By: Anon (no.delete@this.spam.com), August 28, 2022 2:35 pm
Room: Moderated Discussions
--- (---.delete@this.redheron.com) on August 28, 2022 2:16 pm wrote:
> Of course if your branch is truly uncorrelated with the past, the best you can do
> is a statistical predictor; if it's truly uncorrelated with the past AND 50% taken
> then past-based prediction is hopeless. BUT note those *precise* words...
> Branches that are uncorrelated with past branches may be predictable via alternative means that do not
> rely on statistics from the past, and that is one of the newer frontiers of branch research, discovering
> those cases and exploiting them: https://users.elis.ugent.be/~leeckhou/papers/hpca2020.pdf
>
And one of the alternatives to do in those cases is to look at what the software is doing, not all unpredictable branches from hardware perspective are unpredictable from software perspective.
> Of course if your branch is truly uncorrelated with the past, the best you can do
> is a statistical predictor; if it's truly uncorrelated with the past AND 50% taken
> then past-based prediction is hopeless. BUT note those *precise* words...
> Branches that are uncorrelated with past branches may be predictable via alternative means that do not
> rely on statistics from the past, and that is one of the newer frontiers of branch research, discovering
> those cases and exploiting them: https://users.elis.ugent.be/~leeckhou/papers/hpca2020.pdf
>
And one of the alternatives to do in those cases is to look at what the software is doing, not all unpredictable branches from hardware perspective are unpredictable from software perspective.