I'm with Joe on this. I think we should keep all cases from regressing.
Having that, the patch is not really acceptable if it is degrading the
mag.length < 8 cases.

-Aleksey.

On 10/18/2013 03:52 AM, Brian Burkhalter wrote:
> Hi Joe,
> 
> Will do.
> 
> Thanks,
> 
> Brian
> 
> On Oct 17, 2013, at 4:08 PM, Joseph Darcy wrote:
> 
>> Can you test the performance of a patch with:
>>
>>    if (val == this and mag.length > 8)
>>
>> Do you have any measurements of the multiply performance of values that are 
>> not ==? (Making sure there is no general regression.)
> 

Reply via email to