> if the glibc headers make the kernel not work, that's a bug in the AC code I > think... I'm not responsible for what the glibc folks ship, nor for the RH glibc package - if it is mismatched its an RH packaging bug