mutex_lock(&root->log_mutex);
if (root->log_root) {
+ if (!root->log_start_pid) {
+ root->log_start_pid = current->pid;
+ root->log_multiple_pids = false;
+ } else if (root->log_start_pid != current->pid) {
+ root->log_multiple_pids = true;
+ }
+
root->log_batch++;
atomic_inc(&root->log_writers);
mutex_unlock(&root->log_mutex);
return 0;
}
+ root->log_multiple_pids = false;
+ root->log_start_pid = current->pid;
mutex_lock(&root->fs_info->tree_log_mutex);
if (!root->fs_info->log_root_tree) {
ret = btrfs_init_log_root_tree(trans, root->fs_info);
struct walk_control *wc, u64 gen)
{
if (wc->pin)
- btrfs_update_pinned_extents(log->fs_info->extent_root,
- eb->start, eb->len, 1);
+ btrfs_pin_extent(log->fs_info->extent_root,
+ eb->start, eb->len, 0);
if (btrfs_buffer_uptodate(eb, gen)) {
if (wc->write)
static noinline struct inode *read_one_inode(struct btrfs_root *root,
u64 objectid)
{
+ struct btrfs_key key;
struct inode *inode;
- inode = btrfs_iget_locked(root->fs_info->sb, objectid, root);
- if (inode->i_state & I_NEW) {
- BTRFS_I(inode)->root = root;
- BTRFS_I(inode)->location.objectid = objectid;
- BTRFS_I(inode)->location.type = BTRFS_INODE_ITEM_KEY;
- BTRFS_I(inode)->location.offset = 0;
- btrfs_read_locked_inode(inode);
- unlock_new_inode(inode);
- }
- if (is_bad_inode(inode)) {
+ key.objectid = objectid;
+ key.type = BTRFS_INODE_ITEM_KEY;
+ key.offset = 0;
+ inode = btrfs_iget(root->fs_info->sb, &key, root);
+ if (IS_ERR(inode)) {
+ inode = NULL;
+ } else if (is_bad_inode(inode)) {
iput(inode);
inode = NULL;
}
saved_nbytes = inode_get_bytes(inode);
/* drop any overlapping extents */
- ret = btrfs_drop_extents(trans, root, inode,
- start, extent_end, extent_end, start, &alloc_hint);
+ ret = btrfs_drop_extents(trans, inode, start, extent_end,
+ &alloc_hint, 1);
BUG_ON(ret);
if (found_type == BTRFS_FILE_EXTENT_REG ||
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
+ u64 offset;
unsigned long dest_offset;
struct btrfs_key ins;
ins.objectid = btrfs_file_extent_disk_bytenr(eb, item);
ins.offset = btrfs_file_extent_disk_num_bytes(eb, item);
ins.type = BTRFS_EXTENT_ITEM_KEY;
+ offset = key->offset - btrfs_file_extent_offset(eb, item);
if (ins.objectid > 0) {
u64 csum_start;
if (ret == 0) {
ret = btrfs_inc_extent_ref(trans, root,
ins.objectid, ins.offset,
- path->nodes[0]->start,
- root->root_key.objectid,
- trans->transid, key->objectid);
+ 0, root->root_key.objectid,
+ key->objectid, offset);
} else {
/*
* insert the extent pointer in the extent
* allocation tree
*/
- ret = btrfs_alloc_logged_extent(trans, root,
- path->nodes[0]->start,
- root->root_key.objectid,
- trans->transid, key->objectid,
- &ins);
+ ret = btrfs_alloc_logged_file_extent(trans,
+ root, root->root_key.objectid,
+ key->objectid, offset, &ins);
BUG_ON(ret);
}
btrfs_release_path(root, path);
return -ENOENT;
inode = read_one_inode(root, key->objectid);
- BUG_ON(!dir);
+ BUG_ON(!inode);
ref_ptr = btrfs_item_ptr_offset(eb, slot);
ref_end = ref_ptr + btrfs_item_size_nr(eb, slot);
return 0;
}
+static int insert_orphan_item(struct btrfs_trans_handle *trans,
+ struct btrfs_root *root, u64 offset)
+{
+ int ret;
+ ret = btrfs_find_orphan_item(root, offset);
+ if (ret > 0)
+ ret = btrfs_insert_orphan_item(trans, root, offset);
+ return ret;
+}
+
+
/*
* There are a few corners where the link count of the file can't
* be properly maintained during replay. So, instead of adding
}
BTRFS_I(inode)->index_cnt = (u64)-1;
- if (inode->i_nlink == 0 && S_ISDIR(inode->i_mode)) {
- ret = replay_dir_deletes(trans, root, NULL, path,
- inode->i_ino, 1);
+ if (inode->i_nlink == 0) {
+ if (S_ISDIR(inode->i_mode)) {
+ ret = replay_dir_deletes(trans, root, NULL, path,
+ inode->i_ino, 1);
+ BUG_ON(ret);
+ }
+ ret = insert_orphan_item(trans, root, inode->i_ino);
BUG_ON(ret);
}
btrfs_free_path(path);
/* inode keys are done during the first stage */
if (key.type == BTRFS_INODE_ITEM_KEY &&
wc->stage == LOG_WALK_REPLAY_INODES) {
- struct inode *inode;
struct btrfs_inode_item *inode_item;
u32 mode;
eb, i, &key);
BUG_ON(ret);
- /* for regular files, truncate away
- * extents past the new EOF
+ /* for regular files, make sure corresponding
+ * orhpan item exist. extents past the new EOF
+ * will be truncated later by orphan cleanup.
*/
if (S_ISREG(mode)) {
- inode = read_one_inode(root,
- key.objectid);
- BUG_ON(!inode);
-
- ret = btrfs_truncate_inode_items(wc->trans,
- root, inode, inode->i_size,
- BTRFS_EXTENT_DATA_KEY);
+ ret = insert_orphan_item(wc->trans, root,
+ key.objectid);
BUG_ON(ret);
-
- /* if the nlink count is zero here, the iput
- * will free the inode. We bump it to make
- * sure it doesn't get freed until the link
- * count fixup is done
- */
- if (inode->i_nlink == 0) {
- btrfs_inc_nlink(inode);
- btrfs_update_inode(wc->trans,
- root, inode);
- }
- iput(inode);
}
+
ret = link_to_fixup_dir(wc->trans, root,
path, key.objectid);
BUG_ON(ret);
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
- ret = btrfs_drop_leaf_ref(trans, root, next);
- BUG_ON(ret);
-
WARN_ON(root_owner !=
BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_free_reserved_extent(root,
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
- if (*level == 0) {
- ret = btrfs_drop_leaf_ref(trans, root, next);
- BUG_ON(ret);
- }
WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_free_reserved_extent(root, bytenr, blocksize);
BUG_ON(ret);
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
- if (*level == 0) {
- ret = btrfs_drop_leaf_ref(trans, root,
- next);
- BUG_ON(ret);
- }
-
WARN_ON(root_owner != BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_free_reserved_extent(root,
path->nodes[*level]->start,
btrfs_wait_tree_block_writeback(next);
btrfs_tree_unlock(next);
- if (orig_level == 0) {
- ret = btrfs_drop_leaf_ref(trans, log,
- next);
- BUG_ON(ret);
- }
WARN_ON(log->root_key.objectid !=
BTRFS_TREE_LOG_OBJECTID);
ret = btrfs_free_reserved_extent(log, next->start,
{
int index1;
int index2;
+ int mark;
int ret;
struct btrfs_root *log = root->log_root;
struct btrfs_root *log_root_tree = root->fs_info->log_root_tree;
+ unsigned long log_transid = 0;
mutex_lock(&root->log_mutex);
index1 = root->log_transid % 2;
while (1) {
unsigned long batch = root->log_batch;
- mutex_unlock(&root->log_mutex);
- schedule_timeout_uninterruptible(1);
- mutex_lock(&root->log_mutex);
-
+ if (root->log_multiple_pids) {
+ mutex_unlock(&root->log_mutex);
+ schedule_timeout_uninterruptible(1);
+ mutex_lock(&root->log_mutex);
+ }
wait_for_writer(trans, root);
if (batch == root->log_batch)
break;
goto out;
}
- ret = btrfs_write_and_wait_marked_extents(log, &log->dirty_log_pages);
+ log_transid = root->log_transid;
+ if (log_transid % 2 == 0)
+ mark = EXTENT_DIRTY;
+ else
+ mark = EXTENT_NEW;
+
+ /* we start IO on all the marked extents here, but we don't actually
+ * wait for them until later.
+ */
+ ret = btrfs_write_marked_extents(log, &log->dirty_log_pages, mark);
BUG_ON(ret);
- btrfs_set_root_bytenr(&log->root_item, log->node->start);
- btrfs_set_root_generation(&log->root_item, trans->transid);
- btrfs_set_root_level(&log->root_item, btrfs_header_level(log->node));
+ btrfs_set_root_node(&log->root_item, log->node);
root->log_batch = 0;
root->log_transid++;
log->log_transid = root->log_transid;
+ root->log_start_pid = 0;
smp_mb();
/*
- * log tree has been flushed to disk, new modifications of
- * the log will be written to new positions. so it's safe to
- * allow log writers to go in.
+ * IO has been started, blocks of the log tree have WRITTEN flag set
+ * in their headers. new modifications of the log will be written to
+ * new positions. so it's safe to allow log writers to go in.
*/
mutex_unlock(&root->log_mutex);
index2 = log_root_tree->log_transid % 2;
if (atomic_read(&log_root_tree->log_commit[index2])) {
+ btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
wait_log_commit(trans, log_root_tree,
log_root_tree->log_transid);
mutex_unlock(&log_root_tree->log_mutex);
* check the full commit flag again
*/
if (root->fs_info->last_trans_log_full_commit == trans->transid) {
+ btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
mutex_unlock(&log_root_tree->log_mutex);
ret = -EAGAIN;
goto out_wake_log_root;
}
ret = btrfs_write_and_wait_marked_extents(log_root_tree,
- &log_root_tree->dirty_log_pages);
+ &log_root_tree->dirty_log_pages,
+ EXTENT_DIRTY | EXTENT_NEW);
BUG_ON(ret);
+ btrfs_wait_marked_extents(log, &log->dirty_log_pages, mark);
btrfs_set_super_log_root(&root->fs_info->super_for_commit,
log_root_tree->node->start);
* the running transaction open, so a full commit can't hop
* in and cause problems either.
*/
- write_ctree_super(trans, root->fs_info->tree_root, 2);
+ write_ctree_super(trans, root->fs_info->tree_root, 1);
ret = 0;
+ mutex_lock(&root->log_mutex);
+ if (root->last_log_commit < log_transid)
+ root->last_log_commit = log_transid;
+ mutex_unlock(&root->log_mutex);
+
out_wake_log_root:
atomic_set(&log_root_tree->log_commit[index2], 0);
smp_mb();
while (1) {
ret = find_first_extent_bit(&log->dirty_log_pages,
- 0, &start, &end, EXTENT_DIRTY);
+ 0, &start, &end, EXTENT_DIRTY | EXTENT_NEW);
if (ret)
break;
- clear_extent_dirty(&log->dirty_log_pages,
- start, end, GFP_NOFS);
+ clear_extent_bits(&log->dirty_log_pages, start, end,
+ EXTENT_DIRTY | EXTENT_NEW, GFP_NOFS);
}
if (log->log_transid > 0) {
ins_keys, ins_sizes, nr);
BUG_ON(ret);
- for (i = 0; i < nr; i++) {
+ for (i = 0; i < nr; i++, dst_path->slots[0]++) {
dst_offset = btrfs_item_ptr_offset(dst_path->nodes[0],
dst_path->slots[0]);
found_type = btrfs_file_extent_type(src, extent);
if (found_type == BTRFS_FILE_EXTENT_REG ||
found_type == BTRFS_FILE_EXTENT_PREALLOC) {
- u64 ds = btrfs_file_extent_disk_bytenr(src,
- extent);
- u64 dl = btrfs_file_extent_disk_num_bytes(src,
- extent);
- u64 cs = btrfs_file_extent_offset(src, extent);
- u64 cl = btrfs_file_extent_num_bytes(src,
- extent);;
+ u64 ds, dl, cs, cl;
+ ds = btrfs_file_extent_disk_bytenr(src,
+ extent);
+ /* ds == 0 is a hole */
+ if (ds == 0)
+ continue;
+
+ dl = btrfs_file_extent_disk_num_bytes(src,
+ extent);
+ cs = btrfs_file_extent_offset(src, extent);
+ cl = btrfs_file_extent_num_bytes(src,
+ extent);
if (btrfs_file_extent_compression(src,
extent)) {
cs = 0;
cl = dl;
}
- /* ds == 0 is a hole */
- if (ds != 0) {
- ret = btrfs_inc_extent_ref(trans, log,
- ds, dl,
- dst_path->nodes[0]->start,
- BTRFS_TREE_LOG_OBJECTID,
- trans->transid,
- ins_keys[i].objectid);
- BUG_ON(ret);
- ret = btrfs_lookup_csums_range(
- log->fs_info->csum_root,
- ds + cs, ds + cs + cl - 1,
- &ordered_sums);
- BUG_ON(ret);
- }
+
+ ret = btrfs_lookup_csums_range(
+ log->fs_info->csum_root,
+ ds + cs, ds + cs + cl - 1,
+ &ordered_sums);
+ BUG_ON(ret);
}
}
- dst_path->slots[0]++;
}
btrfs_mark_buffer_dirty(dst_path->nodes[0]);
if (!parent || !parent->d_inode || sb != parent->d_inode->i_sb)
break;
- if (parent == sb->s_root)
+ if (IS_ROOT(parent))
break;
parent = parent->d_parent;
return ret;
}
+static int inode_in_log(struct btrfs_trans_handle *trans,
+ struct inode *inode)
+{
+ struct btrfs_root *root = BTRFS_I(inode)->root;
+ int ret = 0;
+
+ mutex_lock(&root->log_mutex);
+ if (BTRFS_I(inode)->logged_trans == trans->transid &&
+ BTRFS_I(inode)->last_sub_trans <= root->last_log_commit)
+ ret = 1;
+ mutex_unlock(&root->log_mutex);
+ return ret;
+}
+
+
/*
* helper function around btrfs_log_inode to make sure newly created
* parent directories also end up in the log. A minimal inode and backref
goto end_no_trans;
}
+ if (root != BTRFS_I(inode)->root ||
+ btrfs_root_refs(&root->root_item) == 0) {
+ ret = 1;
+ goto end_no_trans;
+ }
+
ret = check_parent_dirs_for_sync(trans, inode, parent,
sb, last_committed);
if (ret)
goto end_no_trans;
+ if (inode_in_log(trans, inode)) {
+ ret = BTRFS_NO_LOG_SYNC;
+ goto end_no_trans;
+ }
+
start_log_trans(trans, root);
ret = btrfs_log_inode(trans, root, inode, inode_only);
break;
inode = parent->d_inode;
+ if (root != BTRFS_I(inode)->root)
+ break;
+
if (BTRFS_I(inode)->generation >
root->fs_info->last_trans_committed) {
ret = btrfs_log_inode(trans, root, inode, inode_only);
BUG_ON(ret);
}
- if (parent == sb->s_root)
+ if (IS_ROOT(parent))
break;
parent = parent->d_parent;
struct btrfs_key tmp_key;
struct btrfs_root *log;
struct btrfs_fs_info *fs_info = log_root_tree->fs_info;
- u64 highest_inode;
struct walk_control wc = {
.process_func = process_one_buffer,
.stage = 0,
BUG_ON(!wc.replay_dest);
wc.replay_dest->log_root = log;
- mutex_lock(&fs_info->trans_mutex);
- btrfs_record_root_in_trans(wc.replay_dest);
- mutex_unlock(&fs_info->trans_mutex);
+ btrfs_record_root_in_trans(trans, wc.replay_dest);
ret = walk_log_tree(trans, log, &wc);
BUG_ON(ret);
path);
BUG_ON(ret);
}
- ret = btrfs_find_highest_inode(wc.replay_dest, &highest_inode);
- if (ret == 0) {
- wc.replay_dest->highest_inode = highest_inode;
- wc.replay_dest->last_inode_alloc = highest_inode;
- }
key.offset = found_key.offset - 1;
wc.replay_dest->log_root = NULL;
free_extent_buffer(log->node);
+ free_extent_buffer(log->commit_root);
kfree(log);
if (found_key.offset == 0)
git://ftp.safe.ca / safe / jmp / linux-2.6 / blobdiff