On Sat, Apr 05, 2014 at 12:22:23PM -0700, Todd Fiala wrote: > I suspect one way to handle is to disable on gcc and just expect the clang > builds to catch the real issues in that case. Correct, that was my suggestion. Joerg