https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/156127
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov auto_merge_enabled
https://github.com/llvm/llvm-project/pull/156127
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/156127
Use the castAs acessor for the type for a UsingEnumDecl, as it can be sugar for
an EnumType.
Fixes a regression reported here:
https://github.com/llvm/llvm-project/pull/155313#issuecomment-3238482327
Since t
mizvekov wrote:
> The most important part for me, is that if
> `TagDecl::isThisDeclarationADefinition()` returns true, then I need
> `getSourceRange()` to return the content of the definition.
That's a more interesting limited part of the problem.
It seems strange to me that we produce a temp
mizvekov wrote:
> Thanks for the feedback @mizvekov. I propose this change for the following
> reasons:
>
> * Class template implicit instantiations have technically no source range
> anywhere, which is why I think having the same behavior as function template
> implicit instantiations or var
mizvekov wrote:
> Which file is appropriate to add it to? Or should I create a new file?
For common-type concerns, you can add it to `Sema/sugar-common-types.c` for C
code, or `clang/test/SemaCXX/sugar-common-types.cpp` for C++.
For AST dumps, I don't see a great fit anywhere, so if you create
mizvekov wrote:
You could include your opening snippet as a test case within the patch. Or even
a simple AST dump test case showing these types contain __size_t et al as
bottom-level sugar.
https://github.com/llvm/llvm-project/pull/155979
___
cfe-com
https://github.com/mizvekov commented:
Why is this the correct behavior, rather than the other way around?
The patch seems to leave no way to query the source range of this declaration,
rather than the definition.
On the other hand, querying the source range of the definition is currently
pos
mizvekov wrote:
> If they are including `Type.h` in files that only exist downstream that will
> force them to deal with this both times around, unless I am missing something.
No, they would only have had to do that if they had rebased in between this
commit landing and the next one, but that
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155965
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov auto_merge_disabled
https://github.com/llvm/llvm-project/pull/155965
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov auto_merge_enabled
https://github.com/llvm/llvm-project/pull/155965
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155965
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155965
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155965
…context
Fix an error in the logic meant to handle a redeclaration such as:
```C++
struct A {
struct __attribute__((foo)) A *ptr;
};
```
In the declaration of ptr, we must introduce a new redeclaration of A i
https://github.com/mizvekov auto_merge_enabled
https://github.com/llvm/llvm-project/pull/155959
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155959
The code being removed became non-functional when injected class names were
removed from the redecl chain in 470c454a6176ef31474553e408c90f5ee630df89, but
that commit didn't clean the code base out of all the
mizvekov wrote:
> Hi -- sidenote, but I noticed that this commit -- and many other of your
> commits -- are missing the entire commit message other than the PR title. I
> don't know if that's a github preference or something you're doing manually,
> but it's really unfortunate because then I c
mizvekov wrote:
I don't think that helps, however the patch is a bit simple, so you could maybe
sort of reverse engineer what could have gone wrong from it.
For example, before this patch we would have produced non-identical
CXXRecordTypes from basically the same information.
There is probab
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155904
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -18544,8 +18544,14 @@ Sema::ActOnTag(Scope *S, unsigned TagSpec, TagUseKind
TUK, SourceLocation KWLoc,
if (PrevDecl)
CheckRedeclarationInModule(New, PrevDecl);
- if (TUK == TagUseKind::Definition && (!SkipBody || !SkipBody->ShouldSkip))
-New->startDefinition();
mizvekov wrote:
@jyknight thanks, this will be fixed here:
https://github.com/llvm/llvm-project/pull/155900
https://github.com/llvm/llvm-project/pull/155028
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155900
Clang skips parsing a TagDecl definition in case a definition was already
parsed in another module.
In those cases, an EnumDecl might be left without an IntegerType. Take this
into account when getting the un
@@ -3246,13 +3246,17 @@ void
MicrosoftCXXNameMangler::mangleTagTypeKind(TagTypeKind TTK) {
}
void MicrosoftCXXNameMangler::mangleType(const EnumType *T, Qualifiers,
SourceRange) {
- mangleType(cast(T)->getOriginalDecl()->getDefinitionO
mizvekov wrote:
It was split off from another PR, which had reviewers.
https://github.com/llvm/llvm-project/pull/155313
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mizvekov wrote:
> Should this be a new issue?
I vaguely remember having created such an issue in the past, but I can't find
it, so I created a new one in any case:
https://github.com/llvm/llvm-project/issues/155709
https://github.com/llvm/llvm-project/pull/155688
_
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155688
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155688
This makes the type printer not qualify constructor and destructor names. These
are represented as canonical types and the type printer is used, but unlike
canonical types which we normally print as fully qual
mizvekov wrote:
The MSVC mangling issue is resolved here:
https://github.com/llvm/llvm-project/pull/155662
https://github.com/llvm/llvm-project/pull/147835
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155662
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155051
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155049
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mizvekov wrote:
> (I think this was covered by tests even, right?)
No, this patch didn't change any existing mangling tests at all, the behavior
in question here was untested.
https://github.com/llvm/llvm-project/pull/147835
___
cfe-commits mailing l
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155051
>From c5d57c558d63380f726260d1c14d6f40162e9114 Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Fri, 22 Aug 2025 15:16:27 -0300
Subject: [PATCH] [clang] NFC: Provide inline definitions for {get,cast}TagDecl
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155050
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155051
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155050
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155028
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155028
>From 63542d5cdeccb17b9125430d0a984f45d58e208c Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Fri, 22 Aug 2025 19:10:05 -0300
Subject: [PATCH 1/2] meat and potatoes
---
clang/include/clang/AST/Type.h
mizvekov wrote:
@aeubanks
I see, so that's unfortunate MSVC mangles the struct / class difference.
This is a different problem than I had in mind.
So MSVC mangles according to the tag kind of the first declaration, clang after
this changes now mangles it according to thedeclaration found
mizvekov wrote:
I did another offloading of NFC changes to another patch.
This PR is tiny now in comparison.
https://github.com/llvm/llvm-project/pull/155028
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailma
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155028
>From 7a334282f87fa5a2135cc35771cc7d99724d89ac Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Fri, 22 Aug 2025 19:10:05 -0300
Subject: [PATCH 1/2] meat and potatoes
---
clang/include/clang/AST/Type.h
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155463
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov auto_merge_disabled
https://github.com/llvm/llvm-project/pull/155463
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov auto_merge_enabled
https://github.com/llvm/llvm-project/pull/155463
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mizvekov wrote:
> Great see commit
> [`af63956`](https://github.com/llvm/llvm-project/pull/148914/commits/af6395691ad98c3763d89e4794f353e77f0e3d9b).
> I just stole the pattern from the array canonicalization code. Added tests
> too.
Great, there are other places that need to be updated to han
mizvekov wrote:
> I'm getting conflicting ideas from this sentence. Do you mean to say that the
> two types from your example should be the _same_ type?
Yes.
https://github.com/llvm/llvm-project/pull/148914
___
cfe-commits mailing list
cfe-commits@li
@@ -5254,153 +5253,297 @@
ASTContext::getPredefinedSugarType(PredefinedSugarType::Kind KD) const {
return QualType(New, 0);
}
-#ifndef NDEBUG
-static bool NeedsInjectedClassNameType(const RecordDecl *D) {
- if (!isa(D)) return false;
- const auto *RD = cast(D);
- if (isa
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155347
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155342
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -5254,153 +5253,297 @@
ASTContext::getPredefinedSugarType(PredefinedSugarType::Kind KD) const {
return QualType(New, 0);
}
-#ifndef NDEBUG
-static bool NeedsInjectedClassNameType(const RecordDecl *D) {
- if (!isa(D)) return false;
- const auto *RD = cast(D);
- if (isa
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155347
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155347
This fixes a bug in the fast path for the creation of TagTypes from injected
class names.
The creation of TagTypes has a fast path which, when there is no elaboration,
uses storage in the declaration itself f
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155342
This makes sure NestedNameSpecifierLocs don't apply to the replacement
TemplateName of SubstTemplate* nodes.
Also removes improper name qualification over these Subst Nodes, causing some
canonical TemplateNam
mizvekov wrote:
@michaelrj-google
> ```
>::sapi::v::IntBase v_type__(type__);
> ```
>
> which clang errors on.
And that enum type, is that created by Sema, or is that synthesized by code
from within your project?
If you are creating that type as a canonical type, or that type is being
c
mizvekov wrote:
@aeubanks can you get us a reproducer?
https://github.com/llvm/llvm-project/pull/147835
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mizvekov wrote:
> If the change breaks LLDB then you should also fix that break as well.
Yes of course, but the parts already in the PR can be reviewed independently of
that.
https://github.com/llvm/llvm-project/pull/155120
___
cfe-commits mailing li
mizvekov wrote:
> The previous revision compiles the reduced successfully. Can you please take
> a look?
Fixed in https://github.com/llvm/llvm-project/pull/155272
https://github.com/llvm/llvm-project/pull/147835
___
cfe-commits mailing list
cfe-commi
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/155272
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155272
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155272
Through alias templates, followed by canonicalization, any canonical dependent
type can appear in a NestedNameSpecifier. Remove the list as a practical matter.
This fixes a regression reported here:
https://g
mizvekov wrote:
> The only thing that "leaf type" means is that there's no point in doing
> anything except `dyn_cast`ing the canonical type. It's not a real language
> concept, and we don't use the type property for any purpose except these
> `getAs` specializations. So if the specializations
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/154541
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -95,24 +95,25 @@ TC
https://github.com/llvm/llvm-project/pull/155120
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155120
>From e1c4ead49256ce93ca35dda58b57d96fdb7e001d Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Sat, 9 Aug 2025 18:14:46 -0300
Subject: [PATCH] [clang] remove isDefaulted bit from TemplateArgument
The IsDe
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155120
>From 6f039d0a53ef53436624b6493bfcf3b6c9ba36cb Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Sat, 9 Aug 2025 18:14:46 -0300
Subject: [PATCH] [clang] remove isDefaulted bit from TemplateArgument
The IsDe
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155120
>From 85ba1ffd8726a1596dfd3aebfd56ebed0460cf77 Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Sat, 9 Aug 2025 18:14:46 -0300
Subject: [PATCH] [clang] remove isDefaulted bit from TemplateArgument
The IsDe
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155120
>From f31c4e385a66cf3e6c7aab499bcba7bd271902fb Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Sat, 9 Aug 2025 18:14:46 -0300
Subject: [PATCH] [clang] remove isDefaulted bit from TemplateArgument
The IsDe
https://github.com/mizvekov updated
https://github.com/llvm/llvm-project/pull/155120
>From 3996f1f23d55e5d3fa121a75e015ce7fbd3624ca Mon Sep 17 00:00:00 2001
From: Matheus Izvekov
Date: Sat, 9 Aug 2025 18:14:46 -0300
Subject: [PATCH] [clang] remove isDefaulted bit from TemplateArgument
The IsDe
mizvekov wrote:
As I explained in the commit message, this is not removing the feature.
The information about which arguments were defaulted was already available in
the AST before this bit was added.
And the bit is not correct as I said, being part of the canonical type means
that this infor
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/155120
The IsDefaulted bit being part of a canonical TemplateArgument doesn't make
sense, as that information is not information a canonical type should have.
In C++, all template specialization types for ther same t
mizvekov wrote:
> Seems like `getOriginalDecl()` always returns the full definition for
> enumerations (and not an opaque declaration for enums with fixed underlying
> type), at least in the C language mode. Is it guaranteed?
I don't see that happening. Example: https://compiler-explorer.com/z
mizvekov wrote:
> Related, I think the change to [`void TypePrinter::printUsingBefore(const
> UsingType *T, raw_ostream
> &OS)`](https://github.com/llvm/llvm-project/blame/d439c9ea4a19bf783120d8efee5e67131550/clang/lib/AST/TypePrinter.cpp#L1227)
> might be causing similar issues. The `getK
mizvekov wrote:
> Hi, I'm helping fix some clang-as-a-library users and I'm running into some
> issues with the TypePrinter after this commit. Specifically the clang
> generator in sandboxed-api is broken
> (https://github.com/google/sandboxed-api/tree/main). After doing the trivial
> fixes (
https://github.com/mizvekov commented:
How about inner qualifiers? The patch doesn't address the situation.
For a test like:
```C++
using ConstInt = const int;
using WrapConstInt1 = ConstInt __attribute__((overflow_behavior(wrap)));
using WrapConstInt2 = const int __attribute__((overflow_behavio
@@ -5776,6 +5826,51 @@ QualType ASTContext::getBTFTagAttributedType(const
BTFTypeTagAttr *BTFAttr,
return QualType(Ty, 0);
}
+QualType ASTContext::getOverflowBehaviorType(const OverflowBehaviorAttr *Attr,
+ QualType Underlying) co
@@ -2844,6 +2872,8 @@ inline T TypeLoc::getAsAdjusted() const {
Cur = ATL.getModifiedLoc();
else if (auto ATL = Cur.getAs())
Cur = ATL.getWrappedLoc();
+// else if (auto ATL = Cur.getAs())
+// Cur = ATL.getWrappedLoc();
mizvekov wrote:
@@ -8914,6 +8975,12 @@ inline bool Type::isIntegerType() const {
return IsEnumDeclComplete(ET->getOriginalDecl()) &&
!IsEnumDeclScoped(ET->getOriginalDecl());
}
+
+ if (isOverflowBehaviorType())
+return cast(CanonicalType)
+->getUnderlyingType()
+
@@ -9120,6 +9191,8 @@ template const T *Type::getAsAdjusted()
const {
Ty = A->getModifiedType().getTypePtr();
else if (const auto *A = dyn_cast(Ty))
Ty = A->getWrappedType().getTypePtr();
+// else if (const auto *A = dyn_cast(Ty))
+// Ty = A->getWrap
@@ -405,6 +405,10 @@ ArgType::matchesType(ASTContext &C, QualType argTy) const {
argTy = PT->getPointeeType();
}
+ if (const OverflowBehaviorType *OBT =
+ dyn_cast(argTy.getCanonicalType()))
+argTy = OBT->getUnderlyingType();
+
mizvekov wro
@@ -74,6 +74,13 @@ QualType clang::desugarForDiagnostic(ASTContext &Context,
QualType QT,
QT = AT->desugar();
continue;
}
+// ... or an overflow behavior type.
+if (const OverflowBehaviorType *OBT = dyn_cast(Ty)) {
+ if (!OBT->isSugared())
+
@@ -6660,6 +6672,51 @@ class BTFTagAttributedType : public Type, public
llvm::FoldingSetNode {
}
};
+class OverflowBehaviorType : public Type, public llvm::FoldingSetNode {
+public:
+ enum OverflowBehaviorKind { Wrap, NoWrap };
+
+private:
+ friend class ASTContext; // AS
@@ -11710,6 +11809,48 @@ QualType ASTContext::mergeTagDefinitions(QualType LHS,
QualType RHS) {
return Ctx.IsEquivalent(LHS, RHS) ? LHS : QualType{};
}
+std::optional ASTContext::tryMergeOverflowBehaviorTypes(
+QualType LHS, QualType RHS, bool OfBlockPointer, bool Unqua
@@ -870,6 +870,42 @@
ASTContext::insertCanonicalTemplateTemplateParmDeclInternal(
return CanonTTP;
}
+/// For the purposes of overflow pattern exclusion, does this match the
+/// while(i--) pattern?
+static bool matchesPostDecrInWhile(const UnaryOperator *UO, ASTContext &Ct
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/148914
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
@@ -649,6 +649,10 @@ template <> const CountAttributedType *Type::getAs() const
{
return getAsSugar(this);
}
+template <> const OverflowBehaviorType *Type::getAs() const {
+ return getAsSugar(this);
+}
mizvekov wrote:
The `OverflowBehaviorType` is never a
@@ -14308,6 +14455,12 @@ static QualType getCommonNonSugarTypeNode(const
ASTContext &Ctx, const Type *X,
getCommonTypeKeyword(NX, NY, /*IsSame=*/true),
getCommonQualifier(Ctx, NX, NY, /*IsSame=*/true), NX->getIdentifier());
}
+ case Type::OverflowBehavior: {
https://github.com/mizvekov commented:
Some comments, I will try a more through review later.
https://github.com/llvm/llvm-project/pull/148914
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-c
@@ -3780,6 +3780,11 @@ void MicrosoftCXXNameMangler::mangleType(const
HLSLInlineSpirvType *T,
llvm_unreachable("HLSL uses Itanium name mangling");
}
+void MicrosoftCXXNameMangler::mangleType(const OverflowBehaviorType *T,
+ Qualifiers
@@ -2586,8 +2561,9 @@ bool
CXXNameMangler::mangleUnresolvedTypeOrSimpleId(QualType Ty,
}
case Type::InjectedClassName:
-mangleSourceNameWithAbiTags(
-cast(Ty)->getDecl());
+mangleSourceNameWithAbiTags(cast(Ty)
+->getOrig
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155028
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/155028
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mizvekov wrote:
> I'm trying to reduce a crash that only reproduces w/ in module builds. Does
> this sound familiar?
>
> ```
> assertion failed at clang/lib/AST/RecordLayoutBuilder.cpp:3380 in const
> ASTRecordLayout &clang::ASTContext::getASTRecordLayout(const RecordDecl *)
> const: D && "Ca
mizvekov wrote:
FWIW all of these switches have "not handled" cases, where it would be easy to
just throw another case for the new type node in there.
https://github.com/llvm/llvm-project/pull/154606
___
cfe-commits mailing list
cfe-commits@lists.llvm
https://github.com/mizvekov closed
https://github.com/llvm/llvm-project/pull/154592
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
https://github.com/mizvekov created
https://github.com/llvm/llvm-project/pull/154592
When building the base type for constructor initializer, the case of an
UnresolvedUsingType was not being handled.
For the non-dependent case, we are also skipping adding the UsingType, but this
is just missi
@@ -126,7 +130,7 @@ const TensorSpec llvm::DefaultDecisionSpec =
TensorSpec::createSpec(DefaultDecisionName, {1});
const char *const llvm::RewardName = "delta_size";
-CallBase *getInlinableCS(Instruction &I) {
+static CallBase *getInlinableCS(Instruction &I) {
https://github.com/mizvekov approved this pull request.
LGTM, Thanks.
The changes are a strict improvement.
Pre-existing problem, but it sure looks iffy to me that these data structures
are global, not attached to any context.
https://github.com/llvm/llvm-project/pull/154541
_
https://github.com/mizvekov edited
https://github.com/llvm/llvm-project/pull/154541
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
mizvekov wrote:
That's also better than disabling the test, but we don't need to test that we
produce this warning, this test is not about that. It's a regression test for a
crash and the warning is irrelevant.
https://github.com/llvm/llvm-project/pull/154463
1 - 100 of 1705 matches
Mail list logo
