[deliverable/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;
	em->in_tree = 0;
	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(em->in_tree);
		WARN_ON(!list_empty(&em->list));
		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);

		WARN_ON(!entry->in_tree);

		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;

	em->in_tree = 1;
	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;

		WARN_ON(!entry->in_tree);

		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;
			merge->in_tree = 0;
			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);
			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->len;
		rb_erase(&merge->rb_node, &tree->map);
		merge->in_tree = 0;
		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;

	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
		list_move(&em->list, &tree->modified_extents);
	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 (em->in_tree)
		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;

	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);
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);
	em->in_tree = 0;
	return ret;
}
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;
a52d9a80	54	em->in_tree = 0;
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 CM	75	if (atomic_dec_and_test(&em->refs)) {
a52d9a80 CM	76	WARN_ON(em->in_tree);
5dc562c5	77	WARN_ON(!list_empty(&em->list));
a52d9a80 CM	78	kmem_cache_free(extent_map_cache, em);
	79	}
	80	}
a52d9a80	81
32193c14 FDBM	82	/* simple helper to do math around the end of an extent, handling wrap */
	83	static u64 range_end(u64 start, u64 len)
	84	{
	85	if (start + len < start)
	86	return (u64)-1;
	87	return start + len;
	88	}
	89
	90	static int tree_insert(struct rb_root root, struct extent_map em)
a52d9a80	91	{
d397712b CM	92	struct rb_node **p = &root->rb_node;
d397712b CM	93	struct rb_node *parent = NULL;
32193c14 FDBM	94	struct extent_map *entry = NULL;
	95	struct rb_node *orig_parent = NULL;
	96	u64 end = range_end(em->start, em->len);
a52d9a80	97
d397712b	98	while (*p) {
a52d9a80	99	parent = *p;
d1310b2e CM	100	entry = rb_entry(parent, struct extent_map, rb_node);
	101
	102	WARN_ON(!entry->in_tree);
a52d9a80	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
	131	em->in_tree = 1;
	132	rb_link_node(&em->rb_node, orig_parent, p);
	133	rb_insert_color(&em->rb_node, root);
	134	return 0;
a52d9a80 CM	135	}
a52d9a80 CM	136
d352ac68 CM	137	/*
	138	* search through the tree for an extent_map with a given offset. If
	139	* it can't be found, try to find some neighboring extents
	140	*/
a52d9a80	141	static struct rb_node __tree_search(struct rb_root root, u64 offset,
5f56406a CM	142	struct rb_node **prev_ret,
5f56406a CM	143	struct rb_node **next_ret)
a52d9a80	144	{
d397712b	145	struct rb_node *n = root->rb_node;
a52d9a80	146	struct rb_node *prev = NULL;
5f56406a	147	struct rb_node *orig_prev = NULL;
d1310b2e CM	148	struct extent_map *entry;
d1310b2e CM	149	struct extent_map *prev_entry = NULL;
a52d9a80	150
d397712b	151	while (n) {
d1310b2e	152	entry = rb_entry(n, struct extent_map, rb_node);
a52d9a80 CM	153	prev = n;
	154	prev_entry = entry;
	155
d1310b2e CM	156	WARN_ON(!entry->in_tree);
d1310b2e CM	157
a52d9a80 CM	158	if (offset < entry->start)
a52d9a80 CM	159	n = n->rb_left;
d1310b2e	160	else if (offset >= extent_map_end(entry))
a52d9a80 CM	161	n = n->rb_right;
	162	else
	163	return n;
	164	}
5f56406a CM	165
	166	if (prev_ret) {
	167	orig_prev = prev;
d397712b	168	while (prev && offset >= extent_map_end(prev_entry)) {
5f56406a	169	prev = rb_next(prev);
d1310b2e	170	prev_entry = rb_entry(prev, struct extent_map, rb_node);
5f56406a CM	171	}
	172	*prev_ret = prev;
	173	prev = orig_prev;
	174	}
	175
	176	if (next_ret) {
d1310b2e	177	prev_entry = rb_entry(prev, struct extent_map, rb_node);
d397712b	178	while (prev && offset < prev_entry->start) {
5f56406a	179	prev = rb_prev(prev);
d1310b2e	180	prev_entry = rb_entry(prev, struct extent_map, rb_node);
5f56406a CM	181	}
5f56406a CM	182	*next_ret = prev;
a52d9a80	183	}
a52d9a80 CM	184	return NULL;
	185	}
	186
d352ac68	187	/* check to see if two extent_map structs are adjacent and safe to merge */
d1310b2e	188	static int mergable_maps(struct extent_map prev, struct extent_map next)
a52d9a80	189	{
7f3c74fb CM	190	if (test_bit(EXTENT_FLAG_PINNED, &prev->flags))
	191	return 0;
	192
c8b97818 CM	193	/*
	194	* don't merge compressed extents, we need to know their
	195	* actual size
	196	*/
	197	if (test_bit(EXTENT_FLAG_COMPRESSED, &prev->flags))
	198	return 0;
	199
201a9038 JB	200	if (test_bit(EXTENT_FLAG_LOGGING, &prev->flags) \|\|
	201	test_bit(EXTENT_FLAG_LOGGING, &next->flags))
	202	return 0;
	203
09a2a8f9 JB	204	/*
	205	* We don't want to merge stuff that hasn't been written to the log yet
	206	* since it may not reflect exactly what is on disk, and that would be
	207	* bad.
	208	*/
	209	if (!list_empty(&prev->list) \|\| !list_empty(&next->list))
	210	return 0;
	211
d1310b2e CM	212	if (extent_map_end(prev) == next->start &&
	213	prev->flags == next->flags &&
	214	prev->bdev == next->bdev &&
	215	((next->block_start == EXTENT_MAP_HOLE &&
	216	prev->block_start == EXTENT_MAP_HOLE) \|\|
	217	(next->block_start == EXTENT_MAP_INLINE &&
	218	prev->block_start == EXTENT_MAP_INLINE) \|\|
	219	(next->block_start == EXTENT_MAP_DELALLOC &&
	220	prev->block_start == EXTENT_MAP_DELALLOC) \|\|
	221	(next->block_start < EXTENT_MAP_LAST_BYTE - 1 &&
	222	next->block_start == extent_map_block_end(prev)))) {
	223	return 1;
	224	}
a52d9a80 CM	225	return 0;
	226	}
	227
4d2c8f62	228	static void try_merge_map(struct extent_map_tree tree, struct extent_map em)
a1ed835e	229	{
a1ed835e CM	230	struct extent_map *merge = NULL;
a1ed835e CM	231	struct rb_node *rb;
a1ed835e CM	232
	233	if (em->start != 0) {
	234	rb = rb_prev(&em->rb_node);
	235	if (rb)
	236	merge = rb_entry(rb, struct extent_map, rb_node);
	237	if (rb && mergable_maps(merge, em)) {
	238	em->start = merge->start;
70c8a91c	239	em->orig_start = merge->orig_start;
a1ed835e CM	240	em->len += merge->len;
	241	em->block_len += merge->block_len;
	242	em->block_start = merge->block_start;
	243	merge->in_tree = 0;
70c8a91c JB	244	em->mod_len = (em->mod_len + em->mod_start) - merge->mod_start;
	245	em->mod_start = merge->mod_start;
	246	em->generation = max(em->generation, merge->generation);
5dc562c5	247
a1ed835e CM	248	rb_erase(&merge->rb_node, &tree->map);
	249	free_extent_map(merge);
	250	}
	251	}
	252
	253	rb = rb_next(&em->rb_node);
	254	if (rb)
	255	merge = rb_entry(rb, struct extent_map, rb_node);
	256	if (rb && mergable_maps(em, merge)) {
	257	em->len += merge->len;
	258	em->block_len += merge->len;
	259	rb_erase(&merge->rb_node, &tree->map);
	260	merge->in_tree = 0;
70c8a91c JB	261	em->mod_len = (merge->mod_start + merge->mod_len) - em->mod_start;
70c8a91c JB	262	em->generation = max(em->generation, merge->generation);
a1ed835e CM	263	free_extent_map(merge);
a1ed835e CM	264	}
4d2c8f62 LZ	265	}
4d2c8f62 LZ	266
5dc562c5	267	/**
52b1de91	268	* unpin_extent_cache - unpin an extent from the cache
5dc562c5 JB	269	* @tree: tree to unpin the extent in
	270	* @start: logical offset in the file
	271	* @len: length of the extent
	272	* @gen: generation that this extent has been modified in
5dc562c5 JB	273	*
	274	* Called after an extent has been written to disk properly. Set the generation
	275	* to the generation that actually added the file item to the inode so we know
	276	* we need to sync this extent when we call fsync().
	277	*/
	278	int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len,
	279	u64 gen)
4d2c8f62 LZ	280	{
	281	int ret = 0;
	282	struct extent_map *em;
4e2f84e6	283	bool prealloc = false;
4d2c8f62 LZ	284
	285	write_lock(&tree->lock);
	286	em = lookup_extent_mapping(tree, start, len);
	287
	288	WARN_ON(!em \|\| em->start != start);
	289
	290	if (!em)
	291	goto out;
	292
201a9038 JB	293	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
201a9038 JB	294	list_move(&em->list, &tree->modified_extents);
5dc562c5	295	em->generation = gen;
4d2c8f62	296	clear_bit(EXTENT_FLAG_PINNED, &em->flags);
4e2f84e6 LB	297	em->mod_start = em->start;
	298	em->mod_len = em->len;
	299
b11e234d	300	if (test_bit(EXTENT_FLAG_FILLING, &em->flags)) {
4e2f84e6	301	prealloc = true;
b11e234d	302	clear_bit(EXTENT_FLAG_FILLING, &em->flags);
4e2f84e6	303	}
4d2c8f62 LZ	304
4d2c8f62 LZ	305	try_merge_map(tree, em);
4e2f84e6 LB	306
	307	if (prealloc) {
	308	em->mod_start = em->start;
	309	em->mod_len = em->len;
	310	}
	311
a1ed835e CM	312	free_extent_map(em);
	313	out:
	314	write_unlock(&tree->lock);
	315	return ret;
	316
	317	}
	318
201a9038 JB	319	void clear_em_logging(struct extent_map_tree tree, struct extent_map em)
	320	{
	321	clear_bit(EXTENT_FLAG_LOGGING, &em->flags);
222c81dc JB	322	if (em->in_tree)
222c81dc JB	323	try_merge_map(tree, em);
201a9038 JB	324	}
201a9038 JB	325
9d2423c5 CH	326	/**
	327	* add_extent_mapping - add new extent map to the extent tree
	328	* @tree: tree to insert new map in
	329	* @em: map to insert
	330	*
	331	* Insert @em into @tree or perform a simple forward/backward merge with
	332	* existing mappings. The extent_map struct passed in will be inserted
	333	* into the tree directly, with an additional reference taken, or a
25985edc	334	* reference dropped if the merge attempt was successful.
a52d9a80 CM	335	*/
a52d9a80 CM	336	int add_extent_mapping(struct extent_map_tree *tree,
09a2a8f9	337	struct extent_map *em, int modified)
a52d9a80 CM	338	{
a52d9a80 CM	339	int ret = 0;
a52d9a80	340
32193c14 FDBM	341	ret = tree_insert(&tree->map, em);
32193c14 FDBM	342	if (ret)
a52d9a80	343	goto out;
32193c14	344
a52d9a80	345	atomic_inc(&em->refs);
4d2c8f62	346
4e2f84e6 LB	347	em->mod_start = em->start;
	348	em->mod_len = em->len;
	349
09a2a8f9 JB	350	if (modified)
	351	list_move(&em->list, &tree->modified_extents);
	352	else
	353	try_merge_map(tree, em);
a52d9a80	354	out:
a52d9a80 CM	355	return ret;
a52d9a80 CM	356	}
a52d9a80	357
48a3b636 ES	358	static struct extent_map *
	359	__lookup_extent_mapping(struct extent_map_tree *tree,
	360	u64 start, u64 len, int strict)
a52d9a80 CM	361	{
	362	struct extent_map *em;
	363	struct rb_node *rb_node;
306929f3 CH	364	struct rb_node *prev = NULL;
	365	struct rb_node *next = NULL;
	366	u64 end = range_end(start, len);
	367
5f56406a	368	rb_node = __tree_search(&tree->map, start, &prev, &next);
a52d9a80	369	if (!rb_node) {
ed64f066 LZ	370	if (prev)
	371	rb_node = prev;
	372	else if (next)
	373	rb_node = next;
	374	else
	375	return NULL;
a52d9a80	376	}
ed64f066	377
a52d9a80	378	em = rb_entry(rb_node, struct extent_map, rb_node);
d1310b2e	379
ed64f066 LZ	380	if (strict && !(end > em->start && start < extent_map_end(em)))
ed64f066 LZ	381	return NULL;
d1310b2e	382
a52d9a80	383	atomic_inc(&em->refs);
a52d9a80 CM	384	return em;
a52d9a80 CM	385	}
a52d9a80	386
ed64f066 LZ	387	/**
	388	* lookup_extent_mapping - lookup extent_map
	389	* @tree: tree to lookup in
	390	* @start: byte offset to start the search
	391	* @len: length of the lookup range
	392	*
	393	* Find and return the first extent_map struct in @tree that intersects the
	394	* [start, len] range. There may be additional objects in the tree that
	395	* intersect, so check the object returned carefully to make sure that no
	396	* additional lookups are needed.
	397	*/
	398	struct extent_map lookup_extent_mapping(struct extent_map_tree tree,
	399	u64 start, u64 len)
	400	{
	401	return __lookup_extent_mapping(tree, start, len, 1);
	402	}
	403
b917b7c3 CM	404	/**
	405	* search_extent_mapping - find a nearby extent map
	406	* @tree: tree to lookup in
	407	* @start: byte offset to start the search
	408	* @len: length of the lookup range
	409	*
	410	* Find and return the first extent_map struct in @tree that intersects the
	411	* [start, len] range.
	412	*
	413	* If one can't be found, any nearby extent may be returned
	414	*/
	415	struct extent_map search_extent_mapping(struct extent_map_tree tree,
	416	u64 start, u64 len)
	417	{
ed64f066	418	return __lookup_extent_mapping(tree, start, len, 0);
b917b7c3 CM	419	}
b917b7c3 CM	420
9d2423c5 CH	421	/**
	422	* remove_extent_mapping - removes an extent_map from the extent tree
	423	* @tree: extent tree to remove from
	424	* @em: extent map beeing removed
	425	*
	426	* Removes @em from @tree. No reference counts are dropped, and no checks
	427	* are done to see if the range is in use
a52d9a80 CM	428	*/
	429	int remove_extent_mapping(struct extent_map_tree tree, struct extent_map em)
	430	{
d1310b2e	431	int ret = 0;
a52d9a80	432
7f3c74fb	433	WARN_ON(test_bit(EXTENT_FLAG_PINNED, &em->flags));
d1310b2e	434	rb_erase(&em->rb_node, &tree->map);
ff44c6e3 JB	435	if (!test_bit(EXTENT_FLAG_LOGGING, &em->flags))
ff44c6e3 JB	436	list_del_init(&em->list);
d1310b2e	437	em->in_tree = 0;
a52d9a80 CM	438	return ret;
a52d9a80 CM	439	}