Home Home > GIT Browse
summaryrefslogtreecommitdiff
blob: 280384bf34f13b20b7a0817a36fe49a53d618285 (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
/*
 * Copyright (C) 2008 Oracle.  All rights reserved.
 *
 * This program is free software; you can redistribute it and/or
 * modify it under the terms of the GNU General Public
 * License v2 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., 59 Temple Place - Suite 330,
 * Boston, MA 021110-1307, USA.
 */

#include <linux/kernel.h>
#include <linux/bio.h>
#include <linux/buffer_head.h>
#include <linux/file.h>
#include <linux/fs.h>
#include <linux/pagemap.h>
#include <linux/highmem.h>
#include <linux/time.h>
#include <linux/init.h>
#include <linux/string.h>
#include <linux/backing-dev.h>
#include <linux/mpage.h>
#include <linux/swap.h>
#include <linux/writeback.h>
#include <linux/bit_spinlock.h>
#include <linux/slab.h>
#include <linux/sched/mm.h>
#include "ctree.h"
#include "disk-io.h"
#include "transaction.h"
#include "btrfs_inode.h"
#include "volumes.h"
#include "ordered-data.h"
#include "compression.h"
#include "extent_io.h"
#include "extent_map.h"

static int btrfs_decompress_bio(struct compressed_bio *cb);

static inline int compressed_bio_size(struct btrfs_fs_info *fs_info,
				      unsigned long disk_size)
{
	u16 csum_size = btrfs_super_csum_size(fs_info->super_copy);

	return sizeof(struct compressed_bio) +
		(DIV_ROUND_UP(disk_size, fs_info->sectorsize)) * csum_size;
}

static int check_compressed_csum(struct btrfs_inode *inode,
				 struct compressed_bio *cb,
				 u64 disk_start)
{
	int ret;
	struct page *page;
	unsigned long i;
	char *kaddr;
	u32 csum;
	u32 *cb_sum = &cb->sums;

	if (inode->flags & BTRFS_INODE_NODATASUM)
		return 0;

	for (i = 0; i < cb->nr_pages; i++) {
		page = cb->compressed_pages[i];
		csum = ~(u32)0;

		kaddr = kmap_atomic(page);
		csum = btrfs_csum_data(kaddr, csum, PAGE_SIZE);
		btrfs_csum_final(csum, (u8 *)&csum);
		kunmap_atomic(kaddr);

		if (csum != *cb_sum) {
			btrfs_print_data_csum_error(inode, disk_start, csum,
					*cb_sum, cb->mirror_num);
			ret = -EIO;
			goto fail;
		}
		cb_sum++;

	}
	ret = 0;
fail:
	return ret;
}

/* when we finish reading compressed pages from the disk, we
 * decompress them and then run the bio end_io routines on the
 * decompressed pages (in the inode address space).
 *
 * This allows the checksumming and other IO error handling routines
 * to work normally
 *
 * The compressed pages are freed here, and it must be run
 * in process context
 */
static void end_compressed_bio_read(struct bio *bio)
{
	struct compressed_bio *cb = bio->bi_private;
	struct inode *inode;
	struct page *page;
	unsigned long index;
	unsigned int mirror = btrfs_io_bio(bio)->mirror_num;
	int ret = 0;

	if (bio->bi_status)
		cb->errors = 1;

	/* if there are more bios still pending for this compressed
	 * extent, just exit
	 */
	if (!refcount_dec_and_test(&cb->pending_bios))
		goto out;

	/*
	 * Record the correct mirror_num in cb->orig_bio so that
	 * read-repair can work properly.
	 */
	ASSERT(btrfs_io_bio(cb->orig_bio));
	btrfs_io_bio(cb->orig_bio)->mirror_num = mirror;
	cb->mirror_num = mirror;

	/*
	 * Some IO in this cb have failed, just skip checksum as there
	 * is no way it could be correct.
	 */
	if (cb->errors == 1)
		goto csum_failed;

	inode = cb->inode;
	ret = check_compressed_csum(BTRFS_I(inode), cb,
				    (u64)bio->bi_iter.bi_sector << 9);
	if (ret)
		goto csum_failed;

	/* ok, we're the last bio for this extent, lets start
	 * the decompression.
	 */
	ret = btrfs_decompress_bio(cb);

csum_failed:
	if (ret)
		cb->errors = 1;

	/* release the compressed pages */
	index = 0;
	for (index = 0; index < cb->nr_pages; index++) {
		page = cb->compressed_pages[index];
		page->mapping = NULL;
		put_page(page);
	}

	/* do io completion on the original bio */
	if (cb->errors) {
		bio_io_error(cb->orig_bio);
	} else {
		int i;
		struct bio_vec *bvec;

		/*
		 * we have verified the checksum already, set page
		 * checked so the end_io handlers know about it
		 */
		ASSERT(!bio_flagged(bio, BIO_CLONED));
		bio_for_each_segment_all(bvec, cb->orig_bio, i)
			SetPageChecked(bvec->bv_page);

		bio_endio(cb->orig_bio);
	}

	/* finally free the cb struct */
	kfree(cb->compressed_pages);
	kfree(cb);
out:
	bio_put(bio);
}

/*
 * Clear the writeback bits on all of the file
 * pages for a compressed write
 */
static noinline void end_compressed_writeback(struct inode *inode,
					      const struct compressed_bio *cb)
{
	unsigned long index = cb->start >> PAGE_SHIFT;
	unsigned long end_index = (cb->start + cb->len - 1) >> PAGE_SHIFT;
	struct page *pages[16];
	unsigned long nr_pages = end_index - index + 1;
	int i;
	int ret;

	if (cb->errors)
		mapping_set_error(inode->i_mapping, -EIO);

	while (nr_pages > 0) {
		ret = find_get_pages_contig(inode->i_mapping, index,
				     min_t(unsigned long,
				     nr_pages, ARRAY_SIZE(pages)), pages);
		if (ret == 0) {
			nr_pages -= 1;
			index += 1;
			continue;
		}
		for (i = 0; i < ret; i++) {
			if (cb->errors)
				SetPageError(pages[i]);
			end_page_writeback(pages[i]);
			put_page(pages[i]);
		}
		nr_pages -= ret;
		index += ret;
	}
	/* the inode may be gone now */
}

/*
 * do the cleanup once all the compressed pages hit the disk.
 * This will clear writeback on the file pages and free the compressed
 * pages.
 *
 * This also calls the writeback end hooks for the file pages so that
 * metadata and checksums can be updated in the file.
 */
static void end_compressed_bio_write(struct bio *bio)
{
	struct extent_io_tree *tree;
	struct compressed_bio *cb = bio->bi_private;
	struct inode *inode;
	struct page *page;
	unsigned long index;

	if (bio->bi_status)
		cb->errors = 1;

	/* if there are more bios still pending for this compressed
	 * extent, just exit
	 */
	if (!refcount_dec_and_test(&cb->pending_bios))
		goto out;

	/* ok, we're the last bio for this extent, step one is to
	 * call back into the FS and do all the end_io operations
	 */
	inode = cb->inode;
	tree = &BTRFS_I(inode)->io_tree;
	cb->compressed_pages[0]->mapping = cb->inode->i_mapping;
	tree->ops->writepage_end_io_hook(cb->compressed_pages[0],
					 cb->start,
					 cb->start + cb->len - 1,
					 NULL,
					 bio->bi_status ? 0 : 1);
	cb->compressed_pages[0]->mapping = NULL;

	end_compressed_writeback(inode, cb);
	/* note, our inode could be gone now */

	/*
	 * release the compressed pages, these came from alloc_page and
	 * are not attached to the inode at all
	 */
	index = 0;
	for (index = 0; index < cb->nr_pages; index++) {
		page = cb->compressed_pages[index];
		page->mapping = NULL;
		put_page(page);
	}

	/* finally free the cb struct */
	kfree(cb->compressed_pages);
	kfree(cb);
out:
	bio_put(bio);
}

/*
 * worker function to build and submit bios for previously compressed pages.
 * The corresponding pages in the inode should be marked for writeback
 * and the compressed pages should have a reference on them for dropping
 * when the IO is complete.
 *
 * This also checksums the file bytes and gets things ready for
 * the end io hooks.
 */
blk_status_t btrfs_submit_compressed_write(struct inode *inode, u64 start,
				 unsigned long len, u64 disk_start,
				 unsigned long compressed_len,
				 struct page **compressed_pages,
				 unsigned long nr_pages)
{
	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
	struct bio *bio = NULL;
	struct compressed_bio *cb;
	unsigned long bytes_left;
	struct extent_io_tree *io_tree = &BTRFS_I(inode)->io_tree;
	int pg_index = 0;
	struct page *page;
	u64 first_byte = disk_start;
	struct block_device *bdev;
	blk_status_t ret;
	int skip_sum = BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM;

	WARN_ON(start & ((u64)PAGE_SIZE - 1));
	cb = kmalloc(compressed_bio_size(fs_info, compressed_len), GFP_NOFS);
	if (!cb)
		return BLK_STS_RESOURCE;
	refcount_set(&cb->pending_bios, 0);
	cb->errors = 0;
	cb->inode = inode;
	cb->start = start;
	cb->len = len;
	cb->mirror_num = 0;
	cb->compressed_pages = compressed_pages;
	cb->compressed_len = compressed_len;
	cb->orig_bio = NULL;
	cb->nr_pages = nr_pages;

	bdev = fs_info->fs_devices->latest_bdev;

	bio = btrfs_bio_alloc(bdev, first_byte);
	bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
	bio->bi_private = cb;
	bio->bi_end_io = end_compressed_bio_write;
	refcount_set(&cb->pending_bios, 1);

	/* create and submit bios for the compressed pages */
	bytes_left = compressed_len;
	for (pg_index = 0; pg_index < cb->nr_pages; pg_index++) {
		int submit = 0;

		page = compressed_pages[pg_index];
		page->mapping = inode->i_mapping;
		if (bio->bi_iter.bi_size)
			submit = io_tree->ops->merge_bio_hook(page, 0,
							   PAGE_SIZE,
							   bio, 0);

		page->mapping = NULL;
		if (submit || bio_add_page(bio, page, PAGE_SIZE, 0) <
		    PAGE_SIZE) {
			bio_get(bio);

			/*
			 * inc the count before we submit the bio so
			 * we know the end IO handler won't happen before
			 * we inc the count.  Otherwise, the cb might get
			 * freed before we're done setting it up
			 */
			refcount_inc(&cb->pending_bios);
			ret = btrfs_bio_wq_end_io(fs_info, bio,
						  BTRFS_WQ_ENDIO_DATA);
			BUG_ON(ret); /* -ENOMEM */

			if (!skip_sum) {
				ret = btrfs_csum_one_bio(inode, bio, start, 1);
				BUG_ON(ret); /* -ENOMEM */
			}

			ret = btrfs_map_bio(fs_info, bio, 0, 1);
			if (ret) {
				bio->bi_status = ret;
				bio_endio(bio);
			}

			bio_put(bio);

			bio = btrfs_bio_alloc(bdev, first_byte);
			bio_set_op_attrs(bio, REQ_OP_WRITE, 0);
			bio->bi_private = cb;
			bio->bi_end_io = end_compressed_bio_write;
			bio_add_page(bio, page, PAGE_SIZE, 0);
		}
		if (bytes_left < PAGE_SIZE) {
			btrfs_info(fs_info,
					"bytes left %lu compress len %lu nr %lu",
			       bytes_left, cb->compressed_len, cb->nr_pages);
		}
		bytes_left -= PAGE_SIZE;
		first_byte += PAGE_SIZE;
		cond_resched();
	}
	bio_get(bio);

	ret = btrfs_bio_wq_end_io(fs_info, bio, BTRFS_WQ_ENDIO_DATA);
	BUG_ON(ret); /* -ENOMEM */

	if (!skip_sum) {
		ret = btrfs_csum_one_bio(inode, bio, start, 1);
		BUG_ON(ret); /* -ENOMEM */
	}

	ret = btrfs_map_bio(fs_info, bio, 0, 1);
	if (ret) {
		bio->bi_status = ret;
		bio_endio(bio);
	}

	bio_put(bio);
	return 0;
}

static u64 bio_end_offset(struct bio *bio)
{
	struct bio_vec *last = &bio->bi_io_vec[bio->bi_vcnt - 1];

	return page_offset(last->bv_page) + last->bv_len + last->bv_offset;
}

static noinline int add_ra_bio_pages(struct inode *inode,
				     u64 compressed_end,
				     struct compressed_bio *cb)
{
	unsigned long end_index;
	unsigned long pg_index;
	u64 last_offset;
	u64 isize = i_size_read(inode);
	int ret;
	struct page *page;
	unsigned long nr_pages = 0;
	struct extent_map *em;
	struct address_space *mapping = inode->i_mapping;
	struct extent_map_tree *em_tree;
	struct extent_io_tree *tree;
	u64 end;
	int misses = 0;

	last_offset = bio_end_offset(cb->orig_bio);
	em_tree = &BTRFS_I(inode)->extent_tree;
	tree = &BTRFS_I(inode)->io_tree;

	if (isize == 0)
		return 0;

	end_index = (i_size_read(inode) - 1) >> PAGE_SHIFT;

	while (last_offset < compressed_end) {
		pg_index = last_offset >> PAGE_SHIFT;

		if (pg_index > end_index)
			break;

		rcu_read_lock();
		page = radix_tree_lookup(&mapping->page_tree, pg_index);
		rcu_read_unlock();
		if (page && !radix_tree_exceptional_entry(page)) {
			misses++;
			if (misses > 4)
				break;
			goto next;
		}

		page = __page_cache_alloc(mapping_gfp_constraint(mapping,
								 ~__GFP_FS));
		if (!page)
			break;

		if (add_to_page_cache_lru(page, mapping, pg_index, GFP_NOFS)) {
			put_page(page);
			goto next;
		}

		end = last_offset + PAGE_SIZE - 1;
		/*
		 * at this point, we have a locked page in the page cache
		 * for these bytes in the file.  But, we have to make
		 * sure they map to this compressed extent on disk.
		 */
		set_page_extent_mapped(page);
		lock_extent(tree, last_offset, end);
		read_lock(&em_tree->lock);
		em = lookup_extent_mapping(em_tree, last_offset,
					   PAGE_SIZE);
		read_unlock(&em_tree->lock);

		if (!em || last_offset < em->start ||
		    (last_offset + PAGE_SIZE > extent_map_end(em)) ||
		    (em->block_start >> 9) != cb->orig_bio->bi_iter.bi_sector) {
			free_extent_map(em);
			unlock_extent(tree, last_offset, end);
			unlock_page(page);
			put_page(page);
			break;
		}
		free_extent_map(em);

		if (page->index == end_index) {
			char *userpage;
			size_t zero_offset = isize & (PAGE_SIZE - 1);

			if (zero_offset) {
				int zeros;
				zeros = PAGE_SIZE - zero_offset;
				userpage = kmap_atomic(page);
				memset(userpage + zero_offset, 0, zeros);
				flush_dcache_page(page);
				kunmap_atomic(userpage);
			}
		}

		ret = bio_add_page(cb->orig_bio, page,
				   PAGE_SIZE, 0);

		if (ret == PAGE_SIZE) {
			nr_pages++;
			put_page(page);
		} else {
			unlock_extent(tree, last_offset, end);
			unlock_page(page);
			put_page(page);
			break;
		}
next:
		last_offset += PAGE_SIZE;
	}
	return 0;
}

/*
 * for a compressed read, the bio we get passed has all the inode pages
 * in it.  We don't actually do IO on those pages but allocate new ones
 * to hold the compressed pages on disk.
 *
 * bio->bi_iter.bi_sector points to the compressed extent on disk
 * bio->bi_io_vec points to all of the inode pages
 *
 * After the compressed pages are read, we copy the bytes into the
 * bio we were passed and then call the bio end_io calls
 */
blk_status_t btrfs_submit_compressed_read(struct inode *inode, struct bio *bio,
				 int mirror_num, unsigned long bio_flags)
{
	struct btrfs_fs_info *fs_info = btrfs_sb(inode->i_sb);
	struct extent_io_tree *tree;
	struct extent_map_tree *em_tree;
	struct compressed_bio *cb;
	unsigned long compressed_len;
	unsigned long nr_pages;
	unsigned long pg_index;
	struct page *page;
	struct block_device *bdev;
	struct bio *comp_bio;
	u64 cur_disk_byte = (u64)bio->bi_iter.bi_sector << 9;
	u64 em_len;
	u64 em_start;
	struct extent_map *em;
	blk_status_t ret = BLK_STS_RESOURCE;
	int faili = 0;
	u32 *sums;

	tree = &BTRFS_I(inode)->io_tree;
	em_tree = &BTRFS_I(inode)->extent_tree;

	/* we need the actual starting offset of this extent in the file */
	read_lock(&em_tree->lock);
	em = lookup_extent_mapping(em_tree,
				   page_offset(bio->bi_io_vec->bv_page),
				   PAGE_SIZE);
	read_unlock(&em_tree->lock);
	if (!em)
		return BLK_STS_IOERR;

	compressed_len = em->block_len;
	cb = kmalloc(compressed_bio_size(fs_info, compressed_len), GFP_NOFS);
	if (!cb)
		goto out;

	refcount_set(&cb->pending_bios, 0);
	cb->errors = 0;
	cb->inode = inode;
	cb->mirror_num = mirror_num;
	sums = &cb->sums;

	cb->start = em->orig_start;
	em_len = em->len;
	em_start = em->start;

	free_extent_map(em);
	em = NULL;

	cb->len = bio->bi_iter.bi_size;
	cb->compressed_len = compressed_len;
	cb->compress_type = extent_compress_type(bio_flags);
	cb->orig_bio = bio;

	nr_pages = DIV_ROUND_UP(compressed_len, PAGE_SIZE);
	cb->compressed_pages = kcalloc(nr_pages, sizeof(struct page *),
				       GFP_NOFS);
	if (!cb->compressed_pages)
		goto fail1;

	bdev = fs_info->fs_devices->latest_bdev;

	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
		cb->compressed_pages[pg_index] = alloc_page(GFP_NOFS |
							      __GFP_HIGHMEM);
		if (!cb->compressed_pages[pg_index]) {
			faili = pg_index - 1;
			ret = BLK_STS_RESOURCE;
			goto fail2;
		}
	}
	faili = nr_pages - 1;
	cb->nr_pages = nr_pages;

	add_ra_bio_pages(inode, em_start + em_len, cb);

	/* include any pages we added in add_ra-bio_pages */
	cb->len = bio->bi_iter.bi_size;

	comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
	bio_set_op_attrs (comp_bio, REQ_OP_READ, 0);
	comp_bio->bi_private = cb;
	comp_bio->bi_end_io = end_compressed_bio_read;
	refcount_set(&cb->pending_bios, 1);

	for (pg_index = 0; pg_index < nr_pages; pg_index++) {
		int submit = 0;

		page = cb->compressed_pages[pg_index];
		page->mapping = inode->i_mapping;
		page->index = em_start >> PAGE_SHIFT;

		if (comp_bio->bi_iter.bi_size)
			submit = tree->ops->merge_bio_hook(page, 0,
							PAGE_SIZE,
							comp_bio, 0);

		page->mapping = NULL;
		if (submit || bio_add_page(comp_bio, page, PAGE_SIZE, 0) <
		    PAGE_SIZE) {
			bio_get(comp_bio);

			ret = btrfs_bio_wq_end_io(fs_info, comp_bio,
						  BTRFS_WQ_ENDIO_DATA);
			BUG_ON(ret); /* -ENOMEM */

			/*
			 * inc the count before we submit the bio so
			 * we know the end IO handler won't happen before
			 * we inc the count.  Otherwise, the cb might get
			 * freed before we're done setting it up
			 */
			refcount_inc(&cb->pending_bios);

			if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
				ret = btrfs_lookup_bio_sums(inode, comp_bio,
							    sums);
				BUG_ON(ret); /* -ENOMEM */
			}
			sums += DIV_ROUND_UP(comp_bio->bi_iter.bi_size,
					     fs_info->sectorsize);

			ret = btrfs_map_bio(fs_info, comp_bio, mirror_num, 0);
			if (ret) {
				comp_bio->bi_status = ret;
				bio_endio(comp_bio);
			}

			bio_put(comp_bio);

			comp_bio = btrfs_bio_alloc(bdev, cur_disk_byte);
			bio_set_op_attrs(comp_bio, REQ_OP_READ, 0);
			comp_bio->bi_private = cb;
			comp_bio->bi_end_io = end_compressed_bio_read;

			bio_add_page(comp_bio, page, PAGE_SIZE, 0);
		}
		cur_disk_byte += PAGE_SIZE;
	}
	bio_get(comp_bio);

	ret = btrfs_bio_wq_end_io(fs_info, comp_bio, BTRFS_WQ_ENDIO_DATA);
	BUG_ON(ret); /* -ENOMEM */

	if (!(BTRFS_I(inode)->flags & BTRFS_INODE_NODATASUM)) {
		ret = btrfs_lookup_bio_sums(inode, comp_bio, sums);
		BUG_ON(ret); /* -ENOMEM */
	}

	ret = btrfs_map_bio(fs_info, comp_bio, mirror_num, 0);
	if (ret) {
		comp_bio->bi_status = ret;
		bio_endio(comp_bio);
	}

	bio_put(comp_bio);
	return 0;

fail2:
	while (faili >= 0) {
		__free_page(cb->compressed_pages[faili]);
		faili--;
	}

	kfree(cb->compressed_pages);
fail1:
	kfree(cb);
out:
	free_extent_map(em);
	return ret;
}

static struct {
	struct list_head idle_ws;
	spinlock_t ws_lock;
	/* Number of free workspaces */
	int free_ws;
	/* Total number of allocated workspaces */
	atomic_t total_ws;
	/* Waiters for a free workspace */
	wait_queue_head_t ws_wait;
} btrfs_comp_ws[BTRFS_COMPRESS_TYPES];

static const struct btrfs_compress_op * const btrfs_compress_op[] = {
	&btrfs_zlib_compress,
	&btrfs_lzo_compress,
	&btrfs_zstd_compress,
};

void __init btrfs_init_compress(void)
{
	int i;

	for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
		struct list_head *workspace;

		INIT_LIST_HEAD(&btrfs_comp_ws[i].idle_ws);
		spin_lock_init(&btrfs_comp_ws[i].ws_lock);
		atomic_set(&btrfs_comp_ws[i].total_ws, 0);
		init_waitqueue_head(&btrfs_comp_ws[i].ws_wait);

		/*
		 * Preallocate one workspace for each compression type so
		 * we can guarantee forward progress in the worst case
		 */
		workspace = btrfs_compress_op[i]->alloc_workspace();
		if (IS_ERR(workspace)) {
			pr_warn("BTRFS: cannot preallocate compression workspace, will try later\n");
		} else {
			atomic_set(&btrfs_comp_ws[i].total_ws, 1);
			btrfs_comp_ws[i].free_ws = 1;
			list_add(workspace, &btrfs_comp_ws[i].idle_ws);
		}
	}
}

/*
 * This finds an available workspace or allocates a new one.
 * If it's not possible to allocate a new one, waits until there's one.
 * Preallocation makes a forward progress guarantees and we do not return
 * errors.
 */
static struct list_head *find_workspace(int type)
{
	struct list_head *workspace;
	int cpus = num_online_cpus();
	int idx = type - 1;
	unsigned nofs_flag;

	struct list_head *idle_ws	= &btrfs_comp_ws[idx].idle_ws;
	spinlock_t *ws_lock		= &btrfs_comp_ws[idx].ws_lock;
	atomic_t *total_ws		= &btrfs_comp_ws[idx].total_ws;
	wait_queue_head_t *ws_wait	= &btrfs_comp_ws[idx].ws_wait;
	int *free_ws			= &btrfs_comp_ws[idx].free_ws;
again:
	spin_lock(ws_lock);
	if (!list_empty(idle_ws)) {
		workspace = idle_ws->next;
		list_del(workspace);
		(*free_ws)--;
		spin_unlock(ws_lock);
		return workspace;

	}
	if (atomic_read(total_ws) > cpus) {
		DEFINE_WAIT(wait);

		spin_unlock(ws_lock);
		prepare_to_wait(ws_wait, &wait, TASK_UNINTERRUPTIBLE);
		if (atomic_read(total_ws) > cpus && !*free_ws)
			schedule();
		finish_wait(ws_wait, &wait);
		goto again;
	}
	atomic_inc(total_ws);
	spin_unlock(ws_lock);

	/*
	 * Allocation helpers call vmalloc that can't use GFP_NOFS, so we have
	 * to turn it off here because we might get called from the restricted
	 * context of btrfs_compress_bio/btrfs_compress_pages
	 */
	nofs_flag = memalloc_nofs_save();
	workspace = btrfs_compress_op[idx]->alloc_workspace();
	memalloc_nofs_restore(nofs_flag);

	if (IS_ERR(workspace)) {
		atomic_dec(total_ws);
		wake_up(ws_wait);

		/*
		 * Do not return the error but go back to waiting. There's a
		 * workspace preallocated for each type and the compression
		 * time is bounded so we get to a workspace eventually. This
		 * makes our caller's life easier.
		 *
		 * To prevent silent and low-probability deadlocks (when the
		 * initial preallocation fails), check if there are any
		 * workspaces at all.
		 */
		if (atomic_read(total_ws) == 0) {
			static DEFINE_RATELIMIT_STATE(_rs,
					/* once per minute */ 60 * HZ,
					/* no burst */ 1);

			if (__ratelimit(&_rs)) {
				pr_warn("BTRFS: no compression workspaces, low memory, retrying\n");
			}
		}
		goto again;
	}
	return workspace;
}

/*
 * put a workspace struct back on the list or free it if we have enough
 * idle ones sitting around
 */
static void free_workspace(int type, struct list_head *workspace)
{
	int idx = type - 1;
	struct list_head *idle_ws	= &btrfs_comp_ws[idx].idle_ws;
	spinlock_t *ws_lock		= &btrfs_comp_ws[idx].ws_lock;
	atomic_t *total_ws		= &btrfs_comp_ws[idx].total_ws;
	wait_queue_head_t *ws_wait	= &btrfs_comp_ws[idx].ws_wait;
	int *free_ws			= &btrfs_comp_ws[idx].free_ws;

	spin_lock(ws_lock);
	if (*free_ws <= num_online_cpus()) {
		list_add(workspace, idle_ws);
		(*free_ws)++;
		spin_unlock(ws_lock);
		goto wake;
	}
	spin_unlock(ws_lock);

	btrfs_compress_op[idx]->free_workspace(workspace);
	atomic_dec(total_ws);
wake:
	/*
	 * Make sure counter is updated before we wake up waiters.
	 */
	smp_mb();
	if (waitqueue_active(ws_wait))
		wake_up(ws_wait);
}

/*
 * cleanup function for module exit
 */
static void free_workspaces(void)
{
	struct list_head *workspace;
	int i;

	for (i = 0; i < BTRFS_COMPRESS_TYPES; i++) {
		while (!list_empty(&btrfs_comp_ws[i].idle_ws)) {
			workspace = btrfs_comp_ws[i].idle_ws.next;
			list_del(workspace);
			btrfs_compress_op[i]->free_workspace(workspace);
			atomic_dec(&btrfs_comp_ws[i].total_ws);
		}
	}
}

/*
 * Given an address space and start and length, compress the bytes into @pages
 * that are allocated on demand.
 *
 * @out_pages is an in/out parameter, holds maximum number of pages to allocate
 * and returns number of actually allocated pages
 *
 * @total_in is used to return the number of bytes actually read.  It
 * may be smaller than the input length if we had to exit early because we
 * ran out of room in the pages array or because we cross the
 * max_out threshold.
 *
 * @total_out is an in/out parameter, must be set to the input length and will
 * be also used to return the total number of compressed bytes
 *
 * @max_out tells us the max number of bytes that we're allowed to
 * stuff into pages
 */
int btrfs_compress_pages(int type, struct address_space *mapping,
			 u64 start, struct page **pages,
			 unsigned long *out_pages,
			 unsigned long *total_in,
			 unsigned long *total_out)
{
	struct list_head *workspace;
	int ret;

	workspace = find_workspace(type);

	ret = btrfs_compress_op[type-1]->compress_pages(workspace, mapping,
						      start, pages,
						      out_pages,
						      total_in, total_out);
	free_workspace(type, workspace);
	return ret;
}

/*
 * pages_in is an array of pages with compressed data.
 *
 * disk_start is the starting logical offset of this array in the file
 *
 * orig_bio contains the pages from the file that we want to decompress into
 *
 * srclen is the number of bytes in pages_in
 *
 * The basic idea is that we have a bio that was created by readpages.
 * The pages in the bio are for the uncompressed data, and they may not
 * be contiguous.  They all correspond to the range of bytes covered by
 * the compressed extent.
 */
static int btrfs_decompress_bio(struct compressed_bio *cb)
{
	struct list_head *workspace;
	int ret;
	int type = cb->compress_type;

	workspace = find_workspace(type);
	ret = btrfs_compress_op[type - 1]->decompress_bio(workspace, cb);
	free_workspace(type, workspace);

	return ret;
}

/*
 * a less complex decompression routine.  Our compressed data fits in a
 * single page, and we want to read a single page out of it.
 * start_byte tells us the offset into the compressed data we're interested in
 */
int btrfs_decompress(int type, unsigned char *data_in, struct page *dest_page,
		     unsigned long start_byte, size_t srclen, size_t destlen)
{
	struct list_head *workspace;
	int ret;

	workspace = find_workspace(type);

	ret = btrfs_compress_op[type-1]->decompress(workspace, data_in,
						  dest_page, start_byte,
						  srclen, destlen);

	free_workspace(type, workspace);
	return ret;
}

void btrfs_exit_compress(void)
{
	free_workspaces();
}

/*
 * Copy uncompressed data from working buffer to pages.
 *
 * buf_start is the byte offset we're of the start of our workspace buffer.
 *
 * total_out is the last byte of the buffer
 */
int btrfs_decompress_buf2page(const char *buf, unsigned long buf_start,
			      unsigned long total_out, u64 disk_start,
			      struct bio *bio)
{
	unsigned long buf_offset;
	unsigned long current_buf_start;
	unsigned long start_byte;
	unsigned long prev_start_byte;
	unsigned long working_bytes = total_out - buf_start;
	unsigned long bytes;
	char *kaddr;
	struct bio_vec bvec = bio_iter_iovec(bio, bio->bi_iter);

	/*
	 * start byte is the first byte of the page we're currently
	 * copying into relative to the start of the compressed data.
	 */
	start_byte = page_offset(bvec.bv_page) - disk_start;

	/* we haven't yet hit data corresponding to this page */
	if (total_out <= start_byte)
		return 1;

	/*
	 * the start of the data we care about is offset into
	 * the middle of our working buffer
	 */
	if (total_out > start_byte && buf_start < start_byte) {
		buf_offset = start_byte - buf_start;
		working_bytes -= buf_offset;
	} else {
		buf_offset = 0;
	}
	current_buf_start = buf_start;

	/* copy bytes from the working buffer into the pages */
	while (working_bytes > 0) {
		bytes = min_t(unsigned long, bvec.bv_len,
				PAGE_SIZE - buf_offset);
		bytes = min(bytes, working_bytes);

		kaddr = kmap_atomic(bvec.bv_page);
		memcpy(kaddr + bvec.bv_offset, buf + buf_offset, bytes);
		kunmap_atomic(kaddr);
		flush_dcache_page(bvec.bv_page);

		buf_offset += bytes;
		working_bytes -= bytes;
		current_buf_start += bytes;

		/* check if we need to pick another page */
		bio_advance(bio, bytes);
		if (!bio->bi_iter.bi_size)
			return 0;
		bvec = bio_iter_iovec(bio, bio->bi_iter);
		prev_start_byte = start_byte;
		start_byte = page_offset(bvec.bv_page) - disk_start;

		/*
		 * We need to make sure we're only adjusting
		 * our offset into compression working buffer when
		 * we're switching pages.  Otherwise we can incorrectly
		 * keep copying when we were actually done.
		 */
		if (start_byte != prev_start_byte) {
			/*
			 * make sure our new page is covered by this
			 * working buffer
			 */
			if (total_out <= start_byte)
				return 1;

			/*
			 * the next page in the biovec might not be adjacent
			 * to the last page, but it might still be found
			 * inside this working buffer. bump our offset pointer
			 */
			if (total_out > start_byte &&
			    current_buf_start < start_byte) {
				buf_offset = start_byte - buf_start;
				working_bytes = total_out - start_byte;
				current_buf_start = buf_start + buf_offset;
			}
		}
	}

	return 1;
}

/*
 * Compression heuristic.
 *
 * For now is's a naive and optimistic 'return true', we'll extend the logic to
 * quickly (compared to direct compression) detect data characteristics
 * (compressible/uncompressible) to avoid wasting CPU time on uncompressible
 * data.
 *
 * The following types of analysis can be performed:
 * - detect mostly zero data
 * - detect data with low "byte set" size (text, etc)
 * - detect data with low/high "core byte" set
 *
 * Return non-zero if the compression should be done, 0 otherwise.
 */
int btrfs_compress_heuristic(struct inode *inode, u64 start, u64 end)
{
	u64 index = start >> PAGE_SHIFT;
	u64 end_index = end >> PAGE_SHIFT;
	struct page *page;
	int ret = 1;

	while (index <= end_index) {
		page = find_get_page(inode->i_mapping, index);
		kmap(page);
		kunmap(page);
		put_page(page);
		index++;
	}

	return ret;
}