float32_to_int64_round_to_zero
#define float32_to_int64_round_to_zero __fixsfdi
int64 float32_to_int64_round_to_zero( float32 a )