1	// SPDX-License-Identifier: GPL-2.0
2	/*
3	* Copyright (C) 2009 Oracle. All rights reserved.
4	*/
5
6	#include <linux/sched.h>
7	#include <linux/pagemap.h>
8	#include <linux/writeback.h>
9	#include <linux/blkdev.h>
10	#include <linux/rbtree.h>
11	#include <linux/slab.h>
12	#include <linux/error-injection.h>
13	#include "ctree.h"
14	#include "disk-io.h"
15	#include "transaction.h"
16	#include "volumes.h"
17	#include "locking.h"
18	#include "btrfs_inode.h"
19	#include "async-thread.h"
20	#include "free-space-cache.h"
21	#include "qgroup.h"
22	#include "print-tree.h"
23	#include "delalloc-space.h"
24	#include "block-group.h"
25	#include "backref.h"
26	#include "misc.h"
27	#include "subpage.h"
28	#include "zoned.h"
29	#include "inode-item.h"
30	#include "space-info.h"
31	#include "fs.h"
32	#include "accessors.h"
33	#include "extent-tree.h"
34	#include "root-tree.h"
35	#include "file-item.h"
36	#include "relocation.h"
37	#include "super.h"
38	#include "tree-checker.h"
39
40	/*
41	* Relocation overview
42	*
43	* [What does relocation do]
44	*
45	* The objective of relocation is to relocate all extents of the target block
46	* group to other block groups.
47	* This is utilized by resize (shrink only), profile converting, compacting
48	* space, or balance routine to spread chunks over devices.
49	*
50	* Before | After
51	* ------------------------------------------------------------------
52	* BG A: 10 data extents | BG A: deleted
53	* BG B: 2 data extents | BG B: 10 data extents (2 old + 8 relocated)
54	* BG C: 1 extents | BG C: 3 data extents (1 old + 2 relocated)
55	*
56	* [How does relocation work]
57	*
58	* 1. Mark the target block group read-only
59	* New extents won't be allocated from the target block group.
60	*
61	* 2.1 Record each extent in the target block group
62	* To build a proper map of extents to be relocated.
63	*
64	* 2.2 Build data reloc tree and reloc trees
65	* Data reloc tree will contain an inode, recording all newly relocated
66	* data extents.
67	* There will be only one data reloc tree for one data block group.
68	*
69	* Reloc tree will be a special snapshot of its source tree, containing
70	* relocated tree blocks.
71	* Each tree referring to a tree block in target block group will get its
72	* reloc tree built.
73	*
74	* 2.3 Swap source tree with its corresponding reloc tree
75	* Each involved tree only refers to new extents after swap.
76	*
77	* 3. Cleanup reloc trees and data reloc tree.
78	* As old extents in the target block group are still referenced by reloc
79	* trees, we need to clean them up before really freeing the target block
80	* group.
81	*
82	* The main complexity is in steps 2.2 and 2.3.
83	*
84	* The entry point of relocation is relocate_block_group() function.
85	*/
86
87	#define RELOCATION_RESERVED_NODES 256
88	/*
89	* map address of tree root to tree
90	*/
91	struct mapping_node {
92	struct {
93	struct rb_node rb_node;
94	u64 bytenr;
95	}; /* Use rb_simle_node for search/insert */
96	void *data;
97	};
98
99	struct mapping_tree {
100	struct rb_root rb_root;
101	spinlock_t lock;
102	};
103
104	/*
105	* present a tree block to process
106	*/
107	struct tree_block {
108	struct {
109	struct rb_node rb_node;
110	u64 bytenr;
111	}; /* Use rb_simple_node for search/insert */
112	u64 owner;
113	struct btrfs_key key;
114	u8 level;
115	bool key_ready;
116	};
117
118	#define MAX_EXTENTS 128
119
120	struct file_extent_cluster {
121	u64 start;
122	u64 end;
123	u64 boundary[MAX_EXTENTS];
124	unsigned int nr;
125	u64 owning_root;
126	};
127
128	/* Stages of data relocation. */
129	enum reloc_stage {
130	MOVE_DATA_EXTENTS,
131	UPDATE_DATA_PTRS
132	};
133
134	struct reloc_control {
135	/* block group to relocate */
136	struct btrfs_block_group *block_group;
137	/* extent tree */
138	struct btrfs_root *extent_root;
139	/* inode for moving data */
140	struct inode *data_inode;
141
142	struct btrfs_block_rsv *block_rsv;
143
144	struct btrfs_backref_cache backref_cache;
145
146	struct file_extent_cluster cluster;
147	/* tree blocks have been processed */
148	struct extent_io_tree processed_blocks;
149	/* map start of tree root to corresponding reloc tree */
150	struct mapping_tree reloc_root_tree;
151	/* list of reloc trees */
152	struct list_head reloc_roots;
153	/* list of subvolume trees that get relocated */
154	struct list_head dirty_subvol_roots;
155	/* size of metadata reservation for merging reloc trees */
156	u64 merging_rsv_size;
157	/* size of relocated tree nodes */
158	u64 nodes_relocated;
159	/* reserved size for block group relocation*/
160	u64 reserved_bytes;
161
162	u64 search_start;
163	u64 extents_found;
164
165	enum reloc_stage stage;
166	bool create_reloc_tree;
167	bool merge_reloc_tree;
168	bool found_file_extent;
169	};
170
171	static void mark_block_processed(struct reloc_control *rc,
172	struct btrfs_backref_node *node)
173	{
174	u32 blocksize;
175
176	if (node->level == 0 ||
177	in_range(node->bytenr, rc->block_group->start,
178	rc->block_group->length)) {
179	blocksize = rc->extent_root->fs_info->nodesize;
180	set_extent_bit(tree: &rc->processed_blocks, start: node->bytenr,
181	end: node->bytenr + blocksize - 1, bits: EXTENT_DIRTY, NULL);
182	}
183	node->processed = 1;
184	}
185
186	/*
187	* walk up backref nodes until reach node presents tree root
188	*/
189	static struct btrfs_backref_node *walk_up_backref(
190	struct btrfs_backref_node *node,
191	struct btrfs_backref_edge *edges[], int *index)
192	{
193	struct btrfs_backref_edge *edge;
194	int idx = *index;
195
196	while (!list_empty(head: &node->upper)) {
197	edge = list_entry(node->upper.next,
198	struct btrfs_backref_edge, list[LOWER]);
199	edges[idx++] = edge;
200	node = edge->node[UPPER];
201	}
202	BUG_ON(node->detached);
203	*index = idx;
204	return node;
205	}
206
207	/*
208	* walk down backref nodes to find start of next reference path
209	*/
210	static struct btrfs_backref_node *walk_down_backref(
211	struct btrfs_backref_edge *edges[], int *index)
212	{
213	struct btrfs_backref_edge *edge;
214	struct btrfs_backref_node *lower;
215	int idx = *index;
216
217	while (idx > 0) {
218	edge = edges[idx - 1];
219	lower = edge->node[LOWER];
220	if (list_is_last(list: &edge->list[LOWER], head: &lower->upper)) {
221	idx--;
222	continue;
223	}
224	edge = list_entry(edge->list[LOWER].next,
225	struct btrfs_backref_edge, list[LOWER]);
226	edges[idx - 1] = edge;
227	*index = idx;
228	return edge->node[UPPER];
229	}
230	*index = 0;
231	return NULL;
232	}
233
234	static void update_backref_node(struct btrfs_backref_cache *cache,
235	struct btrfs_backref_node *node, u64 bytenr)
236	{
237	struct rb_node *rb_node;
238	rb_erase(&node->rb_node, &cache->rb_root);
239	node->bytenr = bytenr;
240	rb_node = rb_simple_insert(root: &cache->rb_root, bytenr: node->bytenr, node: &node->rb_node);
241	if (rb_node)
242	btrfs_backref_panic(fs_info: cache->fs_info, bytenr, error: -EEXIST);
243	}
244
245	/*
246	* update backref cache after a transaction commit
247	*/
248	static int update_backref_cache(struct btrfs_trans_handle *trans,
249	struct btrfs_backref_cache *cache)
250	{
251	struct btrfs_backref_node *node;
252	int level = 0;
253
254	if (cache->last_trans == 0) {
255	cache->last_trans = trans->transid;
256	return 0;
257	}
258
259	if (cache->last_trans == trans->transid)
260	return 0;
261
262	/*
263	* detached nodes are used to avoid unnecessary backref
264	* lookup. transaction commit changes the extent tree.
265	* so the detached nodes are no longer useful.
266	*/
267	while (!list_empty(head: &cache->detached)) {
268	node = list_entry(cache->detached.next,
269	struct btrfs_backref_node, list);
270	btrfs_backref_cleanup_node(cache, node);
271	}
272
273	while (!list_empty(head: &cache->changed)) {
274	node = list_entry(cache->changed.next,
275	struct btrfs_backref_node, list);
276	list_del_init(entry: &node->list);
277	BUG_ON(node->pending);
278	update_backref_node(cache, node, bytenr: node->new_bytenr);
279	}
280
281	/*
282	* some nodes can be left in the pending list if there were
283	* errors during processing the pending nodes.
284	*/
285	for (level = 0; level < BTRFS_MAX_LEVEL; level++) {
286	list_for_each_entry(node, &cache->pending[level], list) {
287	BUG_ON(!node->pending);
288	if (node->bytenr == node->new_bytenr)
289	continue;
290	update_backref_node(cache, node, bytenr: node->new_bytenr);
291	}
292	}
293
294	cache->last_trans = 0;
295	return 1;
296	}
297
298	static bool reloc_root_is_dead(const struct btrfs_root *root)
299	{
300	/*
301	* Pair with set_bit/clear_bit in clean_dirty_subvols and
302	* btrfs_update_reloc_root. We need to see the updated bit before
303	* trying to access reloc_root
304	*/
305	smp_rmb();
306	if (test_bit(BTRFS_ROOT_DEAD_RELOC_TREE, &root->state))
307	return true;
308	return false;
309	}
310
311	/*
312	* Check if this subvolume tree has valid reloc tree.
313	*
314	* Reloc tree after swap is considered dead, thus not considered as valid.
315	* This is enough for most callers, as they don't distinguish dead reloc root
316	* from no reloc root. But btrfs_should_ignore_reloc_root() below is a
317	* special case.
318	*/
319	static bool have_reloc_root(const struct btrfs_root *root)
320	{
321	if (reloc_root_is_dead(root))
322	return false;
323	if (!root->reloc_root)
324	return false;
325	return true;
326	}
327
328	bool btrfs_should_ignore_reloc_root(const struct btrfs_root *root)
329	{
330	struct btrfs_root *reloc_root;
331
332	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
333	return false;
334
335	/* This root has been merged with its reloc tree, we can ignore it */
336	if (reloc_root_is_dead(root))
337	return true;
338
339	reloc_root = root->reloc_root;
340	if (!reloc_root)
341	return false;
342
343	if (btrfs_header_generation(eb: reloc_root->commit_root) ==
344	root->fs_info->running_transaction->transid)
345	return false;
346	/*
347	* If there is reloc tree and it was created in previous transaction
348	* backref lookup can find the reloc tree, so backref node for the fs
349	* tree root is useless for relocation.
350	*/
351	return true;
352	}
353
354	/*
355	* find reloc tree by address of tree root
356	*/
357	struct btrfs_root *find_reloc_root(struct btrfs_fs_info *fs_info, u64 bytenr)
358	{
359	struct reloc_control *rc = fs_info->reloc_ctl;
360	struct rb_node *rb_node;
361	struct mapping_node *node;
362	struct btrfs_root *root = NULL;
363
364	ASSERT(rc);
365	spin_lock(lock: &rc->reloc_root_tree.lock);
366	rb_node = rb_simple_search(root: &rc->reloc_root_tree.rb_root, bytenr);
367	if (rb_node) {
368	node = rb_entry(rb_node, struct mapping_node, rb_node);
369	root = node->data;
370	}
371	spin_unlock(lock: &rc->reloc_root_tree.lock);
372	return btrfs_grab_root(root);
373	}
374
375	/*
376	* For useless nodes, do two major clean ups:
377	*
378	* - Cleanup the children edges and nodes
379	* If child node is also orphan (no parent) during cleanup, then the child
380	* node will also be cleaned up.
381	*
382	* - Freeing up leaves (level 0), keeps nodes detached
383	* For nodes, the node is still cached as "detached"
384	*
385	* Return false if @node is not in the @useless_nodes list.
386	* Return true if @node is in the @useless_nodes list.
387	*/
388	static bool handle_useless_nodes(struct reloc_control *rc,
389	struct btrfs_backref_node *node)
390	{
391	struct btrfs_backref_cache *cache = &rc->backref_cache;
392	struct list_head *useless_node = &cache->useless_node;
393	bool ret = false;
394
395	while (!list_empty(head: useless_node)) {
396	struct btrfs_backref_node *cur;
397
398	cur = list_first_entry(useless_node, struct btrfs_backref_node,
399	list);
400	list_del_init(entry: &cur->list);
401
402	/* Only tree root nodes can be added to @useless_nodes */
403	ASSERT(list_empty(&cur->upper));
404
405	if (cur == node)
406	ret = true;
407
408	/* The node is the lowest node */
409	if (cur->lowest) {
410	list_del_init(entry: &cur->lower);
411	cur->lowest = 0;
412	}
413
414	/* Cleanup the lower edges */
415	while (!list_empty(head: &cur->lower)) {
416	struct btrfs_backref_edge *edge;
417	struct btrfs_backref_node *lower;
418
419	edge = list_entry(cur->lower.next,
420	struct btrfs_backref_edge, list[UPPER]);
421	list_del(entry: &edge->list[UPPER]);
422	list_del(entry: &edge->list[LOWER]);
423	lower = edge->node[LOWER];
424	btrfs_backref_free_edge(cache, edge);
425
426	/* Child node is also orphan, queue for cleanup */
427	if (list_empty(head: &lower->upper))
428	list_add(new: &lower->list, head: useless_node);
429	}
430	/* Mark this block processed for relocation */
431	mark_block_processed(rc, node: cur);
432
433	/*
434	* Backref nodes for tree leaves are deleted from the cache.
435	* Backref nodes for upper level tree blocks are left in the
436	* cache to avoid unnecessary backref lookup.
437	*/
438	if (cur->level > 0) {
439	list_add(new: &cur->list, head: &cache->detached);
440	cur->detached = 1;
441	} else {
442	rb_erase(&cur->rb_node, &cache->rb_root);
443	btrfs_backref_free_node(cache, node: cur);
444	}
445	}
446	return ret;
447	}
448
449	/*
450	* Build backref tree for a given tree block. Root of the backref tree
451	* corresponds the tree block, leaves of the backref tree correspond roots of
452	* b-trees that reference the tree block.
453	*
454	* The basic idea of this function is check backrefs of a given block to find
455	* upper level blocks that reference the block, and then check backrefs of
456	* these upper level blocks recursively. The recursion stops when tree root is
457	* reached or backrefs for the block is cached.
458	*
459	* NOTE: if we find that backrefs for a block are cached, we know backrefs for
460	* all upper level blocks that directly/indirectly reference the block are also
461	* cached.
462	*/
463	static noinline_for_stack struct btrfs_backref_node *build_backref_tree(
464	struct btrfs_trans_handle *trans,
465	struct reloc_control *rc, struct btrfs_key *node_key,
466	int level, u64 bytenr)
467	{
468	struct btrfs_backref_iter *iter;
469	struct btrfs_backref_cache *cache = &rc->backref_cache;
470	/* For searching parent of TREE_BLOCK_REF */
471	struct btrfs_path *path;
472	struct btrfs_backref_node *cur;
473	struct btrfs_backref_node *node = NULL;
474	struct btrfs_backref_edge *edge;
475	int ret;
476	int err = 0;
477
478	iter = btrfs_backref_iter_alloc(fs_info: rc->extent_root->fs_info);
479	if (!iter)
480	return ERR_PTR(error: -ENOMEM);
481	path = btrfs_alloc_path();
482	if (!path) {
483	err = -ENOMEM;
484	goto out;
485	}
486
487	node = btrfs_backref_alloc_node(cache, bytenr, level);
488	if (!node) {
489	err = -ENOMEM;
490	goto out;
491	}
492
493	node->lowest = 1;
494	cur = node;
495
496	/* Breadth-first search to build backref cache */
497	do {
498	ret = btrfs_backref_add_tree_node(trans, cache, path, iter,
499	node_key, cur);
500	if (ret < 0) {
501	err = ret;
502	goto out;
503	}
504	edge = list_first_entry_or_null(&cache->pending_edge,
505	struct btrfs_backref_edge, list[UPPER]);
506	/*
507	* The pending list isn't empty, take the first block to
508	* process
509	*/
510	if (edge) {
511	list_del_init(entry: &edge->list[UPPER]);
512	cur = edge->node[UPPER];
513	}
514	} while (edge);
515
516	/* Finish the upper linkage of newly added edges/nodes */
517	ret = btrfs_backref_finish_upper_links(cache, start: node);
518	if (ret < 0) {
519	err = ret;
520	goto out;
521	}
522
523	if (handle_useless_nodes(rc, node))
524	node = NULL;
525	out:
526	btrfs_backref_iter_free(iter);
527	btrfs_free_path(p: path);
528	if (err) {
529	btrfs_backref_error_cleanup(cache, node);
530	return ERR_PTR(error: err);
531	}
532	ASSERT(!node || !node->detached);
533	ASSERT(list_empty(&cache->useless_node) &&
534	list_empty(&cache->pending_edge));
535	return node;
536	}
537
538	/*
539	* helper to add backref node for the newly created snapshot.
540	* the backref node is created by cloning backref node that
541	* corresponds to root of source tree
542	*/
543	static int clone_backref_node(struct btrfs_trans_handle *trans,
544	struct reloc_control *rc,
545	const struct btrfs_root *src,
546	struct btrfs_root *dest)
547	{
548	struct btrfs_root *reloc_root = src->reloc_root;
549	struct btrfs_backref_cache *cache = &rc->backref_cache;
550	struct btrfs_backref_node *node = NULL;
551	struct btrfs_backref_node *new_node;
552	struct btrfs_backref_edge *edge;
553	struct btrfs_backref_edge *new_edge;
554	struct rb_node *rb_node;
555
556	if (cache->last_trans > 0)
557	update_backref_cache(trans, cache);
558
559	rb_node = rb_simple_search(root: &cache->rb_root, bytenr: src->commit_root->start);
560	if (rb_node) {
561	node = rb_entry(rb_node, struct btrfs_backref_node, rb_node);
562	if (node->detached)
563	node = NULL;
564	else
565	BUG_ON(node->new_bytenr != reloc_root->node->start);
566	}
567
568	if (!node) {
569	rb_node = rb_simple_search(root: &cache->rb_root,
570	bytenr: reloc_root->commit_root->start);
571	if (rb_node) {
572	node = rb_entry(rb_node, struct btrfs_backref_node,
573	rb_node);
574	BUG_ON(node->detached);
575	}
576	}
577
578	if (!node)
579	return 0;
580
581	new_node = btrfs_backref_alloc_node(cache, bytenr: dest->node->start,
582	level: node->level);
583	if (!new_node)
584	return -ENOMEM;
585
586	new_node->lowest = node->lowest;
587	new_node->checked = 1;
588	new_node->root = btrfs_grab_root(root: dest);
589	ASSERT(new_node->root);
590
591	if (!node->lowest) {
592	list_for_each_entry(edge, &node->lower, list[UPPER]) {
593	new_edge = btrfs_backref_alloc_edge(cache);
594	if (!new_edge)
595	goto fail;
596
597	btrfs_backref_link_edge(edge: new_edge, lower: edge->node[LOWER],
598	upper: new_node, LINK_UPPER);
599	}
600	} else {
601	list_add_tail(new: &new_node->lower, head: &cache->leaves);
602	}
603
604	rb_node = rb_simple_insert(root: &cache->rb_root, bytenr: new_node->bytenr,
605	node: &new_node->rb_node);
606	if (rb_node)
607	btrfs_backref_panic(fs_info: trans->fs_info, bytenr: new_node->bytenr, error: -EEXIST);
608
609	if (!new_node->lowest) {
610	list_for_each_entry(new_edge, &new_node->lower, list[UPPER]) {
611	list_add_tail(new: &new_edge->list[LOWER],
612	head: &new_edge->node[LOWER]->upper);
613	}
614	}
615	return 0;
616	fail:
617	while (!list_empty(head: &new_node->lower)) {
618	new_edge = list_entry(new_node->lower.next,
619	struct btrfs_backref_edge, list[UPPER]);
620	list_del(entry: &new_edge->list[UPPER]);
621	btrfs_backref_free_edge(cache, edge: new_edge);
622	}
623	btrfs_backref_free_node(cache, node: new_node);
624	return -ENOMEM;
625	}
626
627	/*
628	* helper to add 'address of tree root -> reloc tree' mapping
629	*/
630	static int __add_reloc_root(struct btrfs_root *root)
631	{
632	struct btrfs_fs_info *fs_info = root->fs_info;
633	struct rb_node *rb_node;
634	struct mapping_node *node;
635	struct reloc_control *rc = fs_info->reloc_ctl;
636
637	node = kmalloc(size: sizeof(*node), GFP_NOFS);
638	if (!node)
639	return -ENOMEM;
640
641	node->bytenr = root->commit_root->start;
642	node->data = root;
643
644	spin_lock(lock: &rc->reloc_root_tree.lock);
645	rb_node = rb_simple_insert(root: &rc->reloc_root_tree.rb_root,
646	bytenr: node->bytenr, node: &node->rb_node);
647	spin_unlock(lock: &rc->reloc_root_tree.lock);
648	if (rb_node) {
649	btrfs_err(fs_info,
650	"Duplicate root found for start=%llu while inserting into relocation tree",
651	node->bytenr);
652	return -EEXIST;
653	}
654
655	list_add_tail(new: &root->root_list, head: &rc->reloc_roots);
656	return 0;
657	}
658
659	/*
660	* helper to delete the 'address of tree root -> reloc tree'
661	* mapping
662	*/
663	static void __del_reloc_root(struct btrfs_root *root)
664	{
665	struct btrfs_fs_info *fs_info = root->fs_info;
666	struct rb_node *rb_node;
667	struct mapping_node *node = NULL;
668	struct reloc_control *rc = fs_info->reloc_ctl;
669	bool put_ref = false;
670
671	if (rc && root->node) {
672	spin_lock(lock: &rc->reloc_root_tree.lock);
673	rb_node = rb_simple_search(root: &rc->reloc_root_tree.rb_root,
674	bytenr: root->commit_root->start);
675	if (rb_node) {
676	node = rb_entry(rb_node, struct mapping_node, rb_node);
677	rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
678	RB_CLEAR_NODE(&node->rb_node);
679	}
680	spin_unlock(lock: &rc->reloc_root_tree.lock);
681	ASSERT(!node || (struct btrfs_root *)node->data == root);
682	}
683
684	/*
685	* We only put the reloc root here if it's on the list. There's a lot
686	* of places where the pattern is to splice the rc->reloc_roots, process
687	* the reloc roots, and then add the reloc root back onto
688	* rc->reloc_roots. If we call __del_reloc_root while it's off of the
689	* list we don't want the reference being dropped, because the guy
690	* messing with the list is in charge of the reference.
691	*/
692	spin_lock(lock: &fs_info->trans_lock);
693	if (!list_empty(head: &root->root_list)) {
694	put_ref = true;
695	list_del_init(entry: &root->root_list);
696	}
697	spin_unlock(lock: &fs_info->trans_lock);
698	if (put_ref)
699	btrfs_put_root(root);
700	kfree(objp: node);
701	}
702
703	/*
704	* helper to update the 'address of tree root -> reloc tree'
705	* mapping
706	*/
707	static int __update_reloc_root(struct btrfs_root *root)
708	{
709	struct btrfs_fs_info *fs_info = root->fs_info;
710	struct rb_node *rb_node;
711	struct mapping_node *node = NULL;
712	struct reloc_control *rc = fs_info->reloc_ctl;
713
714	spin_lock(lock: &rc->reloc_root_tree.lock);
715	rb_node = rb_simple_search(root: &rc->reloc_root_tree.rb_root,
716	bytenr: root->commit_root->start);
717	if (rb_node) {
718	node = rb_entry(rb_node, struct mapping_node, rb_node);
719	rb_erase(&node->rb_node, &rc->reloc_root_tree.rb_root);
720	}
721	spin_unlock(lock: &rc->reloc_root_tree.lock);
722
723	if (!node)
724	return 0;
725	BUG_ON((struct btrfs_root *)node->data != root);
726
727	spin_lock(lock: &rc->reloc_root_tree.lock);
728	node->bytenr = root->node->start;
729	rb_node = rb_simple_insert(root: &rc->reloc_root_tree.rb_root,
730	bytenr: node->bytenr, node: &node->rb_node);
731	spin_unlock(lock: &rc->reloc_root_tree.lock);
732	if (rb_node)
733	btrfs_backref_panic(fs_info, bytenr: node->bytenr, error: -EEXIST);
734	return 0;
735	}
736
737	static struct btrfs_root *create_reloc_root(struct btrfs_trans_handle *trans,
738	struct btrfs_root *root, u64 objectid)
739	{
740	struct btrfs_fs_info *fs_info = root->fs_info;
741	struct btrfs_root *reloc_root;
742	struct extent_buffer *eb;
743	struct btrfs_root_item *root_item;
744	struct btrfs_key root_key;
745	int ret = 0;
746	bool must_abort = false;
747
748	root_item = kmalloc(size: sizeof(*root_item), GFP_NOFS);
749	if (!root_item)
750	return ERR_PTR(error: -ENOMEM);
751
752	root_key.objectid = BTRFS_TREE_RELOC_OBJECTID;
753	root_key.type = BTRFS_ROOT_ITEM_KEY;
754	root_key.offset = objectid;
755
756	if (root->root_key.objectid == objectid) {
757	u64 commit_root_gen;
758
759	/* called by btrfs_init_reloc_root */
760	ret = btrfs_copy_root(trans, root, buf: root->commit_root, cow_ret: &eb,
761	BTRFS_TREE_RELOC_OBJECTID);
762	if (ret)
763	goto fail;
764
765	/*
766	* Set the last_snapshot field to the generation of the commit
767	* root - like this ctree.c:btrfs_block_can_be_shared() behaves
768	* correctly (returns true) when the relocation root is created
769	* either inside the critical section of a transaction commit
770	* (through transaction.c:qgroup_account_snapshot()) and when
771	* it's created before the transaction commit is started.
772	*/
773	commit_root_gen = btrfs_header_generation(eb: root->commit_root);
774	btrfs_set_root_last_snapshot(s: &root->root_item, val: commit_root_gen);
775	} else {
776	/*
777	* called by btrfs_reloc_post_snapshot_hook.
778	* the source tree is a reloc tree, all tree blocks
779	* modified after it was created have RELOC flag
780	* set in their headers. so it's OK to not update
781	* the 'last_snapshot'.
782	*/
783	ret = btrfs_copy_root(trans, root, buf: root->node, cow_ret: &eb,
784	BTRFS_TREE_RELOC_OBJECTID);
785	if (ret)
786	goto fail;
787	}
788
789	/*
790	* We have changed references at this point, we must abort the
791	* transaction if anything fails.
792	*/
793	must_abort = true;
794
795	memcpy(root_item, &root->root_item, sizeof(*root_item));
796	btrfs_set_root_bytenr(s: root_item, val: eb->start);
797	btrfs_set_root_level(s: root_item, val: btrfs_header_level(eb));
798	btrfs_set_root_generation(s: root_item, val: trans->transid);
799
800	if (root->root_key.objectid == objectid) {
801	btrfs_set_root_refs(s: root_item, val: 0);
802	memset(&root_item->drop_progress, 0,
803	sizeof(struct btrfs_disk_key));
804	btrfs_set_root_drop_level(s: root_item, val: 0);
805	}
806
807	btrfs_tree_unlock(eb);
808	free_extent_buffer(eb);
809
810	ret = btrfs_insert_root(trans, root: fs_info->tree_root,
811	key: &root_key, item: root_item);
812	if (ret)
813	goto fail;
814
815	kfree(objp: root_item);
816
817	reloc_root = btrfs_read_tree_root(tree_root: fs_info->tree_root, key: &root_key);
818	if (IS_ERR(ptr: reloc_root)) {
819	ret = PTR_ERR(ptr: reloc_root);
820	goto abort;
821	}
822	set_bit(nr: BTRFS_ROOT_SHAREABLE, addr: &reloc_root->state);
823	reloc_root->last_trans = trans->transid;
824	return reloc_root;
825	fail:
826	kfree(objp: root_item);
827	abort:
828	if (must_abort)
829	btrfs_abort_transaction(trans, ret);
830	return ERR_PTR(error: ret);
831	}
832
833	/*
834	* create reloc tree for a given fs tree. reloc tree is just a
835	* snapshot of the fs tree with special root objectid.
836	*
837	* The reloc_root comes out of here with two references, one for
838	* root->reloc_root, and another for being on the rc->reloc_roots list.
839	*/
840	int btrfs_init_reloc_root(struct btrfs_trans_handle *trans,
841	struct btrfs_root *root)
842	{
843	struct btrfs_fs_info *fs_info = root->fs_info;
844	struct btrfs_root *reloc_root;
845	struct reloc_control *rc = fs_info->reloc_ctl;
846	struct btrfs_block_rsv *rsv;
847	int clear_rsv = 0;
848	int ret;
849
850	if (!rc)
851	return 0;
852
853	/*
854	* The subvolume has reloc tree but the swap is finished, no need to
855	* create/update the dead reloc tree
856	*/
857	if (reloc_root_is_dead(root))
858	return 0;
859
860	/*
861	* This is subtle but important. We do not do
862	* record_root_in_transaction for reloc roots, instead we record their
863	* corresponding fs root, and then here we update the last trans for the
864	* reloc root. This means that we have to do this for the entire life
865	* of the reloc root, regardless of which stage of the relocation we are
866	* in.
867	*/
868	if (root->reloc_root) {
869	reloc_root = root->reloc_root;
870	reloc_root->last_trans = trans->transid;
871	return 0;
872	}
873
874	/*
875	* We are merging reloc roots, we do not need new reloc trees. Also
876	* reloc trees never need their own reloc tree.
877	*/
878	if (!rc->create_reloc_tree ||
879	root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
880	return 0;
881
882	if (!trans->reloc_reserved) {
883	rsv = trans->block_rsv;
884	trans->block_rsv = rc->block_rsv;
885	clear_rsv = 1;
886	}
887	reloc_root = create_reloc_root(trans, root, objectid: root->root_key.objectid);
888	if (clear_rsv)
889	trans->block_rsv = rsv;
890	if (IS_ERR(ptr: reloc_root))
891	return PTR_ERR(ptr: reloc_root);
892
893	ret = __add_reloc_root(root: reloc_root);
894	ASSERT(ret != -EEXIST);
895	if (ret) {
896	/* Pairs with create_reloc_root */
897	btrfs_put_root(root: reloc_root);
898	return ret;
899	}
900	root->reloc_root = btrfs_grab_root(root: reloc_root);
901	return 0;
902	}
903
904	/*
905	* update root item of reloc tree
906	*/
907	int btrfs_update_reloc_root(struct btrfs_trans_handle *trans,
908	struct btrfs_root *root)
909	{
910	struct btrfs_fs_info *fs_info = root->fs_info;
911	struct btrfs_root *reloc_root;
912	struct btrfs_root_item *root_item;
913	int ret;
914
915	if (!have_reloc_root(root))
916	return 0;
917
918	reloc_root = root->reloc_root;
919	root_item = &reloc_root->root_item;
920
921	/*
922	* We are probably ok here, but __del_reloc_root() will drop its ref of
923	* the root. We have the ref for root->reloc_root, but just in case
924	* hold it while we update the reloc root.
925	*/
926	btrfs_grab_root(root: reloc_root);
927
928	/* root->reloc_root will stay until current relocation finished */
929	if (fs_info->reloc_ctl->merge_reloc_tree &&
930	btrfs_root_refs(s: root_item) == 0) {
931	set_bit(nr: BTRFS_ROOT_DEAD_RELOC_TREE, addr: &root->state);
932	/*
933	* Mark the tree as dead before we change reloc_root so
934	* have_reloc_root will not touch it from now on.
935	*/
936	smp_wmb();
937	__del_reloc_root(root: reloc_root);
938	}
939
940	if (reloc_root->commit_root != reloc_root->node) {
941	__update_reloc_root(root: reloc_root);
942	btrfs_set_root_node(item: root_item, node: reloc_root->node);
943	free_extent_buffer(eb: reloc_root->commit_root);
944	reloc_root->commit_root = btrfs_root_node(root: reloc_root);
945	}
946
947	ret = btrfs_update_root(trans, root: fs_info->tree_root,
948	key: &reloc_root->root_key, item: root_item);
949	btrfs_put_root(root: reloc_root);
950	return ret;
951	}
952
953	/*
954	* helper to find first cached inode with inode number >= objectid
955	* in a subvolume
956	*/
957	static struct inode *find_next_inode(struct btrfs_root *root, u64 objectid)
958	{
959	struct rb_node *node;
960	struct rb_node *prev;
961	struct btrfs_inode *entry;
962	struct inode *inode;
963
964	spin_lock(lock: &root->inode_lock);
965	again:
966	node = root->inode_tree.rb_node;
967	prev = NULL;
968	while (node) {
969	prev = node;
970	entry = rb_entry(node, struct btrfs_inode, rb_node);
971
972	if (objectid < btrfs_ino(inode: entry))
973	node = node->rb_left;
974	else if (objectid > btrfs_ino(inode: entry))
975	node = node->rb_right;
976	else
977	break;
978	}
979	if (!node) {
980	while (prev) {
981	entry = rb_entry(prev, struct btrfs_inode, rb_node);
982	if (objectid <= btrfs_ino(inode: entry)) {
983	node = prev;
984	break;
985	}
986	prev = rb_next(prev);
987	}
988	}
989	while (node) {
990	entry = rb_entry(node, struct btrfs_inode, rb_node);
991	inode = igrab(&entry->vfs_inode);
992	if (inode) {
993	spin_unlock(lock: &root->inode_lock);
994	return inode;
995	}
996
997	objectid = btrfs_ino(inode: entry) + 1;
998	if (cond_resched_lock(&root->inode_lock))
999	goto again;
1000
1001	node = rb_next(node);
1002	}
1003	spin_unlock(lock: &root->inode_lock);
1004	return NULL;
1005	}
1006
1007	/*
1008	* get new location of data
1009	*/
1010	static int get_new_location(struct inode *reloc_inode, u64 *new_bytenr,
1011	u64 bytenr, u64 num_bytes)
1012	{
1013	struct btrfs_root *root = BTRFS_I(inode: reloc_inode)->root;
1014	struct btrfs_path *path;
1015	struct btrfs_file_extent_item *fi;
1016	struct extent_buffer *leaf;
1017	int ret;
1018
1019	path = btrfs_alloc_path();
1020	if (!path)
1021	return -ENOMEM;
1022
1023	bytenr -= BTRFS_I(inode: reloc_inode)->index_cnt;
1024	ret = btrfs_lookup_file_extent(NULL, root, path,
1025	objectid: btrfs_ino(inode: BTRFS_I(inode: reloc_inode)), bytenr, mod: 0);
1026	if (ret < 0)
1027	goto out;
1028	if (ret > 0) {
1029	ret = -ENOENT;
1030	goto out;
1031	}
1032
1033	leaf = path->nodes[0];
1034	fi = btrfs_item_ptr(leaf, path->slots[0],
1035	struct btrfs_file_extent_item);
1036
1037	BUG_ON(btrfs_file_extent_offset(leaf, fi) ||
1038	btrfs_file_extent_compression(leaf, fi) ||
1039	btrfs_file_extent_encryption(leaf, fi) ||
1040	btrfs_file_extent_other_encoding(leaf, fi));
1041
1042	if (num_bytes != btrfs_file_extent_disk_num_bytes(eb: leaf, s: fi)) {
1043	ret = -EINVAL;
1044	goto out;
1045	}
1046
1047	*new_bytenr = btrfs_file_extent_disk_bytenr(eb: leaf, s: fi);
1048	ret = 0;
1049	out:
1050	btrfs_free_path(p: path);
1051	return ret;
1052	}
1053
1054	/*
1055	* update file extent items in the tree leaf to point to
1056	* the new locations.
1057	*/
1058	static noinline_for_stack
1059	int replace_file_extents(struct btrfs_trans_handle *trans,
1060	struct reloc_control *rc,
1061	struct btrfs_root *root,
1062	struct extent_buffer *leaf)
1063	{
1064	struct btrfs_fs_info *fs_info = root->fs_info;
1065	struct btrfs_key key;
1066	struct btrfs_file_extent_item *fi;
1067	struct inode *inode = NULL;
1068	u64 parent;
1069	u64 bytenr;
1070	u64 new_bytenr = 0;
1071	u64 num_bytes;
1072	u64 end;
1073	u32 nritems;
1074	u32 i;
1075	int ret = 0;
1076	int first = 1;
1077	int dirty = 0;
1078
1079	if (rc->stage != UPDATE_DATA_PTRS)
1080	return 0;
1081
1082	/* reloc trees always use full backref */
1083	if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID)
1084	parent = leaf->start;
1085	else
1086	parent = 0;
1087
1088	nritems = btrfs_header_nritems(eb: leaf);
1089	for (i = 0; i < nritems; i++) {
1090	struct btrfs_ref ref = { 0 };
1091
1092	cond_resched();
1093	btrfs_item_key_to_cpu(eb: leaf, cpu_key: &key, nr: i);
1094	if (key.type != BTRFS_EXTENT_DATA_KEY)
1095	continue;
1096	fi = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
1097	if (btrfs_file_extent_type(eb: leaf, s: fi) ==
1098	BTRFS_FILE_EXTENT_INLINE)
1099	continue;
1100	bytenr = btrfs_file_extent_disk_bytenr(eb: leaf, s: fi);
1101	num_bytes = btrfs_file_extent_disk_num_bytes(eb: leaf, s: fi);
1102	if (bytenr == 0)
1103	continue;
1104	if (!in_range(bytenr, rc->block_group->start,
1105	rc->block_group->length))
1106	continue;
1107
1108	/*
1109	* if we are modifying block in fs tree, wait for read_folio
1110	* to complete and drop the extent cache
1111	*/
1112	if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
1113	if (first) {
1114	inode = find_next_inode(root, objectid: key.objectid);
1115	first = 0;
1116	} else if (inode && btrfs_ino(inode: BTRFS_I(inode)) < key.objectid) {
1117	btrfs_add_delayed_iput(inode: BTRFS_I(inode));
1118	inode = find_next_inode(root, objectid: key.objectid);
1119	}
1120	if (inode && btrfs_ino(inode: BTRFS_I(inode)) == key.objectid) {
1121	struct extent_state *cached_state = NULL;
1122
1123	end = key.offset +
1124	btrfs_file_extent_num_bytes(eb: leaf, s: fi);
1125	WARN_ON(!IS_ALIGNED(key.offset,
1126	fs_info->sectorsize));
1127	WARN_ON(!IS_ALIGNED(end, fs_info->sectorsize));
1128	end--;
1129	ret = try_lock_extent(tree: &BTRFS_I(inode)->io_tree,
1130	start: key.offset, end,
1131	cached: &cached_state);
1132	if (!ret)
1133	continue;
1134
1135	btrfs_drop_extent_map_range(inode: BTRFS_I(inode),
1136	start: key.offset, end, skip_pinned: true);
1137	unlock_extent(tree: &BTRFS_I(inode)->io_tree,
1138	start: key.offset, end, cached: &cached_state);
1139	}
1140	}
1141
1142	ret = get_new_location(reloc_inode: rc->data_inode, new_bytenr: &new_bytenr,
1143	bytenr, num_bytes);
1144	if (ret) {
1145	/*
1146	* Don't have to abort since we've not changed anything
1147	* in the file extent yet.
1148	*/
1149	break;
1150	}
1151
1152	btrfs_set_file_extent_disk_bytenr(eb: leaf, s: fi, val: new_bytenr);
1153	dirty = 1;
1154
1155	key.offset -= btrfs_file_extent_offset(eb: leaf, s: fi);
1156	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_ADD_DELAYED_REF, bytenr: new_bytenr,
1157	len: num_bytes, parent, owning_root: root->root_key.objectid);
1158	btrfs_init_data_ref(generic_ref: &ref, ref_root: btrfs_header_owner(eb: leaf),
1159	ino: key.objectid, offset: key.offset,
1160	mod_root: root->root_key.objectid, skip_qgroup: false);
1161	ret = btrfs_inc_extent_ref(trans, generic_ref: &ref);
1162	if (ret) {
1163	btrfs_abort_transaction(trans, ret);
1164	break;
1165	}
1166
1167	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_DROP_DELAYED_REF, bytenr,
1168	len: num_bytes, parent, owning_root: root->root_key.objectid);
1169	btrfs_init_data_ref(generic_ref: &ref, ref_root: btrfs_header_owner(eb: leaf),
1170	ino: key.objectid, offset: key.offset,
1171	mod_root: root->root_key.objectid, skip_qgroup: false);
1172	ret = btrfs_free_extent(trans, ref: &ref);
1173	if (ret) {
1174	btrfs_abort_transaction(trans, ret);
1175	break;
1176	}
1177	}
1178	if (dirty)
1179	btrfs_mark_buffer_dirty(trans, buf: leaf);
1180	if (inode)
1181	btrfs_add_delayed_iput(inode: BTRFS_I(inode));
1182	return ret;
1183	}
1184
1185	static noinline_for_stack int memcmp_node_keys(const struct extent_buffer *eb,
1186	int slot, const struct btrfs_path *path,
1187	int level)
1188	{
1189	struct btrfs_disk_key key1;
1190	struct btrfs_disk_key key2;
1191	btrfs_node_key(eb, disk_key: &key1, nr: slot);
1192	btrfs_node_key(eb: path->nodes[level], disk_key: &key2, nr: path->slots[level]);
1193	return memcmp(p: &key1, q: &key2, size: sizeof(key1));
1194	}
1195
1196	/*
1197	* try to replace tree blocks in fs tree with the new blocks
1198	* in reloc tree. tree blocks haven't been modified since the
1199	* reloc tree was create can be replaced.
1200	*
1201	* if a block was replaced, level of the block + 1 is returned.
1202	* if no block got replaced, 0 is returned. if there are other
1203	* errors, a negative error number is returned.
1204	*/
1205	static noinline_for_stack
1206	int replace_path(struct btrfs_trans_handle *trans, struct reloc_control *rc,
1207	struct btrfs_root *dest, struct btrfs_root *src,
1208	struct btrfs_path *path, struct btrfs_key *next_key,
1209	int lowest_level, int max_level)
1210	{
1211	struct btrfs_fs_info *fs_info = dest->fs_info;
1212	struct extent_buffer *eb;
1213	struct extent_buffer *parent;
1214	struct btrfs_ref ref = { 0 };
1215	struct btrfs_key key;
1216	u64 old_bytenr;
1217	u64 new_bytenr;
1218	u64 old_ptr_gen;
1219	u64 new_ptr_gen;
1220	u64 last_snapshot;
1221	u32 blocksize;
1222	int cow = 0;
1223	int level;
1224	int ret;
1225	int slot;
1226
1227	ASSERT(src->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID);
1228	ASSERT(dest->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
1229
1230	last_snapshot = btrfs_root_last_snapshot(s: &src->root_item);
1231	again:
1232	slot = path->slots[lowest_level];
1233	btrfs_node_key_to_cpu(eb: path->nodes[lowest_level], cpu_key: &key, nr: slot);
1234
1235	eb = btrfs_lock_root_node(root: dest);
1236	level = btrfs_header_level(eb);
1237
1238	if (level < lowest_level) {
1239	btrfs_tree_unlock(eb);
1240	free_extent_buffer(eb);
1241	return 0;
1242	}
1243
1244	if (cow) {
1245	ret = btrfs_cow_block(trans, root: dest, buf: eb, NULL, parent_slot: 0, cow_ret: &eb,
1246	nest: BTRFS_NESTING_COW);
1247	if (ret) {
1248	btrfs_tree_unlock(eb);
1249	free_extent_buffer(eb);
1250	return ret;
1251	}
1252	}
1253
1254	if (next_key) {
1255	next_key->objectid = (u64)-1;
1256	next_key->type = (u8)-1;
1257	next_key->offset = (u64)-1;
1258	}
1259
1260	parent = eb;
1261	while (1) {
1262	level = btrfs_header_level(eb: parent);
1263	ASSERT(level >= lowest_level);
1264
1265	ret = btrfs_bin_search(eb: parent, first_slot: 0, key: &key, slot: &slot);
1266	if (ret < 0)
1267	break;
1268	if (ret && slot > 0)
1269	slot--;
1270
1271	if (next_key && slot + 1 < btrfs_header_nritems(eb: parent))
1272	btrfs_node_key_to_cpu(eb: parent, cpu_key: next_key, nr: slot + 1);
1273
1274	old_bytenr = btrfs_node_blockptr(eb: parent, nr: slot);
1275	blocksize = fs_info->nodesize;
1276	old_ptr_gen = btrfs_node_ptr_generation(eb: parent, nr: slot);
1277
1278	if (level <= max_level) {
1279	eb = path->nodes[level];
1280	new_bytenr = btrfs_node_blockptr(eb,
1281	nr: path->slots[level]);
1282	new_ptr_gen = btrfs_node_ptr_generation(eb,
1283	nr: path->slots[level]);
1284	} else {
1285	new_bytenr = 0;
1286	new_ptr_gen = 0;
1287	}
1288
1289	if (WARN_ON(new_bytenr > 0 && new_bytenr == old_bytenr)) {
1290	ret = level;
1291	break;
1292	}
1293
1294	if (new_bytenr == 0 || old_ptr_gen > last_snapshot ||
1295	memcmp_node_keys(eb: parent, slot, path, level)) {
1296	if (level <= lowest_level) {
1297	ret = 0;
1298	break;
1299	}
1300
1301	eb = btrfs_read_node_slot(parent, slot);
1302	if (IS_ERR(ptr: eb)) {
1303	ret = PTR_ERR(ptr: eb);
1304	break;
1305	}
1306	btrfs_tree_lock(eb);
1307	if (cow) {
1308	ret = btrfs_cow_block(trans, root: dest, buf: eb, parent,
1309	parent_slot: slot, cow_ret: &eb,
1310	nest: BTRFS_NESTING_COW);
1311	if (ret) {
1312	btrfs_tree_unlock(eb);
1313	free_extent_buffer(eb);
1314	break;
1315	}
1316	}
1317
1318	btrfs_tree_unlock(eb: parent);
1319	free_extent_buffer(eb: parent);
1320
1321	parent = eb;
1322	continue;
1323	}
1324
1325	if (!cow) {
1326	btrfs_tree_unlock(eb: parent);
1327	free_extent_buffer(eb: parent);
1328	cow = 1;
1329	goto again;
1330	}
1331
1332	btrfs_node_key_to_cpu(eb: path->nodes[level], cpu_key: &key,
1333	nr: path->slots[level]);
1334	btrfs_release_path(p: path);
1335
1336	path->lowest_level = level;
1337	set_bit(nr: BTRFS_ROOT_RESET_LOCKDEP_CLASS, addr: &src->state);
1338	ret = btrfs_search_slot(trans, root: src, key: &key, p: path, ins_len: 0, cow: 1);
1339	clear_bit(nr: BTRFS_ROOT_RESET_LOCKDEP_CLASS, addr: &src->state);
1340	path->lowest_level = 0;
1341	if (ret) {
1342	if (ret > 0)
1343	ret = -ENOENT;
1344	break;
1345	}
1346
1347	/*
1348	* Info qgroup to trace both subtrees.
1349	*
1350	* We must trace both trees.
1351	* 1) Tree reloc subtree
1352	* If not traced, we will leak data numbers
1353	* 2) Fs subtree
1354	* If not traced, we will double count old data
1355	*
1356	* We don't scan the subtree right now, but only record
1357	* the swapped tree blocks.
1358	* The real subtree rescan is delayed until we have new
1359	* CoW on the subtree root node before transaction commit.
1360	*/
1361	ret = btrfs_qgroup_add_swapped_blocks(trans, subvol_root: dest,
1362	bg: rc->block_group, subvol_parent: parent, subvol_slot: slot,
1363	reloc_parent: path->nodes[level], reloc_slot: path->slots[level],
1364	last_snapshot);
1365	if (ret < 0)
1366	break;
1367	/*
1368	* swap blocks in fs tree and reloc tree.
1369	*/
1370	btrfs_set_node_blockptr(eb: parent, nr: slot, val: new_bytenr);
1371	btrfs_set_node_ptr_generation(eb: parent, nr: slot, val: new_ptr_gen);
1372	btrfs_mark_buffer_dirty(trans, buf: parent);
1373
1374	btrfs_set_node_blockptr(eb: path->nodes[level],
1375	nr: path->slots[level], val: old_bytenr);
1376	btrfs_set_node_ptr_generation(eb: path->nodes[level],
1377	nr: path->slots[level], val: old_ptr_gen);
1378	btrfs_mark_buffer_dirty(trans, buf: path->nodes[level]);
1379
1380	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_ADD_DELAYED_REF, bytenr: old_bytenr,
1381	len: blocksize, parent: path->nodes[level]->start,
1382	owning_root: src->root_key.objectid);
1383	btrfs_init_tree_ref(generic_ref: &ref, level: level - 1, root: src->root_key.objectid,
1384	mod_root: 0, skip_qgroup: true);
1385	ret = btrfs_inc_extent_ref(trans, generic_ref: &ref);
1386	if (ret) {
1387	btrfs_abort_transaction(trans, ret);
1388	break;
1389	}
1390	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_ADD_DELAYED_REF, bytenr: new_bytenr,
1391	len: blocksize, parent: 0, owning_root: dest->root_key.objectid);
1392	btrfs_init_tree_ref(generic_ref: &ref, level: level - 1, root: dest->root_key.objectid, mod_root: 0,
1393	skip_qgroup: true);
1394	ret = btrfs_inc_extent_ref(trans, generic_ref: &ref);
1395	if (ret) {
1396	btrfs_abort_transaction(trans, ret);
1397	break;
1398	}
1399
1400	/* We don't know the real owning_root, use 0. */
1401	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_DROP_DELAYED_REF, bytenr: new_bytenr,
1402	len: blocksize, parent: path->nodes[level]->start, owning_root: 0);
1403	btrfs_init_tree_ref(generic_ref: &ref, level: level - 1, root: src->root_key.objectid,
1404	mod_root: 0, skip_qgroup: true);
1405	ret = btrfs_free_extent(trans, ref: &ref);
1406	if (ret) {
1407	btrfs_abort_transaction(trans, ret);
1408	break;
1409	}
1410
1411	/* We don't know the real owning_root, use 0. */
1412	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_DROP_DELAYED_REF, bytenr: old_bytenr,
1413	len: blocksize, parent: 0, owning_root: 0);
1414	btrfs_init_tree_ref(generic_ref: &ref, level: level - 1, root: dest->root_key.objectid,
1415	mod_root: 0, skip_qgroup: true);
1416	ret = btrfs_free_extent(trans, ref: &ref);
1417	if (ret) {
1418	btrfs_abort_transaction(trans, ret);
1419	break;
1420	}
1421
1422	btrfs_unlock_up_safe(path, level: 0);
1423
1424	ret = level;
1425	break;
1426	}
1427	btrfs_tree_unlock(eb: parent);
1428	free_extent_buffer(eb: parent);
1429	return ret;
1430	}
1431
1432	/*
1433	* helper to find next relocated block in reloc tree
1434	*/
1435	static noinline_for_stack
1436	int walk_up_reloc_tree(struct btrfs_root *root, struct btrfs_path *path,
1437	int *level)
1438	{
1439	struct extent_buffer *eb;
1440	int i;
1441	u64 last_snapshot;
1442	u32 nritems;
1443
1444	last_snapshot = btrfs_root_last_snapshot(s: &root->root_item);
1445
1446	for (i = 0; i < *level; i++) {
1447	free_extent_buffer(eb: path->nodes[i]);
1448	path->nodes[i] = NULL;
1449	}
1450
1451	for (i = *level; i < BTRFS_MAX_LEVEL && path->nodes[i]; i++) {
1452	eb = path->nodes[i];
1453	nritems = btrfs_header_nritems(eb);
1454	while (path->slots[i] + 1 < nritems) {
1455	path->slots[i]++;
1456	if (btrfs_node_ptr_generation(eb, nr: path->slots[i]) <=
1457	last_snapshot)
1458	continue;
1459
1460	*level = i;
1461	return 0;
1462	}
1463	free_extent_buffer(eb: path->nodes[i]);
1464	path->nodes[i] = NULL;
1465	}
1466	return 1;
1467	}
1468
1469	/*
1470	* walk down reloc tree to find relocated block of lowest level
1471	*/
1472	static noinline_for_stack
1473	int walk_down_reloc_tree(struct btrfs_root *root, struct btrfs_path *path,
1474	int *level)
1475	{
1476	struct extent_buffer *eb = NULL;
1477	int i;
1478	u64 ptr_gen = 0;
1479	u64 last_snapshot;
1480	u32 nritems;
1481
1482	last_snapshot = btrfs_root_last_snapshot(s: &root->root_item);
1483
1484	for (i = *level; i > 0; i--) {
1485	eb = path->nodes[i];
1486	nritems = btrfs_header_nritems(eb);
1487	while (path->slots[i] < nritems) {
1488	ptr_gen = btrfs_node_ptr_generation(eb, nr: path->slots[i]);
1489	if (ptr_gen > last_snapshot)
1490	break;
1491	path->slots[i]++;
1492	}
1493	if (path->slots[i] >= nritems) {
1494	if (i == *level)
1495	break;
1496	*level = i + 1;
1497	return 0;
1498	}
1499	if (i == 1) {
1500	*level = i;
1501	return 0;
1502	}
1503
1504	eb = btrfs_read_node_slot(parent: eb, slot: path->slots[i]);
1505	if (IS_ERR(ptr: eb))
1506	return PTR_ERR(ptr: eb);
1507	BUG_ON(btrfs_header_level(eb) != i - 1);
1508	path->nodes[i - 1] = eb;
1509	path->slots[i - 1] = 0;
1510	}
1511	return 1;
1512	}
1513
1514	/*
1515	* invalidate extent cache for file extents whose key in range of
1516	* [min_key, max_key)
1517	*/
1518	static int invalidate_extent_cache(struct btrfs_root *root,
1519	const struct btrfs_key *min_key,
1520	const struct btrfs_key *max_key)
1521	{
1522	struct btrfs_fs_info *fs_info = root->fs_info;
1523	struct inode *inode = NULL;
1524	u64 objectid;
1525	u64 start, end;
1526	u64 ino;
1527
1528	objectid = min_key->objectid;
1529	while (1) {
1530	struct extent_state *cached_state = NULL;
1531
1532	cond_resched();
1533	iput(inode);
1534
1535	if (objectid > max_key->objectid)
1536	break;
1537
1538	inode = find_next_inode(root, objectid);
1539	if (!inode)
1540	break;
1541	ino = btrfs_ino(inode: BTRFS_I(inode));
1542
1543	if (ino > max_key->objectid) {
1544	iput(inode);
1545	break;
1546	}
1547
1548	objectid = ino + 1;
1549	if (!S_ISREG(inode->i_mode))
1550	continue;
1551
1552	if (unlikely(min_key->objectid == ino)) {
1553	if (min_key->type > BTRFS_EXTENT_DATA_KEY)
1554	continue;
1555	if (min_key->type < BTRFS_EXTENT_DATA_KEY)
1556	start = 0;
1557	else {
1558	start = min_key->offset;
1559	WARN_ON(!IS_ALIGNED(start, fs_info->sectorsize));
1560	}
1561	} else {
1562	start = 0;
1563	}
1564
1565	if (unlikely(max_key->objectid == ino)) {
1566	if (max_key->type < BTRFS_EXTENT_DATA_KEY)
1567	continue;
1568	if (max_key->type > BTRFS_EXTENT_DATA_KEY) {
1569	end = (u64)-1;
1570	} else {
1571	if (max_key->offset == 0)
1572	continue;
1573	end = max_key->offset;
1574	WARN_ON(!IS_ALIGNED(end, fs_info->sectorsize));
1575	end--;
1576	}
1577	} else {
1578	end = (u64)-1;
1579	}
1580
1581	/* the lock_extent waits for read_folio to complete */
1582	lock_extent(tree: &BTRFS_I(inode)->io_tree, start, end, cached: &cached_state);
1583	btrfs_drop_extent_map_range(inode: BTRFS_I(inode), start, end, skip_pinned: true);
1584	unlock_extent(tree: &BTRFS_I(inode)->io_tree, start, end, cached: &cached_state);
1585	}
1586	return 0;
1587	}
1588
1589	static int find_next_key(struct btrfs_path *path, int level,
1590	struct btrfs_key *key)
1591
1592	{
1593	while (level < BTRFS_MAX_LEVEL) {
1594	if (!path->nodes[level])
1595	break;
1596	if (path->slots[level] + 1 <
1597	btrfs_header_nritems(eb: path->nodes[level])) {
1598	btrfs_node_key_to_cpu(eb: path->nodes[level], cpu_key: key,
1599	nr: path->slots[level] + 1);
1600	return 0;
1601	}
1602	level++;
1603	}
1604	return 1;
1605	}
1606
1607	/*
1608	* Insert current subvolume into reloc_control::dirty_subvol_roots
1609	*/
1610	static int insert_dirty_subvol(struct btrfs_trans_handle *trans,
1611	struct reloc_control *rc,
1612	struct btrfs_root *root)
1613	{
1614	struct btrfs_root *reloc_root = root->reloc_root;
1615	struct btrfs_root_item *reloc_root_item;
1616	int ret;
1617
1618	/* @root must be a subvolume tree root with a valid reloc tree */
1619	ASSERT(root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID);
1620	ASSERT(reloc_root);
1621
1622	reloc_root_item = &reloc_root->root_item;
1623	memset(&reloc_root_item->drop_progress, 0,
1624	sizeof(reloc_root_item->drop_progress));
1625	btrfs_set_root_drop_level(s: reloc_root_item, val: 0);
1626	btrfs_set_root_refs(s: reloc_root_item, val: 0);
1627	ret = btrfs_update_reloc_root(trans, root);
1628	if (ret)
1629	return ret;
1630
1631	if (list_empty(head: &root->reloc_dirty_list)) {
1632	btrfs_grab_root(root);
1633	list_add_tail(new: &root->reloc_dirty_list, head: &rc->dirty_subvol_roots);
1634	}
1635
1636	return 0;
1637	}
1638
1639	static int clean_dirty_subvols(struct reloc_control *rc)
1640	{
1641	struct btrfs_root *root;
1642	struct btrfs_root *next;
1643	int ret = 0;
1644	int ret2;
1645
1646	list_for_each_entry_safe(root, next, &rc->dirty_subvol_roots,
1647	reloc_dirty_list) {
1648	if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID) {
1649	/* Merged subvolume, cleanup its reloc root */
1650	struct btrfs_root *reloc_root = root->reloc_root;
1651
1652	list_del_init(entry: &root->reloc_dirty_list);
1653	root->reloc_root = NULL;
1654	/*
1655	* Need barrier to ensure clear_bit() only happens after
1656	* root->reloc_root = NULL. Pairs with have_reloc_root.
1657	*/
1658	smp_wmb();
1659	clear_bit(nr: BTRFS_ROOT_DEAD_RELOC_TREE, addr: &root->state);
1660	if (reloc_root) {
1661	/*
1662	* btrfs_drop_snapshot drops our ref we hold for
1663	* ->reloc_root. If it fails however we must
1664	* drop the ref ourselves.
1665	*/
1666	ret2 = btrfs_drop_snapshot(root: reloc_root, update_ref: 0, for_reloc: 1);
1667	if (ret2 < 0) {
1668	btrfs_put_root(root: reloc_root);
1669	if (!ret)
1670	ret = ret2;
1671	}
1672	}
1673	btrfs_put_root(root);
1674	} else {
1675	/* Orphan reloc tree, just clean it up */
1676	ret2 = btrfs_drop_snapshot(root, update_ref: 0, for_reloc: 1);
1677	if (ret2 < 0) {
1678	btrfs_put_root(root);
1679	if (!ret)
1680	ret = ret2;
1681	}
1682	}
1683	}
1684	return ret;
1685	}
1686
1687	/*
1688	* merge the relocated tree blocks in reloc tree with corresponding
1689	* fs tree.
1690	*/
1691	static noinline_for_stack int merge_reloc_root(struct reloc_control *rc,
1692	struct btrfs_root *root)
1693	{
1694	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
1695	struct btrfs_key key;
1696	struct btrfs_key next_key;
1697	struct btrfs_trans_handle *trans = NULL;
1698	struct btrfs_root *reloc_root;
1699	struct btrfs_root_item *root_item;
1700	struct btrfs_path *path;
1701	struct extent_buffer *leaf;
1702	int reserve_level;
1703	int level;
1704	int max_level;
1705	int replaced = 0;
1706	int ret = 0;
1707	u32 min_reserved;
1708
1709	path = btrfs_alloc_path();
1710	if (!path)
1711	return -ENOMEM;
1712	path->reada = READA_FORWARD;
1713
1714	reloc_root = root->reloc_root;
1715	root_item = &reloc_root->root_item;
1716
1717	if (btrfs_disk_key_objectid(s: &root_item->drop_progress) == 0) {
1718	level = btrfs_root_level(s: root_item);
1719	atomic_inc(v: &reloc_root->node->refs);
1720	path->nodes[level] = reloc_root->node;
1721	path->slots[level] = 0;
1722	} else {
1723	btrfs_disk_key_to_cpu(cpu_key: &key, disk_key: &root_item->drop_progress);
1724
1725	level = btrfs_root_drop_level(s: root_item);
1726	BUG_ON(level == 0);
1727	path->lowest_level = level;
1728	ret = btrfs_search_slot(NULL, root: reloc_root, key: &key, p: path, ins_len: 0, cow: 0);
1729	path->lowest_level = 0;
1730	if (ret < 0) {
1731	btrfs_free_path(p: path);
1732	return ret;
1733	}
1734
1735	btrfs_node_key_to_cpu(eb: path->nodes[level], cpu_key: &next_key,
1736	nr: path->slots[level]);
1737	WARN_ON(memcmp(&key, &next_key, sizeof(key)));
1738
1739	btrfs_unlock_up_safe(path, level: 0);
1740	}
1741
1742	/*
1743	* In merge_reloc_root(), we modify the upper level pointer to swap the
1744	* tree blocks between reloc tree and subvolume tree. Thus for tree
1745	* block COW, we COW at most from level 1 to root level for each tree.
1746	*
1747	* Thus the needed metadata size is at most root_level * nodesize,
1748	* and * 2 since we have two trees to COW.
1749	*/
1750	reserve_level = max_t(int, 1, btrfs_root_level(root_item));
1751	min_reserved = fs_info->nodesize * reserve_level * 2;
1752	memset(&next_key, 0, sizeof(next_key));
1753
1754	while (1) {
1755	ret = btrfs_block_rsv_refill(fs_info, block_rsv: rc->block_rsv,
1756	num_bytes: min_reserved,
1757	flush: BTRFS_RESERVE_FLUSH_LIMIT);
1758	if (ret)
1759	goto out;
1760	trans = btrfs_start_transaction(root, num_items: 0);
1761	if (IS_ERR(ptr: trans)) {
1762	ret = PTR_ERR(ptr: trans);
1763	trans = NULL;
1764	goto out;
1765	}
1766
1767	/*
1768	* At this point we no longer have a reloc_control, so we can't
1769	* depend on btrfs_init_reloc_root to update our last_trans.
1770	*
1771	* But that's ok, we started the trans handle on our
1772	* corresponding fs_root, which means it's been added to the
1773	* dirty list. At commit time we'll still call
1774	* btrfs_update_reloc_root() and update our root item
1775	* appropriately.
1776	*/
1777	reloc_root->last_trans = trans->transid;
1778	trans->block_rsv = rc->block_rsv;
1779
1780	replaced = 0;
1781	max_level = level;
1782
1783	ret = walk_down_reloc_tree(root: reloc_root, path, level: &level);
1784	if (ret < 0)
1785	goto out;
1786	if (ret > 0)
1787	break;
1788
1789	if (!find_next_key(path, level, key: &key) &&
1790	btrfs_comp_cpu_keys(k1: &next_key, k2: &key) >= 0) {
1791	ret = 0;
1792	} else {
1793	ret = replace_path(trans, rc, dest: root, src: reloc_root, path,
1794	next_key: &next_key, lowest_level: level, max_level);
1795	}
1796	if (ret < 0)
1797	goto out;
1798	if (ret > 0) {
1799	level = ret;
1800	btrfs_node_key_to_cpu(eb: path->nodes[level], cpu_key: &key,
1801	nr: path->slots[level]);
1802	replaced = 1;
1803	}
1804
1805	ret = walk_up_reloc_tree(root: reloc_root, path, level: &level);
1806	if (ret > 0)
1807	break;
1808
1809	BUG_ON(level == 0);
1810	/*
1811	* save the merging progress in the drop_progress.
1812	* this is OK since root refs == 1 in this case.
1813	*/
1814	btrfs_node_key(eb: path->nodes[level], disk_key: &root_item->drop_progress,
1815	nr: path->slots[level]);
1816	btrfs_set_root_drop_level(s: root_item, val: level);
1817
1818	btrfs_end_transaction_throttle(trans);
1819	trans = NULL;
1820
1821	btrfs_btree_balance_dirty(fs_info);
1822
1823	if (replaced && rc->stage == UPDATE_DATA_PTRS)
1824	invalidate_extent_cache(root, min_key: &key, max_key: &next_key);
1825	}
1826
1827	/*
1828	* handle the case only one block in the fs tree need to be
1829	* relocated and the block is tree root.
1830	*/
1831	leaf = btrfs_lock_root_node(root);
1832	ret = btrfs_cow_block(trans, root, buf: leaf, NULL, parent_slot: 0, cow_ret: &leaf,
1833	nest: BTRFS_NESTING_COW);
1834	btrfs_tree_unlock(eb: leaf);
1835	free_extent_buffer(eb: leaf);
1836	out:
1837	btrfs_free_path(p: path);
1838
1839	if (ret == 0) {
1840	ret = insert_dirty_subvol(trans, rc, root);
1841	if (ret)
1842	btrfs_abort_transaction(trans, ret);
1843	}
1844
1845	if (trans)
1846	btrfs_end_transaction_throttle(trans);
1847
1848	btrfs_btree_balance_dirty(fs_info);
1849
1850	if (replaced && rc->stage == UPDATE_DATA_PTRS)
1851	invalidate_extent_cache(root, min_key: &key, max_key: &next_key);
1852
1853	return ret;
1854	}
1855
1856	static noinline_for_stack
1857	int prepare_to_merge(struct reloc_control *rc, int err)
1858	{
1859	struct btrfs_root *root = rc->extent_root;
1860	struct btrfs_fs_info *fs_info = root->fs_info;
1861	struct btrfs_root *reloc_root;
1862	struct btrfs_trans_handle *trans;
1863	LIST_HEAD(reloc_roots);
1864	u64 num_bytes = 0;
1865	int ret;
1866
1867	mutex_lock(&fs_info->reloc_mutex);
1868	rc->merging_rsv_size += fs_info->nodesize * (BTRFS_MAX_LEVEL - 1) * 2;
1869	rc->merging_rsv_size += rc->nodes_relocated * 2;
1870	mutex_unlock(lock: &fs_info->reloc_mutex);
1871
1872	again:
1873	if (!err) {
1874	num_bytes = rc->merging_rsv_size;
1875	ret = btrfs_block_rsv_add(fs_info, block_rsv: rc->block_rsv, num_bytes,
1876	flush: BTRFS_RESERVE_FLUSH_ALL);
1877	if (ret)
1878	err = ret;
1879	}
1880
1881	trans = btrfs_join_transaction(root: rc->extent_root);
1882	if (IS_ERR(ptr: trans)) {
1883	if (!err)
1884	btrfs_block_rsv_release(fs_info, block_rsv: rc->block_rsv,
1885	num_bytes, NULL);
1886	return PTR_ERR(ptr: trans);
1887	}
1888
1889	if (!err) {
1890	if (num_bytes != rc->merging_rsv_size) {
1891	btrfs_end_transaction(trans);
1892	btrfs_block_rsv_release(fs_info, block_rsv: rc->block_rsv,
1893	num_bytes, NULL);
1894	goto again;
1895	}
1896	}
1897
1898	rc->merge_reloc_tree = true;
1899
1900	while (!list_empty(head: &rc->reloc_roots)) {
1901	reloc_root = list_entry(rc->reloc_roots.next,
1902	struct btrfs_root, root_list);
1903	list_del_init(entry: &reloc_root->root_list);
1904
1905	root = btrfs_get_fs_root(fs_info, objectid: reloc_root->root_key.offset,
1906	check_ref: false);
1907	if (IS_ERR(ptr: root)) {
1908	/*
1909	* Even if we have an error we need this reloc root
1910	* back on our list so we can clean up properly.
1911	*/
1912	list_add(new: &reloc_root->root_list, head: &reloc_roots);
1913	btrfs_abort_transaction(trans, (int)PTR_ERR(root));
1914	if (!err)
1915	err = PTR_ERR(ptr: root);
1916	break;
1917	}
1918
1919	if (unlikely(root->reloc_root != reloc_root)) {
1920	if (root->reloc_root) {
1921	btrfs_err(fs_info,
1922	"reloc tree mismatch, root %lld has reloc root key (%lld %u %llu) gen %llu, expect reloc root key (%lld %u %llu) gen %llu",
1923	root->root_key.objectid,
1924	root->reloc_root->root_key.objectid,
1925	root->reloc_root->root_key.type,
1926	root->reloc_root->root_key.offset,
1927	btrfs_root_generation(
1928	&root->reloc_root->root_item),
1929	reloc_root->root_key.objectid,
1930	reloc_root->root_key.type,
1931	reloc_root->root_key.offset,
1932	btrfs_root_generation(
1933	&reloc_root->root_item));
1934	} else {
1935	btrfs_err(fs_info,
1936	"reloc tree mismatch, root %lld has no reloc root, expect reloc root key (%lld %u %llu) gen %llu",
1937	root->root_key.objectid,
1938	reloc_root->root_key.objectid,
1939	reloc_root->root_key.type,
1940	reloc_root->root_key.offset,
1941	btrfs_root_generation(
1942	&reloc_root->root_item));
1943	}
1944	list_add(new: &reloc_root->root_list, head: &reloc_roots);
1945	btrfs_put_root(root);
1946	btrfs_abort_transaction(trans, -EUCLEAN);
1947	if (!err)
1948	err = -EUCLEAN;
1949	break;
1950	}
1951
1952	/*
1953	* set reference count to 1, so btrfs_recover_relocation
1954	* knows it should resumes merging
1955	*/
1956	if (!err)
1957	btrfs_set_root_refs(s: &reloc_root->root_item, val: 1);
1958	ret = btrfs_update_reloc_root(trans, root);
1959
1960	/*
1961	* Even if we have an error we need this reloc root back on our
1962	* list so we can clean up properly.
1963	*/
1964	list_add(new: &reloc_root->root_list, head: &reloc_roots);
1965	btrfs_put_root(root);
1966
1967	if (ret) {
1968	btrfs_abort_transaction(trans, ret);
1969	if (!err)
1970	err = ret;
1971	break;
1972	}
1973	}
1974
1975	list_splice(list: &reloc_roots, head: &rc->reloc_roots);
1976
1977	if (!err)
1978	err = btrfs_commit_transaction(trans);
1979	else
1980	btrfs_end_transaction(trans);
1981	return err;
1982	}
1983
1984	static noinline_for_stack
1985	void free_reloc_roots(struct list_head *list)
1986	{
1987	struct btrfs_root *reloc_root, *tmp;
1988
1989	list_for_each_entry_safe(reloc_root, tmp, list, root_list)
1990	__del_reloc_root(root: reloc_root);
1991	}
1992
1993	static noinline_for_stack
1994	void merge_reloc_roots(struct reloc_control *rc)
1995	{
1996	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
1997	struct btrfs_root *root;
1998	struct btrfs_root *reloc_root;
1999	LIST_HEAD(reloc_roots);
2000	int found = 0;
2001	int ret = 0;
2002	again:
2003	root = rc->extent_root;
2004
2005	/*
2006	* this serializes us with btrfs_record_root_in_transaction,
2007	* we have to make sure nobody is in the middle of
2008	* adding their roots to the list while we are
2009	* doing this splice
2010	*/
2011	mutex_lock(&fs_info->reloc_mutex);
2012	list_splice_init(list: &rc->reloc_roots, head: &reloc_roots);
2013	mutex_unlock(lock: &fs_info->reloc_mutex);
2014
2015	while (!list_empty(head: &reloc_roots)) {
2016	found = 1;
2017	reloc_root = list_entry(reloc_roots.next,
2018	struct btrfs_root, root_list);
2019
2020	root = btrfs_get_fs_root(fs_info, objectid: reloc_root->root_key.offset,
2021	check_ref: false);
2022	if (btrfs_root_refs(s: &reloc_root->root_item) > 0) {
2023	if (WARN_ON(IS_ERR(root))) {
2024	/*
2025	* For recovery we read the fs roots on mount,
2026	* and if we didn't find the root then we marked
2027	* the reloc root as a garbage root. For normal
2028	* relocation obviously the root should exist in
2029	* memory. However there's no reason we can't
2030	* handle the error properly here just in case.
2031	*/
2032	ret = PTR_ERR(ptr: root);
2033	goto out;
2034	}
2035	if (WARN_ON(root->reloc_root != reloc_root)) {
2036	/*
2037	* This can happen if on-disk metadata has some
2038	* corruption, e.g. bad reloc tree key offset.
2039	*/
2040	ret = -EINVAL;
2041	goto out;
2042	}
2043	ret = merge_reloc_root(rc, root);
2044	btrfs_put_root(root);
2045	if (ret) {
2046	if (list_empty(head: &reloc_root->root_list))
2047	list_add_tail(new: &reloc_root->root_list,
2048	head: &reloc_roots);
2049	goto out;
2050	}
2051	} else {
2052	if (!IS_ERR(ptr: root)) {
2053	if (root->reloc_root == reloc_root) {
2054	root->reloc_root = NULL;
2055	btrfs_put_root(root: reloc_root);
2056	}
2057	clear_bit(nr: BTRFS_ROOT_DEAD_RELOC_TREE,
2058	addr: &root->state);
2059	btrfs_put_root(root);
2060	}
2061
2062	list_del_init(entry: &reloc_root->root_list);
2063	/* Don't forget to queue this reloc root for cleanup */
2064	list_add_tail(new: &reloc_root->reloc_dirty_list,
2065	head: &rc->dirty_subvol_roots);
2066	}
2067	}
2068
2069	if (found) {
2070	found = 0;
2071	goto again;
2072	}
2073	out:
2074	if (ret) {
2075	btrfs_handle_fs_error(fs_info, ret, NULL);
2076	free_reloc_roots(list: &reloc_roots);
2077
2078	/* new reloc root may be added */
2079	mutex_lock(&fs_info->reloc_mutex);
2080	list_splice_init(list: &rc->reloc_roots, head: &reloc_roots);
2081	mutex_unlock(lock: &fs_info->reloc_mutex);
2082	free_reloc_roots(list: &reloc_roots);
2083	}
2084
2085	/*
2086	* We used to have
2087	*
2088	* BUG_ON(!RB_EMPTY_ROOT(&rc->reloc_root_tree.rb_root));
2089	*
2090	* here, but it's wrong. If we fail to start the transaction in
2091	* prepare_to_merge() we will have only 0 ref reloc roots, none of which
2092	* have actually been removed from the reloc_root_tree rb tree. This is
2093	* fine because we're bailing here, and we hold a reference on the root
2094	* for the list that holds it, so these roots will be cleaned up when we
2095	* do the reloc_dirty_list afterwards. Meanwhile the root->reloc_root
2096	* will be cleaned up on unmount.
2097	*
2098	* The remaining nodes will be cleaned up by free_reloc_control.
2099	*/
2100	}
2101
2102	static void free_block_list(struct rb_root *blocks)
2103	{
2104	struct tree_block *block;
2105	struct rb_node *rb_node;
2106	while ((rb_node = rb_first(blocks))) {
2107	block = rb_entry(rb_node, struct tree_block, rb_node);
2108	rb_erase(rb_node, blocks);
2109	kfree(objp: block);
2110	}
2111	}
2112
2113	static int record_reloc_root_in_trans(struct btrfs_trans_handle *trans,
2114	struct btrfs_root *reloc_root)
2115	{
2116	struct btrfs_fs_info *fs_info = reloc_root->fs_info;
2117	struct btrfs_root *root;
2118	int ret;
2119
2120	if (reloc_root->last_trans == trans->transid)
2121	return 0;
2122
2123	root = btrfs_get_fs_root(fs_info, objectid: reloc_root->root_key.offset, check_ref: false);
2124
2125	/*
2126	* This should succeed, since we can't have a reloc root without having
2127	* already looked up the actual root and created the reloc root for this
2128	* root.
2129	*
2130	* However if there's some sort of corruption where we have a ref to a
2131	* reloc root without a corresponding root this could return ENOENT.
2132	*/
2133	if (IS_ERR(ptr: root)) {
2134	ASSERT(0);
2135	return PTR_ERR(ptr: root);
2136	}
2137	if (root->reloc_root != reloc_root) {
2138	ASSERT(0);
2139	btrfs_err(fs_info,
2140	"root %llu has two reloc roots associated with it",
2141	reloc_root->root_key.offset);
2142	btrfs_put_root(root);
2143	return -EUCLEAN;
2144	}
2145	ret = btrfs_record_root_in_trans(trans, root);
2146	btrfs_put_root(root);
2147
2148	return ret;
2149	}
2150
2151	static noinline_for_stack
2152	struct btrfs_root *select_reloc_root(struct btrfs_trans_handle *trans,
2153	struct reloc_control *rc,
2154	struct btrfs_backref_node *node,
2155	struct btrfs_backref_edge *edges[])
2156	{
2157	struct btrfs_backref_node *next;
2158	struct btrfs_root *root;
2159	int index = 0;
2160	int ret;
2161
2162	next = node;
2163	while (1) {
2164	cond_resched();
2165	next = walk_up_backref(node: next, edges, index: &index);
2166	root = next->root;
2167
2168	/*
2169	* If there is no root, then our references for this block are
2170	* incomplete, as we should be able to walk all the way up to a
2171	* block that is owned by a root.
2172	*
2173	* This path is only for SHAREABLE roots, so if we come upon a
2174	* non-SHAREABLE root then we have backrefs that resolve
2175	* improperly.
2176	*
2177	* Both of these cases indicate file system corruption, or a bug
2178	* in the backref walking code.
2179	*/
2180	if (!root) {
2181	ASSERT(0);
2182	btrfs_err(trans->fs_info,
2183	"bytenr %llu doesn't have a backref path ending in a root",
2184	node->bytenr);
2185	return ERR_PTR(error: -EUCLEAN);
2186	}
2187	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
2188	ASSERT(0);
2189	btrfs_err(trans->fs_info,
2190	"bytenr %llu has multiple refs with one ending in a non-shareable root",
2191	node->bytenr);
2192	return ERR_PTR(error: -EUCLEAN);
2193	}
2194
2195	if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID) {
2196	ret = record_reloc_root_in_trans(trans, reloc_root: root);
2197	if (ret)
2198	return ERR_PTR(error: ret);
2199	break;
2200	}
2201
2202	ret = btrfs_record_root_in_trans(trans, root);
2203	if (ret)
2204	return ERR_PTR(error: ret);
2205	root = root->reloc_root;
2206
2207	/*
2208	* We could have raced with another thread which failed, so
2209	* root->reloc_root may not be set, return ENOENT in this case.
2210	*/
2211	if (!root)
2212	return ERR_PTR(error: -ENOENT);
2213
2214	if (next->new_bytenr != root->node->start) {
2215	/*
2216	* We just created the reloc root, so we shouldn't have
2217	* ->new_bytenr set and this shouldn't be in the changed
2218	* list. If it is then we have multiple roots pointing
2219	* at the same bytenr which indicates corruption, or
2220	* we've made a mistake in the backref walking code.
2221	*/
2222	ASSERT(next->new_bytenr == 0);
2223	ASSERT(list_empty(&next->list));
2224	if (next->new_bytenr || !list_empty(head: &next->list)) {
2225	btrfs_err(trans->fs_info,
2226	"bytenr %llu possibly has multiple roots pointing at the same bytenr %llu",
2227	node->bytenr, next->bytenr);
2228	return ERR_PTR(error: -EUCLEAN);
2229	}
2230
2231	next->new_bytenr = root->node->start;
2232	btrfs_put_root(root: next->root);
2233	next->root = btrfs_grab_root(root);
2234	ASSERT(next->root);
2235	list_add_tail(new: &next->list,
2236	head: &rc->backref_cache.changed);
2237	mark_block_processed(rc, node: next);
2238	break;
2239	}
2240
2241	WARN_ON(1);
2242	root = NULL;
2243	next = walk_down_backref(edges, index: &index);
2244	if (!next || next->level <= node->level)
2245	break;
2246	}
2247	if (!root) {
2248	/*
2249	* This can happen if there's fs corruption or if there's a bug
2250	* in the backref lookup code.
2251	*/
2252	ASSERT(0);
2253	return ERR_PTR(error: -ENOENT);
2254	}
2255
2256	next = node;
2257	/* setup backref node path for btrfs_reloc_cow_block */
2258	while (1) {
2259	rc->backref_cache.path[next->level] = next;
2260	if (--index < 0)
2261	break;
2262	next = edges[index]->node[UPPER];
2263	}
2264	return root;
2265	}
2266
2267	/*
2268	* Select a tree root for relocation.
2269	*
2270	* Return NULL if the block is not shareable. We should use do_relocation() in
2271	* this case.
2272	*
2273	* Return a tree root pointer if the block is shareable.
2274	* Return -ENOENT if the block is root of reloc tree.
2275	*/
2276	static noinline_for_stack
2277	struct btrfs_root *select_one_root(struct btrfs_backref_node *node)
2278	{
2279	struct btrfs_backref_node *next;
2280	struct btrfs_root *root;
2281	struct btrfs_root *fs_root = NULL;
2282	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
2283	int index = 0;
2284
2285	next = node;
2286	while (1) {
2287	cond_resched();
2288	next = walk_up_backref(node: next, edges, index: &index);
2289	root = next->root;
2290
2291	/*
2292	* This can occur if we have incomplete extent refs leading all
2293	* the way up a particular path, in this case return -EUCLEAN.
2294	*/
2295	if (!root)
2296	return ERR_PTR(error: -EUCLEAN);
2297
2298	/* No other choice for non-shareable tree */
2299	if (!test_bit(BTRFS_ROOT_SHAREABLE, &root->state))
2300	return root;
2301
2302	if (root->root_key.objectid != BTRFS_TREE_RELOC_OBJECTID)
2303	fs_root = root;
2304
2305	if (next != node)
2306	return NULL;
2307
2308	next = walk_down_backref(edges, index: &index);
2309	if (!next || next->level <= node->level)
2310	break;
2311	}
2312
2313	if (!fs_root)
2314	return ERR_PTR(error: -ENOENT);
2315	return fs_root;
2316	}
2317
2318	static noinline_for_stack
2319	u64 calcu_metadata_size(struct reloc_control *rc,
2320	struct btrfs_backref_node *node, int reserve)
2321	{
2322	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
2323	struct btrfs_backref_node *next = node;
2324	struct btrfs_backref_edge *edge;
2325	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
2326	u64 num_bytes = 0;
2327	int index = 0;
2328
2329	BUG_ON(reserve && node->processed);
2330
2331	while (next) {
2332	cond_resched();
2333	while (1) {
2334	if (next->processed && (reserve || next != node))
2335	break;
2336
2337	num_bytes += fs_info->nodesize;
2338
2339	if (list_empty(head: &next->upper))
2340	break;
2341
2342	edge = list_entry(next->upper.next,
2343	struct btrfs_backref_edge, list[LOWER]);
2344	edges[index++] = edge;
2345	next = edge->node[UPPER];
2346	}
2347	next = walk_down_backref(edges, index: &index);
2348	}
2349	return num_bytes;
2350	}
2351
2352	static int reserve_metadata_space(struct btrfs_trans_handle *trans,
2353	struct reloc_control *rc,
2354	struct btrfs_backref_node *node)
2355	{
2356	struct btrfs_root *root = rc->extent_root;
2357	struct btrfs_fs_info *fs_info = root->fs_info;
2358	u64 num_bytes;
2359	int ret;
2360	u64 tmp;
2361
2362	num_bytes = calcu_metadata_size(rc, node, reserve: 1) * 2;
2363
2364	trans->block_rsv = rc->block_rsv;
2365	rc->reserved_bytes += num_bytes;
2366
2367	/*
2368	* We are under a transaction here so we can only do limited flushing.
2369	* If we get an enospc just kick back -EAGAIN so we know to drop the
2370	* transaction and try to refill when we can flush all the things.
2371	*/
2372	ret = btrfs_block_rsv_refill(fs_info, block_rsv: rc->block_rsv, num_bytes,
2373	flush: BTRFS_RESERVE_FLUSH_LIMIT);
2374	if (ret) {
2375	tmp = fs_info->nodesize * RELOCATION_RESERVED_NODES;
2376	while (tmp <= rc->reserved_bytes)
2377	tmp <<= 1;
2378	/*
2379	* only one thread can access block_rsv at this point,
2380	* so we don't need hold lock to protect block_rsv.
2381	* we expand more reservation size here to allow enough
2382	* space for relocation and we will return earlier in
2383	* enospc case.
2384	*/
2385	rc->block_rsv->size = tmp + fs_info->nodesize *
2386	RELOCATION_RESERVED_NODES;
2387	return -EAGAIN;
2388	}
2389
2390	return 0;
2391	}
2392
2393	/*
2394	* relocate a block tree, and then update pointers in upper level
2395	* blocks that reference the block to point to the new location.
2396	*
2397	* if called by link_to_upper, the block has already been relocated.
2398	* in that case this function just updates pointers.
2399	*/
2400	static int do_relocation(struct btrfs_trans_handle *trans,
2401	struct reloc_control *rc,
2402	struct btrfs_backref_node *node,
2403	struct btrfs_key *key,
2404	struct btrfs_path *path, int lowest)
2405	{
2406	struct btrfs_backref_node *upper;
2407	struct btrfs_backref_edge *edge;
2408	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
2409	struct btrfs_root *root;
2410	struct extent_buffer *eb;
2411	u32 blocksize;
2412	u64 bytenr;
2413	int slot;
2414	int ret = 0;
2415
2416	/*
2417	* If we are lowest then this is the first time we're processing this
2418	* block, and thus shouldn't have an eb associated with it yet.
2419	*/
2420	ASSERT(!lowest || !node->eb);
2421
2422	path->lowest_level = node->level + 1;
2423	rc->backref_cache.path[node->level] = node;
2424	list_for_each_entry(edge, &node->upper, list[LOWER]) {
2425	struct btrfs_ref ref = { 0 };
2426
2427	cond_resched();
2428
2429	upper = edge->node[UPPER];
2430	root = select_reloc_root(trans, rc, node: upper, edges);
2431	if (IS_ERR(ptr: root)) {
2432	ret = PTR_ERR(ptr: root);
2433	goto next;
2434	}
2435
2436	if (upper->eb && !upper->locked) {
2437	if (!lowest) {
2438	ret = btrfs_bin_search(eb: upper->eb, first_slot: 0, key, slot: &slot);
2439	if (ret < 0)
2440	goto next;
2441	BUG_ON(ret);
2442	bytenr = btrfs_node_blockptr(eb: upper->eb, nr: slot);
2443	if (node->eb->start == bytenr)
2444	goto next;
2445	}
2446	btrfs_backref_drop_node_buffer(node: upper);
2447	}
2448
2449	if (!upper->eb) {
2450	ret = btrfs_search_slot(trans, root, key, p: path, ins_len: 0, cow: 1);
2451	if (ret) {
2452	if (ret > 0)
2453	ret = -ENOENT;
2454
2455	btrfs_release_path(p: path);
2456	break;
2457	}
2458
2459	if (!upper->eb) {
2460	upper->eb = path->nodes[upper->level];
2461	path->nodes[upper->level] = NULL;
2462	} else {
2463	BUG_ON(upper->eb != path->nodes[upper->level]);
2464	}
2465
2466	upper->locked = 1;
2467	path->locks[upper->level] = 0;
2468
2469	slot = path->slots[upper->level];
2470	btrfs_release_path(p: path);
2471	} else {
2472	ret = btrfs_bin_search(eb: upper->eb, first_slot: 0, key, slot: &slot);
2473	if (ret < 0)
2474	goto next;
2475	BUG_ON(ret);
2476	}
2477
2478	bytenr = btrfs_node_blockptr(eb: upper->eb, nr: slot);
2479	if (lowest) {
2480	if (bytenr != node->bytenr) {
2481	btrfs_err(root->fs_info,
2482	"lowest leaf/node mismatch: bytenr %llu node->bytenr %llu slot %d upper %llu",
2483	bytenr, node->bytenr, slot,
2484	upper->eb->start);
2485	ret = -EIO;
2486	goto next;
2487	}
2488	} else {
2489	if (node->eb->start == bytenr)
2490	goto next;
2491	}
2492
2493	blocksize = root->fs_info->nodesize;
2494	eb = btrfs_read_node_slot(parent: upper->eb, slot);
2495	if (IS_ERR(ptr: eb)) {
2496	ret = PTR_ERR(ptr: eb);
2497	goto next;
2498	}
2499	btrfs_tree_lock(eb);
2500
2501	if (!node->eb) {
2502	ret = btrfs_cow_block(trans, root, buf: eb, parent: upper->eb,
2503	parent_slot: slot, cow_ret: &eb, nest: BTRFS_NESTING_COW);
2504	btrfs_tree_unlock(eb);
2505	free_extent_buffer(eb);
2506	if (ret < 0)
2507	goto next;
2508	/*
2509	* We've just COWed this block, it should have updated
2510	* the correct backref node entry.
2511	*/
2512	ASSERT(node->eb == eb);
2513	} else {
2514	btrfs_set_node_blockptr(eb: upper->eb, nr: slot,
2515	val: node->eb->start);
2516	btrfs_set_node_ptr_generation(eb: upper->eb, nr: slot,
2517	val: trans->transid);
2518	btrfs_mark_buffer_dirty(trans, buf: upper->eb);
2519
2520	btrfs_init_generic_ref(generic_ref: &ref, action: BTRFS_ADD_DELAYED_REF,
2521	bytenr: node->eb->start, len: blocksize,
2522	parent: upper->eb->start,
2523	owning_root: btrfs_header_owner(eb: upper->eb));
2524	btrfs_init_tree_ref(generic_ref: &ref, level: node->level,
2525	root: btrfs_header_owner(eb: upper->eb),
2526	mod_root: root->root_key.objectid, skip_qgroup: false);
2527	ret = btrfs_inc_extent_ref(trans, generic_ref: &ref);
2528	if (!ret)
2529	ret = btrfs_drop_subtree(trans, root, node: eb,
2530	parent: upper->eb);
2531	if (ret)
2532	btrfs_abort_transaction(trans, ret);
2533	}
2534	next:
2535	if (!upper->pending)
2536	btrfs_backref_drop_node_buffer(node: upper);
2537	else
2538	btrfs_backref_unlock_node_buffer(node: upper);
2539	if (ret)
2540	break;
2541	}
2542
2543	if (!ret && node->pending) {
2544	btrfs_backref_drop_node_buffer(node);
2545	list_move_tail(list: &node->list, head: &rc->backref_cache.changed);
2546	node->pending = 0;
2547	}
2548
2549	path->lowest_level = 0;
2550
2551	/*
2552	* We should have allocated all of our space in the block rsv and thus
2553	* shouldn't ENOSPC.
2554	*/
2555	ASSERT(ret != -ENOSPC);
2556	return ret;
2557	}
2558
2559	static int link_to_upper(struct btrfs_trans_handle *trans,
2560	struct reloc_control *rc,
2561	struct btrfs_backref_node *node,
2562	struct btrfs_path *path)
2563	{
2564	struct btrfs_key key;
2565
2566	btrfs_node_key_to_cpu(eb: node->eb, cpu_key: &key, nr: 0);
2567	return do_relocation(trans, rc, node, key: &key, path, lowest: 0);
2568	}
2569
2570	static int finish_pending_nodes(struct btrfs_trans_handle *trans,
2571	struct reloc_control *rc,
2572	struct btrfs_path *path, int err)
2573	{
2574	LIST_HEAD(list);
2575	struct btrfs_backref_cache *cache = &rc->backref_cache;
2576	struct btrfs_backref_node *node;
2577	int level;
2578	int ret;
2579
2580	for (level = 0; level < BTRFS_MAX_LEVEL; level++) {
2581	while (!list_empty(head: &cache->pending[level])) {
2582	node = list_entry(cache->pending[level].next,
2583	struct btrfs_backref_node, list);
2584	list_move_tail(list: &node->list, head: &list);
2585	BUG_ON(!node->pending);
2586
2587	if (!err) {
2588	ret = link_to_upper(trans, rc, node, path);
2589	if (ret < 0)
2590	err = ret;
2591	}
2592	}
2593	list_splice_init(list: &list, head: &cache->pending[level]);
2594	}
2595	return err;
2596	}
2597
2598	/*
2599	* mark a block and all blocks directly/indirectly reference the block
2600	* as processed.
2601	*/
2602	static void update_processed_blocks(struct reloc_control *rc,
2603	struct btrfs_backref_node *node)
2604	{
2605	struct btrfs_backref_node *next = node;
2606	struct btrfs_backref_edge *edge;
2607	struct btrfs_backref_edge *edges[BTRFS_MAX_LEVEL - 1];
2608	int index = 0;
2609
2610	while (next) {
2611	cond_resched();
2612	while (1) {
2613	if (next->processed)
2614	break;
2615
2616	mark_block_processed(rc, node: next);
2617
2618	if (list_empty(head: &next->upper))
2619	break;
2620
2621	edge = list_entry(next->upper.next,
2622	struct btrfs_backref_edge, list[LOWER]);
2623	edges[index++] = edge;
2624	next = edge->node[UPPER];
2625	}
2626	next = walk_down_backref(edges, index: &index);
2627	}
2628	}
2629
2630	static int tree_block_processed(u64 bytenr, struct reloc_control *rc)
2631	{
2632	u32 blocksize = rc->extent_root->fs_info->nodesize;
2633
2634	if (test_range_bit(tree: &rc->processed_blocks, start: bytenr,
2635	end: bytenr + blocksize - 1, bit: EXTENT_DIRTY, NULL))
2636	return 1;
2637	return 0;
2638	}
2639
2640	static int get_tree_block_key(struct btrfs_fs_info *fs_info,
2641	struct tree_block *block)
2642	{
2643	struct btrfs_tree_parent_check check = {
2644	.level = block->level,
2645	.owner_root = block->owner,
2646	.transid = block->key.offset
2647	};
2648	struct extent_buffer *eb;
2649
2650	eb = read_tree_block(fs_info, bytenr: block->bytenr, check: &check);
2651	if (IS_ERR(ptr: eb))
2652	return PTR_ERR(ptr: eb);
2653	if (!extent_buffer_uptodate(eb)) {
2654	free_extent_buffer(eb);
2655	return -EIO;
2656	}
2657	if (block->level == 0)
2658	btrfs_item_key_to_cpu(eb, cpu_key: &block->key, nr: 0);
2659	else
2660	btrfs_node_key_to_cpu(eb, cpu_key: &block->key, nr: 0);
2661	free_extent_buffer(eb);
2662	block->key_ready = true;
2663	return 0;
2664	}
2665
2666	/*
2667	* helper function to relocate a tree block
2668	*/
2669	static int relocate_tree_block(struct btrfs_trans_handle *trans,
2670	struct reloc_control *rc,
2671	struct btrfs_backref_node *node,
2672	struct btrfs_key *key,
2673	struct btrfs_path *path)
2674	{
2675	struct btrfs_root *root;
2676	int ret = 0;
2677
2678	if (!node)
2679	return 0;
2680
2681	/*
2682	* If we fail here we want to drop our backref_node because we are going
2683	* to start over and regenerate the tree for it.
2684	*/
2685	ret = reserve_metadata_space(trans, rc, node);
2686	if (ret)
2687	goto out;
2688
2689	BUG_ON(node->processed);
2690	root = select_one_root(node);
2691	if (IS_ERR(ptr: root)) {
2692	ret = PTR_ERR(ptr: root);
2693
2694	/* See explanation in select_one_root for the -EUCLEAN case. */
2695	ASSERT(ret == -ENOENT);
2696	if (ret == -ENOENT) {
2697	ret = 0;
2698	update_processed_blocks(rc, node);
2699	}
2700	goto out;
2701	}
2702
2703	if (root) {
2704	if (test_bit(BTRFS_ROOT_SHAREABLE, &root->state)) {
2705	/*
2706	* This block was the root block of a root, and this is
2707	* the first time we're processing the block and thus it
2708	* should not have had the ->new_bytenr modified and
2709	* should have not been included on the changed list.
2710	*
2711	* However in the case of corruption we could have
2712	* multiple refs pointing to the same block improperly,
2713	* and thus we would trip over these checks. ASSERT()
2714	* for the developer case, because it could indicate a
2715	* bug in the backref code, however error out for a
2716	* normal user in the case of corruption.
2717	*/
2718	ASSERT(node->new_bytenr == 0);
2719	ASSERT(list_empty(&node->list));
2720	if (node->new_bytenr || !list_empty(head: &node->list)) {
2721	btrfs_err(root->fs_info,
2722	"bytenr %llu has improper references to it",
2723	node->bytenr);
2724	ret = -EUCLEAN;
2725	goto out;
2726	}
2727	ret = btrfs_record_root_in_trans(trans, root);
2728	if (ret)
2729	goto out;
2730	/*
2731	* Another thread could have failed, need to check if we
2732	* have reloc_root actually set.
2733	*/
2734	if (!root->reloc_root) {
2735	ret = -ENOENT;
2736	goto out;
2737	}
2738	root = root->reloc_root;
2739	node->new_bytenr = root->node->start;
2740	btrfs_put_root(root: node->root);
2741	node->root = btrfs_grab_root(root);
2742	ASSERT(node->root);
2743	list_add_tail(new: &node->list, head: &rc->backref_cache.changed);
2744	} else {
2745	path->lowest_level = node->level;
2746	if (root == root->fs_info->chunk_root)
2747	btrfs_reserve_chunk_metadata(trans, is_item_insertion: false);
2748	ret = btrfs_search_slot(trans, root, key, p: path, ins_len: 0, cow: 1);
2749	btrfs_release_path(p: path);
2750	if (root == root->fs_info->chunk_root)
2751	btrfs_trans_release_chunk_metadata(trans);
2752	if (ret > 0)
2753	ret = 0;
2754	}
2755	if (!ret)
2756	update_processed_blocks(rc, node);
2757	} else {
2758	ret = do_relocation(trans, rc, node, key, path, lowest: 1);
2759	}
2760	out:
2761	if (ret || node->level == 0 || node->cowonly)
2762	btrfs_backref_cleanup_node(cache: &rc->backref_cache, node);
2763	return ret;
2764	}
2765
2766	/*
2767	* relocate a list of blocks
2768	*/
2769	static noinline_for_stack
2770	int relocate_tree_blocks(struct btrfs_trans_handle *trans,
2771	struct reloc_control *rc, struct rb_root *blocks)
2772	{
2773	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
2774	struct btrfs_backref_node *node;
2775	struct btrfs_path *path;
2776	struct tree_block *block;
2777	struct tree_block *next;
2778	int ret;
2779	int err = 0;
2780
2781	path = btrfs_alloc_path();
2782	if (!path) {
2783	err = -ENOMEM;
2784	goto out_free_blocks;
2785	}
2786
2787	/* Kick in readahead for tree blocks with missing keys */
2788	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
2789	if (!block->key_ready)
2790	btrfs_readahead_tree_block(fs_info, bytenr: block->bytenr,
2791	owner_root: block->owner, gen: 0,
2792	level: block->level);
2793	}
2794
2795	/* Get first keys */
2796	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
2797	if (!block->key_ready) {
2798	err = get_tree_block_key(fs_info, block);
2799	if (err)
2800	goto out_free_path;
2801	}
2802	}
2803
2804	/* Do tree relocation */
2805	rbtree_postorder_for_each_entry_safe(block, next, blocks, rb_node) {
2806	node = build_backref_tree(trans, rc, node_key: &block->key,
2807	level: block->level, bytenr: block->bytenr);
2808	if (IS_ERR(ptr: node)) {
2809	err = PTR_ERR(ptr: node);
2810	goto out;
2811	}
2812
2813	ret = relocate_tree_block(trans, rc, node, key: &block->key,
2814	path);
2815	if (ret < 0) {
2816	err = ret;
2817	break;
2818	}
2819	}
2820	out:
2821	err = finish_pending_nodes(trans, rc, path, err);
2822
2823	out_free_path:
2824	btrfs_free_path(p: path);
2825	out_free_blocks:
2826	free_block_list(blocks);
2827	return err;
2828	}
2829
2830	static noinline_for_stack int prealloc_file_extent_cluster(
2831	struct btrfs_inode *inode,
2832	const struct file_extent_cluster *cluster)
2833	{
2834	u64 alloc_hint = 0;
2835	u64 start;
2836	u64 end;
2837	u64 offset = inode->index_cnt;
2838	u64 num_bytes;
2839	int nr;
2840	int ret = 0;
2841	u64 i_size = i_size_read(inode: &inode->vfs_inode);
2842	u64 prealloc_start = cluster->start - offset;
2843	u64 prealloc_end = cluster->end - offset;
2844	u64 cur_offset = prealloc_start;
2845
2846	/*
2847	* For subpage case, previous i_size may not be aligned to PAGE_SIZE.
2848	* This means the range [i_size, PAGE_END + 1) is filled with zeros by
2849	* btrfs_do_readpage() call of previously relocated file cluster.
2850	*
2851	* If the current cluster starts in the above range, btrfs_do_readpage()
2852	* will skip the read, and relocate_one_page() will later writeback
2853	* the padding zeros as new data, causing data corruption.
2854	*
2855	* Here we have to manually invalidate the range (i_size, PAGE_END + 1).
2856	*/
2857	if (!PAGE_ALIGNED(i_size)) {
2858	struct address_space *mapping = inode->vfs_inode.i_mapping;
2859	struct btrfs_fs_info *fs_info = inode->root->fs_info;
2860	const u32 sectorsize = fs_info->sectorsize;
2861	struct page *page;
2862
2863	ASSERT(sectorsize < PAGE_SIZE);
2864	ASSERT(IS_ALIGNED(i_size, sectorsize));
2865
2866	/*
2867	* Subpage can't handle page with DIRTY but without UPTODATE
2868	* bit as it can lead to the following deadlock:
2869	*
2870	* btrfs_read_folio()
2871	* | Page already *locked*
2872	* |- btrfs_lock_and_flush_ordered_range()
2873	* |- btrfs_start_ordered_extent()
2874	* |- extent_write_cache_pages()
2875	* |- lock_page()
2876	* We try to lock the page we already hold.
2877	*
2878	* Here we just writeback the whole data reloc inode, so that
2879	* we will be ensured to have no dirty range in the page, and
2880	* are safe to clear the uptodate bits.
2881	*
2882	* This shouldn't cause too much overhead, as we need to write
2883	* the data back anyway.
2884	*/
2885	ret = filemap_write_and_wait(mapping);
2886	if (ret < 0)
2887	return ret;
2888
2889	clear_extent_bits(tree: &inode->io_tree, start: i_size,
2890	round_up(i_size, PAGE_SIZE) - 1,
2891	bits: EXTENT_UPTODATE);
2892	page = find_lock_page(mapping, index: i_size >> PAGE_SHIFT);
2893	/*
2894	* If page is freed we don't need to do anything then, as we
2895	* will re-read the whole page anyway.
2896	*/
2897	if (page) {
2898	btrfs_subpage_clear_uptodate(fs_info, page, start: i_size,
2899	round_up(i_size, PAGE_SIZE) - i_size);
2900	unlock_page(page);
2901	put_page(page);
2902	}
2903	}
2904
2905	BUG_ON(cluster->start != cluster->boundary[0]);
2906	ret = btrfs_alloc_data_chunk_ondemand(inode,
2907	bytes: prealloc_end + 1 - prealloc_start);
2908	if (ret)
2909	return ret;
2910
2911	btrfs_inode_lock(inode, ilock_flags: 0);
2912	for (nr = 0; nr < cluster->nr; nr++) {
2913	struct extent_state *cached_state = NULL;
2914
2915	start = cluster->boundary[nr] - offset;
2916	if (nr + 1 < cluster->nr)
2917	end = cluster->boundary[nr + 1] - 1 - offset;
2918	else
2919	end = cluster->end - offset;
2920
2921	lock_extent(tree: &inode->io_tree, start, end, cached: &cached_state);
2922	num_bytes = end + 1 - start;
2923	ret = btrfs_prealloc_file_range(inode: &inode->vfs_inode, mode: 0, start,
2924	num_bytes, min_size: num_bytes,
2925	actual_len: end + 1, alloc_hint: &alloc_hint);
2926	cur_offset = end + 1;
2927	unlock_extent(tree: &inode->io_tree, start, end, cached: &cached_state);
2928	if (ret)
2929	break;
2930	}
2931	btrfs_inode_unlock(inode, ilock_flags: 0);
2932
2933	if (cur_offset < prealloc_end)
2934	btrfs_free_reserved_data_space_noquota(fs_info: inode->root->fs_info,
2935	len: prealloc_end + 1 - cur_offset);
2936	return ret;
2937	}
2938
2939	static noinline_for_stack int setup_relocation_extent_mapping(struct inode *inode,
2940	u64 start, u64 end, u64 block_start)
2941	{
2942	struct extent_map *em;
2943	struct extent_state *cached_state = NULL;
2944	int ret = 0;
2945
2946	em = alloc_extent_map();
2947	if (!em)
2948	return -ENOMEM;
2949
2950	em->start = start;
2951	em->len = end + 1 - start;
2952	em->block_len = em->len;
2953	em->block_start = block_start;
2954	set_bit(nr: EXTENT_FLAG_PINNED, addr: &em->flags);
2955
2956	lock_extent(tree: &BTRFS_I(inode)->io_tree, start, end, cached: &cached_state);
2957	ret = btrfs_replace_extent_map_range(inode: BTRFS_I(inode), new_em: em, modified: false);
2958	unlock_extent(tree: &BTRFS_I(inode)->io_tree, start, end, cached: &cached_state);
2959	free_extent_map(em);
2960
2961	return ret;
2962	}
2963
2964	/*
2965	* Allow error injection to test balance/relocation cancellation
2966	*/
2967	noinline int btrfs_should_cancel_balance(const struct btrfs_fs_info *fs_info)
2968	{
2969	return atomic_read(v: &fs_info->balance_cancel_req) ||
2970	atomic_read(v: &fs_info->reloc_cancel_req) ||
2971	fatal_signal_pending(current);
2972	}
2973	ALLOW_ERROR_INJECTION(btrfs_should_cancel_balance, TRUE);
2974
2975	static u64 get_cluster_boundary_end(const struct file_extent_cluster *cluster,
2976	int cluster_nr)
2977	{
2978	/* Last extent, use cluster end directly */
2979	if (cluster_nr >= cluster->nr - 1)
2980	return cluster->end;
2981
2982	/* Use next boundary start*/
2983	return cluster->boundary[cluster_nr + 1] - 1;
2984	}
2985
2986	static int relocate_one_page(struct inode *inode, struct file_ra_state *ra,
2987	const struct file_extent_cluster *cluster,
2988	int *cluster_nr, unsigned long page_index)
2989	{
2990	struct btrfs_fs_info *fs_info = btrfs_sb(sb: inode->i_sb);
2991	u64 offset = BTRFS_I(inode)->index_cnt;
2992	const unsigned long last_index = (cluster->end - offset) >> PAGE_SHIFT;
2993	gfp_t mask = btrfs_alloc_write_mask(mapping: inode->i_mapping);
2994	struct page *page;
2995	u64 page_start;
2996	u64 page_end;
2997	u64 cur;
2998	int ret;
2999
3000	ASSERT(page_index <= last_index);
3001	page = find_lock_page(mapping: inode->i_mapping, index: page_index);
3002	if (!page) {
3003	page_cache_sync_readahead(mapping: inode->i_mapping, ra, NULL,
3004	index: page_index, req_count: last_index + 1 - page_index);
3005	page = find_or_create_page(mapping: inode->i_mapping, index: page_index, gfp_mask: mask);
3006	if (!page)
3007	return -ENOMEM;
3008	}
3009
3010	if (PageReadahead(page))
3011	page_cache_async_readahead(mapping: inode->i_mapping, ra, NULL,
3012	page_folio(page), index: page_index,
3013	req_count: last_index + 1 - page_index);
3014
3015	if (!PageUptodate(page)) {
3016	btrfs_read_folio(NULL, page_folio(page));
3017	lock_page(page);
3018	if (!PageUptodate(page)) {
3019	ret = -EIO;
3020	goto release_page;
3021	}
3022	}
3023
3024	/*
3025	* We could have lost page private when we dropped the lock to read the
3026	* page above, make sure we set_page_extent_mapped here so we have any
3027	* of the subpage blocksize stuff we need in place.
3028	*/
3029	ret = set_page_extent_mapped(page);
3030	if (ret < 0)
3031	goto release_page;
3032
3033	page_start = page_offset(page);
3034	page_end = page_start + PAGE_SIZE - 1;
3035
3036	/*
3037	* Start from the cluster, as for subpage case, the cluster can start
3038	* inside the page.
3039	*/
3040	cur = max(page_start, cluster->boundary[*cluster_nr] - offset);
3041	while (cur <= page_end) {
3042	struct extent_state *cached_state = NULL;
3043	u64 extent_start = cluster->boundary[*cluster_nr] - offset;
3044	u64 extent_end = get_cluster_boundary_end(cluster,
3045	cluster_nr: *cluster_nr) - offset;
3046	u64 clamped_start = max(page_start, extent_start);
3047	u64 clamped_end = min(page_end, extent_end);
3048	u32 clamped_len = clamped_end + 1 - clamped_start;
3049
3050	/* Reserve metadata for this range */
3051	ret = btrfs_delalloc_reserve_metadata(inode: BTRFS_I(inode),
3052	num_bytes: clamped_len, disk_num_bytes: clamped_len,
3053	noflush: false);
3054	if (ret)
3055	goto release_page;
3056
3057	/* Mark the range delalloc and dirty for later writeback */
3058	lock_extent(tree: &BTRFS_I(inode)->io_tree, start: clamped_start, end: clamped_end,
3059	cached: &cached_state);
3060	ret = btrfs_set_extent_delalloc(inode: BTRFS_I(inode), start: clamped_start,
3061	end: clamped_end, extra_bits: 0, cached_state: &cached_state);
3062	if (ret) {
3063	clear_extent_bit(tree: &BTRFS_I(inode)->io_tree,
3064	start: clamped_start, end: clamped_end,
3065	bits: EXTENT_LOCKED | EXTENT_BOUNDARY,
3066	cached: &cached_state);
3067	btrfs_delalloc_release_metadata(inode: BTRFS_I(inode),
3068	num_bytes: clamped_len, qgroup_free: true);
3069	btrfs_delalloc_release_extents(inode: BTRFS_I(inode),
3070	num_bytes: clamped_len);
3071	goto release_page;
3072	}
3073	btrfs_page_set_dirty(fs_info, page, start: clamped_start, len: clamped_len);
3074
3075	/*
3076	* Set the boundary if it's inside the page.
3077	* Data relocation requires the destination extents to have the
3078	* same size as the source.
3079	* EXTENT_BOUNDARY bit prevents current extent from being merged
3080	* with previous extent.
3081	*/
3082	if (in_range(cluster->boundary[*cluster_nr] - offset,
3083	page_start, PAGE_SIZE)) {
3084	u64 boundary_start = cluster->boundary[*cluster_nr] -
3085	offset;
3086	u64 boundary_end = boundary_start +
3087	fs_info->sectorsize - 1;
3088
3089	set_extent_bit(tree: &BTRFS_I(inode)->io_tree,
3090	start: boundary_start, end: boundary_end,
3091	bits: EXTENT_BOUNDARY, NULL);
3092	}
3093	unlock_extent(tree: &BTRFS_I(inode)->io_tree, start: clamped_start, end: clamped_end,
3094	cached: &cached_state);
3095	btrfs_delalloc_release_extents(inode: BTRFS_I(inode), num_bytes: clamped_len);
3096	cur += clamped_len;
3097
3098	/* Crossed extent end, go to next extent */
3099	if (cur >= extent_end) {
3100	(*cluster_nr)++;
3101	/* Just finished the last extent of the cluster, exit. */
3102	if (*cluster_nr >= cluster->nr)
3103	break;
3104	}
3105	}
3106	unlock_page(page);
3107	put_page(page);
3108
3109	balance_dirty_pages_ratelimited(mapping: inode->i_mapping);
3110	btrfs_throttle(fs_info);
3111	if (btrfs_should_cancel_balance(fs_info))
3112	ret = -ECANCELED;
3113	return ret;
3114
3115	release_page:
3116	unlock_page(page);
3117	put_page(page);
3118	return ret;
3119	}
3120
3121	static int relocate_file_extent_cluster(struct inode *inode,
3122	const struct file_extent_cluster *cluster)
3123	{
3124	u64 offset = BTRFS_I(inode)->index_cnt;
3125	unsigned long index;
3126	unsigned long last_index;
3127	struct file_ra_state *ra;
3128	int cluster_nr = 0;
3129	int ret = 0;
3130
3131	if (!cluster->nr)
3132	return 0;
3133
3134	ra = kzalloc(size: sizeof(*ra), GFP_NOFS);
3135	if (!ra)
3136	return -ENOMEM;
3137
3138	ret = prealloc_file_extent_cluster(inode: BTRFS_I(inode), cluster);
3139	if (ret)
3140	goto out;
3141
3142	file_ra_state_init(ra, mapping: inode->i_mapping);
3143
3144	ret = setup_relocation_extent_mapping(inode, start: cluster->start - offset,
3145	end: cluster->end - offset, block_start: cluster->start);
3146	if (ret)
3147	goto out;
3148
3149	last_index = (cluster->end - offset) >> PAGE_SHIFT;
3150	for (index = (cluster->start - offset) >> PAGE_SHIFT;
3151	index <= last_index && !ret; index++)
3152	ret = relocate_one_page(inode, ra, cluster, cluster_nr: &cluster_nr, page_index: index);
3153	if (ret == 0)
3154	WARN_ON(cluster_nr != cluster->nr);
3155	out:
3156	kfree(objp: ra);
3157	return ret;
3158	}
3159
3160	static noinline_for_stack int relocate_data_extent(struct inode *inode,
3161	const struct btrfs_key *extent_key,
3162	struct file_extent_cluster *cluster)
3163	{
3164	int ret;
3165	struct btrfs_root *root = BTRFS_I(inode)->root;
3166
3167	if (cluster->nr > 0 && extent_key->objectid != cluster->end + 1) {
3168	ret = relocate_file_extent_cluster(inode, cluster);
3169	if (ret)
3170	return ret;
3171	cluster->nr = 0;
3172	}
3173
3174	/*
3175	* Under simple quotas, we set root->relocation_src_root when we find
3176	* the extent. If adjacent extents have different owners, we can't merge
3177	* them while relocating. Handle this by storing the owning root that
3178	* started a cluster and if we see an extent from a different root break
3179	* cluster formation (just like the above case of non-adjacent extents).
3180	*
3181	* Without simple quotas, relocation_src_root is always 0, so we should
3182	* never see a mismatch, and it should have no effect on relocation
3183	* clusters.
3184	*/
3185	if (cluster->nr > 0 && cluster->owning_root != root->relocation_src_root) {
3186	u64 tmp = root->relocation_src_root;
3187
3188	/*
3189	* root->relocation_src_root is the state that actually affects
3190	* the preallocation we do here, so set it to the root owning
3191	* the cluster we need to relocate.
3192	*/
3193	root->relocation_src_root = cluster->owning_root;
3194	ret = relocate_file_extent_cluster(inode, cluster);
3195	if (ret)
3196	return ret;
3197	cluster->nr = 0;
3198	/* And reset it back for the current extent's owning root. */
3199	root->relocation_src_root = tmp;
3200	}
3201
3202	if (!cluster->nr) {
3203	cluster->start = extent_key->objectid;
3204	cluster->owning_root = root->relocation_src_root;
3205	}
3206	else
3207	BUG_ON(cluster->nr >= MAX_EXTENTS);
3208	cluster->end = extent_key->objectid + extent_key->offset - 1;
3209	cluster->boundary[cluster->nr] = extent_key->objectid;
3210	cluster->nr++;
3211
3212	if (cluster->nr >= MAX_EXTENTS) {
3213	ret = relocate_file_extent_cluster(inode, cluster);
3214	if (ret)
3215	return ret;
3216	cluster->nr = 0;
3217	}
3218	return 0;
3219	}
3220
3221	/*
3222	* helper to add a tree block to the list.
3223	* the major work is getting the generation and level of the block
3224	*/
3225	static int add_tree_block(struct reloc_control *rc,
3226	const struct btrfs_key *extent_key,
3227	struct btrfs_path *path,
3228	struct rb_root *blocks)
3229	{
3230	struct extent_buffer *eb;
3231	struct btrfs_extent_item *ei;
3232	struct btrfs_tree_block_info *bi;
3233	struct tree_block *block;
3234	struct rb_node *rb_node;
3235	u32 item_size;
3236	int level = -1;
3237	u64 generation;
3238	u64 owner = 0;
3239
3240	eb = path->nodes[0];
3241	item_size = btrfs_item_size(eb, slot: path->slots[0]);
3242
3243	if (extent_key->type == BTRFS_METADATA_ITEM_KEY ||
3244	item_size >= sizeof(*ei) + sizeof(*bi)) {
3245	unsigned long ptr = 0, end;
3246
3247	ei = btrfs_item_ptr(eb, path->slots[0],
3248	struct btrfs_extent_item);
3249	end = (unsigned long)ei + item_size;
3250	if (extent_key->type == BTRFS_EXTENT_ITEM_KEY) {
3251	bi = (struct btrfs_tree_block_info *)(ei + 1);
3252	level = btrfs_tree_block_level(eb, s: bi);
3253	ptr = (unsigned long)(bi + 1);
3254	} else {
3255	level = (int)extent_key->offset;
3256	ptr = (unsigned long)(ei + 1);
3257	}
3258	generation = btrfs_extent_generation(eb, s: ei);
3259
3260	/*
3261	* We're reading random blocks without knowing their owner ahead
3262	* of time. This is ok most of the time, as all reloc roots and
3263	* fs roots have the same lock type. However normal trees do
3264	* not, and the only way to know ahead of time is to read the
3265	* inline ref offset. We know it's an fs root if
3266	*
3267	* 1. There's more than one ref.
3268	* 2. There's a SHARED_DATA_REF_KEY set.
3269	* 3. FULL_BACKREF is set on the flags.
3270	*
3271	* Otherwise it's safe to assume that the ref offset == the
3272	* owner of this block, so we can use that when calling
3273	* read_tree_block.
3274	*/
3275	if (btrfs_extent_refs(eb, s: ei) == 1 &&
3276	!(btrfs_extent_flags(eb, s: ei) &
3277	BTRFS_BLOCK_FLAG_FULL_BACKREF) &&
3278	ptr < end) {
3279	struct btrfs_extent_inline_ref *iref;
3280	int type;
3281
3282	iref = (struct btrfs_extent_inline_ref *)ptr;
3283	type = btrfs_get_extent_inline_ref_type(eb, iref,
3284	is_data: BTRFS_REF_TYPE_BLOCK);
3285	if (type == BTRFS_REF_TYPE_INVALID)
3286	return -EINVAL;
3287	if (type == BTRFS_TREE_BLOCK_REF_KEY)
3288	owner = btrfs_extent_inline_ref_offset(eb, s: iref);
3289	}
3290	} else {
3291	btrfs_print_leaf(l: eb);
3292	btrfs_err(rc->block_group->fs_info,
3293	"unrecognized tree backref at tree block %llu slot %u",
3294	eb->start, path->slots[0]);
3295	btrfs_release_path(p: path);
3296	return -EUCLEAN;
3297	}
3298
3299	btrfs_release_path(p: path);
3300
3301	BUG_ON(level == -1);
3302
3303	block = kmalloc(size: sizeof(*block), GFP_NOFS);
3304	if (!block)
3305	return -ENOMEM;
3306
3307	block->bytenr = extent_key->objectid;
3308	block->key.objectid = rc->extent_root->fs_info->nodesize;
3309	block->key.offset = generation;
3310	block->level = level;
3311	block->key_ready = false;
3312	block->owner = owner;
3313
3314	rb_node = rb_simple_insert(root: blocks, bytenr: block->bytenr, node: &block->rb_node);
3315	if (rb_node)
3316	btrfs_backref_panic(fs_info: rc->extent_root->fs_info, bytenr: block->bytenr,
3317	error: -EEXIST);
3318
3319	return 0;
3320	}
3321
3322	/*
3323	* helper to add tree blocks for backref of type BTRFS_SHARED_DATA_REF_KEY
3324	*/
3325	static int __add_tree_block(struct reloc_control *rc,
3326	u64 bytenr, u32 blocksize,
3327	struct rb_root *blocks)
3328	{
3329	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
3330	struct btrfs_path *path;
3331	struct btrfs_key key;
3332	int ret;
3333	bool skinny = btrfs_fs_incompat(fs_info, SKINNY_METADATA);
3334
3335	if (tree_block_processed(bytenr, rc))
3336	return 0;
3337
3338	if (rb_simple_search(root: blocks, bytenr))
3339	return 0;
3340
3341	path = btrfs_alloc_path();
3342	if (!path)
3343	return -ENOMEM;
3344	again:
3345	key.objectid = bytenr;
3346	if (skinny) {
3347	key.type = BTRFS_METADATA_ITEM_KEY;
3348	key.offset = (u64)-1;
3349	} else {
3350	key.type = BTRFS_EXTENT_ITEM_KEY;
3351	key.offset = blocksize;
3352	}
3353
3354	path->search_commit_root = 1;
3355	path->skip_locking = 1;
3356	ret = btrfs_search_slot(NULL, root: rc->extent_root, key: &key, p: path, ins_len: 0, cow: 0);
3357	if (ret < 0)
3358	goto out;
3359
3360	if (ret > 0 && skinny) {
3361	if (path->slots[0]) {
3362	path->slots[0]--;
3363	btrfs_item_key_to_cpu(eb: path->nodes[0], cpu_key: &key,
3364	nr: path->slots[0]);
3365	if (key.objectid == bytenr &&
3366	(key.type == BTRFS_METADATA_ITEM_KEY ||
3367	(key.type == BTRFS_EXTENT_ITEM_KEY &&
3368	key.offset == blocksize)))
3369	ret = 0;
3370	}
3371
3372	if (ret) {
3373	skinny = false;
3374	btrfs_release_path(p: path);
3375	goto again;
3376	}
3377	}
3378	if (ret) {
3379	ASSERT(ret == 1);
3380	btrfs_print_leaf(l: path->nodes[0]);
3381	btrfs_err(fs_info,
3382	"tree block extent item (%llu) is not found in extent tree",
3383	bytenr);
3384	WARN_ON(1);
3385	ret = -EINVAL;
3386	goto out;
3387	}
3388
3389	ret = add_tree_block(rc, extent_key: &key, path, blocks);
3390	out:
3391	btrfs_free_path(p: path);
3392	return ret;
3393	}
3394
3395	static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
3396	struct btrfs_block_group *block_group,
3397	struct inode *inode,
3398	u64 ino)
3399	{
3400	struct btrfs_root *root = fs_info->tree_root;
3401	struct btrfs_trans_handle *trans;
3402	int ret = 0;
3403
3404	if (inode)
3405	goto truncate;
3406
3407	inode = btrfs_iget(s: fs_info->sb, ino, root);
3408	if (IS_ERR(ptr: inode))
3409	return -ENOENT;
3410
3411	truncate:
3412	ret = btrfs_check_trunc_cache_free_space(fs_info,
3413	rsv: &fs_info->global_block_rsv);
3414	if (ret)
3415	goto out;
3416
3417	trans = btrfs_join_transaction(root);
3418	if (IS_ERR(ptr: trans)) {
3419	ret = PTR_ERR(ptr: trans);
3420	goto out;
3421	}
3422
3423	ret = btrfs_truncate_free_space_cache(trans, block_group, inode);
3424
3425	btrfs_end_transaction(trans);
3426	btrfs_btree_balance_dirty(fs_info);
3427	out:
3428	iput(inode);
3429	return ret;
3430	}
3431
3432	/*
3433	* Locate the free space cache EXTENT_DATA in root tree leaf and delete the
3434	* cache inode, to avoid free space cache data extent blocking data relocation.
3435	*/
3436	static int delete_v1_space_cache(struct extent_buffer *leaf,
3437	struct btrfs_block_group *block_group,
3438	u64 data_bytenr)
3439	{
3440	u64 space_cache_ino;
3441	struct btrfs_file_extent_item *ei;
3442	struct btrfs_key key;
3443	bool found = false;
3444	int i;
3445	int ret;
3446
3447	if (btrfs_header_owner(eb: leaf) != BTRFS_ROOT_TREE_OBJECTID)
3448	return 0;
3449
3450	for (i = 0; i < btrfs_header_nritems(eb: leaf); i++) {
3451	u8 type;
3452
3453	btrfs_item_key_to_cpu(eb: leaf, cpu_key: &key, nr: i);
3454	if (key.type != BTRFS_EXTENT_DATA_KEY)
3455	continue;
3456	ei = btrfs_item_ptr(leaf, i, struct btrfs_file_extent_item);
3457	type = btrfs_file_extent_type(eb: leaf, s: ei);
3458
3459	if ((type == BTRFS_FILE_EXTENT_REG ||
3460	type == BTRFS_FILE_EXTENT_PREALLOC) &&
3461	btrfs_file_extent_disk_bytenr(eb: leaf, s: ei) == data_bytenr) {
3462	found = true;
3463	space_cache_ino = key.objectid;
3464	break;
3465	}
3466	}
3467	if (!found)
3468	return -ENOENT;
3469	ret = delete_block_group_cache(fs_info: leaf->fs_info, block_group, NULL,
3470	ino: space_cache_ino);
3471	return ret;
3472	}
3473
3474	/*
3475	* helper to find all tree blocks that reference a given data extent
3476	*/
3477	static noinline_for_stack int add_data_references(struct reloc_control *rc,
3478	const struct btrfs_key *extent_key,
3479	struct btrfs_path *path,
3480	struct rb_root *blocks)
3481	{
3482	struct btrfs_backref_walk_ctx ctx = { 0 };
3483	struct ulist_iterator leaf_uiter;
3484	struct ulist_node *ref_node = NULL;
3485	const u32 blocksize = rc->extent_root->fs_info->nodesize;
3486	int ret = 0;
3487
3488	btrfs_release_path(p: path);
3489
3490	ctx.bytenr = extent_key->objectid;
3491	ctx.skip_inode_ref_list = true;
3492	ctx.fs_info = rc->extent_root->fs_info;
3493
3494	ret = btrfs_find_all_leafs(ctx: &ctx);
3495	if (ret < 0)
3496	return ret;
3497
3498	ULIST_ITER_INIT(&leaf_uiter);
3499	while ((ref_node = ulist_next(ulist: ctx.refs, uiter: &leaf_uiter))) {
3500	struct btrfs_tree_parent_check check = { 0 };
3501	struct extent_buffer *eb;
3502
3503	eb = read_tree_block(fs_info: ctx.fs_info, bytenr: ref_node->val, check: &check);
3504	if (IS_ERR(ptr: eb)) {
3505	ret = PTR_ERR(ptr: eb);
3506	break;
3507	}
3508	ret = delete_v1_space_cache(leaf: eb, block_group: rc->block_group,
3509	data_bytenr: extent_key->objectid);
3510	free_extent_buffer(eb);
3511	if (ret < 0)
3512	break;
3513	ret = __add_tree_block(rc, bytenr: ref_node->val, blocksize, blocks);
3514	if (ret < 0)
3515	break;
3516	}
3517	if (ret < 0)
3518	free_block_list(blocks);
3519	ulist_free(ulist: ctx.refs);
3520	return ret;
3521	}
3522
3523	/*
3524	* helper to find next unprocessed extent
3525	*/
3526	static noinline_for_stack
3527	int find_next_extent(struct reloc_control *rc, struct btrfs_path *path,
3528	struct btrfs_key *extent_key)
3529	{
3530	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
3531	struct btrfs_key key;
3532	struct extent_buffer *leaf;
3533	u64 start, end, last;
3534	int ret;
3535
3536	last = rc->block_group->start + rc->block_group->length;
3537	while (1) {
3538	bool block_found;
3539
3540	cond_resched();
3541	if (rc->search_start >= last) {
3542	ret = 1;
3543	break;
3544	}
3545
3546	key.objectid = rc->search_start;
3547	key.type = BTRFS_EXTENT_ITEM_KEY;
3548	key.offset = 0;
3549
3550	path->search_commit_root = 1;
3551	path->skip_locking = 1;
3552	ret = btrfs_search_slot(NULL, root: rc->extent_root, key: &key, p: path,
3553	ins_len: 0, cow: 0);
3554	if (ret < 0)
3555	break;
3556	next:
3557	leaf = path->nodes[0];
3558	if (path->slots[0] >= btrfs_header_nritems(eb: leaf)) {
3559	ret = btrfs_next_leaf(root: rc->extent_root, path);
3560	if (ret != 0)
3561	break;
3562	leaf = path->nodes[0];
3563	}
3564
3565	btrfs_item_key_to_cpu(eb: leaf, cpu_key: &key, nr: path->slots[0]);
3566	if (key.objectid >= last) {
3567	ret = 1;
3568	break;
3569	}
3570
3571	if (key.type != BTRFS_EXTENT_ITEM_KEY &&
3572	key.type != BTRFS_METADATA_ITEM_KEY) {
3573	path->slots[0]++;
3574	goto next;
3575	}
3576
3577	if (key.type == BTRFS_EXTENT_ITEM_KEY &&
3578	key.objectid + key.offset <= rc->search_start) {
3579	path->slots[0]++;
3580	goto next;
3581	}
3582
3583	if (key.type == BTRFS_METADATA_ITEM_KEY &&
3584	key.objectid + fs_info->nodesize <=
3585	rc->search_start) {
3586	path->slots[0]++;
3587	goto next;
3588	}
3589
3590	block_found = find_first_extent_bit(tree: &rc->processed_blocks,
3591	start: key.objectid, start_ret: &start, end_ret: &end,
3592	bits: EXTENT_DIRTY, NULL);
3593
3594	if (block_found && start <= key.objectid) {
3595	btrfs_release_path(p: path);
3596	rc->search_start = end + 1;
3597	} else {
3598	if (key.type == BTRFS_EXTENT_ITEM_KEY)
3599	rc->search_start = key.objectid + key.offset;
3600	else
3601	rc->search_start = key.objectid +
3602	fs_info->nodesize;
3603	memcpy(extent_key, &key, sizeof(key));
3604	return 0;
3605	}
3606	}
3607	btrfs_release_path(p: path);
3608	return ret;
3609	}
3610
3611	static void set_reloc_control(struct reloc_control *rc)
3612	{
3613	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
3614
3615	mutex_lock(&fs_info->reloc_mutex);
3616	fs_info->reloc_ctl = rc;
3617	mutex_unlock(lock: &fs_info->reloc_mutex);
3618	}
3619
3620	static void unset_reloc_control(struct reloc_control *rc)
3621	{
3622	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
3623
3624	mutex_lock(&fs_info->reloc_mutex);
3625	fs_info->reloc_ctl = NULL;
3626	mutex_unlock(lock: &fs_info->reloc_mutex);
3627	}
3628
3629	static noinline_for_stack
3630	int prepare_to_relocate(struct reloc_control *rc)
3631	{
3632	struct btrfs_trans_handle *trans;
3633	int ret;
3634
3635	rc->block_rsv = btrfs_alloc_block_rsv(fs_info: rc->extent_root->fs_info,
3636	type: BTRFS_BLOCK_RSV_TEMP);
3637	if (!rc->block_rsv)
3638	return -ENOMEM;
3639
3640	memset(&rc->cluster, 0, sizeof(rc->cluster));
3641	rc->search_start = rc->block_group->start;
3642	rc->extents_found = 0;
3643	rc->nodes_relocated = 0;
3644	rc->merging_rsv_size = 0;
3645	rc->reserved_bytes = 0;
3646	rc->block_rsv->size = rc->extent_root->fs_info->nodesize *
3647	RELOCATION_RESERVED_NODES;
3648	ret = btrfs_block_rsv_refill(fs_info: rc->extent_root->fs_info,
3649	block_rsv: rc->block_rsv, num_bytes: rc->block_rsv->size,
3650	flush: BTRFS_RESERVE_FLUSH_ALL);
3651	if (ret)
3652	return ret;
3653
3654	rc->create_reloc_tree = true;
3655	set_reloc_control(rc);
3656
3657	trans = btrfs_join_transaction(root: rc->extent_root);
3658	if (IS_ERR(ptr: trans)) {
3659	unset_reloc_control(rc);
3660	/*
3661	* extent tree is not a ref_cow tree and has no reloc_root to
3662	* cleanup. And callers are responsible to free the above
3663	* block rsv.
3664	*/
3665	return PTR_ERR(ptr: trans);
3666	}
3667
3668	ret = btrfs_commit_transaction(trans);
3669	if (ret)
3670	unset_reloc_control(rc);
3671
3672	return ret;
3673	}
3674
3675	static noinline_for_stack int relocate_block_group(struct reloc_control *rc)
3676	{
3677	struct btrfs_fs_info *fs_info = rc->extent_root->fs_info;
3678	struct rb_root blocks = RB_ROOT;
3679	struct btrfs_key key;
3680	struct btrfs_trans_handle *trans = NULL;
3681	struct btrfs_path *path;
3682	struct btrfs_extent_item *ei;
3683	u64 flags;
3684	int ret;
3685	int err = 0;
3686	int progress = 0;
3687
3688	path = btrfs_alloc_path();
3689	if (!path)
3690	return -ENOMEM;
3691	path->reada = READA_FORWARD;
3692
3693	ret = prepare_to_relocate(rc);
3694	if (ret) {
3695	err = ret;
3696	goto out_free;
3697	}
3698
3699	while (1) {
3700	rc->reserved_bytes = 0;
3701	ret = btrfs_block_rsv_refill(fs_info, block_rsv: rc->block_rsv,
3702	num_bytes: rc->block_rsv->size,
3703	flush: BTRFS_RESERVE_FLUSH_ALL);
3704	if (ret) {
3705	err = ret;
3706	break;
3707	}
3708	progress++;
3709	trans = btrfs_start_transaction(root: rc->extent_root, num_items: 0);
3710	if (IS_ERR(ptr: trans)) {
3711	err = PTR_ERR(ptr: trans);
3712	trans = NULL;
3713	break;
3714	}
3715	restart:
3716	if (update_backref_cache(trans, cache: &rc->backref_cache)) {
3717	btrfs_end_transaction(trans);
3718	trans = NULL;
3719	continue;
3720	}
3721
3722	ret = find_next_extent(rc, path, extent_key: &key);
3723	if (ret < 0)
3724	err = ret;
3725	if (ret != 0)
3726	break;
3727
3728	rc->extents_found++;
3729
3730	ei = btrfs_item_ptr(path->nodes[0], path->slots[0],
3731	struct btrfs_extent_item);
3732	flags = btrfs_extent_flags(eb: path->nodes[0], s: ei);
3733
3734	/*
3735	* If we are relocating a simple quota owned extent item, we
3736	* need to note the owner on the reloc data root so that when
3737	* we allocate the replacement item, we can attribute it to the
3738	* correct eventual owner (rather than the reloc data root).
3739	*/
3740	if (btrfs_qgroup_mode(fs_info) == BTRFS_QGROUP_MODE_SIMPLE) {
3741	struct btrfs_root *root = BTRFS_I(inode: rc->data_inode)->root;
3742	u64 owning_root_id = btrfs_get_extent_owner_root(fs_info,
3743	leaf: path->nodes[0],
3744	slot: path->slots[0]);
3745
3746	root->relocation_src_root = owning_root_id;
3747	}
3748
3749	if (flags & BTRFS_EXTENT_FLAG_TREE_BLOCK) {
3750	ret = add_tree_block(rc, extent_key: &key, path, blocks: &blocks);
3751	} else if (rc->stage == UPDATE_DATA_PTRS &&
3752	(flags & BTRFS_EXTENT_FLAG_DATA)) {
3753	ret = add_data_references(rc, extent_key: &key, path, blocks: &blocks);
3754	} else {
3755	btrfs_release_path(p: path);
3756	ret = 0;
3757	}
3758	if (ret < 0) {
3759	err = ret;
3760	break;
3761	}
3762
3763	if (!RB_EMPTY_ROOT(&blocks)) {
3764	ret = relocate_tree_blocks(trans, rc, blocks: &blocks);
3765	if (ret < 0) {
3766	if (ret != -EAGAIN) {
3767	err = ret;
3768	break;
3769	}
3770	rc->extents_found--;
3771	rc->search_start = key.objectid;
3772	}
3773	}
3774
3775	btrfs_end_transaction_throttle(trans);
3776	btrfs_btree_balance_dirty(fs_info);
3777	trans = NULL;
3778
3779	if (rc->stage == MOVE_DATA_EXTENTS &&
3780	(flags & BTRFS_EXTENT_FLAG_DATA)) {
3781	rc->found_file_extent = true;
3782	ret = relocate_data_extent(inode: rc->data_inode,
3783	extent_key: &key, cluster: &rc->cluster);
3784	if (ret < 0) {
3785	err = ret;
3786	break;
3787	}
3788	}
3789	if (btrfs_should_cancel_balance(fs_info)) {
3790	err = -ECANCELED;
3791	break;
3792	}
3793	}
3794	if (trans && progress && err == -ENOSPC) {
3795	ret = btrfs_force_chunk_alloc(trans, type: rc->block_group->flags);
3796	if (ret == 1) {
3797	err = 0;
3798	progress = 0;
3799	goto restart;
3800	}
3801	}
3802
3803	btrfs_release_path(p: path);
3804	clear_extent_bits(tree: &rc->processed_blocks, start: 0, end: (u64)-1, bits: EXTENT_DIRTY);
3805
3806	if (trans) {
3807	btrfs_end_transaction_throttle(trans);
3808	btrfs_btree_balance_dirty(fs_info);
3809	}
3810
3811	if (!err) {
3812	ret = relocate_file_extent_cluster(inode: rc->data_inode,
3813	cluster: &rc->cluster);
3814	if (ret < 0)
3815	err = ret;
3816	}
3817
3818	rc->create_reloc_tree = false;
3819	set_reloc_control(rc);
3820
3821	btrfs_backref_release_cache(cache: &rc->backref_cache);
3822	btrfs_block_rsv_release(fs_info, block_rsv: rc->block_rsv, num_bytes: (u64)-1, NULL);
3823
3824	/*
3825	* Even in the case when the relocation is cancelled, we should all go
3826	* through prepare_to_merge() and merge_reloc_roots().
3827	*
3828	* For error (including cancelled balance), prepare_to_merge() will
3829	* mark all reloc trees orphan, then queue them for cleanup in
3830	* merge_reloc_roots()
3831	*/
3832	err = prepare_to_merge(rc, err);
3833
3834	merge_reloc_roots(rc);
3835
3836	rc->merge_reloc_tree = false;
3837	unset_reloc_control(rc);
3838	btrfs_block_rsv_release(fs_info, block_rsv: rc->block_rsv, num_bytes: (u64)-1, NULL);
3839
3840	/* get rid of pinned extents */
3841	trans = btrfs_join_transaction(root: rc->extent_root);
3842	if (IS_ERR(ptr: trans)) {
3843	err = PTR_ERR(ptr: trans);
3844	goto out_free;
3845	}
3846	ret = btrfs_commit_transaction(trans);
3847	if (ret && !err)
3848	err = ret;
3849	out_free:
3850	ret = clean_dirty_subvols(rc);
3851	if (ret < 0 && !err)
3852	err = ret;
3853	btrfs_free_block_rsv(fs_info, rsv: rc->block_rsv);
3854	btrfs_free_path(p: path);
3855	return err;
3856	}
3857
3858	static int __insert_orphan_inode(struct btrfs_trans_handle *trans,
3859	struct btrfs_root *root, u64 objectid)
3860	{
3861	struct btrfs_path *path;
3862	struct btrfs_inode_item *item;
3863	struct extent_buffer *leaf;
3864	int ret;
3865
3866	path = btrfs_alloc_path();
3867	if (!path)
3868	return -ENOMEM;
3869
3870	ret = btrfs_insert_empty_inode(trans, root, path, objectid);
3871	if (ret)
3872	goto out;
3873
3874	leaf = path->nodes[0];
3875	item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_inode_item);
3876	memzero_extent_buffer(eb: leaf, start: (unsigned long)item, len: sizeof(*item));
3877	btrfs_set_inode_generation(eb: leaf, s: item, val: 1);
3878	btrfs_set_inode_size(eb: leaf, s: item, val: 0);
3879	btrfs_set_inode_mode(eb: leaf, s: item, S_IFREG | 0600);
3880	btrfs_set_inode_flags(eb: leaf, s: item, BTRFS_INODE_NOCOMPRESS |
3881	BTRFS_INODE_PREALLOC);
3882	btrfs_mark_buffer_dirty(trans, buf: leaf);
3883	out:
3884	btrfs_free_path(p: path);
3885	return ret;
3886	}
3887
3888	static void delete_orphan_inode(struct btrfs_trans_handle *trans,
3889	struct btrfs_root *root, u64 objectid)
3890	{
3891	struct btrfs_path *path;
3892	struct btrfs_key key;
3893	int ret = 0;
3894
3895	path = btrfs_alloc_path();
3896	if (!path) {
3897	ret = -ENOMEM;
3898	goto out;
3899	}
3900
3901	key.objectid = objectid;
3902	key.type = BTRFS_INODE_ITEM_KEY;
3903	key.offset = 0;
3904	ret = btrfs_search_slot(trans, root, key: &key, p: path, ins_len: -1, cow: 1);
3905	if (ret) {
3906	if (ret > 0)
3907	ret = -ENOENT;
3908	goto out;
3909	}
3910	ret = btrfs_del_item(trans, root, path);
3911	out:
3912	if (ret)
3913	btrfs_abort_transaction(trans, ret);
3914	btrfs_free_path(p: path);
3915	}
3916
3917	/*
3918	* helper to create inode for data relocation.
3919	* the inode is in data relocation tree and its link count is 0
3920	*/
3921	static noinline_for_stack struct inode *create_reloc_inode(
3922	struct btrfs_fs_info *fs_info,
3923	const struct btrfs_block_group *group)
3924	{
3925	struct inode *inode = NULL;
3926	struct btrfs_trans_handle *trans;
3927	struct btrfs_root *root;
3928	u64 objectid;
3929	int err = 0;
3930
3931	root = btrfs_grab_root(root: fs_info->data_reloc_root);
3932	trans = btrfs_start_transaction(root, num_items: 6);
3933	if (IS_ERR(ptr: trans)) {
3934	btrfs_put_root(root);
3935	return ERR_CAST(ptr: trans);
3936	}
3937
3938	err = btrfs_get_free_objectid(root, objectid: &objectid);
3939	if (err)
3940	goto out;
3941
3942	err = __insert_orphan_inode(trans, root, objectid);
3943	if (err)
3944	goto out;
3945
3946	inode = btrfs_iget(s: fs_info->sb, ino: objectid, root);
3947	if (IS_ERR(ptr: inode)) {
3948	delete_orphan_inode(trans, root, objectid);
3949	err = PTR_ERR(ptr: inode);
3950	inode = NULL;
3951	goto out;
3952	}
3953	BTRFS_I(inode)->index_cnt = group->start;
3954
3955	err = btrfs_orphan_add(trans, inode: BTRFS_I(inode));
3956	out:
3957	btrfs_put_root(root);
3958	btrfs_end_transaction(trans);
3959	btrfs_btree_balance_dirty(fs_info);
3960	if (err) {
3961	iput(inode);
3962	inode = ERR_PTR(error: err);
3963	}
3964	return inode;
3965	}
3966
3967	/*
3968	* Mark start of chunk relocation that is cancellable. Check if the cancellation
3969	* has been requested meanwhile and don't start in that case.
3970	*
3971	* Return:
3972	* 0 success
3973	* -EINPROGRESS operation is already in progress, that's probably a bug
3974	* -ECANCELED cancellation request was set before the operation started
3975	*/
3976	static int reloc_chunk_start(struct btrfs_fs_info *fs_info)
3977	{
3978	if (test_and_set_bit(nr: BTRFS_FS_RELOC_RUNNING, addr: &fs_info->flags)) {
3979	/* This should not happen */
3980	btrfs_err(fs_info, "reloc already running, cannot start");
3981	return -EINPROGRESS;
3982	}
3983
3984	if (atomic_read(v: &fs_info->reloc_cancel_req) > 0) {
3985	btrfs_info(fs_info, "chunk relocation canceled on start");
3986	/*
3987	* On cancel, clear all requests but let the caller mark
3988	* the end after cleanup operations.
3989	*/
3990	atomic_set(v: &fs_info->reloc_cancel_req, i: 0);
3991	return -ECANCELED;
3992	}
3993	return 0;
3994	}
3995
3996	/*
3997	* Mark end of chunk relocation that is cancellable and wake any waiters.
3998	*/
3999	static void reloc_chunk_end(struct btrfs_fs_info *fs_info)
4000	{
4001	/* Requested after start, clear bit first so any waiters can continue */
4002	if (atomic_read(v: &fs_info->reloc_cancel_req) > 0)
4003	btrfs_info(fs_info, "chunk relocation canceled during operation");
4004	clear_and_wake_up_bit(bit: BTRFS_FS_RELOC_RUNNING, word: &fs_info->flags);
4005	atomic_set(v: &fs_info->reloc_cancel_req, i: 0);
4006	}
4007
4008	static struct reloc_control *alloc_reloc_control(struct btrfs_fs_info *fs_info)
4009	{
4010	struct reloc_control *rc;
4011
4012	rc = kzalloc(size: sizeof(*rc), GFP_NOFS);
4013	if (!rc)
4014	return NULL;
4015
4016	INIT_LIST_HEAD(list: &rc->reloc_roots);
4017	INIT_LIST_HEAD(list: &rc->dirty_subvol_roots);
4018	btrfs_backref_init_cache(fs_info, cache: &rc->backref_cache, is_reloc: true);
4019	rc->reloc_root_tree.rb_root = RB_ROOT;
4020	spin_lock_init(&rc->reloc_root_tree.lock);
4021	extent_io_tree_init(fs_info, tree: &rc->processed_blocks, owner: IO_TREE_RELOC_BLOCKS);
4022	return rc;
4023	}
4024
4025	static void free_reloc_control(struct reloc_control *rc)
4026	{
4027	struct mapping_node *node, *tmp;
4028
4029	free_reloc_roots(list: &rc->reloc_roots);
4030	rbtree_postorder_for_each_entry_safe(node, tmp,
4031	&rc->reloc_root_tree.rb_root, rb_node)
4032	kfree(objp: node);
4033
4034	kfree(objp: rc);
4035	}
4036
4037	/*
4038	* Print the block group being relocated
4039	*/
4040	static void describe_relocation(struct btrfs_fs_info *fs_info,
4041	struct btrfs_block_group *block_group)
4042	{
4043	char buf[128] = {'\0'};
4044
4045	btrfs_describe_block_groups(flags: block_group->flags, buf, size_buf: sizeof(buf));
4046
4047	btrfs_info(fs_info,
4048	"relocating block group %llu flags %s",
4049	block_group->start, buf);
4050	}
4051
4052	static const char *stage_to_string(enum reloc_stage stage)
4053	{
4054	if (stage == MOVE_DATA_EXTENTS)
4055	return "move data extents";
4056	if (stage == UPDATE_DATA_PTRS)
4057	return "update data pointers";
4058	return "unknown";
4059	}
4060
4061	/*
4062	* function to relocate all extents in a block group.
4063	*/
4064	int btrfs_relocate_block_group(struct btrfs_fs_info *fs_info, u64 group_start)
4065	{
4066	struct btrfs_block_group *bg;
4067	struct btrfs_root *extent_root = btrfs_extent_root(fs_info, bytenr: group_start);
4068	struct reloc_control *rc;
4069	struct inode *inode;
4070	struct btrfs_path *path;
4071	int ret;
4072	int rw = 0;
4073	int err = 0;
4074
4075	/*
4076	* This only gets set if we had a half-deleted snapshot on mount. We
4077	* cannot allow relocation to start while we're still trying to clean up
4078	* these pending deletions.
4079	*/
4080	ret = wait_on_bit(word: &fs_info->flags, bit: BTRFS_FS_UNFINISHED_DROPS, TASK_INTERRUPTIBLE);
4081	if (ret)
4082	return ret;
4083
4084	/* We may have been woken up by close_ctree, so bail if we're closing. */
4085	if (btrfs_fs_closing(fs_info))
4086	return -EINTR;
4087
4088	bg = btrfs_lookup_block_group(info: fs_info, bytenr: group_start);
4089	if (!bg)
4090	return -ENOENT;
4091
4092	/*
4093	* Relocation of a data block group creates ordered extents. Without
4094	* sb_start_write(), we can freeze the filesystem while unfinished
4095	* ordered extents are left. Such ordered extents can cause a deadlock
4096	* e.g. when syncfs() is waiting for their completion but they can't
4097	* finish because they block when joining a transaction, due to the
4098	* fact that the freeze locks are being held in write mode.
4099	*/
4100	if (bg->flags & BTRFS_BLOCK_GROUP_DATA)
4101	ASSERT(sb_write_started(fs_info->sb));
4102
4103	if (btrfs_pinned_by_swapfile(fs_info, ptr: bg)) {
4104	btrfs_put_block_group(cache: bg);
4105	return -ETXTBSY;
4106	}
4107
4108	rc = alloc_reloc_control(fs_info);
4109	if (!rc) {
4110	btrfs_put_block_group(cache: bg);
4111	return -ENOMEM;
4112	}
4113
4114	ret = reloc_chunk_start(fs_info);
4115	if (ret < 0) {
4116	err = ret;
4117	goto out_put_bg;
4118	}
4119
4120	rc->extent_root = extent_root;
4121	rc->block_group = bg;
4122
4123	ret = btrfs_inc_block_group_ro(cache: rc->block_group, do_chunk_alloc: true);
4124	if (ret) {
4125	err = ret;
4126	goto out;
4127	}
4128	rw = 1;
4129
4130	path = btrfs_alloc_path();
4131	if (!path) {
4132	err = -ENOMEM;
4133	goto out;
4134	}
4135
4136	inode = lookup_free_space_inode(block_group: rc->block_group, path);
4137	btrfs_free_path(p: path);
4138
4139	if (!IS_ERR(ptr: inode))
4140	ret = delete_block_group_cache(fs_info, block_group: rc->block_group, inode, ino: 0);
4141	else
4142	ret = PTR_ERR(ptr: inode);
4143
4144	if (ret && ret != -ENOENT) {
4145	err = ret;
4146	goto out;
4147	}
4148
4149	rc->data_inode = create_reloc_inode(fs_info, group: rc->block_group);
4150	if (IS_ERR(ptr: rc->data_inode)) {
4151	err = PTR_ERR(ptr: rc->data_inode);
4152	rc->data_inode = NULL;
4153	goto out;
4154	}
4155
4156	describe_relocation(fs_info, block_group: rc->block_group);
4157
4158	btrfs_wait_block_group_reservations(bg: rc->block_group);
4159	btrfs_wait_nocow_writers(bg: rc->block_group);
4160	btrfs_wait_ordered_roots(fs_info, U64_MAX,
4161	range_start: rc->block_group->start,
4162	range_len: rc->block_group->length);
4163
4164	ret = btrfs_zone_finish(block_group: rc->block_group);
4165	WARN_ON(ret && ret != -EAGAIN);
4166
4167	while (1) {
4168	enum reloc_stage finishes_stage;
4169
4170	mutex_lock(&fs_info->cleaner_mutex);
4171	ret = relocate_block_group(rc);
4172	mutex_unlock(lock: &fs_info->cleaner_mutex);
4173	if (ret < 0)
4174	err = ret;
4175
4176	finishes_stage = rc->stage;
4177	/*
4178	* We may have gotten ENOSPC after we already dirtied some
4179	* extents. If writeout happens while we're relocating a
4180	* different block group we could end up hitting the
4181	* BUG_ON(rc->stage == UPDATE_DATA_PTRS) in
4182	* btrfs_reloc_cow_block. Make sure we write everything out
4183	* properly so we don't trip over this problem, and then break
4184	* out of the loop if we hit an error.
4185	*/
4186	if (rc->stage == MOVE_DATA_EXTENTS && rc->found_file_extent) {
4187	ret = btrfs_wait_ordered_range(inode: rc->data_inode, start: 0,
4188	len: (u64)-1);
4189	if (ret)
4190	err = ret;
4191	invalidate_mapping_pages(mapping: rc->data_inode->i_mapping,
4192	start: 0, end: -1);
4193	rc->stage = UPDATE_DATA_PTRS;
4194	}
4195
4196	if (err < 0)
4197	goto out;
4198
4199	if (rc->extents_found == 0)
4200	break;
4201
4202	btrfs_info(fs_info, "found %llu extents, stage: %s",
4203	rc->extents_found, stage_to_string(finishes_stage));
4204	}
4205
4206	WARN_ON(rc->block_group->pinned > 0);
4207	WARN_ON(rc->block_group->reserved > 0);
4208	WARN_ON(rc->block_group->used > 0);
4209	out:
4210	if (err && rw)
4211	btrfs_dec_block_group_ro(cache: rc->block_group);
4212	iput(rc->data_inode);
4213	out_put_bg:
4214	btrfs_put_block_group(cache: bg);
4215	reloc_chunk_end(fs_info);
4216	free_reloc_control(rc);
4217	return err;
4218	}
4219
4220	static noinline_for_stack int mark_garbage_root(struct btrfs_root *root)
4221	{
4222	struct btrfs_fs_info *fs_info = root->fs_info;
4223	struct btrfs_trans_handle *trans;
4224	int ret, err;
4225
4226	trans = btrfs_start_transaction(root: fs_info->tree_root, num_items: 0);
4227	if (IS_ERR(ptr: trans))
4228	return PTR_ERR(ptr: trans);
4229
4230	memset(&root->root_item.drop_progress, 0,
4231	sizeof(root->root_item.drop_progress));
4232	btrfs_set_root_drop_level(s: &root->root_item, val: 0);
4233	btrfs_set_root_refs(s: &root->root_item, val: 0);
4234	ret = btrfs_update_root(trans, root: fs_info->tree_root,
4235	key: &root->root_key, item: &root->root_item);
4236
4237	err = btrfs_end_transaction(trans);
4238	if (err)
4239	return err;
4240	return ret;
4241	}
4242
4243	/*
4244	* recover relocation interrupted by system crash.
4245	*
4246	* this function resumes merging reloc trees with corresponding fs trees.
4247	* this is important for keeping the sharing of tree blocks
4248	*/
4249	int btrfs_recover_relocation(struct btrfs_fs_info *fs_info)
4250	{
4251	LIST_HEAD(reloc_roots);
4252	struct btrfs_key key;
4253	struct btrfs_root *fs_root;
4254	struct btrfs_root *reloc_root;
4255	struct btrfs_path *path;
4256	struct extent_buffer *leaf;
4257	struct reloc_control *rc = NULL;
4258	struct btrfs_trans_handle *trans;
4259	int ret;
4260	int err = 0;
4261
4262	path = btrfs_alloc_path();
4263	if (!path)
4264	return -ENOMEM;
4265	path->reada = READA_BACK;
4266
4267	key.objectid = BTRFS_TREE_RELOC_OBJECTID;
4268	key.type = BTRFS_ROOT_ITEM_KEY;
4269	key.offset = (u64)-1;
4270
4271	while (1) {
4272	ret = btrfs_search_slot(NULL, root: fs_info->tree_root, key: &key,
4273	p: path, ins_len: 0, cow: 0);
4274	if (ret < 0) {
4275	err = ret;
4276	goto out;
4277	}
4278	if (ret > 0) {
4279	if (path->slots[0] == 0)
4280	break;
4281	path->slots[0]--;
4282	}
4283	leaf = path->nodes[0];
4284	btrfs_item_key_to_cpu(eb: leaf, cpu_key: &key, nr: path->slots[0]);
4285	btrfs_release_path(p: path);
4286
4287	if (key.objectid != BTRFS_TREE_RELOC_OBJECTID ||
4288	key.type != BTRFS_ROOT_ITEM_KEY)
4289	break;
4290
4291	reloc_root = btrfs_read_tree_root(tree_root: fs_info->tree_root, key: &key);
4292	if (IS_ERR(ptr: reloc_root)) {
4293	err = PTR_ERR(ptr: reloc_root);
4294	goto out;
4295	}
4296
4297	set_bit(nr: BTRFS_ROOT_SHAREABLE, addr: &reloc_root->state);
4298	list_add(new: &reloc_root->root_list, head: &reloc_roots);
4299
4300	if (btrfs_root_refs(s: &reloc_root->root_item) > 0) {
4301	fs_root = btrfs_get_fs_root(fs_info,
4302	objectid: reloc_root->root_key.offset, check_ref: false);
4303	if (IS_ERR(ptr: fs_root)) {
4304	ret = PTR_ERR(ptr: fs_root);
4305	if (ret != -ENOENT) {
4306	err = ret;
4307	goto out;
4308	}
4309	ret = mark_garbage_root(root: reloc_root);
4310	if (ret < 0) {
4311	err = ret;
4312	goto out;
4313	}
4314	} else {
4315	btrfs_put_root(root: fs_root);
4316	}
4317	}
4318
4319	if (key.offset == 0)
4320	break;
4321
4322	key.offset--;
4323	}
4324	btrfs_release_path(p: path);
4325
4326	if (list_empty(head: &reloc_roots))
4327	goto out;
4328
4329	rc = alloc_reloc_control(fs_info);
4330	if (!rc) {
4331	err = -ENOMEM;
4332	goto out;
4333	}
4334
4335	ret = reloc_chunk_start(fs_info);
4336	if (ret < 0) {
4337	err = ret;
4338	goto out_end;
4339	}
4340
4341	rc->extent_root = btrfs_extent_root(fs_info, bytenr: 0);
4342
4343	set_reloc_control(rc);
4344
4345	trans = btrfs_join_transaction(root: rc->extent_root);
4346	if (IS_ERR(ptr: trans)) {
4347	err = PTR_ERR(ptr: trans);
4348	goto out_unset;
4349	}
4350
4351	rc->merge_reloc_tree = true;
4352
4353	while (!list_empty(head: &reloc_roots)) {
4354	reloc_root = list_entry(reloc_roots.next,
4355	struct btrfs_root, root_list);
4356	list_del(entry: &reloc_root->root_list);
4357
4358	if (btrfs_root_refs(s: &reloc_root->root_item) == 0) {
4359	list_add_tail(new: &reloc_root->root_list,
4360	head: &rc->reloc_roots);
4361	continue;
4362	}
4363
4364	fs_root = btrfs_get_fs_root(fs_info, objectid: reloc_root->root_key.offset,
4365	check_ref: false);
4366	if (IS_ERR(ptr: fs_root)) {
4367	err = PTR_ERR(ptr: fs_root);
4368	list_add_tail(new: &reloc_root->root_list, head: &reloc_roots);
4369	btrfs_end_transaction(trans);
4370	goto out_unset;
4371	}
4372
4373	err = __add_reloc_root(root: reloc_root);
4374	ASSERT(err != -EEXIST);
4375	if (err) {
4376	list_add_tail(new: &reloc_root->root_list, head: &reloc_roots);
4377	btrfs_put_root(root: fs_root);
4378	btrfs_end_transaction(trans);
4379	goto out_unset;
4380	}
4381	fs_root->reloc_root = btrfs_grab_root(root: reloc_root);
4382	btrfs_put_root(root: fs_root);
4383	}
4384
4385	err = btrfs_commit_transaction(trans);
4386	if (err)
4387	goto out_unset;
4388
4389	merge_reloc_roots(rc);
4390
4391	unset_reloc_control(rc);
4392
4393	trans = btrfs_join_transaction(root: rc->extent_root);
4394	if (IS_ERR(ptr: trans)) {
4395	err = PTR_ERR(ptr: trans);
4396	goto out_clean;
4397	}
4398	err = btrfs_commit_transaction(trans);
4399	out_clean:
4400	ret = clean_dirty_subvols(rc);
4401	if (ret < 0 && !err)
4402	err = ret;
4403	out_unset:
4404	unset_reloc_control(rc);
4405	out_end:
4406	reloc_chunk_end(fs_info);
4407	free_reloc_control(rc);
4408	out:
4409	free_reloc_roots(list: &reloc_roots);
4410
4411	btrfs_free_path(p: path);
4412
4413	if (err == 0) {
4414	/* cleanup orphan inode in data relocation tree */
4415	fs_root = btrfs_grab_root(root: fs_info->data_reloc_root);
4416	ASSERT(fs_root);
4417	err = btrfs_orphan_cleanup(root: fs_root);
4418	btrfs_put_root(root: fs_root);
4419	}
4420	return err;
4421	}
4422
4423	/*
4424	* helper to add ordered checksum for data relocation.
4425	*
4426	* cloning checksum properly handles the nodatasum extents.
4427	* it also saves CPU time to re-calculate the checksum.
4428	*/
4429	int btrfs_reloc_clone_csums(struct btrfs_ordered_extent *ordered)
4430	{
4431	struct btrfs_inode *inode = BTRFS_I(inode: ordered->inode);
4432	struct btrfs_fs_info *fs_info = inode->root->fs_info;
4433	u64 disk_bytenr = ordered->file_offset + inode->index_cnt;
4434	struct btrfs_root *csum_root = btrfs_csum_root(fs_info, bytenr: disk_bytenr);
4435	LIST_HEAD(list);
4436	int ret;
4437
4438	ret = btrfs_lookup_csums_list(root: csum_root, start: disk_bytenr,
4439	end: disk_bytenr + ordered->num_bytes - 1,
4440	list: &list, search_commit: 0, nowait: false);
4441	if (ret)
4442	return ret;
4443
4444	while (!list_empty(head: &list)) {
4445	struct btrfs_ordered_sum *sums =
4446	list_entry(list.next, struct btrfs_ordered_sum, list);
4447
4448	list_del_init(entry: &sums->list);
4449
4450	/*
4451	* We need to offset the new_bytenr based on where the csum is.
4452	* We need to do this because we will read in entire prealloc
4453	* extents but we may have written to say the middle of the
4454	* prealloc extent, so we need to make sure the csum goes with
4455	* the right disk offset.
4456	*
4457	* We can do this because the data reloc inode refers strictly
4458	* to the on disk bytes, so we don't have to worry about
4459	* disk_len vs real len like with real inodes since it's all
4460	* disk length.
4461	*/
4462	sums->logical = ordered->disk_bytenr + sums->logical - disk_bytenr;
4463	btrfs_add_ordered_sum(entry: ordered, sum: sums);
4464	}
4465
4466	return 0;
4467	}
4468
4469	int btrfs_reloc_cow_block(struct btrfs_trans_handle *trans,
4470	struct btrfs_root *root,
4471	const struct extent_buffer *buf,
4472	struct extent_buffer *cow)
4473	{
4474	struct btrfs_fs_info *fs_info = root->fs_info;
4475	struct reloc_control *rc;
4476	struct btrfs_backref_node *node;
4477	int first_cow = 0;
4478	int level;
4479	int ret = 0;
4480
4481	rc = fs_info->reloc_ctl;
4482	if (!rc)
4483	return 0;
4484
4485	BUG_ON(rc->stage == UPDATE_DATA_PTRS && btrfs_is_data_reloc_root(root));
4486
4487	level = btrfs_header_level(eb: buf);
4488	if (btrfs_header_generation(eb: buf) <=
4489	btrfs_root_last_snapshot(s: &root->root_item))
4490	first_cow = 1;
4491
4492	if (root->root_key.objectid == BTRFS_TREE_RELOC_OBJECTID &&
4493	rc->create_reloc_tree) {
4494	WARN_ON(!first_cow && level == 0);
4495
4496	node = rc->backref_cache.path[level];
4497	BUG_ON(node->bytenr != buf->start &&
4498	node->new_bytenr != buf->start);
4499
4500	btrfs_backref_drop_node_buffer(node);
4501	atomic_inc(v: &cow->refs);
4502	node->eb = cow;
4503	node->new_bytenr = cow->start;
4504
4505	if (!node->pending) {
4506	list_move_tail(list: &node->list,
4507	head: &rc->backref_cache.pending[level]);
4508	node->pending = 1;
4509	}
4510
4511	if (first_cow)
4512	mark_block_processed(rc, node);
4513
4514	if (first_cow && level > 0)
4515	rc->nodes_relocated += buf->len;
4516	}
4517
4518	if (level == 0 && first_cow && rc->stage == UPDATE_DATA_PTRS)
4519	ret = replace_file_extents(trans, rc, root, leaf: cow);
4520	return ret;
4521	}
4522
4523	/*
4524	* called before creating snapshot. it calculates metadata reservation
4525	* required for relocating tree blocks in the snapshot
4526	*/
4527	void btrfs_reloc_pre_snapshot(struct btrfs_pending_snapshot *pending,
4528	u64 *bytes_to_reserve)
4529	{
4530	struct btrfs_root *root = pending->root;
4531	struct reloc_control *rc = root->fs_info->reloc_ctl;
4532
4533	if (!rc || !have_reloc_root(root))
4534	return;
4535
4536	if (!rc->merge_reloc_tree)
4537	return;
4538
4539	root = root->reloc_root;
4540	BUG_ON(btrfs_root_refs(&root->root_item) == 0);
4541	/*
4542	* relocation is in the stage of merging trees. the space
4543	* used by merging a reloc tree is twice the size of
4544	* relocated tree nodes in the worst case. half for cowing
4545	* the reloc tree, half for cowing the fs tree. the space
4546	* used by cowing the reloc tree will be freed after the
4547	* tree is dropped. if we create snapshot, cowing the fs
4548	* tree may use more space than it frees. so we need
4549	* reserve extra space.
4550	*/
4551	*bytes_to_reserve += rc->nodes_relocated;
4552	}
4553
4554	/*
4555	* called after snapshot is created. migrate block reservation
4556	* and create reloc root for the newly created snapshot
4557	*
4558	* This is similar to btrfs_init_reloc_root(), we come out of here with two
4559	* references held on the reloc_root, one for root->reloc_root and one for
4560	* rc->reloc_roots.
4561	*/
4562	int btrfs_reloc_post_snapshot(struct btrfs_trans_handle *trans,
4563	struct btrfs_pending_snapshot *pending)
4564	{
4565	struct btrfs_root *root = pending->root;
4566	struct btrfs_root *reloc_root;
4567	struct btrfs_root *new_root;
4568	struct reloc_control *rc = root->fs_info->reloc_ctl;
4569	int ret;
4570
4571	if (!rc || !have_reloc_root(root))
4572	return 0;
4573
4574	rc = root->fs_info->reloc_ctl;
4575	rc->merging_rsv_size += rc->nodes_relocated;
4576
4577	if (rc->merge_reloc_tree) {
4578	ret = btrfs_block_rsv_migrate(src_rsv: &pending->block_rsv,
4579	dst_rsv: rc->block_rsv,
4580	num_bytes: rc->nodes_relocated, update_size: true);
4581	if (ret)
4582	return ret;
4583	}
4584
4585	new_root = pending->snap;
4586	reloc_root = create_reloc_root(trans, root: root->reloc_root,
4587	objectid: new_root->root_key.objectid);
4588	if (IS_ERR(ptr: reloc_root))
4589	return PTR_ERR(ptr: reloc_root);
4590
4591	ret = __add_reloc_root(root: reloc_root);
4592	ASSERT(ret != -EEXIST);
4593	if (ret) {
4594	/* Pairs with create_reloc_root */
4595	btrfs_put_root(root: reloc_root);
4596	return ret;
4597	}
4598	new_root->reloc_root = btrfs_grab_root(root: reloc_root);
4599
4600	if (rc->create_reloc_tree)
4601	ret = clone_backref_node(trans, rc, src: root, dest: reloc_root);
4602	return ret;
4603	}
4604
4605	/*
4606	* Get the current bytenr for the block group which is being relocated.
4607	*
4608	* Return U64_MAX if no running relocation.
4609	*/
4610	u64 btrfs_get_reloc_bg_bytenr(const struct btrfs_fs_info *fs_info)
4611	{
4612	u64 logical = U64_MAX;
4613
4614	lockdep_assert_held(&fs_info->reloc_mutex);
4615
4616	if (fs_info->reloc_ctl && fs_info->reloc_ctl->block_group)
4617	logical = fs_info->reloc_ctl->block_group->start;
4618	return logical;
4619	}
4620