summaryrefslogtreecommitdiffstats
path: root/fs/xfs/libxfs/xfs_rmap_btree.c
blob: e0ae6da94fc3efebfef9a924563ce86677c7733a (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
// SPDX-License-Identifier: GPL-2.0
/*
 * Copyright (c) 2014 Red Hat, Inc.
 * All Rights Reserved.
 */
#include "xfs.h"
#include "xfs_fs.h"
#include "xfs_shared.h"
#include "xfs_format.h"
#include "xfs_log_format.h"
#include "xfs_trans_resv.h"
#include "xfs_mount.h"
#include "xfs_trans.h"
#include "xfs_alloc.h"
#include "xfs_btree.h"
#include "xfs_btree_staging.h"
#include "xfs_rmap.h"
#include "xfs_rmap_btree.h"
#include "xfs_trace.h"
#include "xfs_error.h"
#include "xfs_extent_busy.h"
#include "xfs_ag.h"
#include "xfs_ag_resv.h"

static struct kmem_cache	*xfs_rmapbt_cur_cache;

/*
 * Reverse map btree.
 *
 * This is a per-ag tree used to track the owner(s) of a given extent. With
 * reflink it is possible for there to be multiple owners, which is a departure
 * from classic XFS. Owner records for data extents are inserted when the
 * extent is mapped and removed when an extent is unmapped.  Owner records for
 * all other block types (i.e. metadata) are inserted when an extent is
 * allocated and removed when an extent is freed. There can only be one owner
 * of a metadata extent, usually an inode or some other metadata structure like
 * an AG btree.
 *
 * The rmap btree is part of the free space management, so blocks for the tree
 * are sourced from the agfl. Hence we need transaction reservation support for
 * this tree so that the freelist is always large enough. This also impacts on
 * the minimum space we need to leave free in the AG.
 *
 * The tree is ordered by [ag block, owner, offset]. This is a large key size,
 * but it is the only way to enforce unique keys when a block can be owned by
 * multiple files at any offset. There's no need to order/search by extent
 * size for online updating/management of the tree. It is intended that most
 * reverse lookups will be to find the owner(s) of a particular block, or to
 * try to recover tree and file data from corrupt primary metadata.
 */

static struct xfs_btree_cur *
xfs_rmapbt_dup_cursor(
	struct xfs_btree_cur	*cur)
{
	return xfs_rmapbt_init_cursor(cur->bc_mp, cur->bc_tp,
				cur->bc_ag.agbp, cur->bc_ag.pag);
}

STATIC void
xfs_rmapbt_set_root(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_ptr	*ptr,
	int				inc)
{
	struct xfs_buf		*agbp = cur->bc_ag.agbp;
	struct xfs_agf		*agf = agbp->b_addr;
	int			btnum = cur->bc_btnum;

	ASSERT(ptr->s != 0);

	agf->agf_roots[btnum] = ptr->s;
	be32_add_cpu(&agf->agf_levels[btnum], inc);
	cur->bc_ag.pag->pagf_levels[btnum] += inc;

	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS);
}

STATIC int
xfs_rmapbt_alloc_block(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_ptr	*start,
	union xfs_btree_ptr		*new,
	int				*stat)
{
	struct xfs_buf		*agbp = cur->bc_ag.agbp;
	struct xfs_agf		*agf = agbp->b_addr;
	struct xfs_perag	*pag = cur->bc_ag.pag;
	int			error;
	xfs_agblock_t		bno;

	/* Allocate the new block from the freelist. If we can't, give up.  */
	error = xfs_alloc_get_freelist(pag, cur->bc_tp, cur->bc_ag.agbp,
				       &bno, 1);
	if (error)
		return error;
	if (bno == NULLAGBLOCK) {
		*stat = 0;
		return 0;
	}

	xfs_extent_busy_reuse(cur->bc_mp, pag, bno, 1, false);

	new->s = cpu_to_be32(bno);
	be32_add_cpu(&agf->agf_rmap_blocks, 1);
	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);

	xfs_ag_resv_rmapbt_alloc(cur->bc_mp, pag->pag_agno);

	*stat = 1;
	return 0;
}

STATIC int
xfs_rmapbt_free_block(
	struct xfs_btree_cur	*cur,
	struct xfs_buf		*bp)
{
	struct xfs_buf		*agbp = cur->bc_ag.agbp;
	struct xfs_agf		*agf = agbp->b_addr;
	struct xfs_perag	*pag = cur->bc_ag.pag;
	xfs_agblock_t		bno;
	int			error;

	bno = xfs_daddr_to_agbno(cur->bc_mp, xfs_buf_daddr(bp));
	be32_add_cpu(&agf->agf_rmap_blocks, -1);
	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_RMAP_BLOCKS);
	error = xfs_alloc_put_freelist(pag, cur->bc_tp, agbp, NULL, bno, 1);
	if (error)
		return error;

	xfs_extent_busy_insert(cur->bc_tp, pag, bno, 1,
			      XFS_EXTENT_BUSY_SKIP_DISCARD);

	xfs_ag_resv_free_extent(pag, XFS_AG_RESV_RMAPBT, NULL, 1);
	return 0;
}

STATIC int
xfs_rmapbt_get_minrecs(
	struct xfs_btree_cur	*cur,
	int			level)
{
	return cur->bc_mp->m_rmap_mnr[level != 0];
}

STATIC int
xfs_rmapbt_get_maxrecs(
	struct xfs_btree_cur	*cur,
	int			level)
{
	return cur->bc_mp->m_rmap_mxr[level != 0];
}

/*
 * Convert the ondisk record's offset field into the ondisk key's offset field.
 * Fork and bmbt are significant parts of the rmap record key, but written
 * status is merely a record attribute.
 */
static inline __be64 ondisk_rec_offset_to_key(const union xfs_btree_rec *rec)
{
	return rec->rmap.rm_offset & ~cpu_to_be64(XFS_RMAP_OFF_UNWRITTEN);
}

STATIC void
xfs_rmapbt_init_key_from_rec(
	union xfs_btree_key		*key,
	const union xfs_btree_rec	*rec)
{
	key->rmap.rm_startblock = rec->rmap.rm_startblock;
	key->rmap.rm_owner = rec->rmap.rm_owner;
	key->rmap.rm_offset = ondisk_rec_offset_to_key(rec);
}

/*
 * The high key for a reverse mapping record can be computed by shifting
 * the startblock and offset to the highest value that would still map
 * to that record.  In practice this means that we add blockcount-1 to
 * the startblock for all records, and if the record is for a data/attr
 * fork mapping, we add blockcount-1 to the offset too.
 */
STATIC void
xfs_rmapbt_init_high_key_from_rec(
	union xfs_btree_key		*key,
	const union xfs_btree_rec	*rec)
{
	uint64_t			off;
	int				adj;

	adj = be32_to_cpu(rec->rmap.rm_blockcount) - 1;

	key->rmap.rm_startblock = rec->rmap.rm_startblock;
	be32_add_cpu(&key->rmap.rm_startblock, adj);
	key->rmap.rm_owner = rec->rmap.rm_owner;
	key->rmap.rm_offset = ondisk_rec_offset_to_key(rec);
	if (XFS_RMAP_NON_INODE_OWNER(be64_to_cpu(rec->rmap.rm_owner)) ||
	    XFS_RMAP_IS_BMBT_BLOCK(be64_to_cpu(rec->rmap.rm_offset)))
		return;
	off = be64_to_cpu(key->rmap.rm_offset);
	off = (XFS_RMAP_OFF(off) + adj) | (off & ~XFS_RMAP_OFF_MASK);
	key->rmap.rm_offset = cpu_to_be64(off);
}

STATIC void
xfs_rmapbt_init_rec_from_cur(
	struct xfs_btree_cur	*cur,
	union xfs_btree_rec	*rec)
{
	rec->rmap.rm_startblock = cpu_to_be32(cur->bc_rec.r.rm_startblock);
	rec->rmap.rm_blockcount = cpu_to_be32(cur->bc_rec.r.rm_blockcount);
	rec->rmap.rm_owner = cpu_to_be64(cur->bc_rec.r.rm_owner);
	rec->rmap.rm_offset = cpu_to_be64(
			xfs_rmap_irec_offset_pack(&cur->bc_rec.r));
}

STATIC void
xfs_rmapbt_init_ptr_from_cur(
	struct xfs_btree_cur	*cur,
	union xfs_btree_ptr	*ptr)
{
	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;

	ASSERT(cur->bc_ag.pag->pag_agno == be32_to_cpu(agf->agf_seqno));

	ptr->s = agf->agf_roots[cur->bc_btnum];
}

/*
 * Mask the appropriate parts of the ondisk key field for a key comparison.
 * Fork and bmbt are significant parts of the rmap record key, but written
 * status is merely a record attribute.
 */
static inline uint64_t offset_keymask(uint64_t offset)
{
	return offset & ~XFS_RMAP_OFF_UNWRITTEN;
}

STATIC int64_t
xfs_rmapbt_key_diff(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_key	*key)
{
	struct xfs_rmap_irec		*rec = &cur->bc_rec.r;
	const struct xfs_rmap_key	*kp = &key->rmap;
	__u64				x, y;
	int64_t				d;

	d = (int64_t)be32_to_cpu(kp->rm_startblock) - rec->rm_startblock;
	if (d)
		return d;

	x = be64_to_cpu(kp->rm_owner);
	y = rec->rm_owner;
	if (x > y)
		return 1;
	else if (y > x)
		return -1;

	x = offset_keymask(be64_to_cpu(kp->rm_offset));
	y = offset_keymask(xfs_rmap_irec_offset_pack(rec));
	if (x > y)
		return 1;
	else if (y > x)
		return -1;
	return 0;
}

STATIC int64_t
xfs_rmapbt_diff_two_keys(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_key	*k1,
	const union xfs_btree_key	*k2,
	const union xfs_btree_key	*mask)
{
	const struct xfs_rmap_key	*kp1 = &k1->rmap;
	const struct xfs_rmap_key	*kp2 = &k2->rmap;
	int64_t				d;
	__u64				x, y;

	/* Doesn't make sense to mask off the physical space part */
	ASSERT(!mask || mask->rmap.rm_startblock);

	d = (int64_t)be32_to_cpu(kp1->rm_startblock) -
		     be32_to_cpu(kp2->rm_startblock);
	if (d)
		return d;

	if (!mask || mask->rmap.rm_owner) {
		x = be64_to_cpu(kp1->rm_owner);
		y = be64_to_cpu(kp2->rm_owner);
		if (x > y)
			return 1;
		else if (y > x)
			return -1;
	}

	if (!mask || mask->rmap.rm_offset) {
		/* Doesn't make sense to allow offset but not owner */
		ASSERT(!mask || mask->rmap.rm_owner);

		x = offset_keymask(be64_to_cpu(kp1->rm_offset));
		y = offset_keymask(be64_to_cpu(kp2->rm_offset));
		if (x > y)
			return 1;
		else if (y > x)
			return -1;
	}

	return 0;
}

static xfs_failaddr_t
xfs_rmapbt_verify(
	struct xfs_buf		*bp)
{
	struct xfs_mount	*mp = bp->b_mount;
	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
	struct xfs_perag	*pag = bp->b_pag;
	xfs_failaddr_t		fa;
	unsigned int		level;

	/*
	 * magic number and level verification
	 *
	 * During growfs operations, we can't verify the exact level or owner as
	 * the perag is not fully initialised and hence not attached to the
	 * buffer.  In this case, check against the maximum tree depth.
	 *
	 * Similarly, during log recovery we will have a perag structure
	 * attached, but the agf information will not yet have been initialised
	 * from the on disk AGF. Again, we can only check against maximum limits
	 * in this case.
	 */
	if (!xfs_verify_magic(bp, block->bb_magic))
		return __this_address;

	if (!xfs_has_rmapbt(mp))
		return __this_address;
	fa = xfs_btree_sblock_v5hdr_verify(bp);
	if (fa)
		return fa;

	level = be16_to_cpu(block->bb_level);
	if (pag && xfs_perag_initialised_agf(pag)) {
		if (level >= pag->pagf_levels[XFS_BTNUM_RMAPi])
			return __this_address;
	} else if (level >= mp->m_rmap_maxlevels)
		return __this_address;

	return xfs_btree_sblock_verify(bp, mp->m_rmap_mxr[level != 0]);
}

static void
xfs_rmapbt_read_verify(
	struct xfs_buf	*bp)
{
	xfs_failaddr_t	fa;

	if (!xfs_btree_sblock_verify_crc(bp))
		xfs_verifier_error(bp, -EFSBADCRC, __this_address);
	else {
		fa = xfs_rmapbt_verify(bp);
		if (fa)
			xfs_verifier_error(bp, -EFSCORRUPTED, fa);
	}

	if (bp->b_error)
		trace_xfs_btree_corrupt(bp, _RET_IP_);
}

static void
xfs_rmapbt_write_verify(
	struct xfs_buf	*bp)
{
	xfs_failaddr_t	fa;

	fa = xfs_rmapbt_verify(bp);
	if (fa) {
		trace_xfs_btree_corrupt(bp, _RET_IP_);
		xfs_verifier_error(bp, -EFSCORRUPTED, fa);
		return;
	}
	xfs_btree_sblock_calc_crc(bp);

}

const struct xfs_buf_ops xfs_rmapbt_buf_ops = {
	.name			= "xfs_rmapbt",
	.magic			= { 0, cpu_to_be32(XFS_RMAP_CRC_MAGIC) },
	.verify_read		= xfs_rmapbt_read_verify,
	.verify_write		= xfs_rmapbt_write_verify,
	.verify_struct		= xfs_rmapbt_verify,
};

STATIC int
xfs_rmapbt_keys_inorder(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_key	*k1,
	const union xfs_btree_key	*k2)
{
	uint32_t		x;
	uint32_t		y;
	uint64_t		a;
	uint64_t		b;

	x = be32_to_cpu(k1->rmap.rm_startblock);
	y = be32_to_cpu(k2->rmap.rm_startblock);
	if (x < y)
		return 1;
	else if (x > y)
		return 0;
	a = be64_to_cpu(k1->rmap.rm_owner);
	b = be64_to_cpu(k2->rmap.rm_owner);
	if (a < b)
		return 1;
	else if (a > b)
		return 0;
	a = offset_keymask(be64_to_cpu(k1->rmap.rm_offset));
	b = offset_keymask(be64_to_cpu(k2->rmap.rm_offset));
	if (a <= b)
		return 1;
	return 0;
}

STATIC int
xfs_rmapbt_recs_inorder(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_rec	*r1,
	const union xfs_btree_rec	*r2)
{
	uint32_t		x;
	uint32_t		y;
	uint64_t		a;
	uint64_t		b;

	x = be32_to_cpu(r1->rmap.rm_startblock);
	y = be32_to_cpu(r2->rmap.rm_startblock);
	if (x < y)
		return 1;
	else if (x > y)
		return 0;
	a = be64_to_cpu(r1->rmap.rm_owner);
	b = be64_to_cpu(r2->rmap.rm_owner);
	if (a < b)
		return 1;
	else if (a > b)
		return 0;
	a = offset_keymask(be64_to_cpu(r1->rmap.rm_offset));
	b = offset_keymask(be64_to_cpu(r2->rmap.rm_offset));
	if (a <= b)
		return 1;
	return 0;
}

STATIC enum xbtree_key_contig
xfs_rmapbt_keys_contiguous(
	struct xfs_btree_cur		*cur,
	const union xfs_btree_key	*key1,
	const union xfs_btree_key	*key2,
	const union xfs_btree_key	*mask)
{
	ASSERT(!mask || mask->rmap.rm_startblock);

	/*
	 * We only support checking contiguity of the physical space component.
	 * If any callers ever need more specificity than that, they'll have to
	 * implement it here.
	 */
	ASSERT(!mask || (!mask->rmap.rm_owner && !mask->rmap.rm_offset));

	return xbtree_key_contig(be32_to_cpu(key1->rmap.rm_startblock),
				 be32_to_cpu(key2->rmap.rm_startblock));
}

const struct xfs_btree_ops xfs_rmapbt_ops = {
	.geom_flags		= XFS_BTGEO_OVERLAPPING,

	.rec_len		= sizeof(struct xfs_rmap_rec),
	.key_len		= 2 * sizeof(struct xfs_rmap_key),

	.lru_refs		= XFS_RMAP_BTREE_REF,
	.statoff		= XFS_STATS_CALC_INDEX(xs_rmap_2),

	.dup_cursor		= xfs_rmapbt_dup_cursor,
	.set_root		= xfs_rmapbt_set_root,
	.alloc_block		= xfs_rmapbt_alloc_block,
	.free_block		= xfs_rmapbt_free_block,
	.get_minrecs		= xfs_rmapbt_get_minrecs,
	.get_maxrecs		= xfs_rmapbt_get_maxrecs,
	.init_key_from_rec	= xfs_rmapbt_init_key_from_rec,
	.init_high_key_from_rec	= xfs_rmapbt_init_high_key_from_rec,
	.init_rec_from_cur	= xfs_rmapbt_init_rec_from_cur,
	.init_ptr_from_cur	= xfs_rmapbt_init_ptr_from_cur,
	.key_diff		= xfs_rmapbt_key_diff,
	.buf_ops		= &xfs_rmapbt_buf_ops,
	.diff_two_keys		= xfs_rmapbt_diff_two_keys,
	.keys_inorder		= xfs_rmapbt_keys_inorder,
	.recs_inorder		= xfs_rmapbt_recs_inorder,
	.keys_contiguous	= xfs_rmapbt_keys_contiguous,
};

static struct xfs_btree_cur *
xfs_rmapbt_init_common(
	struct xfs_mount	*mp,
	struct xfs_trans	*tp,
	struct xfs_perag	*pag)
{
	struct xfs_btree_cur	*cur;

	/* Overlapping btree; 2 keys per pointer. */
	cur = xfs_btree_alloc_cursor(mp, tp, XFS_BTNUM_RMAP, &xfs_rmapbt_ops,
			mp->m_rmap_maxlevels, xfs_rmapbt_cur_cache);
	cur->bc_ag.pag = xfs_perag_hold(pag);
	return cur;
}

/* Create a new reverse mapping btree cursor. */
struct xfs_btree_cur *
xfs_rmapbt_init_cursor(
	struct xfs_mount	*mp,
	struct xfs_trans	*tp,
	struct xfs_buf		*agbp,
	struct xfs_perag	*pag)
{
	struct xfs_agf		*agf = agbp->b_addr;
	struct xfs_btree_cur	*cur;

	cur = xfs_rmapbt_init_common(mp, tp, pag);
	cur->bc_nlevels = be32_to_cpu(agf->agf_levels[XFS_BTNUM_RMAP]);
	cur->bc_ag.agbp = agbp;
	return cur;
}

/* Create a new reverse mapping btree cursor with a fake root for staging. */
struct xfs_btree_cur *
xfs_rmapbt_stage_cursor(
	struct xfs_mount	*mp,
	struct xbtree_afakeroot	*afake,
	struct xfs_perag	*pag)
{
	struct xfs_btree_cur	*cur;

	cur = xfs_rmapbt_init_common(mp, NULL, pag);
	xfs_btree_stage_afakeroot(cur, afake);
	return cur;
}

/*
 * Install a new reverse mapping btree root.  Caller is responsible for
 * invalidating and freeing the old btree blocks.
 */
void
xfs_rmapbt_commit_staged_btree(
	struct xfs_btree_cur	*cur,
	struct xfs_trans	*tp,
	struct xfs_buf		*agbp)
{
	struct xfs_agf		*agf = agbp->b_addr;
	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;

	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);

	agf->agf_roots[cur->bc_btnum] = cpu_to_be32(afake->af_root);
	agf->agf_levels[cur->bc_btnum] = cpu_to_be32(afake->af_levels);
	agf->agf_rmap_blocks = cpu_to_be32(afake->af_blocks);
	xfs_alloc_log_agf(tp, agbp, XFS_AGF_ROOTS | XFS_AGF_LEVELS |
				    XFS_AGF_RMAP_BLOCKS);
	xfs_btree_commit_afakeroot(cur, tp, agbp, &xfs_rmapbt_ops);
}

/* Calculate number of records in a reverse mapping btree block. */
static inline unsigned int
xfs_rmapbt_block_maxrecs(
	unsigned int		blocklen,
	bool			leaf)
{
	if (leaf)
		return blocklen / sizeof(struct xfs_rmap_rec);
	return blocklen /
		(2 * sizeof(struct xfs_rmap_key) + sizeof(xfs_rmap_ptr_t));
}

/*
 * Calculate number of records in an rmap btree block.
 */
int
xfs_rmapbt_maxrecs(
	int			blocklen,
	int			leaf)
{
	blocklen -= XFS_RMAP_BLOCK_LEN;
	return xfs_rmapbt_block_maxrecs(blocklen, leaf);
}

/* Compute the max possible height for reverse mapping btrees. */
unsigned int
xfs_rmapbt_maxlevels_ondisk(void)
{
	unsigned int		minrecs[2];
	unsigned int		blocklen;

	blocklen = XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN;

	minrecs[0] = xfs_rmapbt_block_maxrecs(blocklen, true) / 2;
	minrecs[1] = xfs_rmapbt_block_maxrecs(blocklen, false) / 2;

	/*
	 * Compute the asymptotic maxlevels for an rmapbt on any reflink fs.
	 *
	 * On a reflink filesystem, each AG block can have up to 2^32 (per the
	 * refcount record format) owners, which means that theoretically we
	 * could face up to 2^64 rmap records.  However, we're likely to run
	 * out of blocks in the AG long before that happens, which means that
	 * we must compute the max height based on what the btree will look
	 * like if it consumes almost all the blocks in the AG due to maximal
	 * sharing factor.
	 */
	return xfs_btree_space_to_height(minrecs, XFS_MAX_CRC_AG_BLOCKS);
}

/* Compute the maximum height of an rmap btree. */
void
xfs_rmapbt_compute_maxlevels(
	struct xfs_mount		*mp)
{
	if (!xfs_has_rmapbt(mp)) {
		mp->m_rmap_maxlevels = 0;
		return;
	}

	if (xfs_has_reflink(mp)) {
		/*
		 * Compute the asymptotic maxlevels for an rmap btree on a
		 * filesystem that supports reflink.
		 *
		 * On a reflink filesystem, each AG block can have up to 2^32
		 * (per the refcount record format) owners, which means that
		 * theoretically we could face up to 2^64 rmap records.
		 * However, we're likely to run out of blocks in the AG long
		 * before that happens, which means that we must compute the
		 * max height based on what the btree will look like if it
		 * consumes almost all the blocks in the AG due to maximal
		 * sharing factor.
		 */
		mp->m_rmap_maxlevels = xfs_btree_space_to_height(mp->m_rmap_mnr,
				mp->m_sb.sb_agblocks);
	} else {
		/*
		 * If there's no block sharing, compute the maximum rmapbt
		 * height assuming one rmap record per AG block.
		 */
		mp->m_rmap_maxlevels = xfs_btree_compute_maxlevels(
				mp->m_rmap_mnr, mp->m_sb.sb_agblocks);
	}
	ASSERT(mp->m_rmap_maxlevels <= xfs_rmapbt_maxlevels_ondisk());
}

/* Calculate the refcount btree size for some records. */
xfs_extlen_t
xfs_rmapbt_calc_size(
	struct xfs_mount	*mp,
	unsigned long long	len)
{
	return xfs_btree_calc_size(mp->m_rmap_mnr, len);
}

/*
 * Calculate the maximum refcount btree size.
 */
xfs_extlen_t
xfs_rmapbt_max_size(
	struct xfs_mount	*mp,
	xfs_agblock_t		agblocks)
{
	/* Bail out if we're uninitialized, which can happen in mkfs. */
	if (mp->m_rmap_mxr[0] == 0)
		return 0;

	return xfs_rmapbt_calc_size(mp, agblocks);
}

/*
 * Figure out how many blocks to reserve and how many are used by this btree.
 */
int
xfs_rmapbt_calc_reserves(
	struct xfs_mount	*mp,
	struct xfs_trans	*tp,
	struct xfs_perag	*pag,
	xfs_extlen_t		*ask,
	xfs_extlen_t		*used)
{
	struct xfs_buf		*agbp;
	struct xfs_agf		*agf;
	xfs_agblock_t		agblocks;
	xfs_extlen_t		tree_len;
	int			error;

	if (!xfs_has_rmapbt(mp))
		return 0;

	error = xfs_alloc_read_agf(pag, tp, 0, &agbp);
	if (error)
		return error;

	agf = agbp->b_addr;
	agblocks = be32_to_cpu(agf->agf_length);
	tree_len = be32_to_cpu(agf->agf_rmap_blocks);
	xfs_trans_brelse(tp, agbp);

	/*
	 * The log is permanently allocated, so the space it occupies will
	 * never be available for the kinds of things that would require btree
	 * expansion.  We therefore can pretend the space isn't there.
	 */
	if (xfs_ag_contains_log(mp, pag->pag_agno))
		agblocks -= mp->m_sb.sb_logblocks;

	/* Reserve 1% of the AG or enough for 1 block per record. */
	*ask += max(agblocks / 100, xfs_rmapbt_max_size(mp, agblocks));
	*used += tree_len;

	return error;
}

int __init
xfs_rmapbt_init_cur_cache(void)
{
	xfs_rmapbt_cur_cache = kmem_cache_create("xfs_rmapbt_cur",
			xfs_btree_cur_sizeof(xfs_rmapbt_maxlevels_ondisk()),
			0, 0, NULL);

	if (!xfs_rmapbt_cur_cache)
		return -ENOMEM;
	return 0;
}

void
xfs_rmapbt_destroy_cur_cache(void)
{
	kmem_cache_destroy(xfs_rmapbt_cur_cache);
	xfs_rmapbt_cur_cache = NULL;
}