llvm-project/clang/test/SemaCXX/attr-selectany.cpp

53 lines
2.1 KiB
C++
Raw Normal View History

Move fixit for const init from note to diag, weaken to warning in MS mode. r235046 turned "extern __declspec(selectany) int a;" from a declaration into a definition to fix PR23242 (required for compatibility with mc.exe output). However, this broke parsing Windows headers: A d3d11 headers contain something like struct SomeStruct {}; extern const __declspec(selectany) SomeStruct some_struct; This is now a definition, and const objects either need an explicit default ctor or an initializer so this errors out with d3d11.h(1065,48) : error: default initialization of an object of const type 'const CD3D11_DEFAULT' without a user-provided default constructor (cl.exe just doesn't implement this rule, independent of selectany.) To work around this, weaken this error into a warning for selectany decls in microsoft mode, and recover with zero-initialization. Doing this is a bit hairy since it adds a fixit on an error emitted by InitializationSequence – this means it needs to build a correct AST, which in turn means InitializationSequence::Failed() cannot return true when this fixit is applied. As a workaround, the patch adds a fixit member to InitializationSequence, and InitializationSequence::Perform() prints the diagnostic if the fixit member is set right after its call to Diagnose. That function is usually called when InitializationSequences are used – InitListChecker::PerformEmptyInit() doesn't call it, but the InitListChecker case never performs default-initialization, so this is technically OK. This is the alternative, original fix for PR20208 that got reviewed in the thread "[patch] Improve diagnostic on default-initializing const variables (PR20208)". This change basically reverts r213725, adds the original fix for PR20208, and makes the error a warning in Microsoft mode. llvm-svn: 235166
2015-04-17 16:32:38 +08:00
// RUN: %clang_cc1 -fms-compatibility -fms-extensions -fsyntax-only -verify -std=c++11 %s
// MSVC produces similar diagnostics.
__declspec(selectany) void foo() { } // expected-error{{'selectany' can only be applied to data items with external linkage}}
__declspec(selectany) int x1 = 1;
const __declspec(selectany) int x2 = 2; // expected-error{{'selectany' can only be applied to data items with external linkage}}
extern const __declspec(selectany) int x3 = 3;
extern const int x4;
const __declspec(selectany) int x4 = 4;
// MSDN says this is incorrect, but MSVC doesn't diagnose it.
extern __declspec(selectany) int x5;
static __declspec(selectany) int x6 = 2; // expected-error{{'selectany' can only be applied to data items with external linkage}}
// FIXME: MSVC accepts this and makes x7 externally visible and comdat, but keep
// it as internal and not weak/linkonce.
static int x7; // expected-note{{previous definition}}
extern __declspec(selectany) int x7; // expected-warning{{attribute declaration must precede definition}}
int asdf() { return x7; }
class X {
public:
X(int i) { i++; };
int i;
};
__declspec(selectany) X x(1);
namespace { class Internal {}; }
__declspec(selectany) auto x8 = Internal(); // expected-error {{'selectany' can only be applied to data items with external linkage}}
Move fixit for const init from note to diag, weaken to warning in MS mode. r235046 turned "extern __declspec(selectany) int a;" from a declaration into a definition to fix PR23242 (required for compatibility with mc.exe output). However, this broke parsing Windows headers: A d3d11 headers contain something like struct SomeStruct {}; extern const __declspec(selectany) SomeStruct some_struct; This is now a definition, and const objects either need an explicit default ctor or an initializer so this errors out with d3d11.h(1065,48) : error: default initialization of an object of const type 'const CD3D11_DEFAULT' without a user-provided default constructor (cl.exe just doesn't implement this rule, independent of selectany.) To work around this, weaken this error into a warning for selectany decls in microsoft mode, and recover with zero-initialization. Doing this is a bit hairy since it adds a fixit on an error emitted by InitializationSequence – this means it needs to build a correct AST, which in turn means InitializationSequence::Failed() cannot return true when this fixit is applied. As a workaround, the patch adds a fixit member to InitializationSequence, and InitializationSequence::Perform() prints the diagnostic if the fixit member is set right after its call to Diagnose. That function is usually called when InitializationSequences are used – InitListChecker::PerformEmptyInit() doesn't call it, but the InitListChecker case never performs default-initialization, so this is technically OK. This is the alternative, original fix for PR20208 that got reviewed in the thread "[patch] Improve diagnostic on default-initializing const variables (PR20208)". This change basically reverts r213725, adds the original fix for PR20208, and makes the error a warning in Microsoft mode. llvm-svn: 235166
2015-04-17 16:32:38 +08:00
// The D3D11 headers do something like this. MSVC doesn't error on this at
// all, even without the __declspec(selectany), in violation of the standard.
// We fall back to a warning for selectany to accept headers.
struct SomeStruct {
int foo;
};
Move fixit for const init from note to diag, weaken to warning in MS mode. r235046 turned "extern __declspec(selectany) int a;" from a declaration into a definition to fix PR23242 (required for compatibility with mc.exe output). However, this broke parsing Windows headers: A d3d11 headers contain something like struct SomeStruct {}; extern const __declspec(selectany) SomeStruct some_struct; This is now a definition, and const objects either need an explicit default ctor or an initializer so this errors out with d3d11.h(1065,48) : error: default initialization of an object of const type 'const CD3D11_DEFAULT' without a user-provided default constructor (cl.exe just doesn't implement this rule, independent of selectany.) To work around this, weaken this error into a warning for selectany decls in microsoft mode, and recover with zero-initialization. Doing this is a bit hairy since it adds a fixit on an error emitted by InitializationSequence – this means it needs to build a correct AST, which in turn means InitializationSequence::Failed() cannot return true when this fixit is applied. As a workaround, the patch adds a fixit member to InitializationSequence, and InitializationSequence::Perform() prints the diagnostic if the fixit member is set right after its call to Diagnose. That function is usually called when InitializationSequences are used – InitListChecker::PerformEmptyInit() doesn't call it, but the InitListChecker case never performs default-initialization, so this is technically OK. This is the alternative, original fix for PR20208 that got reviewed in the thread "[patch] Improve diagnostic on default-initializing const variables (PR20208)". This change basically reverts r213725, adds the original fix for PR20208, and makes the error a warning in Microsoft mode. llvm-svn: 235166
2015-04-17 16:32:38 +08:00
extern const __declspec(selectany) SomeStruct some_struct; // expected-warning {{default initialization of an object of const type 'const SomeStruct' without a user-provided default constructor is a Microsoft extension}}
// It should be possible to redeclare variables that were defined
// __declspec(selectany) previously.
extern const SomeStruct some_struct;
Move fixit for const init from note to diag, weaken to warning in MS mode. r235046 turned "extern __declspec(selectany) int a;" from a declaration into a definition to fix PR23242 (required for compatibility with mc.exe output). However, this broke parsing Windows headers: A d3d11 headers contain something like struct SomeStruct {}; extern const __declspec(selectany) SomeStruct some_struct; This is now a definition, and const objects either need an explicit default ctor or an initializer so this errors out with d3d11.h(1065,48) : error: default initialization of an object of const type 'const CD3D11_DEFAULT' without a user-provided default constructor (cl.exe just doesn't implement this rule, independent of selectany.) To work around this, weaken this error into a warning for selectany decls in microsoft mode, and recover with zero-initialization. Doing this is a bit hairy since it adds a fixit on an error emitted by InitializationSequence – this means it needs to build a correct AST, which in turn means InitializationSequence::Failed() cannot return true when this fixit is applied. As a workaround, the patch adds a fixit member to InitializationSequence, and InitializationSequence::Perform() prints the diagnostic if the fixit member is set right after its call to Diagnose. That function is usually called when InitializationSequences are used – InitListChecker::PerformEmptyInit() doesn't call it, but the InitListChecker case never performs default-initialization, so this is technically OK. This is the alternative, original fix for PR20208 that got reviewed in the thread "[patch] Improve diagnostic on default-initializing const variables (PR20208)". This change basically reverts r213725, adds the original fix for PR20208, and makes the error a warning in Microsoft mode. llvm-svn: 235166
2015-04-17 16:32:38 +08:00
// Without selectany, this should stay an error.
const SomeStruct some_struct2; // expected-error {{default initialization of an object of const type 'const SomeStruct' without a user-provided default constructor}}