[Mesa-dev] V8 Loop unrolling in NIR

[Jason Ekstrand]

I have no real opinion on patch 13.  Patch 1 is mine so you need to review
it. :-)  Patch 3/13 needs a fix or it will break clip and cull.  We don't
need 6/13 or lcssa_phi at all anymore.  Pending fixes to the loop unrolling
pass to get rid of Vulkan regressions, the series (except 13) is

Reviewed-by: Jason Ekstrand <jason at jlekstrand.net>

Feel free to grab patches off my jenkins_vulkan branch and squash them in.

Happy Christmas!

--Jason

On Wed, Dec 21, 2016 at 6:26 PM, Timothy Arceri <
timothy.arceri at collabora.com> wrote:

> V8:
>  - bunch of tidy ups mostly suggestions from Jason.
>  - fixed some regressions since V6 caused by V7 handling less
>    trivial loops. Piglit tests sent to cover these cases.
>
> New piglit tests:
> https://patchwork.freedesktop.org/patch/129066/
>
> This series seems to regress some Vulkan CTS tests but I assume
> jasons series will take care of those? I haven't tested that,
> I haven't been seeing jenkins results from the Vulkan CTS until
> today.
>
> _______________________________________________
> mesa-dev mailing list
> mesa-dev at lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/mesa-dev
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.freedesktop.org/archives/mesa-dev/attachments/20161222/ebaf29b0/attachment.html>