Btrfs: Split the extent_map code into two parts
There is now extent_map for mapping offsets in the file to disk and
extent_io for state tracking, IO submission and extent_bufers.
The new extent_map code shifts from [start,end] pairs to [start,len], and
pushes the locking out into the caller. This allows a few performance
optimizations and is easier to use.
A number of extent_map usage bugs were fixed, mostly with failing
to remove extent_map entries when changing the file.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/btrfs_inode.h b/fs/btrfs/btrfs_inode.h
index 9fc9988..9a0647d 100644
--- a/fs/btrfs/btrfs_inode.h
+++ b/fs/btrfs/btrfs_inode.h
@@ -20,6 +20,7 @@
#define __BTRFS_I__
#include "extent_map.h"
+#include "extent_io.h"
/* in memory btrfs inode */
struct btrfs_inode {
@@ -27,6 +28,7 @@
struct btrfs_block_group_cache *block_group;
struct btrfs_key location;
struct extent_map_tree extent_tree;
+ struct extent_io_tree io_tree;
struct inode vfs_inode;
u64 ordered_trans;
