This reverts commit 7c51a128.
Breaks SemaCXX/modules-ts.cppm in check-clang.

According to [basic.namespace.general]/p2, a namespace declaration
shouldn't have a module linkage.
> A namespace is never attached to a named module and never has a name
> with module linkage.

Without this patch, the compiler would crash for the test in assertion
enabled build due to inconsistent linkage for redeclaration for
namespaces.

Reviewed by: rsmith

Differential Revision: https://reviews.llvm.org/D115132

Based on post-commit review discussion on
2bd84938 with Richard Smith.

Other uses of forcing HasEmptyPlaceHolder to false seem OK to me -
they're all around pointer/reference types where the pointer/reference
token will appear at the rightmost side of the left side of the type
name, so they make nested types (eg: the "int" in "int *") behave as
though there is a non-empty placeholder (because the "*" is essentially
the placeholder as far as the "int" is concerned).

This was originally committed in 277623f4

Reverted in f9ad1d1c due to breakages
outside of clang - lldb seems to have some strange/strong dependence on
"char [N]" versus "char[N]" when printing strings (not due to that name
appearing in DWARF, but probably due to using clang to stringify type
names) that'll need to be addressed, plus a few other odds and ends in
other subprojects (clang-tools-extra, compiler-rt, etc).

Looks like lldb has some issues with this - somehow it causes lldb to
treat a "char[N]" type as an array of chars (prints them out
individually) but a "char [N]" is printed as a string. (even though the
DWARF doesn't have this string in it - it's something to do with the
string lldb generates for itself using clang)

This reverts commit 277623f4.

Based on post-commit review discussion on
2bd84938 with Richard Smith.

Other uses of forcing HasEmptyPlaceHolder to false seem OK to me -
they're all around pointer/reference types where the pointer/reference
token will appear at the rightmost side of the left side of the type
name, so they make nested types (eg: the "int" in "int *") behave as
though there is a non-empty placeholder (because the "*" is essentially
the placeholder as far as the "int" is concerned).

(This relands 59337263 and makes sure comma operator
 diagnostics are suppressed in a SFINAE context.)

While at it, add the diagnosis message "left operand of comma operator has no effect" (used by GCC) for comma operator.

This also makes Clang diagnose in the constant evaluation context which aligns with GCC/MSVC behavior. (https://godbolt.org/z/7zxb8Tx96)

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D103938

This reverts commit cbbf2e8c.
It seems causing diagnoses in SFINAE context.

While at it, add the diagnosis message "left operand of comma operator has no effect" (used by GCC) for comma operator.

This also makes Clang diagnose in the constant evaluation context which aligns with GCC/MSVC behavior. (https://godbolt.org/z/7zxb8Tx96)

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D103938

This reverts commit 63e0d038.

It causes test failures:

http://lab.llvm.org:8011/#/builders/119/builds/5612
https://logs.chromium.org/logs/fuchsia/buildbucket/cr-buildbucket/8835548361443044001/+/u/clang/test/stdout

While at it, add the diagnosis message "left operand of comma operator has no effect" (used by GCC) for comma operator.

This also makes Clang diagnose in the constant evaluation context which aligns with GCC/MSVC behavior. (https://godbolt.org/z/7zxb8Tx96)

Reviewed By: aaron.ballman

Differential Revision: https://reviews.llvm.org/D103938

Since these map to the same effective type - render them the same/in the
more legible way (const x[n]).

In an unevaluated contexts, consteval functions should not be
immediately evaluated.

552c6c23 removed support for promoting VLAs to constant arrays when the bounds
isn't an ICE, since this can result in miscompiling a conforming program that
assumes that the array is a VLA. Promoting VLAs for fields is still supported,
since clang doesn't support VLAs in fields, so no conforming program could have
a field VLA.

This change is really disruptive, so this commit carves out two more cases
where we promote VLAs which can't miscompile a conforming program:

 - When the VLA appears in an ivar -- this seems like a corollary to the field thing
 - When the VLA has an initializer -- VLAs can't have an initializer

Differential revision: https://reviews.llvm.org/D90871

classes into the enclosing block scope.

We weren't properly detecting whether the name would be injected into a
block scope in the case where it was lexically declared in a local
class.

Old GCC used to aggressively fold VLAs to constant-bound arrays at block
scope in GNU mode. That's non-conforming, and more modern versions of
GCC only do this at file scope. Update Clang to do the same.

Also promote the warning for this from off-by-default to on-by-default
in all cases; more recent versions of GCC likewise warn on this by
default.

This is still slightly more permissive than GCC, as pointed out in
PR44406, as we still fold VLAs to constant arrays in structs, but that
seems justifiable given that we don't support VLA-in-struct (and don't
intend to ever support it), but GCC does.

Differential Revision: https://reviews.llvm.org/D89523

Fix a few bugs where we would fail to properly determine header to
module correspondence when determining whether to suggest a #include or
import, and suggest a #include more often in language modes where there
is no import syntax. Generally, if the target is in a header with
include guards or #pragma once, we should suggest either #including or
importing that header, and not importing a module that happens to
textually include it.

In passing, improve the notes we attach to the corresponding
diagnostics: calling an entity that we couldn't see "previous" is
confusing.

The driver enables -fdiagnostics-show-option by default, so flip the CC1
default to reduce the lengths of common CC1 command lines.

This change also makes ParseDiagnosticArgs() consistently enable
-fdiagnostics-show-option by default.

Produce the conventional "to match this '<'" note, so that the user
knows why we expected a '>', and properly handle '>>' in C++11 onwards.

user interface and documentation, and update __cplusplus for C++20.

WG21 considers the C++20 standard to be finished (even though it still
has some more steps to pass through in the ISO process).

The old flag names are accepted for compatibility, as usual, and we
still have lots of references to C++2a in comments and identifiers;
those can be cleaned up separately.

explicit functions that are not candidates.

It's not always obvious that the reason a conversion was not possible is
because the function you wanted to call is 'explicit', so explicitly say
if that's the case.

It would be nice to rank the explicit candidates higher in the
diagnostic if an implicit conversion sequence exists for their
arguments, but unfortunately we can't determine that without potentially
triggering non-immediate-context errors that we're not permitted to
produce.

C++ DR712 and others: handle non-odr-use resulting from an lvalue-to-rvalue conversion applied to a member access or similar not-quite-trivial lvalue expression.

Summary:
When a variable is named in a context where we can't directly emit a
reference to it (because we don't know for sure that it's going to be
defined, or it's from an enclosing function and not captured, or the
reference might not "work" for some reason), we emit a copy of the
variable as a global and use that for the known-to-be-read-only access.

This reinstates r363295, reverted in r363352, with a fix for PR42276:
we now produce a proper name for a non-odr-use reference to a static
constexpr data member. The name <mangled-name>.const is used in that
case; such names are reserved to the implementation for cases such as
this and should demangle nicely.

Reviewers: rjmccall

Subscribers: jdoerfert, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D63157

llvm-svn: 363428

Revert 363340 "Remove unused SK_LValueToRValue initialization step."
Revert 363337 "PR23833, DR2140: an lvalue-to-rvalue conversion on a glvalue of type"
Revert 363295 "C++ DR712 and others: handle non-odr-use resulting from an lvalue-to-rvalue conversion applied to a member access or similar not-quite-trivial lvalue expression."

llvm-svn: 363352

C++ DR712 and others: handle non-odr-use resulting from an lvalue-to-rvalue conversion applied to a member access or similar not-quite-trivial lvalue expression.

Summary:
When a variable is named in a context where we can't directly emit a
reference to it (because we don't know for sure that it's going to be
defined, or it's from an enclosing function and not captured, or the
reference might not "work" for some reason), we emit a copy of the
variable as a global and use that for the known-to-be-read-only access.

Reviewers: rjmccall

Subscribers: jdoerfert, cfe-commits

Tags: #clang

Differential Revision: https://reviews.llvm.org/D63157

llvm-svn: 363295

template name is not visible to unqualified lookup.

In order to support this without a severe degradation in our ability to
diagnose typos in template names, this change significantly restructures
the way we handle template-id-shaped syntax for which lookup of the
template name finds nothing.

Instead of eagerly diagnosing an undeclared template name, we now form a
placeholder template-name representing a name that is known to not find
any templates. When the parser sees such a name, it attempts to
disambiguate whether we have a less-than comparison or a template-id.
Any diagnostics or typo-correction for the name are delayed until its
point of use.

The upshot should be a small improvement of our diagostic quality
overall: we now take more syntactic context into account when trying to
resolve an undeclared identifier on the left hand side of a '<'. In
fact, this works well enough that the backwards-compatible portion (for
an undeclared identifier rather than a lookup that finds functions but
no function templates) is enabled in all language modes.

llvm-svn: 360308

CWG 1691 changed the definition of the namespaces associated with a class
type or enumeration type.

For a class type, the associated namespaces are the innermost enclosing
namespaces of the associated classes. For an enumeration type, the associated
namespace is the innermost enclosing namespace of its declaration.

This also fixes CWG 1690 and CWG 1692.

Differential Revision: https://reviews.llvm.org/D60573

Reviewed By: rjmccall, rsmith

llvm-svn: 358882

The goal here is to exercise each rule in [basic.lookup.argdep] at least once.
These new tests expose what I believe are 2 issues:

1. CWG 1691 needs to be implemented (p2:  [...] Its associated namespaces are
   the innermost enclosing namespaces of its associated classes [...]) The
   corresponding tests are adl_class_type::X2 and adl_class_type::X5.

2. The end of paragraph 2 ([...] Additionally, if the aforementioned set of
   overloaded functions is named with a template-id, its associated classes
   and namespaces also include those of its type template-arguments and its
   template template-arguments.) is not implemented. Closely related, the
   restriction on non-dependent parameter types in this same paragraph needs
   to be removed. The corresponding tests are in adl_overload_set (both issues
   are from CWG 997).

Differential Revision: https://reviews.llvm.org/D60570

Reviewed By: riccibruno, Quuxplusone

llvm-svn: 358881

llvm-svn: 358713

global module fragment.

We know that the declaration in question should have been introduced by
a '#include', so try to figure out which one and suggest it. Don't
suggest importing the global module fragment itself!

llvm-svn: 358631

and the global and private module fragment.

For now, the private module fragment introducer is ignored, but use of
the global module fragment introducer should be properly enforced.

llvm-svn: 358353

members of dependent contexts.

This permits cases where the names before and after the '::' in a
dependent inherited constructor using-declaration do not match, but
where we can nonetheless tell when parsing the template that a
constructor is being named. Under (open) core language DR 2070, such
cases will probably be ill-formed, but r335182 does not quite give
that result and didn't intend to change this, so restore the old
behavior for now.

llvm-svn: 335381

function-style cast.

This fires for cases such as

  T(x);

... where 'x' was previously declared and T is a type. This construct declares
a variable named 'x' rather than the (probably expected) interpretation of a
function-style cast of 'x' to T.

llvm-svn: 314570

We used to give such members no linkage instead of giving them the linkage of
the class.

llvm-svn: 314054

This doesn't affect our code generation in any material way -- we already give
such declarations internal linkage from a codegen perspective -- but it has
some subtle effects on code validity.

We suppress the 'L' (internal linkage) marker for mangled names in anonymous
namespaces, because it is redundant (the information is already carried by the
namespace); this deviates from GCC's behavior if a variable or function in an
anonymous namespace is redundantly declared 'static' (where GCC does include
the 'L'), but GCC's behavior is incoherent because such a declaration can be
validly declared with or without the 'static'.

We still deviate from the standard in one regard here: extern "C" declarations
in anonymous namespaces are still granted external linkage. Changing those does
not appear to have been an intentional consequence of the standard change in
DR1113.

llvm-svn: 314037

purposes), so they never formally have linkage.

llvm-svn: 313957

If a function or variable has a type with no linkage (and is not extern "C"),
any use of it requires a definition within the same translation unit; the idea
is that it is not possible to define the entity elsewhere, so any such use is
necessarily an error.

There is an exception, though: some types formally have no linkage but
nonetheless can be referenced from other translation units (for example, this
happens to anonymous structures defined within inline functions). For entities
with those types, we suppress the diagnostic except under -pedantic.

llvm-svn: 313729

Fixes some accepts-invalids with tags and other declarations declared in the
same scope.

llvm-svn: 312743

Printing typedefs or type aliases using clang_getTypeSpelling() is missing the
namespace they are defined in. This is in contrast to other types that always
yield the full typename including namespaces.

Patch by Michael Reiher!

Differential Revision: https://reviews.llvm.org/D29944

llvm-svn: 297465