Btrfs: reduce CPU usage in the extent_state tree
Btrfs is currently mirroring some of the page state bits into
its extent state tree. The goal behind this was to use it in supporting
blocksizes other than the page size.
But, we don't currently support that, and we're using quite a lot of CPU
on the rb tree and its spin lock. This commit starts a series of
cleanups to reduce the amount of work done in the extent state tree as
part of each IO.
This commit:
* Adds the ability to lock an extent in the state tree and also set
other bits. The idea is to do locking and delalloc in one call
* Removes the EXTENT_WRITEBACK and EXTENT_DIRTY bits. Btrfs is using
a combination of the page bits and the ordered write code for this
instead.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c
index 8e168a4..7c70613 100644
--- a/fs/btrfs/extent_io.c
+++ b/fs/btrfs/extent_io.c
@@ -654,25 +654,24 @@
}
/*
- * set some bits on a range in the tree. This may require allocations
- * or sleeping, so the gfp mask is used to indicate what is allowed.
+ * set some bits on a range in the tree. This may require allocations or
+ * sleeping, so the gfp mask is used to indicate what is allowed.
*
- * If 'exclusive' == 1, this will fail with -EEXIST if some part of the
- * range already has the desired bits set. The start of the existing
- * range is returned in failed_start in this case.
+ * If any of the exclusive bits are set, this will fail with -EEXIST if some
+ * part of the range already has the desired bits set. The start of the
+ * existing range is returned in failed_start in this case.
*
- * [start, end] is inclusive
- * This takes the tree lock.
+ * [start, end] is inclusive This takes the tree lock.
*/
+
static int set_extent_bit(struct extent_io_tree *tree, u64 start, u64 end,
- int bits, int exclusive, u64 *failed_start,
+ int bits, int exclusive_bits, u64 *failed_start,
gfp_t mask)
{
struct extent_state *state;
struct extent_state *prealloc = NULL;
struct rb_node *node;
int err = 0;
- int set;
u64 last_start;
u64 last_end;
again:
@@ -707,8 +706,7 @@
*/
if (state->start == start && state->end <= end) {
struct rb_node *next_node;
- set = state->state & bits;
- if (set && exclusive) {
+ if (state->state & exclusive_bits) {
*failed_start = state->start;
err = -EEXIST;
goto out;
@@ -748,8 +746,7 @@
* desired bit on it.
*/
if (state->start < start) {
- set = state->state & bits;
- if (exclusive && set) {
+ if (state->state & exclusive_bits) {
*failed_start = start;
err = -EEXIST;
goto out;
@@ -799,8 +796,7 @@
* on the first half
*/
if (state->start <= end && state->end > end) {
- set = state->state & bits;
- if (exclusive && set) {
+ if (state->state & exclusive_bits) {
*failed_start = start;
err = -EEXIST;
goto out;
@@ -906,19 +902,6 @@
return clear_extent_bit(tree, start, end, EXTENT_UPTODATE, 0, 0, mask);
}
-static int set_extent_writeback(struct extent_io_tree *tree, u64 start, u64 end,
- gfp_t mask)
-{
- return set_extent_bit(tree, start, end, EXTENT_WRITEBACK,
- 0, NULL, mask);
-}
-
-static int clear_extent_writeback(struct extent_io_tree *tree, u64 start,
- u64 end, gfp_t mask)
-{
- return clear_extent_bit(tree, start, end, EXTENT_WRITEBACK, 1, 0, mask);
-}
-
int wait_on_extent_writeback(struct extent_io_tree *tree, u64 start, u64 end)
{
return wait_extent_bit(tree, start, end, EXTENT_WRITEBACK);
@@ -928,13 +911,14 @@
* either insert or lock state struct between start and end use mask to tell
* us if waiting is desired.
*/
-int lock_extent(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask)
+int lock_extent_bits(struct extent_io_tree *tree, u64 start, u64 end,
+ int bits, gfp_t mask)
{
int err;
u64 failed_start;
while (1) {
- err = set_extent_bit(tree, start, end, EXTENT_LOCKED, 1,
- &failed_start, mask);
+ err = set_extent_bit(tree, start, end, EXTENT_LOCKED | bits,
+ EXTENT_LOCKED, &failed_start, mask);
if (err == -EEXIST && (mask & __GFP_WAIT)) {
wait_extent_bit(tree, failed_start, end, EXTENT_LOCKED);
start = failed_start;
@@ -946,6 +930,11 @@
return err;
}
+int lock_extent(struct extent_io_tree *tree, u64 start, u64 end, gfp_t mask)
+{
+ return lock_extent_bits(tree, start, end, 0, mask);
+}
+
int try_lock_extent(struct extent_io_tree *tree, u64 start, u64 end,
gfp_t mask)
{
@@ -985,7 +974,6 @@
page_cache_release(page);
index++;
}
- set_extent_dirty(tree, start, end, GFP_NOFS);
return 0;
}
@@ -1005,7 +993,6 @@
page_cache_release(page);
index++;
}
- set_extent_writeback(tree, start, end, GFP_NOFS);
return 0;
}
@@ -1563,10 +1550,7 @@
static int check_page_writeback(struct extent_io_tree *tree,
struct page *page)
{
- u64 start = (u64)page->index << PAGE_CACHE_SHIFT;
- u64 end = start + PAGE_CACHE_SIZE - 1;
- if (!test_range_bit(tree, start, end, EXTENT_WRITEBACK, 0))
- end_page_writeback(page);
+ end_page_writeback(page);
return 0;
}
@@ -1624,13 +1608,11 @@
}
if (!uptodate) {
- clear_extent_uptodate(tree, start, end, GFP_ATOMIC);
+ clear_extent_uptodate(tree, start, end, GFP_NOFS);
ClearPageUptodate(page);
SetPageError(page);
}
- clear_extent_writeback(tree, start, end, GFP_ATOMIC);
-
if (whole_page)
end_page_writeback(page);
else
@@ -2208,8 +2190,9 @@
printk(KERN_ERR "btrfs delalloc bits after lock_extent\n");
if (last_byte <= start) {
- clear_extent_dirty(tree, start, page_end, GFP_NOFS);
- unlock_extent(tree, start, page_end, GFP_NOFS);
+ clear_extent_bit(tree, start, page_end,
+ EXTENT_LOCKED | EXTENT_DIRTY,
+ 1, 0, GFP_NOFS);
if (tree->ops && tree->ops->writepage_end_io_hook)
tree->ops->writepage_end_io_hook(page, start,
page_end, NULL, 1);
@@ -2217,12 +2200,10 @@
goto done;
}
- set_extent_uptodate(tree, start, page_end, GFP_NOFS);
blocksize = inode->i_sb->s_blocksize;
while (cur <= end) {
if (cur >= last_byte) {
- clear_extent_dirty(tree, cur, page_end, GFP_NOFS);
unlock_extent(tree, unlock_start, page_end, GFP_NOFS);
if (tree->ops && tree->ops->writepage_end_io_hook)
tree->ops->writepage_end_io_hook(page, cur,
@@ -2255,9 +2236,6 @@
*/
if (compressed || block_start == EXTENT_MAP_HOLE ||
block_start == EXTENT_MAP_INLINE) {
- clear_extent_dirty(tree, cur,
- cur + iosize - 1, GFP_NOFS);
-
unlock_extent(tree, unlock_start, cur + iosize - 1,
GFP_NOFS);
@@ -2291,7 +2269,6 @@
continue;
}
- clear_extent_dirty(tree, cur, cur + iosize - 1, GFP_NOFS);
if (tree->ops && tree->ops->writepage_io_hook) {
ret = tree->ops->writepage_io_hook(page, cur,
cur + iosize - 1);
@@ -2619,7 +2596,7 @@
return 0;
lock_extent(tree, start, end, GFP_NOFS);
- wait_on_extent_writeback(tree, start, end);
+ wait_on_page_writeback(page);
clear_extent_bit(tree, start, end,
EXTENT_LOCKED | EXTENT_DIRTY | EXTENT_DELALLOC,
1, 1, GFP_NOFS);