int btrfs_insert_inline_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 objectid,
u64 offset, char *buffer, size_t size)
{
struct btrfs_key key;
struct btrfs_path *path;
struct extent_buffer *leaf;
unsigned long ptr;
struct btrfs_file_extent_item *ei;
u32 datasize;
int err = 0;
int ret;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
key.objectid = objectid;
key.offset = offset;
btrfs_set_key_type(&key, BTRFS_EXTENT_DATA_KEY);
datasize = btrfs_file_extent_calc_inline_size(size);
ret = btrfs_insert_empty_item(trans, root, path, &key, datasize);
if (ret) {
err = ret;
goto fail;
}
leaf = path->nodes[0];
ei = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
btrfs_set_file_extent_generation(leaf, ei, trans->transid);
btrfs_set_file_extent_type(leaf, ei, BTRFS_FILE_EXTENT_INLINE);
btrfs_set_file_extent_ram_bytes(leaf, ei, size);
btrfs_set_file_extent_compression(leaf, ei, 0);
btrfs_set_file_extent_encryption(leaf, ei, 0);
btrfs_set_file_extent_other_encoding(leaf, ei, 0);
ptr = btrfs_file_extent_inline_start(ei) + offset - key.offset;
write_extent_buffer(leaf, buffer, ptr, size);
btrfs_mark_buffer_dirty(leaf);
fail:
btrfs_free_path(path);
return err;
}
int btrfs_insert_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
u64 objectid, u64 pos,
u64 disk_offset, u64 disk_num_bytes,
u64 num_bytes, u64 offset, u64 ram_bytes,
u8 compression, u8 encryption, u16 other_encoding)
{
int ret = 0;
struct btrfs_file_extent_item *item;
struct btrfs_key file_key;
struct btrfs_path *path;
struct extent_buffer *leaf;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
file_key.objectid = objectid;
file_key.offset = pos;
btrfs_set_key_type(&file_key, BTRFS_EXTENT_DATA_KEY);
path->leave_spinning = 1;
ret = btrfs_insert_empty_item(trans, root, path, &file_key,
sizeof(*item));
if (ret < 0)
goto out;
BUG_ON(ret); /* Can't happen */
leaf = path->nodes[0];
item = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
btrfs_set_file_extent_disk_bytenr(leaf, item, disk_offset);
btrfs_set_file_extent_disk_num_bytes(leaf, item, disk_num_bytes);
btrfs_set_file_extent_offset(leaf, item, offset);
btrfs_set_file_extent_num_bytes(leaf, item, num_bytes);
btrfs_set_file_extent_ram_bytes(leaf, item, ram_bytes);
btrfs_set_file_extent_generation(leaf, item, trans->transid);
btrfs_set_file_extent_type(leaf, item, BTRFS_FILE_EXTENT_REG);
btrfs_set_file_extent_compression(leaf, item, compression);
btrfs_set_file_extent_encryption(leaf, item, encryption);
btrfs_set_file_extent_other_encoding(leaf, item, other_encoding);
btrfs_mark_buffer_dirty(leaf);
out:
btrfs_free_path(path);
return ret;
}
/*
* xattrs work a lot like directories, this inserts an xattr item
* into the tree
*/
int btrfs_insert_xattr_item(struct btrfs_trans_handle *trans,
struct btrfs_root *root,
struct btrfs_path *path, u64 objectid,
const char *name, u16 name_len,
const void *data, u16 data_len)
{
int ret = 0;
struct btrfs_dir_item *dir_item;
unsigned long name_ptr, data_ptr;
struct btrfs_key key, location;
struct btrfs_disk_key disk_key;
struct extent_buffer *leaf;
u32 data_size;
if (name_len + data_len > BTRFS_MAX_XATTR_SIZE(root->fs_info))
return -ENOSPC;
key.objectid = objectid;
key.type = BTRFS_XATTR_ITEM_KEY;
key.offset = btrfs_name_hash(name, name_len);
data_size = sizeof(*dir_item) + name_len + data_len;
dir_item = insert_with_overflow(trans, root, path, &key, data_size,
name, name_len);
if (IS_ERR(dir_item))
return PTR_ERR(dir_item);
memset(&location, 0, sizeof(location));
leaf = path->nodes[0];
btrfs_cpu_key_to_disk(&disk_key, &location);
btrfs_set_dir_item_key(leaf, dir_item, &disk_key);
btrfs_set_dir_type(leaf, dir_item, BTRFS_FT_XATTR);
btrfs_set_dir_name_len(leaf, dir_item, name_len);
btrfs_set_dir_transid(leaf, dir_item, trans->transid);
btrfs_set_dir_data_len(leaf, dir_item, data_len);
name_ptr = (unsigned long)(dir_item + 1);
data_ptr = (unsigned long)((char *)name_ptr + name_len);
write_extent_buffer(leaf, name, name_ptr, name_len);
write_extent_buffer(leaf, data, data_ptr, data_len);
btrfs_mark_buffer_dirty(path->nodes[0]);
return ret;
}
static int update_free_space_extent_count(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path,
int new_extents)
{
struct btrfs_free_space_info *info;
u32 flags;
u32 extent_count;
int ret = 0;
if (new_extents == 0)
return 0;
info = search_free_space_info(trans, fs_info, block_group, path, 1);
if (IS_ERR(info)) {
ret = PTR_ERR(info);
goto out;
}
flags = btrfs_free_space_flags(path->nodes[0], info);
extent_count = btrfs_free_space_extent_count(path->nodes[0], info);
extent_count += new_extents;
btrfs_set_free_space_extent_count(path->nodes[0], info, extent_count);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
if (!(flags & BTRFS_FREE_SPACE_USING_BITMAPS) &&
extent_count > block_group->bitmap_high_thresh) {
ret = convert_free_space_to_bitmaps(trans, fs_info, block_group,
path);
} else if ((flags & BTRFS_FREE_SPACE_USING_BITMAPS) &&
extent_count < block_group->bitmap_low_thresh) {
ret = convert_free_space_to_extents(trans, fs_info, block_group,
path);
}
out:
return ret;
}
/*
* copy the data in 'item' into the btree
*/
int btrfs_update_root(struct btrfs_trans_handle *trans, struct btrfs_root
*root, struct btrfs_key *key, struct btrfs_root_item
*item)
{
struct btrfs_path *path;
struct extent_buffer *l;
int ret;
int slot;
unsigned long ptr;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
ret = btrfs_search_slot(trans, root, key, path, 0, 1);
if (ret < 0) {
btrfs_abort_transaction(trans, root, ret);
goto out;
}
if (ret != 0) {
btrfs_print_leaf(root, path->nodes[0]);
printk(KERN_CRIT "unable to update root key %llu %u %llu\n",
(unsigned long long)key->objectid, key->type,
(unsigned long long)key->offset);
BUG_ON(1);
}
l = path->nodes[0];
slot = path->slots[0];
ptr = btrfs_item_ptr_offset(l, slot);
write_extent_buffer(l, item, ptr, sizeof(*item));
btrfs_mark_buffer_dirty(path->nodes[0]);
out:
btrfs_free_path(path);
return ret;
}
/*
* add a btrfs_root_ref item. type is either BTRFS_ROOT_REF_KEY
* or BTRFS_ROOT_BACKREF_KEY.
*
* The dirid, sequence, name and name_len refer to the directory entry
* that is referencing the root.
*
* For a forward ref, the root_id is the id of the tree referencing
* the root and ref_id is the id of the subvol or snapshot.
*
* For a back ref the root_id is the id of the subvol or snapshot and
* ref_id is the id of the tree referencing it.
*/
int btrfs_add_root_ref(struct btrfs_trans_handle *trans,
struct btrfs_root *tree_root,
u64 root_id, u8 type, u64 ref_id,
u64 dirid, u64 sequence,
const char *name, int name_len)
{
struct btrfs_key key;
int ret;
struct btrfs_path *path;
struct btrfs_root_ref *ref;
struct extent_buffer *leaf;
unsigned long ptr;
path = btrfs_alloc_path();
key.objectid = root_id;
key.type = type;
key.offset = ref_id;
ret = btrfs_insert_empty_item(trans, tree_root, path, &key,
sizeof(*ref) + name_len);
BUG_ON(ret);
leaf = path->nodes[0];
ref = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_root_ref);
btrfs_set_root_ref_dirid(leaf, ref, dirid);
btrfs_set_root_ref_sequence(leaf, ref, sequence);
btrfs_set_root_ref_name_len(leaf, ref, name_len);
ptr = (unsigned long)(ref + 1);
write_extent_buffer(leaf, name, ptr, name_len);
btrfs_mark_buffer_dirty(leaf);
btrfs_free_path(path);
return ret;
}
int btrfs_uuid_tree_add(struct btrfs_trans_handle *trans, u8 *uuid, u8 type,
u64 subid_cpu)
{
struct btrfs_fs_info *fs_info = trans->fs_info;
struct btrfs_root *uuid_root = fs_info->uuid_root;
int ret;
struct btrfs_path *path = NULL;
struct btrfs_key key;
struct extent_buffer *eb;
int slot;
unsigned long offset;
__le64 subid_le;
ret = btrfs_uuid_tree_lookup(uuid_root, uuid, type, subid_cpu);
if (ret != -ENOENT)
return ret;
if (WARN_ON_ONCE(!uuid_root)) {
ret = -EINVAL;
goto out;
}
btrfs_uuid_to_key(uuid, type, &key);
path = btrfs_alloc_path();
if (!path) {
ret = -ENOMEM;
goto out;
}
ret = btrfs_insert_empty_item(trans, uuid_root, path, &key,
sizeof(subid_le));
if (ret >= 0) {
/* Add an item for the type for the first time */
eb = path->nodes[0];
slot = path->slots[0];
offset = btrfs_item_ptr_offset(eb, slot);
} else if (ret == -EEXIST) {
/*
* An item with that type already exists.
* Extend the item and store the new subid at the end.
*/
btrfs_extend_item(path, sizeof(subid_le));
eb = path->nodes[0];
slot = path->slots[0];
offset = btrfs_item_ptr_offset(eb, slot);
offset += btrfs_item_size_nr(eb, slot) - sizeof(subid_le);
} else {
btrfs_warn(fs_info,
"insert uuid item failed %d (0x%016llx, 0x%016llx) type %u!",
ret, (unsigned long long)key.objectid,
(unsigned long long)key.offset, type);
goto out;
}
ret = 0;
subid_le = cpu_to_le64(subid_cpu);
write_extent_buffer(eb, &subid_le, offset, sizeof(subid_le));
btrfs_mark_buffer_dirty(eb);
out:
btrfs_free_path(path);
return ret;
}
static noinline int create_subvol(struct btrfs_root *root,
struct dentry *dentry,
char *name, int namelen)
{
struct btrfs_trans_handle *trans;
struct btrfs_key key;
struct btrfs_root_item root_item;
struct btrfs_inode_item *inode_item;
struct extent_buffer *leaf;
struct btrfs_root *new_root;
struct inode *dir = dentry->d_parent->d_inode;
int ret;
int err;
u64 objectid;
u64 new_dirid = BTRFS_FIRST_FREE_OBJECTID;
u64 index = 0;
ret = btrfs_find_free_objectid(NULL, root->fs_info->tree_root,
0, &objectid);
if (ret)
return ret;
/*
* 1 - inode item
* 2 - refs
* 1 - root item
* 2 - dir items
*/
trans = btrfs_start_transaction(root, 6);
if (IS_ERR(trans))
return PTR_ERR(trans);
leaf = btrfs_alloc_free_block(trans, root, root->leafsize,
0, objectid, NULL, 0, 0, 0);
if (IS_ERR(leaf)) {
ret = PTR_ERR(leaf);
goto fail;
}
memset_extent_buffer(leaf, 0, 0, sizeof(struct btrfs_header));
btrfs_set_header_bytenr(leaf, leaf->start);
btrfs_set_header_generation(leaf, trans->transid);
btrfs_set_header_backref_rev(leaf, BTRFS_MIXED_BACKREF_REV);
btrfs_set_header_owner(leaf, objectid);
write_extent_buffer(leaf, root->fs_info->fsid,
(unsigned long)btrfs_header_fsid(leaf),
BTRFS_FSID_SIZE);
write_extent_buffer(leaf, root->fs_info->chunk_tree_uuid,
(unsigned long)btrfs_header_chunk_tree_uuid(leaf),
BTRFS_UUID_SIZE);
btrfs_mark_buffer_dirty(leaf);
inode_item = &root_item.inode;
memset(inode_item, 0, sizeof(*inode_item));
inode_item->generation = cpu_to_le64(1);
inode_item->size = cpu_to_le64(3);
inode_item->nlink = cpu_to_le32(1);
inode_item->nbytes = cpu_to_le64(root->leafsize);
inode_item->mode = cpu_to_le32(S_IFDIR | 0755);
root_item.flags = 0;
root_item.byte_limit = 0;
inode_item->flags = cpu_to_le64(BTRFS_INODE_ROOT_ITEM_INIT);
btrfs_set_root_bytenr(&root_item, leaf->start);
btrfs_set_root_generation(&root_item, trans->transid);
btrfs_set_root_level(&root_item, 0);
btrfs_set_root_refs(&root_item, 1);
btrfs_set_root_used(&root_item, leaf->len);
btrfs_set_root_last_snapshot(&root_item, 0);
memset(&root_item.drop_progress, 0, sizeof(root_item.drop_progress));
root_item.drop_level = 0;
btrfs_tree_unlock(leaf);
free_extent_buffer(leaf);
leaf = NULL;
btrfs_set_root_dirid(&root_item, new_dirid);
key.objectid = objectid;
key.offset = 0;
btrfs_set_key_type(&key, BTRFS_ROOT_ITEM_KEY);
ret = btrfs_insert_root(trans, root->fs_info->tree_root, &key,
&root_item);
if (ret)
goto fail;
key.offset = (u64)-1;
new_root = btrfs_read_fs_root_no_name(root->fs_info, &key);
BUG_ON(IS_ERR(new_root));
btrfs_record_root_in_trans(trans, new_root);
ret = btrfs_create_subvol_root(trans, new_root, new_dirid,
BTRFS_I(dir)->block_group);
/*
* insert the directory item
*/
ret = btrfs_set_inode_index(dir, &index);
BUG_ON(ret);
ret = btrfs_insert_dir_item(trans, root,
name, namelen, dir->i_ino, &key,
BTRFS_FT_DIR, index);
if (ret)
goto fail;
btrfs_i_size_write(dir, dir->i_size + namelen * 2);
ret = btrfs_update_inode(trans, root, dir);
BUG_ON(ret);
ret = btrfs_add_root_ref(trans, root->fs_info->tree_root,
objectid, root->root_key.objectid,
dir->i_ino, index, name, namelen);
BUG_ON(ret);
d_instantiate(dentry, btrfs_lookup_dentry(dir, dentry));
fail:
err = btrfs_commit_transaction(trans, root);
if (err && !ret)
ret = err;
return ret;
}
int convert_free_space_to_extents(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path)
{
struct btrfs_root *root = fs_info->free_space_root;
struct btrfs_free_space_info *info;
struct btrfs_key key, found_key;
struct extent_buffer *leaf;
unsigned long *bitmap;
u64 start, end;
/* Initialize to silence GCC. */
u64 extent_start = 0;
u64 offset;
u32 bitmap_size, flags, expected_extent_count;
int prev_bit = 0, bit, bitnr;
u32 extent_count = 0;
int done = 0, nr;
int ret;
bitmap_size = free_space_bitmap_size(block_group->key.offset,
block_group->sectorsize);
bitmap = alloc_bitmap(bitmap_size);
if (!bitmap) {
ret = -ENOMEM;
goto out;
}
start = block_group->key.objectid;
end = block_group->key.objectid + block_group->key.offset;
key.objectid = end - 1;
key.type = (u8)-1;
key.offset = (u64)-1;
while (!done) {
ret = btrfs_search_prev_slot(trans, root, &key, path, -1, 1);
if (ret)
goto out;
leaf = path->nodes[0];
nr = 0;
path->slots[0]++;
while (path->slots[0] > 0) {
btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
ASSERT(found_key.objectid == block_group->key.objectid);
ASSERT(found_key.offset == block_group->key.offset);
done = 1;
break;
} else if (found_key.type == BTRFS_FREE_SPACE_BITMAP_KEY) {
unsigned long ptr;
char *bitmap_cursor;
u32 bitmap_pos, data_size;
ASSERT(found_key.objectid >= start);
ASSERT(found_key.objectid < end);
ASSERT(found_key.objectid + found_key.offset <= end);
bitmap_pos = div_u64(found_key.objectid - start,
block_group->sectorsize *
BITS_PER_BYTE);
bitmap_cursor = ((char *)bitmap) + bitmap_pos;
data_size = free_space_bitmap_size(found_key.offset,
block_group->sectorsize);
ptr = btrfs_item_ptr_offset(leaf, path->slots[0] - 1);
read_extent_buffer(leaf, bitmap_cursor, ptr,
data_size);
nr++;
path->slots[0]--;
} else {
ASSERT(0);
}
}
ret = btrfs_del_items(trans, root, path, path->slots[0], nr);
if (ret)
goto out;
btrfs_release_path(path);
}
info = search_free_space_info(trans, fs_info, block_group, path, 1);
if (IS_ERR(info)) {
ret = PTR_ERR(info);
goto out;
}
leaf = path->nodes[0];
flags = btrfs_free_space_flags(leaf, info);
flags &= ~BTRFS_FREE_SPACE_USING_BITMAPS;
btrfs_set_free_space_flags(leaf, info, flags);
expected_extent_count = btrfs_free_space_extent_count(leaf, info);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
offset = start;
bitnr = 0;
while (offset < end) {
bit = !!test_bit(bitnr, bitmap);
if (prev_bit == 0 && bit == 1) {
extent_start = offset;
} else if (prev_bit == 1 && bit == 0) {
key.objectid = extent_start;
key.type = BTRFS_FREE_SPACE_EXTENT_KEY;
key.offset = offset - extent_start;
ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
if (ret)
goto out;
btrfs_release_path(path);
extent_count++;
}
prev_bit = bit;
offset += block_group->sectorsize;
bitnr++;
}
if (prev_bit == 1) {
key.objectid = extent_start;
key.type = BTRFS_FREE_SPACE_EXTENT_KEY;
key.offset = end - extent_start;
ret = btrfs_insert_empty_item(trans, root, path, &key, 0);
if (ret)
goto out;
btrfs_release_path(path);
extent_count++;
}
if (extent_count != expected_extent_count) {
btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
block_group->key.objectid, extent_count,
expected_extent_count);
ASSERT(0);
ret = -EIO;
goto out;
}
ret = 0;
out:
kvfree(bitmap);
if (ret)
btrfs_abort_transaction(trans, ret);
return ret;
}
/*
* called from commit_transaction. Writes changed device replace state to
* disk.
*/
int btrfs_run_dev_replace(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info)
{
int ret;
struct btrfs_root *dev_root = fs_info->dev_root;
struct btrfs_path *path;
struct btrfs_key key;
struct extent_buffer *eb;
struct btrfs_dev_replace_item *ptr;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
btrfs_dev_replace_lock(dev_replace);
if (!dev_replace->is_valid ||
!dev_replace->item_needs_writeback) {
btrfs_dev_replace_unlock(dev_replace);
return 0;
}
btrfs_dev_replace_unlock(dev_replace);
key.objectid = 0;
key.type = BTRFS_DEV_REPLACE_KEY;
key.offset = 0;
path = btrfs_alloc_path();
if (!path) {
ret = -ENOMEM;
goto out;
}
ret = btrfs_search_slot(trans, dev_root, &key, path, -1, 1);
if (ret < 0) {
pr_warn("btrfs: error %d while searching for dev_replace item!\n",
ret);
goto out;
}
if (ret == 0 &&
btrfs_item_size_nr(path->nodes[0], path->slots[0]) < sizeof(*ptr)) {
/*
* need to delete old one and insert a new one.
* Since no attempt is made to recover any old state, if the
* dev_replace state is 'running', the data on the target
* drive is lost.
* It would be possible to recover the state: just make sure
* that the beginning of the item is never changed and always
* contains all the essential information. Then read this
* minimal set of information and use it as a base for the
* new state.
*/
ret = btrfs_del_item(trans, dev_root, path);
if (ret != 0) {
pr_warn("btrfs: delete too small dev_replace item failed %d!\n",
ret);
goto out;
}
ret = 1;
}
if (ret == 1) {
/* need to insert a new item */
btrfs_release_path(path);
ret = btrfs_insert_empty_item(trans, dev_root, path,
&key, sizeof(*ptr));
if (ret < 0) {
pr_warn("btrfs: insert dev_replace item failed %d!\n",
ret);
goto out;
}
}
eb = path->nodes[0];
ptr = btrfs_item_ptr(eb, path->slots[0],
struct btrfs_dev_replace_item);
btrfs_dev_replace_lock(dev_replace);
if (dev_replace->srcdev)
btrfs_set_dev_replace_src_devid(eb, ptr,
dev_replace->srcdev->devid);
else
btrfs_set_dev_replace_src_devid(eb, ptr, (u64)-1);
btrfs_set_dev_replace_cont_reading_from_srcdev_mode(eb, ptr,
dev_replace->cont_reading_from_srcdev_mode);
btrfs_set_dev_replace_replace_state(eb, ptr,
dev_replace->replace_state);
btrfs_set_dev_replace_time_started(eb, ptr, dev_replace->time_started);
btrfs_set_dev_replace_time_stopped(eb, ptr, dev_replace->time_stopped);
btrfs_set_dev_replace_num_write_errors(eb, ptr,
atomic64_read(&dev_replace->num_write_errors));
btrfs_set_dev_replace_num_uncorrectable_read_errors(eb, ptr,
atomic64_read(&dev_replace->num_uncorrectable_read_errors));
dev_replace->cursor_left_last_write_of_item =
dev_replace->cursor_left;
btrfs_set_dev_replace_cursor_left(eb, ptr,
dev_replace->cursor_left_last_write_of_item);
btrfs_set_dev_replace_cursor_right(eb, ptr,
dev_replace->cursor_right);
dev_replace->item_needs_writeback = 0;
btrfs_dev_replace_unlock(dev_replace);
btrfs_mark_buffer_dirty(eb);
out:
btrfs_free_path(path);
return ret;
}
static int ondisk_add(struct btrfs_trans_handle *trans,
struct btrfs_dedup_info *dedup_info,
struct btrfs_dedup_hash *hash)
{
struct btrfs_path *path;
struct btrfs_root *dedup_root = dedup_info->dedup_root;
struct btrfs_key key;
struct btrfs_dedup_hash_item *hash_item;
u64 bytenr;
u32 num_bytes;
int hash_len = btrfs_dedup_sizes[dedup_info->hash_type];
int ret;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
mutex_lock(&dedup_info->lock);
ret = ondisk_search_bytenr(NULL, dedup_info, path, hash->bytenr, 0);
if (ret < 0)
goto out;
if (ret > 0) {
ret = 0;
goto out;
}
btrfs_release_path(path);
ret = ondisk_search_hash(dedup_info, hash->hash, &bytenr, &num_bytes);
if (ret < 0)
goto out;
/* Same hash found, don't re-add to save dedup tree space */
if (ret > 0) {
ret = 0;
goto out;
}
/* Insert hash->bytenr item */
memcpy(&key.objectid, hash->hash + hash_len - 8, 8);
key.type = BTRFS_DEDUP_HASH_ITEM_KEY;
key.offset = hash->bytenr;
ret = btrfs_insert_empty_item(trans, dedup_root, path, &key,
sizeof(*hash_item) + hash_len);
WARN_ON(ret == -EEXIST);
if (ret < 0)
goto out;
hash_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_dedup_hash_item);
btrfs_set_dedup_hash_len(path->nodes[0], hash_item, hash->num_bytes);
write_extent_buffer(path->nodes[0], hash->hash,
(unsigned long)(hash_item + 1), hash_len);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
/* Then bytenr->hash item */
key.objectid = hash->bytenr;
key.type = BTRFS_DEDUP_BYTENR_ITEM_KEY;
memcpy(&key.offset, hash->hash + hash_len - 8, 8);
ret = btrfs_insert_empty_item(trans, dedup_root, path, &key, hash_len);
WARN_ON(ret == -EEXIST);
if (ret < 0)
goto out;
write_extent_buffer(path->nodes[0], hash->hash,
btrfs_item_ptr_offset(path->nodes[0], path->slots[0]),
hash_len);
btrfs_mark_buffer_dirty(path->nodes[0]);
out:
mutex_unlock(&dedup_info->lock);
btrfs_free_path(path);
return ret;
}
static int record_file_extent(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 objectid,
struct btrfs_inode_item *inode,
u64 file_pos, u64 disk_bytenr,
u64 num_bytes)
{
int ret;
struct btrfs_fs_info *info = root->fs_info;
struct btrfs_root *extent_root = info->extent_root;
struct extent_buffer *leaf;
struct btrfs_file_extent_item *fi;
struct btrfs_key ins_key;
struct btrfs_path path;
struct btrfs_extent_item *ei;
btrfs_init_path(&path);
ins_key.objectid = objectid;
ins_key.offset = 0;
btrfs_set_key_type(&ins_key, BTRFS_EXTENT_DATA_KEY);
ret = btrfs_insert_empty_item(trans, root, &path, &ins_key,
sizeof(*fi));
if (ret)
goto fail;
leaf = path.nodes[0];
fi = btrfs_item_ptr(leaf, path.slots[0],
struct btrfs_file_extent_item);
btrfs_set_file_extent_generation(leaf, fi, trans->transid);
btrfs_set_file_extent_type(leaf, fi, BTRFS_FILE_EXTENT_REG);
btrfs_set_file_extent_disk_bytenr(leaf, fi, disk_bytenr);
btrfs_set_file_extent_disk_num_bytes(leaf, fi, num_bytes);
btrfs_set_file_extent_offset(leaf, fi, 0);
btrfs_set_file_extent_num_bytes(leaf, fi, num_bytes);
btrfs_set_file_extent_ram_bytes(leaf, fi, num_bytes);
btrfs_set_file_extent_compression(leaf, fi, 0);
btrfs_set_file_extent_encryption(leaf, fi, 0);
btrfs_set_file_extent_other_encoding(leaf, fi, 0);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(root, &path);
ins_key.objectid = disk_bytenr;
ins_key.offset = num_bytes;
ins_key.type = BTRFS_EXTENT_ITEM_KEY;
ret = btrfs_insert_empty_item(trans, extent_root, &path,
&ins_key, sizeof(*ei));
if (ret == 0) {
leaf = path.nodes[0];
ei = btrfs_item_ptr(leaf, path.slots[0],
struct btrfs_extent_item);
btrfs_set_extent_refs(leaf, ei, 0);
btrfs_set_extent_generation(leaf, ei, trans->transid);
btrfs_set_extent_flags(leaf, ei, BTRFS_EXTENT_FLAG_DATA);
btrfs_mark_buffer_dirty(leaf);
ret = btrfs_update_block_group(trans, root, disk_bytenr,
num_bytes, 1, 0);
if (ret)
goto fail;
} else if (ret != -EEXIST) {
/*
* Unlink an inode, which will remove its backref and corresponding dir_index/
* dir_item if any of them exists.
*
* If an inode's nlink is reduced to 0 and 'add_orphan' is true, it will be
* added to orphan inode and waiting to be deleted by next kernel mount.
*/
int btrfs_unlink(struct btrfs_trans_handle *trans, struct btrfs_root *root,
u64 ino, u64 parent_ino, u64 index, const char *name,
int namelen, int add_orphan)
{
struct btrfs_path *path;
struct btrfs_key key;
struct btrfs_inode_item *inode_item;
struct btrfs_inode_ref *inode_ref;
struct btrfs_dir_item *dir_item;
u64 inode_size;
u32 nlinks;
int del_inode_ref = 0;
int del_dir_item = 0;
int del_dir_index = 0;
int ret = 0;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
/* check the ref and backref exists */
inode_ref = btrfs_lookup_inode_ref(trans, root, path, name, namelen,
ino, parent_ino, index, 0);
if (IS_ERR(inode_ref)) {
ret = PTR_ERR(inode_ref);
goto out;
}
if (inode_ref)
del_inode_ref = 1;
btrfs_release_path(path);
dir_item = btrfs_lookup_dir_item(NULL, root, path, parent_ino,
name, namelen, 0);
if (IS_ERR(dir_item)) {
ret = PTR_ERR(dir_item);
goto out;
}
if (dir_item)
del_dir_item = 1;
btrfs_release_path(path);
dir_item = btrfs_lookup_dir_index(NULL, root, path, parent_ino,
name, namelen, index, 0);
/*
* Since lookup_dir_index() will return -ENOENT when not found,
* we need to do extra check.
*/
if (IS_ERR(dir_item) && PTR_ERR(dir_item) == -ENOENT)
dir_item = NULL;
if (IS_ERR(dir_item)) {
ret = PTR_ERR(dir_item);
goto out;
}
if (dir_item)
del_dir_index = 1;
btrfs_release_path(path);
if (!del_inode_ref && !del_dir_item && !del_dir_index) {
/* All not found, shouldn't happen */
ret = -ENOENT;
goto out;
}
if (del_inode_ref) {
/* Only decrease nlink when deleting inode_ref */
key.objectid = ino;
key.type = BTRFS_INODE_ITEM_KEY;
key.offset = 0;
ret = btrfs_search_slot(trans, root, &key, path, -1, 1);
if (ret) {
if (ret > 0)
ret = -ENOENT;
goto out;
}
inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_item);
nlinks = btrfs_inode_nlink(path->nodes[0], inode_item);
if (nlinks > 0)
nlinks--;
btrfs_set_inode_nlink(path->nodes[0], inode_item, nlinks);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
/* For nlinks == 0, add it to orphan list if needed */
if (nlinks == 0 && add_orphan) {
ret = btrfs_add_orphan_item(trans, root, path, ino);
if (ret < 0)
goto out;
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
}
ret = btrfs_del_inode_ref(trans, root, name, namelen, ino,
parent_ino, &index);
if (ret < 0)
goto out;
}
if (del_dir_index) {
dir_item = btrfs_lookup_dir_index(trans, root, path,
parent_ino, name, namelen,
index, -1);
if (IS_ERR(dir_item)) {
ret = PTR_ERR(dir_item);
goto out;
}
if (!dir_item) {
ret = -ENOENT;
goto out;
}
ret = btrfs_delete_one_dir_name(trans, root, path, dir_item);
if (ret)
goto out;
btrfs_release_path(path);
/* Update inode size of the parent inode */
key.objectid = parent_ino;
key.type = BTRFS_INODE_ITEM_KEY;
key.offset = 0;
ret = btrfs_search_slot(trans, root, &key, path, 1, 1);
if (ret)
goto out;
inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_item);
inode_size = btrfs_inode_size(path->nodes[0], inode_item);
if (inode_size >= namelen)
inode_size -= namelen;
btrfs_set_inode_size(path->nodes[0], inode_item, inode_size);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
}
if (del_dir_item) {
dir_item = btrfs_lookup_dir_item(trans, root, path, parent_ino,
name, namelen, -1);
if (IS_ERR(dir_item)) {
ret = PTR_ERR(dir_item);
goto out;
}
if (!dir_item) {
ret = -ENOENT;
goto out;
}
ret = btrfs_delete_one_dir_name(trans, root, path, dir_item);
if (ret < 0)
goto out;
btrfs_release_path(path);
/* Update inode size of the parent inode */
key.objectid = parent_ino;
key.type = BTRFS_INODE_ITEM_KEY;
key.offset = 0;
ret = btrfs_search_slot(trans, root, &key, path, 1, 1);
if (ret)
goto out;
inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_item);
inode_size = btrfs_inode_size(path->nodes[0], inode_item);
if (inode_size >= namelen)
inode_size -= namelen;
btrfs_set_inode_size(path->nodes[0], inode_item, inode_size);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
}
out:
btrfs_free_path(path);
return ret;
}
/*
* Add dir_item/index for 'parent_ino' if add_backref is true, also insert a
* backref from the ino to parent dir and update the nlink(Kernel version does
* not do this thing)
*
* Currently only supports adding link from an inode to another inode.
*/
int btrfs_add_link(struct btrfs_trans_handle *trans, struct btrfs_root *root,
u64 ino, u64 parent_ino, char *name, int namelen,
u8 type, u64 *index, int add_backref)
{
struct btrfs_path *path;
struct btrfs_key key;
struct btrfs_inode_item *inode_item;
u32 nlink;
u64 inode_size;
u64 ret_index = 0;
int ret = 0;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
if (index && *index) {
ret_index = *index;
} else {
ret = btrfs_find_free_dir_index(root, parent_ino, &ret_index);
if (ret < 0)
goto out;
}
ret = check_dir_conflict(root, name, namelen, parent_ino, ret_index);
if (ret < 0)
goto out;
/* Add inode ref */
if (add_backref) {
ret = btrfs_insert_inode_ref(trans, root, name, namelen,
ino, parent_ino, ret_index);
if (ret < 0)
goto out;
/* Update nlinks for the inode */
key.objectid = ino;
key.type = BTRFS_INODE_ITEM_KEY;
key.offset = 0;
ret = btrfs_search_slot(trans, root, &key, path, 1, 1);
if (ret) {
if (ret > 0)
ret = -ENOENT;
goto out;
}
inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_item);
nlink = btrfs_inode_nlink(path->nodes[0], inode_item);
nlink++;
btrfs_set_inode_nlink(path->nodes[0], inode_item, nlink);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
}
/* Add dir_item and dir_index */
key.objectid = ino;
key.type = BTRFS_INODE_ITEM_KEY;
key.offset = 0;
ret = btrfs_insert_dir_item(trans, root, name, namelen, parent_ino,
&key, type, ret_index);
if (ret < 0)
goto out;
/* Update inode size of the parent inode */
key.objectid = parent_ino;
key.type = BTRFS_INODE_ITEM_KEY;
key.offset = 0;
ret = btrfs_search_slot(trans, root, &key, path, 1, 1);
if (ret)
goto out;
inode_item = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_inode_item);
inode_size = btrfs_inode_size(path->nodes[0], inode_item);
inode_size += namelen * 2;
btrfs_set_inode_size(path->nodes[0], inode_item, inode_size);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_release_path(path);
out:
btrfs_free_path(path);
if (ret == 0 && index)
*index = ret_index;
return ret;
}
static int do_setxattr(struct btrfs_trans_handle *trans,
struct inode *inode, const char *name,
const void *value, size_t size, int flags)
{
struct btrfs_dir_item *di = NULL;
struct btrfs_root *root = BTRFS_I(inode)->root;
struct btrfs_path *path;
size_t name_len = strlen(name);
int ret = 0;
if (name_len + size > BTRFS_MAX_XATTR_SIZE(root))
return -ENOSPC;
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
path->skip_release_on_error = 1;
if (!value) {
di = btrfs_lookup_xattr(trans, root, path, btrfs_ino(inode),
name, name_len, -1);
if (!di && (flags & XATTR_REPLACE))
ret = -ENODATA;
else if (IS_ERR(di))
ret = PTR_ERR(di);
else if (di)
ret = btrfs_delete_one_dir_name(trans, root, path, di);
goto out;
}
/*
* For a replace we can't just do the insert blindly.
* Do a lookup first (read-only btrfs_search_slot), and return if xattr
* doesn't exist. If it exists, fall down below to the insert/replace
* path - we can't race with a concurrent xattr delete, because the VFS
* locks the inode's i_mutex before calling setxattr or removexattr.
*/
if (flags & XATTR_REPLACE) {
ASSERT(inode_is_locked(inode));
di = btrfs_lookup_xattr(NULL, root, path, btrfs_ino(inode),
name, name_len, 0);
if (!di)
ret = -ENODATA;
else if (IS_ERR(di))
ret = PTR_ERR(di);
if (ret)
goto out;
btrfs_release_path(path);
di = NULL;
}
ret = btrfs_insert_xattr_item(trans, root, path, btrfs_ino(inode),
name, name_len, value, size);
if (ret == -EOVERFLOW) {
/*
* We have an existing item in a leaf, split_leaf couldn't
* expand it. That item might have or not a dir_item that
* matches our target xattr, so lets check.
*/
ret = 0;
btrfs_assert_tree_locked(path->nodes[0]);
di = btrfs_match_dir_item_name(root, path, name, name_len);
if (!di && !(flags & XATTR_REPLACE)) {
ret = -ENOSPC;
goto out;
}
} else if (ret == -EEXIST) {
ret = 0;
di = btrfs_match_dir_item_name(root, path, name, name_len);
ASSERT(di); /* logic error */
} else if (ret) {
goto out;
}
if (di && (flags & XATTR_CREATE)) {
ret = -EEXIST;
goto out;
}
if (di) {
/*
* We're doing a replace, and it must be atomic, that is, at
* any point in time we have either the old or the new xattr
* value in the tree. We don't want readers (getxattr and
* listxattrs) to miss a value, this is specially important
* for ACLs.
*/
const int slot = path->slots[0];
struct extent_buffer *leaf = path->nodes[0];
const u16 old_data_len = btrfs_dir_data_len(leaf, di);
const u32 item_size = btrfs_item_size_nr(leaf, slot);
const u32 data_size = sizeof(*di) + name_len + size;
struct btrfs_item *item;
unsigned long data_ptr;
char *ptr;
if (size > old_data_len) {
if (btrfs_leaf_free_space(root, leaf) <
(size - old_data_len)) {
ret = -ENOSPC;
goto out;
}
}
if (old_data_len + name_len + sizeof(*di) == item_size) {
/* No other xattrs packed in the same leaf item. */
if (size > old_data_len)
btrfs_extend_item(root, path,
size - old_data_len);
else if (size < old_data_len)
btrfs_truncate_item(root, path, data_size, 1);
} else {
/* There are other xattrs packed in the same item. */
ret = btrfs_delete_one_dir_name(trans, root, path, di);
if (ret)
goto out;
btrfs_extend_item(root, path, data_size);
}
item = btrfs_item_nr(slot);
ptr = btrfs_item_ptr(leaf, slot, char);
ptr += btrfs_item_size(leaf, item) - data_size;
di = (struct btrfs_dir_item *)ptr;
btrfs_set_dir_data_len(leaf, di, size);
data_ptr = ((unsigned long)(di + 1)) + name_len;
write_extent_buffer(leaf, value, data_ptr, size);
btrfs_mark_buffer_dirty(leaf);
} else {
/*
* Insert, and we had space for the xattr, so path->slots[0] is
* where our xattr dir_item is and btrfs_insert_xattr_item()
* filled it.
*/
}
/*
* insert a directory item in the tree, doing all the magic for
* both indexes. 'dir' indicates which objectid to insert it into,
* 'location' is the key to stuff into the directory item, 'type' is the
* type of the inode we're pointing to, and 'index' is the sequence number
* to use for the second index (if one is created).
* Will return 0 or -ENOMEM
*/
int btrfs_insert_dir_item(struct btrfs_trans_handle *trans, struct btrfs_root
*root, const char *name, int name_len,
struct inode *dir, struct btrfs_key *location,
u8 type, u64 index)
{
int ret = 0;
int ret2 = 0;
struct btrfs_path *path;
struct btrfs_dir_item *dir_item;
struct extent_buffer *leaf;
unsigned long name_ptr;
struct btrfs_key key;
struct btrfs_disk_key disk_key;
u32 data_size;
key.objectid = btrfs_ino(dir);
btrfs_set_key_type(&key, BTRFS_DIR_ITEM_KEY);
key.offset = btrfs_name_hash(name, name_len);
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
path->leave_spinning = 1;
btrfs_cpu_key_to_disk(&disk_key, location);
data_size = sizeof(*dir_item) + name_len;
dir_item = insert_with_overflow(trans, root, path, &key, data_size,
name, name_len);
if (IS_ERR(dir_item)) {
ret = PTR_ERR(dir_item);
if (ret == -EEXIST)
goto second_insert;
goto out_free;
}
leaf = path->nodes[0];
btrfs_set_dir_item_key(leaf, dir_item, &disk_key);
btrfs_set_dir_type(leaf, dir_item, type);
btrfs_set_dir_data_len(leaf, dir_item, 0);
btrfs_set_dir_name_len(leaf, dir_item, name_len);
btrfs_set_dir_transid(leaf, dir_item, trans->transid);
name_ptr = (unsigned long)(dir_item + 1);
write_extent_buffer(leaf, name, name_ptr, name_len);
btrfs_mark_buffer_dirty(leaf);
second_insert:
/* FIXME, use some real flag for selecting the extra index */
if (root == root->fs_info->tree_root) {
ret = 0;
goto out_free;
}
btrfs_release_path(path);
ret2 = btrfs_insert_delayed_dir_index(trans, root, name, name_len, dir,
&disk_key, type, index);
out_free:
btrfs_free_path(path);
if (ret)
return ret;
if (ret2)
return ret2;
return 0;
}
static int traverse_directory(struct btrfs_trans_handle *trans,
struct btrfs_root *root, char *dir_name,
struct directory_name_entry *dir_head, int out_fd)
{
int ret = 0;
struct btrfs_inode_item cur_inode;
struct btrfs_inode_item *inode_item;
int count, i, dir_index_cnt;
struct direct **files;
struct stat st;
struct directory_name_entry *dir_entry, *parent_dir_entry;
struct direct *cur_file;
ino_t parent_inum, cur_inum;
ino_t highest_inum = 0;
char *parent_dir_name;
struct btrfs_path path;
struct extent_buffer *leaf;
struct btrfs_key root_dir_key;
u64 root_dir_inode_size = 0;
/* Add list for source directory */
dir_entry = malloc(sizeof(struct directory_name_entry));
dir_entry->dir_name = dir_name;
dir_entry->path = strdup(dir_name);
parent_inum = highest_inum + BTRFS_FIRST_FREE_OBJECTID;
dir_entry->inum = parent_inum;
list_add_tail(&dir_entry->list, &dir_head->list);
btrfs_init_path(&path);
root_dir_key.objectid = btrfs_root_dirid(&root->root_item);
root_dir_key.offset = 0;
btrfs_set_key_type(&root_dir_key, BTRFS_INODE_ITEM_KEY);
ret = btrfs_lookup_inode(trans, root, &path, &root_dir_key, 1);
if (ret) {
fprintf(stderr, "root dir lookup error\n");
return -1;
}
leaf = path.nodes[0];
inode_item = btrfs_item_ptr(leaf, path.slots[0],
struct btrfs_inode_item);
root_dir_inode_size = calculate_dir_inode_size(dir_name);
btrfs_set_inode_size(leaf, inode_item, root_dir_inode_size);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(&path);
do {
parent_dir_entry = list_entry(dir_head->list.next,
struct directory_name_entry,
list);
list_del(&parent_dir_entry->list);
parent_inum = parent_dir_entry->inum;
parent_dir_name = parent_dir_entry->dir_name;
if (chdir(parent_dir_entry->path)) {
fprintf(stderr, "chdir error for %s\n",
parent_dir_name);
goto fail_no_files;
}
count = scandir(parent_dir_entry->path, &files,
directory_select, NULL);
if (count == -1)
{
fprintf(stderr, "scandir for %s failed: %s\n",
parent_dir_name, strerror (errno));
goto fail;
}
for (i = 0; i < count; i++) {
cur_file = files[i];
if (lstat(cur_file->d_name, &st) == -1) {
fprintf(stderr, "lstat failed for file %s\n",
cur_file->d_name);
goto fail;
}
cur_inum = ++highest_inum + BTRFS_FIRST_FREE_OBJECTID;
ret = add_directory_items(trans, root,
cur_inum, parent_inum,
cur_file->d_name,
&st, &dir_index_cnt);
if (ret) {
fprintf(stderr, "add_directory_items failed\n");
goto fail;
}
ret = add_inode_items(trans, root, &st,
cur_file->d_name, cur_inum,
parent_inum, dir_index_cnt,
&cur_inode);
if (ret) {
fprintf(stderr, "add_inode_items failed\n");
goto fail;
}
ret = add_xattr_item(trans, root,
cur_inum, cur_file->d_name);
if (ret) {
fprintf(stderr, "add_xattr_item failed\n");
if(ret != -ENOTSUP)
goto fail;
}
if (S_ISDIR(st.st_mode)) {
dir_entry = malloc(sizeof(struct directory_name_entry));
dir_entry->dir_name = cur_file->d_name;
dir_entry->path = make_path(parent_dir_entry->path,
cur_file->d_name);
dir_entry->inum = cur_inum;
list_add_tail(&dir_entry->list, &dir_head->list);
} else if (S_ISREG(st.st_mode)) {
ret = add_file_items(trans, root, &cur_inode,
cur_inum, parent_inum, &st,
cur_file->d_name, out_fd);
if (ret) {
fprintf(stderr, "add_file_items failed\n");
goto fail;
}
} else if (S_ISLNK(st.st_mode)) {
ret = add_symbolic_link(trans, root,
cur_inum, cur_file->d_name);
if (ret) {
fprintf(stderr, "add_symbolic_link failed\n");
goto fail;
}
}
}
free_namelist(files, count);
free(parent_dir_entry->path);
free(parent_dir_entry);
index_cnt = 2;
} while (!list_empty(&dir_head->list));
return 0;
fail:
free_namelist(files, count);
fail_no_files:
free(parent_dir_entry->path);
free(parent_dir_entry);
return -1;
}
int btrfs_dedup_enable(struct btrfs_fs_info *fs_info, u16 type, u16 backend,
u64 blocksize, u64 limit)
{
struct btrfs_dedup_info *dedup_info;
struct btrfs_root *dedup_root;
struct btrfs_key key;
struct btrfs_trans_handle *trans;
struct btrfs_path *path;
struct btrfs_dedup_status_item *status;
int create_tree;
u64 compat_ro_flag = btrfs_super_compat_ro_flags(fs_info->super_copy);
int ret = 0;
/* Sanity check */
if (blocksize > BTRFS_DEDUP_BLOCKSIZE_MAX ||
blocksize < BTRFS_DEDUP_BLOCKSIZE_MIN ||
blocksize < fs_info->tree_root->sectorsize ||
!is_power_of_2(blocksize))
return -EINVAL;
if (type > ARRAY_SIZE(btrfs_dedup_sizes))
return -EINVAL;
if (backend >= BTRFS_DEDUP_BACKEND_LAST)
return -EINVAL;
if (backend == BTRFS_DEDUP_BACKEND_INMEMORY && limit == 0)
limit = 4096; /* default value */
if (backend == BTRFS_DEDUP_BACKEND_ONDISK && limit != 0)
limit = 0;
/*
* If current fs doesn't support DEDUP feature, don't enable
* on-disk dedup.
*/
if (!(compat_ro_flag & BTRFS_FEATURE_COMPAT_RO_DEDUP) &&
backend == BTRFS_DEDUP_BACKEND_ONDISK)
return -EINVAL;
/* Meaningless and unable to enable dedup for RO fs */
if (fs_info->sb->s_flags & MS_RDONLY)
return -EINVAL;
if (fs_info->dedup_info) {
dedup_info = fs_info->dedup_info;
/* Check if we are re-enable for different dedup config */
if (dedup_info->blocksize != blocksize ||
dedup_info->hash_type != type ||
dedup_info->backend != backend) {
btrfs_dedup_disable(fs_info);
goto enable;
}
/* On-fly limit change is OK */
mutex_lock(&dedup_info->lock);
fs_info->dedup_info->limit_nr = limit;
mutex_unlock(&dedup_info->lock);
return 0;
}
enable:
create_tree = compat_ro_flag & BTRFS_FEATURE_COMPAT_RO_DEDUP;
ret = init_dedup_info(fs_info, type, backend, blocksize, limit);
dedup_info = fs_info->dedup_info;
if (ret < 0)
goto out;
if (!create_tree)
goto out;
/* Create dedup tree for status at least */
path = btrfs_alloc_path();
if (!path) {
ret = -ENOMEM;
goto out;
}
trans = btrfs_start_transaction(fs_info->tree_root, 2);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
btrfs_free_path(path);
goto out;
}
dedup_root = btrfs_create_tree(trans, fs_info,
BTRFS_DEDUP_TREE_OBJECTID);
if (IS_ERR(dedup_root)) {
ret = PTR_ERR(dedup_root);
btrfs_abort_transaction(trans, fs_info->tree_root, ret);
btrfs_free_path(path);
goto out;
}
dedup_info->dedup_root = dedup_root;
key.objectid = 0;
key.type = BTRFS_DEDUP_STATUS_ITEM_KEY;
key.offset = 0;
ret = btrfs_insert_empty_item(trans, dedup_root, path, &key,
sizeof(*status));
if (ret < 0) {
btrfs_abort_transaction(trans, fs_info->tree_root, ret);
btrfs_free_path(path);
goto out;
}
status = btrfs_item_ptr(path->nodes[0], path->slots[0],
struct btrfs_dedup_status_item);
btrfs_set_dedup_status_blocksize(path->nodes[0], status, blocksize);
btrfs_set_dedup_status_limit(path->nodes[0], status, limit);
btrfs_set_dedup_status_hash_type(path->nodes[0], status, type);
btrfs_set_dedup_status_backend(path->nodes[0], status, backend);
btrfs_mark_buffer_dirty(path->nodes[0]);
btrfs_free_path(path);
ret = btrfs_commit_transaction(trans, fs_info->tree_root);
out:
if (ret < 0) {
kfree(dedup_info);
fs_info->dedup_info = NULL;
}
return ret;
}
/*
* this is very complex, but the basic idea is to drop all extents
* in the range start - end. hint_block is filled in with a block number
* that would be a good hint to the block allocator for this file.
*
* If an extent intersects the range but is not entirely inside the range
* it is either truncated or split. Anything entirely inside the range
* is deleted from the tree.
*/
int btrfs_drop_extents(struct btrfs_trans_handle *trans, struct inode *inode,
u64 start, u64 end, u64 *hint_byte, int drop_cache)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
struct extent_buffer *leaf;
struct btrfs_file_extent_item *fi;
struct btrfs_path *path;
struct btrfs_key key;
struct btrfs_key new_key;
u64 search_start = start;
u64 disk_bytenr = 0;
u64 num_bytes = 0;
u64 extent_offset = 0;
u64 extent_end = 0;
int del_nr = 0;
int del_slot = 0;
int extent_type;
int recow;
int ret;
if (drop_cache)
btrfs_drop_extent_cache(inode, start, end - 1, 0);
path = btrfs_alloc_path();
if (!path)
return -ENOMEM;
while (1) {
recow = 0;
ret = btrfs_lookup_file_extent(trans, root, path, inode->i_ino,
search_start, -1);
if (ret < 0)
break;
if (ret > 0 && path->slots[0] > 0 && search_start == start) {
leaf = path->nodes[0];
btrfs_item_key_to_cpu(leaf, &key, path->slots[0] - 1);
if (key.objectid == inode->i_ino &&
key.type == BTRFS_EXTENT_DATA_KEY)
path->slots[0]--;
}
ret = 0;
next_slot:
leaf = path->nodes[0];
if (path->slots[0] >= btrfs_header_nritems(leaf)) {
BUG_ON(del_nr > 0);
ret = btrfs_next_leaf(root, path);
if (ret < 0)
break;
if (ret > 0) {
ret = 0;
break;
}
leaf = path->nodes[0];
recow = 1;
}
btrfs_item_key_to_cpu(leaf, &key, path->slots[0]);
if (key.objectid > inode->i_ino ||
key.type > BTRFS_EXTENT_DATA_KEY || key.offset >= end)
break;
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
extent_type = btrfs_file_extent_type(leaf, fi);
if (extent_type == BTRFS_FILE_EXTENT_REG ||
extent_type == BTRFS_FILE_EXTENT_PREALLOC) {
disk_bytenr = btrfs_file_extent_disk_bytenr(leaf, fi);
num_bytes = btrfs_file_extent_disk_num_bytes(leaf, fi);
extent_offset = btrfs_file_extent_offset(leaf, fi);
extent_end = key.offset +
btrfs_file_extent_num_bytes(leaf, fi);
} else if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
extent_end = key.offset +
btrfs_file_extent_inline_len(leaf, fi);
} else {
WARN_ON(1);
extent_end = search_start;
}
if (extent_end <= search_start) {
path->slots[0]++;
goto next_slot;
}
search_start = max(key.offset, start);
if (recow) {
btrfs_release_path(root, path);
continue;
}
/*
* | - range to drop - |
* | -------- extent -------- |
*/
if (start > key.offset && end < extent_end) {
BUG_ON(del_nr > 0);
BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
memcpy(&new_key, &key, sizeof(new_key));
new_key.offset = start;
ret = btrfs_duplicate_item(trans, root, path,
&new_key);
if (ret == -EAGAIN) {
btrfs_release_path(root, path);
continue;
}
if (ret < 0)
break;
leaf = path->nodes[0];
fi = btrfs_item_ptr(leaf, path->slots[0] - 1,
struct btrfs_file_extent_item);
btrfs_set_file_extent_num_bytes(leaf, fi,
start - key.offset);
fi = btrfs_item_ptr(leaf, path->slots[0],
struct btrfs_file_extent_item);
extent_offset += start - key.offset;
btrfs_set_file_extent_offset(leaf, fi, extent_offset);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - start);
btrfs_mark_buffer_dirty(leaf);
if (disk_bytenr > 0) {
ret = btrfs_inc_extent_ref(trans, root,
disk_bytenr, num_bytes, 0,
root->root_key.objectid,
new_key.objectid,
start - extent_offset);
BUG_ON(ret);
*hint_byte = disk_bytenr;
}
key.offset = start;
}
/*
* | ---- range to drop ----- |
* | -------- extent -------- |
*/
if (start <= key.offset && end < extent_end) {
BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
memcpy(&new_key, &key, sizeof(new_key));
new_key.offset = end;
btrfs_set_item_key_safe(trans, root, path, &new_key);
extent_offset += end - key.offset;
btrfs_set_file_extent_offset(leaf, fi, extent_offset);
btrfs_set_file_extent_num_bytes(leaf, fi,
extent_end - end);
btrfs_mark_buffer_dirty(leaf);
if (disk_bytenr > 0) {
inode_sub_bytes(inode, end - key.offset);
*hint_byte = disk_bytenr;
}
break;
}
search_start = extent_end;
/*
* | ---- range to drop ----- |
* | -------- extent -------- |
*/
if (start > key.offset && end >= extent_end) {
BUG_ON(del_nr > 0);
BUG_ON(extent_type == BTRFS_FILE_EXTENT_INLINE);
btrfs_set_file_extent_num_bytes(leaf, fi,
start - key.offset);
btrfs_mark_buffer_dirty(leaf);
if (disk_bytenr > 0) {
inode_sub_bytes(inode, extent_end - start);
*hint_byte = disk_bytenr;
}
if (end == extent_end)
break;
path->slots[0]++;
goto next_slot;
}
/*
* | ---- range to drop ----- |
* | ------ extent ------ |
*/
if (start <= key.offset && end >= extent_end) {
if (del_nr == 0) {
del_slot = path->slots[0];
del_nr = 1;
} else {
BUG_ON(del_slot + del_nr != path->slots[0]);
del_nr++;
}
if (extent_type == BTRFS_FILE_EXTENT_INLINE) {
inode_sub_bytes(inode,
extent_end - key.offset);
extent_end = ALIGN(extent_end,
root->sectorsize);
} else if (disk_bytenr > 0) {
ret = btrfs_free_extent(trans, root,
disk_bytenr, num_bytes, 0,
root->root_key.objectid,
key.objectid, key.offset -
extent_offset);
BUG_ON(ret);
inode_sub_bytes(inode,
extent_end - key.offset);
*hint_byte = disk_bytenr;
}
if (end == extent_end)
break;
if (path->slots[0] + 1 < btrfs_header_nritems(leaf)) {
path->slots[0]++;
goto next_slot;
}
ret = btrfs_del_items(trans, root, path, del_slot,
del_nr);
BUG_ON(ret);
del_nr = 0;
del_slot = 0;
btrfs_release_path(root, path);
continue;
}
BUG_ON(1);
}
int btrfs_csum_file_block(struct btrfs_trans_handle *trans,
struct btrfs_root *root, u64 alloc_end,
u64 bytenr, char *data, size_t len)
{
int ret;
struct btrfs_key file_key;
struct btrfs_key found_key;
u64 next_offset = (u64)-1;
int found_next = 0;
struct btrfs_path *path;
struct btrfs_csum_item *item;
struct extent_buffer *leaf = NULL;
u64 csum_offset;
u32 csum_result = ~(u32)0;
u32 nritems;
u32 ins_size;
u16 csum_size =
btrfs_super_csum_size(&root->fs_info->super_copy);
path = btrfs_alloc_path();
BUG_ON(!path);
file_key.objectid = BTRFS_EXTENT_CSUM_OBJECTID;
file_key.offset = bytenr;
file_key.type = BTRFS_EXTENT_CSUM_KEY;
item = btrfs_lookup_csum(trans, root, path, bytenr, 1);
if (!IS_ERR(item)) {
leaf = path->nodes[0];
goto found;
}
ret = PTR_ERR(item);
if (ret == -EFBIG) {
u32 item_size;
/* we found one, but it isn't big enough yet */
leaf = path->nodes[0];
item_size = btrfs_item_size_nr(leaf, path->slots[0]);
if ((item_size / csum_size) >= MAX_CSUM_ITEMS(root, csum_size)) {
/* already at max size, make a new one */
goto insert;
}
} else {
int slot = path->slots[0] + 1;
/* we didn't find a csum item, insert one */
nritems = btrfs_header_nritems(path->nodes[0]);
if (path->slots[0] >= nritems - 1) {
ret = btrfs_next_leaf(root, path);
if (ret == 1)
found_next = 1;
if (ret != 0)
goto insert;
slot = 0;
}
btrfs_item_key_to_cpu(path->nodes[0], &found_key, slot);
if (found_key.objectid != BTRFS_EXTENT_CSUM_OBJECTID ||
found_key.type != BTRFS_EXTENT_CSUM_KEY) {
found_next = 1;
goto insert;
}
next_offset = found_key.offset;
found_next = 1;
goto insert;
}
/*
* at this point, we know the tree has an item, but it isn't big
* enough yet to put our csum in. Grow it
*/
btrfs_release_path(root, path);
ret = btrfs_search_slot(trans, root, &file_key, path,
csum_size, 1);
if (ret < 0)
goto fail;
if (ret == 0) {
BUG();
}
if (path->slots[0] == 0) {
goto insert;
}
path->slots[0]--;
leaf = path->nodes[0];
btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0]);
csum_offset = (file_key.offset - found_key.offset) / root->sectorsize;
if (found_key.objectid != BTRFS_EXTENT_CSUM_OBJECTID ||
found_key.type != BTRFS_EXTENT_CSUM_KEY ||
csum_offset >= MAX_CSUM_ITEMS(root, csum_size)) {
goto insert;
}
if (csum_offset >= btrfs_item_size_nr(leaf, path->slots[0]) /
csum_size) {
u32 diff = (csum_offset + 1) * csum_size;
diff = diff - btrfs_item_size_nr(leaf, path->slots[0]);
if (diff != csum_size)
goto insert;
ret = btrfs_extend_item(trans, root, path, diff);
BUG_ON(ret);
goto csum;
}
insert:
btrfs_release_path(root, path);
csum_offset = 0;
if (found_next) {
u64 tmp = min(alloc_end, next_offset);
tmp -= file_key.offset;
tmp /= root->sectorsize;
tmp = max((u64)1, tmp);
tmp = min(tmp, (u64)MAX_CSUM_ITEMS(root, csum_size));
ins_size = csum_size * tmp;
} else {
ins_size = csum_size;
}
ret = btrfs_insert_empty_item(trans, root, path, &file_key,
ins_size);
if (ret < 0)
goto fail;
if (ret != 0) {
WARN_ON(1);
goto fail;
}
csum:
leaf = path->nodes[0];
item = btrfs_item_ptr(leaf, path->slots[0], struct btrfs_csum_item);
ret = 0;
item = (struct btrfs_csum_item *)((unsigned char *)item +
csum_offset * csum_size);
found:
csum_result = btrfs_csum_data(root, data, csum_result, len);
btrfs_csum_final(csum_result, (char *)&csum_result);
if (csum_result == 0) {
printk("csum result is 0 for block %llu\n",
(unsigned long long)bytenr);
}
write_extent_buffer(leaf, &csum_result, (unsigned long)item,
csum_size);
btrfs_mark_buffer_dirty(path->nodes[0]);
fail:
btrfs_release_path(root, path);
btrfs_free_path(path);
return ret;
}
int convert_free_space_to_bitmaps(struct btrfs_trans_handle *trans,
struct btrfs_fs_info *fs_info,
struct btrfs_block_group_cache *block_group,
struct btrfs_path *path)
{
struct btrfs_root *root = fs_info->free_space_root;
struct btrfs_free_space_info *info;
struct btrfs_key key, found_key;
struct extent_buffer *leaf;
unsigned long *bitmap;
char *bitmap_cursor;
u64 start, end;
u64 bitmap_range, i;
u32 bitmap_size, flags, expected_extent_count;
u32 extent_count = 0;
int done = 0, nr;
int ret;
bitmap_size = free_space_bitmap_size(block_group->key.offset,
block_group->sectorsize);
bitmap = alloc_bitmap(bitmap_size);
if (!bitmap) {
ret = -ENOMEM;
goto out;
}
start = block_group->key.objectid;
end = block_group->key.objectid + block_group->key.offset;
key.objectid = end - 1;
key.type = (u8)-1;
key.offset = (u64)-1;
while (!done) {
ret = btrfs_search_prev_slot(trans, root, &key, path, -1, 1);
if (ret)
goto out;
leaf = path->nodes[0];
nr = 0;
path->slots[0]++;
while (path->slots[0] > 0) {
btrfs_item_key_to_cpu(leaf, &found_key, path->slots[0] - 1);
if (found_key.type == BTRFS_FREE_SPACE_INFO_KEY) {
ASSERT(found_key.objectid == block_group->key.objectid);
ASSERT(found_key.offset == block_group->key.offset);
done = 1;
break;
} else if (found_key.type == BTRFS_FREE_SPACE_EXTENT_KEY) {
u64 first, last;
ASSERT(found_key.objectid >= start);
ASSERT(found_key.objectid < end);
ASSERT(found_key.objectid + found_key.offset <= end);
first = div_u64(found_key.objectid - start,
block_group->sectorsize);
last = div_u64(found_key.objectid + found_key.offset - start,
block_group->sectorsize);
bitmap_set(bitmap, first, last - first);
extent_count++;
nr++;
path->slots[0]--;
} else {
ASSERT(0);
}
}
ret = btrfs_del_items(trans, root, path, path->slots[0], nr);
if (ret)
goto out;
btrfs_release_path(path);
}
info = search_free_space_info(trans, fs_info, block_group, path, 1);
if (IS_ERR(info)) {
ret = PTR_ERR(info);
goto out;
}
leaf = path->nodes[0];
flags = btrfs_free_space_flags(leaf, info);
flags |= BTRFS_FREE_SPACE_USING_BITMAPS;
btrfs_set_free_space_flags(leaf, info, flags);
expected_extent_count = btrfs_free_space_extent_count(leaf, info);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
if (extent_count != expected_extent_count) {
btrfs_err(fs_info, "incorrect extent count for %llu; counted %u, expected %u",
block_group->key.objectid, extent_count,
expected_extent_count);
ASSERT(0);
ret = -EIO;
goto out;
}
bitmap_cursor = (char *)bitmap;
bitmap_range = block_group->sectorsize * BTRFS_FREE_SPACE_BITMAP_BITS;
i = start;
while (i < end) {
unsigned long ptr;
u64 extent_size;
u32 data_size;
extent_size = min(end - i, bitmap_range);
data_size = free_space_bitmap_size(extent_size,
block_group->sectorsize);
key.objectid = i;
key.type = BTRFS_FREE_SPACE_BITMAP_KEY;
key.offset = extent_size;
ret = btrfs_insert_empty_item(trans, root, path, &key,
data_size);
if (ret)
goto out;
leaf = path->nodes[0];
ptr = btrfs_item_ptr_offset(leaf, path->slots[0]);
write_extent_buffer(leaf, bitmap_cursor, ptr,
data_size);
btrfs_mark_buffer_dirty(leaf);
btrfs_release_path(path);
i += extent_size;
bitmap_cursor += data_size;
}
ret = 0;
out:
kvfree(bitmap);
if (ret)
btrfs_abort_transaction(trans, ret);
return ret;
}
