[llvm] [RISCV][TTI] Refine reverse shuffle costing for high LMUL (PR #144155)

[Craig Topper via llvm-commits]

================
@@ -602,6 +602,15 @@ InstructionCost RISCVTTIImpl::getSlideCost(FixedVectorType *Tp,
   return FirstSlideCost + SecondSlideCost + MaskCost;
 }
 
+// Consolidate!
+static MVT getLMUL1VT(MVT VT) {
+  assert(VT.getVectorElementType().getSizeInBits() <= 64 &&
----------------
topperc wrote:

Should 64 here be RISCV::RVVBitsPerBlock?

https://github.com/llvm/llvm-project/pull/144155