summaryrefslogtreecommitdiff
path: root/fs/ubifs/dir.c
blob: 90fe60c2f112da2f9f7f2d2dd14e969d2e04dc14 (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
/* * This file is part of UBIFS.
 *
 * Copyright (C) 2006-2008 Nokia Corporation.
 * Copyright (C) 2006, 2007 University of Szeged, Hungary
 *
 * This program is free software; you can redistribute it and/or modify it
 * under the terms of the GNU General Public License version 2 as published by
 * the Free Software Foundation.
 *
 * 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., 51
 * Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
 *
 * Authors: Artem Bityutskiy (Битюцкий Артём)
 *          Adrian Hunter
 *          Zoltan Sogor
 */

/*
 * This file implements directory operations.
 *
 * All FS operations in this file allocate budget before writing anything to the
 * media. If they fail to allocate it, the error is returned. The only
 * exceptions are 'ubifs_unlink()' and 'ubifs_rmdir()' which keep working even
 * if they unable to allocate the budget, because deletion %-ENOSPC failure is
 * not what users are usually ready to get. UBIFS budgeting subsystem has some
 * space reserved for these purposes.
 *
 * All operations in this file write all inodes which they change straight
 * away, instead of marking them dirty. For example, 'ubifs_link()' changes
 * @i_size of the parent inode and writes the parent inode together with the
 * target inode. This was done to simplify file-system recovery which would
 * otherwise be very difficult to do. The only exception is rename which marks
 * the re-named inode dirty (because its @i_ctime is updated) but does not
 * write it, but just marks it as dirty.
 */

#include "ubifs.h"

/**
 * inherit_flags - inherit flags of the parent inode.
 * @dir: parent inode
 * @mode: new inode mode flags
 *
 * This is a helper function for 'ubifs_new_inode()' which inherits flag of the
 * parent directory inode @dir. UBIFS inodes inherit the following flags:
 * o %UBIFS_COMPR_FL, which is useful to switch compression on/of on
 *   sub-directory basis;
 * o %UBIFS_SYNC_FL - useful for the same reasons;
 * o %UBIFS_DIRSYNC_FL - similar, but relevant only to directories.
 *
 * This function returns the inherited flags.
 */
static int inherit_flags(const struct inode *dir, umode_t mode)
{
	int flags;
	const struct ubifs_inode *ui = ubifs_inode(dir);

	if (!S_ISDIR(dir->i_mode))
		/*
		 * The parent is not a directory, which means that an extended
		 * attribute inode is being created. No flags.
		 */
		return 0;

	flags = ui->flags & (UBIFS_COMPR_FL | UBIFS_SYNC_FL | UBIFS_DIRSYNC_FL);
	if (!S_ISDIR(mode))
		/* The "DIRSYNC" flag only applies to directories */
		flags &= ~UBIFS_DIRSYNC_FL;
	return flags;
}

/**
 * ubifs_new_inode - allocate new UBIFS inode object.
 * @c: UBIFS file-system description object
 * @dir: parent directory inode
 * @mode: inode mode flags
 *
 * This function finds an unused inode number, allocates new inode and
 * initializes it. Returns new inode in case of success and an error code in
 * case of failure.
 */
struct inode *ubifs_new_inode(struct ubifs_info *c, const struct inode *dir,
			      umode_t mode)
{
	struct inode *inode;
	struct ubifs_inode *ui;

	inode = new_inode(c->vfs_sb);
	ui = ubifs_inode(inode);
	if (!inode)
		return ERR_PTR(-ENOMEM);

	/*
	 * Set 'S_NOCMTIME' to prevent VFS form updating [mc]time of inodes and
	 * marking them dirty in file write path (see 'file_update_time()').
	 * UBIFS has to fully control "clean <-> dirty" transitions of inodes
	 * to make budgeting work.
	 */
	inode->i_flags |= S_NOCMTIME;

	inode_init_owner(inode, dir, mode);
	inode->i_mtime = inode->i_atime = inode->i_ctime =
			 ubifs_current_time(inode);
	inode->i_mapping->nrpages = 0;

	switch (mode & S_IFMT) {
	case S_IFREG:
		inode->i_mapping->a_ops = &ubifs_file_address_operations;
		inode->i_op = &ubifs_file_inode_operations;
		inode->i_fop = &ubifs_file_operations;
		break;
	case S_IFDIR:
		inode->i_op  = &ubifs_dir_inode_operations;
		inode->i_fop = &ubifs_dir_operations;
		inode->i_size = ui->ui_size = UBIFS_INO_NODE_SZ;
		break;
	case S_IFLNK:
		inode->i_op = &ubifs_symlink_inode_operations;
		break;
	case S_IFSOCK:
	case S_IFIFO:
	case S_IFBLK:
	case S_IFCHR:
		inode->i_op  = &ubifs_file_inode_operations;
		break;
	default:
		BUG();
	}

	ui->flags = inherit_flags(dir, mode);
	ubifs_set_inode_flags(inode);
	if (S_ISREG(mode))
		ui->compr_type = c->default_compr;
	else
		ui->compr_type = UBIFS_COMPR_NONE;
	ui->synced_i_size = 0;

	spin_lock(&c->cnt_lock);
	/* Inode number overflow is currently not supported */
	if (c->highest_inum >= INUM_WARN_WATERMARK) {
		if (c->highest_inum >= INUM_WATERMARK) {
			spin_unlock(&c->cnt_lock);
			ubifs_err(c, "out of inode numbers");
			make_bad_inode(inode);
			iput(inode);
			return ERR_PTR(-EINVAL);
		}
		ubifs_warn(c, "running out of inode numbers (current %lu, max %d)",
			   (unsigned long)c->highest_inum, INUM_WATERMARK);
	}

	inode->i_ino = ++c->highest_inum;
	/*
	 * The creation sequence number remains with this inode for its
	 * lifetime. All nodes for this inode have a greater sequence number,
	 * and so it is possible to distinguish obsolete nodes belonging to a
	 * previous incarnation of the same inode number - for example, for the
	 * purpose of rebuilding the index.
	 */
	ui->creat_sqnum = ++c->max_sqnum;
	spin_unlock(&c->cnt_lock);
	return inode;
}

static int dbg_check_name(const struct ubifs_info *c,
			  const struct ubifs_dent_node *dent,
			  const struct qstr *nm)
{
	if (!dbg_is_chk_gen(c))
		return 0;
	if (le16_to_cpu(dent->nlen) != nm->len)
		return -EINVAL;
	if (memcmp(dent->name, nm->name, nm->len))
		return -EINVAL;
	return 0;
}

static struct dentry *ubifs_lookup(struct inode *dir, struct dentry *dentry,
				   unsigned int flags)
{
	int err;
	union ubifs_key key;
	struct inode *inode = NULL;
	struct ubifs_dent_node *dent;
	struct ubifs_info *c = dir->i_sb->s_fs_info;

	dbg_gen("'%pd' in dir ino %lu", dentry, dir->i_ino);

	if (dentry->d_name.len > UBIFS_MAX_NLEN)
		return ERR_PTR(-ENAMETOOLONG);

	dent = kmalloc(UBIFS_MAX_DENT_NODE_SZ, GFP_NOFS);
	if (!dent)
		return ERR_PTR(-ENOMEM);

	dent_key_init(c, &key, dir->i_ino, &dentry->d_name);

	err = ubifs_tnc_lookup_nm(c, &key, dent, &dentry->d_name);
	if (err) {
		if (err == -ENOENT) {
			dbg_gen("not found");
			goto done;
		}
		goto out;
	}

	if (dbg_check_name(c, dent, &dentry->d_name)) {
		err = -EINVAL;
		goto out;
	}

	inode = ubifs_iget(dir->i_sb, le64_to_cpu(dent->inum));
	if (IS_ERR(inode)) {
		/*
		 * This should not happen. Probably the file-system needs
		 * checking.
		 */
		err = PTR_ERR(inode);
		ubifs_err(c, "dead directory entry '%pd', error %d",
			  dentry, err);
		ubifs_ro_mode(c, err);
		goto out;
	}

done:
	kfree(dent);
	/*
	 * Note, d_splice_alias() would be required instead if we supported
	 * NFS.
	 */
	d_add(dentry, inode);
	return NULL;

out:
	kfree(dent);
	return ERR_PTR(err);
}

static int ubifs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
			bool excl)
{
	struct inode *inode;
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	int err, sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1,
					.dirtied_ino = 1 };
	struct ubifs_inode *dir_ui = ubifs_inode(dir);

	/*
	 * Budget request settings: new inode, new direntry, changing the
	 * parent directory inode.
	 */

	dbg_gen("dent '%pd', mode %#hx in dir ino %lu",
		dentry, mode, dir->i_ino);

	err = ubifs_budget_space(c, &req);
	if (err)
		return err;

	inode = ubifs_new_inode(c, dir, mode);
	if (IS_ERR(inode)) {
		err = PTR_ERR(inode);
		goto out_budg;
	}

	err = ubifs_init_security(dir, inode, &dentry->d_name);
	if (err)
		goto out_inode;

	mutex_lock(&dir_ui->ui_mutex);
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 0, 0);
	if (err)
		goto out_cancel;
	mutex_unlock(&dir_ui->ui_mutex);

	ubifs_release_budget(c, &req);
	insert_inode_hash(inode);
	d_instantiate(dentry, inode);
	return 0;

out_cancel:
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	mutex_unlock(&dir_ui->ui_mutex);
out_inode:
	make_bad_inode(inode);
	iput(inode);
out_budg:
	ubifs_release_budget(c, &req);
	ubifs_err(c, "cannot create regular file, error %d", err);
	return err;
}

/**
 * vfs_dent_type - get VFS directory entry type.
 * @type: UBIFS directory entry type
 *
 * This function converts UBIFS directory entry type into VFS directory entry
 * type.
 */
static unsigned int vfs_dent_type(uint8_t type)
{
	switch (type) {
	case UBIFS_ITYPE_REG:
		return DT_REG;
	case UBIFS_ITYPE_DIR:
		return DT_DIR;
	case UBIFS_ITYPE_LNK:
		return DT_LNK;
	case UBIFS_ITYPE_BLK:
		return DT_BLK;
	case UBIFS_ITYPE_CHR:
		return DT_CHR;
	case UBIFS_ITYPE_FIFO:
		return DT_FIFO;
	case UBIFS_ITYPE_SOCK:
		return DT_SOCK;
	default:
		BUG();
	}
	return 0;
}

/*
 * The classical Unix view for directory is that it is a linear array of
 * (name, inode number) entries. Linux/VFS assumes this model as well.
 * Particularly, 'readdir()' call wants us to return a directory entry offset
 * which later may be used to continue 'readdir()'ing the directory or to
 * 'seek()' to that specific direntry. Obviously UBIFS does not really fit this
 * model because directory entries are identified by keys, which may collide.
 *
 * UBIFS uses directory entry hash value for directory offsets, so
 * 'seekdir()'/'telldir()' may not always work because of possible key
 * collisions. But UBIFS guarantees that consecutive 'readdir()' calls work
 * properly by means of saving full directory entry name in the private field
 * of the file description object.
 *
 * This means that UBIFS cannot support NFS which requires full
 * 'seekdir()'/'telldir()' support.
 */
static int ubifs_readdir(struct file *file, struct dir_context *ctx)
{
	int err;
	struct qstr nm;
	union ubifs_key key;
	struct ubifs_dent_node *dent;
	struct inode *dir = file_inode(file);
	struct ubifs_info *c = dir->i_sb->s_fs_info;

	dbg_gen("dir ino %lu, f_pos %#llx", dir->i_ino, ctx->pos);

	if (ctx->pos > UBIFS_S_KEY_HASH_MASK || ctx->pos == 2)
		/*
		 * The directory was seek'ed to a senseless position or there
		 * are no more entries.
		 */
		return 0;

	if (file->f_version == 0) {
		/*
		 * The file was seek'ed, which means that @file->private_data
		 * is now invalid. This may also be just the first
		 * 'ubifs_readdir()' invocation, in which case
		 * @file->private_data is NULL, and the below code is
		 * basically a no-op.
		 */
		kfree(file->private_data);
		file->private_data = NULL;
	}

	/*
	 * 'generic_file_llseek()' unconditionally sets @file->f_version to
	 * zero, and we use this for detecting whether the file was seek'ed.
	 */
	file->f_version = 1;

	/* File positions 0 and 1 correspond to "." and ".." */
	if (ctx->pos < 2) {
		ubifs_assert(!file->private_data);
		if (!dir_emit_dots(file, ctx))
			return 0;

		/* Find the first entry in TNC and save it */
		lowest_dent_key(c, &key, dir->i_ino);
		nm.name = NULL;
		dent = ubifs_tnc_next_ent(c, &key, &nm);
		if (IS_ERR(dent)) {
			err = PTR_ERR(dent);
			goto out;
		}

		ctx->pos = key_hash_flash(c, &dent->key);
		file->private_data = dent;
	}

	dent = file->private_data;
	if (!dent) {
		/*
		 * The directory was seek'ed to and is now readdir'ed.
		 * Find the entry corresponding to @ctx->pos or the closest one.
		 */
		dent_key_init_hash(c, &key, dir->i_ino, ctx->pos);
		nm.name = NULL;
		dent = ubifs_tnc_next_ent(c, &key, &nm);
		if (IS_ERR(dent)) {
			err = PTR_ERR(dent);
			goto out;
		}
		ctx->pos = key_hash_flash(c, &dent->key);
		file->private_data = dent;
	}

	while (1) {
		dbg_gen("feed '%s', ino %llu, new f_pos %#x",
			dent->name, (unsigned long long)le64_to_cpu(dent->inum),
			key_hash_flash(c, &dent->key));
		ubifs_assert(le64_to_cpu(dent->ch.sqnum) >
			     ubifs_inode(dir)->creat_sqnum);

		nm.len = le16_to_cpu(dent->nlen);
		if (!dir_emit(ctx, dent->name, nm.len,
			       le64_to_cpu(dent->inum),
			       vfs_dent_type(dent->type)))
			return 0;

		/* Switch to the next entry */
		key_read(c, &dent->key, &key);
		nm.name = dent->name;
		dent = ubifs_tnc_next_ent(c, &key, &nm);
		if (IS_ERR(dent)) {
			err = PTR_ERR(dent);
			goto out;
		}

		kfree(file->private_data);
		ctx->pos = key_hash_flash(c, &dent->key);
		file->private_data = dent;
		cond_resched();
	}

out:
	if (err != -ENOENT) {
		ubifs_err(c, "cannot find next direntry, error %d", err);
		return err;
	}

	kfree(file->private_data);
	file->private_data = NULL;
	/* 2 is a special value indicating that there are no more direntries */
	ctx->pos = 2;
	return 0;
}

/* Free saved readdir() state when the directory is closed */
static int ubifs_dir_release(struct inode *dir, struct file *file)
{
	kfree(file->private_data);
	file->private_data = NULL;
	return 0;
}

/**
 * lock_2_inodes - a wrapper for locking two UBIFS inodes.
 * @inode1: first inode
 * @inode2: second inode
 *
 * We do not implement any tricks to guarantee strict lock ordering, because
 * VFS has already done it for us on the @i_mutex. So this is just a simple
 * wrapper function.
 */
static void lock_2_inodes(struct inode *inode1, struct inode *inode2)
{
	mutex_lock_nested(&ubifs_inode(inode1)->ui_mutex, WB_MUTEX_1);
	mutex_lock_nested(&ubifs_inode(inode2)->ui_mutex, WB_MUTEX_2);
}

/**
 * unlock_2_inodes - a wrapper for unlocking two UBIFS inodes.
 * @inode1: first inode
 * @inode2: second inode
 */
static void unlock_2_inodes(struct inode *inode1, struct inode *inode2)
{
	mutex_unlock(&ubifs_inode(inode2)->ui_mutex);
	mutex_unlock(&ubifs_inode(inode1)->ui_mutex);
}

static int ubifs_link(struct dentry *old_dentry, struct inode *dir,
		      struct dentry *dentry)
{
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	struct inode *inode = old_dentry->d_inode;
	struct ubifs_inode *ui = ubifs_inode(inode);
	struct ubifs_inode *dir_ui = ubifs_inode(dir);
	int err, sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	struct ubifs_budget_req req = { .new_dent = 1, .dirtied_ino = 2,
				.dirtied_ino_d = ALIGN(ui->data_len, 8) };

	/*
	 * Budget request settings: new direntry, changing the target inode,
	 * changing the parent inode.
	 */

	dbg_gen("dent '%pd' to ino %lu (nlink %d) in dir ino %lu",
		dentry, inode->i_ino,
		inode->i_nlink, dir->i_ino);
	ubifs_assert(mutex_is_locked(&dir->i_mutex));
	ubifs_assert(mutex_is_locked(&inode->i_mutex));

	err = dbg_check_synced_i_size(c, inode);
	if (err)
		return err;

	err = ubifs_budget_space(c, &req);
	if (err)
		return err;

	lock_2_inodes(dir, inode);
	inc_nlink(inode);
	ihold(inode);
	inode->i_ctime = ubifs_current_time(inode);
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 0, 0);
	if (err)
		goto out_cancel;
	unlock_2_inodes(dir, inode);

	ubifs_release_budget(c, &req);
	d_instantiate(dentry, inode);
	return 0;

out_cancel:
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	drop_nlink(inode);
	unlock_2_inodes(dir, inode);
	ubifs_release_budget(c, &req);
	iput(inode);
	return err;
}

static int ubifs_unlink(struct inode *dir, struct dentry *dentry)
{
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	struct inode *inode = dentry->d_inode;
	struct ubifs_inode *dir_ui = ubifs_inode(dir);
	int sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	int err, budgeted = 1;
	struct ubifs_budget_req req = { .mod_dent = 1, .dirtied_ino = 2 };
	unsigned int saved_nlink = inode->i_nlink;

	/*
	 * Budget request settings: deletion direntry, deletion inode (+1 for
	 * @dirtied_ino), changing the parent directory inode. If budgeting
	 * fails, go ahead anyway because we have extra space reserved for
	 * deletions.
	 */

	dbg_gen("dent '%pd' from ino %lu (nlink %d) in dir ino %lu",
		dentry, inode->i_ino,
		inode->i_nlink, dir->i_ino);
	ubifs_assert(mutex_is_locked(&dir->i_mutex));
	ubifs_assert(mutex_is_locked(&inode->i_mutex));
	err = dbg_check_synced_i_size(c, inode);
	if (err)
		return err;

	err = ubifs_budget_space(c, &req);
	if (err) {
		if (err != -ENOSPC)
			return err;
		budgeted = 0;
	}

	lock_2_inodes(dir, inode);
	inode->i_ctime = ubifs_current_time(dir);
	drop_nlink(inode);
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 1, 0);
	if (err)
		goto out_cancel;
	unlock_2_inodes(dir, inode);

	if (budgeted)
		ubifs_release_budget(c, &req);
	else {
		/* We've deleted something - clean the "no space" flags */
		c->bi.nospace = c->bi.nospace_rp = 0;
		smp_wmb();
	}
	return 0;

out_cancel:
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	set_nlink(inode, saved_nlink);
	unlock_2_inodes(dir, inode);
	if (budgeted)
		ubifs_release_budget(c, &req);
	return err;
}

/**
 * check_dir_empty - check if a directory is empty or not.
 * @c: UBIFS file-system description object
 * @dir: VFS inode object of the directory to check
 *
 * This function checks if directory @dir is empty. Returns zero if the
 * directory is empty, %-ENOTEMPTY if it is not, and other negative error codes
 * in case of of errors.
 */
static int check_dir_empty(struct ubifs_info *c, struct inode *dir)
{
	struct qstr nm = { .name = NULL };
	struct ubifs_dent_node *dent;
	union ubifs_key key;
	int err;

	lowest_dent_key(c, &key, dir->i_ino);
	dent = ubifs_tnc_next_ent(c, &key, &nm);
	if (IS_ERR(dent)) {
		err = PTR_ERR(dent);
		if (err == -ENOENT)
			err = 0;
	} else {
		kfree(dent);
		err = -ENOTEMPTY;
	}
	return err;
}

static int ubifs_rmdir(struct inode *dir, struct dentry *dentry)
{
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	struct inode *inode = dentry->d_inode;
	int sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	int err, budgeted = 1;
	struct ubifs_inode *dir_ui = ubifs_inode(dir);
	struct ubifs_budget_req req = { .mod_dent = 1, .dirtied_ino = 2 };

	/*
	 * Budget request settings: deletion direntry, deletion inode and
	 * changing the parent inode. If budgeting fails, go ahead anyway
	 * because we have extra space reserved for deletions.
	 */

	dbg_gen("directory '%pd', ino %lu in dir ino %lu", dentry,
		inode->i_ino, dir->i_ino);
	ubifs_assert(mutex_is_locked(&dir->i_mutex));
	ubifs_assert(mutex_is_locked(&inode->i_mutex));
	err = check_dir_empty(c, dentry->d_inode);
	if (err)
		return err;

	err = ubifs_budget_space(c, &req);
	if (err) {
		if (err != -ENOSPC)
			return err;
		budgeted = 0;
	}

	lock_2_inodes(dir, inode);
	inode->i_ctime = ubifs_current_time(dir);
	clear_nlink(inode);
	drop_nlink(dir);
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 1, 0);
	if (err)
		goto out_cancel;
	unlock_2_inodes(dir, inode);

	if (budgeted)
		ubifs_release_budget(c, &req);
	else {
		/* We've deleted something - clean the "no space" flags */
		c->bi.nospace = c->bi.nospace_rp = 0;
		smp_wmb();
	}
	return 0;

out_cancel:
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	inc_nlink(dir);
	set_nlink(inode, 2);
	unlock_2_inodes(dir, inode);
	if (budgeted)
		ubifs_release_budget(c, &req);
	return err;
}

static int ubifs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
{
	struct inode *inode;
	struct ubifs_inode *dir_ui = ubifs_inode(dir);
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	int err, sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1 };

	/*
	 * Budget request settings: new inode, new direntry and changing parent
	 * directory inode.
	 */

	dbg_gen("dent '%pd', mode %#hx in dir ino %lu",
		dentry, mode, dir->i_ino);

	err = ubifs_budget_space(c, &req);
	if (err)
		return err;

	inode = ubifs_new_inode(c, dir, S_IFDIR | mode);
	if (IS_ERR(inode)) {
		err = PTR_ERR(inode);
		goto out_budg;
	}

	err = ubifs_init_security(dir, inode, &dentry->d_name);
	if (err)
		goto out_inode;

	mutex_lock(&dir_ui->ui_mutex);
	insert_inode_hash(inode);
	inc_nlink(inode);
	inc_nlink(dir);
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 0, 0);
	if (err) {
		ubifs_err(c, "cannot create directory, error %d", err);
		goto out_cancel;
	}
	mutex_unlock(&dir_ui->ui_mutex);

	ubifs_release_budget(c, &req);
	d_instantiate(dentry, inode);
	return 0;

out_cancel:
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	drop_nlink(dir);
	mutex_unlock(&dir_ui->ui_mutex);
out_inode:
	make_bad_inode(inode);
	iput(inode);
out_budg:
	ubifs_release_budget(c, &req);
	return err;
}

static int ubifs_mknod(struct inode *dir, struct dentry *dentry,
		       umode_t mode, dev_t rdev)
{
	struct inode *inode;
	struct ubifs_inode *ui;
	struct ubifs_inode *dir_ui = ubifs_inode(dir);
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	union ubifs_dev_desc *dev = NULL;
	int sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	int err, devlen = 0;
	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1,
					.new_ino_d = ALIGN(devlen, 8),
					.dirtied_ino = 1 };

	/*
	 * Budget request settings: new inode, new direntry and changing parent
	 * directory inode.
	 */

	dbg_gen("dent '%pd' in dir ino %lu", dentry, dir->i_ino);

	if (!new_valid_dev(rdev))
		return -EINVAL;

	if (S_ISBLK(mode) || S_ISCHR(mode)) {
		dev = kmalloc(sizeof(union ubifs_dev_desc), GFP_NOFS);
		if (!dev)
			return -ENOMEM;
		devlen = ubifs_encode_dev(dev, rdev);
	}

	err = ubifs_budget_space(c, &req);
	if (err) {
		kfree(dev);
		return err;
	}

	inode = ubifs_new_inode(c, dir, mode);
	if (IS_ERR(inode)) {
		kfree(dev);
		err = PTR_ERR(inode);
		goto out_budg;
	}

	init_special_inode(inode, inode->i_mode, rdev);
	inode->i_size = ubifs_inode(inode)->ui_size = devlen;
	ui = ubifs_inode(inode);
	ui->data = dev;
	ui->data_len = devlen;

	err = ubifs_init_security(dir, inode, &dentry->d_name);
	if (err)
		goto out_inode;

	mutex_lock(&dir_ui->ui_mutex);
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 0, 0);
	if (err)
		goto out_cancel;
	mutex_unlock(&dir_ui->ui_mutex);

	ubifs_release_budget(c, &req);
	insert_inode_hash(inode);
	d_instantiate(dentry, inode);
	return 0;

out_cancel:
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	mutex_unlock(&dir_ui->ui_mutex);
out_inode:
	make_bad_inode(inode);
	iput(inode);
out_budg:
	ubifs_release_budget(c, &req);
	return err;
}

static int ubifs_symlink(struct inode *dir, struct dentry *dentry,
			 const char *symname)
{
	struct inode *inode;
	struct ubifs_inode *ui;
	struct ubifs_inode *dir_ui = ubifs_inode(dir);
	struct ubifs_info *c = dir->i_sb->s_fs_info;
	int err, len = strlen(symname);
	int sz_change = CALC_DENT_SIZE(dentry->d_name.len);
	struct ubifs_budget_req req = { .new_ino = 1, .new_dent = 1,
					.new_ino_d = ALIGN(len, 8),
					.dirtied_ino = 1 };

	/*
	 * Budget request settings: new inode, new direntry and changing parent
	 * directory inode.
	 */

	dbg_gen("dent '%pd', target '%s' in dir ino %lu", dentry,
		symname, dir->i_ino);

	if (len > UBIFS_MAX_INO_DATA)
		return -ENAMETOOLONG;

	err = ubifs_budget_space(c, &req);
	if (err)
		return err;

	inode = ubifs_new_inode(c, dir, S_IFLNK | S_IRWXUGO);
	if (IS_ERR(inode)) {
		err = PTR_ERR(inode);
		goto out_budg;
	}

	ui = ubifs_inode(inode);
	ui->data = kmalloc(len + 1, GFP_NOFS);
	if (!ui->data) {
		err = -ENOMEM;
		goto out_inode;
	}

	memcpy(ui->data, symname, len);
	((char *)ui->data)[len] = '\0';
	/*
	 * The terminating zero byte is not written to the flash media and it
	 * is put just to make later in-memory string processing simpler. Thus,
	 * data length is @len, not @len + %1.
	 */
	ui->data_len = len;
	inode->i_size = ubifs_inode(inode)->ui_size = len;

	err = ubifs_init_security(dir, inode, &dentry->d_name);
	if (err)
		goto out_inode;

	mutex_lock(&dir_ui->ui_mutex);
	dir->i_size += sz_change;
	dir_ui->ui_size = dir->i_size;
	dir->i_mtime = dir->i_ctime = inode->i_ctime;
	err = ubifs_jnl_update(c, dir, &dentry->d_name, inode, 0, 0);
	if (err)
		goto out_cancel;
	mutex_unlock(&dir_ui->ui_mutex);

	ubifs_release_budget(c, &req);
	insert_inode_hash(inode);
	d_instantiate(dentry, inode);
	return 0;

out_cancel:
	dir->i_size -= sz_change;
	dir_ui->ui_size = dir->i_size;
	mutex_unlock(&dir_ui->ui_mutex);
out_inode:
	make_bad_inode(inode);
	iput(inode);
out_budg:
	ubifs_release_budget(c, &req);
	return err;
}

/**
 * lock_3_inodes - a wrapper for locking three UBIFS inodes.
 * @inode1: first inode
 * @inode2: second inode
 * @inode3: third inode
 *
 * This function is used for 'ubifs_rename()' and @inode1 may be the same as
 * @inode2 whereas @inode3 may be %NULL.
 *
 * We do not implement any tricks to guarantee strict lock ordering, because
 * VFS has already done it for us on the @i_mutex. So this is just a simple
 * wrapper function.
 */
static void lock_3_inodes(struct inode *inode1, struct inode *inode2,
			  struct inode *inode3)
{
	mutex_lock_nested(&ubifs_inode(inode1)->ui_mutex, WB_MUTEX_1);
	if (inode2 != inode1)
		mutex_lock_nested(&ubifs_inode(inode2)->ui_mutex, WB_MUTEX_2);
	if (inode3)
		mutex_lock_nested(&ubifs_inode(inode3)->ui_mutex, WB_MUTEX_3);
}

/**
 * unlock_3_inodes - a wrapper for unlocking three UBIFS inodes for rename.
 * @inode1: first inode
 * @inode2: second inode
 * @inode3: third inode
 */
static void unlock_3_inodes(struct inode *inode1, struct inode *inode2,
			    struct inode *inode3)
{
	if (inode3)
		mutex_unlock(&ubifs_inode(inode3)->ui_mutex);
	if (inode1 != inode2)
		mutex_unlock(&ubifs_inode(inode2)->ui_mutex);
	mutex_unlock(&ubifs_inode(inode1)->ui_mutex);
}

static int ubifs_rename(struct inode *old_dir, struct dentry *old_dentry,
			struct inode *new_dir, struct dentry *new_dentry)
{
	struct ubifs_info *c = old_dir->i_sb->s_fs_info;
	struct inode *old_inode = old_dentry->d_inode;
	struct inode *new_inode = new_dentry->d_inode;
	struct ubifs_inode *old_inode_ui = ubifs_inode(old_inode);
	int err, release, sync = 0, move = (new_dir != old_dir);
	int is_dir = S_ISDIR(old_inode->i_mode);
	int unlink = !!new_inode;
	int new_sz = CALC_DENT_SIZE(new_dentry->d_name.len);
	int old_sz = CALC_DENT_SIZE(old_dentry->d_name.len);
	struct ubifs_budget_req req = { .new_dent = 1, .mod_dent = 1,
					.dirtied_ino = 3 };
	struct ubifs_budget_req ino_req = { .dirtied_ino = 1,
			.dirtied_ino_d = ALIGN(old_inode_ui->data_len, 8) };
	struct timespec time;
	unsigned int uninitialized_var(saved_nlink);

	/*
	 * Budget request settings: deletion direntry, new direntry, removing
	 * the old inode, and changing old and new parent directory inodes.
	 *
	 * However, this operation also marks the target inode as dirty and
	 * does not write it, so we allocate budget for the target inode
	 * separately.
	 */

	dbg_gen("dent '%pd' ino %lu in dir ino %lu to dent '%pd' in dir ino %lu",
		old_dentry, old_inode->i_ino, old_dir->i_ino,
		new_dentry, new_dir->i_ino);
	ubifs_assert(mutex_is_locked(&old_dir->i_mutex));
	ubifs_assert(mutex_is_locked(&new_dir->i_mutex));
	if (unlink)
		ubifs_assert(mutex_is_locked(&new_inode->i_mutex));


	if (unlink && is_dir) {
		err = check_dir_empty(c, new_inode);
		if (err)
			return err;
	}

	err = ubifs_budget_space(c, &req);
	if (err)
		return err;
	err = ubifs_budget_space(c, &ino_req);
	if (err) {
		ubifs_release_budget(c, &req);
		return err;
	}

	lock_3_inodes(old_dir, new_dir, new_inode);

	/*
	 * Like most other Unix systems, set the @i_ctime for inodes on a
	 * rename.
	 */
	time = ubifs_current_time(old_dir);
	old_inode->i_ctime = time;

	/* We must adjust parent link count when renaming directories */
	if (is_dir) {
		if (move) {
			/*
			 * @old_dir loses a link because we are moving
			 * @old_inode to a different directory.
			 */
			drop_nlink(old_dir);
			/*
			 * @new_dir only gains a link if we are not also
			 * overwriting an existing directory.
			 */
			if (!unlink)
				inc_nlink(new_dir);
		} else {
			/*
			 * @old_inode is not moving to a different directory,
			 * but @old_dir still loses a link if we are
			 * overwriting an existing directory.
			 */
			if (unlink)
				drop_nlink(old_dir);
		}
	}

	old_dir->i_size -= old_sz;
	ubifs_inode(old_dir)->ui_size = old_dir->i_size;
	old_dir->i_mtime = old_dir->i_ctime = time;
	new_dir->i_mtime = new_dir->i_ctime = time;

	/*
	 * And finally, if we unlinked a direntry which happened to have the
	 * same name as the moved direntry, we have to decrement @i_nlink of
	 * the unlinked inode and change its ctime.
	 */
	if (unlink) {
		/*
		 * Directories cannot have hard-links, so if this is a
		 * directory, just clear @i_nlink.
		 */
		saved_nlink = new_inode->i_nlink;
		if (is_dir)
			clear_nlink(new_inode);
		else
			drop_nlink(new_inode);
		new_inode->i_ctime = time;
	} else {
		new_dir->i_size += new_sz;
		ubifs_inode(new_dir)->ui_size = new_dir->i_size;
	}

	/*
	 * Do not ask 'ubifs_jnl_rename()' to flush write-buffer if @old_inode
	 * is dirty, because this will be done later on at the end of
	 * 'ubifs_rename()'.
	 */
	if (IS_SYNC(old_inode)) {
		sync = IS_DIRSYNC(old_dir) || IS_DIRSYNC(new_dir);
		if (unlink && IS_SYNC(new_inode))
			sync = 1;
	}
	err = ubifs_jnl_rename(c, old_dir, old_dentry, new_dir, new_dentry,
			       sync);
	if (err)
		goto out_cancel;

	unlock_3_inodes(old_dir, new_dir, new_inode);
	ubifs_release_budget(c, &req);

	mutex_lock(&old_inode_ui->ui_mutex);
	release = old_inode_ui->dirty;
	mark_inode_dirty_sync(old_inode);
	mutex_unlock(&old_inode_ui->ui_mutex);

	if (release)
		ubifs_release_budget(c, &ino_req);
	if (IS_SYNC(old_inode))
		err = old_inode->i_sb->s_op->write_inode(old_inode, NULL);
	return err;

out_cancel:
	if (unlink) {
		set_nlink(new_inode, saved_nlink);
	} else {
		new_dir->i_size -= new_sz;
		ubifs_inode(new_dir)->ui_size = new_dir->i_size;
	}
	old_dir->i_size += old_sz;
	ubifs_inode(old_dir)->ui_size = old_dir->i_size;
	if (is_dir) {
		if (move) {
			inc_nlink(old_dir);
			if (!unlink)
				drop_nlink(new_dir);
		} else {
			if (unlink)
				inc_nlink(old_dir);
		}
	}
	unlock_3_inodes(old_dir, new_dir, new_inode);
	ubifs_release_budget(c, &ino_req);
	ubifs_release_budget(c, &req);
	return err;
}

int ubifs_getattr(struct vfsmount *mnt, struct dentry *dentry,
		  struct kstat *stat)
{
	loff_t size;
	struct inode *inode = dentry->d_inode;
	struct ubifs_inode *ui = ubifs_inode(inode);

	mutex_lock(&ui->ui_mutex);
	generic_fillattr(inode, stat);
	stat->blksize = UBIFS_BLOCK_SIZE;
	stat->size = ui->ui_size;

	/*
	 * Unfortunately, the 'stat()' system call was designed for block
	 * device based file systems, and it is not appropriate for UBIFS,
	 * because UBIFS does not have notion of "block". For example, it is
	 * difficult to tell how many block a directory takes - it actually
	 * takes less than 300 bytes, but we have to round it to block size,
	 * which introduces large mistake. This makes utilities like 'du' to
	 * report completely senseless numbers. This is the reason why UBIFS
	 * goes the same way as JFFS2 - it reports zero blocks for everything
	 * but regular files, which makes more sense than reporting completely
	 * wrong sizes.
	 */
	if (S_ISREG(inode->i_mode)) {
		size = ui->xattr_size;
		size += stat->size;
		size = ALIGN(size, UBIFS_BLOCK_SIZE);
		/*
		 * Note, user-space expects 512-byte blocks count irrespectively
		 * of what was reported in @stat->size.
		 */
		stat->blocks = size >> 9;
	} else
		stat->blocks = 0;
	mutex_unlock(&ui->ui_mutex);
	return 0;
}

const struct inode_operations ubifs_dir_inode_operations = {
	.lookup      = ubifs_lookup,
	.create      = ubifs_create,
	.link        = ubifs_link,
	.symlink     = ubifs_symlink,
	.unlink      = ubifs_unlink,
	.mkdir       = ubifs_mkdir,
	.rmdir       = ubifs_rmdir,
	.mknod       = ubifs_mknod,
	.rename      = ubifs_rename,
	.setattr     = ubifs_setattr,
	.getattr     = ubifs_getattr,
	.setxattr    = ubifs_setxattr,
	.getxattr    = ubifs_getxattr,
	.listxattr   = ubifs_listxattr,
	.removexattr = ubifs_removexattr,
};

const struct file_operations ubifs_dir_operations = {
	.llseek         = generic_file_llseek,
	.release        = ubifs_dir_release,
	.read           = generic_read_dir,
	.iterate        = ubifs_readdir,
	.fsync          = ubifs_fsync,
	.unlocked_ioctl = ubifs_ioctl,
#ifdef CONFIG_COMPAT
	.compat_ioctl   = ubifs_compat_ioctl,
#endif
};