Tag
GLib adopts libffi's new ffi_call_plan caching to reduce GClosure invocation overhead by 10-25%, with merge request submitted and frame-pointer patches for profiling.