Commit 06b0c886 authored by Zheng Liu's avatar Zheng Liu Committed by Theodore Ts'o

ext4: refine extent status tree

This commit refines the extent status tree code.

1) A prefix 'es_' is added to to the extent status tree structure
members.

2) Refactored es_remove_extent() so that __es_remove_extent() can be
used by es_insert_extent() to remove the old extent entry(-ies) before
inserting a new one.

3) Rename extent_status_end() to ext4_es_end()

4) ext4_es_can_be_merged() is define to check whether two extents can
be merged or not.

5) Update and clarified comments.
Signed-off-by: default avatarZheng Liu <wenqing.lz@taobao.com>
Signed-off-by: default avatar"Theodore Ts'o" <tytso@mit.edu>
Reviewed-by: default avatarJan Kara <jack@suse.cz>
parent 0f70b406
...@@ -3528,13 +3528,14 @@ static int ext4_find_delalloc_range(struct inode *inode, ...@@ -3528,13 +3528,14 @@ static int ext4_find_delalloc_range(struct inode *inode,
{ {
struct extent_status es; struct extent_status es;
es.start = lblk_start; es.es_lblk = lblk_start;
ext4_es_find_extent(inode, &es); (void)ext4_es_find_extent(inode, &es);
if (es.len == 0) if (es.es_len == 0)
return 0; /* there is no delay extent in this tree */ return 0; /* there is no delay extent in this tree */
else if (es.start <= lblk_start && lblk_start < es.start + es.len) else if (es.es_lblk <= lblk_start &&
lblk_start < es.es_lblk + es.es_len)
return 1; return 1;
else if (lblk_start <= es.start && es.start <= lblk_end) else if (lblk_start <= es.es_lblk && es.es_lblk <= lblk_end)
return 1; return 1;
else else
return 0; return 0;
...@@ -4569,7 +4570,7 @@ static int ext4_find_delayed_extent(struct inode *inode, ...@@ -4569,7 +4570,7 @@ static int ext4_find_delayed_extent(struct inode *inode,
struct extent_status es; struct extent_status es;
ext4_lblk_t next_del; ext4_lblk_t next_del;
es.start = newex->ec_block; es.es_lblk = newex->ec_block;
next_del = ext4_es_find_extent(inode, &es); next_del = ext4_es_find_extent(inode, &es);
if (newex->ec_start == 0) { if (newex->ec_start == 0) {
...@@ -4577,18 +4578,18 @@ static int ext4_find_delayed_extent(struct inode *inode, ...@@ -4577,18 +4578,18 @@ static int ext4_find_delayed_extent(struct inode *inode,
* No extent in extent-tree contains block @newex->ec_start, * No extent in extent-tree contains block @newex->ec_start,
* then the block may stay in 1)a hole or 2)delayed-extent. * then the block may stay in 1)a hole or 2)delayed-extent.
*/ */
if (es.len == 0) if (es.es_len == 0)
/* A hole found. */ /* A hole found. */
return 0; return 0;
if (es.start > newex->ec_block) { if (es.es_lblk > newex->ec_block) {
/* A hole found. */ /* A hole found. */
newex->ec_len = min(es.start - newex->ec_block, newex->ec_len = min(es.es_lblk - newex->ec_block,
newex->ec_len); newex->ec_len);
return 0; return 0;
} }
newex->ec_len = es.start + es.len - newex->ec_block; newex->ec_len = es.es_lblk + es.es_len - newex->ec_block;
} }
return next_del; return next_del;
......
...@@ -23,40 +23,53 @@ ...@@ -23,40 +23,53 @@
* (e.g. Reservation space warning), and provide extent-level locking. * (e.g. Reservation space warning), and provide extent-level locking.
* Delay extent tree is the first step to achieve this goal. It is * Delay extent tree is the first step to achieve this goal. It is
* original built by Yongqiang Yang. At that time it is called delay * original built by Yongqiang Yang. At that time it is called delay
* extent tree, whose goal is only track delay extent in memory to * extent tree, whose goal is only track delayed extents in memory to
* simplify the implementation of fiemap and bigalloc, and introduce * simplify the implementation of fiemap and bigalloc, and introduce
* lseek SEEK_DATA/SEEK_HOLE support. That is why it is still called * lseek SEEK_DATA/SEEK_HOLE support. That is why it is still called
* delay extent tree at the following comment. But for better * delay extent tree at the first commit. But for better understand
* understand what it does, it has been rename to extent status tree. * what it does, it has been rename to extent status tree.
* *
* Currently the first step has been done. All delay extents are * Step1:
* tracked in the tree. It maintains the delay extent when a delay * Currently the first step has been done. All delayed extents are
* allocation is issued, and the delay extent is written out or * tracked in the tree. It maintains the delayed extent when a delayed
* allocation is issued, and the delayed extent is written out or
* invalidated. Therefore the implementation of fiemap and bigalloc * invalidated. Therefore the implementation of fiemap and bigalloc
* are simplified, and SEEK_DATA/SEEK_HOLE are introduced. * are simplified, and SEEK_DATA/SEEK_HOLE are introduced.
* *
* The following comment describes the implemenmtation of extent * The following comment describes the implemenmtation of extent
* status tree and future works. * status tree and future works.
*
* Step2:
* In this step all extent status are tracked by extent status tree.
* Thus, we can first try to lookup a block mapping in this tree before
* finding it in extent tree. Hence, single extent cache can be removed
* because extent status tree can do a better job. Extents in status
* tree are loaded on-demand. Therefore, the extent status tree may not
* contain all of the extents in a file. Meanwhile we define a shrinker
* to reclaim memory from extent status tree because fragmented extent
* tree will make status tree cost too much memory. written/unwritten/-
* hole extents in the tree will be reclaimed by this shrinker when we
* are under high memory pressure. Delayed extents will not be
* reclimed because fiemap, bigalloc, and seek_data/hole need it.
*/ */
/* /*
* extents status tree implementation for ext4. * Extent status tree implementation for ext4.
* *
* *
* ========================================================================== * ==========================================================================
* Extents status encompass delayed extents and extent locks * Extent status tree tracks all extent status.
* *
* 1. Why delayed extent implementation ? * 1. Why we need to implement extent status tree?
* *
* Without delayed extent, ext4 identifies a delayed extent by looking * Without extent status tree, ext4 identifies a delayed extent by looking
* up page cache, this has several deficiencies - complicated, buggy, * up page cache, this has several deficiencies - complicated, buggy,
* and inefficient code. * and inefficient code.
* *
* FIEMAP, SEEK_HOLE/DATA, bigalloc, punch hole and writeout all need * FIEMAP, SEEK_HOLE/DATA, bigalloc, and writeout all need to know if a
* to know if a block or a range of blocks are belonged to a delayed * block or a range of blocks are belonged to a delayed extent.
* extent.
* *
* Let us have a look at how they do without delayed extents implementation. * Let us have a look at how they do without extent status tree.
* -- FIEMAP * -- FIEMAP
* FIEMAP looks up page cache to identify delayed allocations from holes. * FIEMAP looks up page cache to identify delayed allocations from holes.
* *
...@@ -68,47 +81,48 @@ ...@@ -68,47 +81,48 @@
* already under delayed allocation or not to determine whether * already under delayed allocation or not to determine whether
* quota reserving is needed for the cluster. * quota reserving is needed for the cluster.
* *
* -- punch hole
* punch hole looks up page cache to identify a delayed extent.
*
* -- writeout * -- writeout
* Writeout looks up whole page cache to see if a buffer is * Writeout looks up whole page cache to see if a buffer is
* mapped, If there are not very many delayed buffers, then it is * mapped, If there are not very many delayed buffers, then it is
* time comsuming. * time comsuming.
* *
* With delayed extents implementation, FIEMAP, SEEK_HOLE/DATA, * With extent status tree implementation, FIEMAP, SEEK_HOLE/DATA,
* bigalloc and writeout can figure out if a block or a range of * bigalloc and writeout can figure out if a block or a range of
* blocks is under delayed allocation(belonged to a delayed extent) or * blocks is under delayed allocation(belonged to a delayed extent) or
* not by searching the delayed extent tree. * not by searching the extent tree.
* *
* *
* ========================================================================== * ==========================================================================
* 2. ext4 delayed extents impelmentation * 2. Ext4 extent status tree impelmentation
*
* -- extent
* A extent is a range of blocks which are contiguous logically and
* physically. Unlike extent in extent tree, this extent in ext4 is
* a in-memory struct, there is no corresponding on-disk data. There
* is no limit on length of extent, so an extent can contain as many
* blocks as they are contiguous logically and physically.
* *
* -- delayed extent * -- extent status tree
* A delayed extent is a range of blocks which are contiguous * Every inode has an extent status tree and all allocation blocks
* logically and under delayed allocation. Unlike extent in * are added to the tree with different status. The extent in the
* ext4, delayed extent in ext4 is a in-memory struct, there is * tree are ordered by logical block no.
* no corresponding on-disk data. There is no limit on length of
* delayed extent, so a delayed extent can contain as many blocks
* as they are contiguous logically.
* *
* -- delayed extent tree * -- operations on a extent status tree
* Every inode has a delayed extent tree and all under delayed * There are three important operations on a delayed extent tree: find
* allocation blocks are added to the tree as delayed extents. * next extent, adding a extent(a range of blocks) and removing a extent.
* Delayed extents in the tree are ordered by logical block no.
* *
* -- operations on a delayed extent tree * -- race on a extent status tree
* There are three operations on a delayed extent tree: find next * Extent status tree is protected by inode->i_es_lock.
* delayed extent, adding a space(a range of blocks) and removing
* a space.
* *
* -- race on a delayed extent tree * -- memory consumption
* Delayed extent tree is protected inode->i_es_lock. * Fragmented extent tree will make extent status tree cost too much
* memory. Hence, we will reclaim written/unwritten/hole extents from
* the tree under a heavy memory pressure.
* *
* *
* ========================================================================== * ==========================================================================
* 3. performance analysis * 3. Performance analysis
*
* -- overhead * -- overhead
* 1. There is a cache extent for write access, so if writes are * 1. There is a cache extent for write access, so if writes are
* not very random, adding space operaions are in O(1) time. * not very random, adding space operaions are in O(1) time.
...@@ -120,15 +134,19 @@ ...@@ -120,15 +134,19 @@
* *
* ========================================================================== * ==========================================================================
* 4. TODO list * 4. TODO list
* -- Track all extent status
* *
* -- Improve get block process * -- Refactor delayed space reservation
* *
* -- Extent-level locking * -- Extent-level locking
*/ */
static struct kmem_cache *ext4_es_cachep; static struct kmem_cache *ext4_es_cachep;
static int __es_insert_extent(struct ext4_es_tree *tree,
struct extent_status *newes);
static int __es_remove_extent(struct ext4_es_tree *tree, ext4_lblk_t lblk,
ext4_lblk_t end);
int __init ext4_init_es(void) int __init ext4_init_es(void)
{ {
ext4_es_cachep = KMEM_CACHE(extent_status, SLAB_RECLAIM_ACCOUNT); ext4_es_cachep = KMEM_CACHE(extent_status, SLAB_RECLAIM_ACCOUNT);
...@@ -161,7 +179,7 @@ static void ext4_es_print_tree(struct inode *inode) ...@@ -161,7 +179,7 @@ static void ext4_es_print_tree(struct inode *inode)
while (node) { while (node) {
struct extent_status *es; struct extent_status *es;
es = rb_entry(node, struct extent_status, rb_node); es = rb_entry(node, struct extent_status, rb_node);
printk(KERN_DEBUG " [%u/%u)", es->start, es->len); printk(KERN_DEBUG " [%u/%u)", es->es_lblk, es->es_len);
node = rb_next(node); node = rb_next(node);
} }
printk(KERN_DEBUG "\n"); printk(KERN_DEBUG "\n");
...@@ -170,10 +188,10 @@ static void ext4_es_print_tree(struct inode *inode) ...@@ -170,10 +188,10 @@ static void ext4_es_print_tree(struct inode *inode)
#define ext4_es_print_tree(inode) #define ext4_es_print_tree(inode)
#endif #endif
static inline ext4_lblk_t extent_status_end(struct extent_status *es) static inline ext4_lblk_t ext4_es_end(struct extent_status *es)
{ {
BUG_ON(es->start + es->len < es->start); BUG_ON(es->es_lblk + es->es_len < es->es_lblk);
return es->start + es->len - 1; return es->es_lblk + es->es_len - 1;
} }
/* /*
...@@ -181,25 +199,25 @@ static inline ext4_lblk_t extent_status_end(struct extent_status *es) ...@@ -181,25 +199,25 @@ static inline ext4_lblk_t extent_status_end(struct extent_status *es)
* it can't be found, try to find next extent. * it can't be found, try to find next extent.
*/ */
static struct extent_status *__es_tree_search(struct rb_root *root, static struct extent_status *__es_tree_search(struct rb_root *root,
ext4_lblk_t offset) ext4_lblk_t lblk)
{ {
struct rb_node *node = root->rb_node; struct rb_node *node = root->rb_node;
struct extent_status *es = NULL; struct extent_status *es = NULL;
while (node) { while (node) {
es = rb_entry(node, struct extent_status, rb_node); es = rb_entry(node, struct extent_status, rb_node);
if (offset < es->start) if (lblk < es->es_lblk)
node = node->rb_left; node = node->rb_left;
else if (offset > extent_status_end(es)) else if (lblk > ext4_es_end(es))
node = node->rb_right; node = node->rb_right;
else else
return es; return es;
} }
if (es && offset < es->start) if (es && lblk < es->es_lblk)
return es; return es;
if (es && offset > extent_status_end(es)) { if (es && lblk > ext4_es_end(es)) {
node = rb_next(&es->rb_node); node = rb_next(&es->rb_node);
return node ? rb_entry(node, struct extent_status, rb_node) : return node ? rb_entry(node, struct extent_status, rb_node) :
NULL; NULL;
...@@ -209,8 +227,8 @@ static struct extent_status *__es_tree_search(struct rb_root *root, ...@@ -209,8 +227,8 @@ static struct extent_status *__es_tree_search(struct rb_root *root,
} }
/* /*
* ext4_es_find_extent: find the 1st delayed extent covering @es->start * ext4_es_find_extent: find the 1st delayed extent covering @es->lblk
* if it exists, otherwise, the next extent after @es->start. * if it exists, otherwise, the next extent after @es->lblk.
* *
* @inode: the inode which owns delayed extents * @inode: the inode which owns delayed extents
* @es: delayed extent that we found * @es: delayed extent that we found
...@@ -226,7 +244,7 @@ ext4_lblk_t ext4_es_find_extent(struct inode *inode, struct extent_status *es) ...@@ -226,7 +244,7 @@ ext4_lblk_t ext4_es_find_extent(struct inode *inode, struct extent_status *es)
struct rb_node *node; struct rb_node *node;
ext4_lblk_t ret = EXT_MAX_BLOCKS; ext4_lblk_t ret = EXT_MAX_BLOCKS;
trace_ext4_es_find_extent_enter(inode, es->start); trace_ext4_es_find_extent_enter(inode, es->es_lblk);
read_lock(&EXT4_I(inode)->i_es_lock); read_lock(&EXT4_I(inode)->i_es_lock);
tree = &EXT4_I(inode)->i_es_tree; tree = &EXT4_I(inode)->i_es_tree;
...@@ -234,25 +252,25 @@ ext4_lblk_t ext4_es_find_extent(struct inode *inode, struct extent_status *es) ...@@ -234,25 +252,25 @@ ext4_lblk_t ext4_es_find_extent(struct inode *inode, struct extent_status *es)
/* find delay extent in cache firstly */ /* find delay extent in cache firstly */
if (tree->cache_es) { if (tree->cache_es) {
es1 = tree->cache_es; es1 = tree->cache_es;
if (in_range(es->start, es1->start, es1->len)) { if (in_range(es->es_lblk, es1->es_lblk, es1->es_len)) {
es_debug("%u cached by [%u/%u)\n", es_debug("%u cached by [%u/%u)\n",
es->start, es1->start, es1->len); es->es_lblk, es1->es_lblk, es1->es_len);
goto out; goto out;
} }
} }
es->len = 0; es->es_len = 0;
es1 = __es_tree_search(&tree->root, es->start); es1 = __es_tree_search(&tree->root, es->es_lblk);
out: out:
if (es1) { if (es1) {
tree->cache_es = es1; tree->cache_es = es1;
es->start = es1->start; es->es_lblk = es1->es_lblk;
es->len = es1->len; es->es_len = es1->es_len;
node = rb_next(&es1->rb_node); node = rb_next(&es1->rb_node);
if (node) { if (node) {
es1 = rb_entry(node, struct extent_status, rb_node); es1 = rb_entry(node, struct extent_status, rb_node);
ret = es1->start; ret = es1->es_lblk;
} }
} }
...@@ -263,14 +281,14 @@ ext4_lblk_t ext4_es_find_extent(struct inode *inode, struct extent_status *es) ...@@ -263,14 +281,14 @@ ext4_lblk_t ext4_es_find_extent(struct inode *inode, struct extent_status *es)
} }
static struct extent_status * static struct extent_status *
ext4_es_alloc_extent(ext4_lblk_t start, ext4_lblk_t len) ext4_es_alloc_extent(ext4_lblk_t lblk, ext4_lblk_t len)
{ {
struct extent_status *es; struct extent_status *es;
es = kmem_cache_alloc(ext4_es_cachep, GFP_ATOMIC); es = kmem_cache_alloc(ext4_es_cachep, GFP_ATOMIC);
if (es == NULL) if (es == NULL)
return NULL; return NULL;
es->start = start; es->es_lblk = lblk;
es->len = len; es->es_len = len;
return es; return es;
} }
...@@ -279,6 +297,20 @@ static void ext4_es_free_extent(struct extent_status *es) ...@@ -279,6 +297,20 @@ static void ext4_es_free_extent(struct extent_status *es)
kmem_cache_free(ext4_es_cachep, es); kmem_cache_free(ext4_es_cachep, es);
} }
/*
* Check whether or not two extents can be merged
* Condition:
* - logical block number is contiguous
*/
static int ext4_es_can_be_merged(struct extent_status *es1,
struct extent_status *es2)
{
if (es1->es_lblk + es1->es_len != es2->es_lblk)
return 0;
return 1;
}
static struct extent_status * static struct extent_status *
ext4_es_try_to_merge_left(struct ext4_es_tree *tree, struct extent_status *es) ext4_es_try_to_merge_left(struct ext4_es_tree *tree, struct extent_status *es)
{ {
...@@ -290,8 +322,8 @@ ext4_es_try_to_merge_left(struct ext4_es_tree *tree, struct extent_status *es) ...@@ -290,8 +322,8 @@ ext4_es_try_to_merge_left(struct ext4_es_tree *tree, struct extent_status *es)
return es; return es;
es1 = rb_entry(node, struct extent_status, rb_node); es1 = rb_entry(node, struct extent_status, rb_node);
if (es->start == extent_status_end(es1) + 1) { if (ext4_es_can_be_merged(es1, es)) {
es1->len += es->len; es1->es_len += es->es_len;
rb_erase(&es->rb_node, &tree->root); rb_erase(&es->rb_node, &tree->root);
ext4_es_free_extent(es); ext4_es_free_extent(es);
es = es1; es = es1;
...@@ -311,8 +343,8 @@ ext4_es_try_to_merge_right(struct ext4_es_tree *tree, struct extent_status *es) ...@@ -311,8 +343,8 @@ ext4_es_try_to_merge_right(struct ext4_es_tree *tree, struct extent_status *es)
return es; return es;
es1 = rb_entry(node, struct extent_status, rb_node); es1 = rb_entry(node, struct extent_status, rb_node);
if (es1->start == extent_status_end(es) + 1) { if (ext4_es_can_be_merged(es, es1)) {
es->len += es1->len; es->es_len += es1->es_len;
rb_erase(node, &tree->root); rb_erase(node, &tree->root);
ext4_es_free_extent(es1); ext4_es_free_extent(es1);
} }
...@@ -320,60 +352,43 @@ ext4_es_try_to_merge_right(struct ext4_es_tree *tree, struct extent_status *es) ...@@ -320,60 +352,43 @@ ext4_es_try_to_merge_right(struct ext4_es_tree *tree, struct extent_status *es)
return es; return es;
} }
static int __es_insert_extent(struct ext4_es_tree *tree, ext4_lblk_t offset, static int __es_insert_extent(struct ext4_es_tree *tree,
ext4_lblk_t len) struct extent_status *newes)
{ {
struct rb_node **p = &tree->root.rb_node; struct rb_node **p = &tree->root.rb_node;
struct rb_node *parent = NULL; struct rb_node *parent = NULL;
struct extent_status *es; struct extent_status *es;
ext4_lblk_t end = offset + len - 1;
BUG_ON(end < offset);
es = tree->cache_es;
if (es && offset == (extent_status_end(es) + 1)) {
es_debug("cached by [%u/%u)\n", es->start, es->len);
es->len += len;
es = ext4_es_try_to_merge_right(tree, es);
goto out;
} else if (es && es->start == end + 1) {
es_debug("cached by [%u/%u)\n", es->start, es->len);
es->start = offset;
es->len += len;
es = ext4_es_try_to_merge_left(tree, es);
goto out;
} else if (es && es->start <= offset &&
end <= extent_status_end(es)) {
es_debug("cached by [%u/%u)\n", es->start, es->len);
goto out;
}
while (*p) { while (*p) {
parent = *p; parent = *p;
es = rb_entry(parent, struct extent_status, rb_node); es = rb_entry(parent, struct extent_status, rb_node);
if (offset < es->start) { if (newes->es_lblk < es->es_lblk) {
if (es->start == end + 1) { if (ext4_es_can_be_merged(newes, es)) {
es->start = offset; /*
es->len += len; * Here we can modify es_lblk directly
* because it isn't overlapped.
*/
es->es_lblk = newes->es_lblk;
es->es_len += newes->es_len;
es = ext4_es_try_to_merge_left(tree, es); es = ext4_es_try_to_merge_left(tree, es);
goto out; goto out;
} }
p = &(*p)->rb_left; p = &(*p)->rb_left;
} else if (offset > extent_status_end(es)) { } else if (newes->es_lblk > ext4_es_end(es)) {
if (offset == extent_status_end(es) + 1) { if (ext4_es_can_be_merged(es, newes)) {
es->len += len; es->es_len += newes->es_len;
es = ext4_es_try_to_merge_right(tree, es); es = ext4_es_try_to_merge_right(tree, es);
goto out; goto out;
} }
p = &(*p)->rb_right; p = &(*p)->rb_right;
} else { } else {
if (extent_status_end(es) <= end) BUG_ON(1);
es->len = offset - es->start + len; return -EINVAL;
goto out;
} }
} }
es = ext4_es_alloc_extent(offset, len); es = ext4_es_alloc_extent(newes->es_lblk, newes->es_len);
if (!es) if (!es)
return -ENOMEM; return -ENOMEM;
rb_link_node(&es->rb_node, parent, p); rb_link_node(&es->rb_node, parent, p);
...@@ -385,27 +400,38 @@ static int __es_insert_extent(struct ext4_es_tree *tree, ext4_lblk_t offset, ...@@ -385,27 +400,38 @@ static int __es_insert_extent(struct ext4_es_tree *tree, ext4_lblk_t offset,
} }
/* /*
* ext4_es_insert_extent() adds a space to a delayed extent tree. * ext4_es_insert_extent() adds a space to a extent status tree.
* Caller holds inode->i_es_lock.
* *
* ext4_es_insert_extent is called by ext4_da_write_begin and * ext4_es_insert_extent is called by ext4_da_write_begin and
* ext4_es_remove_extent. * ext4_es_remove_extent.
* *
* Return 0 on success, error code on failure. * Return 0 on success, error code on failure.
*/ */
int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t offset, int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
ext4_lblk_t len) ext4_lblk_t len)
{ {
struct ext4_es_tree *tree; struct ext4_es_tree *tree;
struct extent_status newes;
ext4_lblk_t end = lblk + len - 1;
int err = 0; int err = 0;
trace_ext4_es_insert_extent(inode, offset, len); trace_ext4_es_insert_extent(inode, lblk, len);
es_debug("add [%u/%u) to extent status tree of inode %lu\n", es_debug("add [%u/%u) to extent status tree of inode %lu\n",
offset, len, inode->i_ino); lblk, len, inode->i_ino);
BUG_ON(end < lblk);
newes.es_lblk = lblk;
newes.es_len = len;
write_lock(&EXT4_I(inode)->i_es_lock); write_lock(&EXT4_I(inode)->i_es_lock);
tree = &EXT4_I(inode)->i_es_tree; tree = &EXT4_I(inode)->i_es_tree;
err = __es_insert_extent(tree, offset, len); err = __es_remove_extent(tree, lblk, end);
if (err != 0)
goto error;
err = __es_insert_extent(tree, &newes);
error:
write_unlock(&EXT4_I(inode)->i_es_lock); write_unlock(&EXT4_I(inode)->i_es_lock);
ext4_es_print_tree(inode); ext4_es_print_tree(inode);
...@@ -413,57 +439,45 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t offset, ...@@ -413,57 +439,45 @@ int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t offset,
return err; return err;
} }
/* static int __es_remove_extent(struct ext4_es_tree *tree, ext4_lblk_t lblk,
* ext4_es_remove_extent() removes a space from a delayed extent tree. ext4_lblk_t end)
* Caller holds inode->i_es_lock.
*
* Return 0 on success, error code on failure.
*/
int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t offset,
ext4_lblk_t len)
{ {
struct rb_node *node; struct rb_node *node;
struct ext4_es_tree *tree;
struct extent_status *es; struct extent_status *es;
struct extent_status orig_es; struct extent_status orig_es;
ext4_lblk_t len1, len2, end; ext4_lblk_t len1, len2;
int err = 0; int err = 0;
trace_ext4_es_remove_extent(inode, offset, len); es = __es_tree_search(&tree->root, lblk);
es_debug("remove [%u/%u) from extent status tree of inode %lu\n",
offset, len, inode->i_ino);
end = offset + len - 1;
BUG_ON(end < offset);
write_lock(&EXT4_I(inode)->i_es_lock);
tree = &EXT4_I(inode)->i_es_tree;
es = __es_tree_search(&tree->root, offset);
if (!es) if (!es)
goto out; goto out;
if (es->start > end) if (es->es_lblk > end)
goto out; goto out;
/* Simply invalidate cache_es. */ /* Simply invalidate cache_es. */
tree->cache_es = NULL; tree->cache_es = NULL;
orig_es.start = es->start; orig_es.es_lblk = es->es_lblk;
orig_es.len = es->len; orig_es.es_len = es->es_len;
len1 = offset > es->start ? offset - es->start : 0; len1 = lblk > es->es_lblk ? lblk - es->es_lblk : 0;
len2 = extent_status_end(es) > end ? len2 = ext4_es_end(es) > end ? ext4_es_end(es) - end : 0;
extent_status_end(es) - end : 0;
if (len1 > 0) if (len1 > 0)
es->len = len1; es->es_len = len1;
if (len2 > 0) { if (len2 > 0) {
if (len1 > 0) { if (len1 > 0) {
err = __es_insert_extent(tree, end + 1, len2); struct extent_status newes;
newes.es_lblk = end + 1;
newes.es_len = len2;
err = __es_insert_extent(tree, &newes);
if (err) { if (err) {
es->start = orig_es.start; es->es_lblk = orig_es.es_lblk;
es->len = orig_es.len; es->es_len = orig_es.es_len;
goto out; goto out;
} }
} else { } else {
es->start = end + 1; es->es_lblk = end + 1;
es->len = len2; es->es_len = len2;
} }
goto out; goto out;
} }
...@@ -476,7 +490,7 @@ int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t offset, ...@@ -476,7 +490,7 @@ int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t offset,
es = NULL; es = NULL;
} }
while (es && extent_status_end(es) <= end) { while (es && ext4_es_end(es) <= end) {
node = rb_next(&es->rb_node); node = rb_next(&es->rb_node);
rb_erase(&es->rb_node, &tree->root); rb_erase(&es->rb_node, &tree->root);
ext4_es_free_extent(es); ext4_es_free_extent(es);
...@@ -487,13 +501,39 @@ int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t offset, ...@@ -487,13 +501,39 @@ int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t offset,
es = rb_entry(node, struct extent_status, rb_node); es = rb_entry(node, struct extent_status, rb_node);
} }
if (es && es->start < end + 1) { if (es && es->es_lblk < end + 1) {
len1 = extent_status_end(es) - end; len1 = ext4_es_end(es) - end;
es->start = end + 1; es->es_lblk = end + 1;
es->len = len1; es->es_len = len1;
} }
out: out:
return err;
}
/*
* ext4_es_remove_extent() removes a space from a extent status tree.
*
* Return 0 on success, error code on failure.
*/
int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
ext4_lblk_t len)
{
struct ext4_es_tree *tree;
ext4_lblk_t end;
int err = 0;
trace_ext4_es_remove_extent(inode, lblk, len);
es_debug("remove [%u/%u) from extent status tree of inode %lu\n",
lblk, len, inode->i_ino);
end = lblk + len - 1;
BUG_ON(end < lblk);
tree = &EXT4_I(inode)->i_es_tree;
write_lock(&EXT4_I(inode)->i_es_lock);
err = __es_remove_extent(tree, lblk, end);
write_unlock(&EXT4_I(inode)->i_es_lock); write_unlock(&EXT4_I(inode)->i_es_lock);
ext4_es_print_tree(inode); ext4_es_print_tree(inode);
return err; return err;
......
...@@ -22,8 +22,8 @@ ...@@ -22,8 +22,8 @@
struct extent_status { struct extent_status {
struct rb_node rb_node; struct rb_node rb_node;
ext4_lblk_t start; /* first block extent covers */ ext4_lblk_t es_lblk; /* first logical block extent covers */
ext4_lblk_t len; /* length of extent in block */ ext4_lblk_t es_len; /* length of extent in block */
}; };
struct ext4_es_tree { struct ext4_es_tree {
...@@ -35,9 +35,9 @@ extern int __init ext4_init_es(void); ...@@ -35,9 +35,9 @@ extern int __init ext4_init_es(void);
extern void ext4_exit_es(void); extern void ext4_exit_es(void);
extern void ext4_es_init_tree(struct ext4_es_tree *tree); extern void ext4_es_init_tree(struct ext4_es_tree *tree);
extern int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t start, extern int ext4_es_insert_extent(struct inode *inode, ext4_lblk_t lblk,
ext4_lblk_t len); ext4_lblk_t len);
extern int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t start, extern int ext4_es_remove_extent(struct inode *inode, ext4_lblk_t lblk,
ext4_lblk_t len); ext4_lblk_t len);
extern ext4_lblk_t ext4_es_find_extent(struct inode *inode, extern ext4_lblk_t ext4_es_find_extent(struct inode *inode,
struct extent_status *es); struct extent_status *es);
......
...@@ -464,10 +464,9 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize) ...@@ -464,10 +464,9 @@ static loff_t ext4_seek_data(struct file *file, loff_t offset, loff_t maxsize)
* If there is a delay extent at this offset, * If there is a delay extent at this offset,
* it will be as a data. * it will be as a data.
*/ */
es.start = last; es.es_lblk = last;
(void)ext4_es_find_extent(inode, &es); (void)ext4_es_find_extent(inode, &es);
if (last >= es.start && if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) {
last < es.start + es.len) {
if (last != start) if (last != start)
dataoff = last << blkbits; dataoff = last << blkbits;
break; break;
...@@ -549,11 +548,10 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize) ...@@ -549,11 +548,10 @@ static loff_t ext4_seek_hole(struct file *file, loff_t offset, loff_t maxsize)
* If there is a delay extent at this offset, * If there is a delay extent at this offset,
* we will skip this extent. * we will skip this extent.
*/ */
es.start = last; es.es_lblk = last;
(void)ext4_es_find_extent(inode, &es); (void)ext4_es_find_extent(inode, &es);
if (last >= es.start && if (es.es_len != 0 && in_range(last, es.es_lblk, es.es_len)) {
last < es.start + es.len) { last = es.es_lblk + es.es_len;
last = es.start + es.len;
holeoff = last << blkbits; holeoff = last << blkbits;
continue; continue;
} }
......
...@@ -2093,75 +2093,75 @@ TRACE_EVENT(ext4_ext_remove_space_done, ...@@ -2093,75 +2093,75 @@ TRACE_EVENT(ext4_ext_remove_space_done,
); );
TRACE_EVENT(ext4_es_insert_extent, TRACE_EVENT(ext4_es_insert_extent,
TP_PROTO(struct inode *inode, ext4_lblk_t start, ext4_lblk_t len), TP_PROTO(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len),
TP_ARGS(inode, start, len), TP_ARGS(inode, lblk, len),
TP_STRUCT__entry( TP_STRUCT__entry(
__field( dev_t, dev ) __field( dev_t, dev )
__field( ino_t, ino ) __field( ino_t, ino )
__field( loff_t, start ) __field( loff_t, lblk )
__field( loff_t, len ) __field( loff_t, len )
), ),
TP_fast_assign( TP_fast_assign(
__entry->dev = inode->i_sb->s_dev; __entry->dev = inode->i_sb->s_dev;
__entry->ino = inode->i_ino; __entry->ino = inode->i_ino;
__entry->start = start; __entry->lblk = lblk;
__entry->len = len; __entry->len = len;
), ),
TP_printk("dev %d,%d ino %lu es [%lld/%lld)", TP_printk("dev %d,%d ino %lu es [%lld/%lld)",
MAJOR(__entry->dev), MINOR(__entry->dev), MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long) __entry->ino, (unsigned long) __entry->ino,
__entry->start, __entry->len) __entry->lblk, __entry->len)
); );
TRACE_EVENT(ext4_es_remove_extent, TRACE_EVENT(ext4_es_remove_extent,
TP_PROTO(struct inode *inode, ext4_lblk_t start, ext4_lblk_t len), TP_PROTO(struct inode *inode, ext4_lblk_t lblk, ext4_lblk_t len),
TP_ARGS(inode, start, len), TP_ARGS(inode, lblk, len),
TP_STRUCT__entry( TP_STRUCT__entry(
__field( dev_t, dev ) __field( dev_t, dev )
__field( ino_t, ino ) __field( ino_t, ino )
__field( loff_t, start ) __field( loff_t, lblk )
__field( loff_t, len ) __field( loff_t, len )
), ),
TP_fast_assign( TP_fast_assign(
__entry->dev = inode->i_sb->s_dev; __entry->dev = inode->i_sb->s_dev;
__entry->ino = inode->i_ino; __entry->ino = inode->i_ino;
__entry->start = start; __entry->lblk = lblk;
__entry->len = len; __entry->len = len;
), ),
TP_printk("dev %d,%d ino %lu es [%lld/%lld)", TP_printk("dev %d,%d ino %lu es [%lld/%lld)",
MAJOR(__entry->dev), MINOR(__entry->dev), MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long) __entry->ino, (unsigned long) __entry->ino,
__entry->start, __entry->len) __entry->lblk, __entry->len)
); );
TRACE_EVENT(ext4_es_find_extent_enter, TRACE_EVENT(ext4_es_find_extent_enter,
TP_PROTO(struct inode *inode, ext4_lblk_t start), TP_PROTO(struct inode *inode, ext4_lblk_t lblk),
TP_ARGS(inode, start), TP_ARGS(inode, lblk),
TP_STRUCT__entry( TP_STRUCT__entry(
__field( dev_t, dev ) __field( dev_t, dev )
__field( ino_t, ino ) __field( ino_t, ino )
__field( ext4_lblk_t, start ) __field( ext4_lblk_t, lblk )
), ),
TP_fast_assign( TP_fast_assign(
__entry->dev = inode->i_sb->s_dev; __entry->dev = inode->i_sb->s_dev;
__entry->ino = inode->i_ino; __entry->ino = inode->i_ino;
__entry->start = start; __entry->lblk = lblk;
), ),
TP_printk("dev %d,%d ino %lu start %u", TP_printk("dev %d,%d ino %lu lblk %u",
MAJOR(__entry->dev), MINOR(__entry->dev), MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long) __entry->ino, __entry->start) (unsigned long) __entry->ino, __entry->lblk)
); );
TRACE_EVENT(ext4_es_find_extent_exit, TRACE_EVENT(ext4_es_find_extent_exit,
...@@ -2173,7 +2173,7 @@ TRACE_EVENT(ext4_es_find_extent_exit, ...@@ -2173,7 +2173,7 @@ TRACE_EVENT(ext4_es_find_extent_exit,
TP_STRUCT__entry( TP_STRUCT__entry(
__field( dev_t, dev ) __field( dev_t, dev )
__field( ino_t, ino ) __field( ino_t, ino )
__field( ext4_lblk_t, start ) __field( ext4_lblk_t, lblk )
__field( ext4_lblk_t, len ) __field( ext4_lblk_t, len )
__field( ext4_lblk_t, ret ) __field( ext4_lblk_t, ret )
), ),
...@@ -2181,15 +2181,15 @@ TRACE_EVENT(ext4_es_find_extent_exit, ...@@ -2181,15 +2181,15 @@ TRACE_EVENT(ext4_es_find_extent_exit,
TP_fast_assign( TP_fast_assign(
__entry->dev = inode->i_sb->s_dev; __entry->dev = inode->i_sb->s_dev;
__entry->ino = inode->i_ino; __entry->ino = inode->i_ino;
__entry->start = es->start; __entry->lblk = es->es_lblk;
__entry->len = es->len; __entry->len = es->es_len;
__entry->ret = ret; __entry->ret = ret;
), ),
TP_printk("dev %d,%d ino %lu es [%u/%u) ret %u", TP_printk("dev %d,%d ino %lu es [%u/%u) ret %u",
MAJOR(__entry->dev), MINOR(__entry->dev), MAJOR(__entry->dev), MINOR(__entry->dev),
(unsigned long) __entry->ino, (unsigned long) __entry->ino,
__entry->start, __entry->len, __entry->ret) __entry->lblk, __entry->len, __entry->ret)
); );
#endif /* _TRACE_EXT4_H */ #endif /* _TRACE_EXT4_H */
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment