[llvm] [AMDGPU] Add amdgcn.av.global.(load|store).b128 intrinsics (PR #191390)

[Pierre van Houtryve via llvm-commits]

================
@@ -1718,6 +1718,27 @@ void SITargetLowering::getTgtMemIntrinsic(SmallVectorImpl<IntrinsicInfo> &Infos,
     Infos.push_back(Info);
     return;
   }
+  case Intrinsic::amdgcn_av_global_load_b128:
+  case Intrinsic::amdgcn_av_global_store_b128: {
+    bool IsStore = IntrID == Intrinsic::amdgcn_av_global_store_b128;
+    Info.opc = IsStore ? ISD::INTRINSIC_VOID : ISD::INTRINSIC_W_CHAIN;
+    Info.memVT = MVT::v4i32;
+    Info.ptrVal = CI.getArgOperand(0);
+    Info.flags |=
+        IsStore ? MachineMemOperand::MOStore : MachineMemOperand::MOLoad;
+    // Pretend to be atomic so that SIMemoryLegalizer::expandStore sets cache
----------------
Pierre-vh wrote:

I'm not sure about this, is it that much harder to do this the right way ? (add MOFlag operand and handle this separately in SIMemoryLegalizer)

https://github.com/llvm/llvm-project/pull/191390