r252960 - [modules] Simplify and generalize the existing rule for finding hidden

[Richard Smith via cfe-commits]

On Wed, Dec 9, 2015 at 2:00 PM, Ben Langmuir <blangmuir at apple.com> wrote:

>
> On Dec 9, 2015, at 1:36 PM, Richard Smith <richard at metafoo.co.uk> wrote:
>
> On Wed, Dec 9, 2015 at 11:55 AM, Ben Langmuir via cfe-commits <
> cfe-commits at lists.llvm.org> wrote:
>
>> > On Dec 9, 2015, at 11:07 AM, Ben Langmuir <blangmuir at apple.com> wrote:
>> >
>> > Hey Richard,
>> >
>> > This caused a new error for the following code:
>> >
>> >    @import Foo.X; // declaration of ‘struct foo’ from Foo.Y is not
>> visible yet, but the pcm is loaded.
>> >    struct foo *bar; // declares ‘struct foo’
>> >    @import Foo.Y; // also declares ‘struct foo’
>> >
>> >    void useFoo(struct foo *x);  // error: reference to ‘foo’ is
>> ambiguous
>> >
>> > This seems to be specific to declaring the tag with an elaborated type
>> specifier that is not just ‘struct foo;’.  Any idea what went wrong?  I’m
>> trying to track this down and fix it.
>>
>> It’s also specific to non-C++ language modes.  In C++ we seem to cheat
>> and make a second lookup that has ForRedeclaration set (SemaDecl.cpp:12122)
>> which then turns this into a use of the hidden declaration rather than
>> creating a new declaration or redeclaration.  I call this “cheating”
>> because the comments imply this second lookup is for diagnostic purposes,
>> but it clearly has a semantic affect in this case.
>>
>
> Well, this comes back to our handling of C structs and modules being a
> little incoherent. Suppose Foo.Y defines one 'struct foo', and we define a
> completely different 'struct foo':
>
> // Foo.Y
> struct foo { int n; };
>
> // my.c
> @import Foo.X;
> struct foo *bar;
> struct foo { double d; };
> // Don't import Foo.Y
>
> If this is not an error, then our declaration of 'struct foo' cannot be
> the same type as Foo.Y. The usual C rule for structs is that you can have
> multiple conflicting definitions in different TUs. Such declarations
> declare different types, but they're compatible types (and thus can be used
> interchangeably) if they're structurally equivalent.
>
> I think implementing that rule for C is the right way to go here. Either
> that, or we extend C++'s ODR semantics to C modules, but given that the C
> standard gives us a completely reasonable rule here, it seems more
> appropriate to follow it.
>
>
> + Doug
>
> To make sure I understand: if we followed the C rules then in my original
> example we would get a fresh declaration here:
> struct foo *x;
>
> and then when we import
> @import Foo.Y;
>
> we would try to merge the types, and if we had more than one definition we
> would diagnose if they aren’t structurally equivalent.
>

No. At the point of import, the name 'foo' would name two different
structs. If at some later point we look up 'foo' in the tag name space,
name lookup would fold together the lookup results if they refer to
compatible types (and if they're not compatible, you'd get an ambiguity
error at the point of use). Likewise, ASTContext::typesAreCompatible would
be extended to treat such types as compatible in the cases where the C
standard says they are compatible (presumably with a cache so we don't
repeatedly check for the same pair of types).

Whereas if we used the ODR we could use the hidden declaration directly
> like we do in C++,
>

That would cause my example above to silently do the wrong thing. You'd get
a struct foo with a single member 'n', despite no such member being
visible. In C++, the code would already be ill-formed (NDR) due to
violating the ODR, so that's less terrible (but still pretty bad). In C,
where the corresponding code not using modules is valid, such behavior is
pretty hard to defend.

although perhaps looking it up with a for-redeclaration lookup wouldn’t be
> sufficient if we fixed the linkage of the structs as you mention below.
>
> Switching to structural equivalence checking seems like a big change to
> make just to fix this regression.  Even if that’s what we want I’m not sure
> I could sign up for it right now.  Is there a more tactical way we could
> fix this in the interim?
>

Adding the for-redeclaration lookup to C seems fine to me as a short-term
workaround. (This will break again if/when we fix the linkage of C types,
but I don't think we have a reason to do that until we have the type
compatibility rules implemented.)

> I guess the reason this started failing after r252960 is because of this
>> change:
>> > This also removes the prior special case for tag lookup, which made
>> some cases
>> > of this work, but also led to bizarre, bogus "must use 'struct' to
>> refer to type
>> > 'Foo' in this scope" diagnostics in C++.
>>
>> We could do the for-redeclaration lookup outside of C++ too, which will
>> fix this case,
>
>
> I don't think so: structs in C don't have linkage, so the
> for-redeclaration lookup won't find unimported structs (though it's
> possible that we get the linkage calculation for C structs wrong).
>
>
> It does find the struct and the hidden decl is considered “externally
> visible” so I guess that’s incorrect then?  Indeed doing this extra lookup
> does fix my test case (and doesn’t break any other regression tests we
> have).
>

Yes, that seems like a bug. (It would presumably break code such as my
example above...)


>
>
> but I’m not sure if this is the correct fix.
>>
>> Thoughts?
>>
>> Ben
>>
>> >
>> > Ben
>> >
>> >> On Nov 12, 2015, at 2:19 PM, Richard Smith via cfe-commits <
>> cfe-commits at lists.llvm.org> wrote:
>> >>
>> >> Author: rsmith
>> >> Date: Thu Nov 12 16:19:45 2015
>> >> New Revision: 252960
>> >>
>> >> URL: http://llvm.org/viewvc/llvm-project?rev=252960&view=rev
>> >> Log:
>> >> [modules] Simplify and generalize the existing rule for finding hidden
>> >> declarations in redeclaration lookup. A declaration is now visible to
>> >> lookup if:
>> >>
>> >> * It is visible (not in a module, or in an imported module), or
>> >> * We're doing redeclaration lookup and it's externally-visible, or
>> >> * We're doing typo correction and looking for unimported decls.
>> >>
>> >> We now support multiple modules having different internal-linkage or
>> no-linkage
>> >> definitions of the same name for all entities, not just for functions,
>> >> variables, and some typedefs. As previously, if multiple such entities
>> are
>> >> visible, any attempt to use them will result in an ambiguity error.
>> >>
>> >> This patch fixes the linkage calculation for a number of entities
>> where we
>> >> previously didn't need to get it right (using-declarations, namespace
>> aliases,
>> >> and so on).  It also classifies enumerators as always having no
>> linkage, which
>> >> is a slight deviation from the C++ standard's definition, but not an
>> observable
>> >> change outside modules (this change is being discussed on the -core
>> reflector
>> >> currently).
>> >>
>> >> This also removes the prior special case for tag lookup, which made
>> some cases
>> >> of this work, but also led to bizarre, bogus "must use 'struct' to
>> refer to type
>> >> 'Foo' in this scope" diagnostics in C++.
>> >>
>> >> Added:
>> >>   cfe/trunk/test/Modules/Inputs/no-linkage/
>> >>   cfe/trunk/test/Modules/Inputs/no-linkage/decls.h
>> >>   cfe/trunk/test/Modules/Inputs/no-linkage/empty.h
>> >>   cfe/trunk/test/Modules/Inputs/no-linkage/module.modulemap
>> >>   cfe/trunk/test/Modules/no-linkage.cpp
>> >> Modified:
>> >>   cfe/trunk/include/clang/Sema/Lookup.h
>> >>   cfe/trunk/lib/AST/Decl.cpp
>> >>   cfe/trunk/lib/Sema/SemaDecl.cpp
>> >>   cfe/trunk/lib/Sema/SemaDeclCXX.cpp
>> >>   cfe/trunk/test/Index/linkage.c
>> >>   cfe/trunk/test/Index/usrs.m
>> >>   cfe/trunk/test/Modules/decldef.m
>> >>   cfe/trunk/test/Modules/merge-enumerators.cpp
>> >>   cfe/trunk/test/Modules/module-private.cpp
>> >>   cfe/trunk/test/Modules/submodule-visibility-cycles.cpp
>> >>   cfe/trunk/test/Modules/submodules-merge-defs.cpp
>> >>
>> >> Modified: cfe/trunk/include/clang/Sema/Lookup.h
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/include/clang/Sema/Lookup.h?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/include/clang/Sema/Lookup.h (original)
>> >> +++ cfe/trunk/include/clang/Sema/Lookup.h Thu Nov 12 16:19:45 2015
>> >> @@ -139,8 +139,7 @@ public:
>> >>      Redecl(Redecl != Sema::NotForRedeclaration),
>> >>      HideTags(true),
>> >>      Diagnose(Redecl == Sema::NotForRedeclaration),
>> >> -      AllowHidden(Redecl == Sema::ForRedeclaration),
>> >> -      AllowHiddenInternal(AllowHidden),
>> >> +      AllowHidden(false),
>> >>      Shadowed(false)
>> >>  {
>> >>    configure();
>> >> @@ -162,8 +161,7 @@ public:
>> >>      Redecl(Redecl != Sema::NotForRedeclaration),
>> >>      HideTags(true),
>> >>      Diagnose(Redecl == Sema::NotForRedeclaration),
>> >> -      AllowHidden(Redecl == Sema::ForRedeclaration),
>> >> -      AllowHiddenInternal(AllowHidden),
>> >> +      AllowHidden(false),
>> >>      Shadowed(false)
>> >>  {
>> >>    configure();
>> >> @@ -184,7 +182,6 @@ public:
>> >>      HideTags(Other.HideTags),
>> >>      Diagnose(false),
>> >>      AllowHidden(Other.AllowHidden),
>> >> -      AllowHiddenInternal(Other.AllowHiddenInternal),
>> >>      Shadowed(false)
>> >>  {}
>> >>
>> >> @@ -226,27 +223,16 @@ public:
>> >>  /// \brief Specify whether hidden declarations are visible, e.g.,
>> >>  /// for recovery reasons.
>> >>  void setAllowHidden(bool AH) {
>> >> -    AllowHiddenInternal = AllowHidden = AH;
>> >> -  }
>> >> -
>> >> -  /// \brief Specify whether hidden internal declarations are visible.
>> >> -  void setAllowHiddenInternal(bool AHI) {
>> >> -    AllowHiddenInternal = AHI;
>> >> +    AllowHidden = AH;
>> >>  }
>> >>
>> >>  /// \brief Determine whether this lookup is permitted to see hidden
>> >>  /// declarations, such as those in modules that have not yet been
>> imported.
>> >>  bool isHiddenDeclarationVisible(NamedDecl *ND) const {
>> >> -    // If a using-shadow declaration is hidden, it's never visible,
>> not
>> >> -    // even to redeclaration lookup.
>> >> -    // FIXME: Should this apply to typedefs and namespace aliases too?
>> >> -    if (isa<UsingShadowDecl>(ND) && LookupKind !=
>> Sema::LookupUsingDeclName)
>> >> -      return false;
>> >> -    return (AllowHidden &&
>> >> -            (AllowHiddenInternal || ND->isExternallyVisible())) ||
>> >> -           LookupKind == Sema::LookupTagName;
>> >> +    return AllowHidden ||
>> >> +           (isForRedeclaration() && ND->isExternallyVisible());
>> >>  }
>> >> -
>> >> +
>> >>  /// Sets whether tag declarations should be hidden by non-tag
>> >>  /// declarations during resolution.  The default is true.
>> >>  void setHideTags(bool Hide) {
>> >> @@ -317,7 +303,8 @@ public:
>> >>    if (!D->isInIdentifierNamespace(IDNS))
>> >>      return nullptr;
>> >>
>> >> -    if (isHiddenDeclarationVisible(D) || isVisible(getSema(), D))
>> >> +    if (!D->isHidden() || isHiddenDeclarationVisible(D) ||
>> >> +        isVisibleSlow(getSema(), D))
>> >>      return D;
>> >>
>> >>    return getAcceptableDeclSlow(D);
>> >> @@ -526,7 +513,6 @@ public:
>> >>  /// \brief Change this lookup's redeclaration kind.
>> >>  void setRedeclarationKind(Sema::RedeclarationKind RK) {
>> >>    Redecl = RK;
>> >> -    AllowHiddenInternal = AllowHidden = (RK ==
>> Sema::ForRedeclaration);
>> >>    configure();
>> >>  }
>> >>
>> >> @@ -698,9 +684,6 @@ private:
>> >>  /// \brief True if we should allow hidden declarations to be
>> 'visible'.
>> >>  bool AllowHidden;
>> >>
>> >> -  /// \brief True if we should allow hidden internal declarations to
>> be visible.
>> >> -  bool AllowHiddenInternal;
>> >> -
>> >>  /// \brief True if the found declarations were shadowed by some other
>> >>  /// declaration that we skipped. This only happens when \c LookupKind
>> >>  /// is \c LookupRedeclarationWithLinkage.
>> >>
>> >> Modified: cfe/trunk/lib/AST/Decl.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/AST/Decl.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/lib/AST/Decl.cpp (original)
>> >> +++ cfe/trunk/lib/AST/Decl.cpp Thu Nov 12 16:19:45 2015
>> >> @@ -635,6 +635,8 @@ static LinkageInfo getLVForNamespaceScop
>> >>  if (D->isInAnonymousNamespace()) {
>> >>    const auto *Var = dyn_cast<VarDecl>(D);
>> >>    const auto *Func = dyn_cast<FunctionDecl>(D);
>> >> +    // FIXME: In C++11 onwards, anonymous namespaces should give decls
>> >> +    // within them internal linkage, not unique external linkage.
>> >>    if ((!Var || !isFirstInExternCContext(Var)) &&
>> >>        (!Func || !isFirstInExternCContext(Func)))
>> >>      return LinkageInfo::uniqueExternal();
>> >> @@ -821,10 +823,14 @@ static LinkageInfo getLVForNamespaceScop
>> >>  } else if (isa<ObjCInterfaceDecl>(D)) {
>> >>    // fallout
>> >>
>> >> +  } else if (auto *TD = dyn_cast<TypedefNameDecl>(D)) {
>> >> +    // A typedef declaration has linkage if it gives a type a name for
>> >> +    // linkage purposes.
>> >> +    if (!TD->getAnonDeclWithTypedefName(/*AnyRedecl*/true))
>> >> +      return LinkageInfo::none();
>> >> +
>> >>  // Everything not covered here has no linkage.
>> >>  } else {
>> >> -    // FIXME: A typedef declaration has linkage if it gives a type a
>> name for
>> >> -    // linkage purposes.
>> >>    return LinkageInfo::none();
>> >>  }
>> >>
>> >> @@ -1226,8 +1232,32 @@ static LinkageInfo computeLVForDecl(cons
>> >>  switch (D->getKind()) {
>> >>    default:
>> >>      break;
>> >> +
>> >> +    // Per C++ [basic.link]p2, only the names of objects, references,
>> >> +    // functions, types, templates, namespaces, and values ever have
>> linkage.
>> >> +    //
>> >> +    // Note that the name of a typedef, namespace alias, using
>> declaration,
>> >> +    // and so on are not the name of the corresponding type,
>> namespace, or
>> >> +    // declaration, so they do *not* have linkage.
>> >> +    case Decl::EnumConstant: // FIXME: This has linkage, but that's
>> dumb.
>> >> +    case Decl::ImplicitParam:
>> >> +    case Decl::Label:
>> >> +    case Decl::NamespaceAlias:
>> >>    case Decl::ParmVar:
>> >> +    case Decl::Using:
>> >> +    case Decl::UsingShadow:
>> >> +    case Decl::UsingDirective:
>> >>      return LinkageInfo::none();
>> >> +
>> >> +    case Decl::Typedef:
>> >> +    case Decl::TypeAlias:
>> >> +      // A typedef declaration has linkage if it gives a type a name
>> for
>> >> +      // linkage purposes.
>> >> +      if (!cast<TypedefNameDecl>(D)
>> >> +               ->getAnonDeclWithTypedefName(/*AnyRedecl*/true))
>> >> +        return LinkageInfo::none();
>> >> +      break;
>> >> +
>> >>    case Decl::TemplateTemplateParm: // count these as external
>> >>    case Decl::NonTypeTemplateParm:
>> >>    case Decl::ObjCAtDefsField:
>> >>
>> >> Modified: cfe/trunk/lib/Sema/SemaDecl.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Sema/SemaDecl.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/lib/Sema/SemaDecl.cpp (original)
>> >> +++ cfe/trunk/lib/Sema/SemaDecl.cpp Thu Nov 12 16:19:45 2015
>> >> @@ -4819,12 +4819,6 @@ NamedDecl *Sema::HandleDeclarator(Scope
>> >>  LookupResult Previous(*this, NameInfo, LookupOrdinaryName,
>> >>                        ForRedeclaration);
>> >>
>> >> -  // If we're hiding internal-linkage symbols in modules from
>> redeclaration
>> >> -  // lookup, let name lookup know.
>> >> -  if ((getLangOpts().Modules || getLangOpts().ModulesLocalVisibility)
>> &&
>> >> -      D.getDeclSpec().getStorageClassSpec() != DeclSpec::SCS_typedef)
>> >> -    Previous.setAllowHiddenInternal(false);
>> >> -
>> >>  // See if this is a redefinition of a variable in the same scope.
>> >>  if (!D.getCXXScopeSpec().isSet()) {
>> >>    bool IsLinkageLookup = false;
>> >>
>> >> Modified: cfe/trunk/lib/Sema/SemaDeclCXX.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/lib/Sema/SemaDeclCXX.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/lib/Sema/SemaDeclCXX.cpp (original)
>> >> +++ cfe/trunk/lib/Sema/SemaDeclCXX.cpp Thu Nov 12 16:19:45 2015
>> >> @@ -7206,23 +7206,13 @@ Decl *Sema::ActOnStartNamespaceDef(Scope
>> >>    //   treated as an original-namespace-name.
>> >>    //
>> >>    // Since namespace names are unique in their scope, and we don't
>> >> -    // look through using directives, just look for any ordinary
>> names.
>> >> -
>> >> -    const unsigned IDNS = Decl::IDNS_Ordinary | Decl::IDNS_Member |
>> >> -    Decl::IDNS_Type | Decl::IDNS_Using | Decl::IDNS_Tag |
>> >> -    Decl::IDNS_Namespace;
>> >> -    NamedDecl *PrevDecl = nullptr;
>> >> -    DeclContext::lookup_result R =
>> CurContext->getRedeclContext()->lookup(II);
>> >> -    for (DeclContext::lookup_iterator I = R.begin(), E = R.end(); I
>> != E;
>> >> -         ++I) {
>> >> -      if ((*I)->getIdentifierNamespace() & IDNS) {
>> >> -        PrevDecl = *I;
>> >> -        break;
>> >> -      }
>> >> -    }
>> >> -
>> >> +    // look through using directives, just look for any ordinary names
>> >> +    // as if by qualified name lookup.
>> >> +    LookupResult R(*this, II, IdentLoc, LookupOrdinaryName,
>> ForRedeclaration);
>> >> +    LookupQualifiedName(R, CurContext->getRedeclContext());
>> >> +    NamedDecl *PrevDecl = R.getAsSingle<NamedDecl>();
>> >>    PrevNS = dyn_cast_or_null<NamespaceDecl>(PrevDecl);
>> >> -
>> >> +
>> >>    if (PrevNS) {
>> >>      // This is an extended namespace definition.
>> >>      if (IsInline != PrevNS->isInline())
>> >>
>> >> Modified: cfe/trunk/test/Index/linkage.c
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Index/linkage.c?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Index/linkage.c (original)
>> >> +++ cfe/trunk/test/Index/linkage.c Thu Nov 12 16:19:45 2015
>> >> @@ -20,7 +20,7 @@ void f16(void) {
>> >>
>> >>
>> >> // CHECK: EnumDecl=Baz:3:6 (Definition)linkage=External
>> >> -// CHECK: EnumConstantDecl=Qux:3:12 (Definition)linkage=External
>> >> +// CHECK: EnumConstantDecl=Qux:3:12 (Definition)linkage=NoLinkage
>> >> // CHECK: VarDecl=x:4:5linkage=External
>> >> // CHECK: FunctionDecl=foo:5:6linkage=External
>> >> // CHECK: VarDecl=w:6:12linkage=Internal
>> >>
>> >> Modified: cfe/trunk/test/Index/usrs.m
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Index/usrs.m?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Index/usrs.m (original)
>> >> +++ cfe/trunk/test/Index/usrs.m Thu Nov 12 16:19:45 2015
>> >> @@ -119,7 +119,7 @@ int test_multi_declaration(void) {
>> >> // CHECK: usrs.m c:@SA at MyStruct Extent=[15:9 - 18:2]
>> >> // CHECK: usrs.m c:@SA at MyStruct@FI at wa Extent=[16:3 - 16:9]
>> >> // CHECK: usrs.m c:@SA at MyStruct@FI at moo Extent=[17:3 - 17:10]
>> >> -// CHECK: usrs.m c:usrs.m at T@MyStruct Extent=[15:1 - 18:11]
>> >> +// CHECK: usrs.m c:@T at MyStruct Extent=[15:1 - 18:11]
>> >> // CHECK: usrs.m c:@E at Pizza Extent=[20:1 - 23:2]
>> >> // CHECK: usrs.m c:@E at Pizza@CHEESE Extent=[21:3 - 21:9]
>> >> // CHECK: usrs.m c:@E at Pizza@MUSHROOMS Extent=[22:3 - 22:12]
>> >>
>> >> Added: cfe/trunk/test/Modules/Inputs/no-linkage/decls.h
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/Inputs/no-linkage/decls.h?rev=252960&view=auto
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/Inputs/no-linkage/decls.h (added)
>> >> +++ cfe/trunk/test/Modules/Inputs/no-linkage/decls.h Thu Nov 12
>> 16:19:45 2015
>> >> @@ -0,0 +1,6 @@
>> >> +namespace RealNS { int UsingDecl; }
>> >> +namespace NS = RealNS;
>> >> +typedef int Typedef;
>> >> +using AliasDecl = int;
>> >> +enum Enum { Enumerator };
>> >> +using RealNS::UsingDecl;
>> >>
>> >> Added: cfe/trunk/test/Modules/Inputs/no-linkage/empty.h
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/Inputs/no-linkage/empty.h?rev=252960&view=auto
>> >>
>> ==============================================================================
>> >>   (empty)
>> >>
>> >> Added: cfe/trunk/test/Modules/Inputs/no-linkage/module.modulemap
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/Inputs/no-linkage/module.modulemap?rev=252960&view=auto
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/Inputs/no-linkage/module.modulemap (added)
>> >> +++ cfe/trunk/test/Modules/Inputs/no-linkage/module.modulemap Thu Nov
>> 12 16:19:45 2015
>> >> @@ -0,0 +1 @@
>> >> +module M { module E { header "empty.h" } module D { header "decls.h"
>> } }
>> >>
>> >> Modified: cfe/trunk/test/Modules/decldef.m
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/decldef.m?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/decldef.m (original)
>> >> +++ cfe/trunk/test/Modules/decldef.m Thu Nov 12 16:19:45 2015
>> >> @@ -11,7 +11,13 @@ Def *def;
>> >> #ifdef USE_EARLY
>> >> A *a1; // expected-error{{declaration of 'A' must be imported from
>> module 'decldef.Def' before it is required}}
>> >> #endif
>> >> -B *b1; // expected-error{{must use 'struct' tag to refer to type 'B'}}
>> >> +B *b1;
>> >> +#ifdef USE_EARLY
>> >> +// expected-error at -2{{must use 'struct' tag to refer to type 'B'}}
>> >> +#else
>> >> +// expected-error at -4{{declaration of 'B' must be imported from
>> module 'decldef.Decl' before it is required}}
>> >> +// expected-note at Inputs/decl.h:2 {{previous}}
>> >> +#endif
>> >> @import decldef.Decl;
>> >>
>> >> A *a2;
>> >>
>> >> Modified: cfe/trunk/test/Modules/merge-enumerators.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/merge-enumerators.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/merge-enumerators.cpp (original)
>> >> +++ cfe/trunk/test/Modules/merge-enumerators.cpp Thu Nov 12 16:19:45
>> 2015
>> >> @@ -16,11 +16,13 @@
>> >>
>> >> #ifdef MERGE_LATE
>> >> namespace N {
>> >> -  // FIXME: Should we accept this and reject the usage below due to
>> ambiguity instead?
>> >> -  enum { A } a; // expected-error {{redefinition of enumerator 'A'}}
>> >> -  // expected-note at a.h:1 {{here}} (from module B.b)
>> >> +  enum { A } a; // expected-note {{candidate}}
>> >> +  // expected-note at a.h:1 {{candidate}} (from module B.b)
>> >> }
>> >> #include "a.h"
>> >> #endif
>> >>
>> >> N::E e = N::A;
>> >> +#ifdef MERGE_LATE
>> >> +// expected-error at -2 {{ambiguous}}
>> >> +#endif
>> >>
>> >> Modified: cfe/trunk/test/Modules/module-private.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/module-private.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/module-private.cpp (original)
>> >> +++ cfe/trunk/test/Modules/module-private.cpp Thu Nov 12 16:19:45 2015
>> >> @@ -12,11 +12,7 @@ void test() {
>> >> }
>> >>
>> >> int test_broken() {
>> >> -  HiddenStruct hidden; // \
>> >> -  // expected-error{{must use 'struct' tag to refer to type
>> 'HiddenStruct' in this scope}} \
>> >> -  // expected-error{{definition of 'HiddenStruct' must be imported}}
>> >> -  // expected-note at Inputs/module_private_left.h:3 {{previous
>> definition is here}}
>> >> -
>> >> +  HiddenStruct hidden; // expected-error{{unknown type name
>> 'HiddenStruct'}}
>> >>  Integer i; // expected-error{{unknown type name 'Integer'}}
>> >>
>> >>  int *ip = 0;
>> >>
>> >> Added: cfe/trunk/test/Modules/no-linkage.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/no-linkage.cpp?rev=252960&view=auto
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/no-linkage.cpp (added)
>> >> +++ cfe/trunk/test/Modules/no-linkage.cpp Thu Nov 12 16:19:45 2015
>> >> @@ -0,0 +1,35 @@
>> >> +// RUN: rm -rf %t
>> >> +// RUN: %clang_cc1 -fmodules -fmodules-cache-path=%t
>> -fmodules-local-submodule-visibility -I%S/Inputs/no-linkage
>> -fmodule-map-file=%S/Inputs/no-linkage/module.modulemap %s -verify
>> >> +
>> >> +#include "empty.h"
>> >> +
>> >> +namespace NS { int n; } // expected-note {{candidate}}
>> >> +struct Typedef { int n; }; // expected-note {{candidate}}
>> >> +int AliasDecl; // expected-note {{candidate}}
>> >> +enum AlsoAnEnum { Enumerator }; // expected-note {{candidate}}
>> >> +int UsingDecl; // expected-note {{candidate}}
>> >> +
>> >> +// expected-note at decls.h:2 {{candidate}}
>> >> +// expected-note at decls.h:3 {{candidate}}
>> >> +// expected-note at decls.h:4 {{candidate}}
>> >> +// expected-note at decls.h:5 {{candidate}}
>> >> +// expected-note at decls.h:6 {{candidate}}
>> >> +
>> >> +void use(int);
>> >> +void use_things() {
>> >> +  use(Typedef().n);
>> >> +  use(NS::n);
>> >> +  use(AliasDecl);
>> >> +  use(Enumerator);
>> >> +  use(UsingDecl);
>> >> +}
>> >> +
>> >> +#include "decls.h"
>> >> +
>> >> +void use_things_again() {
>> >> +  use(Typedef().n); // expected-error {{ambiguous}}
>> >> +  use(NS::n); // expected-error {{ambiguous}} expected-error{{'NS' is
>> not a class, namespace, or enumeration}}
>> >> +  use(AliasDecl); // expected-error {{ambiguous}}
>> >> +  use(Enumerator); // expected-error {{ambiguous}}
>> >> +  use(UsingDecl); // expected-error {{ambiguous}}
>> >> +}
>> >>
>> >> Modified: cfe/trunk/test/Modules/submodule-visibility-cycles.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/submodule-visibility-cycles.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/submodule-visibility-cycles.cpp (original)
>> >> +++ cfe/trunk/test/Modules/submodule-visibility-cycles.cpp Thu Nov 12
>> 16:19:45 2015
>> >> @@ -3,7 +3,7 @@
>> >>
>> >> #include "cycle1.h"
>> >> C1 c1;
>> >> -C2 c2; // expected-error {{must be imported}} expected-error {{}}
>> >> +C2 c2; // expected-error {{must be imported}}
>> >> // expected-note at cycle2.h:6 {{here}}
>> >>
>> >> #include "cycle2.h"
>> >>
>> >> Modified: cfe/trunk/test/Modules/submodules-merge-defs.cpp
>> >> URL:
>> http://llvm.org/viewvc/llvm-project/cfe/trunk/test/Modules/submodules-merge-defs.cpp?rev=252960&r1=252959&r2=252960&view=diff
>> >>
>> ==============================================================================
>> >> --- cfe/trunk/test/Modules/submodules-merge-defs.cpp (original)
>> >> +++ cfe/trunk/test/Modules/submodules-merge-defs.cpp Thu Nov 12
>> 16:19:45 2015
>> >> @@ -12,7 +12,7 @@
>> >> #include "indirect.h"
>> >> #endif
>> >>
>> >> -A pre_a; // expected-error {{must use 'struct'}}
>> >> +A pre_a;
>> >> #ifdef IMPORT_USE_2
>> >> // expected-error-re at -2 {{must be imported from one of
>> {{.*}}stuff.use{{.*}}stuff.use-2}}
>> >> #elif EARLY_INDIRECT_INCLUDE
>> >> @@ -21,29 +21,28 @@ A pre_a; // expected-error {{must use 's
>> >> // expected-error at -6 {{must be imported from module 'stuff.use'}}
>> >> #endif
>> >> // expected-note at defs.h:1 +{{here}}
>> >> +extern class A pre_a2;
>> >> +int pre_use_a = use_a(pre_a2); // expected-error {{'A' must be
>> imported}} expected-error {{'use_a' must be imported}}
>> >> // expected-note at defs.h:2 +{{here}}
>> >> -int pre_use_a = use_a(pre_a); // expected-error {{'A' must be
>> imported}} expected-error {{'use_a' must be imported}}
>> >>
>> >> B::Inner2 pre_bi; // expected-error +{{must be imported}}
>> >> // expected-note at defs.h:4 +{{here}}
>> >> // expected-note at defs.h:17 +{{here}}
>> >> -void pre_bfi(B b) { // expected-error {{must use 'class'}}
>> expected-error +{{must be imported}}
>> >> -  b.f<int>(); // expected-error +{{must be imported}} expected-error
>> +{{}}
>> >> -  // expected-note at defs.h:19 +{{here}}
>> >> +void pre_bfi(B b) { // expected-error +{{must be imported}}
>> >> +  b.f<int>(); // expected-error +{{}}
>> >> }
>> >>
>> >> C_Base<1> pre_cb1; // expected-error +{{must be imported}}
>> >> // expected-note at defs.h:23 +{{here}}
>> >> -C1 pre_c1; // expected-error +{{must be imported}} expected-error
>> {{must use 'struct'}}
>> >> +C1 pre_c1; // expected-error +{{must be imported}}
>> >> // expected-note at defs.h:25 +{{here}}
>> >> -C2 pre_c2; // expected-error +{{must be imported}} expected-error
>> {{must use 'struct'}}
>> >> +C2 pre_c2; // expected-error +{{must be imported}}
>> >> // expected-note at defs.h:26 +{{here}}
>> >>
>> >> D::X pre_dx; // expected-error +{{must be imported}}
>> >> // expected-note at defs.h:28 +{{here}}
>> >> // expected-note at defs.h:29 +{{here}}
>> >> -// FIXME: We should warn that use_dx is being used without being
>> imported.
>> >> -int pre_use_dx = use_dx(pre_dx);
>> >> +int pre_use_dx = use_dx(pre_dx); // ignored; pre_dx is invalid
>> >>
>> >> int pre_e = E(0); // expected-error {{must be imported}}
>> >> // expected-note at defs.h:32 +{{here}}
>> >> @@ -69,8 +68,9 @@ J<> pre_j; // expected-error {{declarati
>> >> #endif
>> >> // expected-note at defs.h:58 +{{here}}
>> >>
>> >> -ScopedEnum pre_scopedenum; // expected-error {{must be imported}}
>> expected-error {{must use 'enum'}}
>> >> -// expected-note at defs.h:106 {{here}}
>> >> +ScopedEnum pre_scopedenum; // expected-error {{must be imported}}
>> >> +// expected-note at defs.h:105 0-1{{here}}
>> >> +// expected-note at defs.h:106 0-1{{here}}
>> >> enum ScopedEnum : int;
>> >> ScopedEnum pre_scopedenum_declared; // ok
>> >>
>> >>
>> >>
>> >> _______________________________________________
>> >> cfe-commits mailing list
>> >> cfe-commits at lists.llvm.org
>> >> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
>> >
>>
>> _______________________________________________
>> cfe-commits mailing list
>> cfe-commits at lists.llvm.org
>> http://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
>>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20151209/d6140429/attachment-0001.html>