[llvm] [AMDGPU] Enable vectorization of i8 values. (PR #134934)

[Shilei Tian via llvm-commits]

================
@@ -282,6 +282,16 @@ class GCNTTIImpl final : public BasicTTIImplBase<GCNTTIImpl> {
   void collectKernelLaunchBounds(
       const Function &F,
       SmallVectorImpl<std::pair<StringRef, int64_t>> &LB) const;
+
+  /// \return The cost of Load and Store instructions.
+  InstructionCost getMemoryOpCost(
+      unsigned Opcode, Type *Src, Align Alignment, unsigned AddressSpace,
+      TTI::TargetCostKind CostKind,
+      TTI::OperandValueInfo OpInfo = {TTI::OK_AnyValue, TTI::OP_None},
+      const Instruction *I = nullptr);
+
+  /// \return numer of parts in this type.
+  unsigned getNumberOfParts(Type *Tp);
----------------
shiltian wrote:

What do you mean by "parts"?

https://github.com/llvm/llvm-project/pull/134934