bitops for handling 32 bits

Hi all, Is there a kernel API for handling 32 bits ? I see a macro like BIT(nr) http://lxr.free-electrons.com/source/include/linux/bitops.h#L6 #define BIT(nr) (1UL << (nr)) and also #define BIT_ULL(nr) (1ULL << (nr)) However, these macros are using 64 bit semantics, so if I will use them with 32 bits values the results will be wrong. Any ideas ? Regards, Kevin