[llvm] [LV] Support strided memory accesses with a stride of -1 (PR #128718)

[Mel Chen via llvm-commits]

================
@@ -2126,8 +2126,13 @@ void VPVectorPointerRecipe::execute(VPTransformState &State) {
   Value *Ptr = State.get(getOperand(0), VPLane(0));
 
   Value *Increment = createStepForVF(Builder, IndexTy, State.VF, CurrentPart);
+  // TODO: Support non-unit-reverse strided accesses.
+  Value *Index =
+      Strided
+          ? Builder.CreateMul(Increment, ConstantInt::getSigned(IndexTy, -1))
----------------
Mel-Chen wrote:

> I don't think we should add another 'mode' to WidenLoad/StoreRecipe to make it even more complicated.
> 
> It would probably be good to have a separate recipe.

@fhahn I see. Let's first try adding VPStridedLoadRecipe for strided loads. What about VPVectorPointerRecipe? Can we modify it to:
```
VPVectorPointerRecipe(VPValue *Ptr, Type *IndexedTy, VPValue *Stride,  
                      GEPNoWrapFlags GEPFlags, DebugLoc DL) 
```
Or should we introduce a new recipe instead?


https://github.com/llvm/llvm-project/pull/128718