[clang] [llvm] [mlir] [AMDGPU] Add ptr.s.buffer.load intrinsic, use it from Clang (PR #209243)

[Scott Linder via cfe-commits]

================
@@ -10705,33 +10707,58 @@ SDValue SITargetLowering::lowerImage(SDValue Op,
                            NumVDataDwords, IsAtomicPacked16Bit, DL);
 }
 
-SDValue SITargetLowering::lowerSBuffer(EVT VT, SDLoc DL, SDValue Rsrc,
+SDValue SITargetLowering::lowerSBuffer(EVT VT, EVT MemVT, SDLoc DL,
+                                       SDValue Chain, SDValue Rsrc,
                                        SDValue Offset, SDValue CachePolicy,
-                                       SelectionDAG &DAG) const {
+                                       SelectionDAG &DAG,
+                                       MachineMemOperand *MMO) const {
   MachineFunction &MF = DAG.getMachineFunction();
+  bool HasChainResult = MMO != nullptr;
----------------
slinder1 wrote:

Nit: pick either `MMO != nullptr` or `!MMO` for this and the next `if`s test

https://github.com/llvm/llvm-project/pull/209243