[PATCH 07/15] btrfs-progs: do not insert extent items for metadata for extent tree v2
From: Josef Bacik <josef@toxicpanda.com>
Date: 2021-12-15 20:15:07
Subsystem:
the rest · Maintainer:
Linus Torvalds
For extent tree v2 we are no longer tracking metadata blocks in the extent tree, so simply skip this step and remove the space from the free space tree and do the block group accounting. Signed-off-by: Josef Bacik <josef@toxicpanda.com> --- kernel-shared/extent-tree.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
diff --git a/kernel-shared/extent-tree.c b/kernel-shared/extent-tree.c
index 1af3eb06..7beff7ee 100644
--- a/kernel-shared/extent-tree.c
+++ b/kernel-shared/extent-tree.c@@ -2392,6 +2392,9 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, sinfo = __find_space_info(fs_info, BTRFS_BLOCK_GROUP_METADATA); ASSERT(sinfo); + if (btrfs_fs_incompat(fs_info, EXTENT_TREE_V2)) + goto alloc; + ins.objectid = node->bytenr; if (skinny_metadata) { ins.offset = ref->level;
@@ -2445,11 +2448,12 @@ static int alloc_reserved_tree_block(struct btrfs_trans_handle *trans, btrfs_mark_buffer_dirty(leaf); btrfs_free_path(path); - ret = remove_from_free_space_tree(trans, ins.objectid, fs_info->nodesize); +alloc: + ret = remove_from_free_space_tree(trans, node->bytenr, fs_info->nodesize); if (ret) return ret; - ret = update_block_group(trans, ins.objectid, fs_info->nodesize, 1, 0); + ret = update_block_group(trans, node->bytenr, fs_info->nodesize, 1, 0); if (sinfo) { if (fs_info->nodesize > sinfo->bytes_reserved) { WARN_ON(1);
--
2.26.3