r315439 - Revert r314955: "Remove PendingBody mechanism for function and ObjC method deserialization."
Author: djasper
Date: Wed Oct 11 00:47:54 2017
New Revision: 315439
URL: http://llvm.org/viewvc/llvm-project?rev=315439&view=rev
Log:
Revert r314955: "Remove PendingBody mechanism for function and ObjC method
deserialization."
This is breaking a build of https://github.com/abseil/abseil-cpp and so
likely not really NFC. Also reverted subsequent r314956/7.
I'll forward reproduction instructions to Richard.
Removed:
cfe/trunk/test/Modules/merge-lambdas.cpp
Modified:
cfe/trunk/include/clang/Serialization/ASTReader.h
cfe/trunk/lib/Serialization/ASTCommon.cpp
cfe/trunk/lib/Serialization/ASTReader.cpp
cfe/trunk/lib/Serialization/ASTReaderDecl.cpp
Modified: cfe/trunk/include/clang/Serialization/ASTReader.h
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Serialization/ASTReader.h?rev=315439&r1=315438&r2=315439&view=diff
==
--- cfe/trunk/include/clang/Serialization/ASTReader.h (original)
+++ cfe/trunk/include/clang/Serialization/ASTReader.h Wed Oct 11 00:47:54 2017
@@ -559,9 +559,13 @@ private:
/// declarations that have not yet been linked to their definitions.
llvm::SmallPtrSet PendingDefinitions;
- /// \brief Functions or methods that are known to already have a definition
- /// (that might not yet be merged into the redeclaration chain).
- llvm::SmallDenseMap FunctionDefinitions;
+ typedef llvm::MapVector,
+ SmallVector, 4> >
+PendingBodiesMap;
+
+ /// \brief Functions or methods that have bodies that will be attached.
+ PendingBodiesMap PendingBodies;
/// \brief Definitions for which we have added merged definitions but not yet
/// performed deduplication.
@@ -987,13 +991,25 @@ private:
/// the last time we loaded information about this identifier.
llvm::DenseMap IdentifierGeneration;
+ class InterestingDecl {
+Decl *D;
+bool DeclHasPendingBody;
+
+ public:
+InterestingDecl(Decl *D, bool HasBody)
+: D(D), DeclHasPendingBody(HasBody) {}
+Decl *getDecl() { return D; }
+/// Whether the declaration has a pending body.
+bool hasPendingBody() { return DeclHasPendingBody; }
+ };
+
/// \brief Contains declarations and definitions that could be
/// "interesting" to the ASTConsumer, when we get that AST consumer.
///
/// "Interesting" declarations are those that have data that may
/// need to be emitted, such as inline function definitions or
/// Objective-C protocols.
- std::deque PotentiallyInterestingDecls;
+ std::deque PotentiallyInterestingDecls;
/// \brief The list of redeclaration chains that still need to be
/// reconstructed, and the local offset to the corresponding list
Modified: cfe/trunk/lib/Serialization/ASTCommon.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Serialization/ASTCommon.cpp?rev=315439&r1=315438&r2=315439&view=diff
==
--- cfe/trunk/lib/Serialization/ASTCommon.cpp (original)
+++ cfe/trunk/lib/Serialization/ASTCommon.cpp Wed Oct 11 00:47:54 2017
@@ -344,8 +344,8 @@ bool serialization::needsAnonymousDeclar
return true;
}
- // Otherwise, we only care about anonymous class members / block-scope decls.
- if (D->getDeclName() || D->getLexicalDeclContext()->isFileContext())
+ // Otherwise, we only care about anonymous class members.
+ if (D->getDeclName() || !isa(D->getLexicalDeclContext()))
return false;
return isa(D) || isa(D);
}
Modified: cfe/trunk/lib/Serialization/ASTReader.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Serialization/ASTReader.cpp?rev=315439&r1=315438&r2=315439&view=diff
==
--- cfe/trunk/lib/Serialization/ASTReader.cpp (original)
+++ cfe/trunk/lib/Serialization/ASTReader.cpp Wed Oct 11 00:47:54 2017
@@ -9169,6 +9169,30 @@ void ASTReader::finishPendingActions() {
}
PendingDefinitions.clear();
+ // Load the bodies of any functions or methods we've encountered. We do
+ // this now (delayed) so that we can be sure that the declaration chains
+ // have been fully wired up (hasBody relies on this).
+ // FIXME: We shouldn't require complete redeclaration chains here.
+ for (PendingBodiesMap::iterator PB = PendingBodies.begin(),
+ PBEnd = PendingBodies.end();
+ PB != PBEnd; ++PB) {
+if (FunctionDecl *FD = dyn_cast(PB->first)) {
+ // FIXME: Check for =delete/=default?
+ // FIXME: Complain about ODR violations here?
+ const FunctionDecl *Defn = nullptr;
+ if (!getContext().getLangOpts().Modules || !FD->hasBody(Defn)) {
+FD->setLazyBody(PB->second);
+ } else
+mergeDefinitionVisibility(const_cast(Defn), FD);
+ continue;
+}
+
+ObjCMethodDecl *MD = cast(PB->first);
+if (!getContext().getLangOpts().Modules || !MD->hasBody())
+ MD->setLazyBody(
[PATCH] D36973: [libclang] Add support for querying cursor availability
jklaehn marked an inline comment as done. jklaehn added a comment. In https://reviews.llvm.org/D36973#893851, @jbcoe wrote: > LGTM > > Would you like me to commit this for you? Yes, that would be great! https://reviews.llvm.org/D36973 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38402: [clang-refactor] Apply source replacements
hokein accepted this revision. hokein added a comment. LGTM, let's check in it. Repository: rL LLVM https://reviews.llvm.org/D38402 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D36955: [libclang] Visit attributes for function and class templates
jklaehn added a comment. In https://reviews.llvm.org/D36955#893856, @jbcoe wrote: > LGTM > > Would you like me to commit this for you? Yes, I would appreciate it. https://reviews.llvm.org/D36955 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38733: [CodeGen] Generate TBAA info along with LValue base info
kosarev updated this revision to Diff 118551.
kosarev added a comment.
Removed changes related to functions like getNaturalTypeAlignment() that
generate LValue base info objects to simplify the patch. These removed changes
will be addressed with separate patches.
https://reviews.llvm.org/D38733
Files:
lib/CodeGen/CGExpr.cpp
lib/CodeGen/CGOpenMPRuntime.cpp
lib/CodeGen/CodeGenFunction.cpp
lib/CodeGen/CodeGenFunction.h
Index: lib/CodeGen/CodeGenFunction.h
===
--- lib/CodeGen/CodeGenFunction.h
+++ lib/CodeGen/CodeGenFunction.h
@@ -1917,9 +1917,9 @@
CGM.getTBAAAccessInfo(T));
}
- LValue MakeAddrLValue(Address Addr, QualType T, LValueBaseInfo BaseInfo) {
-return LValue::MakeAddr(Addr, T, getContext(), BaseInfo,
-CGM.getTBAAAccessInfo(T));
+ LValue MakeAddrLValue(Address Addr, QualType T, LValueBaseInfo BaseInfo,
+TBAAAccessInfo TBAAInfo) {
+return LValue::MakeAddr(Addr, T, getContext(), BaseInfo, TBAAInfo);
}
LValue MakeAddrLValue(llvm::Value *V, QualType T, CharUnits Alignment,
@@ -1930,9 +1930,9 @@
}
LValue MakeAddrLValue(llvm::Value *V, QualType T, CharUnits Alignment,
-LValueBaseInfo BaseInfo) {
+LValueBaseInfo BaseInfo, TBAAAccessInfo TBAAInfo) {
return LValue::MakeAddr(Address(V, Alignment), T, getContext(),
-BaseInfo, CGM.getTBAAAccessInfo(T));
+BaseInfo, TBAAInfo);
}
LValue MakeNaturalAlignPointeeAddrLValue(llvm::Value *V, QualType T);
Index: lib/CodeGen/CodeGenFunction.cpp
===
--- lib/CodeGen/CodeGenFunction.cpp
+++ lib/CodeGen/CodeGenFunction.cpp
@@ -180,7 +180,8 @@
CodeGenFunction::MakeNaturalAlignPointeeAddrLValue(llvm::Value *V, QualType T) {
LValueBaseInfo BaseInfo;
CharUnits Align = getNaturalTypeAlignment(T, &BaseInfo, /*pointee*/ true);
- return MakeAddrLValue(Address(V, Align), T, BaseInfo);
+ return MakeAddrLValue(Address(V, Align), T, BaseInfo,
+CGM.getTBAAAccessInfo(T));
}
Index: lib/CodeGen/CGOpenMPRuntime.cpp
===
--- lib/CodeGen/CGOpenMPRuntime.cpp
+++ lib/CodeGen/CGOpenMPRuntime.cpp
@@ -980,7 +980,8 @@
SharedLVal = CGF.MakeAddrLValue(
CGF.Builder.CreateElementBitCast(SharedLVal.getAddress(),
CGF.ConvertTypeForMem(SharedType)),
- SharedType, SharedAddresses[N].first.getBaseInfo());
+ SharedType, SharedAddresses[N].first.getBaseInfo(),
+ CGF.CGM.getTBAAAccessInfo(SharedType));
if (isa(ClausesData[N].Ref) ||
CGF.getContext().getAsArrayType(PrivateVD->getType())) {
emitAggregateInitialization(CGF, N, PrivateAddr, SharedLVal, DRD);
@@ -1033,7 +1034,8 @@
return CGF.MakeAddrLValue(
CGF.Builder.CreateElementBitCast(BaseLV.getAddress(),
CGF.ConvertTypeForMem(ElTy)),
- BaseLV.getType(), BaseLV.getBaseInfo());
+ BaseLV.getType(), BaseLV.getBaseInfo(),
+ CGF.CGM.getTBAAAccessInfo(BaseLV.getType()));
}
static Address castToBase(CodeGenFunction &CGF, QualType BaseTy, QualType ElTy,
@@ -4072,7 +4074,8 @@
Address(SharedRefLValue.getPointer(), C.getDeclAlign(OriginalVD)),
SharedRefLValue.getType(),
LValueBaseInfo(AlignmentSource::Decl,
- SharedRefLValue.getBaseInfo().getMayAlias()));
+ SharedRefLValue.getBaseInfo().getMayAlias()),
+CGF.CGM.getTBAAAccessInfo(SharedRefLValue.getType()));
QualType Type = OriginalVD->getType();
if (Type->isArrayType()) {
// Initialize firstprivate array.
Index: lib/CodeGen/CGExpr.cpp
===
--- lib/CodeGen/CGExpr.cpp
+++ lib/CodeGen/CGExpr.cpp
@@ -2159,7 +2159,8 @@
const ReferenceType *RefTy) {
LValueBaseInfo BaseInfo;
Address Addr = EmitLoadOfReference(RefAddr, RefTy, &BaseInfo);
- return MakeAddrLValue(Addr, RefTy->getPointeeType(), BaseInfo);
+ return MakeAddrLValue(Addr, RefTy->getPointeeType(), BaseInfo,
+CGM.getTBAAAccessInfo(RefTy->getPointeeType()));
}
Address CodeGenFunction::EmitLoadOfPointer(Address Ptr,
@@ -2175,7 +2176,8 @@
const PointerType *PtrTy) {
LValueBaseInfo BaseInfo;
Address Addr = EmitLoadOfPointer(PtrAddr, PtrTy, &BaseInfo);
- return MakeAddrLValue(Addr, PtrTy->getPointeeType(), BaseInfo);
+ return MakeAddrLValue(Addr, PtrTy->getPointeeType(), BaseInfo,
+CGM.getTBAAAccessInfo(PtrTy->getPointeeType()));
}
static LValue EmitGlobalVarDeclLValue(CodeGenFunction &CGF,
@@ -2328,7 +2330,8 @
[PATCH] D38772: [refactor] allow the use of refactoring diagnostics
hokein added a comment. The code looks most good to me, a few nits. Comment at: lib/Basic/DiagnosticIDs.cpp:46 unsigned WarnShowInSystemHeader : 1; - unsigned Category : 5; + unsigned Category : 6; just curious: is this change needed? Comment at: tools/clang-refactor/ToolRefactoringResultConsumer.h:19 + +/// A subclass of \c RefactoringResultConsumer that stores the reference to the +/// TU-specific diagnostics engine. I'd name it "interface", because it has unimplemented virtual function (`handleError`), clients can't create an instance of it. or alternatively, does it make more sense to just add these methods and `DiagnosticsEngine` variable to the `tooling::RefactoringResultConsumer` interface? I see you have replaced "RefactoringResultConsumer" with this new interface in many places. Repository: rL LLVM https://reviews.llvm.org/D38772 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38788: [CodeGen] EmitCXXMemberDataPointerAddress() to generate TBAA info along with LValue base info
kosarev created this revision.
kosarev added a project: clang.
Prepared on top of https://reviews.llvm.org/D38733.
Repository:
rL LLVM
https://reviews.llvm.org/D38788
Files:
CodeGen/CGClass.cpp
CodeGen/CGExpr.cpp
CodeGen/CodeGenFunction.h
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -3327,7 +3327,8 @@
Address EmitCXXMemberDataPointerAddress(const Expr *E, Address base,
llvm::Value *memberPtr,
const MemberPointerType
*memberPtrType,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
RValue EmitCXXMemberPointerCallExpr(const CXXMemberCallExpr *E,
ReturnValueSlot ReturnValue);
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -4571,11 +4571,12 @@
= E->getRHS()->getType()->getAs();
LValueBaseInfo BaseInfo;
+ TBAAAccessInfo TBAAInfo;
Address MemberAddr =
-EmitCXXMemberDataPointerAddress(E, BaseAddr, OffsetV, MPT, &BaseInfo);
+EmitCXXMemberDataPointerAddress(E, BaseAddr, OffsetV, MPT, &BaseInfo,
+&TBAAInfo);
- return MakeAddrLValue(MemberAddr, MPT->getPointeeType(), BaseInfo,
-CGM.getTBAAAccessInfo(MPT->getPointeeType()));
+ return MakeAddrLValue(MemberAddr, MPT->getPointeeType(), BaseInfo, TBAAInfo);
}
/// Given the address of a temporary variable, produce an r-value of
Index: CodeGen/CGClass.cpp
===
--- CodeGen/CGClass.cpp
+++ CodeGen/CGClass.cpp
@@ -129,13 +129,16 @@
CodeGenFunction::EmitCXXMemberDataPointerAddress(const Expr *E, Address base,
llvm::Value *memberPtr,
const MemberPointerType *memberPtrType,
- LValueBaseInfo *BaseInfo) {
+ LValueBaseInfo *BaseInfo,
+ TBAAAccessInfo *TBAAInfo) {
// Ask the ABI to compute the actual address.
llvm::Value *ptr =
CGM.getCXXABI().EmitMemberDataPointerAddress(*this, E, base,
memberPtr, memberPtrType);
QualType memberType = memberPtrType->getPointeeType();
+ if (TBAAInfo)
+*TBAAInfo = CGM.getTBAAAccessInfo(memberType);
CharUnits memberAlign = getNaturalTypeAlignment(memberType, BaseInfo);
memberAlign =
CGM.getDynamicOffsetAlignment(base.getAlignment(),
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -3327,7 +3327,8 @@
Address EmitCXXMemberDataPointerAddress(const Expr *E, Address base,
llvm::Value *memberPtr,
const MemberPointerType *memberPtrType,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
RValue EmitCXXMemberPointerCallExpr(const CXXMemberCallExpr *E,
ReturnValueSlot ReturnValue);
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -4571,11 +4571,12 @@
= E->getRHS()->getType()->getAs();
LValueBaseInfo BaseInfo;
+ TBAAAccessInfo TBAAInfo;
Address MemberAddr =
-EmitCXXMemberDataPointerAddress(E, BaseAddr, OffsetV, MPT, &BaseInfo);
+EmitCXXMemberDataPointerAddress(E, BaseAddr, OffsetV, MPT, &BaseInfo,
+&TBAAInfo);
- return MakeAddrLValue(MemberAddr, MPT->getPointeeType(), BaseInfo,
-CGM.getTBAAAccessInfo(MPT->getPointeeType()));
+ return MakeAddrLValue(MemberAddr, MPT->getPointeeType(), BaseInfo, TBAAInfo);
}
/// Given the address of a temporary variable, produce an r-value of
Index: CodeGen/CGClass.cpp
===
--- CodeGen/CGClass.cpp
+++ CodeGen/CGClass.cpp
@@ -129,13 +129,16 @@
CodeGenFunction::EmitCXXMemberDataPointerAddress(const Expr *E, Address base,
llvm::Value *memberPtr,
const MemberPointerType *memberPtrType,
- LValueBaseInfo *BaseInfo) {
+
[PATCH] D38452: Mark test as a long-test
olista01 accepted this revision. olista01 added a comment. This revision is now accepted and ready to land. LGTM. https://reviews.llvm.org/D38452 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38723: [clang-rename] Don't add prefix qualifiers to the declaration and definition of the renamed symbol.
hokein updated this revision to Diff 118573.
hokein marked 3 inline comments as done.
hokein added a comment.
Address review comments.
https://reviews.llvm.org/D38723
Files:
lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
unittests/Rename/RenameClassTest.cpp
Index: unittests/Rename/RenameClassTest.cpp
===
--- unittests/Rename/RenameClassTest.cpp
+++ unittests/Rename/RenameClassTest.cpp
@@ -469,8 +469,6 @@
CompareSnippets(Expected, After);
}
-// FIXME: no prefix qualifiers being added to the class definition and
-// constructor.
TEST_F(ClangRenameTest, RenameClassWithNamespaceWithInlineMembers) {
std::string Before = R"(
namespace ns {
@@ -488,9 +486,9 @@
)";
std::string Expected = R"(
namespace ns {
- class ns::New {
+ class New {
public:
-ns::New() {}
+New() {}
~New() {}
New* next() { return next_; }
@@ -504,8 +502,6 @@
CompareSnippets(Expected, After);
}
-// FIXME: no prefix qualifiers being added to the class definition and
-// constructor.
TEST_F(ClangRenameTest, RenameClassWithNamespaceWithOutOfInlineMembers) {
std::string Before = R"(
namespace ns {
@@ -527,32 +523,32 @@
)";
std::string Expected = R"(
namespace ns {
- class ns::New {
+ class New {
public:
-ns::New();
+New();
~New();
New* next();
private:
New* next_;
};
- New::ns::New() {}
+ New::New() {}
New::~New() {}
New* New::next() { return next_; }
} // namespace ns
)";
std::string After = runClangRenameOnCode(Before, "ns::Old", "ns::New");
CompareSnippets(Expected, After);
}
-// FIXME: no prefix qualifiers being added to the definition.
TEST_F(ClangRenameTest, RenameClassInInheritedConstructor) {
// `using Base::Base;` will generate an implicit constructor containing usage
// of `::ns::Old` which should not be matched.
std::string Before = R"(
namespace ns {
+ class Old;
class Old {
int x;
};
@@ -574,7 +570,8 @@
})";
std::string Expected = R"(
namespace ns {
- class ns::New {
+ class New;
+ class New {
int x;
};
class Base {
@@ -615,7 +612,7 @@
)";
std::string Expected = R"(
namespace ns {
- class ::new_ns::New {
+ class New {
};
} // namespace ns
struct S {
@@ -632,7 +629,6 @@
CompareSnippets(Expected, After);
}
-// FIXME: no prefix qualifiers being adding to the definition.
TEST_F(ClangRenameTest, ReferencesInLambdaFunctionParameters) {
std::string Before = R"(
template
@@ -669,7 +665,7 @@
};
namespace ns {
- class ::new_ns::New {};
+ class New {};
void f() {
function func;
}
Index: lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
===
--- lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
+++ lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
@@ -160,13 +160,15 @@
const Decl *Context;
// The nested name being replaced (can be nullptr).
const NestedNameSpecifier *Specifier;
+// Determine whether the prefix qualifiers of the NewName should be ignored.
+// Normally, we set it to true for the symbol declaration and definition to
+// avoid adding prefix qualifiers.
+// For example, if it is true and NewName is
+// "a::b::foo", then the symbol occurrence which the RenameInfo points to
+// will be renamed to "foo".
+bool IgnorePrefixQualifers;
};
- // FIXME: Currently, prefix qualifiers will be added to the renamed symbol
- // definition (e.g. "class Foo {};" => "class b::Bar {};" when renaming
- // "a::Foo" to "b::Bar").
- // For renaming declarations/definitions, prefix qualifiers should be filtered
- // out.
bool VisitNamedDecl(const NamedDecl *Decl) {
// UsingDecl has been handled in other place.
if (llvm::isa(Decl))
@@ -180,8 +182,12 @@
return true;
if (isInUSRSet(Decl)) {
- RenameInfo Info = {Decl->getLocation(), Decl->getLocation(), nullptr,
- nullptr, nullptr};
+ RenameInfo Info = {Decl->getLocation(),
+ Decl->getLocation(),
+ /*FromDecl=*/nullptr,
+ /*Context=*/nullptr,
+ /*Specifier=*/nullptr,
+ /*IgnorePrefixQualifers=*/true};
RenameInfos.push_back(Info);
}
return true;
@@ -191,8 +197,11 @@
const NamedDecl *Decl = Expr->getFoundDecl();
if (isInUSRSet(Decl)) {
RenameInfo Info = {Expr->getSourceRange().getBegin(),
- Expr->getSourceRange().getEnd(), Decl,
- getClosestAncestorDecl(*Expr), Expr->getQualifier()};
+ Expr->getSourceR
[PATCH] D38791: [CodeGen] EmitLoadOfPointerLValue() to generate TBAA info along with LValue base info
kosarev created this revision.
kosarev added a project: clang.
This patch should not bring in any functional changes. Prepared on top of
https://reviews.llvm.org/D38733.
Repository:
rL LLVM
https://reviews.llvm.org/D38791
Files:
CodeGen/CGExpr.cpp
CodeGen/CodeGenFunction.h
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -1948,7 +1948,8 @@
LValue EmitLoadOfReferenceLValue(Address Ref, const ReferenceType *RefTy);
Address EmitLoadOfPointer(Address Ptr, const PointerType *PtrTy,
-LValueBaseInfo *BaseInfo = nullptr);
+LValueBaseInfo *BaseInfo = nullptr,
+TBAAAccessInfo *TBAAInfo = nullptr);
LValue EmitLoadOfPointerLValue(Address Ptr, const PointerType *PtrTy);
/// CreateTempAlloca - This creates an alloca and inserts it into the entry
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -2165,7 +2165,11 @@
Address CodeGenFunction::EmitLoadOfPointer(Address Ptr,
const PointerType *PtrTy,
- LValueBaseInfo *BaseInfo) {
+ LValueBaseInfo *BaseInfo,
+ TBAAAccessInfo *TBAAInfo) {
+ if (TBAAInfo)
+*TBAAInfo = CGM.getTBAAAccessInfo(PtrTy->getPointeeType());
+
llvm::Value *Addr = Builder.CreateLoad(Ptr);
return Address(Addr, getNaturalTypeAlignment(PtrTy->getPointeeType(),
BaseInfo,
@@ -2175,9 +2179,9 @@
LValue CodeGenFunction::EmitLoadOfPointerLValue(Address PtrAddr,
const PointerType *PtrTy) {
LValueBaseInfo BaseInfo;
- Address Addr = EmitLoadOfPointer(PtrAddr, PtrTy, &BaseInfo);
- return MakeAddrLValue(Addr, PtrTy->getPointeeType(), BaseInfo,
-CGM.getTBAAAccessInfo(PtrTy->getPointeeType()));
+ TBAAAccessInfo TBAAInfo;
+ Address Addr = EmitLoadOfPointer(PtrAddr, PtrTy, &BaseInfo, &TBAAInfo);
+ return MakeAddrLValue(Addr, PtrTy->getPointeeType(), BaseInfo, TBAAInfo);
}
static LValue EmitGlobalVarDeclLValue(CodeGenFunction &CGF,
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -1948,7 +1948,8 @@
LValue EmitLoadOfReferenceLValue(Address Ref, const ReferenceType *RefTy);
Address EmitLoadOfPointer(Address Ptr, const PointerType *PtrTy,
-LValueBaseInfo *BaseInfo = nullptr);
+LValueBaseInfo *BaseInfo = nullptr,
+TBAAAccessInfo *TBAAInfo = nullptr);
LValue EmitLoadOfPointerLValue(Address Ptr, const PointerType *PtrTy);
/// CreateTempAlloca - This creates an alloca and inserts it into the entry
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -2165,7 +2165,11 @@
Address CodeGenFunction::EmitLoadOfPointer(Address Ptr,
const PointerType *PtrTy,
- LValueBaseInfo *BaseInfo) {
+ LValueBaseInfo *BaseInfo,
+ TBAAAccessInfo *TBAAInfo) {
+ if (TBAAInfo)
+*TBAAInfo = CGM.getTBAAAccessInfo(PtrTy->getPointeeType());
+
llvm::Value *Addr = Builder.CreateLoad(Ptr);
return Address(Addr, getNaturalTypeAlignment(PtrTy->getPointeeType(),
BaseInfo,
@@ -2175,9 +2179,9 @@
LValue CodeGenFunction::EmitLoadOfPointerLValue(Address PtrAddr,
const PointerType *PtrTy) {
LValueBaseInfo BaseInfo;
- Address Addr = EmitLoadOfPointer(PtrAddr, PtrTy, &BaseInfo);
- return MakeAddrLValue(Addr, PtrTy->getPointeeType(), BaseInfo,
-CGM.getTBAAAccessInfo(PtrTy->getPointeeType()));
+ TBAAAccessInfo TBAAInfo;
+ Address Addr = EmitLoadOfPointer(PtrAddr, PtrTy, &BaseInfo, &TBAAInfo);
+ return MakeAddrLValue(Addr, PtrTy->getPointeeType(), BaseInfo, TBAAInfo);
}
static LValue EmitGlobalVarDeclLValue(CodeGenFunction &CGF,
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38793: [CodeGen] EmitLoadOfReference() to generate TBAA info along with LValue base info
kosarev created this revision.
kosarev added a project: clang.
Repository:
rL LLVM
https://reviews.llvm.org/D38793
Files:
CodeGen/CGExpr.cpp
CodeGen/CodeGenFunction.h
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -1944,7 +1944,8 @@
LValueBaseInfo *BaseInfo = nullptr);
Address EmitLoadOfReference(Address Ref, const ReferenceType *RefTy,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
LValue EmitLoadOfReferenceLValue(Address Ref, const ReferenceType *RefTy);
Address EmitLoadOfPointer(Address Ptr, const PointerType *PtrTy,
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -2149,18 +2149,22 @@
Address CodeGenFunction::EmitLoadOfReference(Address Addr,
const ReferenceType *RefTy,
- LValueBaseInfo *BaseInfo) {
+ LValueBaseInfo *BaseInfo,
+ TBAAAccessInfo *TBAAInfo) {
+ if (TBAAInfo)
+*TBAAInfo = CGM.getTBAAAccessInfo(RefTy->getPointeeType());
+
llvm::Value *Ptr = Builder.CreateLoad(Addr);
return Address(Ptr, getNaturalTypeAlignment(RefTy->getPointeeType(),
BaseInfo, /*forPointee*/ true));
}
LValue CodeGenFunction::EmitLoadOfReferenceLValue(Address RefAddr,
const ReferenceType *RefTy) {
LValueBaseInfo BaseInfo;
- Address Addr = EmitLoadOfReference(RefAddr, RefTy, &BaseInfo);
- return MakeAddrLValue(Addr, RefTy->getPointeeType(), BaseInfo,
-CGM.getTBAAAccessInfo(RefTy->getPointeeType()));
+ TBAAAccessInfo TBAAInfo;
+ Address Addr = EmitLoadOfReference(RefAddr, RefTy, &BaseInfo, &TBAAInfo);
+ return MakeAddrLValue(Addr, RefTy->getPointeeType(), BaseInfo, TBAAInfo);
}
Address CodeGenFunction::EmitLoadOfPointer(Address Ptr,
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -1944,7 +1944,8 @@
LValueBaseInfo *BaseInfo = nullptr);
Address EmitLoadOfReference(Address Ref, const ReferenceType *RefTy,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
LValue EmitLoadOfReferenceLValue(Address Ref, const ReferenceType *RefTy);
Address EmitLoadOfPointer(Address Ptr, const PointerType *PtrTy,
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -2149,18 +2149,22 @@
Address CodeGenFunction::EmitLoadOfReference(Address Addr,
const ReferenceType *RefTy,
- LValueBaseInfo *BaseInfo) {
+ LValueBaseInfo *BaseInfo,
+ TBAAAccessInfo *TBAAInfo) {
+ if (TBAAInfo)
+*TBAAInfo = CGM.getTBAAAccessInfo(RefTy->getPointeeType());
+
llvm::Value *Ptr = Builder.CreateLoad(Addr);
return Address(Ptr, getNaturalTypeAlignment(RefTy->getPointeeType(),
BaseInfo, /*forPointee*/ true));
}
LValue CodeGenFunction::EmitLoadOfReferenceLValue(Address RefAddr,
const ReferenceType *RefTy) {
LValueBaseInfo BaseInfo;
- Address Addr = EmitLoadOfReference(RefAddr, RefTy, &BaseInfo);
- return MakeAddrLValue(Addr, RefTy->getPointeeType(), BaseInfo,
-CGM.getTBAAAccessInfo(RefTy->getPointeeType()));
+ TBAAAccessInfo TBAAInfo;
+ Address Addr = EmitLoadOfReference(RefAddr, RefTy, &BaseInfo, &TBAAInfo);
+ return MakeAddrLValue(Addr, RefTy->getPointeeType(), BaseInfo, TBAAInfo);
}
Address CodeGenFunction::EmitLoadOfPointer(Address Ptr,
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38794: [CodeGen] getNaturalTypeAlignment() to generate TBAA info along with LValue base info
kosarev created this revision.
kosarev added a project: clang.
This patch should not bring in any functional changes. Prepared on top of
https://reviews.llvm.org/D38733, https://reviews.llvm.org/D38788,
https://reviews.llvm.org/D38791 and https://reviews.llvm.org/D38793.
Repository:
rL LLVM
https://reviews.llvm.org/D38794
Files:
CodeGen/CGClass.cpp
CodeGen/CGExpr.cpp
CodeGen/CodeGenFunction.cpp
CodeGen/CodeGenFunction.h
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -1939,6 +1939,7 @@
LValue MakeNaturalAlignAddrLValue(llvm::Value *V, QualType T);
CharUnits getNaturalTypeAlignment(QualType T,
LValueBaseInfo *BaseInfo = nullptr,
+TBAAAccessInfo *TBAAInfo = nullptr,
bool forPointeeType = false);
CharUnits getNaturalPointeeTypeAlignment(QualType T,
LValueBaseInfo *BaseInfo = nullptr);
Index: CodeGen/CodeGenFunction.cpp
===
--- CodeGen/CodeGenFunction.cpp
+++ CodeGen/CodeGenFunction.cpp
@@ -120,12 +120,17 @@
CharUnits CodeGenFunction::getNaturalPointeeTypeAlignment(QualType T,
LValueBaseInfo *BaseInfo) {
return getNaturalTypeAlignment(T->getPointeeType(), BaseInfo,
- /*forPointee*/ true);
+ /* TBAAInfo= */ nullptr,
+ /* forPointeeType= */ true);
}
CharUnits CodeGenFunction::getNaturalTypeAlignment(QualType T,
LValueBaseInfo *BaseInfo,
+ TBAAAccessInfo *TBAAInfo,
bool forPointeeType) {
+ if (TBAAInfo)
+*TBAAInfo = CGM.getTBAAAccessInfo(T);
+
// Honor alignment typedef attributes even on incomplete types.
// We also honor them straight for C++ class types, even as pointees;
// there's an expressivity gap here.
@@ -169,19 +174,21 @@
LValue CodeGenFunction::MakeNaturalAlignAddrLValue(llvm::Value *V, QualType T) {
LValueBaseInfo BaseInfo;
- CharUnits Alignment = getNaturalTypeAlignment(T, &BaseInfo);
+ TBAAAccessInfo TBAAInfo;
+ CharUnits Alignment = getNaturalTypeAlignment(T, &BaseInfo, &TBAAInfo);
return LValue::MakeAddr(Address(V, Alignment), T, getContext(), BaseInfo,
- CGM.getTBAAAccessInfo(T));
+ TBAAInfo);
}
/// Given a value of type T* that may not be to a complete object,
/// construct an l-value with the natural pointee alignment of T.
LValue
CodeGenFunction::MakeNaturalAlignPointeeAddrLValue(llvm::Value *V, QualType T) {
LValueBaseInfo BaseInfo;
- CharUnits Align = getNaturalTypeAlignment(T, &BaseInfo, /*pointee*/ true);
- return MakeAddrLValue(Address(V, Align), T, BaseInfo,
-CGM.getTBAAAccessInfo(T));
+ TBAAAccessInfo TBAAInfo;
+ CharUnits Align = getNaturalTypeAlignment(T, &BaseInfo, &TBAAInfo,
+/* forPointeeType= */ true);
+ return MakeAddrLValue(Address(V, Align), T, BaseInfo, TBAAInfo);
}
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -2151,12 +2151,10 @@
const ReferenceType *RefTy,
LValueBaseInfo *BaseInfo,
TBAAAccessInfo *TBAAInfo) {
- if (TBAAInfo)
-*TBAAInfo = CGM.getTBAAAccessInfo(RefTy->getPointeeType());
-
llvm::Value *Ptr = Builder.CreateLoad(Addr);
return Address(Ptr, getNaturalTypeAlignment(RefTy->getPointeeType(),
- BaseInfo, /*forPointee*/ true));
+ BaseInfo, TBAAInfo,
+ /* forPointeeType= */ true));
}
LValue CodeGenFunction::EmitLoadOfReferenceLValue(Address RefAddr,
@@ -2171,12 +2169,9 @@
const PointerType *PtrTy,
LValueBaseInfo *BaseInfo,
TBAAAccessInfo *TBAAInfo) {
- if (TBAAInfo)
-*TBAAInfo = CGM.getTBAAAccessInfo(PtrTy->getPointeeType());
-
llvm::Value *Addr = Builder.CreateLoad(Ptr);
return Address(Addr, getNaturalTypeAlignment(PtrTy->getPointeeType(),
- BaseInfo,
+ BaseInfo, TBAAInfo,
/*forPointeeType=*/true));
}
@@ -2315,8 +2310,10 @@
// FIXME: Eventually we will want to emit vector element re
[PATCH] D38723: [clang-rename] Don't add prefix qualifiers to the declaration and definition of the renamed symbol.
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315452: [clang-rename] Don't add prefix qualifiers to the
declaration and definition of… (authored by hokein).
Changed prior to commit:
https://reviews.llvm.org/D38723?vs=118573&id=118587#toc
Repository:
rL LLVM
https://reviews.llvm.org/D38723
Files:
cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
cfe/trunk/unittests/Rename/RenameClassTest.cpp
Index: cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
===
--- cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
+++ cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
@@ -160,13 +160,14 @@
const Decl *Context;
// The nested name being replaced (can be nullptr).
const NestedNameSpecifier *Specifier;
+// Determine whether the prefix qualifiers of the NewName should be ignored.
+// Normally, we set it to true for the symbol declaration and definition to
+// avoid adding prefix qualifiers.
+// For example, if it is true and NewName is "a::b::foo", then the symbol
+// occurrence which the RenameInfo points to will be renamed to "foo".
+bool IgnorePrefixQualifers;
};
- // FIXME: Currently, prefix qualifiers will be added to the renamed symbol
- // definition (e.g. "class Foo {};" => "class b::Bar {};" when renaming
- // "a::Foo" to "b::Bar").
- // For renaming declarations/definitions, prefix qualifiers should be filtered
- // out.
bool VisitNamedDecl(const NamedDecl *Decl) {
// UsingDecl has been handled in other place.
if (llvm::isa(Decl))
@@ -180,8 +181,12 @@
return true;
if (isInUSRSet(Decl)) {
- RenameInfo Info = {Decl->getLocation(), Decl->getLocation(), nullptr,
- nullptr, nullptr};
+ RenameInfo Info = {Decl->getLocation(),
+ Decl->getLocation(),
+ /*FromDecl=*/nullptr,
+ /*Context=*/nullptr,
+ /*Specifier=*/nullptr,
+ /*IgnorePrefixQualifers=*/true};
RenameInfos.push_back(Info);
}
return true;
@@ -191,8 +196,11 @@
const NamedDecl *Decl = Expr->getFoundDecl();
if (isInUSRSet(Decl)) {
RenameInfo Info = {Expr->getSourceRange().getBegin(),
- Expr->getSourceRange().getEnd(), Decl,
- getClosestAncestorDecl(*Expr), Expr->getQualifier()};
+ Expr->getSourceRange().getEnd(),
+ Decl,
+ getClosestAncestorDecl(*Expr),
+ Expr->getQualifier(),
+ /*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
}
@@ -220,8 +228,10 @@
if (isInUSRSet(TargetDecl)) {
RenameInfo Info = {NestedLoc.getBeginLoc(),
EndLocationForType(NestedLoc.getTypeLoc()),
- TargetDecl, getClosestAncestorDecl(NestedLoc),
- NestedLoc.getNestedNameSpecifier()->getPrefix()};
+ TargetDecl,
+ getClosestAncestorDecl(NestedLoc),
+ NestedLoc.getNestedNameSpecifier()->getPrefix(),
+ /*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
}
}
@@ -265,9 +275,12 @@
if (!ParentTypeLoc.isNull() &&
isInUSRSet(getSupportedDeclFromTypeLoc(ParentTypeLoc)))
return true;
-RenameInfo Info = {StartLocationForType(Loc), EndLocationForType(Loc),
- TargetDecl, getClosestAncestorDecl(Loc),
- GetNestedNameForType(Loc)};
+RenameInfo Info = {StartLocationForType(Loc),
+ EndLocationForType(Loc),
+ TargetDecl,
+ getClosestAncestorDecl(Loc),
+ GetNestedNameForType(Loc),
+ /*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
return true;
}
@@ -293,11 +306,13 @@
llvm::isa(ParentTypeLoc.getType()))
TargetLoc = ParentTypeLoc;
RenameInfo Info = {
-StartLocationForType(TargetLoc), EndLocationForType(TargetLoc),
+StartLocationForType(TargetLoc),
+EndLocationForType(TargetLoc),
TemplateSpecType->getTemplateName().getAsTemplateDecl(),
getClosestAncestorDecl(
ast_type_traits::DynTypedNode::create(TargetLoc)),
-GetNestedNameForType(TargetLoc)};
+GetNestedNameForType(TargetLoc),
+/*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
}
}
@@ -423,18 +438,26 @@
for (const auto &RenameInfo : Finder.getRenameInfos()) {
std::string ReplacedName =
r315452 - [clang-rename] Don't add prefix qualifiers to the declaration and definition of the renamed symbol.
Author: hokein
Date: Wed Oct 11 04:15:48 2017
New Revision: 315452
URL: http://llvm.org/viewvc/llvm-project?rev=315452&view=rev
Log:
[clang-rename] Don't add prefix qualifiers to the declaration and definition of
the renamed symbol.
Reviewers: ioeric
Reviewed By: ioeric
Subscribers: klimek, cfe-commits, arphaman
Differential Revision: https://reviews.llvm.org/D38723
Modified:
cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
cfe/trunk/unittests/Rename/RenameClassTest.cpp
Modified: cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp?rev=315452&r1=315451&r2=315452&view=diff
==
--- cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp (original)
+++ cfe/trunk/lib/Tooling/Refactoring/Rename/USRLocFinder.cpp Wed Oct 11
04:15:48 2017
@@ -160,13 +160,14 @@ public:
const Decl *Context;
// The nested name being replaced (can be nullptr).
const NestedNameSpecifier *Specifier;
+// Determine whether the prefix qualifiers of the NewName should be
ignored.
+// Normally, we set it to true for the symbol declaration and definition to
+// avoid adding prefix qualifiers.
+// For example, if it is true and NewName is "a::b::foo", then the symbol
+// occurrence which the RenameInfo points to will be renamed to "foo".
+bool IgnorePrefixQualifers;
};
- // FIXME: Currently, prefix qualifiers will be added to the renamed symbol
- // definition (e.g. "class Foo {};" => "class b::Bar {};" when renaming
- // "a::Foo" to "b::Bar").
- // For renaming declarations/definitions, prefix qualifiers should be
filtered
- // out.
bool VisitNamedDecl(const NamedDecl *Decl) {
// UsingDecl has been handled in other place.
if (llvm::isa(Decl))
@@ -180,8 +181,12 @@ public:
return true;
if (isInUSRSet(Decl)) {
- RenameInfo Info = {Decl->getLocation(), Decl->getLocation(), nullptr,
- nullptr, nullptr};
+ RenameInfo Info = {Decl->getLocation(),
+ Decl->getLocation(),
+ /*FromDecl=*/nullptr,
+ /*Context=*/nullptr,
+ /*Specifier=*/nullptr,
+ /*IgnorePrefixQualifers=*/true};
RenameInfos.push_back(Info);
}
return true;
@@ -191,8 +196,11 @@ public:
const NamedDecl *Decl = Expr->getFoundDecl();
if (isInUSRSet(Decl)) {
RenameInfo Info = {Expr->getSourceRange().getBegin(),
- Expr->getSourceRange().getEnd(), Decl,
- getClosestAncestorDecl(*Expr), Expr->getQualifier()};
+ Expr->getSourceRange().getEnd(),
+ Decl,
+ getClosestAncestorDecl(*Expr),
+ Expr->getQualifier(),
+ /*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
}
@@ -220,8 +228,10 @@ public:
if (isInUSRSet(TargetDecl)) {
RenameInfo Info = {NestedLoc.getBeginLoc(),
EndLocationForType(NestedLoc.getTypeLoc()),
- TargetDecl, getClosestAncestorDecl(NestedLoc),
- NestedLoc.getNestedNameSpecifier()->getPrefix()};
+ TargetDecl,
+ getClosestAncestorDecl(NestedLoc),
+ NestedLoc.getNestedNameSpecifier()->getPrefix(),
+ /*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
}
}
@@ -265,9 +275,12 @@ public:
if (!ParentTypeLoc.isNull() &&
isInUSRSet(getSupportedDeclFromTypeLoc(ParentTypeLoc)))
return true;
-RenameInfo Info = {StartLocationForType(Loc), EndLocationForType(Loc),
- TargetDecl, getClosestAncestorDecl(Loc),
- GetNestedNameForType(Loc)};
+RenameInfo Info = {StartLocationForType(Loc),
+ EndLocationForType(Loc),
+ TargetDecl,
+ getClosestAncestorDecl(Loc),
+ GetNestedNameForType(Loc),
+ /*IgnorePrefixQualifers=*/false};
RenameInfos.push_back(Info);
return true;
}
@@ -293,11 +306,13 @@ public:
llvm::isa(ParentTypeLoc.getType()))
TargetLoc = ParentTypeLoc;
RenameInfo Info = {
-StartLocationForType(TargetLoc), EndLocationForType(TargetLoc),
+StartLocationForType(TargetLoc),
+EndLocationForType(TargetLoc),
TemplateSpecType->getTemplateName().getAsTemplateDecl(),
getClosestAncestorDecl(
ast_type_traits::DynTypedNode::create(TargetLoc)),
-GetNestedNameForTy
r315453 - [OpenCL] Allow function declaration with empty argument list.
Author: bader
Date: Wed Oct 11 04:16:31 2017
New Revision: 315453
URL: http://llvm.org/viewvc/llvm-project?rev=315453&view=rev
Log:
[OpenCL] Allow function declaration with empty argument list.
Treat 'f()' as 'f(void)' rather than a function w/o a prototype.
Reviewers: Anastasia, yaxunl
Reviewed By: Anastasia, yaxunl
Subscribers: cfe-commits, echuraev, chapuni
Differential Revision: https://reviews.llvm.org/D33681
Re-apply revision 306653.
Modified:
cfe/trunk/lib/Parse/ParseDecl.cpp
cfe/trunk/lib/Sema/SemaType.cpp
cfe/trunk/test/SemaOpenCL/func.cl
cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
Modified: cfe/trunk/lib/Parse/ParseDecl.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Parse/ParseDecl.cpp?rev=315453&r1=315452&r2=315453&view=diff
==
--- cfe/trunk/lib/Parse/ParseDecl.cpp (original)
+++ cfe/trunk/lib/Parse/ParseDecl.cpp Wed Oct 11 04:16:31 2017
@@ -5989,7 +5989,8 @@ void Parser::ParseFunctionDeclarator(Dec
else if (RequiresArg)
Diag(Tok, diag::err_argument_required_after_attribute);
-HasProto = ParamInfo.size() || getLangOpts().CPlusPlus;
+HasProto = ParamInfo.size() || getLangOpts().CPlusPlus
+|| getLangOpts().OpenCL;
// If we have the closing ')', eat it.
Tracker.consumeClose();
Modified: cfe/trunk/lib/Sema/SemaType.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Sema/SemaType.cpp?rev=315453&r1=315452&r2=315453&view=diff
==
--- cfe/trunk/lib/Sema/SemaType.cpp (original)
+++ cfe/trunk/lib/Sema/SemaType.cpp Wed Oct 11 04:16:31 2017
@@ -4460,7 +4460,8 @@ static TypeSourceInfo *GetFullTypeForDec
FunctionType::ExtInfo EI(getCCForDeclaratorChunk(S, D, FTI, chunkIndex));
- if (!FTI.NumParams && !FTI.isVariadic && !LangOpts.CPlusPlus) {
+ if (!FTI.NumParams && !FTI.isVariadic && !LangOpts.CPlusPlus
+&& !LangOpts.OpenCL) {
// Simple void foo(), where the incoming T is the result type.
T = Context.getFunctionNoProtoType(T, EI);
} else {
Modified: cfe/trunk/test/SemaOpenCL/func.cl
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/test/SemaOpenCL/func.cl?rev=315453&r1=315452&r2=315453&view=diff
==
--- cfe/trunk/test/SemaOpenCL/func.cl (original)
+++ cfe/trunk/test/SemaOpenCL/func.cl Wed Oct 11 04:16:31 2017
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only
+// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only -triple
spir-unknown-unknown
// Variadic functions
void vararg_f(int, ...);// expected-error {{invalid
prototype, variadic arguments are not allowed in OpenCL}}
@@ -16,6 +16,9 @@ typedef struct s
//Function pointer
void foo(void*);
+// Expect no diagnostics for an empty parameter list.
+void bar();
+
void bar()
{
// declaring a function pointer is an error
Modified: cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl?rev=315453&r1=315452&r2=315453&view=diff
==
--- cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl (original)
+++ cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl Wed Oct 11 04:16:31 2017
@@ -3,7 +3,7 @@
global pipe int gp;// expected-error {{type '__global read_only
pipe int' can only be used as a function parameter in OpenCL}}
global reserve_id_t rid; // expected-error {{the '__global
reserve_id_t' type cannot be used to declare a program scope variable}}
-extern pipe write_only int get_pipe(); // expected-error {{type '__global
write_only pipe int ()' can only be used as a function parameter in OpenCL}}
+extern pipe write_only int get_pipe(); // expected-error {{type '__global
write_only pipe int (void)' can only be used as a function parameter in OpenCL}}
kernel void test_invalid_reserved_id(reserve_id_t ID) { // expected-error
{{'reserve_id_t' cannot be used as the type of a kernel parameter}}
}
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D33681: [OpenCL] Allow function declaration with empty argument list.
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315453: [OpenCL] Allow function declaration with empty
argument list. (authored by bader).
Changed prior to commit:
https://reviews.llvm.org/D33681?vs=118437&id=118588#toc
Repository:
rL LLVM
https://reviews.llvm.org/D33681
Files:
cfe/trunk/lib/Parse/ParseDecl.cpp
cfe/trunk/lib/Sema/SemaType.cpp
cfe/trunk/test/SemaOpenCL/func.cl
cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
Index: cfe/trunk/test/SemaOpenCL/func.cl
===
--- cfe/trunk/test/SemaOpenCL/func.cl
+++ cfe/trunk/test/SemaOpenCL/func.cl
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only
+// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only -triple
spir-unknown-unknown
// Variadic functions
void vararg_f(int, ...);// expected-error {{invalid
prototype, variadic arguments are not allowed in OpenCL}}
@@ -16,6 +16,9 @@
//Function pointer
void foo(void*);
+// Expect no diagnostics for an empty parameter list.
+void bar();
+
void bar()
{
// declaring a function pointer is an error
Index: cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
===
--- cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
+++ cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
@@ -3,7 +3,7 @@
global pipe int gp;// expected-error {{type '__global read_only
pipe int' can only be used as a function parameter in OpenCL}}
global reserve_id_t rid; // expected-error {{the '__global
reserve_id_t' type cannot be used to declare a program scope variable}}
-extern pipe write_only int get_pipe(); // expected-error {{type '__global
write_only pipe int ()' can only be used as a function parameter in OpenCL}}
+extern pipe write_only int get_pipe(); // expected-error {{type '__global
write_only pipe int (void)' can only be used as a function parameter in OpenCL}}
kernel void test_invalid_reserved_id(reserve_id_t ID) { // expected-error
{{'reserve_id_t' cannot be used as the type of a kernel parameter}}
}
Index: cfe/trunk/lib/Sema/SemaType.cpp
===
--- cfe/trunk/lib/Sema/SemaType.cpp
+++ cfe/trunk/lib/Sema/SemaType.cpp
@@ -4460,7 +4460,8 @@
FunctionType::ExtInfo EI(getCCForDeclaratorChunk(S, D, FTI, chunkIndex));
- if (!FTI.NumParams && !FTI.isVariadic && !LangOpts.CPlusPlus) {
+ if (!FTI.NumParams && !FTI.isVariadic && !LangOpts.CPlusPlus
+&& !LangOpts.OpenCL) {
// Simple void foo(), where the incoming T is the result type.
T = Context.getFunctionNoProtoType(T, EI);
} else {
Index: cfe/trunk/lib/Parse/ParseDecl.cpp
===
--- cfe/trunk/lib/Parse/ParseDecl.cpp
+++ cfe/trunk/lib/Parse/ParseDecl.cpp
@@ -5989,7 +5989,8 @@
else if (RequiresArg)
Diag(Tok, diag::err_argument_required_after_attribute);
-HasProto = ParamInfo.size() || getLangOpts().CPlusPlus;
+HasProto = ParamInfo.size() || getLangOpts().CPlusPlus
+|| getLangOpts().OpenCL;
// If we have the closing ')', eat it.
Tracker.consumeClose();
Index: cfe/trunk/test/SemaOpenCL/func.cl
===
--- cfe/trunk/test/SemaOpenCL/func.cl
+++ cfe/trunk/test/SemaOpenCL/func.cl
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only
+// RUN: %clang_cc1 %s -verify -pedantic -fsyntax-only -triple spir-unknown-unknown
// Variadic functions
void vararg_f(int, ...);// expected-error {{invalid prototype, variadic arguments are not allowed in OpenCL}}
@@ -16,6 +16,9 @@
//Function pointer
void foo(void*);
+// Expect no diagnostics for an empty parameter list.
+void bar();
+
void bar()
{
// declaring a function pointer is an error
Index: cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
===
--- cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
+++ cfe/trunk/test/SemaOpenCL/invalid-pipes-cl2.0.cl
@@ -3,7 +3,7 @@
global pipe int gp;// expected-error {{type '__global read_only pipe int' can only be used as a function parameter in OpenCL}}
global reserve_id_t rid; // expected-error {{the '__global reserve_id_t' type cannot be used to declare a program scope variable}}
-extern pipe write_only int get_pipe(); // expected-error {{type '__global write_only pipe int ()' can only be used as a function parameter in OpenCL}}
+extern pipe write_only int get_pipe(); // expected-error {{type '__global write_only pipe int (void)' can only be used as a function parameter in OpenCL}}
kernel void test_invalid_reserved_id(reserve_id_t ID) { // expected-error {{'reserve_id_t' cannot be used as the ty
[PATCH] D38795: [CodeGen] emitOMPArraySectionBase() to generate TBAA info along with LValue base info
kosarev created this revision.
kosarev added a project: clang.
Prepared on top of https://reviews.llvm.org/D38733.
Repository:
rL LLVM
https://reviews.llvm.org/D38795
Files:
CodeGen/CGExpr.cpp
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -3318,8 +3318,11 @@
static Address emitOMPArraySectionBase(CodeGenFunction &CGF, const Expr *Base,
LValueBaseInfo &BaseInfo,
+ TBAAAccessInfo &TBAAInfo,
QualType BaseTy, QualType ElTy,
bool IsLowerBound) {
+ TBAAInfo = CGF.CGM.getTBAAAccessInfo(ElTy);
+
LValue BaseLVal;
if (auto *ASE = dyn_cast(Base->IgnoreParenImpCasts())) {
BaseLVal = CGF.EmitOMPArraySectionExpr(ASE, IsLowerBound);
@@ -3454,13 +3457,14 @@
Address EltPtr = Address::invalid();
LValueBaseInfo BaseInfo;
+ TBAAAccessInfo TBAAInfo;
if (auto *VLA = getContext().getAsVariableArrayType(ResultExprTy)) {
// The base must be a pointer, which is not an aggregate. Emit
// it. It needs to be emitted first in case it's what captures
// the VLA bounds.
Address Base =
-emitOMPArraySectionBase(*this, E->getBase(), BaseInfo, BaseTy,
-VLA->getElementType(), IsLowerBound);
+emitOMPArraySectionBase(*this, E->getBase(), BaseInfo, TBAAInfo,
+BaseTy, VLA->getElementType(), IsLowerBound);
// The element count here is the total number of non-VLA elements.
llvm::Value *NumElements = getVLASize(VLA).first;
@@ -3496,16 +3500,17 @@
ResultExprTy, !getLangOpts().isSignedOverflowDefined(),
/*SignedIndices=*/false, E->getExprLoc());
BaseInfo = ArrayLV.getBaseInfo();
+TBAAInfo = CGM.getTBAAAccessInfo(ResultExprTy);
} else {
Address Base = emitOMPArraySectionBase(*this, E->getBase(), BaseInfo,
- BaseTy, ResultExprTy, IsLowerBound);
+ TBAAInfo, BaseTy, ResultExprTy,
+ IsLowerBound);
EltPtr = emitArraySubscriptGEP(*this, Base, Idx, ResultExprTy,
!getLangOpts().isSignedOverflowDefined(),
/*SignedIndices=*/false, E->getExprLoc());
}
- return MakeAddrLValue(EltPtr, ResultExprTy, BaseInfo,
-CGM.getTBAAAccessInfo(ResultExprTy));
+ return MakeAddrLValue(EltPtr, ResultExprTy, BaseInfo, TBAAInfo);
}
LValue CodeGenFunction::
Index: CodeGen/CGExpr.cpp
===
--- CodeGen/CGExpr.cpp
+++ CodeGen/CGExpr.cpp
@@ -3318,8 +3318,11 @@
static Address emitOMPArraySectionBase(CodeGenFunction &CGF, const Expr *Base,
LValueBaseInfo &BaseInfo,
+ TBAAAccessInfo &TBAAInfo,
QualType BaseTy, QualType ElTy,
bool IsLowerBound) {
+ TBAAInfo = CGF.CGM.getTBAAAccessInfo(ElTy);
+
LValue BaseLVal;
if (auto *ASE = dyn_cast(Base->IgnoreParenImpCasts())) {
BaseLVal = CGF.EmitOMPArraySectionExpr(ASE, IsLowerBound);
@@ -3454,13 +3457,14 @@
Address EltPtr = Address::invalid();
LValueBaseInfo BaseInfo;
+ TBAAAccessInfo TBAAInfo;
if (auto *VLA = getContext().getAsVariableArrayType(ResultExprTy)) {
// The base must be a pointer, which is not an aggregate. Emit
// it. It needs to be emitted first in case it's what captures
// the VLA bounds.
Address Base =
-emitOMPArraySectionBase(*this, E->getBase(), BaseInfo, BaseTy,
-VLA->getElementType(), IsLowerBound);
+emitOMPArraySectionBase(*this, E->getBase(), BaseInfo, TBAAInfo,
+BaseTy, VLA->getElementType(), IsLowerBound);
// The element count here is the total number of non-VLA elements.
llvm::Value *NumElements = getVLASize(VLA).first;
@@ -3496,16 +3500,17 @@
ResultExprTy, !getLangOpts().isSignedOverflowDefined(),
/*SignedIndices=*/false, E->getExprLoc());
BaseInfo = ArrayLV.getBaseInfo();
+TBAAInfo = CGM.getTBAAAccessInfo(ResultExprTy);
} else {
Address Base = emitOMPArraySectionBase(*this, E->getBase(), BaseInfo,
- BaseTy, ResultExprTy, IsLowerBound);
+ TBAAInfo, BaseTy, ResultExprTy,
+ IsLowerBound);
EltPtr = emitArraySubscriptGEP(*this, Base, Idx, ResultExprTy,
!getLangOpts().isSignedOverflowDefined(),
/*SignedIndices=*/false, E->getExprLoc());
}
- return MakeAddrLValue(EltPt
[PATCH] D38728: [analyzer] Use the signature of the primary template for issue hash calculation
xazax.hun added inline comments.
Comment at: lib/StaticAnalyzer/Core/IssueHash.cpp:39
+ // primary template.
+ if (const FunctionDecl *InstantiatedFrom =
+ Target->getInstantiatedFromMemberFunction())
martong wrote:
> Could we use here FunctionDecl::getPrimaryTemplate() ? That seems more
> general, it handles both specializations and instantiations.
Unfortunately `getPrimaryTemplate` is not sufficient. The function might be a
member function of a template class. In this case, there is no primary template
for the function (only for the enclosing class) but it still depends on a
template parameter.
Comment at: test/Analysis/bug_hash_test.cpp:61
-// CHECK: diagnostics
+template
+T f(T i) {
martong wrote:
> We could add a few more test cases:
> - a function template in class template
> - specializations vs instantiations
> - the combination of the above two (?)
>
Good point.
Comment at: test/Analysis/bug_hash_test.cpp:1363
// CHECK-NEXT:
+// CHECK-NEXT:
+// CHECK-NEXT: path
martong wrote:
> I am not sure if this is possible, but could we add unit test just for the
> `GetSignature` function? Instead of these huge plist files?
>
> I am thinking something like this:
> https://github.com/martong/friend-stats/blob/ed0c69ea3669c933204c799f59b85cd7b2507c34/ut/FriendFunctionsTest.cpp#L31
I think it is more convenient to use regression test for this purpose than
unittests. But I replaced the long plist checking with something much more
concise.
https://reviews.llvm.org/D38728
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38728: [analyzer] Use the signature of the primary template for issue hash calculation
martong added inline comments.
Comment at: test/Analysis/bug_hash_test.cpp:105
+void g() {
+ TX x;
+ TX xl;
As we discussed, the checking of the equality of the `IssueString` in case of
`TX` and `TX` is implicit. And as such it is hard to see that it
is really tested. Perhaps a comment here could indicate this implicit checking
mechanism.
https://reviews.llvm.org/D38728
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D37182: [libcxx] Special visibility macros for the experimental library
hamzasood added a comment. I have access now, so I'm able to commit this myself. However it's been a while since it was approved, so I'd be grateful if someone could take another look to make sure nothing has changed in the meantime (besides potentially needing to re-tag some new APIs). https://reviews.llvm.org/D37182 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D35109: [Analyzer] SValBuilder Comparison Rearrangement
baloghadamsoftware added a comment. I tried to extend the type to avoid overflow scenarios. Unfortunately, this breaks essential calculations based on the overflow scenarios (e.g. ProgramSate::assumeInbound()). So I see no other option than to abandon this patch and return to the local solution in the iterator checkers. https://reviews.llvm.org/D35109 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D36111: [COFF, ARM64] Add MS builtins __dmb, __dsb, __isb
mstorsjo added a comment. @mgrang, did you ever get to completing this? I've got a need for this now (only `__dmb` so far), and if you don't have time, I can try to finish it. https://reviews.llvm.org/D36111 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38796: [CodeGen] EmitPointerWithAlignment() to generate TBAA info along with LValue base info
kosarev created this revision.
kosarev added a project: clang.
Prepared on top of https://reviews.llvm.org/D38795.
Repository:
rL LLVM
https://reviews.llvm.org/D38796
Files:
CodeGen/CGExpr.cpp
CodeGen/CodeGenFunction.cpp
CodeGen/CodeGenFunction.h
CodeGen/CodeGenModule.cpp
CodeGen/CodeGenModule.h
CodeGen/CodeGenTBAA.cpp
CodeGen/CodeGenTBAA.h
Index: CodeGen/CodeGenTBAA.h
===
--- CodeGen/CodeGenTBAA.h
+++ CodeGen/CodeGenTBAA.h
@@ -47,6 +47,12 @@
: TBAAAccessInfo(/* AccessType= */ nullptr)
{}
+ bool operator==(const TBAAAccessInfo &Other) const {
+return BaseType == Other.BaseType &&
+ AccessType == Other.AccessType &&
+ Offset == Other.Offset;
+ }
+
/// BaseType - The base/leading access type. May be null if this access
/// descriptor represents an access that is not considered to be an access
/// to an aggregate or union member.
@@ -136,6 +142,11 @@
/// getMayAliasAccessInfo - Get TBAA information that represents may-alias
/// accesses.
TBAAAccessInfo getMayAliasAccessInfo();
+
+ /// mergeTBAAInfoForCast - Get merged TBAA information for the purpose of
+ /// type casts.
+ TBAAAccessInfo mergeTBAAInfoForCast(TBAAAccessInfo SourceInfo,
+ TBAAAccessInfo TargetInfo);
};
} // end namespace CodeGen
Index: CodeGen/CodeGenTBAA.cpp
===
--- CodeGen/CodeGenTBAA.cpp
+++ CodeGen/CodeGenTBAA.cpp
@@ -309,3 +309,11 @@
TBAAAccessInfo CodeGenTBAA::getMayAliasAccessInfo() {
return TBAAAccessInfo(getChar());
}
+
+TBAAAccessInfo CodeGenTBAA::mergeTBAAInfoForCast(TBAAAccessInfo SourceInfo,
+ TBAAAccessInfo TargetInfo) {
+ TBAAAccessInfo MayAliasInfo = getMayAliasAccessInfo();
+ if (SourceInfo == MayAliasInfo || TargetInfo == MayAliasInfo)
+return MayAliasInfo;
+ return TargetInfo;
+}
Index: CodeGen/CodeGenModule.h
===
--- CodeGen/CodeGenModule.h
+++ CodeGen/CodeGenModule.h
@@ -677,6 +677,11 @@
/// may-alias accesses.
TBAAAccessInfo getTBAAMayAliasAccessInfo();
+ /// mergeTBAAInfoForCast - Get merged TBAA information for the purposes of
+ /// type casts.
+ TBAAAccessInfo mergeTBAAInfoForCast(TBAAAccessInfo SourceInfo,
+ TBAAAccessInfo TargetInfo);
+
bool isTypeConstant(QualType QTy, bool ExcludeCtorDtor);
bool isPaddedAtomicType(QualType type);
Index: CodeGen/CodeGenModule.cpp
===
--- CodeGen/CodeGenModule.cpp
+++ CodeGen/CodeGenModule.cpp
@@ -612,6 +612,13 @@
return TBAA->getMayAliasAccessInfo();
}
+TBAAAccessInfo CodeGenModule::mergeTBAAInfoForCast(TBAAAccessInfo SourceInfo,
+ TBAAAccessInfo TargetInfo) {
+ if (!TBAA)
+return TBAAAccessInfo();
+ return TBAA->mergeTBAAInfoForCast(SourceInfo, TargetInfo);
+}
+
void CodeGenModule::DecorateInstructionWithTBAA(llvm::Instruction *Inst,
TBAAAccessInfo TBAAInfo) {
if (llvm::MDNode *Tag = getTBAAAccessTagInfo(TBAAInfo))
Index: CodeGen/CodeGenFunction.h
===
--- CodeGen/CodeGenFunction.h
+++ CodeGen/CodeGenFunction.h
@@ -1942,7 +1942,8 @@
TBAAAccessInfo *TBAAInfo = nullptr,
bool forPointeeType = false);
CharUnits getNaturalPointeeTypeAlignment(QualType T,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
Address EmitLoadOfReference(Address Ref, const ReferenceType *RefTy,
LValueBaseInfo *BaseInfo = nullptr,
@@ -3188,7 +3189,8 @@
RValue EmitRValueForField(LValue LV, const FieldDecl *FD, SourceLocation Loc);
Address EmitArrayToPointerDecay(const Expr *Array,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
class ConstantEmission {
llvm::PointerIntPair ValueAndIsReference;
@@ -3910,7 +3912,8 @@
/// reasonable to just ignore the returned alignment when it isn't from an
/// explicit source.
Address EmitPointerWithAlignment(const Expr *Addr,
- LValueBaseInfo *BaseInfo = nullptr);
+ LValueBaseInfo *BaseInfo = nullptr,
+ TBAAAccessInfo *TBAAInfo = nullptr);
void EmitSanitizerStatReport(llvm::SanitizerStatKind SSK);
Index: CodeGen/CodeGenFuncti
[PATCH] D38797: [analyzer] CStringChecker: pr34460: Admit that some casts are hard to model.
NoQ created this revision.
Herald added a subscriber: szepet.
In https://bugs.llvm.org/show_bug.cgi?id=34460 CStringChecker tries to
`evalCast()` a memory region from `void *` to `char *` for the purposes of
modeling `mempcpy()`. The memory region turned out to be an element region of
type `unsigned int` with a symbolic index. For such regions, even such trivial
casts have turned out to be unsupported (modeled as `UnknownVal`), and the
checker crashed upon asserting them to be supported.
I tried to get such trivial cast working for a few days and ended up believing
that it not worth the effort because of:
1. How we represent casts in the SVal hierarchy.
- In our case, the region is an ElementRegion with element type `unsigned
int`, which //essentially// represents a pointer of type `unsigned int *`, not
`void *`. It means that the cast is definitely not a no-op; the region needs to
be modified.
- Now, if the offset was concrete, we already try to replace ElementRegion
with another ElementRegion of the other type, and recompute element index
accordingly. It is only possible if byte offset of the region is divisible by
size of the target type (in our case it is because the target type is `char`).
If it's not divisible, we make two ElementRegions on top of each other, first
with element-type `char` to represent a raw offset, other of the target type to
represent the cast. It means that for symbolic index case, we need to account
for being able to receive such two-element-regions construct in the first
place, and then investigate divisibility of a linear expression of form `a*N +
b`, where `N` is concrete but `a` and `b` are possibly symbolic, to a concrete
type size, which our range constraint manager would never handle.
- In fact, now we not only unpack the top one or two ElementRegions while
computing the offset (that needs to be divisible), but also we unpack all other
top-level element regions (i.e. if we look into a multi-dimensional array; see
`ElementRegion::getAsArrayOffset()`). This was never tested; i added relevant
tests into `casts.c` in this patch to demonstrate the problems caused by that
(which are unrelated to the rest of the patch). I believe that unpacking
multi-dimensional arrays in this way is not ideal; however, i also added tests
that would fail if this behavior is disabled. This needs much more work. Of
course, this work would be even harder if we intend to support symbolic offsets.
2. How we split the responsibilities between entities. When trying to implement
the plan described in 1., i ended up moving a lot of code around and passing
ProgramState through dozens of APIs. We need the state to perform binary
operations on `SVal` objects through `SValBuilder`. Originally, `castRegion()`
resides in `RegionStore` which is not intended to access the program state. It
is trivial to move `castRegion` to `SimpleSValBuilder`, but passing `State`
into it results in a huge cascade of changes in `SValBuilder` method arguments.
I have given up when i had to pass `State` to
`SValBuilder::getConstantVal(const Expr *);`, which seemed ridiculous (it's
constant, why does it depend on the state?). Introducing all this complexity
only to compute an `SVal` that nobody would be able to understand was not worth
the effort, in my opinion.
Hence the quick fix. Code slightly simplified to use a more high-level API.
https://reviews.llvm.org/D38797
Files:
lib/StaticAnalyzer/Checkers/CStringChecker.cpp
test/Analysis/bstring.cpp
test/Analysis/casts.c
Index: test/Analysis/casts.c
===
--- test/Analysis/casts.c
+++ test/Analysis/casts.c
@@ -123,3 +123,29 @@
int x = (int) p;
clang_analyzer_eval(++x < 10); // no-crash // expected-warning{{UNKNOWN}}
}
+
+void multiDimensionalArrayPointerCasts() {
+ static int x[10][10];
+ int *y1 = &(x[3][5]);
+ char *z = ((char *) y1) + 2;
+ int *y2 = (int *)(z - 2);
+ int *y3 = ((int *)x) + 35; // This is offset for [3][5].
+
+ clang_analyzer_eval(y1 == y2); // expected-warning{{TRUE}}
+
+ // FIXME: should be FALSE (i.e. equal pointers).
+ clang_analyzer_eval(y1 - y2); // expected-warning{{UNKNOWN}}
+ // FIXME: should be TRUE (i.e. same symbol).
+ clang_analyzer_eval(*y1 == *y2); // expected-warning{{UNKNOWN}}
+
+ clang_analyzer_eval(*((char *)y1) == *((char *) y2)); // expected-warning{{TRUE}}
+
+ clang_analyzer_eval(y1 == y3); // expected-warning{{TRUE}}
+
+ // FIXME: should be FALSE (i.e. equal pointers).
+ clang_analyzer_eval(y1 - y3); // expected-warning{{UNKNOWN}}
+ // FIXME: should be TRUE (i.e. same symbol).
+ clang_analyzer_eval(*y1 == *y3); // expected-warning{{UNKNOWN}}
+
+ clang_analyzer_eval(*((char *)y1) == *((char *) y3)); // expected-warning{{TRUE}}
+}
Index: test/Analysis/bstring.cpp
===
--- test/Analysis/bstring.cpp
+++ test/Analysis/bstring.cpp
@@ -1,8 +1,35 @
[PATCH] D38797: [analyzer] CStringChecker: pr34460: Admit that some casts are hard to model.
NoQ added inline comments.
Comment at: test/Analysis/casts.c:134-139
+ clang_analyzer_eval(y1 == y2); // expected-warning{{TRUE}}
+
+ // FIXME: should be FALSE (i.e. equal pointers).
+ clang_analyzer_eval(y1 - y2); // expected-warning{{UNKNOWN}}
+ // FIXME: should be TRUE (i.e. same symbol).
+ clang_analyzer_eval(*y1 == *y2); // expected-warning{{UNKNOWN}}
Here we have `y1` pointing to `element{element{x, 3}, 5}` vs. both `y2` and
`y3` pointing to `element{x, 35}`. While `y3` is inevitably looking like that,
`y2` could have been same as `y1` if we didn't unwrap multi-dimensional array
elements.
https://reviews.llvm.org/D38797
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D37856: [refactor] add support for refactoring options
hokein added a comment. Sorry for the delay. I saw you have reverted this commit somehow. A post commit. Comment at: cfe/trunk/lib/Tooling/Refactoring/Rename/RenamingAction.cpp:113 Rules.push_back(createRefactoringActionRule( -SymbolSelectionRequirement())); +SymbolSelectionRequirement(), OptionRequirement())); return Rules; Thought it a bit more: it requires all of the requirements are satisfied, I think we need to support "one-of" option. For example, we have two option "-a" and "-b", only one of them is allowed to be present at the same time. Repository: rL LLVM https://reviews.llvm.org/D37856 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315459 - [clang-rename] Add more unittest.
Author: hokein
Date: Wed Oct 11 07:00:42 2017
New Revision: 315459
URL: http://llvm.org/viewvc/llvm-project?rev=315459&view=rev
Log:
[clang-rename] Add more unittest.
Modified:
cfe/trunk/unittests/Rename/RenameClassTest.cpp
Modified: cfe/trunk/unittests/Rename/RenameClassTest.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/unittests/Rename/RenameClassTest.cpp?rev=315459&r1=315458&r2=315459&view=diff
==
--- cfe/trunk/unittests/Rename/RenameClassTest.cpp (original)
+++ cfe/trunk/unittests/Rename/RenameClassTest.cpp Wed Oct 11 07:00:42 2017
@@ -674,6 +674,124 @@ TEST_F(ClangRenameTest, ReferencesInLamb
CompareSnippets(Expected, After);
}
+TEST_F(ClangRenameTest, DontChangeIfSameName) {
+ std::string Before = R"(
+ namespace foo {
+ class Old {
+ public:
+ static void foo() {}
+ };
+ }
+
+ void f(foo::Old * x) {
+foo::Old::foo() ;
+ }
+ using foo::Old;)";
+ std::string Expected = R"(
+ namespace foo {
+ class Old {
+ public:
+ static void foo() {}
+ };
+ }
+
+ void f(foo::Old * x) {
+foo::Old::foo() ;
+ }
+ using foo::Old;)";
+ std::string After = runClangRenameOnCode(Before, "foo::Old", "foo::Old");
+ CompareSnippets(Expected, After);
+}
+
+TEST_F(ClangRenameTest, ChangeIfNewNameWithLeadingDotDot) {
+ std::string Before = R"(
+ namespace foo {
+ class Old {
+ public:
+ static void foo() {}
+ };
+ }
+
+ void f(foo::Old * x) {
+foo::Old::foo() ;
+ }
+ using foo::Old;)";
+ std::string Expected = R"(
+ namespace foo {
+ class Old {
+ public:
+ static void foo() {}
+ };
+ }
+
+ void f(::foo::Old * x) {
+::foo::Old::foo() ;
+ }
+ using ::foo::Old;)";
+ std::string After = runClangRenameOnCode(Before, "foo::Old", "::foo::Old");
+ CompareSnippets(Expected, After);
+}
+
+TEST_F(ClangRenameTest, ChangeIfSameNameWithLeadingDotDot) {
+ std::string Before = R"(
+ namespace foo {
+ class Old {
+ public:
+ static void foo() {}
+ };
+ }
+
+ void f(foo::Old * x) {
+foo::Old::foo() ;
+ }
+ using foo::Old;)";
+ std::string Expected = R"(
+ namespace foo {
+ class Old {
+ public:
+ static void foo() {}
+ };
+ }
+
+ void f(::foo::Old * x) {
+::foo::Old::foo() ;
+ }
+ using ::foo::Old;)";
+ std::string After = runClangRenameOnCode(Before, "::foo::Old", "::foo::Old");
+ CompareSnippets(Expected, After);
+}
+
+TEST_F(RenameClassTest, UsingAlias) {
+ std::string Before = R"(
+ namespace a { struct A {}; }
+
+ namespace foo {
+ using Alias = a::A;
+ Alias a;
+ })";
+ std::string Expected = R"(
+ namespace a { struct B {}; }
+
+ namespace foo {
+ using Alias = b::B;
+ Alias a;
+ })";
+ std::string After = runClangRenameOnCode(Before, "a::A", "b::B");
+ CompareSnippets(Expected, After);
+}
+
+TEST_F(ClangRenameTest, NestedTemplates) {
+ std::string Before = R"(
+ namespace a { template struct A {}; }
+ a::A> foo;)";
+ std::string Expected = R"(
+ namespace a { template struct B {}; }
+ b::B> foo;)";
+ std::string After = runClangRenameOnCode(Before, "a::A", "b::B");
+ CompareSnippets(Expected, After);
+}
+
+
} // anonymous namespace
} // namespace test
} // namespace clang_rename
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38798: [OpenMP] Support for implicit "declare target" functions - Sema patch
grokos created this revision.
grokos added a project: clang.
This patch completes the support for the "declare target" directive in Sema.
With this patch Sema handles implicitly used functions (i.e. functions which
are used inside a target region without having been "declared target")
including lambdas, templated functions, functions called from within target
functions and ctors/dtors.
By default, use of implicit declare target functions is enabled. An upcoming
driver patch will change that.
Repository:
rL LLVM
https://reviews.llvm.org/D38798
Files:
include/clang/Basic/LangOptions.def
include/clang/Sema/Sema.h
include/clang/Sema/SemaInternal.h
lib/Parse/ParseOpenMP.cpp
lib/Sema/SemaDecl.cpp
lib/Sema/SemaOpenMP.cpp
Index: lib/Sema/SemaOpenMP.cpp
===
--- lib/Sema/SemaOpenMP.cpp
+++ lib/Sema/SemaOpenMP.cpp
@@ -19,6 +19,7 @@
#include "clang/AST/Decl.h"
#include "clang/AST/DeclCXX.h"
#include "clang/AST/DeclOpenMP.h"
+#include "clang/AST/RecursiveASTVisitor.h"
#include "clang/AST/StmtCXX.h"
#include "clang/AST/StmtOpenMP.h"
#include "clang/AST/StmtVisitor.h"
@@ -1139,6 +1140,124 @@
return false;
}
+namespace {
+/// Visit actual function body and its associated nested functions bodies.
+class ImplicitDeviceFunctionChecker
+: public RecursiveASTVisitor {
+ Sema &SemaRef;
+
+public:
+ ImplicitDeviceFunctionChecker(Sema &SemaReference) : SemaRef(SemaReference){};
+
+ /// Traverse body of lambda, and mark it the with OMPDeclareTargetDeclAttr
+ bool TraverseLambdaCapture(LambdaExpr *LE, const LambdaCapture *C,
+ Expr *Init);
+
+ /// Traverse FunctionDecl and mark it the with OMPDeclareTargetDeclAttr
+ bool VisitFunctionDecl(FunctionDecl *F);
+
+ /// Traverse Callee of Calexpr and mark it the with OMPDeclareTargetDeclAttr
+ bool VisitCallExpr(CallExpr *Call);
+
+ /// Traverse Constructs and mark it the with OMPDeclareTargetDeclAttr
+ bool VisitCXXConstructExpr(CXXConstructExpr *E);
+
+ /// Traverse Destructor and mark it the with OMPDeclareTargetDeclAttr
+ bool VisitCXXDestructorDecl(CXXDestructorDecl *D);
+};
+}
+
+/// Traverse declaration of /param D to check whether it has
+/// OMPDeclareTargetDeclAttr or not. If so, it marks definition with
+/// OMPDeclareTargetDeclAttr.
+static void ImplicitDeclareTargetCheck(Sema &SemaRef, Decl *D) {
+ if (SemaRef.getLangOpts().OpenMPImplicitDeclareTarget) {
+// Structured block of target region is visited to catch function call.
+// Revealed function calls are marked with OMPDeclareTargetDeclAttr
+// attribute,
+// in case -fopenmp-implicit-declare-target extension is enabled.
+ImplicitDeviceFunctionChecker FunctionCallChecker(SemaRef);
+FunctionCallChecker.TraverseDecl(D);
+ }
+}
+
+/// Traverse declaration of /param D to check whether it has
+/// OMPDeclareTargetDeclAttr or not. If so, it marks definition with
+/// OMPDeclareTargetDeclAttr.
+void Sema::checkDeclImplicitlyUsedOpenMPTargetContext(Decl *D) {
+ if (!D || D->isInvalidDecl())
+return;
+
+ if (FunctionDecl *FD = dyn_cast(D)) {
+if (FD->hasBody()) {
+ for (auto RI : FD->redecls()) {
+if (RI->hasAttr()) {
+ Attr *A = OMPDeclareTargetDeclAttr::CreateImplicit(
+ Context, OMPDeclareTargetDeclAttr::MT_To);
+ D->addAttr(A);
+
+ ImplicitDeclareTargetCheck(*this, FD);
+ return;
+}
+ }
+}
+ }
+ return;
+}
+
+bool ImplicitDeviceFunctionChecker::TraverseLambdaCapture(
+LambdaExpr *LE, const LambdaCapture *C, Expr *Init) {
+ if (CXXRecordDecl *Class = LE->getLambdaClass())
+if (!Class->hasAttr()) {
+ Attr *A = OMPDeclareTargetDeclAttr::CreateImplicit(
+ SemaRef.Context, OMPDeclareTargetDeclAttr::MT_To);
+ Class->addAttr(A);
+}
+
+ TraverseStmt(LE->getBody());
+ return true;
+}
+
+bool ImplicitDeviceFunctionChecker::VisitFunctionDecl(FunctionDecl *F) {
+ assert(F);
+ if (!F->hasAttr()) {
+Attr *A = OMPDeclareTargetDeclAttr::CreateImplicit(
+SemaRef.Context, OMPDeclareTargetDeclAttr::MT_To);
+F->addAttr(A);
+TraverseDecl(F);
+ }
+ return true;
+}
+
+bool ImplicitDeviceFunctionChecker::VisitCallExpr(CallExpr *Call) {
+ if (FunctionDecl *Callee = Call->getDirectCallee()) {
+return VisitFunctionDecl(Callee);
+ }
+ return true;
+}
+
+bool ImplicitDeviceFunctionChecker::VisitCXXConstructExpr(CXXConstructExpr *E) {
+ CXXConstructorDecl *Constructor = E->getConstructor();
+ // When constructor is invoked, it is checked whether the object has
+ // destructor or not. In case it has destructor, destructor is automatically
+ // marked with declare target attribute since it is needed to emit for device,
+ QualType Ty = E->getType();
+ const RecordType *RT =
+ SemaRef.Context.getBaseElementType(Ty)->getAs();
+ CXXRecordDecl *RD = cast(RT->getDecl());
+
+ if (auto *Destructor = RD->getDestructor())
+VisitC
[PATCH] D38679: [libunwind] Support dwarf unwinding on i386 windows
jroelofs accepted this revision. jroelofs added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D38679 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38801: [analyzer] In getSVal() API, disable auto-detection of void type as char type.
NoQ created this revision.
Herald added a subscriber: szepet.
In https://reviews.llvm.org/D38358, we ended up believing that reading the
first byte of the void pointer is not the intended behavior of
`ProgramState::getSVal(Loc)`. Hence the fix.
Additionally, allow specifying the type in the `ProgramState::getSVal(const
MemRegion *)` override (i personally like this API better, and it is also used
in the affected checker).
https://reviews.llvm.org/D38801
Files:
include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
lib/StaticAnalyzer/Core/RegionStore.cpp
Index: lib/StaticAnalyzer/Core/RegionStore.cpp
===
--- lib/StaticAnalyzer/Core/RegionStore.cpp
+++ lib/StaticAnalyzer/Core/RegionStore.cpp
@@ -1403,10 +1403,7 @@
T = Ctx.VoidTy;
}
assert(!T.isNull() && "Unable to auto-detect binding type!");
-if (T->isVoidType()) {
- // When trying to dereference a void pointer, read the first byte.
- T = Ctx.CharTy;
-}
+assert(!T->isVoidType() && "Attempting to dereference a void pointer!");
MR = GetElementZeroRegion(cast(MR), T);
}
Index: lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
===
--- lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
+++ lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
@@ -179,7 +179,7 @@
if (const MemRegion *SValMemRegion = V.getAsRegion()) {
const ProgramStateRef State = C.getState();
-const SVal PSV = State->getSVal(SValMemRegion);
+const SVal PSV = State->getSVal(SValMemRegion, C.getASTContext().CharTy);
if (PSV.isUndef()) {
if (ExplodedNode *N = C.generateErrorNode()) {
LazyInit_BT(BD, BT);
Index: include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
===
--- include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
+++ include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
@@ -308,8 +308,12 @@
/// \brief Return the value bound to the specified location.
/// Returns UnknownVal() if none found.
- SVal getSVal(const MemRegion* R) const;
+ SVal getSVal(const MemRegion* R, QualType T = QualType()) const;
+ /// \brief Return the value bound to the specified location, assuming
+ /// that the value is a scalar integer or an enumeration or a pointer.
+ /// Returns UnknownVal() if none found or the region is not known to hold
+ /// a value of such type.
SVal getSValAsScalarOrLoc(const MemRegion *R) const;
/// \brief Visits the symbols reachable from the given SVal using the
provided
@@ -758,9 +762,10 @@
return getStateManager().StoreMgr->getBinding(getStore(), LV, T);
}
-inline SVal ProgramState::getSVal(const MemRegion* R) const {
+inline SVal ProgramState::getSVal(const MemRegion* R, QualType T) const {
return getStateManager().StoreMgr->getBinding(getStore(),
-loc::MemRegionVal(R));
+loc::MemRegionVal(R),
+T);
}
inline BasicValueFactory &ProgramState::getBasicVals() const {
Index: lib/StaticAnalyzer/Core/RegionStore.cpp
===
--- lib/StaticAnalyzer/Core/RegionStore.cpp
+++ lib/StaticAnalyzer/Core/RegionStore.cpp
@@ -1403,10 +1403,7 @@
T = Ctx.VoidTy;
}
assert(!T.isNull() && "Unable to auto-detect binding type!");
-if (T->isVoidType()) {
- // When trying to dereference a void pointer, read the first byte.
- T = Ctx.CharTy;
-}
+assert(!T->isVoidType() && "Attempting to dereference a void pointer!");
MR = GetElementZeroRegion(cast(MR), T);
}
Index: lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
===
--- lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
+++ lib/StaticAnalyzer/Checkers/CallAndMessageChecker.cpp
@@ -179,7 +179,7 @@
if (const MemRegion *SValMemRegion = V.getAsRegion()) {
const ProgramStateRef State = C.getState();
-const SVal PSV = State->getSVal(SValMemRegion);
+const SVal PSV = State->getSVal(SValMemRegion, C.getASTContext().CharTy);
if (PSV.isUndef()) {
if (ExplodedNode *N = C.generateErrorNode()) {
LazyInit_BT(BD, BT);
Index: include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
===
--- include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
+++ include/clang/StaticAnalyzer/Core/PathSensitive/ProgramState.h
@@ -308,8 +308,12 @@
/// \brief Return the value bound to the specified location.
/// Returns UnknownVal() if none found.
- SVal getSVal(const MemRegion* R) const;
+ SVal getSVal(const MemRegion* R, Q
[PATCH] D33722: [clang-tidy] Add checker for undelegated copy of base classes
szdominik updated this revision to Diff 118617.
szdominik marked 4 inline comments as done.
szdominik added a comment.
Small changes after aaron.ballman's comments.
https://reviews.llvm.org/D33722
Files:
clang-tidy/misc/CMakeLists.txt
clang-tidy/misc/CopyConstructorInitCheck.cpp
clang-tidy/misc/CopyConstructorInitCheck.h
clang-tidy/misc/MiscTidyModule.cpp
docs/ReleaseNotes.rst
docs/clang-tidy/checks/list.rst
docs/clang-tidy/checks/misc-copy-constructor-init.rst
test/clang-tidy/misc-copy-constructor-init.cpp
Index: test/clang-tidy/misc-copy-constructor-init.cpp
===
--- /dev/null
+++ test/clang-tidy/misc-copy-constructor-init.cpp
@@ -0,0 +1,95 @@
+// RUN: %check_clang_tidy %s misc-copy-constructor-init %t
+
+class Copyable {
+ public:
+ Copyable() = default;
+ Copyable(const Copyable&) = default;
+};
+class X : public Copyable {
+ X(const X& other) : Copyable(other) {}
+ //Good code: the copy ctor call the ctor of the base class.
+};
+
+class Copyable2 {
+ public:
+ Copyable2() = default;
+ Copyable2(const Copyable2&) = default;
+};
+class X2 : public Copyable2 {
+ X2(const X2& other) {};
+// CHECK-MESSAGES: :[[@LINE-1]]:22: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X2(const X2& other) : Copyable2(other) {};
+};
+
+class X3 : public Copyable, public Copyable2 {
+ X3(const X3& other): Copyable(other) {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:23: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X3(const X3& other): Copyable2(other), Copyable(other) {};
+};
+
+class X4 : public Copyable {
+ X4(const X4& other): Copyable() {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:32: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X4(const X4& other): Copyable(other) {};
+};
+
+class Copyable3 : public Copyable {
+ public:
+ Copyable3() = default;
+ Copyable3(const Copyable3&) = default;
+};
+class X5 : public Copyable3 {
+ X5(const X5& other) {};
+// CHECK-MESSAGES: :[[@LINE-1]]:22: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X5(const X5& other) : Copyable3(other) {};
+};
+
+class X6 : public Copyable2, public Copyable3 {
+ X6(const X6& other) {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:22: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X6(const X6& other) : Copyable2(other), Copyable3(other) {};
+};
+
+class X7 : public Copyable, public Copyable2 {
+ X7(const X7& other): Copyable() {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:32: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-MESSAGES: :[[@LINE-2]]:22: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X7(const X7& other): Copyable2(other), Copyable(other) {};
+};
+
+template
+class Copyable4 {
+ public:
+ Copyable4() = default;
+ Copyable4(const Copyable4&) = default;
+};
+
+class X8 : public Copyable4 {
+ X8(const X8& other): Copyable4(other) {};
+ //Good code: the copy ctor call the ctor of the base class.
+};
+
+class X9 : public Copyable4 {
+ X9(const X9& other): Copyable4() {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:37: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X9(const X9& other): Copyable4(other) {};
+};
+
+class X10 : public Copyable4 {
+ X10(const X10& other) {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:22: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X10(const X10& other) : Copyable4(other) {};
+};
+
+template
+class Copyable5 {
+ public:
+ Copyable5() = default;
+ Copyable5(const Copyable5&) = default;
+};
+
+class X11 : public Copyable5 {
+ X11(const X11& other) {};
+ // CHECK-MESSAGES: :[[@LINE-1]]:22: warning: calling an inherited constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X11(const X11& other) : Copyable5(other) {};
+};
Index: docs/clang-tidy/checks/misc-copy-constructor-init.rst
===
--- /dev/null
+++ docs/clang-tidy/checks/misc-copy-constructor-init.rst
@@ -0,0 +1,30 @@
+.. title:: clang-tidy - misc-copy-constructor-init
+
+misc-copy-constructor-init
+=
+
+Finds copy constructors where the constructor don't call
+the constructor of the base class.
+
+.. code-block:: c++
+
+class Copyable {
+public:
+Copyable() = default;
+Copyable(const Copyable&) = default;
+};
+class X2 : public Copyable {
+X2(const X2& other) {}; // Copyable(other) is missing
+};
+
[PATCH] D33722: [clang-tidy] Add checker for undelegated copy of base classes
szdominik added inline comments.
Comment at: clang-tidy/misc/CopyConstructorInitCheck.cpp:37
+
+ // We match here because we want one warning (and FixIt) for every ctor.
+ const auto Matches = match(
aaron.ballman wrote:
> Wouldn't registering this matcher achieve the same goal instead of needing to
> re-match?
I wanted to create //only one// FixIt to every ctor - if I move the
forEachCtorInitializer to the register part, it would warn us twice.
Comment at: clang-tidy/misc/CopyConstructorInitCheck.cpp:104
+
+ diag(Tok.getLocation(),
+ "calling an inherited constructor other than the copy constructor")
aaron.ballman wrote:
> Insteaad of having to re-lex the physical source, can the AST should be
> modified to carry the information you need if it doesn't already have it? For
> instance, you can tell there is not initializer list by looking at
> `CXXConstructorDecl::getNumCtorInitializers()`.
The getNumCtorInitializers method counts the generated initializers as well,
not just the manually written ones.
My basic problem was that the AST's methods can't really distinguish the
generated and the manually written initializers, so it's quite complicated to
find the locations what we need. I found easier & more understandable if I
start reading the tokens.
Comment at: test/clang-tidy/misc-copy-constructor-init.cpp:27
+ // CHECK-MESSAGES: :[[@LINE-1]]:23: warning: calling an inherited
constructor other than the copy constructor [misc-copy-constructor-init]
+ // CHECK-FIXES: X3(const X3& other): Copyable2(other), Copyable(other)
{};
+};
aaron.ballman wrote:
> Don't we want the ctor-inits to be in the same order as the bases are
> specified?
I think it's more readable if we put the FixIts to the beginning of the
expression - it's easier to check that everyting is correct & it's more obvious
what the changes are.
https://reviews.llvm.org/D33722
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D37856: [refactor] add support for refactoring options
arphaman added a comment. In https://reviews.llvm.org/D37856#894638, @hokein wrote: > Sorry for the delay. I saw you have reverted this commit somehow. A post > commit. I had some issues with ppc64/s390x bots for some reason, so I had to revert. I'm still trying to investigate what went wrong there, it looks like ASAN/UBSAN are not catching anything though. Comment at: cfe/trunk/lib/Tooling/Refactoring/Rename/RenamingAction.cpp:113 Rules.push_back(createRefactoringActionRule( -SymbolSelectionRequirement())); +SymbolSelectionRequirement(), OptionRequirement())); return Rules; hokein wrote: > Thought it a bit more: it requires all of the requirements are satisfied, I > think we need to support "one-of" option. For example, we have two option > "-a" and "-b", only one of them is allowed to be present at the same time. That should be straightforward enough to implement, either with a custom class or with a built-in requirement class. I'll probably add a builtin one when the need comes up. Repository: rL LLVM https://reviews.llvm.org/D37856 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315462 - [Analyzer] Clarify error messages for undefined result
Author: danielmarjamaki
Date: Wed Oct 11 07:49:35 2017
New Revision: 315462
URL: http://llvm.org/viewvc/llvm-project?rev=315462&view=rev
Log:
[Analyzer] Clarify error messages for undefined result
Differential Revision: https://reviews.llvm.org/D30295
Modified:
cfe/trunk/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h
cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
cfe/trunk/lib/StaticAnalyzer/Checkers/UndefResultChecker.cpp
cfe/trunk/lib/StaticAnalyzer/Core/CheckerContext.cpp
cfe/trunk/test/Analysis/bitwise-ops.c
Modified:
cfe/trunk/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h?rev=315462&r1=315461&r2=315462&view=diff
==
--- cfe/trunk/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h
(original)
+++ cfe/trunk/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h
Wed Oct 11 07:49:35 2017
@@ -196,6 +196,13 @@ public:
return getState()->getSVal(S, getLocationContext());
}
+ /// \brief Returns true if the value of \p E is greater than or equal to \p
+ /// Val under unsigned comparison
+ bool isGreaterOrEqual(const Expr *E, unsigned long long Val);
+
+ /// Returns true if the value of \p E is negative.
+ bool isNegative(const Expr *E);
+
/// \brief Generates a new transition in the program state graph
/// (ExplodedGraph). Uses the default CheckerContext predecessor node.
///
Modified: cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp?rev=315462&r1=315461&r2=315462&view=diff
==
--- cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp (original)
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp Wed Oct 11
07:49:35 2017
@@ -123,57 +123,6 @@ void ConversionChecker::reportBug(Explod
C.emitReport(std::move(R));
}
-// Is E value greater or equal than Val?
-static bool isGreaterEqual(CheckerContext &C, const Expr *E,
- unsigned long long Val) {
- ProgramStateRef State = C.getState();
- SVal EVal = C.getSVal(E);
- if (EVal.isUnknownOrUndef())
-return false;
- if (!EVal.getAs() && EVal.getAs()) {
-ProgramStateManager &Mgr = C.getStateManager();
-EVal =
-Mgr.getStoreManager().getBinding(State->getStore(),
EVal.castAs());
- }
- if (EVal.isUnknownOrUndef() || !EVal.getAs())
-return false;
-
- SValBuilder &Bldr = C.getSValBuilder();
- DefinedSVal V = Bldr.makeIntVal(Val, C.getASTContext().LongLongTy);
-
- // Is DefinedEVal greater or equal with V?
- SVal GE = Bldr.evalBinOp(State, BO_GE, EVal, V, Bldr.getConditionType());
- if (GE.isUnknownOrUndef())
-return false;
- ConstraintManager &CM = C.getConstraintManager();
- ProgramStateRef StGE, StLT;
- std::tie(StGE, StLT) = CM.assumeDual(State, GE.castAs());
- return StGE && !StLT;
-}
-
-// Is E value negative?
-static bool isNegative(CheckerContext &C, const Expr *E) {
- ProgramStateRef State = C.getState();
- SVal EVal = State->getSVal(E, C.getLocationContext());
- if (EVal.isUnknownOrUndef() || !EVal.getAs())
-return false;
- DefinedSVal DefinedEVal = EVal.castAs();
-
- SValBuilder &Bldr = C.getSValBuilder();
- DefinedSVal V = Bldr.makeIntVal(0, false);
-
- SVal LT =
- Bldr.evalBinOp(State, BO_LT, DefinedEVal, V, Bldr.getConditionType());
-
- // Is E value greater than MaxVal?
- ConstraintManager &CM = C.getConstraintManager();
- ProgramStateRef StNegative, StPositive;
- std::tie(StNegative, StPositive) =
- CM.assumeDual(State, LT.castAs());
-
- return StNegative && !StPositive;
-}
-
bool ConversionChecker::isLossOfPrecision(const ImplicitCastExpr *Cast,
QualType DestType,
CheckerContext &C) const {
@@ -195,18 +144,18 @@ bool ConversionChecker::isLossOfPrecisio
return false;
unsigned long long MaxVal = 1ULL << W;
- return isGreaterEqual(C, Cast->getSubExpr(), MaxVal);
+ return C.isGreaterOrEqual(Cast->getSubExpr(), MaxVal);
}
bool ConversionChecker::isLossOfSign(const ImplicitCastExpr *Cast,
- CheckerContext &C) const {
+ CheckerContext &C) const {
QualType CastType = Cast->getType();
QualType SubType = Cast->IgnoreParenImpCasts()->getType();
if (!CastType->isUnsignedIntegerType() || !SubType->isSignedIntegerType())
return false;
- return isNegative(C, Cast->getSubExpr());
+ return C.isNegative(Cast->getSubExpr());
}
void ento::registerConversionChecker(CheckerManager &mgr) {
Modified: cfe/trunk/lib/StaticAnalyzer/Checkers/UndefResultChecker.cpp
URL:
http://llv
[PATCH] D30295: [analyzer] clarify undef shift result when shift count is negative or exceeds the bit width
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315462: [Analyzer] Clarify error messages for undefined
result (authored by danielmarjamaki).
Changed prior to commit:
https://reviews.llvm.org/D30295?vs=116865&id=118620#toc
Repository:
rL LLVM
https://reviews.llvm.org/D30295
Files:
cfe/trunk/include/clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h
cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
cfe/trunk/lib/StaticAnalyzer/Checkers/UndefResultChecker.cpp
cfe/trunk/lib/StaticAnalyzer/Core/CheckerContext.cpp
cfe/trunk/test/Analysis/bitwise-ops.c
Index: cfe/trunk/lib/StaticAnalyzer/Core/CheckerContext.cpp
===
--- cfe/trunk/lib/StaticAnalyzer/Core/CheckerContext.cpp
+++ cfe/trunk/lib/StaticAnalyzer/Core/CheckerContext.cpp
@@ -99,3 +99,35 @@
return Lexer::getSpelling(Loc, buf, getSourceManager(), getLangOpts());
}
+/// Evaluate comparison and return true if it's known that condition is true
+static bool evalComparison(SVal LHSVal, BinaryOperatorKind ComparisonOp,
+ SVal RHSVal, ProgramStateRef State) {
+ if (LHSVal.isUnknownOrUndef())
+return false;
+ ProgramStateManager &Mgr = State->getStateManager();
+ if (!LHSVal.getAs()) {
+LHSVal = Mgr.getStoreManager().getBinding(State->getStore(),
+ LHSVal.castAs());
+if (LHSVal.isUnknownOrUndef() || !LHSVal.getAs())
+ return false;
+ }
+
+ SValBuilder &Bldr = Mgr.getSValBuilder();
+ SVal Eval = Bldr.evalBinOp(State, ComparisonOp, LHSVal, RHSVal,
+ Bldr.getConditionType());
+ if (Eval.isUnknownOrUndef())
+return false;
+ ProgramStateRef StTrue, StFalse;
+ std::tie(StTrue, StFalse) = State->assume(Eval.castAs());
+ return StTrue && !StFalse;
+}
+
+bool CheckerContext::isGreaterOrEqual(const Expr *E, unsigned long long Val) {
+ DefinedSVal V = getSValBuilder().makeIntVal(Val, getASTContext().LongLongTy);
+ return evalComparison(getSVal(E), BO_GE, V, getState());
+}
+
+bool CheckerContext::isNegative(const Expr *E) {
+ DefinedSVal V = getSValBuilder().makeIntVal(0, false);
+ return evalComparison(getSVal(E), BO_LT, V, getState());
+}
Index: cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
===
--- cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/ConversionChecker.cpp
@@ -123,57 +123,6 @@
C.emitReport(std::move(R));
}
-// Is E value greater or equal than Val?
-static bool isGreaterEqual(CheckerContext &C, const Expr *E,
- unsigned long long Val) {
- ProgramStateRef State = C.getState();
- SVal EVal = C.getSVal(E);
- if (EVal.isUnknownOrUndef())
-return false;
- if (!EVal.getAs() && EVal.getAs()) {
-ProgramStateManager &Mgr = C.getStateManager();
-EVal =
-Mgr.getStoreManager().getBinding(State->getStore(), EVal.castAs());
- }
- if (EVal.isUnknownOrUndef() || !EVal.getAs())
-return false;
-
- SValBuilder &Bldr = C.getSValBuilder();
- DefinedSVal V = Bldr.makeIntVal(Val, C.getASTContext().LongLongTy);
-
- // Is DefinedEVal greater or equal with V?
- SVal GE = Bldr.evalBinOp(State, BO_GE, EVal, V, Bldr.getConditionType());
- if (GE.isUnknownOrUndef())
-return false;
- ConstraintManager &CM = C.getConstraintManager();
- ProgramStateRef StGE, StLT;
- std::tie(StGE, StLT) = CM.assumeDual(State, GE.castAs());
- return StGE && !StLT;
-}
-
-// Is E value negative?
-static bool isNegative(CheckerContext &C, const Expr *E) {
- ProgramStateRef State = C.getState();
- SVal EVal = State->getSVal(E, C.getLocationContext());
- if (EVal.isUnknownOrUndef() || !EVal.getAs())
-return false;
- DefinedSVal DefinedEVal = EVal.castAs();
-
- SValBuilder &Bldr = C.getSValBuilder();
- DefinedSVal V = Bldr.makeIntVal(0, false);
-
- SVal LT =
- Bldr.evalBinOp(State, BO_LT, DefinedEVal, V, Bldr.getConditionType());
-
- // Is E value greater than MaxVal?
- ConstraintManager &CM = C.getConstraintManager();
- ProgramStateRef StNegative, StPositive;
- std::tie(StNegative, StPositive) =
- CM.assumeDual(State, LT.castAs());
-
- return StNegative && !StPositive;
-}
-
bool ConversionChecker::isLossOfPrecision(const ImplicitCastExpr *Cast,
QualType DestType,
CheckerContext &C) const {
@@ -195,18 +144,18 @@
return false;
unsigned long long MaxVal = 1ULL << W;
- return isGreaterEqual(C, Cast->getSubExpr(), MaxVal);
+ return C.isGreaterOrEqual(Cast->getSubExpr(), MaxVal);
}
bool ConversionChecker::isLossOfSign(const ImplicitCastExpr *Cast,
- CheckerContext &C) const {
+ CheckerContext &C) const {
QualType CastType = Cast->getType();
Q
r315463 - [clang-fuzzer] Fix shared library dependencies.
Author: morehouse
Date: Wed Oct 11 08:13:53 2017
New Revision: 315463
URL: http://llvm.org/viewvc/llvm-project?rev=315463&view=rev
Log:
[clang-fuzzer] Fix shared library dependencies.
Modified:
cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
cfe/trunk/tools/clang-fuzzer/handle-cxx/CMakeLists.txt
cfe/trunk/tools/clang-fuzzer/proto-to-cxx/CMakeLists.txt
Modified: cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/tools/clang-fuzzer/CMakeLists.txt?rev=315463&r1=315462&r2=315463&view=diff
==
--- cfe/trunk/tools/clang-fuzzer/CMakeLists.txt (original)
+++ cfe/trunk/tools/clang-fuzzer/CMakeLists.txt Wed Oct 11 08:13:53 2017
@@ -37,6 +37,7 @@ if( LLVM_USE_SANITIZE_COVERAGE )
add_clang_executable(clang-proto-fuzzer ExampleClangProtoFuzzer.cpp)
target_link_libraries(clang-proto-fuzzer
${ProtobufMutator_LIBRARIES}
+ ${PROTOBUF_LIBRARIES}
clangCXXProto
clangHandleCXX
clangProtoToCXX
Modified: cfe/trunk/tools/clang-fuzzer/handle-cxx/CMakeLists.txt
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/tools/clang-fuzzer/handle-cxx/CMakeLists.txt?rev=315463&r1=315462&r2=315463&view=diff
==
--- cfe/trunk/tools/clang-fuzzer/handle-cxx/CMakeLists.txt (original)
+++ cfe/trunk/tools/clang-fuzzer/handle-cxx/CMakeLists.txt Wed Oct 11 08:13:53
2017
@@ -1,9 +1,10 @@
-set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD})
+set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} Support)
add_clang_library(clangHandleCXX
handle_cxx.cpp
LINK_LIBS
+ clangBasic
clangCodeGen
clangFrontend
clangLex
Modified: cfe/trunk/tools/clang-fuzzer/proto-to-cxx/CMakeLists.txt
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/tools/clang-fuzzer/proto-to-cxx/CMakeLists.txt?rev=315463&r1=315462&r2=315463&view=diff
==
--- cfe/trunk/tools/clang-fuzzer/proto-to-cxx/CMakeLists.txt (original)
+++ cfe/trunk/tools/clang-fuzzer/proto-to-cxx/CMakeLists.txt Wed Oct 11
08:13:53 2017
@@ -5,9 +5,9 @@ set(CMAKE_CXX_FLAGS ${CXX_FLAGS_NOFUZZ})
# an executable built from this directory.
set(LLVM_OPTIONAL_SOURCES proto_to_cxx.cpp proto_to_cxx_main.cpp)
-add_clang_library(clangProtoToCXX proto_to_cxx.cpp
+add_clang_library(clangProtoToCXX proto_to_cxx.cpp
DEPENDS clangCXXProto
- LINK_LIBS clangCXXProto
+ LINK_LIBS clangCXXProto ${PROTOBUF_LIBRARIES}
)
add_clang_executable(clang-proto-to-cxx proto_to_cxx_main.cpp)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D23963: [analyzer] pr28449 - Move literal rvalue construction away from RegionStore.
NoQ updated this revision to Diff 118625.
NoQ added a comment.
Herald added a subscriber: szepet.
Because i didn't get back to this in a while, and similar crashes keep coming,
i decided to leave this refactoring as a FIXME.
https://reviews.llvm.org/D23963
Files:
lib/StaticAnalyzer/Core/RegionStore.cpp
test/Analysis/compound-literals.c
test/Analysis/objc-encode.m
Index: test/Analysis/objc-encode.m
===
--- /dev/null
+++ test/Analysis/objc-encode.m
@@ -0,0 +1,9 @@
+// RUN: %clang_cc1 -analyze -analyzer-checker=debug.ExprInspection -verify %s
+// expected-no-diagnostics
+
+void clang_analyzer_eval(int);
+
+// rdar://problem/34831581: Used to crash.
+void foo(void) {
+ char buf1[] = @encode(int **);
+}
Index: test/Analysis/compound-literals.c
===
--- /dev/null
+++ test/Analysis/compound-literals.c
@@ -0,0 +1,9 @@
+// RUN: %clang_cc1 -triple=i386-apple-darwin10 -analyze
-analyzer-checker=debug.ExprInspection -verify %s
+void clang_analyzer_eval(int);
+
+// pr28449: Used to crash.
+void foo(void) {
+ static const unsigned short array[] = (const unsigned short[]){0x0F00};
+ // FIXME: Should be true.
+ clang_analyzer_eval(array[0] == 0x0F00); // expected-warning{{UNKNOWN}}
+}
Index: lib/StaticAnalyzer/Core/RegionStore.cpp
===
--- lib/StaticAnalyzer/Core/RegionStore.cpp
+++ lib/StaticAnalyzer/Core/RegionStore.cpp
@@ -2085,15 +2085,12 @@
if (const ConstantArrayType* CAT = dyn_cast(AT))
Size = CAT->getSize().getZExtValue();
- // Check if the init expr is a string literal.
+ // Check if the init expr is a literal. If so, bind the rvalue instead.
+ // FIXME: It's not responsibility of the Store to transform this lvalue
+ // to rvalue. ExprEngine or maybe even CFG should do this before binding.
if (Optional MRV = Init.getAs()) {
-const StringRegion *S = cast(MRV->getRegion());
-
-// Treat the string as a lazy compound value.
-StoreRef store(B.asStore(), *this);
-nonloc::LazyCompoundVal LCV = svalBuilder.makeLazyCompoundVal(store, S)
-.castAs();
-return bindAggregate(B, R, LCV);
+SVal V = getBinding(B.asStore(), *MRV, R->getValueType());
+return bindAggregate(B, R, V);
}
// Handle lazy compound values.
Index: test/Analysis/objc-encode.m
===
--- /dev/null
+++ test/Analysis/objc-encode.m
@@ -0,0 +1,9 @@
+// RUN: %clang_cc1 -analyze -analyzer-checker=debug.ExprInspection -verify %s
+// expected-no-diagnostics
+
+void clang_analyzer_eval(int);
+
+// rdar://problem/34831581: Used to crash.
+void foo(void) {
+ char buf1[] = @encode(int **);
+}
Index: test/Analysis/compound-literals.c
===
--- /dev/null
+++ test/Analysis/compound-literals.c
@@ -0,0 +1,9 @@
+// RUN: %clang_cc1 -triple=i386-apple-darwin10 -analyze -analyzer-checker=debug.ExprInspection -verify %s
+void clang_analyzer_eval(int);
+
+// pr28449: Used to crash.
+void foo(void) {
+ static const unsigned short array[] = (const unsigned short[]){0x0F00};
+ // FIXME: Should be true.
+ clang_analyzer_eval(array[0] == 0x0F00); // expected-warning{{UNKNOWN}}
+}
Index: lib/StaticAnalyzer/Core/RegionStore.cpp
===
--- lib/StaticAnalyzer/Core/RegionStore.cpp
+++ lib/StaticAnalyzer/Core/RegionStore.cpp
@@ -2085,15 +2085,12 @@
if (const ConstantArrayType* CAT = dyn_cast(AT))
Size = CAT->getSize().getZExtValue();
- // Check if the init expr is a string literal.
+ // Check if the init expr is a literal. If so, bind the rvalue instead.
+ // FIXME: It's not responsibility of the Store to transform this lvalue
+ // to rvalue. ExprEngine or maybe even CFG should do this before binding.
if (Optional MRV = Init.getAs()) {
-const StringRegion *S = cast(MRV->getRegion());
-
-// Treat the string as a lazy compound value.
-StoreRef store(B.asStore(), *this);
-nonloc::LazyCompoundVal LCV = svalBuilder.makeLazyCompoundVal(store, S)
-.castAs();
-return bindAggregate(B, R, LCV);
+SVal V = getBinding(B.asStore(), *MRV, R->getValueType());
+return bindAggregate(B, R, V);
}
// Handle lazy compound values.
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315464 - [OPENMP] Fix PR34916: Crash on mixing taskloop|tasks directives.
Author: abataev
Date: Wed Oct 11 08:29:40 2017
New Revision: 315464
URL: http://llvm.org/viewvc/llvm-project?rev=315464&view=rev
Log:
[OPENMP] Fix PR34916: Crash on mixing taskloop|tasks directives.
If both taskloop and task directives are used at the same time in one
program, we may ran into the situation when the particular type for task
directive is reused for taskloop directives. Patch fixes this problem.
Modified:
cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp
cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h
cfe/trunk/test/OpenMP/taskloop_codegen.cpp
Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp?rev=315464&r1=315463&r2=315464&view=diff
==
--- cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp (original)
+++ cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp Wed Oct 11 08:29:40 2017
@@ -4268,9 +4268,20 @@ CGOpenMPRuntime::emitTaskInit(CodeGenFun
// Build type kmp_routine_entry_t (if not built yet).
emitKmpRoutineEntryT(KmpInt32Ty);
// Build type kmp_task_t (if not built yet).
- if (KmpTaskTQTy.isNull()) {
-KmpTaskTQTy = C.getRecordType(createKmpTaskTRecordDecl(
-CGM, D.getDirectiveKind(), KmpInt32Ty, KmpRoutineEntryPtrQTy));
+ if (isOpenMPTaskLoopDirective(D.getDirectiveKind())) {
+if (SavedKmpTaskloopTQTy.isNull()) {
+ SavedKmpTaskloopTQTy = C.getRecordType(createKmpTaskTRecordDecl(
+ CGM, D.getDirectiveKind(), KmpInt32Ty, KmpRoutineEntryPtrQTy));
+}
+KmpTaskTQTy = SavedKmpTaskloopTQTy;
+ } else if (D.getDirectiveKind() == OMPD_task) {
+assert(D.getDirectiveKind() == OMPD_task &&
+ "Expected taskloop or task directive");
+if (SavedKmpTaskTQTy.isNull()) {
+ SavedKmpTaskTQTy = C.getRecordType(createKmpTaskTRecordDecl(
+ CGM, D.getDirectiveKind(), KmpInt32Ty, KmpRoutineEntryPtrQTy));
+}
+KmpTaskTQTy = SavedKmpTaskTQTy;
}
auto *KmpTaskTQTyRD = cast(KmpTaskTQTy->getAsTagDecl());
// Build particular struct kmp_task_t for the given task.
Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h?rev=315464&r1=315463&r2=315464&view=diff
==
--- cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h (original)
+++ cfe/trunk/lib/CodeGen/CGOpenMPRuntime.h Wed Oct 11 08:29:40 2017
@@ -318,6 +318,10 @@ private:
///deconstructors of firstprivate C++ objects */
/// } kmp_task_t;
QualType KmpTaskTQTy;
+ /// Saved kmp_task_t for task directive.
+ QualType SavedKmpTaskTQTy;
+ /// Saved kmp_task_t for taskloop-based directive.
+ QualType SavedKmpTaskloopTQTy;
/// \brief Type typedef struct kmp_depend_info {
///kmp_intptr_t base_addr;
///size_t len;
Modified: cfe/trunk/test/OpenMP/taskloop_codegen.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/test/OpenMP/taskloop_codegen.cpp?rev=315464&r1=315463&r2=315464&view=diff
==
--- cfe/trunk/test/OpenMP/taskloop_codegen.cpp (original)
+++ cfe/trunk/test/OpenMP/taskloop_codegen.cpp Wed Oct 11 08:29:40 2017
@@ -8,6 +8,10 @@
// CHECK-LABEL: @main
int main(int argc, char **argv) {
// CHECK: [[GTID:%.+]] = call i32 @__kmpc_global_thread_num(%ident_t*
[[DEFLOC:@.+]])
+// CHECK: call i8* @__kmpc_omp_task_alloc(%ident_t* [[DEFLOC]], i32 [[GTID]],
+// CHECK: call i32 @__kmpc_omp_task(%ident_t* [[DEFLOC]], i32 [[GTID]],
+#pragma omp task
+ ;
// CHECK: call void @__kmpc_taskgroup(%ident_t* [[DEFLOC]], i32 [[GTID]])
// CHECK: [[TASKV:%.+]] = call i8* @__kmpc_omp_task_alloc(%ident_t*
[[DEFLOC]], i32 [[GTID]], i32 33, i64 80, i64 1, i32 (i32, i8*)* bitcast (i32
(i32, [[TDP_TY:%.+]]*)* [[TASK1:@.+]] to i32 (i32, i8*)*))
// CHECK: [[TASK:%.+]] = bitcast i8* [[TASKV]] to [[TDP_TY]]*
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38720: [clangd] Report proper kinds for 'Keyword' and 'Snippet' completion items.
rwols accepted this revision. rwols added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D38720 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38757: [libc++] Fix PR34898 - vector iterator constructors and assign method perform push_back instead of emplace_back.
mclow.lists accepted this revision. mclow.lists added a comment. This revision is now accepted and ready to land. LGTM - thanks! https://reviews.llvm.org/D38757 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315465 - Reland "[clang-fuzzer] Allow building without coverage instrumentation."
Author: morehouse
Date: Wed Oct 11 08:51:12 2017
New Revision: 315465
URL: http://llvm.org/viewvc/llvm-project?rev=315465&view=rev
Log:
Reland "[clang-fuzzer] Allow building without coverage instrumentation."
This relands r315336 after fixing bot breakage.
Added:
cfe/trunk/tools/clang-fuzzer/DummyClangFuzzer.cpp
Modified:
cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
cfe/trunk/tools/clang-fuzzer/ClangFuzzer.cpp
Modified: cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/tools/clang-fuzzer/CMakeLists.txt?rev=315465&r1=315464&r2=315465&view=diff
==
--- cfe/trunk/tools/clang-fuzzer/CMakeLists.txt (original)
+++ cfe/trunk/tools/clang-fuzzer/CMakeLists.txt Wed Oct 11 08:51:12 2017
@@ -1,61 +1,67 @@
-if( LLVM_USE_SANITIZE_COVERAGE )
- set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD})
- set(CXX_FLAGS_NOFUZZ "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer-no-link")
+set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} FuzzMutate)
+set(CXX_FLAGS_NOFUZZ ${CMAKE_CXX_FLAGS})
+set(DUMMY_MAIN DummyClangFuzzer.cpp)
+if(LLVM_USE_SANITIZE_COVERAGE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer")
+ set(CXX_FLAGS_NOFUZZ "${CXX_FLAGS_NOFUZZ} -fsanitize=fuzzer-no-link")
+ unset(DUMMY_MAIN)
+endif()
+
+# Hack to bypass LLVM's cmake sources check and allow multiple libraries and
+# executables from this directory.
+set(LLVM_OPTIONAL_SOURCES
+ ClangFuzzer.cpp
+ DummyClangFuzzer.cpp
+ ExampleClangProtoFuzzer.cpp
+ )
+
+if(CLANG_ENABLE_PROTO_FUZZER)
+ # Create protobuf .h and .cc files, and put them in a library for use by
+ # clang-proto-fuzzer components.
+ find_package(Protobuf REQUIRED)
+ add_definitions(-DGOOGLE_PROTOBUF_NO_RTTI)
+ include_directories(${PROTOBUF_INCLUDE_DIRS})
+ include_directories(${CMAKE_CURRENT_BINARY_DIR})
+ protobuf_generate_cpp(PROTO_SRCS PROTO_HDRS cxx_proto.proto)
+ set(LLVM_OPTIONAL_SOURCES ${LLVM_OPTIONAL_SOURCES} ${PROTO_SRCS})
+ add_clang_library(clangCXXProto
+${PROTO_SRCS}
+${PROTO_HDRS}
+
+LINK_LIBS
+${PROTOBUF_LIBRARIES}
+)
- if(CLANG_ENABLE_PROTO_FUZZER)
-# Create protobuf .h and .cc files, and put them in a library for use by
-# clang-proto-fuzzer components.
-find_package(Protobuf REQUIRED)
-add_definitions(-DGOOGLE_PROTOBUF_NO_RTTI)
-include_directories(${PROTOBUF_INCLUDE_DIRS})
-include_directories(${CMAKE_CURRENT_BINARY_DIR})
-protobuf_generate_cpp(PROTO_SRCS PROTO_HDRS cxx_proto.proto)
-# Hack to bypass LLVM's cmake sources check and allow multiple libraries
and
-# executables from this directory.
-set(LLVM_OPTIONAL_SOURCES
- ClangFuzzer.cpp
- ExampleClangProtoFuzzer.cpp
- ${PROTO_SRCS}
- )
-add_clang_library(clangCXXProto
- ${PROTO_SRCS}
- ${PROTO_HDRS}
-
- LINK_LIBS
- ${PROTOBUF_LIBRARIES}
- )
-
-# Build and include libprotobuf-mutator
-include(ProtobufMutator)
-include_directories(${ProtobufMutator_INCLUDE_DIRS})
-
-# Build the protobuf->C++ translation library and driver.
-add_clang_subdirectory(proto-to-cxx)
-
-# Build the protobuf fuzzer
-add_clang_executable(clang-proto-fuzzer ExampleClangProtoFuzzer.cpp)
-target_link_libraries(clang-proto-fuzzer
- ${ProtobufMutator_LIBRARIES}
- ${PROTOBUF_LIBRARIES}
- clangCXXProto
- clangHandleCXX
- clangProtoToCXX
- )
- else()
-# Hack to bypass LLVM's cmake sources check and allow multiple libraries
and
-# executables from this directory.
-set(LLVM_OPTIONAL_SOURCES ClangFuzzer.cpp ExampleClangProtoFuzzer.cpp)
- endif()
-
- add_clang_subdirectory(handle-cxx)
-
- add_clang_executable(clang-fuzzer
-EXCLUDE_FROM_ALL
-ClangFuzzer.cpp
+ # Build and include libprotobuf-mutator
+ include(ProtobufMutator)
+ include_directories(${ProtobufMutator_INCLUDE_DIRS})
+
+ # Build the protobuf->C++ translation library and driver.
+ add_clang_subdirectory(proto-to-cxx)
+
+ # Build the protobuf fuzzer
+ add_clang_executable(clang-proto-fuzzer
+${DUMMY_MAIN}
+ExampleClangProtoFuzzer.cpp
)
- target_link_libraries(clang-fuzzer
+ target_link_libraries(clang-proto-fuzzer
+${ProtobufMutator_LIBRARIES}
+${PROTOBUF_LIBRARIES}
+clangCXXProto
clangHandleCXX
+clangProtoToCXX
)
endif()
+
+add_clang_subdirectory(handle-cxx)
+
+add_clang_executable(clang-fuzzer
+ EXCLUDE_FROM_ALL
+ ${DUMMY_MAIN}
+ ClangFuzzer.cpp
+ )
+
+target_link_libraries(clang-fuzzer
+ clangHandleCXX
+ )
Modified: cfe/trunk/tools/clang-fuzzer/ClangFuzzer.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/tools/clang-fuzzer/ClangFuzzer.cpp?rev=315465&r1=315464&r2=315465&view=diff
==
--- cfe/trunk/tools/clang-fuzzer/ClangFuzzer.cpp (original)
+++ cfe/trunk/tools/clang-fuzzer/ClangFuzzer.cpp Wed Oct 11 08:51:12 2017
@@ -17,6
r315467 - [OPENMP] Remove extra if, NFC.
Author: abataev
Date: Wed Oct 11 08:56:38 2017
New Revision: 315467
URL: http://llvm.org/viewvc/llvm-project?rev=315467&view=rev
Log:
[OPENMP] Remove extra if, NFC.
Modified:
cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp
Modified: cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp?rev=315467&r1=315466&r2=315467&view=diff
==
--- cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp (original)
+++ cfe/trunk/lib/CodeGen/CGOpenMPRuntime.cpp Wed Oct 11 08:56:38 2017
@@ -4274,7 +4274,7 @@ CGOpenMPRuntime::emitTaskInit(CodeGenFun
CGM, D.getDirectiveKind(), KmpInt32Ty, KmpRoutineEntryPtrQTy));
}
KmpTaskTQTy = SavedKmpTaskloopTQTy;
- } else if (D.getDirectiveKind() == OMPD_task) {
+ } else {
assert(D.getDirectiveKind() == OMPD_task &&
"Expected taskloop or task directive");
if (SavedKmpTaskTQTy.isNull()) {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315470 - [X86] Correct type for argument to clflushopt intrinsic.
Author: ctopper
Date: Wed Oct 11 09:06:08 2017
New Revision: 315470
URL: http://llvm.org/viewvc/llvm-project?rev=315470&view=rev
Log:
[X86] Correct type for argument to clflushopt intrinsic.
Summary: According to Intel docs this should take void const *. We had char*.
The lack of const is the main issue.
Reviewers: RKSimon, zvi, igorb
Reviewed By: igorb
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D38782
Modified:
cfe/trunk/include/clang/Basic/BuiltinsX86.def
cfe/trunk/lib/Headers/clflushoptintrin.h
Modified: cfe/trunk/include/clang/Basic/BuiltinsX86.def
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Basic/BuiltinsX86.def?rev=315470&r1=315469&r2=315470&view=diff
==
--- cfe/trunk/include/clang/Basic/BuiltinsX86.def (original)
+++ cfe/trunk/include/clang/Basic/BuiltinsX86.def Wed Oct 11 09:06:08 2017
@@ -639,7 +639,7 @@ TARGET_BUILTIN(__builtin_ia32_xsavec, "v
TARGET_BUILTIN(__builtin_ia32_xsaves, "vv*ULLi", "", "xsaves")
//CLFLUSHOPT
-TARGET_BUILTIN(__builtin_ia32_clflushopt, "vc*", "", "clflushopt")
+TARGET_BUILTIN(__builtin_ia32_clflushopt, "vvC*", "", "clflushopt")
// ADX
TARGET_BUILTIN(__builtin_ia32_addcarryx_u32, "UcUcUiUiUi*", "", "adx")
Modified: cfe/trunk/lib/Headers/clflushoptintrin.h
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Headers/clflushoptintrin.h?rev=315470&r1=315469&r2=315470&view=diff
==
--- cfe/trunk/lib/Headers/clflushoptintrin.h (original)
+++ cfe/trunk/lib/Headers/clflushoptintrin.h Wed Oct 11 09:06:08 2017
@@ -32,7 +32,7 @@
#define __DEFAULT_FN_ATTRS __attribute__((__always_inline__, __nodebug__,
__target__("clflushopt")))
static __inline__ void __DEFAULT_FN_ATTRS
-_mm_clflushopt(char * __m) {
+_mm_clflushopt(void const * __m) {
__builtin_ia32_clflushopt(__m);
}
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D37856: [refactor] add support for refactoring options
hokein added inline comments. Comment at: cfe/trunk/lib/Tooling/Refactoring/Rename/RenamingAction.cpp:113 Rules.push_back(createRefactoringActionRule( -SymbolSelectionRequirement())); +SymbolSelectionRequirement(), OptionRequirement())); return Rules; arphaman wrote: > hokein wrote: > > Thought it a bit more: it requires all of the requirements are satisfied, I > > think we need to support "one-of" option. For example, we have two option > > "-a" and "-b", only one of them is allowed to be present at the same time. > That should be straightforward enough to implement, either with a custom > class or with a built-in requirement class. I'll probably add a builtin one > when the need comes up. Thanks. I'm asking it because I plan to add the `-qualified-name` option to the clang-refactor rename subtool. And obviously "-qualified-name" can not be present with "-selection". Repository: rL LLVM https://reviews.llvm.org/D37856 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38742: [CUDA] Added __hmma_m16n16k16_* builtins to support mma instructions in sm_70
jlebar accepted this revision.
jlebar added inline comments.
This revision is now accepted and ready to land.
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9726
+ case NVPTX::BI__hmma_m16n16k16_ld_c_f16:
+case NVPTX::BI__hmma_m16n16k16_ld_c_f32:{
+Address Dst = EmitPointerWithAlignment(E->getArg(0));
weird indentation?
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9733
+ return nullptr;
+bool isColMajor = isColMajorArg.getZExtValue();
+unsigned IID;
Urg, this isn't a bool? Do we want it to be?
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9761
+
+//auto EltTy = cast(Src->getType())->getElementType();
+// Returned are 8 16x2 elements.
Accidentally left over?
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9762
+//auto EltTy = cast(Src->getType())->getElementType();
+// Returned are 8 16x2 elements.
+for (unsigned i = 0; i < NumResults; ++i) {
s/8/NumElements/?
s/16/f16/?
Maybe it would be better to write it as "Return value has type [[f16 x 2] x
NumResults]."?
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9784
+unsigned IID;
+unsigned NumResults = 8;
+// PTX Instructions (and LLVM instrinsics) are defined for slice _d_, yet
Nit, at this point it's probably better to assign NumResults in each branch,
since there are only two. clang should make sure that we don't accidentally
use it uninitialized.
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9786
+// PTX Instructions (and LLVM instrinsics) are defined for slice _d_, yet
+// for some reason nvcc buildtins are using _c_.
+switch(BuiltinID) {
s/are using/use/
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9800
+}
+Function * Intrinsic = CGM.getIntrinsic(IID);
+llvm::Type *ParamType = Intrinsic->getFunctionType()->getParamType(1);
spacing. (Probably just worth clang-formatting this and the other patch.)
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9802
+llvm::Type *ParamType = Intrinsic->getFunctionType()->getParamType(1);
+SmallVector Values;
+Values.push_back(Builder.CreatePointerCast(Dst, VoidPtrTy));
Nit, we know that there won't ever be more than 8 elements...
https://reviews.llvm.org/D38742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38742: [CUDA] Added __hmma_m16n16k16_* builtins to support mma instructions in sm_70
tra updated this revision to Diff 118636.
tra marked 6 inline comments as done.
tra added a comment.
Addressed Justin's comments.
https://reviews.llvm.org/D38742
Files:
clang/include/clang/Basic/BuiltinsNVPTX.def
clang/lib/CodeGen/CGBuiltin.cpp
clang/test/CodeGen/builtins-nvptx-sm_70.cu
Index: clang/test/CodeGen/builtins-nvptx-sm_70.cu
===
--- /dev/null
+++ clang/test/CodeGen/builtins-nvptx-sm_70.cu
@@ -0,0 +1,166 @@
+// RUN: %clang_cc1 -triple nvptx64-unknown-unknown -target-cpu sm_70 \
+// RUN:-fcuda-is-device -target-feature +ptx60 \
+// RUN:-S -emit-llvm -o - -x cuda %s \
+// RUN: | FileCheck -check-prefix=CHECK %s
+// RUN: %clang_cc1 -triple nvptx-unknown-unknown -target-cpu sm_60 \
+// RUN: -fcuda-is-device -S -o /dev/null -x cuda -verify %s
+
+#if !defined(CUDA_VERSION)
+#define __device__ __attribute__((device))
+#define __global__ __attribute__((global))
+#define __shared__ __attribute__((shared))
+#define __constant__ __attribute__((constant))
+
+typedef unsigned long long uint64_t;
+#endif
+// We have to keep all builtins that depend on particular target feature in the
+// same function, because the codegen will stop after the very first function
+// that encounters an error, so -verify will not be able to find errors in
+// subsequent functions.
+
+// CHECK-LABEL: nvvm_wmma
+__device__ void nvvm_wmma(int *src, int *dst,
+ float *fsrc, float *fdst,
+ int ldm) {
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.a.sync.row.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_a' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_a(dst, src, ldm, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.a.sync.col.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_a' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_a(dst, src+1, ldm, 1);
+
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.b.sync.row.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_b' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_b(dst, src, ldm, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.b.sync.col.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_b' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_b(dst, src+2, ldm, 1);
+
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.c.sync.row.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_c_f16' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_c_f16(dst, src, ldm, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.c.sync.col.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_c_f16' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_c_f16(dst, src, ldm, 1);
+
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.c.sync.row.m16n16k16.stride.f32
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_c_f32' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_c_f32(fdst, fsrc, ldm, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.load.c.sync.col.m16n16k16.stride.f32
+ // expected-error@+1 {{'__hmma_m16n16k16_ld_c_f32' needs target feature ptx60}}
+ __hmma_m16n16k16_ld_c_f32(fdst, fsrc, ldm, 1);
+
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.store.d.sync.row.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_st_c_f16' needs target feature ptx60}}
+ __hmma_m16n16k16_st_c_f16(dst, src, ldm, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.store.d.sync.col.m16n16k16.stride.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_st_c_f16' needs target feature ptx60}}
+ __hmma_m16n16k16_st_c_f16(dst, src, ldm, 1);
+
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.store.d.sync.row.m16n16k16.stride.f32
+ // expected-error@+1 {{'__hmma_m16n16k16_st_c_f32' needs target feature ptx60}}
+ __hmma_m16n16k16_st_c_f32(fdst, fsrc, ldm, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.store.d.sync.col.m16n16k16.stride.f32
+ // expected-error@+1 {{'__hmma_m16n16k16_st_c_f32' needs target feature ptx60}}
+ __hmma_m16n16k16_st_c_f32(fdst, fsrc, ldm, 1);
+
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.mma.sync.row.row.m16n16k16.f16.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_mma_f16f16' needs target feature ptx60}}
+ __hmma_m16n16k16_mma_f16f16(dst, src, src, src, 0, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.mma.sync.row.row.m16n16k16.f16.f16.satfinite
+ // expected-error@+1 {{'__hmma_m16n16k16_mma_f16f16' needs target feature ptx60}}
+ __hmma_m16n16k16_mma_f16f16(dst, src, src, src, 0, 1);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.mma.sync.row.col.m16n16k16.f16.f16
+ // expected-error@+1 {{'__hmma_m16n16k16_mma_f16f16' needs target feature ptx60}}
+ __hmma_m16n16k16_mma_f16f16(dst, src, src, src, 1, 0);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.mma.sync.row.col.m16n16k16.f16.f16.satfinite
+ // expected-error@+1 {{'__hmma_m16n16k16_mma_f16f16' needs target feature ptx60}}
+ __hmma_m16n16k16_mma_f16f16(dst, src, src, src, 1, 1);
+ // CHECK: call {{.*}} @llvm.nvvm.wmma.mma.sync.c
[PATCH] D38742: [CUDA] Added __hmma_m16n16k16_* builtins to support mma instructions in sm_70
tra added inline comments.
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9726
+ case NVPTX::BI__hmma_m16n16k16_ld_c_f16:
+case NVPTX::BI__hmma_m16n16k16_ld_c_f32:{
+Address Dst = EmitPointerWithAlignment(E->getArg(0));
jlebar wrote:
> weird indentation?
My emacs and clang-format keep fighting case indentation... Fixed.
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9733
+ return nullptr;
+bool isColMajor = isColMajorArg.getZExtValue();
+unsigned IID;
jlebar wrote:
> Urg, this isn't a bool? Do we want it to be?
There are no explicit declarations for these builtins in CUDA headers. Callers
of these builtins pass 0/1 and corresponding intrinsic described in [[
http://docs.nvidia.com/cuda/nvvm-ir-spec/index.html#nvvm-intrin-warp-level-matrix-ld
| NVVM-IR spec ]] shows the argument type as i32, so I've made the type
integer in clang.
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9762
+//auto EltTy = cast(Src->getType())->getElementType();
+// Returned are 8 16x2 elements.
+for (unsigned i = 0; i < NumResults; ++i) {
jlebar wrote:
> s/8/NumElements/?
> s/16/f16/?
>
> Maybe it would be better to write it as "Return value has type [[f16 x 2] x
> NumResults]."?
That was part of the leftover block. Particular types are irrelevant here. All
we care is to store whatever intrinsic call returned ([4 or 8 elements of v2f16
or float] ) in the destination array (which is int[] ).
Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9802
+llvm::Type *ParamType = Intrinsic->getFunctionType()->getParamType(1);
+SmallVector Values;
+Values.push_back(Builder.CreatePointerCast(Dst, VoidPtrTy));
jlebar wrote:
> Nit, we know that there won't ever be more than 8 elements...
We have two extra arguments -- destination buffer and stride.
https://reviews.llvm.org/D38742
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38425: [clangd] Document highlights for clangd
Nebiroth marked 3 inline comments as done. Nebiroth added inline comments. Comment at: clangd/ClangdUnit.cpp:1017 + + auto DeclLocationsFinder = std::make_shared( + llvm::errs(), SourceLocationBeg, AST.getASTContext(), ilya-biryukov wrote: > I wonder if we really need to rerun indexer twice here. Probably ok for the > first version, but we should definitely think about a faster way to get the > `Decl` under cursor than visiting the whole AST. Not sure if it's easy to do > with clang's AST, though, maybe we'll need a separate index for that. Yeah I realise it's definitely not the fastest way to go about it. Maybe there is a way to stop handling occurrences once the highlighted declaration is found. This would be the most straightforward way of improving the performance of the feature without re-writing a new AST indexer. https://reviews.llvm.org/D38425 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38110: [libunwind][MIPS]: Add support for unwinding in O32 and N64 processes.
bsdjhb updated this revision to Diff 118638. bsdjhb added a comment. - Add more soft-float checks. https://reviews.llvm.org/D38110 Files: include/__libunwind_config.h include/libunwind.h src/Registers.hpp src/UnwindCursor.hpp src/UnwindRegistersRestore.S src/UnwindRegistersSave.S src/config.h src/libunwind.cpp Index: src/libunwind.cpp === --- src/libunwind.cpp +++ src/libunwind.cpp @@ -58,8 +58,12 @@ # define REGISTER_KIND Registers_arm #elif defined(__or1k__) # define REGISTER_KIND Registers_or1k +#elif defined(__mips__) && _MIPS_SIM == _ABIO32 && defined(__mips_soft_float) +# define REGISTER_KIND Registers_mips_o32 +#elif defined(__mips__) && _MIPS_SIM == _ABI64 && defined(__mips_soft_float) +# define REGISTER_KIND Registers_mips_n64 #elif defined(__mips__) -# warning The MIPS architecture is not supported. +# warning The MIPS architecture is not supported with this ABI and environment! #else # error Architecture not supported #endif Index: src/config.h === --- src/config.h +++ src/config.h @@ -65,7 +65,7 @@ defined(__ppc__) || defined(__ppc64__) || \ (!defined(__APPLE__) && defined(__arm__)) || \ (defined(__arm64__) || defined(__aarch64__)) ||\ -(defined(__APPLE__) && defined(__mips__)) +defined(__mips__) #define _LIBUNWIND_BUILD_ZERO_COST_APIS #endif Index: src/UnwindRegistersSave.S === --- src/UnwindRegistersSave.S +++ src/UnwindRegistersSave.S @@ -87,6 +87,76 @@ xorl %eax, %eax# return UNW_ESUCCESS ret +#elif defined(__mips__) && _MIPS_SIM == _ABIO32 && defined(__mips_soft_float) + +# +# extern int unw_getcontext(unw_context_t* thread_state) +# +# On entry: +# thread_state pointer is in a0 ($4) +# +# Only save registers preserved across calls. +# +DEFINE_LIBUNWIND_FUNCTION(unw_getcontext) + .set push + .set noat + .set noreorder + .set nomacro + # s0 - s7 + sw$16, (4 * 16)($4) + sw$17, (4 * 17)($4) + sw$18, (4 * 18)($4) + sw$19, (4 * 19)($4) + sw$20, (4 * 20)($4) + sw$21, (4 * 21)($4) + sw$22, (4 * 22)($4) + sw$23, (4 * 23)($4) + # gp + sw$28, (4 * 28)($4) + # sp + sw$29, (4 * 29)($4) + # Store return address to pc + sw$31, (4 * 32)($4) + jr $31 + # fp (in delay slot) + sw$30, (4 * 30)($4) + .set pop + +#elif defined(__mips__) && _MIPS_SIM == _ABI64 && defined(__mips_soft_float) + +# +# extern int unw_getcontext(unw_context_t* thread_state) +# +# On entry: +# thread_state pointer is in a0 ($4) +# +# Only save registers preserved across calls. +# +DEFINE_LIBUNWIND_FUNCTION(unw_getcontext) + .set push + .set noat + .set noreorder + .set nomacro + # s0 - s7 + sd$16, (8 * 16)($4) + sd$17, (8 * 17)($4) + sd$18, (8 * 18)($4) + sd$19, (8 * 19)($4) + sd$20, (8 * 20)($4) + sd$21, (8 * 21)($4) + sd$22, (8 * 22)($4) + sd$23, (8 * 23)($4) + # gp + sd$28, (8 * 28)($4) + # sp + sd$29, (8 * 29)($4) + # Store return address to pc + sd$31, (8 * 32)($4) + jr $31 + # fp (in delay slot) + sd$30, (8 * 30)($4) + .set pop + # elif defined(__mips__) # Index: src/UnwindRegistersRestore.S === --- src/UnwindRegistersRestore.S +++ src/UnwindRegistersRestore.S @@ -489,6 +489,108 @@ l.jr r9 l.nop +#elif defined(__mips__) && _MIPS_SIM == _ABIO32 && defined(__mips_soft_float) + +// +// void libunwind::Registers_mips_o32::jumpto() +// +// On entry: +// thread state pointer is in a0 ($4) +// +DEFINE_LIBUNWIND_PRIVATE_FUNCTION(_ZN9libunwind18Registers_mips_o326jumptoEv) + .set push + .set noat + .set noreorder + .set nomacro + // r0 is zero + lw$1, (4 * 1)($4) + lw$2, (4 * 2)($4) + lw$3, (4 * 3)($4) + // skip a0 for now + lw$5, (4 * 5)($4) + lw$6, (4 * 6)($4) + lw$7, (4 * 7)($4) + lw$8, (4 * 8)($4) + lw$9, (4 * 9)($4) + lw$10, (4 * 10)($4) + lw$11, (4 * 11)($4) + lw$12, (4 * 12)($4) + lw$13, (4 * 13)($4) + lw$14, (4 * 14)($4) + lw$15, (4 * 15)($4) + lw$16, (4 * 16)($4) + lw$17, (4 * 17)($4) + lw$18, (4 * 18)($4) + lw$19, (4 * 19)($4) + lw$20, (4 * 20)($4) + lw$21, (4 * 21)($4) + lw$22, (4 * 22)($4) + lw$23, (4 * 23)($4) + lw$24, (4 * 24)($4) + lw$25, (4 * 25)($4) + lw$26, (4 * 26)($4) + lw$27, (4 * 27)($4) + lw$28, (4 * 28)($4) + lw$29, (4 * 29)($4) + lw$30, (4 * 30)($4) + // load new pc into ra + lw$31, (4 * 32)($4) + // jump to ra, load a0 in the delay slot + jr$31 + lw$4, (4 * 4)($4) + .set pop + +#elif defined(__mips__) && _MIPS_SIM == _ABI64 && defined(__mips_soft_float) + +// +// void libunwind::R
[PATCH] D38742: [CUDA] Added __hmma_m16n16k16_* builtins to support mma instructions in sm_70
jlebar added inline comments. Comment at: clang/lib/CodeGen/CGBuiltin.cpp:9733 + return nullptr; +bool isColMajor = isColMajorArg.getZExtValue(); +unsigned IID; tra wrote: > jlebar wrote: > > Urg, this isn't a bool? Do we want it to be? > There are no explicit declarations for these builtins in CUDA headers. > Callers of these builtins pass 0/1 and corresponding intrinsic described in > [[ > http://docs.nvidia.com/cuda/nvvm-ir-spec/index.html#nvvm-intrin-warp-level-matrix-ld > | NVVM-IR spec ]] shows the argument type as i32, so I've made the type > integer in clang. > > sgtm https://reviews.llvm.org/D38742 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38810: [Analyzer] Support bodyfarming std::call_once for libstdc++
george.karpenkov created this revision.
Herald added subscribers: szepet, kristof.beyls, xazax.hun, javed.absar,
aemerson.
https://reviews.llvm.org/D38810
Files:
lib/Analysis/BodyFarm.cpp
test/Analysis/call_once.cpp
Index: test/Analysis/call_once.cpp
===
--- test/Analysis/call_once.cpp
+++ test/Analysis/call_once.cpp
@@ -1,15 +1,24 @@
-// RUN: %clang_analyze_cc1 -std=c++11 -fblocks -analyzer-checker=core,debug.ExprInspection -w -verify %s
+// RUN: %clang_analyze_cc1 -std=c++11 -fblocks -analyzer-checker=core,debug.ExprInspection -verify %s
+// RUN: %clang_analyze_cc1 -std=c++11 -fblocks -analyzer-checker=core,debug.ExprInspection -DEMULATE_LIBSTDCPP -verify %s
void clang_analyzer_eval(bool);
// Faking std::std::call_once implementation.
namespace std {
+
+#ifndef EMULATE_LIBSTDCPP
typedef struct once_flag_s {
unsigned long __state_ = 0;
} once_flag;
+#else
+typedef struct once_flag_s {
+ int _M_once = 0;
+} once_flag;
+#endif
template
void call_once(once_flag &o, Callable func, Args... args) {};
+
} // namespace std
// Check with Lambdas.
Index: lib/Analysis/BodyFarm.cpp
===
--- lib/Analysis/BodyFarm.cpp
+++ lib/Analysis/BodyFarm.cpp
@@ -108,7 +108,7 @@
/// Returns a *first* member field of a record declaration with a given name.
/// \return an nullptr if no member with such a name exists.
- NamedDecl *findMemberField(const CXXRecordDecl *RD, StringRef Name);
+ NamedDecl *findMemberField(const RecordDecl *RD, StringRef Name);
private:
ASTContext &C;
@@ -234,7 +234,7 @@
OK_Ordinary);
}
-NamedDecl *ASTMaker::findMemberField(const CXXRecordDecl *RD, StringRef Name) {
+NamedDecl *ASTMaker::findMemberField(const RecordDecl *RD, StringRef Name) {
CXXBasePaths Paths(
/* FindAmbiguities=*/false,
@@ -328,28 +328,35 @@
const ParmVarDecl *Callback = D->getParamDecl(1);
QualType CallbackType = Callback->getType().getNonReferenceType();
QualType FlagType = Flag->getType().getNonReferenceType();
- CXXRecordDecl *FlagCXXDecl = FlagType->getAsCXXRecordDecl();
- if (!FlagCXXDecl) {
-DEBUG(llvm::dbgs() << "Flag field is not a CXX record: "
- << "unknown std::call_once implementation."
- << "Ignoring the call.\n");
+ auto *FlagRecordDecl = dyn_cast_or_null(FlagType->getAsTagDecl());
+
+ if (!FlagRecordDecl) {
+DEBUG(llvm::dbgs() << "Flag field is not a record: "
+ << "unknown std::call_once implementation, "
+ << "ignoring the call.\n");
return nullptr;
}
- // Note: here we are assuming libc++ implementation of call_once,
- // which has a struct with a field `__state_`.
- // Body farming might not work for other `call_once` implementations.
- NamedDecl *FoundDecl = M.findMemberField(FlagCXXDecl, "__state_");
- ValueDecl *FieldDecl;
- if (FoundDecl) {
-FieldDecl = dyn_cast(FoundDecl);
- } else {
+ // We initially assume libc++ implementation of call_once,
+ // where the once_flag struct has a field `__state_`.
+ NamedDecl *FoundDecl = M.findMemberField(FlagRecordDecl, "__state_");
+
+ // Otherwise, try libstdc++ implementation, with a field
+ // `_M_once`
+ if (!FoundDecl) {
DEBUG(llvm::dbgs() << "No field __state_ found on std::once_flag struct, "
- << "unable to synthesize call_once body, ignoring "
- << "the call.\n");
+ << "trying libstdc++ implementation\n");
+FoundDecl = M.findMemberField(FlagRecordDecl, "_M_once");
+ }
+
+ if (!FoundDecl) {
+DEBUG(llvm::dbgs() << "No field _M_once found on std::once flag struct: "
+ << "unknown std::call_once implementation, "
+ << "ignoring the call");
return nullptr;
}
+ ValueDecl *FieldDecl = dyn_cast(FoundDecl);
bool isLambdaCall = CallbackType->getAsCXXRecordDecl() &&
CallbackType->getAsCXXRecordDecl()->isLambda();
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38596: Implement attribute target multiversioning
erichkeane updated this revision to Diff 118648.
erichkeane added a comment.
Craig noticed a pair of ordering issues in the TargetArray list, so fixed htose.
https://reviews.llvm.org/D38596
Files:
include/clang/AST/Decl.h
include/clang/Basic/Attr.td
include/clang/Basic/DiagnosticSemaKinds.td
include/clang/Basic/TargetInfo.h
include/clang/Sema/Sema.h
lib/Basic/Targets/X86.cpp
lib/Basic/Targets/X86.h
lib/CodeGen/CGBuiltin.cpp
lib/CodeGen/CodeGenFunction.cpp
lib/CodeGen/CodeGenFunction.h
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/Sema/SemaDecl.cpp
test/CodeGen/attr-target-multiversion.c
test/CodeGenCXX/attr-target-multiversion.cpp
test/Sema/attr-target-multiversion.c
test/SemaCXX/attr-target-multiversion.cpp
Index: test/SemaCXX/attr-target-multiversion.cpp
===
--- /dev/null
+++ test/SemaCXX/attr-target-multiversion.cpp
@@ -0,0 +1,49 @@
+// RUN: %clang_cc1 -triple x86_64-linux -fsyntax-only -verify -emit-llvm-only %s
+
+struct S {
+ __attribute__((target("arch=sandybridge")))
+ void mv(){}
+ __attribute__((target("arch=ivybridge")))
+ void mv(){}
+ __attribute__((target("default")))
+ void mv(){}
+
+ // NOTE: Virtual functions aren't implement for multiversioning in GCC either,
+ // so this is a 'TBD' feature.
+ // expected-error@+2 {{function multiversioning with 'target' doesn't support virtual functions yet}}
+ __attribute__((target("arch=sandybridge")))
+ virtual void mv_2(){}
+ // expected-error@+4 {{function multiversioning with 'target' doesn't support virtual functions yet}}
+ // expected-error@+3 {{class member cannot be redeclared}}
+ // expected-note@-3 {{previous definition is here}}
+ __attribute__((target("arch=ivybridge")))
+ virtual void mv_2(){}
+ // expected-error@+3 {{class member cannot be redeclared}}
+ // expected-note@-7 {{previous definition is here}}
+ __attribute__((target("default")))
+ virtual void mv_2(){}
+};
+
+// Note: Template attribute 'target' isn't implemented in GCC either, and would
+// end up causing some nasty issues attempting it, so ensure that it still gives
+// the same errors as without the attribute.
+
+template
+__attribute__((target("arch=sandybridge")))
+void mv_temp(){}
+
+template
+__attribute__((target("arch=ivybridge")))
+//expected-error@+2 {{redefinition of}}
+//expected-note@-5{{previous definition is here}}
+void mv_temp(){}
+
+template
+__attribute__((target("default")))
+void mv_temp(){}
+
+void foo() {
+ //expected-error@+2{{no matching function for call to}}
+ //expected-note@-8{{candidate template ignored}}
+ mv_temp();
+}
Index: test/Sema/attr-target-multiversion.c
===
--- /dev/null
+++ test/Sema/attr-target-multiversion.c
@@ -0,0 +1,136 @@
+// RUN: %clang_cc1 -triple x86_64-linux -fsyntax-only -verify -emit-llvm-only %s
+// RUN: %clang_cc1 -triple x86_64-linux -fsyntax-only -verify -emit-llvm-only -DCHECK_DEFAULT %s
+
+#if defined(CHECK_DEFAULT)
+__attribute__((target("arch=sandybridge")))
+//expected-error@+1 {{function multiversioning with 'target' requires a 'default' implementation}}
+void no_default(){}
+__attribute__((target("arch=ivybridge")))
+void no_default(){}
+#else
+// Only multiversioning causes issues with redeclaration changing 'target'.
+__attribute__((target("arch=sandybridge")))
+void fine_since_no_mv();
+void fine_since_no_mv();
+
+void also_fine_since_no_mv();
+__attribute__((target("arch=sandybridge")))
+void also_fine_since_no_mv();
+
+__attribute__((target("arch=sandybridge")))
+void also_fine_since_no_mv2();
+__attribute__((target("arch=sandybridge")))
+void also_fine_since_no_mv2();
+void also_fine_since_no_mv2();
+
+__attribute__((target("arch=sandybridge")))
+void mv(){}
+__attribute__((target("arch=ivybridge")))
+void mv(){}
+__attribute__((target("default")))
+void mv(){}
+
+void redecl_causes_mv();
+__attribute__((target("arch=sandybridge")))
+void redecl_causes_mv();
+// expected-error@+3 {{function redeclaration declares a multiversioned function, but a previous declaration lacks a 'target' attribute}}
+// expected-note@-4 {{previous declaration is here}}
+__attribute__((target("arch=ivybridge")))
+void redecl_causes_mv();
+
+__attribute__((target("arch=sandybridge")))
+void redecl_causes_mv2();
+void redecl_causes_mv2();
+// expected-error@+3 {{function redeclaration declares a multiversioned function, but a previous declaration lacks a 'target' attribute}}
+// expected-note@-2 {{previous declaration is here}}
+__attribute__((target("arch=ivybridge")))
+void redecl_causes_mv2();
+
+__attribute__((target("arch=sandybridge")))
+void redecl_without_attr();
+__attribute__((target("arch=ivybridge")))
+void redecl_without_attr();
+// expected-error@+2 {{function redeclaration is missing 'target' attribute in a multiversioned function}}
+// expected-note@-4 {{previous declaration is here}}
+void redecl_without_attr(
[PATCH] D37897: [StaticAnalyzer] Fix ProgramState for static variables that are not written
danielmarjamaki added a comment. In https://reviews.llvm.org/D37897#892667, @dcoughlin wrote: > Apologies for the delay reviewing! As I noted inline, I'm pretty worried > about the performance impact of this. Is it possible to do the analysis in a > single traversal of the translation unit? I agree. I first tried more like that but ran into problems. Don't remember the details. I will try again.. however as far as I see this will mean the LoopUnroller AST matchers can't be reused unless I change them. Repository: rL LLVM https://reviews.llvm.org/D37897 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38812: [clang-fuzzer] Allow linking with any fuzzing engine.
morehouse created this revision.
Herald added a subscriber: mgorny.
Makes clang-[proto-]fuzzer compatible with flags specified by OSS-Fuzz.
https://llvm.org/pr34314
https://reviews.llvm.org/D38812
Files:
clang/tools/clang-fuzzer/CMakeLists.txt
Index: clang/tools/clang-fuzzer/CMakeLists.txt
===
--- clang/tools/clang-fuzzer/CMakeLists.txt
+++ clang/tools/clang-fuzzer/CMakeLists.txt
@@ -1,7 +1,9 @@
set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} FuzzMutate)
set(CXX_FLAGS_NOFUZZ ${CMAKE_CXX_FLAGS})
set(DUMMY_MAIN DummyClangFuzzer.cpp)
-if(LLVM_USE_SANITIZE_COVERAGE)
+if(DEFINED LIB_FUZZING_ENGINE)
+ unset(DUMMY_MAIN)
+elseif(LLVM_USE_SANITIZE_COVERAGE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer")
set(CXX_FLAGS_NOFUZZ "${CXX_FLAGS_NOFUZZ} -fsanitize=fuzzer-no-link")
unset(DUMMY_MAIN)
@@ -48,6 +50,7 @@
target_link_libraries(clang-proto-fuzzer
${ProtobufMutator_LIBRARIES}
${PROTOBUF_LIBRARIES}
+${LIB_FUZZING_ENGINE}
clangCXXProto
clangHandleCXX
clangProtoToCXX
@@ -63,5 +66,6 @@
)
target_link_libraries(clang-fuzzer
+ ${LIB_FUZZING_ENGINE}
clangHandleCXX
)
Index: clang/tools/clang-fuzzer/CMakeLists.txt
===
--- clang/tools/clang-fuzzer/CMakeLists.txt
+++ clang/tools/clang-fuzzer/CMakeLists.txt
@@ -1,7 +1,9 @@
set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} FuzzMutate)
set(CXX_FLAGS_NOFUZZ ${CMAKE_CXX_FLAGS})
set(DUMMY_MAIN DummyClangFuzzer.cpp)
-if(LLVM_USE_SANITIZE_COVERAGE)
+if(DEFINED LIB_FUZZING_ENGINE)
+ unset(DUMMY_MAIN)
+elseif(LLVM_USE_SANITIZE_COVERAGE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer")
set(CXX_FLAGS_NOFUZZ "${CXX_FLAGS_NOFUZZ} -fsanitize=fuzzer-no-link")
unset(DUMMY_MAIN)
@@ -48,6 +50,7 @@
target_link_libraries(clang-proto-fuzzer
${ProtobufMutator_LIBRARIES}
${PROTOBUF_LIBRARIES}
+${LIB_FUZZING_ENGINE}
clangCXXProto
clangHandleCXX
clangProtoToCXX
@@ -63,5 +66,6 @@
)
target_link_libraries(clang-fuzzer
+ ${LIB_FUZZING_ENGINE}
clangHandleCXX
)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38812: [clang-fuzzer] Allow linking with any fuzzing engine.
kcc accepted this revision. kcc added a comment. This revision is now accepted and ready to land. LGTM https://reviews.llvm.org/D38812 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38810: [Analyzer] Support bodyfarming std::call_once for libstdc++
dcoughlin accepted this revision. dcoughlin added a comment. This revision is now accepted and ready to land. LGTM with the dyn_cast mentioned inline changed. Comment at: lib/Analysis/BodyFarm.cpp:359 + ValueDecl *FieldDecl = dyn_cast(FoundDecl); bool isLambdaCall = CallbackType->getAsCXXRecordDecl() && Should this be a cast<>() rather than a dyn_cast<>()? Do you expect this to fail? If so then you should check the result for null. If not then the idiom is to use cast<>(). An alternative would be to change findMemberField() to return a FieldDecl. https://reviews.llvm.org/D38810 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315486 - [clang-fuzzer] Allow linking with any fuzzing engine.
Author: morehouse
Date: Wed Oct 11 11:29:24 2017
New Revision: 315486
URL: http://llvm.org/viewvc/llvm-project?rev=315486&view=rev
Log:
[clang-fuzzer] Allow linking with any fuzzing engine.
Summary:
Makes clang-[proto-]fuzzer compatible with flags specified by OSS-Fuzz.
https://llvm.org/pr34314
Reviewers: vitalybuka, kcc
Reviewed By: kcc
Subscribers: cfe-commits, mgorny
Differential Revision: https://reviews.llvm.org/D38812
Modified:
cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
Modified: cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/tools/clang-fuzzer/CMakeLists.txt?rev=315486&r1=315485&r2=315486&view=diff
==
--- cfe/trunk/tools/clang-fuzzer/CMakeLists.txt (original)
+++ cfe/trunk/tools/clang-fuzzer/CMakeLists.txt Wed Oct 11 11:29:24 2017
@@ -1,7 +1,9 @@
set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} FuzzMutate)
set(CXX_FLAGS_NOFUZZ ${CMAKE_CXX_FLAGS})
set(DUMMY_MAIN DummyClangFuzzer.cpp)
-if(LLVM_USE_SANITIZE_COVERAGE)
+if(DEFINED LIB_FUZZING_ENGINE)
+ unset(DUMMY_MAIN)
+elseif(LLVM_USE_SANITIZE_COVERAGE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer")
set(CXX_FLAGS_NOFUZZ "${CXX_FLAGS_NOFUZZ} -fsanitize=fuzzer-no-link")
unset(DUMMY_MAIN)
@@ -48,6 +50,7 @@ if(CLANG_ENABLE_PROTO_FUZZER)
target_link_libraries(clang-proto-fuzzer
${ProtobufMutator_LIBRARIES}
${PROTOBUF_LIBRARIES}
+${LIB_FUZZING_ENGINE}
clangCXXProto
clangHandleCXX
clangProtoToCXX
@@ -63,5 +66,6 @@ add_clang_executable(clang-fuzzer
)
target_link_libraries(clang-fuzzer
+ ${LIB_FUZZING_ENGINE}
clangHandleCXX
)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38812: [clang-fuzzer] Allow linking with any fuzzing engine.
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315486: [clang-fuzzer] Allow linking with any fuzzing
engine. (authored by morehouse).
Changed prior to commit:
https://reviews.llvm.org/D38812?vs=118651&id=118655#toc
Repository:
rL LLVM
https://reviews.llvm.org/D38812
Files:
cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
Index: cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
===
--- cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
+++ cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
@@ -1,7 +1,9 @@
set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} FuzzMutate)
set(CXX_FLAGS_NOFUZZ ${CMAKE_CXX_FLAGS})
set(DUMMY_MAIN DummyClangFuzzer.cpp)
-if(LLVM_USE_SANITIZE_COVERAGE)
+if(DEFINED LIB_FUZZING_ENGINE)
+ unset(DUMMY_MAIN)
+elseif(LLVM_USE_SANITIZE_COVERAGE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer")
set(CXX_FLAGS_NOFUZZ "${CXX_FLAGS_NOFUZZ} -fsanitize=fuzzer-no-link")
unset(DUMMY_MAIN)
@@ -48,6 +50,7 @@
target_link_libraries(clang-proto-fuzzer
${ProtobufMutator_LIBRARIES}
${PROTOBUF_LIBRARIES}
+${LIB_FUZZING_ENGINE}
clangCXXProto
clangHandleCXX
clangProtoToCXX
@@ -63,5 +66,6 @@
)
target_link_libraries(clang-fuzzer
+ ${LIB_FUZZING_ENGINE}
clangHandleCXX
)
Index: cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
===
--- cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
+++ cfe/trunk/tools/clang-fuzzer/CMakeLists.txt
@@ -1,7 +1,9 @@
set(LLVM_LINK_COMPONENTS ${LLVM_TARGETS_TO_BUILD} FuzzMutate)
set(CXX_FLAGS_NOFUZZ ${CMAKE_CXX_FLAGS})
set(DUMMY_MAIN DummyClangFuzzer.cpp)
-if(LLVM_USE_SANITIZE_COVERAGE)
+if(DEFINED LIB_FUZZING_ENGINE)
+ unset(DUMMY_MAIN)
+elseif(LLVM_USE_SANITIZE_COVERAGE)
set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -fsanitize=fuzzer")
set(CXX_FLAGS_NOFUZZ "${CXX_FLAGS_NOFUZZ} -fsanitize=fuzzer-no-link")
unset(DUMMY_MAIN)
@@ -48,6 +50,7 @@
target_link_libraries(clang-proto-fuzzer
${ProtobufMutator_LIBRARIES}
${PROTOBUF_LIBRARIES}
+${LIB_FUZZING_ENGINE}
clangCXXProto
clangHandleCXX
clangProtoToCXX
@@ -63,5 +66,6 @@
)
target_link_libraries(clang-fuzzer
+ ${LIB_FUZZING_ENGINE}
clangHandleCXX
)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38797: [analyzer] CStringChecker: pr34460: Admit that some casts are hard to model.
dcoughlin accepted this revision. dcoughlin added a comment. This revision is now accepted and ready to land. This looks good to me. https://reviews.llvm.org/D38797 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38764: [Analyzer] Assume const string-like globals are non-null
george.karpenkov marked 9 inline comments as done. george.karpenkov added inline comments. Comment at: lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp:22 +// Checker uses are defined in the test file: +// - test/Analysis/nonnull-string-constants.mm +// dcoughlin wrote: > We generally don't do this kind of cross-reference in comments since they > tend to get stale fast when things get moved around. There is no tool support > to keep them in sync. :( I guess I'm too spoiled by Javadoc. Comment at: lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp:53 +private: + /// Lazily initialize cache for required identifier informations. + void initIdentifierInfo(ASTContext &Ctx) const; dcoughlin wrote: > We usually put the oxygen comments on checkers on the implementation and not > the interface since they aren't API and people reading the code generally > look at the implementations. Can you move them to the implementations? OK. I assume there's no good universal solution here. https://reviews.llvm.org/D38764 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38764: [Analyzer] Assume const string-like globals are non-null
george.karpenkov updated this revision to Diff 118657.
george.karpenkov marked 2 inline comments as done.
https://reviews.llvm.org/D38764
Files:
include/clang/StaticAnalyzer/Checkers/Checkers.td
lib/StaticAnalyzer/Checkers/CMakeLists.txt
lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
test/Analysis/nonnull-string-constants.mm
Index: test/Analysis/nonnull-string-constants.mm
===
--- /dev/null
+++ test/Analysis/nonnull-string-constants.mm
@@ -0,0 +1,90 @@
+// RUN: %clang_analyze_cc1 -analyzer-checker=core,debug.ExprInspection -verify %s
+
+// Nullability of const string-like globals, testing
+// NonnullStringConstantsChecker.
+
+void clang_analyzer_eval(bool);
+
+@class NSString;
+typedef const struct __CFString *CFStringRef;
+
+// Global NSString* is non-null.
+extern NSString *const StringConstGlobal;
+void stringConstGlobal() {
+ clang_analyzer_eval(StringConstGlobal); // expected-warning{{TRUE}}
+}
+
+// The logic does not apply to local variables though.
+extern NSString *stringGetter();
+void stringConstLocal() {
+ NSString *const local = stringGetter();
+ clang_analyzer_eval(local); // expected-warning{{UNKNOWN}}
+}
+
+// Global const CFStringRef's are also assumed to be non-null.
+extern const CFStringRef CFStringConstGlobal;
+void cfStringCheckGlobal() {
+ clang_analyzer_eval(CFStringConstGlobal); // expected-warning{{TRUE}}
+}
+
+// But only "const" ones.
+extern CFStringRef CFStringNonConstGlobal;
+void cfStringCheckMutableGlobal() {
+ clang_analyzer_eval(CFStringNonConstGlobal); // expected-warning{{UNKNOWN}}
+}
+
+// char* const is also assumed to be non-null.
+extern const char *const ConstCharStarConst;
+void constCharStarCheckGlobal() {
+ clang_analyzer_eval(ConstCharStarConst); // expected-warning{{TRUE}}
+}
+
+// Pointer value can be mutable.
+extern char *const CharStarConst;
+void charStarCheckGlobal() {
+ clang_analyzer_eval(CharStarConst); // expected-warning{{TRUE}}
+}
+
+// But the pointer itself should be immutable.
+extern char *CharStar;
+void charStartCheckMutableGlobal() {
+ clang_analyzer_eval(CharStar); // expected-warning{{UNKNOWN}}
+}
+
+// Type definitions should also work across typedefs, for pointers:
+typedef char *const str;
+extern str globalStr;
+void charStarCheckTypedef() {
+ clang_analyzer_eval(globalStr); // expected-warning{{TRUE}}
+}
+
+// And for types.
+typedef NSString *const NStr;
+extern NStr globalNSString;
+void NSStringCheckTypedef() {
+ clang_analyzer_eval(globalNSString); // expected-warning{{TRUE}}
+}
+
+// Note that constness could be either inside
+// the var declaration, or in a typedef.
+typedef NSString *NStr2;
+extern const NStr2 globalNSString2;
+void NSStringCheckConstTypedef() {
+ clang_analyzer_eval(globalNSString2); // expected-warning{{TRUE}}
+}
+
+// Nested typedefs should work as well.
+typedef const CFStringRef str1;
+typedef str1 str2;
+extern str2 globalStr2;
+void testNestedTypedefs() {
+ clang_analyzer_eval(globalStr2); // expected-warning{{TRUE}}
+}
+
+// And for NSString *.
+typedef NSString *const nstr1;
+typedef nstr1 nstr2;
+extern nstr2 nglobalStr2;
+void testNestedTypedefsForNSString() {
+ clang_analyzer_eval(nglobalStr2); // expected-warning{{TRUE}}
+}
Index: lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
===
--- /dev/null
+++ lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
@@ -0,0 +1,134 @@
+//==- NonnullStringConstantsChecker.cpp ---*- C++ -*--//
+//
+// The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===--===//
+//
+// This checker adds an assumption that constant string-like globals are
+// non-null, as otherwise they generally do not convey any useful information.
+// The assumption is useful, as many framework use such global const strings,
+// and the analyzer might not be able to infer the global value if the
+// definition is in a separate translation unit.
+// The following types (and their typedef aliases) are considered string-like:
+// - `char* const`
+// - `const CFStringRef` from CoreFoundation
+// - `NSString* const` from Foundation
+//
+//===--===//
+
+#include "ClangSACheckers.h"
+#include "clang/StaticAnalyzer/Core/BugReporter/BugType.h"
+#include "clang/StaticAnalyzer/Core/Checker.h"
+#include "clang/StaticAnalyzer/Core/CheckerManager.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/ExprEngine.h"
+
+using namespace clang;
+using namespace ento;
+
+namespace {
+
+class NonnullStringConstantsChecker : public Checker {
+ mutable IdentifierInfo *N
r315488 - [Analyzer] Assume that string-like const globals are non-nil.
Author: george.karpenkov
Date: Wed Oct 11 11:39:40 2017
New Revision: 315488
URL: http://llvm.org/viewvc/llvm-project?rev=315488&view=rev
Log:
[Analyzer] Assume that string-like const globals are non-nil.
Differential Revision: https://reviews.llvm.org/D38764
Added:
cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
cfe/trunk/test/Analysis/nonnull-string-constants.mm
Modified:
cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td
cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt
Modified: cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td?rev=315488&r1=315487&r2=315488&view=diff
==
--- cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td (original)
+++ cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td Wed Oct 11
11:39:40 2017
@@ -132,6 +132,10 @@ def DynamicTypePropagation : Checker<"Dy
HelpText<"Generate dynamic type information">,
DescFile<"DynamicTypePropagation.cpp">;
+def NonnullStringConstantsChecker: Checker<"NonnilStringConstants">,
+ HelpText<"Assume that const string-like globals are non-null">,
+ DescFile<"NonilStringConstantsChecker.cpp">;
+
} // end "core"
let ParentPackage = CoreAlpha in {
Modified: cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt?rev=315488&r1=315487&r2=315488&view=diff
==
--- cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt (original)
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt Wed Oct 11 11:39:40
2017
@@ -57,6 +57,7 @@ add_clang_library(clangStaticAnalyzerChe
NSErrorChecker.cpp
NoReturnFunctionChecker.cpp
NonNullParamChecker.cpp
+ NonnullStringConstantsChecker.cpp
NullabilityChecker.cpp
NumberObjectConversionChecker.cpp
ObjCAtSyncChecker.cpp
Added: cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp?rev=315488&view=auto
==
--- cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
(added)
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp Wed
Oct 11 11:39:40 2017
@@ -0,0 +1,134 @@
+//==- NonnullStringConstantsChecker.cpp ---*- C++
-*--//
+//
+// The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===--===//
+//
+// This checker adds an assumption that constant string-like globals are
+// non-null, as otherwise they generally do not convey any useful information.
+// The assumption is useful, as many framework use such global const strings,
+// and the analyzer might not be able to infer the global value if the
+// definition is in a separate translation unit.
+// The following types (and their typedef aliases) are considered string-like:
+// - `char* const`
+// - `const CFStringRef` from CoreFoundation
+// - `NSString* const` from Foundation
+//
+//===--===//
+
+#include "ClangSACheckers.h"
+#include "clang/StaticAnalyzer/Core/BugReporter/BugType.h"
+#include "clang/StaticAnalyzer/Core/Checker.h"
+#include "clang/StaticAnalyzer/Core/CheckerManager.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/ExprEngine.h"
+
+using namespace clang;
+using namespace ento;
+
+namespace {
+
+class NonnullStringConstantsChecker : public Checker {
+ mutable IdentifierInfo *NSStringII = nullptr;
+ mutable IdentifierInfo *CFStringRefII = nullptr;
+
+public:
+ NonnullStringConstantsChecker() {}
+
+ void checkLocation(SVal l, bool isLoad, const Stmt *S,
+ CheckerContext &C) const;
+
+private:
+ void initIdentifierInfo(ASTContext &Ctx) const;
+
+ bool isGlobalConstString(SVal V) const;
+
+ bool isStringlike(QualType Ty) const;
+};
+
+} // namespace
+
+/// Lazily initialize cache for required identifier informations.
+void NonnullStringConstantsChecker::initIdentifierInfo(ASTContext &Ctx) const {
+ if (NSStringII)
+return;
+
+ NSStringII = &Ctx.Idents.get("NSString");
+ CFStringRefII = &Ctx.Idents.get("CFStringRef");
+}
+
+/// Add an assumption that const string-like globals are non-null.
+void NonnullStringConstantsChecker::checkLocation(SVal location, bool isLoad,
+ const Stmt *S,
+
[PATCH] D38764: [Analyzer] Assume const string-like globals are non-null
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315488: [Analyzer] Assume that string-like const globals are
non-nil. (authored by george.karpenkov).
Changed prior to commit:
https://reviews.llvm.org/D38764?vs=118657&id=118660#toc
Repository:
rL LLVM
https://reviews.llvm.org/D38764
Files:
cfe/trunk/include/clang/StaticAnalyzer/Checkers/Checkers.td
cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt
cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
cfe/trunk/test/Analysis/nonnull-string-constants.mm
Index: cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt
===
--- cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/CMakeLists.txt
@@ -57,6 +57,7 @@
NSErrorChecker.cpp
NoReturnFunctionChecker.cpp
NonNullParamChecker.cpp
+ NonnullStringConstantsChecker.cpp
NullabilityChecker.cpp
NumberObjectConversionChecker.cpp
ObjCAtSyncChecker.cpp
Index: cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
===
--- cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
@@ -0,0 +1,134 @@
+//==- NonnullStringConstantsChecker.cpp ---*- C++ -*--//
+//
+// The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===--===//
+//
+// This checker adds an assumption that constant string-like globals are
+// non-null, as otherwise they generally do not convey any useful information.
+// The assumption is useful, as many framework use such global const strings,
+// and the analyzer might not be able to infer the global value if the
+// definition is in a separate translation unit.
+// The following types (and their typedef aliases) are considered string-like:
+// - `char* const`
+// - `const CFStringRef` from CoreFoundation
+// - `NSString* const` from Foundation
+//
+//===--===//
+
+#include "ClangSACheckers.h"
+#include "clang/StaticAnalyzer/Core/BugReporter/BugType.h"
+#include "clang/StaticAnalyzer/Core/Checker.h"
+#include "clang/StaticAnalyzer/Core/CheckerManager.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/CheckerContext.h"
+#include "clang/StaticAnalyzer/Core/PathSensitive/ExprEngine.h"
+
+using namespace clang;
+using namespace ento;
+
+namespace {
+
+class NonnullStringConstantsChecker : public Checker {
+ mutable IdentifierInfo *NSStringII = nullptr;
+ mutable IdentifierInfo *CFStringRefII = nullptr;
+
+public:
+ NonnullStringConstantsChecker() {}
+
+ void checkLocation(SVal l, bool isLoad, const Stmt *S,
+ CheckerContext &C) const;
+
+private:
+ void initIdentifierInfo(ASTContext &Ctx) const;
+
+ bool isGlobalConstString(SVal V) const;
+
+ bool isStringlike(QualType Ty) const;
+};
+
+} // namespace
+
+/// Lazily initialize cache for required identifier informations.
+void NonnullStringConstantsChecker::initIdentifierInfo(ASTContext &Ctx) const {
+ if (NSStringII)
+return;
+
+ NSStringII = &Ctx.Idents.get("NSString");
+ CFStringRefII = &Ctx.Idents.get("CFStringRef");
+}
+
+/// Add an assumption that const string-like globals are non-null.
+void NonnullStringConstantsChecker::checkLocation(SVal location, bool isLoad,
+ const Stmt *S,
+ CheckerContext &C) const {
+ initIdentifierInfo(C.getASTContext());
+ if (!isLoad || !location.isValid())
+return;
+
+ ProgramStateRef State = C.getState();
+ SVal V = State->getSVal(location.castAs());
+
+ if (isGlobalConstString(location)) {
+Optional Constr = V.getAs();
+
+if (Constr) {
+
+ // Assume that the variable is non-null.
+ ProgramStateRef OutputState = State->assume(*Constr, true);
+ C.addTransition(OutputState);
+}
+ }
+}
+
+/// \param V loaded lvalue.
+/// \return whether {@code val} is a string-like const global.
+bool NonnullStringConstantsChecker::isGlobalConstString(SVal V) const {
+ Optional RegionVal = V.getAs();
+ if (!RegionVal)
+return false;
+ auto *Region = dyn_cast(RegionVal->getAsRegion());
+ if (!Region)
+return false;
+ const VarDecl *Decl = Region->getDecl();
+
+ if (!Decl->hasGlobalStorage())
+return false;
+
+ QualType Ty = Decl->getType();
+ bool HasConst = Ty.isConstQualified();
+ if (isStringlike(Ty) && HasConst)
+return true;
+
+ // Look through the typedefs.
+ while (auto *T = dyn_cast(Ty)) {
+Ty = T->getDecl()->getUnderlyingType();
+
+// It is sufficient for any intermediate typ
r315489 - [Analyzer] Remove dead code from testing scripts
Author: george.karpenkov
Date: Wed Oct 11 11:42:39 2017
New Revision: 315489
URL: http://llvm.org/viewvc/llvm-project?rev=315489&view=rev
Log:
[Analyzer] Remove dead code from testing scripts
Differential Revision: https://reviews.llvm.org/D38488
Modified:
cfe/trunk/utils/analyzer/SATestBuild.py
Modified: cfe/trunk/utils/analyzer/SATestBuild.py
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/utils/analyzer/SATestBuild.py?rev=315489&r1=315488&r2=315489&view=diff
==
--- cfe/trunk/utils/analyzer/SATestBuild.py (original)
+++ cfe/trunk/utils/analyzer/SATestBuild.py Wed Oct 11 11:42:39 2017
@@ -653,40 +653,6 @@ def cleanupReferenceResults(SBOutputDir)
removeLogFile(SBOutputDir)
-def updateSVN(Mode, PMapFile):
-"""
-svn delete or svn add (depending on `Mode`) all folders defined in the file
-handler `PMapFile`.
-Commit the result to SVN.
-"""
-try:
-for I in iterateOverProjects(PMapFile):
-ProjName = I[0]
-Path = os.path.join(ProjName, getSBOutputDirName(True))
-
-if Mode == "delete":
-Command = "svn delete '%s'" % (Path,)
-else:
-Command = "svn add '%s'" % (Path,)
-
-if Verbose == 1:
-print " Executing: %s" % (Command,)
-check_call(Command, shell=True)
-
-if Mode == "delete":
-CommitCommand = "svn commit -m \"[analyzer tests] Remove " \
-"reference results.\""
-else:
-CommitCommand = "svn commit -m \"[analyzer tests] Add new " \
-"reference results.\""
-if Verbose == 1:
-print " Executing: %s" % (CommitCommand,)
-check_call(CommitCommand, shell=True)
-except:
-print "Error: SVN update failed."
-sys.exit(-1)
-
-
def testProject(ID, ProjectBuildMode, IsReferenceBuild=False, Strictness=0):
"""
Test a given project.
@@ -757,26 +723,16 @@ def validateProjectFile(PMapFile):
raise Exception()
-def testAll(IsReferenceBuild=False, UpdateSVN=False, Strictness=0):
+def testAll(IsReferenceBuild=False, Strictness=0):
TestsPassed = True
with projectFileHandler() as PMapFile:
validateProjectFile(PMapFile)
-# When we are regenerating the reference results, we might need to
-# update svn. Remove reference results from SVN.
-if UpdateSVN:
-assert(IsReferenceBuild)
-updateSVN("delete", PMapFile)
-
# Test the projects.
for (ProjName, ProjBuildMode) in iterateOverProjects(PMapFile):
TestsPassed &= testProject(
ProjName, int(ProjBuildMode), IsReferenceBuild, Strictness)
-# Re-add reference results to SVN.
-if UpdateSVN:
-updateSVN("add", PMapFile)
-
if __name__ == '__main__':
# Parse command line arguments.
@@ -789,20 +745,13 @@ if __name__ == '__main__':
reference. Default is 0.')
Parser.add_argument('-r', dest='regenerate', action='store_true',
default=False, help='Regenerate reference output.')
-Parser.add_argument('-rs', dest='update_reference', action='store_true',
-default=False,
-help='Regenerate reference output and update svn.')
Args = Parser.parse_args()
IsReference = False
-UpdateSVN = False
Strictness = Args.strictness
if Args.regenerate:
IsReference = True
-elif Args.update_reference:
-IsReference = True
-UpdateSVN = True
-TestsPassed = testAll(IsReference, UpdateSVN, Strictness)
+TestsPassed = testAll(IsReference, Strictness)
if not TestsPassed:
sys.exit(-1)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D35894: [clangd] Code hover for Clangd
Nebiroth added a comment. Bumping this. I've worked on a patch for this feature that currently supports showing the declaration of whatever is being hovered on instead of the raw source code. It also has basic support to distinguish declarations in types ( class/struct, namespace, global variable, typedef, enum, namespace, etc.) so that we can have a 'kind' of a symbol like mentioned in the comments above. The step I am at right now is figuring out a way to display this information inside the box that will be displayed to the client once a response (MarkedString) is sent back by the server. I would like some feedback/suggestions on whether it's possible to "append" information to a MarkedString to be displayed on client? I was also considering extending LSP to include more than just a MarkedString for the response in order to have something a bit more flexible to use in-client. https://reviews.llvm.org/D35894 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38733: [CodeGen] Generate TBAA info along with LValue base info
rjmccall accepted this revision. rjmccall added a comment. This revision is now accepted and ready to land. Sure. https://reviews.llvm.org/D38733 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D23963: [analyzer] pr28449 - Move literal rvalue construction away from RegionStore.
dcoughlin accepted this revision. dcoughlin added a comment. This revision is now accepted and ready to land. OK. Seems reasonable! https://reviews.llvm.org/D23963 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315492 - [Analyzer] Fix introduced regression: properly check for nullable attribute.
Author: george.karpenkov
Date: Wed Oct 11 12:13:15 2017
New Revision: 315492
URL: http://llvm.org/viewvc/llvm-project?rev=315492&view=rev
Log:
[Analyzer] Fix introduced regression: properly check for nullable attribute.
Modified:
cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
Modified:
cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp?rev=315492&r1=315491&r2=315492&view=diff
==
--- cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp
(original)
+++ cfe/trunk/lib/StaticAnalyzer/Checkers/NonnullStringConstantsChecker.cpp Wed
Oct 11 12:13:15 2017
@@ -122,7 +122,8 @@ bool NonnullStringConstantsChecker::isSt
return true;
if (auto *T = dyn_cast(Ty)) {
-return T->getInterfaceDecl()->getIdentifier() == NSStringII;
+return T->getInterfaceDecl() &&
+ T->getInterfaceDecl()->getIdentifier() == NSStringII;
} else if (auto *T = dyn_cast(Ty)) {
return T->getDecl()->getIdentifier() == CFStringRefII;
}
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [clang-tools-extra] r315060 - Renaming a test to start with the name of the check based on post-commit review feedback; NFC.
On Fri, Oct 6, 2017 at 3:27 PM, Aaron Ballman via cfe-commits <
cfe-commits@lists.llvm.org> wrote:
> Author: aaronballman
> Date: Fri Oct 6 06:27:59 2017
> New Revision: 315060
>
> URL: http://llvm.org/viewvc/llvm-project?rev=315060&view=rev
> Log:
> Renaming a test to start with the name of the check based on post-commit
> review feedback; NFC.
>
> Added:
> clang-tools-extra/trunk/test/clang-tidy/google-readability-
> namespace-comments-cxx17
>
Sorry for not being clear. I didn't mean the `.cpp` extension should be
removed. This effectively disables the test, since lit only runs tests in
files with certain extensions (under clang-tools-extra/test these are '.c',
'.cpp', '.hpp', '.m', '.mm', '.cu', '.ll', '.cl', '.s', '.modularize',
'.module-map-checker', '.test').
I've just renamed the file to *.cpp and the test fails for me:
[0/1] Running the Clang extra tools' regression tests
FAIL: Clang Tools ::
clang-tidy/google-readability-namespace-comments-cxx17.cpp (102 of 674)
TEST 'Clang Tools ::
clang-tidy/google-readability-namespace-comments-cxx17.cpp' FAILED
Script:
--
/usr/bin/python2.7
/src/tools/clang/tools/extra/test/../test/clang-tidy/check_clang_tidy.py
/src/tools/clang/tools/extra/test/clang-tidy/google-readability-namespace-comments-cxx17.cpp
google-readability-namespace-comments
/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp
-- -- -std=c++17
--
Exit Code: 1
Command Output (stdout):
--
Running ['clang-tidy',
'/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp',
'-fix', '--checks=-*,google-readability-namespace-comments', '--',
'-std=c++17', '-nostdinc++']...
clang-tidy output ---
--
-- Fixes -
--
FileCheck failed:
/src/tools/clang/tools/extra/test/clang-tidy/google-readability-namespace-comments-cxx17.cpp:13:17:
error: expected string not found in input
// CHECK-FIXES: } // namespace n3
^
/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp:1:1:
note: scanning from here
// RUN: %check_clang_tidy %s google-readability-namespace-comments %t -- --
-std=c++17
^
/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp:5:7:
note: possible intended match here
// So that namespace is not empty.
^
--
Command Output (stderr):
--
Traceback (most recent call last):
File
"/src/tools/clang/tools/extra/test/../test/clang-tidy/check_clang_tidy.py",
line 140, in
main()
File
"/src/tools/clang/tools/extra/test/../test/clang-tidy/check_clang_tidy.py",
line 121, in main
stderr=subprocess.STDOUT)
File "/usr/lib/python2.7/subprocess.py", line 573, in check_output
raise CalledProcessError(retcode, cmd, output=output)
subprocess.CalledProcessError: Command '['FileCheck',
'-input-file=/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp',
'/src/tools/clang/tools/extra/test/clang-tidy/google-readability-namespace-comments-cxx17.cpp',
'-check-prefix=CHECK-FIXES', '-strict-whitespace']' returned non-zero exit
status 1
--
Testing Time: 13.07s
Failing Tests (1):
Clang Tools ::
clang-tidy/google-readability-namespace-comments-cxx17.cpp
Expected Passes: 673
Unexpected Failures: 1
FAILED: tools/clang/tools/extra/test/CMakeFiles/check-clang-tools
Did you experience anything similar? Any ideas?
- copied unchanged from r315059, clang-tools-extra/trunk/test/
> clang-tidy/google-readability-nested-namespace-comments.cpp
> Removed:
> clang-tools-extra/trunk/test/clang-tidy/google-readability-
> nested-namespace-comments.cpp
>
> Removed: clang-tools-extra/trunk/test/clang-tidy/google-readability-
> nested-namespace-comments.cpp
> URL: http://llvm.org/viewvc/llvm-project/clang-tools-extra/
> trunk/test/clang-tidy/google-readability-nested-namespace-
> comments.cpp?rev=315059&view=auto
>
> ==
> ---
> clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp
> (original)
> +++
> clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp
> (removed)
> @@ -1,15 +0,0 @@
> -// RUN: %check_clang_tidy %s google-readability-namespace-comments %t --
> -- -std=c++17
> -
> -namespace n1::n2 {
> -namespace n3 {
> - // So that namespace is not empty.
> - void f();
> -
> -// CHECK-MESSAGES: :[[@LINE+4]]:2: warning: namespace 'n3' not terminated
> with
> -// CHECK-MESSAGES: :[[@LINE-7]]:11: note: namespace 'n3' starts here
> -// CHEC
[PATCH] D38679: [libunwind] Support dwarf unwinding on i386 windows
mstorsjo added a comment. In https://reviews.llvm.org/D38679#894668, @jroelofs wrote: > LGTM Thanks, will push this one without the docs update since it's still buggy in practice on windows until https://reviews.llvm.org/D38680 gets resolved. https://reviews.llvm.org/D38679 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D36111: [COFF, ARM64] Add MS builtins __dmb, __dsb, __isb
mgrang added a comment. In https://reviews.llvm.org/D36111#894604, @mstorsjo wrote: > @mgrang, did you ever get to completing this? I've got a need for this now > (only `__dmb` so far), and if you don't have time, I can try to finish it. Sorry, I never got to complete this as I moved to other priorities. I do not think I have cycles to do this now. Please feel free to take over this patch :) https://reviews.llvm.org/D36111 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
Re: [clang-tools-extra] r315060 - Renaming a test to start with the name of the check based on post-commit review feedback; NFC.
On Wed, Oct 11, 2017 at 3:29 PM, Alexander Kornienko wrote: > On Fri, Oct 6, 2017 at 3:27 PM, Aaron Ballman via cfe-commits > wrote: >> >> Author: aaronballman >> Date: Fri Oct 6 06:27:59 2017 >> New Revision: 315060 >> >> URL: http://llvm.org/viewvc/llvm-project?rev=315060&view=rev >> Log: >> Renaming a test to start with the name of the check based on post-commit >> review feedback; NFC. >> >> Added: >> >> clang-tools-extra/trunk/test/clang-tidy/google-readability-namespace-comments-cxx17 > > > Sorry for not being clear. I didn't mean the `.cpp` extension should be > removed. This effectively disables the test, since lit only runs tests in > files with certain extensions (under clang-tools-extra/test these are '.c', > '.cpp', '.hpp', '.m', '.mm', '.cu', '.ll', '.cl', '.s', '.modularize', > '.module-map-checker', '.test'). That's entirely my fault -- I should have recognized that. Sorry for the trouble! > I've just renamed the file to *.cpp and the test fails for me: > > [0/1] Running the Clang extra tools' regression tests > FAIL: Clang Tools :: > clang-tidy/google-readability-namespace-comments-cxx17.cpp (102 of 674) > TEST 'Clang Tools :: > clang-tidy/google-readability-namespace-comments-cxx17.cpp' FAILED > > Script: > -- > /usr/bin/python2.7 > /src/tools/clang/tools/extra/test/../test/clang-tidy/check_clang_tidy.py > /src/tools/clang/tools/extra/test/clang-tidy/google-readability-namespace-comments-cxx17.cpp > google-readability-namespace-comments > /build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp > -- -- -std=c++17 > -- > Exit Code: 1 > > Command Output (stdout): > -- > Running ['clang-tidy', > '/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp', > '-fix', '--checks=-*,google-readability-namespace-comments', '--', > '-std=c++17', '-nostdinc++']... > clang-tidy output --- > > -- > -- Fixes - > > -- > FileCheck failed: > /src/tools/clang/tools/extra/test/clang-tidy/google-readability-namespace-comments-cxx17.cpp:13:17: > error: expected string not found in input > // CHECK-FIXES: } // namespace n3 > ^ > /build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp:1:1: > note: scanning from here > // RUN: %check_clang_tidy %s google-readability-namespace-comments %t -- -- > -std=c++17 > ^ > /build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp:5:7: > note: possible intended match here > // So that namespace is not empty. > ^ > > > -- > Command Output (stderr): > -- > Traceback (most recent call last): > File > "/src/tools/clang/tools/extra/test/../test/clang-tidy/check_clang_tidy.py", > line 140, in > main() > File > "/src/tools/clang/tools/extra/test/../test/clang-tidy/check_clang_tidy.py", > line 121, in main > stderr=subprocess.STDOUT) > File "/usr/lib/python2.7/subprocess.py", line 573, in check_output > raise CalledProcessError(retcode, cmd, output=output) > subprocess.CalledProcessError: Command '['FileCheck', > '-input-file=/build/tools/clang/tools/extra/test/clang-tidy/Output/google-readability-namespace-comments-cxx17.cpp.tmp.cpp', > '/src/tools/clang/tools/extra/test/clang-tidy/google-readability-namespace-comments-cxx17.cpp', > '-check-prefix=CHECK-FIXES', '-strict-whitespace']' returned non-zero exit > status 1 > > -- > > > Testing Time: 13.07s > > Failing Tests (1): > Clang Tools :: > clang-tidy/google-readability-namespace-comments-cxx17.cpp > > Expected Passes: 673 > Unexpected Failures: 1 > FAILED: tools/clang/tools/extra/test/CMakeFiles/check-clang-tools > > > Did you experience anything similar? Any ideas? I now get the same behavior that you're seeing. I'm not certain what's going on there (and don't have the time to look into it at the moment), but perhaps Jonas has ideas. ~Aaron > >> - copied unchanged from r315059, >> clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp >> Removed: >> >> clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp >> >> Removed: >> clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp >> URL: >> http://llvm.org/viewvc/llvm-project/clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp?rev=315059&view=auto >> >> == >> --- >> clang-tools-extra/trunk/test/clang-tidy/google-readability-nested-namespace-comments.cpp >> (original) >> +++ >> clang-tools-ex
[PATCH] D36111: [COFF, ARM64] Add MS builtins __dmb, __dsb, __isb
mstorsjo added a comment. In https://reviews.llvm.org/D36111#895084, @mgrang wrote: > Sorry, I never got to complete this as I moved to other priorities. I do not > think I have cycles to do this now. Please feel free to take over this patch > :) Ok - will do! Thanks for letting me know! https://reviews.llvm.org/D36111 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38816: Convert clang::LangAS to a strongly typed enum
arichardson created this revision.
Herald added subscribers: Anastasia, nhaehnle, jholewinski.
Currently both clang AST address spaces and target specific address spaces
are represented as unsigned which can lead to subtle errors if the wrong
type is passed. It is especially confusing in the CodeGen files as it is
not possible to see what kind of address space should be passed to a
function without looking at the implementation.
I originally made this change for our LLVM fork for the CHERI architecture
where we make extensive use of address spaces to differentiate between
capabilities and pointers. When merging the upstream changes I usually
run into some test failures or runtime crashes because the wrong kind of
address space is passed to a function. By converting the LangAS enum to a
C++11 we can catch these errors at compile time. Additionally, it is now
obvious from the function signature which kind of address space it expects.
I found the following errors while writing this patch:
- ItaniumRecordLayoutBuilder::LayoutField was passing a clang AST address space
to TargetInfo::getPointer{Width,Align}()
- TypePrinter::printAttributedAfter() was printing the numeric value of the
clang AST address space instead of the target address space
- initializeForBlockHeader() in CGBlocks.cpp was passing LangAS::opencl_generic
to TargetInfo::getPointer{Width,Align}()
- CodeGenFunction::EmitBlockLiteral() was passing a AST address space to
TargetInfo::getPointerWidth()
- clang_getAddressSpace() was returning either a LangAS or a target address
address space. I made it consistently return a target address space
- CGOpenMPRuntimeNVPTX::translateParameter() passed a target address space to
Qualifiers::addAddressSpace()
- CGOpenMPRuntimeNVPTX::getParameterAddress() was using
llvm::Type::getPointerTo() with a AST address space
Other than this the patch should not cause any functional changes.
Repository:
rL LLVM
https://reviews.llvm.org/D38816
Files:
include/clang/AST/ASTContext.h
include/clang/AST/Type.h
include/clang/Basic/AddressSpaces.h
include/clang/Basic/TargetInfo.h
lib/AST/ASTContext.cpp
lib/AST/ItaniumMangle.cpp
lib/AST/RecordLayoutBuilder.cpp
lib/AST/TypePrinter.cpp
lib/Basic/TargetInfo.cpp
lib/Basic/Targets/AMDGPU.cpp
lib/Basic/Targets/AMDGPU.h
lib/CodeGen/CGBlocks.cpp
lib/CodeGen/CGDecl.cpp
lib/CodeGen/CGExprConstant.cpp
lib/CodeGen/CGOpenMPRuntimeNVPTX.cpp
lib/CodeGen/CGValue.h
lib/CodeGen/CodeGenFunction.cpp
lib/CodeGen/CodeGenModule.cpp
lib/CodeGen/CodeGenModule.h
lib/CodeGen/CodeGenTypeCache.h
lib/CodeGen/ConstantEmitter.h
lib/CodeGen/TargetInfo.cpp
lib/CodeGen/TargetInfo.h
lib/Sema/SemaChecking.cpp
lib/Sema/SemaDecl.cpp
lib/Sema/SemaDeclAttr.cpp
lib/Sema/SemaDeclObjC.cpp
lib/Sema/SemaExpr.cpp
lib/Sema/SemaExprCXX.cpp
lib/Sema/SemaTemplateDeduction.cpp
lib/Sema/SemaType.cpp
tools/libclang/CXType.cpp
Index: tools/libclang/CXType.cpp
===
--- tools/libclang/CXType.cpp
+++ tools/libclang/CXType.cpp
@@ -398,12 +398,8 @@
unsigned clang_getAddressSpace(CXType CT) {
QualType T = GetQualType(CT);
-
- // For non language-specific address space, use separate helper function.
- if (T.getAddressSpace() >= LangAS::FirstTargetAddressSpace) {
-return T.getQualifiers().getAddressSpaceAttributePrintValue();
- }
- return T.getAddressSpace();
+ ASTContext &Ctx = cxtu::getASTUnit(GetTU(CT))->getASTContext();
+ return Ctx.getTargetAddressSpace(T);
}
CXString clang_getTypedefName(CXType CT) {
Index: lib/Sema/SemaType.cpp
===
--- lib/Sema/SemaType.cpp
+++ lib/Sema/SemaType.cpp
@@ -5632,7 +5632,7 @@
// If this type is already address space qualified, reject it.
// ISO/IEC TR 18037 S5.3 (amending C99 6.7.3): "No type shall be qualified
// by qualifiers for two or more different address spaces."
-if (T.getAddressSpace()) {
+if (T.getAddressSpace() != LangAS::Default) {
Diag(AttrLoc, diag::err_attribute_address_multiple_qualifiers);
return QualType();
}
@@ -5656,15 +5656,16 @@
}
llvm::APSInt max(addrSpace.getBitWidth());
-max = Qualifiers::MaxAddressSpace - LangAS::FirstTargetAddressSpace;
+max =
+Qualifiers::MaxAddressSpace - (unsigned)LangAS::FirstTargetAddressSpace;
if (addrSpace > max) {
Diag(AttrLoc, diag::err_attribute_address_space_too_high)
<< (unsigned)max.getZExtValue() << AddrSpace->getSourceRange();
return QualType();
}
-unsigned ASIdx = static_cast(addrSpace.getZExtValue()) +
- LangAS::FirstTargetAddressSpace;
+LangAS ASIdx = LanguageAS::fromTargetAS(
+static_cast(addrSpace.getZExtValue()));
return Context.getAddrSpaceQualType(T, ASIdx);
}
@@ -5690,7 +5691,7 @@
// If this type is already address space qualified, reject it.
// ISO/IEC TR 18
[libunwind] r315498 - Support DWARF unwinding on i386 windows
Author: mstorsjo
Date: Wed Oct 11 13:06:18 2017
New Revision: 315498
URL: http://llvm.org/viewvc/llvm-project?rev=315498&view=rev
Log:
Support DWARF unwinding on i386 windows
In practice, with code built with clang, there are still unresolved
issues with DW_CFA_GNU_args_size though.
Differential Revision: https://reviews.llvm.org/D38679
Modified:
libunwind/trunk/src/AddressSpace.hpp
libunwind/trunk/src/UnwindRegistersRestore.S
libunwind/trunk/src/assembly.h
libunwind/trunk/src/config.h
Modified: libunwind/trunk/src/AddressSpace.hpp
URL:
http://llvm.org/viewvc/llvm-project/libunwind/trunk/src/AddressSpace.hpp?rev=315498&r1=315497&r2=315498&view=diff
==
--- libunwind/trunk/src/AddressSpace.hpp (original)
+++ libunwind/trunk/src/AddressSpace.hpp Wed Oct 11 13:06:18 2017
@@ -18,7 +18,7 @@
#include
#include
-#ifndef _LIBUNWIND_IS_BAREMETAL
+#if !defined(_LIBUNWIND_IS_BAREMETAL) && !defined(_WIN32)
#include
#endif
@@ -97,7 +97,12 @@ extern char __exidx_end;
// independent ELF header traversal is not provided by on some
// systems (e.g., FreeBSD). On these systems the data structures are
// just called Elf_XXX. Define ElfW() locally.
+#ifndef _WIN32
#include
+#else
+#include
+#include
+#endif
#if !defined(ElfW)
#define ElfW(type) Elf_##type
#endif
@@ -356,6 +361,43 @@ inline bool LocalAddressSpace::findUnwin
info.arm_section, info.arm_section_length);
if (info.arm_section && info.arm_section_length)
return true;
+#elif defined(_LIBUNWIND_SUPPORT_DWARF_UNWIND) && defined(_WIN32)
+ HMODULE mods[1024];
+ HANDLE process = GetCurrentProcess();
+ DWORD needed;
+
+ if (!EnumProcessModules(process, mods, sizeof(mods), &needed))
+return false;
+
+ for (unsigned i = 0; i < (needed / sizeof(HMODULE)); i++) {
+PIMAGE_DOS_HEADER pidh = (PIMAGE_DOS_HEADER)mods[i];
+PIMAGE_NT_HEADERS pinh = (PIMAGE_NT_HEADERS)((BYTE *)pidh +
pidh->e_lfanew);
+PIMAGE_FILE_HEADER pifh = (PIMAGE_FILE_HEADER)&pinh->FileHeader;
+PIMAGE_SECTION_HEADER pish = IMAGE_FIRST_SECTION(pinh);
+bool found_obj = false;
+bool found_hdr = false;
+
+info.dso_base = (uintptr_t)mods[i];
+for (unsigned j = 0; j < pifh->NumberOfSections; j++, pish++) {
+ uintptr_t begin = pish->VirtualAddress + (uintptr_t)mods[i];
+ uintptr_t end = begin + pish->Misc.VirtualSize;
+ if (!strncmp((const char *)pish->Name, ".text",
+ IMAGE_SIZEOF_SHORT_NAME)) {
+if (targetAddr >= begin && targetAddr < end)
+ found_obj = true;
+ } else if (!strncmp((const char *)pish->Name, ".eh_frame",
+ IMAGE_SIZEOF_SHORT_NAME)) {
+// FIXME: This section name actually is truncated, ideally we
+// should locate and check the full long name instead.
+info.dwarf_section = begin;
+info.dwarf_section_length = pish->Misc.VirtualSize;
+found_hdr = true;
+ }
+ if (found_obj && found_hdr)
+return true;
+}
+ }
+ return false;
#elif defined(_LIBUNWIND_ARM_EHABI) || defined(_LIBUNWIND_SUPPORT_DWARF_UNWIND)
struct dl_iterate_cb_data {
LocalAddressSpace *addressSpace;
@@ -478,7 +520,7 @@ inline bool LocalAddressSpace::findOther
inline bool LocalAddressSpace::findFunctionName(pint_t addr, char *buf,
size_t bufLen,
unw_word_t *offset) {
-#ifndef _LIBUNWIND_IS_BAREMETAL
+#if !defined(_LIBUNWIND_IS_BAREMETAL) && !defined(_WIN32)
Dl_info dyldInfo;
if (dladdr((void *)addr, &dyldInfo)) {
if (dyldInfo.dli_sname != NULL) {
Modified: libunwind/trunk/src/UnwindRegistersRestore.S
URL:
http://llvm.org/viewvc/llvm-project/libunwind/trunk/src/UnwindRegistersRestore.S?rev=315498&r1=315497&r2=315498&view=diff
==
--- libunwind/trunk/src/UnwindRegistersRestore.S (original)
+++ libunwind/trunk/src/UnwindRegistersRestore.S Wed Oct 11 13:06:18 2017
@@ -18,6 +18,10 @@ DEFINE_LIBUNWIND_PRIVATE_FUNCTION(_ZN9li
#
# void libunwind::Registers_x86::jumpto()
#
+#if defined(_WIN32)
+# On windows, the 'this' pointer is passed in ecx instead of on the stack
+ movl %ecx, %eax
+#else
# On entry:
# + +
# +---+
@@ -27,6 +31,7 @@ DEFINE_LIBUNWIND_PRIVATE_FUNCTION(_ZN9li
# +---+ <-- SP
# + +
movl 4(%esp), %eax
+#endif
# set up eax and ret on new stack location
movl 28(%eax), %edx # edx holds new stack pointer
subl $8,%edx
Modified: libunwind/trunk/src/assembly.h
URL:
http://llvm.org/viewvc/llvm-project/libunwind/trunk/src/assembly.h?rev=315498&r1=315497&r2=315498&view=diff
==
--- libunwind/trunk/src/assembly.h (original)
+++ libunwi
[PATCH] D38679: [libunwind] Support dwarf unwinding on i386 windows
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315498: Support DWARF unwinding on i386 windows (authored by
mstorsjo).
Changed prior to commit:
https://reviews.llvm.org/D38679?vs=118263&id=118673#toc
Repository:
rL LLVM
https://reviews.llvm.org/D38679
Files:
libunwind/trunk/src/AddressSpace.hpp
libunwind/trunk/src/UnwindRegistersRestore.S
libunwind/trunk/src/assembly.h
libunwind/trunk/src/config.h
Index: libunwind/trunk/src/UnwindRegistersRestore.S
===
--- libunwind/trunk/src/UnwindRegistersRestore.S
+++ libunwind/trunk/src/UnwindRegistersRestore.S
@@ -18,6 +18,10 @@
#
# void libunwind::Registers_x86::jumpto()
#
+#if defined(_WIN32)
+# On windows, the 'this' pointer is passed in ecx instead of on the stack
+ movl %ecx, %eax
+#else
# On entry:
# + +
# +---+
@@ -27,6 +31,7 @@
# +---+ <-- SP
# + +
movl 4(%esp), %eax
+#endif
# set up eax and ret on new stack location
movl 28(%eax), %edx # edx holds new stack pointer
subl $8,%edx
Index: libunwind/trunk/src/config.h
===
--- libunwind/trunk/src/config.h
+++ libunwind/trunk/src/config.h
@@ -37,6 +37,8 @@
#define _LIBUNWIND_SUPPORT_COMPACT_UNWIND
#define _LIBUNWIND_SUPPORT_DWARF_UNWIND 1
#endif
+#elif defined(_WIN32)
+ #define _LIBUNWIND_SUPPORT_DWARF_UNWIND 1
#else
#if defined(__ARM_DWARF_EH__) || !defined(__arm__)
#define _LIBUNWIND_SUPPORT_DWARF_UNWIND 1
Index: libunwind/trunk/src/AddressSpace.hpp
===
--- libunwind/trunk/src/AddressSpace.hpp
+++ libunwind/trunk/src/AddressSpace.hpp
@@ -18,7 +18,7 @@
#include
#include
-#ifndef _LIBUNWIND_IS_BAREMETAL
+#if !defined(_LIBUNWIND_IS_BAREMETAL) && !defined(_WIN32)
#include
#endif
@@ -97,7 +97,12 @@
// independent ELF header traversal is not provided by on some
// systems (e.g., FreeBSD). On these systems the data structures are
// just called Elf_XXX. Define ElfW() locally.
+#ifndef _WIN32
#include
+#else
+#include
+#include
+#endif
#if !defined(ElfW)
#define ElfW(type) Elf_##type
#endif
@@ -356,6 +361,43 @@
info.arm_section, info.arm_section_length);
if (info.arm_section && info.arm_section_length)
return true;
+#elif defined(_LIBUNWIND_SUPPORT_DWARF_UNWIND) && defined(_WIN32)
+ HMODULE mods[1024];
+ HANDLE process = GetCurrentProcess();
+ DWORD needed;
+
+ if (!EnumProcessModules(process, mods, sizeof(mods), &needed))
+return false;
+
+ for (unsigned i = 0; i < (needed / sizeof(HMODULE)); i++) {
+PIMAGE_DOS_HEADER pidh = (PIMAGE_DOS_HEADER)mods[i];
+PIMAGE_NT_HEADERS pinh = (PIMAGE_NT_HEADERS)((BYTE *)pidh + pidh->e_lfanew);
+PIMAGE_FILE_HEADER pifh = (PIMAGE_FILE_HEADER)&pinh->FileHeader;
+PIMAGE_SECTION_HEADER pish = IMAGE_FIRST_SECTION(pinh);
+bool found_obj = false;
+bool found_hdr = false;
+
+info.dso_base = (uintptr_t)mods[i];
+for (unsigned j = 0; j < pifh->NumberOfSections; j++, pish++) {
+ uintptr_t begin = pish->VirtualAddress + (uintptr_t)mods[i];
+ uintptr_t end = begin + pish->Misc.VirtualSize;
+ if (!strncmp((const char *)pish->Name, ".text",
+ IMAGE_SIZEOF_SHORT_NAME)) {
+if (targetAddr >= begin && targetAddr < end)
+ found_obj = true;
+ } else if (!strncmp((const char *)pish->Name, ".eh_frame",
+ IMAGE_SIZEOF_SHORT_NAME)) {
+// FIXME: This section name actually is truncated, ideally we
+// should locate and check the full long name instead.
+info.dwarf_section = begin;
+info.dwarf_section_length = pish->Misc.VirtualSize;
+found_hdr = true;
+ }
+ if (found_obj && found_hdr)
+return true;
+}
+ }
+ return false;
#elif defined(_LIBUNWIND_ARM_EHABI) || defined(_LIBUNWIND_SUPPORT_DWARF_UNWIND)
struct dl_iterate_cb_data {
LocalAddressSpace *addressSpace;
@@ -478,7 +520,7 @@
inline bool LocalAddressSpace::findFunctionName(pint_t addr, char *buf,
size_t bufLen,
unw_word_t *offset) {
-#ifndef _LIBUNWIND_IS_BAREMETAL
+#if !defined(_LIBUNWIND_IS_BAREMETAL) && !defined(_WIN32)
Dl_info dyldInfo;
if (dladdr((void *)addr, &dyldInfo)) {
if (dyldInfo.dli_sname != NULL) {
Index: libunwind/trunk/src/assembly.h
===
--- libunwind/trunk/src/assembly.h
+++ libunwind/trunk/src/assembly.h
@@ -26,6 +26,14 @@
#if defined(__APPLE__)
#define HIDDEN_DIRECTIVE .private_extern
+#elif defined(_WIN32)
+// In the COFF object file format, there's no attributes for a global,
+// non-static symbol to make it somehow hidden. S
[PATCH] D38134: [OpenCL] Emit enqueued block as kernel
yaxunl updated this revision to Diff 118677.
yaxunl marked 2 inline comments as done.
yaxunl added a comment.
Revised by Anastasia's comments. Get block invoke function by API instead of
iterate through IR's. Pass the block kernel directly to `__enqueu_kernel
functions`.
https://reviews.llvm.org/D38134
Files:
lib/CodeGen/CGBlocks.cpp
lib/CodeGen/CGBuiltin.cpp
lib/CodeGen/CGOpenCLRuntime.cpp
lib/CodeGen/CGOpenCLRuntime.h
lib/CodeGen/CodeGenFunction.h
lib/CodeGen/CodeGenTypes.h
lib/CodeGen/TargetInfo.cpp
lib/CodeGen/TargetInfo.h
test/CodeGenOpenCL/amdgpu-enqueue-kernel.cl
test/CodeGenOpenCL/cl20-device-side-enqueue.cl
Index: test/CodeGenOpenCL/cl20-device-side-enqueue.cl
===
--- test/CodeGenOpenCL/cl20-device-side-enqueue.cl
+++ test/CodeGenOpenCL/cl20-device-side-enqueue.cl
@@ -6,10 +6,30 @@
typedef void (^bl_t)(local void *);
typedef struct {int a;} ndrange_t;
-// N.B. The check here only exists to set BL_GLOBAL
-// COMMON: @block_G = addrspace(1) constant void (i8 addrspace(3)*) addrspace(4)* addrspacecast (void (i8 addrspace(3)*) addrspace(1)* bitcast ({ i32, i32, i8 addrspace(4)* } addrspace(1)* [[BL_GLOBAL:@__block_literal_global(\.[0-9]+)?]] to void (i8 addrspace(3)*) addrspace(1)*) to void (i8 addrspace(3)*) addrspace(4)*)
+// COMMON: %struct.__opencl_block_literal_generic = type { i32, i32, i8 addrspace(4)* }
+
+// For a block global variable, first emit the block literal as a global variable, then emit the block variable itself.
+// COMMON: [[BL_GLOBAL:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* [[INV_G:@[^ ]+]] to i8*) to i8 addrspace(4)*) }
+// COMMON: @block_G = addrspace(1) constant void (i8 addrspace(3)*) addrspace(4)* addrspacecast (void (i8 addrspace(3)*) addrspace(1)* bitcast ({ i32, i32, i8 addrspace(4)* } addrspace(1)* [[BL_GLOBAL]] to void (i8 addrspace(3)*) addrspace(1)*) to void (i8 addrspace(3)*) addrspace(4)*)
+
+// For anonymous blocks without captures, emit block literals as global variable.
+// COMMON: [[BLG1:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG2:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG3:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG4:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG5:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG6:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*, i8 addrspace(3)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG7:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* {{@[^ ]+}} to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG8:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*)* [[INVG8:@[^ ]+]] to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG9:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*, i8 addrspace(3)*)* [[INVG9:@[^ ]+]] to i8*) to i8 addrspace(4)*) }
+// COMMON: [[BLG10:@__block_literal_global[^ ]*]] = internal addrspace(1) constant { i32, i32, i8 addrspace(4)* } { i32 {{[0-9]+}}, i32 {{[0-9]+}}, i8 addrspace(4)* addrspacecast (i8* bitcast (void (i8 addrspace(4)*)* {{@[^ ]+}} to i8*) to i8 add
[PATCH] D38819: [libunwind] Add support for dwarf unwinding on windows on x86_64
mstorsjo created this revision.
Herald added a subscriber: aprantl.
Clang doesn't currently support building for windows/x86_64 with dwarf by
setting command line parameters, but if manually modified to use dwarf, we can
make libunwind work in this configuration as well.
Even if support for this is deemed pointless, I guess one could omit the
assembly changes, but the fixes for the generic code probably are worthwhile at
least.
https://reviews.llvm.org/D38819
Files:
docs/index.rst
include/unwind.h
src/AddressSpace.hpp
src/UnwindLevel1.c
src/UnwindRegistersRestore.S
src/UnwindRegistersSave.S
Index: src/UnwindRegistersSave.S
===
--- src/UnwindRegistersSave.S
+++ src/UnwindRegistersSave.S
@@ -63,6 +63,27 @@
# thread_state pointer is in rdi
#
DEFINE_LIBUNWIND_FUNCTION(unw_getcontext)
+#if defined(_WIN32)
+ movq %rax, (%rcx)
+ movq %rbx, 8(%rcx)
+ movq %rcx, 16(%rcx)
+ movq %rdx, 24(%rcx)
+ movq %rdi, 32(%rcx)
+ movq %rsi, 40(%rcx)
+ movq %rbp, 48(%rcx)
+ movq %rsp, 56(%rcx)
+ addq $8, 56(%rcx)
+ movq %r8, 64(%rcx)
+ movq %r9, 72(%rcx)
+ movq %r10, 80(%rcx)
+ movq %r11, 88(%rcx)
+ movq %r12, 96(%rcx)
+ movq %r13,104(%rcx)
+ movq %r14,112(%rcx)
+ movq %r15,120(%rcx)
+ movq (%rsp),%rdx
+ movq %rdx,128(%rcx) # store return address as rip
+#else
movq %rax, (%rdi)
movq %rbx, 8(%rdi)
movq %rcx, 16(%rdi)
@@ -82,6 +103,7 @@
movq %r15,120(%rdi)
movq (%rsp),%rsi
movq %rsi,128(%rdi) # store return address as rip
+#endif
# skip rflags
# skip cs
# skip fs
Index: src/UnwindRegistersRestore.S
===
--- src/UnwindRegistersRestore.S
+++ src/UnwindRegistersRestore.S
@@ -65,6 +65,40 @@
#
# void libunwind::Registers_x86_64::jumpto()
#
+#if defined(_WIN32)
+# On entry, thread_state pointer is in rcx
+
+ movq 56(%rcx), %rax # rax holds new stack pointer
+ subq $16, %rax
+ movq %rax, 56(%rcx)
+ movq 16(%rcx), %rdx # store new rcx on new stack
+ movq %rdx, 0(%rax)
+ movq 128(%rcx), %rdx # store new rip on new stack
+ movq %rdx, 8(%rax)
+ # restore all registers
+ movq0(%rcx), %rax
+ movq8(%rcx), %rbx
+ # restore rcx later
+ movq 24(%rcx), %rdx
+ movq 32(%rcx), %rdi
+ movq 40(%rcx), %rsi
+ movq 48(%rcx), %rbp
+ # restore rsp later
+ movq 64(%rcx), %r8
+ movq 72(%rcx), %r9
+ movq 80(%rcx), %r10
+ movq 88(%rcx), %r11
+ movq 96(%rcx), %r12
+ movq 104(%rcx), %r13
+ movq 112(%rcx), %r14
+ movq 120(%rcx), %r15
+ # skip rflags
+ # skip cs
+ # skip fs
+ # skip gs
+ movq 56(%rcx), %rsp # cut back rsp to new location
+ pop%rcx # rcx was saved here earlier
+#else
# On entry, thread_state pointer is in rdi
movq 56(%rdi), %rax # rax holds new stack pointer
@@ -97,6 +131,7 @@
# skip gs
movq 56(%rdi), %rsp # cut back rsp to new location
pop%rdi # rdi was saved here earlier
+#endif
ret# rip was saved here
Index: src/UnwindLevel1.c
===
--- src/UnwindLevel1.c
+++ src/UnwindLevel1.c
@@ -86,7 +86,7 @@
// this frame.
if (frameInfo.handler != 0) {
__personality_routine p =
- (__personality_routine)(long)(frameInfo.handler);
+ (__personality_routine)(uintptr_t)(frameInfo.handler);
_LIBUNWIND_TRACE_UNWINDING(
"unwind_phase1(ex_ojb=%p): calling personality function %p",
(void *)exception_object, (void *)(uintptr_t)p);
@@ -181,7 +181,7 @@
// If there is a personality routine, tell it we are unwinding.
if (frameInfo.handler != 0) {
__personality_routine p =
- (__personality_routine)(long)(frameInfo.handler);
+ (__personality_routine)(uintptr_t)(frameInfo.handler);
_Unwind_Action action = _UA_CLEANUP_PHASE;
if (sp == exception_object->private_2) {
// Tell personality this was the frame it marked in phase 1.
Index: src/AddressSpace.hpp
===
--- src/AddressSpace.hpp
+++ src/AddressSpace.hpp
@@ -142,7 +142,7 @@
/// making local unwinds fast.
class __attribute__((visibility("hidden"))) LocalAddressSpace {
public:
-#ifdef __LP64__
+#if defined(__LP64__) || defined(_WIN64)
typedef uint64_t pint_t;
typedef int64_t sint_t;
#else
@@ -194,7 +194,7 @@
};
inline uintptr_t LocalAddressSpace::getP(pint_t addr) {
-#ifdef __LP64__
+#if defined(__LP64__) || defined(_WIN64)
return get64(addr);
#else
return get32(addr);
Index: include/unwind.h
===
--- include/unwind.h
+++ include/unwind.h
@@ -122,7 +122,7 @@
_Unwind_Exception *exc);
uintptr_t private_1; // non-zero means forced unwind
uintptr_t private_2; // holds sp that phase1 found for phase2 to use
-#ifndef __LP
[PATCH] D35082: [OpenCL] Add LangAS::opencl_private to represent private address space in AST
yaxunl added a comment. If there is no other issues. May I commit this patch now? Thanks. https://reviews.llvm.org/D35082 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38820: [CGExprScalar] Add missing types in function GetIntrinsic
Carrot created this revision.
In function GetIntrinsic, not all types are covered. Types double and long long
are missed, type long is wrongly treated same as int, it should be same as long
long. These problems cause compiler crashes when compiling code in PR31161.
This patch fixed the problem.
https://reviews.llvm.org/D38820
Files:
lib/CodeGen/CGExprScalar.cpp
test/CodeGen/ppc-vector-compare.cc
Index: test/CodeGen/ppc-vector-compare.cc
===
--- test/CodeGen/ppc-vector-compare.cc
+++ test/CodeGen/ppc-vector-compare.cc
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -target-feature +altivec -triple powerpc64-unknown-unknown
-emit-llvm %s \
+// RUN: %clang_cc1 -target-feature +vsx -triple powerpc64-unknown-unknown
-emit-llvm %s \
// RUN:-o - | FileCheck %s
#include
@@ -9,3 +9,26 @@
return v1 == v2;
}
+// CHECK-LABEL: @_Z5test2Dv2_mS_Dv2_lS0_Dv2_yS1_Dv2_xS2_Dv2_dS3_
+bool test2(vector unsigned long v1, vector unsigned long v2,
+ vector long v3, vector long v4,
+ vector unsigned long long v5, vector unsigned long long v6,
+ vector long long v7, vector long long v8,
+ vector double v9, vector double v10) {
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ bool res = v1 == v2;
+
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ res |= v3 == v4;
+
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ res |= v5 == v6;
+
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ res |= v7 == v8;
+
+ // CHECK: @llvm.ppc.vsx.xvcmpeqdp.p
+ res |= v9 == v10;
+ return res;
+}
+
Index: lib/CodeGen/CGExprScalar.cpp
===
--- lib/CodeGen/CGExprScalar.cpp
+++ lib/CodeGen/CGExprScalar.cpp
@@ -3120,16 +3120,25 @@
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequh_p :
llvm::Intrinsic::ppc_altivec_vcmpgtsh_p;
case BuiltinType::UInt:
- case BuiltinType::ULong:
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequw_p :
llvm::Intrinsic::ppc_altivec_vcmpgtuw_p;
case BuiltinType::Int:
- case BuiltinType::Long:
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequw_p :
llvm::Intrinsic::ppc_altivec_vcmpgtsw_p;
+ case BuiltinType::ULong:
+ case BuiltinType::ULongLong:
+return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequd_p :
+llvm::Intrinsic::ppc_altivec_vcmpgtud_p;
+ case BuiltinType::Long:
+ case BuiltinType::LongLong:
+return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequd_p :
+llvm::Intrinsic::ppc_altivec_vcmpgtsd_p;
case BuiltinType::Float:
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpeqfp_p :
llvm::Intrinsic::ppc_altivec_vcmpgtfp_p;
+ case BuiltinType::Double:
+return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_vsx_xvcmpeqdp_p :
+llvm::Intrinsic::ppc_vsx_xvcmpgtdp_p;
}
}
Index: test/CodeGen/ppc-vector-compare.cc
===
--- test/CodeGen/ppc-vector-compare.cc
+++ test/CodeGen/ppc-vector-compare.cc
@@ -1,4 +1,4 @@
-// RUN: %clang_cc1 -target-feature +altivec -triple powerpc64-unknown-unknown -emit-llvm %s \
+// RUN: %clang_cc1 -target-feature +vsx -triple powerpc64-unknown-unknown -emit-llvm %s \
// RUN:-o - | FileCheck %s
#include
@@ -9,3 +9,26 @@
return v1 == v2;
}
+// CHECK-LABEL: @_Z5test2Dv2_mS_Dv2_lS0_Dv2_yS1_Dv2_xS2_Dv2_dS3_
+bool test2(vector unsigned long v1, vector unsigned long v2,
+ vector long v3, vector long v4,
+ vector unsigned long long v5, vector unsigned long long v6,
+ vector long long v7, vector long long v8,
+ vector double v9, vector double v10) {
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ bool res = v1 == v2;
+
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ res |= v3 == v4;
+
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ res |= v5 == v6;
+
+ // CHECK: @llvm.ppc.altivec.vcmpequd.p
+ res |= v7 == v8;
+
+ // CHECK: @llvm.ppc.vsx.xvcmpeqdp.p
+ res |= v9 == v10;
+ return res;
+}
+
Index: lib/CodeGen/CGExprScalar.cpp
===
--- lib/CodeGen/CGExprScalar.cpp
+++ lib/CodeGen/CGExprScalar.cpp
@@ -3120,16 +3120,25 @@
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequh_p :
llvm::Intrinsic::ppc_altivec_vcmpgtsh_p;
case BuiltinType::UInt:
- case BuiltinType::ULong:
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequw_p :
llvm::Intrinsic::ppc_altivec_vcmpgtuw_p;
case BuiltinType::Int:
- case BuiltinType::Long:
return (IT == VCMPEQ) ? llvm::Intrinsic::ppc_altivec_vcmpequw_p :
llvm::Intrinsic::ppc_altivec_vcmpgtsw_p;
+ case BuiltinType::ULong:
+ case Builti
[PATCH] D38816: Convert clang::LangAS to a strongly typed enum
jlebar added a comment.
My only regret is that I have but one +1 to give to this patch.
Comment at: include/clang/Basic/AddressSpaces.h:51
+namespace LanguageAS {
/// The type of a lookup table which maps from language-specific address spaces
I wonder if you need this namespace? LangAS right next to LanguageAS reads
strangely to me -- "what's the difference?".
I guess you'd need to rename Map and fromTargetAS, but the other two members
are probably OK?
Repository:
rL LLVM
https://reviews.llvm.org/D38816
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38430: Enable -pie and --enable-new-dtags by default on Android.
eugenis added a comment. ping https://reviews.llvm.org/D38430 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315506 - [clang-fuzzer] Build with newer protobuf-mutator.
Author: morehouse
Date: Wed Oct 11 13:45:10 2017
New Revision: 315506
URL: http://llvm.org/viewvc/llvm-project?rev=315506&view=rev
Log:
[clang-fuzzer] Build with newer protobuf-mutator.
Modified:
cfe/trunk/cmake/modules/ProtobufMutator.cmake
Modified: cfe/trunk/cmake/modules/ProtobufMutator.cmake
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/cmake/modules/ProtobufMutator.cmake?rev=315506&r1=315505&r2=315506&view=diff
==
--- cfe/trunk/cmake/modules/ProtobufMutator.cmake (original)
+++ cfe/trunk/cmake/modules/ProtobufMutator.cmake Wed Oct 11 13:45:10 2017
@@ -6,7 +6,7 @@ set(PBM_FUZZ_LIB_PATH ${PBM_PATH}/src/li
ExternalProject_Add(${PBM_PREFIX}
PREFIX ${PBM_PREFIX}
GIT_REPOSITORY https://github.com/google/libprotobuf-mutator.git
- GIT_TAG 17789d1
+ GIT_TAG e4eae60
CONFIGURE_COMMAND ${CMAKE_COMMAND} -G${CMAKE_GENERATOR}
-DCMAKE_C_COMPILER=${CMAKE_C_COMPILER}
-DCMAKE_CXX_COMPILER=${CMAKE_CXX_COMPILER}
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38819: [libunwind] Add support for dwarf unwinding on windows on x86_64
mstorsjo added a comment. As further explanation/justification - libcxxabi and libunwind don't support SEH exceptions yet. https://reviews.llvm.org/D38819 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38810: [Analyzer] Support bodyfarming std::call_once for libstdc++
This revision was automatically updated to reflect the committed changes.
Closed by commit rL315508: [Analyzer] Support bodyfarming libstdc++
implementation of std::call_once. (authored by george.karpenkov).
Changed prior to commit:
https://reviews.llvm.org/D38810?vs=118644&id=118686#toc
Repository:
rL LLVM
https://reviews.llvm.org/D38810
Files:
cfe/trunk/lib/Analysis/BodyFarm.cpp
cfe/trunk/test/Analysis/call_once.cpp
Index: cfe/trunk/test/Analysis/call_once.cpp
===
--- cfe/trunk/test/Analysis/call_once.cpp
+++ cfe/trunk/test/Analysis/call_once.cpp
@@ -1,15 +1,24 @@
-// RUN: %clang_analyze_cc1 -std=c++11 -fblocks -analyzer-checker=core,debug.ExprInspection -w -verify %s
+// RUN: %clang_analyze_cc1 -std=c++11 -fblocks -analyzer-checker=core,debug.ExprInspection -verify %s
+// RUN: %clang_analyze_cc1 -std=c++11 -fblocks -analyzer-checker=core,debug.ExprInspection -DEMULATE_LIBSTDCPP -verify %s
void clang_analyzer_eval(bool);
// Faking std::std::call_once implementation.
namespace std {
+
+#ifndef EMULATE_LIBSTDCPP
typedef struct once_flag_s {
unsigned long __state_ = 0;
} once_flag;
+#else
+typedef struct once_flag_s {
+ int _M_once = 0;
+} once_flag;
+#endif
template
void call_once(once_flag &o, Callable func, Args... args) {};
+
} // namespace std
// Check with Lambdas.
Index: cfe/trunk/lib/Analysis/BodyFarm.cpp
===
--- cfe/trunk/lib/Analysis/BodyFarm.cpp
+++ cfe/trunk/lib/Analysis/BodyFarm.cpp
@@ -108,7 +108,7 @@
/// Returns a *first* member field of a record declaration with a given name.
/// \return an nullptr if no member with such a name exists.
- NamedDecl *findMemberField(const CXXRecordDecl *RD, StringRef Name);
+ ValueDecl *findMemberField(const RecordDecl *RD, StringRef Name);
private:
ASTContext &C;
@@ -234,7 +234,7 @@
OK_Ordinary);
}
-NamedDecl *ASTMaker::findMemberField(const CXXRecordDecl *RD, StringRef Name) {
+ValueDecl *ASTMaker::findMemberField(const RecordDecl *RD, StringRef Name) {
CXXBasePaths Paths(
/* FindAmbiguities=*/false,
@@ -246,7 +246,7 @@
DeclContextLookupResult Decls = RD->lookup(DeclName);
for (NamedDecl *FoundDecl : Decls)
if (!FoundDecl->getDeclContext()->isFunctionOrMethod())
- return FoundDecl;
+ return cast(FoundDecl);
return nullptr;
}
@@ -328,25 +328,31 @@
const ParmVarDecl *Callback = D->getParamDecl(1);
QualType CallbackType = Callback->getType().getNonReferenceType();
QualType FlagType = Flag->getType().getNonReferenceType();
- CXXRecordDecl *FlagCXXDecl = FlagType->getAsCXXRecordDecl();
- if (!FlagCXXDecl) {
-DEBUG(llvm::dbgs() << "Flag field is not a CXX record: "
- << "unknown std::call_once implementation."
- << "Ignoring the call.\n");
+ auto *FlagRecordDecl = dyn_cast_or_null(FlagType->getAsTagDecl());
+
+ if (!FlagRecordDecl) {
+DEBUG(llvm::dbgs() << "Flag field is not a record: "
+ << "unknown std::call_once implementation, "
+ << "ignoring the call.\n");
return nullptr;
}
- // Note: here we are assuming libc++ implementation of call_once,
- // which has a struct with a field `__state_`.
- // Body farming might not work for other `call_once` implementations.
- NamedDecl *FoundDecl = M.findMemberField(FlagCXXDecl, "__state_");
- ValueDecl *FieldDecl;
- if (FoundDecl) {
-FieldDecl = dyn_cast(FoundDecl);
- } else {
+ // We initially assume libc++ implementation of call_once,
+ // where the once_flag struct has a field `__state_`.
+ ValueDecl *FlagFieldDecl = M.findMemberField(FlagRecordDecl, "__state_");
+
+ // Otherwise, try libstdc++ implementation, with a field
+ // `_M_once`
+ if (!FlagFieldDecl) {
DEBUG(llvm::dbgs() << "No field __state_ found on std::once_flag struct, "
- << "unable to synthesize call_once body, ignoring "
- << "the call.\n");
+ << "assuming libstdc++ implementation\n");
+FlagFieldDecl = M.findMemberField(FlagRecordDecl, "_M_once");
+ }
+
+ if (!FlagFieldDecl) {
+DEBUG(llvm::dbgs() << "No field _M_once found on std::once flag struct: "
+ << "unknown std::call_once implementation, "
+ << "ignoring the call");
return nullptr;
}
@@ -383,7 +389,7 @@
/* GetNonReferenceType=*/true);
- MemberExpr *Deref = M.makeMemberExpression(FlagDecl, FieldDecl);
+ MemberExpr *Deref = M.makeMemberExpression(FlagDecl, FlagFieldDecl);
assert(Deref->isLValue());
QualType DerefType = Deref->getType();
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315508 - [Analyzer] Support bodyfarming libstdc++ implementation of std::call_once.
Author: george.karpenkov
Date: Wed Oct 11 13:53:01 2017
New Revision: 315508
URL: http://llvm.org/viewvc/llvm-project?rev=315508&view=rev
Log:
[Analyzer] Support bodyfarming libstdc++ implementation of std::call_once.
Differential Revision: https://reviews.llvm.org/D38810
Modified:
cfe/trunk/lib/Analysis/BodyFarm.cpp
cfe/trunk/test/Analysis/call_once.cpp
Modified: cfe/trunk/lib/Analysis/BodyFarm.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Analysis/BodyFarm.cpp?rev=315508&r1=315507&r2=315508&view=diff
==
--- cfe/trunk/lib/Analysis/BodyFarm.cpp (original)
+++ cfe/trunk/lib/Analysis/BodyFarm.cpp Wed Oct 11 13:53:01 2017
@@ -108,7 +108,7 @@ public:
/// Returns a *first* member field of a record declaration with a given name.
/// \return an nullptr if no member with such a name exists.
- NamedDecl *findMemberField(const CXXRecordDecl *RD, StringRef Name);
+ ValueDecl *findMemberField(const RecordDecl *RD, StringRef Name);
private:
ASTContext &C;
@@ -234,7 +234,7 @@ MemberExpr *ASTMaker::makeMemberExpressi
OK_Ordinary);
}
-NamedDecl *ASTMaker::findMemberField(const CXXRecordDecl *RD, StringRef Name) {
+ValueDecl *ASTMaker::findMemberField(const RecordDecl *RD, StringRef Name) {
CXXBasePaths Paths(
/* FindAmbiguities=*/false,
@@ -246,7 +246,7 @@ NamedDecl *ASTMaker::findMemberField(con
DeclContextLookupResult Decls = RD->lookup(DeclName);
for (NamedDecl *FoundDecl : Decls)
if (!FoundDecl->getDeclContext()->isFunctionOrMethod())
- return FoundDecl;
+ return cast(FoundDecl);
return nullptr;
}
@@ -328,25 +328,31 @@ static Stmt *create_call_once(ASTContext
const ParmVarDecl *Callback = D->getParamDecl(1);
QualType CallbackType = Callback->getType().getNonReferenceType();
QualType FlagType = Flag->getType().getNonReferenceType();
- CXXRecordDecl *FlagCXXDecl = FlagType->getAsCXXRecordDecl();
- if (!FlagCXXDecl) {
-DEBUG(llvm::dbgs() << "Flag field is not a CXX record: "
- << "unknown std::call_once implementation."
- << "Ignoring the call.\n");
+ auto *FlagRecordDecl =
dyn_cast_or_null(FlagType->getAsTagDecl());
+
+ if (!FlagRecordDecl) {
+DEBUG(llvm::dbgs() << "Flag field is not a record: "
+ << "unknown std::call_once implementation, "
+ << "ignoring the call.\n");
return nullptr;
}
- // Note: here we are assuming libc++ implementation of call_once,
- // which has a struct with a field `__state_`.
- // Body farming might not work for other `call_once` implementations.
- NamedDecl *FoundDecl = M.findMemberField(FlagCXXDecl, "__state_");
- ValueDecl *FieldDecl;
- if (FoundDecl) {
-FieldDecl = dyn_cast(FoundDecl);
- } else {
+ // We initially assume libc++ implementation of call_once,
+ // where the once_flag struct has a field `__state_`.
+ ValueDecl *FlagFieldDecl = M.findMemberField(FlagRecordDecl, "__state_");
+
+ // Otherwise, try libstdc++ implementation, with a field
+ // `_M_once`
+ if (!FlagFieldDecl) {
DEBUG(llvm::dbgs() << "No field __state_ found on std::once_flag struct, "
- << "unable to synthesize call_once body, ignoring "
- << "the call.\n");
+ << "assuming libstdc++ implementation\n");
+FlagFieldDecl = M.findMemberField(FlagRecordDecl, "_M_once");
+ }
+
+ if (!FlagFieldDecl) {
+DEBUG(llvm::dbgs() << "No field _M_once found on std::once flag struct: "
+ << "unknown std::call_once implementation, "
+ << "ignoring the call");
return nullptr;
}
@@ -383,7 +389,7 @@ static Stmt *create_call_once(ASTContext
/* GetNonReferenceType=*/true);
- MemberExpr *Deref = M.makeMemberExpression(FlagDecl, FieldDecl);
+ MemberExpr *Deref = M.makeMemberExpression(FlagDecl, FlagFieldDecl);
assert(Deref->isLValue());
QualType DerefType = Deref->getType();
Modified: cfe/trunk/test/Analysis/call_once.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Analysis/call_once.cpp?rev=315508&r1=315507&r2=315508&view=diff
==
--- cfe/trunk/test/Analysis/call_once.cpp (original)
+++ cfe/trunk/test/Analysis/call_once.cpp Wed Oct 11 13:53:01 2017
@@ -1,15 +1,24 @@
-// RUN: %clang_analyze_cc1 -std=c++11 -fblocks
-analyzer-checker=core,debug.ExprInspection -w -verify %s
+// RUN: %clang_analyze_cc1 -std=c++11 -fblocks
-analyzer-checker=core,debug.ExprInspection -verify %s
+// RUN: %clang_analyze_cc1 -std=c++11 -fblocks
-analyzer-checker=core,debug.ExprInspection -DEMULATE_LIBSTDCPP -verify %s
void clang_analyzer_eval(bool);
// Faking std::std::call_once implementation.
namespace std {
+
+#ifndef EMULATE_LIBSTDCPP
typedef struct once_flag_s {
unsigned long __state_ = 0;
}
[PATCH] D38425: [clangd] Document highlights for clangd
Nebiroth marked an inline comment as done.
Nebiroth added inline comments.
Comment at: clangd/ClangdUnit.cpp:784
+/// Finds declarations locations that a given source location refers to.
+class TargetDeclarationFinder : public index::IndexDataConsumer {
+ std::vector DeclarationLocations;
ilya-biryukov wrote:
> This `IndexDataConsumer` effectively does the same thing as
> `DeclarationLocationsFinder`.
> We should reuse the code between those two.
>
> To do that we could:
> 1. Change `DeclarationLocationsFinder` to return a list of `Decl*`s and
> `MacroDef*`s that are under caret.
> 2. In `findDefinition` we should return locations of found `Decl` and
> `MacroDef`.
> 3. In `documentHighlights` we would rerun `DocumentHighlightsFinder`to
> capture locations of `Decls` referenced in step 1.
>
> Do you think this approach would work?
I have something similar to this implemented. Will upstream it soon.
https://reviews.llvm.org/D38425
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38821: [COFF, ARM64] Add MS builtins __dmb, __dsb, __isb
mstorsjo created this revision.
Herald added subscribers: kristof.beyls, javed.absar, aemerson.
This is an update to https://reviews.llvm.org/D36111 by @mgrang, taking over
finishing of this patch. Compared to his version, this makes the intrinsics
conditional to ALL_MS_LANGUAGES.
https://reviews.llvm.org/D38821
Files:
include/clang/Basic/BuiltinsAArch64.def
lib/Basic/Targets/AArch64.cpp
test/CodeGen/arm64-microsoft-intrinsics.c
Index: test/CodeGen/arm64-microsoft-intrinsics.c
===
--- /dev/null
+++ test/CodeGen/arm64-microsoft-intrinsics.c
@@ -0,0 +1,26 @@
+// RUN: %clang_cc1 -triple arm64-windows -fms-compatibility -emit-llvm -o - %s
\
+// RUN:| FileCheck %s -check-prefix CHECK-MSVC
+
+// RUN: not %clang_cc1 -triple arm64-linux -Werror -S -o /dev/null %s 2>&1 \
+// RUN:| FileCheck %s -check-prefix CHECK-LINUX
+
+void check__dmb(void) {
+ __dmb(0);
+}
+
+// CHECK-MSVC: @llvm.aarch64.dmb(i32 0)
+// CHECK-LINUX: error: implicit declaration of function '__dmb'
+
+void check__dsb(void) {
+ __dsb(0);
+}
+
+// CHECK-MSVC: @llvm.aarch64.dsb(i32 0)
+// CHECK-LINUX: error: implicit declaration of function '__dsb'
+
+void check__isb(void) {
+ __isb(0);
+}
+
+// CHECK-MSVC: @llvm.aarch64.isb(i32 0)
+// CHECK-LINUX: error: implicit declaration of function '__isb'
Index: lib/Basic/Targets/AArch64.cpp
===
--- lib/Basic/Targets/AArch64.cpp
+++ lib/Basic/Targets/AArch64.cpp
@@ -27,6 +27,8 @@
#define BUILTIN(ID, TYPE, ATTRS)
\
{#ID, TYPE, ATTRS, nullptr, ALL_LANGUAGES, nullptr},
+#define LANGBUILTIN(ID, TYPE, ATTRS, LANG)
\
+ {#ID, TYPE, ATTRS, nullptr, LANG, nullptr},
#include "clang/Basic/BuiltinsAArch64.def"
};
Index: include/clang/Basic/BuiltinsAArch64.def
===
--- include/clang/Basic/BuiltinsAArch64.def
+++ include/clang/Basic/BuiltinsAArch64.def
@@ -14,6 +14,10 @@
// The format of this database matches clang/Basic/Builtins.def.
+#if defined(BUILTIN) && !defined(LANGBUILTIN)
+# define LANGBUILTIN(ID, TYPE, ATTRS, BUILTIN_LANG) BUILTIN(ID, TYPE, ATTRS)
+#endif
+
// In libgcc
BUILTIN(__clear_cache, "vv*v*", "i")
@@ -61,4 +65,9 @@
BUILTIN(__builtin_arm_wsr64, "vcC*LUi", "nc")
BUILTIN(__builtin_arm_wsrp, "vcC*vC*", "nc")
+LANGBUILTIN(__dmb, "vUi", "nc", ALL_MS_LANGUAGES)
+LANGBUILTIN(__dsb, "vUi", "nc", ALL_MS_LANGUAGES)
+LANGBUILTIN(__isb, "vUi", "nc", ALL_MS_LANGUAGES)
+
#undef BUILTIN
+#undef LANGBUILTIN
Index: test/CodeGen/arm64-microsoft-intrinsics.c
===
--- /dev/null
+++ test/CodeGen/arm64-microsoft-intrinsics.c
@@ -0,0 +1,26 @@
+// RUN: %clang_cc1 -triple arm64-windows -fms-compatibility -emit-llvm -o - %s \
+// RUN:| FileCheck %s -check-prefix CHECK-MSVC
+
+// RUN: not %clang_cc1 -triple arm64-linux -Werror -S -o /dev/null %s 2>&1 \
+// RUN:| FileCheck %s -check-prefix CHECK-LINUX
+
+void check__dmb(void) {
+ __dmb(0);
+}
+
+// CHECK-MSVC: @llvm.aarch64.dmb(i32 0)
+// CHECK-LINUX: error: implicit declaration of function '__dmb'
+
+void check__dsb(void) {
+ __dsb(0);
+}
+
+// CHECK-MSVC: @llvm.aarch64.dsb(i32 0)
+// CHECK-LINUX: error: implicit declaration of function '__dsb'
+
+void check__isb(void) {
+ __isb(0);
+}
+
+// CHECK-MSVC: @llvm.aarch64.isb(i32 0)
+// CHECK-LINUX: error: implicit declaration of function '__isb'
Index: lib/Basic/Targets/AArch64.cpp
===
--- lib/Basic/Targets/AArch64.cpp
+++ lib/Basic/Targets/AArch64.cpp
@@ -27,6 +27,8 @@
#define BUILTIN(ID, TYPE, ATTRS) \
{#ID, TYPE, ATTRS, nullptr, ALL_LANGUAGES, nullptr},
+#define LANGBUILTIN(ID, TYPE, ATTRS, LANG) \
+ {#ID, TYPE, ATTRS, nullptr, LANG, nullptr},
#include "clang/Basic/BuiltinsAArch64.def"
};
Index: include/clang/Basic/BuiltinsAArch64.def
===
--- include/clang/Basic/BuiltinsAArch64.def
+++ include/clang/Basic/BuiltinsAArch64.def
@@ -14,6 +14,10 @@
// The format of this database matches clang/Basic/Builtins.def.
+#if defined(BUILTIN) && !defined(LANGBUILTIN)
+# define LANGBUILTIN(ID, TYPE, ATTRS, BUILTIN_LANG) BUILTIN(ID, TYPE, ATTRS)
+#endif
+
// In libgcc
BUILTIN(__clear_cache, "vv*v*", "i")
@@ -61,4 +65,9 @@
BUILTIN(__builtin_arm_wsr64, "vcC*LUi", "nc")
BUILTIN(__builtin_arm_wsrp, "vcC*vC*", "nc")
+LANGBUILTIN(__dmb, "vUi", "nc", ALL_MS_LANGUAGES)
+LANGBUILTIN(__dsb, "vUi", "nc", ALL_MS_LANGUAGES)
+LANGBUILTIN(__isb, "vUi", "nc", ALL_MS_LANGUAGES)
+
#undef BUILTIN
+#undef LANGBUILTIN
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/c
[PATCH] D35082: [OpenCL] Add LangAS::opencl_private to represent private address space in AST
rjmccall added a comment. It sounds like there's agreement about the basic technical direction of introducing LangAS::opencl_private. Please introduce isAddressSpaceImplicit() in a different patch and make this patch just about the introduction of LangAS::opencl_private. You can have the pretty-printer just ignore __private for now, which should avoid gratuitous diagnostic changes. I would like you to investigate using AttributedType for the pretty-printing and address-space semantic checks before adding isAddressSpaceImplicit(). Comment at: include/clang/AST/Type.h:562 + static const uint32_t IMask = 0x200; + static const uint32_t IShift = 9; static const uint32_t AddressSpaceMask = "I" is not an appropriate abbreviation for "AddressSpaceImplicit". Comment at: include/clang/Basic/AddressSpaces.h:34 // OpenCL specific address spaces. opencl_global, I think you need a real comment about the design of OpenCL address spaces here. Specifically, it is important to note that OpenCL no longer uses LangAS::Default for anything except r-values. https://reviews.llvm.org/D35082 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38824: [X86] Synchronize the CPU predefined macros with gcc
craig.topper created this revision. Herald added a subscriber: krytarowski. We were using corei7 for a large swatch of Intel CPUs. gcc has a different defines that more closely match the march flags. This updates to match. It also fixes skylake-avx512 and adds silvermont in addition to slm. https://reviews.llvm.org/D38824 Files: lib/Basic/Targets/X86.cpp test/Preprocessor/predefined-arch-macros.c Index: test/Preprocessor/predefined-arch-macros.c === --- test/Preprocessor/predefined-arch-macros.c +++ test/Preprocessor/predefined-arch-macros.c @@ -427,11 +427,14 @@ // CHECK_COREI7_AVX_M32: #define __SSSE3__ 1 // CHECK_COREI7_AVX_M32: #define __XSAVEOPT__ 1 // CHECK_COREI7_AVX_M32: #define __XSAVE__ 1 -// CHECK_COREI7_AVX_M32: #define __corei7 1 -// CHECK_COREI7_AVX_M32: #define __corei7__ 1 +// CHECK_COREI7_AVX_M32: #define __corei7_avx 1 +// CHECK_COREI7_AVX_M32: #define __corei7_avx__ 1 // CHECK_COREI7_AVX_M32: #define __i386 1 // CHECK_COREI7_AVX_M32: #define __i386__ 1 -// CHECK_COREI7_AVX_M32: #define __tune_corei7__ 1 +// CHECK_COREI7_AVX_M32: #define __sandybridge 1 +// CHECK_COREI7_AVX_M32: #define __sandybridge__ 1 +// CHECK_COREI7_AVX_M32: #define __tune_corei7_avx__ 1 +// CHECK_COREI7_AVX_M32: #define __tune_sandybridge__ 1 // CHECK_COREI7_AVX_M32: #define i386 1 // RUN: %clang -march=corei7-avx -m64 -E -dM %s -o - 2>&1 \ // RUN: -target i386-unknown-linux \ @@ -454,9 +457,12 @@ // CHECK_COREI7_AVX_M64: #define __XSAVE__ 1 // CHECK_COREI7_AVX_M64: #define __amd64 1 // CHECK_COREI7_AVX_M64: #define __amd64__ 1 -// CHECK_COREI7_AVX_M64: #define __corei7 1 -// CHECK_COREI7_AVX_M64: #define __corei7__ 1 -// CHECK_COREI7_AVX_M64: #define __tune_corei7__ 1 +// CHECK_COREI7_AVX_M64: #define __corei7_avx 1 +// CHECK_COREI7_AVX_M64: #define __corei7_avx__ 1 +// CHECK_COREI7_AVX_M64: #define __sandybridge 1 +// CHECK_COREI7_AVX_M64: #define __sandybridge__ 1 +// CHECK_COREI7_AVX_M64: #define __tune_corei7_avx__ 1 +// CHECK_COREI7_AVX_M64: #define __tune_sandybridge__ 1 // CHECK_COREI7_AVX_M64: #define __x86_64 1 // CHECK_COREI7_AVX_M64: #define __x86_64__ 1 // @@ -477,11 +483,14 @@ // CHECK_CORE_AVX_I_M32: #define __SSSE3__ 1 // CHECK_CORE_AVX_I_M32: #define __XSAVEOPT__ 1 // CHECK_CORE_AVX_I_M32: #define __XSAVE__ 1 -// CHECK_CORE_AVX_I_M32: #define __corei7 1 -// CHECK_CORE_AVX_I_M32: #define __corei7__ 1 +// CHECK_CORE_AVX_I_M32: #define __corei7_avx 1 +// CHECK_CORE_AVX_I_M32: #define __corei7_avx__ 1 // CHECK_CORE_AVX_I_M32: #define __i386 1 // CHECK_CORE_AVX_I_M32: #define __i386__ 1 -// CHECK_CORE_AVX_I_M32: #define __tune_corei7__ 1 +// CHECK_CORE_AVX_I_M32: #define __sandybridge 1 +// CHECK_CORE_AVX_I_M32: #define __sandybridge__ 1 +// CHECK_CORE_AVX_I_M32: #define __tune_corei7_avx__ 1 +// CHECK_CORE_AVX_I_M32: #define __tune_sandybridge__ 1 // CHECK_CORE_AVX_I_M32: #define i386 1 // RUN: %clang -march=core-avx-i -m64 -E -dM %s -o - 2>&1 \ // RUN: -target i386-unknown-linux \ @@ -504,9 +513,12 @@ // CHECK_CORE_AVX_I_M64: #define __XSAVE__ 1 // CHECK_CORE_AVX_I_M64: #define __amd64 1 // CHECK_CORE_AVX_I_M64: #define __amd64__ 1 -// CHECK_CORE_AVX_I_M64: #define __corei7 1 -// CHECK_CORE_AVX_I_M64: #define __corei7__ 1 -// CHECK_CORE_AVX_I_M64: #define __tune_corei7__ 1 +// CHECK_CORE_AVX_I_M64: #define __corei7_avx 1 +// CHECK_CORE_AVX_I_M64: #define __corei7_avx__ 1 +// CHECK_CORE_AVX_I_M64: #define __sandybridge 1 +// CHECK_CORE_AVX_I_M64: #define __sandybridge__ 1 +// CHECK_CORE_AVX_I_M64: #define __tune_corei7_avx__ 1 +// CHECK_CORE_AVX_I_M64: #define __tune_sandybridge__ 1 // CHECK_CORE_AVX_I_M64: #define __x86_64 1 // CHECK_CORE_AVX_I_M64: #define __x86_64__ 1 // @@ -533,11 +545,14 @@ // CHECK_CORE_AVX2_M32: #define __SSSE3__ 1 // CHECK_CORE_AVX2_M32: #define __XSAVEOPT__ 1 // CHECK_CORE_AVX2_M32: #define __XSAVE__ 1 -// CHECK_CORE_AVX2_M32: #define __corei7 1 -// CHECK_CORE_AVX2_M32: #define __corei7__ 1 +// CHECK_CORE_AVX2_M32: #define __core_avx2 1 +// CHECK_CORE_AVX2_M32: #define __core_avx2__ 1 +// CHECK_CORE_AVX2_M32: #define __haswell 1 +// CHECK_CORE_AVX2_M32: #define __haswell__ 1 // CHECK_CORE_AVX2_M32: #define __i386 1 // CHECK_CORE_AVX2_M32: #define __i386__ 1 -// CHECK_CORE_AVX2_M32: #define __tune_corei7__ 1 +// CHECK_CORE_AVX2_M32: #define __tune_core_avx2__ 1 +// CHECK_CORE_AVX2_M32: #define __tune_haswell__ 1 // CHECK_CORE_AVX2_M32: #define i386 1 // RUN: %clang -march=core-avx2 -m64 -E -dM %s -o - 2>&1 \ // RUN: -target i386-unknown-linux \ @@ -566,9 +581,12 @@ // CHECK_CORE_AVX2_M64: #define __XSAVE__ 1 // CHECK_CORE_AVX2_M64: #define __amd64 1 // CHECK_CORE_AVX2_M64: #define __amd64__ 1 -// CHECK_CORE_AVX2_M64: #define __corei7 1 -// CHECK_CORE_AVX2_M64: #define __corei7__ 1 -// CHECK_CORE_AVX2_M64: #define __tune_corei7__ 1 +// CHECK_CORE_AVX2_M64: #define __core_avx2 1 +// CHECK_CORE_AVX2_M64: #define __core_avx2__ 1 +// CHECK_CORE_AVX2_M64: #def
[PATCH] D38772: [refactor] allow the use of refactoring diagnostics
arphaman added inline comments. Comment at: lib/Basic/DiagnosticIDs.cpp:46 unsigned WarnShowInSystemHeader : 1; - unsigned Category : 5; + unsigned Category : 6; hokein wrote: > just curious: is this change needed? I get a build warning without this change as the bitfield becomes too narrow with the new category, so yeah. Comment at: tools/clang-refactor/ToolRefactoringResultConsumer.h:19 + +/// A subclass of \c RefactoringResultConsumer that stores the reference to the +/// TU-specific diagnostics engine. hokein wrote: > I'd name it "interface", because it has unimplemented virtual function > (`handleError`), clients can't create an instance of it. > > or alternatively, does it make more sense to just add these methods and > `DiagnosticsEngine` variable to the `tooling::RefactoringResultConsumer` > interface? I see you have replaced "RefactoringResultConsumer" with this new > interface in many places. Right now I don't think having the diagnostics engine will be useful for clients outside of tool, so I'd prefer to keep it here. We can reconsider this decision in the future if we need to. Repository: rL LLVM https://reviews.llvm.org/D38772 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38772: [refactor] allow the use of refactoring diagnostics
arphaman updated this revision to Diff 118701.
arphaman marked 2 inline comments as done.
arphaman added a comment.
- rename the common consumer class.
Repository:
rL LLVM
https://reviews.llvm.org/D38772
Files:
include/clang/Basic/AllDiagnostics.h
include/clang/Basic/CMakeLists.txt
include/clang/Basic/Diagnostic.td
include/clang/Basic/DiagnosticIDs.h
include/clang/Basic/DiagnosticRefactoringKinds.td
include/clang/Tooling/Refactoring/RefactoringActionRuleRequirements.h
include/clang/Tooling/Refactoring/RefactoringDiagnostic.h
include/clang/Tooling/Refactoring/RefactoringRuleContext.h
lib/Basic/DiagnosticIDs.cpp
lib/Tooling/Refactoring/Rename/RenamingAction.cpp
test/Refactor/LocalRename/NoSymbolSelectedError.cpp
tools/clang-refactor/ClangRefactor.cpp
tools/clang-refactor/TestSupport.cpp
tools/clang-refactor/TestSupport.h
tools/clang-refactor/ToolRefactoringResultConsumer.h
tools/diagtool/DiagnosticNames.cpp
Index: tools/diagtool/DiagnosticNames.cpp
===
--- tools/diagtool/DiagnosticNames.cpp
+++ tools/diagtool/DiagnosticNames.cpp
@@ -42,6 +42,7 @@
#include "clang/Basic/DiagnosticCommentKinds.inc"
#include "clang/Basic/DiagnosticSemaKinds.inc"
#include "clang/Basic/DiagnosticAnalysisKinds.inc"
+#include "clang/Basic/DiagnosticRefactoringKinds.inc"
#undef DIAG
};
Index: tools/clang-refactor/ToolRefactoringResultConsumer.h
===
--- /dev/null
+++ tools/clang-refactor/ToolRefactoringResultConsumer.h
@@ -0,0 +1,48 @@
+//===--- ToolRefactoringResultConsumer.h - --*- C++ -*-===//
+//
+// The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===--===//
+
+#ifndef LLVM_CLANG_TOOLS_CLANG_REFACTOR_TOOL_REFACTORING_RESULT_CONSUMER_H
+#define LLVM_CLANG_TOOLS_CLANG_REFACTOR_TOOL_REFACTORING_RESULT_CONSUMER_H
+
+#include "clang/AST/ASTContext.h"
+#include "clang/Tooling/Refactoring/RefactoringResultConsumer.h"
+
+namespace clang {
+namespace refactor {
+
+/// An interface that subclasses the \c RefactoringResultConsumer interface
+/// that stores the reference to the TU-specific diagnostics engine.
+class ClangRefactorToolConsumerInterface
+: public tooling::RefactoringResultConsumer {
+public:
+ /// Called when a TU is entered.
+ void beginTU(ASTContext &Context) {
+assert(!Diags && "Diags has been set");
+Diags = &Context.getDiagnostics();
+ }
+
+ /// Called when the tool is done with a TU.
+ void endTU() {
+assert(Diags && "Diags unset");
+Diags = nullptr;
+ }
+
+ DiagnosticsEngine &getDiags() const {
+assert(Diags && "no diags");
+return *Diags;
+ }
+
+private:
+ DiagnosticsEngine *Diags = nullptr;
+};
+
+} // end namespace refactor
+} // end namespace clang
+
+#endif // LLVM_CLANG_TOOLS_CLANG_REFACTOR_TOOL_REFACTORING_RESULT_CONSUMER_H
Index: tools/clang-refactor/TestSupport.h
===
--- tools/clang-refactor/TestSupport.h
+++ tools/clang-refactor/TestSupport.h
@@ -16,9 +16,9 @@
#ifndef LLVM_CLANG_TOOLS_CLANG_REFACTOR_TEST_SUPPORT_H
#define LLVM_CLANG_TOOLS_CLANG_REFACTOR_TEST_SUPPORT_H
+#include "ToolRefactoringResultConsumer.h"
#include "clang/Basic/LLVM.h"
#include "clang/Basic/SourceLocation.h"
-#include "clang/Tooling/Refactoring/RefactoringResultConsumer.h"
#include "llvm/ADT/Optional.h"
#include "llvm/ADT/SmallVector.h"
#include "llvm/Support/Error.h"
@@ -65,7 +65,7 @@
bool foreachRange(const SourceManager &SM,
llvm::function_ref Callback) const;
- std::unique_ptr createConsumer() const;
+ std::unique_ptr createConsumer() const;
void dump(llvm::raw_ostream &OS) const;
};
Index: tools/clang-refactor/TestSupport.cpp
===
--- tools/clang-refactor/TestSupport.cpp
+++ tools/clang-refactor/TestSupport.cpp
@@ -14,6 +14,7 @@
//===--===//
#include "TestSupport.h"
+#include "clang/Basic/DiagnosticError.h"
#include "clang/Basic/SourceManager.h"
#include "clang/Lex/Lexer.h"
#include "llvm/ADT/STLExtras.h"
@@ -106,7 +107,7 @@
}
class TestRefactoringResultConsumer final
-: public tooling::RefactoringResultConsumer {
+: public ClangRefactorToolConsumerInterface {
public:
TestRefactoringResultConsumer(const TestSelectionRangesInFile &TestRanges)
: TestRanges(TestRanges) {
@@ -182,10 +183,15 @@
std::string ErrorMessage;
bool HasResult = !!Result;
if (!HasResult) {
-// FIXME: Handle diagnostic error as well.
-handleAllErrors(Result.takeError(), [&](StringError &Err) {
- ErrorMessage = Err.getMessage();
-
[PATCH] D38816: Convert clang::LangAS to a strongly typed enum
arichardson added inline comments.
Comment at: include/clang/Basic/AddressSpaces.h:51
+namespace LanguageAS {
/// The type of a lookup table which maps from language-specific address spaces
jlebar wrote:
> I wonder if you need this namespace? LangAS right next to LanguageAS reads
> strangely to me -- "what's the difference?".
>
> I guess you'd need to rename Map and fromTargetAS, but the other two members
> are probably OK?
The only reason I added this namespace is that I wasn't sure whether having
those functions in the clang namespace is acceptable.
Not quite sure what to call the functions though. `langASFromTargetAS`?
Repository:
rL LLVM
https://reviews.llvm.org/D38816
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38816: Convert clang::LangAS to a strongly typed enum
jlebar added a comment. > The only reason I added this namespace is that I wasn't sure whether having > those functions in the clang namespace is acceptable. Maybe someone else will object, or suggest an existing namespace they should be in. FWIW I think it's fine. > Not quite sure what to call the functions though. langASFromTargetAS? sgtm! Repository: rL LLVM https://reviews.llvm.org/D38816 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315517 - [X86] Add support for 'amdfam17h' to __builtin_cpu_is to match gcc.
Author: ctopper
Date: Wed Oct 11 14:42:02 2017
New Revision: 315517
URL: http://llvm.org/viewvc/llvm-project?rev=315517&view=rev
Log:
[X86] Add support for 'amdfam17h' to __builtin_cpu_is to match gcc.
The compiler-rt implementation already supported it, it just wasn't exposed.
Modified:
cfe/trunk/lib/Basic/Targets/X86.cpp
cfe/trunk/lib/CodeGen/CGBuiltin.cpp
cfe/trunk/test/CodeGen/target-builtin-noerror.c
Modified: cfe/trunk/lib/Basic/Targets/X86.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Basic/Targets/X86.cpp?rev=315517&r1=315516&r2=315517&view=diff
==
--- cfe/trunk/lib/Basic/Targets/X86.cpp (original)
+++ cfe/trunk/lib/Basic/Targets/X86.cpp Wed Oct 11 14:42:02 2017
@@ -1293,6 +1293,7 @@ bool X86TargetInfo::validateCpuIs(String
.Case("amd", true)
.Case("amdfam10h", true)
.Case("amdfam15h", true)
+ .Case("amdfam17h", true)
.Case("atom", true)
.Case("barcelona", true)
.Case("bdver1", true)
Modified: cfe/trunk/lib/CodeGen/CGBuiltin.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/CodeGen/CGBuiltin.cpp?rev=315517&r1=315516&r2=315517&view=diff
==
--- cfe/trunk/lib/CodeGen/CGBuiltin.cpp (original)
+++ cfe/trunk/lib/CodeGen/CGBuiltin.cpp Wed Oct 11 14:42:02 2017
@@ -7501,6 +7501,7 @@ Value *CodeGenFunction::EmitX86CpuIs(Str
INTEL_KNL,
AMD_BTVER1,
AMD_BTVER2,
+AMDFAM17H,
CPU_SUBTYPE_START,
INTEL_COREI7_NEHALEM,
INTEL_COREI7_WESTMERE,
@@ -7527,6 +7528,7 @@ Value *CodeGenFunction::EmitX86CpuIs(Str
.Case("amdfam10", AMDFAM10H)
.Case("amdfam15h", AMDFAM15H)
.Case("amdfam15", AMDFAM15H)
+ .Case("amdfam17h", AMDFAM17H)
.Case("atom", INTEL_BONNELL)
.Case("barcelona", AMDFAM10H_BARCELONA)
.Case("bdver1", AMDFAM15H_BDVER1)
Modified: cfe/trunk/test/CodeGen/target-builtin-noerror.c
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/test/CodeGen/target-builtin-noerror.c?rev=315517&r1=315516&r2=315517&view=diff
==
--- cfe/trunk/test/CodeGen/target-builtin-noerror.c (original)
+++ cfe/trunk/test/CodeGen/target-builtin-noerror.c Wed Oct 11 14:42:02 2017
@@ -81,6 +81,7 @@ void verifycpustrings() {
(void)__builtin_cpu_is("amd");
(void)__builtin_cpu_is("amdfam10h");
(void)__builtin_cpu_is("amdfam15h");
+ (void)__builtin_cpu_is("amdfam17h");
(void)__builtin_cpu_is("atom");
(void)__builtin_cpu_is("barcelona");
(void)__builtin_cpu_is("bdver1");
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38680: [libunwind] Fix handling of DW_CFA_GNU_args_size
rnk added inline comments. Comment at: src/libunwind.cpp:188 + co->getInfo(&info); + pint_t orgArgSize = (pint_t)info.gp; + uint64_t orgFuncStart = info.start_ip; I think it makes sense to have this here: the contract is that if the personality sets the IP when the cursor pointed to a PC with a non-zero arg size, we should adjust SP for the personality. However, it's not clear to me that we don't need the same adjustment when stepping across frames that use arg size without a frame pointer. https://reviews.llvm.org/D38680 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
r315518 - [Driver] Export symbols needed to use profile runtime
Author: vedantk
Date: Wed Oct 11 14:54:09 2017
New Revision: 315518
URL: http://llvm.org/viewvc/llvm-project?rev=315518&view=rev
Log:
[Driver] Export symbols needed to use profile runtime
Apple's API verification tool (tapi) checks that the symbols exported
from a program match a whitelist. When the program uses the profile
runtime, some additional symbols which are typically not part of the
regular whitelist must be exported.
If we're using symbol export directives along with the profile runtime
on Darwin, the driver needs to export the additional symbols to avoid
verification failures.
rdar://problem/30067753
Modified:
cfe/trunk/lib/Driver/ToolChains/Darwin.cpp
cfe/trunk/test/Driver/darwin-ld.c
Modified: cfe/trunk/lib/Driver/ToolChains/Darwin.cpp
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Driver/ToolChains/Darwin.cpp?rev=315518&r1=315517&r2=315518&view=diff
==
--- cfe/trunk/lib/Driver/ToolChains/Darwin.cpp (original)
+++ cfe/trunk/lib/Driver/ToolChains/Darwin.cpp Wed Oct 11 14:54:09 2017
@@ -986,6 +986,25 @@ StringRef Darwin::getOSLibraryNameSuffix
llvm_unreachable("Unsupported platform");
}
+/// Check if the link command contains a symbol export directive.
+static bool hasExportSymbolDirective(const ArgList &Args) {
+ for (Arg *A : Args) {
+if (!A->getOption().matches(options::OPT_Wl_COMMA) &&
+!A->getOption().matches(options::OPT_Xlinker))
+ continue;
+if (A->containsValue("-exported_symbols_list") ||
+A->containsValue("-exported_symbol"))
+ return true;
+ }
+ return false;
+}
+
+/// Add an export directive for \p Symbol to the link command.
+static void addExportedSymbol(ArgStringList &CmdArgs, const char *Symbol) {
+ CmdArgs.push_back("-exported_symbol");
+ CmdArgs.push_back(Symbol);
+}
+
void Darwin::addProfileRTLibs(const ArgList &Args,
ArgStringList &CmdArgs) const {
if (!needsProfileRT(Args)) return;
@@ -994,6 +1013,16 @@ void Darwin::addProfileRTLibs(const ArgL
Args, CmdArgs,
(Twine("libclang_rt.profile_") + getOSLibraryNameSuffix() + ".a").str(),
RuntimeLinkOptions(RLO_AlwaysLink | RLO_FirstLink));
+
+ // If we have a symbol export directive and we're linking in the profile
+ // runtime, automatically export symbols necessary to implement some of the
+ // runtime's functionality.
+ if (hasExportSymbolDirective(Args)) {
+addExportedSymbol(CmdArgs, "_VPMergeHook");
+addExportedSymbol(CmdArgs, "___llvm_profile_filename");
+addExportedSymbol(CmdArgs, "___llvm_profile_raw_version");
+addExportedSymbol(CmdArgs, "_lprofCurFilename");
+ }
}
void DarwinClang::AddLinkSanitizerLibArgs(const ArgList &Args,
Modified: cfe/trunk/test/Driver/darwin-ld.c
URL:
http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Driver/darwin-ld.c?rev=315518&r1=315517&r2=315518&view=diff
==
--- cfe/trunk/test/Driver/darwin-ld.c (original)
+++ cfe/trunk/test/Driver/darwin-ld.c Wed Oct 11 14:54:09 2017
@@ -351,3 +351,17 @@
// RUN: %clang -target arm64-apple-ios5.0 -miphoneos-version-min=5.0
-fprofile-instr-generate -### %t.o 2> %t.log
// RUN: FileCheck -check-prefix=LINK_PROFILE_FIRST %s < %t.log
// LINK_PROFILE_FIRST: {{ld(.exe)?"}}
"{{[^"]+}}libclang_rt.profile_{{[a-z]+}}.a"
+
+// RUN: %clang -target x86_64-apple-darwin12 -fprofile-instr-generate
-Wl,-exported_symbols_list,/dev/null -### %t.o 2> %t.log
+// RUN: FileCheck -check-prefix=PROFILE_EXPORT %s < %t.log
+// RUN: %clang -target x86_64-apple-darwin12 -fprofile-instr-generate
-Wl,-exported_symbol,foo -### %t.o 2> %t.log
+// RUN: FileCheck -check-prefix=PROFILE_EXPORT %s < %t.log
+// RUN: %clang -target x86_64-apple-darwin12 -fprofile-instr-generate -Xlinker
-exported_symbol -Xlinker foo -### %t.o 2> %t.log
+// RUN: FileCheck -check-prefix=PROFILE_EXPORT %s < %t.log
+// RUN: %clang -target x86_64-apple-darwin12 -fprofile-instr-generate -Xlinker
-exported_symbols_list -Xlinker /dev/null -### %t.o 2> %t.log
+// RUN: FileCheck -check-prefix=PROFILE_EXPORT %s < %t.log
+// PROFILE_EXPORT: "-exported_symbol" "_VPMergeHook" "-exported_symbol"
"___llvm_profile_filename" "-exported_symbol" "___llvm_profile_raw_version"
"-exported_symbol" "_lprofCurFilename"
+//
+// RUN: %clang -target x86_64-apple-darwin12 -fprofile-instr-generate -###
%t.o 2> %t.log
+// RUN: FileCheck -check-prefix=NO_PROFILE_EXPORT %s < %t.log
+// NO_PROFILE_EXPORT-NOT: "-exported_symbol"
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D35082: [OpenCL] Add LangAS::opencl_private to represent private address space in AST
yaxunl marked 2 inline comments as done. yaxunl added a comment. In https://reviews.llvm.org/D35082#895230, @rjmccall wrote: > It sounds like there's agreement about the basic technical direction of > introducing LangAS::opencl_private. Please introduce > isAddressSpaceImplicit() in a different patch and make this patch just about > the introduction of LangAS::opencl_private. You can have the pretty-printer > just ignore __private for now, which should avoid gratuitous diagnostic > changes. > > I would like you to investigate using AttributedType for the pretty-printing > and address-space semantic checks before adding isAddressSpaceImplicit(). Thanks. I will separate the implicit addr space flag to another patch. Comment at: include/clang/AST/Type.h:562 + static const uint32_t IMask = 0x200; + static const uint32_t IShift = 9; static const uint32_t AddressSpaceMask = rjmccall wrote: > "I" is not an appropriate abbreviation for "AddressSpaceImplicit". Will change it to ImplictAddrSpace. Comment at: include/clang/Basic/AddressSpaces.h:34 // OpenCL specific address spaces. opencl_global, rjmccall wrote: > I think you need a real comment about the design of OpenCL address spaces > here. Specifically, it is important to note that OpenCL no longer uses > LangAS::Default for anything except r-values. Will do. https://reviews.llvm.org/D35082 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38821: [COFF, ARM64] Add MS builtins __dmb, __dsb, __isb
mgrang added a comment. LGTM. https://reviews.llvm.org/D38821 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
[PATCH] D38824: [X86] Synchronize the CPU predefined macros with gcc
chandlerc added inline comments. Comment at: lib/Basic/Targets/X86.cpp:844-845 -// FIXME: Historically, we defined this legacy name, it would be nice to -// remove it at some point. We've never exposed fine-grained names for -// recent primary x86 CPUs, and we should keep it that way. -defineCPUMacros(Builder, "corei7"); This seems to undo the idea that we should keep avoiding exposing fine-grained CPU names? What's new that changes this? Comment at: lib/Basic/Targets/X86.cpp:852 +defineCPUMacros(Builder, "core_avx2"); +defineCPUMacros(Builder, "haswell"); break; I find calling a Westmere CPU `nehalem` a little odd. Calling IvyBridge a `sandybridge' CPU seems quite confusing. But calling Skylake (client) and Cannonlake (all? client?) `haswell` seems deeply weird. https://reviews.llvm.org/D38824 ___ cfe-commits mailing list cfe-commits@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
