On Wed, Mar 20, 2024 at 02:29:22PM +0800, Hongbo Li wrote: > First, the old structure cannot clearly represent the state changes > of btree_path (such as BTREE_ITER_xxx). Secondly, the member ( > btree_path->uptodate) cannot express its purpose intuitively. It's > essentially a state value if I understand correctly. Using this way > can makes the representation of member variables more reasonable.
path->uptodate was fine, I don't think I'll be taking this patch. The helpers you introduced in the other patch were better, if you want to rework that patch.
