summaryrefslogtreecommitdiff
path: root/drivers/mtd/ubi/eba.c
blob: 61356a9b13032b2ae1304984758a5d94fb670255 (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
/*
 * Copyright (c) International Business Machines Corp., 2006
 *
 * 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 02111-1307 USA
 *
 * Author: Artem Bityutskiy (Битюцкий Артём)
 */

/*
 * The UBI Eraseblock Association (EBA) sub-system.
 *
 * This sub-system is responsible for I/O to/from logical eraseblock.
 *
 * Although in this implementation the EBA table is fully kept and managed in
 * RAM, which assumes poor scalability, it might be (partially) maintained on
 * flash in future implementations.
 *
 * The EBA sub-system implements per-logical eraseblock locking. Before
 * accessing a logical eraseblock it is locked for reading or writing. The
 * per-logical eraseblock locking is implemented by means of the lock tree. The
 * lock tree is an RB-tree which refers all the currently locked logical
 * eraseblocks. The lock tree elements are &struct ubi_ltree_entry objects.
 * They are indexed by (@vol_id, @lnum) pairs.
 *
 * EBA also maintains the global sequence counter which is incremented each
 * time a logical eraseblock is mapped to a physical eraseblock and it is
 * stored in the volume identifier header. This means that each VID header has
 * a unique sequence number. The sequence number is only increased an we assume
 * 64 bits is enough to never overflow.
 */

#include <linux/slab.h>
#include <linux/crc32.h>
#include <linux/err.h>
#include "ubi.h"

/* Number of physical eraseblocks reserved for atomic LEB change operation */
#define EBA_RESERVED_PEBS 1

/**
 * next_sqnum - get next sequence number.
 * @ubi: UBI device description object
 *
 * This function returns next sequence number to use, which is just the current
 * global sequence counter value. It also increases the global sequence
 * counter.
 */
unsigned long long ubi_next_sqnum(struct ubi_device *ubi)
{
	unsigned long long sqnum;

	spin_lock(&ubi->ltree_lock);
	sqnum = ubi->global_sqnum++;
	spin_unlock(&ubi->ltree_lock);

	return sqnum;
}

/**
 * ubi_get_compat - get compatibility flags of a volume.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 *
 * This function returns compatibility flags for an internal volume. User
 * volumes have no compatibility flags, so %0 is returned.
 */
static int ubi_get_compat(const struct ubi_device *ubi, int vol_id)
{
	if (vol_id == UBI_LAYOUT_VOLUME_ID)
		return UBI_LAYOUT_VOLUME_COMPAT;
	return 0;
}

/**
 * ltree_lookup - look up the lock tree.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 *
 * This function returns a pointer to the corresponding &struct ubi_ltree_entry
 * object if the logical eraseblock is locked and %NULL if it is not.
 * @ubi->ltree_lock has to be locked.
 */
static struct ubi_ltree_entry *ltree_lookup(struct ubi_device *ubi, int vol_id,
					    int lnum)
{
	struct rb_node *p;

	p = ubi->ltree.rb_node;
	while (p) {
		struct ubi_ltree_entry *le;

		le = rb_entry(p, struct ubi_ltree_entry, rb);

		if (vol_id < le->vol_id)
			p = p->rb_left;
		else if (vol_id > le->vol_id)
			p = p->rb_right;
		else {
			if (lnum < le->lnum)
				p = p->rb_left;
			else if (lnum > le->lnum)
				p = p->rb_right;
			else
				return le;
		}
	}

	return NULL;
}

/**
 * ltree_add_entry - add new entry to the lock tree.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 *
 * This function adds new entry for logical eraseblock (@vol_id, @lnum) to the
 * lock tree. If such entry is already there, its usage counter is increased.
 * Returns pointer to the lock tree entry or %-ENOMEM if memory allocation
 * failed.
 */
static struct ubi_ltree_entry *ltree_add_entry(struct ubi_device *ubi,
					       int vol_id, int lnum)
{
	struct ubi_ltree_entry *le, *le1, *le_free;

	le = kmalloc(sizeof(struct ubi_ltree_entry), GFP_NOFS);
	if (!le)
		return ERR_PTR(-ENOMEM);

	le->users = 0;
	init_rwsem(&le->mutex);
	le->vol_id = vol_id;
	le->lnum = lnum;

	spin_lock(&ubi->ltree_lock);
	le1 = ltree_lookup(ubi, vol_id, lnum);

	if (le1) {
		/*
		 * This logical eraseblock is already locked. The newly
		 * allocated lock entry is not needed.
		 */
		le_free = le;
		le = le1;
	} else {
		struct rb_node **p, *parent = NULL;

		/*
		 * No lock entry, add the newly allocated one to the
		 * @ubi->ltree RB-tree.
		 */
		le_free = NULL;

		p = &ubi->ltree.rb_node;
		while (*p) {
			parent = *p;
			le1 = rb_entry(parent, struct ubi_ltree_entry, rb);

			if (vol_id < le1->vol_id)
				p = &(*p)->rb_left;
			else if (vol_id > le1->vol_id)
				p = &(*p)->rb_right;
			else {
				ubi_assert(lnum != le1->lnum);
				if (lnum < le1->lnum)
					p = &(*p)->rb_left;
				else
					p = &(*p)->rb_right;
			}
		}

		rb_link_node(&le->rb, parent, p);
		rb_insert_color(&le->rb, &ubi->ltree);
	}
	le->users += 1;
	spin_unlock(&ubi->ltree_lock);

	kfree(le_free);
	return le;
}

/**
 * leb_read_lock - lock logical eraseblock for reading.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 *
 * This function locks a logical eraseblock for reading. Returns zero in case
 * of success and a negative error code in case of failure.
 */
static int leb_read_lock(struct ubi_device *ubi, int vol_id, int lnum)
{
	struct ubi_ltree_entry *le;

	le = ltree_add_entry(ubi, vol_id, lnum);
	if (IS_ERR(le))
		return PTR_ERR(le);
	down_read(&le->mutex);
	return 0;
}

/**
 * leb_read_unlock - unlock logical eraseblock.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 */
static void leb_read_unlock(struct ubi_device *ubi, int vol_id, int lnum)
{
	struct ubi_ltree_entry *le;

	spin_lock(&ubi->ltree_lock);
	le = ltree_lookup(ubi, vol_id, lnum);
	le->users -= 1;
	ubi_assert(le->users >= 0);
	up_read(&le->mutex);
	if (le->users == 0) {
		rb_erase(&le->rb, &ubi->ltree);
		kfree(le);
	}
	spin_unlock(&ubi->ltree_lock);
}

/**
 * leb_write_lock - lock logical eraseblock for writing.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 *
 * This function locks a logical eraseblock for writing. Returns zero in case
 * of success and a negative error code in case of failure.
 */
static int leb_write_lock(struct ubi_device *ubi, int vol_id, int lnum)
{
	struct ubi_ltree_entry *le;

	le = ltree_add_entry(ubi, vol_id, lnum);
	if (IS_ERR(le))
		return PTR_ERR(le);
	down_write(&le->mutex);
	return 0;
}

/**
 * leb_write_lock - lock logical eraseblock for writing.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 *
 * This function locks a logical eraseblock for writing if there is no
 * contention and does nothing if there is contention. Returns %0 in case of
 * success, %1 in case of contention, and and a negative error code in case of
 * failure.
 */
static int leb_write_trylock(struct ubi_device *ubi, int vol_id, int lnum)
{
	struct ubi_ltree_entry *le;

	le = ltree_add_entry(ubi, vol_id, lnum);
	if (IS_ERR(le))
		return PTR_ERR(le);
	if (down_write_trylock(&le->mutex))
		return 0;

	/* Contention, cancel */
	spin_lock(&ubi->ltree_lock);
	le->users -= 1;
	ubi_assert(le->users >= 0);
	if (le->users == 0) {
		rb_erase(&le->rb, &ubi->ltree);
		kfree(le);
	}
	spin_unlock(&ubi->ltree_lock);

	return 1;
}

/**
 * leb_write_unlock - unlock logical eraseblock.
 * @ubi: UBI device description object
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 */
static void leb_write_unlock(struct ubi_device *ubi, int vol_id, int lnum)
{
	struct ubi_ltree_entry *le;

	spin_lock(&ubi->ltree_lock);
	le = ltree_lookup(ubi, vol_id, lnum);
	le->users -= 1;
	ubi_assert(le->users >= 0);
	up_write(&le->mutex);
	if (le->users == 0) {
		rb_erase(&le->rb, &ubi->ltree);
		kfree(le);
	}
	spin_unlock(&ubi->ltree_lock);
}

/**
 * ubi_eba_unmap_leb - un-map logical eraseblock.
 * @ubi: UBI device description object
 * @vol: volume description object
 * @lnum: logical eraseblock number
 *
 * This function un-maps logical eraseblock @lnum and schedules corresponding
 * physical eraseblock for erasure. Returns zero in case of success and a
 * negative error code in case of failure.
 */
int ubi_eba_unmap_leb(struct ubi_device *ubi, struct ubi_volume *vol,
		      int lnum)
{
	int err, pnum, vol_id = vol->vol_id;

	if (ubi->ro_mode)
		return -EROFS;

	err = leb_write_lock(ubi, vol_id, lnum);
	if (err)
		return err;

	pnum = vol->eba_tbl[lnum];
	if (pnum < 0)
		/* This logical eraseblock is already unmapped */
		goto out_unlock;

	dbg_eba("erase LEB %d:%d, PEB %d", vol_id, lnum, pnum);

	down_read(&ubi->fm_eba_sem);
	vol->eba_tbl[lnum] = UBI_LEB_UNMAPPED;
	up_read(&ubi->fm_eba_sem);
	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 0);

out_unlock:
	leb_write_unlock(ubi, vol_id, lnum);
	return err;
}

/**
 * ubi_eba_read_leb - read data.
 * @ubi: UBI device description object
 * @vol: volume description object
 * @lnum: logical eraseblock number
 * @buf: buffer to store the read data
 * @offset: offset from where to read
 * @len: how many bytes to read
 * @check: data CRC check flag
 *
 * If the logical eraseblock @lnum is unmapped, @buf is filled with 0xFF
 * bytes. The @check flag only makes sense for static volumes and forces
 * eraseblock data CRC checking.
 *
 * In case of success this function returns zero. In case of a static volume,
 * if data CRC mismatches - %-EBADMSG is returned. %-EBADMSG may also be
 * returned for any volume type if an ECC error was detected by the MTD device
 * driver. Other negative error cored may be returned in case of other errors.
 */
int ubi_eba_read_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
		     void *buf, int offset, int len, int check)
{
	int err, pnum, scrub = 0, vol_id = vol->vol_id;
	struct ubi_vid_hdr *vid_hdr;
	uint32_t uninitialized_var(crc);

	err = leb_read_lock(ubi, vol_id, lnum);
	if (err)
		return err;

	pnum = vol->eba_tbl[lnum];
	if (pnum < 0) {
		/*
		 * The logical eraseblock is not mapped, fill the whole buffer
		 * with 0xFF bytes. The exception is static volumes for which
		 * it is an error to read unmapped logical eraseblocks.
		 */
		dbg_eba("read %d bytes from offset %d of LEB %d:%d (unmapped)",
			len, offset, vol_id, lnum);
		leb_read_unlock(ubi, vol_id, lnum);
		ubi_assert(vol->vol_type != UBI_STATIC_VOLUME);
		memset(buf, 0xFF, len);
		return 0;
	}

	dbg_eba("read %d bytes from offset %d of LEB %d:%d, PEB %d",
		len, offset, vol_id, lnum, pnum);

	if (vol->vol_type == UBI_DYNAMIC_VOLUME)
		check = 0;

retry:
	if (check) {
		vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
		if (!vid_hdr) {
			err = -ENOMEM;
			goto out_unlock;
		}

		err = ubi_io_read_vid_hdr(ubi, pnum, vid_hdr, 1);
		if (err && err != UBI_IO_BITFLIPS) {
			if (err > 0) {
				/*
				 * The header is either absent or corrupted.
				 * The former case means there is a bug -
				 * switch to read-only mode just in case.
				 * The latter case means a real corruption - we
				 * may try to recover data. FIXME: but this is
				 * not implemented.
				 */
				if (err == UBI_IO_BAD_HDR_EBADMSG ||
				    err == UBI_IO_BAD_HDR) {
					ubi_warn(ubi, "corrupted VID header at PEB %d, LEB %d:%d",
						 pnum, vol_id, lnum);
					err = -EBADMSG;
				} else {
					err = -EINVAL;
					ubi_ro_mode(ubi);
				}
			}
			goto out_free;
		} else if (err == UBI_IO_BITFLIPS)
			scrub = 1;

		ubi_assert(lnum < be32_to_cpu(vid_hdr->used_ebs));
		ubi_assert(len == be32_to_cpu(vid_hdr->data_size));

		crc = be32_to_cpu(vid_hdr->data_crc);
		ubi_free_vid_hdr(ubi, vid_hdr);
	}

	err = ubi_io_read_data(ubi, buf, pnum, offset, len);
	if (err) {
		if (err == UBI_IO_BITFLIPS)
			scrub = 1;
		else if (mtd_is_eccerr(err)) {
			if (vol->vol_type == UBI_DYNAMIC_VOLUME)
				goto out_unlock;
			scrub = 1;
			if (!check) {
				ubi_msg(ubi, "force data checking");
				check = 1;
				goto retry;
			}
		} else
			goto out_unlock;
	}

	if (check) {
		uint32_t crc1 = crc32(UBI_CRC32_INIT, buf, len);
		if (crc1 != crc) {
			ubi_warn(ubi, "CRC error: calculated %#08x, must be %#08x",
				 crc1, crc);
			err = -EBADMSG;
			goto out_unlock;
		}
	}

	if (scrub)
		err = ubi_wl_scrub_peb(ubi, pnum);

	leb_read_unlock(ubi, vol_id, lnum);
	return err;

out_free:
	ubi_free_vid_hdr(ubi, vid_hdr);
out_unlock:
	leb_read_unlock(ubi, vol_id, lnum);
	return err;
}

/**
 * ubi_eba_read_leb_sg - read data into a scatter gather list.
 * @ubi: UBI device description object
 * @vol: volume description object
 * @lnum: logical eraseblock number
 * @sgl: UBI scatter gather list to store the read data
 * @offset: offset from where to read
 * @len: how many bytes to read
 * @check: data CRC check flag
 *
 * This function works exactly like ubi_eba_read_leb(). But instead of
 * storing the read data into a buffer it writes to an UBI scatter gather
 * list.
 */
int ubi_eba_read_leb_sg(struct ubi_device *ubi, struct ubi_volume *vol,
			struct ubi_sgl *sgl, int lnum, int offset, int len,
			int check)
{
	int to_read;
	int ret;
	struct scatterlist *sg;

	for (;;) {
		ubi_assert(sgl->list_pos < UBI_MAX_SG_COUNT);
		sg = &sgl->sg[sgl->list_pos];
		if (len < sg->length - sgl->page_pos)
			to_read = len;
		else
			to_read = sg->length - sgl->page_pos;

		ret = ubi_eba_read_leb(ubi, vol, lnum,
				       sg_virt(sg) + sgl->page_pos, offset,
				       to_read, check);
		if (ret < 0)
			return ret;

		offset += to_read;
		len -= to_read;
		if (!len) {
			sgl->page_pos += to_read;
			if (sgl->page_pos == sg->length) {
				sgl->list_pos++;
				sgl->page_pos = 0;
			}

			break;
		}

		sgl->list_pos++;
		sgl->page_pos = 0;
	}

	return ret;
}

/**
 * recover_peb - recover from write failure.
 * @ubi: UBI device description object
 * @pnum: the physical eraseblock to recover
 * @vol_id: volume ID
 * @lnum: logical eraseblock number
 * @buf: data which was not written because of the write failure
 * @offset: offset of the failed write
 * @len: how many bytes should have been written
 *
 * This function is called in case of a write failure and moves all good data
 * from the potentially bad physical eraseblock to a good physical eraseblock.
 * This function also writes the data which was not written due to the failure.
 * Returns new physical eraseblock number in case of success, and a negative
 * error code in case of failure.
 */
static int recover_peb(struct ubi_device *ubi, int pnum, int vol_id, int lnum,
		       const void *buf, int offset, int len)
{
	int err, idx = vol_id2idx(ubi, vol_id), new_pnum, data_size, tries = 0;
	struct ubi_volume *vol = ubi->volumes[idx];
	struct ubi_vid_hdr *vid_hdr;

	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
	if (!vid_hdr)
		return -ENOMEM;

retry:
	new_pnum = ubi_wl_get_peb(ubi);
	if (new_pnum < 0) {
		ubi_free_vid_hdr(ubi, vid_hdr);
		up_read(&ubi->fm_eba_sem);
		return new_pnum;
	}

	ubi_msg(ubi, "recover PEB %d, move data to PEB %d",
		pnum, new_pnum);

	err = ubi_io_read_vid_hdr(ubi, pnum, vid_hdr, 1);
	if (err && err != UBI_IO_BITFLIPS) {
		if (err > 0)
			err = -EIO;
		up_read(&ubi->fm_eba_sem);
		goto out_put;
	}

	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	err = ubi_io_write_vid_hdr(ubi, new_pnum, vid_hdr);
	if (err) {
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	data_size = offset + len;
	mutex_lock(&ubi->buf_mutex);
	memset(ubi->peb_buf + offset, 0xFF, len);

	/* Read everything before the area where the write failure happened */
	if (offset > 0) {
		err = ubi_io_read_data(ubi, ubi->peb_buf, pnum, 0, offset);
		if (err && err != UBI_IO_BITFLIPS) {
			up_read(&ubi->fm_eba_sem);
			goto out_unlock;
		}
	}

	memcpy(ubi->peb_buf + offset, buf, len);

	err = ubi_io_write_data(ubi, ubi->peb_buf, new_pnum, 0, data_size);
	if (err) {
		mutex_unlock(&ubi->buf_mutex);
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	mutex_unlock(&ubi->buf_mutex);
	ubi_free_vid_hdr(ubi, vid_hdr);

	vol->eba_tbl[lnum] = new_pnum;
	up_read(&ubi->fm_eba_sem);
	ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);

	ubi_msg(ubi, "data was successfully recovered");
	return 0;

out_unlock:
	mutex_unlock(&ubi->buf_mutex);
out_put:
	ubi_wl_put_peb(ubi, vol_id, lnum, new_pnum, 1);
	ubi_free_vid_hdr(ubi, vid_hdr);
	return err;

write_error:
	/*
	 * Bad luck? This physical eraseblock is bad too? Crud. Let's try to
	 * get another one.
	 */
	ubi_warn(ubi, "failed to write to PEB %d", new_pnum);
	ubi_wl_put_peb(ubi, vol_id, lnum, new_pnum, 1);
	if (++tries > UBI_IO_RETRIES) {
		ubi_free_vid_hdr(ubi, vid_hdr);
		return err;
	}
	ubi_msg(ubi, "try again");
	goto retry;
}

/**
 * ubi_eba_write_leb - write data to dynamic volume.
 * @ubi: UBI device description object
 * @vol: volume description object
 * @lnum: logical eraseblock number
 * @buf: the data to write
 * @offset: offset within the logical eraseblock where to write
 * @len: how many bytes to write
 *
 * This function writes data to logical eraseblock @lnum of a dynamic volume
 * @vol. Returns zero in case of success and a negative error code in case
 * of failure. In case of error, it is possible that something was still
 * written to the flash media, but may be some garbage.
 */
int ubi_eba_write_leb(struct ubi_device *ubi, struct ubi_volume *vol, int lnum,
		      const void *buf, int offset, int len)
{
	int err, pnum, tries = 0, vol_id = vol->vol_id;
	struct ubi_vid_hdr *vid_hdr;

	if (ubi->ro_mode)
		return -EROFS;

	err = leb_write_lock(ubi, vol_id, lnum);
	if (err)
		return err;

	pnum = vol->eba_tbl[lnum];
	if (pnum >= 0) {
		dbg_eba("write %d bytes at offset %d of LEB %d:%d, PEB %d",
			len, offset, vol_id, lnum, pnum);

		err = ubi_io_write_data(ubi, buf, pnum, offset, len);
		if (err) {
			ubi_warn(ubi, "failed to write data to PEB %d", pnum);
			if (err == -EIO && ubi->bad_allowed)
				err = recover_peb(ubi, pnum, vol_id, lnum, buf,
						  offset, len);
			if (err)
				ubi_ro_mode(ubi);
		}
		leb_write_unlock(ubi, vol_id, lnum);
		return err;
	}

	/*
	 * The logical eraseblock is not mapped. We have to get a free physical
	 * eraseblock and write the volume identifier header there first.
	 */
	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
	if (!vid_hdr) {
		leb_write_unlock(ubi, vol_id, lnum);
		return -ENOMEM;
	}

	vid_hdr->vol_type = UBI_VID_DYNAMIC;
	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	vid_hdr->vol_id = cpu_to_be32(vol_id);
	vid_hdr->lnum = cpu_to_be32(lnum);
	vid_hdr->compat = ubi_get_compat(ubi, vol_id);
	vid_hdr->data_pad = cpu_to_be32(vol->data_pad);

retry:
	pnum = ubi_wl_get_peb(ubi);
	if (pnum < 0) {
		ubi_free_vid_hdr(ubi, vid_hdr);
		leb_write_unlock(ubi, vol_id, lnum);
		up_read(&ubi->fm_eba_sem);
		return pnum;
	}

	dbg_eba("write VID hdr and %d bytes at offset %d of LEB %d:%d, PEB %d",
		len, offset, vol_id, lnum, pnum);

	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
	if (err) {
		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
			 vol_id, lnum, pnum);
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	if (len) {
		err = ubi_io_write_data(ubi, buf, pnum, offset, len);
		if (err) {
			ubi_warn(ubi, "failed to write %d bytes at offset %d of LEB %d:%d, PEB %d",
				 len, offset, vol_id, lnum, pnum);
			up_read(&ubi->fm_eba_sem);
			goto write_error;
		}
	}

	vol->eba_tbl[lnum] = pnum;
	up_read(&ubi->fm_eba_sem);

	leb_write_unlock(ubi, vol_id, lnum);
	ubi_free_vid_hdr(ubi, vid_hdr);
	return 0;

write_error:
	if (err != -EIO || !ubi->bad_allowed) {
		ubi_ro_mode(ubi);
		leb_write_unlock(ubi, vol_id, lnum);
		ubi_free_vid_hdr(ubi, vid_hdr);
		return err;
	}

	/*
	 * Fortunately, this is the first write operation to this physical
	 * eraseblock, so just put it and request a new one. We assume that if
	 * this physical eraseblock went bad, the erase code will handle that.
	 */
	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
	if (err || ++tries > UBI_IO_RETRIES) {
		ubi_ro_mode(ubi);
		leb_write_unlock(ubi, vol_id, lnum);
		ubi_free_vid_hdr(ubi, vid_hdr);
		return err;
	}

	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	ubi_msg(ubi, "try another PEB");
	goto retry;
}

/**
 * ubi_eba_write_leb_st - write data to static volume.
 * @ubi: UBI device description object
 * @vol: volume description object
 * @lnum: logical eraseblock number
 * @buf: data to write
 * @len: how many bytes to write
 * @used_ebs: how many logical eraseblocks will this volume contain
 *
 * This function writes data to logical eraseblock @lnum of static volume
 * @vol. The @used_ebs argument should contain total number of logical
 * eraseblock in this static volume.
 *
 * When writing to the last logical eraseblock, the @len argument doesn't have
 * to be aligned to the minimal I/O unit size. Instead, it has to be equivalent
 * to the real data size, although the @buf buffer has to contain the
 * alignment. In all other cases, @len has to be aligned.
 *
 * It is prohibited to write more than once to logical eraseblocks of static
 * volumes. This function returns zero in case of success and a negative error
 * code in case of failure.
 */
int ubi_eba_write_leb_st(struct ubi_device *ubi, struct ubi_volume *vol,
			 int lnum, const void *buf, int len, int used_ebs)
{
	int err, pnum, tries = 0, data_size = len, vol_id = vol->vol_id;
	struct ubi_vid_hdr *vid_hdr;
	uint32_t crc;

	if (ubi->ro_mode)
		return -EROFS;

	if (lnum == used_ebs - 1)
		/* If this is the last LEB @len may be unaligned */
		len = ALIGN(data_size, ubi->min_io_size);
	else
		ubi_assert(!(len & (ubi->min_io_size - 1)));

	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
	if (!vid_hdr)
		return -ENOMEM;

	err = leb_write_lock(ubi, vol_id, lnum);
	if (err) {
		ubi_free_vid_hdr(ubi, vid_hdr);
		return err;
	}

	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	vid_hdr->vol_id = cpu_to_be32(vol_id);
	vid_hdr->lnum = cpu_to_be32(lnum);
	vid_hdr->compat = ubi_get_compat(ubi, vol_id);
	vid_hdr->data_pad = cpu_to_be32(vol->data_pad);

	crc = crc32(UBI_CRC32_INIT, buf, data_size);
	vid_hdr->vol_type = UBI_VID_STATIC;
	vid_hdr->data_size = cpu_to_be32(data_size);
	vid_hdr->used_ebs = cpu_to_be32(used_ebs);
	vid_hdr->data_crc = cpu_to_be32(crc);

retry:
	pnum = ubi_wl_get_peb(ubi);
	if (pnum < 0) {
		ubi_free_vid_hdr(ubi, vid_hdr);
		leb_write_unlock(ubi, vol_id, lnum);
		up_read(&ubi->fm_eba_sem);
		return pnum;
	}

	dbg_eba("write VID hdr and %d bytes at LEB %d:%d, PEB %d, used_ebs %d",
		len, vol_id, lnum, pnum, used_ebs);

	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
	if (err) {
		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
			 vol_id, lnum, pnum);
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	err = ubi_io_write_data(ubi, buf, pnum, 0, len);
	if (err) {
		ubi_warn(ubi, "failed to write %d bytes of data to PEB %d",
			 len, pnum);
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	ubi_assert(vol->eba_tbl[lnum] < 0);
	vol->eba_tbl[lnum] = pnum;
	up_read(&ubi->fm_eba_sem);

	leb_write_unlock(ubi, vol_id, lnum);
	ubi_free_vid_hdr(ubi, vid_hdr);
	return 0;

write_error:
	if (err != -EIO || !ubi->bad_allowed) {
		/*
		 * This flash device does not admit of bad eraseblocks or
		 * something nasty and unexpected happened. Switch to read-only
		 * mode just in case.
		 */
		ubi_ro_mode(ubi);
		leb_write_unlock(ubi, vol_id, lnum);
		ubi_free_vid_hdr(ubi, vid_hdr);
		return err;
	}

	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
	if (err || ++tries > UBI_IO_RETRIES) {
		ubi_ro_mode(ubi);
		leb_write_unlock(ubi, vol_id, lnum);
		ubi_free_vid_hdr(ubi, vid_hdr);
		return err;
	}

	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	ubi_msg(ubi, "try another PEB");
	goto retry;
}

/*
 * ubi_eba_atomic_leb_change - change logical eraseblock atomically.
 * @ubi: UBI device description object
 * @vol: volume description object
 * @lnum: logical eraseblock number
 * @buf: data to write
 * @len: how many bytes to write
 *
 * This function changes the contents of a logical eraseblock atomically. @buf
 * has to contain new logical eraseblock data, and @len - the length of the
 * data, which has to be aligned. This function guarantees that in case of an
 * unclean reboot the old contents is preserved. Returns zero in case of
 * success and a negative error code in case of failure.
 *
 * UBI reserves one LEB for the "atomic LEB change" operation, so only one
 * LEB change may be done at a time. This is ensured by @ubi->alc_mutex.
 */
int ubi_eba_atomic_leb_change(struct ubi_device *ubi, struct ubi_volume *vol,
			      int lnum, const void *buf, int len)
{
	int err, pnum, old_pnum, tries = 0, vol_id = vol->vol_id;
	struct ubi_vid_hdr *vid_hdr;
	uint32_t crc;

	if (ubi->ro_mode)
		return -EROFS;

	if (len == 0) {
		/*
		 * Special case when data length is zero. In this case the LEB
		 * has to be unmapped and mapped somewhere else.
		 */
		err = ubi_eba_unmap_leb(ubi, vol, lnum);
		if (err)
			return err;
		return ubi_eba_write_leb(ubi, vol, lnum, NULL, 0, 0);
	}

	vid_hdr = ubi_zalloc_vid_hdr(ubi, GFP_NOFS);
	if (!vid_hdr)
		return -ENOMEM;

	mutex_lock(&ubi->alc_mutex);
	err = leb_write_lock(ubi, vol_id, lnum);
	if (err)
		goto out_mutex;

	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	vid_hdr->vol_id = cpu_to_be32(vol_id);
	vid_hdr->lnum = cpu_to_be32(lnum);
	vid_hdr->compat = ubi_get_compat(ubi, vol_id);
	vid_hdr->data_pad = cpu_to_be32(vol->data_pad);

	crc = crc32(UBI_CRC32_INIT, buf, len);
	vid_hdr->vol_type = UBI_VID_DYNAMIC;
	vid_hdr->data_size = cpu_to_be32(len);
	vid_hdr->copy_flag = 1;
	vid_hdr->data_crc = cpu_to_be32(crc);

retry:
	pnum = ubi_wl_get_peb(ubi);
	if (pnum < 0) {
		err = pnum;
		up_read(&ubi->fm_eba_sem);
		goto out_leb_unlock;
	}

	dbg_eba("change LEB %d:%d, PEB %d, write VID hdr to PEB %d",
		vol_id, lnum, vol->eba_tbl[lnum], pnum);

	err = ubi_io_write_vid_hdr(ubi, pnum, vid_hdr);
	if (err) {
		ubi_warn(ubi, "failed to write VID header to LEB %d:%d, PEB %d",
			 vol_id, lnum, pnum);
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	err = ubi_io_write_data(ubi, buf, pnum, 0, len);
	if (err) {
		ubi_warn(ubi, "failed to write %d bytes of data to PEB %d",
			 len, pnum);
		up_read(&ubi->fm_eba_sem);
		goto write_error;
	}

	old_pnum = vol->eba_tbl[lnum];
	vol->eba_tbl[lnum] = pnum;
	up_read(&ubi->fm_eba_sem);

	if (old_pnum >= 0) {
		err = ubi_wl_put_peb(ubi, vol_id, lnum, old_pnum, 0);
		if (err)
			goto out_leb_unlock;
	}

out_leb_unlock:
	leb_write_unlock(ubi, vol_id, lnum);
out_mutex:
	mutex_unlock(&ubi->alc_mutex);
	ubi_free_vid_hdr(ubi, vid_hdr);
	return err;

write_error:
	if (err != -EIO || !ubi->bad_allowed) {
		/*
		 * This flash device does not admit of bad eraseblocks or
		 * something nasty and unexpected happened. Switch to read-only
		 * mode just in case.
		 */
		ubi_ro_mode(ubi);
		goto out_leb_unlock;
	}

	err = ubi_wl_put_peb(ubi, vol_id, lnum, pnum, 1);
	if (err || ++tries > UBI_IO_RETRIES) {
		ubi_ro_mode(ubi);
		goto out_leb_unlock;
	}

	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));
	ubi_msg(ubi, "try another PEB");
	goto retry;
}

/**
 * is_error_sane - check whether a read error is sane.
 * @err: code of the error happened during reading
 *
 * This is a helper function for 'ubi_eba_copy_leb()' which is called when we
 * cannot read data from the target PEB (an error @err happened). If the error
 * code is sane, then we treat this error as non-fatal. Otherwise the error is
 * fatal and UBI will be switched to R/O mode later.
 *
 * The idea is that we try not to switch to R/O mode if the read error is
 * something which suggests there was a real read problem. E.g., %-EIO. Or a
 * memory allocation failed (-%ENOMEM). Otherwise, it is safer to switch to R/O
 * mode, simply because we do not know what happened at the MTD level, and we
 * cannot handle this. E.g., the underlying driver may have become crazy, and
 * it is safer to switch to R/O mode to preserve the data.
 *
 * And bear in mind, this is about reading from the target PEB, i.e. the PEB
 * which we have just written.
 */
static int is_error_sane(int err)
{
	if (err == -EIO || err == -ENOMEM || err == UBI_IO_BAD_HDR ||
	    err == UBI_IO_BAD_HDR_EBADMSG || err == -ETIMEDOUT)
		return 0;
	return 1;
}

/**
 * ubi_eba_copy_leb - copy logical eraseblock.
 * @ubi: UBI device description object
 * @from: physical eraseblock number from where to copy
 * @to: physical eraseblock number where to copy
 * @vid_hdr: VID header of the @from physical eraseblock
 *
 * This function copies logical eraseblock from physical eraseblock @from to
 * physical eraseblock @to. The @vid_hdr buffer may be changed by this
 * function. Returns:
 *   o %0 in case of success;
 *   o %MOVE_CANCEL_RACE, %MOVE_TARGET_WR_ERR, %MOVE_TARGET_BITFLIPS, etc;
 *   o a negative error code in case of failure.
 */
int ubi_eba_copy_leb(struct ubi_device *ubi, int from, int to,
		     struct ubi_vid_hdr *vid_hdr)
{
	int err, vol_id, lnum, data_size, aldata_size, idx;
	struct ubi_volume *vol;
	uint32_t crc;

	vol_id = be32_to_cpu(vid_hdr->vol_id);
	lnum = be32_to_cpu(vid_hdr->lnum);

	dbg_wl("copy LEB %d:%d, PEB %d to PEB %d", vol_id, lnum, from, to);

	if (vid_hdr->vol_type == UBI_VID_STATIC) {
		data_size = be32_to_cpu(vid_hdr->data_size);
		aldata_size = ALIGN(data_size, ubi->min_io_size);
	} else
		data_size = aldata_size =
			    ubi->leb_size - be32_to_cpu(vid_hdr->data_pad);

	idx = vol_id2idx(ubi, vol_id);
	spin_lock(&ubi->volumes_lock);
	/*
	 * Note, we may race with volume deletion, which means that the volume
	 * this logical eraseblock belongs to might be being deleted. Since the
	 * volume deletion un-maps all the volume's logical eraseblocks, it will
	 * be locked in 'ubi_wl_put_peb()' and wait for the WL worker to finish.
	 */
	vol = ubi->volumes[idx];
	spin_unlock(&ubi->volumes_lock);
	if (!vol) {
		/* No need to do further work, cancel */
		dbg_wl("volume %d is being removed, cancel", vol_id);
		return MOVE_CANCEL_RACE;
	}

	/*
	 * We do not want anybody to write to this logical eraseblock while we
	 * are moving it, so lock it.
	 *
	 * Note, we are using non-waiting locking here, because we cannot sleep
	 * on the LEB, since it may cause deadlocks. Indeed, imagine a task is
	 * unmapping the LEB which is mapped to the PEB we are going to move
	 * (@from). This task locks the LEB and goes sleep in the
	 * 'ubi_wl_put_peb()' function on the @ubi->move_mutex. In turn, we are
	 * holding @ubi->move_mutex and go sleep on the LEB lock. So, if the
	 * LEB is already locked, we just do not move it and return
	 * %MOVE_RETRY. Note, we do not return %MOVE_CANCEL_RACE here because
	 * we do not know the reasons of the contention - it may be just a
	 * normal I/O on this LEB, so we want to re-try.
	 */
	err = leb_write_trylock(ubi, vol_id, lnum);
	if (err) {
		dbg_wl("contention on LEB %d:%d, cancel", vol_id, lnum);
		return MOVE_RETRY;
	}

	/*
	 * The LEB might have been put meanwhile, and the task which put it is
	 * probably waiting on @ubi->move_mutex. No need to continue the work,
	 * cancel it.
	 */
	if (vol->eba_tbl[lnum] != from) {
		dbg_wl("LEB %d:%d is no longer mapped to PEB %d, mapped to PEB %d, cancel",
		       vol_id, lnum, from, vol->eba_tbl[lnum]);
		err = MOVE_CANCEL_RACE;
		goto out_unlock_leb;
	}

	/*
	 * OK, now the LEB is locked and we can safely start moving it. Since
	 * this function utilizes the @ubi->peb_buf buffer which is shared
	 * with some other functions - we lock the buffer by taking the
	 * @ubi->buf_mutex.
	 */
	mutex_lock(&ubi->buf_mutex);
	dbg_wl("read %d bytes of data", aldata_size);
	err = ubi_io_read_data(ubi, ubi->peb_buf, from, 0, aldata_size);
	if (err && err != UBI_IO_BITFLIPS) {
		ubi_warn(ubi, "error %d while reading data from PEB %d",
			 err, from);
		err = MOVE_SOURCE_RD_ERR;
		goto out_unlock_buf;
	}

	/*
	 * Now we have got to calculate how much data we have to copy. In
	 * case of a static volume it is fairly easy - the VID header contains
	 * the data size. In case of a dynamic volume it is more difficult - we
	 * have to read the contents, cut 0xFF bytes from the end and copy only
	 * the first part. We must do this to avoid writing 0xFF bytes as it
	 * may have some side-effects. And not only this. It is important not
	 * to include those 0xFFs to CRC because later the they may be filled
	 * by data.
	 */
	if (vid_hdr->vol_type == UBI_VID_DYNAMIC)
		aldata_size = data_size =
			ubi_calc_data_len(ubi, ubi->peb_buf, data_size);

	cond_resched();
	crc = crc32(UBI_CRC32_INIT, ubi->peb_buf, data_size);
	cond_resched();

	/*
	 * It may turn out to be that the whole @from physical eraseblock
	 * contains only 0xFF bytes. Then we have to only write the VID header
	 * and do not write any data. This also means we should not set
	 * @vid_hdr->copy_flag, @vid_hdr->data_size, and @vid_hdr->data_crc.
	 */
	if (data_size > 0) {
		vid_hdr->copy_flag = 1;
		vid_hdr->data_size = cpu_to_be32(data_size);
		vid_hdr->data_crc = cpu_to_be32(crc);
	}
	vid_hdr->sqnum = cpu_to_be64(ubi_next_sqnum(ubi));

	err = ubi_io_write_vid_hdr(ubi, to, vid_hdr);
	if (err) {
		if (err == -EIO)
			err = MOVE_TARGET_WR_ERR;
		goto out_unlock_buf;
	}

	cond_resched();

	/* Read the VID header back and check if it was written correctly */
	err = ubi_io_read_vid_hdr(ubi, to, vid_hdr, 1);
	if (err) {
		if (err != UBI_IO_BITFLIPS) {
			ubi_warn(ubi, "error %d while reading VID header back from PEB %d",
				 err, to);
			if (is_error_sane(err))
				err = MOVE_TARGET_RD_ERR;
		} else
			err = MOVE_TARGET_BITFLIPS;
		goto out_unlock_buf;
	}

	if (data_size > 0) {
		err = ubi_io_write_data(ubi, ubi->peb_buf, to, 0, aldata_size);
		if (err) {
			if (err == -EIO)
				err = MOVE_TARGET_WR_ERR;
			goto out_unlock_buf;
		}

		cond_resched();
	}

	ubi_assert(vol->eba_tbl[lnum] == from);
	down_read(&ubi->fm_eba_sem);
	vol->eba_tbl[lnum] = to;
	up_read(&ubi->fm_eba_sem);

out_unlock_buf:
	mutex_unlock(&ubi->buf_mutex);
out_unlock_leb:
	leb_write_unlock(ubi, vol_id, lnum);
	return err;
}

/**
 * print_rsvd_warning - warn about not having enough reserved PEBs.
 * @ubi: UBI device description object
 *
 * This is a helper function for 'ubi_eba_init()' which is called when UBI
 * cannot reserve enough PEBs for bad block handling. This function makes a
 * decision whether we have to print a warning or not. The algorithm is as
 * follows:
 *   o if this is a new UBI image, then just print the warning
 *   o if this is an UBI image which has already been used for some time, print
 *     a warning only if we can reserve less than 10% of the expected amount of
 *     the reserved PEB.
 *
 * The idea is that when UBI is used, PEBs become bad, and the reserved pool
 * of PEBs becomes smaller, which is normal and we do not want to scare users
 * with a warning every time they attach the MTD device. This was an issue
 * reported by real users.
 */
static void print_rsvd_warning(struct ubi_device *ubi,
			       struct ubi_attach_info *ai)
{
	/*
	 * The 1 << 18 (256KiB) number is picked randomly, just a reasonably
	 * large number to distinguish between newly flashed and used images.
	 */
	if (ai->max_sqnum > (1 << 18)) {
		int min = ubi->beb_rsvd_level / 10;

		if (!min)
			min = 1;
		if (ubi->beb_rsvd_pebs > min)
			return;
	}

	ubi_warn(ubi, "cannot reserve enough PEBs for bad PEB handling, reserved %d, need %d",
		 ubi->beb_rsvd_pebs, ubi->beb_rsvd_level);
	if (ubi->corr_peb_count)
		ubi_warn(ubi, "%d PEBs are corrupted and not used",
			 ubi->corr_peb_count);
}

/**
 * self_check_eba - run a self check on the EBA table constructed by fastmap.
 * @ubi: UBI device description object
 * @ai_fastmap: UBI attach info object created by fastmap
 * @ai_scan: UBI attach info object created by scanning
 *
 * Returns < 0 in case of an internal error, 0 otherwise.
 * If a bad EBA table entry was found it will be printed out and
 * ubi_assert() triggers.
 */
int self_check_eba(struct ubi_device *ubi, struct ubi_attach_info *ai_fastmap,
		   struct ubi_attach_info *ai_scan)
{
	int i, j, num_volumes, ret = 0;
	int **scan_eba, **fm_eba;
	struct ubi_ainf_volume *av;
	struct ubi_volume *vol;
	struct ubi_ainf_peb *aeb;
	struct rb_node *rb;

	num_volumes = ubi->vtbl_slots + UBI_INT_VOL_COUNT;

	scan_eba = kmalloc(sizeof(*scan_eba) * num_volumes, GFP_KERNEL);
	if (!scan_eba)
		return -ENOMEM;

	fm_eba = kmalloc(sizeof(*fm_eba) * num_volumes, GFP_KERNEL);
	if (!fm_eba) {
		kfree(scan_eba);
		return -ENOMEM;
	}

	for (i = 0; i < num_volumes; i++) {
		vol = ubi->volumes[i];
		if (!vol)
			continue;

		scan_eba[i] = kmalloc(vol->reserved_pebs * sizeof(**scan_eba),
				      GFP_KERNEL);
		if (!scan_eba[i]) {
			ret = -ENOMEM;
			goto out_free;
		}

		fm_eba[i] = kmalloc(vol->reserved_pebs * sizeof(**fm_eba),
				    GFP_KERNEL);
		if (!fm_eba[i]) {
			ret = -ENOMEM;
			goto out_free;
		}

		for (j = 0; j < vol->reserved_pebs; j++)
			scan_eba[i][j] = fm_eba[i][j] = UBI_LEB_UNMAPPED;

		av = ubi_find_av(ai_scan, idx2vol_id(ubi, i));
		if (!av)
			continue;

		ubi_rb_for_each_entry(rb, aeb, &av->root, u.rb)
			scan_eba[i][aeb->lnum] = aeb->pnum;

		av = ubi_find_av(ai_fastmap, idx2vol_id(ubi, i));
		if (!av)
			continue;

		ubi_rb_for_each_entry(rb, aeb, &av->root, u.rb)
			fm_eba[i][aeb->lnum] = aeb->pnum;

		for (j = 0; j < vol->reserved_pebs; j++) {
			if (scan_eba[i][j] != fm_eba[i][j]) {
				if (scan_eba[i][j] == UBI_LEB_UNMAPPED ||
					fm_eba[i][j] == UBI_LEB_UNMAPPED)
					continue;

				ubi_err(ubi, "LEB:%i:%i is PEB:%i instead of %i!",
					vol->vol_id, j, fm_eba[i][j],
					scan_eba[i][j]);
				ubi_assert(0);
			}
		}
	}

out_free:
	for (i = 0; i < num_volumes; i++) {
		if (!ubi->volumes[i])
			continue;

		kfree(scan_eba[i]);
		kfree(fm_eba[i]);
	}

	kfree(scan_eba);
	kfree(fm_eba);
	return ret;
}

/**
 * ubi_eba_init - initialize the EBA sub-system using attaching information.
 * @ubi: UBI device description object
 * @ai: attaching information
 *
 * This function returns zero in case of success and a negative error code in
 * case of failure.
 */
int ubi_eba_init(struct ubi_device *ubi, struct ubi_attach_info *ai)
{
	int i, j, err, num_volumes;
	struct ubi_ainf_volume *av;
	struct ubi_volume *vol;
	struct ubi_ainf_peb *aeb;
	struct rb_node *rb;

	dbg_eba("initialize EBA sub-system");

	spin_lock_init(&ubi->ltree_lock);
	mutex_init(&ubi->alc_mutex);
	ubi->ltree = RB_ROOT;

	ubi->global_sqnum = ai->max_sqnum + 1;
	num_volumes = ubi->vtbl_slots + UBI_INT_VOL_COUNT;

	for (i = 0; i < num_volumes; i++) {
		vol = ubi->volumes[i];
		if (!vol)
			continue;

		cond_resched();

		vol->eba_tbl = kmalloc(vol->reserved_pebs * sizeof(int),
				       GFP_KERNEL);
		if (!vol->eba_tbl) {
			err = -ENOMEM;
			goto out_free;
		}

		for (j = 0; j < vol->reserved_pebs; j++)
			vol->eba_tbl[j] = UBI_LEB_UNMAPPED;

		av = ubi_find_av(ai, idx2vol_id(ubi, i));
		if (!av)
			continue;

		ubi_rb_for_each_entry(rb, aeb, &av->root, u.rb) {
			if (aeb->lnum >= vol->reserved_pebs)
				/*
				 * This may happen in case of an unclean reboot
				 * during re-size.
				 */
				ubi_move_aeb_to_list(av, aeb, &ai->erase);
			else
				vol->eba_tbl[aeb->lnum] = aeb->pnum;
		}
	}

	if (ubi->avail_pebs < EBA_RESERVED_PEBS) {
		ubi_err(ubi, "no enough physical eraseblocks (%d, need %d)",
			ubi->avail_pebs, EBA_RESERVED_PEBS);
		if (ubi->corr_peb_count)
			ubi_err(ubi, "%d PEBs are corrupted and not used",
				ubi->corr_peb_count);
		err = -ENOSPC;
		goto out_free;
	}
	ubi->avail_pebs -= EBA_RESERVED_PEBS;
	ubi->rsvd_pebs += EBA_RESERVED_PEBS;

	if (ubi->bad_allowed) {
		ubi_calculate_reserved(ubi);

		if (ubi->avail_pebs < ubi->beb_rsvd_level) {
			/* No enough free physical eraseblocks */
			ubi->beb_rsvd_pebs = ubi->avail_pebs;
			print_rsvd_warning(ubi, ai);
		} else
			ubi->beb_rsvd_pebs = ubi->beb_rsvd_level;

		ubi->avail_pebs -= ubi->beb_rsvd_pebs;
		ubi->rsvd_pebs  += ubi->beb_rsvd_pebs;
	}

	dbg_eba("EBA sub-system is initialized");
	return 0;

out_free:
	for (i = 0; i < num_volumes; i++) {
		if (!ubi->volumes[i])
			continue;
		kfree(ubi->volumes[i]->eba_tbl);
		ubi->volumes[i]->eba_tbl = NULL;
	}
	return err;
}