[PATCH] D41292: [AMDGPU] Fixed incorrect uniform branch condition

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added a comment.

In https://reviews.llvm.org/D41292#956933, @arsenm wrote:

> In https://reviews.llvm.org/D41292#956918, @tpr wrote:
>
> > Ah, I didn't see r286134.
> >
> > How do you mean, partially revert it? It looks like I want to completely revert it and fix the comment for semantic correctness, then think about optimizing the cases when the s_and is not needed.
>
>
> I mean mostly keep it, just re-insert the and when the branch condition isn't known to be a compare like it implicitly assumes


I suppose there are a few other special cases, like CMP_CLASS and other known-to-be-i1 produced instructions


https://reviews.llvm.org/D41292