ocfs2_reserve_new_metadata(osb, di, meta_ac);
		if (ret < 0) {
			if (ret != -ENOSPC)
				mlog_errno(ret);
			goto out;
		}
	}

	ret = ocfs2_reserve_clusters(osb, clusters_to_add, data_ac);
	if (ret < 0) {
		if (ret != -ENOSPC)
			mlog_errno(ret);
		goto out;
	}

out:
	if (ret) {
		if (*meta_ac) {
			ocfs2_free_alloc_context(*meta_ac);
			*meta_ac = NULL;
		}

		/*
		 * We cannot have an error and a non null *data_ac.
		 */
	}

	return ret;
}

static int ocfs2_extend_allocation(struct inode *inode,
				   u32 clusters_to_add)
{
	int status = 0;
	int restart_func = 0;
	int drop_alloc_sem = 0;
	int credits;
	u32 prev_clusters, logical_start;
	struct buffer_head *bh = NULL;
	struct ocfs2_dinode *fe = NULL;
	handle_t *handle = NULL;
	struct ocfs2_alloc_context *data_ac = NULL;
	struct ocfs2_alloc_context *meta_ac = NULL;
	enum ocfs2_alloc_restarted why;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);

	mlog_entry("(clusters_to_add = %u)\n", clusters_to_add);

	/*
	 * This function only exists for file systems which don't
	 * support holes.
	 */
	BUG_ON(ocfs2_sparse_alloc(osb));

	status = ocfs2_read_block(osb, OCFS2_I(inode)->ip_blkno, &bh,
				  OCFS2_BH_CACHED, inode);
	if (status < 0) {
		mlog_errno(status);
		goto leave;
	}

	fe = (struct ocfs2_dinode *) bh->b_data;
	if (!OCFS2_IS_VALID_DINODE(fe)) {
		OCFS2_RO_ON_INVALID_DINODE(inode->i_sb, fe);
		status = -EIO;
		goto leave;
	}

	logical_start = OCFS2_I(inode)->ip_clusters;

restart_all:
	BUG_ON(le32_to_cpu(fe->i_clusters) != OCFS2_I(inode)->ip_clusters);

	/* blocks peope in read/write from reading our allocation
	 * until we're done changing it. We depend on i_mutex to block
	 * other extend/truncate calls while we're here. Ordering wrt
	 * start_trans is important here -- always do it before! */
	down_write(&OCFS2_I(inode)->ip_alloc_sem);
	drop_alloc_sem = 1;

	status = ocfs2_lock_allocators(inode, fe, clusters_to_add, &data_ac,
				       &meta_ac);
	if (status) {
		mlog_errno(status);
		goto leave;
	}

	credits = ocfs2_calc_extend_credits(osb->sb, fe, clusters_to_add);
	handle = ocfs2_start_trans(osb, credits);
	if (IS_ERR(handle)) {
		status = PTR_ERR(handle);
		handle = NULL;
		mlog_errno(status);
		goto leave;
	}

restarted_transaction:
	/* reserve a write to the file entry early on - that we if we
	 * run out of credits in the allocation path, we can still
	 * update i_size. */
	status = ocfs2_journal_access(handle, inode, bh,
				      OCFS2_JOURNAL_ACCESS_WRITE);
	if (status < 0) {
		mlog_errno(status);
		goto leave;
	}

	prev_clusters = OCFS2_I(inode)->ip_clusters;

	status = ocfs2_do_extend_allocation(osb,
					    inode,
					    &logical_start,
					    clusters_to_add,
					    bh,
					    handle,
					    data_ac,
					    meta_ac,
					    &why);
	if ((status < 0) && (status != -EAGAIN)) {
		if (status != -ENOSPC)
			mlog_errno(status);
		goto leave;
	}

	status = ocfs2_journal_dirty(handle, bh);
	if (status < 0) {
		mlog_errno(status);
		goto leave;
	}

	spin_lock(&OCFS2_I(inode)->ip_lock);
	clusters_to_add -= (OCFS2_I(inode)->ip_clusters - prev_clusters);
	spin_unlock(&OCFS2_I(inode)->ip_lock);

	if (why != RESTART_NONE && clusters_to_add) {
		if (why == RESTART_META) {
			mlog(0, "restarting function.\n");
			restart_func = 1;
		} else {
			BUG_ON(why != RESTART_TRANS);

			mlog(0, "restarting transaction.\n");
			/* TODO: This can be more intelligent. */
			credits = ocfs2_calc_extend_credits(osb->sb,
							    fe,
							    clusters_to_add);
			status = ocfs2_extend_trans(handle, credits);
			if (status < 0) {
				/* handle still has to be committed at
				 * this point. */
				status = -ENOMEM;
				mlog_errno(status);
				goto leave;
			}
			goto restarted_transaction;
		}
	}

	mlog(0, "fe: i_clusters = %u, i_size=%llu\n",
	     le32_to_cpu(fe->i_clusters),
	     (unsigned long long)le64_to_cpu(fe->i_size));
	mlog(0, "inode: ip_clusters=%u, i_size=%lld\n",
	     OCFS2_I(inode)->ip_clusters, i_size_read(inode));

leave:
	if (drop_alloc_sem) {
		up_write(&OCFS2_I(inode)->ip_alloc_sem);
		drop_alloc_sem = 0;
	}
	if (handle) {
		ocfs2_commit_trans(osb, handle);
		handle = NULL;
	}
	if (data_ac) {
		ocfs2_free_alloc_context(data_ac);
		data_ac = NULL;
	}
	if (meta_ac) {
		ocfs2_free_alloc_context(meta_ac);
		meta_ac = NULL;
	}
	if ((!status) && restart_func) {
		restart_func = 0;
		goto restart_all;
	}
	if (bh) {
		brelse(bh);
		bh = NULL;
	}

	mlog_exit(status);
	return status;
}

/* Some parts of this taken from generic_cont_expand, which turned out
 * to be too fragile to do exactly what we need without us having to
 * worry about recursive locking in ->prepare_write() and
 * ->commit_write(). */
static int ocfs2_write_zero_page(struct inode *inode,
				 u64 size)
{
	struct address_space *mapping = inode->i_mapping;
	struct page *page;
	unsigned long index;
	unsigned int offset;
	handle_t *handle = NULL;
	int ret;

	offset = (size & (PAGE_CACHE_SIZE-1)); /* Within page */
	/* ugh.  in prepare/commit_write, if from==to==start of block, we 
	** skip the prepare.  make sure we never send an offset for the start
	** of a block
	*/
	if ((offset & (inode->i_sb->s_blocksize - 1)) == 0) {
		offset++;
	}
	index = size >> PAGE_CACHE_SHIFT;

	page = grab_cache_page(mapping, index);
	if (!page) {
		ret = -ENOMEM;
		mlog_errno(ret);
		goto out;
	}

	ret = ocfs2_prepare_write_nolock(inode, page, offset, offset);
	if (ret < 0) {
		mlog_errno(ret);
		goto out_unlock;
	}

	if (ocfs2_should_order_data(inode)) {
		handle = ocfs2_start_walk_page_trans(inode, page, offset,
						     offset);
		if (IS_ERR(handle)) {
			ret = PTR_ERR(handle);
			handle = NULL;
			goto out_unlock;
		}
	}

	/* must not update i_size! */
	ret = block_commit_write(page, offset, offset);
	if (ret < 0)
		mlog_errno(ret);
	else
		ret = 0;

	if (handle)
		ocfs2_commit_trans(OCFS2_SB(inode->i_sb), handle);
out_unlock:
	unlock_page(page);
	page_cache_release(page);
out:
	return ret;
}

static int ocfs2_zero_extend(struct inode *inode,
			     u64 zero_to_size)
{
	int ret = 0;
	u64 start_off;
	struct super_block *sb = inode->i_sb;

	start_off = ocfs2_align_bytes_to_blocks(sb, i_size_read(inode));
	while (start_off < zero_to_size) {
		ret = ocfs2_write_zero_page(inode, start_off);
		if (ret < 0) {
			mlog_errno(ret);
			goto out;
		}

		start_off += sb->s_blocksize;

		/*
		 * Very large extends have the potential to lock up
		 * the cpu for extended periods of time.
		 */
		cond_resched();
	}

out:
	return ret;
}

/* 
 * A tail_to_skip value > 0 indicates that we're being called from
 * ocfs2_file_aio_write(). This has the following implications:
 *
 * - we don't want to update i_size
 * - di_bh will be NULL, which is fine because it's only used in the
 *   case where we want to update i_size.
 * - ocfs2_zero_extend() will then only be filling the hole created
 *   between i_size and the start of the write.
 */
static int ocfs2_extend_file(struct inode *inode,
			     struct buffer_head *di_bh,
			     u64 new_i_size,
			     size_t tail_to_skip)
{
	int ret = 0;
	u32 clusters_to_add = 0;

	BUG_ON(!tail_to_skip && !di_bh);

	/* setattr sometimes calls us like this. */
	if (new_i_size == 0)
		goto out;

	if (i_size_read(inode) == new_i_size)
  		goto out;
	BUG_ON(new_i_size < i_size_read(inode));

	if (ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb))) {
		BUG_ON(tail_to_skip != 0);
		goto out_update_size;
	}

	clusters_to_add = ocfs2_clusters_for_bytes(inode->i_sb, new_i_size) - 
		OCFS2_I(inode)->ip_clusters;

	/* 
	 * protect the pages that ocfs2_zero_extend is going to be
	 * pulling into the page cache.. we do this before the
	 * metadata extend so that we don't get into the situation
	 * where we've extended the metadata but can't get the data
	 * lock to zero.
	 */
	ret = ocfs2_data_lock(inode, 1);
	if (ret < 0) {
		mlog_errno(ret);
		goto out;
	}

	if (clusters_to_add) {
		ret = ocfs2_extend_allocation(inode, clusters_to_add);
		if (ret < 0) {
			mlog_errno(ret);
			goto out_unlock;
		}
	}

	/*
	 * Call this even if we don't add any clusters to the tree. We
	 * still need to zero the area between the old i_size and the
	 * new i_size.
	 */
	ret = ocfs2_zero_extend(inode, (u64)new_i_size - tail_to_skip);
	if (ret < 0) {
		mlog_errno(ret);
		goto out_unlock;
	}

out_update_size:
	if (!tail_to_skip) {
		/* We're being called from ocfs2_setattr() which wants
		 * us to update i_size */
		ret = ocfs2_simple_size_update(inode, di_bh, new_i_size);
		if (ret < 0)
			mlog_errno(ret);
	}

out_unlock:
	if (!ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb)))
		ocfs2_data_unlock(inode, 1);

out:
	return ret;
}

int ocfs2_setattr(struct dentry *dentry, struct iattr *attr)
{
	int status = 0, size_change;
	struct inode *inode = dentry->d_inode;
	struct super_block *sb = inode->i_sb;
	struct ocfs2_super *osb = OCFS2_SB(sb);
	struct buffer_head *bh = NULL;
	handle_t *handle = NULL;

	mlog_entry("(0x%p, '%.*s')\n", dentry,
	           dentry->d_name.len, dentry->d_name.name);

	if (attr->ia_valid & ATTR_MODE)
		mlog(0, "mode change: %d\n", attr->ia_mode);
	if (attr->ia_valid & ATTR_UID)
		mlog(0, "uid change: %d\n", attr->ia_uid);
	if (attr->ia_valid & ATTR_GID)
		mlog(0, "gid change: %d\n", attr->ia_gid);
	if (attr->ia_valid & ATTR_SIZE)
		mlog(0, "size change...\n");
	if (attr->ia_valid & (ATTR_ATIME | ATTR_MTIME | ATTR_CTIME))
		mlog(0, "time change...\n");

#define OCFS2_VALID_ATTRS (ATTR_ATIME | ATTR_MTIME | ATTR_CTIME | ATTR_SIZE \
			   | ATTR_GID | ATTR_UID | ATTR_MODE)
	if (!(attr->ia_valid & OCFS2_VALID_ATTRS)) {
		mlog(0, "can't handle attrs: 0x%x\n", attr->ia_valid);
		return 0;
	}

	status = inode_change_ok(inode, attr);
	if (status)
		return status;

	size_change = S_ISREG(inode->i_mode) && attr->ia_valid & ATTR_SIZE;
	if (size_change) {
		status = ocfs2_rw_lock(inode, 1);
		if (status < 0) {
			mlog_errno(status);
			goto bail;
		}
	}

	status = ocfs2_meta_lock(inode, &bh, 1);
	if (status < 0) {
		if (status != -ENOENT)
			mlog_errno(status);
		goto bail_unlock_rw;
	}

	if (size_change && attr->ia_size != i_size_read(inode)) {
		if (i_size_read(inode) > attr->ia_size)
			status = ocfs2_truncate_file(inode, bh, attr->ia_size);
		else
			status = ocfs2_extend_file(inode, bh, attr->ia_size, 0);
		if (status < 0) {
			if (status != -ENOSPC)
				mlog_errno(status);
			status = -ENOSPC;
			goto bail_unlock;
		}
	}

	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
	if (IS_ERR(handle)) {
		status = PTR_ERR(handle);
		mlog_errno(status);
		goto bail_unlock;
	}

	/*
	 * This will intentionally not wind up calling vmtruncate(),
	 * since all the work for a size change has been done above.
	 * Otherwise, we could get into problems with truncate as
	 * ip_alloc_sem is used there to protect against i_size
	 * changes.
	 */
	status = inode_setattr(inode, attr);
	if (status < 0) {
		mlog_errno(status);
		goto bail_commit;
	}

	status = ocfs2_mark_inode_dirty(handle, inode, bh);
	if (status < 0)
		mlog_errno(status);

bail_commit:
	ocfs2_commit_trans(osb, handle);
bail_unlock:
	ocfs2_meta_unlock(inode, 1);
bail_unlock_rw:
	if (size_change)
		ocfs2_rw_unlock(inode, 1);
bail:
	if (bh)
		brelse(bh);

	mlog_exit(status);
	return status;
}

int ocfs2_getattr(struct vfsmount *mnt,
		  struct dentry *dentry,
		  struct kstat *stat)
{
	struct inode *inode = dentry->d_inode;
	struct super_block *sb = dentry->d_inode->i_sb;
	struct ocfs2_super *osb = sb->s_fs_info;
	int err;

	mlog_entry_void();

	err = ocfs2_inode_revalidate(dentry);
	if (err) {
		if (err != -ENOENT)
			mlog_errno(err);
		goto bail;
	}

	generic_fillattr(inode, stat);

	/* We set the blksize from the cluster size for performance */
	stat->blksize = osb->s_clustersize;

bail:
	mlog_exit(err);

	return err;
}

int ocfs2_permission(struct inode *inode, int mask, struct nameidata *nd)
{
	int ret;

	mlog_entry_void();

	ret = ocfs2_meta_lock(inode, NULL, 0);
	if (ret) {
		if (ret != -ENOENT)
			mlog_errno(ret);
		goto out;
	}

	ret = generic_permission(inode, mask, NULL);

	ocfs2_meta_unlock(inode, 0);
out:
	mlog_exit(ret);
	return ret;
}

static int ocfs2_write_remove_suid(struct inode *inode)
{
	int ret;
	struct buffer_head *bh = NULL;
	struct ocfs2_inode_info *oi = OCFS2_I(inode);
	handle_t *handle;
	struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
	struct ocfs2_dinode *di;

	mlog_entry("(Inode %llu, mode 0%o)\n",
		   (unsigned long long)oi->ip_blkno, inode->i_mode);

	handle = ocfs2_start_trans(osb, OCFS2_INODE_UPDATE_CREDITS);
	if (handle == NULL) {
		ret = -ENOMEM;
		mlog_errno(ret);
		goto out;
	}

	ret = ocfs2_read_block(osb, oi->ip_blkno, &bh, OCFS2_BH_CACHED, inode);
	if (ret < 0) {
		mlog_errno(ret);
		goto out_trans;
	}

	ret = ocfs2_journal_access(handle, inode, bh,
				   OCFS2_JOURNAL_ACCESS_WRITE);
	if (ret < 0) {
		mlog_errno(ret);
		goto out_bh;
	}

	inode->i_mode &= ~S_ISUID;
	if ((inode->i_mode & S_ISGID) && (inode->i_mode & S_IXGRP))
		inode->i_mode &= ~S_ISGID;

	di = (struct ocfs2_dinode *) bh->b_data;
	di->i_mode = cpu_to_le16(inode->i_mode);

	ret = ocfs2_journal_dirty(handle, bh);
	if (ret < 0)
		mlog_errno(ret);
out_bh:
	brelse(bh);
out_trans:
	ocfs2_commit_trans(osb, handle);
out:
	mlog_exit(ret);
	return ret;
}

/*
 * Will look for holes and unwritten extents in the range starting at
 * pos for count bytes (inclusive).
 */
static int ocfs2_check_range_for_holes(struct inode *inode, loff_t pos,
				       size_t count)
{
	int ret = 0;
	unsigned int extent_flags;
	u32 cpos, clusters, extent_len, phys_cpos;
	struct super_block *sb = inode->i_sb;

	cpos = pos >> OCFS2_SB(sb)->s_clustersize_bits;
	clusters = ocfs2_clusters_for_bytes(sb, pos + count) - cpos;

	while (clusters) {
		ret = ocfs2_get_clusters(inode, cpos, &phys_cpos, &extent_len,
					 &extent_flags);
		if (ret < 0) {
			mlog_errno(ret);
			goto out;
		}

		if (phys_cpos == 0 || (extent_flags & OCFS2_EXT_UNWRITTEN)) {
			ret = 1;
			break;
		}

		if (extent_len > clusters)
			extent_len = clusters;

		clusters -= extent_len;
		cpos += extent_len;
	}
out:
	return ret;
}

static int ocfs2_prepare_inode_for_write(struct dentry *dentry,
					 loff_t *ppos,
					 size_t count,
					 int appending,
					 int *direct_io)
{
	int ret = 0, meta_level = appending;
	struct inode *inode = dentry->d_inode;
	u32 clusters;
	loff_t newsize, saved_pos;

	/* 
	 * We sample i_size under a read level meta lock to see if our write
	 * is extending the file, if it is we back off and get a write level
	 * meta lock.
	 */
	for(;;) {
		ret = ocfs2_meta_lock(inode, NULL, meta_level);
		if (ret < 0) {
			meta_level = -1;
			mlog_errno(ret);
			goto out;
		}

		/* Clear suid / sgid if necessary. We do this here
		 * instead of later in the write path because
		 * remove_suid() calls ->setattr without any hint that
		 * we may have already done our cluster locking. Since
		 * ocfs2_setattr() *must* take cluster locks to
		 * proceeed, this will lead us to recursively lock the
		 * inode. There's also the dinode i_size state which
		 * can be lost via setattr during extending writes (we
		 * set inode->i_size at the end of a write. */
		if (should_remove_suid(dentry)) {
			if (meta_level == 0) {
				ocfs2_meta_unlock(inode, meta_level);
				meta_level = 1;
				continue;
			}

			ret = ocfs2_write_remove_suid(inode);
			if (ret < 0) {
				mlog_errno(ret);
				goto out_unlock;
			}
		}

		/* work on a copy of ppos until we're sure that we won't have
		 * to recalculate it due to relocking. */
		if (appending) {
			saved_pos = i_size_read(inode);
			mlog(0, "O_APPEND: inode->i_size=%llu\n", saved_pos);
		} else {
			saved_pos = *ppos;
		}

		if (ocfs2_sparse_alloc(OCFS2_SB(inode->i_sb))) {
			loff_t end = saved_pos + count;

			/*
			 * Skip the O_DIRECT checks if we don't need
			 * them.
			 */
			if (!direct_io || !(*direct_io))
				break;

			/*
			 * Allowing concurrent direct writes means
			 * i_size changes wouldn't be synchronized, so
			 * one node could wind up truncating another
			 * nodes writes.
			 */
			if (end > i_size_read(inode)) {
				*direct_io = 0;
				break;
			}

			/*
			 * We don't fill holes during direct io, so
			 * check for them here. If any are found, the
			 * caller will have to retake some cluster
			 * locks and initiate the io as buffered.
			 */
			ret = ocfs2_check_range_for_holes(inode, saved_pos,
							  count);
			if (ret == 1) {
				*direct_io = 0;
				ret = 0;
			} else if (ret < 0)
				mlog_errno(ret);
			break;
		}

		/*
		 * The rest of this loop is concerned with legacy file
		 * systems which don't support sparse files.
		 */

		newsize = count + saved_pos;

		mlog(0, "pos=%lld newsize=%lld cursize=%lld\n",
		     (long long) saved_pos, (long long) newsize,
		     (long long) i_size_read(inode));

		/* No need for a higher level metadata lock if we're
		 * never going past i_size. */
		if (newsize <= i_size_read(inode))
			break;

		if (meta_level == 0) {
			ocfs2_meta_unlock(inode, meta_level);
			meta_level = 1;
			continue;
		}

		spin_lock(&OCFS2_I(inode)->ip_lock);
		clusters = ocfs2_clusters_for_bytes(inode->i_sb, newsize) -
			OCFS2_I(inode)->ip_clusters;
		spin_unlock(&OCFS2_I(inode)->ip_lock);

		mlog(0, "Writing at EOF, may need more allocation: "
		     "i_size = %lld, newsize = %lld, need %u clusters\n",
		     (long long) i_size_read(inode), (long long) newsize,
		     clusters);

		/* We only want to continue the rest of this loop if
		 * our extend will actually require more
		 * allocation. */
		if (!clusters)
			break;

		ret = ocfs2_extend_file(inode, NULL, newsize, count);
		if (ret < 0) {
			if (ret != -ENOSPC)
				mlog_errno(ret);
			goto out_unlock;
		}
		break;
	}

	if (appending)
		*ppos = saved_pos;

out_unlock:
	ocfs2_meta_unlock(inode, meta_level);

out:
	return ret;
}

static inline void
ocfs2_set_next_iovec(const struct iovec **iovp, size_t *basep, size_t bytes)
{
	const struct iovec *iov = *iovp;
	size_t base = *basep;

	do {
		int copy = min(bytes, iov->iov_len - base);

		bytes -= copy;
		base += copy;
		if (iov->iov_len == base) {
			iov++;
			base = 0;
		}
	} while (bytes);
	*iovp = iov;
	*basep = base;
}

static struct page * ocfs2_get_write_source(char **ret_src_buf,
					    const struct iovec *cur_iov,
					    size_t iov_offset)
{
	int ret;
	char *buf = cur_iov->iov_base + iov_offset;
	struct page *src_page = NULL;
	unsigned long off;

	off = (unsigned long)(buf) & ~PAGE_CACHE_MASK;

	if (!segment_eq(get_fs(), KERNEL_DS)) {
		/*
		 * Pull in the user page. We want to do this outside
		 * of the meta data locks in order to preserve locking
		 * order in case of page fault.
		 */
		ret = get_user_pages(current, current->mm,
				     (unsigned long)buf & PAGE_CACHE_MASK, 1,
				     0, 0, &src_page, NULL);
		if (ret == 1)
			*ret_src_buf = kmap(src_page) + off;
		else
			src_page = ERR_PTR(-EFAULT);
	} else {
		*ret_src_buf = buf;
	}

	return src_page;
}

static void ocfs2_put_write_source(struct page *page)
{
	if (page) {
		kunmap(page);
		page_cache_release(page);
	}
}

static ssize_t ocfs2_file_buffered_write(struct file *file, loff_t *ppos,
					 const struct iovec *iov,
					 unsigned long nr_segs,
					 size_t count,
					 ssize_t o_direct_written)
{
	int ret = 0;
	ssize_t copied, total = 0;
	size_t iov_offset = 0, bytes;
	loff_t pos;
	const struct iovec *cur_iov = iov;
	struct page *user_page, *page;
	char *buf, *dst;
	void *fsdata;

	/*
	 * handle partial DIO write.  Adjust cur_iov if needed.
	 */
	ocfs2_set_next_iovec(&cur_iov, &iov_offset, o_direct_written);

	do {
		pos = *ppos;

		user_page = ocfs2_get_write_source(&buf, cur_iov, iov_offset);
		if (IS_ERR(user_page)) {
			ret = PTR_ERR(user_page);
			goto out;
		}

		/* Stay within our page boundaries */
		bytes = min((PAGE_CACHE_SIZE - ((unsigned long)pos & ~PAGE_CACHE_MASK)),
			    (PAGE_CACHE_SIZE - ((unsigned long)buf & ~PAGE_CACHE_MASK)));
		/* Stay within the vector boundary */
		bytes = min_t(size_t, bytes, cur_iov->iov_len - iov_offset);
		/* Stay within count */
		bytes = min(bytes, count);

		page = NULL;
		ret = ocfs2_write_begin(file, file->f_mapping, pos, bytes, 0,
					&page, &fsdata);
		if (ret) {
			mlog_errno(ret);
			goto out;
		}

		dst = kmap_atomic(page, KM_USER0);
		memcpy(dst + (pos & (PAGE_CACHE_SIZE - 1)), buf, bytes);
		kunmap_atomic(dst, KM_USER0);
		flush_dcache_page(page);
		ocfs2_put_write_source(user_page);

		copied = ocfs2_write_end(file, file->f_mapping, pos, bytes,
					 bytes, page, fsdata);
		if (copied < 0) {
			mlog_errno(copied);
			ret = copied;
			goto out;
		}

		total += copied;
		*ppos = pos + copied;
		count -= copied;

		ocfs2_set_next_iovec(&cur_iov, &iov_offset, copied);
	} while(count);

out:
	return total ? total : ret;
}

static ssize_t ocfs2_file_aio_write(struct kiocb *iocb,
				    const struct iovec *iov,
				    unsigned long nr_segs,
				    loff_t pos)
{
	int ret, direct_io, appending, rw_level, have_alloc_sem  = 0;
	int can_do_direct, sync = 0;
	ssize_t written = 0;
	size_t ocount;		/* original count */
	size_t count;		/* after file limit checks */
	loff_t *ppos = &iocb->ki_pos;
	struct file *file = iocb->ki_filp;
	struct inode *inode = file->f_path.dentry->d_inode;

	mlog_entry("(0x%p, %u, '%.*s')\n", file,
		   (unsigned int)nr_segs,
		   file->f_path.dentry->d_name.len,
		   file->f_path.dentry->d_name.name);

	if (iocb->ki_left == 0)
		return 0;

	ret = generic_segment_checks(iov, &nr_segs, &ocount, VERIFY_READ);
	if (ret)
		return ret;

	count = ocount;

	vfs_check_frozen(inode->i_sb, SB_FREEZE_WRITE);

	appending = file->f_flags & O_APPEND ? 1 : 0;
	direct_io = file->f_flags & O_DIRECT ? 1 : 0;

	mutex_lock(&inode->i_mutex);

relock:
	/* to match setattr's i_mutex -> i_alloc_sem -> rw_lock ordering */
	if (direct_io) {
		down_read(&inode->i_alloc_sem);
		have_alloc_sem = 1;
	}

	/* concurrent O_DIRECT writes are allowed */
	rw_level = !direct_io;
	ret = ocfs2_rw_lock(inode, rw_level);
	if (ret < 0) {
		mlog_errno(ret);
		goto out_sems;
	}

	can_do_direct = direct_io;
	ret = ocfs2_prepare_inode_for_write(file->f_path.dentry, ppos,
					    iocb->ki_left, appending,
					    &can_do_direct);
	if (ret < 0) {
		mlog_errno(ret);
		goto out;
	}

	/*
	 * We can't complete the direct I/O as requested, fall back to
	 * buffered I/O.
	 */
	if (direct_io && !can_do_direct) {
		ocfs2_rw_unlock(inode, rw_level);
		up_read(&inode->i_alloc_sem);

		have_alloc_sem = 0;
		rw_level = -1;

		direct_io = 0;
		sync = 1;
		goto relock;
	}

	if (!sync && ((file->f_flags & O_SYNC) || IS_SYNC(inode)))
		sync = 1;

	/*
	 * XXX: Is it ok to execute these checks a second time?
	 */
	ret = generic_write_checks(file, ppos, &count, S_ISBLK(inode->i_mode));
	if (ret)
		goto out;

	/*
	 * Set pos so that sync_page_range_nolock() below understands
	 * where to start from. We might've moved it around via the
	 * calls above. The range we want to actually sync starts from
	 * *ppos here.
	 *
	 */
	pos = *ppos;

	/* communicate with ocfs2_dio_end_io */
	ocfs2_iocb_set_rw_locked(iocb, rw_level);

	if (direct_io) {
		written = generic_file_direct_write(iocb, iov, &nr_segs, *ppos,
						    ppos, count, ocount);
		if (written < 0) {
			ret = written;
			goto out_dio;
		}
	} else {
		written = ocfs2_file_buffered_write(file, ppos, iov, nr_segs,
						    count, written);
		if (written < 0) {
			ret = written;
			if (ret != -EFAULT || ret != -ENOSPC)
				mlog_errno(ret);
			goto out;
		}
	}

out_dio:
	/* buffered aio wouldn't have proper lock coverage today */
	BUG_ON(ret == -EIOCBQUEUED && !(file->f_flags & O_DIRECT));

	/* 
	 * deep in g_f_a_w_n()->ocfs2_direct_IO we pass in a ocfs2_dio_end_io
	 * function pointer which is called when o_direct io completes so that
	 * it can unlock our rw lock.  (it's the clustered equivalent of
	 * i_alloc_sem; protects truncate from racing with pending ios).
	 * Unfortunately there are error cases which call end_io and others
	 * that don't.  so we don't have to unlock the rw_lock if either an
	 * async dio is going to do it in the future or an end_io after an
	 * error has already done it.
	 */
	if (ret == -EIOCBQUEUED || !ocfs2_iocb_is_rw_locked(iocb)) {
		rw_level = -1;
		have_alloc_sem = 0;
	}

out:
	if (rw_level != -1)
		ocfs2_rw_unlock(inode, rw_level);

out_sems:
	if (have_alloc_sem)
		up_read(&inode->i_alloc_sem);

	if (written > 0 && sync) {
		ssize_t err;

		err = sync_page_range_nolock(inode, file->f_mapping, pos, count);
		if (err < 0)
			written = err;
	}

	mutex_unlock(&inode->i_mutex);

	mlog_exit(ret);
	return written ? written : ret;
}

static int ocfs2_splice_write_actor(struct pipe_inode_info *pipe,
				    struct pipe_buffer *buf,
				    struct splice_desc *sd)
{
	int ret, count;
	ssize_t copied = 0;
	struct file *file = sd->u.file;
	unsigned int offset;
	struct page *page = NULL;
	void *fsdata;
	char *src, *dst;

	ret = buf->ops->confirm(pipe, buf);
	if (ret)
		goto out;

	offset = sd->pos & ~PAGE_CACHE_MASK;
	count = sd->len;
	if (count + offset > PAGE_CACHE_SIZE)
		count = PAGE_CACHE_SIZE - offset;

	ret = ocfs2_write_begin(file, file->f_mapping, sd->pos, count, 0,
				&page, &fsdata);
	if (ret) {
		mlog_errno(ret);
		goto out;
	}

	src = buf->ops->map(pipe, buf, 1);
	dst = kmap_atomic(page, KM_USER1);
	memcpy(dst + offset, src + buf->offset, count);
	kunmap_atomic(page, KM_USER1);
	buf->ops->unmap(pipe, buf, src);

	copied = ocfs2_write_end(file, file->f_mapping, sd->pos, count, count,
				 page, fsdata);
	if (copied < 0) {
		mlog_errno(copied);
		ret = copied;
		goto out;
	}
out:

	return copied ? copied : ret;
}

static ssize_t __ocfs2_file_splice_write(struct pipe_inode_info *pipe,
					 struct file *out,
					 loff_t *ppos,
					 size_t len,
					 unsigned int flags)
{
	int ret, err;
	struct address_space *mapping = out->f_mapping;
	struct inode *inode = mapping->host;
	struct splice_desc sd = {
		.total_len = len,
		.flags = flags,
		.pos = *ppos,
		.u.file = out,
	};

	ret = __splice_from_pipe(pipe, &sd, ocfs2_splice_write_actor);
	if (ret > 0) {
		*ppos += ret;

		if (unlikely((out->f_flags & O_SYNC) || IS_SYNC(inode))) {
			err = generic_osync_inode(inode, mapping,
						  OSYNC_METADATA|OSYNC_DATA);
			if (err)
				ret = err;
		}
	}

	return ret;
}

static ssize_t ocfs2_file_splice_write(struct pipe_inode_info *pipe,
				       struct file *out,
				       loff_t *ppos,
				       size_t len,
				       unsigned int flags)
{
	int ret;
	struct inode *inode = out->f_path.dentry->d_inode;

	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", out, pipe,
		   (unsigned int)len,
		   out->f_path.dentry->d_name.len,
		   out->f_path.dentry->d_name.name);

	inode_double_lock(inode, pipe->inode);

	ret = ocfs2_rw_lock(inode, 1);
	if (ret < 0) {
		mlog_errno(ret);
		goto out;
	}

	ret = ocfs2_prepare_inode_for_write(out->f_path.dentry, ppos, len, 0,
					    NULL);
	if (ret < 0) {
		mlog_errno(ret);
		goto out_unlock;
	}

	/* ok, we're done with i_size and alloc work */
	ret = __ocfs2_file_splice_write(pipe, out, ppos, len, flags);

out_unlock:
	ocfs2_rw_unlock(inode, 1);
out:
	inode_double_unlock(inode, pipe->inode);

	mlog_exit(ret);
	return ret;
}

static ssize_t ocfs2_file_splice_read(struct file *in,
				      loff_t *ppos,
				      struct pipe_inode_info *pipe,
				      size_t len,
				      unsigned int flags)
{
	int ret = 0;
	struct inode *inode = in->f_path.dentry->d_inode;

	mlog_entry("(0x%p, 0x%p, %u, '%.*s')\n", in, pipe,
		   (unsigned int)len,
		   in->f_path.dentry->d_name.len,
		   in->f_path.dentry->d_name.name);

	/*
	 * See the comment in ocfs2_file_aio_read()
	 */
	ret = ocfs2_meta_lock(inode, NULL, 0);
	if (ret < 0) {
		mlog_errno(ret);
		goto bail;
	}
	ocfs2_meta_unlock(inode, 0);

	ret = generic_file_splice_read(in, ppos, pipe, len, flags);

bail:
	mlog_exit(ret);
	return ret;
}

static ssize_t ocfs2_file_aio_read(struct kiocb *iocb,
				   const struct iovec *iov,
				   unsigned long nr_segs,
				   loff_t pos)
{
	int ret = 0, rw_level = -1, have_alloc_sem = 0, lock_level = 0;
	struct file *filp = iocb->ki_filp;
	struct inode *inode = filp->f_path.dentry->d_inode;

	mlog_entry("(0x%p, %u, '%.*s')\n", filp,
		   (unsigned int)nr_segs,
		   filp->f_path.dentry->d_name.len,
		   filp->f_path.dentry->d_name.name);

	if (!inode) {
		ret = -EINVAL;
		mlog_errno(ret);
		goto bail;
	}

	/* 
	 * buffered reads protect themselves in ->readpage().  O_DIRECT reads
	 * need locks to protect pending reads from racing with truncate.
	 */
	if (filp->f_flags & O_DIRECT) {
		down_read(&inode->i_alloc_sem);
		have_alloc_sem = 1;

		ret = ocfs2_rw_lock(inode, 0);
		if (ret < 0) {
			mlog_errno(ret);
			goto bail;
		}
		rw_level = 0;
		/* communicate with ocfs2_dio_end_io */
		ocfs2_iocb_set_rw_locked(iocb, rw_level);
	}

	/*
	 * We're fine letting folks race truncates and extending
	 * writes with read across the cluster, just like they can
	 * locally. Hence no rw_lock during read.
	 * 
	 * Take and drop the meta data lock to update inode fields
	 * like i_size. This allows the checks down below
	 * generic_file_aio_read() a chance of actually working. 
	 */
	ret = ocfs2_meta_lock_atime(inode, filp->f_vfsmnt, &lock_level);
	if (ret < 0) {
		mlog_errno(ret);
		goto bail;
	}
	ocfs2_meta_unlock(inode, lock_level);

	ret = generic_file_aio_read(iocb, iov, nr_segs, iocb->ki_pos);
	if (ret == -EINVAL)
		mlog(ML_ERROR, "generic_file_aio_read returned -EINVAL\n");

	/* buffered aio wouldn't have proper lock coverage today */
	BUG_ON(ret == -EIOCBQUEUED && !(filp->f_flags & O_DIRECT));

	/* see ocfs2_file_aio_write */
	if (ret == -EIOCBQUEUED || !ocfs2_iocb_is_rw_locked(iocb)) {
		rw_level = -1;
		have_alloc_sem = 0;
	}

bail:
	if (have_alloc_sem)
		up_read(&inode->i_alloc_sem);
	if (rw_level != -1) 
		ocfs2_rw_unlock(inode, rw_level);
	mlog_exit(ret);

	return ret;
}

const struct inode_operations ocfs2_file_iops = {
	.setattr	= ocfs2_setattr,
	.getattr	= ocfs2_getattr,
	.permission	= ocfs2_permission,
};

const struct inode_operations ocfs2_special_file_iops = {
	.setattr	= ocfs2_setattr,
	.getattr	= ocfs2_getattr,
	.permission	= ocfs2_permission,
};

const struct file_operations ocfs2_fops = {
	.read		= do_sync_read,
	.write		= do_sync_write,
	.mmap		= ocfs2_mmap,
	.fsync		= ocfs2_sync_file,
	.release	= ocfs2_file_release,
	.open		= ocfs2_file_open,
	.aio_read	= ocfs2_file_aio_read,
	.aio_write	= ocfs2_file_aio_write,
	.ioctl		= ocfs2_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl   = ocfs2_compat_ioctl,
#endif
	.splice_read	= ocfs2_file_splice_read,
	.splice_write	= ocfs2_file_splice_write,
};

const struct file_operations ocfs2_dops = {
	.read		= generic_read_dir,
	.readdir	= ocfs2_readdir,
	.fsync		= ocfs2_sync_file,
	.ioctl		= ocfs2_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl   = ocfs2_compat_ioctl,
#endif
};