[llvm-commits] Vectorization Factor review

[Renato Golin]

Sorry. Here it is.


On 28 January 2013 15:24, Nadav Rotem <nrotem at apple.com> wrote:

> Can you please send the patch to the mailing list ?
>
> Thanks,
> Nadav
>
> On Jan 28, 2013, at 4:04 AM, Renato Golin <renato.golin at linaro.org> wrote:
>
> http://llvm-reviews.chandlerc.com/D335
>
> Small patch to (hopefully) clarify what the parameters mean (width/cost)
> from the std::pair<>'s first/second approach.
>
> There's also a potentially problematic (and wrong) change by the end. I'm
> assuming the cost we want to return is the cost of the cheapest width
> multiplied by the width where that cost was calculated, not by the maximum
> width.
>
> It's possible that this was never a problem, if on all tests, the maximum
> width is the one always returned (in which case, Width == VF).
>
> cheers,
> --renato
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130128/7ee38ac1/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: vect-factor.patch
Type: application/octet-stream
Size: 4722 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130128/7ee38ac1/attachment.obj>