AMDGPU: Unify LowerGlobalAddress

[jvesely]

Summary

Now that EG/CM supports constant loads the code (in this place) is identical
effectively reverts D19784

[• tstellarAMD]

LGTM.

lib/Target/AMDGPU/AMDGPUISelLowering.cpp

SDValue AMDGPUTargetLowering::LowerGlobalAddress(AMDGPUMachineFunction* MFI,
                                                  SDValue Op,
                                                  SelectionDAG &DAG) const {
 
   const DataLayout &DL = DAG.getDataLayout();
   GlobalAddressSDNode *G = cast<GlobalAddressSDNode>(Op);
   const GlobalValue *GV = G->getGlobal();
 
   switch (G->getAddressSpace()) {
+  case AMDGPUAS::CONSTANT_ADDRESS: {
+    MVT ConstPtrVT = getPointerTy(DL, AMDGPUAS::CONSTANT_ADDRESS);
+    SDValue GA = DAG.getTargetGlobalAddress(GV, SDLoc(G), ConstPtrVT);
+    return DAG.getNode(AMDGPUISD::CONST_DATA_PTR, SDLoc(G), ConstPtrVT, GA);
+  }
   case AMDGPUAS::LOCAL_ADDRESS: {
     // XXX: What does the value of G->getOffset() mean?
     assert(G->getOffset() == 0 &&
          "Do not know what to do with an non-zero offset");
 
     // TODO: We could emit code to handle the initialization somewhere.
     if (hasDefinedInitializer(GV))
       break;

lib/Target/AMDGPU/R600ISelLowering.h

private:
   SDValue lowerPrivateExtLoad(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerLOAD(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerBRCOND(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerTrig(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerSHLParts(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerSRXParts(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerUADDSUBO(SDValue Op, SelectionDAG &DAG,
                         unsigned mainop, unsigned ovf) const;
-  SDValue LowerGlobalAddress(SDValue Op, SelectionDAG &DAG) const;
 
   SDValue stackPtrToRegIndex(SDValue Ptr, unsigned StackWidth,
                                           SelectionDAG &DAG) const;
   void getStackAddress(unsigned StackWidth, unsigned ElemIdx,
                        unsigned &Channel, unsigned &PtrIncr) const;
   bool isZero(SDValue Op) const;
   bool isHWTrueValue(SDValue Op) const;
   bool isHWFalseValue(SDValue Op) const;
 
   SDNode *PostISelFolding(MachineSDNode *N, SelectionDAG &DAG) const override;
 };
 
 } // End namespace llvm;
 
 #endif

lib/Target/AMDGPU/R600ISelLowering.cpp

case ISD::LOAD: {
     SDValue Result = LowerLOAD(Op, DAG);
     assert((!Result.getNode() ||
             Result.getNode()->getNumValues() == 2) &&
            "Load should return a value and a chain");
     return Result;
   }
 
   case ISD::BRCOND: return LowerBRCOND(Op, DAG);
-  case ISD::GlobalAddress: return LowerGlobalAddress(Op, DAG);
+  case ISD::GlobalAddress: return LowerGlobalAddress(MFI, Op, DAG);
   case ISD::FrameIndex: return lowerFrameIndex(Op, DAG);
   case ISD::INTRINSIC_VOID: {
     SDValue Chain = Op.getOperand(0);
     unsigned IntrinsicID =
                          cast<ConstantSDNode>(Op.getOperand(1))->getZExtValue();
     switch (IntrinsicID) {
     case AMDGPUIntrinsic::R600_store_swizzle: {
       SDLoc DL(Op);
return DAG.getNode(
       ISD::SETCC,
       DL,
       MVT::i1,
       Op, DAG.getConstantFP(0.0f, DL, MVT::f32),
       DAG.getCondCode(ISD::SETNE)
       );
 }
 
-SDValue R600TargetLowering::LowerGlobalAddress(SDValue Op, SelectionDAG &DAG) const
-{
-  MachineFunction &MF = DAG.getMachineFunction();
-  R600MachineFunctionInfo *MFI = MF.getInfo<R600MachineFunctionInfo>();
-  GlobalAddressSDNode *G = cast<GlobalAddressSDNode>(Op);
-  if (G->getAddressSpace() != AMDGPUAS::CONSTANT_ADDRESS)
-    return AMDGPUTargetLowering::LowerGlobalAddress(MFI, Op, DAG);
-
-  const DataLayout &DL = DAG.getDataLayout();
-  const GlobalValue *GV = G->getGlobal();
-  MVT ConstPtrVT = getPointerTy(DL, AMDGPUAS::CONSTANT_ADDRESS);
-
-  SDValue GA = DAG.getTargetGlobalAddress(GV, SDLoc(G), ConstPtrVT);
-  return DAG.getNode(AMDGPUISD::CONST_DATA_PTR, SDLoc(G), ConstPtrVT, GA);
-}
-
 SDValue R600TargetLowering::LowerImplicitParameter(SelectionDAG &DAG, EVT VT,
                                                    SDLoc DL,
                                                    unsigned DwordOffset) const {
   unsigned ByteOffset = DwordOffset * 4;
   PointerType * PtrType = PointerType::get(VT.getTypeForEVT(*DAG.getContext()),
                                       AMDGPUAS::CONSTANT_BUFFER_0);
 
   // We shouldn't be using an offset wider than 16-bits for implicit parameters.

lib/Target/AMDGPU/SIISelLowering.h

 #include "AMDGPUISelLowering.h"
 #include "SIInstrInfo.h"
 
 namespace llvm {
 
 class SITargetLowering final : public AMDGPUTargetLowering {
   SDValue LowerParameter(SelectionDAG &DAG, EVT VT, EVT MemVT, SDLoc DL,
                          SDValue Chain, unsigned Offset, bool Signed) const;
-  SDValue LowerGlobalAddress(AMDGPUMachineFunction *MFI, SDValue Op,
-                             SelectionDAG &DAG) const override;
 
   SDValue lowerImplicitZextParam(SelectionDAG &DAG, SDValue Op,
                                  MVT VT, unsigned Offset) const;
 
   SDValue LowerINTRINSIC_WO_CHAIN(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerINTRINSIC_W_CHAIN(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerINTRINSIC_VOID(SDValue Op, SelectionDAG &DAG) const;
   SDValue LowerFrameIndex(SDValue Op, SelectionDAG &DAG) const;

lib/Target/AMDGPU/SIISelLowering.cpp

SDValue SITargetLowering::lowerADDRSPACECAST(SDValue Op,
   const MachineFunction &MF = DAG.getMachineFunction();
   DiagnosticInfoUnsupported InvalidAddrSpaceCast(
     *MF.getFunction(), "invalid addrspacecast", SL.getDebugLoc());
   DAG.getContext()->diagnose(InvalidAddrSpaceCast);
 
   return DAG.getUNDEF(ASC->getValueType(0));
 }
 
-SDValue SITargetLowering::LowerGlobalAddress(AMDGPUMachineFunction *MFI,
-                                             SDValue Op,
-                                             SelectionDAG &DAG) const {
-  GlobalAddressSDNode *GSD = cast<GlobalAddressSDNode>(Op);
-
-  if (GSD->getAddressSpace() != AMDGPUAS::CONSTANT_ADDRESS)
-    return AMDGPUTargetLowering::LowerGlobalAddress(MFI, Op, DAG);
-
-  SDLoc DL(GSD);
-  const GlobalValue *GV = GSD->getGlobal();
-  MVT PtrVT = getPointerTy(DAG.getDataLayout(), GSD->getAddressSpace());
-
-  SDValue GA = DAG.getTargetGlobalAddress(GV, DL, MVT::i32);
-  return DAG.getNode(AMDGPUISD::CONST_DATA_PTR, DL, PtrVT, GA);
-}
-
 SDValue SITargetLowering::copyToM0(SelectionDAG &DAG, SDValue Chain, SDLoc DL,
                                    SDValue V) const {
   // We can't use S_MOV_B32 directly, because there is no way to specify m0 as
   // the destination register.
   //
   // We can't use CopyToReg, because MachineCSE won't combine COPY instructions,
   // so we will end up with redundant moves to m0.
   //