> It was disabled because it made asserting whether or not execution was > currently in a critical section fail when the program failed to use > threads. Yeah, I don't object to that at all. It seemed like a fine tradeoff in favor of maintenance sanity over performance.