> Sorry, but that does not make any sense. An (non-experimental) option that > leads to breakage is still broken even if it's documented. Sure, sure, but on the other hand I pushed one more patch! --tml