Memory held for too long by the BRC queue while the owning thread is detached

On the free-threaded build, when a non-owner thread decrefs an object and ob_ref_shared drops to zero, the object is queued for its owning thread and is only freed when that thread next runs Python code. If the owner stays detached for a long time — blocked on a lock, in a syscall, waiting on a pipe — the queued objects stay alive for as long as the owner is blocked. The GC drains the queue, but if the objects holds large amounts of memory then it is ineffective.

This was reported in Py314t uses 2× memory: a PyTorch training run uses 10GB on 3.14 and 20GB on 3.14t. The objects are large tensors enqueued on a multiprocessing.Queue, whose feeder thread does the cross-thread decref while the owner sits detached.

I propose to let the queueing thread do the merge itself. When it tries to queue a object of detached owner, it moves the owner from detached to suspended (the same transition stop-the-world uses), merges the refcounts, resumes the owner, and deallocates whatever reached zero outside the bucket mutex.

Linked PRs

• gh-157839

Issue không nêu tên tệp hay test nào. Hãy bắt đầu bằng cách theo dõi free-threaded BRC queue và quá trình chuyển từ detached sang suspended, sau đó xem xét cách GC xả các đối tượng đang xếp hàng. Công việc được xem là hoàn tất khi queueing thread có thể hợp nhất các tham chiếu cho một detached owner và giải phóng các đối tượng mà không cần chờ owner đó chạy mã Python; hãy xác thực bằng workload multiprocessing.Queue được mô tả trong báo cáo.