xref: /linux/fs/btrfs/ctree.h (revision 79997eda0d31bc68203c95ecb978773ee6ce7a1f)
1 /* SPDX-License-Identifier: GPL-2.0 */
2 /*
3  * Copyright (C) 2007 Oracle.  All rights reserved.
4  */
5 
6 #ifndef BTRFS_CTREE_H
7 #define BTRFS_CTREE_H
8 
9 #include <linux/pagemap.h>
10 #include "locking.h"
11 #include "fs.h"
12 #include "accessors.h"
13 
14 struct btrfs_trans_handle;
15 struct btrfs_transaction;
16 struct btrfs_pending_snapshot;
17 struct btrfs_delayed_ref_root;
18 struct btrfs_space_info;
19 struct btrfs_block_group;
20 struct btrfs_ordered_sum;
21 struct btrfs_ref;
22 struct btrfs_bio;
23 struct btrfs_ioctl_encoded_io_args;
24 struct btrfs_device;
25 struct btrfs_fs_devices;
26 struct btrfs_balance_control;
27 struct btrfs_delayed_root;
28 struct reloc_control;
29 
30 /* Read ahead values for struct btrfs_path.reada */
31 enum {
32 	READA_NONE,
33 	READA_BACK,
34 	READA_FORWARD,
35 	/*
36 	 * Similar to READA_FORWARD but unlike it:
37 	 *
38 	 * 1) It will trigger readahead even for leaves that are not close to
39 	 *    each other on disk;
40 	 * 2) It also triggers readahead for nodes;
41 	 * 3) During a search, even when a node or leaf is already in memory, it
42 	 *    will still trigger readahead for other nodes and leaves that follow
43 	 *    it.
44 	 *
45 	 * This is meant to be used only when we know we are iterating over the
46 	 * entire tree or a very large part of it.
47 	 */
48 	READA_FORWARD_ALWAYS,
49 };
50 
51 /*
52  * btrfs_paths remember the path taken from the root down to the leaf.
53  * level 0 is always the leaf, and nodes[1...BTRFS_MAX_LEVEL] will point
54  * to any other levels that are present.
55  *
56  * The slots array records the index of the item or block pointer
57  * used while walking the tree.
58  */
59 struct btrfs_path {
60 	struct extent_buffer *nodes[BTRFS_MAX_LEVEL];
61 	int slots[BTRFS_MAX_LEVEL];
62 	/* if there is real range locking, this locks field will change */
63 	u8 locks[BTRFS_MAX_LEVEL];
64 	u8 reada;
65 	/* keep some upper locks as we walk down */
66 	u8 lowest_level;
67 
68 	/*
69 	 * set by btrfs_split_item, tells search_slot to keep all locks
70 	 * and to force calls to keep space in the nodes
71 	 */
72 	unsigned int search_for_split:1;
73 	unsigned int keep_locks:1;
74 	unsigned int skip_locking:1;
75 	unsigned int search_commit_root:1;
76 	unsigned int need_commit_sem:1;
77 	unsigned int skip_release_on_error:1;
78 	/*
79 	 * Indicate that new item (btrfs_search_slot) is extending already
80 	 * existing item and ins_len contains only the data size and not item
81 	 * header (ie. sizeof(struct btrfs_item) is not included).
82 	 */
83 	unsigned int search_for_extension:1;
84 	/* Stop search if any locks need to be taken (for read) */
85 	unsigned int nowait:1;
86 };
87 
88 /*
89  * The state of btrfs root
90  */
91 enum {
92 	/*
93 	 * btrfs_record_root_in_trans is a multi-step process, and it can race
94 	 * with the balancing code.   But the race is very small, and only the
95 	 * first time the root is added to each transaction.  So IN_TRANS_SETUP
96 	 * is used to tell us when more checks are required
97 	 */
98 	BTRFS_ROOT_IN_TRANS_SETUP,
99 
100 	/*
101 	 * Set if tree blocks of this root can be shared by other roots.
102 	 * Only subvolume trees and their reloc trees have this bit set.
103 	 * Conflicts with TRACK_DIRTY bit.
104 	 *
105 	 * This affects two things:
106 	 *
107 	 * - How balance works
108 	 *   For shareable roots, we need to use reloc tree and do path
109 	 *   replacement for balance, and need various pre/post hooks for
110 	 *   snapshot creation to handle them.
111 	 *
112 	 *   While for non-shareable trees, we just simply do a tree search
113 	 *   with COW.
114 	 *
115 	 * - How dirty roots are tracked
116 	 *   For shareable roots, btrfs_record_root_in_trans() is needed to
117 	 *   track them, while non-subvolume roots have TRACK_DIRTY bit, they
118 	 *   don't need to set this manually.
119 	 */
120 	BTRFS_ROOT_SHAREABLE,
121 	BTRFS_ROOT_TRACK_DIRTY,
122 	BTRFS_ROOT_IN_RADIX,
123 	BTRFS_ROOT_ORPHAN_ITEM_INSERTED,
124 	BTRFS_ROOT_DEFRAG_RUNNING,
125 	BTRFS_ROOT_FORCE_COW,
126 	BTRFS_ROOT_MULTI_LOG_TASKS,
127 	BTRFS_ROOT_DIRTY,
128 	BTRFS_ROOT_DELETING,
129 
130 	/*
131 	 * Reloc tree is orphan, only kept here for qgroup delayed subtree scan
132 	 *
133 	 * Set for the subvolume tree owning the reloc tree.
134 	 */
135 	BTRFS_ROOT_DEAD_RELOC_TREE,
136 	/* Mark dead root stored on device whose cleanup needs to be resumed */
137 	BTRFS_ROOT_DEAD_TREE,
138 	/* The root has a log tree. Used for subvolume roots and the tree root. */
139 	BTRFS_ROOT_HAS_LOG_TREE,
140 	/* Qgroup flushing is in progress */
141 	BTRFS_ROOT_QGROUP_FLUSHING,
142 	/* We started the orphan cleanup for this root. */
143 	BTRFS_ROOT_ORPHAN_CLEANUP,
144 	/* This root has a drop operation that was started previously. */
145 	BTRFS_ROOT_UNFINISHED_DROP,
146 	/* This reloc root needs to have its buffers lockdep class reset. */
147 	BTRFS_ROOT_RESET_LOCKDEP_CLASS,
148 };
149 
150 /*
151  * Record swapped tree blocks of a subvolume tree for delayed subtree trace
152  * code. For detail check comment in fs/btrfs/qgroup.c.
153  */
154 struct btrfs_qgroup_swapped_blocks {
155 	spinlock_t lock;
156 	/* RM_EMPTY_ROOT() of above blocks[] */
157 	bool swapped;
158 	struct rb_root blocks[BTRFS_MAX_LEVEL];
159 };
160 
161 /*
162  * in ram representation of the tree.  extent_root is used for all allocations
163  * and for the extent tree extent_root root.
164  */
165 struct btrfs_root {
166 	struct rb_node rb_node;
167 
168 	struct extent_buffer *node;
169 
170 	struct extent_buffer *commit_root;
171 	struct btrfs_root *log_root;
172 	struct btrfs_root *reloc_root;
173 
174 	unsigned long state;
175 	struct btrfs_root_item root_item;
176 	struct btrfs_key root_key;
177 	struct btrfs_fs_info *fs_info;
178 	struct extent_io_tree dirty_log_pages;
179 
180 	struct mutex objectid_mutex;
181 
182 	spinlock_t accounting_lock;
183 	struct btrfs_block_rsv *block_rsv;
184 
185 	struct mutex log_mutex;
186 	wait_queue_head_t log_writer_wait;
187 	wait_queue_head_t log_commit_wait[2];
188 	struct list_head log_ctxs[2];
189 	/* Used only for log trees of subvolumes, not for the log root tree */
190 	atomic_t log_writers;
191 	atomic_t log_commit[2];
192 	/* Used only for log trees of subvolumes, not for the log root tree */
193 	atomic_t log_batch;
194 	/*
195 	 * Protected by the 'log_mutex' lock but can be read without holding
196 	 * that lock to avoid unnecessary lock contention, in which case it
197 	 * should be read using btrfs_get_root_log_transid() except if it's a
198 	 * log tree in which case it can be directly accessed. Updates to this
199 	 * field should always use btrfs_set_root_log_transid(), except for log
200 	 * trees where the field can be updated directly.
201 	 */
202 	int log_transid;
203 	/* No matter the commit succeeds or not*/
204 	int log_transid_committed;
205 	/*
206 	 * Just be updated when the commit succeeds. Use
207 	 * btrfs_get_root_last_log_commit() and btrfs_set_root_last_log_commit()
208 	 * to access this field.
209 	 */
210 	int last_log_commit;
211 	pid_t log_start_pid;
212 
213 	u64 last_trans;
214 
215 	u32 type;
216 
217 	u64 free_objectid;
218 
219 	struct btrfs_key defrag_progress;
220 	struct btrfs_key defrag_max;
221 
222 	/* The dirty list is only used by non-shareable roots */
223 	struct list_head dirty_list;
224 
225 	struct list_head root_list;
226 
227 	spinlock_t log_extents_lock[2];
228 	struct list_head logged_list[2];
229 
230 	spinlock_t inode_lock;
231 	/* red-black tree that keeps track of in-memory inodes */
232 	struct rb_root inode_tree;
233 
234 	/*
235 	 * radix tree that keeps track of delayed nodes of every inode,
236 	 * protected by inode_lock
237 	 */
238 	struct radix_tree_root delayed_nodes_tree;
239 	/*
240 	 * right now this just gets used so that a root has its own devid
241 	 * for stat.  It may be used for more later
242 	 */
243 	dev_t anon_dev;
244 
245 	spinlock_t root_item_lock;
246 	refcount_t refs;
247 
248 	struct mutex delalloc_mutex;
249 	spinlock_t delalloc_lock;
250 	/*
251 	 * all of the inodes that have delalloc bytes.  It is possible for
252 	 * this list to be empty even when there is still dirty data=ordered
253 	 * extents waiting to finish IO.
254 	 */
255 	struct list_head delalloc_inodes;
256 	struct list_head delalloc_root;
257 	u64 nr_delalloc_inodes;
258 
259 	struct mutex ordered_extent_mutex;
260 	/*
261 	 * this is used by the balancing code to wait for all the pending
262 	 * ordered extents
263 	 */
264 	spinlock_t ordered_extent_lock;
265 
266 	/*
267 	 * all of the data=ordered extents pending writeback
268 	 * these can span multiple transactions and basically include
269 	 * every dirty data page that isn't from nodatacow
270 	 */
271 	struct list_head ordered_extents;
272 	struct list_head ordered_root;
273 	u64 nr_ordered_extents;
274 
275 	/*
276 	 * Not empty if this subvolume root has gone through tree block swap
277 	 * (relocation)
278 	 *
279 	 * Will be used by reloc_control::dirty_subvol_roots.
280 	 */
281 	struct list_head reloc_dirty_list;
282 
283 	/*
284 	 * Number of currently running SEND ioctls to prevent
285 	 * manipulation with the read-only status via SUBVOL_SETFLAGS
286 	 */
287 	int send_in_progress;
288 	/*
289 	 * Number of currently running deduplication operations that have a
290 	 * destination inode belonging to this root. Protected by the lock
291 	 * root_item_lock.
292 	 */
293 	int dedupe_in_progress;
294 	/* For exclusion of snapshot creation and nocow writes */
295 	struct btrfs_drew_lock snapshot_lock;
296 
297 	atomic_t snapshot_force_cow;
298 
299 	/* For qgroup metadata reserved space */
300 	spinlock_t qgroup_meta_rsv_lock;
301 	u64 qgroup_meta_rsv_pertrans;
302 	u64 qgroup_meta_rsv_prealloc;
303 	wait_queue_head_t qgroup_flush_wait;
304 
305 	/* Number of active swapfiles */
306 	atomic_t nr_swapfiles;
307 
308 	/* Record pairs of swapped blocks for qgroup */
309 	struct btrfs_qgroup_swapped_blocks swapped_blocks;
310 
311 	/* Used only by log trees, when logging csum items */
312 	struct extent_io_tree log_csum_range;
313 
314 	/* Used in simple quotas, track root during relocation. */
315 	u64 relocation_src_root;
316 
317 #ifdef CONFIG_BTRFS_FS_RUN_SANITY_TESTS
318 	u64 alloc_bytenr;
319 #endif
320 
321 #ifdef CONFIG_BTRFS_DEBUG
322 	struct list_head leak_list;
323 #endif
324 };
325 
326 static inline bool btrfs_root_readonly(const struct btrfs_root *root)
327 {
328 	/* Byte-swap the constant at compile time, root_item::flags is LE */
329 	return (root->root_item.flags & cpu_to_le64(BTRFS_ROOT_SUBVOL_RDONLY)) != 0;
330 }
331 
332 static inline bool btrfs_root_dead(const struct btrfs_root *root)
333 {
334 	/* Byte-swap the constant at compile time, root_item::flags is LE */
335 	return (root->root_item.flags & cpu_to_le64(BTRFS_ROOT_SUBVOL_DEAD)) != 0;
336 }
337 
338 static inline u64 btrfs_root_id(const struct btrfs_root *root)
339 {
340 	return root->root_key.objectid;
341 }
342 
343 static inline int btrfs_get_root_log_transid(const struct btrfs_root *root)
344 {
345 	return READ_ONCE(root->log_transid);
346 }
347 
348 static inline void btrfs_set_root_log_transid(struct btrfs_root *root, int log_transid)
349 {
350 	WRITE_ONCE(root->log_transid, log_transid);
351 }
352 
353 static inline int btrfs_get_root_last_log_commit(const struct btrfs_root *root)
354 {
355 	return READ_ONCE(root->last_log_commit);
356 }
357 
358 static inline void btrfs_set_root_last_log_commit(struct btrfs_root *root, int commit_id)
359 {
360 	WRITE_ONCE(root->last_log_commit, commit_id);
361 }
362 
363 /*
364  * Structure that conveys information about an extent that is going to replace
365  * all the extents in a file range.
366  */
367 struct btrfs_replace_extent_info {
368 	u64 disk_offset;
369 	u64 disk_len;
370 	u64 data_offset;
371 	u64 data_len;
372 	u64 file_offset;
373 	/* Pointer to a file extent item of type regular or prealloc. */
374 	char *extent_buf;
375 	/*
376 	 * Set to true when attempting to replace a file range with a new extent
377 	 * described by this structure, set to false when attempting to clone an
378 	 * existing extent into a file range.
379 	 */
380 	bool is_new_extent;
381 	/* Indicate if we should update the inode's mtime and ctime. */
382 	bool update_times;
383 	/* Meaningful only if is_new_extent is true. */
384 	int qgroup_reserved;
385 	/*
386 	 * Meaningful only if is_new_extent is true.
387 	 * Used to track how many extent items we have already inserted in a
388 	 * subvolume tree that refer to the extent described by this structure,
389 	 * so that we know when to create a new delayed ref or update an existing
390 	 * one.
391 	 */
392 	int insertions;
393 };
394 
395 /* Arguments for btrfs_drop_extents() */
396 struct btrfs_drop_extents_args {
397 	/* Input parameters */
398 
399 	/*
400 	 * If NULL, btrfs_drop_extents() will allocate and free its own path.
401 	 * If 'replace_extent' is true, this must not be NULL. Also the path
402 	 * is always released except if 'replace_extent' is true and
403 	 * btrfs_drop_extents() sets 'extent_inserted' to true, in which case
404 	 * the path is kept locked.
405 	 */
406 	struct btrfs_path *path;
407 	/* Start offset of the range to drop extents from */
408 	u64 start;
409 	/* End (exclusive, last byte + 1) of the range to drop extents from */
410 	u64 end;
411 	/* If true drop all the extent maps in the range */
412 	bool drop_cache;
413 	/*
414 	 * If true it means we want to insert a new extent after dropping all
415 	 * the extents in the range. If this is true, the 'extent_item_size'
416 	 * parameter must be set as well and the 'extent_inserted' field will
417 	 * be set to true by btrfs_drop_extents() if it could insert the new
418 	 * extent.
419 	 * Note: when this is set to true the path must not be NULL.
420 	 */
421 	bool replace_extent;
422 	/*
423 	 * Used if 'replace_extent' is true. Size of the file extent item to
424 	 * insert after dropping all existing extents in the range
425 	 */
426 	u32 extent_item_size;
427 
428 	/* Output parameters */
429 
430 	/*
431 	 * Set to the minimum between the input parameter 'end' and the end
432 	 * (exclusive, last byte + 1) of the last dropped extent. This is always
433 	 * set even if btrfs_drop_extents() returns an error.
434 	 */
435 	u64 drop_end;
436 	/*
437 	 * The number of allocated bytes found in the range. This can be smaller
438 	 * than the range's length when there are holes in the range.
439 	 */
440 	u64 bytes_found;
441 	/*
442 	 * Only set if 'replace_extent' is true. Set to true if we were able
443 	 * to insert a replacement extent after dropping all extents in the
444 	 * range, otherwise set to false by btrfs_drop_extents().
445 	 * Also, if btrfs_drop_extents() has set this to true it means it
446 	 * returned with the path locked, otherwise if it has set this to
447 	 * false it has returned with the path released.
448 	 */
449 	bool extent_inserted;
450 };
451 
452 struct btrfs_file_private {
453 	void *filldir_buf;
454 	u64 last_index;
455 	struct extent_state *llseek_cached_state;
456 };
457 
458 static inline u32 BTRFS_LEAF_DATA_SIZE(const struct btrfs_fs_info *info)
459 {
460 	return info->nodesize - sizeof(struct btrfs_header);
461 }
462 
463 static inline u32 BTRFS_MAX_ITEM_SIZE(const struct btrfs_fs_info *info)
464 {
465 	return BTRFS_LEAF_DATA_SIZE(info) - sizeof(struct btrfs_item);
466 }
467 
468 static inline u32 BTRFS_NODEPTRS_PER_BLOCK(const struct btrfs_fs_info *info)
469 {
470 	return BTRFS_LEAF_DATA_SIZE(info) / sizeof(struct btrfs_key_ptr);
471 }
472 
473 static inline u32 BTRFS_MAX_XATTR_SIZE(const struct btrfs_fs_info *info)
474 {
475 	return BTRFS_MAX_ITEM_SIZE(info) - sizeof(struct btrfs_dir_item);
476 }
477 
478 #define BTRFS_BYTES_TO_BLKS(fs_info, bytes) \
479 				((bytes) >> (fs_info)->sectorsize_bits)
480 
481 static inline gfp_t btrfs_alloc_write_mask(struct address_space *mapping)
482 {
483 	return mapping_gfp_constraint(mapping, ~__GFP_FS);
484 }
485 
486 int btrfs_error_unpin_extent_range(struct btrfs_fs_info *fs_info,
487 				   u64 start, u64 end);
488 int btrfs_discard_extent(struct btrfs_fs_info *fs_info, u64 bytenr,
489 			 u64 num_bytes, u64 *actual_bytes);
490 int btrfs_trim_fs(struct btrfs_fs_info *fs_info, struct fstrim_range *range);
491 
492 /* ctree.c */
493 int __init btrfs_ctree_init(void);
494 void __cold btrfs_ctree_exit(void);
495 
496 int btrfs_bin_search(struct extent_buffer *eb, int first_slot,
497 		     const struct btrfs_key *key, int *slot);
498 
499 int __pure btrfs_comp_cpu_keys(const struct btrfs_key *k1, const struct btrfs_key *k2);
500 
501 #ifdef __LITTLE_ENDIAN
502 
503 /*
504  * Compare two keys, on little-endian the disk order is same as CPU order and
505  * we can avoid the conversion.
506  */
507 static inline int btrfs_comp_keys(const struct btrfs_disk_key *disk_key,
508 				  const struct btrfs_key *k2)
509 {
510 	const struct btrfs_key *k1 = (const struct btrfs_key *)disk_key;
511 
512 	return btrfs_comp_cpu_keys(k1, k2);
513 }
514 
515 #else
516 
517 /* Compare two keys in a memcmp fashion. */
518 static inline int btrfs_comp_keys(const struct btrfs_disk_key *disk,
519 				  const struct btrfs_key *k2)
520 {
521 	struct btrfs_key k1;
522 
523 	btrfs_disk_key_to_cpu(&k1, disk);
524 
525 	return btrfs_comp_cpu_keys(&k1, k2);
526 }
527 
528 #endif
529 
530 int btrfs_previous_item(struct btrfs_root *root,
531 			struct btrfs_path *path, u64 min_objectid,
532 			int type);
533 int btrfs_previous_extent_item(struct btrfs_root *root,
534 			struct btrfs_path *path, u64 min_objectid);
535 void btrfs_set_item_key_safe(struct btrfs_trans_handle *trans,
536 			     struct btrfs_path *path,
537 			     const struct btrfs_key *new_key);
538 struct extent_buffer *btrfs_root_node(struct btrfs_root *root);
539 int btrfs_find_next_key(struct btrfs_root *root, struct btrfs_path *path,
540 			struct btrfs_key *key, int lowest_level,
541 			u64 min_trans);
542 int btrfs_search_forward(struct btrfs_root *root, struct btrfs_key *min_key,
543 			 struct btrfs_path *path,
544 			 u64 min_trans);
545 struct extent_buffer *btrfs_read_node_slot(struct extent_buffer *parent,
546 					   int slot);
547 
548 int btrfs_cow_block(struct btrfs_trans_handle *trans,
549 		    struct btrfs_root *root, struct extent_buffer *buf,
550 		    struct extent_buffer *parent, int parent_slot,
551 		    struct extent_buffer **cow_ret,
552 		    enum btrfs_lock_nesting nest);
553 int btrfs_force_cow_block(struct btrfs_trans_handle *trans,
554 			  struct btrfs_root *root,
555 			  struct extent_buffer *buf,
556 			  struct extent_buffer *parent, int parent_slot,
557 			  struct extent_buffer **cow_ret,
558 			  u64 search_start, u64 empty_size,
559 			  enum btrfs_lock_nesting nest);
560 int btrfs_copy_root(struct btrfs_trans_handle *trans,
561 		      struct btrfs_root *root,
562 		      struct extent_buffer *buf,
563 		      struct extent_buffer **cow_ret, u64 new_root_objectid);
564 int btrfs_block_can_be_shared(struct btrfs_trans_handle *trans,
565 			      struct btrfs_root *root,
566 			      struct extent_buffer *buf);
567 int btrfs_del_ptr(struct btrfs_trans_handle *trans, struct btrfs_root *root,
568 		  struct btrfs_path *path, int level, int slot);
569 void btrfs_extend_item(struct btrfs_trans_handle *trans,
570 		       struct btrfs_path *path, u32 data_size);
571 void btrfs_truncate_item(struct btrfs_trans_handle *trans,
572 			 struct btrfs_path *path, u32 new_size, int from_end);
573 int btrfs_split_item(struct btrfs_trans_handle *trans,
574 		     struct btrfs_root *root,
575 		     struct btrfs_path *path,
576 		     const struct btrfs_key *new_key,
577 		     unsigned long split_offset);
578 int btrfs_duplicate_item(struct btrfs_trans_handle *trans,
579 			 struct btrfs_root *root,
580 			 struct btrfs_path *path,
581 			 const struct btrfs_key *new_key);
582 int btrfs_find_item(struct btrfs_root *fs_root, struct btrfs_path *path,
583 		u64 inum, u64 ioff, u8 key_type, struct btrfs_key *found_key);
584 int btrfs_search_slot(struct btrfs_trans_handle *trans, struct btrfs_root *root,
585 		      const struct btrfs_key *key, struct btrfs_path *p,
586 		      int ins_len, int cow);
587 int btrfs_search_old_slot(struct btrfs_root *root, const struct btrfs_key *key,
588 			  struct btrfs_path *p, u64 time_seq);
589 int btrfs_search_slot_for_read(struct btrfs_root *root,
590 			       const struct btrfs_key *key,
591 			       struct btrfs_path *p, int find_higher,
592 			       int return_any);
593 void btrfs_release_path(struct btrfs_path *p);
594 struct btrfs_path *btrfs_alloc_path(void);
595 void btrfs_free_path(struct btrfs_path *p);
596 
597 int btrfs_del_items(struct btrfs_trans_handle *trans, struct btrfs_root *root,
598 		   struct btrfs_path *path, int slot, int nr);
599 static inline int btrfs_del_item(struct btrfs_trans_handle *trans,
600 				 struct btrfs_root *root,
601 				 struct btrfs_path *path)
602 {
603 	return btrfs_del_items(trans, root, path, path->slots[0], 1);
604 }
605 
606 /*
607  * Describes a batch of items to insert in a btree. This is used by
608  * btrfs_insert_empty_items().
609  */
610 struct btrfs_item_batch {
611 	/*
612 	 * Pointer to an array containing the keys of the items to insert (in
613 	 * sorted order).
614 	 */
615 	const struct btrfs_key *keys;
616 	/* Pointer to an array containing the data size for each item to insert. */
617 	const u32 *data_sizes;
618 	/*
619 	 * The sum of data sizes for all items. The caller can compute this while
620 	 * setting up the data_sizes array, so it ends up being more efficient
621 	 * than having btrfs_insert_empty_items() or setup_item_for_insert()
622 	 * doing it, as it would avoid an extra loop over a potentially large
623 	 * array, and in the case of setup_item_for_insert(), we would be doing
624 	 * it while holding a write lock on a leaf and often on upper level nodes
625 	 * too, unnecessarily increasing the size of a critical section.
626 	 */
627 	u32 total_data_size;
628 	/* Size of the keys and data_sizes arrays (number of items in the batch). */
629 	int nr;
630 };
631 
632 void btrfs_setup_item_for_insert(struct btrfs_trans_handle *trans,
633 				 struct btrfs_root *root,
634 				 struct btrfs_path *path,
635 				 const struct btrfs_key *key,
636 				 u32 data_size);
637 int btrfs_insert_item(struct btrfs_trans_handle *trans, struct btrfs_root *root,
638 		      const struct btrfs_key *key, void *data, u32 data_size);
639 int btrfs_insert_empty_items(struct btrfs_trans_handle *trans,
640 			     struct btrfs_root *root,
641 			     struct btrfs_path *path,
642 			     const struct btrfs_item_batch *batch);
643 
644 static inline int btrfs_insert_empty_item(struct btrfs_trans_handle *trans,
645 					  struct btrfs_root *root,
646 					  struct btrfs_path *path,
647 					  const struct btrfs_key *key,
648 					  u32 data_size)
649 {
650 	struct btrfs_item_batch batch;
651 
652 	batch.keys = key;
653 	batch.data_sizes = &data_size;
654 	batch.total_data_size = data_size;
655 	batch.nr = 1;
656 
657 	return btrfs_insert_empty_items(trans, root, path, &batch);
658 }
659 
660 int btrfs_next_old_leaf(struct btrfs_root *root, struct btrfs_path *path,
661 			u64 time_seq);
662 
663 int btrfs_search_backwards(struct btrfs_root *root, struct btrfs_key *key,
664 			   struct btrfs_path *path);
665 
666 int btrfs_get_next_valid_item(struct btrfs_root *root, struct btrfs_key *key,
667 			      struct btrfs_path *path);
668 
669 /*
670  * Search in @root for a given @key, and store the slot found in @found_key.
671  *
672  * @root:	The root node of the tree.
673  * @key:	The key we are looking for.
674  * @found_key:	Will hold the found item.
675  * @path:	Holds the current slot/leaf.
676  * @iter_ret:	Contains the value returned from btrfs_search_slot or
677  * 		btrfs_get_next_valid_item, whichever was executed last.
678  *
679  * The @iter_ret is an output variable that will contain the return value of
680  * btrfs_search_slot, if it encountered an error, or the value returned from
681  * btrfs_get_next_valid_item otherwise. That return value can be 0, if a valid
682  * slot was found, 1 if there were no more leaves, and <0 if there was an error.
683  *
684  * It's recommended to use a separate variable for iter_ret and then use it to
685  * set the function return value so there's no confusion of the 0/1/errno
686  * values stemming from btrfs_search_slot.
687  */
688 #define btrfs_for_each_slot(root, key, found_key, path, iter_ret)		\
689 	for (iter_ret = btrfs_search_slot(NULL, (root), (key), (path), 0, 0);	\
690 		(iter_ret) >= 0 &&						\
691 		(iter_ret = btrfs_get_next_valid_item((root), (found_key), (path))) == 0; \
692 		(path)->slots[0]++						\
693 	)
694 
695 int btrfs_next_old_item(struct btrfs_root *root, struct btrfs_path *path, u64 time_seq);
696 
697 /*
698  * Search the tree again to find a leaf with greater keys.
699  *
700  * Returns 0 if it found something or 1 if there are no greater leaves.
701  * Returns < 0 on error.
702  */
703 static inline int btrfs_next_leaf(struct btrfs_root *root, struct btrfs_path *path)
704 {
705 	return btrfs_next_old_leaf(root, path, 0);
706 }
707 
708 static inline int btrfs_next_item(struct btrfs_root *root, struct btrfs_path *p)
709 {
710 	return btrfs_next_old_item(root, p, 0);
711 }
712 int btrfs_leaf_free_space(const struct extent_buffer *leaf);
713 
714 static inline int is_fstree(u64 rootid)
715 {
716 	if (rootid == BTRFS_FS_TREE_OBJECTID ||
717 	    ((s64)rootid >= (s64)BTRFS_FIRST_FREE_OBJECTID &&
718 	      !btrfs_qgroup_level(rootid)))
719 		return 1;
720 	return 0;
721 }
722 
723 static inline bool btrfs_is_data_reloc_root(const struct btrfs_root *root)
724 {
725 	return root->root_key.objectid == BTRFS_DATA_RELOC_TREE_OBJECTID;
726 }
727 
728 u16 btrfs_csum_type_size(u16 type);
729 int btrfs_super_csum_size(const struct btrfs_super_block *s);
730 const char *btrfs_super_csum_name(u16 csum_type);
731 const char *btrfs_super_csum_driver(u16 csum_type);
732 size_t __attribute_const__ btrfs_get_num_csums(void);
733 
734 /*
735  * We use page status Private2 to indicate there is an ordered extent with
736  * unfinished IO.
737  *
738  * Rename the Private2 accessors to Ordered, to improve readability.
739  */
740 #define PageOrdered(page)		PagePrivate2(page)
741 #define SetPageOrdered(page)		SetPagePrivate2(page)
742 #define ClearPageOrdered(page)		ClearPagePrivate2(page)
743 #define folio_test_ordered(folio)	folio_test_private_2(folio)
744 #define folio_set_ordered(folio)	folio_set_private_2(folio)
745 #define folio_clear_ordered(folio)	folio_clear_private_2(folio)
746 
747 #endif
748