xref: /linux/fs/xfs/libxfs/xfs_refcount_btree.c (revision 4b660dbd9ee2059850fd30e0df420ca7a38a1856)
1 // SPDX-License-Identifier: GPL-2.0+
2 /*
3  * Copyright (C) 2016 Oracle.  All Rights Reserved.
4  * Author: Darrick J. Wong <darrick.wong@oracle.com>
5  */
6 #include "xfs.h"
7 #include "xfs_fs.h"
8 #include "xfs_shared.h"
9 #include "xfs_format.h"
10 #include "xfs_log_format.h"
11 #include "xfs_trans_resv.h"
12 #include "xfs_mount.h"
13 #include "xfs_btree.h"
14 #include "xfs_btree_staging.h"
15 #include "xfs_refcount_btree.h"
16 #include "xfs_refcount.h"
17 #include "xfs_alloc.h"
18 #include "xfs_error.h"
19 #include "xfs_health.h"
20 #include "xfs_trace.h"
21 #include "xfs_trans.h"
22 #include "xfs_bit.h"
23 #include "xfs_rmap.h"
24 #include "xfs_ag.h"
25 
26 static struct kmem_cache	*xfs_refcountbt_cur_cache;
27 
28 static struct xfs_btree_cur *
29 xfs_refcountbt_dup_cursor(
30 	struct xfs_btree_cur	*cur)
31 {
32 	return xfs_refcountbt_init_cursor(cur->bc_mp, cur->bc_tp,
33 			cur->bc_ag.agbp, cur->bc_ag.pag);
34 }
35 
36 STATIC void
37 xfs_refcountbt_set_root(
38 	struct xfs_btree_cur		*cur,
39 	const union xfs_btree_ptr	*ptr,
40 	int				inc)
41 {
42 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
43 	struct xfs_agf		*agf = agbp->b_addr;
44 	struct xfs_perag	*pag = agbp->b_pag;
45 
46 	ASSERT(ptr->s != 0);
47 
48 	agf->agf_refcount_root = ptr->s;
49 	be32_add_cpu(&agf->agf_refcount_level, inc);
50 	pag->pagf_refcount_level += inc;
51 
52 	xfs_alloc_log_agf(cur->bc_tp, agbp,
53 			XFS_AGF_REFCOUNT_ROOT | XFS_AGF_REFCOUNT_LEVEL);
54 }
55 
56 STATIC int
57 xfs_refcountbt_alloc_block(
58 	struct xfs_btree_cur		*cur,
59 	const union xfs_btree_ptr	*start,
60 	union xfs_btree_ptr		*new,
61 	int				*stat)
62 {
63 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
64 	struct xfs_agf		*agf = agbp->b_addr;
65 	struct xfs_alloc_arg	args;		/* block allocation args */
66 	int			error;		/* error return value */
67 
68 	memset(&args, 0, sizeof(args));
69 	args.tp = cur->bc_tp;
70 	args.mp = cur->bc_mp;
71 	args.pag = cur->bc_ag.pag;
72 	args.oinfo = XFS_RMAP_OINFO_REFC;
73 	args.minlen = args.maxlen = args.prod = 1;
74 	args.resv = XFS_AG_RESV_METADATA;
75 
76 	error = xfs_alloc_vextent_near_bno(&args,
77 			XFS_AGB_TO_FSB(args.mp, args.pag->pag_agno,
78 					xfs_refc_block(args.mp)));
79 	if (error)
80 		goto out_error;
81 	if (args.fsbno == NULLFSBLOCK) {
82 		*stat = 0;
83 		return 0;
84 	}
85 	ASSERT(args.agno == cur->bc_ag.pag->pag_agno);
86 	ASSERT(args.len == 1);
87 
88 	new->s = cpu_to_be32(args.agbno);
89 	be32_add_cpu(&agf->agf_refcount_blocks, 1);
90 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_REFCOUNT_BLOCKS);
91 
92 	*stat = 1;
93 	return 0;
94 
95 out_error:
96 	return error;
97 }
98 
99 STATIC int
100 xfs_refcountbt_free_block(
101 	struct xfs_btree_cur	*cur,
102 	struct xfs_buf		*bp)
103 {
104 	struct xfs_mount	*mp = cur->bc_mp;
105 	struct xfs_buf		*agbp = cur->bc_ag.agbp;
106 	struct xfs_agf		*agf = agbp->b_addr;
107 	xfs_fsblock_t		fsbno = XFS_DADDR_TO_FSB(mp, xfs_buf_daddr(bp));
108 
109 	be32_add_cpu(&agf->agf_refcount_blocks, -1);
110 	xfs_alloc_log_agf(cur->bc_tp, agbp, XFS_AGF_REFCOUNT_BLOCKS);
111 	return xfs_free_extent_later(cur->bc_tp, fsbno, 1,
112 			&XFS_RMAP_OINFO_REFC, XFS_AG_RESV_METADATA, false);
113 }
114 
115 STATIC int
116 xfs_refcountbt_get_minrecs(
117 	struct xfs_btree_cur	*cur,
118 	int			level)
119 {
120 	return cur->bc_mp->m_refc_mnr[level != 0];
121 }
122 
123 STATIC int
124 xfs_refcountbt_get_maxrecs(
125 	struct xfs_btree_cur	*cur,
126 	int			level)
127 {
128 	return cur->bc_mp->m_refc_mxr[level != 0];
129 }
130 
131 STATIC void
132 xfs_refcountbt_init_key_from_rec(
133 	union xfs_btree_key		*key,
134 	const union xfs_btree_rec	*rec)
135 {
136 	key->refc.rc_startblock = rec->refc.rc_startblock;
137 }
138 
139 STATIC void
140 xfs_refcountbt_init_high_key_from_rec(
141 	union xfs_btree_key		*key,
142 	const union xfs_btree_rec	*rec)
143 {
144 	__u32				x;
145 
146 	x = be32_to_cpu(rec->refc.rc_startblock);
147 	x += be32_to_cpu(rec->refc.rc_blockcount) - 1;
148 	key->refc.rc_startblock = cpu_to_be32(x);
149 }
150 
151 STATIC void
152 xfs_refcountbt_init_rec_from_cur(
153 	struct xfs_btree_cur	*cur,
154 	union xfs_btree_rec	*rec)
155 {
156 	const struct xfs_refcount_irec *irec = &cur->bc_rec.rc;
157 	uint32_t		start;
158 
159 	start = xfs_refcount_encode_startblock(irec->rc_startblock,
160 			irec->rc_domain);
161 	rec->refc.rc_startblock = cpu_to_be32(start);
162 	rec->refc.rc_blockcount = cpu_to_be32(cur->bc_rec.rc.rc_blockcount);
163 	rec->refc.rc_refcount = cpu_to_be32(cur->bc_rec.rc.rc_refcount);
164 }
165 
166 STATIC void
167 xfs_refcountbt_init_ptr_from_cur(
168 	struct xfs_btree_cur	*cur,
169 	union xfs_btree_ptr	*ptr)
170 {
171 	struct xfs_agf		*agf = cur->bc_ag.agbp->b_addr;
172 
173 	ASSERT(cur->bc_ag.pag->pag_agno == be32_to_cpu(agf->agf_seqno));
174 
175 	ptr->s = agf->agf_refcount_root;
176 }
177 
178 STATIC int64_t
179 xfs_refcountbt_key_diff(
180 	struct xfs_btree_cur		*cur,
181 	const union xfs_btree_key	*key)
182 {
183 	const struct xfs_refcount_key	*kp = &key->refc;
184 	const struct xfs_refcount_irec	*irec = &cur->bc_rec.rc;
185 	uint32_t			start;
186 
187 	start = xfs_refcount_encode_startblock(irec->rc_startblock,
188 			irec->rc_domain);
189 	return (int64_t)be32_to_cpu(kp->rc_startblock) - start;
190 }
191 
192 STATIC int64_t
193 xfs_refcountbt_diff_two_keys(
194 	struct xfs_btree_cur		*cur,
195 	const union xfs_btree_key	*k1,
196 	const union xfs_btree_key	*k2,
197 	const union xfs_btree_key	*mask)
198 {
199 	ASSERT(!mask || mask->refc.rc_startblock);
200 
201 	return (int64_t)be32_to_cpu(k1->refc.rc_startblock) -
202 			be32_to_cpu(k2->refc.rc_startblock);
203 }
204 
205 STATIC xfs_failaddr_t
206 xfs_refcountbt_verify(
207 	struct xfs_buf		*bp)
208 {
209 	struct xfs_mount	*mp = bp->b_mount;
210 	struct xfs_btree_block	*block = XFS_BUF_TO_BLOCK(bp);
211 	struct xfs_perag	*pag = bp->b_pag;
212 	xfs_failaddr_t		fa;
213 	unsigned int		level;
214 
215 	if (!xfs_verify_magic(bp, block->bb_magic))
216 		return __this_address;
217 
218 	if (!xfs_has_reflink(mp))
219 		return __this_address;
220 	fa = xfs_btree_agblock_v5hdr_verify(bp);
221 	if (fa)
222 		return fa;
223 
224 	level = be16_to_cpu(block->bb_level);
225 	if (pag && xfs_perag_initialised_agf(pag)) {
226 		unsigned int	maxlevel = pag->pagf_refcount_level;
227 
228 #ifdef CONFIG_XFS_ONLINE_REPAIR
229 		/*
230 		 * Online repair could be rewriting the refcount btree, so
231 		 * we'll validate against the larger of either tree while this
232 		 * is going on.
233 		 */
234 		maxlevel = max_t(unsigned int, maxlevel,
235 				pag->pagf_repair_refcount_level);
236 #endif
237 		if (level >= maxlevel)
238 			return __this_address;
239 	} else if (level >= mp->m_refc_maxlevels)
240 		return __this_address;
241 
242 	return xfs_btree_agblock_verify(bp, mp->m_refc_mxr[level != 0]);
243 }
244 
245 STATIC void
246 xfs_refcountbt_read_verify(
247 	struct xfs_buf	*bp)
248 {
249 	xfs_failaddr_t	fa;
250 
251 	if (!xfs_btree_agblock_verify_crc(bp))
252 		xfs_verifier_error(bp, -EFSBADCRC, __this_address);
253 	else {
254 		fa = xfs_refcountbt_verify(bp);
255 		if (fa)
256 			xfs_verifier_error(bp, -EFSCORRUPTED, fa);
257 	}
258 
259 	if (bp->b_error)
260 		trace_xfs_btree_corrupt(bp, _RET_IP_);
261 }
262 
263 STATIC void
264 xfs_refcountbt_write_verify(
265 	struct xfs_buf	*bp)
266 {
267 	xfs_failaddr_t	fa;
268 
269 	fa = xfs_refcountbt_verify(bp);
270 	if (fa) {
271 		trace_xfs_btree_corrupt(bp, _RET_IP_);
272 		xfs_verifier_error(bp, -EFSCORRUPTED, fa);
273 		return;
274 	}
275 	xfs_btree_agblock_calc_crc(bp);
276 
277 }
278 
279 const struct xfs_buf_ops xfs_refcountbt_buf_ops = {
280 	.name			= "xfs_refcountbt",
281 	.magic			= { 0, cpu_to_be32(XFS_REFC_CRC_MAGIC) },
282 	.verify_read		= xfs_refcountbt_read_verify,
283 	.verify_write		= xfs_refcountbt_write_verify,
284 	.verify_struct		= xfs_refcountbt_verify,
285 };
286 
287 STATIC int
288 xfs_refcountbt_keys_inorder(
289 	struct xfs_btree_cur		*cur,
290 	const union xfs_btree_key	*k1,
291 	const union xfs_btree_key	*k2)
292 {
293 	return be32_to_cpu(k1->refc.rc_startblock) <
294 	       be32_to_cpu(k2->refc.rc_startblock);
295 }
296 
297 STATIC int
298 xfs_refcountbt_recs_inorder(
299 	struct xfs_btree_cur		*cur,
300 	const union xfs_btree_rec	*r1,
301 	const union xfs_btree_rec	*r2)
302 {
303 	return  be32_to_cpu(r1->refc.rc_startblock) +
304 		be32_to_cpu(r1->refc.rc_blockcount) <=
305 		be32_to_cpu(r2->refc.rc_startblock);
306 }
307 
308 STATIC enum xbtree_key_contig
309 xfs_refcountbt_keys_contiguous(
310 	struct xfs_btree_cur		*cur,
311 	const union xfs_btree_key	*key1,
312 	const union xfs_btree_key	*key2,
313 	const union xfs_btree_key	*mask)
314 {
315 	ASSERT(!mask || mask->refc.rc_startblock);
316 
317 	return xbtree_key_contig(be32_to_cpu(key1->refc.rc_startblock),
318 				 be32_to_cpu(key2->refc.rc_startblock));
319 }
320 
321 const struct xfs_btree_ops xfs_refcountbt_ops = {
322 	.name			= "refcount",
323 	.type			= XFS_BTREE_TYPE_AG,
324 
325 	.rec_len		= sizeof(struct xfs_refcount_rec),
326 	.key_len		= sizeof(struct xfs_refcount_key),
327 	.ptr_len		= XFS_BTREE_SHORT_PTR_LEN,
328 
329 	.lru_refs		= XFS_REFC_BTREE_REF,
330 	.statoff		= XFS_STATS_CALC_INDEX(xs_refcbt_2),
331 	.sick_mask		= XFS_SICK_AG_REFCNTBT,
332 
333 	.dup_cursor		= xfs_refcountbt_dup_cursor,
334 	.set_root		= xfs_refcountbt_set_root,
335 	.alloc_block		= xfs_refcountbt_alloc_block,
336 	.free_block		= xfs_refcountbt_free_block,
337 	.get_minrecs		= xfs_refcountbt_get_minrecs,
338 	.get_maxrecs		= xfs_refcountbt_get_maxrecs,
339 	.init_key_from_rec	= xfs_refcountbt_init_key_from_rec,
340 	.init_high_key_from_rec	= xfs_refcountbt_init_high_key_from_rec,
341 	.init_rec_from_cur	= xfs_refcountbt_init_rec_from_cur,
342 	.init_ptr_from_cur	= xfs_refcountbt_init_ptr_from_cur,
343 	.key_diff		= xfs_refcountbt_key_diff,
344 	.buf_ops		= &xfs_refcountbt_buf_ops,
345 	.diff_two_keys		= xfs_refcountbt_diff_two_keys,
346 	.keys_inorder		= xfs_refcountbt_keys_inorder,
347 	.recs_inorder		= xfs_refcountbt_recs_inorder,
348 	.keys_contiguous	= xfs_refcountbt_keys_contiguous,
349 };
350 
351 /*
352  * Create a new refcount btree cursor.
353  *
354  * For staging cursors tp and agbp are NULL.
355  */
356 struct xfs_btree_cur *
357 xfs_refcountbt_init_cursor(
358 	struct xfs_mount	*mp,
359 	struct xfs_trans	*tp,
360 	struct xfs_buf		*agbp,
361 	struct xfs_perag	*pag)
362 {
363 	struct xfs_btree_cur	*cur;
364 
365 	ASSERT(pag->pag_agno < mp->m_sb.sb_agcount);
366 
367 	cur = xfs_btree_alloc_cursor(mp, tp, &xfs_refcountbt_ops,
368 			mp->m_refc_maxlevels, xfs_refcountbt_cur_cache);
369 	cur->bc_ag.pag = xfs_perag_hold(pag);
370 	cur->bc_refc.nr_ops = 0;
371 	cur->bc_refc.shape_changes = 0;
372 	cur->bc_ag.agbp = agbp;
373 	if (agbp) {
374 		struct xfs_agf		*agf = agbp->b_addr;
375 
376 		cur->bc_nlevels = be32_to_cpu(agf->agf_refcount_level);
377 	}
378 	return cur;
379 }
380 
381 /*
382  * Swap in the new btree root.  Once we pass this point the newly rebuilt btree
383  * is in place and we have to kill off all the old btree blocks.
384  */
385 void
386 xfs_refcountbt_commit_staged_btree(
387 	struct xfs_btree_cur	*cur,
388 	struct xfs_trans	*tp,
389 	struct xfs_buf		*agbp)
390 {
391 	struct xfs_agf		*agf = agbp->b_addr;
392 	struct xbtree_afakeroot	*afake = cur->bc_ag.afake;
393 
394 	ASSERT(cur->bc_flags & XFS_BTREE_STAGING);
395 
396 	agf->agf_refcount_root = cpu_to_be32(afake->af_root);
397 	agf->agf_refcount_level = cpu_to_be32(afake->af_levels);
398 	agf->agf_refcount_blocks = cpu_to_be32(afake->af_blocks);
399 	xfs_alloc_log_agf(tp, agbp, XFS_AGF_REFCOUNT_BLOCKS |
400 				    XFS_AGF_REFCOUNT_ROOT |
401 				    XFS_AGF_REFCOUNT_LEVEL);
402 	xfs_btree_commit_afakeroot(cur, tp, agbp);
403 }
404 
405 /* Calculate number of records in a refcount btree block. */
406 static inline unsigned int
407 xfs_refcountbt_block_maxrecs(
408 	unsigned int		blocklen,
409 	bool			leaf)
410 {
411 	if (leaf)
412 		return blocklen / sizeof(struct xfs_refcount_rec);
413 	return blocklen / (sizeof(struct xfs_refcount_key) +
414 			   sizeof(xfs_refcount_ptr_t));
415 }
416 
417 /*
418  * Calculate the number of records in a refcount btree block.
419  */
420 int
421 xfs_refcountbt_maxrecs(
422 	int			blocklen,
423 	bool			leaf)
424 {
425 	blocklen -= XFS_REFCOUNT_BLOCK_LEN;
426 	return xfs_refcountbt_block_maxrecs(blocklen, leaf);
427 }
428 
429 /* Compute the max possible height of the maximally sized refcount btree. */
430 unsigned int
431 xfs_refcountbt_maxlevels_ondisk(void)
432 {
433 	unsigned int		minrecs[2];
434 	unsigned int		blocklen;
435 
436 	blocklen = XFS_MIN_CRC_BLOCKSIZE - XFS_BTREE_SBLOCK_CRC_LEN;
437 
438 	minrecs[0] = xfs_refcountbt_block_maxrecs(blocklen, true) / 2;
439 	minrecs[1] = xfs_refcountbt_block_maxrecs(blocklen, false) / 2;
440 
441 	return xfs_btree_compute_maxlevels(minrecs, XFS_MAX_CRC_AG_BLOCKS);
442 }
443 
444 /* Compute the maximum height of a refcount btree. */
445 void
446 xfs_refcountbt_compute_maxlevels(
447 	struct xfs_mount		*mp)
448 {
449 	if (!xfs_has_reflink(mp)) {
450 		mp->m_refc_maxlevels = 0;
451 		return;
452 	}
453 
454 	mp->m_refc_maxlevels = xfs_btree_compute_maxlevels(
455 			mp->m_refc_mnr, mp->m_sb.sb_agblocks);
456 	ASSERT(mp->m_refc_maxlevels <= xfs_refcountbt_maxlevels_ondisk());
457 }
458 
459 /* Calculate the refcount btree size for some records. */
460 xfs_extlen_t
461 xfs_refcountbt_calc_size(
462 	struct xfs_mount	*mp,
463 	unsigned long long	len)
464 {
465 	return xfs_btree_calc_size(mp->m_refc_mnr, len);
466 }
467 
468 /*
469  * Calculate the maximum refcount btree size.
470  */
471 xfs_extlen_t
472 xfs_refcountbt_max_size(
473 	struct xfs_mount	*mp,
474 	xfs_agblock_t		agblocks)
475 {
476 	/* Bail out if we're uninitialized, which can happen in mkfs. */
477 	if (mp->m_refc_mxr[0] == 0)
478 		return 0;
479 
480 	return xfs_refcountbt_calc_size(mp, agblocks);
481 }
482 
483 /*
484  * Figure out how many blocks to reserve and how many are used by this btree.
485  */
486 int
487 xfs_refcountbt_calc_reserves(
488 	struct xfs_mount	*mp,
489 	struct xfs_trans	*tp,
490 	struct xfs_perag	*pag,
491 	xfs_extlen_t		*ask,
492 	xfs_extlen_t		*used)
493 {
494 	struct xfs_buf		*agbp;
495 	struct xfs_agf		*agf;
496 	xfs_agblock_t		agblocks;
497 	xfs_extlen_t		tree_len;
498 	int			error;
499 
500 	if (!xfs_has_reflink(mp))
501 		return 0;
502 
503 	error = xfs_alloc_read_agf(pag, tp, 0, &agbp);
504 	if (error)
505 		return error;
506 
507 	agf = agbp->b_addr;
508 	agblocks = be32_to_cpu(agf->agf_length);
509 	tree_len = be32_to_cpu(agf->agf_refcount_blocks);
510 	xfs_trans_brelse(tp, agbp);
511 
512 	/*
513 	 * The log is permanently allocated, so the space it occupies will
514 	 * never be available for the kinds of things that would require btree
515 	 * expansion.  We therefore can pretend the space isn't there.
516 	 */
517 	if (xfs_ag_contains_log(mp, pag->pag_agno))
518 		agblocks -= mp->m_sb.sb_logblocks;
519 
520 	*ask += xfs_refcountbt_max_size(mp, agblocks);
521 	*used += tree_len;
522 
523 	return error;
524 }
525 
526 int __init
527 xfs_refcountbt_init_cur_cache(void)
528 {
529 	xfs_refcountbt_cur_cache = kmem_cache_create("xfs_refcbt_cur",
530 			xfs_btree_cur_sizeof(xfs_refcountbt_maxlevels_ondisk()),
531 			0, 0, NULL);
532 
533 	if (!xfs_refcountbt_cur_cache)
534 		return -ENOMEM;
535 	return 0;
536 }
537 
538 void
539 xfs_refcountbt_destroy_cur_cache(void)
540 {
541 	kmem_cache_destroy(xfs_refcountbt_cur_cache);
542 	xfs_refcountbt_cur_cache = NULL;
543 }
544