btrfs: no need to run delayed refs after commit_fs_roots
David Sterba
maintainerPATCH 1/1 btrfs: no need to run delayed refs after commit_fs_roots
The inode number cache has been removed in this dev cycle, there's one more leftover. We don't need to run the delayed refs again after commit_fs_roots as stated in the comment, because btrfs_save_ino_cache is no more since 5297199a8bca ("btrfs: remove inode number cache feature").
Nothing else between commit_fs_roots and btrfs_qgroup_account_extents could create new delayed refs so the qgroup consistency should be safe.
---
There's the patchset
https://lore.kernel.org/linux-btrfs/[email protected]/
to remove several other run delayed ref calls. I haven't spotted any problems so this patch can go in now as it's related to the inode cache removal. The patchset unifies the location of delayed refs calls so this one would be an outlier anyway.
fs/btrfs/transaction.c | 8 -------- 1 file changed, 8 deletions(-)
fs/btrfs/transaction.c +0 -8
@@ -2265,14 +2265,6 @@ int btrfs_commit_transaction(struct btrfs_trans_handle *trans)*/ btrfs_free_log_root_tree (trans, fs_info); -/* -* commit_fs_roots() can call btrfs_save_ino_cache(), which generates -* new delayed refs. Must handle them or qgroup can be wrong. -*/ - ret =btrfs_run_delayed_refs (trans, (unsigned long )-1 ); -if (ret) - goto unlock_tree_log; -/* * Since fs roots are all committed, we can get a quite accurate * new_roots. So let's do quota accounting.
Signature · 1 line
2.25.0
Quoted 10 lines · David Sterba
The inode number cache has been removed in this dev cycle, there's one more leftover. We don't need to run the delayed refs again after
commit_fs_rootsas stated in the comment, becausebtrfs_save_ino_cacheis no more since5297199a8bca("btrfs: remove inode number cache feature").Nothing else between
commit_fs_rootsandbtrfs_qgroup_account_extentscould create new delayed refs so the qgroup consistency should be safe.Signed-off-by David Sterba <[email protected]>