[clang] [llvm] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (PR #120753)

[Tomas Matheson via cfe-commits]

https://github.com/tmatheson-arm approved this pull request.

Makes sense to me.

https://github.com/llvm/llvm-project/pull/120753
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (PR #120753)

[Jonathan Thackray via cfe-commits]

https://github.com/jthackray updated 
https://github.com/llvm/llvm-project/pull/120753

>From 519bcca2359ee9c89b12bac6e58eb955c79cd7bc Mon Sep 17 00:00:00 2001
From: Jonathan Thackray 
Date: Fri, 20 Dec 2024 14:50:37 +
Subject: [PATCH 1/2] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and
 later.

The ArmARM says:
```
"In an Armv9.4 implementation, if FEAT_SVE2 is implemented,
 FEAT_SVE2p1 is implemented."
```
Since FEAT_SVE2 is already enabled for Armv9.0-A and later,
then FEAT_SVE2p1 should also be enabled by default.
---
 clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c | 1 +
 clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c | 1 +
 llvm/lib/Target/AArch64/AArch64Features.td | 2 +-
 3 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c 
b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
index 0032c926c22d96..1cfda6c996b9e3 100644
--- a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
+++ b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
@@ -58,6 +58,7 @@
 // CHECK-NEXT: FEAT_SSBS, FEAT_SSBS2  
Enable Speculative Store Bypass Safe bit
 // CHECK-NEXT: FEAT_SVE   
Enable Scalable Vector Extension (SVE) instructions
 // CHECK-NEXT: FEAT_SVE2  
Enable Scalable Vector Extension 2 (SVE2) instructions
+// CHECK-NEXT: FEAT_SVE2p1
Enable Scalable Vector Extension 2.1 instructions
 // CHECK-NEXT: FEAT_TLBIOS, FEAT_TLBIRANGE
Enable Armv8.4-A TLB Range and Maintenance instructions
 // CHECK-NEXT: FEAT_TRBE  
Enable Trace Buffer Extension
 // CHECK-NEXT: FEAT_TRF   
Enable Armv8.4-A Trace extension
diff --git a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c 
b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
index be24bd0bbddb68..76c8b34a56b75b 100644
--- a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
+++ b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
@@ -61,6 +61,7 @@
 // CHECK-NEXT: FEAT_SSBS, FEAT_SSBS2  
Enable Speculative Store Bypass Safe bit
 // CHECK-NEXT: FEAT_SVE   
Enable Scalable Vector Extension (SVE) instructions
 // CHECK-NEXT: FEAT_SVE2  
Enable Scalable Vector Extension 2 (SVE2) instructions
+// CHECK-NEXT: FEAT_SVE2p1
Enable Scalable Vector Extension 2.1 instructions
 // CHECK-NEXT: FEAT_TLBIOS, FEAT_TLBIRANGE
Enable Armv8.4-A TLB Range and Maintenance instructions
 // CHECK-NEXT: FEAT_TRBE  
Enable Trace Buffer Extension
 // CHECK-NEXT: FEAT_TRF   
Enable Armv8.4-A Trace extension
diff --git a/llvm/lib/Target/AArch64/AArch64Features.td 
b/llvm/lib/Target/AArch64/AArch64Features.td
index 3a6bef81f4a0d8..96473e55f01d8e 100644
--- a/llvm/lib/Target/AArch64/AArch64Features.td
+++ b/llvm/lib/Target/AArch64/AArch64Features.td
@@ -872,7 +872,7 @@ def HasV9_3aOps : Architecture64<9, 3, "a", "v9.3a",
   [HasV8_8aOps, HasV9_2aOps],
   !listconcat(HasV9_2aOps.DefaultExts, [FeatureMOPS, FeatureHBC])>;
 def HasV9_4aOps : Architecture64<9, 4, "a", "v9.4a",
-  [HasV8_9aOps, HasV9_3aOps],
+  [HasV8_9aOps, HasV9_3aOps, FeatureSVE2p1],
   !listconcat(HasV9_3aOps.DefaultExts, [FeatureSPECRES2, FeatureCSSC,
 FeatureRASv2])>;
 def HasV9_5aOps : Architecture64<9, 5, "a", "v9.5a",

>From 15a2f3922bec4ccc6472f3eb2029c66f9ef6d834 Mon Sep 17 00:00:00 2001
From: Jonathan Thackray 
Date: Fri, 20 Dec 2024 16:13:53 +
Subject: [PATCH 2/2] fixup! [AArch64] Enable FEAT_SVE2p1 by default for
 Armv9.4-A and later.

---
 llvm/lib/Target/AArch64/AArch64Features.td | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/llvm/lib/Target/AArch64/AArch64Features.td 
b/llvm/lib/Target/AArch64/AArch64Features.td
index 96473e55f01d8e..41eb9a73bd013d 100644
--- a/llvm/lib/Target/AArch64/AArch64Features.td
+++ b/llvm/lib/Target/AArch64/AArch64Features.td
@@ -872,9 +872,9 @@ def HasV9_3aOps : Architecture64<9, 3, "a", "v9.3a",
   [HasV8_8aOps, HasV9_2aOps],
   !listconcat(HasV9_2aOps.DefaultExts, [FeatureMOPS, FeatureHBC])>;
 def HasV9_4aOps : Architecture64<9, 4, "a", "v9.4a",
-  [HasV8_9aOps, HasV9_3aOps, FeatureSVE2p1],
+  [HasV8_9aOps, HasV9_3aOps],
   !listconcat(HasV9_3aOps.DefaultExts, [FeatureSPECRES2, FeatureCSSC,
-FeatureRASv2])>;
+FeatureRASv2, FeatureSVE2p1])>;
 def HasV9_5aOps : Architecture64<9, 5, "a", "v9.5a",
   [HasV9_4aOps, FeatureCPA],
   !listco

[clang] [llvm] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (PR #120753)

[Tomas Matheson via cfe-commits]

tmatheson-arm wrote:

Actually this is being added as a mandatory dependency of 9.4, should it not be 
added as one of the default features instead?

https://github.com/llvm/llvm-project/pull/120753
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [HLSL] Codegen for `cbuffer` declarations without embedded arrays or structs (PR #119755)

[Chris B via cfe-commits]

https://github.com/llvm-beanz requested changes to this pull request.


https://github.com/llvm/llvm-project/pull/119755
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [HLSL] Codegen for `cbuffer` declarations without embedded arrays or structs (PR #119755)

[Chris B via cfe-commits]

@@ -54,78 +54,108 @@ void addDxilValVersion(StringRef ValVersionStr, 
llvm::Module &M) {
   auto *DXILValMD = M.getOrInsertNamedMetadata(DXILValKey);
   DXILValMD->addOperand(Val);
 }
+
 void addDisableOptimizations(llvm::Module &M) {
   StringRef Key = "dx.disable_optimizations";
   M.addModuleFlag(llvm::Module::ModFlagBehavior::Override, Key, 1);
 }
-// cbuffer will be translated into global variable in special address space.
-// If translate into C,
-// cbuffer A {
-//   float a;
-//   float b;
-// }
-// float foo() { return a + b; }
+
+// Creates resource handle representing the constant buffer.
+// For cbuffer declaration:
 //
-// will be translated into
+//   cbuffer MyConstants {
+// float a;
+//   }
 //
-// struct A {
-//   float a;
-//   float b;
-// } cbuffer_A __attribute__((address_space(4)));
-// float foo() { return cbuffer_A.a + cbuffer_A.b; }
+// creates a structure type MyConstants and then returns the resource handle
+// that would be spelled as:
 //
-// layoutBuffer will create the struct A type.
-// replaceBuffer will replace use of global variable a and b with cbuffer_A.a
-// and cbuffer_A.b.
+//   __hlsl_resource_t [[hlsl::resource_class(CBuffer)]]
+//   [[contained_type(MyConstants)]]
 //
-void layoutBuffer(CGHLSLRuntime::Buffer &Buf, const DataLayout &DL) {
-  if (Buf.Constants.empty())
-return;
-
-  std::vector EltTys;
-  for (auto &Const : Buf.Constants) {
-GlobalVariable *GV = Const.first;
-Const.second = EltTys.size();
-llvm::Type *Ty = GV->getValueType();
-EltTys.emplace_back(Ty);
+static const clang::Type *getBufferHandleType(CGHLSLRuntime::Buffer &Buf) {
+  HLSLBufferDecl *BD = Buf.Decl;
+  ASTContext &AST = BD->getASTContext();
+
+  // create struct type for the constant buffer; filter out any declarations
+  // that are not a VarDecls or that are static
+  CXXRecordDecl *StructDecl = CXXRecordDecl::Create(
+  BD->getASTContext(), TagDecl::TagKind::Class, BD->getDeclContext(),
+  BD->getLocation(), BD->getLocation(), BD->getIdentifier());
+  StructDecl->startDefinition();
+  for (Decl *it : Buf.Decl->decls()) {
+const VarDecl *VD = dyn_cast(it);
+if (!VD || VD->getStorageClass() == SC_Static)
+  continue;
+auto *Field = FieldDecl::Create(
+AST, StructDecl, VD->getLocation(), VD->getLocation(),
+VD->getIdentifier(), VD->getType(), VD->getTypeSourceInfo(), nullptr,
+false, InClassInitStyle::ICIS_NoInit);
+Field->setAccess(AccessSpecifier::AS_private);
+StructDecl->addDecl(Field);
   }
-  Buf.LayoutStruct = llvm::StructType::get(EltTys[0]->getContext(), EltTys);
+  StructDecl->completeDefinition();
+  assert(!StructDecl->fields().empty() && "empty cbuffer should not get here");
+
+  // create the resource handle type
+  HLSLAttributedResourceType::Attributes ResAttrs(dxil::ResourceClass::CBuffer,
+  false, false);
+  QualType ContainedTy = QualType(StructDecl->getTypeForDecl(), 0);
+  return AST
+  .getHLSLAttributedResourceType(AST.HLSLResourceTy, ContainedTy, ResAttrs)
+  .getTypePtr();
 }
 
-GlobalVariable *replaceBuffer(CGHLSLRuntime::Buffer &Buf) {
-  // Create global variable for CB.
-  GlobalVariable *CBGV = new GlobalVariable(
-  Buf.LayoutStruct, /*isConstant*/ true,
-  GlobalValue::LinkageTypes::ExternalLinkage, nullptr,
-  llvm::formatv("{0}{1}", Buf.Name, Buf.IsCBuffer ? ".cb." : ".tb."),
-  GlobalValue::NotThreadLocal);
-
-  IRBuilder<> B(CBGV->getContext());
-  Value *ZeroIdx = B.getInt32(0);
-  // Replace Const use with CB use.
-  for (auto &[GV, Offset] : Buf.Constants) {
-Value *GEP =
-B.CreateGEP(Buf.LayoutStruct, CBGV, {ZeroIdx, B.getInt32(Offset)});
-
-assert(Buf.LayoutStruct->getElementType(Offset) == GV->getValueType() &&
-   "constant type mismatch");
-
-// Replace.
-GV->replaceAllUsesWith(GEP);
-// Erase GV.
-GV->removeDeadConstantUsers();
-GV->eraseFromParent();
+// Replaces all uses of the temporary constant buffer global variables with
+// buffer access intrinsic resource.getpointer and GEP.
+static void replaceBufferGlobals(CodeGenModule &CGM,
+ CGHLSLRuntime::Buffer &Buf) {
+  assert(Buf.IsCBuffer && "tbuffer codegen is not yet supported");
+
+  GlobalVariable *BufGV = Buf.GlobalVar;
+  llvm::Type *TargetTy = BufGV->getValueType();
+  llvm::Type *BufStructTy = cast(TargetTy)->getTypeParameter(0);
+  unsigned Index = 0;
+  for (auto ConstIt = Buf.Constants.begin(); ConstIt != Buf.Constants.end();
+   ++ConstIt, ++Index) {
+GlobalVariable *ConstGV = *ConstIt;
+
+// TODO: Map to an hlsl_device address space.

llvm-beanz wrote:

```suggestion
// TODO: Map to an hlsl_device address space (#109877).
```

https://github.com/llvm/llvm-project/pull/119755
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/m

[clang] [llvm] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (PR #120753)

[Jonathan Thackray via cfe-commits]

jthackray wrote:

> Actually this is being added as a mandatory dependency of 9.4, should it not 
> be added as one of the default features instead?

Good point. I'll amend.

https://github.com/llvm/llvm-project/pull/120753
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [HLSL] Codegen for `cbuffer` declarations without embedded arrays or structs (PR #119755)

[Chris B via cfe-commits]

@@ -54,78 +54,108 @@ void addDxilValVersion(StringRef ValVersionStr, 
llvm::Module &M) {
   auto *DXILValMD = M.getOrInsertNamedMetadata(DXILValKey);
   DXILValMD->addOperand(Val);
 }
+
 void addDisableOptimizations(llvm::Module &M) {
   StringRef Key = "dx.disable_optimizations";
   M.addModuleFlag(llvm::Module::ModFlagBehavior::Override, Key, 1);
 }
-// cbuffer will be translated into global variable in special address space.
-// If translate into C,
-// cbuffer A {
-//   float a;
-//   float b;
-// }
-// float foo() { return a + b; }
+
+// Creates resource handle representing the constant buffer.
+// For cbuffer declaration:
 //
-// will be translated into
+//   cbuffer MyConstants {
+// float a;
+//   }
 //
-// struct A {
-//   float a;
-//   float b;
-// } cbuffer_A __attribute__((address_space(4)));
-// float foo() { return cbuffer_A.a + cbuffer_A.b; }
+// creates a structure type MyConstants and then returns the resource handle
+// that would be spelled as:
 //
-// layoutBuffer will create the struct A type.
-// replaceBuffer will replace use of global variable a and b with cbuffer_A.a
-// and cbuffer_A.b.
+//   __hlsl_resource_t [[hlsl::resource_class(CBuffer)]]
+//   [[contained_type(MyConstants)]]
 //
-void layoutBuffer(CGHLSLRuntime::Buffer &Buf, const DataLayout &DL) {
-  if (Buf.Constants.empty())
-return;
-
-  std::vector EltTys;
-  for (auto &Const : Buf.Constants) {
-GlobalVariable *GV = Const.first;
-Const.second = EltTys.size();
-llvm::Type *Ty = GV->getValueType();
-EltTys.emplace_back(Ty);
+static const clang::Type *getBufferHandleType(CGHLSLRuntime::Buffer &Buf) {
+  HLSLBufferDecl *BD = Buf.Decl;
+  ASTContext &AST = BD->getASTContext();
+
+  // create struct type for the constant buffer; filter out any declarations
+  // that are not a VarDecls or that are static
+  CXXRecordDecl *StructDecl = CXXRecordDecl::Create(
+  BD->getASTContext(), TagDecl::TagKind::Class, BD->getDeclContext(),
+  BD->getLocation(), BD->getLocation(), BD->getIdentifier());
+  StructDecl->startDefinition();
+  for (Decl *it : Buf.Decl->decls()) {
+const VarDecl *VD = dyn_cast(it);
+if (!VD || VD->getStorageClass() == SC_Static)
+  continue;
+auto *Field = FieldDecl::Create(
+AST, StructDecl, VD->getLocation(), VD->getLocation(),
+VD->getIdentifier(), VD->getType(), VD->getTypeSourceInfo(), nullptr,
+false, InClassInitStyle::ICIS_NoInit);
+Field->setAccess(AccessSpecifier::AS_private);
+StructDecl->addDecl(Field);
   }
-  Buf.LayoutStruct = llvm::StructType::get(EltTys[0]->getContext(), EltTys);
+  StructDecl->completeDefinition();
+  assert(!StructDecl->fields().empty() && "empty cbuffer should not get here");
+
+  // create the resource handle type
+  HLSLAttributedResourceType::Attributes ResAttrs(dxil::ResourceClass::CBuffer,
+  false, false);
+  QualType ContainedTy = QualType(StructDecl->getTypeForDecl(), 0);
+  return AST
+  .getHLSLAttributedResourceType(AST.HLSLResourceTy, ContainedTy, ResAttrs)
+  .getTypePtr();
 }
 
-GlobalVariable *replaceBuffer(CGHLSLRuntime::Buffer &Buf) {
-  // Create global variable for CB.
-  GlobalVariable *CBGV = new GlobalVariable(
-  Buf.LayoutStruct, /*isConstant*/ true,
-  GlobalValue::LinkageTypes::ExternalLinkage, nullptr,
-  llvm::formatv("{0}{1}", Buf.Name, Buf.IsCBuffer ? ".cb." : ".tb."),
-  GlobalValue::NotThreadLocal);
-
-  IRBuilder<> B(CBGV->getContext());
-  Value *ZeroIdx = B.getInt32(0);
-  // Replace Const use with CB use.
-  for (auto &[GV, Offset] : Buf.Constants) {
-Value *GEP =
-B.CreateGEP(Buf.LayoutStruct, CBGV, {ZeroIdx, B.getInt32(Offset)});
-
-assert(Buf.LayoutStruct->getElementType(Offset) == GV->getValueType() &&
-   "constant type mismatch");
-
-// Replace.
-GV->replaceAllUsesWith(GEP);
-// Erase GV.
-GV->removeDeadConstantUsers();
-GV->eraseFromParent();
+// Replaces all uses of the temporary constant buffer global variables with
+// buffer access intrinsic resource.getpointer and GEP.
+static void replaceBufferGlobals(CodeGenModule &CGM,

llvm-beanz wrote:

This is a big no. This is falling into the mess that was CGHLSLMSFinishCodeGen 
in DXC.

We should not be manipulating IR in Clang CodeGen. We should be able to inspect 
and dump the post-codegen IR, and test every IR transformation independently.

https://github.com/llvm/llvm-project/pull/119755
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [HLSL] Codegen for `cbuffer` declarations without embedded arrays or structs (PR #119755)

[Chris B via cfe-commits]

@@ -54,78 +54,108 @@ void addDxilValVersion(StringRef ValVersionStr, 
llvm::Module &M) {
   auto *DXILValMD = M.getOrInsertNamedMetadata(DXILValKey);
   DXILValMD->addOperand(Val);
 }
+
 void addDisableOptimizations(llvm::Module &M) {
   StringRef Key = "dx.disable_optimizations";
   M.addModuleFlag(llvm::Module::ModFlagBehavior::Override, Key, 1);
 }
-// cbuffer will be translated into global variable in special address space.
-// If translate into C,
-// cbuffer A {
-//   float a;
-//   float b;
-// }
-// float foo() { return a + b; }
+
+// Creates resource handle representing the constant buffer.
+// For cbuffer declaration:
 //
-// will be translated into
+//   cbuffer MyConstants {
+// float a;
+//   }
 //
-// struct A {
-//   float a;
-//   float b;
-// } cbuffer_A __attribute__((address_space(4)));
-// float foo() { return cbuffer_A.a + cbuffer_A.b; }
+// creates a structure type MyConstants and then returns the resource handle
+// that would be spelled as:
 //
-// layoutBuffer will create the struct A type.
-// replaceBuffer will replace use of global variable a and b with cbuffer_A.a
-// and cbuffer_A.b.
+//   __hlsl_resource_t [[hlsl::resource_class(CBuffer)]]
+//   [[contained_type(MyConstants)]]
 //
-void layoutBuffer(CGHLSLRuntime::Buffer &Buf, const DataLayout &DL) {
-  if (Buf.Constants.empty())
-return;
-
-  std::vector EltTys;
-  for (auto &Const : Buf.Constants) {
-GlobalVariable *GV = Const.first;
-Const.second = EltTys.size();
-llvm::Type *Ty = GV->getValueType();
-EltTys.emplace_back(Ty);
+static const clang::Type *getBufferHandleType(CGHLSLRuntime::Buffer &Buf) {
+  HLSLBufferDecl *BD = Buf.Decl;
+  ASTContext &AST = BD->getASTContext();
+
+  // create struct type for the constant buffer; filter out any declarations
+  // that are not a VarDecls or that are static
+  CXXRecordDecl *StructDecl = CXXRecordDecl::Create(
+  BD->getASTContext(), TagDecl::TagKind::Class, BD->getDeclContext(),
+  BD->getLocation(), BD->getLocation(), BD->getIdentifier());
+  StructDecl->startDefinition();

llvm-beanz wrote:

This concerns me. We really shouldn't be generating AST types during CodeGen.

We should generate the struct decl during Sema and attach it to the buffer decl 
so that we can dump and inspect the AST.


https://github.com/llvm/llvm-project/pull/119755
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [HLSL] Codegen for `cbuffer` declarations without embedded arrays or structs (PR #119755)

[Chris B via cfe-commits]

@@ -164,18 +164,18 @@ Decl *SemaHLSL::ActOnStartBuffer(Scope *BufferScope, bool 
CBuffer,
   return Result;
 }
 
-// Calculate the size of a legacy cbuffer type based on
+// Calculate the size of a legacy cbuffer type in bytes based on
 // 
https://learn.microsoft.com/en-us/windows/win32/direct3dhlsl/dx-graphics-hlsl-packing-rules
 static unsigned calculateLegacyCbufferSize(const ASTContext &Context,
QualType T) {
   unsigned Size = 0;
-  constexpr unsigned CBufferAlign = 128;
+  constexpr unsigned CBufferAlign = 16;
   if (const RecordType *RT = T->getAs()) {
 const RecordDecl *RD = RT->getDecl();
 for (const FieldDecl *Field : RD->fields()) {
   QualType Ty = Field->getType();
   unsigned FieldSize = calculateLegacyCbufferSize(Context, Ty);
-  unsigned FieldAlign = 32;
+  unsigned FieldAlign = 4;
   if (Ty->isAggregateType())

llvm-beanz wrote:

Can you put a comment there and reference the issue just to make it easier to 
keep track?

https://github.com/llvm/llvm-project/pull/119755
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [HLSL] Codegen for `cbuffer` declarations without embedded arrays or structs (PR #119755)

[Chris B via cfe-commits]

@@ -890,6 +890,10 @@ CodeGenModule::EmitCXXGlobalInitFunc() {
   ModuleInits.push_back(Fn);
 }
 
+  if (getLangOpts().HLSL && getHLSLRuntime().needsResourceBindingInitFn()) {
+CXXGlobalInits.push_back(getHLSLRuntime().createResourceBindingInitFn());
+  }

llvm-beanz wrote:

You probably haven't seen this, but I filed an issue yesterday about how we 
were connecting initializers in the wrong place 
(https://github.com/llvm/llvm-project/issues/120636).

This is definitely better. I do wonder if we should be generating one binding 
function for all bindings (which would require collecting them and generating 
them late), or if we should be generating an initializer per binding and 
appending them to the `CXXGlobalInits` list as we process each binding. While 
the later produces more functions that need to be inlined, I'm inclined to 
prefer it because it is simpler and aligns more with a traditional RAII-model 
for resource initialization.

The issue I filed is probably redundant since you're still reworking all of 
this.

https://github.com/llvm/llvm-project/pull/119755
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[Ikhlas Ajbar via cfe-commits]

https://github.com/iajbar created 
https://github.com/llvm/llvm-project/pull/120773

This patch introduces support for the Hexagon V75 architecture. It includes 
instruction formats, definitions, encodings, scheduling classes, and 
builtins/intrinsics.

>From 1d3ac02da5bfb00d86b123334e3a75628e47ef42 Mon Sep 17 00:00:00 2001
From: Ikhlas Ajbar 
Date: Fri, 13 Dec 2024 14:54:44 -0800
Subject: [PATCH] Add V75 support to compiler and assembler

This patch introduces support for the Hexagon V75 architecture.
It includes instruction formats, definitions, encodings, scheduling classes,
and builtins/intrinsics.
---
 clang/include/clang/Basic/BuiltinsHexagon.def |  10 +-
 clang/include/clang/Driver/Options.td |   2 +
 clang/lib/Basic/Targets/Hexagon.cpp   |   7 +-
 clang/test/Driver/hexagon-toolchain-elf.c |   7 +
 .../Misc/target-invalid-cpu-note/hexagon.c|   1 +
 clang/test/Preprocessor/hexagon-predefines.c  |  16 +
 llvm/include/llvm/BinaryFormat/ELF.h  |   1 +
 llvm/lib/Object/ELFObjectFile.cpp |   2 +
 llvm/lib/ObjectYAML/ELFYAML.cpp   |   2 +
 llvm/lib/Target/Hexagon/Hexagon.td|  14 +
 llvm/lib/Target/Hexagon/HexagonDepArch.h  |   4 +-
 llvm/lib/Target/Hexagon/HexagonDepArch.td |   2 +
 llvm/lib/Target/Hexagon/HexagonDepIICHVX.td   | 592 
 .../lib/Target/Hexagon/HexagonDepIICScalar.td | 888 ++
 llvm/lib/Target/Hexagon/HexagonSchedule.td|   1 +
 llvm/lib/Target/Hexagon/HexagonScheduleV75.td |  39 +
 llvm/lib/Target/Hexagon/HexagonSubtarget.h|   6 +
 .../MCTargetDesc/HexagonMCELFStreamer.cpp |   3 +
 .../MCTargetDesc/HexagonMCTargetDesc.cpp  |  41 +-
 llvm/test/MC/Hexagon/arch-support.s   |   9 +
 llvm/test/MC/Hexagon/v75_arch.s   |  10 +
 21 files changed, 1638 insertions(+), 19 deletions(-)
 create mode 100644 llvm/lib/Target/Hexagon/HexagonScheduleV75.td
 create mode 100644 llvm/test/MC/Hexagon/v75_arch.s

diff --git a/clang/include/clang/Basic/BuiltinsHexagon.def 
b/clang/include/clang/Basic/BuiltinsHexagon.def
index 0dc0f4567dd413..ca3f99176110e3 100644
--- a/clang/include/clang/Basic/BuiltinsHexagon.def
+++ b/clang/include/clang/Basic/BuiltinsHexagon.def
@@ -17,8 +17,10 @@
 #   define TARGET_BUILTIN(ID, TYPE, ATTRS, FEATURE) BUILTIN(ID, TYPE, ATTRS)
 #endif
 
+#pragma push_macro("V75")
+#define V75 "v75"
 #pragma push_macro("V73")
-#define V73 "v73"
+#define V73 "v73|" V75
 #pragma push_macro("V71")
 #define V71 "v71|" V73
 #pragma push_macro("V69")
@@ -40,8 +42,10 @@
 #pragma push_macro("V5")
 #define V5 "v5|" V55
 
+#pragma push_macro("HVXV75")
+#define HVXV75 "hvxv75"
 #pragma push_macro("HVXV73")
-#define HVXV73 "hvxv73"
+#define HVXV73 "hvxv73|" HVXV75
 #pragma push_macro("HVXV71")
 #define HVXV71 "hvxv71|" HVXV73
 #pragma push_macro("HVXV69")
@@ -143,6 +147,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("HVXV69")
 #pragma pop_macro("HVXV71")
 #pragma pop_macro("HVXV73")
+#pragma pop_macro("HVXV75")
 
 #pragma pop_macro("V5")
 #pragma pop_macro("V55")
@@ -155,6 +160,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("V69")
 #pragma pop_macro("V71")
 #pragma pop_macro("V73")
+#pragma pop_macro("V75")
 
 #undef BUILTIN
 #undef TARGET_BUILTIN
diff --git a/clang/include/clang/Driver/Options.td 
b/clang/include/clang/Driver/Options.td
index 638f8c52053ec5..9edf308cc685ea 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -6224,6 +6224,8 @@ def mv71t : Flag<["-"], "mv71t">, 
Group,
   Alias, AliasArgs<["hexagonv71t"]>;
 def mv73 : Flag<["-"], "mv73">, Group,
   Alias, AliasArgs<["hexagonv73"]>;
+def mv75 : Flag<["-"], "mv75">, Group,
+  Alias, AliasArgs<["hexagonv75"]>;
 def mhexagon_hvx : Flag<["-"], "mhvx">, Group,
   HelpText<"Enable Hexagon Vector eXtensions">;
 def mhexagon_hvx_EQ : Joined<["-"], "mhvx=">,
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp 
b/clang/lib/Basic/Targets/Hexagon.cpp
index 0282ac812c306f..1fce0d0ed006ec 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -78,6 +78,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions 
&Opts,
   } else if (CPU == "hexagonv73") {
 Builder.defineMacro("__HEXAGON_V73__");
 Builder.defineMacro("__HEXAGON_ARCH__", "73");
+  } else if (CPU == "hexagonv75") {
+Builder.defineMacro("__HEXAGON_V75__");
+Builder.defineMacro("__HEXAGON_ARCH__", "75");
   }
 
   if (hasFeature("hvx-length64b")) {
@@ -234,8 +237,8 @@ static constexpr CPUSuffix Suffixes[] = {
 {{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
 {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
 {{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
-{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"},  {"71t"}},
-{{"hexagonv73"}, {"73"}},
+{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"}, {"71t"}},
+{{"hexagonv73

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-backend-hexagon

Author: Ikhlas Ajbar (iajbar)


Changes

This patch introduces support for the Hexagon V75 architecture. It includes 
instruction formats, definitions, encodings, scheduling classes, and 
builtins/intrinsics.

---

Patch is 74.75 KiB, truncated to 20.00 KiB below, full version: 
https://github.com/llvm/llvm-project/pull/120773.diff


21 Files Affected:

- (modified) clang/include/clang/Basic/BuiltinsHexagon.def (+8-2) 
- (modified) clang/include/clang/Driver/Options.td (+2) 
- (modified) clang/lib/Basic/Targets/Hexagon.cpp (+5-2) 
- (modified) clang/test/Driver/hexagon-toolchain-elf.c (+7) 
- (modified) clang/test/Misc/target-invalid-cpu-note/hexagon.c (+1) 
- (modified) clang/test/Preprocessor/hexagon-predefines.c (+16) 
- (modified) llvm/include/llvm/BinaryFormat/ELF.h (+1) 
- (modified) llvm/lib/Object/ELFObjectFile.cpp (+2) 
- (modified) llvm/lib/ObjectYAML/ELFYAML.cpp (+2) 
- (modified) llvm/lib/Target/Hexagon/Hexagon.td (+14) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepArch.h (+3-1) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepArch.td (+2) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepIICHVX.td (+592) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepIICScalar.td (+888) 
- (modified) llvm/lib/Target/Hexagon/HexagonSchedule.td (+1) 
- (added) llvm/lib/Target/Hexagon/HexagonScheduleV75.td (+39) 
- (modified) llvm/lib/Target/Hexagon/HexagonSubtarget.h (+6) 
- (modified) llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCELFStreamer.cpp (+3) 
- (modified) llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp 
(+27-14) 
- (modified) llvm/test/MC/Hexagon/arch-support.s (+9) 
- (added) llvm/test/MC/Hexagon/v75_arch.s (+10) 


``diff
diff --git a/clang/include/clang/Basic/BuiltinsHexagon.def 
b/clang/include/clang/Basic/BuiltinsHexagon.def
index 0dc0f4567dd413..ca3f99176110e3 100644
--- a/clang/include/clang/Basic/BuiltinsHexagon.def
+++ b/clang/include/clang/Basic/BuiltinsHexagon.def
@@ -17,8 +17,10 @@
 #   define TARGET_BUILTIN(ID, TYPE, ATTRS, FEATURE) BUILTIN(ID, TYPE, ATTRS)
 #endif
 
+#pragma push_macro("V75")
+#define V75 "v75"
 #pragma push_macro("V73")
-#define V73 "v73"
+#define V73 "v73|" V75
 #pragma push_macro("V71")
 #define V71 "v71|" V73
 #pragma push_macro("V69")
@@ -40,8 +42,10 @@
 #pragma push_macro("V5")
 #define V5 "v5|" V55
 
+#pragma push_macro("HVXV75")
+#define HVXV75 "hvxv75"
 #pragma push_macro("HVXV73")
-#define HVXV73 "hvxv73"
+#define HVXV73 "hvxv73|" HVXV75
 #pragma push_macro("HVXV71")
 #define HVXV71 "hvxv71|" HVXV73
 #pragma push_macro("HVXV69")
@@ -143,6 +147,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("HVXV69")
 #pragma pop_macro("HVXV71")
 #pragma pop_macro("HVXV73")
+#pragma pop_macro("HVXV75")
 
 #pragma pop_macro("V5")
 #pragma pop_macro("V55")
@@ -155,6 +160,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("V69")
 #pragma pop_macro("V71")
 #pragma pop_macro("V73")
+#pragma pop_macro("V75")
 
 #undef BUILTIN
 #undef TARGET_BUILTIN
diff --git a/clang/include/clang/Driver/Options.td 
b/clang/include/clang/Driver/Options.td
index 638f8c52053ec5..9edf308cc685ea 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -6224,6 +6224,8 @@ def mv71t : Flag<["-"], "mv71t">, 
Group,
   Alias, AliasArgs<["hexagonv71t"]>;
 def mv73 : Flag<["-"], "mv73">, Group,
   Alias, AliasArgs<["hexagonv73"]>;
+def mv75 : Flag<["-"], "mv75">, Group,
+  Alias, AliasArgs<["hexagonv75"]>;
 def mhexagon_hvx : Flag<["-"], "mhvx">, Group,
   HelpText<"Enable Hexagon Vector eXtensions">;
 def mhexagon_hvx_EQ : Joined<["-"], "mhvx=">,
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp 
b/clang/lib/Basic/Targets/Hexagon.cpp
index 0282ac812c306f..1fce0d0ed006ec 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -78,6 +78,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions 
&Opts,
   } else if (CPU == "hexagonv73") {
 Builder.defineMacro("__HEXAGON_V73__");
 Builder.defineMacro("__HEXAGON_ARCH__", "73");
+  } else if (CPU == "hexagonv75") {
+Builder.defineMacro("__HEXAGON_V75__");
+Builder.defineMacro("__HEXAGON_ARCH__", "75");
   }
 
   if (hasFeature("hvx-length64b")) {
@@ -234,8 +237,8 @@ static constexpr CPUSuffix Suffixes[] = {
 {{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
 {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
 {{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
-{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"},  {"71t"}},
-{{"hexagonv73"}, {"73"}},
+{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"}, {"71t"}},
+{{"hexagonv73"}, {"73"}}, {{"hexagonv75"},  {"75"}},
 };
 
 std::optional HexagonTargetInfo::getHexagonCPURev(StringRef Name) {
diff --git a/clang/test/Driver/hexagon-toolchain-elf.c 
b/clang/test/Driver/hexagon-toolchain-elf.c
index 716d82bcf316b5..37d2

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang

Author: Ikhlas Ajbar (iajbar)


Changes

This patch introduces support for the Hexagon V75 architecture. It includes 
instruction formats, definitions, encodings, scheduling classes, and 
builtins/intrinsics.

---

Patch is 74.75 KiB, truncated to 20.00 KiB below, full version: 
https://github.com/llvm/llvm-project/pull/120773.diff


21 Files Affected:

- (modified) clang/include/clang/Basic/BuiltinsHexagon.def (+8-2) 
- (modified) clang/include/clang/Driver/Options.td (+2) 
- (modified) clang/lib/Basic/Targets/Hexagon.cpp (+5-2) 
- (modified) clang/test/Driver/hexagon-toolchain-elf.c (+7) 
- (modified) clang/test/Misc/target-invalid-cpu-note/hexagon.c (+1) 
- (modified) clang/test/Preprocessor/hexagon-predefines.c (+16) 
- (modified) llvm/include/llvm/BinaryFormat/ELF.h (+1) 
- (modified) llvm/lib/Object/ELFObjectFile.cpp (+2) 
- (modified) llvm/lib/ObjectYAML/ELFYAML.cpp (+2) 
- (modified) llvm/lib/Target/Hexagon/Hexagon.td (+14) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepArch.h (+3-1) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepArch.td (+2) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepIICHVX.td (+592) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepIICScalar.td (+888) 
- (modified) llvm/lib/Target/Hexagon/HexagonSchedule.td (+1) 
- (added) llvm/lib/Target/Hexagon/HexagonScheduleV75.td (+39) 
- (modified) llvm/lib/Target/Hexagon/HexagonSubtarget.h (+6) 
- (modified) llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCELFStreamer.cpp (+3) 
- (modified) llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp 
(+27-14) 
- (modified) llvm/test/MC/Hexagon/arch-support.s (+9) 
- (added) llvm/test/MC/Hexagon/v75_arch.s (+10) 


``diff
diff --git a/clang/include/clang/Basic/BuiltinsHexagon.def 
b/clang/include/clang/Basic/BuiltinsHexagon.def
index 0dc0f4567dd413..ca3f99176110e3 100644
--- a/clang/include/clang/Basic/BuiltinsHexagon.def
+++ b/clang/include/clang/Basic/BuiltinsHexagon.def
@@ -17,8 +17,10 @@
 #   define TARGET_BUILTIN(ID, TYPE, ATTRS, FEATURE) BUILTIN(ID, TYPE, ATTRS)
 #endif
 
+#pragma push_macro("V75")
+#define V75 "v75"
 #pragma push_macro("V73")
-#define V73 "v73"
+#define V73 "v73|" V75
 #pragma push_macro("V71")
 #define V71 "v71|" V73
 #pragma push_macro("V69")
@@ -40,8 +42,10 @@
 #pragma push_macro("V5")
 #define V5 "v5|" V55
 
+#pragma push_macro("HVXV75")
+#define HVXV75 "hvxv75"
 #pragma push_macro("HVXV73")
-#define HVXV73 "hvxv73"
+#define HVXV73 "hvxv73|" HVXV75
 #pragma push_macro("HVXV71")
 #define HVXV71 "hvxv71|" HVXV73
 #pragma push_macro("HVXV69")
@@ -143,6 +147,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("HVXV69")
 #pragma pop_macro("HVXV71")
 #pragma pop_macro("HVXV73")
+#pragma pop_macro("HVXV75")
 
 #pragma pop_macro("V5")
 #pragma pop_macro("V55")
@@ -155,6 +160,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("V69")
 #pragma pop_macro("V71")
 #pragma pop_macro("V73")
+#pragma pop_macro("V75")
 
 #undef BUILTIN
 #undef TARGET_BUILTIN
diff --git a/clang/include/clang/Driver/Options.td 
b/clang/include/clang/Driver/Options.td
index 638f8c52053ec5..9edf308cc685ea 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -6224,6 +6224,8 @@ def mv71t : Flag<["-"], "mv71t">, 
Group,
   Alias, AliasArgs<["hexagonv71t"]>;
 def mv73 : Flag<["-"], "mv73">, Group,
   Alias, AliasArgs<["hexagonv73"]>;
+def mv75 : Flag<["-"], "mv75">, Group,
+  Alias, AliasArgs<["hexagonv75"]>;
 def mhexagon_hvx : Flag<["-"], "mhvx">, Group,
   HelpText<"Enable Hexagon Vector eXtensions">;
 def mhexagon_hvx_EQ : Joined<["-"], "mhvx=">,
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp 
b/clang/lib/Basic/Targets/Hexagon.cpp
index 0282ac812c306f..1fce0d0ed006ec 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -78,6 +78,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions 
&Opts,
   } else if (CPU == "hexagonv73") {
 Builder.defineMacro("__HEXAGON_V73__");
 Builder.defineMacro("__HEXAGON_ARCH__", "73");
+  } else if (CPU == "hexagonv75") {
+Builder.defineMacro("__HEXAGON_V75__");
+Builder.defineMacro("__HEXAGON_ARCH__", "75");
   }
 
   if (hasFeature("hvx-length64b")) {
@@ -234,8 +237,8 @@ static constexpr CPUSuffix Suffixes[] = {
 {{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
 {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
 {{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
-{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"},  {"71t"}},
-{{"hexagonv73"}, {"73"}},
+{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"}, {"71t"}},
+{{"hexagonv73"}, {"73"}}, {{"hexagonv75"},  {"75"}},
 };
 
 std::optional HexagonTargetInfo::getHexagonCPURev(StringRef Name) {
diff --git a/clang/test/Driver/hexagon-toolchain-elf.c 
b/clang/test/Driver/hexagon-toolchain-elf.c
index 716d82bcf316b5..37d24d4b2b99dc

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[via cfe-commits]

llvmbot wrote:



@llvm/pr-subscribers-objectyaml

@llvm/pr-subscribers-clang-driver

Author: Ikhlas Ajbar (iajbar)


Changes

This patch introduces support for the Hexagon V75 architecture. It includes 
instruction formats, definitions, encodings, scheduling classes, and 
builtins/intrinsics.

---

Patch is 74.75 KiB, truncated to 20.00 KiB below, full version: 
https://github.com/llvm/llvm-project/pull/120773.diff


21 Files Affected:

- (modified) clang/include/clang/Basic/BuiltinsHexagon.def (+8-2) 
- (modified) clang/include/clang/Driver/Options.td (+2) 
- (modified) clang/lib/Basic/Targets/Hexagon.cpp (+5-2) 
- (modified) clang/test/Driver/hexagon-toolchain-elf.c (+7) 
- (modified) clang/test/Misc/target-invalid-cpu-note/hexagon.c (+1) 
- (modified) clang/test/Preprocessor/hexagon-predefines.c (+16) 
- (modified) llvm/include/llvm/BinaryFormat/ELF.h (+1) 
- (modified) llvm/lib/Object/ELFObjectFile.cpp (+2) 
- (modified) llvm/lib/ObjectYAML/ELFYAML.cpp (+2) 
- (modified) llvm/lib/Target/Hexagon/Hexagon.td (+14) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepArch.h (+3-1) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepArch.td (+2) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepIICHVX.td (+592) 
- (modified) llvm/lib/Target/Hexagon/HexagonDepIICScalar.td (+888) 
- (modified) llvm/lib/Target/Hexagon/HexagonSchedule.td (+1) 
- (added) llvm/lib/Target/Hexagon/HexagonScheduleV75.td (+39) 
- (modified) llvm/lib/Target/Hexagon/HexagonSubtarget.h (+6) 
- (modified) llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCELFStreamer.cpp (+3) 
- (modified) llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp 
(+27-14) 
- (modified) llvm/test/MC/Hexagon/arch-support.s (+9) 
- (added) llvm/test/MC/Hexagon/v75_arch.s (+10) 


``diff
diff --git a/clang/include/clang/Basic/BuiltinsHexagon.def 
b/clang/include/clang/Basic/BuiltinsHexagon.def
index 0dc0f4567dd413..ca3f99176110e3 100644
--- a/clang/include/clang/Basic/BuiltinsHexagon.def
+++ b/clang/include/clang/Basic/BuiltinsHexagon.def
@@ -17,8 +17,10 @@
 #   define TARGET_BUILTIN(ID, TYPE, ATTRS, FEATURE) BUILTIN(ID, TYPE, ATTRS)
 #endif
 
+#pragma push_macro("V75")
+#define V75 "v75"
 #pragma push_macro("V73")
-#define V73 "v73"
+#define V73 "v73|" V75
 #pragma push_macro("V71")
 #define V71 "v71|" V73
 #pragma push_macro("V69")
@@ -40,8 +42,10 @@
 #pragma push_macro("V5")
 #define V5 "v5|" V55
 
+#pragma push_macro("HVXV75")
+#define HVXV75 "hvxv75"
 #pragma push_macro("HVXV73")
-#define HVXV73 "hvxv73"
+#define HVXV73 "hvxv73|" HVXV75
 #pragma push_macro("HVXV71")
 #define HVXV71 "hvxv71|" HVXV73
 #pragma push_macro("HVXV69")
@@ -143,6 +147,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("HVXV69")
 #pragma pop_macro("HVXV71")
 #pragma pop_macro("HVXV73")
+#pragma pop_macro("HVXV75")
 
 #pragma pop_macro("V5")
 #pragma pop_macro("V55")
@@ -155,6 +160,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("V69")
 #pragma pop_macro("V71")
 #pragma pop_macro("V73")
+#pragma pop_macro("V75")
 
 #undef BUILTIN
 #undef TARGET_BUILTIN
diff --git a/clang/include/clang/Driver/Options.td 
b/clang/include/clang/Driver/Options.td
index 638f8c52053ec5..9edf308cc685ea 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -6224,6 +6224,8 @@ def mv71t : Flag<["-"], "mv71t">, 
Group,
   Alias, AliasArgs<["hexagonv71t"]>;
 def mv73 : Flag<["-"], "mv73">, Group,
   Alias, AliasArgs<["hexagonv73"]>;
+def mv75 : Flag<["-"], "mv75">, Group,
+  Alias, AliasArgs<["hexagonv75"]>;
 def mhexagon_hvx : Flag<["-"], "mhvx">, Group,
   HelpText<"Enable Hexagon Vector eXtensions">;
 def mhexagon_hvx_EQ : Joined<["-"], "mhvx=">,
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp 
b/clang/lib/Basic/Targets/Hexagon.cpp
index 0282ac812c306f..1fce0d0ed006ec 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -78,6 +78,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions 
&Opts,
   } else if (CPU == "hexagonv73") {
 Builder.defineMacro("__HEXAGON_V73__");
 Builder.defineMacro("__HEXAGON_ARCH__", "73");
+  } else if (CPU == "hexagonv75") {
+Builder.defineMacro("__HEXAGON_V75__");
+Builder.defineMacro("__HEXAGON_ARCH__", "75");
   }
 
   if (hasFeature("hvx-length64b")) {
@@ -234,8 +237,8 @@ static constexpr CPUSuffix Suffixes[] = {
 {{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
 {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
 {{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
-{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"},  {"71t"}},
-{{"hexagonv73"}, {"73"}},
+{{"hexagonv71"}, {"71"}}, {{"hexagonv71t"}, {"71t"}},
+{{"hexagonv73"}, {"73"}}, {{"hexagonv75"},  {"75"}},
 };
 
 std::optional HexagonTargetInfo::getHexagonCPURev(StringRef Name) {
diff --git a/clang/test/Driver/hexagon-toolchain-elf.c 
b/clang/test/Driver/hexagon-toolchain-elf

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[via cfe-commits]

github-actions[bot] wrote:




:warning: C/C++ code formatter, clang-format found issues in your code. 
:warning:



You can test this locally with the following command:


``bash
git-clang-format --diff 7739380643718bc912bc05b969e4be525a85c0d2 
1d3ac02da5bfb00d86b123334e3a75628e47ef42 --extensions h,c,cpp -- 
clang/lib/Basic/Targets/Hexagon.cpp clang/test/Driver/hexagon-toolchain-elf.c 
clang/test/Misc/target-invalid-cpu-note/hexagon.c 
clang/test/Preprocessor/hexagon-predefines.c 
llvm/include/llvm/BinaryFormat/ELF.h llvm/lib/Object/ELFObjectFile.cpp 
llvm/lib/ObjectYAML/ELFYAML.cpp llvm/lib/Target/Hexagon/HexagonDepArch.h 
llvm/lib/Target/Hexagon/HexagonSubtarget.h 
llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCELFStreamer.cpp 
llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
``





View the diff from clang-format here.


``diff
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp 
b/clang/lib/Basic/Targets/Hexagon.cpp
index 1fce0d0ed0..931327bd86 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -232,13 +232,13 @@ struct CPUSuffix {
 };
 
 static constexpr CPUSuffix Suffixes[] = {
-{{"hexagonv5"},  {"5"}},  {{"hexagonv55"},  {"55"}},
-{{"hexagonv60"}, {"60"}}, {{"hexagonv62"},  {"62"}},
-{{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
+{{"hexagonv5"}, {"5"}},   {{"hexagonv55"}, {"55"}},
+{{"hexagonv60"}, {"60"}}, {{"hexagonv62"}, {"62"}},
+{{"hexagonv65"}, {"65"}}, {{"hexagonv66"}, {"66"}},
 {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
-{{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
+{{"hexagonv68"}, {"68"}}, {{"hexagonv69"}, {"69"}},
 {{"hexagonv71"}, {"71"}}, {{"hexagonv71t"}, {"71t"}},
-{{"hexagonv73"}, {"73"}}, {{"hexagonv75"},  {"75"}},
+{{"hexagonv73"}, {"73"}}, {{"hexagonv75"}, {"75"}},
 };
 
 std::optional HexagonTargetInfo::getHexagonCPURev(StringRef Name) {
diff --git a/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp 
b/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
index 40d2b122d4..dfa1f5471a 100644
--- a/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
+++ b/llvm/lib/Target/Hexagon/MCTargetDesc/HexagonMCTargetDesc.cpp
@@ -453,18 +453,18 @@ std::string selectHexagonFS(StringRef CPU, StringRef FS) {
 break;
   case Hexagon::ArchEnum::Generic: {
 Result.push_back(StringSwitch(CPU)
- .Case("hexagonv60", "+hvxv60")
- .Case("hexagonv62", "+hvxv62")
- .Case("hexagonv65", "+hvxv65")
- .Case("hexagonv66", "+hvxv66")
- .Case("hexagonv67", "+hvxv67")
- .Case("hexagonv67t", "+hvxv67")
- .Case("hexagonv68", "+hvxv68")
- .Case("hexagonv69", "+hvxv69")
- .Case("hexagonv71", "+hvxv71")
- .Case("hexagonv71t", "+hvxv71")
- .Case("hexagonv73", "+hvxv73")
- .Case("hexagonv75", "+hvxv75"));
+ .Case("hexagonv60", "+hvxv60")
+ .Case("hexagonv62", "+hvxv62")
+ .Case("hexagonv65", "+hvxv65")
+ .Case("hexagonv66", "+hvxv66")
+ .Case("hexagonv67", "+hvxv67")
+ .Case("hexagonv67t", "+hvxv67")
+ .Case("hexagonv68", "+hvxv68")
+ .Case("hexagonv69", "+hvxv69")
+ .Case("hexagonv71", "+hvxv71")
+ .Case("hexagonv71t", "+hvxv71")
+ .Case("hexagonv73", "+hvxv73")
+ .Case("hexagonv75", "+hvxv75"));
 break;
   }
   case Hexagon::ArchEnum::NoArch:
@@ -512,8 +512,8 @@ FeatureBitset Hexagon_MC::completeHVXFeatures(const 
FeatureBitset &S) {
   // turns on hvxvNN, corresponding to the existing ArchVNN.
   FeatureBitset FB = S;
   unsigned CpuArch = ArchV5;
-  for (unsigned F : {ArchV75, ArchV73, ArchV71, ArchV69, ArchV68, ArchV67, 
ArchV66,
- ArchV65, ArchV62, ArchV60, ArchV55, ArchV5}) {
+  for (unsigned F : {ArchV75, ArchV73, ArchV71, ArchV69, ArchV68, ArchV67,
+ ArchV66, ArchV65, ArchV62, ArchV60, ArchV55, ArchV5}) {
 if (!FB.test(F))
   continue;
 CpuArch = F;
@@ -527,9 +527,10 @@ FeatureBitset Hexagon_MC::completeHVXFeatures(const 
FeatureBitset &S) {
 break;
   }
   bool HasHvxVer = false;
-  for (unsigned F : {ExtensionHVXV60, ExtensionHVXV62, ExtensionHVXV65,
- ExtensionHVXV66, ExtensionHVXV67, ExtensionHVXV68,
-ExtensionHVXV69, ExtensionHVXV71, ExtensionHVXV73, ExtensionHVXV75}) {
+  for (unsigned F :
+   {ExtensionHVXV60, ExtensionHVXV62, ExtensionHVXV65, ExtensionHVXV66,
+ExtensionHVXV67, ExtensionHVXV68, ExtensionHVXV69, ExtensionHVXV71,
+ExtensionHVXV73, ExtensionHVXV75}) {
 if (!FB.test(F))
   continue;
 HasHvxVer = true;

``




https://github.com/llvm/llvm-project/pull/120773

[clang] [clang][ObjectiveC] Fix Parsing Types with the `::` Optional Scope Specifier (PR #119908)

[Qiongsi Wu via cfe-commits]

qiongsiwu wrote:

Gentle ping for review. Thanks so much! 

https://github.com/llvm/llvm-project/pull/119908
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (PR #119908)

[Qiongsi Wu via cfe-commits]

https://github.com/qiongsiwu edited 
https://github.com/llvm/llvm-project/pull/119908
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Fix non-deterministic infinite recursion... (PR #118288)

[Hubert Tong via cfe-commits]

Alejandro =?utf-8?q?Álvarez_Ayllón?Message-ID:
In-Reply-To: 


hubert-reinterpretcast wrote:

> LLVM patch releases cannot break API or ABI compatibility

Thanks @nikic for the explanation. I thought you meant ABI compatibility in the 
"user program" sense.

https://github.com/llvm/llvm-project/pull/118288
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (PR #120753)

[Paul Walker via cfe-commits]

https://github.com/paulwalker-arm approved this pull request.


https://github.com/llvm/llvm-project/pull/120753
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86][AVX10.2] Add comments for the avx10_2convertintrin.h file (PR #120766)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang

Author: Mikołaj Piróg (mikolaj-pirog)


Changes

As in title. I will create a sibling pr with comments to the 512 in a separate 
PR.

---

Patch is 111.62 KiB, truncated to 20.00 KiB below, full version: 
https://github.com/llvm/llvm-project/pull/120766.diff


1 Files Affected:

- (modified) clang/lib/Headers/avx10_2convertintrin.h (+2696) 


``diff
diff --git a/clang/lib/Headers/avx10_2convertintrin.h 
b/clang/lib/Headers/avx10_2convertintrin.h
index 134adb2850c8de..52a64a5309763d 100644
--- a/clang/lib/Headers/avx10_2convertintrin.h
+++ b/clang/lib/Headers/avx10_2convertintrin.h
@@ -24,24 +24,146 @@
   __attribute__((__always_inline__, __nodebug__, __target__("avx10.2-256"),
\
  __min_vector_width__(256)))
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed
+/// single-precision (32-bit) floating-point elements to a 128-bit vector
+/// containing FP16 elements.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+///IF i < 4
+///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+///ELSE
+///dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 4])
+///FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile 
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __A
+///A 128-bit vector of [4 x float].
+/// \param __B
+///A 128-bit vector of [4 x float].
+/// \returns
+///A 128-bit vector of [8 x fp16]. Lower elements correspond to the
+///(converted) elements from \a __B; higher order elements correspond to 
the
+///(converted) elements from \a __A.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128 _mm_cvtx2ps_ph(__m128 __A,
__m128 __B) {
   return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
   (__v4sf)__A, (__v4sf)__B, (__v8hf)_mm_setzero_ph(), (__mmask8)(-1));
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed
+/// single-precision (32-bit) floating-point elements to a 128-bit vector
+/// containing FP16 elements. Merging mask \a __U is used to determine if given
+/// element should be taken from \a __W instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+///IF mask[i]
+///dst.fp16[i] := __W[i]
+///ELSE
+///IF i < 4
+///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+///ELSE
+///dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 4])
+///FI
+///FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile 
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __W
+///A 128-bit vector of [8 x fp16].
+/// \param __U
+///A 8-bit merging mask.
+/// \param __A
+///A 128-bit vector of [4 x float].
+/// \param __B
+///A 128-bit vector of [4 x float].
+/// \returns
+///A 128-bit vector of [8 x fp16]. Lower elements correspond to the
+///(converted) elements from \a __B; higher order elements correspond to 
the
+///(converted) elements from \a __A. If corresponding mask bit is set, then
+///element from \a __W is taken instead.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
 _mm_mask_cvtx2ps_ph(__m128h __W, __mmask8 __U, __m128 __A, __m128 __B) {
   return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
   (__v4sf)__A, (__v4sf)__B, (__v8hf)__W, (__mmask8)__U);
 }
 
+/// Convert two 128-bit vectors, \a __A and \a __B, containing packed
+/// single-precision (32-bit) floating-point elements to a 128-bit vector
+/// containing FP16 elements. Zeroing mask \a __U is used to determine if given
+/// element should be zeroed instead.
+///
+/// \code{.operation}
+/// FOR i := 0 to 7
+///IF mask[i]
+///IF i < 4
+///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
+///ELSE
+///dst.fp16[i] := convert_fp32_to_fp16(__A.fp32[i - 4])
+///FI
+///ELSE
+///dst.fp16[i] := 0
+///FI
+/// ENDFOR
+/// \endcode
+///
+/// \headerfile 
+///
+/// This intrinsic corresponds to the \c VCVT2PS2PHX instruction.
+///
+/// \param __U
+///A 8-bit zeroing mask.
+/// \param __A
+///A 128-bit vector of [4 x float].
+/// \param __B
+///A 128-bit vector of [4 x float].
+/// \returns
+///A 128-bit vector of [8 x fp16]. Lower elements correspond to the
+///(converted) elements from \a __B; higher order elements correspond to 
the
+///(converted) elements from \a __A. If corresponding mask bit is not set,
+///then zero is taken instead.
 static __inline__ __m128h __DEFAULT_FN_ATTRS128
 _mm_maskz_cvtx2ps_ph(__mmask8 __U, __m128 __A, __m128 __B) {
   return (__m128h)__builtin_ia32_vcvt2ps2phx128_mask(
   (__v4sf)__A, (__v4sf)__B, (__v8hf)_mm_setzero_ph(), (__mmask8)__U);
 }
 
+/// Convert two 256-bit vectors, \a __A and \a __B, containing packed
+/// single-precision (32-bit) floating-point elements to a

[clang] [X86][AVX10.2] Add comments for the avx10_2convertintrin.h file (PR #120766)

[via cfe-commits]

github-actions[bot] wrote:




:warning: C/C++ code formatter, clang-format found issues in your code. 
:warning:



You can test this locally with the following command:


``bash
git-clang-format --diff d5fe6332c9f25590b9878ad31d6461ea581e0fa5 
b069337d461da42883093aedeb00c019131d3559 --extensions h -- 
clang/lib/Headers/avx10_2convertintrin.h
``





View the diff from clang-format here.


``diff
diff --git a/clang/lib/Headers/avx10_2convertintrin.h 
b/clang/lib/Headers/avx10_2convertintrin.h
index 52a64a5309..38d9f1982b 100644
--- a/clang/lib/Headers/avx10_2convertintrin.h
+++ b/clang/lib/Headers/avx10_2convertintrin.h
@@ -143,7 +143,7 @@ _mm_maskz_cvtx2ps_ph(__mmask8 __U, __m128 __A, __m128 __B) {
 /// containing FP16 elements.
 ///
 /// \code{.operation}
-/// FOR i := 0 to 15 
+/// FOR i := 0 to 15
 ///IF i < 8
 ///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
 ///ELSE
@@ -220,7 +220,7 @@ _mm256_mask_cvtx2ps_ph(__m256h __W, __mmask16 __U, __m256 
__A, __m256 __B) {
 /// element should be zeroed instead.
 ///
 /// \code{.operation}
-/// FOR i := 0 to 15 
+/// FOR i := 0 to 15
 ///IF mask[i]
 ///IF i < 8
 ///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
@@ -260,7 +260,7 @@ _mm256_maskz_cvtx2ps_ph(__mmask16 __U, __m256 __A, __m256 
__B) {
 /// containing FP16 elements. Rounding mode \a __R needs to be provided.
 ///
 /// \code{.operation}
-/// FOR i := 0 to 15 
+/// FOR i := 0 to 15
 ///IF i < 8
 ///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
 ///ELSE
@@ -279,9 +279,9 @@ _mm256_maskz_cvtx2ps_ph(__mmask16 __U, __m256 __A, __m256 
__B) {
 ///A 256-bit vector of [8 x float].
 /// \param __R
 ///Rounding mode. Valid inputs are: _MM_FROUND_CUR_DIRECTION or
-///result of bitwise or of _MM_FROUND_NO_EXC with at most one of the 
following:
-///_MM_FROUND_TO_NEAREST_INT, _MM_FROUND_TO_NEG_INF, _MM_FROUND_TO_POS_INF,
-///_MM_FROUND_TO_ZERO.
+///result of bitwise or of _MM_FROUND_NO_EXC with at most one of the
+///following: _MM_FROUND_TO_NEAREST_INT, _MM_FROUND_TO_NEG_INF,
+///_MM_FROUND_TO_POS_INF, _MM_FROUND_TO_ZERO.
 /// \returns
 ///A 256-bit vector of [16 x fp16]. Lower elements correspond to the
 ///(converted) elements from \a __B; higher order elements correspond to 
the
@@ -325,9 +325,9 @@ _mm256_maskz_cvtx2ps_ph(__mmask16 __U, __m256 __A, __m256 
__B) {
 ///A 256-bit vector of [8 x float].
 /// \param __R
 ///Rounding mode. Valid inputs are: _MM_FROUND_CUR_DIRECTION or
-///result of bitwise or of _MM_FROUND_NO_EXC with at most one of the 
following:
-///_MM_FROUND_TO_NEAREST_INT, _MM_FROUND_TO_NEG_INF, _MM_FROUND_TO_POS_INF,
-///_MM_FROUND_TO_ZERO.
+///result of bitwise or of _MM_FROUND_NO_EXC with at most one of the
+///following: _MM_FROUND_TO_NEAREST_INT, _MM_FROUND_TO_NEG_INF,
+///_MM_FROUND_TO_POS_INF, _MM_FROUND_TO_ZERO.
 /// \returns
 ///A 256-bit vector of [16 x fp16]. Lower elements correspond to the
 ///(converted) elements from \a __B; higher order elements correspond to 
the
@@ -343,7 +343,7 @@ _mm256_maskz_cvtx2ps_ph(__mmask16 __U, __m256 __A, __m256 
__B) {
 /// element should be zeroed instead. Rounding mode \a __R needs to be 
provided.
 ///
 /// \code{.operation}
-/// FOR i := 0 to 15 
+/// FOR i := 0 to 15
 ///IF mask[i]
 ///IF i < 8
 ///dst.fp16[i] := convert_fp32_to_fp16(__B.fp32[i])
@@ -368,9 +368,9 @@ _mm256_maskz_cvtx2ps_ph(__mmask16 __U, __m256 __A, __m256 
__B) {
 ///A 256-bit vector of [8 x float].
 /// \param __R
 ///Rounding mode. Valid inputs are: _MM_FROUND_CUR_DIRECTION or
-///result of bitwise or of _MM_FROUND_NO_EXC with at most one of the 
following:
-///_MM_FROUND_TO_NEAREST_INT, _MM_FROUND_TO_NEG_INF, _MM_FROUND_TO_POS_INF,
-///_MM_FROUND_TO_ZERO.
+///result of bitwise or of _MM_FROUND_NO_EXC with at most one of the
+///following: _MM_FROUND_TO_NEAREST_INT, _MM_FROUND_TO_NEG_INF,
+///_MM_FROUND_TO_POS_INF, _MM_FROUND_TO_ZERO.
 /// \returns
 ///A 256-bit vector of [16 x fp16]. Lower elements correspond to the
 ///(converted) elements from \a __B; higher order elements correspond to 
the
@@ -421,9 +421,8 @@ _mm_cvtbiasph_pbf8(__m128i __A, __m128h __B) {
 ///IF __U[i]
 ///dst.fp8[i] := _W[i]
 ///ELSE
-///dst.fp8[i] := add_convert_fp16_to_fp8_bias(__A.fp16[i], 
__B.int8[2i])
-///FI
-/// ENDFOR
+///dst.fp8[i] := add_convert_fp16_to_fp8_bias(__A.fp16[i],
+/// __B.int8[2i])  FI ENDFOR
 ///
 /// dst[127:64] := 0
 /// \endcode
@@ -459,10 +458,8 @@ _mm_mask_cvtbiasph_pbf8(__m128i __W, __mmask8 __U, __m128i 
__A, __m128h __B) {
 /// \code{.operation}
 /// FOR i := 0 to 7
 ///IF __U[i]
-///dst.fp8[i] := add_convert_fp16_to_fp8_bias(__A.fp16[i], 
__B.int8[2i])
-/// ELSE
-///dst.fp8[i] := 0
-/// FI
+///dst.fp8[i] := add_c

[clang-tools-extra] [clang-tidy][NFC] extract options verify to sperately function (PR #120768)

[Congcong Cai via cfe-commits]

https://github.com/HerrCai0907 created 
https://github.com/llvm/llvm-project/pull/120768

None

>From 0543d52cd0af5ff645bfef0f41d4c1ffcd4b4962 Mon Sep 17 00:00:00 2001
From: Congcong Cai 
Date: Sat, 21 Dec 2024 00:25:00 +0800
Subject: [PATCH] [clang-tidy][NFC] extract options verify to sperately
 function

---
 clang-tools-extra/clang-tidy/ClangTidy.cpp|  8 ++---
 clang-tools-extra/clang-tidy/ClangTidy.h  |  6 ++--
 .../clang-tidy/tool/ClangTidyMain.cpp | 36 +++
 3 files changed, 28 insertions(+), 22 deletions(-)

diff --git a/clang-tools-extra/clang-tidy/ClangTidy.cpp 
b/clang-tools-extra/clang-tidy/ClangTidy.cpp
index 9c8c93c5d16c72..959b11777e88d4 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.cpp
+++ b/clang-tools-extra/clang-tidy/ClangTidy.cpp
@@ -646,9 +646,9 @@ void exportReplacements(const llvm::StringRef MainFilePath,
   YAML << TUD;
 }
 
-NamesAndOptions
+ChecksAndOptions
 getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers) {
-  NamesAndOptions Result;
+  ChecksAndOptions Result;
   ClangTidyOptions Opts;
   Opts.Checks = "*";
   clang::tidy::ClangTidyContext Context(
@@ -661,7 +661,7 @@ getAllChecksAndOptions(bool 
AllowEnablingAnalyzerAlphaCheckers) {
   }
 
   for (const auto &Factory : Factories)
-Result.Names.insert(Factory.getKey());
+Result.Checks.insert(Factory.getKey());
 
 #if CLANG_TIDY_ENABLE_STATIC_ANALYZER
   SmallString<64> Buffer(AnalyzerCheckNamePrefix);
@@ -670,7 +670,7 @@ getAllChecksAndOptions(bool 
AllowEnablingAnalyzerAlphaCheckers) {
AllowEnablingAnalyzerAlphaCheckers)) {
 Buffer.truncate(DefSize);
 Buffer.append(AnalyzerCheck);
-Result.Names.insert(Buffer);
+Result.Checks.insert(Buffer);
   }
   for (std::string OptionName : {
 #define GET_CHECKER_OPTIONS
diff --git a/clang-tools-extra/clang-tidy/ClangTidy.h 
b/clang-tools-extra/clang-tidy/ClangTidy.h
index 51d9e226c79465..4ffd49f6ebf50f 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.h
+++ b/clang-tools-extra/clang-tidy/ClangTidy.h
@@ -58,12 +58,12 @@ class ClangTidyASTConsumerFactory {
 std::vector getCheckNames(const ClangTidyOptions &Options,
bool 
AllowEnablingAnalyzerAlphaCheckers);
 
-struct NamesAndOptions {
-  llvm::StringSet<> Names;
+struct ChecksAndOptions {
+  llvm::StringSet<> Checks;
   llvm::StringSet<> Options;
 };
 
-NamesAndOptions
+ChecksAndOptions
 getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers = true);
 
 /// Returns the effective check-specific options.
diff --git a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp 
b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
index d42dafa8ffc362..b8d843cba71330 100644
--- a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
+++ b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
@@ -526,6 +526,24 @@ static bool verifyFileExtensions(
   return AnyInvalid;
 }
 
+static bool verifyOptions(const llvm::StringSet<> &ValidOptions,
+  const ClangTidyOptions::OptionMap &OptionMap,
+  StringRef Source) {
+  bool AnyInvalid = false;
+  for (auto Key : OptionMap.keys()) {
+if (ValidOptions.contains(Key))
+  continue;
+AnyInvalid = true;
+auto &Output = llvm::WithColor::warning(llvm::errs(), Source)
+   << "unknown check option '" << Key << '\'';
+llvm::StringRef Closest = closest(Key, ValidOptions);
+if (!Closest.empty())
+  Output << "; did you mean '" << Closest << '\'';
+Output << VerifyConfigWarningEnd;
+  }
+  return AnyInvalid;
+}
+
 static SmallString<256> makeAbsolute(llvm::StringRef Input) {
   if (Input.empty())
 return {};
@@ -629,29 +647,17 @@ int clangTidyMain(int argc, const char **argv) {
   if (VerifyConfig) {
 std::vector RawOptions =
 OptionsProvider->getRawOptions(FileName);
-NamesAndOptions Valid =
+ChecksAndOptions Valid =
 getAllChecksAndOptions(AllowEnablingAnalyzerAlphaCheckers);
 bool AnyInvalid = false;
 for (const auto &[Opts, Source] : RawOptions) {
   if (Opts.Checks)
-AnyInvalid |= verifyChecks(Valid.Names, *Opts.Checks, Source);
-
+AnyInvalid |= verifyChecks(Valid.Checks, *Opts.Checks, Source);
   if (Opts.HeaderFileExtensions && Opts.ImplementationFileExtensions)
 AnyInvalid |=
 verifyFileExtensions(*Opts.HeaderFileExtensions,
  *Opts.ImplementationFileExtensions, Source);
-
-  for (auto Key : Opts.CheckOptions.keys()) {
-if (Valid.Options.contains(Key))
-  continue;
-AnyInvalid = true;
-auto &Output = llvm::WithColor::warning(llvm::errs(), Source)
-   << "unknown check option '" << Key << '\'';
-llvm::StringRef Closest = closest(Key, Valid.Options);
-if (!Closest.empty())
-  Output << "; did you mean '" << Closest << '\'';
-Output << VerifyConfigWarningEnd;
-  }
+  AnyInvalid |= verify

[clang-tools-extra] [clang-tidy][NFC] extract options verify to sperately function (PR #120768)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-tools-extra

Author: Congcong Cai (HerrCai0907)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/120768.diff


3 Files Affected:

- (modified) clang-tools-extra/clang-tidy/ClangTidy.cpp (+4-4) 
- (modified) clang-tools-extra/clang-tidy/ClangTidy.h (+3-3) 
- (modified) clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp (+21-15) 


``diff
diff --git a/clang-tools-extra/clang-tidy/ClangTidy.cpp 
b/clang-tools-extra/clang-tidy/ClangTidy.cpp
index 9c8c93c5d16c72..959b11777e88d4 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.cpp
+++ b/clang-tools-extra/clang-tidy/ClangTidy.cpp
@@ -646,9 +646,9 @@ void exportReplacements(const llvm::StringRef MainFilePath,
   YAML << TUD;
 }
 
-NamesAndOptions
+ChecksAndOptions
 getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers) {
-  NamesAndOptions Result;
+  ChecksAndOptions Result;
   ClangTidyOptions Opts;
   Opts.Checks = "*";
   clang::tidy::ClangTidyContext Context(
@@ -661,7 +661,7 @@ getAllChecksAndOptions(bool 
AllowEnablingAnalyzerAlphaCheckers) {
   }
 
   for (const auto &Factory : Factories)
-Result.Names.insert(Factory.getKey());
+Result.Checks.insert(Factory.getKey());
 
 #if CLANG_TIDY_ENABLE_STATIC_ANALYZER
   SmallString<64> Buffer(AnalyzerCheckNamePrefix);
@@ -670,7 +670,7 @@ getAllChecksAndOptions(bool 
AllowEnablingAnalyzerAlphaCheckers) {
AllowEnablingAnalyzerAlphaCheckers)) {
 Buffer.truncate(DefSize);
 Buffer.append(AnalyzerCheck);
-Result.Names.insert(Buffer);
+Result.Checks.insert(Buffer);
   }
   for (std::string OptionName : {
 #define GET_CHECKER_OPTIONS
diff --git a/clang-tools-extra/clang-tidy/ClangTidy.h 
b/clang-tools-extra/clang-tidy/ClangTidy.h
index 51d9e226c79465..4ffd49f6ebf50f 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.h
+++ b/clang-tools-extra/clang-tidy/ClangTidy.h
@@ -58,12 +58,12 @@ class ClangTidyASTConsumerFactory {
 std::vector getCheckNames(const ClangTidyOptions &Options,
bool 
AllowEnablingAnalyzerAlphaCheckers);
 
-struct NamesAndOptions {
-  llvm::StringSet<> Names;
+struct ChecksAndOptions {
+  llvm::StringSet<> Checks;
   llvm::StringSet<> Options;
 };
 
-NamesAndOptions
+ChecksAndOptions
 getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers = true);
 
 /// Returns the effective check-specific options.
diff --git a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp 
b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
index d42dafa8ffc362..b8d843cba71330 100644
--- a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
+++ b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
@@ -526,6 +526,24 @@ static bool verifyFileExtensions(
   return AnyInvalid;
 }
 
+static bool verifyOptions(const llvm::StringSet<> &ValidOptions,
+  const ClangTidyOptions::OptionMap &OptionMap,
+  StringRef Source) {
+  bool AnyInvalid = false;
+  for (auto Key : OptionMap.keys()) {
+if (ValidOptions.contains(Key))
+  continue;
+AnyInvalid = true;
+auto &Output = llvm::WithColor::warning(llvm::errs(), Source)
+   << "unknown check option '" << Key << '\'';
+llvm::StringRef Closest = closest(Key, ValidOptions);
+if (!Closest.empty())
+  Output << "; did you mean '" << Closest << '\'';
+Output << VerifyConfigWarningEnd;
+  }
+  return AnyInvalid;
+}
+
 static SmallString<256> makeAbsolute(llvm::StringRef Input) {
   if (Input.empty())
 return {};
@@ -629,29 +647,17 @@ int clangTidyMain(int argc, const char **argv) {
   if (VerifyConfig) {
 std::vector RawOptions =
 OptionsProvider->getRawOptions(FileName);
-NamesAndOptions Valid =
+ChecksAndOptions Valid =
 getAllChecksAndOptions(AllowEnablingAnalyzerAlphaCheckers);
 bool AnyInvalid = false;
 for (const auto &[Opts, Source] : RawOptions) {
   if (Opts.Checks)
-AnyInvalid |= verifyChecks(Valid.Names, *Opts.Checks, Source);
-
+AnyInvalid |= verifyChecks(Valid.Checks, *Opts.Checks, Source);
   if (Opts.HeaderFileExtensions && Opts.ImplementationFileExtensions)
 AnyInvalid |=
 verifyFileExtensions(*Opts.HeaderFileExtensions,
  *Opts.ImplementationFileExtensions, Source);
-
-  for (auto Key : Opts.CheckOptions.keys()) {
-if (Valid.Options.contains(Key))
-  continue;
-AnyInvalid = true;
-auto &Output = llvm::WithColor::warning(llvm::errs(), Source)
-   << "unknown check option '" << Key << '\'';
-llvm::StringRef Closest = closest(Key, Valid.Options);
-if (!Closest.empty())
-  Output << "; did you mean '" << Closest << '\'';
-Output << VerifyConfigWarningEnd;
-  }
+  AnyInvalid |= verifyOptions(Valid.Options, Opts.CheckOptions, Source);
 }
 if (AnyInvalid)
   return 1;

``




https://github.com/llvm/l

[clang-tools-extra] [clang-tidy][NFC] extract options verify to sperately function (PR #120768)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-tidy

Author: Congcong Cai (HerrCai0907)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/120768.diff


3 Files Affected:

- (modified) clang-tools-extra/clang-tidy/ClangTidy.cpp (+4-4) 
- (modified) clang-tools-extra/clang-tidy/ClangTidy.h (+3-3) 
- (modified) clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp (+21-15) 


``diff
diff --git a/clang-tools-extra/clang-tidy/ClangTidy.cpp 
b/clang-tools-extra/clang-tidy/ClangTidy.cpp
index 9c8c93c5d16c72..959b11777e88d4 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.cpp
+++ b/clang-tools-extra/clang-tidy/ClangTidy.cpp
@@ -646,9 +646,9 @@ void exportReplacements(const llvm::StringRef MainFilePath,
   YAML << TUD;
 }
 
-NamesAndOptions
+ChecksAndOptions
 getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers) {
-  NamesAndOptions Result;
+  ChecksAndOptions Result;
   ClangTidyOptions Opts;
   Opts.Checks = "*";
   clang::tidy::ClangTidyContext Context(
@@ -661,7 +661,7 @@ getAllChecksAndOptions(bool 
AllowEnablingAnalyzerAlphaCheckers) {
   }
 
   for (const auto &Factory : Factories)
-Result.Names.insert(Factory.getKey());
+Result.Checks.insert(Factory.getKey());
 
 #if CLANG_TIDY_ENABLE_STATIC_ANALYZER
   SmallString<64> Buffer(AnalyzerCheckNamePrefix);
@@ -670,7 +670,7 @@ getAllChecksAndOptions(bool 
AllowEnablingAnalyzerAlphaCheckers) {
AllowEnablingAnalyzerAlphaCheckers)) {
 Buffer.truncate(DefSize);
 Buffer.append(AnalyzerCheck);
-Result.Names.insert(Buffer);
+Result.Checks.insert(Buffer);
   }
   for (std::string OptionName : {
 #define GET_CHECKER_OPTIONS
diff --git a/clang-tools-extra/clang-tidy/ClangTidy.h 
b/clang-tools-extra/clang-tidy/ClangTidy.h
index 51d9e226c79465..4ffd49f6ebf50f 100644
--- a/clang-tools-extra/clang-tidy/ClangTidy.h
+++ b/clang-tools-extra/clang-tidy/ClangTidy.h
@@ -58,12 +58,12 @@ class ClangTidyASTConsumerFactory {
 std::vector getCheckNames(const ClangTidyOptions &Options,
bool 
AllowEnablingAnalyzerAlphaCheckers);
 
-struct NamesAndOptions {
-  llvm::StringSet<> Names;
+struct ChecksAndOptions {
+  llvm::StringSet<> Checks;
   llvm::StringSet<> Options;
 };
 
-NamesAndOptions
+ChecksAndOptions
 getAllChecksAndOptions(bool AllowEnablingAnalyzerAlphaCheckers = true);
 
 /// Returns the effective check-specific options.
diff --git a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp 
b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
index d42dafa8ffc362..b8d843cba71330 100644
--- a/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
+++ b/clang-tools-extra/clang-tidy/tool/ClangTidyMain.cpp
@@ -526,6 +526,24 @@ static bool verifyFileExtensions(
   return AnyInvalid;
 }
 
+static bool verifyOptions(const llvm::StringSet<> &ValidOptions,
+  const ClangTidyOptions::OptionMap &OptionMap,
+  StringRef Source) {
+  bool AnyInvalid = false;
+  for (auto Key : OptionMap.keys()) {
+if (ValidOptions.contains(Key))
+  continue;
+AnyInvalid = true;
+auto &Output = llvm::WithColor::warning(llvm::errs(), Source)
+   << "unknown check option '" << Key << '\'';
+llvm::StringRef Closest = closest(Key, ValidOptions);
+if (!Closest.empty())
+  Output << "; did you mean '" << Closest << '\'';
+Output << VerifyConfigWarningEnd;
+  }
+  return AnyInvalid;
+}
+
 static SmallString<256> makeAbsolute(llvm::StringRef Input) {
   if (Input.empty())
 return {};
@@ -629,29 +647,17 @@ int clangTidyMain(int argc, const char **argv) {
   if (VerifyConfig) {
 std::vector RawOptions =
 OptionsProvider->getRawOptions(FileName);
-NamesAndOptions Valid =
+ChecksAndOptions Valid =
 getAllChecksAndOptions(AllowEnablingAnalyzerAlphaCheckers);
 bool AnyInvalid = false;
 for (const auto &[Opts, Source] : RawOptions) {
   if (Opts.Checks)
-AnyInvalid |= verifyChecks(Valid.Names, *Opts.Checks, Source);
-
+AnyInvalid |= verifyChecks(Valid.Checks, *Opts.Checks, Source);
   if (Opts.HeaderFileExtensions && Opts.ImplementationFileExtensions)
 AnyInvalid |=
 verifyFileExtensions(*Opts.HeaderFileExtensions,
  *Opts.ImplementationFileExtensions, Source);
-
-  for (auto Key : Opts.CheckOptions.keys()) {
-if (Valid.Options.contains(Key))
-  continue;
-AnyInvalid = true;
-auto &Output = llvm::WithColor::warning(llvm::errs(), Source)
-   << "unknown check option '" << Key << '\'';
-llvm::StringRef Closest = closest(Key, Valid.Options);
-if (!Closest.empty())
-  Output << "; did you mean '" << Closest << '\'';
-Output << VerifyConfigWarningEnd;
-  }
+  AnyInvalid |= verifyOptions(Valid.Options, Opts.CheckOptions, Source);
 }
 if (AnyInvalid)
   return 1;

``




https://github.com/llvm/llvm-pro

[clang] [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (PR #119908)

[Cyndy Ishida via cfe-commits]

https://github.com/cyndyishida approved this pull request.


https://github.com/llvm/llvm-project/pull/119908
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [analyzer] Don't assume third iteration in loops (PR #119388)

[Donát Nagy via cfe-commits]

NagyDonat wrote:

> > [...] To fix this problem, it would be sufficient to e.g. ensure that 
> > evalEagerlyAssumeBifurcation sets LastEagerlyAssumeBifurcationAt to nullptr 
> > [...]
> 
> Sounds good to me. Let's zero it out after it's "used"/"consumed".

I took a different approach (which was already mostly implemented when I've 
seen this suggestion): I'm setting the stored `Expr *` to nullpointer when I 
encounter an `EagerlyAssume` which does not succeed (i.e. at most one branch is 
feasible). I also added a dedicated test file which shows that this change 
works (and demonstrates other features and limitations of the commit).

> > [...] there might be better solutions to implement this "did EagerlyAssume 
> > split the state right now directly before this processBranch callback?" 
> > check that I need. (E.g. perhaps we could walk a few steps backwards on the 
> > ExplodedGraph -- but I don't know what kinds of nodes should I expect and 
> > I'm afraid that what I could write would run into pitfalls in unusual 
> > cases.)
> 
> Yes, it could get messy. For instance if we have multiple parent nodes due to 
> a merge. Let's not do traversals.

OK, if you say so, I'm very happy to avoid it :grinning: 

> I'm good with the change once the last remark is fixed. BTW have you measured 
> the running time implications of this patch? How much we spare?

I don't expect running _time_ implications, because as we discard these 
execution paths, the analyzer will traverse other execution paths instead of 
them -- so the performance gains should appear as "more results" instead of 
"faster analysis". Unfortunately the improvements that I hoped didn't appear in 
the experimental results: the "final" version of this commit doesn't produce 
significantly more results compared to the "intermediate" version which just 
suppresses the results (after "wasting time" on traversing the unjustified 
paths).

I think we don't see visible performance improvements because my observation 
that "difference between 2 or 3 iterations is irrelevant on the later parts of 
the execution path" does not imply that we can save a significant amount of 
calculations by driving each of those paths through 2 iterations (instead of 
roughly half 2 iterations, half 3 iterations) -- because those paths differ in 
many assumptions, so we cannot merge them even if we ignore the iteration count 
in that particular loop.

Additionally, my measurements are on stable open source projects, where real 
results are much rarer than false positives (even on stable checkers), so the 
number of results that we get is an imperfect and somewhat noisy proxy of the 
actual improvements.

https://github.com/llvm/llvm-project/pull/119388
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy][NFC] extract options verify to separately function (PR #120768)

[Congcong Cai via cfe-commits]

https://github.com/HerrCai0907 edited 
https://github.com/llvm/llvm-project/pull/120768
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (PR #119908)

[Qiongsi Wu via cfe-commits]

https://github.com/qiongsiwu updated 
https://github.com/llvm/llvm-project/pull/119908

>From 63c424414c1814ec9b4c3c5a459bfe1be684586d Mon Sep 17 00:00:00 2001
From: Qiongsi Wu 
Date: Fri, 13 Dec 2024 09:41:41 -0800
Subject: [PATCH 1/4] Fix parsing :: in method parameter type.

---
 clang/lib/Parse/Parser.cpp  | 8 +++-
 clang/test/Parser/objc-coloncolon.m | 5 +
 2 files changed, 12 insertions(+), 1 deletion(-)
 create mode 100644 clang/test/Parser/objc-coloncolon.m

diff --git a/clang/lib/Parse/Parser.cpp b/clang/lib/Parse/Parser.cpp
index 36e56a92c3092e..aa78d702553172 100644
--- a/clang/lib/Parse/Parser.cpp
+++ b/clang/lib/Parse/Parser.cpp
@@ -,8 +,14 @@ bool Parser::TryAnnotateTypeOrScopeTokenAfterScopeSpec(
 }
   }
 
-  if (SS.isEmpty())
+  if (SS.isEmpty()) {
+if (getLangOpts().ObjC && Tok.is(tok::coloncolon)) {
+  // ObjectiveC does not allow :: as as a scope token.
+  Diag(ConsumeToken(), diag::err_expected_type);
+  return true;
+}
 return false;
+  }
 
   // A C++ scope specifier that isn't followed by a typename.
   AnnotateScopeToken(SS, IsNewScope);
diff --git a/clang/test/Parser/objc-coloncolon.m 
b/clang/test/Parser/objc-coloncolon.m
new file mode 100644
index 00..e8a09898263bb3
--- /dev/null
+++ b/clang/test/Parser/objc-coloncolon.m
@@ -0,0 +1,5 @@
+// RUN: %clang_cc1 -x objective-c -fsyntax-only -verify %s
+
+@interface A
+- (instancetype)init:(::A *) foo; // expected-error {{expected a type}} 
+@end

>From ec903eb3fcd18ba53af901582060bd61b13cf324 Mon Sep 17 00:00:00 2001
From: Qiongsi Wu 
Date: Mon, 16 Dec 2024 09:34:52 -0800
Subject: [PATCH 2/4] Fix ObjectiveC++

---
 clang/lib/Parse/Parser.cpp | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/clang/lib/Parse/Parser.cpp b/clang/lib/Parse/Parser.cpp
index aa78d702553172..8ba6a5dce8a994 100644
--- a/clang/lib/Parse/Parser.cpp
+++ b/clang/lib/Parse/Parser.cpp
@@ -2223,7 +2223,8 @@ bool Parser::TryAnnotateTypeOrScopeTokenAfterScopeSpec(
   }
 
   if (SS.isEmpty()) {
-if (getLangOpts().ObjC && Tok.is(tok::coloncolon)) {
+if (getLangOpts().ObjC && !getLangOpts().CPlusPlus &&
+Tok.is(tok::coloncolon)) {
   // ObjectiveC does not allow :: as as a scope token.
   Diag(ConsumeToken(), diag::err_expected_type);
   return true;

>From 66c3b0b874f5a962ff902c43d33fbb22c658780c Mon Sep 17 00:00:00 2001
From: Qiongsi Wu 
Date: Wed, 18 Dec 2024 09:49:15 -0800
Subject: [PATCH 3/4] Address code review.

---
 clang/docs/ReleaseNotes.rst|  3 +++
 clang/test/Parser/objc-coloncolon.m| 12 ++--
 clang/test/Parser/objcxx-coloncolon.mm |  9 +
 3 files changed, 22 insertions(+), 2 deletions(-)
 create mode 100644 clang/test/Parser/objcxx-coloncolon.mm

diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 26fa986810a4b8..d6d3149df07d87 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -675,6 +675,9 @@ Improvements to Clang's diagnostics
   views.push_back(std::string("123")); // warning
 }
 
+- Fixed a bug where Clang hangs on unsupported optional scope specifier ``::`` 
when parsing
+  Objective-C. Clang now emits a diagnostic message instead of hanging.
+
 Improvements to Clang's time-trace
 --
 
diff --git a/clang/test/Parser/objc-coloncolon.m 
b/clang/test/Parser/objc-coloncolon.m
index e8a09898263bb3..68b54ef5229af5 100644
--- a/clang/test/Parser/objc-coloncolon.m
+++ b/clang/test/Parser/objc-coloncolon.m
@@ -1,5 +1,13 @@
-// RUN: %clang_cc1 -x objective-c -fsyntax-only -verify %s
+// RUN: %clang_cc1 -x objective-c -fsyntax-only -Wno-objc-root-class -verify %s
+
+int GV = 42;
 
 @interface A
-- (instancetype)init:(::A *) foo; // expected-error {{expected a type}} 
++ (int) getGV;
+- (instancetype)init:(::A *) foo; // expected-error {{expected a type}}
+@end
+
+@implementation A
++ (int) getGV { return ::GV; } // expected-error {{expected a type}}
+- (instancetype)init:(::A *) foo { return self; } // expected-error {{expected 
a type}}
 @end
diff --git a/clang/test/Parser/objcxx-coloncolon.mm 
b/clang/test/Parser/objcxx-coloncolon.mm
new file mode 100644
index 00..864a7df8400c1c
--- /dev/null
+++ b/clang/test/Parser/objcxx-coloncolon.mm
@@ -0,0 +1,9 @@
+// Test to make sure the parser does not get stuck on the optional
+// scope specifier on the type B.
+// RUN: %clang_cc1 -fsyntax-only %s
+
+class B;
+
+@interface A
+- (void) init:(::B *) foo;
+@end

>From 2e0667d1f21f406be8c9ae1f10eb12f60296ddd8 Mon Sep 17 00:00:00 2001
From: Qiongsi Wu 
Date: Wed, 18 Dec 2024 11:10:16 -0800
Subject: [PATCH 4/4] Modifying the test to include valid Objective-C with

---
 clang/test/Parser/objc-coloncolon.m | 6 ++
 1 file changed, 6 insertions(+)

diff --git a/clang/test/Parser/objc-coloncolon.m 
b/clang/test/Parser/objc-coloncolon.m
index 68b54ef5229af5..04a24fd81ec08f 100644
--- a/clang/test/Parser/objc-coloncolon.m
+++ b/clang/t

[clang-tools-extra] [clang-tidy][NFC][doc] mention some range algorithms do not work for `vector` in C++20 (PR #120774)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-tidy

Author: Congcong Cai (HerrCai0907)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/120774.diff


1 Files Affected:

- (modified) clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst 
(+3) 


``diff
diff --git a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst 
b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
index 1ce866ca1f66a0..912b42b33f9194 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
@@ -104,6 +104,9 @@ Calls to the following std library algorithms are checked:
 ``std::unique``,
 ``std::upper_bound``.
 
+Note: some range algorithms for ``vector`` require C++23 because it uses
+proxy iterators.
+
 Reverse Iteration
 -
 

``




https://github.com/llvm/llvm-project/pull/120774
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy][NFC][doc] mention some range algorithms do not work for `vector` in C++20 (PR #120774)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-tools-extra

Author: Congcong Cai (HerrCai0907)


Changes



---
Full diff: https://github.com/llvm/llvm-project/pull/120774.diff


1 Files Affected:

- (modified) clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst 
(+3) 


``diff
diff --git a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst 
b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
index 1ce866ca1f66a0..912b42b33f9194 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
@@ -104,6 +104,9 @@ Calls to the following std library algorithms are checked:
 ``std::unique``,
 ``std::upper_bound``.
 
+Note: some range algorithms for ``vector`` require C++23 because it uses
+proxy iterators.
+
 Reverse Iteration
 -
 

``




https://github.com/llvm/llvm-project/pull/120774
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy][NFC][doc] mention some range algorithms do not work for `vector` in C++20 (PR #120774)

[Congcong Cai via cfe-commits]

https://github.com/HerrCai0907 created 
https://github.com/llvm/llvm-project/pull/120774

None

>From b8f686aa8fa45ab50c45c16d6b4eeea99de64bcb Mon Sep 17 00:00:00 2001
From: Congcong Cai 
Date: Sat, 21 Dec 2024 01:01:14 +0800
Subject: [PATCH] [clang-tidy][NFC][doc] mention some range algorithms do not
 work for `vector` in C++20

---
 .../docs/clang-tidy/checks/modernize/use-ranges.rst| 3 +++
 1 file changed, 3 insertions(+)

diff --git a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst 
b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
index 1ce866ca1f66a0..912b42b33f9194 100644
--- a/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
+++ b/clang-tools-extra/docs/clang-tidy/checks/modernize/use-ranges.rst
@@ -104,6 +104,9 @@ Calls to the following std library algorithms are checked:
 ``std::unique``,
 ``std::upper_bound``.
 
+Note: some range algorithms for ``vector`` require C++23 because it uses
+proxy iterators.
+
 Reverse Iteration
 -
 

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang] Re-write codegen for atomic_test_and_set and atomic_clear (PR #120449)

[Mikhail Goncharov via cfe-commits]

metaflow wrote:

@ostannard :christmas_tree: :santa: happy holidays

https://github.com/llvm/llvm-project/pull/120449
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [webkit.UncountedLambdaCapturesChecker] Fix a nullptr deference. (PR #120702)

[Gábor Horváth via cfe-commits]

https://github.com/Xazax-hun approved this pull request.


https://github.com/llvm/llvm-project/pull/120702
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Patch series to reapply #118734 and substantially improve it (PR #120534)

[via cfe-commits]

h-vetinari wrote:

I've [tested](https://github.com/conda-forge/clangdev-feedstock/pull/333) this 
PR (as of 2bcc4e5f7043dab1ef673dd20b38009363db51db) in our infrastructure and 
can confirm that things run fine with VS2019 again. Thanks a lot for reworking 
this! 🙏 

https://github.com/llvm/llvm-project/pull/120534
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] cf7b3f8 - Fix double-quotes in diagnostic when attempting to access a ext_vector of bools (#118186)

[via cfe-commits]

Author: William Tran-Viet
Date: 2024-12-20T13:40:17+01:00
New Revision: cf7b3f8d827abba49930202e51702714349c716d

URL: 
https://github.com/llvm/llvm-project/commit/cf7b3f8d827abba49930202e51702714349c716d
DIFF: 
https://github.com/llvm/llvm-project/commit/cf7b3f8d827abba49930202e51702714349c716d.diff

LOG: Fix double-quotes in diagnostic when attempting to access a ext_vector of 
bools (#118186)

Fixes #116932 

- Remove the quotation marks in the diagnostic message for
err_ext_vector_component_name_illegal
- Pass in the quotation marks directly when reporting an illegal vector
component name inside `CheckExtVectorComponent`
- Add an offset to the `OpLoc` passed into `S.Diag` so the error message
arrow points directly to the offending illegal component rather than to
the '.' at the start of the component identifier.
- Modify the `vector-bool.cpp` element-wise access test case so it
(correctly) now only expects a single set of quotes.

Added: 


Modified: 
clang/include/clang/Basic/DiagnosticSemaKinds.td
clang/lib/Sema/SemaExprMember.cpp
clang/test/SemaCXX/vector-bool.cpp

Removed: 




diff  --git a/clang/include/clang/Basic/DiagnosticSemaKinds.td 
b/clang/include/clang/Basic/DiagnosticSemaKinds.td
index 8d19e9030ac2e3..491bc83c1e1297 100644
--- a/clang/include/clang/Basic/DiagnosticSemaKinds.td
+++ b/clang/include/clang/Basic/DiagnosticSemaKinds.td
@@ -3423,7 +3423,7 @@ def warn_typecheck_vector_element_sizes_not_equal : 
Warning<
 def err_ext_vector_component_exceeds_length : Error<
   "vector component access exceeds type %0">;
 def err_ext_vector_component_name_illegal : Error<
-  "illegal vector component name '%0'">;
+  "illegal vector component name %0">;
 def err_attribute_address_space_negative : Error<
   "address space is negative">;
 def err_attribute_address_space_too_high : Error<

diff  --git a/clang/lib/Sema/SemaExprMember.cpp 
b/clang/lib/Sema/SemaExprMember.cpp
index bcc1b92ffdec73..d130e8b86bc56d 100644
--- a/clang/lib/Sema/SemaExprMember.cpp
+++ b/clang/lib/Sema/SemaExprMember.cpp
@@ -434,8 +434,11 @@ CheckExtVectorComponent(Sema &S, QualType baseType, 
ExprValueKind &VK,
   if (!HalvingSwizzle && *compStr) {
 // We didn't get to the end of the string. This means the component names
 // didn't come from the same set *or* we encountered an illegal name.
-S.Diag(OpLoc, diag::err_ext_vector_component_name_illegal)
-  << StringRef(compStr, 1) << SourceRange(CompLoc);
+size_t Offset = compStr - CompName->getNameStart() + 1;
+char Fmt[3] = {'\'', *compStr, '\''};
+S.Diag(OpLoc.getLocWithOffset(Offset),
+   diag::err_ext_vector_component_name_illegal)
+<< StringRef(Fmt, 3) << SourceRange(CompLoc);
 return QualType();
   }
 

diff  --git a/clang/test/SemaCXX/vector-bool.cpp 
b/clang/test/SemaCXX/vector-bool.cpp
index e99d420e73fab2..cd638056f348b0 100644
--- a/clang/test/SemaCXX/vector-bool.cpp
+++ b/clang/test/SemaCXX/vector-bool.cpp
@@ -85,10 +85,10 @@ void foo(const bool& X);
 
 // Disallow element-wise access.
 bool* ElementRefs() {
-  eight_bools.y = false; // expected-error@88 {{illegal vector component name 
''y''}}
-  &eight_bools.z;// expected-error@89 {{illegal vector component name 
''z''}}
-  foo(eight_bools.w);// expected-error@90 {{illegal vector component name 
''w''}}
-  foo(eight_bools.wyx);  // expected-error@91 {{illegal vector component name 
''wyx''}}
+  eight_bools.y = false; // expected-error@88 {{illegal vector component name 
'y'}}
+  &eight_bools.z;// expected-error@89 {{illegal vector component name 
'z'}}
+  foo(eight_bools.w);// expected-error@90 {{illegal vector component name 
'w'}}
+  foo(eight_bools.wyx);  // expected-error@91 {{illegal vector component name 
'wyx'}}
 }
 
 void Sizeof() {



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Fix double-quotes in diagnostic when attempting to access a ext_vector of bools (PR #118186)

[Timm Baeder via cfe-commits]

https://github.com/tbaederr closed 
https://github.com/llvm/llvm-project/pull/118186
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [ConstantRange] Estimate tighter lower (upper) bounds for masked binary and (or) (PR #120352)

[Yingwei Zheng via cfe-commits]

@@ -1520,15 +1520,101 @@ ConstantRange ConstantRange::binaryNot() const {
   return ConstantRange(APInt::getAllOnes(getBitWidth())).sub(*this);
 }
 
+/// Estimate the 'bit-masked AND' operation's lower bound.
+///
+/// E.g., given two ranges as follows (single quotes are separators and
+/// have no meaning here),
+///
+///   LHS = [10'001'010,  ; LLo
+///  10'100'000]  ; LHi
+///   RHS = [10'111'010,  ; RLo
+///  10'111'100]  ; RHi
+///
+/// we know that the higher 2 bits of the result is always '10'; and note that
+/// there's at least one bit is 1 in LHS[3:6] (since the range is continuous),
+/// and all bits in RHS[3:6] are 1, so we know the lower bound of the result is
+/// 10'001'000.
+///
+/// The algorithm is as follows,
+/// 1. we first calculate a mask to mask out the higher common bits by
+///   Mask = (LLo ^ LHi) | (LLo ^ LHi) | (LLo ^ RLo);
+///   Mask = set all non-leading-zero bits to 1 for Mask;
+/// 2. find the bit field with at least 1 in LHS (i.e., bit 3:6 in the example)
+///after applying the mask, with
+///   StartBit = BitWidth - (LLo & Mask).clz() - 1;
+///   EndBit = BitWidth - (LHi & Mask).clz();
+/// 3. check if all bits in [StartBit:EndBit] in RHS are 1, and all bits of
+///RLo and RHi in [StartBit:BitWidth] are same, and if so, the lower bound
+///can be updated to
+///   LowerBound = LLo & Keep;
+///where Keep is a mask to mask out trailing bits (the lower 3 bits in the
+///example);
+/// 4. repeat the step 2 and 3 with LHS and RHS swapped, and update the lower
+///bound with the larger one.
+static APInt estimateBitMaskedAndLowerBound(const ConstantRange &LHS,
+const ConstantRange &RHS) {
+  auto BitWidth = LHS.getBitWidth();
+  // If either is full set or unsigned wrapped, then the range must contain '0'
+  // which leads the lower bound to 0.
+  if ((LHS.isFullSet() || RHS.isFullSet()) ||
+  (LHS.isWrappedSet() || RHS.isWrappedSet()))
+return APInt::getZero(BitWidth);
+
+  auto LLo = LHS.getLower();
+  auto LHi = LHS.getUpper() - 1;
+  auto RLo = RHS.getLower();
+  auto RHi = RHS.getUpper() - 1;
+
+  // Calculate the mask that mask out the higher common bits.
+  auto Mask = (LLo ^ LHi) | (RLo ^ RHi) | (LLo ^ RLo);
+  unsigned LeadingZeros = Mask.countLeadingZeros();
+  Mask.setLowBits(BitWidth - LeadingZeros);
+
+  auto estimateBound =
+  [BitWidth, &Mask](const APInt &ALo, const APInt &AHi, const APInt &BLo,
+const APInt &BHi) -> std::optional {
+unsigned LeadingZeros = (ALo & Mask).countLeadingZeros();
+if (LeadingZeros == BitWidth)
+  return std::nullopt;
+
+unsigned StartBit = BitWidth - LeadingZeros - 1;

dtcxzyw wrote:

Can we extend this algorithm to make the result optimal for all non-wrapped 
ranges?
An example: `[7, 14) & [-1, 0)` should produce `[7, 14)`. But currently it 
gives `[4, 14)`.

You can apply the following patch to get more sub-optimal cases :)
```
diff --git a/llvm/unittests/IR/ConstantRangeTest.cpp 
b/llvm/unittests/IR/ConstantRangeTest.cpp
index e1d9b3e387b2..1a616269e3b7 100644
--- a/llvm/unittests/IR/ConstantRangeTest.cpp
+++ b/llvm/unittests/IR/ConstantRangeTest.cpp
@@ -2725,7 +2725,7 @@ TEST_F(ConstantRangeTest, binaryAnd) {
 return CR1.binaryAnd(CR2);
   },
   [](const APInt &N1, const APInt &N2) { return N1 & N2; }, PreferSmallest,
-  CheckSingleElementsOnly);
+  CheckNonWrappedOnly);
 }
 
 TEST_F(ConstantRangeTest, binaryOr) {
```


https://github.com/llvm/llvm-project/pull/120352
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang-CL][DXC] Expose -fdiagnostic-color= to clang-cl and clang-dxc (PR #120644)

[Mészáros Gergely via cfe-commits]

https://github.com/Maetveis requested changes to this pull request.

Thank you, mostly looks good, I think the tests can be made simpler. And I 
think adding release notes would benefit users.
You can add a new point to "New Compiler Flags" in the clang release notes that 
clang-cl and clang-dxc now also accept`-fdiagnostics-color=[auto|never|always]` 
in addition to `-f[no-]color-diagnostics`.

https://github.com/llvm/llvm-project/pull/120644
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang-CL][DXC] Expose -fdiagnostic-color= to clang-cl and clang-dxc (PR #120644)

[Mészáros Gergely via cfe-commits]

https://github.com/Maetveis edited 
https://github.com/llvm/llvm-project/pull/120644
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang-CL][DXC] Expose -fdiagnostic-color= to clang-cl and clang-dxc (PR #120644)

[Mészáros Gergely via cfe-commits]

@@ -0,0 +1,8 @@
+// The test doesn't run in a PTY, so "auto" defaults to off.

Maetveis wrote:

Similarly here I think its enough to check that the option is accepted, no real 
need to verify what it does.

```c
// RUN: %clang_dxc \
// RUN: -fcolor-diagnostics \
// RUN: -fno-color-diagnostics \
// RUN: -fdiagnostics-color \
// RUN: -fno-diagnostics-color \
// RUN: -fdiagnostics-color=auto \
// RUN: -Tlib_6_7 -Vd -fdriver-only -- %s 2>&1 | count 0
```

(Might as well test all the other color flags while at it too).
(`-fdriver-only` and `count 0` should make sure we don't get any warnings not 
just errors).

Also NIT: Make sure to have a final newline in the file.

https://github.com/llvm/llvm-project/pull/120644
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang-CL][DXC] Expose -fdiagnostic-color= to clang-cl and clang-dxc (PR #120644)

[Mészáros Gergely via cfe-commits]

@@ -664,6 +664,15 @@
 // RUN: not %clang_cl /guard:foo -### -- %s 2>&1 | FileCheck 
-check-prefix=CFGUARDINVALID %s
 // CFGUARDINVALID: invalid value 'foo' in '/guard:'
 
+// The test doesn't run in a PTY, so "auto" defaults to off.
+// RUN: %clang_cl -fdiagnostics-color=auto -### -- %s 2>&1 | FileCheck 
-check-prefix=NO_COLOR %s
+
+// RUN: %clang_cl -fdiagnostics-color -### -- %s 2>&1 | FileCheck 
-check-prefix=COLOR %s
+// RUN: %clang_cl -fdiagnostics-color=always -### -- %s 2>&1 | FileCheck 
-check-prefix=COLOR %s
+// RUN: %clang_cl -fdiagnostics-color=never -### -- %s 2>&1 | FileCheck 
-check-prefix=NO_COLOR %s
+// COLOR: "-fcolor-diagnostics"

Maetveis wrote:

I think you should instead add `-fcolor-diagnostics=auto` to the list below, no 
need to check the functionality again in clang-cl, it is already sufficiently 
tested IMO by 
[clang/test/Driver/color-diagnostics.c](https://github.com/llvm/llvm-project/blob/2405c5fb3ed49b928bb2816ace7b67c8979cd9d7/clang/test/Driver/color-diagnostics.c)

```diff
 // RUN: -fno-coverage-mapping \
 // RUN: -fdiagnostics-color \
+// RUN: -fdiagnostics-color=auto \
 // RUN: -fno-diagnostics-color \
 // RUN: -fdebug-compilation-dir . \
 // RUN: -fdebug-compilation-dir=. \
 // RUN: -ffile-compilation-dir=. \
 ```

https://github.com/llvm/llvm-project/pull/120644
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang] Adjust pointer-overflow sanitizer for N3322 (PR #120719)

[Nikita Popov via cfe-commits]

https://github.com/nikic updated 
https://github.com/llvm/llvm-project/pull/120719

>From bd7a8273d07725d898188df00c708b5a52d68ac7 Mon Sep 17 00:00:00 2001
From: Nikita Popov 
Date: Fri, 20 Dec 2024 12:41:01 +0100
Subject: [PATCH 1/2] [Clang] Adjust pointer-overflow sanitizer for N3322

N3322 makes NULL + 0 well-defined in C, matching the C++ semantics.
Adjust the pointer-overflow sanitizer to no longer report NULL + 0
as a pointer overflow in any language mode. NULL + nonzero will of
course continue to be reported.

As N3322 is part of https://www.open-std.org/jtc1/sc22/wg14/www/previous.html,
and we never performed any optimizations based on NULL + 0 being
undefined in the first place, I'm applying this change to all C
versions.
---
 clang/lib/CodeGen/CGExprScalar.cpp| 16 +---
 ...d-nonzero-offset-when-nullptr-is-defined.c | 15 ++-
 .../catch-nullptr-and-nonzero-offset.c| 95 ---
 .../CodeGen/catch-pointer-overflow-volatile.c | 15 ++-
 clang/test/CodeGen/catch-pointer-overflow.c   | 36 +++
 clang/test/CodeGen/ubsan-pointer-overflow.c   |  7 +-
 6 files changed, 54 insertions(+), 130 deletions(-)

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp 
b/clang/lib/CodeGen/CGExprScalar.cpp
index 4b71bd730ce12c..0724b4265e96b2 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -5839,9 +5839,8 @@ CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type 
*ElemTy, Value *Ptr,
 
   auto *Zero = llvm::ConstantInt::getNullValue(IntPtrTy);
 
-  // Common case: if the total offset is zero, and we are using C++ semantics,
-  // where nullptr+0 is defined, don't emit a check.
-  if (EvaluatedGEP.TotalOffset == Zero && CGM.getLangOpts().CPlusPlus)
+  // Common case: if the total offset is zero, don't emit a check.
+  if (EvaluatedGEP.TotalOffset == Zero)
 return GEPVal;
 
   // Now that we've computed the total offset, add it to the base pointer (with
@@ -5852,23 +5851,16 @@ CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type 
*ElemTy, Value *Ptr,
   llvm::SmallVector, 2> Checks;
 
   if (PerformNullCheck) {
-// In C++, if the base pointer evaluates to a null pointer value,
+// If the base pointer evaluates to a null pointer value,
 // the only valid  pointer this inbounds GEP can produce is also
 // a null pointer, so the offset must also evaluate to zero.
 // Likewise, if we have non-zero base pointer, we can not get null pointer
 // as a result, so the offset can not be -intptr_t(BasePtr).
 // In other words, both pointers are either null, or both are non-null,
 // or the behaviour is undefined.
-//
-// C, however, is more strict in this regard, and gives more
-// optimization opportunities: in C, additionally, nullptr+0 is undefined.
-// So both the input to the 'gep inbounds' AND the output must not be null.
 auto *BaseIsNotNullptr = Builder.CreateIsNotNull(Ptr);
 auto *ResultIsNotNullptr = Builder.CreateIsNotNull(ComputedGEP);
-auto *Valid =
-CGM.getLangOpts().CPlusPlus
-? Builder.CreateICmpEQ(BaseIsNotNullptr, ResultIsNotNullptr)
-: Builder.CreateAnd(BaseIsNotNullptr, ResultIsNotNullptr);
+auto *Valid = Builder.CreateICmpEQ(BaseIsNotNullptr, ResultIsNotNullptr);
 Checks.emplace_back(Valid, SanitizerKind::PointerOverflow);
   }
 
diff --git 
a/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c 
b/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
index 8a560a47ad1e10..00198b4faf8bcc 100644
--- 
a/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
+++ 
b/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -x c -emit-llvm %s -o - -triple x86_64-linux-gnu | 
FileCheck %s
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fno-sanitize-recover=pointer-overflow -emit-llvm %s -o - -triple 
x86_64-linux-gnu | FileCheck %s -implicit-check-not="call void 
@__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-NORECOVER,CHECK-SANITIZE-UNREACHABLE
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fsanitize-recover=pointer-overflow -emit-llvm %s -o - -triple x86_64-linux-gnu 
| FileCheck %s -implicit-check-not="call void @__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-RECOVER
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fsanitize-trap=pointer-overflow -emit-llvm %s -o - -triple x86_64-linux-gnu | 
FileCheck %s -implicit-check-not="call void @__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-TRAP,CHECK-SANITIZE-UNREACHABLE
+// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fno-saniti

[clang] [lld] [llvm] Remove `urllib2` and `SocketServer` (PR #120723)

[Vinay Deshmukh via cfe-commits]

@@ -39,7 +36,6 @@ def url_is_up(url):
 
 
 def start_browser(port, options):
-import webbrowser

vinay-deshmukh wrote:

Due to this import, the import on L17 is unused.

let me know if keeping this one and removing the import from L17 is more 
preferable.

https://github.com/llvm/llvm-project/pull/120723
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [lld] [llvm] Remove `urllib2` and `SocketServer` (PR #120723)

[via cfe-commits]

github-actions[bot] wrote:

⚠️ We detected that you are using a GitHub private e-mail address to contribute 
to the repo. Please turn off [Keep my email addresses 
private](https://github.com/settings/emails) setting in your account. See 
[LLVM 
Discourse](https://discourse.llvm.org/t/hidden-emails-on-github-should-we-do-something-about-it)
 for more information.

https://github.com/llvm/llvm-project/pull/120723
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [Clang] Adjust pointer-overflow sanitizer for N3322 (PR #120719)

[Nikita Popov via cfe-commits]

https://github.com/nikic updated 
https://github.com/llvm/llvm-project/pull/120719

>From bd7a8273d07725d898188df00c708b5a52d68ac7 Mon Sep 17 00:00:00 2001
From: Nikita Popov 
Date: Fri, 20 Dec 2024 12:41:01 +0100
Subject: [PATCH 1/3] [Clang] Adjust pointer-overflow sanitizer for N3322

N3322 makes NULL + 0 well-defined in C, matching the C++ semantics.
Adjust the pointer-overflow sanitizer to no longer report NULL + 0
as a pointer overflow in any language mode. NULL + nonzero will of
course continue to be reported.

As N3322 is part of https://www.open-std.org/jtc1/sc22/wg14/www/previous.html,
and we never performed any optimizations based on NULL + 0 being
undefined in the first place, I'm applying this change to all C
versions.
---
 clang/lib/CodeGen/CGExprScalar.cpp| 16 +---
 ...d-nonzero-offset-when-nullptr-is-defined.c | 15 ++-
 .../catch-nullptr-and-nonzero-offset.c| 95 ---
 .../CodeGen/catch-pointer-overflow-volatile.c | 15 ++-
 clang/test/CodeGen/catch-pointer-overflow.c   | 36 +++
 clang/test/CodeGen/ubsan-pointer-overflow.c   |  7 +-
 6 files changed, 54 insertions(+), 130 deletions(-)

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp 
b/clang/lib/CodeGen/CGExprScalar.cpp
index 4b71bd730ce12c..0724b4265e96b2 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -5839,9 +5839,8 @@ CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type 
*ElemTy, Value *Ptr,
 
   auto *Zero = llvm::ConstantInt::getNullValue(IntPtrTy);
 
-  // Common case: if the total offset is zero, and we are using C++ semantics,
-  // where nullptr+0 is defined, don't emit a check.
-  if (EvaluatedGEP.TotalOffset == Zero && CGM.getLangOpts().CPlusPlus)
+  // Common case: if the total offset is zero, don't emit a check.
+  if (EvaluatedGEP.TotalOffset == Zero)
 return GEPVal;
 
   // Now that we've computed the total offset, add it to the base pointer (with
@@ -5852,23 +5851,16 @@ CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type 
*ElemTy, Value *Ptr,
   llvm::SmallVector, 2> Checks;
 
   if (PerformNullCheck) {
-// In C++, if the base pointer evaluates to a null pointer value,
+// If the base pointer evaluates to a null pointer value,
 // the only valid  pointer this inbounds GEP can produce is also
 // a null pointer, so the offset must also evaluate to zero.
 // Likewise, if we have non-zero base pointer, we can not get null pointer
 // as a result, so the offset can not be -intptr_t(BasePtr).
 // In other words, both pointers are either null, or both are non-null,
 // or the behaviour is undefined.
-//
-// C, however, is more strict in this regard, and gives more
-// optimization opportunities: in C, additionally, nullptr+0 is undefined.
-// So both the input to the 'gep inbounds' AND the output must not be null.
 auto *BaseIsNotNullptr = Builder.CreateIsNotNull(Ptr);
 auto *ResultIsNotNullptr = Builder.CreateIsNotNull(ComputedGEP);
-auto *Valid =
-CGM.getLangOpts().CPlusPlus
-? Builder.CreateICmpEQ(BaseIsNotNullptr, ResultIsNotNullptr)
-: Builder.CreateAnd(BaseIsNotNullptr, ResultIsNotNullptr);
+auto *Valid = Builder.CreateICmpEQ(BaseIsNotNullptr, ResultIsNotNullptr);
 Checks.emplace_back(Valid, SanitizerKind::PointerOverflow);
   }
 
diff --git 
a/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c 
b/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
index 8a560a47ad1e10..00198b4faf8bcc 100644
--- 
a/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
+++ 
b/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -x c -emit-llvm %s -o - -triple x86_64-linux-gnu | 
FileCheck %s
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fno-sanitize-recover=pointer-overflow -emit-llvm %s -o - -triple 
x86_64-linux-gnu | FileCheck %s -implicit-check-not="call void 
@__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-NORECOVER,CHECK-SANITIZE-UNREACHABLE
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fsanitize-recover=pointer-overflow -emit-llvm %s -o - -triple x86_64-linux-gnu 
| FileCheck %s -implicit-check-not="call void @__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-RECOVER
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fsanitize-trap=pointer-overflow -emit-llvm %s -o - -triple x86_64-linux-gnu | 
FileCheck %s -implicit-check-not="call void @__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-TRAP,CHECK-SANITIZE-UNREACHABLE
+// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fno-saniti

[clang] [lld] [llvm] Remove `urllib2` and `SocketServer` (PR #120723)

[Vinay Deshmukh via cfe-commits]

https://github.com/vinay-deshmukh created 
https://github.com/llvm/llvm-project/pull/120723

Resolves #120299 

1. Testing in CI

>From 5f7a46d117053b582b8ee4c91a46e5edd84fbad2 Mon Sep 17 00:00:00 2001
From: Vinay Deshmukh <32487576+vinay-deshm...@users.noreply.github.com>
Date: Fri, 20 Dec 2024 16:56:26 +0530
Subject: [PATCH 1/3] Remove `urllib2`

---
 clang/docs/tools/dump_ast_matchers.py | 5 +
 clang/tools/scan-view/bin/scan-view   | 5 +
 lld/utils/benchmark.py| 8 ++--
 llvm/utils/unicode-case-fold.py   | 5 +
 4 files changed, 5 insertions(+), 18 deletions(-)

diff --git a/clang/docs/tools/dump_ast_matchers.py 
b/clang/docs/tools/dump_ast_matchers.py
index b6f00657ec914c..4c4a2bd1e9d5a7 100755
--- a/clang/docs/tools/dump_ast_matchers.py
+++ b/clang/docs/tools/dump_ast_matchers.py
@@ -7,10 +7,7 @@
 import re
 import os
 
-try:
-from urllib.request import urlopen
-except ImportError:
-from urllib2 import urlopen
+from urllib.request import urlopen
 
 CLASS_INDEX_PAGE_URL = "https://clang.llvm.org/doxygen/classes.html";
 try:
diff --git a/clang/tools/scan-view/bin/scan-view 
b/clang/tools/scan-view/bin/scan-view
index d01aebb4029a75..f8dc6c18018de1 100755
--- a/clang/tools/scan-view/bin/scan-view
+++ b/clang/tools/scan-view/bin/scan-view
@@ -10,10 +10,7 @@ import os
 import posixpath
 import threading
 import time
-try:
-from urllib.request import urlopen
-except ImportError:
-from urllib2 import urlopen
+from urllib.request import urlopen
 import webbrowser
 
 # How long to wait for server to start.
diff --git a/lld/utils/benchmark.py b/lld/utils/benchmark.py
index 7202e07ec438d2..4257f6bb160565 100755
--- a/lld/utils/benchmark.py
+++ b/lld/utils/benchmark.py
@@ -14,12 +14,8 @@
 import datetime
 import argparse
 
-try:
-from urllib.parse import urlencode
-from urllib.request import urlopen, Request
-except ImportError:
-from urllib import urlencode
-from urllib2 import urlopen, Request
+from urllib.parse import urlencode
+from urllib.request import urlopen, Request
 
 
 parser = argparse.ArgumentParser()
diff --git a/llvm/utils/unicode-case-fold.py b/llvm/utils/unicode-case-fold.py
index 9639aa0dc44b41..3ef12a0ef881a8 100755
--- a/llvm/utils/unicode-case-fold.py
+++ b/llvm/utils/unicode-case-fold.py
@@ -22,10 +22,7 @@
 import sys
 import re
 
-try:
-from urllib.request import urlopen
-except ImportError:
-from urllib2 import urlopen
+from urllib.request import urlopen
 
 
 # This variable will body of the mappings function

>From d73faddf37c8d880f9f44a02a4df4290aac0f916 Mon Sep 17 00:00:00 2001
From: Vinay Deshmukh <32487576+vinay-deshm...@users.noreply.github.com>
Date: Fri, 20 Dec 2024 17:00:30 +0530
Subject: [PATCH 2/3] Cleanup import

---
 clang/tools/scan-view/bin/scan-view | 1 -
 1 file changed, 1 deletion(-)

diff --git a/clang/tools/scan-view/bin/scan-view 
b/clang/tools/scan-view/bin/scan-view
index f8dc6c18018de1..d6b8c446ccfe8a 100755
--- a/clang/tools/scan-view/bin/scan-view
+++ b/clang/tools/scan-view/bin/scan-view
@@ -36,7 +36,6 @@ def url_is_up(url):
 
 
 def start_browser(port, options):
-import webbrowser
 
 url = 'http://%s:%d' % (options.host, port)
 

>From d377097e7df5a4933112e69cd49d1fe8f8980d05 Mon Sep 17 00:00:00 2001
From: Vinay Deshmukh <32487576+vinay-deshm...@users.noreply.github.com>
Date: Fri, 20 Dec 2024 17:01:30 +0530
Subject: [PATCH 3/3] Update socketserver to python3

---
 clang/tools/scan-view/bin/scan-view | 5 +
 1 file changed, 1 insertion(+), 4 deletions(-)

diff --git a/clang/tools/scan-view/bin/scan-view 
b/clang/tools/scan-view/bin/scan-view
index d6b8c446ccfe8a..b63f494adfbe76 100755
--- a/clang/tools/scan-view/bin/scan-view
+++ b/clang/tools/scan-view/bin/scan-view
@@ -79,10 +79,7 @@ def run(port, options, root):
 
 
 def port_is_open(port):
-try:
-import socketserver
-except ImportError:
-import SocketServer as socketserver
+import socketserver
 try:
 t = socketserver.TCPServer((kDefaultHost, port), None)
 except:

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [AMDGPU][True16][MC] true16 for v_alignbyte_b32 (PR #119750)

[Matt Arsenault via cfe-commits]

@@ -2353,8 +2353,8 @@ def int_amdgcn_writelane :
   [IntrNoMem, IntrConvergent, IntrWillReturn, IntrNoCallback, IntrNoFree]
 >;
 
-def int_amdgcn_alignbyte : ClangBuiltin<"__builtin_amdgcn_alignbyte">,
-  DefaultAttrsIntrinsic<[llvm_i32_ty], [llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+def int_amdgcn_alignbyte : DefaultAttrsIntrinsic<[llvm_i32_ty],
+  [llvm_i32_ty, llvm_i32_ty, llvm_anyint_ty],

arsenm wrote:

This should be done separately 

https://github.com/llvm/llvm-project/pull/119750
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [AMDGPU][True16][MC] true16 for v_alignbyte_b32 (PR #119750)

[Matt Arsenault via cfe-commits]

@@ -2353,8 +2353,8 @@ def int_amdgcn_writelane :
   [IntrNoMem, IntrConvergent, IntrWillReturn, IntrNoCallback, IntrNoFree]
 >;
 
-def int_amdgcn_alignbyte : ClangBuiltin<"__builtin_amdgcn_alignbyte">,
-  DefaultAttrsIntrinsic<[llvm_i32_ty], [llvm_i32_ty, llvm_i32_ty, llvm_i32_ty],
+def int_amdgcn_alignbyte : DefaultAttrsIntrinsic<[llvm_i32_ty],
+  [llvm_i32_ty, llvm_i32_ty, llvm_anyint_ty],

arsenm wrote:

Does this even really need to be done? Only the low bits are read anyway, so we 
could just truncate 

https://github.com/llvm/llvm-project/pull/119750
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [compiler-rt] [Clang] Adjust pointer-overflow sanitizer for N3322 (PR #120719)

[Nikita Popov via cfe-commits]

https://github.com/nikic updated 
https://github.com/llvm/llvm-project/pull/120719

>From bd7a8273d07725d898188df00c708b5a52d68ac7 Mon Sep 17 00:00:00 2001
From: Nikita Popov 
Date: Fri, 20 Dec 2024 12:41:01 +0100
Subject: [PATCH 1/4] [Clang] Adjust pointer-overflow sanitizer for N3322

N3322 makes NULL + 0 well-defined in C, matching the C++ semantics.
Adjust the pointer-overflow sanitizer to no longer report NULL + 0
as a pointer overflow in any language mode. NULL + nonzero will of
course continue to be reported.

As N3322 is part of https://www.open-std.org/jtc1/sc22/wg14/www/previous.html,
and we never performed any optimizations based on NULL + 0 being
undefined in the first place, I'm applying this change to all C
versions.
---
 clang/lib/CodeGen/CGExprScalar.cpp| 16 +---
 ...d-nonzero-offset-when-nullptr-is-defined.c | 15 ++-
 .../catch-nullptr-and-nonzero-offset.c| 95 ---
 .../CodeGen/catch-pointer-overflow-volatile.c | 15 ++-
 clang/test/CodeGen/catch-pointer-overflow.c   | 36 +++
 clang/test/CodeGen/ubsan-pointer-overflow.c   |  7 +-
 6 files changed, 54 insertions(+), 130 deletions(-)

diff --git a/clang/lib/CodeGen/CGExprScalar.cpp 
b/clang/lib/CodeGen/CGExprScalar.cpp
index 4b71bd730ce12c..0724b4265e96b2 100644
--- a/clang/lib/CodeGen/CGExprScalar.cpp
+++ b/clang/lib/CodeGen/CGExprScalar.cpp
@@ -5839,9 +5839,8 @@ CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type 
*ElemTy, Value *Ptr,
 
   auto *Zero = llvm::ConstantInt::getNullValue(IntPtrTy);
 
-  // Common case: if the total offset is zero, and we are using C++ semantics,
-  // where nullptr+0 is defined, don't emit a check.
-  if (EvaluatedGEP.TotalOffset == Zero && CGM.getLangOpts().CPlusPlus)
+  // Common case: if the total offset is zero, don't emit a check.
+  if (EvaluatedGEP.TotalOffset == Zero)
 return GEPVal;
 
   // Now that we've computed the total offset, add it to the base pointer (with
@@ -5852,23 +5851,16 @@ CodeGenFunction::EmitCheckedInBoundsGEP(llvm::Type 
*ElemTy, Value *Ptr,
   llvm::SmallVector, 2> Checks;
 
   if (PerformNullCheck) {
-// In C++, if the base pointer evaluates to a null pointer value,
+// If the base pointer evaluates to a null pointer value,
 // the only valid  pointer this inbounds GEP can produce is also
 // a null pointer, so the offset must also evaluate to zero.
 // Likewise, if we have non-zero base pointer, we can not get null pointer
 // as a result, so the offset can not be -intptr_t(BasePtr).
 // In other words, both pointers are either null, or both are non-null,
 // or the behaviour is undefined.
-//
-// C, however, is more strict in this regard, and gives more
-// optimization opportunities: in C, additionally, nullptr+0 is undefined.
-// So both the input to the 'gep inbounds' AND the output must not be null.
 auto *BaseIsNotNullptr = Builder.CreateIsNotNull(Ptr);
 auto *ResultIsNotNullptr = Builder.CreateIsNotNull(ComputedGEP);
-auto *Valid =
-CGM.getLangOpts().CPlusPlus
-? Builder.CreateICmpEQ(BaseIsNotNullptr, ResultIsNotNullptr)
-: Builder.CreateAnd(BaseIsNotNullptr, ResultIsNotNullptr);
+auto *Valid = Builder.CreateICmpEQ(BaseIsNotNullptr, ResultIsNotNullptr);
 Checks.emplace_back(Valid, SanitizerKind::PointerOverflow);
   }
 
diff --git 
a/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c 
b/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
index 8a560a47ad1e10..00198b4faf8bcc 100644
--- 
a/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
+++ 
b/clang/test/CodeGen/catch-nullptr-and-nonzero-offset-when-nullptr-is-defined.c
@@ -1,7 +1,7 @@
 // RUN: %clang_cc1 -x c -emit-llvm %s -o - -triple x86_64-linux-gnu | 
FileCheck %s
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fno-sanitize-recover=pointer-overflow -emit-llvm %s -o - -triple 
x86_64-linux-gnu | FileCheck %s -implicit-check-not="call void 
@__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-NORECOVER,CHECK-SANITIZE-UNREACHABLE
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fsanitize-recover=pointer-overflow -emit-llvm %s -o - -triple x86_64-linux-gnu 
| FileCheck %s -implicit-check-not="call void @__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-ANYRECOVER,CHECK-SANITIZE-RECOVER
-// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fsanitize-trap=pointer-overflow -emit-llvm %s -o - -triple x86_64-linux-gnu | 
FileCheck %s -implicit-check-not="call void @__ubsan_handle_pointer_overflow" 
--check-prefixes=CHECK,CHECK-SANITIZE,CHECK-SANITIZE-NULLNOTOK,CHECK-SANITIZE-NULLNOTOK-C,CHECK-SANITIZE-TRAP,CHECK-SANITIZE-UNREACHABLE
+// RUN: %clang_cc1 -x c -fsanitize=pointer-overflow 
-fno-saniti

[clang] [compiler-rt] [Clang] Adjust pointer-overflow sanitizer for N3322 (PR #120719)

[Nikita Popov via cfe-commits]

nikic wrote:

@vitalybuka Should I also be dropping checks like 
https://github.com/llvm/llvm-project/blob/5f0db7c11264fa235d73730b2b93a31407dfbef3/compiler-rt/lib/ubsan/ubsan_handlers.cpp#L817-L818
 from the ubsan runtime? (Not sure whether the runtime is supposed to also be 
compatible with older clang versions or something like that.)

https://github.com/llvm/llvm-project/pull/120719
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (PR #120753)

[Jonathan Thackray via cfe-commits]

https://github.com/jthackray closed 
https://github.com/llvm/llvm-project/pull/120753
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] 952c8d3 - [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (#120753)

[via cfe-commits]

Author: Jonathan Thackray
Date: 2024-12-20T17:27:08Z
New Revision: 952c8d305459d3f1218d1ce12c7f20b6a10ba046

URL: 
https://github.com/llvm/llvm-project/commit/952c8d305459d3f1218d1ce12c7f20b6a10ba046
DIFF: 
https://github.com/llvm/llvm-project/commit/952c8d305459d3f1218d1ce12c7f20b6a10ba046.diff

LOG: [AArch64] Enable FEAT_SVE2p1 by default for Armv9.4-A and later (#120753)

The ArmARM says:
```
"In an Armv9.4 implementation, if FEAT_SVE2 is implemented,
 FEAT_SVE2p1 is implemented."
```
Since FEAT_SVE2 is already enabled for Armv9.0-A and later, then
FEAT_SVE2p1 should also be enabled by default.

Added: 


Modified: 
clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
llvm/lib/Target/AArch64/AArch64Features.td

Removed: 




diff  --git a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c 
b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
index 0032c926c22d96..1cfda6c996b9e3 100644
--- a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
+++ b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.4-a.c
@@ -58,6 +58,7 @@
 // CHECK-NEXT: FEAT_SSBS, FEAT_SSBS2  
Enable Speculative Store Bypass Safe bit
 // CHECK-NEXT: FEAT_SVE   
Enable Scalable Vector Extension (SVE) instructions
 // CHECK-NEXT: FEAT_SVE2  
Enable Scalable Vector Extension 2 (SVE2) instructions
+// CHECK-NEXT: FEAT_SVE2p1
Enable Scalable Vector Extension 2.1 instructions
 // CHECK-NEXT: FEAT_TLBIOS, FEAT_TLBIRANGE
Enable Armv8.4-A TLB Range and Maintenance instructions
 // CHECK-NEXT: FEAT_TRBE  
Enable Trace Buffer Extension
 // CHECK-NEXT: FEAT_TRF   
Enable Armv8.4-A Trace extension

diff  --git a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c 
b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
index be24bd0bbddb68..76c8b34a56b75b 100644
--- a/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
+++ b/clang/test/Driver/print-enabled-extensions/aarch64-armv9.5-a.c
@@ -61,6 +61,7 @@
 // CHECK-NEXT: FEAT_SSBS, FEAT_SSBS2  
Enable Speculative Store Bypass Safe bit
 // CHECK-NEXT: FEAT_SVE   
Enable Scalable Vector Extension (SVE) instructions
 // CHECK-NEXT: FEAT_SVE2  
Enable Scalable Vector Extension 2 (SVE2) instructions
+// CHECK-NEXT: FEAT_SVE2p1
Enable Scalable Vector Extension 2.1 instructions
 // CHECK-NEXT: FEAT_TLBIOS, FEAT_TLBIRANGE
Enable Armv8.4-A TLB Range and Maintenance instructions
 // CHECK-NEXT: FEAT_TRBE  
Enable Trace Buffer Extension
 // CHECK-NEXT: FEAT_TRF   
Enable Armv8.4-A Trace extension

diff  --git a/llvm/lib/Target/AArch64/AArch64Features.td 
b/llvm/lib/Target/AArch64/AArch64Features.td
index 3a6bef81f4a0d8..41eb9a73bd013d 100644
--- a/llvm/lib/Target/AArch64/AArch64Features.td
+++ b/llvm/lib/Target/AArch64/AArch64Features.td
@@ -874,7 +874,7 @@ def HasV9_3aOps : Architecture64<9, 3, "a", "v9.3a",
 def HasV9_4aOps : Architecture64<9, 4, "a", "v9.4a",
   [HasV8_9aOps, HasV9_3aOps],
   !listconcat(HasV9_3aOps.DefaultExts, [FeatureSPECRES2, FeatureCSSC,
-FeatureRASv2])>;
+FeatureRASv2, FeatureSVE2p1])>;
 def HasV9_5aOps : Architecture64<9, 5, "a", "v9.5a",
   [HasV9_4aOps, FeatureCPA],
   !listconcat(HasV9_4aOps.DefaultExts, [FeatureCPA,  FeatureLUT, 
FeatureFAMINMAX])>;



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Use smart pointer caching in unchecked optional accessor (PR #120249)

[Jan Voung via cfe-commits]

https://github.com/jvoung updated 
https://github.com/llvm/llvm-project/pull/120249

>From c526263a7accc434dbf6e93c2995ceb2f95873b8 Mon Sep 17 00:00:00 2001
From: Jan Voung 
Date: Tue, 17 Dec 2024 15:38:19 +
Subject: [PATCH] [clang][dataflow] Use smart pointer caching in unchecked
 optional accessor

Part 2 (and final part) following 
https://github.com/llvm/llvm-project/pull/120102
Allows users to do things like:

```
if (o->x.has_value()) {
  ((*o).x).value();
}
```
where the `->` and `*` are operator overload calls.

A user could instead extract the nested optional into a local variable
once instead of doing two accessor calls back to back, but currently
they are unsure why the code is flagged.
---
 .../CachedConstAccessorsLattice.h |  41 
 .../SmartPointerAccessorCaching.h | 168 +++
 .../lib/Analysis/FlowSensitive/CMakeLists.txt |   1 +
 .../Models/UncheckedOptionalAccessModel.cpp   |  58 +-
 .../SmartPointerAccessorCaching.cpp   | 153 ++
 .../Analysis/FlowSensitive/CMakeLists.txt |   1 +
 .../CachedConstAccessorsLatticeTest.cpp   |  32 +++
 .../SmartPointerAccessorCachingTest.cpp   | 194 ++
 .../UncheckedOptionalAccessModelTest.cpp  |  50 +
 9 files changed, 692 insertions(+), 6 deletions(-)
 create mode 100644 
clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h
 create mode 100644 
clang/lib/Analysis/FlowSensitive/SmartPointerAccessorCaching.cpp
 create mode 100644 
clang/unittests/Analysis/FlowSensitive/SmartPointerAccessorCachingTest.cpp

diff --git 
a/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h 
b/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h
index 48c5287367739a..6b5dacf9f66d2d 100644
--- a/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h
+++ b/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h
@@ -13,7 +13,9 @@
 #ifndef LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_CACHED_CONST_ACCESSORS_LATTICE_H
 #define LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_CACHED_CONST_ACCESSORS_LATTICE_H
 
+#include "clang/AST/Decl.h"
 #include "clang/AST/Expr.h"
+#include "clang/AST/Type.h"
 #include "clang/Analysis/FlowSensitive/DataflowEnvironment.h"
 #include "clang/Analysis/FlowSensitive/DataflowLattice.h"
 #include "clang/Analysis/FlowSensitive/StorageLocation.h"
@@ -71,10 +73,28 @@ template  class CachedConstAccessorsLattice 
: public Base {
   /// Requirements:
   ///
   ///  - `CE` should return a location (GLValue or a record type).
+  ///
+  /// DEPRECATED: switch users to the below overload which takes Callee and 
Type
+  /// directly.
   StorageLocation *getOrCreateConstMethodReturnStorageLocation(
   const RecordStorageLocation &RecordLoc, const CallExpr *CE,
   Environment &Env, llvm::function_ref 
Initialize);
 
+  /// Creates or returns a previously created `StorageLocation` associated with
+  /// a const method call `obj.getFoo()` where `RecordLoc` is the
+  /// `RecordStorageLocation` of `obj`, `Callee` is the decl for `getFoo`,
+  /// and `Type` is the return type of `getFoo`.
+  ///
+  /// The callback `Initialize` runs on the storage location if newly created.
+  ///
+  /// Requirements:
+  ///
+  ///  - `Type` should return a location (GLValue or a record type).
+  StorageLocation &getOrCreateConstMethodReturnStorageLocation(
+  const RecordStorageLocation &RecordLoc, const FunctionDecl *Callee,
+  QualType Type, Environment &Env,
+  llvm::function_ref Initialize);
+
   void clearConstMethodReturnValues(const RecordStorageLocation &RecordLoc) {
 ConstMethodReturnValues.erase(&RecordLoc);
   }
@@ -212,6 +232,27 @@ 
CachedConstAccessorsLattice::getOrCreateConstMethodReturnStorageLocation(
   return &Loc;
 }
 
+template 
+StorageLocation &
+CachedConstAccessorsLattice::getOrCreateConstMethodReturnStorageLocation(
+const RecordStorageLocation &RecordLoc, const FunctionDecl *Callee,
+QualType Type, Environment &Env,
+llvm::function_ref Initialize) {
+  assert(Callee != nullptr);
+  assert(!Type.isNull());
+  assert(Type->isReferenceType() || Type->isRecordType());
+  auto &ObjMap = ConstMethodReturnStorageLocations[&RecordLoc];
+  auto it = ObjMap.find(Callee);
+  if (it != ObjMap.end())
+return *it->second;
+
+  StorageLocation &Loc = Env.createStorageLocation(Type.getNonReferenceType());
+  Initialize(Loc);
+
+  ObjMap.insert({Callee, &Loc});
+  return Loc;
+}
+
 } // namespace dataflow
 } // namespace clang
 
diff --git 
a/clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h 
b/clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h
new file mode 100644
index 00..e89e82c893d8cb
--- /dev/null
+++ b/clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h
@@ -0,0 +1,168 @@
+//===-- SmartPointerAccessorCaching.h ---*- C++ 
-*-===//
+//
+// Part of the LLVM Project, under

[clang] [llvm] Remove -bounds-checking-unique-traps (replace with -fno-sanitize-merge=local-bounds) (PR #120682)

[Vitaly Buka via cfe-commits]

@@ -1208,6 +1202,11 @@ Sanitizers
 
 - Implemented ``-f[no-]sanitize-trap=local-bounds``, and 
``-f[no-]sanitize-recover=local-bounds``.
 
+- ``-fsanitize-merge`` (default) and ``-fno-sanitize-merge`` have been added 
for
+  fine-grained, unified control of which UBSan checks can potentially be merged
+  by the compiler (for example,

vitalybuka wrote:

Please mention on  
https://clang.llvm.org/docs/UndefinedBehaviorSanitizer.html#stack-traces-and-report-symbolization
 as well

https://github.com/llvm/llvm-project/pull/120682
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] 1418018 - [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (#119908)

[via cfe-commits]

Author: Qiongsi Wu
Date: 2024-12-20T09:47:26-08:00
New Revision: 14180185026b8ed793cc7dfd037620e5150ccc35

URL: 
https://github.com/llvm/llvm-project/commit/14180185026b8ed793cc7dfd037620e5150ccc35
DIFF: 
https://github.com/llvm/llvm-project/commit/14180185026b8ed793cc7dfd037620e5150ccc35.diff

LOG: [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (#119908)

The parser hangs when processing types/variables prefixed by `::` as an
optional scope specifier. For example,
```
- (instancetype)init:(::A *) foo;
```

The parser should not hang, and it should emit an error. This PR
implements the error check.

rdar://140885078

Added: 
clang/test/Parser/objc-coloncolon.m
clang/test/Parser/objcxx-coloncolon.mm

Modified: 
clang/docs/ReleaseNotes.rst
clang/lib/Parse/Parser.cpp

Removed: 




diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index a85ef60b7b58ba..38c35aea3cf220 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -708,6 +708,9 @@ Improvements to Clang's diagnostics
 
 - Fix -Wdangling false positives on conditional operators (#120206).
 
+- Fixed a bug where Clang hung on an unsupported optional scope specifier 
``::`` when parsing
+  Objective-C. Clang now emits a diagnostic message instead of hanging.
+
 Improvements to Clang's time-trace
 --
 

diff  --git a/clang/lib/Parse/Parser.cpp b/clang/lib/Parse/Parser.cpp
index 36e56a92c3092e..8ba6a5dce8a994 100644
--- a/clang/lib/Parse/Parser.cpp
+++ b/clang/lib/Parse/Parser.cpp
@@ -,8 +,15 @@ bool Parser::TryAnnotateTypeOrScopeTokenAfterScopeSpec(
 }
   }
 
-  if (SS.isEmpty())
+  if (SS.isEmpty()) {
+if (getLangOpts().ObjC && !getLangOpts().CPlusPlus &&
+Tok.is(tok::coloncolon)) {
+  // ObjectiveC does not allow :: as as a scope token.
+  Diag(ConsumeToken(), diag::err_expected_type);
+  return true;
+}
 return false;
+  }
 
   // A C++ scope specifier that isn't followed by a typename.
   AnnotateScopeToken(SS, IsNewScope);

diff  --git a/clang/test/Parser/objc-coloncolon.m 
b/clang/test/Parser/objc-coloncolon.m
new file mode 100644
index 00..04a24fd81ec08f
--- /dev/null
+++ b/clang/test/Parser/objc-coloncolon.m
@@ -0,0 +1,19 @@
+// RUN: %clang_cc1 -x objective-c -fsyntax-only -Wno-objc-root-class -verify %s
+
+int GV = 42;
+
+@interface A
++ (int) getGV;
+- (instancetype)init:(::A *) foo; // expected-error {{expected a type}}
+@end
+
+@implementation A
+- (void)performSelector:(SEL)selector {}
+- (void)double:(int)firstArg :(int)secondArg colon:(int)thirdArg {}
+- (void)test {
+  // The `::` below should not trigger an error.
+  [self performSelector:@selector(double::colon:)];
+}
++ (int) getGV { return ::GV; } // expected-error {{expected a type}}
+- (instancetype)init:(::A *) foo { return self; } // expected-error {{expected 
a type}}
+@end

diff  --git a/clang/test/Parser/objcxx-coloncolon.mm 
b/clang/test/Parser/objcxx-coloncolon.mm
new file mode 100644
index 00..864a7df8400c1c
--- /dev/null
+++ b/clang/test/Parser/objcxx-coloncolon.mm
@@ -0,0 +1,9 @@
+// Test to make sure the parser does not get stuck on the optional
+// scope specifier on the type B.
+// RUN: %clang_cc1 -fsyntax-only %s
+
+class B;
+
+@interface A
+- (void) init:(::B *) foo;
+@end



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (PR #119908)

[Qiongsi Wu via cfe-commits]

https://github.com/qiongsiwu closed 
https://github.com/llvm/llvm-project/pull/119908
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] Remove -bounds-checking-unique-traps (replace with -fno-sanitize-merge=local-bounds) (PR #120682)

[Vitaly Buka via cfe-commits]

https://github.com/vitalybuka approved this pull request.


https://github.com/llvm/llvm-project/pull/120682
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][ObjectiveC] Fix Parsing the `::` Optional Scope Specifier (PR #119908)

[Qiongsi Wu via cfe-commits]

https://github.com/qiongsiwu edited 
https://github.com/llvm/llvm-project/pull/119908
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy][NFC] extract options verify to separately function (PR #120768)

[Julian Schmidt via cfe-commits]

https://github.com/5chmidti approved this pull request.

LGTM

https://github.com/llvm/llvm-project/pull/120768
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][NFC] Fix the static assertion in 4797437 (PR #120643)

[Jon Roelofs via cfe-commits]

jroelofs wrote:

may as well address @nikic's post-commit suggestion to move this out of the 
header, and into a source file, while you're here: 
https://github.com/llvm/llvm-project/pull/120341#discussion_r1893822845

https://github.com/llvm/llvm-project/pull/120643
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][dataflow] Use smart pointer caching in unchecked optional accessor (PR #120249)

[Jan Voung via cfe-commits]

https://github.com/jvoung updated 
https://github.com/llvm/llvm-project/pull/120249

>From c526263a7accc434dbf6e93c2995ceb2f95873b8 Mon Sep 17 00:00:00 2001
From: Jan Voung 
Date: Tue, 17 Dec 2024 15:38:19 +
Subject: [PATCH 1/2] [clang][dataflow] Use smart pointer caching in unchecked
 optional accessor

Part 2 (and final part) following 
https://github.com/llvm/llvm-project/pull/120102
Allows users to do things like:

```
if (o->x.has_value()) {
  ((*o).x).value();
}
```
where the `->` and `*` are operator overload calls.

A user could instead extract the nested optional into a local variable
once instead of doing two accessor calls back to back, but currently
they are unsure why the code is flagged.
---
 .../CachedConstAccessorsLattice.h |  41 
 .../SmartPointerAccessorCaching.h | 168 +++
 .../lib/Analysis/FlowSensitive/CMakeLists.txt |   1 +
 .../Models/UncheckedOptionalAccessModel.cpp   |  58 +-
 .../SmartPointerAccessorCaching.cpp   | 153 ++
 .../Analysis/FlowSensitive/CMakeLists.txt |   1 +
 .../CachedConstAccessorsLatticeTest.cpp   |  32 +++
 .../SmartPointerAccessorCachingTest.cpp   | 194 ++
 .../UncheckedOptionalAccessModelTest.cpp  |  50 +
 9 files changed, 692 insertions(+), 6 deletions(-)
 create mode 100644 
clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h
 create mode 100644 
clang/lib/Analysis/FlowSensitive/SmartPointerAccessorCaching.cpp
 create mode 100644 
clang/unittests/Analysis/FlowSensitive/SmartPointerAccessorCachingTest.cpp

diff --git 
a/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h 
b/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h
index 48c5287367739a..6b5dacf9f66d2d 100644
--- a/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h
+++ b/clang/include/clang/Analysis/FlowSensitive/CachedConstAccessorsLattice.h
@@ -13,7 +13,9 @@
 #ifndef LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_CACHED_CONST_ACCESSORS_LATTICE_H
 #define LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_CACHED_CONST_ACCESSORS_LATTICE_H
 
+#include "clang/AST/Decl.h"
 #include "clang/AST/Expr.h"
+#include "clang/AST/Type.h"
 #include "clang/Analysis/FlowSensitive/DataflowEnvironment.h"
 #include "clang/Analysis/FlowSensitive/DataflowLattice.h"
 #include "clang/Analysis/FlowSensitive/StorageLocation.h"
@@ -71,10 +73,28 @@ template  class CachedConstAccessorsLattice 
: public Base {
   /// Requirements:
   ///
   ///  - `CE` should return a location (GLValue or a record type).
+  ///
+  /// DEPRECATED: switch users to the below overload which takes Callee and 
Type
+  /// directly.
   StorageLocation *getOrCreateConstMethodReturnStorageLocation(
   const RecordStorageLocation &RecordLoc, const CallExpr *CE,
   Environment &Env, llvm::function_ref 
Initialize);
 
+  /// Creates or returns a previously created `StorageLocation` associated with
+  /// a const method call `obj.getFoo()` where `RecordLoc` is the
+  /// `RecordStorageLocation` of `obj`, `Callee` is the decl for `getFoo`,
+  /// and `Type` is the return type of `getFoo`.
+  ///
+  /// The callback `Initialize` runs on the storage location if newly created.
+  ///
+  /// Requirements:
+  ///
+  ///  - `Type` should return a location (GLValue or a record type).
+  StorageLocation &getOrCreateConstMethodReturnStorageLocation(
+  const RecordStorageLocation &RecordLoc, const FunctionDecl *Callee,
+  QualType Type, Environment &Env,
+  llvm::function_ref Initialize);
+
   void clearConstMethodReturnValues(const RecordStorageLocation &RecordLoc) {
 ConstMethodReturnValues.erase(&RecordLoc);
   }
@@ -212,6 +232,27 @@ 
CachedConstAccessorsLattice::getOrCreateConstMethodReturnStorageLocation(
   return &Loc;
 }
 
+template 
+StorageLocation &
+CachedConstAccessorsLattice::getOrCreateConstMethodReturnStorageLocation(
+const RecordStorageLocation &RecordLoc, const FunctionDecl *Callee,
+QualType Type, Environment &Env,
+llvm::function_ref Initialize) {
+  assert(Callee != nullptr);
+  assert(!Type.isNull());
+  assert(Type->isReferenceType() || Type->isRecordType());
+  auto &ObjMap = ConstMethodReturnStorageLocations[&RecordLoc];
+  auto it = ObjMap.find(Callee);
+  if (it != ObjMap.end())
+return *it->second;
+
+  StorageLocation &Loc = Env.createStorageLocation(Type.getNonReferenceType());
+  Initialize(Loc);
+
+  ObjMap.insert({Callee, &Loc});
+  return Loc;
+}
+
 } // namespace dataflow
 } // namespace clang
 
diff --git 
a/clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h 
b/clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h
new file mode 100644
index 00..e89e82c893d8cb
--- /dev/null
+++ b/clang/include/clang/Analysis/FlowSensitive/SmartPointerAccessorCaching.h
@@ -0,0 +1,168 @@
+//===-- SmartPointerAccessorCaching.h ---*- C++ 
-*-===//
+//
+// Part of the LLVM Project, u

[clang] [llvm] [LLVM][Clang][AArch64] Implement AArch64 build attributes (PR #118771)

[via cfe-commits]

sivan-shani wrote:

- Fixed according to comments above
- Fixed some errors
- Ability to add attributes numerically added
- Tests for pauthabi platform/version in codegen
- Tests for assembly parser errors
- Not added: Tests for attributes with NTBS values - currently all attributes 
takes only ULEB128 as values.

https://github.com/llvm/llvm-project/pull/118771
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Fix a cmake error when using the Xcode generator. (PR #119403)

[Thomas Fransham via cfe-commits]

fsfod wrote:

I guess I must of broke build for the MSVC MSBuild/vcxproj CMake generator as 
well, maybe it can use the same kind checks as earlier code like this
``` cmake
if(NOT XCODE AND NOT MSVC_IDE)
  target_compile_definitions("obj.${name}" PUBLIC CLANG_EXPORTS)
else()
  target_compile_definitions("${name}" PUBLIC CLANG_EXPORTS)
endif()
```

https://github.com/llvm/llvm-project/pull/119403
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [AArch64] Fix argument passing for SVE tuples (PR #118961)

[Eli Friedman via cfe-commits]

https://github.com/efriedma-quic approved this pull request.

LGTM

I somehow thought you could use arm_sve_vector_bits with tuples, but I guess 
not.

https://github.com/llvm/llvm-project/pull/118961
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] Remove -bounds-checking-unique-traps (replace with -fno-sanitize-merge=local-bounds) (PR #120682)

[Thurston Dang via cfe-commits]

https://github.com/thurstond updated 
https://github.com/llvm/llvm-project/pull/120682

>From ee51ed7bd68df7b2dae3f1426471b34d0388a42f Mon Sep 17 00:00:00 2001
From: Thurston Dang 
Date: Fri, 20 Dec 2024 04:11:36 +
Subject: [PATCH 1/5] Remove -bounds-checking-unique-traps (replace with
 -fno-sanitize-merge=local-bounds)

-fno-sanitize-merge (introduced in #120511) combines the functionality of 
-ubsan-unique-traps
and -bounds-checking-unique-traps, while allowing fine-grained control
of which UBSan checks to prevent merging. #120613 removed -ubsan-unique-traps.
This patch removes -bound-checking-unique-traps, which can be controlled
via -fno-sanitize-merge=local-bounds.

Note: this patch subtly changes -fsanitize-merge (the default) to also
include -fsanitize-merge=local-bounds. This is different from the
previous behavior, where -fsanitize-merge (or the old
-ubsan-unique-traps) did not affect local-bounds (requiring the separate
-bounds-checking-unique-traps). However, we argue that the new behavior
is more intuitive.

Removing -bounds-checking-unique-traps and merging its functionality into 
-fsanitize-merge breaks backwards compatibility; we hope that this is 
acceptable since '-mllvm -bounds-checking-unique-traps' was an experimental 
flag.
---
 clang/docs/ReleaseNotes.rst   |  14 ++-
 clang/lib/CodeGen/BackendUtil.cpp |   5 +-
 clang/test/CodeGen/bounds-checking.c  |  14 ++-
 .../Instrumentation/BoundsChecking.h  |  15 ++-
 llvm/lib/Passes/PassBuilder.cpp   |  29 +++--
 llvm/lib/Passes/PassRegistry.def  |   4 +-
 .../Instrumentation/BoundsChecking.cpp|  24 ++--
 .../BoundsChecking/runtimes.ll| 107 ++
 .../BoundsChecking/ubsan-unique-traps.ll  |   5 +-
 9 files changed, 178 insertions(+), 39 deletions(-)

diff --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index b8d92a6c881c68..0c6c894e17416a 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -445,9 +445,10 @@ New Compiler Flags
 - The ``-Warray-compare-cxx26`` warning has been added to warn about array 
comparison
   starting from C++26, this warning is enabled as an error by default.
 
-- '-fsanitize-merge' (default) and '-fno-sanitize-merge' have been added for
-  fine-grained control of which UBSan checks are allowed to be merged by the
-  backend (for example, -fno-sanitize-merge=bool,enum).
+- ``-fsanitize-merge`` (default) and ``-fno-sanitize-merge`` have been added 
for
+  fine-grained, unified control of which UBSan checks can potentially be merged
+  by the compiler (for example,
+  ``-fno-sanitize-merge=bool,enum,array-bounds,local-bounds``).
 
 Deprecated Compiler Flags
 -
@@ -488,8 +489,11 @@ Removed Compiler Flags
   derivatives) is now removed, since it's no longer possible to suppress the
   diagnostic (see above). Users can expect an `unknown warning` diagnostic if
   it's still in use.
-- The experimental flag '-ubsan-unique-traps' has been removed. It is
-  superseded by '-fno-sanitize-merge'.
+- The experimental flags '-ubsan-unique-traps' and
+  '-bounds-checking-unique-traps' have been removed. The combination of the
+  two flags is equivalent to '-fno-sanitize-merge' with no parameters.
+  '-bounds-checking-unique-traps' can be selectively controlled via
+  '-f(no-)sanitize-merge=local-bounds'.
 
 Attribute Changes in Clang
 --
diff --git a/clang/lib/CodeGen/BackendUtil.cpp 
b/clang/lib/CodeGen/BackendUtil.cpp
index e6c9d77d29f6f1..bfb73aa51b9b5b 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -1030,6 +1030,8 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
   PB.registerScalarOptimizerLateEPCallback(
   [this](FunctionPassManager &FPM, OptimizationLevel Level) {
 BoundsCheckingPass::ReportingMode Mode;
+bool Merge = 
CodeGenOpts.SanitizeMergeHandlers.has(SanitizerKind::LocalBounds);
+
 if (CodeGenOpts.SanitizeTrap.has(SanitizerKind::LocalBounds)) {
   Mode = BoundsCheckingPass::ReportingMode::Trap;
 } else if (CodeGenOpts.SanitizeMinimalRuntime) {
@@ -1041,7 +1043,8 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
  ? BoundsCheckingPass::ReportingMode::FullRuntime
  : BoundsCheckingPass::ReportingMode::FullRuntimeAbort;
 }
-FPM.addPass(BoundsCheckingPass(Mode));
+BoundsCheckingPass::BoundsCheckingOptions Options(Mode, Merge);
+FPM.addPass(BoundsCheckingPass(Options));
   });
 
 // Don't add sanitizers if we are here from ThinLTO PostLink. That already
diff --git a/clang/test/CodeGen/bounds-checking.c 
b/clang/test/CodeGen/bounds-checking.c
index f9319ca61670c3..5e6b317a99969e 100644
--- a/clang/test/CodeGen/bounds-checking.c
+++ b/clang/test/CodeGen/bounds-checking.c
@@ -1,12 +1,14 @@
-// RUN: %c

[clang] 5bb6503 - Remove -bounds-checking-unique-traps (replace with -fno-sanitize-merge=local-bounds) (#120682)

[via cfe-commits]

Author: Thurston Dang
Date: 2024-12-20T10:07:44-08:00
New Revision: 5bb650345d83669434713146aaa431c1f7ad43d6

URL: 
https://github.com/llvm/llvm-project/commit/5bb650345d83669434713146aaa431c1f7ad43d6
DIFF: 
https://github.com/llvm/llvm-project/commit/5bb650345d83669434713146aaa431c1f7ad43d6.diff

LOG: Remove -bounds-checking-unique-traps (replace with 
-fno-sanitize-merge=local-bounds) (#120682)

#120613 removed -ubsan-unique-traps and replaced it with
-fno-sanitize-merge (introduced in #120511), which allows fine-grained
control of which UBSan checks to prevent merging. This analogous patch
removes -bound-checking-unique-traps, and allows it to be controlled via
-fno-sanitize-merge=local-bounds.

Most of this patch is simply plumbing through the compiler flags into
the bounds checking pass.

Note: this patch subtly changes -fsanitize-merge (the default) to also
include -fsanitize-merge=local-bounds. This is different from the
previous behavior, where -fsanitize-merge (or the old
-ubsan-unique-traps) did not affect local-bounds (requiring the separate
-bounds-checking-unique-traps). However, we argue that the new behavior
is more intuitive.

Removing -bounds-checking-unique-traps and merging its functionality
into -fsanitize-merge breaks backwards compatibility; we hope that this
is acceptable since '-mllvm -bounds-checking-unique-traps' was an
experimental flag.

Added: 


Modified: 
clang/docs/ReleaseNotes.rst
clang/docs/UndefinedBehaviorSanitizer.rst
clang/lib/CodeGen/BackendUtil.cpp
clang/test/CodeGen/bounds-checking.c
llvm/include/llvm/Transforms/Instrumentation/BoundsChecking.h
llvm/lib/Passes/PassBuilder.cpp
llvm/lib/Passes/PassRegistry.def
llvm/lib/Transforms/Instrumentation/BoundsChecking.cpp
llvm/test/Instrumentation/BoundsChecking/many-trap.ll
llvm/test/Instrumentation/BoundsChecking/runtimes.ll
llvm/test/Instrumentation/BoundsChecking/simple-32.ll
llvm/test/Instrumentation/BoundsChecking/simple.ll
llvm/test/Instrumentation/BoundsChecking/ubsan-unique-traps.ll

Removed: 




diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index 38c35aea3cf220..d9d016b30f93d6 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -445,10 +445,6 @@ New Compiler Flags
 - The ``-Warray-compare-cxx26`` warning has been added to warn about array 
comparison
   starting from C++26, this warning is enabled as an error by default.
 
-- '-fsanitize-merge' (default) and '-fno-sanitize-merge' have been added for
-  fine-grained control of which UBSan checks are allowed to be merged by the
-  backend (for example, -fno-sanitize-merge=bool,enum).
-
 Deprecated Compiler Flags
 -
 
@@ -488,8 +484,6 @@ Removed Compiler Flags
   derivatives) is now removed, since it's no longer possible to suppress the
   diagnostic (see above). Users can expect an `unknown warning` diagnostic if
   it's still in use.
-- The experimental flag '-ubsan-unique-traps' has been removed. It is
-  superseded by '-fno-sanitize-merge'.
 
 Attribute Changes in Clang
 --
@@ -1213,6 +1207,11 @@ Sanitizers
 
 - Implemented ``-f[no-]sanitize-trap=local-bounds``, and 
``-f[no-]sanitize-recover=local-bounds``.
 
+- ``-fsanitize-merge`` (default) and ``-fno-sanitize-merge`` have been added 
for
+  fine-grained, unified control of which UBSan checks can potentially be merged
+  by the compiler (for example,
+  ``-fno-sanitize-merge=bool,enum,array-bounds,local-bounds``).
+
 Python Binding Changes
 --
 - Fixed an issue that led to crashes when calling 
``Type.get_exception_specification_kind``.

diff  --git a/clang/docs/UndefinedBehaviorSanitizer.rst 
b/clang/docs/UndefinedBehaviorSanitizer.rst
index 671db7f9f36714..b9ee4484fb9aec 100644
--- a/clang/docs/UndefinedBehaviorSanitizer.rst
+++ b/clang/docs/UndefinedBehaviorSanitizer.rst
@@ -276,8 +276,8 @@ Stack traces and report symbolization
 If you want UBSan to print symbolized stack trace for each error report, you
 will need to:
 
-#. Compile with ``-g`` and ``-fno-omit-frame-pointer`` to get proper debug
-   information in your binary.
+#. Compile with ``-g``, ``-fno-sanitize-merge`` and ``-fno-omit-frame-pointer``
+   to get proper debug information in your binary.
 #. Run your program with environment variable
``UBSAN_OPTIONS=print_stacktrace=1``.
 #. Make sure ``llvm-symbolizer`` binary is in ``PATH``.

diff  --git a/clang/lib/CodeGen/BackendUtil.cpp 
b/clang/lib/CodeGen/BackendUtil.cpp
index e6c9d77d29f6f1..04358cd6d7c232 100644
--- a/clang/lib/CodeGen/BackendUtil.cpp
+++ b/clang/lib/CodeGen/BackendUtil.cpp
@@ -1030,6 +1030,9 @@ void EmitAssemblyHelper::RunOptimizationPipeline(
   PB.registerScalarOptimizerLateEPCallback(
   [this](FunctionPassManager &FPM, OptimizationLevel Level) {
 BoundsCheckingPass::ReportingMode Mode;
+

[clang] [llvm] Remove -bounds-checking-unique-traps (replace with -fno-sanitize-merge=local-bounds) (PR #120682)

[Thurston Dang via cfe-commits]

https://github.com/thurstond closed 
https://github.com/llvm/llvm-project/pull/120682
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [AArch64] Fix argument passing for SVE tuples (PR #118961)

[Momchil Velikov via cfe-commits]

momchil-velikov wrote:

Thanks!

https://github.com/llvm/llvm-project/pull/118961
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[Brian Cain via cfe-commits]

androm3da wrote:

LGTM (after the `clang-format` fix)

https://github.com/llvm/llvm-project/pull/120773
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] 5162fde - [webkit.UncountedLambdaCapturesChecker] Fix a nullptr deference. (#120702)

[via cfe-commits]

Author: Ryosuke Niwa
Date: 2024-12-20T10:18:08-08:00
New Revision: 5162fde6ee6565d39511e451c04865e7b53bcdcc

URL: 
https://github.com/llvm/llvm-project/commit/5162fde6ee6565d39511e451c04865e7b53bcdcc
DIFF: 
https://github.com/llvm/llvm-project/commit/5162fde6ee6565d39511e451c04865e7b53bcdcc.diff

LOG: [webkit.UncountedLambdaCapturesChecker] Fix a nullptr deference. (#120702)

Added a nullptr check.

Added: 

clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp

Modified: 
clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp

Removed: 




diff  --git 
a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp 
b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
index da9698e327562e..a57499d52acd0c 100644
--- 
a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
+++ 
b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
@@ -155,6 +155,8 @@ class UncountedLambdaCapturesChecker
 if (!Init)
   return nullptr;
 TempExpr = dyn_cast(Init->IgnoreParenCasts());
+if (!TempExpr)
+  return nullptr;
 return dyn_cast_or_null(TempExpr->getSubExpr());
   }
 

diff  --git 
a/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
 
b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
new file mode 100644
index 00..4d9edb75b7ff36
--- /dev/null
+++ 
b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
@@ -0,0 +1,16 @@
+// RUN: %clang_analyze_cc1 
-analyzer-checker=webkit.UncountedLambdaCapturesChecker -verify %s
+// expected-no-diagnostics
+
+struct Foo {
+  int x;
+  int y;
+  Foo(int x, int y) : x(x) , y(y) { }
+  ~Foo() { }
+};
+
+Foo bar(const Foo&);
+void foo() {
+  int x = 7;
+  int y = 5;
+  bar(Foo(x, y));
+}



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [webkit.UncountedLambdaCapturesChecker] Fix a nullptr deference. (PR #120702)

[Ryosuke Niwa via cfe-commits]

https://github.com/rniwa closed https://github.com/llvm/llvm-project/pull/120702
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Return larger CXX records in memory (PR #120670)

[Eli Friedman via cfe-commits]

efriedma-quic wrote:

Missing regression test in clang/test/CodeGen.

We probably want a release note for this (clang/docs/ReleaseNotes.rst).

(If you're interested, there's another x86-64 ABI bug which nobody got around 
to fixing: #76017.)

https://github.com/llvm/llvm-project/pull/120670
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Introduce a new UEFI target predefine. (PR #111719)

[via cfe-commits]

https://github.com/Prabhuk updated 
https://github.com/llvm/llvm-project/pull/111719

>From 05eab970b59c61b22e8afa1ed9381906ae925c03 Mon Sep 17 00:00:00 2001
From: prabhukr 
Date: Wed, 9 Oct 2024 10:16:58 -0700
Subject: [PATCH 1/4] [clang] Introduce a new UEFI target predefine.

---
 clang/lib/Basic/Targets/OSTargets.h | 5 -
 1 file changed, 4 insertions(+), 1 deletion(-)

diff --git a/clang/lib/Basic/Targets/OSTargets.h 
b/clang/lib/Basic/Targets/OSTargets.h
index a83d6464e789d6..92c76b303def65 100644
--- a/clang/lib/Basic/Targets/OSTargets.h
+++ b/clang/lib/Basic/Targets/OSTargets.h
@@ -783,7 +783,10 @@ template 
 class LLVM_LIBRARY_VISIBILITY UEFITargetInfo : public OSTargetInfo {
 protected:
   void getOSDefines(const LangOptions &Opts, const llvm::Triple &Triple,
-MacroBuilder &Builder) const override {}
+MacroBuilder &Builder) const override {
+Builder.defineMacro("__UEFI__");
+Builder.defineMacro("__PECOFF__");
+  }
 
 public:
   UEFITargetInfo(const llvm::Triple &Triple, const TargetOptions &Opts)

>From 4ef8bcd3a9019299145b59d4f26cd4307171bed5 Mon Sep 17 00:00:00 2001
From: Prabhuk 
Date: Wed, 13 Nov 2024 10:30:50 -0800
Subject: [PATCH 2/4] Update OSTargets.h

Remove PE_COFF
---
 clang/lib/Basic/Targets/OSTargets.h | 1 -
 1 file changed, 1 deletion(-)

diff --git a/clang/lib/Basic/Targets/OSTargets.h 
b/clang/lib/Basic/Targets/OSTargets.h
index 92c76b303def65..5a23eef0117294 100644
--- a/clang/lib/Basic/Targets/OSTargets.h
+++ b/clang/lib/Basic/Targets/OSTargets.h
@@ -785,7 +785,6 @@ class LLVM_LIBRARY_VISIBILITY UEFITargetInfo : public 
OSTargetInfo {
   void getOSDefines(const LangOptions &Opts, const llvm::Triple &Triple,
 MacroBuilder &Builder) const override {
 Builder.defineMacro("__UEFI__");
-Builder.defineMacro("__PECOFF__");
   }
 
 public:

>From bcccf1f86609a828130c1a9474d10b78c8f212c9 Mon Sep 17 00:00:00 2001
From: Prabhuk 
Date: Wed, 13 Nov 2024 10:36:15 -0800
Subject: [PATCH 3/4] add __UEFI__ to windows predefines

Updating the PE COFF generating windows code path to predefine the new __UEFI__ 
macro.
---
 clang/lib/Basic/Targets/OSTargets.cpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/clang/lib/Basic/Targets/OSTargets.cpp 
b/clang/lib/Basic/Targets/OSTargets.cpp
index b56e2c7ca9c494..b9edeefa60148e 100644
--- a/clang/lib/Basic/Targets/OSTargets.cpp
+++ b/clang/lib/Basic/Targets/OSTargets.cpp
@@ -259,6 +259,7 @@ static void addVisualCDefines(const LangOptions &Opts, 
MacroBuilder &Builder) {
 
 void addWindowsDefines(const llvm::Triple &Triple, const LangOptions &Opts,
MacroBuilder &Builder) {
+  Builder.defineMacro("__UEFI__");
   Builder.defineMacro("_WIN32");
   if (Triple.isArch64Bit())
 Builder.defineMacro("_WIN64");

>From 2f6af5664bbd6f3cc01dfb7b46f67190dc4efba4 Mon Sep 17 00:00:00 2001
From: prabhukr 
Date: Fri, 20 Dec 2024 10:18:18 -0800
Subject: [PATCH 4/4] Removing the predefine from Windows targets. Keeping it
 only on UEFI.

---
 clang/lib/Basic/Targets/OSTargets.cpp | 1 -
 1 file changed, 1 deletion(-)

diff --git a/clang/lib/Basic/Targets/OSTargets.cpp 
b/clang/lib/Basic/Targets/OSTargets.cpp
index b9edeefa60148e..b56e2c7ca9c494 100644
--- a/clang/lib/Basic/Targets/OSTargets.cpp
+++ b/clang/lib/Basic/Targets/OSTargets.cpp
@@ -259,7 +259,6 @@ static void addVisualCDefines(const LangOptions &Opts, 
MacroBuilder &Builder) {
 
 void addWindowsDefines(const llvm::Triple &Triple, const LangOptions &Opts,
MacroBuilder &Builder) {
-  Builder.defineMacro("__UEFI__");
   Builder.defineMacro("_WIN32");
   if (Triple.isArch64Bit())
 Builder.defineMacro("_WIN64");

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Introduce a new UEFI target predefine. (PR #111719)

[via cfe-commits]

https://github.com/Prabhuk edited 
https://github.com/llvm/llvm-project/pull/111719
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Introduce a new UEFI target predefine. (PR #111719)

[via cfe-commits]

https://github.com/Prabhuk edited 
https://github.com/llvm/llvm-project/pull/111719
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Introduce a new UEFI target predefine. (PR #111719)

[via cfe-commits]

@@ -259,6 +259,7 @@ static void addVisualCDefines(const LangOptions &Opts, 
MacroBuilder &Builder) {
 
 void addWindowsDefines(const llvm::Triple &Triple, const LangOptions &Opts,
MacroBuilder &Builder) {
+  Builder.defineMacro("__UEFI__");

Prabhuk wrote:

Removed it from Windows target.

https://github.com/llvm/llvm-project/pull/111719
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [Hexagon] Add V75 support to compiler and assembler (PR #120773)

[Ikhlas Ajbar via cfe-commits]

https://github.com/iajbar updated 
https://github.com/llvm/llvm-project/pull/120773

>From a57f928e545e4bd398fe0ae3fca8702c83a79516 Mon Sep 17 00:00:00 2001
From: Ikhlas Ajbar 
Date: Fri, 13 Dec 2024 14:54:44 -0800
Subject: [PATCH] [Hexagon] Add V75 support to compiler and assembler

This patch introduces support for the Hexagon V75 architecture.
It includes instruction formats, definitions, encodings, scheduling classes,
and builtins/intrinsics.
---
 clang/include/clang/Basic/BuiltinsHexagon.def |  10 +-
 clang/include/clang/Driver/Options.td |   2 +
 clang/lib/Basic/Targets/Hexagon.cpp   |  15 +-
 clang/test/Driver/hexagon-toolchain-elf.c |   7 +
 .../Misc/target-invalid-cpu-note/hexagon.c|   1 +
 clang/test/Preprocessor/hexagon-predefines.c  |  16 +
 llvm/include/llvm/BinaryFormat/ELF.h  |   1 +
 llvm/lib/Object/ELFObjectFile.cpp |   2 +
 llvm/lib/ObjectYAML/ELFYAML.cpp   |   2 +
 llvm/lib/Target/Hexagon/Hexagon.td|  14 +
 llvm/lib/Target/Hexagon/HexagonDepArch.h  |   4 +-
 llvm/lib/Target/Hexagon/HexagonDepArch.td |   2 +
 llvm/lib/Target/Hexagon/HexagonDepIICHVX.td   | 592 
 .../lib/Target/Hexagon/HexagonDepIICScalar.td | 888 ++
 llvm/lib/Target/Hexagon/HexagonSchedule.td|   1 +
 llvm/lib/Target/Hexagon/HexagonScheduleV75.td |  39 +
 llvm/lib/Target/Hexagon/HexagonSubtarget.h|   6 +
 .../MCTargetDesc/HexagonMCELFStreamer.cpp |   3 +
 .../MCTargetDesc/HexagonMCTargetDesc.cpp  |  68 +-
 llvm/test/MC/Hexagon/arch-support.s   |   9 +
 llvm/test/MC/Hexagon/v75_arch.s   |  10 +
 21 files changed, 1656 insertions(+), 36 deletions(-)
 create mode 100644 llvm/lib/Target/Hexagon/HexagonScheduleV75.td
 create mode 100644 llvm/test/MC/Hexagon/v75_arch.s

diff --git a/clang/include/clang/Basic/BuiltinsHexagon.def 
b/clang/include/clang/Basic/BuiltinsHexagon.def
index 0dc0f4567dd413..ca3f99176110e3 100644
--- a/clang/include/clang/Basic/BuiltinsHexagon.def
+++ b/clang/include/clang/Basic/BuiltinsHexagon.def
@@ -17,8 +17,10 @@
 #   define TARGET_BUILTIN(ID, TYPE, ATTRS, FEATURE) BUILTIN(ID, TYPE, ATTRS)
 #endif
 
+#pragma push_macro("V75")
+#define V75 "v75"
 #pragma push_macro("V73")
-#define V73 "v73"
+#define V73 "v73|" V75
 #pragma push_macro("V71")
 #define V71 "v71|" V73
 #pragma push_macro("V69")
@@ -40,8 +42,10 @@
 #pragma push_macro("V5")
 #define V5 "v5|" V55
 
+#pragma push_macro("HVXV75")
+#define HVXV75 "hvxv75"
 #pragma push_macro("HVXV73")
-#define HVXV73 "hvxv73"
+#define HVXV73 "hvxv73|" HVXV75
 #pragma push_macro("HVXV71")
 #define HVXV71 "hvxv71|" HVXV73
 #pragma push_macro("HVXV69")
@@ -143,6 +147,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("HVXV69")
 #pragma pop_macro("HVXV71")
 #pragma pop_macro("HVXV73")
+#pragma pop_macro("HVXV75")
 
 #pragma pop_macro("V5")
 #pragma pop_macro("V55")
@@ -155,6 +160,7 @@ 
TARGET_BUILTIN(__builtin_HEXAGON_V6_vrmpyub_rtt_acc_128B,"V64iV64iV32iLLi","", "
 #pragma pop_macro("V69")
 #pragma pop_macro("V71")
 #pragma pop_macro("V73")
+#pragma pop_macro("V75")
 
 #undef BUILTIN
 #undef TARGET_BUILTIN
diff --git a/clang/include/clang/Driver/Options.td 
b/clang/include/clang/Driver/Options.td
index 638f8c52053ec5..9edf308cc685ea 100644
--- a/clang/include/clang/Driver/Options.td
+++ b/clang/include/clang/Driver/Options.td
@@ -6224,6 +6224,8 @@ def mv71t : Flag<["-"], "mv71t">, 
Group,
   Alias, AliasArgs<["hexagonv71t"]>;
 def mv73 : Flag<["-"], "mv73">, Group,
   Alias, AliasArgs<["hexagonv73"]>;
+def mv75 : Flag<["-"], "mv75">, Group,
+  Alias, AliasArgs<["hexagonv75"]>;
 def mhexagon_hvx : Flag<["-"], "mhvx">, Group,
   HelpText<"Enable Hexagon Vector eXtensions">;
 def mhexagon_hvx_EQ : Joined<["-"], "mhvx=">,
diff --git a/clang/lib/Basic/Targets/Hexagon.cpp 
b/clang/lib/Basic/Targets/Hexagon.cpp
index 0282ac812c306f..931327bd8657b0 100644
--- a/clang/lib/Basic/Targets/Hexagon.cpp
+++ b/clang/lib/Basic/Targets/Hexagon.cpp
@@ -78,6 +78,9 @@ void HexagonTargetInfo::getTargetDefines(const LangOptions 
&Opts,
   } else if (CPU == "hexagonv73") {
 Builder.defineMacro("__HEXAGON_V73__");
 Builder.defineMacro("__HEXAGON_ARCH__", "73");
+  } else if (CPU == "hexagonv75") {
+Builder.defineMacro("__HEXAGON_V75__");
+Builder.defineMacro("__HEXAGON_ARCH__", "75");
   }
 
   if (hasFeature("hvx-length64b")) {
@@ -229,13 +232,13 @@ struct CPUSuffix {
 };
 
 static constexpr CPUSuffix Suffixes[] = {
-{{"hexagonv5"},  {"5"}},  {{"hexagonv55"},  {"55"}},
-{{"hexagonv60"}, {"60"}}, {{"hexagonv62"},  {"62"}},
-{{"hexagonv65"}, {"65"}}, {{"hexagonv66"},  {"66"}},
+{{"hexagonv5"}, {"5"}},   {{"hexagonv55"}, {"55"}},
+{{"hexagonv60"}, {"60"}}, {{"hexagonv62"}, {"62"}},
+{{"hexagonv65"}, {"65"}}, {{"hexagonv66"}, {"66"}},
 {{"hexagonv67"}, {"67"}}, {{"hexagonv67t"}, {"67t"}},
-{{"hexagonv68"}, {"68"}}, {{"hexagonv69"},  {"69"}},
-{{"hexa

[clang] [Darwin][Driver][clang] apple-none-macho orders the resource directory after internal-externc-isystem when nostdlibinc is used (PR #120507)

[Ian Anderson via cfe-commits]

@@ -1018,13 +1018,19 @@ bool Darwin::hasBlocksRuntime() const {
   }
 }
 
-void Darwin::AddCudaIncludeArgs(const ArgList &DriverArgs,
-ArgStringList &CC1Args) const {
+void MachO::AddCudaIncludeArgs(const ArgList &DriverArgs,
+   ArgStringList &CC1Args) const {
+  if (!isTargetAppleBased())
+return ToolChain::AddCudaIncludeArgs(DriverArgs, CC1Args);

ian-twilightcoder wrote:

>From `Driver::getToolChain` you get it for these
arch-vendor-win32-macho
arch-vendor-windows-macho
arch-vendor-none-macho

The only one I really want to change is arch-apple-none-macho, I don't want to 
disturb any of the others. But that's a good point, it would probably be 
cleaner to make an AppleMachO toolchain in between the existing MachO and 
Darwin ones

https://github.com/llvm/llvm-project/pull/120507
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [CIR] floating-point, pointer, and function types (PR #120484)

[Bruno Cardoso Lopes via cfe-commits]

@@ -18,6 +21,87 @@ mlir::MLIRContext &CIRGenTypes::getMLIRContext() const {
   return *builder.getContext();
 }
 
+/// Return true if the specified type in a function parameter or result 
position
+/// can be converted to a CIR type at this point. This boils down to being
+/// whether it is complete, as well as whether we've temporarily deferred
+/// expanding the type because we're in a recursive context.
+bool CIRGenTypes::isFuncParamTypeConvertible(clang::QualType type) {
+  // Some ABIs cannot have their member pointers represented in LLVM IR unless
+  // certain circumstances have been reached.
+  assert(!type->getAs() && "NYI");
+
+  // If this isn't a tagged type, we can convert it!

bcardosolopes wrote:

I don't remember the details on why it was necessary at the time to deviate - 
so I just made this be equivalent to traditional codegen and everything passes: 
https://github.com/llvm/clangir/commit/04d7dcfb2582753f3eccbf01ec900d60297cbf4b

That said, I don't think we need to introduce any asserts here, can't get 
better than OG for the sake of completeness. Not supported types will crash 
given the other mechanisms David mentioned, but the recursive logic is sound 
with pre-existing work.

Unless Erich has any other concerns, this LGTM to land.

https://github.com/llvm/llvm-project/pull/120484
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [CIR] floating-point, pointer, and function types (PR #120484)

[Erich Keane via cfe-commits]

@@ -18,6 +21,87 @@ mlir::MLIRContext &CIRGenTypes::getMLIRContext() const {
   return *builder.getContext();
 }
 
+/// Return true if the specified type in a function parameter or result 
position
+/// can be converted to a CIR type at this point. This boils down to being
+/// whether it is complete, as well as whether we've temporarily deferred
+/// expanding the type because we're in a recursive context.
+bool CIRGenTypes::isFuncParamTypeConvertible(clang::QualType type) {
+  // Some ABIs cannot have their member pointers represented in LLVM IR unless
+  // certain circumstances have been reached.
+  assert(!type->getAs() && "NYI");
+
+  // If this isn't a tagged type, we can convert it!

erichkeane wrote:

If this includes llvm/clangir@04d7dcf in it, rather than the current 
implementation, I think I'm OK without any asserts.  I'll Approve, but 
conditional on that patch being included in this one.

https://github.com/llvm/llvm-project/pull/120484
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [CIR] floating-point, pointer, and function types (PR #120484)

[Erich Keane via cfe-commits]

https://github.com/erichkeane approved this pull request.

Approve, conditional on isFuncParamTypeConvertible change being pulled from the 
incubator.

https://github.com/llvm/llvm-project/pull/120484
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][Sema] Fixes for %b printf extension handling (PR #120689)

[Brad Smith via cfe-commits]

https://github.com/brad0 updated 
https://github.com/llvm/llvm-project/pull/120689

>From 1da99c0d32e996219a4b4dcad67b0321cab753e3 Mon Sep 17 00:00:00 2001
From: Brad Smith 
Date: Thu, 19 Dec 2024 21:35:57 -0500
Subject: [PATCH] [clang][Sema] Fixes for %b printf extension handling

The %b printf extension in the kernel is not fixed to a int type. On
sparc64 there are various %llb formats. Adjust the code to handle the
length specifiers and type check like it is used by the regular case.
---
 clang/include/clang/AST/FormatString.h   |  6 ++--
 clang/lib/Sema/SemaChecking.cpp  | 38 +---
 clang/test/Sema/format-strings-freebsd.c |  8 -
 3 files changed, 38 insertions(+), 14 deletions(-)

diff --git a/clang/include/clang/AST/FormatString.h 
b/clang/include/clang/AST/FormatString.h
index a074dd23e2ad4c..1ff0d036b70d1a 100644
--- a/clang/include/clang/AST/FormatString.h
+++ b/clang/include/clang/AST/FormatString.h
@@ -240,8 +240,10 @@ class ConversionSpecifier {
 
   bool isIntArg() const { return (kind >= IntArgBeg && kind <= IntArgEnd) ||
 kind == FreeBSDrArg || kind == FreeBSDyArg; }
-  bool isUIntArg() const { return kind >= UIntArgBeg && kind <= UIntArgEnd; }
-  bool isAnyIntArg() const { return kind >= IntArgBeg && kind <= UIntArgEnd; }
+  bool isUIntArg() const { return (kind >= UIntArgBeg && kind <= UIntArgEnd) ||
+kind == FreeBSDbArg; }
+  bool isAnyIntArg() const { return (kind >= IntArgBeg && kind <= UIntArgEnd) 
||
+kind == FreeBSDbArg; }
   bool isDoubleArg() const {
 return kind >= DoubleArgBeg && kind <= DoubleArgEnd;
   }
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index e703a62ff9cf18..e50118762fff32 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -7170,18 +7170,34 @@ bool CheckPrintfHandler::HandlePrintfSpecifier(
 // Claim the second argument.
 CoveredArgs.set(argIndex + 1);
 
-// Type check the first argument (int for %b, pointer for %D)
 const Expr *Ex = getDataArg(argIndex);
-const analyze_printf::ArgType &AT =
-  (CS.getKind() == ConversionSpecifier::FreeBSDbArg) ?
-ArgType(S.Context.IntTy) : ArgType::CPointerTy;
-if (AT.isValid() && !AT.matchesType(S.Context, Ex->getType()))
-  EmitFormatDiagnostic(
-  S.PDiag(diag::warn_format_conversion_argument_type_mismatch)
-  << AT.getRepresentativeTypeName(S.Context) << Ex->getType()
-  << false << Ex->getSourceRange(),
-  Ex->getBeginLoc(), /*IsStringLocation*/ false,
-  getSpecifierRange(startSpecifier, specifierLen));
+if (CS.getKind() == ConversionSpecifier::FreeBSDDArg) {
+  // Type check the first argument (pointer for %D)
+  const analyze_printf::ArgType &AT = ArgType::CPointerTy;
+  if (AT.isValid() && !AT.matchesType(S.Context, Ex->getType()))
+EmitFormatDiagnostic(
+S.PDiag(diag::warn_format_conversion_argument_type_mismatch)
+<< AT.getRepresentativeTypeName(S.Context) << Ex->getType()
+<< false << Ex->getSourceRange(),
+Ex->getBeginLoc(), /*IsStringLocation*/ false,
+getSpecifierRange(startSpecifier, specifierLen));
+} else {
+  // Check the length modifier for %b
+  if (!FS.hasValidLengthModifier(S.getASTContext().getTargetInfo(),
+ S.getLangOpts()))
+HandleInvalidLengthModifier(FS, CS, startSpecifier, specifierLen,
+diag::warn_format_nonsensical_length);
+  else if (!FS.hasStandardLengthModifier())
+HandleNonStandardLengthModifier(FS, startSpecifier, specifierLen);
+  else if (!FS.hasStandardLengthConversionCombination())
+HandleInvalidLengthModifier(
+FS, CS, startSpecifier, specifierLen,
+diag::warn_format_non_standard_conversion_spec);
+
+  // Type check the first argument of %b
+  if (!checkFormatExpr(FS, startSpecifier, specifierLen, Ex))
+return false;
+}
 
 // Type check the second argument (char * for both %b and %D)
 Ex = getDataArg(argIndex + 1);
diff --git a/clang/test/Sema/format-strings-freebsd.c 
b/clang/test/Sema/format-strings-freebsd.c
index 3d3ed6b40f48dc..3540dca2cfabcd 100644
--- a/clang/test/Sema/format-strings-freebsd.c
+++ b/clang/test/Sema/format-strings-freebsd.c
@@ -8,13 +8,19 @@ int freebsd_kernel_printf(const char *, ...) 
__attribute__((__format__(__freebsd
 
 void check_freebsd_kernel_extensions(int i, long l, char *s, short h)
 {
-  // %b expects an int and a char *
+  // %b expects a char *
   freebsd_kernel_printf("reg=%b\n", i, "\10\2BITTWO\1BITONE\n"); // no-warning
   freebsd_kernel_printf("reg=%b\n", l, "\10\2BITTWO\1BITONE\n"); // 
expected-warning{{format specifies type 'int' but the argument has type 'long'}}
   freebsd_kernel_printf("reg=%b\n", i, l); // expected-warning{{format 
specifies type 'char *' but the argument has type 'long

[clang] Fix double-quotes in diagnostic when attempting to access a ext_vector of bools (PR #118186)

[Timm Baeder via cfe-commits]

tbaederr wrote:

Do you need someone to push this?

https://github.com/llvm/llvm-project/pull/118186
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Patch series to reapply #118734 and substantially improve it (PR #120534)

[via cfe-commits]

dyung wrote:

> @dyung -- Could you try out 
> https://github.com/chandlerc/llvm-project/tree/shard-loongarch and see if 
> that works?
> 
> Notably, it includes one additional patch on top of this series: 
> [chandlerc@2d59328](https://github.com/chandlerc/llvm-project/commit/2d593288dc18c55307779ae82a18d024761356ad)
> 
> This won't be enough to fully resolve the issue, but if the LoongArch errors 
> go away with this patch, then it gives me a strong theory about what is still 
> going wrong and I can likely find a workaround for the rest as well.

Sorry for the delay, here are the results of running the branch 
shard-loongarch. Main takeaway is that the loongarch failures seem to have been 
fixed with the patch at the HEAD of that branch.

I have uploaded the test logfile here (another 300MB file when expanded):
https://www.dropbox.com/scl/fi/8v8toveo8h45c4iuwjgo8/log_test.f6426abc224b140bfb3b6fdc45443bc4e16057aa.zip?rlkey=6rj3vygil705ddd63vk4ykqdl&dl=0

https://github.com/llvm/llvm-project/pull/120534
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [webkit.UncountedLambdaCapturesChecker] Fix a nullptr deference. (PR #120702)

[Ryosuke Niwa via cfe-commits]

https://github.com/rniwa created 
https://github.com/llvm/llvm-project/pull/120702

Added a nullptr check.

>From 527fd23a6bd58cada036fe4a47c13ca93eabf507 Mon Sep 17 00:00:00 2001
From: Ryosuke Niwa 
Date: Fri, 20 Dec 2024 01:32:40 -0800
Subject: [PATCH] [webkit.UncountedLambdaCapturesChecker] Fix a nullptr
 deference.

Added a nullptr check.
---
 .../WebKit/UncountedLambdaCapturesChecker.cpp|  2 ++
 ...counted-lambda-captures-find-lambda-crash.cpp | 16 
 2 files changed, 18 insertions(+)
 create mode 100644 
clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp

diff --git 
a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp 
b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
index da9698e327562e..a57499d52acd0c 100644
--- 
a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
+++ 
b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
@@ -155,6 +155,8 @@ class UncountedLambdaCapturesChecker
 if (!Init)
   return nullptr;
 TempExpr = dyn_cast(Init->IgnoreParenCasts());
+if (!TempExpr)
+  return nullptr;
 return dyn_cast_or_null(TempExpr->getSubExpr());
   }
 
diff --git 
a/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
 
b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
new file mode 100644
index 00..4d9edb75b7ff36
--- /dev/null
+++ 
b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
@@ -0,0 +1,16 @@
+// RUN: %clang_analyze_cc1 
-analyzer-checker=webkit.UncountedLambdaCapturesChecker -verify %s
+// expected-no-diagnostics
+
+struct Foo {
+  int x;
+  int y;
+  Foo(int x, int y) : x(x) , y(y) { }
+  ~Foo() { }
+};
+
+Foo bar(const Foo&);
+void foo() {
+  int x = 7;
+  int y = 5;
+  bar(Foo(x, y));
+}

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy] Mention std::forward_list in container-size-empty doc (PR #120701)

[Niels Dekker via cfe-commits]

https://github.com/N-Dekker created 
https://github.com/llvm/llvm-project/pull/120701

Mentioned `std::forward_list` as example of a container without `size()`.

>From e58124c491f46238538bb06f14de31f4d5f25d2a Mon Sep 17 00:00:00 2001
From: Niels Dekker 
Date: Fri, 20 Dec 2024 10:30:54 +0100
Subject: [PATCH] [clang-tidy] Mention std::forward_list in
 container-size-empty doc

Mentioned `std::forward_list` as example of a container without `size()`.
---
 .../clang-tidy/readability/ContainerSizeEmptyCheck.h  | 8 
 .../checks/readability/container-size-empty.rst   | 7 ---
 2 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h 
b/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
index 3aa4bdc496194b..e449686f77566d 100644
--- a/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
+++ b/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
@@ -19,10 +19,10 @@ namespace clang::tidy::readability {
 ///
 /// The emptiness of a container should be checked using the `empty()` method
 /// instead of the `size()`/`length()` method. It shows clearer intent to use
-/// `empty()`. Furthermore some containers may implement the `empty()` method
-/// but not implement the `size()` or `length()` method. Using `empty()`
-/// whenever possible makes it easier to switch to another container in the
-/// future.
+/// `empty()`. Furthermore some containers (for example, a `std::forward_list`)
+/// may implement the `empty()` method but not implement the `size()` or
+/// `length()` method. Using `empty()` whenever possible makes it easier to
+/// switch to another container in the future.
 class ContainerSizeEmptyCheck : public ClangTidyCheck {
 public:
   ContainerSizeEmptyCheck(StringRef Name, ClangTidyContext *Context);
diff --git 
a/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst 
b/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
index 6a007f69767abe..43ad74f60dbe57 100644
--- 
a/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
+++ 
b/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
@@ -9,9 +9,10 @@ with a call to ``empty()``.
 
 The emptiness of a container should be checked using the ``empty()`` method
 instead of the ``size()``/``length()`` method. It shows clearer intent to use
-``empty()``. Furthermore some containers may implement the ``empty()`` method
-but not implement the ``size()`` or ``length()`` method. Using ``empty()``
-whenever possible makes it easier to switch to another container in the future.
+``empty()``. Furthermore some containers (for example, a ``std::forward_list``)
+may implement the ``empty()`` method but not implement the ``size()`` or
+``length()`` method. Using ``empty()`` whenever possible makes it easier to
+switch to another container in the future.
 
 The check issues warning if a container has ``empty()`` and ``size()`` or
 ``length()`` methods matching following signatures:

___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][CodeGen][SPIRV] Translate `amdgpu_flat_work_group_size` into `reqd_work_group_size`. (PR #116820)

[Alex Voicu via cfe-commits]

@@ -245,6 +247,41 @@ 
SPIRVTargetCodeGenInfo::getGlobalVarAddressSpace(CodeGenModule &CGM,
   return DefaultGlobalAS;
 }
 
+void SPIRVTargetCodeGenInfo::setTargetAttributes(
+const Decl *D, llvm::GlobalValue *GV, CodeGen::CodeGenModule &M) const {
+  if (!M.getLangOpts().HIP ||
+  M.getTarget().getTriple().getVendor() != llvm::Triple::AMD)
+return;
+  if (GV->isDeclaration())
+return;
+
+  auto F = dyn_cast(GV);
+  if (!F)
+return;
+
+  auto FD = dyn_cast_or_null(D);
+  if (!FD)
+return;
+  if (!FD->hasAttr())
+return;
+
+  unsigned N = M.getLangOpts().GPUMaxThreadsPerBlock;
+  if (auto FlatWGS = FD->getAttr())
+N = FlatWGS->getMax()->EvaluateKnownConstInt(M.getContext()).getExtValue();
+
+  // We encode the maximum flat WG size in the first component of the 3D
+  // max_work_group_size attribute, which will get reverse translated into the
+  // original AMDGPU attribute when targeting AMDGPU.

AlexVlx wrote:

We are talking across eachother. I am saying that the SPIR-V attribute cannot 
be generated via Clang, i.e. that you cannot write `__attribute__((foo))` in 
your source and obtain `max_work_group_size` metadata, at the moment. 
Furthermore, from the implementation of Clang's `__launch_bounds__`:
 
```cpp
// An AST node is created for this attribute, but is not used by other parts
// of the compiler. However, this node needs to exist in the AST because
// non-LLVM backends may be relying on the attribute's presence.
```

So this is a glorified annotation / we'd still have to decide on how to lower 
it into IR, which would likely end up atop flat workgroup size, unless we 
choose to spam yet another attribute. We also use flat workgroup size 
implicitly to control / implement `--gpu-max-threads-per-block`, which is 
important for correctness, and is in a fairly similar place with 
`__launch_bounds__` (it's always 1D, doesn't have a minimum etc.). It's also 
not handled by this patch, so I'll have to add it:)

That being said, the idea in #91468 is sound, but it will require a bit of work 
to get done; I think we'd still have to choose a way to pass the info through 
SPIR-V (what this PR tries to do).

https://github.com/llvm/llvm-project/pull/116820
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [webkit.UncountedLambdaCapturesChecker] Fix a nullptr deference. (PR #120702)

[via cfe-commits]

llvmbot wrote:



@llvm/pr-subscribers-clang

@llvm/pr-subscribers-clang-static-analyzer-1

Author: Ryosuke Niwa (rniwa)


Changes

Added a nullptr check.

---
Full diff: https://github.com/llvm/llvm-project/pull/120702.diff


2 Files Affected:

- (modified) 
clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp 
(+2) 
- (added) 
clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
 (+16) 


``diff
diff --git 
a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp 
b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
index da9698e327562e..a57499d52acd0c 100644
--- 
a/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
+++ 
b/clang/lib/StaticAnalyzer/Checkers/WebKit/UncountedLambdaCapturesChecker.cpp
@@ -155,6 +155,8 @@ class UncountedLambdaCapturesChecker
 if (!Init)
   return nullptr;
 TempExpr = dyn_cast(Init->IgnoreParenCasts());
+if (!TempExpr)
+  return nullptr;
 return dyn_cast_or_null(TempExpr->getSubExpr());
   }
 
diff --git 
a/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
 
b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
new file mode 100644
index 00..4d9edb75b7ff36
--- /dev/null
+++ 
b/clang/test/Analysis/Checkers/WebKit/uncounted-lambda-captures-find-lambda-crash.cpp
@@ -0,0 +1,16 @@
+// RUN: %clang_analyze_cc1 
-analyzer-checker=webkit.UncountedLambdaCapturesChecker -verify %s
+// expected-no-diagnostics
+
+struct Foo {
+  int x;
+  int y;
+  Foo(int x, int y) : x(x) , y(y) { }
+  ~Foo() { }
+};
+
+Foo bar(const Foo&);
+void foo() {
+  int x = 7;
+  int y = 5;
+  bar(Foo(x, y));
+}

``




https://github.com/llvm/llvm-project/pull/120702
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy] Mention std::forward_list in container-size-empty doc (PR #120701)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-tidy

Author: Niels Dekker (N-Dekker)


Changes

Mentioned `std::forward_list` as example of a container without `size()`.

---
Full diff: https://github.com/llvm/llvm-project/pull/120701.diff


2 Files Affected:

- (modified) clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h 
(+4-4) 
- (modified) 
clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst 
(+4-3) 


``diff
diff --git a/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h 
b/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
index 3aa4bdc496194b..e449686f77566d 100644
--- a/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
+++ b/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
@@ -19,10 +19,10 @@ namespace clang::tidy::readability {
 ///
 /// The emptiness of a container should be checked using the `empty()` method
 /// instead of the `size()`/`length()` method. It shows clearer intent to use
-/// `empty()`. Furthermore some containers may implement the `empty()` method
-/// but not implement the `size()` or `length()` method. Using `empty()`
-/// whenever possible makes it easier to switch to another container in the
-/// future.
+/// `empty()`. Furthermore some containers (for example, a `std::forward_list`)
+/// may implement the `empty()` method but not implement the `size()` or
+/// `length()` method. Using `empty()` whenever possible makes it easier to
+/// switch to another container in the future.
 class ContainerSizeEmptyCheck : public ClangTidyCheck {
 public:
   ContainerSizeEmptyCheck(StringRef Name, ClangTidyContext *Context);
diff --git 
a/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst 
b/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
index 6a007f69767abe..43ad74f60dbe57 100644
--- 
a/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
+++ 
b/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
@@ -9,9 +9,10 @@ with a call to ``empty()``.
 
 The emptiness of a container should be checked using the ``empty()`` method
 instead of the ``size()``/``length()`` method. It shows clearer intent to use
-``empty()``. Furthermore some containers may implement the ``empty()`` method
-but not implement the ``size()`` or ``length()`` method. Using ``empty()``
-whenever possible makes it easier to switch to another container in the future.
+``empty()``. Furthermore some containers (for example, a ``std::forward_list``)
+may implement the ``empty()`` method but not implement the ``size()`` or
+``length()`` method. Using ``empty()`` whenever possible makes it easier to
+switch to another container in the future.
 
 The check issues warning if a container has ``empty()`` and ``size()`` or
 ``length()`` methods matching following signatures:

``




https://github.com/llvm/llvm-project/pull/120701
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy] Mention std::forward_list in container-size-empty doc (PR #120701)

[via cfe-commits]

llvmbot wrote:




@llvm/pr-subscribers-clang-tools-extra

Author: Niels Dekker (N-Dekker)


Changes

Mentioned `std::forward_list` as example of a container without `size()`.

---
Full diff: https://github.com/llvm/llvm-project/pull/120701.diff


2 Files Affected:

- (modified) clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h 
(+4-4) 
- (modified) 
clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst 
(+4-3) 


``diff
diff --git a/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h 
b/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
index 3aa4bdc496194b..e449686f77566d 100644
--- a/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
+++ b/clang-tools-extra/clang-tidy/readability/ContainerSizeEmptyCheck.h
@@ -19,10 +19,10 @@ namespace clang::tidy::readability {
 ///
 /// The emptiness of a container should be checked using the `empty()` method
 /// instead of the `size()`/`length()` method. It shows clearer intent to use
-/// `empty()`. Furthermore some containers may implement the `empty()` method
-/// but not implement the `size()` or `length()` method. Using `empty()`
-/// whenever possible makes it easier to switch to another container in the
-/// future.
+/// `empty()`. Furthermore some containers (for example, a `std::forward_list`)
+/// may implement the `empty()` method but not implement the `size()` or
+/// `length()` method. Using `empty()` whenever possible makes it easier to
+/// switch to another container in the future.
 class ContainerSizeEmptyCheck : public ClangTidyCheck {
 public:
   ContainerSizeEmptyCheck(StringRef Name, ClangTidyContext *Context);
diff --git 
a/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst 
b/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
index 6a007f69767abe..43ad74f60dbe57 100644
--- 
a/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
+++ 
b/clang-tools-extra/docs/clang-tidy/checks/readability/container-size-empty.rst
@@ -9,9 +9,10 @@ with a call to ``empty()``.
 
 The emptiness of a container should be checked using the ``empty()`` method
 instead of the ``size()``/``length()`` method. It shows clearer intent to use
-``empty()``. Furthermore some containers may implement the ``empty()`` method
-but not implement the ``size()`` or ``length()`` method. Using ``empty()``
-whenever possible makes it easier to switch to another container in the future.
+``empty()``. Furthermore some containers (for example, a ``std::forward_list``)
+may implement the ``empty()`` method but not implement the ``size()`` or
+``length()`` method. Using ``empty()`` whenever possible makes it easier to
+switch to another container in the future.
 
 The check issues warning if a container has ``empty()`` and ``size()`` or
 ``length()`` methods matching following signatures:

``




https://github.com/llvm/llvm-project/pull/120701
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] Patch series to reapply #118734 and substantially improve it (PR #120534)

[Chandler Carruth via cfe-commits]

chandlerc wrote:

No worries about delay, this gives me a credible target to resolve the rest of 
the issues. I'll update this PR both to address review comments but also to try 
and address the rest of the failures. Appreciate runs to validate these 
updates. =]

https://github.com/llvm/llvm-project/pull/120534
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang-tools-extra] [clang-tidy] Mention std::forward_list in container-size-empty doc (PR #120701)

[Niels Dekker via cfe-commits]

https://github.com/N-Dekker ready_for_review 
https://github.com/llvm/llvm-project/pull/120701
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Fix non-deterministic infinite recursion... (PR #118288)

[Nikita Popov via cfe-commits]

Alejandro =?utf-8?q?Álvarez_Ayllón?Message-ID:
In-Reply-To: 


nikic wrote:

@hubert-reinterpretcast This patch modifies structure layout in exported 
headers, changing the libclang-cpp ABI. LLVM patch releases cannot break API or 
ABI compatibility, see the last point in 
https://llvm.org/docs/HowToReleaseLLVM.html#release-patch-rules.

https://github.com/llvm/llvm-project/pull/118288
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Fix dangling false positives for conditional operators. (PR #120233)

[Haojian Wu via cfe-commits]

https://github.com/hokein edited 
https://github.com/llvm/llvm-project/pull/120233
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] a6d26c5 - [clang] Fix dangling false positives for conditional operators. (#120233)

[via cfe-commits]

Author: Haojian Wu
Date: 2024-12-20T09:26:38+01:00
New Revision: a6d26c56ff066c8e8f92f4ca169fcf40ae0db537

URL: 
https://github.com/llvm/llvm-project/commit/a6d26c56ff066c8e8f92f4ca169fcf40ae0db537
DIFF: 
https://github.com/llvm/llvm-project/commit/a6d26c56ff066c8e8f92f4ca169fcf40ae0db537.diff

LOG: [clang] Fix dangling false positives for conditional operators. (#120233)

When analyzing a dangling gsl pointer, we currently filter out all field
access `MemberExpr` to avoid common false positives (`string_view sv =
Temp().sv`), However, this filter only applies to direct MemberExpr
instances, leaving the conditional operator as an escaping example
(`GSLPointer pointer(Cond ? Owner().ptr : GSLPointer());`).

This patch extends the MemberExpr logic to handle the conditional
operator. The heuristic is intentionally simple, which may result in
some false negatives. However, it effectively covers common cases like
`std::string_view sv = cond ? "123" : std::string();`, which is a
reasonable trade-off.

Fixes https://github.com/llvm/llvm-project/issues/120206

Added: 


Modified: 
clang/docs/ReleaseNotes.rst
clang/lib/Sema/CheckExprLifetime.cpp
clang/test/Sema/warn-lifetime-analysis-nocfg.cpp

Removed: 




diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index b8d92a6c881c68..a85ef60b7b58ba 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -706,6 +706,8 @@ Improvements to Clang's diagnostics
   return ptr + index < ptr; // warning
 }
 
+- Fix -Wdangling false positives on conditional operators (#120206).
+
 Improvements to Clang's time-trace
 --
 

diff  --git a/clang/lib/Sema/CheckExprLifetime.cpp 
b/clang/lib/Sema/CheckExprLifetime.cpp
index add6d7506bd6f0..7109de03cadd12 100644
--- a/clang/lib/Sema/CheckExprLifetime.cpp
+++ b/clang/lib/Sema/CheckExprLifetime.cpp
@@ -582,6 +582,15 @@ static void visitFunctionCallArguments(IndirectLocalPath 
&Path, Expr *Call,
 //   Temp().ptr; // Here ptr might not dangle.
 if (isa(Arg->IgnoreImpCasts()))
   return;
+// Avoid false positives when the object is constructed from a conditional
+// operator argument. A common case is:
+//   // 'ptr' might not be owned by the Owner object.
+//   std::string_view s = cond() ? Owner().ptr : sv;
+if (const auto *Cond =
+dyn_cast(Arg->IgnoreImpCasts());
+Cond && isPointerLikeType(Cond->getType()))
+  return;
+
 auto ReturnType = Callee->getReturnType();
 
 // Once we initialized a value with a non gsl-owner reference, it can no

diff  --git a/clang/test/Sema/warn-lifetime-analysis-nocfg.cpp 
b/clang/test/Sema/warn-lifetime-analysis-nocfg.cpp
index 45b4dc838f44ed..4c19367bb7f3dd 100644
--- a/clang/test/Sema/warn-lifetime-analysis-nocfg.cpp
+++ b/clang/test/Sema/warn-lifetime-analysis-nocfg.cpp
@@ -777,3 +777,32 @@ void test4() {
 }
 
 } // namespace LifetimeboundInterleave
+
+namespace GH120206 {
+struct S {
+  std::string_view s;
+};
+
+struct [[gsl::Owner]] Q1 {
+  const S* get() const [[clang::lifetimebound]];
+};
+std::string_view test1(int c, std::string_view sv) {
+  std::string_view k = c > 1 ? Q1().get()->s : sv;
+  if (c == 1)
+return  c > 1 ? Q1().get()->s : sv;
+  Q1 q;
+  return c > 1 ? q.get()->s : sv;
+}
+
+struct Q2 {
+  const S* get() const [[clang::lifetimebound]];
+};
+std::string_view test2(int c, std::string_view sv) {
+  std::string_view k = c > 1 ? Q2().get()->s : sv;
+  if (c == 1)
+return c > 1 ? Q2().get()->s : sv;
+  Q2 q;
+  return c > 1 ? q.get()->s : sv;
+}
+
+} // namespace GH120206



___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Fix dangling false positives for conditional operators. (PR #120233)

[Haojian Wu via cfe-commits]

https://github.com/hokein closed 
https://github.com/llvm/llvm-project/pull/120233
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang] Fix dangling false positives for conditional operators. (PR #120233)

[Haojian Wu via cfe-commits]

hokein wrote:

> Please add more details to the summary, especially a note that there 
> theoretically could be false negatives but it fixes the overall issue and the 
> tradeoff good.

Done.

https://github.com/llvm/llvm-project/pull/120233
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [clang][Sema] Fixes for %b printf extension handling (PR #120689)

[Brad Smith via cfe-commits]

https://github.com/brad0 updated 
https://github.com/llvm/llvm-project/pull/120689

>From 3c205ebace308c6d37dff289fab38f93bcb76309 Mon Sep 17 00:00:00 2001
From: Brad Smith 
Date: Thu, 19 Dec 2024 21:35:57 -0500
Subject: [PATCH] [clang][Sema] Fixes for %b printf extension handling

The %b printf extension in the kernel is not fixed to a int type. On
sparc64 there are various %llb formats. Adjust the code to handle the
length specifiers and type check like it is used by the regular case.
---
 clang/include/clang/AST/FormatString.h   |  6 ++--
 clang/lib/AST/FormatString.cpp   |  9 ++
 clang/lib/Sema/SemaChecking.cpp  | 38 +---
 clang/test/Sema/format-strings-freebsd.c |  8 -
 4 files changed, 47 insertions(+), 14 deletions(-)

diff --git a/clang/include/clang/AST/FormatString.h 
b/clang/include/clang/AST/FormatString.h
index a074dd23e2ad4c..1ff0d036b70d1a 100644
--- a/clang/include/clang/AST/FormatString.h
+++ b/clang/include/clang/AST/FormatString.h
@@ -240,8 +240,10 @@ class ConversionSpecifier {
 
   bool isIntArg() const { return (kind >= IntArgBeg && kind <= IntArgEnd) ||
 kind == FreeBSDrArg || kind == FreeBSDyArg; }
-  bool isUIntArg() const { return kind >= UIntArgBeg && kind <= UIntArgEnd; }
-  bool isAnyIntArg() const { return kind >= IntArgBeg && kind <= UIntArgEnd; }
+  bool isUIntArg() const { return (kind >= UIntArgBeg && kind <= UIntArgEnd) ||
+kind == FreeBSDbArg; }
+  bool isAnyIntArg() const { return (kind >= IntArgBeg && kind <= UIntArgEnd) 
||
+kind == FreeBSDbArg; }
   bool isDoubleArg() const {
 return kind >= DoubleArgBeg && kind <= DoubleArgEnd;
   }
diff --git a/clang/lib/AST/FormatString.cpp b/clang/lib/AST/FormatString.cpp
index e892c1592df986..6ccc0c91c025f6 100644
--- a/clang/lib/AST/FormatString.cpp
+++ b/clang/lib/AST/FormatString.cpp
@@ -879,6 +879,10 @@ bool FormatSpecifier::hasValidLengthModifier(const 
TargetInfo &Target,
 case ConversionSpecifier::XArg:
 case ConversionSpecifier::nArg:
   return true;
+case ConversionSpecifier::FreeBSDbArg:
+  return Target.getTriple().isOSFreeBSD() ||
+ Target.getTriple().isPS() ||
+ Target.getTriple().isOSOpenBSD();
 case ConversionSpecifier::FreeBSDrArg:
 case ConversionSpecifier::FreeBSDyArg:
   return Target.getTriple().isOSFreeBSD() || Target.getTriple().isPS();
@@ -919,6 +923,10 @@ bool FormatSpecifier::hasValidLengthModifier(const 
TargetInfo &Target,
 case ConversionSpecifier::ScanListArg:
 case ConversionSpecifier::ZArg:
   return true;
+case ConversionSecifier::FreeBSDbArg:
+  return Target.getTriple().isOSFreeBSD() ||
+ Target.getTriple().isPS() ||
+ Target.getTriple().isOSOpenBSD();
 case ConversionSpecifier::FreeBSDrArg:
 case ConversionSpecifier::FreeBSDyArg:
   return Target.getTriple().isOSFreeBSD() || Target.getTriple().isPS();
@@ -1085,6 +1093,7 @@ bool 
FormatSpecifier::hasStandardLengthConversionCombination() const {
 case ConversionSpecifier::uArg:
 case ConversionSpecifier::xArg:
 case ConversionSpecifier::XArg:
+case ConversionSpecifier::FreeBSDbArg:
   return false;
 default:
   return true;
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index e703a62ff9cf18..e50118762fff32 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -7170,18 +7170,34 @@ bool CheckPrintfHandler::HandlePrintfSpecifier(
 // Claim the second argument.
 CoveredArgs.set(argIndex + 1);
 
-// Type check the first argument (int for %b, pointer for %D)
 const Expr *Ex = getDataArg(argIndex);
-const analyze_printf::ArgType &AT =
-  (CS.getKind() == ConversionSpecifier::FreeBSDbArg) ?
-ArgType(S.Context.IntTy) : ArgType::CPointerTy;
-if (AT.isValid() && !AT.matchesType(S.Context, Ex->getType()))
-  EmitFormatDiagnostic(
-  S.PDiag(diag::warn_format_conversion_argument_type_mismatch)
-  << AT.getRepresentativeTypeName(S.Context) << Ex->getType()
-  << false << Ex->getSourceRange(),
-  Ex->getBeginLoc(), /*IsStringLocation*/ false,
-  getSpecifierRange(startSpecifier, specifierLen));
+if (CS.getKind() == ConversionSpecifier::FreeBSDDArg) {
+  // Type check the first argument (pointer for %D)
+  const analyze_printf::ArgType &AT = ArgType::CPointerTy;
+  if (AT.isValid() && !AT.matchesType(S.Context, Ex->getType()))
+EmitFormatDiagnostic(
+S.PDiag(diag::warn_format_conversion_argument_type_mismatch)
+<< AT.getRepresentativeTypeName(S.Context) << Ex->getType()
+<< false << Ex->getSourceRange(),
+Ex->getBeginLoc(), /*IsStringLocation*/ false,
+getSpecifierRange(startSpecifier, specifierLen));
+} else {
+   

[clang] [clang][Sema] Fixes for %b printf extension handling (PR #120689)

[Brad Smith via cfe-commits]

https://github.com/brad0 updated 
https://github.com/llvm/llvm-project/pull/120689

>From 67a87a492c0b84f4d76ae4818658969cd11a379e Mon Sep 17 00:00:00 2001
From: Brad Smith 
Date: Thu, 19 Dec 2024 21:35:57 -0500
Subject: [PATCH] [clang][Sema] Fixes for %b printf extension handling

The %b printf extension in the kernel is not fixed to a int type. On
sparc64 there are various %llb formats. Adjust the code to handle the
length specifiers and type check like it is used by the regular case.
---
 clang/include/clang/AST/FormatString.h   |  6 ++--
 clang/lib/AST/FormatString.cpp   |  9 ++
 clang/lib/Sema/SemaChecking.cpp  | 38 +---
 clang/test/Sema/format-strings-freebsd.c |  8 -
 4 files changed, 47 insertions(+), 14 deletions(-)

diff --git a/clang/include/clang/AST/FormatString.h 
b/clang/include/clang/AST/FormatString.h
index a074dd23e2ad4c..1ff0d036b70d1a 100644
--- a/clang/include/clang/AST/FormatString.h
+++ b/clang/include/clang/AST/FormatString.h
@@ -240,8 +240,10 @@ class ConversionSpecifier {
 
   bool isIntArg() const { return (kind >= IntArgBeg && kind <= IntArgEnd) ||
 kind == FreeBSDrArg || kind == FreeBSDyArg; }
-  bool isUIntArg() const { return kind >= UIntArgBeg && kind <= UIntArgEnd; }
-  bool isAnyIntArg() const { return kind >= IntArgBeg && kind <= UIntArgEnd; }
+  bool isUIntArg() const { return (kind >= UIntArgBeg && kind <= UIntArgEnd) ||
+kind == FreeBSDbArg; }
+  bool isAnyIntArg() const { return (kind >= IntArgBeg && kind <= UIntArgEnd) 
||
+kind == FreeBSDbArg; }
   bool isDoubleArg() const {
 return kind >= DoubleArgBeg && kind <= DoubleArgEnd;
   }
diff --git a/clang/lib/AST/FormatString.cpp b/clang/lib/AST/FormatString.cpp
index e892c1592df986..422fee7ab5d857 100644
--- a/clang/lib/AST/FormatString.cpp
+++ b/clang/lib/AST/FormatString.cpp
@@ -879,6 +879,10 @@ bool FormatSpecifier::hasValidLengthModifier(const 
TargetInfo &Target,
 case ConversionSpecifier::XArg:
 case ConversionSpecifier::nArg:
   return true;
+case ConversionSpecifier::FreeBSDbArg:
+  return Target.getTriple().isOSFreeBSD() ||
+ Target.getTriple().isPS() ||
+ Target.getTriple().isOSOpenBSD();
 case ConversionSpecifier::FreeBSDrArg:
 case ConversionSpecifier::FreeBSDyArg:
   return Target.getTriple().isOSFreeBSD() || Target.getTriple().isPS();
@@ -919,6 +923,10 @@ bool FormatSpecifier::hasValidLengthModifier(const 
TargetInfo &Target,
 case ConversionSpecifier::ScanListArg:
 case ConversionSpecifier::ZArg:
   return true;
+case ConversionSpecifier::FreeBSDbArg:
+  return Target.getTriple().isOSFreeBSD() ||
+ Target.getTriple().isPS() ||
+ Target.getTriple().isOSOpenBSD();
 case ConversionSpecifier::FreeBSDrArg:
 case ConversionSpecifier::FreeBSDyArg:
   return Target.getTriple().isOSFreeBSD() || Target.getTriple().isPS();
@@ -1085,6 +1093,7 @@ bool 
FormatSpecifier::hasStandardLengthConversionCombination() const {
 case ConversionSpecifier::uArg:
 case ConversionSpecifier::xArg:
 case ConversionSpecifier::XArg:
+case ConversionSpecifier::FreeBSDbArg:
   return false;
 default:
   return true;
diff --git a/clang/lib/Sema/SemaChecking.cpp b/clang/lib/Sema/SemaChecking.cpp
index e703a62ff9cf18..e50118762fff32 100644
--- a/clang/lib/Sema/SemaChecking.cpp
+++ b/clang/lib/Sema/SemaChecking.cpp
@@ -7170,18 +7170,34 @@ bool CheckPrintfHandler::HandlePrintfSpecifier(
 // Claim the second argument.
 CoveredArgs.set(argIndex + 1);
 
-// Type check the first argument (int for %b, pointer for %D)
 const Expr *Ex = getDataArg(argIndex);
-const analyze_printf::ArgType &AT =
-  (CS.getKind() == ConversionSpecifier::FreeBSDbArg) ?
-ArgType(S.Context.IntTy) : ArgType::CPointerTy;
-if (AT.isValid() && !AT.matchesType(S.Context, Ex->getType()))
-  EmitFormatDiagnostic(
-  S.PDiag(diag::warn_format_conversion_argument_type_mismatch)
-  << AT.getRepresentativeTypeName(S.Context) << Ex->getType()
-  << false << Ex->getSourceRange(),
-  Ex->getBeginLoc(), /*IsStringLocation*/ false,
-  getSpecifierRange(startSpecifier, specifierLen));
+if (CS.getKind() == ConversionSpecifier::FreeBSDDArg) {
+  // Type check the first argument (pointer for %D)
+  const analyze_printf::ArgType &AT = ArgType::CPointerTy;
+  if (AT.isValid() && !AT.matchesType(S.Context, Ex->getType()))
+EmitFormatDiagnostic(
+S.PDiag(diag::warn_format_conversion_argument_type_mismatch)
+<< AT.getRepresentativeTypeName(S.Context) << Ex->getType()
+<< false << Ex->getSourceRange(),
+Ex->getBeginLoc(), /*IsStringLocation*/ false,
+getSpecifierRange(startSpecifier, specifierLen));
+} else {
+  

[clang] [llvm] [TargetLibraryInfo] Add libmvec support for risc-v (PR #119844)

[Mark Goncharov via cfe-commits]

@@ -236,6 +236,79 @@ TLI_DEFINE_VECFUNC("llvm.log.f64", "_ZGVdN4v_log", 
FIXED(4), "_ZGV_LLVM_N4v")
 TLI_DEFINE_VECFUNC("llvm.log.f32", "_ZGVbN4v_logf", FIXED(4), "_ZGV_LLVM_N4v")
 TLI_DEFINE_VECFUNC("llvm.log.f32", "_ZGVdN8v_logf", FIXED(8), "_ZGV_LLVM_N8v")
 
+#elif defined(TLI_DEFINE_LIBMVEC_RVV_VECFUNCS)
+// GLIBC Vector math Functions for RISC-V
+
+TLI_DEFINE_VECFUNC("sin", "_ZGV1Nxv_sin", SCALABLE(1), "_ZGVr1Nxv")

mga-sc wrote:

You have missed `r` letter in mangled vector name`after ZGV according to 
[mention 
MR](https://github.com/riscv-non-isa/riscv-elf-psabi-doc/pull/455/files)

https://github.com/llvm/llvm-project/pull/119844
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits