[llvm][NFC] CallSite removal from inliner-related files

[mtrofin]

Summary

This removes CallSite from inliner files. Some dependencies where thus affected.

[craig.topper]

llvm/lib/CodeGen/SafeStack.cpp

324

Shouldn't this be CallBase? Instruction::Call uses this code too.

707–709

Use CallInst to match the only caller?

719–720

I think this should be CI not CB since its CallInst not CallBase.

llvm/lib/Transforms/IPO/PartialInlining.cpp

306–307

Since you're here anyway, drop the else

307–311

Drop this since it should be obviously unreachable without the elses?

769

Why do we need this variable?

779–780

Can we just use CB instead of *Call

llvm/lib/Transforms/Utils/InlineFunction.cpp

1658–1659

Drop this variable?

[mtrofin]

feedback

[mtrofin]

typos

[mtrofin]

llvm/lib/CodeGen/SafeStack.cpp

324

True, thanks for the catch!

llvm/lib/Transforms/IPO/PartialInlining.cpp

306–307

There's still the case of a CallBrInst. Rewrote as it could be simpler, though.

[dblaikie]

llvm/lib/CodeGen/SafeStack.cpp

346–348

Probably commit things like this independently (before/after this refactor, whichever suits better) without precommit review.

llvm/lib/Transforms/Utils/InlineFunction.cpp

1662

You don't need to add a & here, I think?

[mtrofin]

small fixes

llvm/lib/CodeGen/SafeStack.cpp

346–348

In this case, we needed ImmutableCallSite::arg_iterator -> auto because we don't want ImmutableCallSite

[dblaikie]

Sounds good!

llvm/include/llvm/Transforms/IPO/Inliner.h

 //===- Inliner.h - Inliner pass and infrastructure --------------*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #ifndef LLVM_TRANSFORMS_IPO_INLINER_H
 #define LLVM_TRANSFORMS_IPO_INLINER_H
 
 #include "llvm/Analysis/CGSCCPassManager.h"
 #include "llvm/Analysis/CallGraphSCCPass.h"
 #include "llvm/Analysis/InlineCost.h"
 #include "llvm/Analysis/LazyCallGraph.h"
-#include "llvm/IR/CallSite.h"
 #include "llvm/IR/PassManager.h"
 #include "llvm/Transforms/Utils/ImportedFunctionsInliningStatistics.h"
 #include <utility>
 
 namespace llvm {
 
 class AssumptionCacheTracker;
 class CallGraph;
struct LegacyInlinerBase : public CallGraphSCCPass {
 
   /// Remove now-dead linkonce functions at the end of processing to avoid
   /// breaking the SCC traversal.
   bool doFinalization(CallGraph &CG) override;
 
   /// This method must be implemented by the subclass to determine the cost of
   /// inlining the specified call site.  If the cost returned is greater than
   /// the current inline threshold, the call site is not inlined.
-  // FIXME(mtrofin): remove this in favor of the CallBase-based one
-  virtual InlineCost getInlineCost(CallSite CS) = 0;
-
-  /// This method must be implemented by the subclass to determine the cost of
-  /// inlining the specified call site.  If the cost returned is greater than
-  /// the current inline threshold, the call site is not inlined.
-  virtual InlineCost getInlineCost(CallBase &CB) {
-    return getInlineCost(CallSite(&CB));
-  }
+  virtual InlineCost getInlineCost(CallBase &CB) = 0;
 
   /// Remove dead functions.
   ///
   /// This also includes a hack in the form of the 'AlwaysInlineOnly' flag
   /// which restricts it to deleting functions with an 'AlwaysInline'
   /// attribute. This is useful for the InlineAlways pass that only wants to
   /// deal with that subset of the functions.
   bool removeDeadFunctions(CallGraph &CG, bool AlwaysInlineOnly = false);

llvm/include/llvm/Transforms/Utils/Cloning.h

 /// *inlined* code to minimize the actual inserted code, it must not delete
 /// code in the caller as users of this routine may have pointers to
 /// instructions in the caller that need to remain stable.
 ///
 /// If ForwardVarArgsTo is passed, inlining a function with varargs is allowed
 /// and all varargs at the callsite will be passed to any calls to
 /// ForwardVarArgsTo. The caller of InlineFunction has to make sure any varargs
 /// are only used by ForwardVarArgsTo.
-InlineResult InlineFunction(CallBase *CB, InlineFunctionInfo &IFI,
-                            AAResults *CalleeAAR = nullptr,
-                            bool InsertLifetime = true);
-InlineResult InlineFunction(CallSite CS, InlineFunctionInfo &IFI,
+InlineResult InlineFunction(CallBase &CB, InlineFunctionInfo &IFI,
                             AAResults *CalleeAAR = nullptr,
                             bool InsertLifetime = true,
                             Function *ForwardVarArgsTo = nullptr);
 
 /// Clones a loop \p OrigLoop.  Returns the loop and the blocks in \p
 /// Blocks.
 ///
 /// Updates LoopInfo and DominatorTree assuming the loop is dominated by block

llvm/lib/CodeGen/SafeStack.cpp

 #include "llvm/Analysis/ScalarEvolution.h"
 #include "llvm/Analysis/ScalarEvolutionExpressions.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
 #include "llvm/CodeGen/TargetLowering.h"
 #include "llvm/CodeGen/TargetPassConfig.h"
 #include "llvm/CodeGen/TargetSubtargetInfo.h"
 #include "llvm/IR/Argument.h"
 #include "llvm/IR/Attributes.h"
-#include "llvm/IR/CallSite.h"
 #include "llvm/IR/ConstantRange.h"
 #include "llvm/IR/Constants.h"
 #include "llvm/IR/DIBuilder.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/DerivedTypes.h"
 #include "llvm/IR/Dominators.h"
 #include "llvm/IR/Function.h"
 #include "llvm/IR/IRBuilder.h"
class SafeStack {
 
   bool IsSafeStackAlloca(const Value *AllocaPtr, uint64_t AllocaSize);
 
   bool IsMemIntrinsicSafe(const MemIntrinsic *MI, const Use &U,
                           const Value *AllocaPtr, uint64_t AllocaSize);
   bool IsAccessSafe(Value *Addr, uint64_t Size, const Value *AllocaPtr,
                     uint64_t AllocaSize);
 
-  bool ShouldInlinePointerAddress(CallSite &CS);
+  bool ShouldInlinePointerAddress(CallInst &CI);
   void TryInlinePointerAddress();
 
 public:
   SafeStack(Function &F, const TargetLoweringBase &TL, const DataLayout &DL,
             ScalarEvolution &SE)
       : F(F), TL(TL), DL(DL), SE(SE),
         StackPtrTy(Type::getInt8PtrTy(F.getContext())),
         IntPtrTy(DL.getIntPtrType(F.getContext())),
for (const Use &UI : V->uses()) {
         break;
 
       case Instruction::Ret:
         // Information leak.
         return false;
 
       case Instruction::Call:
       case Instruction::Invoke: {
-        ImmutableCallSite CS(I);
+        const CallBase &CS = *cast<CallBase>(I);
 
         if (I->isLifetimeStartOrEnd())
           continue;
 
         if (const MemIntrinsic *MI = dyn_cast<MemIntrinsic>(I)) {
           if (!IsMemIntrinsicSafe(MI, UI, AllocaPtr, AllocaSize)) {
             LLVM_DEBUG(dbgs()
                        << "[SafeStack] Unsafe alloca: " << *AllocaPtr
                        << "\n            unsafe memintrinsic: " << *I << "\n");
             return false;
           }
           continue;
         }
 
         // LLVM 'nocapture' attribute is only set for arguments whose address
         // is not stored, passed around, or used in any other non-trivial way.
         // We assume that passing a pointer to an object as a 'nocapture
         // readnone' argument is safe.
         // FIXME: a more precise solution would require an interprocedural
         // analysis here, which would look at all uses of an argument inside
         // the function being called.
-        ImmutableCallSite::arg_iterator B = CS.arg_begin(), E = CS.arg_end();
-        for (ImmutableCallSite::arg_iterator A = B; A != E; ++A)
+        auto B = CS.arg_begin(), E = CS.arg_end();
+        for (auto A = B; A != E; ++A)
           if (A->get() == V)
             if (!(CS.doesNotCapture(A - B) && (CS.doesNotAccessMemory(A - B) ||
                                                CS.doesNotAccessMemory()))) {
               LLVM_DEBUG(dbgs() << "[SafeStack] Unsafe alloca: " << *AllocaPtr
                                 << "\n            unsafe call: " << *I << "\n");
               return false;
             }
         continue;
       }
for (inst_iterator It = inst_begin(&F), Ie = inst_end(&F); It != Ie;) {
         SI->takeName(II);
         assert(II->use_empty());
         II->eraseFromParent();
       }
     }
   }
 }
 
-bool SafeStack::ShouldInlinePointerAddress(CallSite &CS) {
-  Function *Callee = CS.getCalledFunction();
-  if (CS.hasFnAttr(Attribute::AlwaysInline) &&
+bool SafeStack::ShouldInlinePointerAddress(CallInst &CI) {
+  Function *Callee = CI.getCalledFunction();
+  if (CI.hasFnAttr(Attribute::AlwaysInline) &&
       isInlineViable(*Callee).isSuccess())
     return true;
   if (Callee->isInterposable() || Callee->hasFnAttribute(Attribute::NoInline) ||
-      CS.isNoInline())
+      CI.isNoInline())
     return false;
   return true;
 }
 
 void SafeStack::TryInlinePointerAddress() {
-  if (!isa<CallInst>(UnsafeStackPtr))
+  auto *CI = dyn_cast<CallInst>(UnsafeStackPtr);
+  if (!CI)
     return;
 
   if(F.hasOptNone())
     return;
 
-  CallSite CS(UnsafeStackPtr);
-  Function *Callee = CS.getCalledFunction();
+  Function *Callee = CI->getCalledFunction();
   if (!Callee || Callee->isDeclaration())
     return;
 
-  if (!ShouldInlinePointerAddress(CS))
+  if (!ShouldInlinePointerAddress(*CI))
     return;
 
   InlineFunctionInfo IFI;
-  InlineFunction(CS, IFI);
+  InlineFunction(*CI, IFI);
 }
 
 bool SafeStack::run() {
   assert(F.hasFnAttribute(Attribute::SafeStack) &&
          "Can't run SafeStack on a function without the attribute");
   assert(!F.isDeclaration() && "Can't run SafeStack on a function declaration");
 
   ++NumFunctions;

llvm/lib/Target/AMDGPU/AMDGPUInline.cpp

 //===----------------------------------------------------------------------===//
 
 #include "AMDGPU.h"
 #include "llvm/Analysis/AssumptionCache.h"
 #include "llvm/Analysis/CallGraph.h"
 #include "llvm/Analysis/InlineCost.h"
 #include "llvm/Analysis/TargetTransformInfo.h"
 #include "llvm/Analysis/ValueTracking.h"
-#include "llvm/IR/CallSite.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/Type.h"
 #include "llvm/InitializePasses.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Transforms/IPO.h"

 public:
   AMDGPUInliner() : LegacyInlinerBase(ID) {
     initializeAMDGPUInlinerPass(*PassRegistry::getPassRegistry());
     Params = getInlineParams();
   }
 
   static char ID; // Pass identification, replacement for typeid
 
-  unsigned getInlineThreshold(CallSite CS) const;
+  unsigned getInlineThreshold(CallBase &CB) const;
 
-  InlineCost getInlineCost(CallSite CS) override;
+  InlineCost getInlineCost(CallBase &CB) override;
 
   bool runOnSCC(CallGraphSCC &SCC) override;
 
   void getAnalysisUsage(AnalysisUsage &AU) const override;
 
 private:
   TargetTransformInfoWrapperPass *TTIWP;
 
bool AMDGPUInliner::runOnSCC(CallGraphSCC &SCC) {
   return LegacyInlinerBase::runOnSCC(SCC);
 }
 
 void AMDGPUInliner::getAnalysisUsage(AnalysisUsage &AU) const {
   AU.addRequired<TargetTransformInfoWrapperPass>();
   LegacyInlinerBase::getAnalysisUsage(AU);
 }
 
-unsigned AMDGPUInliner::getInlineThreshold(CallSite CS) const {
+unsigned AMDGPUInliner::getInlineThreshold(CallBase &CB) const {
   int Thres = Params.DefaultThreshold;
 
-  Function *Caller = CS.getCaller();
+  Function *Caller = CB.getCaller();
   // Listen to the inlinehint attribute when it would increase the threshold
   // and the caller does not need to minimize its size.
-  Function *Callee = CS.getCalledFunction();
+  Function *Callee = CB.getCalledFunction();
   bool InlineHint = Callee && !Callee->isDeclaration() &&
     Callee->hasFnAttribute(Attribute::InlineHint);
   if (InlineHint && Params.HintThreshold && Params.HintThreshold > Thres
       && !Caller->hasFnAttribute(Attribute::MinSize))
     Thres = Params.HintThreshold.getValue() *
             TTIWP->getTTI(*Callee).getInliningThresholdMultiplier();
 
   const DataLayout &DL = Caller->getParent()->getDataLayout();
   if (!Callee)
     return (unsigned)Thres;
 
   // If we have a pointer to private array passed into a function
   // it will not be optimized out, leaving scratch usage.
   // Increase the inline threshold to allow inliniting in this case.
   uint64_t AllocaSize = 0;
   SmallPtrSet<const AllocaInst *, 8> AIVisited;
-  for (Value *PtrArg : CS.args()) {
+  for (Value *PtrArg : CB.args()) {
     PointerType *Ty = dyn_cast<PointerType>(PtrArg->getType());
     if (!Ty || (Ty->getAddressSpace() != AMDGPUAS::PRIVATE_ADDRESS &&
                 Ty->getAddressSpace() != AMDGPUAS::FLAT_ADDRESS))
       continue;
 
     PtrArg = GetUnderlyingObject(PtrArg, DL);
     if (const AllocaInst *AI = dyn_cast<AllocaInst>(PtrArg)) {
       if (!AI->isStaticAlloca() || !AIVisited.insert(AI).second)
unsigned AMDGPUInliner::getInlineThreshold(CallBase &CB) const {
   if (AllocaSize)
     Thres += ArgAllocaCost;
 
   return (unsigned)Thres;
 }
 
 // Check if call is just a wrapper around another call.
 // In this case we only have call and ret instructions.
-static bool isWrapperOnlyCall(CallSite CS) {
-  Function *Callee = CS.getCalledFunction();
+static bool isWrapperOnlyCall(CallBase &CB) {
+  Function *Callee = CB.getCalledFunction();
   if (!Callee || Callee->size() != 1)
     return false;
   const BasicBlock &BB = Callee->getEntryBlock();
   if (const Instruction *I = BB.getFirstNonPHI()) {
     if (!isa<CallInst>(I)) {
       return false;
     }
     if (isa<ReturnInst>(*std::next(I->getIterator()))) {
       LLVM_DEBUG(dbgs() << "    Wrapper only call detected: "
                         << Callee->getName() << '\n');
       return true;
     }
   }
   return false;
 }
 
-InlineCost AMDGPUInliner::getInlineCost(CallSite CS) {
-  Function *Callee = CS.getCalledFunction();
-  Function *Caller = CS.getCaller();
+InlineCost AMDGPUInliner::getInlineCost(CallBase &CB) {
+  Function *Callee = CB.getCalledFunction();
+  Function *Caller = CB.getCaller();
 
   if (!Callee || Callee->isDeclaration())
     return llvm::InlineCost::getNever("undefined callee");
 
-  if (CS.isNoInline())
+  if (CB.isNoInline())
     return llvm::InlineCost::getNever("noinline");
 
   TargetTransformInfo &TTI = TTIWP->getTTI(*Callee);
   if (!TTI.areInlineCompatible(Caller, Callee))
     return llvm::InlineCost::getNever("incompatible");
 
-  if (CS.hasFnAttr(Attribute::AlwaysInline)) {
+  if (CB.hasFnAttr(Attribute::AlwaysInline)) {
     auto IsViable = isInlineViable(*Callee);
     if (IsViable.isSuccess())
       return llvm::InlineCost::getAlways("alwaysinline viable");
     return llvm::InlineCost::getNever(IsViable.getFailureReason());
   }
 
-  if (isWrapperOnlyCall(CS))
+  if (isWrapperOnlyCall(CB))
     return llvm::InlineCost::getAlways("wrapper-only call");
 
   InlineParams LocalParams = Params;
-  LocalParams.DefaultThreshold = (int)getInlineThreshold(CS);
+  LocalParams.DefaultThreshold = (int)getInlineThreshold(CB);
   bool RemarksEnabled = false;
   const auto &BBs = Caller->getBasicBlockList();
   if (!BBs.empty()) {
     auto DI = OptimizationRemark(DEBUG_TYPE, "", DebugLoc(), &BBs.front());
     if (DI.isEnabled())
       RemarksEnabled = true;
   }
 
   OptimizationRemarkEmitter ORE(Caller);
   std::function<AssumptionCache &(Function &)> GetAssumptionCache =
       [this](Function &F) -> AssumptionCache & {
     return ACT->getAssumptionCache(F);
   };
 
-  auto IC = llvm::getInlineCost(cast<CallBase>(*CS.getInstruction()), Callee,
-                                LocalParams, TTI, GetAssumptionCache, None,
-                                GetTLI, PSI, RemarksEnabled ? &ORE : nullptr);
+  auto IC =
+      llvm::getInlineCost(CB, Callee, LocalParams, TTI, GetAssumptionCache,
+                          None, GetTLI, PSI, RemarksEnabled ? &ORE : nullptr);
 
   if (IC && !IC.isAlways() && !Callee->hasFnAttribute(Attribute::InlineHint)) {
     // Single BB does not increase total BB amount, thus subtract 1
     size_t Size = Caller->size() + Callee->size() - 1;
     if (MaxBB && Size > MaxBB)
       return llvm::InlineCost::getNever("max number of bb exceeded");
   }
   return IC;
 }

llvm/lib/Transforms/IPO/AlwaysInliner.cpp

 //
 //===----------------------------------------------------------------------===//
 
 #include "llvm/Transforms/IPO/AlwaysInliner.h"
 #include "llvm/ADT/SetVector.h"
 #include "llvm/Analysis/AssumptionCache.h"
 #include "llvm/Analysis/InlineCost.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
-#include "llvm/IR/CallSite.h"
 #include "llvm/IR/CallingConv.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/Type.h"
 #include "llvm/InitializePasses.h"
 #include "llvm/Transforms/IPO.h"
 #include "llvm/Transforms/IPO/Inliner.h"
PreservedAnalyses AlwaysInlinerPass::run(Module &M,
   FunctionAnalysisManager &FAM =
       MAM.getResult<FunctionAnalysisManagerModuleProxy>(M).getManager();
   std::function<AssumptionCache &(Function &)> GetAssumptionCache =
       [&](Function &F) -> AssumptionCache & {
     return FAM.getResult<AssumptionAnalysis>(F);
   };
   InlineFunctionInfo IFI(/*cg=*/nullptr, &GetAssumptionCache);
 
-  SmallSetVector<CallSite, 16> Calls;
+  SmallSetVector<CallBase *, 16> Calls;
   bool Changed = false;
   SmallVector<Function *, 16> InlinedFunctions;
   for (Function &F : M)
     if (!F.isDeclaration() && F.hasFnAttribute(Attribute::AlwaysInline) &&
         isInlineViable(F).isSuccess()) {
       Calls.clear();
 
       for (User *U : F.users())
-        if (auto CS = CallSite(U))
-          if (CS.getCalledFunction() == &F)
-            Calls.insert(CS);
+        if (auto *CB = dyn_cast<CallBase>(U))
+          if (CB->getCalledFunction() == &F)
+            Calls.insert(CB);
 
-      for (CallSite CS : Calls)
+      for (CallBase *CB : Calls)
         // FIXME: We really shouldn't be able to fail to inline at this point!
         // We should do something to log or check the inline failures here.
         Changed |=
-            InlineFunction(CS, IFI, /*CalleeAAR=*/nullptr, InsertLifetime)
+            InlineFunction(*CB, IFI, /*CalleeAAR=*/nullptr, InsertLifetime)
                 .isSuccess();
 
       // Remember to try and delete this function afterward. This both avoids
       // re-walking the rest of the module and avoids dealing with any iterator
       // invalidation issues while deleting functions.
       InlinedFunctions.push_back(&F);
     }
 
AlwaysInlinerLegacyPass(bool InsertLifetime)
     initializeAlwaysInlinerLegacyPassPass(*PassRegistry::getPassRegistry());
   }
 
   /// Main run interface method.  We override here to avoid calling skipSCC().
   bool runOnSCC(CallGraphSCC &SCC) override { return inlineCalls(SCC); }
 
   static char ID; // Pass identification, replacement for typeid
 
-  InlineCost getInlineCost(CallSite CS) override;
+  InlineCost getInlineCost(CallBase &CB) override;
 
   using llvm::Pass::doFinalization;
   bool doFinalization(CallGraph &CG) override {
     return removeDeadFunctions(CG, /*AlwaysInlineOnly=*/true);
   }
 };
 }
 

 /// using the powerful (but expensive) inline cost analysis. Instead it uses
 /// a very simple and boring direct walk of the instructions looking for
 /// impossible-to-inline constructs.
 ///
 /// Note, it would be possible to go to some lengths to cache the information
 /// computed here, but as we only expect to do this for relatively few and
 /// small functions which have the explicit attribute to force inlining, it is
 /// likely not worth it in practice.
-InlineCost AlwaysInlinerLegacyPass::getInlineCost(CallSite CS) {
-  Function *Callee = CS.getCalledFunction();
+InlineCost AlwaysInlinerLegacyPass::getInlineCost(CallBase &CB) {
+  Function *Callee = CB.getCalledFunction();
 
   // Only inline direct calls to functions with always-inline attributes
   // that are viable for inlining.
   if (!Callee)
     return InlineCost::getNever("indirect call");
 
   // FIXME: We shouldn't even get here for declarations.
   if (Callee->isDeclaration())
     return InlineCost::getNever("no definition");
 
-  if (!CS.hasFnAttr(Attribute::AlwaysInline))
+  if (!CB.hasFnAttr(Attribute::AlwaysInline))
     return InlineCost::getNever("no alwaysinline attribute");
 
   auto IsViable = isInlineViable(*Callee);
   if (!IsViable.isSuccess())
     return InlineCost::getNever(IsViable.getFailureReason());
 
   return InlineCost::getAlways("always inliner");
 }

llvm/lib/Transforms/IPO/InlineSimple.cpp

 //
 //===----------------------------------------------------------------------===//
 
 #include "llvm/Analysis/AssumptionCache.h"
 #include "llvm/Analysis/InlineCost.h"
 #include "llvm/Analysis/ProfileSummaryInfo.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
 #include "llvm/Analysis/TargetTransformInfo.h"
-#include "llvm/IR/CallSite.h"
 #include "llvm/IR/CallingConv.h"
 #include "llvm/IR/DataLayout.h"
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/Module.h"
 #include "llvm/IR/Type.h"
 #include "llvm/InitializePasses.h"
 #include "llvm/Transforms/IPO.h"
 #include "llvm/Transforms/IPO/Inliner.h"
public:
 
   explicit SimpleInliner(InlineParams Params)
       : LegacyInlinerBase(ID), Params(std::move(Params)) {
     initializeSimpleInlinerPass(*PassRegistry::getPassRegistry());
   }
 
   static char ID; // Pass identification, replacement for typeid
 
-  InlineCost getInlineCost(CallSite CS) override {
-    Function *Callee = CS.getCalledFunction();
+  InlineCost getInlineCost(CallBase &CB) override {
+    Function *Callee = CB.getCalledFunction();
     TargetTransformInfo &TTI = TTIWP->getTTI(*Callee);
 
     bool RemarksEnabled = false;
-    const auto &BBs = CS.getCaller()->getBasicBlockList();
+    const auto &BBs = CB.getCaller()->getBasicBlockList();
     if (!BBs.empty()) {
       auto DI = OptimizationRemark(DEBUG_TYPE, "", DebugLoc(), &BBs.front());
       if (DI.isEnabled())
         RemarksEnabled = true;
     }
-    OptimizationRemarkEmitter ORE(CS.getCaller());
+    OptimizationRemarkEmitter ORE(CB.getCaller());
 
     std::function<AssumptionCache &(Function &)> GetAssumptionCache =
         [&](Function &F) -> AssumptionCache & {
       return ACT->getAssumptionCache(F);
     };
-    return llvm::getInlineCost(
-        cast<CallBase>(*CS.getInstruction()), Params, TTI, GetAssumptionCache,
-        /*GetBFI=*/None, GetTLI, PSI, RemarksEnabled ? &ORE : nullptr);
+    return llvm::getInlineCost(CB, Params, TTI, GetAssumptionCache,
+                               /*GetBFI=*/None, GetTLI, PSI,
+                               RemarksEnabled ? &ORE : nullptr);
   }
 
   bool runOnSCC(CallGraphSCC &SCC) override;
   void getAnalysisUsage(AnalysisUsage &AU) const override;
 
 private:
   TargetTransformInfoWrapperPass *TTIWP;
 

llvm/lib/Transforms/IPO/Inliner.cpp

static InlineResult inlineCallIfPossible(
     ImportedFunctionsInliningStatistics &ImportedFunctionsStats) {
   Function *Callee = CS.getCalledFunction();
   Function *Caller = CS.getCaller();
 
   AAResults &AAR = AARGetter(*Callee);
 
   // Try to inline the function.  Get the list of static allocas that were
   // inlined.
-  InlineResult IR = InlineFunction(&CS, IFI, &AAR, InsertLifetime);
+  InlineResult IR = InlineFunction(CS, IFI, &AAR, InsertLifetime);
   if (!IR.isSuccess())
     return IR;
 
   if (InlinerFunctionImportStats != InlinerFunctionImportStatsOpts::No)
     ImportedFunctionsStats.recordInline(*Caller, *Callee);
 
   AttributeFuncs::mergeAttributesForInlining(*Caller, *Callee);
 
for (; I < (int)Calls.size() && Calls[I].first->getCaller() == &F; ++I) {
           &FAM.getResult<BlockFrequencyAnalysis>(Callee));
 
       // Get DebugLoc to report. CS will be invalid after Inliner.
       DebugLoc DLoc = CS->getDebugLoc();
       BasicBlock *Block = CS->getParent();
 
       using namespace ore;
 
-      InlineResult IR = InlineFunction(CS, IFI);
+      InlineResult IR = InlineFunction(*CS, IFI);
       if (!IR.isSuccess()) {
         setInlineRemark(*CS, std::string(IR.getFailureReason()) + "; " +
                                  inlineCostStr(*OIC));
         ORE.emit([&]() {
           return OptimizationRemarkMissed(DEBUG_TYPE, "NotInlined", DLoc, Block)
                  << NV("Callee", &Callee) << " will not be inlined into "
                  << NV("Caller", &F) << ": "
                  << NV("Reason", IR.getFailureReason());

llvm/lib/Transforms/IPO/PartialInlining.cpp

 #include "llvm/Analysis/LoopInfo.h"
 #include "llvm/Analysis/OptimizationRemarkEmitter.h"
 #include "llvm/Analysis/ProfileSummaryInfo.h"
 #include "llvm/Analysis/TargetLibraryInfo.h"
 #include "llvm/Analysis/TargetTransformInfo.h"
 #include "llvm/IR/Attributes.h"
 #include "llvm/IR/BasicBlock.h"
 #include "llvm/IR/CFG.h"
-#include "llvm/IR/CallSite.h"
 #include "llvm/IR/DebugLoc.h"
 #include "llvm/IR/DiagnosticInfo.h"
 #include "llvm/IR/Dominators.h"
 #include "llvm/IR/Function.h"
 #include "llvm/IR/InstrTypes.h"
 #include "llvm/IR/Instruction.h"
 #include "llvm/IR/Instructions.h"
 #include "llvm/IR/IntrinsicInst.h"
private:
   ProfileSummaryInfo *PSI;
 
   // Return the frequency of the OutlininingBB relative to F's entry point.
   // The result is no larger than 1 and is represented using BP.
   // (Note that the outlined region's 'head' block can only have incoming
   // edges from the guarding entry blocks).
   BranchProbability getOutliningCallBBRelativeFreq(FunctionCloner &Cloner);
 
-  // Return true if the callee of CS should be partially inlined with
+  // Return true if the callee of CB should be partially inlined with
   // profit.
-  bool shouldPartialInline(CallSite CS, FunctionCloner &Cloner,
+  bool shouldPartialInline(CallBase &CB, FunctionCloner &Cloner,
                            BlockFrequency WeightedOutliningRcost,
                            OptimizationRemarkEmitter &ORE);
 
   // Try to inline DuplicateFunction (cloned from F with call to
   // the OutlinedFunction into its callers. Return true
   // if there is any successful inlining.
   bool tryPartialInline(FunctionCloner &Cloner);
 
   // Compute the mapping from use site of DuplicationFunction to the enclosing
   // BB's profile count.
   void computeCallsiteToProfCountMap(Function *DuplicateFunction,
                                      DenseMap<User *, uint64_t> &SiteCountMap);
 
   bool IsLimitReached() {
     return (MaxNumPartialInlining != -1 &&
             NumPartialInlining >= MaxNumPartialInlining);
   }
 
-  static CallSite getCallSite(User *U) {
-    CallSite CS;
-    if (CallInst *CI = dyn_cast<CallInst>(U))
-      CS = CallSite(CI);
-    else if (InvokeInst *II = dyn_cast<InvokeInst>(U))
-      CS = CallSite(II);
-    else
-      llvm_unreachable("All uses must be calls");
-    return CS;
+  static CallBase *getSupportedCallBase(User *U) {
+    if (isa<CallInst>(U) || isa<InvokeInst>(U))
+      return cast<CallBase>(U);
+    return nullptr;
   }
 
-  static CallSite getOneCallSiteTo(Function *F) {
+  static CallBase *getOneCallSiteTo(Function *F) {
     User *User = *F->user_begin();
-    return getCallSite(User);
+    return getSupportedCallBase(User);
   }
 
   std::tuple<DebugLoc, BasicBlock *> getOneDebugLoc(Function *F) {
-    CallSite CS = getOneCallSiteTo(F);
-    DebugLoc DLoc = CS.getInstruction()->getDebugLoc();
-    BasicBlock *Block = CS.getParent();
+    CallBase *CB = getOneCallSiteTo(F);
+    DebugLoc DLoc = CB->getDebugLoc();
+    BasicBlock *Block = CB->getParent();
     return std::make_tuple(DLoc, Block);
   }
 
   // Returns the costs associated with function outlining:
   // - The first value is the non-weighted runtime cost for making the call
   //   to the outlined function, including the addtional  setup cost in the
   //    outlined function itself;
   // - The second value is the estimated size of the new call sequence in
PartialInlinerImpl::getOutliningCallBBRelativeFreq(FunctionCloner &Cloner) {
 
   OutlineRegionRelFreq = std::max(
       OutlineRegionRelFreq, BranchProbability(OutlineRegionFreqPercent, 100));
 
   return OutlineRegionRelFreq;
 }
 
 bool PartialInlinerImpl::shouldPartialInline(
-    CallSite CS, FunctionCloner &Cloner,
-    BlockFrequency WeightedOutliningRcost,
+    CallBase &CB, FunctionCloner &Cloner, BlockFrequency WeightedOutliningRcost,
     OptimizationRemarkEmitter &ORE) {
   using namespace ore;
 
-  Instruction *Call = CS.getInstruction();
-  Function *Callee = CS.getCalledFunction();
+  Function *Callee = CB.getCalledFunction();
   assert(Callee == Cloner.ClonedFunc);
 
   if (SkipCostAnalysis)
     return isInlineViable(*Callee).isSuccess();
 
-  Function *Caller = CS.getCaller();
+  Function *Caller = CB.getCaller();
   auto &CalleeTTI = (*GetTTI)(*Callee);
   bool RemarksEnabled =
       Callee->getContext().getDiagHandlerPtr()->isMissedOptRemarkEnabled(
           DEBUG_TYPE);
-  assert(Call && "invalid callsite for partial inline");
-  InlineCost IC = getInlineCost(cast<CallBase>(*Call), getInlineParams(),
-                                CalleeTTI, *GetAssumptionCache, GetBFI, *GetTLI,
-                                PSI, RemarksEnabled ? &ORE : nullptr);
+  InlineCost IC =
+      getInlineCost(CB, getInlineParams(), CalleeTTI, *GetAssumptionCache,
+                    GetBFI, *GetTLI, PSI, RemarksEnabled ? &ORE : nullptr);
 
   if (IC.isAlways()) {
     ORE.emit([&]() {
-      return OptimizationRemarkAnalysis(DEBUG_TYPE, "AlwaysInline", Call)
+      return OptimizationRemarkAnalysis(DEBUG_TYPE, "AlwaysInline", &CB)
              << NV("Callee", Cloner.OrigFunc)
              << " should always be fully inlined, not partially";
     });
     return false;
   }
 
   if (IC.isNever()) {
     ORE.emit([&]() {
-      return OptimizationRemarkMissed(DEBUG_TYPE, "NeverInline", Call)
+      return OptimizationRemarkMissed(DEBUG_TYPE, "NeverInline", &CB)
              << NV("Callee", Cloner.OrigFunc) << " not partially inlined into "
              << NV("Caller", Caller)
              << " because it should never be inlined (cost=never)";
     });
     return false;
   }
 
   if (!IC) {
     ORE.emit([&]() {
-      return OptimizationRemarkAnalysis(DEBUG_TYPE, "TooCostly", Call)
+      return OptimizationRemarkAnalysis(DEBUG_TYPE, "TooCostly", &CB)
              << NV("Callee", Cloner.OrigFunc) << " not partially inlined into "
              << NV("Caller", Caller) << " because too costly to inline (cost="
              << NV("Cost", IC.getCost()) << ", threshold="
              << NV("Threshold", IC.getCostDelta() + IC.getCost()) << ")";
     });
     return false;
   }
   const DataLayout &DL = Caller->getParent()->getDataLayout();
 
   // The savings of eliminating the call:
-  int NonWeightedSavings = getCallsiteCost(cast<CallBase>(*Call), DL);
+  int NonWeightedSavings = getCallsiteCost(CB, DL);
   BlockFrequency NormWeightedSavings(NonWeightedSavings);
 
   // Weighted saving is smaller than weighted cost, return false
   if (NormWeightedSavings < WeightedOutliningRcost) {
     ORE.emit([&]() {
       return OptimizationRemarkAnalysis(DEBUG_TYPE, "OutliningCallcostTooHigh",
-                                        Call)
+                                        &CB)
              << NV("Callee", Cloner.OrigFunc) << " not partially inlined into "
              << NV("Caller", Caller) << " runtime overhead (overhead="
              << NV("Overhead", (unsigned)WeightedOutliningRcost.getFrequency())
              << ", savings="
              << NV("Savings", (unsigned)NormWeightedSavings.getFrequency())
              << ")"
              << " of making the outlined call is too high";
     });
 
     return false;
   }
 
   ORE.emit([&]() {
-    return OptimizationRemarkAnalysis(DEBUG_TYPE, "CanBePartiallyInlined", Call)
+    return OptimizationRemarkAnalysis(DEBUG_TYPE, "CanBePartiallyInlined", &CB)
            << NV("Callee", Cloner.OrigFunc) << " can be partially inlined into "
            << NV("Caller", Caller) << " with cost=" << NV("Cost", IC.getCost())
            << " (threshold="
            << NV("Threshold", IC.getCostDelta() + IC.getCost()) << ")";
   });
   return true;
 }
 
auto ComputeCurrBFI = [&,this](Function *Caller) {
         CurrentCallerBFI = TempBFI.get();
       } else {
         // New pass manager:
         CurrentCallerBFI = &(*GetBFI)(*Caller);
       }
   };
 
   for (User *User : Users) {
-    CallSite CS = getCallSite(User);
-    Function *Caller = CS.getCaller();
+    CallBase *CB = getSupportedCallBase(User);
+    Function *Caller = CB->getCaller();
     if (CurrentCaller != Caller) {
       CurrentCaller = Caller;
       ComputeCurrBFI(Caller);
     } else {
       assert(CurrentCallerBFI && "CallerBFI is not set");
     }
-    BasicBlock *CallBB = CS.getInstruction()->getParent();
+    BasicBlock *CallBB = CB->getParent();
     auto Count = CurrentCallerBFI->getBlockProfileCount(CallBB);
     if (Count)
       CallSiteToProfCountMap[User] = *Count;
     else
       CallSiteToProfCountMap[User] = 0;
   }
 }
 

 #endif
     // Do not extract regions that have live exit variables.
     if (Outputs.size() > 0 && !ForceLiveExit)
       continue;
 
     Function *OutlinedFunc = CE.extractCodeRegion(CEAC);
 
     if (OutlinedFunc) {
-      CallSite OCS = PartialInlinerImpl::getOneCallSiteTo(OutlinedFunc);
-      BasicBlock *OutliningCallBB = OCS.getInstruction()->getParent();
+      CallBase *OCS = PartialInlinerImpl::getOneCallSiteTo(OutlinedFunc);
+      BasicBlock *OutliningCallBB = OCS->getParent();
       assert(OutliningCallBB->getParent() == ClonedFunc);
       OutlinedFunctions.push_back(std::make_pair(OutlinedFunc,OutliningCallBB));
       NumColdRegionsOutlined++;
       OutlinedRegionCost += CurrentOutlinedRegionCost;
 
       if (MarkOutlinedColdCC) {
         OutlinedFunc->setCallingConv(CallingConv::Cold);
-        OCS.setCallingConv(CallingConv::Cold);
+        OCS->setCallingConv(CallingConv::Cold);
       }
     } else
       ORE.emit([&]() {
         return OptimizationRemarkMissed(DEBUG_TYPE, "ExtractFailed",
                                         &RegionInfo.Region.front()->front())
                << "Failed to extract region at block "
                << ore::NV("Block", RegionInfo.Region.front());
       });
Function *OutlinedFunc =
       CodeExtractor(ToExtract, &DT, /*AggregateArgs*/ false,
                     ClonedFuncBFI.get(), &BPI, LookupAC(*ClonedFunc),
                     /* AllowVarargs */ true)
           .extractCodeRegion(CEAC);
 
   if (OutlinedFunc) {
     BasicBlock *OutliningCallBB =
         PartialInlinerImpl::getOneCallSiteTo(OutlinedFunc)
-            .getInstruction()
             ->getParent();
     assert(OutliningCallBB->getParent() == ClonedFunc);
     OutlinedFunctions.push_back(std::make_pair(OutlinedFunc, OutliningCallBB));
   } else
     ORE.emit([&]() {
       return OptimizationRemarkMissed(DEBUG_TYPE, "ExtractFailed",
                                       &ToExtract.front()->front())
              << "Failed to extract region at block "
bool PartialInlinerImpl::tryPartialInline(FunctionCloner &Cloner) {
   if (CalleeEntryCount)
     computeCallsiteToProfCountMap(Cloner.ClonedFunc, CallSiteToProfCountMap);
 
   uint64_t CalleeEntryCountV =
       (CalleeEntryCount ? CalleeEntryCount.getCount() : 0);
 
   bool AnyInline = false;
   for (User *User : Users) {
-    CallSite CS = getCallSite(User);
+    CallBase *CB = getSupportedCallBase(User);
 
     if (IsLimitReached())
       continue;
 
-    OptimizationRemarkEmitter CallerORE(CS.getCaller());
-    if (!shouldPartialInline(CS, Cloner, WeightedRcost, CallerORE))
+    OptimizationRemarkEmitter CallerORE(CB->getCaller());
+    if (!shouldPartialInline(*CB, Cloner, WeightedRcost, CallerORE))
       continue;
 
     // Construct remark before doing the inlining, as after successful inlining
     // the callsite is removed.
-    OptimizationRemark OR(DEBUG_TYPE, "PartiallyInlined", CS.getInstruction());
+    OptimizationRemark OR(DEBUG_TYPE, "PartiallyInlined", CB);
     OR << ore::NV("Callee", Cloner.OrigFunc) << " partially inlined into "
-       << ore::NV("Caller", CS.getCaller());
+       << ore::NV("Caller", CB->getCaller());
 
     InlineFunctionInfo IFI(nullptr, GetAssumptionCache, PSI);
     // We can only forward varargs when we outlined a single region, else we
     // bail on vararg functions.
-    if (!InlineFunction(CS, IFI, nullptr, true,
+    if (!InlineFunction(*CB, IFI, nullptr, true,
                         (Cloner.ClonedOI ? Cloner.OutlinedFunctions.back().first
                                          : nullptr))
              .isSuccess())
       continue;
 
     CallerORE.emit(OR);
 
     // Now update the entry count:

llvm/lib/Transforms/IPO/SampleProfile.cpp

if (!DIL)
     return Samples;
 
   auto it = DILocation2SampleMap.try_emplace(DIL,nullptr);
   if (it.second)
     it.first->second = Samples->findFunctionSamples(DIL);
   return it.first->second;
 }
 
+// FIXME(CallSite): Parameter should be CallBase&, as it's assumed to be that,
+// and non-null.
 bool SampleProfileLoader::inlineCallInstruction(Instruction *I) {
   assert(isa<CallInst>(I) || isa<InvokeInst>(I));
-  CallSite CS(I);
+  CallBase &CS = *cast<CallBase>(I);
   Function *CalledFunction = CS.getCalledFunction();
   assert(CalledFunction);
   DebugLoc DLoc = I->getDebugLoc();
   BasicBlock *BB = I->getParent();
   InlineParams Params = getInlineParams();
   Params.ComputeFullInlineCost = true;
   // Checks if there is anything in the reachable portion of the callee at
   // this callsite that makes this inlining potentially illegal. Need to

llvm/lib/Transforms/Scalar/PlaceSafepoints.cpp

InsertSafepointPoll(Instruction *InsertBefore,
   else
     Before--;
 
   After++;
   assert(After != OrigBB->end() && "must have successor");
 
   // Do the actual inlining
   InlineFunctionInfo IFI;
-  bool InlineStatus = InlineFunction(PollCall, IFI).isSuccess();
+  bool InlineStatus = InlineFunction(*PollCall, IFI).isSuccess();
   assert(InlineStatus && "inline must succeed");
   (void)InlineStatus; // suppress warning in release-asserts
 
   // Check post-conditions
   assert(IFI.StaticAllocas.empty() && "can't have allocs");
 
   std::vector<CallInst *> Calls; // new calls
   DenseSet<BasicBlock *> BBs;    // new BBs + insertee

llvm/lib/Transforms/Utils/InlineFunction.cpp

static cl::opt<bool> UpdateLoadMetadataDuringInlining(
             cl::desc("Update metadata on loads within inlined body"));
 
 static cl::opt<unsigned> InlinerAttributeWindow(
     "max-inst-checked-for-throw-during-inlining", cl::Hidden,
     cl::desc("the maximum number of instructions analyzed for may throw during "
              "attribute inference in inlined body"),
     cl::init(4));
 
-llvm::InlineResult llvm::InlineFunction(CallBase *CB, InlineFunctionInfo &IFI,
-                                        AAResults *CalleeAAR,
-                                        bool InsertLifetime) {
-  return InlineFunction(CallSite(CB), IFI, CalleeAAR, InsertLifetime);
-}
-
 namespace {
 
   /// A class for recording information about inlining a landing pad.
   class LandingPadInliningInfo {
     /// Destination of the invoke's unwind.
     BasicBlock *OuterResumeDest;
 
     /// Destination for the callee's resume.
static void HandleInlinedEHPad(InvokeInst *II, BasicBlock *FirstNewBlock,
   // invoke instruction. Eliminate these entries (which might even delete the
   // PHI node) now.
   UnwindDest->removePredecessor(InvokeBB);
 }
 
 /// When inlining a call site that has !llvm.mem.parallel_loop_access or
 /// llvm.access.group metadata, that metadata should be propagated to all
 /// memory-accessing cloned instructions.
-static void PropagateParallelLoopAccessMetadata(CallSite CS,
+static void PropagateParallelLoopAccessMetadata(CallBase &CB,
                                                 ValueToValueMapTy &VMap) {
-  MDNode *M =
-    CS.getInstruction()->getMetadata(LLVMContext::MD_mem_parallel_loop_access);
-  MDNode *CallAccessGroup =
-      CS.getInstruction()->getMetadata(LLVMContext::MD_access_group);
+  MDNode *M = CB.getMetadata(LLVMContext::MD_mem_parallel_loop_access);
+  MDNode *CallAccessGroup = CB.getMetadata(LLVMContext::MD_access_group);
   if (!M && !CallAccessGroup)
     return;
 
   for (ValueToValueMapTy::iterator VMI = VMap.begin(), VMIE = VMap.end();
        VMI != VMIE; ++VMI) {
     if (!VMI->second)
       continue;
 

 
 /// When inlining a function that contains noalias scope metadata,
 /// this metadata needs to be cloned so that the inlined blocks
 /// have different "unique scopes" at every call site. Were this not done, then
 /// aliasing scopes from a function inlined into a caller multiple times could
 /// not be differentiated (and this would lead to miscompiles because the
 /// non-aliasing property communicated by the metadata could have
 /// call-site-specific control dependencies).
-static void CloneAliasScopeMetadata(CallSite CS, ValueToValueMapTy &VMap) {
-  const Function *CalledFunc = CS.getCalledFunction();
+static void CloneAliasScopeMetadata(CallBase &CB, ValueToValueMapTy &VMap) {
+  const Function *CalledFunc = CB.getCalledFunction();
   SetVector<const MDNode *> MD;
 
   // Note: We could only clone the metadata if it is already used in the
   // caller. I'm omitting that check here because it might confuse
   // inter-procedural alias analysis passes. We can revisit this if it becomes
   // an efficiency or overhead problem.
 
   for (const BasicBlock &I : *CalledFunc)
for (ValueToValueMapTy::iterator VMI = VMap.begin(), VMIE = VMap.end();
     if (!NI)
       continue;
 
     if (MDNode *M = NI->getMetadata(LLVMContext::MD_alias_scope)) {
       MDNode *NewMD = MDMap[M];
       // If the call site also had alias scope metadata (a list of scopes to
       // which instructions inside it might belong), propagate those scopes to
       // the inlined instructions.
-      if (MDNode *CSM =
-              CS.getInstruction()->getMetadata(LLVMContext::MD_alias_scope))
+      if (MDNode *CSM = CB.getMetadata(LLVMContext::MD_alias_scope))
         NewMD = MDNode::concatenate(NewMD, CSM);
       NI->setMetadata(LLVMContext::MD_alias_scope, NewMD);
     } else if (NI->mayReadOrWriteMemory()) {
-      if (MDNode *M =
-              CS.getInstruction()->getMetadata(LLVMContext::MD_alias_scope))
+      if (MDNode *M = CB.getMetadata(LLVMContext::MD_alias_scope))
         NI->setMetadata(LLVMContext::MD_alias_scope, M);
     }
 
     if (MDNode *M = NI->getMetadata(LLVMContext::MD_noalias)) {
       MDNode *NewMD = MDMap[M];
       // If the call site also had noalias metadata (a list of scopes with
       // which instructions inside it don't alias), propagate those scopes to
       // the inlined instructions.
-      if (MDNode *CSM =
-              CS.getInstruction()->getMetadata(LLVMContext::MD_noalias))
+      if (MDNode *CSM = CB.getMetadata(LLVMContext::MD_noalias))
         NewMD = MDNode::concatenate(NewMD, CSM);
       NI->setMetadata(LLVMContext::MD_noalias, NewMD);
     } else if (NI->mayReadOrWriteMemory()) {
-      if (MDNode *M = CS.getInstruction()->getMetadata(LLVMContext::MD_noalias))
+      if (MDNode *M = CB.getMetadata(LLVMContext::MD_noalias))
         NI->setMetadata(LLVMContext::MD_noalias, M);
     }
   }
 }
 
 /// If the inlined function has noalias arguments,
 /// then add new alias scopes for each noalias argument, tag the mapped noalias
 /// parameters with noalias metadata specifying the new scope, and tag all
 /// non-derived loads, stores and memory intrinsics with the new alias scopes.
-static void AddAliasScopeMetadata(CallSite CS, ValueToValueMapTy &VMap,
+static void AddAliasScopeMetadata(CallBase &CB, ValueToValueMapTy &VMap,
                                   const DataLayout &DL, AAResults *CalleeAAR) {
   if (!EnableNoAliasConversion)
     return;
 
-  const Function *CalledFunc = CS.getCalledFunction();
+  const Function *CalledFunc = CB.getCalledFunction();
   SmallVector<const Argument *, 4> NoAliasArgs;
 
   for (const Argument &Arg : CalledFunc->args())
-    if (CS.paramHasAttr(Arg.getArgNo(), Attribute::NoAlias) && !Arg.use_empty())
+    if (CB.paramHasAttr(Arg.getArgNo(), Attribute::NoAlias) && !Arg.use_empty())
       NoAliasArgs.push_back(&Arg);
 
   if (NoAliasArgs.empty())
     return;
 
   // To do a good job, if a noalias variable is captured, we need to know if
   // the capture point dominates the particular use we're considering.
   DominatorTree DT;
if (const Instruction *I = dyn_cast<Instruction>(VMI->first)) {
                              isa<ConstantDataVector>(V) || isa<UndefValue>(V);
         if (IsNonPtrConst)
           continue;
 
         // If this is anything other than a noalias argument, then we cannot
         // completely describe the aliasing properties using alias.scope
         // metadata (and, thus, won't add any).
         if (const Argument *A = dyn_cast<Argument>(V)) {
-          if (!CS.paramHasAttr(A->getArgNo(), Attribute::NoAlias))
+          if (!CB.paramHasAttr(A->getArgNo(), Attribute::NoAlias))
             UsesAliasingPtr = true;
         } else {
           UsesAliasingPtr = true;
         }
 
         // If this is not some identified function-local object (which cannot
         // directly alias a noalias argument), or some other argument (which,
         // by definition, also cannot alias a noalias argument), then we could
static bool MayContainThrowingOrExitingCall(Instruction *Begin,
   // transfer execution to successor.
   for (auto &I : make_range(Begin->getIterator(), End->getIterator()))
     if (NumInstChecked++ > InlinerAttributeWindow ||
         !isGuaranteedToTransferExecutionToSuccessor(&I))
       return true;
   return false;
 }
 
-static AttrBuilder IdentifyValidAttributes(CallSite CS) {
+static AttrBuilder IdentifyValidAttributes(CallBase &CB) {
 
-  AttrBuilder AB(CS.getAttributes(), AttributeList::ReturnIndex);
+  AttrBuilder AB(CB.getAttributes(), AttributeList::ReturnIndex);
   if (AB.empty())
     return AB;
   AttrBuilder Valid;
   // Only allow these white listed attributes to be propagated back to the
   // callee. This is because other attributes may only be valid on the call
   // itself, i.e. attributes such as signext and zeroext.
   if (auto DerefBytes = AB.getDereferenceableBytes())
     Valid.addDereferenceableAttr(DerefBytes);
   if (auto DerefOrNullBytes = AB.getDereferenceableOrNullBytes())
     Valid.addDereferenceableOrNullAttr(DerefOrNullBytes);
   if (AB.contains(Attribute::NoAlias))
     Valid.addAttribute(Attribute::NoAlias);
   if (AB.contains(Attribute::NonNull))
     Valid.addAttribute(Attribute::NonNull);
   return Valid;
 }
 
-static void AddReturnAttributes(CallSite CS, ValueToValueMapTy &VMap) {
+static void AddReturnAttributes(CallBase &CB, ValueToValueMapTy &VMap) {
   if (!UpdateReturnAttributes && !UpdateLoadMetadataDuringInlining)
     return;
 
-  AttrBuilder Valid = IdentifyValidAttributes(CS);
+  AttrBuilder Valid = IdentifyValidAttributes(CB);
   if (Valid.empty())
     return;
-  auto *CalledFunction = CS.getCalledFunction();
+  auto *CalledFunction = CB.getCalledFunction();
   auto &Context = CalledFunction->getContext();
 
   auto getExpectedRV = [&](Value *V) -> Instruction * {
     if (UpdateReturnAttributes && isa<CallBase>(V))
       return dyn_cast_or_null<CallBase>(VMap.lookup(V));
     if (UpdateLoadMetadataDuringInlining && isa<LoadInst>(V))
       return dyn_cast_or_null<LoadInst>(VMap.lookup(V));
     return nullptr;
if (RI->getParent() != RetVal->getParent() ||
         MayContainThrowingOrExitingCall(RetVal, RI))
       continue;
     // Add to the existing attributes of NewRetVal, i.e. the cloned call
     // instruction.
     // NB! When we have the same attribute already existing on NewRetVal, but
     // with a differing value, the AttributeList's merge API honours the already
     // existing attribute value (i.e. attributes such as dereferenceable,
     // dereferenceable_or_null etc). See AttrBuilder::merge for more details.
-    if (auto *CB = dyn_cast<CallBase>(NewRetVal)) {
-      AttributeList AL = CB->getAttributes();
+    if (auto *NewRetValCB = dyn_cast<CallBase>(NewRetVal)) {
+      AttributeList AL = NewRetValCB->getAttributes();
       AttributeList NewAL =
           AL.addAttributes(Context, AttributeList::ReturnIndex, Valid);
-      CB->setAttributes(NewAL);
+      NewRetValCB->setAttributes(NewAL);
     } else {
       auto *NewLI = cast<LoadInst>(NewRetVal);
-      if (CS.isReturnNonNull())
+      if (CB.isReturnNonNull())
         NewLI->setMetadata(LLVMContext::MD_nonnull, CreateMDNode(1));
       // If the load already has a dereferenceable/dereferenceable_or_null
       // metadata, we should honour it.
       if (uint64_t DerefBytes = Valid.getDereferenceableBytes())
        if(!NewLI->getMetadata(LLVMContext::MD_dereferenceable))
          NewLI->setMetadata(LLVMContext::MD_dereferenceable,
                             CreateMDNode(DerefBytes));
       if (uint64_t DerefOrNullBytes = Valid.getDereferenceableOrNullBytes())
        if (!NewLI->getMetadata(LLVMContext::MD_dereferenceable_or_null))
          NewLI->setMetadata(LLVMContext::MD_dereferenceable_or_null,
                             CreateMDNode(DerefOrNullBytes));
     }
-
   }
 }
 
 /// If the inlined function has non-byval align arguments, then
 /// add @llvm.assume-based alignment assumptions to preserve this information.
-static void AddAlignmentAssumptions(CallSite CS, InlineFunctionInfo &IFI) {
+static void AddAlignmentAssumptions(CallBase &CB, InlineFunctionInfo &IFI) {
   if (!PreserveAlignmentAssumptions || !IFI.GetAssumptionCache)
     return;
 
-  AssumptionCache *AC = &(*IFI.GetAssumptionCache)(*CS.getCaller());
-  auto &DL = CS.getCaller()->getParent()->getDataLayout();
+  AssumptionCache *AC = &(*IFI.GetAssumptionCache)(*CB.getCaller());
+  auto &DL = CB.getCaller()->getParent()->getDataLayout();
 
   // To avoid inserting redundant assumptions, we should check for assumptions
   // already in the caller. To do this, we might need a DT of the caller.
   DominatorTree DT;
   bool DTCalculated = false;
 
-  Function *CalledFunc = CS.getCalledFunction();
+  Function *CalledFunc = CB.getCalledFunction();
   for (Argument &Arg : CalledFunc->args()) {
     unsigned Align = Arg.getType()->isPointerTy() ? Arg.getParamAlignment() : 0;
     if (Align && !Arg.hasByValOrInAllocaAttr() && !Arg.hasNUses(0)) {
       if (!DTCalculated) {
-        DT.recalculate(*CS.getCaller());
+        DT.recalculate(*CB.getCaller());
         DTCalculated = true;
       }
 
       // If we can already prove the asserted alignment in the context of the
       // caller, then don't bother inserting the assumption.
-      Value *ArgVal = CS.getArgument(Arg.getArgNo());
-      if (getKnownAlignment(ArgVal, DL, CS.getInstruction(), AC, &DT) >= Align)
+      Value *ArgVal = CB.getArgOperand(Arg.getArgNo());
+      if (getKnownAlignment(ArgVal, DL, &CB, AC, &DT) >= Align)
         continue;
 
-      CallInst *NewAsmp = IRBuilder<>(CS.getInstruction())
-                              .CreateAlignmentAssumption(DL, ArgVal, Align);
+      CallInst *NewAsmp =
+          IRBuilder<>(&CB).CreateAlignmentAssumption(DL, ArgVal, Align);
       AC->registerAssumption(NewAsmp);
     }
   }
 }
 
 /// Once we have cloned code over from a callee into the caller,
 /// update the specified callgraph to reflect the changes we made.
 /// Note that it's possible that not all code was copied over, so only
 /// some edges of the callgraph may remain.
-static void UpdateCallGraphAfterInlining(CallSite CS,
+static void UpdateCallGraphAfterInlining(CallBase &CB,
                                          Function::iterator FirstNewBlock,
                                          ValueToValueMapTy &VMap,
                                          InlineFunctionInfo &IFI) {
   CallGraph &CG = *IFI.CG;
-  const Function *Caller = CS.getCaller();
-  const Function *Callee = CS.getCalledFunction();
+  const Function *Caller = CB.getCaller();
+  const Function *Callee = CB.getCalledFunction();
   CallGraphNode *CalleeNode = CG[Callee];
   CallGraphNode *CallerNode = CG[Caller];
 
   // Since we inlined some uninlined call sites in the callee into the caller,
   // add edges from the caller to all of the callees of the callee.
   CallGraphNode::iterator I = CalleeNode->begin(), E = CalleeNode->end();
 
   // Consider the case where CalleeNode == CallerNode.
if (!I->second->getFunction())
         continue;
       }
 
     CallerNode->addCalledFunction(NewCall, I->second);
   }
 
   // Update the call graph by deleting the edge from Callee to Caller.  We must
   // do this after the loop above in case Caller and Callee are the same.
-  CallerNode->removeCallEdgeFor(*cast<CallBase>(CS.getInstruction()));
+  CallerNode->removeCallEdgeFor(*cast<CallBase>(&CB));
 }
 
 static void HandleByValArgumentInit(Value *Dst, Value *Src, Module *M,
                                     BasicBlock *InsertBlock,
                                     InlineFunctionInfo &IFI) {
   Type *AggTy = cast<PointerType>(Src->getType())->getElementType();
   IRBuilder<> Builder(InsertBlock, InsertBlock->begin());
 

 /// This function inlines the called function into the basic block of the
 /// caller. This returns false if it is not possible to inline this call.
 /// The program is still in a well defined state if this occurs though.
 ///
 /// Note that this only does one level of inlining.  For example, if the
 /// instruction 'call B' is inlined, and 'B' calls 'C', then the call to 'C' now
 /// exists in the instruction stream.  Similarly this will inline a recursive
 /// function by one level.
-llvm::InlineResult llvm::InlineFunction(CallSite CS, InlineFunctionInfo &IFI,
+llvm::InlineResult llvm::InlineFunction(CallBase &CB, InlineFunctionInfo &IFI,
                                         AAResults *CalleeAAR,
                                         bool InsertLifetime,
                                         Function *ForwardVarArgsTo) {
-  Instruction *TheCall = CS.getInstruction();
-  assert(TheCall->getParent() && TheCall->getFunction()
-         && "Instruction not in function!");
+  assert(CB.getParent() && CB.getFunction() && "Instruction not in function!");
 
   // FIXME: we don't inline callbr yet.
-  if (isa<CallBrInst>(TheCall))
+  if (isa<CallBrInst>(CB))
     return InlineResult::failure("We don't inline callbr yet.");
 
   // If IFI has any state in it, zap it before we fill it in.
   IFI.reset();
 
-  Function *CalledFunc = CS.getCalledFunction();
+  Function *CalledFunc = CB.getCalledFunction();
   if (!CalledFunc ||               // Can't inline external function or indirect
       CalledFunc->isDeclaration()) // call!
     return InlineResult::failure("external or indirect");
 
   // The inliner does not know how to inline through calls with operand bundles
   // in general ...
-  if (CS.hasOperandBundles()) {
-    for (int i = 0, e = CS.getNumOperandBundles(); i != e; ++i) {
-      uint32_t Tag = CS.getOperandBundleAt(i).getTagID();
+  if (CB.hasOperandBundles()) {
+    for (int i = 0, e = CB.getNumOperandBundles(); i != e; ++i) {
+      uint32_t Tag = CB.getOperandBundleAt(i).getTagID();
       // ... but it knows how to inline through "deopt" operand bundles ...
       if (Tag == LLVMContext::OB_deopt)
         continue;
       // ... and "funclet" operand bundles.
       if (Tag == LLVMContext::OB_funclet)
         continue;
 
       return InlineResult::failure("unsupported operand bundle");
     }
   }
 
   // If the call to the callee cannot throw, set the 'nounwind' flag on any
   // calls that we inline.
-  bool MarkNoUnwind = CS.doesNotThrow();
+  bool MarkNoUnwind = CB.doesNotThrow();
 
-  BasicBlock *OrigBB = TheCall->getParent();
+  BasicBlock *OrigBB = CB.getParent();
   Function *Caller = OrigBB->getParent();
 
   // GC poses two hazards to inlining, which only occur when the callee has GC:
   //  1. If the caller has no GC, then the callee's GC must be propagated to the
   //     caller.
   //  2. If the caller has a differing GC, it is invalid to inline.
   if (CalledFunc->hasGC()) {
     if (!Caller->hasGC())
llvm::InlineResult llvm::InlineFunction(CallBase &CB, InlineFunctionInfo &IFI,
 
   // We need to figure out which funclet the callsite was in so that we may
   // properly nest the callee.
   Instruction *CallSiteEHPad = nullptr;
   if (CallerPersonality) {
     EHPersonality Personality = classifyEHPersonality(CallerPersonality);
     if (isScopedEHPersonality(Personality)) {
       Optional<OperandBundleUse> ParentFunclet =
-          CS.getOperandBundle(LLVMContext::OB_funclet);
+          CB.getOperandBundle(LLVMContext::OB_funclet);
       if (ParentFunclet)
         CallSiteEHPad = cast<FuncletPadInst>(ParentFunclet->Inputs.front());
 
       // OK, the inlining site is legal.  What about the target function?
 
       if (CallSiteEHPad) {
         if (Personality == EHPersonality::MSVC_CXX) {
           // The MSVC personality cannot tolerate catches getting inlined into
if (isScopedEHPersonality(Personality)) {
         }
       }
     }
   }
 
   // Determine if we are dealing with a call in an EHPad which does not unwind
   // to caller.
   bool EHPadForCallUnwindsLocally = false;
-  if (CallSiteEHPad && CS.isCall()) {
+  if (CallSiteEHPad && isa<CallInst>(CB)) {
     UnwindDestMemoTy FuncletUnwindMap;
     Value *CallSiteUnwindDestToken =
         getUnwindDestToken(CallSiteEHPad, FuncletUnwindMap);
 
     EHPadForCallUnwindsLocally =
         CallSiteUnwindDestToken &&
         !isa<ConstantTokenNone>(CallSiteUnwindDestToken);
   }
Function::iterator FirstNewBlock;
     ValueToValueMapTy VMap;
     // Keep a list of pair (dst, src) to emit byval initializations.
     SmallVector<std::pair<Value*, Value*>, 4> ByValInit;
 
     auto &DL = Caller->getParent()->getDataLayout();
 
     // Calculate the vector of arguments to pass into the function cloner, which
     // matches up the formal to the actual argument values.
-    CallSite::arg_iterator AI = CS.arg_begin();
+    auto AI = CB.arg_begin();
     unsigned ArgNo = 0;
     for (Function::arg_iterator I = CalledFunc->arg_begin(),
          E = CalledFunc->arg_end(); I != E; ++I, ++AI, ++ArgNo) {
       Value *ActualArg = *AI;
 
       // When byval arguments actually inlined, we need to make the copy implied
       // by them explicit.  However, we don't do this if the callee is readonly
       // or readnone, because the copy would be unneeded: the callee doesn't
       // modify the struct.
-      if (CS.isByValArgument(ArgNo)) {
-        ActualArg = HandleByValArgument(ActualArg, TheCall, CalledFunc, IFI,
+      if (CB.isByValArgument(ArgNo)) {
+        ActualArg = HandleByValArgument(ActualArg, &CB, CalledFunc, IFI,
                                         CalledFunc->getParamAlignment(ArgNo));
         if (ActualArg != *AI)
           ByValInit.push_back(std::make_pair(ActualArg, (Value*) *AI));
       }
 
       VMap[&*I] = ActualArg;
     }
 
     // TODO: Remove this when users have been updated to the assume bundles.
     // Add alignment assumptions if necessary. We do this before the inlined
     // instructions are actually cloned into the caller so that we can easily
     // check what will be known at the start of the inlined code.
-    AddAlignmentAssumptions(CS, IFI);
+    AddAlignmentAssumptions(CB, IFI);
 
     AssumptionCache *AC =
         IFI.GetAssumptionCache ? &(*IFI.GetAssumptionCache)(*Caller) : nullptr;
 
     /// Preserve all attributes on of the call and its parameters.
-    salvageKnowledge(CS.getInstruction(), AC);
+    salvageKnowledge(&CB, AC);
 
     // We want the inliner to prune the code as it copies.  We would LOVE to
     // have no dead or constant instructions leftover after inlining occurs
     // (which can happen, e.g., because an argument was constant), but we'll be
     // happy with whatever the cloner can do.
     CloneAndPruneFunctionInto(Caller, CalledFunc, VMap,
                               /*ModuleLevelChanges=*/false, Returns, ".i",
-                              &InlinedFunctionInfo, TheCall);
+                              &InlinedFunctionInfo, &CB);
     // Remember the first block that is newly cloned over.
     FirstNewBlock = LastBlock; ++FirstNewBlock;
 
     if (IFI.CallerBFI != nullptr && IFI.CalleeBFI != nullptr)
       // Update the BFI of blocks cloned into the caller.
       updateCallerBFI(OrigBB, VMap, IFI.CallerBFI, IFI.CalleeBFI,
                       CalledFunc->front());
 
-    updateCallProfile(CalledFunc, VMap, CalledFunc->getEntryCount(), TheCall,
+    updateCallProfile(CalledFunc, VMap, CalledFunc->getEntryCount(), &CB,
                       IFI.PSI, IFI.CallerBFI);
 
     // Inject byval arguments initialization.
     for (std::pair<Value*, Value*> &Init : ByValInit)
       HandleByValArgumentInit(Init.first, Init.second, Caller->getParent(),
                               &*FirstNewBlock, IFI);
 
     Optional<OperandBundleUse> ParentDeopt =
-        CS.getOperandBundle(LLVMContext::OB_deopt);
+        CB.getOperandBundle(LLVMContext::OB_deopt);
     if (ParentDeopt) {
       SmallVector<OperandBundleDef, 2> OpDefs;
 
       for (auto &VH : InlinedFunctionInfo.OperandBundleCallSites) {
-        Instruction *I = dyn_cast_or_null<Instruction>(VH);
-        if (!I) continue;  // instruction was DCE'd or RAUW'ed to undef
+        CallBase *ICS = dyn_cast_or_null<CallBase>(VH);
+        if (!ICS)
+          continue; // instruction was DCE'd or RAUW'ed to undef
 
         OpDefs.clear();
 
-        CallSite ICS(I);
-        OpDefs.reserve(ICS.getNumOperandBundles());
+        OpDefs.reserve(ICS->getNumOperandBundles());
 
-        for (unsigned i = 0, e = ICS.getNumOperandBundles(); i < e; ++i) {
-          auto ChildOB = ICS.getOperandBundleAt(i);
+        for (unsigned COBi = 0, COBe = ICS->getNumOperandBundles(); COBi < COBe;
+             ++COBi) {
+          auto ChildOB = ICS->getOperandBundleAt(COBi);
           if (ChildOB.getTagID() != LLVMContext::OB_deopt) {
             // If the inlined call has other operand bundles, let them be
             OpDefs.emplace_back(ChildOB);
             continue;
           }
 
           // It may be useful to separate this logic (of handling operand
           // bundles) out to a separate "policy" component if this gets crowded.
           // Prepend the parent's deoptimization continuation to the newly
           // inlined call's deoptimization continuation.
           std::vector<Value *> MergedDeoptArgs;
           MergedDeoptArgs.reserve(ParentDeopt->Inputs.size() +
                                   ChildOB.Inputs.size());
 
           MergedDeoptArgs.insert(MergedDeoptArgs.end(),
                                  ParentDeopt->Inputs.begin(),
                                  ParentDeopt->Inputs.end());
           MergedDeoptArgs.insert(MergedDeoptArgs.end(), ChildOB.Inputs.begin(),
                                  ChildOB.Inputs.end());
 
           OpDefs.emplace_back("deopt", std::move(MergedDeoptArgs));
         }
 
         Instruction *NewI = nullptr;
-        if (isa<CallInst>(I))
-          NewI = CallInst::Create(cast<CallInst>(I), OpDefs, I);
-        else if (isa<CallBrInst>(I))
-          NewI = CallBrInst::Create(cast<CallBrInst>(I), OpDefs, I);
+        if (isa<CallInst>(ICS))
+          NewI = CallInst::Create(cast<CallInst>(ICS), OpDefs, ICS);
+        else if (isa<CallBrInst>(ICS))
+          NewI = CallBrInst::Create(cast<CallBrInst>(ICS), OpDefs, ICS);
         else
-          NewI = InvokeInst::Create(cast<InvokeInst>(I), OpDefs, I);
+          NewI = InvokeInst::Create(cast<InvokeInst>(ICS), OpDefs, ICS);
 
         // Note: the RAUW does the appropriate fixup in VMap, so we need to do
         // this even if the call returns void.
-        I->replaceAllUsesWith(NewI);
+        ICS->replaceAllUsesWith(NewI);
 
         VH = nullptr;
-        I->eraseFromParent();
+        ICS->eraseFromParent();
       }
     }
 
     // Update the callgraph if requested.
     if (IFI.CG)
-      UpdateCallGraphAfterInlining(CS, FirstNewBlock, VMap, IFI);
+      UpdateCallGraphAfterInlining(CB, FirstNewBlock, VMap, IFI);
 
     // For 'nodebug' functions, the associated DISubprogram is always null.
     // Conservatively avoid propagating the callsite debug location to
     // instructions inlined from a function whose DISubprogram is not null.
-    fixupLineNumbers(Caller, FirstNewBlock, TheCall,
+    fixupLineNumbers(Caller, FirstNewBlock, &CB,
                      CalledFunc->getSubprogram() != nullptr);
 
     // Clone existing noalias metadata if necessary.
-    CloneAliasScopeMetadata(CS, VMap);
+    CloneAliasScopeMetadata(CB, VMap);
 
     // Add noalias metadata if necessary.
-    AddAliasScopeMetadata(CS, VMap, DL, CalleeAAR);
+    AddAliasScopeMetadata(CB, VMap, DL, CalleeAAR);
 
     // Clone return attributes on the callsite into the calls within the inlined
     // function which feed into its return value.
-    AddReturnAttributes(CS, VMap);
+    AddReturnAttributes(CB, VMap);
 
     // Propagate llvm.mem.parallel_loop_access if necessary.
-    PropagateParallelLoopAccessMetadata(CS, VMap);
+    PropagateParallelLoopAccessMetadata(CB, VMap);
 
     // Register any cloned assumptions.
     if (IFI.GetAssumptionCache)
       for (BasicBlock &NewBlock :
            make_range(FirstNewBlock->getIterator(), Caller->end()))
         for (Instruction &I : NewBlock) {
           if (auto *II = dyn_cast<IntrinsicInst>(&I))
             if (II->getIntrinsicID() == Intrinsic::assume)
for (BasicBlock::iterator I = FirstNewBlock->begin(),
       Caller->getEntryBlock().getInstList().splice(
           InsertPoint, FirstNewBlock->getInstList(), AI->getIterator(), I);
     }
   }
 
   SmallVector<Value*,4> VarArgsToForward;
   SmallVector<AttributeSet, 4> VarArgsAttrs;
   for (unsigned i = CalledFunc->getFunctionType()->getNumParams();
-       i < CS.getNumArgOperands(); i++) {
-    VarArgsToForward.push_back(CS.getArgOperand(i));
-    VarArgsAttrs.push_back(CS.getAttributes().getParamAttributes(i));
+       i < CB.getNumArgOperands(); i++) {
+    VarArgsToForward.push_back(CB.getArgOperand(i));
+    VarArgsAttrs.push_back(CB.getAttributes().getParamAttributes(i));
   }
 
   bool InlinedMustTailCalls = false, InlinedDeoptimizeCalls = false;
   if (InlinedFunctionInfo.ContainsCalls) {
     CallInst::TailCallKind CallSiteTailKind = CallInst::TCK_None;
-    if (CallInst *CI = dyn_cast<CallInst>(TheCall))
+    if (CallInst *CI = dyn_cast<CallInst>(&CB))
       CallSiteTailKind = CI->getTailCallKind();
 
     // For inlining purposes, the "notail" marker is the same as no marker.
     if (CallSiteTailKind == CallInst::TCK_NoTail)
       CallSiteTailKind = CallInst::TCK_None;
 
     for (Function::iterator BB = FirstNewBlock, E = Caller->end(); BB != E;
          ++BB) {
for (ReturnInst *RI : Returns) {
       IRBuilder<>(RI).CreateCall(StackRestore, SavedPtr);
     }
   }
 
   // If we are inlining for an invoke instruction, we must make sure to rewrite
   // any call instructions into invoke instructions.  This is sensitive to which
   // funclet pads were top-level in the inlinee, so must be done before
   // rewriting the "parent pad" links.
-  if (auto *II = dyn_cast<InvokeInst>(TheCall)) {
+  if (auto *II = dyn_cast<InvokeInst>(&CB)) {
     BasicBlock *UnwindDest = II->getUnwindDest();
     Instruction *FirstNonPHI = UnwindDest->getFirstNonPHI();
     if (isa<LandingPadInst>(FirstNonPHI)) {
       HandleInlinedLandingPad(II, &*FirstNewBlock, InlinedFunctionInfo);
     } else {
       HandleInlinedEHPad(II, &*FirstNewBlock, InlinedFunctionInfo);
     }
   }
 
   // Update the lexical scopes of the new funclets and callsites.
   // Anything that had 'none' as its parent is now nested inside the callsite's
   // EHPad.
 
   if (CallSiteEHPad) {
     for (Function::iterator BB = FirstNewBlock->getIterator(),
                             E = Caller->end();
          BB != E; ++BB) {
       // Add bundle operands to any top-level call sites.
       SmallVector<OperandBundleDef, 1> OpBundles;
       for (BasicBlock::iterator BBI = BB->begin(), E = BB->end(); BBI != E;) {
-        Instruction *I = &*BBI++;
-        CallSite CS(I);
-        if (!CS)
+        CallBase *I = dyn_cast<CallBase>(&*BBI++);
+        if (!I)
           continue;
 
         // Skip call sites which are nounwind intrinsics.
         auto *CalledFn =
-            dyn_cast<Function>(CS.getCalledValue()->stripPointerCasts());
-        if (CalledFn && CalledFn->isIntrinsic() && CS.doesNotThrow())
+            dyn_cast<Function>(I->getCalledValue()->stripPointerCasts());
+        if (CalledFn && CalledFn->isIntrinsic() && I->doesNotThrow())
           continue;
 
         // Skip call sites which already have a "funclet" bundle.
-        if (CS.getOperandBundle(LLVMContext::OB_funclet))
+        if (I->getOperandBundle(LLVMContext::OB_funclet))
           continue;
 
-        CS.getOperandBundlesAsDefs(OpBundles);
+        I->getOperandBundlesAsDefs(OpBundles);
         OpBundles.emplace_back("funclet", CallSiteEHPad);
 
         Instruction *NewInst;
-        if (CS.isCall())
-          NewInst = CallInst::Create(cast<CallInst>(I), OpBundles, I);
-        else if (CS.isCallBr())
-          NewInst = CallBrInst::Create(cast<CallBrInst>(I), OpBundles, I);
+        if (auto *CallI = dyn_cast<CallInst>(I))
+          NewInst = CallInst::Create(CallI, OpBundles, CallI);
+        else if (auto *CallBrI = dyn_cast<CallBrInst>(I))
+          NewInst = CallBrInst::Create(CallBrI, OpBundles, CallBrI);
         else
           NewInst = InvokeInst::Create(cast<InvokeInst>(I), OpBundles, I);
         NewInst->takeName(I);
         I->replaceAllUsesWith(NewInst);
         I->eraseFromParent();
 
         OpBundles.clear();
       }
llvm::InlineResult llvm::InlineFunction(CallBase &CB, InlineFunctionInfo &IFI,
   }
 
   if (InlinedDeoptimizeCalls) {
     // We need to at least remove the deoptimizing returns from the Return set,
     // so that the control flow from those returns does not get merged into the
     // caller (but terminate it instead).  If the caller's return type does not
     // match the callee's return type, we also need to change the return type of
     // the intrinsic.
-    if (Caller->getReturnType() == TheCall->getType()) {
+    if (Caller->getReturnType() == CB.getType()) {
       auto NewEnd = llvm::remove_if(Returns, [](ReturnInst *RI) {
         return RI->getParent()->getTerminatingDeoptimizeCall() != nullptr;
       });
       Returns.erase(NewEnd, Returns.end());
     } else {
       SmallVector<ReturnInst *, 8> NormalReturns;
       Function *NewDeoptIntrinsic = Intrinsic::getDeclaration(
           Caller->getParent(), Intrinsic::experimental_deoptimize,
llvm::InlineResult llvm::InlineFunction(CallBase &CB, InlineFunctionInfo &IFI,
 
   // Handle any inlined musttail call sites.  In order for a new call site to be
   // musttail, the source of the clone and the inlined call site must have been
   // musttail.  Therefore it's safe to return without merging control into the
   // phi below.
   if (InlinedMustTailCalls) {
     // Check if we need to bitcast the result of any musttail calls.
     Type *NewRetTy = Caller->getReturnType();
-    bool NeedBitCast = !TheCall->use_empty() && TheCall->getType() != NewRetTy;
+    bool NeedBitCast = !CB.use_empty() && CB.getType() != NewRetTy;
 
     // Handle the returns preceded by musttail calls separately.
     SmallVector<ReturnInst *, 8> NormalReturns;
     for (ReturnInst *RI : Returns) {
       CallInst *ReturnedMustTail =
           RI->getParent()->getTerminatingMustTailCall();
       if (!ReturnedMustTail) {
         NormalReturns.push_back(RI);
for (BasicBlock &NewBB :
           IFI.InlinedCallSites.push_back(CB);
   }
 
   // If we cloned in _exactly one_ basic block, and if that block ends in a
   // return instruction, we splice the body of the inlined callee directly into
   // the calling basic block.
   if (Returns.size() == 1 && std::distance(FirstNewBlock, Caller->end()) == 1) {
     // Move all of the instructions right before the call.
-    OrigBB->getInstList().splice(TheCall->getIterator(),
-                                 FirstNewBlock->getInstList(),
+    OrigBB->getInstList().splice(CB.getIterator(), FirstNewBlock->getInstList(),
                                  FirstNewBlock->begin(), FirstNewBlock->end());
     // Remove the cloned basic block.
     Caller->getBasicBlockList().pop_back();
 
     // If the call site was an invoke instruction, add a branch to the normal
     // destination.
-    if (InvokeInst *II = dyn_cast<InvokeInst>(TheCall)) {
-      BranchInst *NewBr = BranchInst::Create(II->getNormalDest(), TheCall);
+    if (InvokeInst *II = dyn_cast<InvokeInst>(&CB)) {
+      BranchInst *NewBr = BranchInst::Create(II->getNormalDest(), &CB);
       NewBr->setDebugLoc(Returns[0]->getDebugLoc());
     }
 
     // If the return instruction returned a value, replace uses of the call with
     // uses of the returned value.
-    if (!TheCall->use_empty()) {
+    if (!CB.use_empty()) {
       ReturnInst *R = Returns[0];
-      if (TheCall == R->getReturnValue())
-        TheCall->replaceAllUsesWith(UndefValue::get(TheCall->getType()));
+      if (&CB == R->getReturnValue())
+        CB.replaceAllUsesWith(UndefValue::get(CB.getType()));
       else
-        TheCall->replaceAllUsesWith(R->getReturnValue());
+        CB.replaceAllUsesWith(R->getReturnValue());
     }
     // Since we are now done with the Call/Invoke, we can delete it.
-    TheCall->eraseFromParent();
+    CB.eraseFromParent();
 
     // Since we are now done with the return instruction, delete it also.
     Returns[0]->eraseFromParent();
 
     // We are now done with the inlining.
     return InlineResult::success();
   }
 
   // Otherwise, we have the normal case, of more than one block to inline or
   // multiple return sites.
 
   // We want to clone the entire callee function into the hole between the
   // "starter" and "ender" blocks.  How we accomplish this depends on whether
   // this is an invoke instruction or a call instruction.
   BasicBlock *AfterCallBB;
   BranchInst *CreatedBranchToNormalDest = nullptr;
-  if (InvokeInst *II = dyn_cast<InvokeInst>(TheCall)) {
+  if (InvokeInst *II = dyn_cast<InvokeInst>(&CB)) {
 
     // Add an unconditional branch to make this look like the CallInst case...
-    CreatedBranchToNormalDest = BranchInst::Create(II->getNormalDest(), TheCall);
+    CreatedBranchToNormalDest = BranchInst::Create(II->getNormalDest(), &CB);
 
     // Split the basic block.  This guarantees that no PHI nodes will have to be
     // updated due to new incoming edges, and make the invoke case more
     // symmetric to the call case.
     AfterCallBB =
         OrigBB->splitBasicBlock(CreatedBranchToNormalDest->getIterator(),
                                 CalledFunc->getName() + ".exit");
 
-  } else {  // It's a call
+  } else { // It's a call
     // If this is a call instruction, we need to split the basic block that
     // the call lives in.
     //
-    AfterCallBB = OrigBB->splitBasicBlock(TheCall->getIterator(),
+    AfterCallBB = OrigBB->splitBasicBlock(CB.getIterator(),
                                           CalledFunc->getName() + ".exit");
   }
 
   if (IFI.CallerBFI) {
     // Copy original BB's block frequency to AfterCallBB
     IFI.CallerBFI->setBlockFreq(
         AfterCallBB, IFI.CallerBFI->getBlockFreq(OrigBB).getFrequency());
   }
llvm::InlineResult llvm::InlineFunction(CallBase &CB, InlineFunctionInfo &IFI,
   // Handle all of the return instructions that we just cloned in, and eliminate
   // any users of the original call/invoke instruction.
   Type *RTy = CalledFunc->getReturnType();
 
   PHINode *PHI = nullptr;
   if (Returns.size() > 1) {
     // The PHI node should go at the front of the new basic block to merge all
     // possible incoming values.
-    if (!TheCall->use_empty()) {
-      PHI = PHINode::Create(RTy, Returns.size(), TheCall->getName(),
+    if (!CB.use_empty()) {
+      PHI = PHINode::Create(RTy, Returns.size(), CB.getName(),
                             &AfterCallBB->front());
       // Anything that used the result of the function call should now use the
       // PHI node as their operand.
-      TheCall->replaceAllUsesWith(PHI);
+      CB.replaceAllUsesWith(PHI);
     }
 
     // Loop over all of the return instructions adding entries to the PHI node
     // as appropriate.
     if (PHI) {
       for (unsigned i = 0, e = Returns.size(); i != e; ++i) {
         ReturnInst *RI = Returns[i];
         assert(RI->getReturnValue()->getType() == PHI->getType() &&
if (Returns.size() > 1) {
     // inlined function. The line number may be nonsensical, but the
     // instruction will at least be associated with the right
     // function.
     if (CreatedBranchToNormalDest)
       CreatedBranchToNormalDest->setDebugLoc(Loc);
   } else if (!Returns.empty()) {
     // Otherwise, if there is exactly one return value, just replace anything
     // using the return value of the call with the computed value.
-    if (!TheCall->use_empty()) {
-      if (TheCall == Returns[0]->getReturnValue())
-        TheCall->replaceAllUsesWith(UndefValue::get(TheCall->getType()));
+    if (!CB.use_empty()) {
+      if (&CB == Returns[0]->getReturnValue())
+        CB.replaceAllUsesWith(UndefValue::get(CB.getType()));
       else
-        TheCall->replaceAllUsesWith(Returns[0]->getReturnValue());
+        CB.replaceAllUsesWith(Returns[0]->getReturnValue());
     }
 
     // Update PHI nodes that use the ReturnBB to use the AfterCallBB.
     BasicBlock *ReturnBB = Returns[0]->getParent();
     ReturnBB->replaceAllUsesWith(AfterCallBB);
 
     // Splice the code from the return block into the block that it will return
     // to, which contains the code that was after the call.
     AfterCallBB->getInstList().splice(AfterCallBB->begin(),
                                       ReturnBB->getInstList());
 
     if (CreatedBranchToNormalDest)
       CreatedBranchToNormalDest->setDebugLoc(Returns[0]->getDebugLoc());
 
     // Delete the return instruction now and empty ReturnBB now.
     Returns[0]->eraseFromParent();
     ReturnBB->eraseFromParent();
-  } else if (!TheCall->use_empty()) {
+  } else if (!CB.use_empty()) {
     // No returns, but something is using the return value of the call.  Just
     // nuke the result.
-    TheCall->replaceAllUsesWith(UndefValue::get(TheCall->getType()));
+    CB.replaceAllUsesWith(UndefValue::get(CB.getType()));
   }
 
   // Since we are now done with the Call/Invoke, we can delete it.
-  TheCall->eraseFromParent();
+  CB.eraseFromParent();
 
   // If we inlined any musttail calls and the original return is now
   // unreachable, delete it.  It can only contain a bitcast and ret.
   if (InlinedMustTailCalls && pred_begin(AfterCallBB) == pred_end(AfterCallBB))
     AfterCallBB->eraseFromParent();
 
   // We should always be able to fold the entry block of the function into the
   // single predecessor of the block...