aboutsummaryrefslogtreecommitdiff
path: root/fs/btrfs/disk-io.c
blob: 7eaa7e949604f2b3f0e8c4a750f70974095b1e0f (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
// SPDX-License-Identifier: GPL-2.0+
#include <common.h>
#include <fs_internal.h>
#include <log.h>
#include <uuid.h>
#include <memalign.h>
#include "kernel-shared/btrfs_tree.h"
#include "common/rbtree-utils.h"
#include "disk-io.h"
#include "ctree.h"
#include "btrfs.h"
#include "volumes.h"
#include "extent-io.h"
#include "crypto/hash.h"

/* specified errno for check_tree_block */
#define BTRFS_BAD_BYTENR		(-1)
#define BTRFS_BAD_FSID			(-2)
#define BTRFS_BAD_LEVEL			(-3)
#define BTRFS_BAD_NRITEMS		(-4)

/* Calculate max possible nritems for a leaf/node */
static u32 max_nritems(u8 level, u32 nodesize)
{

	if (level == 0)
		return ((nodesize - sizeof(struct btrfs_header)) /
			sizeof(struct btrfs_item));
	return ((nodesize - sizeof(struct btrfs_header)) /
		sizeof(struct btrfs_key_ptr));
}

static int check_tree_block(struct btrfs_fs_info *fs_info,
			    struct extent_buffer *buf)
{

	struct btrfs_fs_devices *fs_devices = fs_info->fs_devices;
	u32 nodesize = fs_info->nodesize;
	bool fsid_match = false;
	int ret = BTRFS_BAD_FSID;

	if (buf->start != btrfs_header_bytenr(buf))
		return BTRFS_BAD_BYTENR;
	if (btrfs_header_level(buf) >= BTRFS_MAX_LEVEL)
		return BTRFS_BAD_LEVEL;
	if (btrfs_header_nritems(buf) > max_nritems(btrfs_header_level(buf),
						    nodesize))
		return BTRFS_BAD_NRITEMS;

	/* Only leaf can be empty */
	if (btrfs_header_nritems(buf) == 0 &&
	    btrfs_header_level(buf) != 0)
		return BTRFS_BAD_NRITEMS;

	while (fs_devices) {
		/*
		 * Checking the incompat flag is only valid for the current
		 * fs. For seed devices it's forbidden to have their uuid
		 * changed so reading ->fsid in this case is fine
		 */
		if (fs_devices == fs_info->fs_devices &&
		    btrfs_fs_incompat(fs_info, METADATA_UUID))
			fsid_match = !memcmp_extent_buffer(buf,
						   fs_devices->metadata_uuid,
						   btrfs_header_fsid(),
						   BTRFS_FSID_SIZE);
		else
			fsid_match = !memcmp_extent_buffer(buf,
						    fs_devices->fsid,
						    btrfs_header_fsid(),
						    BTRFS_FSID_SIZE);


		if (fsid_match) {
			ret = 0;
			break;
		}
		fs_devices = fs_devices->seed;
	}
	return ret;
}

static void print_tree_block_error(struct btrfs_fs_info *fs_info,
				struct extent_buffer *eb,
				int err)
{
	char fs_uuid[BTRFS_UUID_UNPARSED_SIZE] = {'\0'};
	char found_uuid[BTRFS_UUID_UNPARSED_SIZE] = {'\0'};
	u8 buf[BTRFS_UUID_SIZE];

	if (!err)
		return;

	fprintf(stderr, "bad tree block %llu, ", eb->start);
	switch (err) {
	case BTRFS_BAD_FSID:
		read_extent_buffer(eb, buf, btrfs_header_fsid(),
				   BTRFS_UUID_SIZE);
		uuid_unparse(buf, found_uuid);
		uuid_unparse(fs_info->fs_devices->metadata_uuid, fs_uuid);
		fprintf(stderr, "fsid mismatch, want=%s, have=%s\n",
			fs_uuid, found_uuid);
		break;
	case BTRFS_BAD_BYTENR:
		fprintf(stderr, "bytenr mismatch, want=%llu, have=%llu\n",
			eb->start, btrfs_header_bytenr(eb));
		break;
	case BTRFS_BAD_LEVEL:
		fprintf(stderr, "bad level, %u > %d\n",
			btrfs_header_level(eb), BTRFS_MAX_LEVEL);
		break;
	case BTRFS_BAD_NRITEMS:
		fprintf(stderr, "invalid nr_items: %u\n",
			btrfs_header_nritems(eb));
		break;
	}
}

int btrfs_csum_data(u16 csum_type, const u8 *data, u8 *out, size_t len)
{
	memset(out, 0, BTRFS_CSUM_SIZE);

	switch (csum_type) {
	case BTRFS_CSUM_TYPE_CRC32:
		return hash_crc32c(data, len, out);
	case BTRFS_CSUM_TYPE_XXHASH:
		return hash_xxhash(data, len, out);
	case BTRFS_CSUM_TYPE_SHA256:
		return hash_sha256(data, len, out);
	case BTRFS_CSUM_TYPE_BLAKE2:
		return hash_blake2(data, len, out);
	default:
		printf("Unknown csum type %d\n", csum_type);
		return -EINVAL;
	}
}

/*
 * Check if the super is valid:
 * - nodesize/sectorsize - minimum, maximum, alignment
 * - tree block starts   - alignment
 * - number of devices   - something sane
 * - sys array size      - maximum
 */
static int btrfs_check_super(struct btrfs_super_block *sb)
{
	u8 result[BTRFS_CSUM_SIZE];
	u16 csum_type;
	int csum_size;
	u8 *metadata_uuid;

	if (btrfs_super_magic(sb) != BTRFS_MAGIC)
		return -EIO;

	csum_type = btrfs_super_csum_type(sb);
	if (csum_type >= btrfs_super_num_csums()) {
		error("unsupported checksum algorithm %u", csum_type);
		return -EIO;
	}
	csum_size = btrfs_super_csum_size(sb);

	btrfs_csum_data(csum_type, (u8 *)sb + BTRFS_CSUM_SIZE,
			result, BTRFS_SUPER_INFO_SIZE - BTRFS_CSUM_SIZE);

	if (memcmp(result, sb->csum, csum_size)) {
		error("superblock checksum mismatch");
		return -EIO;
	}
	if (btrfs_super_root_level(sb) >= BTRFS_MAX_LEVEL) {
		error("tree_root level too big: %d >= %d",
			btrfs_super_root_level(sb), BTRFS_MAX_LEVEL);
		goto error_out;
	}
	if (btrfs_super_chunk_root_level(sb) >= BTRFS_MAX_LEVEL) {
		error("chunk_root level too big: %d >= %d",
			btrfs_super_chunk_root_level(sb), BTRFS_MAX_LEVEL);
		goto error_out;
	}
	if (btrfs_super_log_root_level(sb) >= BTRFS_MAX_LEVEL) {
		error("log_root level too big: %d >= %d",
			btrfs_super_log_root_level(sb), BTRFS_MAX_LEVEL);
		goto error_out;
	}

	if (!IS_ALIGNED(btrfs_super_root(sb), 4096)) {
		error("tree_root block unaligned: %llu", btrfs_super_root(sb));
		goto error_out;
	}
	if (!IS_ALIGNED(btrfs_super_chunk_root(sb), 4096)) {
		error("chunk_root block unaligned: %llu",
			btrfs_super_chunk_root(sb));
		goto error_out;
	}
	if (!IS_ALIGNED(btrfs_super_log_root(sb), 4096)) {
		error("log_root block unaligned: %llu",
			btrfs_super_log_root(sb));
		goto error_out;
	}
	if (btrfs_super_nodesize(sb) < 4096) {
		error("nodesize too small: %u < 4096",
			btrfs_super_nodesize(sb));
		goto error_out;
	}
	if (!IS_ALIGNED(btrfs_super_nodesize(sb), 4096)) {
		error("nodesize unaligned: %u", btrfs_super_nodesize(sb));
		goto error_out;
	}
	if (btrfs_super_sectorsize(sb) < 4096) {
		error("sectorsize too small: %u < 4096",
			btrfs_super_sectorsize(sb));
		goto error_out;
	}
	if (!IS_ALIGNED(btrfs_super_sectorsize(sb), 4096)) {
		error("sectorsize unaligned: %u", btrfs_super_sectorsize(sb));
		goto error_out;
	}
	if (btrfs_super_total_bytes(sb) == 0) {
		error("invalid total_bytes 0");
		goto error_out;
	}
	if (btrfs_super_bytes_used(sb) < 6 * btrfs_super_nodesize(sb)) {
		error("invalid bytes_used %llu", btrfs_super_bytes_used(sb));
		goto error_out;
	}
	if ((btrfs_super_stripesize(sb) != 4096)
		&& (btrfs_super_stripesize(sb) != btrfs_super_sectorsize(sb))) {
		error("invalid stripesize %u", btrfs_super_stripesize(sb));
		goto error_out;
	}

	if (btrfs_super_incompat_flags(sb) & BTRFS_FEATURE_INCOMPAT_METADATA_UUID)
		metadata_uuid = sb->metadata_uuid;
	else
		metadata_uuid = sb->fsid;

	if (memcmp(metadata_uuid, sb->dev_item.fsid, BTRFS_FSID_SIZE) != 0) {
		char fsid[BTRFS_UUID_UNPARSED_SIZE];
		char dev_fsid[BTRFS_UUID_UNPARSED_SIZE];

		uuid_unparse(sb->metadata_uuid, fsid);
		uuid_unparse(sb->dev_item.fsid, dev_fsid);
		error("dev_item UUID does not match fsid: %s != %s",
			dev_fsid, fsid);
		goto error_out;
	}

	/*
	 * Hint to catch really bogus numbers, bitflips or so
	 */
	if (btrfs_super_num_devices(sb) > (1UL << 31)) {
		error("suspicious number of devices: %llu",
			btrfs_super_num_devices(sb));
	}

	if (btrfs_super_num_devices(sb) == 0) {
		error("number of devices is 0");
		goto error_out;
	}

	/*
	 * Obvious sys_chunk_array corruptions, it must hold at least one key
	 * and one chunk
	 */
	if (btrfs_super_sys_array_size(sb) > BTRFS_SYSTEM_CHUNK_ARRAY_SIZE) {
		error("system chunk array too big %u > %u",
		      btrfs_super_sys_array_size(sb),
		      BTRFS_SYSTEM_CHUNK_ARRAY_SIZE);
		goto error_out;
	}
	if (btrfs_super_sys_array_size(sb) < sizeof(struct btrfs_disk_key)
			+ sizeof(struct btrfs_chunk)) {
		error("system chunk array too small %u < %zu",
		      btrfs_super_sys_array_size(sb),
		      sizeof(struct btrfs_disk_key) +
		      sizeof(struct btrfs_chunk));
		goto error_out;
	}

	return 0;

error_out:
	error("superblock checksum matches but it has invalid members");
	return -EIO;
}

/*
 * btrfs_read_dev_super - read a valid primary superblock from a block device
 * @desc,@part:	file descriptor of the device
 * @sb:		buffer where the superblock is going to be read in
 *
 * Unlike the btrfs-progs/kernel version, here we ony care about the first
 * super block, thus it's much simpler.
 */
int btrfs_read_dev_super(struct blk_desc *desc, struct disk_partition *part,
			 struct btrfs_super_block *sb)
{
	ALLOC_CACHE_ALIGN_BUFFER(char, tmp, BTRFS_SUPER_INFO_SIZE);
	struct btrfs_super_block *buf = (struct btrfs_super_block *)tmp;
	int ret;

	ret = __btrfs_devread(desc, part, tmp, BTRFS_SUPER_INFO_SIZE,
			      BTRFS_SUPER_INFO_OFFSET);
	if (ret < BTRFS_SUPER_INFO_SIZE)
		return -EIO;

	if (btrfs_super_bytenr(buf) != BTRFS_SUPER_INFO_OFFSET)
		return -EIO;

	if (btrfs_check_super(buf))
		return -EIO;

	memcpy(sb, buf, BTRFS_SUPER_INFO_SIZE);
	return 0;
}

static int __csum_tree_block_size(struct extent_buffer *buf, u16 csum_size,
				  int verify, int silent, u16 csum_type)
{
	u8 result[BTRFS_CSUM_SIZE];
	u32 len;

	len = buf->len - BTRFS_CSUM_SIZE;
	btrfs_csum_data(csum_type, (u8 *)buf->data + BTRFS_CSUM_SIZE,
			result, len);

	if (verify) {
		if (memcmp_extent_buffer(buf, result, 0, csum_size)) {
			/* FIXME: format */
			if (!silent)
				printk("checksum verify failed on %llu found %08X wanted %08X\n",
				       (unsigned long long)buf->start,
				       result[0],
				       buf->data[0]);
			return 1;
		}
	} else {
		write_extent_buffer(buf, result, 0, csum_size);
	}
	return 0;
}

int csum_tree_block_size(struct extent_buffer *buf, u16 csum_size, int verify,
			 u16 csum_type)
{
	return __csum_tree_block_size(buf, csum_size, verify, 0, csum_type);
}

static int csum_tree_block(struct btrfs_fs_info *fs_info,
			   struct extent_buffer *buf, int verify)
{
	u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);
	u16 csum_type = btrfs_super_csum_type(fs_info->super_copy);

	return csum_tree_block_size(buf, csum_size, verify, csum_type);
}

struct extent_buffer *btrfs_find_tree_block(struct btrfs_fs_info *fs_info,
					    u64 bytenr, u32 blocksize)
{
	return find_extent_buffer(&fs_info->extent_cache,
				  bytenr, blocksize);
}

struct extent_buffer* btrfs_find_create_tree_block(
		struct btrfs_fs_info *fs_info, u64 bytenr)
{
	return alloc_extent_buffer(fs_info, bytenr, fs_info->nodesize);
}

static int verify_parent_transid(struct extent_io_tree *io_tree,
				 struct extent_buffer *eb, u64 parent_transid,
				 int ignore)
{
	int ret;

	if (!parent_transid || btrfs_header_generation(eb) == parent_transid)
		return 0;

	if (extent_buffer_uptodate(eb) &&
	    btrfs_header_generation(eb) == parent_transid) {
		ret = 0;
		goto out;
	}
	printk("parent transid verify failed on %llu wanted %llu found %llu\n",
	       (unsigned long long)eb->start,
	       (unsigned long long)parent_transid,
	       (unsigned long long)btrfs_header_generation(eb));
	if (ignore) {
		eb->flags |= EXTENT_BAD_TRANSID;
		printk("Ignoring transid failure\n");
		return 0;
	}

	ret = 1;
out:
	clear_extent_buffer_uptodate(eb);
	return ret;

}

int read_whole_eb(struct btrfs_fs_info *info, struct extent_buffer *eb, int mirror)
{
	unsigned long offset = 0;
	struct btrfs_multi_bio *multi = NULL;
	struct btrfs_device *device;
	int ret = 0;
	u64 read_len;
	unsigned long bytes_left = eb->len;

	while (bytes_left) {
		read_len = bytes_left;
		device = NULL;

		ret = btrfs_map_block(info, READ, eb->start + offset,
				      &read_len, &multi, mirror, NULL);
		if (ret) {
			printk("Couldn't map the block %Lu\n", eb->start + offset);
			kfree(multi);
			return -EIO;
		}
		device = multi->stripes[0].dev;

		if (!device->desc || !device->part) {
			kfree(multi);
			return -EIO;
		}

		if (read_len > bytes_left)
			read_len = bytes_left;

		ret = read_extent_from_disk(device->desc, device->part,
					    multi->stripes[0].physical, eb,
					    offset, read_len);
		kfree(multi);
		multi = NULL;

		if (ret)
			return -EIO;
		offset += read_len;
		bytes_left -= read_len;
	}
	return 0;
}

struct extent_buffer* read_tree_block(struct btrfs_fs_info *fs_info, u64 bytenr,
		u64 parent_transid)
{
	int ret;
	struct extent_buffer *eb;
	u64 best_transid = 0;
	u32 sectorsize = fs_info->sectorsize;
	int mirror_num = 1;
	int good_mirror = 0;
	int candidate_mirror = 0;
	int num_copies;
	int ignore = 0;

	/*
	 * Don't even try to create tree block for unaligned tree block
	 * bytenr.
	 * Such unaligned tree block will free overlapping extent buffer,
	 * causing use-after-free bugs for fuzzed images.
	 */
	if (bytenr < sectorsize || !IS_ALIGNED(bytenr, sectorsize)) {
		error("tree block bytenr %llu is not aligned to sectorsize %u",
		      bytenr, sectorsize);
		return ERR_PTR(-EIO);
	}

	eb = btrfs_find_create_tree_block(fs_info, bytenr);
	if (!eb)
		return ERR_PTR(-ENOMEM);

	if (btrfs_buffer_uptodate(eb, parent_transid))
		return eb;

	num_copies = btrfs_num_copies(fs_info, eb->start, eb->len);
	while (1) {
		ret = read_whole_eb(fs_info, eb, mirror_num);
		if (ret == 0 && csum_tree_block(fs_info, eb, 1) == 0 &&
		    check_tree_block(fs_info, eb) == 0 &&
		    verify_parent_transid(&fs_info->extent_cache, eb,
					  parent_transid, ignore) == 0) {
			/*
			 * check_tree_block() is less strict to allow btrfs
			 * check to get raw eb with bad key order and fix it.
			 * But we still need to try to get a good copy if
			 * possible, or bad key order can go into tools like
			 * btrfs ins dump-tree.
			 */
			if (btrfs_header_level(eb))
				ret = btrfs_check_node(fs_info, NULL, eb);
			else
				ret = btrfs_check_leaf(fs_info, NULL, eb);
			if (!ret || candidate_mirror == mirror_num) {
				btrfs_set_buffer_uptodate(eb);
				return eb;
			}
			if (candidate_mirror <= 0)
				candidate_mirror = mirror_num;
		}
		if (ignore) {
			if (candidate_mirror > 0) {
				mirror_num = candidate_mirror;
				continue;
			}
			if (check_tree_block(fs_info, eb))
				print_tree_block_error(fs_info, eb,
						check_tree_block(fs_info, eb));
			else
				fprintf(stderr, "Csum didn't match\n");
			ret = -EIO;
			break;
		}
		if (num_copies == 1) {
			ignore = 1;
			continue;
		}
		if (btrfs_header_generation(eb) > best_transid) {
			best_transid = btrfs_header_generation(eb);
			good_mirror = mirror_num;
		}
		mirror_num++;
		if (mirror_num > num_copies) {
			if (candidate_mirror > 0)
				mirror_num = candidate_mirror;
			else
				mirror_num = good_mirror;
			ignore = 1;
			continue;
		}
	}
	/*
	 * We failed to read this tree block, it be should deleted right now
	 * to avoid stale cache populate the cache.
	 */
	free_extent_buffer(eb);
	return ERR_PTR(ret);
}

int read_extent_data(struct btrfs_fs_info *fs_info, char *data, u64 logical,
		     u64 *len, int mirror)
{
	u64 orig_len = *len;
	u64 cur = logical;
	struct btrfs_multi_bio *multi = NULL;
	struct btrfs_device *device;
	int ret = 0;

	while (cur < logical + orig_len) {
		u64 cur_len = logical + orig_len - cur;

		ret = btrfs_map_block(fs_info, READ, cur, &cur_len, &multi,
				      mirror, NULL);
		if (ret) {
			error("Couldn't map the block %llu", cur);
			goto err;
		}
		device = multi->stripes[0].dev;
		if (!device->desc || !device->part) {
			error("devid %llu is missing", device->devid);
			ret = -EIO;
			goto err;
		}
		ret = __btrfs_devread(device->desc, device->part,
				data + (cur - logical), cur_len,
				multi->stripes[0].physical);
		if (ret != cur_len) {
			error("read failed on devid %llu physical %llu",
			      device->devid, multi->stripes[0].physical);
			ret = -EIO;
			goto err;
		}
		cur += cur_len;
		ret = 0;
	}
err:
	kfree(multi);
	return ret;
}

void btrfs_setup_root(struct btrfs_root *root, struct btrfs_fs_info *fs_info,
		      u64 objectid)
{
	root->node = NULL;
	root->track_dirty = 0;

	root->fs_info = fs_info;
	root->objectid = objectid;
	root->last_trans = 0;
	root->last_inode_alloc = 0;

	memset(&root->root_key, 0, sizeof(root->root_key));
	memset(&root->root_item, 0, sizeof(root->root_item));
	root->root_key.objectid = objectid;
}

static int find_and_setup_root(struct btrfs_root *tree_root,
			       struct btrfs_fs_info *fs_info,
			       u64 objectid, struct btrfs_root *root)
{
	int ret;
	u64 generation;

	btrfs_setup_root(root, fs_info, objectid);
	ret = btrfs_find_last_root(tree_root, objectid,
				   &root->root_item, &root->root_key);
	if (ret)
		return ret;

	generation = btrfs_root_generation(&root->root_item);
	root->node = read_tree_block(fs_info,
			btrfs_root_bytenr(&root->root_item), generation);
	if (!extent_buffer_uptodate(root->node))
		return -EIO;

	return 0;
}

int btrfs_free_fs_root(struct btrfs_root *root)
{
	if (root->node)
		free_extent_buffer(root->node);
	kfree(root);
	return 0;
}

static void __free_fs_root(struct rb_node *node)
{
	struct btrfs_root *root;

	root = container_of(node, struct btrfs_root, rb_node);
	btrfs_free_fs_root(root);
}

FREE_RB_BASED_TREE(fs_roots, __free_fs_root);

struct btrfs_root *btrfs_read_fs_root_no_cache(struct btrfs_fs_info *fs_info,
					       struct btrfs_key *location)
{
	struct btrfs_root *root;
	struct btrfs_root *tree_root = fs_info->tree_root;
	struct btrfs_path *path;
	struct extent_buffer *l;
	u64 generation;
	int ret = 0;

	root = calloc(1, sizeof(*root));
	if (!root)
		return ERR_PTR(-ENOMEM);
	if (location->offset == (u64)-1) {
		ret = find_and_setup_root(tree_root, fs_info,
					  location->objectid, root);
		if (ret) {
			free(root);
			return ERR_PTR(ret);
		}
		goto insert;
	}

	btrfs_setup_root(root, fs_info,
			 location->objectid);

	path = btrfs_alloc_path();
	if (!path) {
		free(root);
		return ERR_PTR(-ENOMEM);
	}

	ret = btrfs_search_slot(NULL, tree_root, location, path, 0, 0);
	if (ret != 0) {
		if (ret > 0)
			ret = -ENOENT;
		goto out;
	}
	l = path->nodes[0];
	read_extent_buffer(l, &root->root_item,
	       btrfs_item_ptr_offset(l, path->slots[0]),
	       sizeof(root->root_item));
	memcpy(&root->root_key, location, sizeof(*location));

	/* If this root is already an orphan, no need to read */
	if (btrfs_root_refs(&root->root_item) == 0) {
		ret = -ENOENT;
		goto out;
	}
	ret = 0;
out:
	btrfs_free_path(path);
	if (ret) {
		free(root);
		return ERR_PTR(ret);
	}
	generation = btrfs_root_generation(&root->root_item);
	root->node = read_tree_block(fs_info,
			btrfs_root_bytenr(&root->root_item), generation);
	if (!extent_buffer_uptodate(root->node)) {
		free(root);
		return ERR_PTR(-EIO);
	}
insert:
	root->ref_cows = 1;
	return root;
}

static int btrfs_fs_roots_compare_objectids(struct rb_node *node,
					    void *data)
{
	u64 objectid = *((u64 *)data);
	struct btrfs_root *root;

	root = rb_entry(node, struct btrfs_root, rb_node);
	if (objectid > root->objectid)
		return 1;
	else if (objectid < root->objectid)
		return -1;
	else
		return 0;
}

int btrfs_fs_roots_compare_roots(struct rb_node *node1, struct rb_node *node2)
{
	struct btrfs_root *root;

	root = rb_entry(node2, struct btrfs_root, rb_node);
	return btrfs_fs_roots_compare_objectids(node1, (void *)&root->objectid);
}

struct btrfs_root *btrfs_read_fs_root(struct btrfs_fs_info *fs_info,
				      struct btrfs_key *location)
{
	struct btrfs_root *root;
	struct rb_node *node;
	int ret;
	u64 objectid = location->objectid;

	if (location->objectid == BTRFS_ROOT_TREE_OBJECTID)
		return fs_info->tree_root;
	if (location->objectid == BTRFS_CHUNK_TREE_OBJECTID)
		return fs_info->chunk_root;
	if (location->objectid == BTRFS_CSUM_TREE_OBJECTID)
		return fs_info->csum_root;
	BUG_ON(location->objectid == BTRFS_TREE_RELOC_OBJECTID);

	node = rb_search(&fs_info->fs_root_tree, (void *)&objectid,
			 btrfs_fs_roots_compare_objectids, NULL);
	if (node)
		return container_of(node, struct btrfs_root, rb_node);

	root = btrfs_read_fs_root_no_cache(fs_info, location);
	if (IS_ERR(root))
		return root;

	ret = rb_insert(&fs_info->fs_root_tree, &root->rb_node,
			btrfs_fs_roots_compare_roots);
	BUG_ON(ret);
	return root;
}

void btrfs_free_fs_info(struct btrfs_fs_info *fs_info)
{
	free(fs_info->tree_root);
	free(fs_info->chunk_root);
	free(fs_info->csum_root);
	free(fs_info->super_copy);
	free(fs_info);
}

struct btrfs_fs_info *btrfs_new_fs_info(void)
{
	struct btrfs_fs_info *fs_info;

	fs_info = calloc(1, sizeof(struct btrfs_fs_info));
	if (!fs_info)
		return NULL;

	fs_info->tree_root = calloc(1, sizeof(struct btrfs_root));
	fs_info->chunk_root = calloc(1, sizeof(struct btrfs_root));
	fs_info->csum_root = calloc(1, sizeof(struct btrfs_root));
	fs_info->super_copy = calloc(1, BTRFS_SUPER_INFO_SIZE);

	if (!fs_info->tree_root || !fs_info->chunk_root ||
	    !fs_info->csum_root || !fs_info->super_copy)
		goto free_all;

	extent_io_tree_init(&fs_info->extent_cache);

	fs_info->fs_root_tree = RB_ROOT;
	cache_tree_init(&fs_info->mapping_tree.cache_tree);

	return fs_info;
free_all:
	btrfs_free_fs_info(fs_info);
	return NULL;
}

static int setup_root_or_create_block(struct btrfs_fs_info *fs_info,
				      struct btrfs_root *info_root,
				      u64 objectid, char *str)
{
	struct btrfs_root *root = fs_info->tree_root;
	int ret;

	ret = find_and_setup_root(root, fs_info, objectid, info_root);
	if (ret) {
		error("could not setup %s tree", str);
		return -EIO;
	}

	return 0;
}

static int get_default_subvolume(struct btrfs_fs_info *fs_info,
				 struct btrfs_key *key_ret)
{
	struct btrfs_root *root = fs_info->tree_root;
	struct btrfs_dir_item *dir_item;
	struct btrfs_path path;
	int ret = 0;

	btrfs_init_path(&path);

	dir_item = btrfs_lookup_dir_item(NULL, root, &path,
					 BTRFS_ROOT_TREE_DIR_OBJECTID,
					 "default", 7, 0);
	if (IS_ERR(dir_item)) {
		ret = PTR_ERR(dir_item);
		goto out;
	}

	btrfs_dir_item_key_to_cpu(path.nodes[0], dir_item, key_ret);
out:
	btrfs_release_path(&path);
	return ret;
}

int btrfs_setup_all_roots(struct btrfs_fs_info *fs_info)
{
	struct btrfs_super_block *sb = fs_info->super_copy;
	struct btrfs_root *root;
	struct btrfs_key key;
	u64 root_tree_bytenr;
	u64 generation;
	int ret;

	root = fs_info->tree_root;
	btrfs_setup_root(root, fs_info, BTRFS_ROOT_TREE_OBJECTID);
	generation = btrfs_super_generation(sb);

	root_tree_bytenr = btrfs_super_root(sb);

	root->node = read_tree_block(fs_info, root_tree_bytenr, generation);
	if (!extent_buffer_uptodate(root->node)) {
		fprintf(stderr, "Couldn't read tree root\n");
		return -EIO;
	}

	ret = setup_root_or_create_block(fs_info, fs_info->csum_root,
					 BTRFS_CSUM_TREE_OBJECTID, "csum");
	if (ret)
		return ret;
	fs_info->csum_root->track_dirty = 1;

	fs_info->last_trans_committed = generation;

	ret = get_default_subvolume(fs_info, &key);
	if (ret) {
		/*
		 * The default dir item isn't there. Linux kernel behaviour is
		 * to silently use the top-level subvolume in this case.
		 */
		key.objectid = BTRFS_FS_TREE_OBJECTID;
		key.type = BTRFS_ROOT_ITEM_KEY;
		key.offset = (u64)-1;
	}

	fs_info->fs_root = btrfs_read_fs_root(fs_info, &key);

	if (IS_ERR(fs_info->fs_root))
		return -EIO;
	return 0;
}

void btrfs_release_all_roots(struct btrfs_fs_info *fs_info)
{
	if (fs_info->csum_root)
		free_extent_buffer(fs_info->csum_root->node);
	if (fs_info->tree_root)
		free_extent_buffer(fs_info->tree_root->node);
	if (fs_info->chunk_root)
		free_extent_buffer(fs_info->chunk_root->node);
}

static void free_map_lookup(struct cache_extent *ce)
{
	struct map_lookup *map;

	map = container_of(ce, struct map_lookup, ce);
	kfree(map);
}

FREE_EXTENT_CACHE_BASED_TREE(mapping_cache, free_map_lookup);

void btrfs_cleanup_all_caches(struct btrfs_fs_info *fs_info)
{
	free_mapping_cache_tree(&fs_info->mapping_tree.cache_tree);
	extent_io_tree_cleanup(&fs_info->extent_cache);
}

static int btrfs_scan_fs_devices(struct blk_desc *desc,
				 struct disk_partition *part,
				 struct btrfs_fs_devices **fs_devices)
{
	u64 total_devs;
	int ret;

	if (round_up(BTRFS_SUPER_INFO_SIZE + BTRFS_SUPER_INFO_OFFSET,
		     desc->blksz) > (part->size << desc->log2blksz)) {
		log_debug("superblock end %u is larger than device size " LBAFU,
			  BTRFS_SUPER_INFO_SIZE + BTRFS_SUPER_INFO_OFFSET,
			  part->size << desc->log2blksz);
		return -EINVAL;
	}

	ret = btrfs_scan_one_device(desc, part, fs_devices, &total_devs);
	if (ret) {
		/*
		 * Avoid showing this when probing for a possible Btrfs
		 *
		 * fprintf(stderr, "No valid Btrfs found\n");
		 */
		return ret;
	}
	return 0;
}

int btrfs_check_fs_compatibility(struct btrfs_super_block *sb)
{
	u64 features;

	features = btrfs_super_incompat_flags(sb) &
		   ~BTRFS_FEATURE_INCOMPAT_SUPP;
	if (features) {
		printk("couldn't open because of unsupported "
		       "option features (%llx).\n",
		       (unsigned long long)features);
		return -ENOTSUPP;
	}

	features = btrfs_super_incompat_flags(sb);
	if (!(features & BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF)) {
		features |= BTRFS_FEATURE_INCOMPAT_MIXED_BACKREF;
		btrfs_set_super_incompat_flags(sb, features);
	}

	return 0;
}

static int btrfs_setup_chunk_tree_and_device_map(struct btrfs_fs_info *fs_info)
{
	struct btrfs_super_block *sb = fs_info->super_copy;
	u64 chunk_root_bytenr;
	u64 generation;
	int ret;

	btrfs_setup_root(fs_info->chunk_root, fs_info,
			BTRFS_CHUNK_TREE_OBJECTID);

	ret = btrfs_read_sys_array(fs_info);
	if (ret)
		return ret;

	generation = btrfs_super_chunk_root_generation(sb);
	chunk_root_bytenr = btrfs_super_chunk_root(sb);

	fs_info->chunk_root->node = read_tree_block(fs_info,
						    chunk_root_bytenr,
						    generation);
	if (!extent_buffer_uptodate(fs_info->chunk_root->node)) {
		error("cannot read chunk root");
		return -EIO;
	}

	ret = btrfs_read_chunk_tree(fs_info);
	if (ret) {
		fprintf(stderr, "Couldn't read chunk tree\n");
		return ret;
	}
	return 0;
}

struct btrfs_fs_info *open_ctree_fs_info(struct blk_desc *desc,
					 struct disk_partition *part)
{
	struct btrfs_fs_info *fs_info;
	struct btrfs_super_block *disk_super;
	struct btrfs_fs_devices *fs_devices = NULL;
	struct extent_buffer *eb;
	int ret;

	fs_info = btrfs_new_fs_info();
	if (!fs_info) {
		fprintf(stderr, "Failed to allocate memory for fs_info\n");
		return NULL;
	}

	ret = btrfs_scan_fs_devices(desc, part, &fs_devices);
	if (ret)
		goto out;

	fs_info->fs_devices = fs_devices;

	ret = btrfs_open_devices(fs_devices);
	if (ret)
		goto out;

	disk_super = fs_info->super_copy;
	ret = btrfs_read_dev_super(desc, part, disk_super);
	if (ret) {
		debug("No valid btrfs found\n");
		goto out_devices;
	}

	if (btrfs_super_flags(disk_super) & BTRFS_SUPER_FLAG_CHANGING_FSID) {
		fprintf(stderr, "ERROR: Filesystem UUID change in progress\n");
		goto out_devices;
	}

	ASSERT(!memcmp(disk_super->fsid, fs_devices->fsid, BTRFS_FSID_SIZE));
	if (btrfs_fs_incompat(fs_info, METADATA_UUID))
		ASSERT(!memcmp(disk_super->metadata_uuid,
			       fs_devices->metadata_uuid, BTRFS_FSID_SIZE));

	fs_info->sectorsize = btrfs_super_sectorsize(disk_super);
	fs_info->nodesize = btrfs_super_nodesize(disk_super);
	fs_info->stripesize = btrfs_super_stripesize(disk_super);

	ret = btrfs_check_fs_compatibility(fs_info->super_copy);
	if (ret)
		goto out_devices;

	ret = btrfs_setup_chunk_tree_and_device_map(fs_info);
	if (ret)
		goto out_chunk;

	/* Chunk tree root is unable to read, return directly */
	if (!fs_info->chunk_root)
		return fs_info;

	eb = fs_info->chunk_root->node;
	read_extent_buffer(eb, fs_info->chunk_tree_uuid,
			   btrfs_header_chunk_tree_uuid(eb),
			   BTRFS_UUID_SIZE);

	ret = btrfs_setup_all_roots(fs_info);
	if (ret)
		goto out_chunk;

	return fs_info;

out_chunk:
	btrfs_release_all_roots(fs_info);
	btrfs_cleanup_all_caches(fs_info);
out_devices:
	btrfs_close_devices(fs_devices);
out:
	btrfs_free_fs_info(fs_info);
	return NULL;
}

int close_ctree_fs_info(struct btrfs_fs_info *fs_info)
{
	int ret;

	free_fs_roots_tree(&fs_info->fs_root_tree);

	btrfs_release_all_roots(fs_info);
	ret = btrfs_close_devices(fs_info->fs_devices);
	btrfs_cleanup_all_caches(fs_info);
	btrfs_free_fs_info(fs_info);
	return ret;
}

int btrfs_buffer_uptodate(struct extent_buffer *buf, u64 parent_transid)
{
	int ret;

	ret = extent_buffer_uptodate(buf);
	if (!ret)
		return ret;

	ret = verify_parent_transid(&buf->fs_info->extent_cache, buf,
				    parent_transid, 1);
	return !ret;
}

int btrfs_set_buffer_uptodate(struct extent_buffer *eb)
{
	return set_extent_buffer_uptodate(eb);
}