AMDGPU/SI: Select mad patterns to v_mac_f32

The two-address instruction pass will convert these back to v_mad_f32
if necessary.

Differential Revision: http://reviews.llvm.org/D11060

llvm-svn: 242038

llvm/lib/Target/AMDGPU/AMDGPUISelDAGToDAG.cpp

@@ -110,8 +110,11 @@ class AMDGPUDAGToDAGISel : public SelectionDAGISel {
                          SDValue &Offset, SDValue &GLC) const;
   SDNode *SelectAddrSpaceCast(SDNode *N);
   bool SelectVOP3Mods(SDValue In, SDValue &Src, SDValue &SrcMods) const;
+  bool SelectVOP3NoMods(SDValue In, SDValue &Src, SDValue &SrcMods) const;
   bool SelectVOP3Mods0(SDValue In, SDValue &Src, SDValue &SrcMods,
                        SDValue &Clamp, SDValue &Omod) const;
+  bool SelectVOP3NoMods0(SDValue In, SDValue &Src, SDValue &SrcMods,
+                         SDValue &Clamp, SDValue &Omod) const;
 
   bool SelectVOP3Mods0Clamp(SDValue In, SDValue &Src, SDValue &SrcMods,
                             SDValue &Omod) const;
@@ -1317,6 +1320,12 @@ bool AMDGPUDAGToDAGISel::SelectVOP3Mods(SDValue In, SDValue &Src,
   return true;
 }
 
+bool AMDGPUDAGToDAGISel::SelectVOP3NoMods(SDValue In, SDValue &Src,
+                                         SDValue &SrcMods) const {
+  bool Res = SelectVOP3Mods(In, Src, SrcMods);
+  return Res && cast<ConstantSDNode>(SrcMods)->isNullValue();
+}
+
 bool AMDGPUDAGToDAGISel::SelectVOP3Mods0(SDValue In, SDValue &Src,
                                          SDValue &SrcMods, SDValue &Clamp,
                                          SDValue &Omod) const {
@@ -1328,6 +1337,16 @@ bool AMDGPUDAGToDAGISel::SelectVOP3Mods0(SDValue In, SDValue &Src,
   return SelectVOP3Mods(In, Src, SrcMods);
 }
 
+bool AMDGPUDAGToDAGISel::SelectVOP3NoMods0(SDValue In, SDValue &Src,
+                                           SDValue &SrcMods, SDValue &Clamp,
+                                           SDValue &Omod) const {
+  bool Res = SelectVOP3Mods0(In, Src, SrcMods, Clamp, Omod);
+
+  return Res && cast<ConstantSDNode>(SrcMods)->isNullValue() &&
+                cast<ConstantSDNode>(Clamp)->isNullValue() &&
+                cast<ConstantSDNode>(Omod)->isNullValue();
+}
+
 bool AMDGPUDAGToDAGISel::SelectVOP3Mods0Clamp(SDValue In, SDValue &Src,
                                               SDValue &SrcMods,
                                               SDValue &Omod) const {

llvm/lib/Target/AMDGPU/SIFoldOperands.cpp

@@ -126,11 +126,42 @@ static bool updateOperand(FoldCandidate &Fold,
   return false;
 }
 
+static bool isUseMIInFoldList(const std::vector<FoldCandidate> &FoldList,
+                              const MachineInstr *MI) {
+  for (auto Candidate : FoldList) {
+    if (Candidate.UseMI == MI)
+      return true;
+  }
+  return false;
+}
+
 static bool tryAddToFoldList(std::vector<FoldCandidate> &FoldList,
                              MachineInstr *MI, unsigned OpNo,
                              MachineOperand *OpToFold,
                              const SIInstrInfo *TII) {
   if (!TII->isOperandLegal(MI, OpNo, OpToFold)) {
+
+    // Special case for v_mac_f32_e64 if we are trying to fold into src2
+    unsigned Opc = MI->getOpcode();
+    if (Opc == AMDGPU::V_MAC_F32_e64 &&
+        (int)OpNo == AMDGPU::getNamedOperandIdx(Opc, AMDGPU::OpName::src2)) {
+      // Check if changing this to a v_mad_f32 instruction will allow us to
+      // fold the operand.
+      MI->setDesc(TII->get(AMDGPU::V_MAD_F32));
+      bool FoldAsMAD = tryAddToFoldList(FoldList, MI, OpNo, OpToFold, TII);
+      if (FoldAsMAD) {
+        MI->untieRegOperand(OpNo);
+        return true;
+      }
+      MI->setDesc(TII->get(Opc));
+    }
+
+    // If we are already folding into another operand of MI, then
+    // we can't commute the instruction, otherwise we risk making the
+    // other fold illegal.
+    if (isUseMIInFoldList(FoldList, MI))
+      return false;
+
     // Operand is not legal, so try to commute the instruction to
     // see if this makes it possible to fold.
     unsigned CommuteIdx0;

llvm/lib/Target/AMDGPU/SIInstrInfo.cpp

@@ -924,7 +924,7 @@ bool SIInstrInfo::FoldImmediate(MachineInstr *UseMI, MachineInstr *DefMI,
     return false;
 
   unsigned Opc = UseMI->getOpcode();
-  if (Opc == AMDGPU::V_MAD_F32) {
+  if (Opc == AMDGPU::V_MAD_F32 || Opc == AMDGPU::V_MAC_F32_e64) {
     // Don't fold if we are using source modifiers. The new VOP2 instructions
     // don't have them.
     if (hasModifiersSet(*UseMI, AMDGPU::OpName::src0_modifiers) ||
@@ -963,9 +963,9 @@ bool SIInstrInfo::FoldImmediate(MachineInstr *UseMI, MachineInstr *DefMI,
       // instead of having to modify in place.
 
       // Remove these first since they are at the end.
-      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(AMDGPU::V_MAD_F32,
+      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(Opc,
                                                       AMDGPU::OpName::omod));
-      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(AMDGPU::V_MAD_F32,
+      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(Opc,
                                                       AMDGPU::OpName::clamp));
 
       unsigned Src1Reg = Src1->getReg();
@@ -980,6 +980,14 @@ bool SIInstrInfo::FoldImmediate(MachineInstr *UseMI, MachineInstr *DefMI,
       Src1->setSubReg(Src2SubReg);
       Src1->setIsKill(Src2->isKill());
 
+      if (Opc == AMDGPU::V_MAC_F32_e64) {
+        UseMI->untieRegOperand(
+          AMDGPU::getNamedOperandIdx(Opc, AMDGPU::OpName::src2));
+      }
+
+      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(Opc,
+                                                      AMDGPU::OpName::src2));
+      // ChangingToImmediate adds Src2 back to the instruction.
       Src2->ChangeToImmediate(Imm);
 
       removeModOperands(*UseMI);
@@ -1010,11 +1018,17 @@ bool SIInstrInfo::FoldImmediate(MachineInstr *UseMI, MachineInstr *DefMI,
       // instead of having to modify in place.
 
       // Remove these first since they are at the end.
-      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(AMDGPU::V_MAD_F32,
+      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(Opc,
                                                       AMDGPU::OpName::omod));
-      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(AMDGPU::V_MAD_F32,
+      UseMI->RemoveOperand(AMDGPU::getNamedOperandIdx(Opc,
                                                       AMDGPU::OpName::clamp));
 
+      if (Opc == AMDGPU::V_MAC_F32_e64) {
+        UseMI->untieRegOperand(
+          AMDGPU::getNamedOperandIdx(Opc, AMDGPU::OpName::src2));
+      }
+
+      // ChangingToImmediate adds Src2 back to the instruction.
       Src2->ChangeToImmediate(Imm);
 
       // These come before src2.
@@ -1126,6 +1140,38 @@ bool SIInstrInfo::areMemAccessesTriviallyDisjoint(MachineInstr *MIa,
   return false;
 }
 
+MachineInstr *SIInstrInfo::convertToThreeAddress(MachineFunction::iterator &MBB,
+                                                MachineBasicBlock::iterator &MI,
+                                                LiveVariables *LV) const {
+
+  switch (MI->getOpcode()) {
+    default: return nullptr;
+    case AMDGPU::V_MAC_F32_e64: break;
+    case AMDGPU::V_MAC_F32_e32: {
+      const MachineOperand *Src0 = getNamedOperand(*MI, AMDGPU::OpName::src0);
+      if (Src0->isImm() && !isInlineConstant(*Src0, 4))
+        return nullptr;
+      break;
+    }
+  }
+
+  const MachineOperand *Dst = getNamedOperand(*MI, AMDGPU::OpName::dst);
+  const MachineOperand *Src0 = getNamedOperand(*MI, AMDGPU::OpName::src0);
+  const MachineOperand *Src1 = getNamedOperand(*MI, AMDGPU::OpName::src1);
+  const MachineOperand *Src2 = getNamedOperand(*MI, AMDGPU::OpName::src2);
+
+  return BuildMI(*MBB, MI, MI->getDebugLoc(), get(AMDGPU::V_MAD_F32))
+                 .addOperand(*Dst)
+                 .addImm(0) // Src0 mods
+                 .addOperand(*Src0)
+                 .addImm(0) // Src1 mods
+                 .addOperand(*Src1)
+                 .addImm(0) // Src mods
+                 .addOperand(*Src2)
+                 .addImm(0)  // clamp
+                 .addImm(0); // omod
+}
+
 bool SIInstrInfo::isInlineConstant(const APInt &Imm) const {
   int64_t SVal = Imm.getSExtValue();
   if (SVal >= -16 && SVal <= 64)

llvm/lib/Target/AMDGPU/SIInstrInfo.h

@@ -144,6 +144,10 @@ class SIInstrInfo : public AMDGPUInstrInfo {
 
   unsigned getMachineCSELookAheadLimit() const override { return 500; }
 
+  MachineInstr *convertToThreeAddress(MachineFunction::iterator &MBB,
+                                      MachineBasicBlock::iterator &MI,
+                                      LiveVariables *LV) const override;
+
   bool isSALU(uint16_t Opcode) const {
     return get(Opcode).TSFlags & SIInstrFlags::SALU;
   }

llvm/lib/Target/AMDGPU/SIInstrInfo.td

@@ -529,9 +529,11 @@ def MUBUFOffset : ComplexPattern<i64, 6, "SelectMUBUFOffset">;
 def MUBUFOffsetAtomic : ComplexPattern<i64, 4, "SelectMUBUFOffset">;
 
 def VOP3Mods0 : ComplexPattern<untyped, 4, "SelectVOP3Mods0">;
+def VOP3NoMods0 : ComplexPattern<untyped, 4, "SelectVOP3NoMods0">;
 def VOP3Mods0Clamp : ComplexPattern<untyped, 3, "SelectVOP3Mods0Clamp">;
 def VOP3Mods0Clamp0OMod : ComplexPattern<untyped, 4, "SelectVOP3Mods0Clamp0OMod">;
 def VOP3Mods  : ComplexPattern<untyped, 2, "SelectVOP3Mods">;
+def VOP3NoMods : ComplexPattern<untyped, 2, "SelectVOP3NoMods">;
 
 //===----------------------------------------------------------------------===//
 // SI assembler operands
@@ -1113,6 +1115,13 @@ def VOP_MADK : VOPProfile <[f32, f32, f32, f32]> {
   field dag Ins = (ins VCSrc_32:$src0, VGPR_32:$vsrc1, u32imm:$src2);
   field string Asm = "$dst, $src0, $vsrc1, $src2";
 }
+def VOP_MAC : VOPProfile <[f32, f32, f32, f32]> {
+  let Ins32 = (ins Src0RC32:$src0, Src1RC32:$src1, VGPR_32:$src2);
+  let Ins64 = getIns64<Src0RC64, Src1RC64, RegisterOperand<VGPR_32>, 3,
+                             HasModifiers>.ret;
+  let Asm32 = getAsm32<2>.ret;
+  let Asm64 = getAsm64<2, HasModifiers>.ret;
+}
 def VOP_F64_F64_F64_F64 : VOPProfile <[f64, f64, f64, f64]>;
 def VOP_I32_I32_I32_I32 : VOPProfile <[i32, i32, i32, i32]>;
 def VOP_I64_I32_I32_I64 : VOPProfile <[i64, i32, i32, i64]>;

llvm/lib/Target/AMDGPU/SIInstructions.td

@@ -1488,7 +1488,10 @@ defm V_AND_B32 : VOP2Inst <vop2<0x1b, 0x13>, "v_and_b32", VOP_I32_I32_I32>;
 defm V_OR_B32 : VOP2Inst <vop2<0x1c, 0x14>, "v_or_b32", VOP_I32_I32_I32>;
 defm V_XOR_B32 : VOP2Inst <vop2<0x1d, 0x15>, "v_xor_b32", VOP_I32_I32_I32>;
 
-defm V_MAC_F32 : VOP2Inst <vop2<0x1f, 0x16>, "v_mac_f32", VOP_F32_F32_F32>;
+let Constraints = "$dst = $src2", DisableEncoding="$src2",
+    isConvertibleToThreeAddress = 1 in {
+defm V_MAC_F32 : VOP2Inst <vop2<0x1f, 0x16>, "v_mac_f32", VOP_MAC>;
+}
 } // End isCommutable = 1
 
 defm V_MADMK_F32 : VOP2MADK <vop2<0x20, 0x17>, "v_madmk_f32">;
@@ -2206,6 +2209,15 @@ def : Pat <
   (V_CNDMASK_B32_e64 $src2, $src1, $src0)
 >;
 
+// Pattern for V_MAC_F32
+def : Pat <
+  (fmad  (VOP3NoMods0 f32:$src0, i32:$src0_modifiers, i1:$clamp, i32:$omod),
+         (VOP3NoMods f32:$src1, i32:$src1_modifiers),
+         (VOP3NoMods f32:$src2, i32:$src2_modifiers)),
+  (V_MAC_F32_e64 $src0_modifiers, $src0, $src1_modifiers, $src1,
+                 $src2_modifiers, $src2, $clamp, $omod)
+>;
+
 /********** ======================= **********/
 /********** Image sampling patterns **********/
 /********** ======================= **********/

llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp

@@ -94,8 +94,15 @@ static bool canShrink(MachineInstr &MI, const SIInstrInfo *TII,
   // is vcc.  We should handle this the same way we handle vopc, by addding
   // a register allocation hint pre-regalloc and then do the shrining
   // post-regalloc.
-  if (Src2)
-    return false;
+  if (Src2) {
+    if (MI.getOpcode() != AMDGPU::V_MAC_F32_e64)
+      return false;
+
+    const MachineOperand *Src2Mod =
+        TII->getNamedOperand(MI, AMDGPU::OpName::src2_modifiers);
+    if (!isVGPR(Src2, TRI, MRI) || (Src2Mod && Src2Mod->getImm() != 0))
+      return false;
+  }
 
   const MachineOperand *Src1 = TII->getNamedOperand(MI, AMDGPU::OpName::src1);
   const MachineOperand *Src1Mod =
@@ -259,6 +266,11 @@ bool SIShrinkInstructions::runOnMachineFunction(MachineFunction &MF) {
       if (Src1)
         Inst32.addOperand(*Src1);
 
+      const MachineOperand *Src2 =
+          TII->getNamedOperand(MI, AMDGPU::OpName::src2);
+      if (Src2)
+        Inst32.addOperand(*Src2);
+
       ++NumInstructionsShrunk;
       MI.eraseFromParent();
 

llvm/test/CodeGen/AMDGPU/fmuladd.ll

@@ -6,7 +6,7 @@ declare i32 @llvm.r600.read.tidig.x() nounwind readnone
 declare float @llvm.fabs.f32(float) nounwind readnone
 
 ; CHECK-LABEL: {{^}}fmuladd_f32:
-; CHECK: v_mad_f32 {{v[0-9]+, v[0-9]+, v[0-9]+, v[0-9]+}}
+; CHECK: v_mac_f32_e32 {{v[0-9]+, v[0-9]+, v[0-9]+}}
 
 define void @fmuladd_f32(float addrspace(1)* %out, float addrspace(1)* %in1,
                          float addrspace(1)* %in2, float addrspace(1)* %in3) {
@@ -34,8 +34,8 @@ define void @fmuladd_f64(double addrspace(1)* %out, double addrspace(1)* %in1,
 ; CHECK-LABEL: {{^}}fmuladd_2.0_a_b_f32
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], 2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], 2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fmuladd_2.0_a_b_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
   %tid = call i32 @llvm.r600.read.tidig.x() nounwind readnone
   %gep.0 = getelementptr float, float addrspace(1)* %out, i32 %tid
@@ -53,8 +53,8 @@ define void @fmuladd_2.0_a_b_f32(float addrspace(1)* %out, float addrspace(1)* %
 ; CHECK-LABEL: {{^}}fmuladd_a_2.0_b_f32
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], 2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], 2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fmuladd_a_2.0_b_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
   %tid = call i32 @llvm.r600.read.tidig.x() nounwind readnone
   %gep.0 = getelementptr float, float addrspace(1)* %out, i32 %tid
@@ -72,8 +72,8 @@ define void @fmuladd_a_2.0_b_f32(float addrspace(1)* %out, float addrspace(1)* %
 ; CHECK-LABEL: {{^}}fadd_a_a_b_f32:
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], 2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], 2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fadd_a_a_b_f32(float addrspace(1)* %out,
                             float addrspace(1)* %in1,
                             float addrspace(1)* %in2) {
@@ -94,8 +94,8 @@ define void @fadd_a_a_b_f32(float addrspace(1)* %out,
 ; CHECK-LABEL: {{^}}fadd_b_a_a_f32:
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], 2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], 2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fadd_b_a_a_f32(float addrspace(1)* %out,
                             float addrspace(1)* %in1,
                             float addrspace(1)* %in2) {
@@ -116,8 +116,8 @@ define void @fadd_b_a_a_f32(float addrspace(1)* %out,
 ; CHECK-LABEL: {{^}}fmuladd_neg_2.0_a_b_f32
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], -2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], -2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fmuladd_neg_2.0_a_b_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
   %tid = call i32 @llvm.r600.read.tidig.x() nounwind readnone
   %gep.0 = getelementptr float, float addrspace(1)* %out, i32 %tid
@@ -136,8 +136,8 @@ define void @fmuladd_neg_2.0_a_b_f32(float addrspace(1)* %out, float addrspace(1
 ; CHECK-LABEL: {{^}}fmuladd_neg_2.0_neg_a_b_f32
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], 2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], 2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fmuladd_neg_2.0_neg_a_b_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
   %tid = call i32 @llvm.r600.read.tidig.x() nounwind readnone
   %gep.0 = getelementptr float, float addrspace(1)* %out, i32 %tid
@@ -158,8 +158,8 @@ define void @fmuladd_neg_2.0_neg_a_b_f32(float addrspace(1)* %out, float addrspa
 ; CHECK-LABEL: {{^}}fmuladd_2.0_neg_a_b_f32
 ; CHECK-DAG: buffer_load_dword [[R1:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64{{$}}
 ; CHECK-DAG: buffer_load_dword [[R2:v[0-9]+]], {{v\[[0-9]+:[0-9]+\]}}, {{s\[[0-9]+:[0-9]+\]}}, 0 addr64 offset:4
-; CHECK: v_mad_f32 [[RESULT:v[0-9]+]], -2.0, [[R1]], [[R2]]
-; CHECK: buffer_store_dword [[RESULT]]
+; CHECK: v_mac_f32_e32 [[R2]], -2.0, [[R1]]
+; CHECK: buffer_store_dword [[R2]]
 define void @fmuladd_2.0_neg_a_b_f32(float addrspace(1)* %out, float addrspace(1)* %in) {
   %tid = call i32 @llvm.r600.read.tidig.x() nounwind readnone
   %gep.0 = getelementptr float, float addrspace(1)* %out, i32 %tid

llvm/test/CodeGen/AMDGPU/llvm.amdgpu.lrp.ll

@@ -5,7 +5,7 @@ declare float @llvm.AMDGPU.lrp(float, float, float) nounwind readnone
 
 ; FUNC-LABEL: {{^}}test_lrp:
 ; SI: v_sub_f32
-; SI: v_mad_f32
+; SI: v_mac_f32_e32
 define void @test_lrp(float addrspace(1)* %out, float %src0, float %src1, float %src2) nounwind {
   %mad = call float @llvm.AMDGPU.lrp(float %src0, float %src1, float %src2) nounwind readnone
   store float %mad, float addrspace(1)* %out, align 4