Commit 040ee612 authored by Robbie Ko's avatar Robbie Ko Committed by David Sterba
Browse files

Btrfs: send, improve clone range 



Improve clone_range in two scenarios.

1. Remove the limit of inode size when find clone inodes We can do
   partial clone, so there is no need to limit the size of the candidate
   inode.  When clone a range, we clone the legal range only by bytenr,
   offset, len, inode size.

2. In the scenarios of rewrite or clone_range, data_offset rarely
   matches exactly, so the chance of a clone is missed.

e.g.
    1. Write a 1M file
        dd if=/dev/zero of=1M bs=1M count=1

    2. Clone 1M file
       cp --reflink 1M clone

    3. Rewrite 4k on the clone file
       dd if=/dev/zero of=clone bs=4k count=1 conv=notrunc

    The disk layout is as follows:
    item 16 key (257 EXTENT_DATA 0) itemoff 15353 itemsize 53
	extent data disk byte 1103101952 nr 1048576
	extent data offset 0 nr 1048576 ram 1048576
	extent compression(none)
    ...
    item 22 key (258 EXTENT_DATA 0) itemoff 14959 itemsize 53
	extent data disk byte 1104150528 nr 4096
	extent data offset 0 nr 4096 ram 4096
	extent compression(none)
    item 23 key (258 EXTENT_DATA 4096) itemoff 14906 itemsize 53
	extent data disk byte 1103101952 nr 1048576
	extent data offset 4096 nr 1044480 ram 1048576
	extent compression(none)

When send, inode 258 file offset 4096~1048576 (item 23) has a chance to
clone_range, but because data_offset does not match inode 257 (item 16),
it causes missed clone and can only transfer actual data.

Improve the problem by judging whether the current data_offset has
overlap with the file extent item, and if so, adjusting offset and
extent_len so that we can clone correctly.

Reviewed-by: default avatarFilipe Manana <fdmanana@suse.com>
Signed-off-by: default avatarRobbie Ko <robbieko@synology.com>
Signed-off-by: default avatarDavid Sterba <dsterba@suse.com>
parent 8b4d1efc

fs/btrfs/send.c

+33 −19
Original line number Diff line number Diff line
@@ -1160,7 +1160,6 @@ out: 
struct backref_ctx {

	struct send_ctx *sctx;



	struct btrfs_path *path;

	/* number of total found references */

	u64 found;
@@ -1213,8 +1212,6 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) 
{

	struct backref_ctx *bctx = ctx_;

	struct clone_root *found;

	int ret;

	u64 i_size;



	/* First check if the root is in the list of accepted clone sources */

	found = bsearch((void *)(uintptr_t)root, bctx->sctx->clone_roots,
@@ -1230,19 +1227,6 @@ static int __iterate_backrefs(u64 ino, u64 offset, u64 root, void *ctx_) 
		bctx->found_itself = 1;

	}



	/*

	 * There are inodes that have extents that lie behind its i_size. Don't

	 * accept clones from these extents.

	 */

	ret = __get_inode_info(found->root, bctx->path, ino, &i_size, NULL, NULL,

			       NULL, NULL, NULL);

	btrfs_release_path(bctx->path);

	if (ret < 0)

		return ret;



	if (offset + bctx->data_offset + bctx->extent_len > i_size)

		return 0;



	/*

	 * Make sure we don't consider clones from send_root that are

	 * behind the current inode/offset.
@@ -1319,8 +1303,6 @@ static int find_extent_clone(struct send_ctx *sctx, 
		goto out;

	}



	backref_ctx->path = tmp_path;



	if (data_offset >= ino_size) {

		/*

		 * There may be extents that lie behind the file's size.
@@ -5082,6 +5064,7 @@ static int clone_range(struct send_ctx *sctx, 
	struct btrfs_path *path;

	struct btrfs_key key;

	int ret;

	u64 clone_src_i_size;



	/*

	 * Prevent cloning from a zero offset with a length matching the sector
@@ -5106,6 +5089,16 @@ static int clone_range(struct send_ctx *sctx, 
	if (!path)

		return -ENOMEM;



	/*

	 * There are inodes that have extents that lie behind its i_size. Don't

	 * accept clones from these extents.

	 */

	ret = __get_inode_info(clone_root->root, path, clone_root->ino,

			       &clone_src_i_size, NULL, NULL, NULL, NULL, NULL);

	btrfs_release_path(path);

	if (ret < 0)

		goto out;



	/*

	 * We can't send a clone operation for the entire range if we find

	 * extent items in the respective range in the source file that
@@ -5148,6 +5141,7 @@ static int clone_range(struct send_ctx *sctx, 
		u8 type;

		u64 ext_len;

		u64 clone_len;

		u64 clone_data_offset;



		if (slot >= btrfs_header_nritems(leaf)) {

			ret = btrfs_next_leaf(clone_root->root, path);
@@ -5201,10 +5195,30 @@ static int clone_range(struct send_ctx *sctx, 
		if (key.offset >= clone_root->offset + len)

			break;



		if (key.offset >= clone_src_i_size)

			break;



		if (key.offset + ext_len > clone_src_i_size)

			ext_len = clone_src_i_size - key.offset;



		clone_data_offset = btrfs_file_extent_offset(leaf, ei);

		if (btrfs_file_extent_disk_bytenr(leaf, ei) == disk_byte) {

			clone_root->offset = key.offset;

			if (clone_data_offset < data_offset &&

				clone_data_offset + ext_len > data_offset) {

				u64 extent_offset;



				extent_offset = data_offset - clone_data_offset;

				ext_len -= extent_offset;

				clone_data_offset += extent_offset;

				clone_root->offset += extent_offset;

			}

		}



		clone_len = min_t(u64, ext_len, len);



		if (btrfs_file_extent_disk_bytenr(leaf, ei) == disk_byte &&

		    btrfs_file_extent_offset(leaf, ei) == data_offset)

		    clone_data_offset == data_offset)

			ret = send_clone(sctx, offset, clone_len, clone_root);

		else

			ret = send_extent_data(sctx, offset, clone_len);

CRA Git | Maintained and supported by SUSTech CRA and CCSE