[linux.git] / fs / btrfs / extent_map.c

#include <linux/err.h>
#include <linux/slab.h>
#include <linux/spinlock.h>
#include <linux/hardirq.h>
#include "ctree.h"
#include "extent_map.h"


static struct kmem_cache *extent_map_cache;

int __init extent_map_init(void)
{
	extent_map_cache = kmem_cache_create("btrfs_extent_map",
			sizeof(struct extent_map), 0,
			SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
	if (!extent_map_cache)
		return -ENOMEM;
	return 0;
}

void extent_map_exit(void)
{
	if (extent_map_cache)
		kmem_cache_destroy(extent_map_cache);
}

/**
 * extent_map_tree_init - initialize extent map tree
 * @tree:		tree to initialize
 *
 * Initialize the extent tree @tree.  Should be called for each new inode
 * or other user of the extent_map interface.
 */
void extent_map_tree_init(struct extent_map_tree *tree)
{
	tree->map = RB_ROOT;
	INIT_LIST_HEAD(&tree->modified_extents);
	rwlock_init(&tree->lock);
}

/**
 * alloc_extent_map - allocate new extent map structure
 *
 * Allocate a new extent_map structure.  The new structure is
 * returned with a reference count of one and needs to be
 * freed using free_extent_map()
 */
struct extent_map *alloc_extent_map(void)
{
	struct extent_map *em;
	em = kmem_cache_zalloc(extent_map_cache, GFP_NOFS);
	if (!em)
		return NULL;
	RB_CLEAR_NODE(&em->rb_node);
	em->flags = 0;
	em->compress_type = BTRFS_COMPRESS_NONE;
	em->generation = 0;
	atomic_set(&em->refs, 1);
	INIT_LIST_HEAD(&em->list);
	return em;
}

/**
 * free_extent_map - drop reference count of an extent_map
 * @em:		extent map beeing releasead
 *
 * Drops the reference out on @em by one and free the structure
 * if the reference count hits zero.
 */
void free_extent_map(struct extent_map *em)
{
	if (!em)
		return;
	WARN_ON(atomic_read(&em->refs) == 0);
	if (atomic_dec_and_test(&em->refs)) {
		WARN_ON(extent_map_in_tree(em));
		WARN_ON(!list_empty(&em->list));
		if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
			kfree(em->map_lookup);
		kmem_cache_free(extent_map_cache, em);
	}
}

/* simple helper to do math around the end of an extent, handling wrap */
static u64 range_end(u64 start, u64 len)
{
	if (start + len < start)
		return (u64)-1;
	return start + len;
}

static int tree_insert(struct rb_root *root, struct extent_map *em)
{
	struct rb_node **p = &root->rb_node;
	struct rb_node *parent = NULL;
	struct extent_map *entry = NULL;
	struct rb_node *orig_parent = NULL;
	u64 end = range_end(em->start, em->len);

	while (*p) {
		parent = *p;
		entry = rb_entry(parent, struct extent_map, rb_node);

		if (em->start < entry->start)
			p = &(*p)->rb_left;
		else if (em->start >= extent_map_end(entry))
			p = &(*p)->rb_right;
		else
			return -EEXIST;
	}

	orig_parent = parent;
	while (parent && em->start >= extent_map_end(entry)) {
		parent = rb_next(parent);
		entry = rb_entry(parent, struct extent_map, rb_node);
	}
	if (parent)
		if (end > entry->start && em->start < extent_map_end(entry))
			return -EEXIST;

	parent = orig_parent;
	entry = rb_entry(parent, struct extent_map, rb_node);
	while (parent && em->start < entry->start) {
		parent = rb_prev(parent);
		entry = rb_entry(parent, struct extent_map, rb_node);
	}
	if (parent)
		if (end > entry->start && em->start < extent_map_end(entry))
			return -EEXIST;

	rb_link_node(&em->rb_node, orig_parent, p);
	rb_insert_color(&em->rb_node, root);
	return 0;
}

/*
 * search through the tree for an extent_map with a given offset.  If
 * it can't be found, try to find some neighboring extents
 */
static struct rb_node *__tree_search(struct rb_root *root, u64 offset,
				     struct rb_node **prev_ret,
				     struct rb_node **next_ret)
{
	struct rb_node *n = root->rb_node;
	struct rb_node *prev = NULL;
	struct rb_node *orig_prev = NULL;
	struct extent_map *entry;
	struct extent_map *prev_entry = NULL;

	while (n) {
		entry = rb_entry(n, struct extent_map, rb_node);
		prev = n;
		prev_entry = entry;

		if (offset < entry->start)
			n = n->rb_left;
		else if (offset >= extent_map_end(entry))
			n = n->rb_right;
		else
			return n;
	}

	if (prev_ret) {
		orig_prev = prev;
		while (prev && offset >= extent_map_end(prev_entry)) {
			prev = rb_next(prev);
			prev_entry = rb_entry(prev, struct extent_map, rb_node);
		}
		*prev_ret = prev;
		prev = orig_prev;
	}

	if (next_ret) {
		prev_entry = rb_entry(prev, struct extent_map, rb_node);
		while (prev && offset < prev_entry->start) {
			prev = rb_prev(prev);
			prev_entry = rb_entry(prev, struct extent_map, rb_node);
		}
		*next_ret = prev;
	}
	return NULL;
}

/* check to see if two extent_map structs are adjacent and safe to merge */
static int mergable_maps(struct extent_map *prev, struct extent_map *next)
{
	if (test_bit(EXTENT_FLAG_PINNED, &prev->flags))
		return 0;

	/*
	 * don't merge compressed extents, we need to know their
	 * actual size
	 */
	if (test_bit(EXTENT_FLAG_COMPRESSED, &prev->flags))
		return 0;

	if (test_bit(EXTENT_FLAG_LOGGING, &prev->flags) ||
	    test_bit(EXTENT_FLAG_LOGGING, &next->flags))
		return 0;

	/*
	 * We don't want to merge stuff that hasn't been written to the log yet
	 * since it may not reflect exactly what is on disk, and that would be
	 * bad.
	 */
	if (!list_empty(&prev->list) || !list_empty(&next->list))
		return 0;

	if (extent_map_end(prev) == next->start &&
	    prev->flags == next->flags &&
	    prev->bdev == next->bdev &&
	    ((next->block_start == EXTENT_MAP_HOLE &&
	      prev->block_start == EXTENT_MAP_HOLE) ||
	     (next->block_start == EXTENT_MAP_INLINE &&
	      prev->block_start == EXTENT_MAP_INLINE) ||
	     (next->block_start == EXTENT_MAP_DELALLOC &&
	      prev->block_start == EXTENT_MAP_DELALLOC) ||
	     (next->block_start < EXTENT_MAP_LAST_BYTE - 1 &&
	      next->block_start == extent_map_block_end(prev)))) {
		return 1;
	}
	return 0;
}

static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
{
	struct extent_map *merge = NULL;
	struct rb_node *rb;

	if (em->start != 0) {
		rb = rb_prev(&em->rb_node);
		if (rb)
			merge = rb_entry(rb, struct extent_map, rb_node);
		if (rb && mergable_maps(merge, em)) {
			em->start = merge->start;
			em->orig_start = merge->orig_start;
			em->len += merge->len;
			em->block_len += merge->block_len;
			em->block_start = merge->block_start;
			em->mod_len = (em->mod_len + em->mod_start) - merge->mod_start;
			em->mod_start = merge->mod_start;
			em->generation = max(em->generation, merge->generation);

			rb_erase(&merge->rb_node, &tree->map);
			RB_CLEAR_NODE(&merge->rb_node);
			free_extent_map(merge);
		}
	}

	rb = rb_next(&em->rb_node);
	if (rb)
		merge = rb_entry(rb, struct extent_map, rb_node);
	if (rb && mergable_maps(em, merge)) {
		em->len += merge->len;
		em->block_len += merge->block_len;
		rb_erase(&merge->rb_node, &tree->map);
		RB_CLEAR_NODE(&merge->rb_node);
		em->mod_len = (merge->mod_start + merge->mod_len) - em->mod_start;
		em->generation = max(em->generation, merge->generation);
		free_extent_map(merge);
	}
}

/**
 * unpin_extent_cache - unpin an extent from the cache
 * @tree:	tree to unpin the extent in
 * @start:	logical offset in the file
 * @len:	length of the extent
 * @gen:	generation that this extent has been modified in
 *
 * Called after an extent has been written to disk properly.  Set the generation
 * to the generation that actually added the file item to the inode so we know
 * we need to sync this extent when we call fsync().
 */
int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len,
		       u64 gen)
{
	int ret = 0;
	struct extent_map *em;
	bool prealloc = false;

	write_lock(&tree->lock);
	em = lookup_extent_mapping(tree, start, len);

	WARN_ON(!em || em->start != start);

	if (!em)
		goto out;

	em->generation = gen;
	clear_bit(EXTENT_FLAG_PINNED, &em->flags);
	em->mod_start = em->start;
	em->mod_len = em->len;

	if (test_bit(EXTENT_FLAG_FILLING, &em->flags)) {
		prealloc = true;
		clear_bit(EXTENT_FLAG_FILLING, &em->flags);
	}

	try_merge_map(tree, em);

	if (prealloc) {
		em->mod_start = em->start;
		em->mod_len = em->len;
	}

	free_extent_map(em);
out:
	write_unlock(&tree->lock);
	return ret;

}

void clear_em_logging(struct extent_map_tree *tree, struct extent_map *em)
{
	clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
	if (extent_map_in_tree(em))
		try_merge_map(tree, em);
}

static inline void setup_extent_mapping(struct extent_map_tree *tree,
					struct extent_map *em,
					int modified)
{
	atomic_inc(&em->refs);
	em->mod_start = em->start;
	em->mod_len = em->len;

	if (modified)
		list_move(&em->list, &tree->modified_extents);
	else
		try_merge_map(tree, em);
}

/**
 * add_extent_mapping - add new extent map to the extent tree
 * @tree:	tree to insert new map in
 * @em:		map to insert
 *
 * Insert @em into @tree or perform a simple forward/backward merge with
 * existing mappings.  The extent_map struct passed in will be inserted
 * into the tree directly, with an additional reference taken, or a
 * reference dropped if the merge attempt was successful.
 */
int add_extent_mapping(struct extent_map_tree *tree,
		       struct extent_map *em, int modified)
{
	int ret = 0;

	ret = tree_insert(&tree->map, em);
	if (ret)
		goto out;

	setup_extent_mapping(tree, em, modified);
out:
	return ret;
}

static struct extent_map *
__lookup_extent_mapping(struct extent_map_tree *tree,
			u64 start, u64 len, int strict)
{
	struct extent_map *em;
	struct rb_node *rb_node;
	struct rb_node *prev = NULL;
	struct rb_node *next = NULL;
	u64 end = range_end(start, len);

	rb_node = __tree_search(&tree->map, start, &prev, &next);
	if (!rb_node) {
		if (prev)
			rb_node = prev;
		else if (next)
			rb_node = next;
		else
			return NULL;
	}

	em = rb_entry(rb_node, struct extent_map, rb_node);

	if (strict && !(end > em->start && start < extent_map_end(em)))
		return NULL;

	atomic_inc(&em->refs);
	return em;
}

/**
 * lookup_extent_mapping - lookup extent_map
 * @tree:	tree to lookup in
 * @start:	byte offset to start the search
 * @len:	length of the lookup range
 *
 * Find and return the first extent_map struct in @tree that intersects the
 * [start, len] range.  There may be additional objects in the tree that
 * intersect, so check the object returned carefully to make sure that no
 * additional lookups are needed.
 */
struct extent_map *lookup_extent_mapping(struct extent_map_tree *tree,
					 u64 start, u64 len)
{
	return __lookup_extent_mapping(tree, start, len, 1);
}

/**
 * search_extent_mapping - find a nearby extent map
 * @tree:	tree to lookup in
 * @start:	byte offset to start the search
 * @len:	length of the lookup range
 *
 * Find and return the first extent_map struct in @tree that intersects the
 * [start, len] range.
 *
 * If one can't be found, any nearby extent may be returned
 */
struct extent_map *search_extent_mapping(struct extent_map_tree *tree,
					 u64 start, u64 len)
{
	return __lookup_extent_mapping(tree, start, len, 0);
}

/**
 * remove_extent_mapping - removes an extent_map from the extent tree
 * @tree:	extent tree to remove from
 * @em:		extent map beeing removed
 *
 * Removes @em from @tree.  No reference counts are dropped, and no checks
 * are done to see if the range is in use
 */
int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em)
{
	int ret = 0;

	WARN_ON(test_bit(EXTENT_FLAG_PINNED, &em->flags));
	rb_erase(&em->rb_node, &tree->map);
	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
		list_del_init(&em->list);
	RB_CLEAR_NODE(&em->rb_node);
	return ret;
}

void replace_extent_mapping(struct extent_map_tree *tree,
			    struct extent_map *cur,
			    struct extent_map *new,
			    int modified)
{
	WARN_ON(test_bit(EXTENT_FLAG_PINNED, &cur->flags));
	ASSERT(extent_map_in_tree(cur));
	if (!test_bit(EXTENT_FLAG_LOGGING, &cur->flags))
		list_del_init(&cur->list);
	rb_replace_node(&cur->rb_node, &new->rb_node, &tree->map);
	RB_CLEAR_NODE(&cur->rb_node);

	setup_extent_mapping(tree, new, modified);
}
Commit	Line	Data
d1310b2e	1	#include <linux/err.h>
d1310b2e	2	#include <linux/slab.h>
a52d9a80	3	#include <linux/spinlock.h>
d1310b2e	4	#include <linux/hardirq.h>
261507a0	5	#include "ctree.h"
a52d9a80 CM	6	#include "extent_map.h"
a52d9a80 CM	7
86479a04	8
a52d9a80	9	static struct kmem_cache *extent_map_cache;
ca664626	10
2f4cbe64	11	int __init extent_map_init(void)
a52d9a80	12	{
837e1972	13	extent_map_cache = kmem_cache_create("btrfs_extent_map",
9601e3f6 CH	14	sizeof(struct extent_map), 0,
9601e3f6 CH	15	SLAB_RECLAIM_ACCOUNT \| SLAB_MEM_SPREAD, NULL);
2f4cbe64 WB	16	if (!extent_map_cache)
2f4cbe64 WB	17	return -ENOMEM;
2f4cbe64	18	return 0;
a52d9a80 CM	19	}
a52d9a80 CM	20
17636e03	21	void extent_map_exit(void)
a52d9a80	22	{
a52d9a80 CM	23	if (extent_map_cache)
a52d9a80 CM	24	kmem_cache_destroy(extent_map_cache);
a52d9a80 CM	25	}
a52d9a80 CM	26
9d2423c5 CH	27	/**
	28	* extent_map_tree_init - initialize extent map tree
	29	* @tree: tree to initialize
9d2423c5 CH	30	*
	31	* Initialize the extent tree @tree. Should be called for each new inode
	32	* or other user of the extent_map interface.
	33	*/
a8067e02	34	void extent_map_tree_init(struct extent_map_tree *tree)
a52d9a80	35	{
6bef4d31	36	tree->map = RB_ROOT;
5dc562c5	37	INIT_LIST_HEAD(&tree->modified_extents);
890871be	38	rwlock_init(&tree->lock);
a52d9a80	39	}
a52d9a80	40
9d2423c5 CH	41	/**
9d2423c5 CH	42	* alloc_extent_map - allocate new extent map structure
9d2423c5 CH	43	*
	44	* Allocate a new extent_map structure. The new structure is
	45	* returned with a reference count of one and needs to be
	46	* freed using free_extent_map()
	47	*/
172ddd60	48	struct extent_map *alloc_extent_map(void)
a52d9a80 CM	49	{
a52d9a80 CM	50	struct extent_map *em;
70c8a91c	51	em = kmem_cache_zalloc(extent_map_cache, GFP_NOFS);
c26a9203 TI	52	if (!em)
c26a9203 TI	53	return NULL;
cbc0e928	54	RB_CLEAR_NODE(&em->rb_node);
d1310b2e	55	em->flags = 0;
261507a0	56	em->compress_type = BTRFS_COMPRESS_NONE;
5dc562c5	57	em->generation = 0;
a52d9a80	58	atomic_set(&em->refs, 1);
5dc562c5	59	INIT_LIST_HEAD(&em->list);
a52d9a80 CM	60	return em;
a52d9a80 CM	61	}
a52d9a80	62
9d2423c5 CH	63	/**
	64	* free_extent_map - drop reference count of an extent_map
	65	* @em: extent map beeing releasead
	66	*
	67	* Drops the reference out on @em by one and free the structure
	68	* if the reference count hits zero.
	69	*/
a52d9a80 CM	70	void free_extent_map(struct extent_map *em)
a52d9a80 CM	71	{
2bf5a725 CM	72	if (!em)
2bf5a725 CM	73	return;
d1310b2e	74	WARN_ON(atomic_read(&em->refs) == 0);
a52d9a80	75	if (atomic_dec_and_test(&em->refs)) {
cbc0e928	76	WARN_ON(extent_map_in_tree(em));
5dc562c5	77	WARN_ON(!list_empty(&em->list));
298a8f9c	78	if (test_bit(EXTENT_FLAG_FS_MAPPING, &em->flags))
95617d69	79	kfree(em->map_lookup);
a52d9a80 CM	80	kmem_cache_free(extent_map_cache, em);
	81	}
	82	}
a52d9a80	83
32193c14 FDBM	84	/* simple helper to do math around the end of an extent, handling wrap */
	85	static u64 range_end(u64 start, u64 len)
	86	{
	87	if (start + len < start)
	88	return (u64)-1;
	89	return start + len;
	90	}
	91
	92	static int tree_insert(struct rb_root root, struct extent_map em)
a52d9a80	93	{
d397712b CM	94	struct rb_node **p = &root->rb_node;
d397712b CM	95	struct rb_node *parent = NULL;
32193c14 FDBM	96	struct extent_map *entry = NULL;
	97	struct rb_node *orig_parent = NULL;
	98	u64 end = range_end(em->start, em->len);
a52d9a80	99
d397712b	100	while (*p) {
a52d9a80	101	parent = *p;
d1310b2e CM	102	entry = rb_entry(parent, struct extent_map, rb_node);
d1310b2e CM	103
32193c14	104	if (em->start < entry->start)
a52d9a80	105	p = &(*p)->rb_left;
32193c14	106	else if (em->start >= extent_map_end(entry))
a52d9a80 CM	107	p = &(*p)->rb_right;
a52d9a80 CM	108	else
32193c14	109	return -EEXIST;
a52d9a80 CM	110	}
a52d9a80 CM	111
32193c14 FDBM	112	orig_parent = parent;
	113	while (parent && em->start >= extent_map_end(entry)) {
	114	parent = rb_next(parent);
	115	entry = rb_entry(parent, struct extent_map, rb_node);
	116	}
	117	if (parent)
	118	if (end > entry->start && em->start < extent_map_end(entry))
	119	return -EEXIST;
	120
	121	parent = orig_parent;
	122	entry = rb_entry(parent, struct extent_map, rb_node);
	123	while (parent && em->start < entry->start) {
	124	parent = rb_prev(parent);
	125	entry = rb_entry(parent, struct extent_map, rb_node);
	126	}
	127	if (parent)
	128	if (end > entry->start && em->start < extent_map_end(entry))
	129	return -EEXIST;
	130
32193c14 FDBM	131	rb_link_node(&em->rb_node, orig_parent, p);
	132	rb_insert_color(&em->rb_node, root);
	133	return 0;
a52d9a80 CM	134	}
a52d9a80 CM	135
d352ac68 CM	136	/*
	137	* search through the tree for an extent_map with a given offset. If
	138	* it can't be found, try to find some neighboring extents
	139	*/
a52d9a80	140	static struct rb_node __tree_search(struct rb_root root, u64 offset,
5f56406a CM	141	struct rb_node **prev_ret,
5f56406a CM	142	struct rb_node **next_ret)
a52d9a80	143	{
d397712b	144	struct rb_node *n = root->rb_node;
a52d9a80	145	struct rb_node *prev = NULL;
5f56406a	146	struct rb_node *orig_prev = NULL;
d1310b2e CM	147	struct extent_map *entry;
d1310b2e CM	148	struct extent_map *prev_entry = NULL;
a52d9a80	149
d397712b	150	while (n) {
d1310b2e	151	entry = rb_entry(n, struct extent_map, rb_node);
a52d9a80 CM	152	prev = n;
	153	prev_entry = entry;
	154
	155	if (offset < entry->start)
	156	n = n->rb_left;
d1310b2e	157	else if (offset >= extent_map_end(entry))
a52d9a80 CM	158	n = n->rb_right;
	159	else
	160	return n;
	161	}
5f56406a CM	162
	163	if (prev_ret) {
	164	orig_prev = prev;
d397712b	165	while (prev && offset >= extent_map_end(prev_entry)) {
5f56406a	166	prev = rb_next(prev);
d1310b2e	167	prev_entry = rb_entry(prev, struct extent_map, rb_node);
5f56406a CM	168	}
	169	*prev_ret = prev;
	170	prev = orig_prev;
	171	}
	172
	173	if (next_ret) {
d1310b2e	174	prev_entry = rb_entry(prev, struct extent_map, rb_node);
d397712b	175	while (prev && offset < prev_entry->start) {
5f56406a	176	prev = rb_prev(prev);
d1310b2e	177	prev_entry = rb_entry(prev, struct extent_map, rb_node);
5f56406a CM	178	}
5f56406a CM	179	*next_ret = prev;
a52d9a80	180	}
a52d9a80 CM	181	return NULL;
	182	}
	183
d352ac68	184	/* check to see if two extent_map structs are adjacent and safe to merge */
d1310b2e	185	static int mergable_maps(struct extent_map prev, struct extent_map next)
a52d9a80	186	{
7f3c74fb CM	187	if (test_bit(EXTENT_FLAG_PINNED, &prev->flags))
	188	return 0;
	189
c8b97818 CM	190	/*
	191	* don't merge compressed extents, we need to know their
	192	* actual size
	193	*/
	194	if (test_bit(EXTENT_FLAG_COMPRESSED, &prev->flags))
	195	return 0;
	196
201a9038 JB	197	if (test_bit(EXTENT_FLAG_LOGGING, &prev->flags) \|\|
	198	test_bit(EXTENT_FLAG_LOGGING, &next->flags))
	199	return 0;
	200
09a2a8f9 JB	201	/*
	202	* We don't want to merge stuff that hasn't been written to the log yet
	203	* since it may not reflect exactly what is on disk, and that would be
	204	* bad.
	205	*/
	206	if (!list_empty(&prev->list) \|\| !list_empty(&next->list))
	207	return 0;
	208
d1310b2e CM	209	if (extent_map_end(prev) == next->start &&
	210	prev->flags == next->flags &&
	211	prev->bdev == next->bdev &&
	212	((next->block_start == EXTENT_MAP_HOLE &&
	213	prev->block_start == EXTENT_MAP_HOLE) \|\|
	214	(next->block_start == EXTENT_MAP_INLINE &&
	215	prev->block_start == EXTENT_MAP_INLINE) \|\|
	216	(next->block_start == EXTENT_MAP_DELALLOC &&
	217	prev->block_start == EXTENT_MAP_DELALLOC) \|\|
	218	(next->block_start < EXTENT_MAP_LAST_BYTE - 1 &&
	219	next->block_start == extent_map_block_end(prev)))) {
	220	return 1;
	221	}
a52d9a80 CM	222	return 0;
	223	}
	224
4d2c8f62	225	static void try_merge_map(struct extent_map_tree tree, struct extent_map em)
a1ed835e	226	{
a1ed835e CM	227	struct extent_map *merge = NULL;
a1ed835e CM	228	struct rb_node *rb;
a1ed835e CM	229
	230	if (em->start != 0) {
	231	rb = rb_prev(&em->rb_node);
	232	if (rb)
	233	merge = rb_entry(rb, struct extent_map, rb_node);
	234	if (rb && mergable_maps(merge, em)) {
	235	em->start = merge->start;
70c8a91c	236	em->orig_start = merge->orig_start;
a1ed835e CM	237	em->len += merge->len;
	238	em->block_len += merge->block_len;
	239	em->block_start = merge->block_start;
70c8a91c JB	240	em->mod_len = (em->mod_len + em->mod_start) - merge->mod_start;
	241	em->mod_start = merge->mod_start;
	242	em->generation = max(em->generation, merge->generation);
5dc562c5	243
a1ed835e	244	rb_erase(&merge->rb_node, &tree->map);
cbc0e928	245	RB_CLEAR_NODE(&merge->rb_node);
a1ed835e CM	246	free_extent_map(merge);
	247	}
	248	}
	249
	250	rb = rb_next(&em->rb_node);
	251	if (rb)
	252	merge = rb_entry(rb, struct extent_map, rb_node);
	253	if (rb && mergable_maps(em, merge)) {
	254	em->len += merge->len;
d527afe1	255	em->block_len += merge->block_len;
a1ed835e	256	rb_erase(&merge->rb_node, &tree->map);
cbc0e928	257	RB_CLEAR_NODE(&merge->rb_node);
70c8a91c JB	258	em->mod_len = (merge->mod_start + merge->mod_len) - em->mod_start;
70c8a91c JB	259	em->generation = max(em->generation, merge->generation);
a1ed835e CM	260	free_extent_map(merge);
a1ed835e CM	261	}
4d2c8f62 LZ	262	}
4d2c8f62 LZ	263
5dc562c5	264	/**
52b1de91	265	* unpin_extent_cache - unpin an extent from the cache
5dc562c5 JB	266	* @tree: tree to unpin the extent in
	267	* @start: logical offset in the file
	268	* @len: length of the extent
	269	* @gen: generation that this extent has been modified in
5dc562c5 JB	270	*
	271	* Called after an extent has been written to disk properly. Set the generation
	272	* to the generation that actually added the file item to the inode so we know
	273	* we need to sync this extent when we call fsync().
	274	*/
	275	int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len,
	276	u64 gen)
4d2c8f62 LZ	277	{
	278	int ret = 0;
	279	struct extent_map *em;
4e2f84e6	280	bool prealloc = false;
4d2c8f62 LZ	281
	282	write_lock(&tree->lock);
	283	em = lookup_extent_mapping(tree, start, len);
	284
	285	WARN_ON(!em \|\| em->start != start);
	286
	287	if (!em)
	288	goto out;
	289
5dc562c5	290	em->generation = gen;
4d2c8f62	291	clear_bit(EXTENT_FLAG_PINNED, &em->flags);
4e2f84e6 LB	292	em->mod_start = em->start;
	293	em->mod_len = em->len;
	294
b11e234d	295	if (test_bit(EXTENT_FLAG_FILLING, &em->flags)) {
4e2f84e6	296	prealloc = true;
b11e234d	297	clear_bit(EXTENT_FLAG_FILLING, &em->flags);
4e2f84e6	298	}
4d2c8f62 LZ	299
4d2c8f62 LZ	300	try_merge_map(tree, em);
4e2f84e6 LB	301
	302	if (prealloc) {
	303	em->mod_start = em->start;
	304	em->mod_len = em->len;
	305	}
	306
a1ed835e CM	307	free_extent_map(em);
	308	out:
	309	write_unlock(&tree->lock);
	310	return ret;
	311
	312	}
	313
201a9038 JB	314	void clear_em_logging(struct extent_map_tree tree, struct extent_map em)
	315	{
	316	clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
cbc0e928	317	if (extent_map_in_tree(em))
222c81dc	318	try_merge_map(tree, em);
201a9038 JB	319	}
201a9038 JB	320
176840b3 FM	321	static inline void setup_extent_mapping(struct extent_map_tree *tree,
	322	struct extent_map *em,
	323	int modified)
	324	{
	325	atomic_inc(&em->refs);
	326	em->mod_start = em->start;
	327	em->mod_len = em->len;
	328
	329	if (modified)
	330	list_move(&em->list, &tree->modified_extents);
	331	else
	332	try_merge_map(tree, em);
	333	}
	334
9d2423c5 CH	335	/**
	336	* add_extent_mapping - add new extent map to the extent tree
	337	* @tree: tree to insert new map in
	338	* @em: map to insert
	339	*
	340	* Insert @em into @tree or perform a simple forward/backward merge with
	341	* existing mappings. The extent_map struct passed in will be inserted
	342	* into the tree directly, with an additional reference taken, or a
25985edc	343	* reference dropped if the merge attempt was successful.
a52d9a80 CM	344	*/
a52d9a80 CM	345	int add_extent_mapping(struct extent_map_tree *tree,
09a2a8f9	346	struct extent_map *em, int modified)
a52d9a80 CM	347	{
a52d9a80 CM	348	int ret = 0;
a52d9a80	349
32193c14 FDBM	350	ret = tree_insert(&tree->map, em);
32193c14 FDBM	351	if (ret)
a52d9a80	352	goto out;
32193c14	353
176840b3	354	setup_extent_mapping(tree, em, modified);
a52d9a80	355	out:
a52d9a80 CM	356	return ret;
a52d9a80 CM	357	}
a52d9a80	358
48a3b636 ES	359	static struct extent_map *
	360	__lookup_extent_mapping(struct extent_map_tree *tree,
	361	u64 start, u64 len, int strict)
a52d9a80 CM	362	{
	363	struct extent_map *em;
	364	struct rb_node *rb_node;
306929f3 CH	365	struct rb_node *prev = NULL;
	366	struct rb_node *next = NULL;
	367	u64 end = range_end(start, len);
	368
5f56406a	369	rb_node = __tree_search(&tree->map, start, &prev, &next);
a52d9a80	370	if (!rb_node) {
ed64f066 LZ	371	if (prev)
	372	rb_node = prev;
	373	else if (next)
	374	rb_node = next;
	375	else
	376	return NULL;
a52d9a80	377	}
ed64f066	378
a52d9a80	379	em = rb_entry(rb_node, struct extent_map, rb_node);
d1310b2e	380
ed64f066 LZ	381	if (strict && !(end > em->start && start < extent_map_end(em)))
ed64f066 LZ	382	return NULL;
d1310b2e	383
a52d9a80	384	atomic_inc(&em->refs);
a52d9a80 CM	385	return em;
a52d9a80 CM	386	}
a52d9a80	387
ed64f066 LZ	388	/**
	389	* lookup_extent_mapping - lookup extent_map
	390	* @tree: tree to lookup in
	391	* @start: byte offset to start the search
	392	* @len: length of the lookup range
	393	*
	394	* Find and return the first extent_map struct in @tree that intersects the
	395	* [start, len] range. There may be additional objects in the tree that
	396	* intersect, so check the object returned carefully to make sure that no
	397	* additional lookups are needed.
	398	*/
	399	struct extent_map lookup_extent_mapping(struct extent_map_tree tree,
	400	u64 start, u64 len)
	401	{
	402	return __lookup_extent_mapping(tree, start, len, 1);
	403	}
	404
b917b7c3 CM	405	/**
	406	* search_extent_mapping - find a nearby extent map
	407	* @tree: tree to lookup in
	408	* @start: byte offset to start the search
	409	* @len: length of the lookup range
	410	*
	411	* Find and return the first extent_map struct in @tree that intersects the
	412	* [start, len] range.
	413	*
	414	* If one can't be found, any nearby extent may be returned
	415	*/
	416	struct extent_map search_extent_mapping(struct extent_map_tree tree,
	417	u64 start, u64 len)
	418	{
ed64f066	419	return __lookup_extent_mapping(tree, start, len, 0);
b917b7c3 CM	420	}
b917b7c3 CM	421
9d2423c5 CH	422	/**
	423	* remove_extent_mapping - removes an extent_map from the extent tree
	424	* @tree: extent tree to remove from
	425	* @em: extent map beeing removed
	426	*
	427	* Removes @em from @tree. No reference counts are dropped, and no checks
	428	* are done to see if the range is in use
a52d9a80 CM	429	*/
	430	int remove_extent_mapping(struct extent_map_tree tree, struct extent_map em)
	431	{
d1310b2e	432	int ret = 0;
a52d9a80	433
7f3c74fb	434	WARN_ON(test_bit(EXTENT_FLAG_PINNED, &em->flags));
d1310b2e	435	rb_erase(&em->rb_node, &tree->map);
ff44c6e3 JB	436	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
ff44c6e3 JB	437	list_del_init(&em->list);
cbc0e928	438	RB_CLEAR_NODE(&em->rb_node);
a52d9a80 CM	439	return ret;
a52d9a80 CM	440	}
176840b3 FM	441
	442	void replace_extent_mapping(struct extent_map_tree *tree,
	443	struct extent_map *cur,
	444	struct extent_map *new,
	445	int modified)
	446	{
	447	WARN_ON(test_bit(EXTENT_FLAG_PINNED, &cur->flags));
	448	ASSERT(extent_map_in_tree(cur));
	449	if (!test_bit(EXTENT_FLAG_LOGGING, &cur->flags))
	450	list_del_init(&cur->list);
	451	rb_replace_node(&cur->rb_node, &new->rb_node, &tree->map);
	452	RB_CLEAR_NODE(&cur->rb_node);
	453
	454	setup_extent_mapping(tree, new, modified);
	455	}