summaryrefslogtreecommitdiffstats
path: root/fs/btrfs/ctree.h
diff options
context:
space:
mode:
authorChris Mason2008-12-08 22:58:54 +0100
committerChris Mason2008-12-08 22:58:54 +0100
commitd20f7043fa65659136c1a7c3c456eeeb5c6f431f (patch)
tree05d1031cadec6d440a97221e3a32adb504a51699 /fs/btrfs/ctree.h
parentBtrfs: Fix sparse endian warnings in struct-funcs.c (diff)
downloadkernel-qcow2-linux-d20f7043fa65659136c1a7c3c456eeeb5c6f431f.tar.gz
kernel-qcow2-linux-d20f7043fa65659136c1a7c3c456eeeb5c6f431f.tar.xz
kernel-qcow2-linux-d20f7043fa65659136c1a7c3c456eeeb5c6f431f.zip
Btrfs: move data checksumming into a dedicated tree
Btrfs stores checksums for each data block. Until now, they have been stored in the subvolume trees, indexed by the inode that is referencing the data block. This means that when we read the inode, we've probably read in at least some checksums as well. But, this has a few problems: * The checksums are indexed by logical offset in the file. When compression is on, this means we have to do the expensive checksumming on the uncompressed data. It would be faster if we could checksum the compressed data instead. * If we implement encryption, we'll be checksumming the plain text and storing that on disk. This is significantly less secure. * For either compression or encryption, we have to get the plain text back before we can verify the checksum as correct. This makes the raid layer balancing and extent moving much more expensive. * It makes the front end caching code more complex, as we have touch the subvolume and inodes as we cache extents. * There is potentitally one copy of the checksum in each subvolume referencing an extent. The solution used here is to store the extent checksums in a dedicated tree. This allows us to index the checksums by phyiscal extent start and length. It means: * The checksum is against the data stored on disk, after any compression or encryption is done. * The checksum is stored in a central location, and can be verified without following back references, or reading inodes. This makes compression significantly faster by reducing the amount of data that needs to be checksummed. It will also allow much faster raid management code in general. The checksums are indexed by a key with a fixed objectid (a magic value in ctree.h) and offset set to the starting byte of the extent. This allows us to copy the checksum items into the fsync log tree directly (or any other tree), without having to invent a second format for them. Signed-off-by: Chris Mason <chris.mason@oracle.com>
Diffstat (limited to 'fs/btrfs/ctree.h')
-rw-r--r--fs/btrfs/ctree.h30
1 files changed, 20 insertions, 10 deletions
diff --git a/fs/btrfs/ctree.h b/fs/btrfs/ctree.h
index 96f2ec7ad5bd..242b961ae6de 100644
--- a/fs/btrfs/ctree.h
+++ b/fs/btrfs/ctree.h
@@ -73,6 +73,9 @@ struct btrfs_ordered_sum;
/* directory objectid inside the root tree */
#define BTRFS_ROOT_TREE_DIR_OBJECTID 6ULL
+/* holds checksums of all the data extents */
+#define BTRFS_CSUM_TREE_OBJECTID 7ULL
+
/* orhpan objectid for tracking unlinked/truncated files */
#define BTRFS_ORPHAN_OBJECTID -5ULL
@@ -84,6 +87,13 @@ struct btrfs_ordered_sum;
#define BTRFS_TREE_RELOC_OBJECTID -8ULL
#define BTRFS_DATA_RELOC_TREE_OBJECTID -9ULL
+/*
+ * extent checksums all have this objectid
+ * this allows them to share the logging tree
+ * for fsyncs
+ */
+#define BTRFS_EXTENT_CSUM_OBJECTID -10ULL
+
/* dummy objectid represents multiple objectids */
#define BTRFS_MULTIPLE_OBJECTIDS -255ULL
@@ -634,6 +644,7 @@ struct btrfs_fs_info {
struct btrfs_root *chunk_root;
struct btrfs_root *dev_root;
struct btrfs_root *fs_root;
+ struct btrfs_root *csum_root;
/* the log root tree is a directory of all the other log roots */
struct btrfs_root *log_root_tree;
@@ -716,6 +727,7 @@ struct btrfs_fs_info {
struct btrfs_workers workers;
struct btrfs_workers delalloc_workers;
struct btrfs_workers endio_workers;
+ struct btrfs_workers endio_meta_workers;
struct btrfs_workers endio_write_workers;
struct btrfs_workers submit_workers;
/*
@@ -858,13 +870,12 @@ struct btrfs_root {
* extent data is for file data
*/
#define BTRFS_EXTENT_DATA_KEY 108
+
/*
- * csum items have the checksums for data in the extents
+ * extent csums are stored in a separate tree and hold csums for
+ * an entire extent on disk.
*/
-#define BTRFS_CSUM_ITEM_KEY 120
-
-
-/* reserve 21-31 for other file/dir stuff */
+#define BTRFS_EXTENT_CSUM_KEY 128
/*
* root items point to tree roots. There are typically in the root
@@ -1917,7 +1928,7 @@ int btrfs_lookup_inode(struct btrfs_trans_handle *trans, struct btrfs_root
/* file-item.c */
int btrfs_lookup_bio_sums(struct btrfs_root *root, struct inode *inode,
- struct bio *bio);
+ struct bio *bio, u32 *dst);
int btrfs_insert_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 objectid, u64 pos,
@@ -1929,17 +1940,16 @@ int btrfs_lookup_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_path *path, u64 objectid,
u64 bytenr, int mod);
int btrfs_csum_file_blocks(struct btrfs_trans_handle *trans,
- struct btrfs_root *root, struct inode *inode,
+ struct btrfs_root *root,
struct btrfs_ordered_sum *sums);
int btrfs_csum_one_bio(struct btrfs_root *root, struct inode *inode,
- struct bio *bio);
+ struct bio *bio, u64 file_start, int contig);
int btrfs_csum_file_bytes(struct btrfs_root *root, struct inode *inode,
u64 start, unsigned long len);
struct btrfs_csum_item *btrfs_lookup_csum(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path,
- u64 objectid, u64 offset,
- int cow);
+ u64 bytenr, int cow);
int btrfs_csum_truncate(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct btrfs_path *path,
u64 isize);