[LLVMdev] Predictive Commoning / Scalar Replacement

[Krzysztof Parzyszek]

On 12/2/2012 10:39 AM, Daniel Berlin wrote:
>
> To be clear, you are saying TPO's implementation was always 2nd order?
> If so, that would make sense :)

Sort of.  If I remember correctly, the first working implementation 
might have been simpler, but the final version came shortly thereafter. 
  For all intents and purposes, I consider it to have been implemented 
in one shot.  Look it up in CMVC if you want all the gory details...  ;)

-Krzysztof


-- 
Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, 
hosted by The Linux Foundation