Like gh-126491 but implemented for the free-threaded version of the cyclic GC. This is likely a performance win for mostly the same reasons. In the free-threaded version, the marking phase avoids iterating over the mimalloc arenas as well.
Has this already been discussed elsewhere?
This is a minor feature, which does not need previous discussion elsewhere
Activity
gh-128807: Add marking phase for free-threaded cyclic GC (gh-128808)