[LLVMdev] Predictive Commoning / Scalar Replacement
dberlin at dberlin.org
Sun Dec 2 08:39:57 PST 2012
On Sun, Dec 2, 2012 at 7:48 AM, Krzysztof Parzyszek
<kparzysz at codeaurora.org> wrote:
> On 12/1/2012 10:38 PM, Daniel Berlin wrote:
>> There is a non-internal techpub for this somewhere, though it escapes me.
>> My recollection is that TPO (the middle end of XLC) had moved on to
>> 2nd order predictive quite a while ago.
>> At least, the implementation I saw 6+ years ago in TPO was much closer
>> to what is described in the 2nd order paper.
> TPO never had any other predictive commoning. The TPO's implementation was
> written by the guy who wrote the paper (Arie).
To be clear, you are saying TPO's implementation was always 2nd order?
If so, that would make sense :)
More information about the llvm-dev