https://bugs.freedesktop.org/show_bug.cgi?id=94747
--- Comment #3 from Ian Romanick <i...@freedesktop.org> ---
(In reply to Timothy Arceri from comment #2)
> As (In reply to Jason Ekstrand from comment #1)
> > The code for this already exists. It's called nir_opt_peephole_select. The
> > only problem is that it only triggers if both sides of the if are empty.
> > I've been wanting to add some sort of heuristic to it for some time now.
> > The only problem is that it's really back-end specific.
>
> Although i965 doesn't use it I believe this is otherwise fixed by the
> following commit. We should probably close this bug:
Thanks for reminding me about these "don't forget" bugs that I filed. :) I
clearly forgot about them anyway.
I have a patch for this that has been on the list for almost 3 weeks.
https://patchwork.freedesktop.org/patch/233182/
That will just eliminate the flow control (but it might not since there are 2
instructions... I wish I could remember which shader this was). Even then
we'll get something like
ssa_177 = ssa_153 ? true : abs(ssa_171 - ssa_135) >= ssa_145
when we really want (ssa_153 || (abs(ssa_171 - ssa_135) >= ssa_145)). I have
another patch series in progress that will take care of that.
--
You are receiving this mail because:
You are the QA Contact for the bug._______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/mesa-dev
