[cfe-commits] r49144 - in /cfe/trunk: lib/CodeGen/CodeGenTypes.cpp lib/CodeGen/CodeGenTypes.h test/CodeGen/types.c
Chris Lattner
sabre at nondot.org
Wed Apr 2 22:50:42 PDT 2008
Author: lattner
Date: Thu Apr 3 00:50:42 2008
New Revision: 49144
URL: http://llvm.org/viewvc/llvm-project?rev=49144&view=rev
Log:
Fix PR2081 (problems codegen'ing some recursive structures) patch
by Lauro Venancio!
Modified:
cfe/trunk/lib/CodeGen/CodeGenTypes.cpp
cfe/trunk/lib/CodeGen/CodeGenTypes.h
cfe/trunk/test/CodeGen/types.c
Modified: cfe/trunk/lib/CodeGen/CodeGenTypes.cpp
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CodeGenTypes.cpp?rev=49144&r1=49143&r2=49144&view=diff
==============================================================================
--- cfe/trunk/lib/CodeGen/CodeGenTypes.cpp (original)
+++ cfe/trunk/lib/CodeGen/CodeGenTypes.cpp Thu Apr 3 00:50:42 2008
@@ -92,6 +92,27 @@
/// ConvertType - Convert the specified type to its LLVM form.
const llvm::Type *CodeGenTypes::ConvertType(QualType T) {
+ llvm::PATypeHolder Result = ConvertTypeRecursive(T);
+
+ // Any pointers that were converted defered evaluation of their pointee type,
+ // creating an opaque type instead. This is in order to avoid problems with
+ // circular types. Loop through all these defered pointees, if any, and
+ // resolve them now.
+ while (!PointersToResolve.empty()) {
+ std::pair<const PointerLikeType *, llvm::OpaqueType*> P =
+ PointersToResolve.back();
+ PointersToResolve.pop_back();
+ // We can handle bare pointers here because we know that the only pointers
+ // to the Opaque type are P.second and from other types. Refining the
+ // opqaue type away will invalidate P.second, but we don't mind :).
+ const llvm::Type *NT = ConvertTypeRecursive(P.first->getPointeeType());
+ P.second->refineAbstractTypeTo(NT);
+ }
+
+ return Result;
+}
+
+const llvm::Type *CodeGenTypes::ConvertTypeRecursive(QualType T) {
// See if type is already cached.
llvm::DenseMap<Type *, llvm::PATypeHolder>::iterator
I = TypeCache.find(T.getCanonicalType().getTypePtr());
@@ -203,13 +224,16 @@
}
case Type::Complex: {
const llvm::Type *EltTy =
- ConvertType(cast<ComplexType>(Ty).getElementType());
+ ConvertTypeRecursive(cast<ComplexType>(Ty).getElementType());
return llvm::StructType::get(EltTy, EltTy, NULL);
}
case Type::Reference:
case Type::Pointer: {
- QualType ETy = cast<PointerLikeType>(Ty).getPointeeType();
- return llvm::PointerType::get(ConvertType(ETy), ETy.getAddressSpace());
+ const PointerLikeType &PTy = cast<PointerLikeType>(Ty);
+ QualType ETy = PTy.getPointeeType();
+ llvm::OpaqueType *PointeeType = llvm::OpaqueType::get();
+ PointersToResolve.push_back(std::make_pair(&PTy, PointeeType));
+ return llvm::PointerType::get(PointeeType, ETy.getAddressSpace());
}
case Type::VariableArray: {
@@ -218,24 +242,24 @@
"FIXME: We only handle trivial array types so far!");
// VLAs resolve to the innermost element type; this matches
// the return of alloca, and there isn't any obviously better choice.
- return ConvertType(A.getElementType());
+ return ConvertTypeRecursive(A.getElementType());
}
case Type::IncompleteArray: {
const IncompleteArrayType &A = cast<IncompleteArrayType>(Ty);
assert(A.getIndexTypeQualifier() == 0 &&
"FIXME: We only handle trivial array types so far!");
// int X[] -> [0 x int]
- return llvm::ArrayType::get(ConvertType(A.getElementType()), 0);
+ return llvm::ArrayType::get(ConvertTypeRecursive(A.getElementType()), 0);
}
case Type::ConstantArray: {
const ConstantArrayType &A = cast<ConstantArrayType>(Ty);
- const llvm::Type *EltTy = ConvertType(A.getElementType());
+ const llvm::Type *EltTy = ConvertTypeRecursive(A.getElementType());
return llvm::ArrayType::get(EltTy, A.getSize().getZExtValue());
}
case Type::OCUVector:
case Type::Vector: {
const VectorType &VT = cast<VectorType>(Ty);
- return llvm::VectorType::get(ConvertType(VT.getElementType()),
+ return llvm::VectorType::get(ConvertTypeRecursive(VT.getElementType()),
VT.getNumElements());
}
case Type::FunctionNoProto:
@@ -246,7 +270,7 @@
if (FP.getResultType()->isVoidType())
ResultType = llvm::Type::VoidTy; // Result of function uses llvm void.
else
- ResultType = ConvertType(FP.getResultType());
+ ResultType = ConvertTypeRecursive(FP.getResultType());
// FIXME: Convert argument types.
bool isVarArg;
@@ -270,7 +294,8 @@
}
case Type::ASQual:
- return ConvertType(QualType(cast<ASQualType>(Ty).getBaseType(), 0));
+ return
+ ConvertTypeRecursive(QualType(cast<ASQualType>(Ty).getBaseType(), 0));
case Type::ObjCInterface: {
// Warning: Use of this is strongly discouraged. Late binding of instance
@@ -324,7 +349,7 @@
void CodeGenTypes::DecodeArgumentTypes(const FunctionTypeProto &FTP,
std::vector<const llvm::Type*> &ArgTys) {
for (unsigned i = 0, e = FTP.getNumArgs(); i != e; ++i) {
- const llvm::Type *Ty = ConvertType(FTP.getArgType(i));
+ const llvm::Type *Ty = ConvertTypeRecursive(FTP.getArgType(i));
if (Ty->isFirstClassType())
ArgTys.push_back(Ty);
else
@@ -355,7 +380,7 @@
if (TD->getKind() == Decl::Enum) {
// Don't bother storing enums in TagDeclTypes.
- return ConvertType(cast<EnumDecl>(TD)->getIntegerType());
+ return ConvertTypeRecursive(cast<EnumDecl>(TD)->getIntegerType());
}
// This decl could well be recursive. In this case, insert an opaque
@@ -484,7 +509,7 @@
if (FD->isBitField())
placeBitField(FD);
else {
- const llvm::Type *Ty = CGT.ConvertType(FD->getType());
+ const llvm::Type *Ty = CGT.ConvertTypeRecursive(FD->getType());
addLLVMField(Ty);
CGT.addFieldInfo(FD, llvmFieldNo - 1);
Cursor = llvmSize;
@@ -562,7 +587,8 @@
}
std::vector<const llvm::Type*> Fields;
- const llvm::Type *Ty = CGT.ConvertType(FieldDecls[PrimaryEltNo]->getType());
+ const llvm::Type *Ty =
+ CGT.ConvertTypeRecursive(FieldDecls[PrimaryEltNo]->getType());
Fields.push_back(Ty);
STy = llvm::StructType::get(Fields);
}
@@ -580,7 +606,7 @@
assert (isBitField && "Invalid BitField size expression");
uint64_t BitFieldSize = FieldSize.getZExtValue();
- const llvm::Type *Ty = CGT.ConvertType(FD->getType());
+ const llvm::Type *Ty = CGT.ConvertTypeRecursive(FD->getType());
uint64_t TySize = CGT.getTargetData().getABITypeSizeInBits(Ty);
unsigned Idx = Cursor / TySize;
Modified: cfe/trunk/lib/CodeGen/CodeGenTypes.h
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CodeGenTypes.h?rev=49144&r1=49143&r2=49144&view=diff
==============================================================================
--- cfe/trunk/lib/CodeGen/CodeGenTypes.h (original)
+++ cfe/trunk/lib/CodeGen/CodeGenTypes.h Thu Apr 3 00:50:42 2008
@@ -21,6 +21,7 @@
namespace llvm {
class Module;
class Type;
+ class OpaqueType;
class PATypeHolder;
class TargetData;
}
@@ -30,6 +31,8 @@
class TagDecl;
class TargetInfo;
class QualType;
+ class PointerType;
+ class PointerLikeType;
class Type;
class FunctionTypeProto;
class FieldDecl;
@@ -77,6 +80,10 @@
llvm::Module& TheModule;
const llvm::TargetData& TheTargetData;
+
+ llvm::SmallVector<std::pair<const PointerLikeType *,
+ llvm::OpaqueType *>, 8> PointersToResolve;
+
llvm::DenseMap<const TagDecl*, llvm::PATypeHolder> TagDeclTypes;
/// CGRecordLayouts - This maps llvm struct type with corresponding
@@ -124,6 +131,7 @@
/// ConvertType - Convert type T into a llvm::Type.
const llvm::Type *ConvertType(QualType T);
+ const llvm::Type *ConvertTypeRecursive(QualType T);
/// ConvertTypeForMem - Convert type T into a llvm::Type. This differs from
/// ConvertType in that it is used to convert to the memory representation for
Modified: cfe/trunk/test/CodeGen/types.c
URL: http://llvm.org/viewvc/llvm-project/cfe/trunk/test/CodeGen/types.c?rev=49144&r1=49143&r2=49144&view=diff
==============================================================================
--- cfe/trunk/test/CodeGen/types.c (original)
+++ cfe/trunk/test/CodeGen/types.c Thu Apr 3 00:50:42 2008
@@ -19,3 +19,16 @@
typedef struct MpegEncContext {int pb;} MpegEncContext;
static void test2(void) {MpegEncContext s; s.pb;}
+
+struct Village;
+
+struct List {
+ struct Village *v;
+};
+
+struct Village {
+ struct List returned;
+};
+
+void test3(struct List a) {
+}
More information about the cfe-commits
mailing list