On 09/21/2015 10:54 PM, Chen Gang wrote: > On 2015年09月19日 10:34, Richard Henderson wrote: >> >> There's a trick for this that's more efficient for 4 or more elements >> per vector (i.e. good for v2 and v1, but not v4): >> >> a + b = (a & 0x7f7f7f7f) + (b & 0x7f7f7f7f)) ^ ((a ^ b) & 0x80808080) >> >> a - b = (a | 0x80808080) - (b & 0x7f7f7f7f)) ^ ((a ^ ~b) & 0x80808080) >> > > For me, we need use "(a ^ b) & 0x80..." instead of "(a ^ ~b) & 0x80...".
No. What you did wrong was not use (a | 0x80808080). r~
