[llvm-branch-commits] [llvm] DAG: Avoid forming shufflevector from a single extract_vector_elt (PR #122672)

[Philip Reames via llvm-branch-commits]

preames wrote:

> > * BuildVector w/one non-zero non-undef source, repeated 100 times (i.e. splat or select of two splats)
> 
> I don't follow, this is a 2 element vector, how can you have 100 variants?

Isn't the condition in code in terms of VecIn.size() == 2?  I believe that VecIn is the *unique* input elements, right?  Which is distinct from the number of elements in the destination type?  (Am I just misreading?  I only skimmed this.)

> > If the target isn't optimally lowering the splat or select of splat case in the shuffle lowering, maybe we should just adjust the target lowering to do so?t
> 
> It's not a lowering issue, it's the effect on every other combine. We'd have to special case 1 element + 1 undef shuffles everywhere we handle extract_vector_elt now, which is just excessive complexity. #122671 is almost an alternative in one instance, but still shows expanding complexity of handling this edge case.

Honestly,  #122671 (from the review description only) sounds like a worthwhile change.  That's not a hugely compelling argument here.  Let's settle the prior point, and then return to this.  If I'm just misreading something, let's not waste time discussing this.  



https://github.com/llvm/llvm-project/pull/122672