mul 128bitx128bit = 256bit result2312 cycles for 100 * mul128x128_256 (AW)val1=0xa548ddeb93f93d70479983e499807805val2=0x538999ea78812af3c92329fd26552a91val1*val2=0x35ef77521316318c1b59f6f041c08950a5ee764d3dee56f11425a9139d1dccd5