float128_to_uint64_round_to_zero
uint64 float128_to_uint64_round_to_zero( float128 a )
#define float128_to_uint64_round_to_zero __fixunstfdi