[NFCI][CostModel] Refactor getIntrinsicInstrCost

[samparker]

Summary

Combine the two API calls into one by introducing a structure to hold the relevant data. This has the added benefit of moving the boiler plate code for arguments and flags, into the constructors and hopefully making the scalar cost business easier to read.

[samparker]

A note to @anwel and @dmgreen - the only test that I had trouble with was mve-gather-scatters, as this patch can make it easy to provide more context, so I wondered what you thought of the changes that occurred:

diff --git a/llvm/test/Analysis/CostModel/ARM/mve-gather-scatter-cost.ll b/llvm/test/Analysis/CostModel/ARM/mve-gather-scatter-cost.ll
index b86b2abc88f..a5f2f325cf0 100644
--- a/llvm/test/Analysis/CostModel/ARM/mve-gather-scatter-cost.ll
+++ b/llvm/test/Analysis/CostModel/ARM/mve-gather-scatter-cost.ll
@@ -271,14 +271,14 @@ define void @gep_v4i16(i16* %base, <4 x i32> %ind32, <4 x i16> %ind16, <4 x i1>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %res3 = call <4 x i16> @llvm.masked.gather.v4i16.v4p0i16(<4 x i16*> %gep3, i32 2, <4 x i1> %mask, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: call void @llvm.masked.scatter.v4i16.v4p0i16(<4 x i16> %res3, <4 x i16*> %gep3, i32 2, <4 x i1> %mask)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %gep5 = getelementptr i16, i16* %base, <4 x i16> %ind16
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %res5 = call <4 x i16> @llvm.masked.gather.v4i16.v4p0i16(<4 x i16*> %gep5, i32 2, <4 x i1> %mask, <4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %res5 = call <4 x i16> @llvm.masked.gather.v4i16.v4p0i16(<4 x i16*> %gep5, i32 2, <4 x i1> %mask, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %res5zext = zext <4 x i16> %res5 to <4 x i32>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %res5trunc = trunc <4 x i32> %res5zext to <4 x i16>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.masked.scatter.v4i16.v4p0i16(<4 x i16> %res5trunc, <4 x i16*> %gep5, i32 4, <4 x i1> %mask)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %res6 = call <4 x i16> @llvm.masked.gather.v4i16.v4p0i16(<4 x i16*> %gep5, i32 2, <4 x i1> %mask, <4 x i16> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: call void @llvm.masked.scatter.v4i16.v4p0i16(<4 x i16> %res5trunc, <4 x i16*> %gep5, i32 4, <4 x i1> %mask)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %res6 = call <4 x i16> @llvm.masked.gather.v4i16.v4p0i16(<4 x i16*> %gep5, i32 2, <4 x i1> %mask, <4 x i16> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %res6sext = sext <4 x i16> %res6 to <4 x i32>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %res6trunc = trunc <4 x i32> %res6sext to <4 x i16>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.masked.scatter.v4i16.v4p0i16(<4 x i16> %res6trunc, <4 x i16*> %gep5, i32 4, <4 x i1> %mask)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: call void @llvm.masked.scatter.v4i16.v4p0i16(<4 x i16> %res6trunc, <4 x i16*> %gep5, i32 4, <4 x i1> %mask)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
 
@@ -318,14 +318,14 @@ define void @gep_v4i16(i16* %base, <4 x i32> %ind32, <4 x i16> %ind16, <4 x i1>
 define void @gep_v4i8(i8* %base, <4 x i8> %ind8, <4 x i1> %mask)  {
 ; CHECK-LABEL: 'gep_v4i8'
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %gep5 = getelementptr i8, i8* %base, <4 x i8> %ind8
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %res5 = call <4 x i8> @llvm.masked.gather.v4i8.v4p0i8(<4 x i8*> %gep5, i32 2, <4 x i1> %mask, <4 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %res5 = call <4 x i8> @llvm.masked.gather.v4i8.v4p0i8(<4 x i8*> %gep5, i32 2, <4 x i1> %mask, <4 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %res5zext = zext <4 x i8> %res5 to <4 x i32>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %res5trunc = trunc <4 x i32> %res5zext to <4 x i8>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.masked.scatter.v4i8.v4p0i8(<4 x i8> %res5trunc, <4 x i8*> %gep5, i32 4, <4 x i1> %mask)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %res6 = call <4 x i8> @llvm.masked.gather.v4i8.v4p0i8(<4 x i8*> %gep5, i32 2, <4 x i1> %mask, <4 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: call void @llvm.masked.scatter.v4i8.v4p0i8(<4 x i8> %res5trunc, <4 x i8*> %gep5, i32 4, <4 x i1> %mask)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: %res6 = call <4 x i8> @llvm.masked.gather.v4i8.v4p0i8(<4 x i8*> %gep5, i32 2, <4 x i1> %mask, <4 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: %res6sext = sext <4 x i8> %res6 to <4 x i32>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %res6trunc = trunc <4 x i32> %res6sext to <4 x i8>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 4 for instruction: call void @llvm.masked.scatter.v4i8.v4p0i8(<4 x i8> %res6trunc, <4 x i8*> %gep5, i32 4, <4 x i1> %mask)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 36 for instruction: call void @llvm.masked.scatter.v4i8.v4p0i8(<4 x i8> %res6trunc, <4 x i8*> %gep5, i32 4, <4 x i1> %mask)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;
 
@@ -507,14 +507,14 @@ define void @gep_v8i8(i8* %base, <8 x i8> %ind8, <8 x i1> %mask)  {
 ; CHECK-LABEL: 'gep_v8i8'
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 18 for instruction: %indzext = zext <8 x i8> %ind8 to <8 x i32>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %gep5 = getelementptr i8, i8* %base, <8 x i32> %indzext
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %res5 = call <8 x i8> @llvm.masked.gather.v8i8.v8p0i8(<8 x i8*> %gep5, i32 2, <8 x i1> %mask, <8 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %res5 = call <8 x i8> @llvm.masked.gather.v8i8.v8p0i8(<8 x i8*> %gep5, i32 2, <8 x i1> %mask, <8 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %res5zext = zext <8 x i8> %res5 to <8 x i16>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %res5trunc = trunc <8 x i16> %res5zext to <8 x i8>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v8i8.v8p0i8(<8 x i8> %res5trunc, <8 x i8*> %gep5, i32 4, <8 x i1> %mask)
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: %res6 = call <8 x i8> @llvm.masked.gather.v8i8.v8p0i8(<8 x i8*> %gep5, i32 2, <8 x i1> %mask, <8 x i8> undef)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: call void @llvm.masked.scatter.v8i8.v8p0i8(<8 x i8> %res5trunc, <8 x i8*> %gep5, i32 4, <8 x i1> %mask)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: %res6 = call <8 x i8> @llvm.masked.gather.v8i8.v8p0i8(<8 x i8*> %gep5, i32 2, <8 x i1> %mask, <8 x i8> undef)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 2 for instruction: %res6sext = sext <8 x i8> %res6 to <8 x i16>
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 0 for instruction: %res6trunc = trunc <8 x i16> %res6sext to <8 x i8>
-; CHECK-NEXT:  Cost Model: Found an estimated cost of 8 for instruction: call void @llvm.masked.scatter.v8i8.v8p0i8(<8 x i8> %res6trunc, <8 x i8*> %gep5, i32 4, <8 x i1> %mask)
+; CHECK-NEXT:  Cost Model: Found an estimated cost of 136 for instruction: call void @llvm.masked.scatter.v8i8.v8p0i8(<8 x i8> %res6trunc, <8 x i8*> %gep5, i32 4, <8 x i1> %mask)
 ; CHECK-NEXT:  Cost Model: Found an estimated cost of 1 for instruction: ret void
 ;

[anwel]

Hi Sam, those tests are checking whether the cost model calculates the costs for zext(gather(...)) and scatter(trunc(..)) correctly, because these can be integrated in the MVE intrinsics.
That calculation relies on I being available. I see that you did integrate I in IntrinsicCostAttributes, as an IntrinsicInst; the gather/scatter cost function was taking an Instruction because when it is called we might either be looking at a gather/scatter intrinsic or still at an LLVM load/store, which I think is no intrinsic?
I can't see where it happens, but my guess is that the instruction that ends up being handed over to getGatherScatterOpCost in BasicTTIImpl::getIntrinsicInstrCost doesn't match what was expected anymore.

[samparker]

I think it's also because there's two cost calculation paths, one that is purely typed based and one that looks at the arguments too. I think because of all the nesting, the arguments can get lost too with the original code. If those cost changes look reasonable to you I'll add them in the follow-up patch.

[dmgreen]

I'm not sure I understand. Are you getting those changes with this patch? They would seem worse to me (but MVE gathers are a bit finicky so there might be something subtly wrong with them).

The ones you point to changing look like they should be cheap, not scalarized.

[samparker]

Are you getting those changes with this patch?

Sorry, that definitely wasn't clear! No, what the patch shows is what you get, hopefully NFC, it's just I found some test changes before fixing them and didn't know if they were really 'fixes' or not.

[dfukalov]

I like the idea to simplify cost model implementation since it was not trivial to make even a small changes.

llvm/include/llvm/Analysis/TargetTransformInfo.h

153–154

Regarding the way previous implementation used arrays of types and args, it seems these two functions can return references to const vectors and may be const too?

1054–1056

And almost the same question as previous: are these structures intended to be modifiable?
Perhaps, if there are no plans to use them to pass data back from calls, these references should be const too?

[samparker]

Thanks for taking a look @dfukalov, I originally tried consts and it was a bit (more) pain. But I will have another go.

[samparker]

Well that wasn't painful in the end. I've changed the two remaining methods to const and now passing the object as const too.

llvm/include/llvm/Analysis/TargetTransformInfo.h

bool PerformEntryTest = false;  // Generate the intrinsic which also performs
                                   // icmp ne zero on the loop counter value and
                                   // produces an i1 to guard the loop entry.
   bool isHardwareLoopCandidate(ScalarEvolution &SE, LoopInfo &LI,
                                DominatorTree &DT, bool ForceNestedLoop = false,
                                bool ForceHardwareLoopPHI = false);
   bool canAnalyze(LoopInfo &LI);
 };
 
+class IntrinsicCostAttributes {
+  const IntrinsicInst *II = nullptr;
+  Type *RetTy = nullptr;
+  Intrinsic::ID IID;
+  SmallVector<Type *, 4> ParamTys;
+  SmallVector<Value *, 4> Arguments;
+  FastMathFlags FMF;
+  unsigned VF = 1;
+  // If ScalarizationCost is UINT_MAX, the cost of scalarizing the
+  // arguments and the return value will be computed based on types.
+  unsigned ScalarizationCost = std::numeric_limits<unsigned>::max();
+
+public:
+  IntrinsicCostAttributes(const IntrinsicInst &I);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, CallInst &CI,
+                          unsigned Factor);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, CallInst &CI,
+                          unsigned Factor, unsigned ScalarCost);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                          ArrayRef<Type *> Tys, FastMathFlags Flags);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                          ArrayRef<Type *> Tys, FastMathFlags Flags,
+                          unsigned ScalarCost);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                          ArrayRef<Type *> Tys, FastMathFlags Flags,
+                          unsigned ScalarCost,
+                          const IntrinsicInst *I);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                          ArrayRef<Type *> Tys);
+
+  IntrinsicCostAttributes(Intrinsic::ID Id, Type *Ty,
+                          ArrayRef<Value *> Args);
+
+  Intrinsic::ID getID() const { return IID; }
+  const IntrinsicInst *getInst() const { return II; }
+  Type *getReturnType() const { return RetTy; }
+  unsigned getVectorFactor() const { return VF; }
+  FastMathFlags getFlags() const { return FMF; }
+  unsigned getScalarizationCost() const { return ScalarizationCost; }
+  const SmallVectorImpl<Value *> &getArgs() const { return Arguments; }
+  const SmallVectorImpl<Type *> &getArgTypes() const { return ParamTys; }
+
+  bool isTypeBasedOnly() const {
+    return Arguments.empty();
+  }
+
+  bool skipScalarizationCost() const {
+    return ScalarizationCost != std::numeric_limits<unsigned>::max();
+  }
+};
+
 class TargetTransformInfo;
 typedef TargetTransformInfo TTI;
 
 /// This pass provides access to the codegen interfaces that are needed
 /// for IR-level transformations.
 class TargetTransformInfo {
 public:
   /// Construct a TTI object using a type implementing the \c Concept
int getArithmeticReductionCost(
     TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput) const;
 
   int getMinMaxReductionCost(
     VectorType *Ty, VectorType *CondTy, bool IsPairwiseForm, bool IsUnsigned,
     TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput) const;
 
   /// \returns The cost of Intrinsic instructions. Analyses the real arguments.
   /// Three cases are handled: 1. scalar instruction 2. vector instruction
-  /// 3. scalar instruction which is to be vectorized with VF.
-  /// I is the optional original context instruction holding the call to the
-  /// intrinsic
-  int getIntrinsicInstrCost(
-    Intrinsic::ID ID, Type *RetTy, ArrayRef<Value *> Args,
-    FastMathFlags FMF, unsigned VF = 1,
-    TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-    const Instruction *I = nullptr) const;
-
-  /// \returns The cost of Intrinsic instructions. Types analysis only.
-  /// If ScalarizationCostPassed is UINT_MAX, the cost of scalarizing the
-  /// arguments and the return value will be computed based on types.
-  /// I is the optional original context instruction holding the call to the
-  /// intrinsic
-  int getIntrinsicInstrCost(
-    Intrinsic::ID ID, Type *RetTy, ArrayRef<Type *> Tys, FastMathFlags FMF,
-    unsigned ScalarizationCostPassed = UINT_MAX,
-    TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-    const Instruction *I = nullptr) const;
+  /// 3. scalar instruction which is to be vectorized.
+  int getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                            TTI::TargetCostKind CostKind) const;
 
   /// \returns The cost of Call instructions.
   int getCallInstrCost(Function *F, Type *RetTy, ArrayRef<Type *> Tys,
                  TTI::TargetCostKind CostKind = TTI::TCK_SizeAndLatency) const;
 
   /// \returns The number of pieces into which the provided type must be
   /// split during legalization. Zero is returned when the answer is unknown.
   unsigned getNumberOfParts(Type *Tp) const;
getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy, unsigned Factor,
                              bool UseMaskForCond = false,
                              bool UseMaskForGaps = false) = 0;
   virtual int getArithmeticReductionCost(unsigned Opcode, VectorType *Ty,
                                          bool IsPairwiseForm,
                                          TTI::TargetCostKind CostKind) = 0;
   virtual int getMinMaxReductionCost(VectorType *Ty, VectorType *CondTy,
                                      bool IsPairwiseForm, bool IsUnsigned,
                                      TTI::TargetCostKind CostKind) = 0;
-  virtual int getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                    ArrayRef<Type *> Tys, FastMathFlags FMF,
-                                    unsigned ScalarizationCostPassed,
-                                    TTI::TargetCostKind CostKind,
-                                    const Instruction *I) = 0;
-  virtual int getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                    ArrayRef<Value *> Args, FastMathFlags FMF,
-                                    unsigned VF,
-                                    TTI::TargetCostKind CostKind,
-                                    const Instruction *I) = 0;
+  virtual int getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                    TTI::TargetCostKind CostKind) = 0;
   virtual int getCallInstrCost(Function *F, Type *RetTy,
                                ArrayRef<Type *> Tys,
                                TTI::TargetCostKind CostKind) = 0;
   virtual unsigned getNumberOfParts(Type *Tp) = 0;
   virtual int getAddressComputationCost(Type *Ty, ScalarEvolution *SE,
                                         const SCEV *Ptr) = 0;
   virtual unsigned getCostOfKeepingLiveOverCall(ArrayRef<Type *> Tys) = 0;
   virtual bool getTgtMemIntrinsic(IntrinsicInst *Inst,
return Impl.getArithmeticReductionCost(Opcode, Ty, IsPairwiseForm,
                                            CostKind);
   }
   int getMinMaxReductionCost(VectorType *Ty, VectorType *CondTy,
                              bool IsPairwiseForm, bool IsUnsigned,
                              TTI::TargetCostKind CostKind) override {
     return Impl.getMinMaxReductionCost(Ty, CondTy, IsPairwiseForm, IsUnsigned,
                                        CostKind);
   }
-  int getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy, ArrayRef<Type *> Tys,
-                            FastMathFlags FMF, unsigned ScalarizationCostPassed,
-                            TTI::TargetCostKind CostKind,
-                            const Instruction *I) override {
-    return Impl.getIntrinsicInstrCost(ID, RetTy, Tys, FMF,
-                                      ScalarizationCostPassed, CostKind, I);
-  }
-  int getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                            ArrayRef<Value *> Args, FastMathFlags FMF,
-                            unsigned VF,
-                            TTI::TargetCostKind CostKind,
-                            const Instruction *I) override {
-    return Impl.getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF, CostKind, I);
+  int getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                            TTI::TargetCostKind CostKind) override {
+    return Impl.getIntrinsicInstrCost(ICA, CostKind);
   }
   int getCallInstrCost(Function *F, Type *RetTy,
                        ArrayRef<Type *> Tys,
                        TTI::TargetCostKind CostKind) override {
     return Impl.getCallInstrCost(F, RetTy, Tys, CostKind);
   }
   unsigned getNumberOfParts(Type *Tp) override {
     return Impl.getNumberOfParts(Tp);

llvm/include/llvm/Analysis/TargetTransformInfoImpl.h

unsigned getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
                                       ArrayRef<unsigned> Indices,
                                       unsigned Alignment, unsigned AddressSpace,
                                       TTI::TargetCostKind CostKind,
                                       bool UseMaskForCond,
                                       bool UseMaskForGaps) {
     return 1;
   }
 
-  unsigned getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                 ArrayRef<Type *> Tys, FastMathFlags FMF,
-                                 unsigned ScalarizationCostPassed,
-                                 TTI::TargetCostKind CostKind,
-                                 const Instruction *I) {
-    return 1;
-  }
-  unsigned getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                 ArrayRef<Value *> Args, FastMathFlags FMF,
-                                 unsigned VF, TTI::TargetCostKind CostKind,
-                                 const Instruction *I) {
+  unsigned getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                 TTI::TargetCostKind CostKind) {
     return 1;
   }
 
   unsigned getCallInstrCost(Function *F, Type *RetTy, ArrayRef<Type *> Tys,
                             TTI::TargetCostKind CostKind) {
     return 1;
   }
 

llvm/include/llvm/CodeGen/BasicTTIImpl.h

unsigned getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
     if (UseMaskForGaps)
       Cost += static_cast<T *>(this)->getArithmeticInstrCost(
           BinaryOperator::And, MaskVT, CostKind);
 
     return Cost;
   }
 
   /// Get intrinsic cost based on arguments.
-  unsigned getIntrinsicInstrCost(
-      Intrinsic::ID IID, Type *RetTy, ArrayRef<Value *> Args,
-      FastMathFlags FMF, unsigned VF = 1,
-      TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-      const Instruction *I = nullptr) {
+  unsigned getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                 TTI::TargetCostKind CostKind) {
+
+    // TODO: Combine these two logic paths.
+    if (ICA.isTypeBasedOnly())
+      return getTypeBasedIntrinsicInstrCost(ICA, CostKind);
+
+    Intrinsic::ID IID = ICA.getID();
+    const IntrinsicInst *I = ICA.getInst();
+    Type *RetTy = ICA.getReturnType();
+    const SmallVectorImpl<Value *> &Args = ICA.getArgs();
+    unsigned VF = ICA.getVectorFactor();
+    FastMathFlags FMF = ICA.getFlags();
 
     unsigned RetVF =
         (RetTy->isVectorTy() ? cast<VectorType>(RetTy)->getNumElements() : 1);
     assert((RetVF == 1 || VF == 1) && "VF > 1 and RetVF is a vector type");
     auto *ConcreteTTI = static_cast<T *>(this);
 
     switch (IID) {
     default: {
default: {
       if (RetVF > 1 || VF > 1) {
         ScalarizationCost = 0;
         if (!RetTy->isVoidTy())
           ScalarizationCost +=
               getScalarizationOverhead(cast<VectorType>(RetTy), true, false);
         ScalarizationCost += getOperandsScalarizationOverhead(Args, VF);
       }
 
-      return ConcreteTTI->getIntrinsicInstrCost(IID, RetTy, Types, FMF,
-                                                ScalarizationCost, CostKind,
-                                                I);
+      IntrinsicCostAttributes Attrs(IID, RetTy, Types, FMF,
+                                    ScalarizationCost, I);
+      return ConcreteTTI->getIntrinsicInstrCost(Attrs, CostKind);
     }
     case Intrinsic::masked_scatter: {
       assert(VF == 1 && "Can't vectorize types here.");
       Value *Mask = Args[3];
       bool VarMask = !isa<Constant>(Mask);
       unsigned Alignment = cast<ConstantInt>(Args[2])->getZExtValue();
       return ConcreteTTI->getGatherScatterOpCost(Instruction::Store,
                                                  Args[0]->getType(), Args[1],
unsigned getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
     case Intrinsic::experimental_vector_reduce_xor:
     case Intrinsic::experimental_vector_reduce_v2_fadd:
     case Intrinsic::experimental_vector_reduce_v2_fmul:
     case Intrinsic::experimental_vector_reduce_smax:
     case Intrinsic::experimental_vector_reduce_smin:
     case Intrinsic::experimental_vector_reduce_fmax:
     case Intrinsic::experimental_vector_reduce_fmin:
     case Intrinsic::experimental_vector_reduce_umax:
-    case Intrinsic::experimental_vector_reduce_umin:
-      return getIntrinsicInstrCost(IID, RetTy, Args[0]->getType(), FMF, 1,
-                                   CostKind, I);
+    case Intrinsic::experimental_vector_reduce_umin: {
+      IntrinsicCostAttributes Attrs(IID, RetTy, Args[0]->getType(), FMF, 1, I);
+      return getIntrinsicInstrCost(Attrs, CostKind);
+    }
     case Intrinsic::fshl:
     case Intrinsic::fshr: {
       Value *X = Args[0];
       Value *Y = Args[1];
       Value *Z = Args[2];
       TTI::OperandValueProperties OpPropsX, OpPropsY, OpPropsZ, OpPropsBW;
       TTI::OperandValueKind OpKindX = TTI::getOperandInfo(X, OpPropsX);
       TTI::OperandValueKind OpKindY = TTI::getOperandInfo(Y, OpPropsY);
unsigned getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
     }
     }
   }
 
   /// Get intrinsic cost based on argument types.
   /// If ScalarizationCostPassed is std::numeric_limits<unsigned>::max(), the
   /// cost of scalarizing the arguments and the return value will be computed
   /// based on types.
-  unsigned getIntrinsicInstrCost(
-      Intrinsic::ID IID, Type *RetTy, ArrayRef<Type *> Tys, FastMathFlags FMF,
-      unsigned ScalarizationCostPassed = std::numeric_limits<unsigned>::max(),
-      TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-      const Instruction *I = nullptr) {
+  unsigned getTypeBasedIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                          TTI::TargetCostKind CostKind) {
+
     auto *ConcreteTTI = static_cast<T *>(this);
+
+    Intrinsic::ID IID = ICA.getID();
+    Type *RetTy = ICA.getReturnType();
+    const SmallVectorImpl<Type *> &Tys = ICA.getArgTypes();
+    FastMathFlags FMF = ICA.getFlags();
+    unsigned ScalarizationCostPassed = ICA.getScalarizationCost();
+    bool SkipScalarizationCost = ICA.skipScalarizationCost();
+
     auto *VecOpTy = Tys.empty() ? nullptr : dyn_cast<VectorType>(Tys[0]);
 
     SmallVector<unsigned, 2> ISDs;
     unsigned SingleCallCost = 10; // Library call cost. Make it expensive.
     switch (IID) {
     default: {
       // Assume that we need to scalarize this intrinsic.
       unsigned ScalarizationCost = ScalarizationCostPassed;
       unsigned ScalarCalls = 1;
       Type *ScalarRetTy = RetTy;
       if (auto *RetVTy = dyn_cast<VectorType>(RetTy)) {
-        if (ScalarizationCostPassed == std::numeric_limits<unsigned>::max())
+        if (!SkipScalarizationCost)
           ScalarizationCost = getScalarizationOverhead(RetVTy, true, false);
         ScalarCalls = std::max(ScalarCalls, RetVTy->getNumElements());
         ScalarRetTy = RetTy->getScalarType();
       }
       SmallVector<Type *, 4> ScalarTys;
       for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) {
         Type *Ty = Tys[i];
         if (auto *VTy = dyn_cast<VectorType>(Ty)) {
-          if (ScalarizationCostPassed == std::numeric_limits<unsigned>::max())
+          if (!SkipScalarizationCost)
             ScalarizationCost += getScalarizationOverhead(VTy, false, true);
           ScalarCalls = std::max(ScalarCalls, VTy->getNumElements());
           Ty = Ty->getScalarType();
         }
         ScalarTys.push_back(Ty);
       }
       if (ScalarCalls == 1)
         return 1; // Return cost of a scalar intrinsic. Assume it to be cheap.
 
+      IntrinsicCostAttributes ScalarAttrs(IID, ScalarRetTy, ScalarTys, FMF);
       unsigned ScalarCost =
-          ConcreteTTI->getIntrinsicInstrCost(IID, ScalarRetTy, ScalarTys, FMF,
-                                             CostKind);
+          ConcreteTTI->getIntrinsicInstrCost(ScalarAttrs, CostKind);
 
       return ScalarCalls * ScalarCost + ScalarizationCost;
     }
     // Look for intrinsics that can be lowered directly or turned into a scalar
     // intrinsic call.
     case Intrinsic::sqrt:
       ISDs.push_back(ISD::FSQRT);
       break;
case Intrinsic::ssub_sat: {
       Type *OpTy = StructType::create({RetTy, CondTy});
       Intrinsic::ID OverflowOp = IID == Intrinsic::sadd_sat
                                      ? Intrinsic::sadd_with_overflow
                                      : Intrinsic::ssub_with_overflow;
 
       // SatMax -> Overflow && SumDiff < 0
       // SatMin -> Overflow && SumDiff >= 0
       unsigned Cost = 0;
-      Cost += ConcreteTTI->getIntrinsicInstrCost(
-          OverflowOp, OpTy, {RetTy, RetTy}, FMF, ScalarizationCostPassed,
-          CostKind);
+      IntrinsicCostAttributes Attrs(OverflowOp, OpTy, {RetTy, RetTy}, FMF,
+                                    ScalarizationCostPassed);
+      Cost += ConcreteTTI->getIntrinsicInstrCost(Attrs, CostKind);
       Cost += ConcreteTTI->getCmpSelInstrCost(BinaryOperator::ICmp, RetTy,
                                               CondTy, CostKind);
       Cost += 2 * ConcreteTTI->getCmpSelInstrCost(BinaryOperator::Select, RetTy,
                                                   CondTy, CostKind);
       return Cost;
     }
     case Intrinsic::uadd_sat:
     case Intrinsic::usub_sat: {
       Type *CondTy = RetTy->getWithNewBitWidth(1);
 
       Type *OpTy = StructType::create({RetTy, CondTy});
       Intrinsic::ID OverflowOp = IID == Intrinsic::uadd_sat
                                      ? Intrinsic::uadd_with_overflow
                                      : Intrinsic::usub_with_overflow;
 
       unsigned Cost = 0;
-      Cost += ConcreteTTI->getIntrinsicInstrCost(
-          OverflowOp, OpTy, {RetTy, RetTy}, FMF, ScalarizationCostPassed,
-          CostKind);
+      IntrinsicCostAttributes Attrs(OverflowOp, OpTy, {RetTy, RetTy}, FMF,
+                                    ScalarizationCostPassed);
+      Cost += ConcreteTTI->getIntrinsicInstrCost(Attrs, CostKind);
       Cost += ConcreteTTI->getCmpSelInstrCost(BinaryOperator::Select, RetTy,
                                               CondTy, CostKind);
       return Cost;
     }
     case Intrinsic::smul_fix:
     case Intrinsic::umul_fix: {
       unsigned ExtSize = RetTy->getScalarSizeInBits() * 2;
       Type *ExtTy = RetTy->getWithNewBitWidth(ExtSize);
if (IID == Intrinsic::experimental_constrained_fmuladd)
              ConcreteTTI->getIntrinsicCost(
                  Intrinsic::experimental_constrained_fadd, RetTy, Tys, nullptr,
                  CostKind);
 
     // Else, assume that we need to scalarize this intrinsic. For math builtins
     // this will emit a costly libcall, adding call overhead and spills. Make it
     // very expensive.
     if (auto *RetVTy = dyn_cast<VectorType>(RetTy)) {
-      unsigned ScalarizationCost =
-          ((ScalarizationCostPassed != std::numeric_limits<unsigned>::max())
-               ? ScalarizationCostPassed
-               : getScalarizationOverhead(RetVTy, true, false));
+      unsigned ScalarizationCost = SkipScalarizationCost ?
+        ScalarizationCostPassed : getScalarizationOverhead(RetVTy, true, false);
+
       unsigned ScalarCalls = RetVTy->getNumElements();
       SmallVector<Type *, 4> ScalarTys;
       for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) {
         Type *Ty = Tys[i];
         if (Ty->isVectorTy())
           Ty = Ty->getScalarType();
         ScalarTys.push_back(Ty);
       }
-      unsigned ScalarCost = ConcreteTTI->getIntrinsicInstrCost(
-          IID, RetTy->getScalarType(), ScalarTys, FMF, CostKind);
+      IntrinsicCostAttributes Attrs(IID, RetTy->getScalarType(), ScalarTys, FMF);
+      unsigned ScalarCost = ConcreteTTI->getIntrinsicInstrCost(Attrs, CostKind);
       for (unsigned i = 0, ie = Tys.size(); i != ie; ++i) {
         if (auto *VTy = dyn_cast<VectorType>(Tys[i])) {
-          if (ScalarizationCostPassed == std::numeric_limits<unsigned>::max())
+          if (!ICA.skipScalarizationCost())
             ScalarizationCost += getScalarizationOverhead(VTy, false, true);
           ScalarCalls = std::max(ScalarCalls, VTy->getNumElements());
         }
       }
       return ScalarCalls * ScalarCost + ScalarizationCost;
     }
 
     // This is going to be turned into a library call, make it expensive.

llvm/lib/Analysis/TargetTransformInfo.cpp

bool HardwareLoopInfo::canAnalyze(LoopInfo &LI) {
   // Hardware loop.
   LoopBlocksRPO RPOT(L);
   RPOT.perform(&LI);
   if (containsIrreducibleCFG<const BasicBlock *>(RPOT, LI))
     return false;
   return true;
 }
 
+IntrinsicCostAttributes::IntrinsicCostAttributes(const IntrinsicInst &I) :
+    II(&I), RetTy(I.getType()), IID(I.getIntrinsicID()) {
+
+ FunctionType *FTy = I.getCalledFunction()->getFunctionType();
+ ParamTys.insert(ParamTys.begin(), FTy->param_begin(), FTy->param_end());
+ Arguments.insert(Arguments.begin(), I.arg_begin(), I.arg_end());
+ if (auto *FPMO = dyn_cast<FPMathOperator>(&I))
+   FMF = FPMO->getFastMathFlags();
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, CallInst &CI,
+                                                 unsigned Factor) :
+    RetTy(CI.getType()), IID(Id), VF(Factor) {
+
+  if (auto *FPMO = dyn_cast<FPMathOperator>(&CI))
+    FMF = FPMO->getFastMathFlags();
+
+  Arguments.insert(Arguments.begin(), CI.arg_begin(), CI.arg_end());
+  FunctionType *FTy =
+    CI.getCalledFunction()->getFunctionType();
+  ParamTys.insert(ParamTys.begin(), FTy->param_begin(), FTy->param_end());
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, CallInst &CI,
+                                                 unsigned Factor,
+                                                 unsigned ScalarCost) :
+    RetTy(CI.getType()), IID(Id), VF(Factor), ScalarizationCost(ScalarCost) {
+
+  if (auto *FPMO = dyn_cast<FPMathOperator>(&CI))
+    FMF = FPMO->getFastMathFlags();
+
+  Arguments.insert(Arguments.begin(), CI.arg_begin(), CI.arg_end());
+  FunctionType *FTy =
+    CI.getCalledFunction()->getFunctionType();
+  ParamTys.insert(ParamTys.begin(), FTy->param_begin(), FTy->param_end());
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                                                 ArrayRef<Type *> Tys,
+                                                 FastMathFlags Flags) :
+    RetTy(RTy), IID(Id), FMF(Flags) {
+  ParamTys.insert(ParamTys.begin(), Tys.begin(), Tys.end());
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                                                 ArrayRef<Type *> Tys,
+                                                 FastMathFlags Flags,
+                                                 unsigned ScalarCost) :
+    RetTy(RTy), IID(Id), FMF(Flags), ScalarizationCost(ScalarCost) {
+  ParamTys.insert(ParamTys.begin(), Tys.begin(), Tys.end());
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                                                 ArrayRef<Type *> Tys,
+                                                 FastMathFlags Flags,
+                                                 unsigned ScalarCost,
+                                                 const IntrinsicInst *I) :
+    II(I), RetTy(RTy), IID(Id), FMF(Flags), ScalarizationCost(ScalarCost) {
+  ParamTys.insert(ParamTys.begin(), Tys.begin(), Tys.end());
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, Type *RTy,
+                                                 ArrayRef<Type *> Tys) :
+    RetTy(RTy), IID(Id) {
+  ParamTys.insert(ParamTys.begin(), Tys.begin(), Tys.end());
+}
+
+IntrinsicCostAttributes::IntrinsicCostAttributes(Intrinsic::ID Id, Type *Ty,
+                                                 ArrayRef<Value *> Args) :
+    RetTy(Ty), IID(Id) {
+
+  Arguments.insert(Arguments.begin(), Args.begin(), Args.end());
+  ParamTys.reserve(Arguments.size());
+  for (unsigned Idx = 0, Size = Arguments.size(); Idx != Size; ++Idx)
+    ParamTys.push_back(Arguments[Idx]->getType());
+}
+
 bool HardwareLoopInfo::isHardwareLoopCandidate(ScalarEvolution &SE,
                                                LoopInfo &LI, DominatorTree &DT,
                                                bool ForceNestedLoop,
                                                bool ForceHardwareLoopPHI) {
   SmallVector<BasicBlock *, 4> ExitingBlocks;
   L->getExitingBlocks(ExitingBlocks);
 
   for (BasicBlock *BB : ExitingBlocks) {
int TargetTransformInfo::getInterleavedMemoryOpCost(
     bool UseMaskForCond, bool UseMaskForGaps) const {
   int Cost = TTIImpl->getInterleavedMemoryOpCost(
       Opcode, VecTy, Factor, Indices, Alignment, AddressSpace, CostKind,
       UseMaskForCond, UseMaskForGaps);
   assert(Cost >= 0 && "TTI should not produce negative costs!");
   return Cost;
 }
 
-int TargetTransformInfo::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                               ArrayRef<Type *> Tys,
-                                               FastMathFlags FMF,
-                                               unsigned ScalarizationCostPassed,
-                                               TTI::TargetCostKind CostKind,
-                                               const Instruction *I) const {
-  int Cost = TTIImpl->getIntrinsicInstrCost(ID, RetTy, Tys, FMF,
-                                            ScalarizationCostPassed, CostKind,
-                                            I);
-  assert(Cost >= 0 && "TTI should not produce negative costs!");
-  return Cost;
-}
-
-int TargetTransformInfo::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                               ArrayRef<Value *> Args,
-                                               FastMathFlags FMF, unsigned VF,
-                                               TTI::TargetCostKind CostKind,
-                                               const Instruction *I) const {
-  int Cost = TTIImpl->getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF,
-                                            CostKind, I);
+int
+TargetTransformInfo::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                           TTI::TargetCostKind CostKind) const {
+  int Cost = TTIImpl->getIntrinsicInstrCost(ICA, CostKind);
   assert(Cost >= 0 && "TTI should not produce negative costs!");
   return Cost;
 }
 
 int TargetTransformInfo::getCallInstrCost(Function *F, Type *RetTy,
                                           ArrayRef<Type *> Tys,
                                           TTI::TargetCostKind CostKind) const {
   int Cost = TTIImpl->getCallInstrCost(F, RetTy, Tys, CostKind);
case Instruction::ShuffleVector: {
 
     if (Shuffle->isSingleSource())
       return TTIImpl->getShuffleCost(SK_PermuteSingleSrc, Ty, 0, nullptr);
 
     return TTIImpl->getShuffleCost(SK_PermuteTwoSrc, Ty, 0, nullptr);
   }
   case Instruction::Call:
     if (const IntrinsicInst *II = dyn_cast<IntrinsicInst>(I)) {
-      SmallVector<Value *, 4> Args(II->arg_operands());
-
-      FastMathFlags FMF;
-      if (auto *FPMO = dyn_cast<FPMathOperator>(II))
-        FMF = FPMO->getFastMathFlags();
-
-      return getIntrinsicInstrCost(II->getIntrinsicID(), II->getType(), Args,
-                                   FMF, 1, CostKind, II);
+      IntrinsicCostAttributes CostAttrs(*II);
+      return getIntrinsicInstrCost(CostAttrs, CostKind);
     }
     return -1;
   default:
     // We don't have any information on this instruction.
     return -1;
   }
 }
 

llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.h

public:
   int getInlinerVectorBonusPercent() { return 0; }
 
   int getArithmeticReductionCost(
       unsigned Opcode,
       VectorType *Ty,
       bool IsPairwise,
       TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput);
 
-  template <typename T>
-  int getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy, ArrayRef<T *> Args,
-                            FastMathFlags FMF, unsigned VF,
-                            TTI::TargetCostKind CostKind,
-                            const Instruction *I = nullptr);
-  int getIntrinsicInstrCost(
-    Intrinsic::ID IID, Type *RetTy, ArrayRef<Type *> Tys, FastMathFlags FMF,
-    unsigned ScalarizationCostPassed = UINT_MAX,
-    TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-    const Instruction *I = nullptr);
-  int getIntrinsicInstrCost(
-    Intrinsic::ID IID, Type *RetTy, ArrayRef<Value *> Args, FastMathFlags FMF,
-    unsigned VF = 1, TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-    const Instruction *I = nullptr);
+  int getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                            TTI::TargetCostKind CostKind);
   int getMinMaxReductionCost(
     VectorType *Ty, VectorType *CondTy, bool IsPairwiseForm, bool IsUnsigned,
     TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput);
 
   unsigned getUserCost(const User *U, ArrayRef<const Value *> Operands,
                        TTI::TargetCostKind CostKind);
 };
 

llvm/lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp

static bool intrinsicHasPackedVectorBenefit(Intrinsic::ID ID) {
   // There's a small benefit to using vector ops in the legalized code.
   case Intrinsic::round:
     return true;
   default:
     return false;
   }
 }
 
-template <typename T>
-int GCNTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                      ArrayRef<T *> Args, FastMathFlags FMF,
-                                      unsigned VF,
-                                      TTI::TargetCostKind CostKind,
-                                      const Instruction *I) {
-  if (!intrinsicHasPackedVectorBenefit(ID))
-    return BaseT::getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF, CostKind, I);
+int GCNTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                      TTI::TargetCostKind CostKind) {
+  if (!intrinsicHasPackedVectorBenefit(ICA.getID()))
+    return BaseT::getIntrinsicInstrCost(ICA, CostKind);
 
+  Type *RetTy = ICA.getReturnType();
   EVT OrigTy = TLI->getValueType(DL, RetTy);
   if (!OrigTy.isSimple()) {
-    return BaseT::getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF, CostKind, I);
+    return BaseT::getIntrinsicInstrCost(ICA, CostKind);
   }
 
   // Legalize the type.
   std::pair<int, MVT> LT = TLI->getTypeLegalizationCost(DL, RetTy);
 
   unsigned NElts = LT.second.isVector() ?
     LT.second.getVectorNumElements() : 1;
 
   MVT::SimpleValueType SLT = LT.second.getScalarType().SimpleTy;
 
   if (SLT == MVT::f64)
     return LT.first * NElts * get64BitInstrCost();
 
   if (ST->has16BitInsts() && SLT == MVT::f16)
     NElts = (NElts + 1) / 2;
 
   // TODO: Get more refined intrinsic costs?
   unsigned InstRate = getQuarterRateInstrCost();
-  if (ID == Intrinsic::fma) {
+  if (ICA.getID() == Intrinsic::fma) {
     InstRate = ST->hasFastFMAF32() ? getHalfRateInstrCost()
                                    : getQuarterRateInstrCost();
   }
 
   return LT.first * NElts * InstRate;
 }
 
-int GCNTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                      ArrayRef<Value *> Args, FastMathFlags FMF,
-                                      unsigned VF,
-                                      TTI::TargetCostKind CostKind,
-                                      const Instruction *I) {
-  return getIntrinsicInstrCost<Value>(ID, RetTy, Args, FMF, VF, CostKind, I);
-}
-
-int GCNTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                      ArrayRef<Type *> Tys, FastMathFlags FMF,
-                                      unsigned ScalarizationCostPassed,
-                                      TTI::TargetCostKind CostKind,
-                                      const Instruction *I) {
-  return getIntrinsicInstrCost<Type>(ID, RetTy, Tys, FMF,
-                                     ScalarizationCostPassed, CostKind, I);
-}
-
 unsigned GCNTTIImpl::getCFInstrCost(unsigned Opcode,
                                     TTI::TargetCostKind CostKind) {
   // XXX - For some reason this isn't called for switch.
   switch (Opcode) {
   case Instruction::Br:
   case Instruction::Ret:
     return 10;
   default:
case Instruction::InsertElement: {
     ConstantInt *CI = dyn_cast<ConstantInt>(I->getOperand(2));
     unsigned Idx = -1;
     if (CI)
       Idx = CI->getZExtValue();
     return getVectorInstrCost(I->getOpcode(), I->getType(), Idx);
   }
   case Instruction::Call: {
     if (const IntrinsicInst *II = dyn_cast<IntrinsicInst>(U)) {
-      SmallVector<Value *, 4> Args(II->arg_operands());
-      FastMathFlags FMF;
-      if (auto *FPMO = dyn_cast<FPMathOperator>(II))
-        FMF = FPMO->getFastMathFlags();
-      return getIntrinsicInstrCost(II->getIntrinsicID(), II->getType(), Args,
-                                   FMF, 1, CostKind, II);
+      IntrinsicCostAttributes CostAttrs(*II);
+      return getIntrinsicInstrCost(CostAttrs, CostKind);
     } else {
       return BaseT::getUserCost(U, Operands, CostKind);
     }
   }
   case Instruction::ShuffleVector: {
     const ShuffleVectorInst *Shuffle = cast<ShuffleVectorInst>(I);
     auto *Ty = cast<VectorType>(Shuffle->getType());
     auto *SrcTy = cast<VectorType>(Shuffle->getOperand(0)->getType());

llvm/lib/Target/Hexagon/HexagonTargetTransformInfo.h

public:
   }
 
   unsigned getScalarizationOverhead(VectorType *Ty, const APInt &DemandedElts,
                                     bool Insert, bool Extract);
   unsigned getOperandsScalarizationOverhead(ArrayRef<const Value *> Args,
                                             unsigned VF);
   unsigned getCallInstrCost(Function *F, Type *RetTy, ArrayRef<Type*> Tys,
                             TTI::TargetCostKind CostKind);
-  unsigned getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                 ArrayRef<Value *> Args, FastMathFlags FMF,
-                                 unsigned VF,
-                                 TTI::TargetCostKind CostKind,
-                                 const Instruction *I);
-  unsigned getIntrinsicInstrCost(
-    Intrinsic::ID ID, Type *RetTy, ArrayRef<Type *> Tys, FastMathFlags FMF,
-    unsigned ScalarizationCostPassed = UINT_MAX,
-    TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-    const Instruction *I = nullptr);
+  unsigned getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                 TTI::TargetCostKind CostKind);
   unsigned getAddressComputationCost(Type *Tp, ScalarEvolution *SE,
             const SCEV *S);
   unsigned getMemoryOpCost(unsigned Opcode, Type *Src, MaybeAlign Alignment,
                            unsigned AddressSpace,
                            TTI::TargetCostKind CostKind,
                            const Instruction *I = nullptr);
   unsigned getMaskedMemoryOpCost(unsigned Opcode, Type *Src, unsigned Alignment,
                                  unsigned AddressSpace,

llvm/lib/Target/Hexagon/HexagonTargetTransformInfo.cpp

unsigned HexagonTTIImpl::getOperandsScalarizationOverhead(
   return BaseT::getOperandsScalarizationOverhead(Args, VF);
 }
 
 unsigned HexagonTTIImpl::getCallInstrCost(Function *F, Type *RetTy,
       ArrayRef<Type*> Tys, TTI::TargetCostKind CostKind) {
   return BaseT::getCallInstrCost(F, RetTy, Tys, CostKind);
 }
 
-unsigned HexagonTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                               ArrayRef<Value *> Args,
-                                               FastMathFlags FMF, unsigned VF,
-                                               TTI::TargetCostKind CostKind,
-                                               const Instruction *I) {
-  return BaseT::getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF, CostKind, I);
-}
-
-unsigned HexagonTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                               ArrayRef<Type *> Tys,
-                                               FastMathFlags FMF,
-                                               unsigned ScalarizationCostPassed,
-                                               TTI::TargetCostKind CostKind,
-                                               const Instruction *I) {
-  if (ID == Intrinsic::bswap) {
-    std::pair<int, MVT> LT = TLI.getTypeLegalizationCost(DL, RetTy);
+unsigned
+HexagonTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                      TTI::TargetCostKind CostKind) {
+  if (ICA.getID() == Intrinsic::bswap) {
+    std::pair<int, MVT> LT = TLI.getTypeLegalizationCost(DL, ICA.getReturnType());
     return LT.first + 2;
   }
-  return BaseT::getIntrinsicInstrCost(ID, RetTy, Tys, FMF,
-                                      ScalarizationCostPassed, CostKind, I);
+  return BaseT::getIntrinsicInstrCost(ICA, CostKind);
 }
 
 unsigned HexagonTTIImpl::getAddressComputationCost(Type *Tp,
       ScalarEvolution *SE, const SCEV *S) {
   return 0;
 }
 
 unsigned HexagonTTIImpl::getMemoryOpCost(unsigned Opcode, Type *Src,

llvm/lib/Target/PowerPC/PPCTargetTransformInfo.h

public:
   int getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
                                  unsigned Factor,
                                  ArrayRef<unsigned> Indices,
                                  unsigned Alignment,
                                  unsigned AddressSpace,
                                  TTI::TargetCostKind CostKind = TTI::TCK_SizeAndLatency,
                                  bool UseMaskForCond = false,
                                  bool UseMaskForGaps = false);
-  unsigned getIntrinsicInstrCost(
-    Intrinsic::ID ID, Type *RetTy,
-                                 ArrayRef<Value *> Args, FastMathFlags FMF,
-                                 unsigned VF,
-                                 TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-                                 const Instruction *I = nullptr);
-  unsigned getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                 ArrayRef<Type *> Tys, FastMathFlags FMF,
-                                 unsigned ScalarizationCostPassed = UINT_MAX,
-                                 TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-                                 const Instruction *I = nullptr);
+  unsigned getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                 TTI::TargetCostKind CostKind);
 
   /// @}
 };
 
 } // end namespace llvm
 
 #endif

llvm/lib/Target/PowerPC/PPCTargetTransformInfo.cpp

int PPCTTIImpl::getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
   // instruction). For each result vector, we need one shuffle per incoming
   // vector (except that the first shuffle can take two incoming vectors
   // because it does not need to take itself).
   Cost += Factor*(LT.first-1);
 
   return Cost;
 }
 
-unsigned PPCTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                           ArrayRef<Value *> Args,
-                                           FastMathFlags FMF, unsigned VF,
-                                           TTI::TargetCostKind CostKind,
-                                           const Instruction *I) {
-  return BaseT::getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF, CostKind, I);
-}
-
-unsigned PPCTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                           ArrayRef<Type *> Tys,
-                                           FastMathFlags FMF,
-                                           unsigned ScalarizationCostPassed,
-                                           TTI::TargetCostKind CostKind,
-                                           const Instruction *I) {
-  if (ID == Intrinsic::bswap && ST->hasP9Vector())
-    return TLI->getTypeLegalizationCost(DL, RetTy).first;
-  return BaseT::getIntrinsicInstrCost(ID, RetTy, Tys, FMF,
-                                      ScalarizationCostPassed, CostKind, I);
+unsigned PPCTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                           TTI::TargetCostKind CostKind) {
+  return BaseT::getIntrinsicInstrCost(ICA, CostKind);
 }
 
 bool PPCTTIImpl::canSaveCmp(Loop *L, BranchInst **BI, ScalarEvolution *SE,
                             LoopInfo *LI, DominatorTree *DT,
                             AssumptionCache *AC, TargetLibraryInfo *LibInfo) {
   // Process nested loops first.
   for (Loop::iterator I = L->begin(), E = L->end(); I != E; ++I)
     if (canSaveCmp(*I, BI, SE, LI, DT, AC, LibInfo))

llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.h

int getInterleavedMemoryOpCost(unsigned Opcode, Type *VecTy,
                                  unsigned Factor,
                                  ArrayRef<unsigned> Indices,
                                  unsigned Alignment,
                                  unsigned AddressSpace,
                                  TTI::TargetCostKind CostKind = TTI::TCK_SizeAndLatency,
                                  bool UseMaskForCond = false,
                                  bool UseMaskForGaps = false);
 
-  int getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                            ArrayRef<Value *> Args, FastMathFlags FMF,
-                            unsigned VF = 1,
-                            TTI::TargetCostKind CostKind = TTI::TCK_SizeAndLatency,
-                            const Instruction *I = nullptr);
-  int getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy, ArrayRef<Type *> Tys,
-                            FastMathFlags FMF,
-                            unsigned ScalarizationCostPassed = UINT_MAX,
-                            TTI::TargetCostKind CostKind = TTI::TCK_SizeAndLatency,
-                            const Instruction *I = nullptr);
+  int getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                            TTI::TargetCostKind CostKind);
   /// @}
 };
 
 } // end namespace llvm
 
 #endif

llvm/lib/Target/SystemZ/SystemZTargetTransformInfo.cpp

 }
 
 static int getVectorIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy) {
   if (RetTy->isVectorTy() && ID == Intrinsic::bswap)
     return getNumVectorRegs(RetTy); // VPERM
   return -1;
 }
 
-int SystemZTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                          ArrayRef<Value *> Args,
-                                          FastMathFlags FMF, unsigned VF,
-                                          TTI::TargetCostKind CostKind,
-                                          const Instruction *I) {
-  int Cost = getVectorIntrinsicInstrCost(ID, RetTy);
-  if (Cost != -1)
-    return Cost;
-  return BaseT::getIntrinsicInstrCost(ID, RetTy, Args, FMF, VF, CostKind, I);
-}
-
-int SystemZTTIImpl::getIntrinsicInstrCost(Intrinsic::ID ID, Type *RetTy,
-                                          ArrayRef<Type *> Tys,
-                                          FastMathFlags FMF,
-                                          unsigned ScalarizationCostPassed,
-                                          TTI::TargetCostKind CostKind,
-                                          const Instruction *I) {
-  int Cost = getVectorIntrinsicInstrCost(ID, RetTy);
+int SystemZTTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                          TTI::TargetCostKind CostKind) {
+  int Cost = getVectorIntrinsicInstrCost(ICA.getID(), ICA.getReturnType());
   if (Cost != -1)
     return Cost;
-  return BaseT::getIntrinsicInstrCost(ID, RetTy, Tys, FMF,
-                                      ScalarizationCostPassed, CostKind, I);
+  return BaseT::getIntrinsicInstrCost(ICA, CostKind);
 }

llvm/lib/Target/X86/X86TargetTransformInfo.h

int getGatherScatterOpCost(unsigned Opcode, Type *DataTy, Value *Ptr,
                              bool VariableMask, unsigned Alignment,
                              TTI::TargetCostKind CostKind,
                              const Instruction *I);
   int getAddressComputationCost(Type *PtrTy, ScalarEvolution *SE,
                                 const SCEV *Ptr);
 
   unsigned getAtomicMemIntrinsicMaxElementSize() const;
 
-  int getIntrinsicInstrCost(
-    Intrinsic::ID IID, Type *RetTy, ArrayRef<Type *> Tys,
-    FastMathFlags FMF, unsigned ScalarizationCostPassed = UINT_MAX,
-    TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-    const Instruction *I = nullptr);
-
-  int getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
-                            ArrayRef<Value *> Args, FastMathFlags FMF,
-                            unsigned VF = 1,
-                            TTI::TargetCostKind CostKind = TTI::TCK_RecipThroughput,
-                            const Instruction *I = nullptr);
+  int getTypeBasedIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                     TTI::TargetCostKind CostKind);
+  int getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                            TTI::TargetCostKind CostKind);
 
   int getArithmeticReductionCost(unsigned Opcode, VectorType *Ty,
                                  bool IsPairwiseForm,
                                  TTI::TargetCostKind CostKind = TTI::TCK_SizeAndLatency);
 
   int getMinMaxCost(Type *Ty, Type *CondTy, bool IsUnsigned);
 
   int getMinMaxReductionCost(VectorType *Ty, VectorType *CondTy,

llvm/lib/Target/X86/X86TargetTransformInfo.cpp

if (ST->hasSSE1())
     if (const auto *Entry = CostTableLookup(SSE1CostTbl, ISD, MTy))
       return LT.first * (ExtraCost + Entry->Cost);
 
   return BaseT::getCmpSelInstrCost(Opcode, ValTy, CondTy, CostKind, I);
 }
 
 unsigned X86TTIImpl::getAtomicMemIntrinsicMaxElementSize() const { return 16; }
 
-int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
-                                      ArrayRef<Type *> Tys, FastMathFlags FMF,
-                                      unsigned ScalarizationCostPassed,
-                                      TTI::TargetCostKind CostKind,
-                                      const Instruction *I) {
+int X86TTIImpl::getTypeBasedIntrinsicInstrCost(
+  const IntrinsicCostAttributes &ICA, TTI::TargetCostKind CostKind) {
+
   // Costs should match the codegen from:
   // BITREVERSE: llvm\test\CodeGen\X86\vector-bitreverse.ll
   // BSWAP: llvm\test\CodeGen\X86\bswap-vector.ll
   // CTLZ: llvm\test\CodeGen\X86\vector-lzcnt-*.ll
   // CTPOP: llvm\test\CodeGen\X86\vector-popcnt-*.ll
   // CTTZ: llvm\test\CodeGen\X86\vector-tzcnt-*.ll
   static const CostTblEntry AVX512CDCostTbl[] = {
     { ISD::CTLZ,       MVT::v8i64,   1 },
static const CostTblEntry X86CostTbl[] = { // 32 or 64-bit targets
     { ISD::SADDO,      MVT::i32,     1 },
     { ISD::SADDO,      MVT::i16,     1 },
     { ISD::SADDO,      MVT::i8,      1 },
     { ISD::UADDO,      MVT::i32,     1 },
     { ISD::UADDO,      MVT::i16,     1 },
     { ISD::UADDO,      MVT::i8,      1 },
   };
 
+  Type *RetTy = ICA.getReturnType();
   Type *OpTy = RetTy;
+  Intrinsic::ID IID = ICA.getID();
   unsigned ISD = ISD::DELETED_NODE;
   switch (IID) {
   default:
     break;
   case Intrinsic::bitreverse:
     ISD = ISD::BITREVERSE;
     break;
   case Intrinsic::bswap:
if (ISD != ISD::DELETED_NODE) {
     if (ST->is64Bit())
       if (const auto *Entry = CostTableLookup(X64CostTbl, ISD, MTy))
         return LT.first * Entry->Cost;
 
     if (const auto *Entry = CostTableLookup(X86CostTbl, ISD, MTy))
       return LT.first * Entry->Cost;
   }
 
-  return BaseT::getIntrinsicInstrCost(IID, RetTy, Tys, FMF,
-                                      ScalarizationCostPassed, CostKind, I);
+  return BaseT::getIntrinsicInstrCost(ICA, CostKind);
 }
 
-int X86TTIImpl::getIntrinsicInstrCost(Intrinsic::ID IID, Type *RetTy,
-                                      ArrayRef<Value *> Args, FastMathFlags FMF,
-                                      unsigned VF,
-                                      TTI::TargetCostKind CostKind,
-                                      const Instruction *I) {
+int X86TTIImpl::getIntrinsicInstrCost(const IntrinsicCostAttributes &ICA,
+                                      TTI::TargetCostKind CostKind) {
+  if (ICA.isTypeBasedOnly())
+    return getTypeBasedIntrinsicInstrCost(ICA, CostKind);
+
   static const CostTblEntry AVX512CostTbl[] = {
     { ISD::ROTL,       MVT::v8i64,   1 },
     { ISD::ROTL,       MVT::v4i64,   1 },
     { ISD::ROTL,       MVT::v2i64,   1 },
     { ISD::ROTL,       MVT::v16i32,  1 },
     { ISD::ROTL,       MVT::v8i32,   1 },
     { ISD::ROTL,       MVT::v4i32,   1 },
     { ISD::ROTR,       MVT::v8i64,   1 },
static const CostTblEntry X86CostTbl[] = { // 32 or 64-bit targets
     { ISD::ROTR,       MVT::i32,     1 },
     { ISD::ROTR,       MVT::i16,     1 },
     { ISD::ROTR,       MVT::i8,      1 },
     { ISD::FSHL,       MVT::i32,     4 },
     { ISD::FSHL,       MVT::i16,     4 },
     { ISD::FSHL,       MVT::i8,      4 }
   };
 
+  Intrinsic::ID IID = ICA.getID();
+  Type *RetTy = ICA.getReturnType();
+  const SmallVectorImpl<Value *> &Args = ICA.getArgs();
   unsigned ISD = ISD::DELETED_NODE;
   switch (IID) {
   default:
     break;
   case Intrinsic::fshl:
     ISD = ISD::FSHL;
     if (Args[0] == Args[1])
       ISD = ISD::ROTL;
if (ISD != ISD::DELETED_NODE) {
     if (ST->is64Bit())
       if (const auto *Entry = CostTableLookup(X64CostTbl, ISD, MTy))
         return LT.first * Entry->Cost;
 
     if (const auto *Entry = CostTableLookup(X86CostTbl, ISD, MTy))
       return LT.first * Entry->Cost;
   }
 
-  return BaseT::getIntrinsicInstrCost(IID, RetTy, Args, FMF, VF, CostKind, I);
+  return BaseT::getIntrinsicInstrCost(ICA, CostKind);
 }
 
 int X86TTIImpl::getVectorInstrCost(unsigned Opcode, Type *Val, unsigned Index) {
   static const CostTblEntry SLMCostTbl[] = {
      { ISD::EXTRACT_VECTOR_ELT,       MVT::i8,      4 },
      { ISD::EXTRACT_VECTOR_ELT,       MVT::i16,     4 },
      { ISD::EXTRACT_VECTOR_ELT,       MVT::i32,     4 },
      { ISD::EXTRACT_VECTOR_ELT,       MVT::i64,     7 }

llvm/lib/Transforms/Vectorize/LoopVectorize.cpp

unsigned LoopVectorizationCostModel::getVectorCallCost(CallInst *CI,
   return Cost;
 }
 
 unsigned LoopVectorizationCostModel::getVectorIntrinsicCost(CallInst *CI,
                                                             unsigned VF) {
   Intrinsic::ID ID = getVectorIntrinsicIDForCall(CI, TLI);
   assert(ID && "Expected intrinsic call!");
 
-  FastMathFlags FMF;
-  if (auto *FPMO = dyn_cast<FPMathOperator>(CI))
-    FMF = FPMO->getFastMathFlags();
-
-  SmallVector<Value *, 4> Operands(CI->arg_operands());
-  return TTI.getIntrinsicInstrCost(ID, CI->getType(), Operands, FMF, VF,
-                                   TargetTransformInfo::TCK_RecipThroughput,
-                                   CI);
+  IntrinsicCostAttributes CostAttrs(ID, *CI, VF);
+  return TTI.getIntrinsicInstrCost(CostAttrs,
+                                   TargetTransformInfo::TCK_RecipThroughput);
 }
 
 static Type *smallestIntegerVectorType(Type *T1, Type *T2) {
   auto *I1 = cast<IntegerType>(cast<VectorType>(T1)->getElementType());
   auto *I2 = cast<IntegerType>(cast<VectorType>(T2)->getElementType());
   return I1->getBitWidth() < I2->getBitWidth() ? T1 : T2;
 }
 

llvm/lib/Transforms/Vectorize/SLPVectorizer.cpp

 }
 
 static std::pair<unsigned, unsigned>
 getVectorCallCosts(CallInst *CI, VectorType *VecTy, TargetTransformInfo *TTI,
                    TargetLibraryInfo *TLI) {
   Intrinsic::ID ID = getVectorIntrinsicIDForCall(CI, TLI);
 
   // Calculate the cost of the scalar and vector calls.
-  FastMathFlags FMF;
-  if (auto *FPMO = dyn_cast<FPMathOperator>(CI))
-    FMF = FPMO->getFastMathFlags();
-
-  SmallVector<Value *, 4> Args(CI->arg_operands());
-  int IntrinsicCost = TTI->getIntrinsicInstrCost(ID, CI->getType(), Args, FMF,
-                                                 VecTy->getNumElements());
+  IntrinsicCostAttributes CostAttrs(ID, *CI, VecTy->getNumElements());
+  int IntrinsicCost =
+    TTI->getIntrinsicInstrCost(CostAttrs, TTI::TCK_RecipThroughput);
 
   auto Shape =
       VFShape::get(*CI, {static_cast<unsigned>(VecTy->getNumElements()), false},
                    false /*HasGlobalPred*/);
   Function *VecFunc = VFDatabase(*CI).getVectorizedFunction(Shape);
   int LibCost = IntrinsicCost;
   if (!CI->isNoBuiltin() && VecFunc) {
     // Calculate the cost of the vector library call.
case Instruction::Store: {
       }
       return ReuseShuffleCost + VecStCost - ScalarStCost;
     }
     case Instruction::Call: {
       CallInst *CI = cast<CallInst>(VL0);
       Intrinsic::ID ID = getVectorIntrinsicIDForCall(CI, TLI);
 
       // Calculate the cost of the scalar and vector calls.
-      SmallVector<Type *, 4> ScalarTys;
-      for (unsigned op = 0, opc = CI->getNumArgOperands(); op != opc; ++op)
-        ScalarTys.push_back(CI->getArgOperand(op)->getType());
-
-      FastMathFlags FMF;
-      if (auto *FPMO = dyn_cast<FPMathOperator>(CI))
-        FMF = FPMO->getFastMathFlags();
-
-      int ScalarEltCost =
-          TTI->getIntrinsicInstrCost(ID, ScalarTy, ScalarTys, FMF, 1, CostKind);
+      IntrinsicCostAttributes CostAttrs(ID, *CI, 1, 1);
+      int ScalarEltCost = TTI->getIntrinsicInstrCost(CostAttrs, CostKind);
       if (NeedToShuffleReuses) {
         ReuseShuffleCost -= (ReuseShuffleNumbers - VL.size()) * ScalarEltCost;
       }
       int ScalarCallCost = VecTy->getNumElements() * ScalarEltCost;
 
       auto VecCallCosts = getVectorCallCosts(CI, VecTy, TTI, TLI);
       int VecCallCost = std::min(VecCallCosts.first, VecCallCosts.second);