Btrfs: allow block group cache writeout outside critical section in commit
We loop through all of the dirty block groups during commit and write
the free space cache. In order to make sure the cache is currect, we do
this while no other writers are allowed in the commit.
If a large number of block groups are dirty, this can introduce long
stalls during the final stages of the commit, which can block new procs
trying to change the filesystem.
This commit changes the block group cache writeout to take appropriate
locks and allow it to run earlier in the commit. We'll still have to
redo some of the block groups, but it means we can get most of the work
out of the way without blocking the entire FS.
Signed-off-by: Chris Mason <clm@fb.com>
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c
index d830853..840a4eb 100644
--- a/fs/btrfs/relocation.c
+++ b/fs/btrfs/relocation.c
@@ -3430,7 +3430,9 @@
}
static int delete_block_group_cache(struct btrfs_fs_info *fs_info,
- struct inode *inode, u64 ino)
+ struct btrfs_block_group_cache *block_group,
+ struct inode *inode,
+ u64 ino)
{
struct btrfs_key key;
struct btrfs_root *root = fs_info->tree_root;
@@ -3463,7 +3465,7 @@
goto out;
}
- ret = btrfs_truncate_free_space_cache(root, trans, inode);
+ ret = btrfs_truncate_free_space_cache(root, trans, block_group, inode);
btrfs_end_transaction(trans, root);
btrfs_btree_balance_dirty(root);
@@ -3509,6 +3511,7 @@
*/
if (ref_root == BTRFS_ROOT_TREE_OBJECTID) {
ret = delete_block_group_cache(rc->extent_root->fs_info,
+ rc->block_group,
NULL, ref_objectid);
if (ret != -ENOENT)
return ret;
@@ -4223,7 +4226,7 @@
btrfs_free_path(path);
if (!IS_ERR(inode))
- ret = delete_block_group_cache(fs_info, inode, 0);
+ ret = delete_block_group_cache(fs_info, rc->block_group, inode, 0);
else
ret = PTR_ERR(inode);
