[FastISel][X86] Add MachineMemOperand to load/store instructions.

This commit adds MachineMemOperands to load and store instructions. This allows
the peephole optimizer to fold load instructions. Unfortunatelly the peephole
optimizer currently doesn't run at -O0.

llvm-svn: 210858

include/llvm/CodeGen/FastISel.h

@@ -377,6 +377,9 @@ protected:
   /// Test whether the given value has exactly one use.
   bool hasTrivialKill(const Value *V) const;
 
+  /// \brief Create a machine mem operand from the given instruction.
+  MachineMemOperand *createMachineMemOperandFor(const Instruction *I) const;
+
 private:
   bool SelectBinaryOp(const User *I, unsigned ISDOpcode);
 

lib/CodeGen/SelectionDAG/FastISel.cpp

@@ -1737,3 +1737,47 @@ bool FastISel::canFoldAddIntoGEP(const User *GEP, const Value *Add) {
   return isa<ConstantInt>(cast<AddOperator>(Add)->getOperand(1));
 }
 
+MachineMemOperand *
+FastISel::createMachineMemOperandFor(const Instruction *I) const {
+  const Value *Ptr;
+  Type *ValTy;
+  unsigned Alignment;
+  unsigned Flags;
+  bool IsVolatile;
+
+  if (const auto *LI = dyn_cast<LoadInst>(I)) {
+    Alignment = LI->getAlignment();
+    IsVolatile = LI->isVolatile();
+    Flags = MachineMemOperand::MOLoad;
+    Ptr = LI->getPointerOperand();
+    ValTy = LI->getType();
+  } else if (const auto *SI = dyn_cast<StoreInst>(I)) {
+    Alignment = SI->getAlignment();
+    IsVolatile = SI->isVolatile();
+    Flags = MachineMemOperand::MOStore;
+    Ptr = SI->getPointerOperand();
+    ValTy = SI->getValueOperand()->getType();
+  } else {
+    return nullptr;
+  }
+
+  bool IsNonTemporal = I->getMetadata("nontemporal") != nullptr;
+  bool IsInvariant = I->getMetadata("invariant.load") != nullptr;
+  const MDNode *TBAAInfo = I->getMetadata(LLVMContext::MD_tbaa);
+  const MDNode *Ranges = I->getMetadata(LLVMContext::MD_range);
+
+  if (Alignment == 0)  // Ensure that codegen never sees alignment 0.
+    Alignment = DL.getABITypeAlignment(ValTy);
+
+  unsigned Size = TM.getDataLayout()->getTypeStoreSize(ValTy);
+
+  if (IsVolatile)
+    Flags |= MachineMemOperand::MOVolatile;
+  if (IsNonTemporal)
+    Flags |= MachineMemOperand::MONonTemporal;
+  if (IsInvariant)
+    Flags |= MachineMemOperand::MOInvariant;
+
+  return FuncInfo.MF->getMachineMemOperand(MachinePointerInfo(Ptr), Flags, Size,
+                                           Alignment, TBAAInfo, Ranges);
+}

lib/Target/X86/X86FastISel.cpp

@@ -78,12 +78,14 @@ public:
 private:
   bool X86FastEmitCompare(const Value *LHS, const Value *RHS, EVT VT);
 
-  bool X86FastEmitLoad(EVT VT, const X86AddressMode &AM, unsigned &RR);
+  bool X86FastEmitLoad(EVT VT, const X86AddressMode &AM, MachineMemOperand *MMO,
+                       unsigned &ResultReg);
 
   bool X86FastEmitStore(EVT VT, const Value *Val, const X86AddressMode &AM,
-                        bool Aligned = false);
+                        MachineMemOperand *MMO = nullptr, bool Aligned = false);
-  bool X86FastEmitStore(EVT VT, unsigned ValReg, const X86AddressMode &AM,
+  bool X86FastEmitStore(EVT VT, unsigned ValReg, bool ValIsKill,
-                        bool Aligned = false);
+                        const X86AddressMode &AM,
+                        MachineMemOperand *MMO = nullptr, bool Aligned = false);
 
   bool X86FastEmitExtend(ISD::NodeType Opc, EVT DstVT, unsigned Src, EVT SrcVT,
                          unsigned &ResultReg);
@@ -180,7 +182,7 @@ bool X86FastISel::isTypeLegal(Type *Ty, MVT &VT, bool AllowI1) {
 /// The address is either pre-computed, i.e. Ptr, or a GlobalAddress, i.e. GV.
 /// Return true and the result register by reference if it is possible.
 bool X86FastISel::X86FastEmitLoad(EVT VT, const X86AddressMode &AM,
-                                  unsigned &ResultReg) {
+                                  MachineMemOperand *MMO, unsigned &ResultReg) {
   // Get opcode and regclass of the output for the given load instruction.
   unsigned Opc = 0;
   const TargetRegisterClass *RC = nullptr;
@@ -228,8 +230,11 @@ bool X86FastISel::X86FastEmitLoad(EVT VT, const X86AddressMode &AM,
   }
 
   ResultReg = createResultReg(RC);
-  addFullAddress(BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt,
+  MachineInstrBuilder MIB =
-                         DbgLoc, TII.get(Opc), ResultReg), AM);
+    BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt, DbgLoc, TII.get(Opc), ResultReg);
+  addFullAddress(MIB, AM);
+  if (MMO)
+    MIB->addMemOperand(*FuncInfo.MF, MMO);
   return true;
 }
 
@@ -237,9 +242,9 @@ bool X86FastISel::X86FastEmitLoad(EVT VT, const X86AddressMode &AM,
 /// type VT. The address is either pre-computed, consisted of a base ptr, Ptr
 /// and a displacement offset, or a GlobalAddress,
 /// i.e. V. Return true if it is possible.
-bool
+bool X86FastISel::X86FastEmitStore(EVT VT, unsigned ValReg, bool ValIsKill,
-X86FastISel::X86FastEmitStore(EVT VT, unsigned ValReg,
+                                   const X86AddressMode &AM,
-                              const X86AddressMode &AM, bool Aligned) {
+                                   MachineMemOperand *MMO, bool Aligned) {
   // Get opcode and regclass of the output for the given store instruction.
   unsigned Opc = 0;
   switch (VT.getSimpleVT().SimpleTy) {
@@ -249,7 +254,8 @@ X86FastISel::X86FastEmitStore(EVT VT, unsigned ValReg,
     // Mask out all but lowest bit.
     unsigned AndResult = createResultReg(&X86::GR8RegClass);
     BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt, DbgLoc,
-            TII.get(X86::AND8ri), AndResult).addReg(ValReg).addImm(1);
+            TII.get(X86::AND8ri), AndResult)
+      .addReg(ValReg, getKillRegState(ValIsKill)).addImm(1);
     ValReg = AndResult;
   }
   // FALLTHROUGH, handling i1 as i8.
@@ -288,13 +294,18 @@ X86FastISel::X86FastEmitStore(EVT VT, unsigned ValReg,
     break;
   }
 
-  addFullAddress(BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt,
+  MachineInstrBuilder MIB =
-                         DbgLoc, TII.get(Opc)), AM).addReg(ValReg);
+    BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt, DbgLoc, TII.get(Opc));
+  addFullAddress(MIB, AM).addReg(ValReg, getKillRegState(ValIsKill));
+  if (MMO)
+    MIB->addMemOperand(*FuncInfo.MF, MMO);
+
   return true;
 }
 
 bool X86FastISel::X86FastEmitStore(EVT VT, const Value *Val,
-                                   const X86AddressMode &AM, bool Aligned) {
+                                   const X86AddressMode &AM,
+                                   MachineMemOperand *MMO, bool Aligned) {
   // Handle 'null' like i32/i64 0.
   if (isa<ConstantPointerNull>(Val))
     Val = Constant::getNullValue(DL.getIntPtrType(Val->getContext()));
@@ -317,10 +328,12 @@ bool X86FastISel::X86FastEmitStore(EVT VT, const Value *Val,
     }
 
     if (Opc) {
-      addFullAddress(BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt,
+      MachineInstrBuilder MIB =
-                             DbgLoc, TII.get(Opc)), AM)
+        BuildMI(*FuncInfo.MBB, FuncInfo.InsertPt, DbgLoc, TII.get(Opc));
-                             .addImm(Signed ? (uint64_t) CI->getSExtValue() :
+      addFullAddress(MIB, AM).addImm(Signed ? (uint64_t) CI->getSExtValue()
-                                              CI->getZExtValue());
+                                            : CI->getZExtValue());
+      if (MMO)
+        MIB->addMemOperand(*FuncInfo.MF, MMO);
       return true;
     }
   }
@@ -329,7 +342,8 @@ bool X86FastISel::X86FastEmitStore(EVT VT, const Value *Val,
   if (ValReg == 0)
     return false;
 
-  return X86FastEmitStore(VT, ValReg, AM, Aligned);
+  bool ValKill = hasTrivialKill(Val);
+  return X86FastEmitStore(VT, ValReg, ValKill, AM, MMO, Aligned);
 }
 
 /// X86FastEmitExtend - Emit a machine instruction to extend a value Src of
@@ -740,19 +754,24 @@ bool X86FastISel::X86SelectStore(const Instruction *I) {
   if (S->isAtomic())
     return false;
 
-  unsigned SABIAlignment =
+  const Value *Val = S->getValueOperand();
-    DL.getABITypeAlignment(S->getValueOperand()->getType());
+  const Value *Ptr = S->getPointerOperand();
-  bool Aligned = S->getAlignment() == 0 || S->getAlignment() >= SABIAlignment;
 
   MVT VT;
-  if (!isTypeLegal(I->getOperand(0)->getType(), VT, /*AllowI1=*/true))
+  if (!isTypeLegal(Val->getType(), VT, /*AllowI1=*/true))
     return false;
 
+  unsigned Alignment = S->getAlignment();
+  unsigned ABIAlignment = DL.getABITypeAlignment(Val->getType());
+  if (Alignment == 0)  // Ensure that codegen never sees alignment 0
+    Alignment = ABIAlignment;
+  bool Aligned = Alignment >= ABIAlignment;
+
   X86AddressMode AM;
-  if (!X86SelectAddress(I->getOperand(1), AM))
+  if (!X86SelectAddress(Ptr, AM))
     return false;
 
-  return X86FastEmitStore(VT, I->getOperand(0), AM, Aligned);
+  return X86FastEmitStore(VT, Val, AM, createMachineMemOperandFor(I), Aligned);
 }
 
 /// X86SelectRet - Select and emit code to implement ret instructions.
@@ -887,25 +906,29 @@ bool X86FastISel::X86SelectRet(const Instruction *I) {
 
 /// X86SelectLoad - Select and emit code to implement load instructions.
 ///
-bool X86FastISel::X86SelectLoad(const Instruction *I)  {
+bool X86FastISel::X86SelectLoad(const Instruction *I) {
+  const LoadInst *LI = cast<LoadInst>(I);
+
   // Atomic loads need special handling.
-  if (cast<LoadInst>(I)->isAtomic())
+  if (LI->isAtomic())
     return false;
 
   MVT VT;
-  if (!isTypeLegal(I->getType(), VT, /*AllowI1=*/true))
+  if (!isTypeLegal(LI->getType(), VT, /*AllowI1=*/true))
     return false;
 
+  const Value *Ptr = LI->getPointerOperand();
+
   X86AddressMode AM;
-  if (!X86SelectAddress(I->getOperand(0), AM))
+  if (!X86SelectAddress(Ptr, AM))
     return false;
 
   unsigned ResultReg = 0;
-  if (X86FastEmitLoad(VT, AM, ResultReg)) {
+  if (!X86FastEmitLoad(VT, AM, createMachineMemOperandFor(LI), ResultReg))
-    UpdateValueMap(I, ResultReg);
+    return false;
-    return true;
+
-  }
+  UpdateValueMap(I, ResultReg);
-  return false;
+  return true;
 }
 
 static unsigned X86ChooseCmpOpcode(EVT VT, const X86Subtarget *Subtarget) {
@@ -1624,8 +1647,8 @@ bool X86FastISel::TryEmitSmallMemcpy(X86AddressMode DestAM,
     }
 
     unsigned Reg;
-    bool RV = X86FastEmitLoad(VT, SrcAM, Reg);
+    bool RV = X86FastEmitLoad(VT, SrcAM, nullptr, Reg);
-    RV &= X86FastEmitStore(VT, Reg, DestAM);
+    RV &= X86FastEmitStore(VT, Reg, /*Kill=*/true, DestAM);
     assert(RV && "Failed to emit load or store??");
 
     unsigned Size = VT.getSizeInBits()/8;
@@ -2322,7 +2345,7 @@ bool X86FastISel::DoSelectCall(const Instruction *I, const char *MemIntName) {
         if (!X86FastEmitStore(ArgVT, ArgVal, AM))
           return false;
       } else {
-        if (!X86FastEmitStore(ArgVT, Arg, AM))
+        if (!X86FastEmitStore(ArgVT, Arg, /*ValIsKill=*/false, AM))
           return false;
       }
     }
@@ -2719,8 +2742,9 @@ unsigned X86FastISel::TargetMaterializeFloatZero(const ConstantFP *CF) {
 
 bool X86FastISel::tryToFoldLoadIntoMI(MachineInstr *MI, unsigned OpNo,
                                       const LoadInst *LI) {
+  const Value *Ptr = LI->getPointerOperand();
   X86AddressMode AM;
-  if (!X86SelectAddress(LI->getOperand(0), AM))
+  if (!X86SelectAddress(Ptr, AM))
     return false;
 
   const X86InstrInfo &XII = (const X86InstrInfo&)TII;
@@ -2728,13 +2752,18 @@ bool X86FastISel::tryToFoldLoadIntoMI(MachineInstr *MI, unsigned OpNo,
   unsigned Size = DL.getTypeAllocSize(LI->getType());
   unsigned Alignment = LI->getAlignment();
 
+  if (Alignment == 0)  // Ensure that codegen never sees alignment 0
+    Alignment = DL.getABITypeAlignment(LI->getType());
+
   SmallVector<MachineOperand, 8> AddrOps;
   AM.getFullAddress(AddrOps);
 
   MachineInstr *Result =
     XII.foldMemoryOperandImpl(*FuncInfo.MF, MI, OpNo, AddrOps, Size, Alignment);
-  if (!Result) return false;
+  if (!Result)
+    return false;
 
+  Result->addMemOperand(*FuncInfo.MF, createMachineMemOperandFor(LI));
   FuncInfo.MBB->insert(FuncInfo.InsertPt, Result);
   MI->eraseFromParent();
   return true;

test/CodeGen/X86/fast-isel-fold-mem.ll

@@ -0,0 +1,12 @@
+; RUN: llc < %s                             -mtriple=x86_64-apple-darwin | FileCheck %s
+; RUN: llc < %s -fast-isel -fast-isel-abort -mtriple=x86_64-apple-darwin | FileCheck %s
+
+define i64 @fold_load(i64* %a, i64 %b) {
+; CHECK-LABEL: fold_load
+; CHECK:       addq  (%rdi), %rsi
+; CHECK-NEXT:  movq  %rsi, %rax
+  %1 = load i64* %a, align 8
+  %2 = add i64 %1, %b
+  ret i64 %2
+}
+