int32_to_float32
fpa11->fpreg[getFn(opcode)].fSingle = int32_to_float32(&roundData, readRegister(getRd(opcode)));
float32 int32_to_float32( struct roundingData *, signed int );