[CIR] Initialize vptr in dynamic classes (#152574)

This adds support for initializing the vptr member of a dynamic class in
the constructor of that class.

This does not include support for lowering the
`cir.vtable.address_point` operation to the LLVM dialect. That handling
will be added in a follow-up patch.

Author: andykaylor

clang/include/clang/CIR/MissingFeatures.h

@@ -192,6 +192,7 @@ struct MissingFeatures {
   static bool constantFoldSwitchStatement() { return false; }
   static bool constructABIArgDirectExtend() { return false; }
   static bool coverageMapping() { return false; }
+  static bool createInvariantGroup() { return false; }
   static bool createProfileWeightsForLoop() { return false; }
   static bool ctorMemcpyizer() { return false; }
   static bool cudaSupport() { return false; }
@@ -263,6 +264,7 @@ struct MissingFeatures {
   static bool appleKext() { return false; }
   static bool dtorCleanups() { return false; }
   static bool vtableInitialization() { return false; }
+  static bool vtableRelativeLayout() { return false; }
   static bool msvcBuiltins() { return false; }
   static bool vlas() { return false; }
 

clang/lib/CIR/CodeGen/CIRGenBuilder.h

@@ -244,6 +244,17 @@ class CIRGenBuilderTy : public cir::CIRBaseBuilderTy {
   }
   bool isInt(mlir::Type i) { return mlir::isa<cir::IntType>(i); }
 
+  // Fetch the type representing a pointer to unsigned int8 values.
+  cir::PointerType getUInt8PtrTy() { return typeCache.UInt8PtrTy; }
+
+  /// Get a CIR anonymous record type.
+  cir::RecordType getAnonRecordTy(llvm::ArrayRef<mlir::Type> members,
+                                  bool packed = false, bool padded = false) {
+    assert(!cir::MissingFeatures::astRecordDeclAttr());
+    auto kind = cir::RecordType::RecordKind::Struct;
+    return getType<cir::RecordType>(members, packed, padded, kind);
+  }
+
   //
   // Constant creation helpers
   // -------------------------

clang/lib/CIR/CodeGen/CIRGenCXXABI.h

@@ -80,6 +80,11 @@ class CIRGenCXXABI {
                                   bool forVirtualBase, bool delegating,
                                   Address thisAddr, QualType thisTy) = 0;
 
+  /// Checks if ABI requires extra virtual offset for vtable field.
+  virtual bool
+  isVirtualOffsetNeededForVTableField(CIRGenFunction &cgf,
+                                      CIRGenFunction::VPtr vptr) = 0;
+
   /// Returns true if the given destructor type should be emitted as a linkonce
   /// delegating thunk, regardless of whether the dtor is defined in this TU or
   /// not.
@@ -90,6 +95,26 @@ class CIRGenCXXABI {
   getCXXDestructorLinkage(GVALinkage linkage, const CXXDestructorDecl *dtor,
                           CXXDtorType dt) const;
 
+  /// Get the address of the vtable for the given record decl which should be
+  /// used for the vptr at the given offset in RD.
+  virtual cir::GlobalOp getAddrOfVTable(const CXXRecordDecl *rd,
+                                        CharUnits vptrOffset) = 0;
+
+  /// Get the address point of the vtable for the given base subobject.
+  virtual mlir::Value
+  getVTableAddressPoint(BaseSubobject base,
+                        const CXXRecordDecl *vtableClass) = 0;
+
+  /// Get the address point of the vtable for the given base subobject while
+  /// building a constructor or a destructor.
+  virtual mlir::Value getVTableAddressPointInStructor(
+      CIRGenFunction &cgf, const CXXRecordDecl *vtableClass, BaseSubobject base,
+      const CXXRecordDecl *nearestVBase) = 0;
+
+  /// Checks if ABI requires to initialize vptrs for given dynamic class.
+  virtual bool
+  doStructorsInitializeVPtrs(const clang::CXXRecordDecl *vtableClass) = 0;
+
   /// Returns true if the given constructor or destructor is one of the kinds
   /// that the ABI says returns 'this' (only applies when called non-virtually
   /// for destructors).

clang/lib/CIR/CodeGen/CIRGenClass.cpp

@@ -199,10 +199,6 @@ void CIRGenFunction::emitCtorPrologue(const CXXConstructorDecl *cd,
     return;
   }
 
-  // If there are no member initializers, we can just return.
-  if (cd->getNumCtorInitializers() == 0)
-    return;
-
   const CXXRecordDecl *classDecl = cd->getParent();
 
   // This code doesn't use range-based iteration because we may need to emit
@@ -227,7 +223,8 @@ void CIRGenFunction::emitCtorPrologue(const CXXConstructorDecl *cd,
   }
 
   const mlir::Value oldThisValue = cxxThisValue;
-  if (!constructVBases && (*b)->isBaseInitializer() && (*b)->isBaseVirtual()) {
+  if (!constructVBases && b != e && (*b)->isBaseInitializer() &&
+      (*b)->isBaseVirtual()) {
     cgm.errorNYI(cd->getSourceRange(),
                  "emitCtorPrologue: virtual base initializer");
     return;
@@ -249,11 +246,7 @@ void CIRGenFunction::emitCtorPrologue(const CXXConstructorDecl *cd,
 
   cxxThisValue = oldThisValue;
 
-  if (classDecl->isDynamicClass()) {
-    cgm.errorNYI(cd->getSourceRange(),
-                 "emitCtorPrologue: initialize vtable pointers");
-    return;
-  }
+  initializeVTablePointers(getLoc(cd->getBeginLoc()), classDecl);
 
   // Finally, initialize class members.
   FieldConstructionScope fcs(*this, loadCXXThisAddress());
@@ -271,6 +264,95 @@ void CIRGenFunction::emitCtorPrologue(const CXXConstructorDecl *cd,
   }
 }
 
+void CIRGenFunction::initializeVTablePointer(mlir::Location loc,
+                                             const VPtr &vptr) {
+  // Compute the address point.
+  mlir::Value vtableAddressPoint =
+      cgm.getCXXABI().getVTableAddressPointInStructor(
+          *this, vptr.vtableClass, vptr.base, vptr.nearestVBase);
+
+  if (!vtableAddressPoint)
+    return;
+
+  // Compute where to store the address point.
+  mlir::Value virtualOffset{};
+  CharUnits nonVirtualOffset = CharUnits::Zero();
+
+  mlir::Type baseValueTy;
+  if (cgm.getCXXABI().isVirtualOffsetNeededForVTableField(*this, vptr)) {
+    cgm.errorNYI(loc, "initializeVTablePointer: virtual offset for vtable");
+  } else {
+    // We can just use the base offset in the complete class.
+    nonVirtualOffset = vptr.base.getBaseOffset();
+    baseValueTy = convertType(getContext().getTagDeclType(vptr.base.getBase()));
+  }
+
+  // Apply the offsets.
+  Address vtableField = loadCXXThisAddress();
+  if (!nonVirtualOffset.isZero() || virtualOffset) {
+    cgm.errorNYI(loc,
+                 "initializeVTablePointer: non-virtual and virtual offset");
+  }
+
+  // Finally, store the address point. Use the same CIR types as the field.
+  //
+  // vtable field is derived from `this` pointer, therefore they should be in
+  // the same addr space.
+  assert(!cir::MissingFeatures::addressSpace());
+  // TODO(cir): This should be cir.vtable.get_vptr.
+  vtableField = builder.createElementBitCast(loc, vtableField,
+                                             vtableAddressPoint.getType());
+  builder.createStore(loc, vtableAddressPoint, vtableField);
+  assert(!cir::MissingFeatures::opTBAA());
+  assert(!cir::MissingFeatures::createInvariantGroup());
+}
+
+void CIRGenFunction::initializeVTablePointers(mlir::Location loc,
+                                              const CXXRecordDecl *rd) {
+  // Ignore classes without a vtable.
+  if (!rd->isDynamicClass())
+    return;
+
+  // Initialize the vtable pointers for this class and all of its bases.
+  if (cgm.getCXXABI().doStructorsInitializeVPtrs(rd))
+    for (const auto &vptr : getVTablePointers(rd))
+      initializeVTablePointer(loc, vptr);
+
+  if (rd->getNumVBases())
+    cgm.errorNYI(loc, "initializeVTablePointers: virtual base");
+}
+
+CIRGenFunction::VPtrsVector
+CIRGenFunction::getVTablePointers(const CXXRecordDecl *vtableClass) {
+  CIRGenFunction::VPtrsVector vptrsResult;
+  getVTablePointers(BaseSubobject(vtableClass, CharUnits::Zero()),
+                    /*NearestVBase=*/nullptr,
+                    /*OffsetFromNearestVBase=*/CharUnits::Zero(),
+                    /*BaseIsNonVirtualPrimaryBase=*/false, vtableClass,
+                    vptrsResult);
+  return vptrsResult;
+}
+
+void CIRGenFunction::getVTablePointers(BaseSubobject base,
+                                       const CXXRecordDecl *nearestVBase,
+                                       CharUnits offsetFromNearestVBase,
+                                       bool baseIsNonVirtualPrimaryBase,
+                                       const CXXRecordDecl *vtableClass,
+                                       VPtrsVector &vptrs) {
+  // If this base is a non-virtual primary base the address point has already
+  // been set.
+  if (!baseIsNonVirtualPrimaryBase) {
+    // Initialize the vtable pointer for this base.
+    VPtr vptr = {base, nearestVBase, offsetFromNearestVBase, vtableClass};
+    vptrs.push_back(vptr);
+  }
+
+  const CXXRecordDecl *rd = base.getBase();
+
+  if (rd->getNumBases())
+    cgm.errorNYI(rd->getSourceRange(), "getVTablePointers: traverse bases");
+}
+
 Address CIRGenFunction::loadCXXThisAddress() {
   assert(curFuncDecl && "loading 'this' without a func declaration?");
   assert(isa<CXXMethodDecl>(curFuncDecl));

clang/lib/CIR/CodeGen/CIRGenFunction.h

@@ -23,6 +23,7 @@
 #include "Address.h"
 
 #include "clang/AST/ASTContext.h"
+#include "clang/AST/BaseSubobject.h"
 #include "clang/AST/CharUnits.h"
 #include "clang/AST/Decl.h"
 #include "clang/AST/Stmt.h"
@@ -509,6 +510,25 @@ class CIRGenFunction : public CIRGenTypeCache {
   static bool
   isConstructorDelegationValid(const clang::CXXConstructorDecl *ctor);
 
+  struct VPtr {
+    clang::BaseSubobject base;
+    const clang::CXXRecordDecl *nearestVBase;
+    clang::CharUnits offsetFromNearestVBase;
+    const clang::CXXRecordDecl *vtableClass;
+  };
+
+  using VPtrsVector = llvm::SmallVector<VPtr, 4>;
+  VPtrsVector getVTablePointers(const clang::CXXRecordDecl *vtableClass);
+  void getVTablePointers(clang::BaseSubobject base,
+                         const clang::CXXRecordDecl *nearestVBase,
+                         clang::CharUnits offsetFromNearestVBase,
+                         bool baseIsNonVirtualPrimaryBase,
+                         const clang::CXXRecordDecl *vtableClass,
+                         VPtrsVector &vptrs);
+  /// Return the Value of the vtable pointer member pointed to by thisAddr.
+  mlir::Value getVTablePtr(mlir::Location loc, Address thisAddr,
+                           const clang::CXXRecordDecl *vtableClass);
+
   /// A scope within which we are constructing the fields of an object which
   /// might use a CXXDefaultInitExpr. This stashes away a 'this' value to use if
   /// we need to evaluate the CXXDefaultInitExpr within the evaluation.
@@ -557,6 +577,10 @@ class CIRGenFunction : public CIRGenTypeCache {
     return LValue::makeAddr(addr, ty, baseInfo);
   }
 
+  void initializeVTablePointers(mlir::Location loc,
+                                const clang::CXXRecordDecl *rd);
+  void initializeVTablePointer(mlir::Location loc, const VPtr &vptr);
+
   /// Return the address of a local variable.
   Address getAddrOfLocalVar(const clang::VarDecl *vd) {
     auto it = localDeclMap.find(vd);

clang/lib/CIR/CodeGen/CIRGenItaniumCXXABI.cpp

@@ -22,6 +22,7 @@
 
 #include "clang/AST/ExprCXX.h"
 #include "clang/AST/GlobalDecl.h"
+#include "clang/AST/VTableBuilder.h"
 #include "clang/CIR/MissingFeatures.h"
 #include "llvm/Support/ErrorHandling.h"
 
@@ -31,6 +32,10 @@ using namespace clang::CIRGen;
 namespace {
 
 class CIRGenItaniumCXXABI : public CIRGenCXXABI {
+protected:
+  /// All the vtables which have been defined.
+  llvm::DenseMap<const CXXRecordDecl *, cir::GlobalOp> vtables;
+
 public:
   CIRGenItaniumCXXABI(CIRGenModule &cgm) : CIRGenCXXABI(cgm) {
     assert(!cir::MissingFeatures::cxxabiUseARMMethodPtrABI());
@@ -58,6 +63,24 @@ class CIRGenItaniumCXXABI : public CIRGenCXXABI {
     // emitted with external linkage or as linkonce if they are inline and used.
     return false;
   }
+
+  bool isVirtualOffsetNeededForVTableField(CIRGenFunction &cgf,
+                                           CIRGenFunction::VPtr vptr) override;
+
+  cir::GlobalOp getAddrOfVTable(const CXXRecordDecl *rd,
+                                CharUnits vptrOffset) override;
+
+  mlir::Value getVTableAddressPoint(BaseSubobject base,
+                                    const CXXRecordDecl *vtableClass) override;
+
+  mlir::Value getVTableAddressPointInStructor(
+      CIRGenFunction &cgf, const clang::CXXRecordDecl *vtableClass,
+      clang::BaseSubobject base,
+      const clang::CXXRecordDecl *nearestVBase) override;
+
+  bool doStructorsInitializeVPtrs(const CXXRecordDecl *vtableClass) override {
+    return true;
+  }
 };
 
 } // namespace
@@ -278,3 +301,91 @@ CIRGenCXXABI *clang::CIRGen::CreateCIRGenItaniumCXXABI(CIRGenModule &cgm) {
     llvm_unreachable("bad or NYI ABI kind");
   }
 }
+
+cir::GlobalOp CIRGenItaniumCXXABI::getAddrOfVTable(const CXXRecordDecl *rd,
+                                                   CharUnits vptrOffset) {
+  assert(vptrOffset.isZero() && "Itanium ABI only supports zero vptr offsets");
+  cir::GlobalOp &vtable = vtables[rd];
+  if (vtable)
+    return vtable;
+
+  // Queue up this vtable for possible deferred emission.
+  assert(!cir::MissingFeatures::deferredVtables());
+
+  SmallString<256> name;
+  llvm::raw_svector_ostream out(name);
+  getMangleContext().mangleCXXVTable(rd, out);
+
+  const VTableLayout &vtLayout =
+      cgm.getItaniumVTableContext().getVTableLayout(rd);
+  mlir::Type vtableType = cgm.getVTables().getVTableType(vtLayout);
+
+  // Use pointer alignment for the vtable. Otherwise we would align them based
+  // on the size of the initializer which doesn't make sense as only single
+  // values are read.
+  unsigned ptrAlign = cgm.getItaniumVTableContext().isRelativeLayout()
+                          ? 32
+                          : cgm.getTarget().getPointerAlign(LangAS::Default);
+
+  vtable = cgm.createOrReplaceCXXRuntimeVariable(
+      cgm.getLoc(rd->getSourceRange()), name, vtableType,
+      cir::GlobalLinkageKind::ExternalLinkage,
+      cgm.getASTContext().toCharUnitsFromBits(ptrAlign));
+  // LLVM codegen handles unnamedAddr
+  assert(!cir::MissingFeatures::opGlobalUnnamedAddr());
+
+  // In MS C++ if you have a class with virtual functions in which you are using
+  // selective member import/export, then all virtual functions must be exported
+  // unless they are inline, otherwise a link error will result. To match this
+  // behavior, for such classes, we dllimport the vtable if it is defined
+  // externally and all the non-inline virtual methods are marked dllimport, and
+  // we dllexport the vtable if it is defined in this TU and all the non-inline
+  // virtual methods are marked dllexport.
+  if (cgm.getTarget().hasPS4DLLImportExport())
+    cgm.errorNYI(rd->getSourceRange(),
+                 "getAddrOfVTable: PS4 DLL import/export");
+
+  cgm.setGVProperties(vtable, rd);
+  return vtable;
+}
+
+mlir::Value
+CIRGenItaniumCXXABI::getVTableAddressPoint(BaseSubobject base,
+                                           const CXXRecordDecl *vtableClass) {
+  cir::GlobalOp vtable = getAddrOfVTable(vtableClass, CharUnits());
+
+  // Find the appropriate vtable within the vtable group, and the address point
+  // within that vtable.
+  VTableLayout::AddressPointLocation addressPoint =
+      cgm.getItaniumVTableContext()
+          .getVTableLayout(vtableClass)
+          .getAddressPoint(base);
+
+  mlir::OpBuilder &builder = cgm.getBuilder();
+  auto vtablePtrTy = cir::VPtrType::get(builder.getContext());
+
+  return builder.create<cir::VTableAddrPointOp>(
+      cgm.getLoc(vtableClass->getSourceRange()), vtablePtrTy,
+      mlir::FlatSymbolRefAttr::get(vtable.getSymNameAttr()),
+      cir::AddressPointAttr::get(cgm.getBuilder().getContext(),
+                                 addressPoint.VTableIndex,
+                                 addressPoint.AddressPointIndex));
+}
+
+mlir::Value CIRGenItaniumCXXABI::getVTableAddressPointInStructor(
+    CIRGenFunction &cgf, const clang::CXXRecordDecl *vtableClass,
+    clang::BaseSubobject base, const clang::CXXRecordDecl *nearestVBase) {
+
+  if (base.getBase()->getNumVBases()) {
+    cgm.errorNYI(cgf.curFuncDecl->getLocation(),
+                 "getVTableAddressPointInStructor: virtual base");
+  }
+  return getVTableAddressPoint(base, vtableClass);
+}
+
+bool CIRGenItaniumCXXABI::isVirtualOffsetNeededForVTableField(
+    CIRGenFunction &cgf, CIRGenFunction::VPtr vptr) {
+  if (vptr.nearestVBase == nullptr)
+    return false;
+  return needsVTTParameter(cgf.curGD);
+}