}
}
+static void clear_btree_io_tree(struct extent_io_tree *tree)
+{
+ spin_lock(&tree->lock);
+ while (!RB_EMPTY_ROOT(&tree->state)) {
+ struct rb_node *node;
+ struct extent_state *state;
+
+ node = rb_first(&tree->state);
+ state = rb_entry(node, struct extent_state, rb_node);
+ rb_erase(&state->rb_node, &tree->state);
+ RB_CLEAR_NODE(&state->rb_node);
+ /*
+ * btree io trees aren't supposed to have tasks waiting for
+ * changes in the flags of extent states ever.
+ */
+ ASSERT(!waitqueue_active(&state->wq));
+ free_extent_state(state);
+ if (need_resched()) {
+ spin_unlock(&tree->lock);
+ cond_resched();
+ spin_lock(&tree->lock);
+ }
+ }
+ spin_unlock(&tree->lock);
+}
+
static noinline void switch_commit_roots(struct btrfs_transaction *trans,
struct btrfs_fs_info *fs_info)
{
root->commit_root = btrfs_root_node(root);
if (is_fstree(root->objectid))
btrfs_unpin_free_ino(root);
+ clear_btree_io_tree(&root->dirty_log_pages);
}
up_write(&fs_info->commit_root_sem);
}
* commit the transaction.
*/
atomic_set(&cur_trans->use_count, 2);
+ cur_trans->have_free_bgs = 0;
cur_trans->start_time = get_seconds();
cur_trans->delayed_refs.href_root = RB_ROOT;
INIT_LIST_HEAD(&cur_trans->pending_snapshots);
INIT_LIST_HEAD(&cur_trans->pending_chunks);
INIT_LIST_HEAD(&cur_trans->switch_commits);
+ INIT_LIST_HEAD(&cur_trans->pending_ordered);
+ INIT_LIST_HEAD(&cur_trans->dirty_bgs);
+ spin_lock_init(&cur_trans->dirty_bgs_lock);
list_add_tail(&cur_trans->list, &fs_info->trans_list);
extent_io_tree_init(&cur_trans->dirty_pages,
fs_info->btree_inode->i_mapping);
int ret;
/* Send isn't supposed to start transactions. */
- ASSERT(current->journal_info != (void *)BTRFS_SEND_TRANS_STUB);
+ ASSERT(current->journal_info != BTRFS_SEND_TRANS_STUB);
if (test_bit(BTRFS_FS_STATE_ERROR, &root->fs_info->fs_state))
return ERR_PTR(-EROFS);
if (num_items > 0 && root != root->fs_info->chunk_root) {
if (root->fs_info->quota_enabled &&
is_fstree(root->root_key.objectid)) {
- qgroup_reserved = num_items * root->leafsize;
+ qgroup_reserved = num_items * root->nodesize;
ret = btrfs_qgroup_reserve(root, qgroup_reserved);
if (ret)
return ERR_PTR(ret);
/*
* Do the reservation for the relocation root creation
*/
- if (unlikely(need_reserve_reloc_root(root))) {
+ if (need_reserve_reloc_root(root)) {
num_bytes += root->nodesize;
reloc_reserved = true;
}
h->sync = false;
INIT_LIST_HEAD(&h->qgroup_ref_list);
INIT_LIST_HEAD(&h->new_bgs);
+ INIT_LIST_HEAD(&h->ordered);
smp_mb();
if (cur_trans->state >= TRANS_STATE_BLOCKED &&
if (transid <= root->fs_info->last_trans_committed)
goto out;
- ret = -EINVAL;
/* find specified transaction */
spin_lock(&root->fs_info->trans_lock);
list_for_each_entry(t, &root->fs_info->trans_list, list) {
}
}
spin_unlock(&root->fs_info->trans_lock);
- /* The specified transaction doesn't exist */
- if (!cur_trans)
+
+ /*
+ * The specified transaction doesn't exist, or we
+ * raced with btrfs_commit_transaction
+ */
+ if (!cur_trans) {
+ if (transid > root->fs_info->last_trans_committed)
+ ret = -EINVAL;
goto out;
+ }
} else {
/* find newest transaction that is committing | committed */
spin_lock(&root->fs_info->trans_lock);
if (!list_empty(&trans->new_bgs))
btrfs_create_pending_block_groups(trans, root);
+ if (!list_empty(&trans->ordered)) {
+ spin_lock(&info->trans_lock);
+ list_splice(&trans->ordered, &cur_trans->pending_ordered);
+ spin_unlock(&info->trans_lock);
+ }
+
trans->delayed_ref_updates = 0;
if (!trans->sync) {
must_run_delayed_refs =
while (!find_first_extent_bit(dirty_pages, start, &start, &end,
mark, &cached_state)) {
- convert_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT,
- mark, &cached_state, GFP_NOFS);
- cached_state = NULL;
- err = filemap_fdatawrite_range(mapping, start, end);
+ bool wait_writeback = false;
+
+ err = convert_extent_bit(dirty_pages, start, end,
+ EXTENT_NEED_WAIT,
+ mark, &cached_state, GFP_NOFS);
+ /*
+ * convert_extent_bit can return -ENOMEM, which is most of the
+ * time a temporary error. So when it happens, ignore the error
+ * and wait for writeback of this range to finish - because we
+ * failed to set the bit EXTENT_NEED_WAIT for the range, a call
+ * to btrfs_wait_marked_extents() would not know that writeback
+ * for this range started and therefore wouldn't wait for it to
+ * finish - we don't want to commit a superblock that points to
+ * btree nodes/leafs for which writeback hasn't finished yet
+ * (and without errors).
+ * We cleanup any entries left in the io tree when committing
+ * the transaction (through clear_btree_io_tree()).
+ */
+ if (err == -ENOMEM) {
+ err = 0;
+ wait_writeback = true;
+ }
+ if (!err)
+ err = filemap_fdatawrite_range(mapping, start, end);
if (err)
werr = err;
+ else if (wait_writeback)
+ werr = filemap_fdatawait_range(mapping, start, end);
+ free_extent_state(cached_state);
+ cached_state = NULL;
cond_resched();
start = end + 1;
}
- if (err)
- werr = err;
return werr;
}
struct extent_state *cached_state = NULL;
u64 start = 0;
u64 end;
+ struct btrfs_inode *btree_ino = BTRFS_I(root->fs_info->btree_inode);
+ bool errors = false;
while (!find_first_extent_bit(dirty_pages, start, &start, &end,
EXTENT_NEED_WAIT, &cached_state)) {
- clear_extent_bit(dirty_pages, start, end, EXTENT_NEED_WAIT,
- 0, 0, &cached_state, GFP_NOFS);
- err = filemap_fdatawait_range(mapping, start, end);
+ /*
+ * Ignore -ENOMEM errors returned by clear_extent_bit().
+ * When committing the transaction, we'll remove any entries
+ * left in the io tree. For a log commit, we don't remove them
+ * after committing the log because the tree can be accessed
+ * concurrently - we do it only at transaction commit time when
+ * it's safe to do it (through clear_btree_io_tree()).
+ */
+ err = clear_extent_bit(dirty_pages, start, end,
+ EXTENT_NEED_WAIT,
+ 0, 0, &cached_state, GFP_NOFS);
+ if (err == -ENOMEM)
+ err = 0;
+ if (!err)
+ err = filemap_fdatawait_range(mapping, start, end);
if (err)
werr = err;
+ free_extent_state(cached_state);
+ cached_state = NULL;
cond_resched();
start = end + 1;
}
if (err)
werr = err;
+
+ if (root->root_key.objectid == BTRFS_TREE_LOG_OBJECTID) {
+ if ((mark & EXTENT_DIRTY) &&
+ test_and_clear_bit(BTRFS_INODE_BTREE_LOG1_ERR,
+ &btree_ino->runtime_flags))
+ errors = true;
+
+ if ((mark & EXTENT_NEW) &&
+ test_and_clear_bit(BTRFS_INODE_BTREE_LOG2_ERR,
+ &btree_ino->runtime_flags))
+ errors = true;
+ } else {
+ if (test_and_clear_bit(BTRFS_INODE_BTREE_ERR,
+ &btree_ino->runtime_flags))
+ errors = true;
+ }
+
+ if (errors && !werr)
+ werr = -EIO;
+
return werr;
}
return 0;
}
-int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
+static int btrfs_write_and_wait_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
- if (!trans || !trans->transaction) {
- struct inode *btree_inode;
- btree_inode = root->fs_info->btree_inode;
- return filemap_write_and_wait(btree_inode->i_mapping);
- }
- return btrfs_write_and_wait_marked_extents(root,
+ int ret;
+
+ ret = btrfs_write_and_wait_marked_extents(root,
&trans->transaction->dirty_pages,
EXTENT_DIRTY);
+ clear_btree_io_tree(&trans->transaction->dirty_pages);
+
+ return ret;
}
/*
struct btrfs_root *tree_root = root->fs_info->tree_root;
old_root_used = btrfs_root_used(&root->root_item);
- btrfs_write_dirty_block_groups(trans, root);
while (1) {
old_root_bytenr = btrfs_root_bytenr(&root->root_item);
return ret;
old_root_used = btrfs_root_used(&root->root_item);
- ret = btrfs_write_dirty_block_groups(trans, root);
- if (ret)
- return ret;
}
return 0;
struct btrfs_root *root)
{
struct btrfs_fs_info *fs_info = root->fs_info;
+ struct list_head *dirty_bgs = &trans->transaction->dirty_bgs;
struct list_head *next;
struct extent_buffer *eb;
int ret;
- ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
- if (ret)
- return ret;
-
eb = btrfs_lock_root_node(fs_info->tree_root);
ret = btrfs_cow_block(trans, fs_info->tree_root, eb, NULL,
0, &eb);
if (ret)
return ret;
+ ret = btrfs_setup_space_cache(trans, root);
+ if (ret)
+ return ret;
+
/* run_qgroups might have added some more refs */
ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
if (ret)
return ret;
-
+again:
while (!list_empty(&fs_info->dirty_cowonly_roots)) {
next = fs_info->dirty_cowonly_roots.next;
list_del_init(next);
root = list_entry(next, struct btrfs_root, dirty_list);
+ clear_bit(BTRFS_ROOT_DIRTY, &root->state);
if (root != fs_info->extent_root)
list_add_tail(&root->dirty_list,
ret = update_cowonly_root(trans, root);
if (ret)
return ret;
+ ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
+ if (ret)
+ return ret;
}
+ while (!list_empty(dirty_bgs)) {
+ ret = btrfs_write_dirty_block_groups(trans, root);
+ if (ret)
+ return ret;
+ ret = btrfs_run_delayed_refs(trans, root, (unsigned long)-1);
+ if (ret)
+ return ret;
+ }
+
+ if (!list_empty(&fs_info->dirty_cowonly_roots))
+ goto again;
+
list_add_tail(&fs_info->extent_root->dirty_list,
&trans->transaction->switch_commits);
btrfs_after_dev_replace_commit(fs_info);
btrfs_wait_ordered_roots(fs_info, -1);
}
+static inline void
+btrfs_wait_pending_ordered(struct btrfs_transaction *cur_trans,
+ struct btrfs_fs_info *fs_info)
+{
+ struct btrfs_ordered_extent *ordered;
+
+ spin_lock(&fs_info->trans_lock);
+ while (!list_empty(&cur_trans->pending_ordered)) {
+ ordered = list_first_entry(&cur_trans->pending_ordered,
+ struct btrfs_ordered_extent,
+ trans_list);
+ list_del_init(&ordered->trans_list);
+ spin_unlock(&fs_info->trans_lock);
+
+ wait_event(ordered->wait, test_bit(BTRFS_ORDERED_COMPLETE,
+ &ordered->flags));
+ btrfs_put_ordered_extent(ordered);
+ spin_lock(&fs_info->trans_lock);
+ }
+ spin_unlock(&fs_info->trans_lock);
+}
+
int btrfs_commit_transaction(struct btrfs_trans_handle *trans,
struct btrfs_root *root)
{
struct btrfs_transaction *cur_trans = trans->transaction;
struct btrfs_transaction *prev_trans = NULL;
+ struct btrfs_inode *btree_ino = BTRFS_I(root->fs_info->btree_inode);
int ret;
/* Stop the commit early if ->aborted is set */
}
spin_lock(&root->fs_info->trans_lock);
+ list_splice(&trans->ordered, &cur_trans->pending_ordered);
if (cur_trans->state >= TRANS_STATE_COMMIT_START) {
spin_unlock(&root->fs_info->trans_lock);
atomic_inc(&cur_trans->use_count);
wait_for_commit(root, cur_trans);
+ if (unlikely(cur_trans->aborted))
+ ret = cur_trans->aborted;
+
btrfs_put_transaction(cur_trans);
return ret;
btrfs_wait_delalloc_flush(root->fs_info);
+ btrfs_wait_pending_ordered(cur_trans, root->fs_info);
+
btrfs_scrub_pause(root);
/*
* Ok now we need to make sure to block out any other joins while we
}
/*
- * Since the transaction is done, we should set the inode map cache flag
- * before any other comming transaction.
+ * Since the transaction is done, we can apply the pending changes
+ * before the next transaction.
*/
- if (btrfs_test_opt(root, CHANGE_INODE_CACHE))
- btrfs_set_opt(root->fs_info->mount_opt, INODE_MAP_CACHE);
- else
- btrfs_clear_opt(root->fs_info->mount_opt, INODE_MAP_CACHE);
+ btrfs_apply_pending_changes(root->fs_info);
/* commit_fs_roots gets rid of all the tree log roots, it is now
* safe to free the root of tree log roots
switch_commit_roots(cur_trans, root->fs_info);
assert_qgroups_uptodate(trans);
+ ASSERT(list_empty(&cur_trans->dirty_bgs));
update_super_roots(root);
btrfs_set_super_log_root(root->fs_info->super_copy, 0);
memcpy(root->fs_info->super_for_commit, root->fs_info->super_copy,
sizeof(*root->fs_info->super_copy));
+ btrfs_update_commit_device_size(root->fs_info);
+ btrfs_update_commit_device_bytes_used(root, cur_trans);
+
+ clear_bit(BTRFS_INODE_BTREE_LOG1_ERR, &btree_ino->runtime_flags);
+ clear_bit(BTRFS_INODE_BTREE_LOG2_ERR, &btree_ino->runtime_flags);
+
spin_lock(&root->fs_info->trans_lock);
cur_trans->state = TRANS_STATE_UNBLOCKED;
root->fs_info->running_transaction = NULL;
btrfs_finish_extent_commit(trans, root);
+ if (cur_trans->have_free_bgs)
+ btrfs_clear_space_info_full(root->fs_info);
+
root->fs_info->last_trans_committed = cur_trans->transid;
/*
* We needn't acquire the lock here because there is no other task
ret = btrfs_drop_snapshot(root, NULL, 0, 0);
else
ret = btrfs_drop_snapshot(root, NULL, 1, 0);
- /*
- * If we encounter a transaction abort during snapshot cleaning, we
- * don't want to crash here
- */
+
return (ret < 0) ? 0 : 1;
}
+
+void btrfs_apply_pending_changes(struct btrfs_fs_info *fs_info)
+{
+ unsigned long prev;
+ unsigned long bit;
+
+ prev = cmpxchg(&fs_info->pending_changes, 0, 0);
+ if (!prev)
+ return;
+
+ bit = 1 << BTRFS_PENDING_SET_INODE_MAP_CACHE;
+ if (prev & bit)
+ btrfs_set_opt(fs_info->mount_opt, INODE_MAP_CACHE);
+ prev &= ~bit;
+
+ bit = 1 << BTRFS_PENDING_CLEAR_INODE_MAP_CACHE;
+ if (prev & bit)
+ btrfs_clear_opt(fs_info->mount_opt, INODE_MAP_CACHE);
+ prev &= ~bit;
+
+ bit = 1 << BTRFS_PENDING_COMMIT;
+ if (prev & bit)
+ btrfs_debug(fs_info, "pending commit done");
+ prev &= ~bit;
+
+ if (prev)
+ btrfs_warn(fs_info,
+ "unknown pending changes left 0x%lx, ignoring", prev);
+}