/*
* defrag a given btree. If cacheonly == 1, this won't read from the disk,
* otherwise every leaf in the btree is read and defragged.
*/
int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
{
struct btrfs_fs_info *info = root->fs_info;
int ret;
struct btrfs_trans_handle *trans;
unsigned long nr;
smp_mb();
if (root->defrag_running)
return 0;
trans = btrfs_start_transaction(root, 1);
while (1) {
root->defrag_running = 1;
ret = btrfs_defrag_leaves(trans, root, cacheonly);
nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
btrfs_btree_balance_dirty(info->tree_root, nr);
cond_resched();
trans = btrfs_start_transaction(root, 1);
if (root->fs_info->closing || ret != -EAGAIN)
break;
}
root->defrag_running = 0;
smp_mb();
btrfs_end_transaction(trans, root);
return 0;
}
static int update_seeding_flag(struct btrfs_root *root, int set_flag)
{
struct btrfs_trans_handle *trans;
struct btrfs_super_block *disk_super;
u64 super_flags;
int ret;
disk_super = root->fs_info->super_copy;
super_flags = btrfs_super_flags(disk_super);
if (set_flag) {
if (super_flags & BTRFS_SUPER_FLAG_SEEDING) {
if (force)
return 0;
else
warning("seeding flag is already set on %s",
device);
return 1;
}
super_flags |= BTRFS_SUPER_FLAG_SEEDING;
} else {
if (!(super_flags & BTRFS_SUPER_FLAG_SEEDING)) {
warning("seeding flag is not set on %s", device);
return 1;
}
super_flags &= ~BTRFS_SUPER_FLAG_SEEDING;
warning("seeding flag cleared on %s", device);
}
trans = btrfs_start_transaction(root, 1);
btrfs_set_super_flags(disk_super, super_flags);
ret = btrfs_commit_transaction(trans, root);
return ret;
}
int main(int ac, char **av)
{
struct btrfs_root *root;
struct btrfs_trans_handle *trans;
int ret;
if (ac != 2)
print_usage();
radix_tree_init();
if((ret = check_mounted(av[1])) < 0) {
fprintf(stderr, "Could not check mount status: %s\n", strerror(-ret));
goto out;
} else if(ret) {
fprintf(stderr, "%s is currently mounted. Aborting.\n", av[1]);
ret = -EBUSY;
goto out;
}
root = open_ctree(av[1], 0, OPEN_CTREE_WRITES);
if (root == NULL)
return 1;
trans = btrfs_start_transaction(root, 1);
btrfs_set_super_log_root(root->fs_info->super_copy, 0);
btrfs_set_super_log_root_level(root->fs_info->super_copy, 0);
btrfs_commit_transaction(trans, root);
close_ctree(root);
out:
return !!ret;
}
int __btrfs_setxattr(struct btrfs_trans_handle *trans,
struct inode *inode, const char *name,
const void *value, size_t size, int flags)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret;
if (trans)
return do_setxattr(trans, inode, name, value, size, flags);
ret = btrfs_reserve_metadata_space(root, 2);
if (ret)
return ret;
trans = btrfs_start_transaction(root, 1);
if (!trans) {
ret = -ENOMEM;
goto out;
}
btrfs_set_trans_block_group(trans, inode);
ret = do_setxattr(trans, inode, name, value, size, flags);
if (ret)
goto out;
inode->i_ctime = CURRENT_TIME;
ret = btrfs_update_inode(trans, root, inode);
BUG_ON(ret);
out:
btrfs_end_transaction_throttle(trans, root);
btrfs_unreserve_metadata_space(root, 2);
return ret;
}
/*
* @value: "" makes the attribute to empty, NULL removes it
*/
int __btrfs_setxattr(struct btrfs_trans_handle *trans,
struct inode *inode, const char *name,
const void *value, size_t size, int flags)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret;
if (trans)
return do_setxattr(trans, inode, name, value, size, flags);
trans = btrfs_start_transaction(root, 2);
if (IS_ERR(trans))
return PTR_ERR(trans);
ret = do_setxattr(trans, inode, name, value, size, flags);
if (ret)
goto out;
inode_inc_iversion(inode);
inode->i_ctime = current_fs_time(inode->i_sb);
set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags);
ret = btrfs_update_inode(trans, root, inode);
BUG_ON(ret);
out:
btrfs_end_transaction(trans, root);
return ret;
}
int btrfs_defrag_root(struct btrfs_root *root, int cacheonly)
{
struct btrfs_fs_info *info = root->fs_info;
struct btrfs_trans_handle *trans;
int ret;
unsigned long nr;
if (xchg(&root->defrag_running, 1))
return 0;
while (1) {
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans))
return PTR_ERR(trans);
ret = btrfs_defrag_leaves(trans, root, cacheonly);
nr = trans->blocks_used;
btrfs_end_transaction(trans, root);
btrfs_btree_balance_dirty(info->tree_root, nr);
cond_resched();
if (root->fs_info->closing || ret != -EAGAIN)
break;
}
root->defrag_running = 0;
return ret;
}
int update_seeding_flag(struct btrfs_root *root, int set_flag)
{
struct btrfs_trans_handle *trans;
struct btrfs_super_block *disk_super;
u64 super_flags;
disk_super = root->fs_info->super_copy;
super_flags = btrfs_super_flags(disk_super);
if (set_flag) {
if (super_flags & BTRFS_SUPER_FLAG_SEEDING) {
fprintf(stderr, "seeding flag is already set on %s\n",
device);
return 1;
}
super_flags |= BTRFS_SUPER_FLAG_SEEDING;
} else {
if (!(super_flags & BTRFS_SUPER_FLAG_SEEDING)) {
fprintf(stderr, "seeding flag is not set on %s\n",
device);
return 1;
}
super_flags &= ~BTRFS_SUPER_FLAG_SEEDING;
}
trans = btrfs_start_transaction(root, 1);
btrfs_set_super_flags(disk_super, super_flags);
btrfs_commit_transaction(trans, root);
return 0;
}
/*
* @value: "" makes the attribute to empty, NULL removes it
*/
int __btrfs_setxattr(struct btrfs_trans_handle *trans,
struct inode *inode, const char *name,
const void *value, size_t size, int flags)
{
struct btrfs_root *root = BTRFS_I(inode)->root;
int ret;
if (trans)
return do_setxattr(trans, inode, name, value, size, flags);
trans = btrfs_start_transaction(root, 2);
if (IS_ERR(trans))
return PTR_ERR(trans);
ret = do_setxattr(trans, inode, name, value, size, flags);
if (ret)
goto out;
inode->i_ctime = CURRENT_TIME;
ret = btrfs_update_inode(trans, root, inode);
BUG_ON(ret);
out:
btrfs_end_transaction_throttle(trans, root);
return ret;
}
/*
* Given a list of roots that need to be deleted, call btrfs_drop_snapshot on
* all of them
*/
int btrfs_drop_dead_root(struct btrfs_root *root)
{
struct btrfs_trans_handle *trans;
struct btrfs_root *tree_root = root->fs_info->tree_root;
unsigned long nr;
int ret;
while (1) {
/*
* we don't want to jump in and create a bunch of
* delayed refs if the transaction is starting to close
*/
wait_transaction_pre_flush(tree_root->fs_info);
trans = btrfs_start_transaction(tree_root, 1);
/*
* we've joined a transaction, make sure it isn't
* closing right now
*/
if (trans->transaction->delayed_refs.flushing) {
btrfs_end_transaction(trans, tree_root);
continue;
}
ret = btrfs_drop_snapshot(trans, root);
if (ret != -EAGAIN)
break;
ret = btrfs_update_root(trans, tree_root,
&root->root_key,
&root->root_item);
if (ret)
break;
nr = trans->blocks_used;
ret = btrfs_end_transaction(trans, tree_root);
BUG_ON(ret);
btrfs_btree_balance_dirty(tree_root, nr);
cond_resched();
}
BUG_ON(ret);
ret = btrfs_del_root(trans, tree_root, &root->root_key);
BUG_ON(ret);
nr = trans->blocks_used;
ret = btrfs_end_transaction(trans, tree_root);
BUG_ON(ret);
free_extent_buffer(root->node);
free_extent_buffer(root->commit_root);
kfree(root);
btrfs_btree_balance_dirty(tree_root, nr);
return ret;
}
static int create_snapshot(struct btrfs_root *root, struct dentry *dentry,
char *name, int namelen)
{
struct inode *inode;
struct btrfs_pending_snapshot *pending_snapshot;
struct btrfs_trans_handle *trans;
int ret;
if (!root->ref_cows)
return -EINVAL;
/*
* 1 - inode item
* 2 - refs
* 1 - root item
* 2 - dir items
*/
ret = btrfs_reserve_metadata_space(root, 6);
if (ret)
goto fail;
pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
if (!pending_snapshot) {
ret = -ENOMEM;
btrfs_unreserve_metadata_space(root, 6);
goto fail;
}
pending_snapshot->name = kmalloc(namelen + 1, GFP_NOFS);
if (!pending_snapshot->name) {
ret = -ENOMEM;
kfree(pending_snapshot);
btrfs_unreserve_metadata_space(root, 6);
goto fail;
}
memcpy(pending_snapshot->name, name, namelen);
pending_snapshot->name[namelen] = '\0';
pending_snapshot->dentry = dentry;
trans = btrfs_start_transaction(root, 1);
BUG_ON(!trans);
pending_snapshot->root = root;
list_add(&pending_snapshot->list,
&trans->transaction->pending_snapshots);
ret = btrfs_commit_transaction(trans, root);
BUG_ON(ret);
btrfs_unreserve_metadata_space(root, 6);
inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
goto fail;
}
BUG_ON(!inode);
d_instantiate(dentry, inode);
ret = 0;
fail:
return ret;
}
static int create_snapshot(struct btrfs_root *root, struct dentry *dentry)
{
struct inode *inode;
struct btrfs_pending_snapshot *pending_snapshot;
struct btrfs_trans_handle *trans;
int ret;
if (!root->ref_cows)
return -EINVAL;
pending_snapshot = kzalloc(sizeof(*pending_snapshot), GFP_NOFS);
if (!pending_snapshot)
return -ENOMEM;
btrfs_init_block_rsv(&pending_snapshot->block_rsv);
pending_snapshot->dentry = dentry;
pending_snapshot->root = root;
trans = btrfs_start_transaction(root->fs_info->extent_root, 5);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto fail;
}
ret = btrfs_snap_reserve_metadata(trans, pending_snapshot);
BUG_ON(ret);
list_add(&pending_snapshot->list,
&trans->transaction->pending_snapshots);
ret = btrfs_commit_transaction(trans, root->fs_info->extent_root);
BUG_ON(ret);
ret = pending_snapshot->error;
if (ret)
goto fail;
btrfs_orphan_cleanup(pending_snapshot->snap);
inode = btrfs_lookup_dentry(dentry->d_parent->d_inode, dentry);
if (IS_ERR(inode)) {
ret = PTR_ERR(inode);
goto fail;
}
BUG_ON(!inode);
d_instantiate(dentry, inode);
ret = 0;
fail:
kfree(pending_snapshot);
return ret;
}
static u64 __btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
struct btrfs_device *tgt_device = NULL;
struct btrfs_trans_handle *trans;
struct btrfs_root *root = fs_info->tree_root;
u64 result;
int ret;
if (fs_info->sb->s_flags & MS_RDONLY)
return -EROFS;
mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
btrfs_dev_replace_lock(dev_replace);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED;
btrfs_dev_replace_unlock(dev_replace);
goto leave;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
tgt_device = dev_replace->tgtdev;
dev_replace->tgtdev = NULL;
dev_replace->srcdev = NULL;
break;
}
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED;
dev_replace->time_stopped = get_seconds();
dev_replace->item_needs_writeback = 1;
btrfs_dev_replace_unlock(dev_replace);
btrfs_scrub_cancel(fs_info);
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return PTR_ERR(trans);
}
ret = btrfs_commit_transaction(trans, root);
WARN_ON(ret);
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
leave:
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return result;
}
int enable_extrefs_flag(struct btrfs_root *root)
{
struct btrfs_trans_handle *trans;
struct btrfs_super_block *disk_super;
u64 super_flags;
disk_super = root->fs_info->super_copy;
super_flags = btrfs_super_incompat_flags(disk_super);
super_flags |= BTRFS_FEATURE_INCOMPAT_EXTENDED_IREF;
trans = btrfs_start_transaction(root, 1);
btrfs_set_super_incompat_flags(disk_super, super_flags);
btrfs_commit_transaction(trans, root);
return 0;
}
static int set_super_incompat_flags(struct btrfs_root *root, u64 flags)
{
struct btrfs_trans_handle *trans;
struct btrfs_super_block *disk_super;
u64 super_flags;
int ret;
disk_super = root->fs_info->super_copy;
super_flags = btrfs_super_incompat_flags(disk_super);
super_flags |= flags;
trans = btrfs_start_transaction(root, 1);
btrfs_set_super_incompat_flags(disk_super, super_flags);
ret = btrfs_commit_transaction(trans, root);
return ret;
}
int btrfs_create_free_space_tree(struct btrfs_fs_info *fs_info)
{
struct btrfs_trans_handle *trans;
struct btrfs_root *tree_root = fs_info->tree_root;
struct btrfs_root *free_space_root;
struct btrfs_block_group_cache *block_group;
struct rb_node *node;
int ret;
trans = btrfs_start_transaction(tree_root, 0);
if (IS_ERR(trans))
return PTR_ERR(trans);
fs_info->creating_free_space_tree = 1;
free_space_root = btrfs_create_tree(trans, fs_info,
BTRFS_FREE_SPACE_TREE_OBJECTID);
if (IS_ERR(free_space_root)) {
ret = PTR_ERR(free_space_root);
goto abort;
}
fs_info->free_space_root = free_space_root;
node = rb_first(&fs_info->block_group_cache_tree);
while (node) {
block_group = rb_entry(node, struct btrfs_block_group_cache,
cache_node);
ret = populate_free_space_tree(trans, fs_info, block_group);
if (ret)
goto abort;
node = rb_next(node);
}
btrfs_set_fs_compat_ro(fs_info, FREE_SPACE_TREE);
fs_info->creating_free_space_tree = 0;
ret = btrfs_commit_transaction(trans, tree_root);
if (ret)
return ret;
return 0;
abort:
fs_info->creating_free_space_tree = 0;
btrfs_abort_transaction(trans, ret);
btrfs_end_transaction(trans, tree_root);
return ret;
}
static int cmd_rescue_zero_log(int argc, char **argv)
{
struct btrfs_root *root;
struct btrfs_trans_handle *trans;
struct btrfs_super_block *sb;
char *devname;
int ret;
clean_args_no_options(argc, argv, cmd_rescue_zero_log_usage);
if (check_argc_exact(argc, 2))
usage(cmd_rescue_zero_log_usage);
devname = argv[optind];
ret = check_mounted(devname);
if (ret < 0) {
errno = -ret;
error("could not check mount status: %m");
goto out;
} else if (ret) {
error("%s is currently mounted", devname);
ret = -EBUSY;
goto out;
}
root = open_ctree(devname, 0, OPEN_CTREE_WRITES | OPEN_CTREE_PARTIAL);
if (!root) {
error("could not open ctree");
return 1;
}
sb = root->fs_info->super_copy;
printf("Clearing log on %s, previous log_root %llu, level %u\n",
devname,
(unsigned long long)btrfs_super_log_root(sb),
(unsigned)btrfs_super_log_root_level(sb));
trans = btrfs_start_transaction(root, 1);
BUG_ON(IS_ERR(trans));
btrfs_set_super_log_root(sb, 0);
btrfs_set_super_log_root_level(sb, 0);
btrfs_commit_transaction(trans, root);
close_ctree(root);
out:
return !!ret;
}
int btrfs_clear_free_space_tree(struct btrfs_fs_info *fs_info)
{
struct btrfs_trans_handle *trans;
struct btrfs_root *tree_root = fs_info->tree_root;
struct btrfs_root *free_space_root = fs_info->free_space_root;
int ret;
trans = btrfs_start_transaction(tree_root, 0);
if (IS_ERR(trans))
return PTR_ERR(trans);
btrfs_clear_fs_compat_ro(fs_info, FREE_SPACE_TREE);
fs_info->free_space_root = NULL;
ret = clear_free_space_tree(trans, free_space_root);
if (ret)
goto abort;
ret = btrfs_del_root(trans, tree_root, &free_space_root->root_key);
if (ret)
goto abort;
list_del(&free_space_root->dirty_list);
btrfs_tree_lock(free_space_root->node);
clean_tree_block(trans, tree_root->fs_info, free_space_root->node);
btrfs_tree_unlock(free_space_root->node);
btrfs_free_tree_block(trans, free_space_root, free_space_root->node,
0, 1);
free_extent_buffer(free_space_root->node);
free_extent_buffer(free_space_root->commit_root);
kfree(free_space_root);
ret = btrfs_commit_transaction(trans, tree_root);
if (ret)
return ret;
return 0;
abort:
btrfs_abort_transaction(trans, ret);
btrfs_end_transaction(trans, tree_root);
return ret;
}
int main(int ac, char **av)
{
struct btrfs_root *root;
struct btrfs_trans_handle *trans;
struct btrfs_super_block *sb;
int ret;
set_argv0(av);
if (check_argc_exact(ac, 2))
print_usage();
radix_tree_init();
printf("WARNING: this utility is deprecated, please use 'btrfs rescue zero-log'\n\n");
if ((ret = check_mounted(av[1])) < 0) {
fprintf(stderr, "ERROR: could not check mount status: %s\n", strerror(-ret));
goto out;
} else if (ret) {
fprintf(stderr, "ERROR: %s is currently mounted\n", av[1]);
ret = -EBUSY;
goto out;
}
root = open_ctree(av[1], 0, OPEN_CTREE_WRITES | OPEN_CTREE_PARTIAL);
if (!root) {
fprintf(stderr, "ERROR: cannot open ctree\n");
return 1;
}
sb = root->fs_info->super_copy;
printf("Clearing log on %s, previous log_root %llu, level %u\n",
av[1],
(unsigned long long)btrfs_super_log_root(sb),
(unsigned)btrfs_super_log_root_level(sb));
trans = btrfs_start_transaction(root, 1);
btrfs_set_super_log_root(root->fs_info->super_copy, 0);
btrfs_set_super_log_root_level(root->fs_info->super_copy, 0);
btrfs_commit_transaction(trans, root);
close_ctree(root);
out:
return !!ret;
}
static int btrfs_uuid_iter_rem(struct btrfs_root *uuid_root, u8 *uuid, u8 type,
u64 subid)
{
struct btrfs_trans_handle *trans;
int ret;
/* 1 - for the uuid item */
trans = btrfs_start_transaction(uuid_root, 1);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
goto out;
}
ret = btrfs_uuid_tree_remove(trans, uuid, type, subid);
btrfs_end_transaction(trans);
out:
return ret;
}
static void change_label_unmounted(char *dev, char *nLabel)
{
struct btrfs_root *root;
struct btrfs_trans_handle *trans;
/* Open the super_block at the default location
* and as read-write.
*/
root = open_ctree(dev, 0, 1);
if (!root) /* errors are printed by open_ctree() */
return;
trans = btrfs_start_transaction(root, 1);
strncpy(root->fs_info->super_copy.label, nLabel, BTRFS_LABEL_SIZE);
root->fs_info->super_copy.label[BTRFS_LABEL_SIZE-1] = 0;
btrfs_commit_transaction(trans, root);
/* Now we close it since we are done. */
close_ctree(root);
}
static int make_image(char *source_dir, struct btrfs_root *root, int out_fd)
{
int ret;
struct btrfs_trans_handle *trans;
struct stat root_st;
struct directory_name_entry dir_head;
struct directory_name_entry *dir_entry = NULL;
ret = lstat(source_dir, &root_st);
if (ret) {
fprintf(stderr, "unable to lstat the %s\n", source_dir);
goto fail;
}
INIT_LIST_HEAD(&dir_head.list);
trans = btrfs_start_transaction(root, 1);
ret = traverse_directory(trans, root, source_dir, &dir_head, out_fd);
if (ret) {
fprintf(stderr, "unable to traverse_directory\n");
goto fail;
}
btrfs_commit_transaction(trans, root);
printf("Making image is completed.\n");
return 0;
fail:
while (!list_empty(&dir_head.list)) {
dir_entry = list_entry(dir_head.list.next,
struct directory_name_entry, list);
list_del(&dir_entry->list);
free(dir_entry);
}
fprintf(stderr, "Making image is aborted.\n");
return -1;
}
int btrfs_sync_fs(struct super_block *sb, int wait)
{
struct btrfs_trans_handle *trans;
struct btrfs_root *root;
int ret;
root = btrfs_sb(sb);
if (sb->s_flags & MS_RDONLY)
return 0;
sb->s_dirt = 0;
if (!wait) {
filemap_flush(root->fs_info->btree_inode->i_mapping);
return 0;
}
btrfs_start_delalloc_inodes(root);
btrfs_wait_ordered_extents(root, 0);
trans = btrfs_start_transaction(root, 1);
ret = btrfs_commit_transaction(trans, root);
sb->s_dirt = 0;
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;
}
static int make_root_dir(struct btrfs_root *root, int mixed)
{
struct btrfs_trans_handle *trans;
struct btrfs_key location;
u64 bytes_used;
u64 chunk_start = 0;
u64 chunk_size = 0;
int ret;
trans = btrfs_start_transaction(root, 1);
bytes_used = btrfs_super_bytes_used(root->fs_info->super_copy);
root->fs_info->system_allocs = 1;
ret = btrfs_make_block_group(trans, root, bytes_used,
BTRFS_BLOCK_GROUP_SYSTEM,
BTRFS_FIRST_CHUNK_TREE_OBJECTID,
0, BTRFS_MKFS_SYSTEM_GROUP_SIZE);
BUG_ON(ret);
if (mixed) {
ret = btrfs_alloc_chunk(trans, root->fs_info->extent_root,
&chunk_start, &chunk_size,
BTRFS_BLOCK_GROUP_METADATA |
BTRFS_BLOCK_GROUP_DATA);
if (ret == -ENOSPC) {
fprintf(stderr,
"no space to alloc data/metadata chunk\n");
goto err;
}
BUG_ON(ret);
ret = btrfs_make_block_group(trans, root, 0,
BTRFS_BLOCK_GROUP_METADATA |
BTRFS_BLOCK_GROUP_DATA,
BTRFS_FIRST_CHUNK_TREE_OBJECTID,
chunk_start, chunk_size);
BUG_ON(ret);
printf("Created a data/metadata chunk of size %llu\n", chunk_size);
} else {
ret = btrfs_alloc_chunk(trans, root->fs_info->extent_root,
&chunk_start, &chunk_size,
BTRFS_BLOCK_GROUP_METADATA);
if (ret == -ENOSPC) {
fprintf(stderr, "no space to alloc metadata chunk\n");
goto err;
}
BUG_ON(ret);
ret = btrfs_make_block_group(trans, root, 0,
BTRFS_BLOCK_GROUP_METADATA,
BTRFS_FIRST_CHUNK_TREE_OBJECTID,
chunk_start, chunk_size);
BUG_ON(ret);
}
root->fs_info->system_allocs = 0;
btrfs_commit_transaction(trans, root);
trans = btrfs_start_transaction(root, 1);
BUG_ON(!trans);
if (!mixed) {
ret = btrfs_alloc_chunk(trans, root->fs_info->extent_root,
&chunk_start, &chunk_size,
BTRFS_BLOCK_GROUP_DATA);
if (ret == -ENOSPC) {
fprintf(stderr, "no space to alloc data chunk\n");
goto err;
}
BUG_ON(ret);
ret = btrfs_make_block_group(trans, root, 0,
BTRFS_BLOCK_GROUP_DATA,
BTRFS_FIRST_CHUNK_TREE_OBJECTID,
chunk_start, chunk_size);
BUG_ON(ret);
}
ret = btrfs_make_root_dir(trans, root->fs_info->tree_root,
BTRFS_ROOT_TREE_DIR_OBJECTID);
if (ret)
goto err;
ret = btrfs_make_root_dir(trans, root, BTRFS_FIRST_FREE_OBJECTID);
if (ret)
goto err;
memcpy(&location, &root->fs_info->fs_root->root_key, sizeof(location));
location.offset = (u64)-1;
ret = btrfs_insert_dir_item(trans, root->fs_info->tree_root,
"default", 7,
btrfs_super_root_dir(root->fs_info->super_copy),
&location, BTRFS_FT_DIR, 0);
if (ret)
goto err;
ret = btrfs_insert_inode_ref(trans, root->fs_info->tree_root,
"default", 7, location.objectid,
BTRFS_ROOT_TREE_DIR_OBJECTID, 0);
if (ret)
goto err;
btrfs_commit_transaction(trans, root);
err:
return ret;
}
static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
int scrub_ret)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
struct btrfs_device *tgt_device;
struct btrfs_device *src_device;
struct btrfs_root *root = fs_info->tree_root;
u8 uuid_tmp[BTRFS_UUID_SIZE];
struct btrfs_trans_handle *trans;
int ret = 0;
/* don't allow cancel or unmount to disturb the finishing procedure */
mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
btrfs_dev_replace_lock(dev_replace);
/* was the operation canceled, or is it finished? */
if (dev_replace->replace_state !=
BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED) {
btrfs_dev_replace_unlock(dev_replace);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return 0;
}
tgt_device = dev_replace->tgtdev;
src_device = dev_replace->srcdev;
btrfs_dev_replace_unlock(dev_replace);
/* replace old device with new one in mapping tree */
if (!scrub_ret)
btrfs_dev_replace_update_device_in_mapping_tree(fs_info,
src_device,
tgt_device);
/*
* flush all outstanding I/O and inode extent mappings before the
* copy operation is declared as being finished
*/
ret = btrfs_start_all_delalloc_inodes(root->fs_info, 0);
if (ret) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return ret;
}
btrfs_wait_all_ordered_extents(root->fs_info, 0);
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return PTR_ERR(trans);
}
ret = btrfs_commit_transaction(trans, root);
WARN_ON(ret);
/* keep away write_all_supers() during the finishing procedure */
mutex_lock(&root->fs_info->fs_devices->device_list_mutex);
btrfs_dev_replace_lock(dev_replace);
dev_replace->replace_state =
scrub_ret ? BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED
: BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED;
dev_replace->tgtdev = NULL;
dev_replace->srcdev = NULL;
dev_replace->time_stopped = btrfs_get_seconds_since_1970();
dev_replace->item_needs_writeback = 1;
if (scrub_ret) {
printk_in_rcu(KERN_ERR
"btrfs: btrfs_scrub_dev(%s, %llu, %s) failed %d\n",
src_device->missing ? "<missing disk>" :
rcu_str_deref(src_device->name),
src_device->devid,
rcu_str_deref(tgt_device->name), scrub_ret);
btrfs_dev_replace_unlock(dev_replace);
mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return 0;
}
printk_in_rcu(KERN_INFO
"btrfs: dev_replace from %s (devid %llu) to %s) finished\n",
src_device->missing ? "<missing disk>" :
rcu_str_deref(src_device->name),
src_device->devid,
rcu_str_deref(tgt_device->name));
tgt_device->is_tgtdev_for_dev_replace = 0;
tgt_device->devid = src_device->devid;
src_device->devid = BTRFS_DEV_REPLACE_DEVID;
tgt_device->bytes_used = src_device->bytes_used;
memcpy(uuid_tmp, tgt_device->uuid, sizeof(uuid_tmp));
memcpy(tgt_device->uuid, src_device->uuid, sizeof(tgt_device->uuid));
memcpy(src_device->uuid, uuid_tmp, sizeof(src_device->uuid));
tgt_device->total_bytes = src_device->total_bytes;
tgt_device->disk_total_bytes = src_device->disk_total_bytes;
tgt_device->bytes_used = src_device->bytes_used;
if (fs_info->sb->s_bdev == src_device->bdev)
fs_info->sb->s_bdev = tgt_device->bdev;
if (fs_info->fs_devices->latest_bdev == src_device->bdev)
fs_info->fs_devices->latest_bdev = tgt_device->bdev;
list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
btrfs_rm_dev_replace_srcdev(fs_info, src_device);
if (src_device->bdev) {
/* zero out the old super */
btrfs_scratch_superblock(src_device);
}
/*
* this is again a consistent state where no dev_replace procedure
* is running, the target device is part of the filesystem, the
* source device is not part of the filesystem anymore and its 1st
* superblock is scratched out so that it is no longer marked to
* belong to this filesystem.
*/
btrfs_dev_replace_unlock(dev_replace);
mutex_unlock(&root->fs_info->fs_devices->device_list_mutex);
/* write back the superblocks */
trans = btrfs_start_transaction(root, 0);
if (!IS_ERR(trans))
btrfs_commit_transaction(trans, root);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return 0;
}
int btrfs_dev_replace_start(struct btrfs_root *root,
struct btrfs_ioctl_dev_replace_args *args)
{
struct btrfs_trans_handle *trans;
struct btrfs_fs_info *fs_info = root->fs_info;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
if (btrfs_fs_incompat(fs_info, RAID56)) {
pr_warn("btrfs: dev_replace cannot yet handle RAID5/RAID6\n");
return -EINVAL;
}
switch (args->start.cont_reading_from_srcdev_mode) {
case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_ALWAYS:
case BTRFS_IOCTL_DEV_REPLACE_CONT_READING_FROM_SRCDEV_MODE_AVOID:
break;
default:
return -EINVAL;
}
if ((args->start.srcdevid == 0 && args->start.srcdev_name[0] == '\0') ||
args->start.tgtdev_name[0] == '\0')
return -EINVAL;
mutex_lock(&fs_info->volume_mutex);
ret = btrfs_init_dev_replace_tgtdev(root, args->start.tgtdev_name,
&tgt_device);
if (ret) {
pr_err("btrfs: target device %s is invalid!\n",
args->start.tgtdev_name);
mutex_unlock(&fs_info->volume_mutex);
return -EINVAL;
}
ret = btrfs_dev_replace_find_srcdev(root, args->start.srcdevid,
args->start.srcdev_name,
&src_device);
mutex_unlock(&fs_info->volume_mutex);
if (ret) {
ret = -EINVAL;
goto leave_no_lock;
}
if (tgt_device->total_bytes < src_device->total_bytes) {
pr_err("btrfs: target device is smaller than source device!\n");
ret = -EINVAL;
goto leave_no_lock;
}
btrfs_dev_replace_lock(dev_replace);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
goto leave;
}
dev_replace->cont_reading_from_srcdev_mode =
args->start.cont_reading_from_srcdev_mode;
WARN_ON(!src_device);
dev_replace->srcdev = src_device;
WARN_ON(!tgt_device);
dev_replace->tgtdev = tgt_device;
printk_in_rcu(KERN_INFO
"btrfs: dev_replace from %s (devid %llu) to %s) started\n",
src_device->missing ? "<missing disk>" :
rcu_str_deref(src_device->name),
src_device->devid,
rcu_str_deref(tgt_device->name));
tgt_device->total_bytes = src_device->total_bytes;
tgt_device->disk_total_bytes = src_device->disk_total_bytes;
tgt_device->bytes_used = src_device->bytes_used;
/*
* from now on, the writes to the srcdev are all duplicated to
* go to the tgtdev as well (refer to btrfs_map_block()).
*/
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
dev_replace->time_started = btrfs_get_seconds_since_1970();
dev_replace->cursor_left = 0;
dev_replace->committed_cursor_left = 0;
dev_replace->cursor_left_last_write_of_item = 0;
dev_replace->cursor_right = 0;
dev_replace->is_valid = 1;
dev_replace->item_needs_writeback = 1;
args->result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
btrfs_dev_replace_unlock(dev_replace);
btrfs_wait_all_ordered_extents(root->fs_info, 0);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
btrfs_dev_replace_lock(dev_replace);
goto leave;
}
ret = btrfs_commit_transaction(trans, root);
WARN_ON(ret);
/* the disk copy procedure reuses the scrub code */
ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
src_device->total_bytes,
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(root->fs_info, ret);
WARN_ON(ret);
return 0;
leave:
dev_replace->srcdev = NULL;
dev_replace->tgtdev = NULL;
btrfs_dev_replace_unlock(dev_replace);
leave_no_lock:
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(fs_info, tgt_device);
return ret;
}
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;
}
static int btrfs_dev_replace_start(struct btrfs_fs_info *fs_info,
const char *tgtdev_name, u64 srcdevid, const char *srcdev_name,
int read_src)
{
struct btrfs_root *root = fs_info->dev_root;
struct btrfs_trans_handle *trans;
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
int ret;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
bool need_unlock;
src_device = btrfs_find_device_by_devspec(fs_info, srcdevid,
srcdev_name);
if (IS_ERR(src_device))
return PTR_ERR(src_device);
if (btrfs_pinned_by_swapfile(fs_info, src_device)) {
btrfs_warn_in_rcu(fs_info,
"cannot replace device %s (devid %llu) due to active swapfile",
btrfs_dev_name(src_device), src_device->devid);
return -ETXTBSY;
}
ret = btrfs_init_dev_replace_tgtdev(fs_info, tgtdev_name,
src_device, &tgt_device);
if (ret)
return ret;
/*
* Here we commit the transaction to make sure commit_total_bytes
* of all the devices are updated.
*/
trans = btrfs_attach_transaction(root);
if (!IS_ERR(trans)) {
ret = btrfs_commit_transaction(trans);
if (ret)
return ret;
} else if (PTR_ERR(trans) != -ENOENT) {
return PTR_ERR(trans);
}
need_unlock = true;
down_write(&dev_replace->rwsem);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
ASSERT(0);
ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_ALREADY_STARTED;
goto leave;
}
dev_replace->cont_reading_from_srcdev_mode = read_src;
WARN_ON(!src_device);
dev_replace->srcdev = src_device;
dev_replace->tgtdev = tgt_device;
btrfs_info_in_rcu(fs_info,
"dev_replace from %s (devid %llu) to %s started",
btrfs_dev_name(src_device),
src_device->devid,
rcu_str_deref(tgt_device->name));
/*
* from now on, the writes to the srcdev are all duplicated to
* go to the tgtdev as well (refer to btrfs_map_block()).
*/
dev_replace->replace_state = BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED;
dev_replace->time_started = ktime_get_real_seconds();
dev_replace->cursor_left = 0;
dev_replace->committed_cursor_left = 0;
dev_replace->cursor_left_last_write_of_item = 0;
dev_replace->cursor_right = 0;
dev_replace->is_valid = 1;
dev_replace->item_needs_writeback = 1;
atomic64_set(&dev_replace->num_write_errors, 0);
atomic64_set(&dev_replace->num_uncorrectable_read_errors, 0);
up_write(&dev_replace->rwsem);
need_unlock = false;
ret = btrfs_sysfs_add_device_link(tgt_device->fs_devices, tgt_device);
if (ret)
btrfs_err(fs_info, "kobj add dev failed %d", ret);
btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
/* force writing the updated state information to disk */
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
ret = PTR_ERR(trans);
need_unlock = true;
down_write(&dev_replace->rwsem);
dev_replace->replace_state =
BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED;
dev_replace->srcdev = NULL;
dev_replace->tgtdev = NULL;
goto leave;
}
ret = btrfs_commit_transaction(trans);
WARN_ON(ret);
/* the disk copy procedure reuses the scrub code */
ret = btrfs_scrub_dev(fs_info, src_device->devid, 0,
btrfs_device_get_total_bytes(src_device),
&dev_replace->scrub_progress, 0, 1);
ret = btrfs_dev_replace_finishing(fs_info, ret);
if (ret == -EINPROGRESS) {
ret = BTRFS_IOCTL_DEV_REPLACE_RESULT_SCRUB_INPROGRESS;
} else if (ret != -ECANCELED) {
WARN_ON(ret);
}
return ret;
leave:
if (need_unlock)
up_write(&dev_replace->rwsem);
btrfs_destroy_dev_replace_tgtdev(tgt_device);
return ret;
}
static int btrfs_dev_replace_finishing(struct btrfs_fs_info *fs_info,
int scrub_ret)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
struct btrfs_device *tgt_device;
struct btrfs_device *src_device;
struct btrfs_root *root = fs_info->tree_root;
u8 uuid_tmp[BTRFS_UUID_SIZE];
struct btrfs_trans_handle *trans;
int ret = 0;
/* don't allow cancel or unmount to disturb the finishing procedure */
mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
down_read(&dev_replace->rwsem);
/* was the operation canceled, or is it finished? */
if (dev_replace->replace_state !=
BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED) {
up_read(&dev_replace->rwsem);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return 0;
}
tgt_device = dev_replace->tgtdev;
src_device = dev_replace->srcdev;
up_read(&dev_replace->rwsem);
/*
* flush all outstanding I/O and inode extent mappings before the
* copy operation is declared as being finished
*/
ret = btrfs_start_delalloc_roots(fs_info, -1);
if (ret) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return ret;
}
btrfs_wait_ordered_roots(fs_info, U64_MAX, 0, (u64)-1);
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return PTR_ERR(trans);
}
ret = btrfs_commit_transaction(trans);
WARN_ON(ret);
/* keep away write_all_supers() during the finishing procedure */
mutex_lock(&fs_info->fs_devices->device_list_mutex);
mutex_lock(&fs_info->chunk_mutex);
down_write(&dev_replace->rwsem);
dev_replace->replace_state =
scrub_ret ? BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED
: BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED;
dev_replace->tgtdev = NULL;
dev_replace->srcdev = NULL;
dev_replace->time_stopped = ktime_get_real_seconds();
dev_replace->item_needs_writeback = 1;
/* replace old device with new one in mapping tree */
if (!scrub_ret) {
btrfs_dev_replace_update_device_in_mapping_tree(fs_info,
src_device,
tgt_device);
} else {
if (scrub_ret != -ECANCELED)
btrfs_err_in_rcu(fs_info,
"btrfs_scrub_dev(%s, %llu, %s) failed %d",
btrfs_dev_name(src_device),
src_device->devid,
rcu_str_deref(tgt_device->name), scrub_ret);
up_write(&dev_replace->rwsem);
mutex_unlock(&fs_info->chunk_mutex);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
btrfs_rm_dev_replace_blocked(fs_info);
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(tgt_device);
btrfs_rm_dev_replace_unblocked(fs_info);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return scrub_ret;
}
btrfs_info_in_rcu(fs_info,
"dev_replace from %s (devid %llu) to %s finished",
btrfs_dev_name(src_device),
src_device->devid,
rcu_str_deref(tgt_device->name));
clear_bit(BTRFS_DEV_STATE_REPLACE_TGT, &tgt_device->dev_state);
tgt_device->devid = src_device->devid;
src_device->devid = BTRFS_DEV_REPLACE_DEVID;
memcpy(uuid_tmp, tgt_device->uuid, sizeof(uuid_tmp));
memcpy(tgt_device->uuid, src_device->uuid, sizeof(tgt_device->uuid));
memcpy(src_device->uuid, uuid_tmp, sizeof(src_device->uuid));
btrfs_device_set_total_bytes(tgt_device, src_device->total_bytes);
btrfs_device_set_disk_total_bytes(tgt_device,
src_device->disk_total_bytes);
btrfs_device_set_bytes_used(tgt_device, src_device->bytes_used);
ASSERT(list_empty(&src_device->resized_list));
tgt_device->commit_total_bytes = src_device->commit_total_bytes;
tgt_device->commit_bytes_used = src_device->bytes_used;
btrfs_assign_next_active_device(src_device, tgt_device);
list_add(&tgt_device->dev_alloc_list, &fs_info->fs_devices->alloc_list);
fs_info->fs_devices->rw_devices++;
up_write(&dev_replace->rwsem);
btrfs_rm_dev_replace_blocked(fs_info);
btrfs_rm_dev_replace_remove_srcdev(src_device);
btrfs_rm_dev_replace_unblocked(fs_info);
/*
* Increment dev_stats_ccnt so that btrfs_run_dev_stats() will
* update on-disk dev stats value during commit transaction
*/
atomic_inc(&tgt_device->dev_stats_ccnt);
/*
* this is again a consistent state where no dev_replace procedure
* is running, the target device is part of the filesystem, the
* source device is not part of the filesystem anymore and its 1st
* superblock is scratched out so that it is no longer marked to
* belong to this filesystem.
*/
mutex_unlock(&fs_info->chunk_mutex);
mutex_unlock(&fs_info->fs_devices->device_list_mutex);
/* replace the sysfs entry */
btrfs_sysfs_rm_device_link(fs_info->fs_devices, src_device);
btrfs_rm_dev_replace_free_srcdev(fs_info, src_device);
/* write back the superblocks */
trans = btrfs_start_transaction(root, 0);
if (!IS_ERR(trans))
btrfs_commit_transaction(trans);
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return 0;
}
int btrfs_dev_replace_cancel(struct btrfs_fs_info *fs_info)
{
struct btrfs_dev_replace *dev_replace = &fs_info->dev_replace;
struct btrfs_device *tgt_device = NULL;
struct btrfs_device *src_device = NULL;
struct btrfs_trans_handle *trans;
struct btrfs_root *root = fs_info->tree_root;
int result;
int ret;
if (sb_rdonly(fs_info->sb))
return -EROFS;
mutex_lock(&dev_replace->lock_finishing_cancel_unmount);
down_write(&dev_replace->rwsem);
switch (dev_replace->replace_state) {
case BTRFS_IOCTL_DEV_REPLACE_STATE_NEVER_STARTED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_FINISHED:
case BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED:
result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED;
up_write(&dev_replace->rwsem);
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_STARTED:
tgt_device = dev_replace->tgtdev;
src_device = dev_replace->srcdev;
up_write(&dev_replace->rwsem);
ret = btrfs_scrub_cancel(fs_info);
if (ret < 0) {
result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NOT_STARTED;
} else {
result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
/*
* btrfs_dev_replace_finishing() will handle the
* cleanup part
*/
btrfs_info_in_rcu(fs_info,
"dev_replace from %s (devid %llu) to %s canceled",
btrfs_dev_name(src_device), src_device->devid,
btrfs_dev_name(tgt_device));
}
break;
case BTRFS_IOCTL_DEV_REPLACE_STATE_SUSPENDED:
/*
* Scrub doing the replace isn't running so we need to do the
* cleanup step of btrfs_dev_replace_finishing() here
*/
result = BTRFS_IOCTL_DEV_REPLACE_RESULT_NO_ERROR;
tgt_device = dev_replace->tgtdev;
src_device = dev_replace->srcdev;
dev_replace->tgtdev = NULL;
dev_replace->srcdev = NULL;
dev_replace->replace_state =
BTRFS_IOCTL_DEV_REPLACE_STATE_CANCELED;
dev_replace->time_stopped = ktime_get_real_seconds();
dev_replace->item_needs_writeback = 1;
up_write(&dev_replace->rwsem);
/* Scrub for replace must not be running in suspended state */
ret = btrfs_scrub_cancel(fs_info);
ASSERT(ret != -ENOTCONN);
trans = btrfs_start_transaction(root, 0);
if (IS_ERR(trans)) {
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return PTR_ERR(trans);
}
ret = btrfs_commit_transaction(trans);
WARN_ON(ret);
btrfs_info_in_rcu(fs_info,
"suspended dev_replace from %s (devid %llu) to %s canceled",
btrfs_dev_name(src_device), src_device->devid,
btrfs_dev_name(tgt_device));
if (tgt_device)
btrfs_destroy_dev_replace_tgtdev(tgt_device);
break;
default:
up_write(&dev_replace->rwsem);
result = -EINVAL;
}
mutex_unlock(&dev_replace->lock_finishing_cancel_unmount);
return result;
}
