Backport of gh-152826, merged as f156510, for gh-151518.
A tight loop of stop-the-world pauses can repeatedly suspend a thread trying to reattach after sleeping, preventing it from returning and stopping the loop. This was found during intentional numba-cuda GC stress on many-core machines. The fix preserves active attach waiters across successive pauses so they have a chance to run.
This branch predates gh-157839, so the waiter-aware resume operation is a private helper in pystate.c. The attachment changes, regression test, and guard that avoids blocking waits on WASI are carried over from the merged fix.
Validation:
- Free-threaded debug and TSan: all 31 focused threading, GC, and asyncio tests passed.
- With only the test changes, the unfixed branch hits the 30-second watchdog with 4 and 64 CPUs. The backport passes three runs each with 1, 4, and 64 CPUs.
- Internal C API helper tests, including
test_stop_the_world: all 16 passed in the ordinary GIL build and all 13 available tests passed on WASI, using this branch's SDK 24 and Wasmtime 38.0.3 toolchain.
Rebased onto current 3.14 to include the Windows toolset fix from gh-158188. The backport patch is unchanged; after rebuilding, all 31 focused free-threaded debug tests passed again. Fresh CI is pending, including both Windows ARM64 configurations that previously failed before compilation.