2008-01-25 00:13:08 +03:00
|
|
|
#include <linux/err.h>
|
|
|
|
#include <linux/slab.h>
|
2007-08-28 00:49:44 +04:00
|
|
|
#include <linux/module.h>
|
|
|
|
#include <linux/spinlock.h>
|
2008-01-25 00:13:08 +03:00
|
|
|
#include <linux/hardirq.h>
|
2010-12-17 09:21:50 +03:00
|
|
|
#include "ctree.h"
|
2007-08-28 00:49:44 +04:00
|
|
|
#include "extent_map.h"
|
|
|
|
|
2007-09-11 03:58:16 +04:00
|
|
|
|
2007-08-28 00:49:44 +04:00
|
|
|
static struct kmem_cache *extent_map_cache;
|
2007-11-27 19:16:35 +03:00
|
|
|
|
2007-11-19 18:22:33 +03:00
|
|
|
int __init extent_map_init(void)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
2009-04-13 17:33:09 +04:00
|
|
|
extent_map_cache = kmem_cache_create("extent_map",
|
|
|
|
sizeof(struct extent_map), 0,
|
|
|
|
SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD, NULL);
|
2007-11-19 18:22:33 +03:00
|
|
|
if (!extent_map_cache)
|
|
|
|
return -ENOMEM;
|
|
|
|
return 0;
|
2007-08-28 00:49:44 +04:00
|
|
|
}
|
|
|
|
|
2007-12-11 17:25:06 +03:00
|
|
|
void extent_map_exit(void)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
|
|
|
if (extent_map_cache)
|
|
|
|
kmem_cache_destroy(extent_map_cache);
|
|
|
|
}
|
|
|
|
|
2008-06-12 05:52:17 +04:00
|
|
|
/**
|
|
|
|
* 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.
|
|
|
|
*/
|
2011-04-21 02:34:43 +04:00
|
|
|
void extent_map_tree_init(struct extent_map_tree *tree)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
2010-02-23 22:43:04 +03:00
|
|
|
tree->map = RB_ROOT;
|
2009-09-03 00:24:52 +04:00
|
|
|
rwlock_init(&tree->lock);
|
2007-08-28 00:49:44 +04:00
|
|
|
}
|
|
|
|
|
2008-06-12 05:52:17 +04:00
|
|
|
/**
|
|
|
|
* 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()
|
|
|
|
*/
|
2011-04-21 02:48:27 +04:00
|
|
|
struct extent_map *alloc_extent_map(void)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
|
|
|
struct extent_map *em;
|
2011-04-21 02:48:27 +04:00
|
|
|
em = kmem_cache_alloc(extent_map_cache, GFP_NOFS);
|
2011-02-14 03:45:29 +03:00
|
|
|
if (!em)
|
|
|
|
return NULL;
|
2007-08-28 00:49:44 +04:00
|
|
|
em->in_tree = 0;
|
2008-01-25 00:13:08 +03:00
|
|
|
em->flags = 0;
|
2010-12-17 09:21:50 +03:00
|
|
|
em->compress_type = BTRFS_COMPRESS_NONE;
|
2007-08-28 00:49:44 +04:00
|
|
|
atomic_set(&em->refs, 1);
|
|
|
|
return em;
|
|
|
|
}
|
|
|
|
|
2008-06-12 05:52:17 +04:00
|
|
|
/**
|
|
|
|
* 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.
|
|
|
|
*/
|
2007-08-28 00:49:44 +04:00
|
|
|
void free_extent_map(struct extent_map *em)
|
|
|
|
{
|
2007-08-30 19:54:02 +04:00
|
|
|
if (!em)
|
|
|
|
return;
|
2008-01-25 00:13:08 +03:00
|
|
|
WARN_ON(atomic_read(&em->refs) == 0);
|
2007-08-28 00:49:44 +04:00
|
|
|
if (atomic_dec_and_test(&em->refs)) {
|
|
|
|
WARN_ON(em->in_tree);
|
|
|
|
kmem_cache_free(extent_map_cache, em);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct rb_node *tree_insert(struct rb_root *root, u64 offset,
|
|
|
|
struct rb_node *node)
|
|
|
|
{
|
2009-01-06 05:25:51 +03:00
|
|
|
struct rb_node **p = &root->rb_node;
|
|
|
|
struct rb_node *parent = NULL;
|
2008-01-25 00:13:08 +03:00
|
|
|
struct extent_map *entry;
|
2007-08-28 00:49:44 +04:00
|
|
|
|
2009-01-06 05:25:51 +03:00
|
|
|
while (*p) {
|
2007-08-28 00:49:44 +04:00
|
|
|
parent = *p;
|
2008-01-25 00:13:08 +03:00
|
|
|
entry = rb_entry(parent, struct extent_map, rb_node);
|
|
|
|
|
|
|
|
WARN_ON(!entry->in_tree);
|
2007-08-28 00:49:44 +04:00
|
|
|
|
|
|
|
if (offset < entry->start)
|
|
|
|
p = &(*p)->rb_left;
|
2008-01-25 00:13:08 +03:00
|
|
|
else if (offset >= extent_map_end(entry))
|
2007-08-28 00:49:44 +04:00
|
|
|
p = &(*p)->rb_right;
|
|
|
|
else
|
|
|
|
return parent;
|
|
|
|
}
|
|
|
|
|
2008-01-25 00:13:08 +03:00
|
|
|
entry = rb_entry(node, struct extent_map, rb_node);
|
2007-08-28 00:49:44 +04:00
|
|
|
entry->in_tree = 1;
|
|
|
|
rb_link_node(node, parent, p);
|
|
|
|
rb_insert_color(node, root);
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2008-09-29 23:18:18 +04:00
|
|
|
/*
|
|
|
|
* search through the tree for an extent_map with a given offset. If
|
|
|
|
* it can't be found, try to find some neighboring extents
|
|
|
|
*/
|
2007-08-28 00:49:44 +04:00
|
|
|
static struct rb_node *__tree_search(struct rb_root *root, u64 offset,
|
2008-01-23 00:47:59 +03:00
|
|
|
struct rb_node **prev_ret,
|
|
|
|
struct rb_node **next_ret)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
2009-01-06 05:25:51 +03:00
|
|
|
struct rb_node *n = root->rb_node;
|
2007-08-28 00:49:44 +04:00
|
|
|
struct rb_node *prev = NULL;
|
2008-01-23 00:47:59 +03:00
|
|
|
struct rb_node *orig_prev = NULL;
|
2008-01-25 00:13:08 +03:00
|
|
|
struct extent_map *entry;
|
|
|
|
struct extent_map *prev_entry = NULL;
|
2007-08-28 00:49:44 +04:00
|
|
|
|
2009-01-06 05:25:51 +03:00
|
|
|
while (n) {
|
2008-01-25 00:13:08 +03:00
|
|
|
entry = rb_entry(n, struct extent_map, rb_node);
|
2007-08-28 00:49:44 +04:00
|
|
|
prev = n;
|
|
|
|
prev_entry = entry;
|
|
|
|
|
2008-01-25 00:13:08 +03:00
|
|
|
WARN_ON(!entry->in_tree);
|
|
|
|
|
2007-08-28 00:49:44 +04:00
|
|
|
if (offset < entry->start)
|
|
|
|
n = n->rb_left;
|
2008-01-25 00:13:08 +03:00
|
|
|
else if (offset >= extent_map_end(entry))
|
2007-08-28 00:49:44 +04:00
|
|
|
n = n->rb_right;
|
|
|
|
else
|
|
|
|
return n;
|
|
|
|
}
|
2008-01-23 00:47:59 +03:00
|
|
|
|
|
|
|
if (prev_ret) {
|
|
|
|
orig_prev = prev;
|
2009-01-06 05:25:51 +03:00
|
|
|
while (prev && offset >= extent_map_end(prev_entry)) {
|
2008-01-23 00:47:59 +03:00
|
|
|
prev = rb_next(prev);
|
2008-01-25 00:13:08 +03:00
|
|
|
prev_entry = rb_entry(prev, struct extent_map, rb_node);
|
2008-01-23 00:47:59 +03:00
|
|
|
}
|
|
|
|
*prev_ret = prev;
|
|
|
|
prev = orig_prev;
|
|
|
|
}
|
|
|
|
|
|
|
|
if (next_ret) {
|
2008-01-25 00:13:08 +03:00
|
|
|
prev_entry = rb_entry(prev, struct extent_map, rb_node);
|
2009-01-06 05:25:51 +03:00
|
|
|
while (prev && offset < prev_entry->start) {
|
2008-01-23 00:47:59 +03:00
|
|
|
prev = rb_prev(prev);
|
2008-01-25 00:13:08 +03:00
|
|
|
prev_entry = rb_entry(prev, struct extent_map, rb_node);
|
2008-01-23 00:47:59 +03:00
|
|
|
}
|
|
|
|
*next_ret = prev;
|
2007-08-28 00:49:44 +04:00
|
|
|
}
|
|
|
|
return NULL;
|
|
|
|
}
|
|
|
|
|
2008-09-29 23:18:18 +04:00
|
|
|
/* check to see if two extent_map structs are adjacent and safe to merge */
|
2008-01-25 00:13:08 +03:00
|
|
|
static int mergable_maps(struct extent_map *prev, struct extent_map *next)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
2008-07-18 20:01:11 +04:00
|
|
|
if (test_bit(EXTENT_FLAG_PINNED, &prev->flags))
|
|
|
|
return 0;
|
|
|
|
|
Btrfs: Add zlib compression support
This is a large change for adding compression on reading and writing,
both for inline and regular extents. It does some fairly large
surgery to the writeback paths.
Compression is off by default and enabled by mount -o compress. Even
when the -o compress mount option is not used, it is possible to read
compressed extents off the disk.
If compression for a given set of pages fails to make them smaller, the
file is flagged to avoid future compression attempts later.
* While finding delalloc extents, the pages are locked before being sent down
to the delalloc handler. This allows the delalloc handler to do complex things
such as cleaning the pages, marking them writeback and starting IO on their
behalf.
* Inline extents are inserted at delalloc time now. This allows us to compress
the data before inserting the inline extent, and it allows us to insert
an inline extent that spans multiple pages.
* All of the in-memory extent representations (extent_map.c, ordered-data.c etc)
are changed to record both an in-memory size and an on disk size, as well
as a flag for compression.
From a disk format point of view, the extent pointers in the file are changed
to record the on disk size of a given extent and some encoding flags.
Space in the disk format is allocated for compression encoding, as well
as encryption and a generic 'other' field. Neither the encryption or the
'other' field are currently used.
In order to limit the amount of data read for a single random read in the
file, the size of a compressed extent is limited to 128k. This is a
software only limit, the disk format supports u64 sized compressed extents.
In order to limit the ram consumed while processing extents, the uncompressed
size of a compressed extent is limited to 256k. This is a software only limit
and will be subject to tuning later.
Checksumming is still done on compressed extents, and it is done on the
uncompressed version of the data. This way additional encodings can be
layered on without having to figure out which encoding to checksum.
Compression happens at delalloc time, which is basically singled threaded because
it is usually done by a single pdflush thread. This makes it tricky to
spread the compression load across all the cpus on the box. We'll have to
look at parallel pdflush walks of dirty inodes at a later time.
Decompression is hooked into readpages and it does spread across CPUs nicely.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
2008-10-29 21:49:59 +03:00
|
|
|
/*
|
|
|
|
* don't merge compressed extents, we need to know their
|
|
|
|
* actual size
|
|
|
|
*/
|
|
|
|
if (test_bit(EXTENT_FLAG_COMPRESSED, &prev->flags))
|
|
|
|
return 0;
|
|
|
|
|
2008-01-25 00:13:08 +03:00
|
|
|
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;
|
|
|
|
}
|
2007-08-28 00:49:44 +04:00
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
2011-07-14 07:18:33 +04:00
|
|
|
static void try_merge_map(struct extent_map_tree *tree, struct extent_map *em)
|
2009-09-11 20:27:37 +04:00
|
|
|
{
|
|
|
|
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->len += merge->len;
|
|
|
|
em->block_len += merge->block_len;
|
|
|
|
em->block_start = merge->block_start;
|
|
|
|
merge->in_tree = 0;
|
|
|
|
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;
|
|
|
|
free_extent_map(merge);
|
|
|
|
}
|
2011-07-14 07:18:33 +04:00
|
|
|
}
|
|
|
|
|
|
|
|
int unpin_extent_cache(struct extent_map_tree *tree, u64 start, u64 len)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
struct extent_map *em;
|
|
|
|
|
|
|
|
write_lock(&tree->lock);
|
|
|
|
em = lookup_extent_mapping(tree, start, len);
|
|
|
|
|
|
|
|
WARN_ON(!em || em->start != start);
|
|
|
|
|
|
|
|
if (!em)
|
|
|
|
goto out;
|
|
|
|
|
|
|
|
clear_bit(EXTENT_FLAG_PINNED, &em->flags);
|
|
|
|
|
|
|
|
try_merge_map(tree, em);
|
2009-09-11 20:27:37 +04:00
|
|
|
|
|
|
|
free_extent_map(em);
|
|
|
|
out:
|
|
|
|
write_unlock(&tree->lock);
|
|
|
|
return ret;
|
|
|
|
|
|
|
|
}
|
|
|
|
|
2008-06-12 05:52:17 +04:00
|
|
|
/**
|
|
|
|
* 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
|
2011-03-31 05:57:33 +04:00
|
|
|
* reference dropped if the merge attempt was successful.
|
2007-08-28 00:49:44 +04:00
|
|
|
*/
|
|
|
|
int add_extent_mapping(struct extent_map_tree *tree,
|
|
|
|
struct extent_map *em)
|
|
|
|
{
|
|
|
|
int ret = 0;
|
|
|
|
struct rb_node *rb;
|
2008-08-20 16:51:50 +04:00
|
|
|
struct extent_map *exist;
|
2007-08-28 00:49:44 +04:00
|
|
|
|
2008-08-20 16:51:50 +04:00
|
|
|
exist = lookup_extent_mapping(tree, em->start, em->len);
|
|
|
|
if (exist) {
|
|
|
|
free_extent_map(exist);
|
|
|
|
ret = -EEXIST;
|
|
|
|
goto out;
|
|
|
|
}
|
2008-01-25 00:13:08 +03:00
|
|
|
rb = tree_insert(&tree->map, em->start, &em->rb_node);
|
2007-08-28 00:49:44 +04:00
|
|
|
if (rb) {
|
|
|
|
ret = -EEXIST;
|
|
|
|
goto out;
|
|
|
|
}
|
|
|
|
atomic_inc(&em->refs);
|
2011-07-14 07:18:33 +04:00
|
|
|
|
|
|
|
try_merge_map(tree, em);
|
2007-08-28 00:49:44 +04:00
|
|
|
out:
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
|
2008-09-29 23:18:18 +04:00
|
|
|
/* simple helper to do math around the end of an extent, handling wrap */
|
2008-01-25 00:13:08 +03:00
|
|
|
static u64 range_end(u64 start, u64 len)
|
|
|
|
{
|
|
|
|
if (start + len < start)
|
|
|
|
return (u64)-1;
|
|
|
|
return start + len;
|
|
|
|
}
|
|
|
|
|
2011-07-14 07:18:15 +04:00
|
|
|
struct extent_map *__lookup_extent_mapping(struct extent_map_tree *tree,
|
|
|
|
u64 start, u64 len, int strict)
|
2007-08-28 00:49:44 +04:00
|
|
|
{
|
|
|
|
struct extent_map *em;
|
|
|
|
struct rb_node *rb_node;
|
2008-06-10 18:21:04 +04:00
|
|
|
struct rb_node *prev = NULL;
|
|
|
|
struct rb_node *next = NULL;
|
|
|
|
u64 end = range_end(start, len);
|
|
|
|
|
2008-01-23 00:47:59 +03:00
|
|
|
rb_node = __tree_search(&tree->map, start, &prev, &next);
|
2007-08-28 00:49:44 +04:00
|
|
|
if (!rb_node) {
|
2011-07-14 07:18:15 +04:00
|
|
|
if (prev)
|
|
|
|
rb_node = prev;
|
|
|
|
else if (next)
|
|
|
|
rb_node = next;
|
|
|
|
else
|
|
|
|
return NULL;
|
2007-08-28 00:49:44 +04:00
|
|
|
}
|
2011-07-14 07:18:15 +04:00
|
|
|
|
2007-08-28 00:49:44 +04:00
|
|
|
em = rb_entry(rb_node, struct extent_map, rb_node);
|
2008-01-25 00:13:08 +03:00
|
|
|
|
2011-07-14 07:18:15 +04:00
|
|
|
if (strict && !(end > em->start && start < extent_map_end(em)))
|
|
|
|
return NULL;
|
2008-01-25 00:13:08 +03:00
|
|
|
|
2007-08-28 00:49:44 +04:00
|
|
|
atomic_inc(&em->refs);
|
|
|
|
return em;
|
|
|
|
}
|
|
|
|
|
2011-07-14 07:18:15 +04:00
|
|
|
/**
|
|
|
|
* 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);
|
|
|
|
}
|
|
|
|
|
2009-09-19 00:07:03 +04:00
|
|
|
/**
|
|
|
|
* 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)
|
|
|
|
{
|
2011-07-14 07:18:15 +04:00
|
|
|
return __lookup_extent_mapping(tree, start, len, 0);
|
2009-09-19 00:07:03 +04:00
|
|
|
}
|
|
|
|
|
2008-06-12 05:52:17 +04:00
|
|
|
/**
|
|
|
|
* 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
|
2007-08-28 00:49:44 +04:00
|
|
|
*/
|
|
|
|
int remove_extent_mapping(struct extent_map_tree *tree, struct extent_map *em)
|
|
|
|
{
|
2008-01-25 00:13:08 +03:00
|
|
|
int ret = 0;
|
2007-08-28 00:49:44 +04:00
|
|
|
|
2008-07-18 20:01:11 +04:00
|
|
|
WARN_ON(test_bit(EXTENT_FLAG_PINNED, &em->flags));
|
2008-01-25 00:13:08 +03:00
|
|
|
rb_erase(&em->rb_node, &tree->map);
|
|
|
|
em->in_tree = 0;
|
2007-08-28 00:49:44 +04:00
|
|
|
return ret;
|
|
|
|
}
|