Btrfs: Introduce contexts for metadata reservation

diff -urp 2/fs/btrfs/ctree.c 3/fs/btrfs/ctree.c
@@ -279,7 +279,8 @@ int btrfs_block_can_be_shared(struct btr
 static noinline int update_ref_for_cow(struct btrfs_trans_handle *trans,
 				       struct btrfs_root *root,
 				       struct extent_buffer *buf,
-				       struct extent_buffer *cow)
+				       struct extent_buffer *cow,
+				       int *last_ref)
 {
 	u64 refs;
 	u64 owner;

@@ -365,6 +366,7 @@ static noinline int update_ref_for_cow(s
 			BUG_ON(ret);
 		}
 		clean_tree_block(trans, root, buf);
+		*last_ref = 1;
 	}
 	return 0;
 }

@@ -391,6 +393,7 @@ static noinline int __btrfs_cow_block(st
 	struct btrfs_disk_key disk_key;
 	struct extent_buffer *cow;
 	int level;
+	int last_ref = 0;
 	int unlock_orig = 0;
 	u64 parent_start;
 

@@ -441,7 +444,7 @@ static noinline int __btrfs_cow_block(st
 			    (unsigned long)btrfs_header_fsid(cow),
 			    BTRFS_FSID_SIZE);
 
-	update_ref_for_cow(trans, root, buf, cow);
+	update_ref_for_cow(trans, root, buf, cow, &last_ref);
 
 	if (buf == root->node) {
 		WARN_ON(parent && parent != buf);

@@ -456,8 +459,8 @@ static noinline int __btrfs_cow_block(st
 		extent_buffer_get(cow);
 		spin_unlock(&root->node_lock);
 
-		btrfs_free_tree_block(trans, root, buf->start, buf->len,
-				parent_start, root->root_key.objectid, level);
+		btrfs_free_tree_block(trans, root, buf, parent_start,
+				      last_ref);
 		free_extent_buffer(buf);
 		add_root_to_dirty_list(root);
 	} else {

@@ -472,8 +475,8 @@ static noinline int __btrfs_cow_block(st
 		btrfs_set_node_ptr_generation(parent, parent_slot,
 					      trans->transid);
 		btrfs_mark_buffer_dirty(parent);
-		btrfs_free_tree_block(trans, root, buf->start, buf->len,
-				parent_start, root->root_key.objectid, level);
+		btrfs_free_tree_block(trans, root, buf, parent_start,
+				      last_ref);
 	}
 	if (unlock_orig)
 		btrfs_tree_unlock(buf);

@@ -948,6 +951,22 @@ int btrfs_bin_search(struct extent_buffe
 	return bin_search(eb, key, level, slot);
 }
 
+static void root_add_used(struct btrfs_root *root, u32 size)
+{
+	spin_lock(&root->accounting_lock);
+	btrfs_set_root_used(&root->root_item,
+			    btrfs_root_used(&root->root_item) + size);
+	spin_unlock(&root->accounting_lock);
+}
+
+static void root_sub_used(struct btrfs_root *root, u32 size)
+{
+	spin_lock(&root->accounting_lock);
+	btrfs_set_root_used(&root->root_item,
+			    btrfs_root_used(&root->root_item) - size);
+	spin_unlock(&root->accounting_lock);
+}
+
 /* given a node and slot number, this reads the blocks it points to.  The
  * extent buffer is returned with a reference taken (but unlocked).
  * NULL is returned on error.

@@ -1018,7 +1037,11 @@ static noinline int balance_level(struct
 		btrfs_tree_lock(child);
 		btrfs_set_lock_blocking(child);
 		ret = btrfs_cow_block(trans, root, child, mid, 0, &child);
-		BUG_ON(ret);
+		if (ret) {
+			btrfs_tree_unlock(child);
+			free_extent_buffer(child);
+			goto enospc;
+		}
 
 		spin_lock(&root->node_lock);
 		root->node = child;

@@ -1033,11 +1056,12 @@ static noinline int balance_level(struct
 		btrfs_tree_unlock(mid);
 		/* once for the path */
 		free_extent_buffer(mid);
-		ret = btrfs_free_tree_block(trans, root, mid->start, mid->len,
-					    0, root->root_key.objectid, level);
+
+		root_sub_used(root, mid->len);
+		btrfs_free_tree_block(trans, root, mid, 0, 1);
 		/* once for the root ptr */
 		free_extent_buffer(mid);
-		return ret;
+		return 0;
 	}
 	if (btrfs_header_nritems(mid) >
 	    BTRFS_NODEPTRS_PER_BLOCK(root) / 4)

@@ -1087,23 +1111,16 @@ static noinline int balance_level(struct
 		if (wret < 0 && wret != -ENOSPC)
 			ret = wret;
 		if (btrfs_header_nritems(right) == 0) {
-			u64 bytenr = right->start;
-			u32 blocksize = right->len;
-
 			clean_tree_block(trans, root, right);
 			btrfs_tree_unlock(right);
-			free_extent_buffer(right);
-			right = NULL;
 			wret = del_ptr(trans, root, path, level + 1, pslot +
 				       1);
 			if (wret)
 				ret = wret;
-			wret = btrfs_free_tree_block(trans, root,
-						     bytenr, blocksize, 0,
-						     root->root_key.objectid,
-						     level);
-			if (wret)
-				ret = wret;
+			root_sub_used(root, right->len);
+			btrfs_free_tree_block(trans, root, right, 0, 1);
+			free_extent_buffer(right);
+			right = NULL;
 		} else {
 			struct btrfs_disk_key right_key;
 			btrfs_node_key(right, &right_key, 0);

@@ -1135,21 +1152,15 @@ static noinline int balance_level(struct
 		BUG_ON(wret == 1);
 	}
 	if (btrfs_header_nritems(mid) == 0) {
-		/* we've managed to empty the middle node, drop it */
-		u64 bytenr = mid->start;
-		u32 blocksize = mid->len;
-
 		clean_tree_block(trans, root, mid);
 		btrfs_tree_unlock(mid);
-		free_extent_buffer(mid);
-		mid = NULL;
 		wret = del_ptr(trans, root, path, level + 1, pslot);
 		if (wret)
 			ret = wret;
-		wret = btrfs_free_tree_block(trans, root, bytenr, blocksize,
-					 0, root->root_key.objectid, level);
-		if (wret)
-			ret = wret;
+		root_sub_used(root, mid->len);
+		btrfs_free_tree_block(trans, root, mid, 0, 1);
+		free_extent_buffer(mid);
+		mid = NULL;
 	} else {
 		/* update the parent key to reflect our changes */
 		struct btrfs_disk_key mid_key;

@@ -1739,7 +1750,6 @@ again:
 					      p->nodes[level + 1],
 					      p->slots[level + 1], &b);
 			if (err) {
-				free_extent_buffer(b);
 				ret = err;
 				goto done;
 			}

@@ -2075,6 +2085,8 @@ static noinline int insert_new_root(stru
 	if (IS_ERR(c))
 		return PTR_ERR(c);
 
+	root_add_used(root, root->nodesize);
+
 	memset_extent_buffer(c, 0, 0, sizeof(struct btrfs_header));
 	btrfs_set_header_nritems(c, 1);
 	btrfs_set_header_level(c, level);

@@ -2133,6 +2145,7 @@ static int insert_ptr(struct btrfs_trans
 	int nritems;
 
 	BUG_ON(!path->nodes[level]);
+	btrfs_assert_tree_locked(path->nodes[level]);
 	lower = path->nodes[level];
 	nritems = btrfs_header_nritems(lower);
 	BUG_ON(slot > nritems);

@@ -2201,6 +2214,8 @@ static noinline int split_node(struct bt
 	if (IS_ERR(split))
 		return PTR_ERR(split);
 
+	root_add_used(root, root->nodesize);
+
 	memset_extent_buffer(split, 0, 0, sizeof(struct btrfs_header));
 	btrfs_set_header_level(split, btrfs_header_level(c));
 	btrfs_set_header_bytenr(split, split->start);

@@ -2414,6 +2429,9 @@ static noinline int __push_leaf_right(st
 
 	if (left_nritems)
 		btrfs_mark_buffer_dirty(left);
+	else
+		clean_tree_block(trans, root, left);
+
 	btrfs_mark_buffer_dirty(right);
 
 	btrfs_item_key(right, &disk_key, 0);

@@ -2659,6 +2677,8 @@ static noinline int __push_leaf_left(str
 	btrfs_mark_buffer_dirty(left);
 	if (right_nritems)
 		btrfs_mark_buffer_dirty(right);
+	else
+		clean_tree_block(trans, root, right);
 
 	btrfs_item_key(right, &disk_key, 0);
 	wret = fixup_low_keys(trans, root, path, &disk_key, 1);

@@ -2668,8 +2688,6 @@ static noinline int __push_leaf_left(str
 	/* then fixup the leaf pointer in the path */
 	if (path->slots[0] < push_items) {
 		path->slots[0] += old_left_nritems;
-		if (btrfs_header_nritems(path->nodes[0]) == 0)
-			clean_tree_block(trans, root, path->nodes[0]);
 		btrfs_tree_unlock(path->nodes[0]);
 		free_extent_buffer(path->nodes[0]);
 		path->nodes[0] = left;

@@ -2931,10 +2949,10 @@ again:
 	right = btrfs_alloc_free_block(trans, root, root->leafsize, 0,
 					root->root_key.objectid,
 					&disk_key, 0, l->start, 0);
-	if (IS_ERR(right)) {
-		BUG_ON(1);
+	if (IS_ERR(right))
 		return PTR_ERR(right);
-	}
+
+	root_add_used(root, root->leafsize);
 
 	memset_extent_buffer(right, 0, 0, sizeof(struct btrfs_header));
 	btrfs_set_header_bytenr(right, right->start);

@@ -3053,7 +3071,8 @@ static noinline int setup_leaf_for_split
 
 	btrfs_set_path_blocking(path);
 	ret = split_leaf(trans, root, &key, path, ins_len, 1);
-	BUG_ON(ret);
+	if (ret)
+		goto err;
 
 	path->keep_locks = 0;
 	btrfs_unlock_up_safe(path, 1);

@@ -3795,9 +3814,10 @@ static noinline int btrfs_del_leaf(struc
 	 */
 	btrfs_unlock_up_safe(path, 0);
 
-	ret = btrfs_free_tree_block(trans, root, leaf->start, leaf->len,
-				    0, root->root_key.objectid, 0);
-	return ret;
+	root_sub_used(root, leaf->len);
+
+	btrfs_free_tree_block(trans, root, leaf, 0, 1);
+	return 0;
 }
 /*
  * delete the item at the leaf level in path.  If that empties

@@ -3864,6 +3884,8 @@ int btrfs_del_items(struct btrfs_trans_h
 		if (leaf == root->node) {
 			btrfs_set_header_level(leaf, 0);
 		} else {
+			btrfs_set_path_blocking(path);
+			clean_tree_block(trans, root, leaf);
 			ret = btrfs_del_leaf(trans, root, path, leaf);
 			BUG_ON(ret);
 		}

111 lines not on this page · raw message

diff -urp 2/fs/btrfs/disk-io.c 3/fs/btrfs/disk-io.c
@@ -904,6 +904,7 @@ static int __setup_root(u32 nodesize, u3
 	root->name = NULL;
 	root->in_sysfs = 0;
 	root->inode_tree = RB_ROOT;
+	root->block_rsv = NULL;
 
 	INIT_LIST_HEAD(&root->dirty_list);
 	INIT_LIST_HEAD(&root->orphan_list);

@@ -911,6 +912,7 @@ static int __setup_root(u32 nodesize, u3
 	spin_lock_init(&root->node_lock);
 	spin_lock_init(&root->list_lock);
 	spin_lock_init(&root->inode_lock);
+	spin_lock_init(&root->accounting_lock);
 	mutex_init(&root->objectid_mutex);
 	mutex_init(&root->log_mutex);
 	init_waitqueue_head(&root->log_writer_wait);

@@ -1629,6 +1631,13 @@ struct btrfs_root *open_ctree(struct sup
 	INIT_LIST_HEAD(&fs_info->dirty_cowonly_roots);
 	INIT_LIST_HEAD(&fs_info->space_info);
 	btrfs_mapping_init(&fs_info->mapping_tree);
+	btrfs_init_block_rsv(&fs_info->global_block_rsv);
+	btrfs_init_block_rsv(&fs_info->delalloc_block_rsv);
+	btrfs_init_block_rsv(&fs_info->trans_block_rsv);
+	btrfs_init_block_rsv(&fs_info->chunk_block_rsv);
+	btrfs_init_block_rsv(&fs_info->empty_block_rsv);
+	INIT_LIST_HEAD(&fs_info->durable_block_rsv_list);
+	mutex_init(&fs_info->durable_block_rsv_mutex);
 	atomic_set(&fs_info->nr_async_submits, 0);
 	atomic_set(&fs_info->async_delalloc_pages, 0);
 	atomic_set(&fs_info->async_submit_draining, 0);

1,389 lines not on this page · raw message

diff -urp 2/fs/btrfs/relocation.c 3/fs/btrfs/relocation.c
@@ -3513,6 +3513,7 @@ int btrfs_relocate_block_group(struct bt
 	struct btrfs_fs_info *fs_info = extent_root->fs_info;
 	struct reloc_control *rc;
 	int ret;
+	int rw = 0;
 	int err = 0;
 
 	rc = kzalloc(sizeof(*rc), GFP_NOFS);

@@ -3523,15 +3524,22 @@ int btrfs_relocate_block_group(struct bt
 	extent_io_tree_init(&rc->processed_blocks, NULL, GFP_NOFS);
 	INIT_LIST_HEAD(&rc->reloc_roots);
 
+	rc->extent_root = extent_root;
 	rc->block_group = btrfs_lookup_block_group(fs_info, group_start);
 	BUG_ON(!rc->block_group);
 
+	if (!rc->block_group->ro) {
+		ret = btrfs_set_block_group_ro(extent_root, rc->block_group);
+		if (ret) {
+			err = ret;
+			goto out;
+		}
+		rw = 1;
+	}
+
 	btrfs_init_workers(&rc->workers, "relocate",
 			   fs_info->thread_pool_size, NULL);
 
-	rc->extent_root = extent_root;
-	btrfs_prepare_block_group_relocation(extent_root, rc->block_group);
-
 	rc->data_inode = create_reloc_inode(fs_info, rc->block_group);
 	if (IS_ERR(rc->data_inode)) {
 		err = PTR_ERR(rc->data_inode);

@@ -3596,6 +3604,8 @@ int btrfs_relocate_block_group(struct bt
 	WARN_ON(rc->block_group->reserved > 0);
 	WARN_ON(btrfs_block_group_used(&rc->block_group->item) > 0);
 out:
+	if (err && rw)
+		btrfs_set_block_group_rw(extent_root, rc->block_group);
 	iput(rc->data_inode);
 	btrfs_stop_workers(&rc->workers);
 	btrfs_put_block_group(rc->block_group);

diff -urp 2/fs/btrfs/transaction.c 3/fs/btrfs/transaction.c
@@ -184,9 +184,8 @@ static struct btrfs_trans_handle *start_
 	h->blocks_reserved = num_blocks;
 	h->blocks_used = 0;
 	h->block_group = 0;
-	h->alloc_exclude_nr = 0;
-	h->alloc_exclude_start = 0;
 	h->delayed_ref_updates = 0;
+	h->block_rsv = NULL;
 
 	if (!current->journal_info && type != TRANS_USERSPACE)
 		current->journal_info = h;

diff -urp 2/fs/btrfs/transaction.h 3/fs/btrfs/transaction.h
@@ -45,13 +45,13 @@ struct btrfs_transaction {
 
 struct btrfs_trans_handle {
 	u64 transid;
+	u64 block_group;
+	u64 bytes_reserved;
 	unsigned long blocks_reserved;
 	unsigned long blocks_used;
-	struct btrfs_transaction *transaction;
-	u64 block_group;
-	u64 alloc_exclude_start;
-	u64 alloc_exclude_nr;
 	unsigned long delayed_ref_updates;
+	struct btrfs_transaction *transaction;
+	struct btrfs_block_rsv *block_rsv;
 };
 
 struct btrfs_pending_snapshot {