[PATCH] D114505: [clang][unittests]Fix a clang unittest linking issue

[Lu Weining via Phabricator via cfe-commits]

SixWeining created this revision.
SixWeining added reviewers: cfe-commits, dexonsmith.
Herald added a subscriber: mgorny.
SixWeining requested review of this revision.
Herald added a project: clang.

Currently the clang/unittests/Basic/CMakeLists.txt links LLVMTestingSupport in 
an incorrect way that would cause `ninja check-clang` failing with a linking 
error like below:
/usr/bin/ld: 
tools/clang/unittests/Basic/CMakeFiles/BasicTests.dir/FileManagerTest.cpp.o: in 
function `(anonymous 
namespace)::FileManagerTest_getBypassFile_Test::TestBody()':
FileManagerTest.cpp:(.text._ZN12_GLOBAL__N_134FileManagerTest_getBypassFile_Test8TestBodyEv+0x3a3):
 undefined reference to `llvm::detail::TakeError(llvm::Error)'

This patch changes the linking method of LLVMTestingSupport from 
clang_target_link_libraries to target_link_libraries just like other tests do.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114505

Files:
  clang/unittests/Basic/CMakeLists.txt


Index: clang/unittests/Basic/CMakeLists.txt
===
--- clang/unittests/Basic/CMakeLists.txt
+++ clang/unittests/Basic/CMakeLists.txt
@@ -18,5 +18,9 @@
   clangAST
   clangBasic
   clangLex
-  LLVMTestingSupport
   )
+
+target_link_libraries(BasicTests
+  PRIVATE
+  LLVMTestingSupport
+)


Index: clang/unittests/Basic/CMakeLists.txt
===
--- clang/unittests/Basic/CMakeLists.txt
+++ clang/unittests/Basic/CMakeLists.txt
@@ -18,5 +18,9 @@
   clangAST
   clangBasic
   clangLex
-  LLVMTestingSupport
   )
+
+target_link_libraries(BasicTests
+  PRIVATE
+  LLVMTestingSupport
+)
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D114317: [clang-tidy][WIP] Do not run perfect alias checks

[Carlos Galvez via Phabricator via cfe-commits]

carlosgalvezp added a comment.

> it is likely that the options were tuned, but most likely only for a single 
> check.

Yes, if the alias check has a different configuration than the primary check 
then they are considered as "different" checks (as they should be, since they 
produce potentially different outputs and some people might lose coverage). 
Perhaps it would be interesting to add something to the "--explain-config" flag 
so the user understands why some checks are/aren't enabled.

> I still maintain that there should be a single opt-in option to completely 
> disable all aliasee checks.

By "aliasee" do you mean "alias" checks (as opposed to "primary" check)? I'm a 
bit confused by the terminology :) In order words, are you proposing to opt-in 
to disable all alias checks, leaving only the primary check (regardless of 
configuration)? I suppose this should be fairly easy to implement, just don't 
check the configuration when removing aliases. We can support both options for 
people having different use cases.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114317/new/

https://reviews.llvm.org/D114317

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

[clang-tools-extra] ba4411e - [clang-tidy] performance-unnecessary-copy-initialization: Fix false negative.

[Clement Courbet via cfe-commits]

Author: Clement Courbet
Date: 2021-11-24T08:07:21+01:00
New Revision: ba4411e7c6a5879ce8acf246b0cd03ec738d9d6b

URL: 
https://github.com/llvm/llvm-project/commit/ba4411e7c6a5879ce8acf246b0cd03ec738d9d6b
DIFF: 
https://github.com/llvm/llvm-project/commit/ba4411e7c6a5879ce8acf246b0cd03ec738d9d6b.diff

LOG: [clang-tidy] performance-unnecessary-copy-initialization: Fix false 
negative.

`isConstRefReturningMethodCall` should be considering
`CXXOperatorCallExpr` in addition to `CXXMemberCallExpr`. Clang considers
these to be distinct (`CXXOperatorCallExpr` derives from `CallExpr`, not
`CXXMemberCallExpr`), but we don't care in the context of this
check.

This is important because of
`std::vector::operator[](size_t) const`.

Differential Revision: https://reviews.llvm.org/D114249

Added: 


Modified: 
clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp

clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp

clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp

Removed: 




diff  --git 
a/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp 
b/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp
index 514ce6f6e3b87..c1514aed88f70 100644
--- a/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp
+++ b/clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp
@@ -83,13 +83,19 @@ AST_MATCHER_FUNCTION_P(StatementMatcher, 
isConstRefReturningMethodCall,
   // variable being declared. The assumption is that the const reference being
   // returned either points to a global static variable or to a member of the
   // called object.
-  return cxxMemberCallExpr(
-  callee(cxxMethodDecl(
- returns(hasCanonicalType(matchers::isReferenceToConst(
- .bind(MethodDeclId)),
-  on(declRefExpr(to(varDecl().bind(ObjectArgId,
-  thisPointerType(namedDecl(
-  unless(matchers::matchesAnyListedName(ExcludedContainerTypes);
+  const auto MethodDecl =
+  cxxMethodDecl(returns(hasCanonicalType(matchers::isReferenceToConst(
+  .bind(MethodDeclId);
+  const auto ReceiverExpr = declRefExpr(to(varDecl().bind(ObjectArgId)));
+  const auto ReceiverType =
+  hasCanonicalType(recordType(hasDeclaration(namedDecl(
+  unless(matchers::matchesAnyListedName(ExcludedContainerTypes));
+
+  return expr(anyOf(
+  cxxMemberCallExpr(callee(MethodDecl), on(ReceiverExpr),
+thisPointerType(ReceiverType)),
+  cxxOperatorCallExpr(callee(MethodDecl), hasArgument(0, ReceiverExpr),
+  hasArgument(0, hasType(ReceiverType);
 }
 
 AST_MATCHER_FUNCTION(StatementMatcher, isConstRefReturningFunctionCall) {

diff  --git 
a/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp
 
b/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp
index 88b850fe2ff82..96c7ca8a460c2 100644
--- 
a/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp
+++ 
b/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp
@@ -21,6 +21,7 @@ struct ExpensiveToCopy {
 
 struct ConstInCorrectType {
   const ExpensiveToCopy () const;
+  const ExpensiveToCopy [](int) const;
 };
 
 using NSVTE = ns::ViewType;
@@ -59,12 +60,28 @@ void excludedConstIncorrectType() {
   E.constMethod();
 }
 
+void excludedConstIncorrectTypeOperator() {
+  ConstInCorrectType C;
+  const auto E = C[42];
+  E.constMethod();
+}
+
 void excludedConstIncorrectTypeAsPointer(ConstInCorrectType *C) {
   const auto E = C->secretlyMutates();
   E.constMethod();
 }
 
+void excludedConstIncorrectTypeAsPointerOperator(ConstInCorrectType *C) {
+  const auto E = (*C)[42];
+  E.constMethod();
+}
+
 void excludedConstIncorrectTypeAsReference(const ConstInCorrectType ) {
   const auto E = C.secretlyMutates();
   E.constMethod();
 }
+
+void excludedConstIncorrectTypeAsReferenceOperator(const ConstInCorrectType 
) {
+  const auto E = C[42];
+  E.constMethod();
+}

diff  --git 
a/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
 
b/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
index 4c469c966860b..5eac571c2afa7 100644
--- 
a/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
+++ 
b/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
@@ -3,11 +3,19 @@
 template 
 struct Iterator {
   void operator++();
-  const T *() const;
+  T *() const;
   bool operator!=(const Iterator &) const;
   

[PATCH] D114249: [clang-tidy] performance-unnecessary-copy-initialization: Fix false negative.

[Clement Courbet via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rGba4411e7c6a5: [clang-tidy] 
performance-unnecessary-copy-initialization: Fix false negative. (authored by 
courbet).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114249/new/

https://reviews.llvm.org/D114249

Files:
  clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp
  
clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp
  
clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp

Index: clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
===
--- clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
+++ clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
@@ -3,11 +3,19 @@
 template 
 struct Iterator {
   void operator++();
-  const T *() const;
+  T *() const;
   bool operator!=(const Iterator &) const;
   typedef const T _reference;
 };
 
+template 
+struct ConstIterator {
+  void operator++();
+  const T *() const;
+  bool operator!=(const ConstIterator &) const;
+  typedef const T _reference;
+};
+
 struct ExpensiveToCopyType {
   ExpensiveToCopyType();
   virtual ~ExpensiveToCopyType();
@@ -15,8 +23,6 @@
   using ConstRef = const ExpensiveToCopyType &;
   ConstRef referenceWithAlias() const;
   const ExpensiveToCopyType *pointer() const;
-  Iterator begin() const;
-  Iterator end() const;
   void nonConstMethod();
   bool constMethod() const;
   template 
@@ -24,6 +30,21 @@
   operator int() const; // Implicit conversion to int.
 };
 
+template 
+struct Container {
+  bool empty() const;
+  const T& operator[](int) const;
+  const T& operator[](int);
+  Iterator begin();
+  Iterator end();
+  ConstIterator begin() const;
+  ConstIterator end() const;
+  void nonConstMethod();
+  bool constMethod() const;
+};
+
+using ExpensiveToCopyContainerAlias = Container;
+
 struct TrivialToCopyType {
   const TrivialToCopyType () const;
 };
@@ -138,6 +159,94 @@
   VarCopyConstructed.constMethod();
 }
 
+void PositiveOperatorCallConstReferenceParam(const Container ) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarCopyConstructed'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarCopyConstructed(C[42]);
+  VarCopyConstructed.constMethod();
+}
+
+void PositiveOperatorCallConstValueParam(const Container C) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarCopyConstructed'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarCopyConstructed(C[42]);
+  VarCopyConstructed.constMethod();
+}
+
+void PositiveOperatorCallConstValueParamAlias(const ExpensiveToCopyContainerAlias C) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+ 

[PATCH] D114058: [clangd] Add ObjC method support to prepareCallHierarchy

[Kadir Cetinkaya via Phabricator via cfe-commits]

kadircet accepted this revision.
kadircet added a comment.
This revision is now accepted and ready to land.

thanks, lgtm! let me know of your email address (for commit attribution) if you 
want me to land this for you.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114058/new/

https://reviews.llvm.org/D114058

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

[PATCH] D114504: [clang][DebugInfo] Debug support for private variables inside an OpenMP task construct

[Alok Kumar Sharma via Phabricator via cfe-commits]

alok created this revision.
alok added reviewers: aprantl, djtodoro.
alok added a project: debug-info.
Herald added subscribers: pengfei, guansong, yaxunl.
alok requested review of this revision.
Herald added a reviewer: jdoerfert.
Herald added subscribers: cfe-commits, sstefan1.
Herald added a project: clang.

Currently variables appearing inside private/firstprivate/lastprivate
clause of openmp task construct are not visible inside lldb debugger.
This is because compiler does not generate debug info for it.

Please consider the testcase debug_private.c attached with patch.

  ```

(lldb) b 34
Breakpoint 1: where = a.out`.omp_task_entry. + 243 [inlined] .omp_outlined. + 
157 at debug_private.c:28, address = 0x004008f3
(lldb) r
Process 18475 launched: '/tmp/a.out' (x86_64)
Task n=10,priv1=10,priv2=12,fpriv=14
Process 18475 stopped

- thread #1, name = 'a.out', stop reason = breakpoint 1.1 frame #0: 
0x004008f3 a.out`.omp_task_entry. [inlined] 
.omp_outlined.(.global_tid.=0, .part_id.=0x0070d3d0, 
.privates.=0x0070d3e8, .copy_fn.=(a.out`.omp_task_privates_map. at 
debug_private.c:28), .task_t.=0x0070d3c0, __context=0x0070d3f8) 
at debug_private.c:34:13 31 priv2 = n + 2; 32 printf("Task 
n=%d,priv1=%d,priv2=%d,fpriv=%d\n",n,priv1,priv2,fpriv); 33

-> 34 res = priv1 + priv2 + fpriv + foo(n - 1);

  35   }
  36   #pragma omp taskwait
  37   return res;

(lldb) p priv1
error: :1:1: use of undeclared identifier 'priv1'
priv1
^
(lldb) p priv2
error: :1:1: use of undeclared identifier 'priv2'
priv2
^
(lldb) p fpriv
error: :1:1: use of undeclared identifier 'fpriv'
fpriv
^

  

After the current patch, lldb is able to show the variables

  

(lldb) p priv1
(int) $0 = 10
(lldb) p priv2
(int) $1 = 12
(lldb) p fpriv
(int) $2 = 14

  ``


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114504

Files:
  clang/lib/CodeGen/CGStmtOpenMP.cpp
  clang/test/OpenMP/debug_private.c


Index: clang/test/OpenMP/debug_private.c
===
--- /dev/null
+++ clang/test/OpenMP/debug_private.c
@@ -0,0 +1,45 @@
+// This testcase checks emission of debug info for variables inside
+// private/firstprivate/lastprivate.
+
+// REQUIRES: x86_64-linux
+
+// RUN: %clang_cc1 -debug-info-kind=constructor -x c -verify -triple 
x86_64-pc-linux-gnu -fopenmp -emit-llvm %s -o - | FileCheck %s
+// expected-no-diagnostics
+
+// CHECK: define internal i32 @.omp_task_entry.
+
+// CHECK:  call void @llvm.dbg.declare(metadata i32** %.priv.ptr.addr.i, 
metadata [[PRIV1:![0-9]+]], metadata !DIExpression(DW_OP_deref))
+// CHECK:  call void @llvm.dbg.declare(metadata i32** %.priv.ptr.addr1.i, 
metadata [[PRIV2:![0-9]+]], metadata !DIExpression(DW_OP_deref))
+// CHECK:  call void @llvm.dbg.declare(metadata i32** %.firstpriv.ptr.addr.i, 
metadata [[FPRIV:![0-9]+]], metadata !DIExpression(DW_OP_deref))
+
+// CHECK: [[PRIV1]] = !DILocalVariable(name: "priv1"
+// CHECK: [[PRIV2]] = !DILocalVariable(name: "priv2"
+// CHECK: [[FPRIV]] = !DILocalVariable(name: "fpriv"
+
+extern int printf(const char *, ...);
+
+int foo(int n) {
+  int res, priv1, priv2, fpriv;
+  fpriv = n + 4;
+
+  if (n < 2)
+return n;
+  else {
+#pragma omp task shared(res) private(priv1, priv2) firstprivate(fpriv)
+{
+  priv1 = n;
+  priv2 = n + 2;
+  printf("Task n=%d,priv1=%d,priv2=%d,fpriv=%d\n",n,priv1,priv2,fpriv);
+
+  res = priv1 + priv2 + fpriv + foo(n - 1);
+}
+#pragma omp taskwait
+return res;
+  }
+}
+
+int main() {
+  int n = 10;
+  printf("foo(%d) = %d\n", n, foo(n));
+  return 0;
+}
Index: clang/lib/CodeGen/CGStmtOpenMP.cpp
===
--- clang/lib/CodeGen/CGStmtOpenMP.cpp
+++ clang/lib/CodeGen/CGStmtOpenMP.cpp
@@ -4510,6 +4510,10 @@
 Address Replacement(CGF.Builder.CreateLoad(Pair.second),
 CGF.getContext().getDeclAlign(Pair.first));
 Scope.addPrivate(Pair.first, [Replacement]() { return Replacement; });
+if (auto *DI = CGF.getDebugInfo()) {
+  DI->EmitDeclareOfAutoVariable(Pair.first, Pair.second.getPointer(),
+CGF.Builder, true);
+}
   }
   // Adjust mapping for internal locals by mapping actual memory instead of
   // a pointer to this memory.


Index: clang/test/OpenMP/debug_private.c
===
--- /dev/null
+++ clang/test/OpenMP/debug_private.c
@@ -0,0 +1,45 @@
+// This testcase checks emission of debug info for variables inside
+// private/firstprivate/lastprivate.
+
+// REQUIRES: x86_64-linux
+
+// RUN: %clang_cc1 -debug-info-kind=constructor -x c -verify -triple x86_64-pc-linux-gnu -fopenmp -emit-llvm %s -o - | FileCheck %s
+// expected-no-diagnostics
+
+// CHECK: define internal i32 @.omp_task_entry.
+
+// CHECK:  call void @llvm.dbg.declare(metadata i32** 

[PATCH] D114497: [Driver] Stop adding stdlib paths in -ffreestanding

[Zhihao Yuan via Phabricator via cfe-commits]

lichray updated this revision to Diff 389391.
lichray added a comment.
Herald added subscribers: kbarton, nemanjai.

- Adapt tests instead


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114497/new/

https://reviews.llvm.org/D114497

Files:
  clang/test/CodeGen/ppc-xmmintrin.c


Index: clang/test/CodeGen/ppc-xmmintrin.c
===
--- clang/test/CodeGen/ppc-xmmintrin.c
+++ clang/test/CodeGen/ppc-xmmintrin.c
@@ -10,13 +10,13 @@
 // RUN: %clang -x c++ -fsyntax-only -target powerpc64le-unknown-linux-gnu 
-mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns
 
-// RUN: %clang -S -emit-llvm -target powerpc64-unknown-freebsd13.0 -mcpu=pwr8 
-ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -S -emit-llvm -target powerpc64-unknown-freebsd13.0 -mcpu=pwr8 
-ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | 
llvm-cxxfilt -n | FileCheck %s --check-prefixes=CHECK,CHECK-BE
-// RUN: %clang -x c++ -fsyntax-only -target powerpc64-unknown-freebsd13.0 
-mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -x c++ -fsyntax-only -target powerpc64-unknown-freebsd13.0 
-mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns
-// RUN: %clang -S -emit-llvm -target powerpc64le-unknown-freebsd13.0 
-mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -S -emit-llvm -target powerpc64le-unknown-freebsd13.0 
-mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | 
llvm-cxxfilt -n | FileCheck %s --check-prefixes=CHECK,CHECK-LE
-// RUN: %clang -x c++ -fsyntax-only -target powerpc64le-unknown-freebsd13.0 
-mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -x c++ -fsyntax-only -target powerpc64le-unknown-freebsd13.0 
-mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns
 
 #include 


Index: clang/test/CodeGen/ppc-xmmintrin.c
===
--- clang/test/CodeGen/ppc-xmmintrin.c
+++ clang/test/CodeGen/ppc-xmmintrin.c
@@ -10,13 +10,13 @@
 // RUN: %clang -x c++ -fsyntax-only -target powerpc64le-unknown-linux-gnu -mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns
 
-// RUN: %clang -S -emit-llvm -target powerpc64-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -S -emit-llvm -target powerpc64-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | llvm-cxxfilt -n | FileCheck %s --check-prefixes=CHECK,CHECK-BE
-// RUN: %clang -x c++ -fsyntax-only -target powerpc64-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -x c++ -fsyntax-only -target powerpc64-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns
-// RUN: %clang -S -emit-llvm -target powerpc64le-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -S -emit-llvm -target powerpc64le-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | llvm-cxxfilt -n | FileCheck %s --check-prefixes=CHECK,CHECK-LE
-// RUN: %clang -x c++ -fsyntax-only -target powerpc64le-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -DNO_WARN_X86_INTRINSICS %s \
+// RUN: %clang -x c++ -fsyntax-only -target powerpc64le-unknown-freebsd13.0 -mcpu=pwr8 -ffreestanding -nostdlibinc -DNO_WARN_X86_INTRINSICS %s \
 // RUN:   -fno-discard-value-names -mllvm -disable-llvm-optzns
 
 #include 
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D114394: Compile-time computation of string attribute hashes

[Mehdi AMINI via Phabricator via cfe-commits]

mehdi_amini added inline comments.



Comment at: llvm/lib/IR/Attributes.cpp:125
   FoldingSetNodeID ID;
-  ID.AddString(Kind);
+  ID.AddString(Kind.value());
   if (!Val.empty()) ID.AddString(Val);

Carrying over my comment from the original revision: it seem that you're ever 
only using the StringRef from the Kind in this function.
If so changing the API to use an AttributeKey seems pessimizing to me?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114394/new/

https://reviews.llvm.org/D114394

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

[PATCH] D114497: [Driver] Stop adding stdlib paths in -ffreestanding

[Fangrui Song via Phabricator via cfe-commits]

MaskRay added a comment.

GCC -ffreestanding doesn't change libstdc++ search paths. Clang's behavior 
matches GCC.

To drop /usr/include/c++/v1, you need -nostdinc++. If you want to drop 
/usr/include etc, you need to specify --sysroot=


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114497/new/

https://reviews.llvm.org/D114497

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

[PATCH] D114497: [Driver] Stop adding stdlib paths in -ffreestanding

[Zhihao Yuan via Phabricator via cfe-commits]

lichray created this revision.
Herald added subscribers: krytarowski, arichardson, emaste.
lichray requested review of this revision.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

When targeting FreeBSD on a Linux host with a copy
of system libc++, Clang prepends /usr/include/c++/v1
to the search paths even with -ffreestanding, and
fails to compile a program with a
single #include 

The proposed change fixes this issue.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114497

Files:
  clang/lib/Driver/ToolChains/Gnu.cpp


Index: clang/lib/Driver/ToolChains/Gnu.cpp
===
--- clang/lib/Driver/ToolChains/Gnu.cpp
+++ clang/lib/Driver/ToolChains/Gnu.cpp
@@ -2872,7 +2872,7 @@
 void Generic_GCC::AddClangCXXStdlibIncludeArgs(const ArgList ,
ArgStringList ) const {
   if (DriverArgs.hasArg(options::OPT_nostdinc, options::OPT_nostdincxx,
-options::OPT_nostdlibinc))
+options::OPT_nostdlibinc, options::OPT_ffreestanding))
 return;
 
   switch (GetCXXStdlibType(DriverArgs)) {


Index: clang/lib/Driver/ToolChains/Gnu.cpp
===
--- clang/lib/Driver/ToolChains/Gnu.cpp
+++ clang/lib/Driver/ToolChains/Gnu.cpp
@@ -2872,7 +2872,7 @@
 void Generic_GCC::AddClangCXXStdlibIncludeArgs(const ArgList ,
ArgStringList ) const {
   if (DriverArgs.hasArg(options::OPT_nostdinc, options::OPT_nostdincxx,
-options::OPT_nostdlibinc))
+options::OPT_nostdlibinc, options::OPT_ffreestanding))
 return;
 
   switch (GetCXXStdlibType(DriverArgs)) {
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D114411: [WIP][modules] Avoid deserializing Decls from hidden (sub)modules.

[Volodymyr Sapsai via Phabricator via cfe-commits]

vsapsai updated this revision to Diff 389379.
vsapsai added a comment.

Mark identifiers as out-of-date not only on loading a new module but on making 
a (sub)module visible too.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114411/new/

https://reviews.llvm.org/D114411

Files:
  clang/include/clang/Serialization/ASTReader.h
  clang/lib/Lex/Preprocessor.cpp
  clang/lib/Sema/SemaModule.cpp
  clang/lib/Serialization/ASTReader.cpp
  clang/lib/Serialization/ASTReaderDecl.cpp
  clang/lib/Serialization/ASTWriterDecl.cpp
  clang/test/Modules/ambiguous-enum-lookup.m

Index: clang/test/Modules/ambiguous-enum-lookup.m
===
--- /dev/null
+++ clang/test/Modules/ambiguous-enum-lookup.m
@@ -0,0 +1,42 @@
+// RUN: rm -rf %t
+// RUN: split-file %s %t
+// RUN: %clang_cc1 -fsyntax-only -fmodules -fimplicit-module-maps -fmodules-cache-path=%t/modules.cache -F %t/Frameworks %t/test.m
+
+//--- Frameworks/NonModular.framework/Headers/NonModular.h
+#ifndef NonModular_NonModular_h
+#define NonModular_NonModular_h
+struct SomeStruct {
+int x;
+};
+
+enum SomeEnum {
+kEnumValue = 1,
+};
+#endif
+
+//--- Frameworks/PiecewiseVisible.framework/Headers/InitiallyVisible.h
+// empty
+
+//--- Frameworks/PiecewiseVisible.framework/Headers/InitiallyHidden.h
+#include 
+
+//--- Frameworks/PiecewiseVisible.framework/Modules/module.modulemap
+framework module PiecewiseVisible {
+  header "InitiallyVisible.h"
+  export *
+
+  explicit module HiddenPiece {
+header "InitiallyHidden.h"
+export *
+  }
+}
+
+//--- test.m
+#include 
+#include 
+#include 
+
+void test() {
+struct SomeStruct s;
+s.x = kEnumValue;
+}
Index: clang/lib/Serialization/ASTWriterDecl.cpp
===
--- clang/lib/Serialization/ASTWriterDecl.cpp
+++ clang/lib/Serialization/ASTWriterDecl.cpp
@@ -269,6 +269,7 @@
 }
 
 void ASTDeclWriter::Visit(Decl *D) {
+  Record.push_back(Writer.getSubmoduleID(D->getOwningModule()));
   DeclVisitor::Visit(D);
 
   // Source locations require array (variable-length) abbreviations.  The
@@ -1911,6 +1912,7 @@
   // Abbreviation for DECL_FIELD
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_FIELD));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Decl
   Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // DeclContext
   Abv->Add(BitCodeAbbrevOp(0));   // LexicalDeclContext
@@ -1944,6 +1946,7 @@
   // Abbreviation for DECL_OBJC_IVAR
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_OBJC_IVAR));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Decl
   Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // DeclContext
   Abv->Add(BitCodeAbbrevOp(0));   // LexicalDeclContext
@@ -1980,6 +1983,7 @@
   // Abbreviation for DECL_ENUM
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_ENUM));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2030,6 +2034,7 @@
   // Abbreviation for DECL_RECORD
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_RECORD));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2092,6 +2097,7 @@
   // Abbreviation for DECL_PARM_VAR
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_PARM_VAR));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2140,6 +2146,7 @@
   // Abbreviation for DECL_TYPEDEF
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_TYPEDEF));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2169,6 +2176,7 @@
   // Abbreviation for DECL_VAR
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_VAR));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2221,6 +2229,7 @@
   // Abbreviation for DECL_CXX_METHOD
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_CXX_METHOD));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // RedeclarableDecl
   Abv->Add(BitCodeAbbrevOp(0)); // CanonicalDecl
   // Decl
Index: clang/lib/Serialization/ASTReaderDecl.cpp
===
--- 

LLVM build master will be restarted soon

[Galina Kistanova via cfe-commits]

 Hello,

LLVM build master will be restarted in the nearest hour.

Thanks

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

[PATCH] D114394: Compile-time computation of string attribute hashes

[Fangrui Song via Phabricator via cfe-commits]

MaskRay added inline comments.



Comment at: llvm/lib/IR/AttributeImpl.h:214
 
-  DenseMap StringAttrs;
+  std::unordered_map StringAttrs;
 

std::unordered_map is inefficient. Why is this change?



Comment at: llvm/lib/IR/Attributes.cpp:862
+  auto Where = StringAttrs.find(Kind);
+  return Where != StringAttrs.end() ? Where->second : Attribute();
 }

`lookup`



Comment at: llvm/lib/IR/Attributes.cpp:2054
+  auto  = pImpl->AttributeKeys;
+  auto Where = AttributeKeys.find(s);
+  if (Where == AttributeKeys.end()) {

If you change `AttributeKeys` to `DenseMap`, you can 
construct a `CachedHashStringRef` and avoid a hash computation in `AttributeKey 
Key(s);`



Comment at: llvm/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp:1531
 static StringRef getDeoptLowering(CallBase *Call) {
-  const char *DeoptLowering = "deopt-lowering";
+  const char DeoptLowering[] = "deopt-lowering";
   if (Call->hasFnAttr(DeoptLowering)) {

This can be committed separately.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114394/new/

https://reviews.llvm.org/D114394

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

[PATCH] D114411: [WIP][modules] Avoid deserializing Decls from hidden (sub)modules.

[Volodymyr Sapsai via Phabricator via cfe-commits]

vsapsai updated this revision to Diff 389374.
vsapsai added a comment.

Fix some failing tests to demonstrate the magnitued of changes better.

In some places check if a decl is deserialized successfully as it's not 
guaranteed.
Also tried to make sure a module is marked as visible *before* we try to 
deserialize its decls.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114411/new/

https://reviews.llvm.org/D114411

Files:
  clang/include/clang/Serialization/ASTReader.h
  clang/lib/Sema/SemaModule.cpp
  clang/lib/Serialization/ASTReader.cpp
  clang/lib/Serialization/ASTReaderDecl.cpp
  clang/lib/Serialization/ASTWriterDecl.cpp
  clang/test/Modules/ambiguous-enum-lookup.m

Index: clang/test/Modules/ambiguous-enum-lookup.m
===
--- /dev/null
+++ clang/test/Modules/ambiguous-enum-lookup.m
@@ -0,0 +1,42 @@
+// RUN: rm -rf %t
+// RUN: split-file %s %t
+// RUN: %clang_cc1 -fsyntax-only -fmodules -fimplicit-module-maps -fmodules-cache-path=%t/modules.cache -F %t/Frameworks %t/test.m
+
+//--- Frameworks/NonModular.framework/Headers/NonModular.h
+#ifndef NonModular_NonModular_h
+#define NonModular_NonModular_h
+struct SomeStruct {
+int x;
+};
+
+enum SomeEnum {
+kEnumValue = 1,
+};
+#endif
+
+//--- Frameworks/PiecewiseVisible.framework/Headers/InitiallyVisible.h
+// empty
+
+//--- Frameworks/PiecewiseVisible.framework/Headers/InitiallyHidden.h
+#include 
+
+//--- Frameworks/PiecewiseVisible.framework/Modules/module.modulemap
+framework module PiecewiseVisible {
+  header "InitiallyVisible.h"
+  export *
+
+  explicit module HiddenPiece {
+header "InitiallyHidden.h"
+export *
+  }
+}
+
+//--- test.m
+#include 
+#include 
+#include 
+
+void test() {
+struct SomeStruct s;
+s.x = kEnumValue;
+}
Index: clang/lib/Serialization/ASTWriterDecl.cpp
===
--- clang/lib/Serialization/ASTWriterDecl.cpp
+++ clang/lib/Serialization/ASTWriterDecl.cpp
@@ -269,6 +269,7 @@
 }
 
 void ASTDeclWriter::Visit(Decl *D) {
+  Record.push_back(Writer.getSubmoduleID(D->getOwningModule()));
   DeclVisitor::Visit(D);
 
   // Source locations require array (variable-length) abbreviations.  The
@@ -1911,6 +1912,7 @@
   // Abbreviation for DECL_FIELD
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_FIELD));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Decl
   Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // DeclContext
   Abv->Add(BitCodeAbbrevOp(0));   // LexicalDeclContext
@@ -1944,6 +1946,7 @@
   // Abbreviation for DECL_OBJC_IVAR
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_OBJC_IVAR));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Decl
   Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // DeclContext
   Abv->Add(BitCodeAbbrevOp(0));   // LexicalDeclContext
@@ -1980,6 +1983,7 @@
   // Abbreviation for DECL_ENUM
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_ENUM));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2030,6 +2034,7 @@
   // Abbreviation for DECL_RECORD
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_RECORD));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2092,6 +2097,7 @@
   // Abbreviation for DECL_PARM_VAR
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_PARM_VAR));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2140,6 +2146,7 @@
   // Abbreviation for DECL_TYPEDEF
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_TYPEDEF));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2169,6 +2176,7 @@
   // Abbreviation for DECL_VAR
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_VAR));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // Redeclarable
   Abv->Add(BitCodeAbbrevOp(0));   // No redeclaration
   // Decl
@@ -2221,6 +2229,7 @@
   // Abbreviation for DECL_CXX_METHOD
   Abv = std::make_shared();
   Abv->Add(BitCodeAbbrevOp(serialization::DECL_CXX_METHOD));
+  Abv->Add(BitCodeAbbrevOp(BitCodeAbbrevOp::VBR, 6)); // SubmoduleID
   // RedeclarableDecl
   Abv->Add(BitCodeAbbrevOp(0)); // CanonicalDecl
   // Decl
Index: 

[PATCH] D114151: [clang-format] [C++20] [Module] clang-format couldn't recognize partitions

[Owen Pan via Phabricator via cfe-commits]

owenpan added inline comments.



Comment at: clang/lib/Format/UnwrappedLineParser.cpp:1057
+  nextToken();
+  do {
+if (FormatTok->is(tok::colon)) {

A `while (!eof())` or `while (FormatTok->isNot(tok::eof)` would be  safer here 
just in case the last line is `import` followed by eof.



Comment at: clang/lib/Format/UnwrappedLineParser.cpp:1064-1065
+  nextToken();
+  while (FormatTok && FormatTok->isNot(tok::semi) &&
+ FormatTok->isNot(tok::greater)) {
+// Mark tokens up to the trailing line comments as implicit string

Possible infinite loop here as well. Maybe change it to `while 
(!FormatTok->isOneOf(tok::greater, tok::semi, tok::eof) {`?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114151/new/

https://reviews.llvm.org/D114151

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

[PATCH] D112913: Misleading bidirectional detection

[Fangrui Song via Phabricator via cfe-commits]

MaskRay added inline comments.



Comment at: clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp:49
+  if (C == '\n' || C == '\r' || C == '\f' || C == '\v' ||
+  C == 0x85 /*next line*/)
+EmbeddingOverride = Isolate = 0;

`/*next line=*/0x85` is more common



Comment at: clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp:66
+CodePoint == LRE)
+  EmbeddingOverride += 1;
+else if (CodePoint == PDF)





Comment at: clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp:68
+else if (CodePoint == PDF)
+  EmbeddingOverride = std::min(EmbeddingOverride - 1, EmbeddingOverride);
+else if (CodePoint == RLI || CodePoint == LRI || CodePoint == FSI)

More common style is `A = A ? A - 1 : 0;` which avoids unsigned wraparound.

That said, if the state is currently 0, should an attempt to decrease it be 
reported as an error as well?



Comment at: clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp:70
+else if (CodePoint == RLI || CodePoint == LRI || CodePoint == FSI)
+  Isolate += 1;
+else if (CodePoint == PDI)

ditto



Comment at: clang-tools-extra/clang-tidy/misc/MisleadingBidirectional.cpp:128
+
+void clang::tidy::misc::MisleadingBidirectionalCheck::registerMatchers(
+ast_matchers::MatchFinder *Finder) {

If you `using` MisleadingBidirectionalCheck (or `clang::tidy::misc`), then you 
can use

`void MisleadingBidirectionalCheck::registerMatchers`



Comment at: 
clang-tools-extra/test/clang-tidy/checkers/misc-misleading-bidirectional.cpp:6
+  /*‮ }⁦if(admin)⁩ ⁦ begin*/
+  // CHECK-MESSAGES: :[[@LINE-1]]:3: warning: comment contains misleading 
bidirectional Unicode characters [misc-misleading-bidirectional]
+  const char msg[] = "‮⁦if(admin)⁩ ⁦tes";

`[[@LINE-1]]` is a deprecated FileCheck feature.

Use `[[#@LINE-1]]`


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D112913/new/

https://reviews.llvm.org/D112913

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

[Diffusion] rGc93f93b2e3f2: Revert "Revert "Recommit "Revert "[CVP] processSwitch: Remove default case when…

[JunMa via Phabricator via cfe-commits]

junparser added a reverting change: rG07333810caee: Revert "Revert "Revert 
"Recommit "Revert "[CVP] processSwitch: Remove default….

BRANCHES
  EmptyLineAfterFunctionDefinition, fix_asan, main

Users:
  junparser (Author)

https://reviews.llvm.org/rGc93f93b2e3f2

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[John McCall via Phabricator via cfe-commits]

rjmccall added a comment.

Your builtin is using custom type-checking (`t`), which suppresses all the 
normal conversions that happen on expressions.  Specifically, it skips 
lvalue-to-rvalue conversion, so in this example the argument ends up being an 
l-value reference to a variable rather than an r-value loaded from that 
variable.  In addition to confusing constant evaluation, it would also make 
this look like an ODR-use of the variable, which would be subtly wrong in some 
C++ cases.  The fix is to explicitly request the standard l-value conversions, 
which you can do with code like:

  ExprResult Arg = S.DefaultFunctionArrayLvalueConversion(TheCall->getArg(0));
  if (Arg.isInvalid()) return true;
  TheCall->setArg(0, Arg.get());

After that, constant-evaluating the argument expression in your example should 
give you a `FunctionDecl*` as expected.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

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

[Diffusion] rGc93f93b2e3f2: Revert "Revert "Recommit "Revert "[CVP] processSwitch: Remove default case when…

[JunMa via Phabricator via cfe-commits]

junparser added a comment.

In rGc93f93b2e3f28997f794265089fb8138dd5b5f13#1040054 
, 
@lkail wrote:

> Looks a more general way should be implemented in tailduplicator to avoid 
> adding quadratic edges in CFGs.

yep， first I'll revert this.


BRANCHES
  EmptyLineAfterFunctionDefinition, fix_asan, main

Users:
  junparser (Author)

https://reviews.llvm.org/rGc93f93b2e3f2

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

[PATCH] D112024: [clang] diagnose_as attribute for Fortify diagnosing like builtins.

[Michael Benfield via Phabricator via cfe-commits]

mbenfield updated this revision to Diff 389355.
mbenfield added a comment.

Error on member function. Test case for this.

Test case for attribute applied to a non-function declaration.

Handle variadic functions. Document this. Test case for variadic functions.

Remove const from some local values.

Spell out some types.

Change superflous `dyn_cast_or_null` to `dyn_cast`.

Remove some curly braces.

Pass in a `NamedDecl` directly rather than its name to a diagnostic.

Add a test case for types not being identical.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D112024/new/

https://reviews.llvm.org/D112024

Files:
  clang/include/clang/Basic/Attr.td
  clang/include/clang/Basic/AttrDocs.td
  clang/include/clang/Basic/DiagnosticSemaKinds.td
  clang/include/clang/Sema/ParsedAttr.h
  clang/lib/AST/Decl.cpp
  clang/lib/Sema/SemaChecking.cpp
  clang/lib/Sema/SemaDeclAttr.cpp
  clang/test/Sema/attr-diagnose-as-builtin.c

Index: clang/test/Sema/attr-diagnose-as-builtin.c
===
--- /dev/null
+++ clang/test/Sema/attr-diagnose-as-builtin.c
@@ -0,0 +1,107 @@
+// RUN: %clang_cc1 -Wfortify-source -triple x86_64-apple-macosx10.14.0 %s -verify
+// RUN: %clang_cc1 -Wfortify-source -xc++ -triple x86_64-apple-macosx10.14.0 %s -verify
+
+typedef unsigned long size_t;
+
+__attribute__((diagnose_as_builtin(__builtin_memcpy, 3, 1, 2))) int x; // expected-warning {{'diagnose_as_builtin' attribute only applies to functions}}
+
+void *test_memcpy(const void *src, size_t c, void *dst) __attribute__((diagnose_as_builtin(__builtin_memcpy, 3, 1, 2))) {
+  return __builtin_memcpy(dst, src, c);
+}
+
+void call_test_memcpy() {
+  char bufferA[10];
+  char bufferB[11];
+  test_memcpy(bufferB, 10, bufferA);
+  test_memcpy(bufferB, 11, bufferA); // expected-warning {{'memcpy' will always overflow; destination buffer has size 10, but size argument is 11}}
+}
+
+void failure_function_doesnt_exist() __attribute__((diagnose_as_builtin(__function_that_doesnt_exist))) {} // expected-error {{use of undeclared identifier '__function_that_doesnt_exist'}}
+
+void not_a_builtin() {}
+
+void failure_not_a_builtin() __attribute__((diagnose_as_builtin(not_a_builtin))) {} // expected-error {{'diagnose_as_builtin' attribute requires parameter 1 to be a builtin function}}
+
+void failure_too_many_parameters(void *dst, const void *src, size_t count, size_t nothing) __attribute__((diagnose_as_builtin(__builtin_memcpy, 1, 2, 3, 4))) {} // expected-error {{'diagnose_as_builtin' attribute references function '__builtin_memcpy', which takes exactly 3 arguments}}
+
+void failure_too_few_parameters(void *dst, const void *src) __attribute__((diagnose_as_builtin(__builtin_memcpy, 1, 2))) {} // expected-error{{'diagnose_as_builtin' attribute references function '__builtin_memcpy', which takes exactly 3 arguments}}
+
+void failure_not_an_integer(void *dst, const void *src, size_t count) __attribute__((diagnose_as_builtin(__builtin_memcpy, "abc", 2, 3))) {} // expected-error{{'diagnose_as_builtin' attribute requires parameter 2 to be an integer constant}}
+
+void failure_not_a_builtin2() __attribute__((diagnose_as_builtin("abc"))) {} // expected-error{{'diagnose_as_builtin' attribute requires parameter 1 to be a builtin function}}
+
+void failure_parameter_index_bounds(void *dst, const void *src) __attribute__((diagnose_as_builtin(__builtin_memcpy, 1, 2, 3))) {} // expected-error{{'diagnose_as_builtin' attribute references parameter 3, but the function failure_parameter_index_bounds has only 2 parameters}}
+
+void failure_parameter_types(double dst, const void *src, size_t count) __attribute__((diagnose_as_builtin(__builtin_memcpy, 1, 2, 3))) {} // expected-error{{'diagnose_as_builtin' attribute parameter types do not match: parameter 1 of function 'failure_parameter_types' has type 'double', but parameter 1 of function '__builtin_memcpy' has type 'void *'}}
+
+void to_redeclare(void *dst, const void *src, size_t count);
+
+void use_to_redeclare() {
+  char src[10];
+  char dst[9];
+  // We shouldn't get an error yet.
+  to_redeclare(dst, src, 10);
+}
+
+void to_redeclare(void *dst, const void *src, size_t count) __attribute((diagnose_as_builtin(__builtin_memcpy, 1, 2, 3)));
+
+void error_to_redeclare() {
+  char src[10];
+  char dst[9];
+  // Now we get an error.
+  to_redeclare(dst, src, 10); // expected-warning {{'memcpy' will always overflow; destination buffer has size 9, but size argument is 10}}
+}
+
+// Make sure this works even with extra qualifiers and the pass_object_size attribute.
+void *memcpy2(void *const dst __attribute__((pass_object_size(0))), const void *src, size_t copy_amount) __attribute((diagnose_as_builtin(__builtin_memcpy, 1, 2, 3)));
+
+void call_memcpy2() {
+  char buf1[10];
+  char buf2[11];
+  memcpy2(buf1, buf2, 11); // expected-warning {{'memcpy' will always overflow; destination buffer has size 10, but size argument is 11}}
+}
+

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[Sami Tolvanen via Phabricator via cfe-commits]

samitolvanen added inline comments.



Comment at: clang/lib/Sema/SemaChecking.cpp:208
+if (UnaryOp->getOpcode() == UnaryOperator::Opcode::UO_AddrOf)
+  E = UnaryOp->getSubExpr();
+

rjmccall wrote:
> samitolvanen wrote:
> > rjmccall wrote:
> > > samitolvanen wrote:
> > > > rjmccall wrote:
> > > > > It would be more general to allow any expression that we can 
> > > > > constant-evaluate to a specific function / member function reference. 
> > > > >  That allows callers to do stuff like `__builtin_function_start((int 
> > > > > (A::*)() const) ::x)` to resolve overloaded function references.
> > > > > 
> > > > > You should delay this check if the operand is value-dependent.
> > > > > It would be more general to allow any expression that we can 
> > > > > constant-evaluate to a specific function / member function reference. 
> > > > >  That allows callers to do stuff like `__builtin_function_start((int 
> > > > > (A::*)() const) ::x)` to resolve overloaded function references.
> > > > 
> > > > I looked into using `Expr::EvaluateAsConstantExpr` here and while it 
> > > > works, I'm not sure if allowing arbitrary expressions as the argument 
> > > > provides any value. We can allow resolving overloaded function 
> > > > references without constant-evaluating the expression (and I added 
> > > > tests for this). Did you have any other use cases in mind where this 
> > > > might be useful?
> > > I don't see what the advantage of limiting the constant expression would 
> > > be if we can constant-evaluate it.  `switch` doesn't force you to make 
> > > case values be integer literals and/or references to enumerators.  What 
> > > are you trying to achieve with a restriction?
> > > 
> > > Not having arbitrary restrictions is particularly useful in C++, where 
> > > templates and `constexpr` machinery can usefully do a lot of abstraction.
> > > I don't see what the advantage of limiting the constant expression would 
> > > be if we can constant-evaluate it.  `switch` doesn't force you to make 
> > > case values be integer literals and/or references to enumerators.  What 
> > > are you trying to achieve with a restriction?
> > 
> > I'm trying to understand the benefit of allowing arbitrary expressions like 
> > `__builtin_function_start(100 + a)`, which `EvaluateAsConstantExpr` is 
> > happy to evaluate into a reference to `a`.
> > 
> > I can obviously understand why these should be allowed for `switch`, but 
> > here we have a function whose sole purpose is to return the address of a 
> > function and I'm wondering why someone would want pass anything more 
> > complicated to it.
> > 
> > > Not having arbitrary restrictions is particularly useful in C++, where 
> > > templates and `constexpr` machinery can usefully do a lot of abstraction.
> > 
> > Perhaps I'm just not that familiar with the C++ use cases here. Would you 
> > be able to provide me an example I could use as a test case?
> > I can obviously understand why these should be allowed for switch, but here 
> > we have a function whose sole purpose is to return the address of a 
> > function and I'm wondering why someone would want pass anything more 
> > complicated to it.
> 
> The idea is that you might have a complicated way of picking which function 
> you mean.  I agree that this should probably disallow non-zero offsets.
> 
> > Perhaps I'm just not that familiar with the C++ use cases here. Would you 
> > be able to provide me an example I could use as a test case?
> 
> A simple example: `fn` after `constexpr void (*fn)() = `.
> > Perhaps I'm just not that familiar with the C++ use cases here. Would you 
> > be able to provide me an example I could use as a test case?
> 
> A simple example: `fn` after `constexpr void (*fn)() = `.

Thanks for the example. So, something like this should also work?
```
$ cat test.cc 
void a() {}
constexpr void (*fn)() = 
const void *p = __builtin_function_start(fn);
```

Looking at what happens after we evaluate the expression, `APValue` is an 
l-value containing a `VarDecl`:
```
VarDecl 0xe3427a8  col:18 referenced fn 'void (*const)()' 
constexpr cinit
```
Which means I have to look at `VarDecl::getEvaluatedValue` if I want to get the 
function declaration. Or should the evaluation in this case produce a 
`FunctionDecl` directly?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[Sami Tolvanen via Phabricator via cfe-commits]

samitolvanen updated this revision to Diff 389346.
samitolvanen added a comment.

Changed the code to evaluate the argument as a constant expression.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

Files:
  clang/docs/LanguageExtensions.rst
  clang/include/clang/AST/Expr.h
  clang/include/clang/Basic/Builtins.def
  clang/include/clang/Basic/DiagnosticSemaKinds.td
  clang/lib/AST/Expr.cpp
  clang/lib/AST/ExprConstant.cpp
  clang/lib/CodeGen/CGBuiltin.cpp
  clang/lib/CodeGen/CGExprConstant.cpp
  clang/lib/CodeGen/CodeGenModule.cpp
  clang/lib/CodeGen/CodeGenModule.h
  clang/lib/Sema/SemaChecking.cpp
  clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
  clang/test/CodeGen/builtin-function-start.cpp
  clang/test/SemaCXX/builtins.cpp

Index: clang/test/SemaCXX/builtins.cpp
===
--- clang/test/SemaCXX/builtins.cpp
+++ clang/test/SemaCXX/builtins.cpp
@@ -39,6 +39,13 @@
   S *ptmp = __builtin_addressof(S{}); // expected-error {{taking the address of a temporary}}
 }
 
+namespace function_start {
+void a(void) {}
+int n;
+void *p = __builtin_function_start(n);   // expected-error {{argument must be a function}}
+static_assert(__builtin_function_start(a) == a, ""); // expected-error {{static_assert expression is not an integral constant expression}}
+} // namespace function_start
+
 void no_ms_builtins() {
   __assume(1); // expected-error {{use of undeclared}}
   __noop(1); // expected-error {{use of undeclared}}
Index: clang/test/CodeGen/builtin-function-start.cpp
===
--- /dev/null
+++ clang/test/CodeGen/builtin-function-start.cpp
@@ -0,0 +1,51 @@
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -fsanitize=cfi-icall -o - %s | FileCheck %s
+
+#if !__has_builtin(__builtin_function_start)
+#error "missing __builtin_function_start"
+#endif
+
+void a(void) {}
+void b(void) {}
+void c(void *p) {}
+
+// CHECK: @e = global i8* bitcast (void ()* no_cfi @_Z1av to i8*)
+const void *e = __builtin_function_start(a);
+// CHECK: @f = global [2 x i8*] [i8* bitcast (void ()* @_Z1bv to i8*), i8* bitcast (void ()* no_cfi @_Z1bv to i8*)]
+void *f[] = {(void *)b, __builtin_function_start(b)};
+
+class A {
+public:
+  void f();
+  virtual void g();
+  static void h();
+  int i() const;
+  int i(int n) const;
+};
+
+void A::f() {}
+void A::g() {}
+void A::h() {}
+
+// CHECK: define {{.*}}i32 @_ZNK1A1iEv(%class.A* {{.*}}%this)
+int A::i() const { return 0; }
+
+// CHECK: define {{.*}}i32 @_ZNK1A1iEi(%class.A* {{.*}}%this, i32 %n)
+int A::i(int n) const { return 0; }
+
+void d(void) {
+  // CHECK: store i8* bitcast (void ()* no_cfi @_Z1bv to i8*), i8** %g
+  void *g = __builtin_function_start(b);
+  // call void @_Z1cPv(i8* bitcast (void ()* no_cfi @_Z1av to i8*))
+  c(__builtin_function_start(a));
+
+  // CHECK: store i8* bitcast (void (%class.A*)* no_cfi @_ZN1A1fEv to i8*), i8** %Af
+  void *Af = __builtin_function_start(::f);
+  // CHECK: store i8* bitcast (void (%class.A*)* no_cfi @_ZN1A1gEv to i8*), i8** %Ag
+  void *Ag = __builtin_function_start(::g);
+  // CHECK: store i8* bitcast (void ()* no_cfi @_ZN1A1hEv to i8*), i8** %Ah
+  void *Ah = __builtin_function_start(::h);
+  // CHECK: store i8* bitcast (i32 (%class.A*)* no_cfi @_ZNK1A1iEv to i8*), i8** %Ai1
+  void *Ai1 = __builtin_function_start((int(A::*)() const) & A::i);
+  // CHECK: store i8* bitcast (i32 (%class.A*, i32)* no_cfi @_ZNK1A1iEi to i8*), i8** %Ai2
+  void *Ai2 = __builtin_function_start((int(A::*)(int) const) & A::i);
+}
Index: clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
===
--- clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
+++ clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
@@ -66,7 +66,8 @@
   case Builtin::BI__builtin_expect:
   case Builtin::BI__builtin_expect_with_probability:
   case Builtin::BI__builtin_assume_aligned:
-  case Builtin::BI__builtin_addressof: {
+  case Builtin::BI__builtin_addressof:
+  case Builtin::BI__builtin_function_start: {
 // For __builtin_unpredictable, __builtin_expect,
 // __builtin_expect_with_probability and __builtin_assume_aligned,
 // just return the value of the subexpression.
Index: clang/lib/Sema/SemaChecking.cpp
===
--- clang/lib/Sema/SemaChecking.cpp
+++ clang/lib/Sema/SemaChecking.cpp
@@ -195,6 +195,23 @@
   return false;
 }
 
+/// Check that the argument to __builtin_function_start is a function.
+static bool SemaBuiltinFunctionStart(Sema , CallExpr *TheCall) {
+  if (checkArgCount(S, TheCall, 1))
+return true;
+
+  const FunctionDecl *FD = TheCall->getArgFunctionDecl(S.getASTContext());
+
+  if (!FD) {
+S.Diag(TheCall->getBeginLoc(), diag::err_function_start_invalid_type)
+   

[PATCH] D114484: [NFC][AIX]Disable unsupported hip test on AIX

[Steven Wan via Phabricator via cfe-commits]

stevewan created this revision.
Herald added a subscriber: yaxunl.
stevewan requested review of this revision.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

AIX doesn't support GPU. There is no point testing HIP on it.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114484

Files:
  clang/test/Driver/hip-version.hip


Index: clang/test/Driver/hip-version.hip
===
--- clang/test/Driver/hip-version.hip
+++ clang/test/Driver/hip-version.hip
@@ -1,6 +1,7 @@
 // REQUIRES: clang-driver
 // REQUIRES: x86-registered-target
 // REQUIRES: amdgpu-registered-target
+// UNSUPPORTED: aix
 
 // RUN: %clang -v --rocm-path=%S/Inputs/rocm 2>&1 \
 // RUN:   | FileCheck -check-prefixes=FOUND %s


Index: clang/test/Driver/hip-version.hip
===
--- clang/test/Driver/hip-version.hip
+++ clang/test/Driver/hip-version.hip
@@ -1,6 +1,7 @@
 // REQUIRES: clang-driver
 // REQUIRES: x86-registered-target
 // REQUIRES: amdgpu-registered-target
+// UNSUPPORTED: aix
 
 // RUN: %clang -v --rocm-path=%S/Inputs/rocm 2>&1 \
 // RUN:   | FileCheck -check-prefixes=FOUND %s
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D114058: [clangd] Add ObjC method support to prepareCallHierarchy

[Sheldon Neuberger via Phabricator via cfe-commits]

sheldonneuberger-sc added a comment.

Made suggested changes, ready for another review.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114058/new/

https://reviews.llvm.org/D114058

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

[PATCH] D114483: Add support for sycl_special_class attribute

[Zahira Ammarguellat via Phabricator via cfe-commits]

zahiraam created this revision.
zahiraam added reviewers: bader, Naghasan, keryell, Fznamznon, aaron.ballman, 
erichkeane.
Herald added subscribers: jdoerfert, Anastasia, ebevhan, yaxunl.
zahiraam requested review of this revision.
Herald added a project: clang.

Special classes such as accessor, sampler, and stream need additional 
implementation when they are passed from host to device.
This patch is adding a new attribute “sycl_special_class” used to mark SYCL 
classes/struct that need the additional compiler handling.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114483

Files:
  clang/include/clang/Basic/Attr.td
  clang/include/clang/Basic/AttrDocs.td
  clang/lib/Sema/SemaDeclAttr.cpp
  clang/test/Misc/pragma-attribute-supported-attributes-list.test
  clang/test/SemaSYCL/special-class-attribute-on-non-sycl.cpp
  clang/test/SemaSYCL/special-class-attribute.cpp

Index: clang/test/SemaSYCL/special-class-attribute.cpp
===
--- /dev/null
+++ clang/test/SemaSYCL/special-class-attribute.cpp
@@ -0,0 +1,33 @@
+// RUN: %clang_cc1 -std=c++11 -fsyntax-only -fsycl-is-device -verify %s
+
+// No diagnostics
+class [[clang::sycl_special_class]] class1 {};
+class __attribute__((sycl_special_class)) class2 {};
+
+class class3;
+class [[clang::sycl_special_class]] class3 {};
+
+class class4;
+class __attribute__((sycl_special_class)) class4 {};
+
+class [[clang::sycl_special_class]] struct1 {};
+struct __attribute__((sycl_special_class)) struct2 {};
+
+class [[clang::sycl_special_class]] class5 {};
+class special5 {};
+
+class __attribute__((sycl_special_class)) class6;
+class class6 {};
+
+
+// Only classes
+[[clang::sycl_special_class]] int var1 = 0; // expected-warning {{'sycl_special_class' attribute only applies to classes}}
+__attribute__((sycl_special_class)) int var2 = 0; // expected-warning {{'sycl_special_class' attribute only applies to classes}}
+
+[[clang::sycl_special_class]] void foo1(); // expected-warning {{'sycl_special_class' attribute only applies to classes}}
+__attribute__((sycl_special_class)) void foo2(); // expected-warning {{'sycl_special_class' attribute only applies to classes}}
+
+// Attribute takes no arguments
+class [[clang::sycl_special_class(1)]] class7 {}; // expected-error {{'sycl_special_class' attribute takes no arguments}}
+class __attribute__((sycl_special_class(1))) class8 {}; // expected-error {{'sycl_special_class' attribute takes no arguments}}
+
Index: clang/test/SemaSYCL/special-class-attribute-on-non-sycl.cpp
===
--- /dev/null
+++ clang/test/SemaSYCL/special-class-attribute-on-non-sycl.cpp
@@ -0,0 +1,11 @@
+// RUN: %clang_cc1 -std=c++11 -fsyntax-only -fsycl-is-device -verify %s
+// RUN: %clang_cc1 -std=c++11 -fsyntax-only -verify -x c++ %s
+
+#ifndef __SYCL_DEVICE_ONLY__
+// expected-warning@+5 {{'sycl_special_class' attribute ignored}}
+#else
+// expected-no-diagnostics
+#endif
+
+class __attribute__((sycl_special_class)) special_class {};
+
Index: clang/test/Misc/pragma-attribute-supported-attributes-list.test
===
--- clang/test/Misc/pragma-attribute-supported-attributes-list.test
+++ clang/test/Misc/pragma-attribute-supported-attributes-list.test
@@ -154,6 +154,7 @@
 // CHECK-NEXT: ReturnTypestate (SubjectMatchRule_function, SubjectMatchRule_variable_is_parameter)
 // CHECK-NEXT: ReturnsNonNull (SubjectMatchRule_objc_method, SubjectMatchRule_function)
 // CHECK-NEXT: ReturnsTwice (SubjectMatchRule_function)
+// CHECK-NEXT: SYCLSpecialClass (SubjectMatchRule_record)
 // CHECK-NEXT: ScopedLockable (SubjectMatchRule_record)
 // CHECK-NEXT: Section (SubjectMatchRule_function, SubjectMatchRule_variable_is_global, SubjectMatchRule_objc_method, SubjectMatchRule_objc_property)
 // CHECK-NEXT: SetTypestate (SubjectMatchRule_function_is_member)
Index: clang/lib/Sema/SemaDeclAttr.cpp
===
--- clang/lib/Sema/SemaDeclAttr.cpp
+++ clang/lib/Sema/SemaDeclAttr.cpp
@@ -8054,6 +8054,9 @@
   case ParsedAttr::AT_SYCLKernel:
 handleSYCLKernelAttr(S, D, AL);
 break;
+  case ParsedAttr::AT_SYCLSpecialClass:
+handleSimpleAttribute(S, D, AL);
+break;
   case ParsedAttr::AT_Format:
 handleFormatAttr(S, D, AL);
 break;
Index: clang/include/clang/Basic/AttrDocs.td
===
--- clang/include/clang/Basic/AttrDocs.td
+++ clang/include/clang/Basic/AttrDocs.td
@@ -405,6 +405,65 @@
   }];
 }
 
+def SYCLSpecialClassDocs : Documentation {
+  let Category = DocCatStmt;
+  let Content = [{
+The ``__attribute__((sycl_special_class))`` attribute is used in SYCL
+headers to indicate that a class or a struct needs additional implementation when
+it is passed from host to device.
+Special classes/struct will have a mandatory __init method and an optional 

[PATCH] D114058: [clangd] Add ObjC method support to prepareCallHierarchy

[Sheldon Neuberger via Phabricator via cfe-commits]

sheldonneuberger-sc updated this revision to Diff 389334.
sheldonneuberger-sc added a comment.

- revert whitespace change


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114058/new/

https://reviews.llvm.org/D114058

Files:
  clang-tools-extra/clangd/XRefs.cpp
  clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp

Index: clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp
===
--- clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp
+++ clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp
@@ -65,7 +65,7 @@
UnorderedElementsAre(M...));
 }
 
-TEST(CallHierarchy, IncomingOneFile) {
+TEST(CallHierarchy, IncomingOneFileCpp) {
   Annotations Source(R"cpp(
 void call^ee(int);
 void caller1() {
@@ -91,7 +91,51 @@
   ASSERT_THAT(IncomingLevel1,
   ElementsAre(AllOf(From(WithName("caller1")),
 FromRanges(Source.range("Callee");
+  auto IncomingLevel2 = incomingCalls(IncomingLevel1[0].from, Index.get());
+  ASSERT_THAT(IncomingLevel2,
+  ElementsAre(AllOf(From(WithName("caller2")),
+FromRanges(Source.range("Caller1A"),
+   Source.range("Caller1B"))),
+  AllOf(From(WithName("caller3")),
+FromRanges(Source.range("Caller1C");
 
+  auto IncomingLevel3 = incomingCalls(IncomingLevel2[0].from, Index.get());
+  ASSERT_THAT(IncomingLevel3,
+  ElementsAre(AllOf(From(WithName("caller3")),
+FromRanges(Source.range("Caller2");
+
+  auto IncomingLevel4 = incomingCalls(IncomingLevel3[0].from, Index.get());
+  EXPECT_THAT(IncomingLevel4, IsEmpty());
+}
+
+TEST(CallHierarchy, IncomingOneFileObjC) {
+  Annotations Source(R"objc(
+@implementation MyClass {}
+  +(void)call^ee {}
+  +(void) caller1 {
+[MyClass $Callee[[callee]]];
+  }
+  +(void) caller2 {
+[MyClass $Caller1A[[caller1]]];
+[MyClass $Caller1B[[caller1]]];
+  }
+  +(void) caller3 {
+[MyClass $Caller1C[[caller1]]];
+[MyClass $Caller2[[caller2]]];
+  }
+@end
+  )objc");
+  TestTU TU = TestTU::withCode(Source.code());
+  TU.Filename = "TestTU.m";
+  auto AST = TU.build();
+  auto Index = TU.index();
+  std::vector Items =
+  prepareCallHierarchy(AST, Source.point(), testPath(TU.Filename));
+  ASSERT_THAT(Items, ElementsAre(WithName("callee")));
+  auto IncomingLevel1 = incomingCalls(Items[0], Index.get());
+  ASSERT_THAT(IncomingLevel1,
+  ElementsAre(AllOf(From(WithName("caller1")),
+FromRanges(Source.range("Callee");
   auto IncomingLevel2 = incomingCalls(IncomingLevel1[0].from, Index.get());
   ASSERT_THAT(IncomingLevel2,
   ElementsAre(AllOf(From(WithName("caller2")),
@@ -172,7 +216,7 @@
 FromRanges(Source.range("Caller2");
 }
 
-TEST(CallHierarchy, IncomingMultiFile) {
+TEST(CallHierarchy, IncomingMultiFileCpp) {
   // The test uses a .hh suffix for header files to get clang
   // to parse them in C++ mode. .h files are parsed in C mode
   // by default, which causes problems because e.g. symbol
@@ -268,6 +312,115 @@
   CheckCallHierarchy(*AST, CalleeC.point(), testPath("callee.cc"));
 }
 
+TEST(CallHierarchy, IncomingMultiFileObjC) {
+  // The test uses a .mi suffix for header files to get clang
+  // to parse them in ObjC mode. .h files are parsed in C mode
+  // by default, which causes problems because e.g. symbol
+  // USRs are different in C mode (do not include function signatures).
+
+  Annotations CalleeH(R"objc(
+@interface CalleeClass
+  +(void)call^ee;
+@end
+  )objc");
+  Annotations CalleeC(R"objc(
+#import "callee.mi"
+@implementation CalleeClass {}
+  +(void)call^ee {}
+@end
+  )objc");
+  Annotations Caller1H(R"objc(
+@interface Caller1Class
+  +(void)caller1;
+@end
+  )objc");
+  Annotations Caller1C(R"objc(
+#import "callee.mi"
+#import "caller1.mi"
+@implementation Caller1Class {}
+  +(void)caller1 {
+[CalleeClass [[calle^e]]];
+  }
+@end
+  )objc");
+  Annotations Caller2H(R"objc(
+@interface Caller2Class
+  +(void)caller2;
+@end
+  )objc");
+  Annotations Caller2C(R"objc(
+#import "caller1.mi"
+#import "caller2.mi"
+@implementation Caller2Class {}
+  +(void)caller2 {
+[Caller1Class $A[[caller1]]];
+[Caller1Class $B[[caller1]]];
+  }
+@end
+  )objc");
+  Annotations Caller3C(R"objc(
+#import "caller1.mi"
+#import "caller2.mi"
+@implementation Caller3Class {}
+  +(void)caller3 {
+[Caller1Class $Caller1[[caller1]]];
+[Caller2Class $Caller2[[caller2]]];
+  }
+@end
+  )objc");
+
+  TestWorkspace Workspace;
+  

[PATCH] D114058: [clangd] Add ObjC method support to prepareCallHierarchy

[Sheldon Neuberger via Phabricator via cfe-commits]

sheldonneuberger-sc updated this revision to Diff 389333.
sheldonneuberger-sc added a comment.

- inline test helper for readability
- fix indentation


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114058/new/

https://reviews.llvm.org/D114058

Files:
  clang-tools-extra/clangd/XRefs.cpp
  clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp

Index: clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp
===
--- clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp
+++ clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp
@@ -65,7 +65,7 @@
UnorderedElementsAre(M...));
 }
 
-TEST(CallHierarchy, IncomingOneFile) {
+TEST(CallHierarchy, IncomingOneFileCpp) {
   Annotations Source(R"cpp(
 void call^ee(int);
 void caller1() {
@@ -91,7 +91,51 @@
   ASSERT_THAT(IncomingLevel1,
   ElementsAre(AllOf(From(WithName("caller1")),
 FromRanges(Source.range("Callee");
+  auto IncomingLevel2 = incomingCalls(IncomingLevel1[0].from, Index.get());
+  ASSERT_THAT(IncomingLevel2,
+  ElementsAre(AllOf(From(WithName("caller2")),
+FromRanges(Source.range("Caller1A"),
+   Source.range("Caller1B"))),
+  AllOf(From(WithName("caller3")),
+FromRanges(Source.range("Caller1C");
 
+  auto IncomingLevel3 = incomingCalls(IncomingLevel2[0].from, Index.get());
+  ASSERT_THAT(IncomingLevel3,
+  ElementsAre(AllOf(From(WithName("caller3")),
+FromRanges(Source.range("Caller2");
+
+  auto IncomingLevel4 = incomingCalls(IncomingLevel3[0].from, Index.get());
+  EXPECT_THAT(IncomingLevel4, IsEmpty());
+}
+
+TEST(CallHierarchy, IncomingOneFileObjC) {
+  Annotations Source(R"objc(
+@implementation MyClass {}
+  +(void)call^ee {}
+  +(void) caller1 {
+[MyClass $Callee[[callee]]];
+  }
+  +(void) caller2 {
+[MyClass $Caller1A[[caller1]]];
+[MyClass $Caller1B[[caller1]]];
+  }
+  +(void) caller3 {
+[MyClass $Caller1C[[caller1]]];
+[MyClass $Caller2[[caller2]]];
+  }
+@end
+  )objc");
+  TestTU TU = TestTU::withCode(Source.code());
+  TU.Filename = "TestTU.m";
+  auto AST = TU.build();
+  auto Index = TU.index();
+  std::vector Items =
+  prepareCallHierarchy(AST, Source.point(), testPath(TU.Filename));
+  ASSERT_THAT(Items, ElementsAre(WithName("callee")));
+  auto IncomingLevel1 = incomingCalls(Items[0], Index.get());
+  ASSERT_THAT(IncomingLevel1,
+  ElementsAre(AllOf(From(WithName("caller1")),
+FromRanges(Source.range("Callee");
   auto IncomingLevel2 = incomingCalls(IncomingLevel1[0].from, Index.get());
   ASSERT_THAT(IncomingLevel2,
   ElementsAre(AllOf(From(WithName("caller2")),
@@ -172,7 +216,7 @@
 FromRanges(Source.range("Caller2");
 }
 
-TEST(CallHierarchy, IncomingMultiFile) {
+TEST(CallHierarchy, IncomingMultiFileCpp) {
   // The test uses a .hh suffix for header files to get clang
   // to parse them in C++ mode. .h files are parsed in C mode
   // by default, which causes problems because e.g. symbol
@@ -223,7 +267,6 @@
   Workspace.addMainFile("caller1.cc", Caller1C.code());
   Workspace.addMainFile("caller2.cc", Caller2C.code());
   Workspace.addMainFile("caller3.cc", Caller3C.code());
-
   auto Index = Workspace.index();
 
   auto CheckCallHierarchy = [&](ParsedAST , Position Pos, PathRef TUPath) {
@@ -268,6 +311,115 @@
   CheckCallHierarchy(*AST, CalleeC.point(), testPath("callee.cc"));
 }
 
+TEST(CallHierarchy, IncomingMultiFileObjC) {
+  // The test uses a .mi suffix for header files to get clang
+  // to parse them in ObjC mode. .h files are parsed in C mode
+  // by default, which causes problems because e.g. symbol
+  // USRs are different in C mode (do not include function signatures).
+
+  Annotations CalleeH(R"objc(
+@interface CalleeClass
+  +(void)call^ee;
+@end
+  )objc");
+  Annotations CalleeC(R"objc(
+#import "callee.mi"
+@implementation CalleeClass {}
+  +(void)call^ee {}
+@end
+  )objc");
+  Annotations Caller1H(R"objc(
+@interface Caller1Class
+  +(void)caller1;
+@end
+  )objc");
+  Annotations Caller1C(R"objc(
+#import "callee.mi"
+#import "caller1.mi"
+@implementation Caller1Class {}
+  +(void)caller1 {
+[CalleeClass [[calle^e]]];
+  }
+@end
+  )objc");
+  Annotations Caller2H(R"objc(
+@interface Caller2Class
+  +(void)caller2;
+@end
+  )objc");
+  Annotations Caller2C(R"objc(
+#import "caller1.mi"
+#import "caller2.mi"
+@implementation Caller2Class {}
+  +(void)caller2 {
+[Caller1Class $A[[caller1]]];
+[Caller1Class $B[[caller1]]];
+  }
+

[PATCH] D114481: [NFC][AIX]Disable precompiled module file tests on AIX

[Steven Wan via Phabricator via cfe-commits]

stevewan created this revision.
stevewan requested review of this revision.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

The PCH reader looks for `__clangast` section in the precompiled module file, 
which is not present in the file on AIX, and we don't support writing this 
custom section in XCOFF yet.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114481

Files:
  clang/test/ClangScanDeps/modules-pch-common-submodule.c
  clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
  clang/test/ClangScanDeps/modules-pch.c
  clang/test/PCH/debug-info-pch-path.c


Index: clang/test/PCH/debug-info-pch-path.c
===
--- clang/test/PCH/debug-info-pch-path.c
+++ clang/test/PCH/debug-info-pch-path.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // RUN: rm -rf %t
 // RUN: mkdir %t
 // RUN: cd %t
Index: clang/test/ClangScanDeps/modules-pch.c
===
--- clang/test/ClangScanDeps/modules-pch.c
+++ clang/test/ClangScanDeps/modules-pch.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // RUN: rm -rf %t && mkdir %t
 // RUN: cp %S/Inputs/modules-pch/* %t
 
Index: clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
===
--- clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
+++ clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // Check that we discover dependency on a precompiled module (and generate the
 // appropriate `-fmodule-file=` argument) when it's imported by a **submodule**
 // instead of a top-level module.
Index: clang/test/ClangScanDeps/modules-pch-common-submodule.c
===
--- clang/test/ClangScanDeps/modules-pch-common-submodule.c
+++ clang/test/ClangScanDeps/modules-pch-common-submodule.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // Check that when depending on a precompiled module, we depend on the
 // **top-level** module. Submodules don't have some information present (for
 // example the path to the modulemap file) and depending on them might cause


Index: clang/test/PCH/debug-info-pch-path.c
===
--- clang/test/PCH/debug-info-pch-path.c
+++ clang/test/PCH/debug-info-pch-path.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // RUN: rm -rf %t
 // RUN: mkdir %t
 // RUN: cd %t
Index: clang/test/ClangScanDeps/modules-pch.c
===
--- clang/test/ClangScanDeps/modules-pch.c
+++ clang/test/ClangScanDeps/modules-pch.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // RUN: rm -rf %t && mkdir %t
 // RUN: cp %S/Inputs/modules-pch/* %t
 
Index: clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
===
--- clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
+++ clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // Check that we discover dependency on a precompiled module (and generate the
 // appropriate `-fmodule-file=` argument) when it's imported by a **submodule**
 // instead of a top-level module.
Index: clang/test/ClangScanDeps/modules-pch-common-submodule.c
===
--- clang/test/ClangScanDeps/modules-pch-common-submodule.c
+++ clang/test/ClangScanDeps/modules-pch-common-submodule.c
@@ -1,3 +1,5 @@
+// UNSUPPORTED: aix
+
 // Check that when depending on a precompiled module, we depend on the
 // **top-level** module. Submodules don't have some information present (for
 // example the path to the modulemap file) and depending on them might cause
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D114317: [clang-tidy][WIP] Do not run perfect alias checks

[Roman Lebedev via Phabricator via cfe-commits]

lebedev.ri added a comment.

Thank you for looking into it!

In D114317#3149693 , @carlosgalvezp 
wrote:

> In D114317#3149504 , 
> @salman-javed-nz wrote:
>
>> What would you say are the key differences between this patch differ and 
>> previous attempts, e.g. https://reviews.llvm.org/D72566? How does this patch 
>> address the concerns raised in the previous reviews?
>
> The key differences are:
>
> - It has been previously discussed and agreed upon via RFC in the mailing 
> list, as requested as a comment in the previous patch.
>
> - Check developers do not need to manually specify which check is aliasing 
> which check. It's automated and transparent, thus less burden on check 
> developers and less possibility for errors.



> - Only aliases with the same configuration (perfect aliases) are removed. The 
> patch you mention would run only one instance with it's configuration, so 
> "aliases with different configuration" would incorrectly not be run (if I 
> read the commit message correctly).

The leftover caveat of this approach is that it does not help the checks with 
configuration options,
because it is likely that the options were tuned, but most likely only for a 
single check.
I still maintain that there should be a single opt-in option to completely 
disable all aliasee checks.

> I don't have a strong opinion as to which patch to use to continue this work, 
> I just would like to work on making it happen. It's been a standing issue for 
> quite a long time :)




CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114317/new/

https://reviews.llvm.org/D114317

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

[PATCH] D113393: [c++2b] Implement P0849R8 auto(x)

[Zhihao Yuan via Phabricator via cfe-commits]

lichray added a comment.

In D113393#3128840 , @rsmith wrote:

> I also wonder whether we should accept `decltype(auto)(x)` as an extension, 
> but we can discuss that separately.

Implemented. To produce a reference rather than a temporary, I had to 
special-case it, but in the end, it costs fewer lines of diff.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113393/new/

https://reviews.llvm.org/D113393

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

[PATCH] D113393: [c++2b] Implement P0849R8 auto(x)

[Zhihao Yuan via Phabricator via cfe-commits]

lichray added a comment.

In D113393#3128835 , @rsmith wrote:

> It looks like we'll need some additional work on disambiguation to handle 
> cases like:
>
>   struct A { int n; } a;
>   void f() { auto()->n = 0; }
>
> I think that's valid, but right now we misparse it as a declaration of a 
> variable ``. [...]:
>
>   struct A { int n; } a;
>   using T = A*;
>   void f() { T()->n = 1; }

Solved the issue and added test cases.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113393/new/

https://reviews.llvm.org/D113393

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

[PATCH] D114454: [NFC][AIX]Disable unstable CSA tests failing on AIX

[Gabor Marton via Phabricator via cfe-commits]

martong edited reviewers, added: steakhal; removed: vsavchenko.
martong added a comment.

vsavchenko is inactive, presumably he is no longer working with CSA


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114454/new/

https://reviews.llvm.org/D114454

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

[PATCH] D113372: [Driver] Add CLANG_DEFAULT_PIE to emulate GCC --enable-default-pie

[Fangrui Song via Phabricator via cfe-commits]

MaskRay updated this revision to Diff 389305.
MaskRay added a comment.
Herald added a project: LLVM.
Herald added a subscriber: llvm-commits.

Add a release note


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113372/new/

https://reviews.llvm.org/D113372

Files:
  clang/CMakeLists.txt
  clang/docs/ReleaseNotes.rst
  clang/include/clang/Config/config.h.cmake
  clang/lib/Driver/ToolChains/Linux.cpp
  clang/test/Driver/default-pie.c
  clang/test/Driver/linux-ld.c
  clang/test/lit.cfg.py
  clang/test/lit.site.cfg.py.in
  llvm/utils/gn/secondary/clang/include/clang/Config/BUILD.gn
  utils/bazel/llvm-project-overlay/clang/include/clang/Config/config.h

Index: utils/bazel/llvm-project-overlay/clang/include/clang/Config/config.h
===
--- utils/bazel/llvm-project-overlay/clang/include/clang/Config/config.h
+++ utils/bazel/llvm-project-overlay/clang/include/clang/Config/config.h
@@ -22,6 +22,9 @@
 /* Bug report URL. */
 #define BUG_REPORT_URL "https://bugs.llvm.org/;
 
+/* Default to -fPIE and -pie. */
+#define CLANG_DEFAULT_PIE 0
+
 /* Default linker to use. */
 #define CLANG_DEFAULT_LINKER ""
 
Index: llvm/utils/gn/secondary/clang/include/clang/Config/BUILD.gn
===
--- llvm/utils/gn/secondary/clang/include/clang/Config/BUILD.gn
+++ llvm/utils/gn/secondary/clang/include/clang/Config/BUILD.gn
@@ -9,6 +9,7 @@
   output = "$target_gen_dir/config.h"
   values = [
 "BUG_REPORT_URL=https://bugs.llvm.org/;,
+"CLANG_DEFAULT_PIE=0",
 "CLANG_DEFAULT_LINKER=",
 "CLANG_DEFAULT_STD_C=",
 "CLANG_DEFAULT_STD_CXX=",
Index: clang/test/lit.site.cfg.py.in
===
--- clang/test/lit.site.cfg.py.in
+++ clang/test/lit.site.cfg.py.in
@@ -22,6 +22,7 @@
 config.llvm_use_sanitizer = "@LLVM_USE_SANITIZER@"
 config.have_zlib = @LLVM_ENABLE_ZLIB@
 config.clang_arcmt = @CLANG_ENABLE_ARCMT@
+config.clang_default_pie = "@CLANG_DEFAULT_PIE@"
 config.clang_default_cxx_stdlib = "@CLANG_DEFAULT_CXX_STDLIB@"
 config.clang_staticanalyzer = @CLANG_ENABLE_STATIC_ANALYZER@
 config.clang_staticanalyzer_z3 = "@LLVM_WITH_Z3@"
Index: clang/test/lit.cfg.py
===
--- clang/test/lit.cfg.py
+++ clang/test/lit.cfg.py
@@ -121,6 +121,9 @@
 if config.has_plugins and config.llvm_plugin_ext:
 config.available_features.add('plugins')
 
+if config.clang_default_pie == '1':
+config.available_features.add('default-pie')
+
 # Set available features we allow tests to conditionalize on.
 #
 if config.clang_default_cxx_stdlib != '':
Index: clang/test/Driver/linux-ld.c
===
--- clang/test/Driver/linux-ld.c
+++ clang/test/Driver/linux-ld.c
@@ -1,4 +1,4 @@
-// UNSUPPORTED: system-windows
+// UNSUPPORTED: system-windows, default-pie
 // General tests that ld invocations on Linux targets sane. Note that we use
 // sysroot to make these tests independent of the host system.
 //
Index: clang/test/Driver/default-pie.c
===
--- /dev/null
+++ clang/test/Driver/default-pie.c
@@ -0,0 +1,7 @@
+// REQUIRES: default-pie
+/// Test -DCLANG_DEFAULT_PIE=on.
+
+// RUN: %clang -### --target=aarch64-linux-gnu %s 2>&1 | FileCheck %s --check-prefix=PIE2
+
+// PIE2: "-mrelocation-model" "pic" "-pic-level" "2" "-pic-is-pie"
+// PIE2: "-pie"
Index: clang/lib/Driver/ToolChains/Linux.cpp
===
--- clang/lib/Driver/ToolChains/Linux.cpp
+++ clang/lib/Driver/ToolChains/Linux.cpp
@@ -663,7 +663,7 @@
 }
 
 bool Linux::isPIEDefault(const llvm::opt::ArgList ) const {
-  return getTriple().isAndroid() || getTriple().isMusl() ||
+  return CLANG_DEFAULT_PIE || getTriple().isAndroid() || getTriple().isMusl() ||
  getSanitizerArgs(Args).requiresPIE();
 }
 
Index: clang/include/clang/Config/config.h.cmake
===
--- clang/include/clang/Config/config.h.cmake
+++ clang/include/clang/Config/config.h.cmake
@@ -8,6 +8,9 @@
 /* Bug report URL. */
 #define BUG_REPORT_URL "${BUG_REPORT_URL}"
 
+/* Default to -fPIE and -pie. */
+#cmakedefine01 CLANG_DEFAULT_PIE
+
 /* Default linker to use. */
 #define CLANG_DEFAULT_LINKER "${CLANG_DEFAULT_LINKER}"
 
Index: clang/docs/ReleaseNotes.rst
===
--- clang/docs/ReleaseNotes.rst
+++ clang/docs/ReleaseNotes.rst
@@ -226,7 +226,10 @@
 Build System Changes
 
 
-- ...
+- Linux distros can specify ``-DCLANG_DEFAULT_PIE=On`` to use ``-fPIE`` and
+  ``-pie`` by default. This matches GCC installations on many Linux distros
+  (configured with ``--enable-default-pie``).
+  (`D113372 `_)
 
 AST Matchers
 

[PATCH] D113393: [c++2b] Implement P0849R8 auto(x)

[Zhihao Yuan via Phabricator via cfe-commits]

lichray updated this revision to Diff 389304.
lichray added a comment.

- Implement decltype(auto)(x) as well


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113393/new/

https://reviews.llvm.org/D113393

Files:
  clang/docs/ReleaseNotes.rst
  clang/include/clang/Basic/DiagnosticSemaKinds.td
  clang/lib/Parse/ParseDeclCXX.cpp
  clang/lib/Parse/ParseExpr.cpp
  clang/lib/Parse/ParseExprCXX.cpp
  clang/lib/Parse/ParseTentative.cpp
  clang/lib/Sema/SemaExprCXX.cpp
  clang/lib/Sema/SemaType.cpp
  clang/test/CXX/dcl.dcl/dcl.spec/dcl.type/dcl.spec.auto/p5.cpp
  clang/test/CXX/dcl.dcl/dcl.spec/dcl.type/dcl.type.auto.deduct/p2.cpp
  clang/test/CXX/expr/expr.post/expr.type.conv/p1-2b.cpp
  clang/test/CXX/expr/expr.unary/expr.new/p2-cxx0x.cpp
  clang/test/CXX/expr/expr.unary/expr.new/p2-cxx14.cpp
  clang/test/CXX/expr/expr.unary/expr.new/p2-cxx1z.cpp
  clang/test/Parser/cxx2b-auto-x.cpp
  clang/test/SemaCXX/deduced-return-type-cxx14.cpp
  clang/www/cxx_status.html

Index: clang/www/cxx_status.html
===
--- clang/www/cxx_status.html
+++ clang/www/cxx_status.html
@@ -1379,6 +1379,11 @@
   https://wg21.link/P2360R0;>P2360R0
   Clang 14
 
+
+  auto(x): decay-copy in the language
+  https://wg21.link/P0849R8;>P0849R8
+  Clang 14
+
 
 
 
Index: clang/test/SemaCXX/deduced-return-type-cxx14.cpp
===
--- clang/test/SemaCXX/deduced-return-type-cxx14.cpp
+++ clang/test/SemaCXX/deduced-return-type-cxx14.cpp
@@ -442,6 +442,13 @@
   B() : decltype(auto)() {} // expected-error {{'decltype(auto)' not allowed here}}
 };
   }
+
+  namespace Cast {
+void foo() {
+  (void)decltype(auto)(0); // cxx14_20-error{{'decltype(auto)' not allowed here}}
+  (void)decltype(auto){0}; // cxx14_20-error{{'decltype(auto)' not allowed here}}
+}
+  }
 }
 
 namespace CurrentInstantiation {
@@ -638,5 +645,7 @@
   using B = auto (*)() -> auto; // expected-error {{'auto' not allowed in type alias}}
   template auto> struct X {}; // cxx14-error {{'auto' not allowed in template parameter until C++17}}
   template struct Y { T x; };
-  Y auto> y; // expected-error {{'auto' not allowed in template argument}}
+  Y auto> y; // cxx14_20-error {{'auto' not allowed in template argument}} \
+  cxx2b-error {{initializer for functional-style cast to 'auto' is empty}} \
+  cxx2b-error {{expected unqualified-id}}
 }
Index: clang/test/Parser/cxx2b-auto-x.cpp
===
--- /dev/null
+++ clang/test/Parser/cxx2b-auto-x.cpp
@@ -0,0 +1,23 @@
+// RUN: %clang_cc1 -fsyntax-only -verify=expected,cxx2b -std=c++2b %s
+// RUN: %clang_cc1 -fsyntax-only -verify=expected,cxx20 -std=c++20 %s
+
+void looks_like_decltype_auto() {
+  decltype(auto(42)) b = 42; // cxx20-error {{'auto' not allowed here}}
+  decltype(long *) a = 42;   // expected-error {{expected '(' for function-style cast or type construction}} \
+expected-error {{expected expression}}
+  decltype(auto *) a = 42;   // expected-error {{expected '(' for function-style cast or type construction}} \
+expected-error {{expected expression}}
+  decltype(auto()) c = 42;   // cxx2b-error {{initializer for functional-style cast to 'auto' is empty}} \
+cxx20-error {{'auto' not allowed here}}
+}
+
+struct looks_like_declaration {
+  int n;
+} a;
+
+using T = looks_like_declaration *;
+void f() { T()->n = 1; }
+void g() { auto()->n = 0; } // cxx20-error {{declaration of variable 'a' with deduced type 'auto (&)' requires an initializer}} \
+ cxx20-error {{expected ';' at end of declaration}}
+void h() { auto{}->n = 0; } // cxx20-error {{expected unqualified-id}} \
+ cxx20-error {{expected expression}}
Index: clang/test/CXX/expr/expr.unary/expr.new/p2-cxx1z.cpp
===
--- clang/test/CXX/expr/expr.unary/expr.new/p2-cxx1z.cpp
+++ clang/test/CXX/expr/expr.unary/expr.new/p2-cxx1z.cpp
@@ -1,11 +1,30 @@
 // RUN: %clang_cc1 -fsyntax-only -verify %s -std=c++14
 // RUN: %clang_cc1 -fsyntax-only -verify %s -std=c++17 -pedantic
 
+// [expr.new]p2 ... the invented declaration: T x init ;
+// C++2b [dcl.type.auto.deduct]p2.2
+// For a variable declared with a type that contains a placeholder type, T is the declared type of the variable.
 void f() {
+  // - If the initializer is a parenthesized expression-list, the expression-list shall be a single assignmentexpression and E is the assignment-expression.
   new auto('a');
-  new auto {2};
-  new auto {1, 2}; // expected-error{{new expression for type 'auto' contains multiple constructor arguments}}
-  new auto {}; // expected-error{{new expression for type 

[PATCH] D114317: [clang-tidy][WIP] Do not run perfect alias checks

[Carlos Galvez via Phabricator via cfe-commits]

carlosgalvezp added a comment.

In D114317#3149504 , @salman-javed-nz 
wrote:

> What would you say are the key differences between this patch differ and 
> previous attempts, e.g. https://reviews.llvm.org/D72566? How does this patch 
> address the concerns raised in the previous reviews?

The key differences are:

- It has been previously discussed and agreed upon via RFC in the mailing list, 
as requested as a comment in the previous patch.

- Check developers do not need to manually specify which check is aliasing 
which check. It's automated and transparent, thus less burden on check 
developers and less possibility for errors.

- Only aliases with the same configuration (perfect aliases) are removed. The 
patch you mention would run only one instance with it's configuration, so 
"aliases with different configuration" would incorrectly not be run (if I read 
the commit message currently).

I don't have a strong opinion as to which patch to use to continue this work, I 
just would like work to make it happen. It's been a standing issue for quite a 
long time :)


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114317/new/

https://reviews.llvm.org/D114317

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[John McCall via Phabricator via cfe-commits]

rjmccall added inline comments.



Comment at: clang/lib/Sema/SemaChecking.cpp:208
+if (UnaryOp->getOpcode() == UnaryOperator::Opcode::UO_AddrOf)
+  E = UnaryOp->getSubExpr();
+

samitolvanen wrote:
> rjmccall wrote:
> > samitolvanen wrote:
> > > rjmccall wrote:
> > > > It would be more general to allow any expression that we can 
> > > > constant-evaluate to a specific function / member function reference.  
> > > > That allows callers to do stuff like `__builtin_function_start((int 
> > > > (A::*)() const) ::x)` to resolve overloaded function references.
> > > > 
> > > > You should delay this check if the operand is value-dependent.
> > > > It would be more general to allow any expression that we can 
> > > > constant-evaluate to a specific function / member function reference.  
> > > > That allows callers to do stuff like `__builtin_function_start((int 
> > > > (A::*)() const) ::x)` to resolve overloaded function references.
> > > 
> > > I looked into using `Expr::EvaluateAsConstantExpr` here and while it 
> > > works, I'm not sure if allowing arbitrary expressions as the argument 
> > > provides any value. We can allow resolving overloaded function references 
> > > without constant-evaluating the expression (and I added tests for this). 
> > > Did you have any other use cases in mind where this might be useful?
> > I don't see what the advantage of limiting the constant expression would be 
> > if we can constant-evaluate it.  `switch` doesn't force you to make case 
> > values be integer literals and/or references to enumerators.  What are you 
> > trying to achieve with a restriction?
> > 
> > Not having arbitrary restrictions is particularly useful in C++, where 
> > templates and `constexpr` machinery can usefully do a lot of abstraction.
> > I don't see what the advantage of limiting the constant expression would be 
> > if we can constant-evaluate it.  `switch` doesn't force you to make case 
> > values be integer literals and/or references to enumerators.  What are you 
> > trying to achieve with a restriction?
> 
> I'm trying to understand the benefit of allowing arbitrary expressions like 
> `__builtin_function_start(100 + a)`, which `EvaluateAsConstantExpr` is happy 
> to evaluate into a reference to `a`.
> 
> I can obviously understand why these should be allowed for `switch`, but here 
> we have a function whose sole purpose is to return the address of a function 
> and I'm wondering why someone would want pass anything more complicated to it.
> 
> > Not having arbitrary restrictions is particularly useful in C++, where 
> > templates and `constexpr` machinery can usefully do a lot of abstraction.
> 
> Perhaps I'm just not that familiar with the C++ use cases here. Would you be 
> able to provide me an example I could use as a test case?
> I can obviously understand why these should be allowed for switch, but here 
> we have a function whose sole purpose is to return the address of a function 
> and I'm wondering why someone would want pass anything more complicated to it.

The idea is that you might have a complicated way of picking which function you 
mean.  I agree that this should probably disallow non-zero offsets.

> Perhaps I'm just not that familiar with the C++ use cases here. Would you be 
> able to provide me an example I could use as a test case?

A simple example: `fn` after `constexpr void (*fn)() = `.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

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

[PATCH] D114082: [WIP] Normalize String Attributes

[Reid Kleckner via Phabricator via cfe-commits]

rnk added subscribers: void, rnk.
rnk added a comment.

> it is possible to list all internal keys in one place

I think one of the original goals of adding support for string attributes 
(maybe @void will give some input) was specifically to avoid having one giant 
list with all the attributes supported by all the backends. String attributes 
were initially called "target dependent attributes", or something like that. I 
think we had a `td_attrs` accessor at some point. There was this idea that LLVM 
should not know about x86-specific attributes. There should be some kind of 
indirection and delegation to the target backend. LLVM has not always succeeded 
in following this vision, see for example the target-specific intrinsic enums, 
which were one giant enum until D71320 .

I think, even if LLVM IR has to know all of the target-specific internally used 
attributes, we can try to honor that target independent vision by having 
separate headers for different target specific attributes. Does that sound 
reasonable? It's at least consistent with how we handle intrinsics 
(IntrinsicsX86.h, IntrinsicsAArch64.h etc).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114082/new/

https://reviews.llvm.org/D114082

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

[PATCH] D114394: Compile-time computation of string attribute hashes

[Reid Kleckner via Phabricator via cfe-commits]

rnk added subscribers: aeubanks, rnk.
rnk added a comment.

I think @aeubanks might be a good reviewer for this.

I like the performance wins here.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114394/new/

https://reviews.llvm.org/D114394

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

[PATCH] D114317: [clang-tidy][WIP] Do not run perfect alias checks

[Salman Javed via Phabricator via cfe-commits]

salman-javed-nz added a comment.

What would you say are the key differences between this patch differ and 
previous attempts, e.g. https://reviews.llvm.org/D72566? How does this patch 
address the concerns raised in the previous reviews?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114317/new/

https://reviews.llvm.org/D114317

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

[PATCH] D110663: [Driver] Support Debian multiarch style lib/clang/14.0.0/x86_64-linux-gnu runtime path and include/x86_64-linux-gnu/c++/v1 libc++ path

[Fangrui Song via Phabricator via cfe-commits]

MaskRay added a comment.

In D110663#3149389 , @lebedev.ri 
wrote:

> Were you able to actually reproduce the problem that lead to revert of 
> D107799 , or is this based on blind guesses?

The patch will fix the "lib/clang/14.0.0/x86_64-unknown-linux-gnu/" issue on 
Debian/Ubuntu and their derivatives:
if Debian multiarch LLVM_DEFAULT_TARGET_TRIPLE=x86_64-linux-gnu is used, the 
`-unknown` should not be present.
I have noticed the issue on my Debian testing machine.

Whether this helps the revert of D107799  
(`-DLLVM_ENABLE_PER_TARGET_RUNTIME_DIR=on`): blind guess.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110663/new/

https://reviews.llvm.org/D110663

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

[PATCH] D110663: [Driver] Support Debian multiarch style lib/clang/14.0.0/x86_64-linux-gnu runtime path and include/x86_64-linux-gnu/c++/v1 libc++ path

[Roman Lebedev via Phabricator via cfe-commits]

lebedev.ri added a comment.

Were you able to actually reproduce the problem that lead to revert of D107799 
, or is this based on blind guesses?

In D110663#3149364 , @MaskRay wrote:

> In D110663#3148690 , 
> @sylvestre.ledru wrote:
>
>> I am not sure why you are pinging this review often?! Maybe chat with people 
>> who gave you feedback before directly?!
>
> https://llvm.org/docs/CodeReview.html "If it is not urgent, the common 
> courtesy ping rate is one week." I did it less frequently than once per week.
>
> Ping @phosek
>
>> Myself I wasted so much time because of D107799 
>>  that I am very reluctant to enable 
>> LLVM_ENABLE_PER_TARGET_RUNTIME_DIR again.
>> The mismatch of the triple / quadruple between the system triple and what 
>> have been decided in llvm ( x86_64-linux-gnu vs x86_64-pc-linux-gnu)
>>
>>> Moreover, llvm-project has always had the attitude: "downstream projects 
>>> are on their own".
>>
>> I am not sure it is a correct statement. llvm project is a bunch of people 
>> with different focus. 
>> But in general, we are trying to be nice with the ecosystem and people who 
>> rely on this software.
>
> Still correct. For courtesy I typically notify projects but many changes 
> (especially IR and codegen changes) cannot really notify users simply because 
> you cannot enumerate every user and
> in some IR/codegen folks's minds this would unnecessarily slow down 
> development.
> For this case I have notified rust, which may be the most likely impacted 
> user. I think the mostly likely impacted user was probably just Fuchsia, but 
> I have added a hack to exclude Fuchsia.




Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110663/new/

https://reviews.llvm.org/D110663

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

[PATCH] D110663: [Driver] Support Debian multiarch style lib/clang/14.0.0/x86_64-linux-gnu runtime path and include/x86_64-linux-gnu/c++/v1 libc++ path

[Fangrui Song via Phabricator via cfe-commits]

MaskRay added a comment.

In D110663#3148690 , @sylvestre.ledru 
wrote:

> I am not sure why you are pinging this review often?! Maybe chat with people 
> who gave you feedback before directly?!

https://llvm.org/docs/CodeReview.html "If it is not urgent, the common courtesy 
ping rate is one week." I did it less frequently than once per week.

Ping @phosek

> Myself I wasted so much time because of D107799 
>  that I am very reluctant to enable 
> LLVM_ENABLE_PER_TARGET_RUNTIME_DIR again.
> The mismatch of the triple / quadruple between the system triple and what 
> have been decided in llvm ( x86_64-linux-gnu vs x86_64-pc-linux-gnu)
>
>> Moreover, llvm-project has always had the attitude: "downstream projects are 
>> on their own".
>
> I am not sure it is a correct statement. llvm project is a bunch of people 
> with different focus. 
> But in general, we are trying to be nice with the ecosystem and people who 
> rely on this software.

Still correct. For courtesy I typically notify projects but many changes 
(especially IR and codegen changes) cannot really notify users simply because 
you cannot enumerate every user and
in some IR/codegen folks's minds this would unnecessarily slow down development.
For this case I have notified rust, which may be the most likely impacted user. 
I think the mostly likely impacted user is probably just Fuchsia.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110663/new/

https://reviews.llvm.org/D110663

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[Sami Tolvanen via Phabricator via cfe-commits]

samitolvanen planned changes to this revision.
samitolvanen added a comment.

In D108479#3149297 , @rjmccall wrote:

> In D108479#3149228 , @samitolvanen 
> wrote:
>
>> I worked around this for now by explicitly allowing 
>> `__builtin_function_start` in `CheckLValueConstantExpression`, but this 
>> seems terribly hacky. What would be the correct way to solve this issue?
>
> Try to generalize what we do for `__builtin___CFStringMakeConstantString`.

Thanks, I'll take a look.




Comment at: clang/lib/Sema/SemaChecking.cpp:208
+if (UnaryOp->getOpcode() == UnaryOperator::Opcode::UO_AddrOf)
+  E = UnaryOp->getSubExpr();
+

rjmccall wrote:
> samitolvanen wrote:
> > rjmccall wrote:
> > > It would be more general to allow any expression that we can 
> > > constant-evaluate to a specific function / member function reference.  
> > > That allows callers to do stuff like `__builtin_function_start((int 
> > > (A::*)() const) ::x)` to resolve overloaded function references.
> > > 
> > > You should delay this check if the operand is value-dependent.
> > > It would be more general to allow any expression that we can 
> > > constant-evaluate to a specific function / member function reference.  
> > > That allows callers to do stuff like `__builtin_function_start((int 
> > > (A::*)() const) ::x)` to resolve overloaded function references.
> > 
> > I looked into using `Expr::EvaluateAsConstantExpr` here and while it works, 
> > I'm not sure if allowing arbitrary expressions as the argument provides any 
> > value. We can allow resolving overloaded function references without 
> > constant-evaluating the expression (and I added tests for this). Did you 
> > have any other use cases in mind where this might be useful?
> I don't see what the advantage of limiting the constant expression would be 
> if we can constant-evaluate it.  `switch` doesn't force you to make case 
> values be integer literals and/or references to enumerators.  What are you 
> trying to achieve with a restriction?
> 
> Not having arbitrary restrictions is particularly useful in C++, where 
> templates and `constexpr` machinery can usefully do a lot of abstraction.
> I don't see what the advantage of limiting the constant expression would be 
> if we can constant-evaluate it.  `switch` doesn't force you to make case 
> values be integer literals and/or references to enumerators.  What are you 
> trying to achieve with a restriction?

I'm trying to understand the benefit of allowing arbitrary expressions like 
`__builtin_function_start(100 + a)`, which `EvaluateAsConstantExpr` is happy to 
evaluate into a reference to `a`.

I can obviously understand why these should be allowed for `switch`, but here 
we have a function whose sole purpose is to return the address of a function 
and I'm wondering why someone would want pass anything more complicated to it.

> Not having arbitrary restrictions is particularly useful in C++, where 
> templates and `constexpr` machinery can usefully do a lot of abstraction.

Perhaps I'm just not that familiar with the C++ use cases here. Would you be 
able to provide me an example I could use as a test case?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[John McCall via Phabricator via cfe-commits]

rjmccall added a comment.

In D108479#3149228 , @samitolvanen 
wrote:

> I worked around this for now by explicitly allowing 
> `__builtin_function_start` in `CheckLValueConstantExpression`, but this seems 
> terribly hacky. What would be the correct way to solve this issue?

Try to generalize what we do for `__builtin___CFStringMakeConstantString`.




Comment at: clang/lib/Sema/SemaChecking.cpp:208
+if (UnaryOp->getOpcode() == UnaryOperator::Opcode::UO_AddrOf)
+  E = UnaryOp->getSubExpr();
+

samitolvanen wrote:
> rjmccall wrote:
> > It would be more general to allow any expression that we can 
> > constant-evaluate to a specific function / member function reference.  That 
> > allows callers to do stuff like `__builtin_function_start((int (A::*)() 
> > const) ::x)` to resolve overloaded function references.
> > 
> > You should delay this check if the operand is value-dependent.
> > It would be more general to allow any expression that we can 
> > constant-evaluate to a specific function / member function reference.  That 
> > allows callers to do stuff like `__builtin_function_start((int (A::*)() 
> > const) ::x)` to resolve overloaded function references.
> 
> I looked into using `Expr::EvaluateAsConstantExpr` here and while it works, 
> I'm not sure if allowing arbitrary expressions as the argument provides any 
> value. We can allow resolving overloaded function references without 
> constant-evaluating the expression (and I added tests for this). Did you have 
> any other use cases in mind where this might be useful?
I don't see what the advantage of limiting the constant expression would be if 
we can constant-evaluate it.  `switch` doesn't force you to make case values be 
integer literals and/or references to enumerators.  What are you trying to 
achieve with a restriction?

Not having arbitrary restrictions is particularly useful in C++, where 
templates and `constexpr` machinery can usefully do a lot of abstraction.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

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

[PATCH] D110927: [analyzer] Access stored value of a constant array through a pointer to another type

[Denys Petrov via Phabricator via cfe-commits]

ASDenysPetrov abandoned this revision.
ASDenysPetrov added a comment.

Temporary suspended this revision in favor of making a new checker 
//StrictAliasingChecker//, which would define an access to values through 
unpermited types as Unefined Behavior according to certain statements of the 
current Standard.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110927/new/

https://reviews.llvm.org/D110927

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[Sami Tolvanen via Phabricator via cfe-commits]

samitolvanen added a comment.

In D108479#3140688 , @rjmccall wrote:

> In D108479#3140638 , @samitolvanen 
> wrote:
>
>> Sure, I can take a look at how that would work. Basically, in 
>> `PointerExprEvaluator::VisitBuiltinCallExpr` we should not evaluate the 
>> l-value and just leave it at `Result.set(E)`?
>
> Yes, exactly.  Since the builtin already requires a constant operand in 
> non-dependent contexts, that should be enough.

This does indeed solve comparison issues and allows me to drop the changes to 
`APValue`, but it breaks initializing globals in C because Clang doesn't know 
the expression is a compile-time constant:

  $ cat test.c
  void a() {}
  const void *p = __builtin_function_start(a);
  $ clang -c test.c
  test.c:2:17: error: initializer element is not a compile-time constant
  const void *p = __builtin_function_start(a);
  ^~~
  1 error generated.

I worked around this for now by explicitly allowing `__builtin_function_start` 
in `CheckLValueConstantExpression`, but this seems terribly hacky. What would 
be the correct way to solve this issue?




Comment at: clang/lib/Sema/SemaChecking.cpp:208
+if (UnaryOp->getOpcode() == UnaryOperator::Opcode::UO_AddrOf)
+  E = UnaryOp->getSubExpr();
+

rjmccall wrote:
> It would be more general to allow any expression that we can 
> constant-evaluate to a specific function / member function reference.  That 
> allows callers to do stuff like `__builtin_function_start((int (A::*)() 
> const) ::x)` to resolve overloaded function references.
> 
> You should delay this check if the operand is value-dependent.
> It would be more general to allow any expression that we can 
> constant-evaluate to a specific function / member function reference.  That 
> allows callers to do stuff like `__builtin_function_start((int (A::*)() 
> const) ::x)` to resolve overloaded function references.

I looked into using `Expr::EvaluateAsConstantExpr` here and while it works, I'm 
not sure if allowing arbitrary expressions as the argument provides any value. 
We can allow resolving overloaded function references without 
constant-evaluating the expression (and I added tests for this). Did you have 
any other use cases in mind where this might be useful?



Comment at: clang/lib/Sema/SemaChecking.cpp:221
+  if (ResultType.isNull())
+return true;
+

rjmccall wrote:
> I think we decided that the result type should be `void*`.  Are there other 
> semantic checks from `CheckAddressOfOperand` that still meaningfully apply?
I thought `Sema::checkAddressOfFunctionIsAvailable` would be useful.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

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

[PATCH] D108479: [Clang] Add __builtin_addressof_nocfi

[Sami Tolvanen via Phabricator via cfe-commits]

samitolvanen updated this revision to Diff 389242.
samitolvanen marked an inline comment as done.
samitolvanen added a comment.

Refactored to avoid evaluating the expression into an l-value.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D108479/new/

https://reviews.llvm.org/D108479

Files:
  clang/docs/LanguageExtensions.rst
  clang/include/clang/AST/Expr.h
  clang/include/clang/Basic/Builtins.def
  clang/include/clang/Basic/DiagnosticSemaKinds.td
  clang/lib/AST/Expr.cpp
  clang/lib/AST/ExprConstant.cpp
  clang/lib/CodeGen/CGBuiltin.cpp
  clang/lib/CodeGen/CGExprConstant.cpp
  clang/lib/CodeGen/CodeGenModule.cpp
  clang/lib/CodeGen/CodeGenModule.h
  clang/lib/Sema/SemaChecking.cpp
  clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
  clang/test/CodeGen/builtin-function-start.cpp
  clang/test/SemaCXX/builtins.cpp

Index: clang/test/SemaCXX/builtins.cpp
===
--- clang/test/SemaCXX/builtins.cpp
+++ clang/test/SemaCXX/builtins.cpp
@@ -39,6 +39,13 @@
   S *ptmp = __builtin_addressof(S{}); // expected-error {{taking the address of a temporary}}
 }
 
+namespace function_start {
+void a(void) {}
+int n;
+void *p = __builtin_function_start(n);   // expected-error {{argument must be a function}}
+static_assert(__builtin_function_start(a) == a, ""); // expected-error {{static_assert expression is not an integral constant expression}}
+} // namespace function_start
+
 void no_ms_builtins() {
   __assume(1); // expected-error {{use of undeclared}}
   __noop(1); // expected-error {{use of undeclared}}
Index: clang/test/CodeGen/builtin-function-start.cpp
===
--- /dev/null
+++ clang/test/CodeGen/builtin-function-start.cpp
@@ -0,0 +1,51 @@
+// RUN: %clang_cc1 -triple x86_64-unknown-linux-gnu -emit-llvm -fsanitize=cfi-icall -o - %s | FileCheck %s
+
+#if !__has_builtin(__builtin_function_start)
+#error "missing __builtin_function_start"
+#endif
+
+void a(void) {}
+void b(void) {}
+void c(void *p) {}
+
+// CHECK: @e = global i8* bitcast (void ()* no_cfi @_Z1av to i8*)
+const void *e = __builtin_function_start(a);
+// CHECK: @f = global [2 x i8*] [i8* bitcast (void ()* @_Z1bv to i8*), i8* bitcast (void ()* no_cfi @_Z1bv to i8*)]
+void *f[] = {(void *)b, __builtin_function_start(b)};
+
+class A {
+public:
+  void f();
+  virtual void g();
+  static void h();
+  int i() const;
+  int i(int n) const;
+};
+
+void A::f() {}
+void A::g() {}
+void A::h() {}
+
+// CHECK: define {{.*}}i32 @_ZNK1A1iEv(%class.A* {{.*}}%this)
+int A::i() const { return 0; }
+
+// CHECK: define {{.*}}i32 @_ZNK1A1iEi(%class.A* {{.*}}%this, i32 %n)
+int A::i(int n) const { return 0; }
+
+void d(void) {
+  // CHECK: store i8* bitcast (void ()* no_cfi @_Z1bv to i8*), i8** %g
+  void *g = __builtin_function_start(b);
+  // call void @_Z1cPv(i8* bitcast (void ()* no_cfi @_Z1av to i8*))
+  c(__builtin_function_start(a));
+
+  // CHECK: store i8* bitcast (void (%class.A*)* no_cfi @_ZN1A1fEv to i8*), i8** %Af
+  void *Af = __builtin_function_start(::f);
+  // CHECK: store i8* bitcast (void (%class.A*)* no_cfi @_ZN1A1gEv to i8*), i8** %Ag
+  void *Ag = __builtin_function_start(::g);
+  // CHECK: store i8* bitcast (void ()* no_cfi @_ZN1A1hEv to i8*), i8** %Ah
+  void *Ah = __builtin_function_start(::h);
+  // CHECK: store i8* bitcast (i32 (%class.A*)* no_cfi @_ZNK1A1iEv to i8*), i8** %Ai1
+  void *Ai1 = __builtin_function_start((int(A::*)() const) & A::i);
+  // CHECK: store i8* bitcast (i32 (%class.A*, i32)* no_cfi @_ZNK1A1iEi to i8*), i8** %Ai2
+  void *Ai2 = __builtin_function_start((int(A::*)(int) const) & A::i);
+}
Index: clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
===
--- clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
+++ clang/lib/StaticAnalyzer/Checkers/BuiltinFunctionChecker.cpp
@@ -66,7 +66,8 @@
   case Builtin::BI__builtin_expect:
   case Builtin::BI__builtin_expect_with_probability:
   case Builtin::BI__builtin_assume_aligned:
-  case Builtin::BI__builtin_addressof: {
+  case Builtin::BI__builtin_addressof:
+  case Builtin::BI__builtin_function_start: {
 // For __builtin_unpredictable, __builtin_expect,
 // __builtin_expect_with_probability and __builtin_assume_aligned,
 // just return the value of the subexpression.
Index: clang/lib/Sema/SemaChecking.cpp
===
--- clang/lib/Sema/SemaChecking.cpp
+++ clang/lib/Sema/SemaChecking.cpp
@@ -195,6 +195,23 @@
   return false;
 }
 
+/// Check that the argument to __builtin_function_start is a function.
+static bool SemaBuiltinFunctionStart(Sema , CallExpr *TheCall) {
+  if (checkArgCount(S, TheCall, 1))
+return true;
+
+  const FunctionDecl *FD =
+  dyn_cast_or_null(TheCall->getArgDecl(S.getASTContext()));
+  if (!FD) {
+

[PATCH] D114454: [NFC][AIX]Disable unstable CSA tests failing on AIX

[Steven Wan via Phabricator via cfe-commits]

stevewan created this revision.
stevewan requested review of this revision.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

CSA uses bitwidth to infer the integer value type. In the ILP32 model, for 
example 32-bit AIX, any 32-bit integer type will be
considerred as `int`, which isn't always true. In these particular failed 
tests, CSA thinks the pointers should be `int`, while in fact they are long on 
AIX.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114454

Files:
  clang/unittests/StaticAnalyzer/SValTest.cpp


Index: clang/unittests/StaticAnalyzer/SValTest.cpp
===
--- clang/unittests/StaticAnalyzer/SValTest.cpp
+++ clang/unittests/StaticAnalyzer/SValTest.cpp
@@ -143,11 +143,19 @@
 // Actual tests
 
//===--===//
 
+#ifdef _AIX
+SVAL_TEST(DISABLED_GetConstType, R"(
+void foo() {
+  int x = 42;
+  int *y = nullptr;
+})") {
+#else
 SVAL_TEST(GetConstType, R"(
 void foo() {
   int x = 42;
   int *y = nullptr;
 })") {
+#endif
   SVal X = getByName("x");
   ASSERT_FALSE(X.getType(Context).isNull());
   EXPECT_EQ(Context.IntTy, X.getType(Context));
@@ -157,12 +165,21 @@
   EXPECT_EQ(Context.getUIntPtrType(), Y.getType(Context));
 }
 
+#ifdef _AIX
+SVAL_TEST(DISABLED_GetLocAsIntType, R"(
+void foo(int *x) {
+  long int a = (long int)x;
+  unsigned b = (long unsigned)
+  int c = (long int)nullptr;
+})") {
+#else
 SVAL_TEST(GetLocAsIntType, R"(
 void foo(int *x) {
   long int a = (long int)x;
   unsigned b = (long unsigned)
   int c = (long int)nullptr;
 })") {
+#endif
   SVal A = getByName("a");
   ASSERT_FALSE(A.getType(Context).isNull());
   // TODO: Turn it into signed long


Index: clang/unittests/StaticAnalyzer/SValTest.cpp
===
--- clang/unittests/StaticAnalyzer/SValTest.cpp
+++ clang/unittests/StaticAnalyzer/SValTest.cpp
@@ -143,11 +143,19 @@
 // Actual tests
 //===--===//
 
+#ifdef _AIX
+SVAL_TEST(DISABLED_GetConstType, R"(
+void foo() {
+  int x = 42;
+  int *y = nullptr;
+})") {
+#else
 SVAL_TEST(GetConstType, R"(
 void foo() {
   int x = 42;
   int *y = nullptr;
 })") {
+#endif
   SVal X = getByName("x");
   ASSERT_FALSE(X.getType(Context).isNull());
   EXPECT_EQ(Context.IntTy, X.getType(Context));
@@ -157,12 +165,21 @@
   EXPECT_EQ(Context.getUIntPtrType(), Y.getType(Context));
 }
 
+#ifdef _AIX
+SVAL_TEST(DISABLED_GetLocAsIntType, R"(
+void foo(int *x) {
+  long int a = (long int)x;
+  unsigned b = (long unsigned)
+  int c = (long int)nullptr;
+})") {
+#else
 SVAL_TEST(GetLocAsIntType, R"(
 void foo(int *x) {
   long int a = (long int)x;
   unsigned b = (long unsigned)
   int c = (long int)nullptr;
 })") {
+#endif
   SVal A = getByName("a");
   ASSERT_FALSE(A.getType(Context).isNull());
   // TODO: Turn it into signed long
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D114088: [PowerPC] Add BCD add/sub/cmp builtins

[Nemanja Ivanovic via Phabricator via cfe-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rGc933c2eb3346: [PowerPC] Add BCD add/sub/cmp builtins 
(authored by nemanjai).

Changed prior to commit:
  https://reviews.llvm.org/D114088?vs=387933=389238#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114088/new/

https://reviews.llvm.org/D114088

Files:
  llvm/include/llvm/IR/IntrinsicsPowerPC.td
  llvm/lib/Target/PowerPC/P10InstrResources.td
  llvm/lib/Target/PowerPC/P9InstrResources.td
  llvm/lib/Target/PowerPC/PPCISelDAGToDAG.cpp
  llvm/lib/Target/PowerPC/PPCInstrAltivec.td
  llvm/test/CodeGen/PowerPC/bcd-intrinsics.ll

Index: llvm/test/CodeGen/PowerPC/bcd-intrinsics.ll
===
--- /dev/null
+++ llvm/test/CodeGen/PowerPC/bcd-intrinsics.ll
@@ -0,0 +1,212 @@
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64le-unknown-linux-gnu \
+; RUN:   -mcpu=pwr10 -ppc-asm-full-reg-names \
+; RUN:   -ppc-vsr-nums-as-vr < %s | FileCheck %s
+; RUN: llc -verify-machineinstrs -mtriple=powerpc64-unknown-linux-gnu \
+; RUN:   -mcpu=pwr9 -ppc-asm-full-reg-names \
+; RUN:   -ppc-vsr-nums-as-vr < %s | FileCheck %s --check-prefix=CHECK-P9
+
+define dso_local i64 @test_invalid(<16 x i8> %a) local_unnamed_addr #0 {
+; CHECK-LABEL: test_invalid:
+; CHECK:   # %bb.0: # %entry
+; CHECK-NEXT:bcdsub. v2, v2, v2, 0
+; CHECK-NEXT:setbc r3, 4*cr6+un
+; CHECK-NEXT:extsw r3, r3
+; CHECK-NEXT:blr
+;
+; CHECK-P9-LABEL: test_invalid:
+; CHECK-P9:   # %bb.0: # %entry
+; CHECK-P9-NEXT:bcdsub. v2, v2, v2, 0
+; CHECK-P9-NEXT:mfocrf r3, 2
+; CHECK-P9-NEXT:rlwinm r3, r3, 28, 31, 31
+; CHECK-P9-NEXT:extsw r3, r3
+; CHECK-P9-NEXT:blr
+entry:
+  %0 = tail call i32 @llvm.ppc.bcdsub.p(i32 6, <16 x i8> %a, <16 x i8> %a) #2
+  %conv.i = sext i32 %0 to i64
+  ret i64 %conv.i
+}
+
+define dso_local <16 x i8> @test_add(<16 x i8> %a, <16 x i8> %b, i64 %ps) local_unnamed_addr #0 {
+; CHECK-LABEL: test_add:
+; CHECK:   # %bb.0: # %entry
+; CHECK-NEXT:bcdadd. v2, v2, v3, 1
+; CHECK-NEXT:blr
+;
+; CHECK-P9-LABEL: test_add:
+; CHECK-P9:   # %bb.0: # %entry
+; CHECK-P9-NEXT:bcdadd. v2, v2, v3, 1
+; CHECK-P9-NEXT:blr
+entry:
+  %0 = tail call <16 x i8> @llvm.ppc.bcdadd(<16 x i8> %a, <16 x i8> %b, i32 1)
+  ret <16 x i8> %0
+}
+
+define dso_local i64 @test_add_ofl(<16 x i8> %a, <16 x i8> %b, i64 %ps) local_unnamed_addr #0 {
+; CHECK-LABEL: test_add_ofl:
+; CHECK:   # %bb.0: # %entry
+; CHECK-NEXT:bcdadd. v2, v2, v3, 0
+; CHECK-NEXT:setbc r3, 4*cr6+un
+; CHECK-NEXT:extsw r3, r3
+; CHECK-NEXT:blr
+;
+; CHECK-P9-LABEL: test_add_ofl:
+; CHECK-P9:   # %bb.0: # %entry
+; CHECK-P9-NEXT:bcdadd. v2, v2, v3, 0
+; CHECK-P9-NEXT:mfocrf r3, 2
+; CHECK-P9-NEXT:rlwinm r3, r3, 28, 31, 31
+; CHECK-P9-NEXT:extsw r3, r3
+; CHECK-P9-NEXT:blr
+entry:
+  %0 = tail call i32 @llvm.ppc.bcdadd.p(i32 6, <16 x i8> %a, <16 x i8> %b) #2
+  %conv.i = sext i32 %0 to i64
+  ret i64 %conv.i
+}
+
+define dso_local <16 x i8> @test_sub(<16 x i8> %a, <16 x i8> %b, i64 %ps) local_unnamed_addr #0 {
+; CHECK-LABEL: test_sub:
+; CHECK:   # %bb.0: # %entry
+; CHECK-NEXT:bcdsub. v2, v2, v3, 0
+; CHECK-NEXT:blr
+;
+; CHECK-P9-LABEL: test_sub:
+; CHECK-P9:   # %bb.0: # %entry
+; CHECK-P9-NEXT:bcdsub. v2, v2, v3, 0
+; CHECK-P9-NEXT:blr
+entry:
+  %0 = tail call <16 x i8> @llvm.ppc.bcdsub(<16 x i8> %a, <16 x i8> %b, i32 0)
+  ret <16 x i8> %0
+}
+
+define dso_local i64 @test_sub_ofl(<16 x i8> %a, <16 x i8> %b, i64 %ps) local_unnamed_addr #0 {
+; CHECK-LABEL: test_sub_ofl:
+; CHECK:   # %bb.0: # %entry
+; CHECK-NEXT:bcdsub. v2, v2, v3, 0
+; CHECK-NEXT:setbc r3, 4*cr6+un
+; CHECK-NEXT:extsw r3, r3
+; CHECK-NEXT:blr
+;
+; CHECK-P9-LABEL: test_sub_ofl:
+; CHECK-P9:   # %bb.0: # %entry
+; CHECK-P9-NEXT:bcdsub. v2, v2, v3, 0
+; CHECK-P9-NEXT:mfocrf r3, 2
+; CHECK-P9-NEXT:rlwinm r3, r3, 28, 31, 31
+; CHECK-P9-NEXT:extsw r3, r3
+; CHECK-P9-NEXT:blr
+entry:
+  %0 = tail call i32 @llvm.ppc.bcdsub.p(i32 6, <16 x i8> %a, <16 x i8> %b) #2
+  %conv.i = sext i32 %0 to i64
+  ret i64 %conv.i
+}
+
+define dso_local i64 @test_cmplt(<16 x i8> %a, <16 x i8> %b) local_unnamed_addr #0 {
+; CHECK-LABEL: test_cmplt:
+; CHECK:   # %bb.0: # %entry
+; CHECK-NEXT:bcdsub. v2, v2, v3, 0
+; CHECK-NEXT:setbc r3, 4*cr6+lt
+; CHECK-NEXT:extsw r3, r3
+; CHECK-NEXT:blr
+;
+; CHECK-P9-LABEL: test_cmplt:
+; CHECK-P9:   # %bb.0: # %entry
+; CHECK-P9-NEXT:bcdsub. v2, v2, v3, 0
+; CHECK-P9-NEXT:mfocrf r3, 2
+; CHECK-P9-NEXT:rlwinm r3, r3, 25, 31, 31
+; CHECK-P9-NEXT:extsw r3, r3
+; CHECK-P9-NEXT:blr
+entry:
+  %0 = tail call i32 @llvm.ppc.bcdsub.p(i32 2, <16 x i8> %a, <16 x i8> %b) #2
+  %conv.i = sext i32 %0 to i64
+  ret i64 %conv.i
+}
+
+define 

[PATCH] D114439: [Annotation] Allow parameter pack expansions in annotate attribute

[Tyker via Phabricator via cfe-commits]

Tyker added a comment.

This seems like a good change to me. but i don't think my approval is enough


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114439/new/

https://reviews.llvm.org/D114439

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

[PATCH] D114058: [clangd] Add ObjC method support to prepareCallHierarchy

[Kadir Cetinkaya via Phabricator via cfe-commits]

kadircet added inline comments.



Comment at: clang-tools-extra/clangd/unittests/CallHierarchyTests.cpp:54
 
+void verifyIncomingMultiFile(std::string SourceExt, std::string HeaderExt,
+ Annotations , Annotations ,

nridge wrote:
> nit: rather than passing `std::string` by value, let's use `llvm::StringRef`
> 
> (`const std::string&` would also be fine, but I think `llvm::StringRef` is 
> more conventional in this codebase)
i am not sure if these helpers makes tests easier to read (i believe they're 
actually making it harder).

please inline them back. if we want to have better infrastructure for testing 
this, we need to do it in its own patch (and I definitely agree that we need a 
better infra for testing here. but we need to think about something that can 
generalize rather than forcing a particular pattern).


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114058/new/

https://reviews.llvm.org/D114058

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

[PATCH] D113491: [HIP] Fix device stub name for Windows

[Yaxun Liu via Phabricator via cfe-commits]

This revision was landed with ongoing or failed builds.
This revision was automatically updated to reflect the committed changes.
yaxunl marked an inline comment as done.
Closed by commit rG38211bbab1d9: [HIP] Fix device stub name for Windows 
(authored by yaxunl).
Herald added a project: clang.

Changed prior to commit:
  https://reviews.llvm.org/D113491?vs=385836=389228#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113491/new/

https://reviews.llvm.org/D113491

Files:
  clang/include/clang/AST/GlobalDecl.h
  clang/lib/AST/MicrosoftMangle.cpp
  clang/test/CodeGenCUDA/kernel-stub-name.cu


Index: clang/test/CodeGenCUDA/kernel-stub-name.cu
===
--- clang/test/CodeGenCUDA/kernel-stub-name.cu
+++ clang/test/CodeGenCUDA/kernel-stub-name.cu
@@ -28,8 +28,8 @@
 // GNU: @[[HDKERN:_Z11kernel_declv]] = external constant void ()*, align 8
 
 // MSVC: @[[HCKERN:ckernel]] = dso_local constant void ()* 
@[[CSTUB:__device_stub__ckernel]], align 8
-// MSVC: @[[HNSKERN:"\?nskernel@ns@@YAXXZ.*"]] = dso_local constant void ()* 
@[[NSSTUB:"\?nskernel@ns@@YAXXZ"]], align 8
-// MSVC: @[[HTKERN:"\?\?\$kernelfunc@H@@YAXXZ.*"]] = linkonce_odr dso_local 
constant void ()* @[[TSTUB:"\?\?\$kernelfunc@H@@YAXXZ.*"]], comdat, align 8
+// MSVC: @[[HNSKERN:"\?nskernel@ns@@YAXXZ.*"]] = dso_local constant void ()* 
@[[NSSTUB:"\?__device_stub__nskernel@ns@@YAXXZ"]], align 8
+// MSVC: @[[HTKERN:"\?\?\$kernelfunc@H@@YAXXZ.*"]] = linkonce_odr dso_local 
constant void ()* @[[TSTUB:"\?\?\$__device_stub__kernelfunc@H@@YAXXZ.*"]], 
comdat, align 8
 // MSVC: @[[HDKERN:"\?kernel_decl@@YAXXZ.*"]] = external dso_local constant 
void ()*, align 8
 
 extern "C" __global__ void ckernel() {}
@@ -69,7 +69,7 @@
 // CHECK: call void @[[NSSTUB]]()
 // CHECK: call void @[[TSTUB]]()
 // GNU: call void @[[DSTUB:_Z26__device_stub__kernel_declv]]()
-// MSVC: call void @[[DSTUB:"\?kernel_decl@@YAXXZ"]]()
+// MSVC: call void @[[DSTUB:"\?__device_stub__kernel_decl@@YAXXZ"]]()
 
 extern "C" void fun1(void) {
   ckernel<<<1, 1>>>();
Index: clang/lib/AST/MicrosoftMangle.cpp
===
--- clang/lib/AST/MicrosoftMangle.cpp
+++ clang/lib/AST/MicrosoftMangle.cpp
@@ -962,7 +962,19 @@
   switch (Name.getNameKind()) {
 case DeclarationName::Identifier: {
   if (const IdentifierInfo *II = Name.getAsIdentifierInfo()) {
-mangleSourceName(II->getName());
+bool IsDeviceStub =
+ND &&
+((isa(ND) && ND->hasAttr()) ||
+ (isa(ND) &&
+  cast(ND)
+  ->getTemplatedDecl()
+  ->hasAttr())) &&
+GD.getKernelReferenceKind() == KernelReferenceKind::Stub;
+if (IsDeviceStub)
+  mangleSourceName(
+  (llvm::Twine("__device_stub__") + II->getName()).str());
+else
+  mangleSourceName(II->getName());
 break;
   }
 
Index: clang/include/clang/AST/GlobalDecl.h
===
--- clang/include/clang/AST/GlobalDecl.h
+++ clang/include/clang/AST/GlobalDecl.h
@@ -18,6 +18,7 @@
 #include "clang/AST/DeclCXX.h"
 #include "clang/AST/DeclObjC.h"
 #include "clang/AST/DeclOpenMP.h"
+#include "clang/AST/DeclTemplate.h"
 #include "clang/Basic/ABI.h"
 #include "clang/Basic/LLVM.h"
 #include "llvm/ADT/DenseMapInfo.h"
@@ -129,8 +130,12 @@
   }
 
   KernelReferenceKind getKernelReferenceKind() const {
-assert(isa(getDecl()) &&
-   cast(getDecl())->hasAttr() &&
+assert(((isa(getDecl()) &&
+ cast(getDecl())->hasAttr()) ||
+(isa(getDecl()) &&
+ cast(getDecl())
+ ->getTemplatedDecl()
+ ->hasAttr())) &&
"Decl is not a GPU kernel!");
 return static_cast(Value.getInt());
   }


Index: clang/test/CodeGenCUDA/kernel-stub-name.cu
===
--- clang/test/CodeGenCUDA/kernel-stub-name.cu
+++ clang/test/CodeGenCUDA/kernel-stub-name.cu
@@ -28,8 +28,8 @@
 // GNU: @[[HDKERN:_Z11kernel_declv]] = external constant void ()*, align 8
 
 // MSVC: @[[HCKERN:ckernel]] = dso_local constant void ()* @[[CSTUB:__device_stub__ckernel]], align 8
-// MSVC: @[[HNSKERN:"\?nskernel@ns@@YAXXZ.*"]] = dso_local constant void ()* @[[NSSTUB:"\?nskernel@ns@@YAXXZ"]], align 8
-// MSVC: @[[HTKERN:"\?\?\$kernelfunc@H@@YAXXZ.*"]] = linkonce_odr dso_local constant void ()* @[[TSTUB:"\?\?\$kernelfunc@H@@YAXXZ.*"]], comdat, align 8
+// MSVC: @[[HNSKERN:"\?nskernel@ns@@YAXXZ.*"]] = dso_local constant void ()* @[[NSSTUB:"\?__device_stub__nskernel@ns@@YAXXZ"]], align 8
+// MSVC: @[[HTKERN:"\?\?\$kernelfunc@H@@YAXXZ.*"]] = linkonce_odr dso_local constant void ()* @[[TSTUB:"\?\?\$__device_stub__kernelfunc@H@@YAXXZ.*"]], comdat, align 8
 // MSVC: @[[HDKERN:"\?kernel_decl@@YAXXZ.*"]] = external dso_local constant void ()*, 

[clang] 38211bb - [HIP] Fix device stub name for Windows

[Yaxun Liu via cfe-commits]

Author: Yaxun (Sam) Liu
Date: 2021-11-23T12:03:49-05:00
New Revision: 38211bbab1d949f682271abba0171424a5a335ab

URL: 
https://github.com/llvm/llvm-project/commit/38211bbab1d949f682271abba0171424a5a335ab
DIFF: 
https://github.com/llvm/llvm-project/commit/38211bbab1d949f682271abba0171424a5a335ab.diff

LOG: [HIP] Fix device stub name for Windows

This is a follow up of https://reviews.llvm.org/D68578
where device stub name is changed for Itanium
mangling but not Microsoft mangling.

Reviewed by: Artem Belevich

Differential Revision: https://reviews.llvm.org/D113491

Added: 


Modified: 
clang/include/clang/AST/GlobalDecl.h
clang/lib/AST/MicrosoftMangle.cpp
clang/test/CodeGenCUDA/kernel-stub-name.cu

Removed: 




diff  --git a/clang/include/clang/AST/GlobalDecl.h 
b/clang/include/clang/AST/GlobalDecl.h
index 8cb56fb4ae90b..88abba28c991d 100644
--- a/clang/include/clang/AST/GlobalDecl.h
+++ b/clang/include/clang/AST/GlobalDecl.h
@@ -18,6 +18,7 @@
 #include "clang/AST/DeclCXX.h"
 #include "clang/AST/DeclObjC.h"
 #include "clang/AST/DeclOpenMP.h"
+#include "clang/AST/DeclTemplate.h"
 #include "clang/Basic/ABI.h"
 #include "clang/Basic/LLVM.h"
 #include "llvm/ADT/DenseMapInfo.h"
@@ -129,8 +130,12 @@ class GlobalDecl {
   }
 
   KernelReferenceKind getKernelReferenceKind() const {
-assert(isa(getDecl()) &&
-   cast(getDecl())->hasAttr() &&
+assert(((isa(getDecl()) &&
+ cast(getDecl())->hasAttr()) ||
+(isa(getDecl()) &&
+ cast(getDecl())
+ ->getTemplatedDecl()
+ ->hasAttr())) &&
"Decl is not a GPU kernel!");
 return static_cast(Value.getInt());
   }

diff  --git a/clang/lib/AST/MicrosoftMangle.cpp 
b/clang/lib/AST/MicrosoftMangle.cpp
index beca1be78d99e..79a448a2435cd 100644
--- a/clang/lib/AST/MicrosoftMangle.cpp
+++ b/clang/lib/AST/MicrosoftMangle.cpp
@@ -962,7 +962,19 @@ void 
MicrosoftCXXNameMangler::mangleUnqualifiedName(GlobalDecl GD,
   switch (Name.getNameKind()) {
 case DeclarationName::Identifier: {
   if (const IdentifierInfo *II = Name.getAsIdentifierInfo()) {
-mangleSourceName(II->getName());
+bool IsDeviceStub =
+ND &&
+((isa(ND) && ND->hasAttr()) ||
+ (isa(ND) &&
+  cast(ND)
+  ->getTemplatedDecl()
+  ->hasAttr())) &&
+GD.getKernelReferenceKind() == KernelReferenceKind::Stub;
+if (IsDeviceStub)
+  mangleSourceName(
+  (llvm::Twine("__device_stub__") + II->getName()).str());
+else
+  mangleSourceName(II->getName());
 break;
   }
 

diff  --git a/clang/test/CodeGenCUDA/kernel-stub-name.cu 
b/clang/test/CodeGenCUDA/kernel-stub-name.cu
index 8e82c3612e323..71856bee98902 100644
--- a/clang/test/CodeGenCUDA/kernel-stub-name.cu
+++ b/clang/test/CodeGenCUDA/kernel-stub-name.cu
@@ -28,8 +28,8 @@
 // GNU: @[[HDKERN:_Z11kernel_declv]] = external constant void ()*, align 8
 
 // MSVC: @[[HCKERN:ckernel]] = dso_local constant void ()* 
@[[CSTUB:__device_stub__ckernel]], align 8
-// MSVC: @[[HNSKERN:"\?nskernel@ns@@YAXXZ.*"]] = dso_local constant void ()* 
@[[NSSTUB:"\?nskernel@ns@@YAXXZ"]], align 8
-// MSVC: @[[HTKERN:"\?\?\$kernelfunc@H@@YAXXZ.*"]] = linkonce_odr dso_local 
constant void ()* @[[TSTUB:"\?\?\$kernelfunc@H@@YAXXZ.*"]], comdat, align 8
+// MSVC: @[[HNSKERN:"\?nskernel@ns@@YAXXZ.*"]] = dso_local constant void ()* 
@[[NSSTUB:"\?__device_stub__nskernel@ns@@YAXXZ"]], align 8
+// MSVC: @[[HTKERN:"\?\?\$kernelfunc@H@@YAXXZ.*"]] = linkonce_odr dso_local 
constant void ()* @[[TSTUB:"\?\?\$__device_stub__kernelfunc@H@@YAXXZ.*"]], 
comdat, align 8
 // MSVC: @[[HDKERN:"\?kernel_decl@@YAXXZ.*"]] = external dso_local constant 
void ()*, align 8
 
 extern "C" __global__ void ckernel() {}
@@ -69,7 +69,7 @@ extern "C" void launch(void *kern);
 // CHECK: call void @[[NSSTUB]]()
 // CHECK: call void @[[TSTUB]]()
 // GNU: call void @[[DSTUB:_Z26__device_stub__kernel_declv]]()
-// MSVC: call void @[[DSTUB:"\?kernel_decl@@YAXXZ"]]()
+// MSVC: call void @[[DSTUB:"\?__device_stub__kernel_decl@@YAXXZ"]]()
 
 extern "C" void fun1(void) {
   ckernel<<<1, 1>>>();



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

[PATCH] D114256: [clang-tidy] Fix crashing altera-struct-pack-align on invalid RecordDecls

[Balázs Benics via Phabricator via cfe-commits]

steakhal added a comment.

In D114256#3149009 , @martong wrote:

> What happens if this checker runs on a forward declared class?

The matcher `recordDecl(isStruct(), **isDefinition()**, 
unless(isExpansionInSystemHeader()))` filters that case.

Thanks for the review @martong!




Comment at: clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp:54
 
+  // Ignore invalid decls to prevent crashing on calling `getASTRecordLayout`.
+  if (Struct->isInvalidDecl())

martong wrote:
> I'd mention first that packing and alignment info are meaningless for invalid 
> declarations.
Makes sense.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114256/new/

https://reviews.llvm.org/D114256

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

[PATCH] D114256: [clang-tidy] Fix crashing altera-struct-pack-align on invalid RecordDecls

[Balázs Benics via Phabricator via cfe-commits]

steakhal updated this revision to Diff 389225.
steakhal marked an inline comment as done.
steakhal added a comment.

- update comments
- move `no-crash` comment inline to the class member declaration


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114256/new/

https://reviews.llvm.org/D114256

Files:
  clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
  
clang-tools-extra/test/clang-tidy/checkers/altera-struct-pack-align-invalid-decl-no-crash.cpp


Index: 
clang-tools-extra/test/clang-tidy/checkers/altera-struct-pack-align-invalid-decl-no-crash.cpp
===
--- /dev/null
+++ 
clang-tools-extra/test/clang-tidy/checkers/altera-struct-pack-align-invalid-decl-no-crash.cpp
@@ -0,0 +1,6 @@
+// RUN: %check_clang_tidy -expect-clang-tidy-error %s altera-struct-pack-align 
%t -- -header-filter=.*
+
+struct Foo {
+  member; // no-crash
+};
+// CHECK-MESSAGES: :[[@LINE-2]]:3: error: C++ requires a type specifier for 
all declarations [clang-diagnostic-error]
Index: clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
===
--- clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
+++ clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
@@ -51,6 +51,10 @@
   if (Struct->isTemplated())
  return;
 
+  // Packing and alignment requirements for invalid decls are meaningless.
+  if (Struct->isInvalidDecl())
+return;
+
   // Get sizing info for the struct.
   llvm::SmallVector, 10> FieldSizes;
   unsigned int TotalBitSize = 0;


Index: clang-tools-extra/test/clang-tidy/checkers/altera-struct-pack-align-invalid-decl-no-crash.cpp
===
--- /dev/null
+++ clang-tools-extra/test/clang-tidy/checkers/altera-struct-pack-align-invalid-decl-no-crash.cpp
@@ -0,0 +1,6 @@
+// RUN: %check_clang_tidy -expect-clang-tidy-error %s altera-struct-pack-align %t -- -header-filter=.*
+
+struct Foo {
+  member; // no-crash
+};
+// CHECK-MESSAGES: :[[@LINE-2]]:3: error: C++ requires a type specifier for all declarations [clang-diagnostic-error]
Index: clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
===
--- clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
+++ clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp
@@ -51,6 +51,10 @@
   if (Struct->isTemplated())
  return;
 
+  // Packing and alignment requirements for invalid decls are meaningless.
+  if (Struct->isInvalidDecl())
+return;
+
   // Get sizing info for the struct.
   llvm::SmallVector, 10> FieldSizes;
   unsigned int TotalBitSize = 0;
___
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Balázs Benics via Phabricator via cfe-commits]

steakhal added inline comments.



Comment at: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:1105
 
-  // FIXME: Add support for SymExprs.
   return nullptr;

martong wrote:
> steakhal wrote:
> > Where did you address this FIXME?
> I didn't, but this `FIXME` became obsolete at some point in the past. 
> 
> The reasons:
> 1) We do support `SymExprs`. In `simplifySVal` we have a full blown SymExpr 
> visitor implemented that checks for constant values in the symbol subtrees. 
> And at L1103 we query the top symbol of tree as well.
> 
> 2) The git history shows that the FIXME was previously this: `// FIXME: Add 
> support for SymExprs in RangeConstraintManager.` But we do support SymExprs 
> already in RangeConstraintManager ...
> 
Thanks.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

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

[PATCH] D114256: [clang-tidy] Fix crashing altera-struct-pack-align on invalid RecordDecls

[Gabor Marton via Phabricator via cfe-commits]

martong added a comment.

What happens if this checker runs on a forward declared class?

  struct Foo;

I'd expect the pack/alignment info is missing also in that case.




Comment at: clang-tools-extra/clang-tidy/altera/StructPackAlignCheck.cpp:54
 
+  // Ignore invalid decls to prevent crashing on calling `getASTRecordLayout`.
+  if (Struct->isInvalidDecl())

I'd mention first that packing and alignment info are meaningless for invalid 
declarations.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114256/new/

https://reviews.llvm.org/D114256

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

[PATCH] D114446: [clang] Fix wrong -Wunused-local-typedef warning if use is a dependent qialified identifier

[Kristina Bessonova via Phabricator via cfe-commits]

krisb created this revision.
krisb added reviewers: thakis, rtrieu, rsmith.
krisb requested review of this revision.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

Attempt to fix Tom Smeding's example from 
https://bugs.llvm.org/show_bug.cgi?id=24883.

Given the case like this one:

  struct A {
void f() const {}
  };
  
  template 
  void handler(const T ) {
using a_type_t = A;
item.a_type_t::f();
  }
  
  int main() {
handler(A());
  }

there is no way to know whether the typedef is used or not before
the templated context is instantiated.

Having this the patch proposes deffering all the diagnostics for
typedefs defined within a dependent context.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114446

Files:
  clang/lib/Sema/SemaDecl.cpp
  clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
  clang/test/SemaCXX/warn-unused-local-typedef.cpp


Index: clang/test/SemaCXX/warn-unused-local-typedef.cpp
===
--- clang/test/SemaCXX/warn-unused-local-typedef.cpp
+++ clang/test/SemaCXX/warn-unused-local-typedef.cpp
@@ -255,5 +255,20 @@
 }
 } // TypedefInLocalClassOfTemplateClassMember
 
+namespace TypedefInDependentQualifiedIdentifier {
+struct A { void f() const {} };
+
+template 
+void handler(const T ) {
+   using a_type_t = A; // no-diag
+  using b_type_t = A; // expected-warning {{unused type alias 'b_type_t'}}
+   item.a_type_t::f();
+}
+
+void foo() {
+   handler(A());
+}
+} // TypedefInDependentQualifiedIdentifier
+
 // This should not disable any warnings:
 #pragma clang diagnostic ignored "-Wunused-local-typedef"
Index: clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
===
--- clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
+++ clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
@@ -955,6 +955,14 @@
   Typedef->setAccess(D->getAccess());
   Typedef->setReferenced(D->isReferenced());
 
+  // Diagnose unused local typedefs here since diagnostics for typedefs
+  // in a dependent context were deffered).
+  bool ShouldDiagnoseUnused = Typedef->getDeclContext()->isFunctionOrMethod();
+  if (const auto *R = dyn_cast(Typedef->getDeclContext()))
+ShouldDiagnoseUnused = ShouldDiagnoseUnused || R->isLocalClass();
+  if (!Typedef->isInvalidDecl() && ShouldDiagnoseUnused)
+SemaRef.DiagnoseUnusedDecl(Typedef);
+
   return Typedef;
 }
 
@@ -1907,8 +1915,6 @@
 LocalInstantiations.perform();
   }
 
-  SemaRef.DiagnoseUnusedNestedTypedefs(Record);
-
   if (IsInjectedClassName)
 assert(Record->isInjectedClassName() && "Broken injected-class-name");
 
Index: clang/lib/Sema/SemaDecl.cpp
===
--- clang/lib/Sema/SemaDecl.cpp
+++ clang/lib/Sema/SemaDecl.cpp
@@ -1794,16 +1794,19 @@
 
   // Except for labels, we only care about unused decls that are local to
   // functions.
-  bool WithinFunction = D->getDeclContext()->isFunctionOrMethod();
-  if (const auto *R = dyn_cast(D->getDeclContext()))
-// For dependent types, the diagnostic is deferred.
-WithinFunction =
-WithinFunction || (R->isLocalClass() && !R->isDependentType());
+  auto *Context = D->getDeclContext();
+  bool WithinFunction = Context->isFunctionOrMethod();
+  if (const auto *R = dyn_cast(Context))
+WithinFunction = WithinFunction || R->isLocalClass();
   if (!WithinFunction)
 return false;
 
-  if (isa(D))
+  if (const auto *TD = dyn_cast(D)) {
+// Defer warnings for typedefs within a dependent context.
+if (Context->isDependentContext())
+  return false;
 return true;
+  }
 
   // White-list anything that isn't a local variable.
   if (!isa(D) || isa(D) || isa(D))


Index: clang/test/SemaCXX/warn-unused-local-typedef.cpp
===
--- clang/test/SemaCXX/warn-unused-local-typedef.cpp
+++ clang/test/SemaCXX/warn-unused-local-typedef.cpp
@@ -255,5 +255,20 @@
 }
 } // TypedefInLocalClassOfTemplateClassMember
 
+namespace TypedefInDependentQualifiedIdentifier {
+struct A { void f() const {} };
+
+template 
+void handler(const T ) {
+	using a_type_t = A; // no-diag
+  using b_type_t = A; // expected-warning {{unused type alias 'b_type_t'}}
+	item.a_type_t::f();
+}
+
+void foo() {
+	handler(A());
+}
+} // TypedefInDependentQualifiedIdentifier
+
 // This should not disable any warnings:
 #pragma clang diagnostic ignored "-Wunused-local-typedef"
Index: clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
===
--- clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
+++ clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
@@ -955,6 +955,14 @@
   Typedef->setAccess(D->getAccess());
   Typedef->setReferenced(D->isReferenced());
 
+  // Diagnose unused local typedefs here since diagnostics for typedefs
+  // in a dependent context were deffered).
+  bool 

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Gabor Marton via Phabricator via cfe-commits]

martong marked an inline comment as done.
martong added inline comments.



Comment at: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:1105
 
-  // FIXME: Add support for SymExprs.
   return nullptr;

steakhal wrote:
> Where did you address this FIXME?
I didn't, but this `FIXME` became obsolete at some point in the past. 

The reasons:
1) We do support `SymExprs`. In `simplifySVal` we have a full blown SymExpr 
visitor implemented that checks for constant values in the symbol subtrees. And 
at L1103 we query the top symbol of tree as well.

2) The git history shows that the FIXME was previously this: `// FIXME: Add 
support for SymExprs in RangeConstraintManager.` But we do support SymExprs 
already in RangeConstraintManager ...



Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

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

[PATCH] D113490: [NFC] Let Microsoft mangler accept GlobalDecl

[Yaxun Liu via Phabricator via cfe-commits]

This revision was landed with ongoing or failed builds.
This revision was automatically updated to reflect the committed changes.
yaxunl marked an inline comment as done.
Closed by commit rGb472bd855ed8: [NFC] Let Microsoft mangler accept GlobalDecl 
(authored by yaxunl).
Herald added a project: clang.

Changed prior to commit:
  https://reviews.llvm.org/D113490?vs=385834=389217#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113490/new/

https://reviews.llvm.org/D113490

Files:
  clang/lib/AST/MicrosoftMangle.cpp

Index: clang/lib/AST/MicrosoftMangle.cpp
===
--- clang/lib/AST/MicrosoftMangle.cpp
+++ clang/lib/AST/MicrosoftMangle.cpp
@@ -21,6 +21,7 @@
 #include "clang/AST/DeclTemplate.h"
 #include "clang/AST/Expr.h"
 #include "clang/AST/ExprCXX.h"
+#include "clang/AST/GlobalDecl.h"
 #include "clang/AST/Mangle.h"
 #include "clang/AST/VTableBuilder.h"
 #include "clang/Basic/ABI.h"
@@ -39,6 +40,18 @@
 
 namespace {
 
+// Get GlobalDecl of DeclContext of local entities.
+static GlobalDecl getGlobalDeclAsDeclContext(const DeclContext *DC) {
+  GlobalDecl GD;
+  if (auto *CD = dyn_cast(DC))
+GD = GlobalDecl(CD, Ctor_Complete);
+  else if (auto *DD = dyn_cast(DC))
+GD = GlobalDecl(DD, Dtor_Complete);
+  else
+GD = GlobalDecl(cast(DC));
+  return GD;
+}
+
 struct msvc_hashing_ostream : public llvm::raw_svector_ostream {
   raw_ostream 
   llvm::SmallString<64> Buffer;
@@ -345,9 +358,9 @@
 
   raw_ostream () const { return Out; }
 
-  void mangle(const NamedDecl *D, StringRef Prefix = "?");
-  void mangleName(const NamedDecl *ND);
-  void mangleFunctionEncoding(const FunctionDecl *FD, bool ShouldMangle);
+  void mangle(GlobalDecl GD, StringRef Prefix = "?");
+  void mangleName(GlobalDecl GD);
+  void mangleFunctionEncoding(GlobalDecl GD, bool ShouldMangle);
   void mangleVariableEncoding(const VarDecl *VD);
   void mangleMemberDataPointer(const CXXRecordDecl *RD, const ValueDecl *VD,
StringRef Prefix = "$");
@@ -370,7 +383,7 @@
   const FunctionDecl *D = nullptr,
   bool ForceThisQuals = false,
   bool MangleExceptionSpec = true);
-  void mangleNestedName(const NamedDecl *ND);
+  void mangleNestedName(GlobalDecl GD);
 
 private:
   bool isStructorDecl(const NamedDecl *ND) const {
@@ -384,10 +397,10 @@
   AddrSpace == LangAS::ptr32_uptr));
   }
 
-  void mangleUnqualifiedName(const NamedDecl *ND) {
-mangleUnqualifiedName(ND, ND->getDeclName());
+  void mangleUnqualifiedName(GlobalDecl GD) {
+mangleUnqualifiedName(GD, cast(GD.getDecl())->getDeclName());
   }
-  void mangleUnqualifiedName(const NamedDecl *ND, DeclarationName Name);
+  void mangleUnqualifiedName(GlobalDecl GD, DeclarationName Name);
   void mangleSourceName(StringRef Name);
   void mangleOperatorName(OverloadedOperatorKind OO, SourceLocation Loc);
   void mangleCXXDtorType(CXXDtorType T);
@@ -396,9 +409,9 @@
   void manglePointerCVQualifiers(Qualifiers Quals);
   void manglePointerExtQualifiers(Qualifiers Quals, QualType PointeeType);
 
-  void mangleUnscopedTemplateName(const TemplateDecl *ND);
+  void mangleUnscopedTemplateName(GlobalDecl GD);
   void
-  mangleTemplateInstantiationName(const TemplateDecl *TD,
+  mangleTemplateInstantiationName(GlobalDecl GD,
   const TemplateArgumentList );
   void mangleObjCMethodName(const ObjCMethodDecl *MD);
 
@@ -533,7 +546,8 @@
   return true;
 }
 
-void MicrosoftCXXNameMangler::mangle(const NamedDecl *D, StringRef Prefix) {
+void MicrosoftCXXNameMangler::mangle(GlobalDecl GD, StringRef Prefix) {
+  const NamedDecl *D = cast(GD.getDecl());
   // MSVC doesn't mangle C++ names the same way it mangles extern "C" names.
   // Therefore it's really important that we don't decorate the
   // name with leading underscores or leading/trailing at signs. So, by
@@ -542,9 +556,9 @@
 
   //  ::= ?  
   Out << Prefix;
-  mangleName(D);
+  mangleName(GD);
   if (const FunctionDecl *FD = dyn_cast(D))
-mangleFunctionEncoding(FD, Context.shouldMangleDeclName(FD));
+mangleFunctionEncoding(GD, Context.shouldMangleDeclName(FD));
   else if (const VarDecl *VD = dyn_cast(D))
 mangleVariableEncoding(VD);
   else if (isa(D))
@@ -558,8 +572,9 @@
 llvm_unreachable("Tried to mangle unexpected NamedDecl!");
 }
 
-void MicrosoftCXXNameMangler::mangleFunctionEncoding(const FunctionDecl *FD,
+void MicrosoftCXXNameMangler::mangleFunctionEncoding(GlobalDecl GD,
  bool ShouldMangle) {
+  const FunctionDecl *FD = cast(GD.getDecl());
   //  ::=  
 
   // Since MSVC operates on the type as written and not the canonical type, it
@@ -770,13 +785,13 @@
   mangleCallingConvention(MD->getType()->castAs());
 }
 
-void MicrosoftCXXNameMangler::mangleName(const NamedDecl *ND) {
+void 

[clang] b472bd8 - [NFC] Let Microsoft mangler accept GlobalDecl

[Yaxun Liu via cfe-commits]

Author: Yaxun (Sam) Liu
Date: 2021-11-23T10:59:26-05:00
New Revision: b472bd855ed85691d0d03ef1808c82b780d23721

URL: 
https://github.com/llvm/llvm-project/commit/b472bd855ed85691d0d03ef1808c82b780d23721
DIFF: 
https://github.com/llvm/llvm-project/commit/b472bd855ed85691d0d03ef1808c82b780d23721.diff

LOG: [NFC] Let Microsoft mangler accept GlobalDecl

This is a follow up of https://reviews.llvm.org/D75700
where support of GlobalDecl with Microsoft mangler
is incomplete.

Reviewed by: Artem Belevich, Reid Kleckner

Differential Revision: https://reviews.llvm.org/D113490

Added: 


Modified: 
clang/lib/AST/MicrosoftMangle.cpp

Removed: 




diff  --git a/clang/lib/AST/MicrosoftMangle.cpp 
b/clang/lib/AST/MicrosoftMangle.cpp
index 163d4e95386e..beca1be78d99 100644
--- a/clang/lib/AST/MicrosoftMangle.cpp
+++ b/clang/lib/AST/MicrosoftMangle.cpp
@@ -21,6 +21,7 @@
 #include "clang/AST/DeclTemplate.h"
 #include "clang/AST/Expr.h"
 #include "clang/AST/ExprCXX.h"
+#include "clang/AST/GlobalDecl.h"
 #include "clang/AST/Mangle.h"
 #include "clang/AST/VTableBuilder.h"
 #include "clang/Basic/ABI.h"
@@ -39,6 +40,18 @@ using namespace clang;
 
 namespace {
 
+// Get GlobalDecl of DeclContext of local entities.
+static GlobalDecl getGlobalDeclAsDeclContext(const DeclContext *DC) {
+  GlobalDecl GD;
+  if (auto *CD = dyn_cast(DC))
+GD = GlobalDecl(CD, Ctor_Complete);
+  else if (auto *DD = dyn_cast(DC))
+GD = GlobalDecl(DD, Dtor_Complete);
+  else
+GD = GlobalDecl(cast(DC));
+  return GD;
+}
+
 struct msvc_hashing_ostream : public llvm::raw_svector_ostream {
   raw_ostream 
   llvm::SmallString<64> Buffer;
@@ -345,9 +358,9 @@ class MicrosoftCXXNameMangler {
 
   raw_ostream () const { return Out; }
 
-  void mangle(const NamedDecl *D, StringRef Prefix = "?");
-  void mangleName(const NamedDecl *ND);
-  void mangleFunctionEncoding(const FunctionDecl *FD, bool ShouldMangle);
+  void mangle(GlobalDecl GD, StringRef Prefix = "?");
+  void mangleName(GlobalDecl GD);
+  void mangleFunctionEncoding(GlobalDecl GD, bool ShouldMangle);
   void mangleVariableEncoding(const VarDecl *VD);
   void mangleMemberDataPointer(const CXXRecordDecl *RD, const ValueDecl *VD,
StringRef Prefix = "$");
@@ -370,7 +383,7 @@ class MicrosoftCXXNameMangler {
   const FunctionDecl *D = nullptr,
   bool ForceThisQuals = false,
   bool MangleExceptionSpec = true);
-  void mangleNestedName(const NamedDecl *ND);
+  void mangleNestedName(GlobalDecl GD);
 
 private:
   bool isStructorDecl(const NamedDecl *ND) const {
@@ -384,10 +397,10 @@ class MicrosoftCXXNameMangler {
   AddrSpace == LangAS::ptr32_uptr));
   }
 
-  void mangleUnqualifiedName(const NamedDecl *ND) {
-mangleUnqualifiedName(ND, ND->getDeclName());
+  void mangleUnqualifiedName(GlobalDecl GD) {
+mangleUnqualifiedName(GD, cast(GD.getDecl())->getDeclName());
   }
-  void mangleUnqualifiedName(const NamedDecl *ND, DeclarationName Name);
+  void mangleUnqualifiedName(GlobalDecl GD, DeclarationName Name);
   void mangleSourceName(StringRef Name);
   void mangleOperatorName(OverloadedOperatorKind OO, SourceLocation Loc);
   void mangleCXXDtorType(CXXDtorType T);
@@ -396,9 +409,9 @@ class MicrosoftCXXNameMangler {
   void manglePointerCVQualifiers(Qualifiers Quals);
   void manglePointerExtQualifiers(Qualifiers Quals, QualType PointeeType);
 
-  void mangleUnscopedTemplateName(const TemplateDecl *ND);
+  void mangleUnscopedTemplateName(GlobalDecl GD);
   void
-  mangleTemplateInstantiationName(const TemplateDecl *TD,
+  mangleTemplateInstantiationName(GlobalDecl GD,
   const TemplateArgumentList );
   void mangleObjCMethodName(const ObjCMethodDecl *MD);
 
@@ -533,7 +546,8 @@ MicrosoftMangleContextImpl::shouldMangleStringLiteral(const 
StringLiteral *SL) {
   return true;
 }
 
-void MicrosoftCXXNameMangler::mangle(const NamedDecl *D, StringRef Prefix) {
+void MicrosoftCXXNameMangler::mangle(GlobalDecl GD, StringRef Prefix) {
+  const NamedDecl *D = cast(GD.getDecl());
   // MSVC doesn't mangle C++ names the same way it mangles extern "C" names.
   // Therefore it's really important that we don't decorate the
   // name with leading underscores or leading/trailing at signs. So, by
@@ -542,9 +556,9 @@ void MicrosoftCXXNameMangler::mangle(const NamedDecl *D, 
StringRef Prefix) {
 
   //  ::= ?  
   Out << Prefix;
-  mangleName(D);
+  mangleName(GD);
   if (const FunctionDecl *FD = dyn_cast(D))
-mangleFunctionEncoding(FD, Context.shouldMangleDeclName(FD));
+mangleFunctionEncoding(GD, Context.shouldMangleDeclName(FD));
   else if (const VarDecl *VD = dyn_cast(D))
 mangleVariableEncoding(VD);
   else if (isa(D))
@@ -558,8 +572,9 @@ void MicrosoftCXXNameMangler::mangle(const NamedDecl *D, 
StringRef Prefix) {
 

[PATCH] D113118: [clang][AST] Check context of record in structural equivalence.

[Gabor Marton via Phabricator via cfe-commits]

martong accepted this revision.
martong added a comment.

Still LGTM! Let's land it!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113118/new/

https://reviews.llvm.org/D113118

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

[PATCH] D114418: [clang][ASTImporter] Update lookup table correctly at deduction guides.

[Gabor Marton via Phabricator via cfe-commits]

martong added inline comments.



Comment at: clang/lib/AST/ASTImporter.cpp:6081
+  // FunctionTemplateDecl objects are created, but in different order. In this
+  // way DeclContext of these template parameters may change relative to the
+  // "from" context. Because these DeclContext values look already not stable

?



Comment at: clang/lib/AST/ASTImporter.cpp:6082-6085
+  // "from" context. Because these DeclContext values look already not stable
+  // and unimportant this change looks acceptable.
+  // For these reasons the old DeclContext must be saved to change the lookup
+  // table later.

I think this sentence does not provide any meaningful information and does not 
increase the understand-ability. Plus the word `change` is overloaded, first I 
though you meant the patch itself...



Comment at: clang/unittests/AST/ASTImporterTest.cpp:7336
+
+TEST_P(ASTImporterOptionSpecificTestBase, ImportDeductionGuide) {
+  TranslationUnitDecl *FromTU = getTuDecl(

Does this test provide an assertion failure in the base?



Comment at: clang/unittests/AST/ASTImporterTest.cpp:7348-7353
+  // Get the implicit deduction guide for (non-default) constructor of 'B'.
+  auto *FromDG1 = FirstDeclMatcher().match(
+  FromTU, functionTemplateDecl(templateParameterCountIs(3)));
+  // User defined deduction guide.
+  auto *FromDG2 = FirstDeclMatcher().match(
+  FromTU, cxxDeductionGuideDecl(unless(isImplicit(;

Could you please formulate expectations (assertions) on the DeclContext's of 
the two template parameters? I'd expect them to be different.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114418/new/

https://reviews.llvm.org/D114418

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

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Balázs Benics via Phabricator via cfe-commits]

steakhal added a comment.

In D103317#3148868 , @martong wrote:

> In D103317#3127318 , @steakhal 
> wrote:
>
>> To me at least, the patch looks good.
>> Please post some comparative measurements to demonstrate it won't introduce 
>> runtime regression.
>
> Sure!
>
> F20586670: stats.html 
>
> Teaser:
> F20586689: svalbuilder_improvements.png 

Please repeat the measurement for `openssl`. There must have been some 
interference in the memory consumption. Aside from that the results look great.




Comment at: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:1105
 
-  // FIXME: Add support for SymExprs.
   return nullptr;

Where did you address this FIXME?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

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

[clang] aa9b90c - Fix warning due to default switch label

[Yaxun Liu via cfe-commits]

Author: Yaxun (Sam) Liu
Date: 2021-11-23T10:52:51-05:00
New Revision: aa9b90ca441d09969cab158f1db6341de3c1

URL: 
https://github.com/llvm/llvm-project/commit/aa9b90ca441d09969cab158f1db6341de3c1
DIFF: 
https://github.com/llvm/llvm-project/commit/aa9b90ca441d09969cab158f1db6341de3c1.diff

LOG: Fix warning due to default switch label

Fix warning due to default label in switch which covers all enumeration values

Added: 


Modified: 
clang/lib/CodeGen/TargetInfo.cpp

Removed: 




diff  --git a/clang/lib/CodeGen/TargetInfo.cpp 
b/clang/lib/CodeGen/TargetInfo.cpp
index 824a458e8b53d..4360269f8af19 100644
--- a/clang/lib/CodeGen/TargetInfo.cpp
+++ b/clang/lib/CodeGen/TargetInfo.cpp
@@ -9358,9 +9358,6 @@ AMDGPUTargetCodeGenInfo::getLLVMSyncScopeID(const 
LangOptions ,
   case SyncScope::OpenCLAllSVMDevices:
 Name = "";
 break;
-  default:
-assert(false && "NOT IMPLEMENTED");
-break;
   }
 
   if (Ordering != llvm::AtomicOrdering::SequentiallyConsistent) {



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

[PATCH] D113753: [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

[Gabor Marton via Phabricator via cfe-commits]

martong added a comment.

I'm attaching the coverage of the new test file for the related change:

  375 :   // Constraints may have changed since the creation of a 
bound SVal. Check if
  376 :   // the values can be simplified based on those new 
constraints.
  377  12 :   SVal simplifiedLhs = simplifySVal(state, lhs);
  378  12 :   SVal simplifiedRhs = simplifySVal(state, rhs);
  379  12 :   if (auto simplifiedLhsAsNonLoc = 
simplifiedLhs.getAs())
  380  12 : lhs = *simplifiedLhsAsNonLoc;
  381  12 :   if (auto simplifiedRhsAsNonLoc = 
simplifiedRhs.getAs())
  382  12 : rhs = *simplifiedRhsAsNonLoc;
  383 :


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113753/new/

https://reviews.llvm.org/D113753

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

[PATCH] D113753: [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

[Gabor Marton via Phabricator via cfe-commits]

This revision was landed with ongoing or failed builds.
This revision was automatically updated to reflect the committed changes.
Closed by commit rG12887a202404: [Analyzer][Core] Better simplification in 
SimpleSValBuilder::evalBinOpNN (authored by martong).

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113753/new/

https://reviews.llvm.org/D113753

Files:
  clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
  clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp


Index: clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
===
--- /dev/null
+++ clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
@@ -0,0 +1,30 @@
+// RUN: %clang_analyze_cc1 %s \
+// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=debug.ExprInspection \
+// RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -verify
+
+// Here we test whether the SValBuilder is capable to simplify existing
+// SVals based on a newly added constraints when evaluating a BinOp.
+
+void clang_analyzer_eval(bool);
+
+void test_evalBinOp_simplifies_lhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the LHS being simplified.
+  clang_analyzer_eval(x == 1); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
+
+void test_evalBinOp_simplifies_rhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the RHS being simplified.
+  clang_analyzer_eval(1 == x); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
Index: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
===
--- clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -372,6 +372,15 @@
   NonLoc InputLHS = lhs;
   NonLoc InputRHS = rhs;
 
+  // Constraints may have changed since the creation of a bound SVal. Check if
+  // the values can be simplified based on those new constraints.
+  SVal simplifiedLhs = simplifySVal(state, lhs);
+  SVal simplifiedRhs = simplifySVal(state, rhs);
+  if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs())
+lhs = *simplifiedLhsAsNonLoc;
+  if (auto simplifiedRhsAsNonLoc = simplifiedRhs.getAs())
+rhs = *simplifiedRhsAsNonLoc;
+
   // Handle trivial case where left-side and right-side are the same.
   if (lhs == rhs)
 switch (op) {
@@ -619,16 +628,6 @@
 }
   }
 
-  // Does the symbolic expression simplify to a constant?
-  // If so, "fold" the constant by setting 'lhs' to a ConcreteInt
-  // and try again.
-  SVal simplifiedLhs = simplifySVal(state, lhs);
-  if (simplifiedLhs != lhs)
-if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs()) {
-  lhs = *simplifiedLhsAsNonLoc;
-  continue;
-}
-
   // Is the RHS a constant?
   if (const llvm::APSInt *RHSValue = getKnownValue(state, rhs))
 return MakeSymIntVal(Sym, op, *RHSValue, resultTy);


Index: clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
===
--- /dev/null
+++ clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
@@ -0,0 +1,30 @@
+// RUN: %clang_analyze_cc1 %s \
+// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=debug.ExprInspection \
+// RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -verify
+
+// Here we test whether the SValBuilder is capable to simplify existing
+// SVals based on a newly added constraints when evaluating a BinOp.
+
+void clang_analyzer_eval(bool);
+
+void test_evalBinOp_simplifies_lhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the LHS being simplified.
+  clang_analyzer_eval(x == 1); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
+
+void test_evalBinOp_simplifies_rhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the RHS being simplified.
+  clang_analyzer_eval(1 == x); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
Index: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
===
--- clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -372,6 +372,15 @@
   NonLoc InputLHS = lhs;
   NonLoc InputRHS = rhs;
 
+  // Constraints may have changed since the creation of a bound SVal. Check if
+  // the values can be simplified based on those new constraints.
+  SVal simplifiedLhs = simplifySVal(state, lhs);
+  SVal simplifiedRhs = simplifySVal(state, rhs);
+  if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs())
+lhs = *simplifiedLhsAsNonLoc;
+  if (auto simplifiedRhsAsNonLoc = simplifiedRhs.getAs())
+rhs = *simplifiedRhsAsNonLoc;
+
   // Handle trivial case where left-side and right-side are the same.
   if (lhs == rhs)
 switch (op) {
@@ -619,16 +628,6 @@
 }
   }
 
-  // Does the symbolic 

[clang] 12887a2 - [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

[Gabor Marton via cfe-commits]

Author: Gabor Marton
Date: 2021-11-23T16:38:01+01:00
New Revision: 12887a202404471ddf77f9fae658700573cbebe8

URL: 
https://github.com/llvm/llvm-project/commit/12887a202404471ddf77f9fae658700573cbebe8
DIFF: 
https://github.com/llvm/llvm-project/commit/12887a202404471ddf77f9fae658700573cbebe8.diff

LOG: [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

Make the SValBuilder capable to simplify existing
SVals based on a newly added constraints when evaluating a BinOp.

Before this patch, we called `simplify` only in some edge cases.
However, we can and should investigate the constraints in all cases.

Differential Revision: https://reviews.llvm.org/D113753

Added: 
clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp

Modified: 
clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp

Removed: 




diff  --git a/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp 
b/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
index 54a2386fc2ff9..e179ecc33481b 100644
--- a/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ b/clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -372,6 +372,15 @@ SVal SimpleSValBuilder::evalBinOpNN(ProgramStateRef state,
   NonLoc InputLHS = lhs;
   NonLoc InputRHS = rhs;
 
+  // Constraints may have changed since the creation of a bound SVal. Check if
+  // the values can be simplified based on those new constraints.
+  SVal simplifiedLhs = simplifySVal(state, lhs);
+  SVal simplifiedRhs = simplifySVal(state, rhs);
+  if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs())
+lhs = *simplifiedLhsAsNonLoc;
+  if (auto simplifiedRhsAsNonLoc = simplifiedRhs.getAs())
+rhs = *simplifiedRhsAsNonLoc;
+
   // Handle trivial case where left-side and right-side are the same.
   if (lhs == rhs)
 switch (op) {
@@ -619,16 +628,6 @@ SVal SimpleSValBuilder::evalBinOpNN(ProgramStateRef state,
 }
   }
 
-  // Does the symbolic expression simplify to a constant?
-  // If so, "fold" the constant by setting 'lhs' to a ConcreteInt
-  // and try again.
-  SVal simplifiedLhs = simplifySVal(state, lhs);
-  if (simplifiedLhs != lhs)
-if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs()) {
-  lhs = *simplifiedLhsAsNonLoc;
-  continue;
-}
-
   // Is the RHS a constant?
   if (const llvm::APSInt *RHSValue = getKnownValue(state, rhs))
 return MakeSymIntVal(Sym, op, *RHSValue, resultTy);

diff  --git a/clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp 
b/clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
new file mode 100644
index 0..f9a0ee296caf4
--- /dev/null
+++ b/clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
@@ -0,0 +1,30 @@
+// RUN: %clang_analyze_cc1 %s \
+// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=debug.ExprInspection \
+// RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -verify
+
+// Here we test whether the SValBuilder is capable to simplify existing
+// SVals based on a newly added constraints when evaluating a BinOp.
+
+void clang_analyzer_eval(bool);
+
+void test_evalBinOp_simplifies_lhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the LHS being simplified.
+  clang_analyzer_eval(x == 1); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
+
+void test_evalBinOp_simplifies_rhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the RHS being simplified.
+  clang_analyzer_eval(1 == x); // expected-warning{{TRUE}}
+  (void)(x * y);
+}



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

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Gabor Marton via Phabricator via cfe-commits]

martong added inline comments.



Comment at: clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp:32
+  clang_analyzer_eval(x + y * z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(y * z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(x == 0); // expected-warning{{TRUE}}

martong wrote:
> steakhal wrote:
> > You could additionally assert that `y == 0` and `z == 0`.
> `y * z == 0` does not imply that both `y` and `z` are `0`. However, we could 
> expect that one of them is `0`, but currently we don't have such a deduction 
> (this might be done in `ConstraintAssignor` in an independent patch).
> this might be done in ConstraintAssignor in an independent patch
Actually, it is not a good idea to bifurcate there (and architecturally 
impossible ATM), so we should implement that differently.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

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

[PATCH] D113049: [AIX] Disable tests that fail because of no 64-bit XCOFF object file support

[Jinsong Ji via Phabricator via cfe-commits]

jsji added a comment.

Thanks, looks much better now.




Comment at: clang/test/ClangScanDeps/modules-full-by-mod-name.cpp:1
+// UNSUPPORTED: powerpc64-ibm-aix
 // RUN: rm -rf %t.dir

there is no `fmodule-format=obj` here, why are we failing here?




Comment at: clang/test/lit.cfg.py:259
+if 'aix' in config.target_triple:
+for directory in ('/CodeGenCXX', '/Misc', '/Modules', '/PCH'):
+exclude_unsupported_files_for_aix(config.test_source_root + directory)

Why not adding `ASTMerge` and its subfolders?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113049/new/

https://reviews.llvm.org/D113049

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

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Gabor Marton via Phabricator via cfe-commits]

martong added a comment.

In D103317#3127318 , @steakhal wrote:

> To me at least, the patch looks good.
> Please post some comparative measurements to demonstrate it won't introduce 
> runtime regression.

Sure!

F20586670: stats.html 

Teaser:
F20586689: svalbuilder_improvements.png 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

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

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Gabor Marton via Phabricator via cfe-commits]

martong added inline comments.



Comment at: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:1144
+  : (SVal)SVB.makeIntVal(*Const);
+  return SVal();
+}

steakhal wrote:
> Let's be explicit about it.
Good point.



Comment at: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:1148-1151
+  SVal Ret = getConst(Sym);
+  if (Ret.isUndef())
+Ret = Visit(Sym);
+  return Ret;

steakhal wrote:
> 
Makes sense.



Comment at: clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp:23
+  x = y = z = 1;
+  return 0;
+}

steakhal wrote:
> It feels odd that in some of your examples you return some value, but in the 
> rest, you don't.
yep, I agree, updated.



Comment at: clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp:32
+  clang_analyzer_eval(x + y * z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(y * z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(x == 0); // expected-warning{{TRUE}}

steakhal wrote:
> You could additionally assert that `y == 0` and `z == 0`.
`y * z == 0` does not imply that both `y` and `z` are `0`. However, we could 
expect that one of them is `0`, but currently we don't have such a deduction 
(this might be done in `ConstraintAssignor` in an independent patch).



Comment at: clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp:50
+void test_SymInt_constrained(int x, int y, int z) {
+  if (x * y * z != 4)
+return;

steakhal wrote:
> What if `z` were in the middle? Would it still pass?
No, that would not pass. This is because we do not handle commutativity, for 
that we should have a canonical form of the symbol trees (check out point 3 
here https://reviews.llvm.org/D102696#2784624).



Comment at: clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp:65
+  x = 77 / (y + z);
+  if (y + z != 1)
+return;

steakhal wrote:
> Would the test pass if you were using `z + y != 1` here?
No, because we do not handle commutativity.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

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

[PATCH] D103317: [Analyzer][Core] Make SValBuilder to better simplify svals with 3 symbols in the tree

[Gabor Marton via Phabricator via cfe-commits]

martong updated this revision to Diff 389198.
martong marked 6 inline comments as done.
martong added a comment.

- Return explicitly with UndefinedVal
- Unify test cases (return 0 -> return)


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D103317/new/

https://reviews.llvm.org/D103317

Files:
  clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
  clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp
  clang/test/Analysis/taint-tester.c

Index: clang/test/Analysis/taint-tester.c
===
--- clang/test/Analysis/taint-tester.c
+++ clang/test/Analysis/taint-tester.c
@@ -59,7 +59,7 @@
   int tty = xy.y; // expected-warning + {{tainted}}
 }
 
-void BitwiseOp(int in, char inn) {
+void BitwiseOp(int in, char inn, int zz) {
   // Taint on bitwise operations, integer to integer cast.
   int m;
   int x = 0;
@@ -67,11 +67,12 @@
   int y = (in << (x << in)) * 5;// expected-warning + {{tainted}}
   // The next line tests integer to integer cast.
   int z = y & inn; // expected-warning + {{tainted}}
-  if (y == 5) // expected-warning + {{tainted}}
+  if (y == zz) { // expected-warning + {{tainted}}
 m = z | z;// expected-warning + {{tainted}}
+  }
   else
 m = inn;
-  int mm = m; // expected-warning + {{tainted}}
+  int mm = m; // expected-warning 1 {{tainted}}
 }
 
 // Test getenv.
Index: clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp
===
--- /dev/null
+++ clang/test/Analysis/svalbuilder-simplify-compound-svals.cpp
@@ -0,0 +1,69 @@
+// RUN: %clang_analyze_cc1 %s \
+// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=debug.ExprInspection \
+// RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -verify
+
+// Here we test whether the SValBuilder is capable to simplify existing
+// compound SVals (where there are at leaset 3 symbols in the tree) based on
+// newly added constraints.
+
+void clang_analyzer_eval(bool);
+void clang_analyzer_warnIfReached();
+
+void test_left_tree_constrained(int x, int y, int z) {
+  if (x + y + z != 0)
+return;
+  if (x + y != 0)
+return;
+  clang_analyzer_eval(x + y + z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(x + y == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(z == 0); // expected-warning{{TRUE}}
+  x = y = z = 1;
+  return;
+}
+
+void test_right_tree_constrained(int x, int y, int z) {
+  if (x + y * z != 0)
+return;
+  if (y * z != 0)
+return;
+  clang_analyzer_eval(x + y * z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(y * z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(x == 0); // expected-warning{{TRUE}}
+  return;
+}
+
+void test_left_tree_constrained_minus(int x, int y, int z) {
+  if (x - y - z != 0)
+return;
+  if (x - y != 0)
+return;
+  clang_analyzer_eval(x - y - z == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(x - y == 0); // expected-warning{{TRUE}}
+  clang_analyzer_eval(z == 0); // expected-warning{{TRUE}}
+  x = y = z = 1;
+  return;
+}
+
+void test_SymInt_constrained(int x, int y, int z) {
+  if (x * y * z != 4)
+return;
+  if (z != 2)
+return;
+  if (x * y == 3) {
+clang_analyzer_warnIfReached(); // no-warning
+return;
+  }
+  (void)(x * y * z);
+}
+
+void test_SValBuilder_simplifies_IntSym(int x, int y, int z) {
+  // Most IntSym BinOps are transformed to SymInt in SimpleSValBuilder.
+  // Division is one exception.
+  x = 77 / (y + z);
+  if (y + z != 1)
+return;
+  clang_analyzer_eval(x == 77); // expected-warning{{TRUE}}
+  (void)(x * y * z);
+}
Index: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
===
--- clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -1102,7 +1102,6 @@
   if (SymbolRef Sym = V.getAsSymbol())
 return state->getConstraintManager().getSymVal(state, Sym);
 
-  // FIXME: Add support for SymExprs.
   return nullptr;
 }
 
@@ -1134,6 +1133,24 @@
   return cache(Sym, SVB.makeSymbolVal(Sym));
 }
 
+// Return the known const value for the Sym if available, or return Undef
+// otherwise.
+SVal getConst(SymbolRef Sym) {
+  const llvm::APSInt *Const =
+  State->getConstraintManager().getSymVal(State, Sym);
+  if (Const)
+return Loc::isLocType(Sym->getType()) ? (SVal)SVB.makeIntLocVal(*Const)
+  : (SVal)SVB.makeIntVal(*Const);
+  return UndefinedVal();
+}
+
+SVal getConstOrVisit(SymbolRef Sym) {
+  const SVal Ret = getConst(Sym);
+  if (Ret.isUndef())
+return Visit(Sym);
+  return Ret;
+}
+
   public:
 Simplifier(ProgramStateRef State)
 : State(State), SVB(State->getStateManager().getSValBuilder()) {}
@@ -1147,15 +1164,14 @@
   

[PATCH] D114249: [clang-tidy] performance-unnecessary-copy-initialization: Fix false negative.

[Clement Courbet via Phabricator via cfe-commits]

courbet updated this revision to Diff 389197.
courbet added a comment.

add container exclusion tests for operators.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114249/new/

https://reviews.llvm.org/D114249

Files:
  clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp
  
clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp
  
clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp

Index: clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
===
--- clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
+++ clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
@@ -3,11 +3,19 @@
 template 
 struct Iterator {
   void operator++();
-  const T *() const;
+  T *() const;
   bool operator!=(const Iterator &) const;
   typedef const T _reference;
 };
 
+template 
+struct ConstIterator {
+  void operator++();
+  const T *() const;
+  bool operator!=(const ConstIterator &) const;
+  typedef const T _reference;
+};
+
 struct ExpensiveToCopyType {
   ExpensiveToCopyType();
   virtual ~ExpensiveToCopyType();
@@ -15,8 +23,6 @@
   using ConstRef = const ExpensiveToCopyType &;
   ConstRef referenceWithAlias() const;
   const ExpensiveToCopyType *pointer() const;
-  Iterator begin() const;
-  Iterator end() const;
   void nonConstMethod();
   bool constMethod() const;
   template 
@@ -24,6 +30,21 @@
   operator int() const; // Implicit conversion to int.
 };
 
+template 
+struct Container {
+  bool empty() const;
+  const T& operator[](int) const;
+  const T& operator[](int);
+  Iterator begin();
+  Iterator end();
+  ConstIterator begin() const;
+  ConstIterator end() const;
+  void nonConstMethod();
+  bool constMethod() const;
+};
+
+using ExpensiveToCopyContainerAlias = Container;
+
 struct TrivialToCopyType {
   const TrivialToCopyType () const;
 };
@@ -138,6 +159,94 @@
   VarCopyConstructed.constMethod();
 }
 
+void PositiveOperatorCallConstReferenceParam(const Container ) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarCopyConstructed'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarCopyConstructed(C[42]);
+  VarCopyConstructed.constMethod();
+}
+
+void PositiveOperatorCallConstValueParam(const Container C) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarCopyConstructed'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarCopyConstructed(C[42]);
+  VarCopyConstructed.constMethod();
+}
+
+void PositiveOperatorCallConstValueParamAlias(const ExpensiveToCopyContainerAlias C) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  

[PATCH] D113753: [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

[Balázs Benics via Phabricator via cfe-commits]

steakhal accepted this revision.
steakhal added a comment.
This revision is now accepted and ready to land.

Awesome!


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113753/new/

https://reviews.llvm.org/D113753

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

[PATCH] D113049: [AIX] Disable tests that fail because of no 64-bit XCOFF object file support

[Jake Egan via Phabricator via cfe-commits]

Jake-Egan updated this revision to Diff 389196.
Jake-Egan added a comment.

Thanks for the review. I updated the patch to use lit.cfg.py to filter tests 
that use obj options. For tests that don't use the option or has an individual 
folder, I changed them to UNSUPPORTED instead of XFAIL.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113049/new/

https://reviews.llvm.org/D113049

Files:
  clang/test/ASTMerge/anonymous-fields/test.cpp
  clang/test/ASTMerge/codegen-body/test.c
  clang/test/ASTMerge/injected-class-name-decl/test.cpp
  clang/test/ClangScanDeps/modules-full-by-mod-name.cpp
  clang/test/ClangScanDeps/modules-pch-common-submodule.c
  clang/test/ClangScanDeps/modules-pch-common-via-submodule.c
  clang/test/ClangScanDeps/modules-pch.c
  clang/test/ClangScanDeps/resource_directory.c
  clang/test/Driver/as-version.s
  clang/test/Driver/cc-print-proc-stat.c
  clang/test/Driver/compilation_database.c
  clang/test/Driver/modules-ts.cpp
  clang/test/Driver/report-stat.c
  clang/test/lit.cfg.py
  llvm/test/LTO/X86/remangle_intrinsics.ll
  llvm/test/lit.cfg.py
  llvm/unittests/DebugInfo/DWARF/DWARFDebugInfoTest.cpp

Index: llvm/unittests/DebugInfo/DWARF/DWARFDebugInfoTest.cpp
===
--- llvm/unittests/DebugInfo/DWARF/DWARFDebugInfoTest.cpp
+++ llvm/unittests/DebugInfo/DWARF/DWARFDebugInfoTest.cpp
@@ -1131,7 +1131,11 @@
   EXPECT_STREQ(String1, *Extracted3);
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestEmptyStringOffsets) {
+#else
 TEST(DWARFDebugInfo, TestEmptyStringOffsets) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
@@ -1160,7 +1164,11 @@
   DwarfContext->getDWARFObj().getStrOffsetsSection().Data.empty());
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestRelations) {
+#else
 TEST(DWARFDebugInfo, TestRelations) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
@@ -1347,7 +1355,11 @@
   EXPECT_FALSE(DefaultDie.getSibling().isValid());
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestChildIterators) {
+#else
 TEST(DWARFDebugInfo, TestChildIterators) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
@@ -1456,7 +1468,11 @@
   EXPECT_EQ(CUDie.begin(), CUDie.end());
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestAttributeIterators) {
+#else
 TEST(DWARFDebugInfo, TestAttributeIterators) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
@@ -1518,7 +1534,11 @@
   EXPECT_EQ(E, ++I);
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestFindRecurse) {
+#else
 TEST(DWARFDebugInfo, TestFindRecurse) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
@@ -1732,7 +1752,11 @@
   // Test
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestFindAttrs) {
+#else
 TEST(DWARFDebugInfo, TestFindAttrs) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
@@ -1795,7 +1819,11 @@
   EXPECT_EQ(DieMangled, toString(NameOpt, ""));
 }
 
+#if defined(_AIX) && defined(__64BIT__)
+TEST(DWARFDebugInfo, DISABLED_TestImplicitConstAbbrevs) {
+#else
 TEST(DWARFDebugInfo, TestImplicitConstAbbrevs) {
+#endif
   Triple Triple = getNormalizedDefaultTargetTriple();
   if (!isObjectEmissionSupported(Triple))
 return;
Index: llvm/test/lit.cfg.py
===
--- llvm/test/lit.cfg.py
+++ llvm/test/lit.cfg.py
@@ -402,3 +402,22 @@
 
 if "MemoryWithOrigins" in config.llvm_use_sanitizer:
 config.available_features.add('use_msan_with_origins')
+
+def exclude_unsupported_files_for_aix(dirname):
+   for filename in os.listdir(dirname):
+   source_path = os.path.join( dirname, filename)
+   if os.path.isdir(source_path):
+   continue
+   f = open(source_path, 'r')
+   try:
+  data = f.read()
+  # 64-bit object files are not supported on AIX, so exclude the tests.
+  if ('-emit-obj' in data or '-filetype=obj' in data) and '64' in config.target_triple:
+config.excludes += [ filename ]
+   finally:
+  f.close()
+
+if 'aix' in config.target_triple:
+for directory in ('/CodeGen/X86', '/DebugInfo', '/DebugInfo/X86', '/DebugInfo/Generic', '/LTO/X86', '/Linker'):
+exclude_unsupported_files_for_aix(config.test_source_root + directory)
+
Index: llvm/test/LTO/X86/remangle_intrinsics.ll
===
--- llvm/test/LTO/X86/remangle_intrinsics.ll
+++ 

[PATCH] D113925: [HIP] Add HIP scope atomic operations

[Yaxun Liu via Phabricator via cfe-commits]

This revision was landed with ongoing or failed builds.
This revision was automatically updated to reflect the committed changes.
Closed by commit rGe13246a2ec3d: [HIP] Add HIP scope atomic operations 
(authored by yaxunl).

Changed prior to commit:
  https://reviews.llvm.org/D113925?vs=387668=389195#toc

Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113925/new/

https://reviews.llvm.org/D113925

Files:
  clang/include/clang/AST/Expr.h
  clang/include/clang/Basic/Builtins.def
  clang/include/clang/Basic/SyncScope.h
  clang/lib/AST/Expr.cpp
  clang/lib/CodeGen/CGAtomic.cpp
  clang/lib/CodeGen/TargetInfo.cpp
  clang/lib/Frontend/InitPreprocessor.cpp
  clang/lib/Sema/SemaChecking.cpp
  clang/test/CodeGenCUDA/atomic-ops.cu

Index: clang/test/CodeGenCUDA/atomic-ops.cu
===
--- /dev/null
+++ clang/test/CodeGenCUDA/atomic-ops.cu
@@ -0,0 +1,302 @@
+// RUN: %clang_cc1 -x hip -std=c++11 -triple amdgcn -fcuda-is-device -emit-llvm %s -o - | FileCheck %s
+#include "Inputs/cuda.h"
+
+// CHECK-LABEL: @_Z24atomic32_op_singlethreadPiii
+// CHECK: cmpxchg i32* {{%[0-9]+}}, i32 {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw xchg i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw add i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw and i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw or i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw xor i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw min i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw max i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+__device__ int atomic32_op_singlethread(int *ptr, int val, int desired) {
+  bool flag = __hip_atomic_compare_exchange_strong(ptr, , desired, __ATOMIC_RELAXED, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_exchange(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_add(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_and(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_or(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_xor(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_min(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_max(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  return flag ? val : desired;
+}
+
+// CHECK-LABEL: @_Z25atomicu32_op_singlethreadPjjj
+// CHECK: atomicrmw umin i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+// CHECK: atomicrmw umax i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("singlethread-one-as")
+__device__ unsigned int atomicu32_op_singlethread(unsigned int *ptr, unsigned int val, unsigned int desired) {
+  val = __hip_atomic_fetch_min(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  val = __hip_atomic_fetch_max(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_SINGLETHREAD);
+  return val;
+}
+
+// CHECK-LABEL: @_Z21atomic32_op_wavefrontPiii
+// CHECK: cmpxchg i32* {{%[0-9]+}}, i32 {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw xchg i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw add i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw and i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw or i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw xor i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw min i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+// CHECK: atomicrmw max i32* {{%[0-9]+}}, i32 {{%[0-9]+}} syncscope("wavefront-one-as")
+__device__ int atomic32_op_wavefront(int *ptr, int val, int desired) {
+  bool flag = __hip_atomic_compare_exchange_strong(ptr, , desired, __ATOMIC_RELAXED, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_exchange(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_fetch_add(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_fetch_and(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_fetch_or(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_fetch_xor(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_fetch_min(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  val = __hip_atomic_fetch_max(ptr, val, __ATOMIC_RELAXED, __HIP_MEMORY_SCOPE_WAVEFRONT);
+  return flag ? val : 

[clang] e13246a - [HIP] Add HIP scope atomic operations

[Yaxun Liu via cfe-commits]

Author: Yaxun (Sam) Liu
Date: 2021-11-23T10:13:37-05:00
New Revision: e13246a2ec3dfc13838d43099ca9111c780d2c5e

URL: 
https://github.com/llvm/llvm-project/commit/e13246a2ec3dfc13838d43099ca9111c780d2c5e
DIFF: 
https://github.com/llvm/llvm-project/commit/e13246a2ec3dfc13838d43099ca9111c780d2c5e.diff

LOG: [HIP] Add HIP scope atomic operations

Add an AtomicScopeModel for HIP and support for OpenCL builtins
that are missing in HIP.

Patch by: Michael Liao

Revised by: Anshil Ghandi

Reviewed by: Yaxun Liu

Differential Revision: https://reviews.llvm.org/D113925

Added: 
clang/test/CodeGenCUDA/atomic-ops.cu

Modified: 
clang/include/clang/AST/Expr.h
clang/include/clang/Basic/Builtins.def
clang/include/clang/Basic/SyncScope.h
clang/lib/AST/Expr.cpp
clang/lib/CodeGen/CGAtomic.cpp
clang/lib/CodeGen/TargetInfo.cpp
clang/lib/Frontend/InitPreprocessor.cpp
clang/lib/Sema/SemaChecking.cpp

Removed: 




diff  --git a/clang/include/clang/AST/Expr.h b/clang/include/clang/AST/Expr.h
index 991abef733637..246585e1205fa 100644
--- a/clang/include/clang/AST/Expr.h
+++ b/clang/include/clang/AST/Expr.h
@@ -6305,6 +6305,7 @@ class AtomicExpr : public Expr {
   bool isCmpXChg() const {
 return getOp() == AO__c11_atomic_compare_exchange_strong ||
getOp() == AO__c11_atomic_compare_exchange_weak ||
+   getOp() == AO__hip_atomic_compare_exchange_strong ||
getOp() == AO__opencl_atomic_compare_exchange_strong ||
getOp() == AO__opencl_atomic_compare_exchange_weak ||
getOp() == AO__atomic_compare_exchange ||
@@ -6341,7 +6342,10 @@ class AtomicExpr : public Expr {
 auto Kind =
 (Op >= AO__opencl_atomic_load && Op <= AO__opencl_atomic_fetch_max)
 ? AtomicScopeModelKind::OpenCL
-: AtomicScopeModelKind::None;
+: (Op >= AO__hip_atomic_compare_exchange_strong &&
+   Op <= AO__hip_atomic_fetch_max)
+  ? AtomicScopeModelKind::HIP
+  : AtomicScopeModelKind::None;
 return AtomicScopeModel::create(Kind);
   }
 

diff  --git a/clang/include/clang/Basic/Builtins.def 
b/clang/include/clang/Basic/Builtins.def
index b05777889e79a..1f7680e0d923c 100644
--- a/clang/include/clang/Basic/Builtins.def
+++ b/clang/include/clang/Basic/Builtins.def
@@ -854,6 +854,18 @@ ATOMIC_BUILTIN(__opencl_atomic_fetch_max, "v.", "t")
 ATOMIC_BUILTIN(__atomic_fetch_min, "v.", "t")
 ATOMIC_BUILTIN(__atomic_fetch_max, "v.", "t")
 
+// HIP atomic builtins.
+// FIXME: Is `__hip_atomic_compare_exchange_n` or
+// `__hip_atomic_compare_exchange_weak` needed?
+ATOMIC_BUILTIN(__hip_atomic_compare_exchange_strong, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_exchange, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_fetch_add, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_fetch_and, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_fetch_or, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_fetch_xor, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_fetch_min, "v.", "t")
+ATOMIC_BUILTIN(__hip_atomic_fetch_max, "v.", "t")
+
 #undef ATOMIC_BUILTIN
 
 // Non-overloaded atomic builtins.

diff  --git a/clang/include/clang/Basic/SyncScope.h 
b/clang/include/clang/Basic/SyncScope.h
index ce8fb9cbed131..34703310af2b6 100644
--- a/clang/include/clang/Basic/SyncScope.h
+++ b/clang/include/clang/Basic/SyncScope.h
@@ -40,6 +40,11 @@ namespace clang {
 ///   Update getAsString.
 ///
 enum class SyncScope {
+  HIPSingleThread,
+  HIPWavefront,
+  HIPWorkgroup,
+  HIPAgent,
+  HIPSystem,
   OpenCLWorkGroup,
   OpenCLDevice,
   OpenCLAllSVMDevices,
@@ -49,6 +54,16 @@ enum class SyncScope {
 
 inline llvm::StringRef getAsString(SyncScope S) {
   switch (S) {
+  case SyncScope::HIPSingleThread:
+return "hip_singlethread";
+  case SyncScope::HIPWavefront:
+return "hip_wavefront";
+  case SyncScope::HIPWorkgroup:
+return "hip_workgroup";
+  case SyncScope::HIPAgent:
+return "hip_agent";
+  case SyncScope::HIPSystem:
+return "hip_system";
   case SyncScope::OpenCLWorkGroup:
 return "opencl_workgroup";
   case SyncScope::OpenCLDevice:
@@ -62,7 +77,7 @@ inline llvm::StringRef getAsString(SyncScope S) {
 }
 
 /// Defines the kind of atomic scope models.
-enum class AtomicScopeModelKind { None, OpenCL };
+enum class AtomicScopeModelKind { None, OpenCL, HIP };
 
 /// Defines the interface for synch scope model.
 class AtomicScopeModel {
@@ -138,6 +153,58 @@ class AtomicScopeOpenCLModel : public AtomicScopeModel {
   }
 };
 
+/// Defines the synch scope model for HIP.
+class AtomicScopeHIPModel : public AtomicScopeModel {
+public:
+  /// The enum values match the pre-defined macros
+  /// __HIP_MEMORY_SCOPE_*, which are used to define memory_scope_*
+  /// enums in hip-c.h.
+  enum ID {
+SingleThread = 1,
+Wavefront = 2,
+Workgroup = 3,
+Agent = 4,
+System = 5,
+Last = System
+  };
+
+  AtomicScopeHIPModel() {}
+
+  SyncScope map(unsigned 

[PATCH] D113451: [PowerPC] [Clang] Enable Intel intrinsics support on FreeBSD

[Jinsong Ji via Phabricator via cfe-commits]

jsji added a comment.

In D113451#3148815 , @kamaub wrote:

> Test case `clang/test/CodeGen/ppc-mm-malloc.c` fails on powerpc BE buildbots 
> with this changeset
> https://lab.llvm.org/buildbot/#/builders/93/builds/6031
> https://lab.llvm.org/buildbot/#/builders/100/builds/10836
> https://lab.llvm.org/buildbot/#/builders/52/builds/12719
>
>   *** TEST 'Clang :: CodeGen/ppc-mm-malloc.c' FAILED 
> 
>   Script:
>   --
>   : 'RUN: at line 9';   
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/clang
>  -target powerpc64-unknown-linux-gnu -S -emit-llvm 
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
>  -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | llvm-cxxfilt | 
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/FileCheck
>  
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
>   : 'RUN: at line 10';   
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/clang
>  -target powerpc64-unknown-freebsd13.0 -S -emit-llvm 
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
>  -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | llvm-cxxfilt | 
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/FileCheck
>  
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
>   --
>   Exit Code: 2
>   Command Output (stderr):
>   --
>   In file included from 
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c:12:
>   In file included from 
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/lib/clang/14.0.0/include/ppc_wrappers/mm_malloc.h:15:
>   In file included from /usr/include/stdlib.h:25:
>   In file included from /usr/include/bits/libc-header-start.h:33:
>   In file included from /usr/include/features.h:434:
>   /usr/include/gnu/stubs.h:14:11: fatal error: 'gnu/stubs-64-v2.h' file not 
> found
>   # include 
> ^~~
>   1 error generated.
>   FileCheck error: '' is empty.
>   FileCheck command line:  
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/FileCheck
>  
> /home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
>   --
>   

Thanks @kamaub ! Removed the offending line in 
b0784d1d14246e5d662172e9af6e85776b81fdf0 
.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113451/new/

https://reviews.llvm.org/D113451

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

[PATCH] D113451: [PowerPC] [Clang] Enable Intel intrinsics support on FreeBSD

[Kamau Bridgeman via Phabricator via cfe-commits]

kamaub added a comment.

Test case `clang/test/CodeGen/ppc-mm-malloc.c` fails on powerpc BE buildbots 
with this changeset
https://lab.llvm.org/buildbot/#/builders/93/builds/6031
https://lab.llvm.org/buildbot/#/builders/100/builds/10836
https://lab.llvm.org/buildbot/#/builders/52/builds/12719

  *** TEST 'Clang :: CodeGen/ppc-mm-malloc.c' FAILED 

  Script:
  --
  : 'RUN: at line 9';   
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/clang
 -target powerpc64-unknown-linux-gnu -S -emit-llvm 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
 -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | llvm-cxxfilt | 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/FileCheck
 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
  : 'RUN: at line 10';   
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/clang
 -target powerpc64-unknown-freebsd13.0 -S -emit-llvm 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
 -fno-discard-value-names -mllvm -disable-llvm-optzns -o - | llvm-cxxfilt | 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/FileCheck
 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
  --
  Exit Code: 2
  Command Output (stderr):
  --
  In file included from 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c:12:
  In file included from 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/lib/clang/14.0.0/include/ppc_wrappers/mm_malloc.h:15:
  In file included from /usr/include/stdlib.h:25:
  In file included from /usr/include/bits/libc-header-start.h:33:
  In file included from /usr/include/features.h:434:
  /usr/include/gnu/stubs.h:14:11: fatal error: 'gnu/stubs-64-v2.h' file not 
found
  # include 
^~~
  1 error generated.
  FileCheck error: '' is empty.
  FileCheck command line:  
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/stage1/bin/FileCheck
 
/home/buildbots/ppc64be-clang-multistage-test/clang-ppc64be-multistage/llvm/clang/test/CodeGen/ppc-mm-malloc.c
  --
  




Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113451/new/

https://reviews.llvm.org/D113451

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

[PATCH] D114439: [Annotation] Allow parameter pack expansions in annotate attribute

[Steffen Larsen via Phabricator via cfe-commits]

steffenlarsen created this revision.
steffenlarsen added reviewers: erichkeane, aaron.ballman, Tyker, Naghasan.
Herald added a subscriber: jdoerfert.
steffenlarsen requested review of this revision.
Herald added a project: clang.
Herald added a subscriber: cfe-commits.

These changes make the Clang parser recognize parameter pack expansion in 
attribute arguments and consume them if the attribute is marked as supporting 
parameter pack expansions.
Currently only the `clang::annotate` attribute will support parameter pack 
expansions in its arguments. The parser will issue an error diagnostic if other 
attributes are passed parameter pack expansion arguments.


Repository:
  rG LLVM Github Monorepo

https://reviews.llvm.org/D114439

Files:
  clang/include/clang/Basic/Attr.td
  clang/include/clang/Basic/DiagnosticParseKinds.td
  clang/lib/Parse/ParseDecl.cpp
  clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
  clang/test/Parser/cxx0x-attributes.cpp
  clang/test/SemaTemplate/attributes.cpp
  clang/utils/TableGen/ClangAttrEmitter.cpp

Index: clang/utils/TableGen/ClangAttrEmitter.cpp
===
--- clang/utils/TableGen/ClangAttrEmitter.cpp
+++ clang/utils/TableGen/ClangAttrEmitter.cpp
@@ -1697,6 +1697,22 @@
   OS << "#endif // CLANG_ATTR_LATE_PARSED_LIST\n\n";
 }
 
+// Emits the ParmExpansionArgsSupport property for attributes.
+static void emitClangAttrParmExpansionArgsSupportList(RecordKeeper ,
+  raw_ostream ) {
+  OS << "#if defined(CLANG_ATTR_PARM_EXPANSION_ARGS_SUPPORT_LIST)\n";
+  std::vector Attrs = Records.getAllDerivedDefinitions("Attr");
+
+  for (const auto *Attr : Attrs) {
+if (Attr->getValueAsBit("ParmExpansionArgsSupport")) {
+  std::vector Spellings = GetFlattenedSpellings(*Attr);
+  for (const auto  : Spellings)
+OS << ".Case(\"" << I.name() << "\", true)\n";
+}
+  }
+  OS << "#endif // CLANG_ATTR_PARM_EXPANSION_ARGS_SUPPORT_LIST\n\n";
+}
+
 static bool hasGNUorCXX11Spelling(const Record ) {
   std::vector Spellings = GetFlattenedSpellings(Attribute);
   for (const auto  : Spellings) {
@@ -4221,6 +4237,7 @@
   emitClangAttrThisIsaIdentifierArgList(Records, OS);
   emitClangAttrTypeArgList(Records, OS);
   emitClangAttrLateParsedList(Records, OS);
+  emitClangAttrParmExpansionArgsSupportList(Records, OS);
 }
 
 void EmitClangAttrSubjectMatchRulesParserStringSwitches(RecordKeeper ,
Index: clang/test/SemaTemplate/attributes.cpp
===
--- clang/test/SemaTemplate/attributes.cpp
+++ clang/test/SemaTemplate/attributes.cpp
@@ -64,6 +64,23 @@
 template [[clang::annotate("ANNOTATE_FOO"), clang::annotate("ANNOTATE_BAR")]] void HasAnnotations();
 void UseAnnotations() { HasAnnotations(); }
 
+// CHECK:  FunctionTemplateDecl {{.*}} HasPackAnnotations
+// CHECK:AnnotateAttr {{.*}} "ANNOTATE_BAZ"
+// CHECK:  FunctionDecl {{.*}} HasPackAnnotations
+// CHECK:TemplateArgument{{.*}} pack
+// CHECK-NEXT: TemplateArgument{{.*}} integral 1
+// CHECK-NEXT: TemplateArgument{{.*}} integral 2
+// CHECK-NEXT: TemplateArgument{{.*}} integral 3
+// CHECK:AnnotateAttr {{.*}} "ANNOTATE_BAZ"
+// CHECK:  ConstantExpr {{.*}} 'int'
+// CHECK-NEXT:   value: Int 1
+// CHECK:  ConstantExpr {{.*}} 'int'
+// CHECK-NEXT:   value: Int 2
+// CHECK:  ConstantExpr {{.*}} 'int'
+// CHECK-NEXT:   value: Int 3
+template  [[clang::annotate("ANNOTATE_BAZ", Is...)]] void HasPackAnnotations();
+void UsePackAnnotations() { HasPackAnnotations<1, 2, 3>(); }
+
 namespace preferred_name {
   int x [[clang::preferred_name("frank")]]; // expected-error {{expected a type}}
   int y [[clang::preferred_name(int)]]; // expected-warning {{'preferred_name' attribute only applies to class templates}}
Index: clang/test/Parser/cxx0x-attributes.cpp
===
--- clang/test/Parser/cxx0x-attributes.cpp
+++ clang/test/Parser/cxx0x-attributes.cpp
@@ -258,8 +258,10 @@
   [[]] return;
 }
 
-template void variadic() {
+template  void variadic() {
   void bar [[noreturn...]] (); // expected-error {{attribute 'noreturn' cannot be used as an attribute pack}}
+  void baz [[clang::no_sanitize(Is...)]] (); // expected-error {{attribute 'no_sanitize' does not support parameter pack expansion in arguments}}
+  void boo [[unknown::foo(Is...)]] ();   // expected-warning {{unknown attribute 'foo' ignored}}
 }
 
 // Expression tests
Index: clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
===
--- clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
+++ clang/lib/Sema/SemaTemplateInstantiateDecl.cpp
@@ -189,13 +189,9 @@
   EnterExpressionEvaluationContext Unevaluated(
   S, Sema::ExpressionEvaluationContext::ConstantEvaluated);
   SmallVector Args;
-  Args.reserve(Attr->args_size());
-  for 

[PATCH] D114249: [clang-tidy] performance-unnecessary-copy-initialization: Fix false negative.

[Felix Berger via Phabricator via cfe-commits]

flx accepted this revision.
flx added inline comments.
This revision is now accepted and ready to land.



Comment at: 
clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp:228-232
+void PositiveOperatorCallConstValueParam(const Container* 
C) {
+  const auto AutoAssigned = (*C)[42];
+  // TODO-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 
'AutoAssigned'
+  // TODO-FIXES: const auto& AutoAssigned = (*C)[42];
+  AutoAssigned.constMethod();

Would you mind adding this test also to the file testing the exclusion of 
containers:

https://github.com/llvm/llvm-project/blob/main/clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization-excluded-container-types.cpp

This would cover whether the type matching of pointer types works for excluded 
container types.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114249/new/

https://reviews.llvm.org/D114249

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

[PATCH] D114370: [clangd] IncludeCleaner: Attribute symbols from non self-contained headers to their parents

[Kadir Cetinkaya via Phabricator via cfe-commits]

kadircet added inline comments.



Comment at: clang-tools-extra/clangd/IncludeCleaner.cpp:262
+ ID != SM.getMainFileID() && FE &&
+ !isSelfContainedHeader(PP, ID, FE);) {
+  ID = SM.getFileID(SM.getIncludeLoc(ID));

kbobyrev wrote:
> sammccall wrote:
> > it seems like we'd be better off storing the "is-self-contained" in the 
> > IncludeStructure and looking up the HeaderID here rather than asking the 
> > preprocessor. That way we rely on info that's better obtained at preamble 
> > build time.
> I am slightly confused: we don't really have the `IncludeStructure` here and 
> it is logically detached from the `IncludeStructure` processing. We'd still 
> have to unroll the chain of FIDs in here, so the only difference would be 
> querying `IncludeStructure` data for the cached `isSelfContainedHeader` 
> value, is that right? Why is obtaining that info at preamble build time 
> better?
the call-site has access to ParsedAST, hence the IncludeStructure. I believe 
the main reasoning behind Sam's suggestion is performing these lookups once 
while building the preamble and sharing it afterwards (we can even do the IO 
there).



Comment at: clang-tools-extra/clangd/SourceCode.cpp:56
+// Is Line an #if or #ifdef directive?
+static bool isIf(llvm::StringRef Line) {
+  Line = Line.ltrim();

no need for static here (and other places below)



Comment at: clang-tools-extra/clangd/SourceCode.h:332
+bool isSelfContainedHeader(const Preprocessor , FileID FID,
+   const FileEntry *FE, bool ExpensiveCheck = false);
+

i'd suggest `AllowIO` rather than `ExpensiveCheck`



Comment at: clang-tools-extra/clangd/index/SymbolCollector.cpp:283
   // Is Line an #if or #ifdef directive?
   static bool isIf(llvm::StringRef Line) {
 Line = Line.ltrim();

can you also delete this and the other helpers?



Comment at: clang-tools-extra/clangd/unittests/IncludeCleanerTests.cpp:346
+  // headers to header-guarded ones.
+  for (const auto  : ReferencedFileNames.keys()) {
+llvm::outs() << "Key: " << Key << '\n';

looks like a debug artifact


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114370/new/

https://reviews.llvm.org/D114370

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

[PATCH] D114234: [clang][dataflow] Add base types for building dataflow analyses

[Stanislav Gatev via Phabricator via cfe-commits]

sgatev updated this revision to Diff 389189.
sgatev marked an inline comment as done.
sgatev added a comment.

Add a note about asserting the requirements of the CFG object.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114234/new/

https://reviews.llvm.org/D114234

Files:
  clang/include/clang/Analysis/FlowSensitive/DataflowAnalysis.h
  clang/include/clang/Analysis/FlowSensitive/Environment.h
  clang/lib/Analysis/CMakeLists.txt
  clang/lib/Analysis/FlowSensitive/CMakeLists.txt
  clang/lib/Analysis/FlowSensitive/DataflowAnalysis.cpp

Index: clang/lib/Analysis/FlowSensitive/DataflowAnalysis.cpp
===
--- /dev/null
+++ clang/lib/Analysis/FlowSensitive/DataflowAnalysis.cpp
@@ -0,0 +1,34 @@
+//===- DataflowAnalysis.cpp ---===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+//
+//  This file defines base types for building dataflow analyses that run over
+//  Control-Flow Graphs (CFGs).
+//
+//===--===//
+
+#include 
+
+#include "clang/Analysis/CFG.h"
+#include "clang/Analysis/FlowSensitive/DataflowAnalysis.h"
+#include "clang/Analysis/FlowSensitive/Environment.h"
+#include "llvm/ADT/Optional.h"
+
+using namespace clang;
+using namespace dataflow;
+
+std::vector>
+runDataflowAnalysis(const CFG , DataflowAnalysisDynamic ,
+const Environment ) {
+  // FIXME: Consider enforcing that `Cfg` meets the requirements that
+  // are specified in the header. This could be done by remembering
+  // what options were used to build `Cfg` and asserting on them here.
+
+  // FIXME: Implement work list-based algorithm to compute the fixed
+  // point of `Analysis::transform` for every basic block in `Cfg`.
+  return {};
+}
Index: clang/lib/Analysis/FlowSensitive/CMakeLists.txt
===
--- /dev/null
+++ clang/lib/Analysis/FlowSensitive/CMakeLists.txt
@@ -0,0 +1,7 @@
+add_clang_library(clangAnalysisFlowSensitive
+  DataflowAnalysis.cpp
+
+  LINK_LIBS
+  clangAnalysis
+  clangAST
+  )
Index: clang/lib/Analysis/CMakeLists.txt
===
--- clang/lib/Analysis/CMakeLists.txt
+++ clang/lib/Analysis/CMakeLists.txt
@@ -44,3 +44,4 @@
   )
 
 add_subdirectory(plugins)
+add_subdirectory(FlowSensitive)
Index: clang/include/clang/Analysis/FlowSensitive/Environment.h
===
--- /dev/null
+++ clang/include/clang/Analysis/FlowSensitive/Environment.h
@@ -0,0 +1,27 @@
+//===-- Environment.h ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+//
+//  This file defines an Environment class that is used by dataflow analyses
+//  that run over Control-Flow Graphs (CFGs) to keep track of the state of the
+//  program at given program points.
+//
+//===--===//
+
+#ifndef LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_ENVIRONMENT_H
+#define LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_ENVIRONMENT_H
+
+namespace clang {
+namespace dataflow {
+
+/// Holds the state of the program (store and heap) at a given program point.
+class Environment {};
+
+} // namespace dataflow
+} // namespace clang
+
+#endif // LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_ENVIRONMENT_H
Index: clang/include/clang/Analysis/FlowSensitive/DataflowAnalysis.h
===
--- /dev/null
+++ clang/include/clang/Analysis/FlowSensitive/DataflowAnalysis.h
@@ -0,0 +1,160 @@
+//===- DataflowAnalysis.h ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+//
+//  This file defines base types for building dataflow analyses that run over
+//  Control-Flow Graphs (CFGs).
+//
+//===--===//
+
+#ifndef LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_DATAFLOWANALYSIS_H
+#define LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_DATAFLOWANALYSIS_H
+
+#include 
+#include 
+
+#include "clang/AST/ASTContext.h"
+#include 

[PATCH] D114394: Compile-time computation of string attribute hashes

[serge via Phabricator via cfe-commits]

serge-sans-paille updated this revision to Diff 389187.
serge-sans-paille added a comment.

Remove static Dict and replace it by a dict attached to LLVMContext.

Some early benchmarks, on the SQLite amalgamation, through ` valgrind 
--tool=callgrind ./bin/clang -c -o/dev/null sqlite3.c`

Instruction count
Before: 6,072,172,562
After:6,011,551,695


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114394/new/

https://reviews.llvm.org/D114394

Files:
  clang/lib/CodeGen/CGCall.cpp
  clang/lib/CodeGen/CGOpenMPRuntime.cpp
  clang/lib/CodeGen/CodeGenFunction.cpp
  llvm/include/llvm/Analysis/TargetLibraryInfo.h
  llvm/include/llvm/Analysis/VectorUtils.h
  llvm/include/llvm/CodeGen/CommandFlags.h
  llvm/include/llvm/IR/Assumptions.h
  llvm/include/llvm/IR/Attributes.h
  llvm/include/llvm/IR/Function.h
  llvm/include/llvm/IR/GlobalVariable.h
  llvm/include/llvm/IR/InstrTypes.h
  llvm/include/llvm/ProfileData/SampleProf.h
  llvm/lib/Analysis/InlineCost.cpp
  llvm/lib/AsmParser/LLParser.cpp
  llvm/lib/Bitcode/Reader/BitcodeReader.cpp
  llvm/lib/Bitcode/Writer/BitcodeWriter.cpp
  llvm/lib/CodeGen/CommandFlags.cpp
  llvm/lib/IR/AttributeImpl.h
  llvm/lib/IR/Attributes.cpp
  llvm/lib/IR/Core.cpp
  llvm/lib/IR/DiagnosticInfo.cpp
  llvm/lib/IR/Function.cpp
  llvm/lib/IR/Instructions.cpp
  llvm/lib/IR/LLVMContextImpl.h
  llvm/lib/IR/Verifier.cpp
  llvm/lib/Transforms/IPO/Attributor.cpp
  llvm/lib/Transforms/IPO/OpenMPOpt.cpp
  llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
  llvm/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp
  llvm/lib/Transforms/Utils/CodeExtractor.cpp
  llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp

Index: llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp
===
--- llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp
+++ llvm/lib/Transforms/Utils/EntryExitInstrumenter.cpp
@@ -67,11 +67,13 @@
 }
 
 static bool runOnFunction(Function , bool PostInlining) {
-  StringRef EntryAttr = PostInlining ? "instrument-function-entry-inlined"
- : "instrument-function-entry";
+  AttributeKey EntryAttr =
+  PostInlining ? AttributeKey("instrument-function-entry-inlined")
+   : AttributeKey("instrument-function-entry");
 
-  StringRef ExitAttr = PostInlining ? "instrument-function-exit-inlined"
-: "instrument-function-exit";
+  AttributeKey ExitAttr = PostInlining
+  ? AttributeKey("instrument-function-exit-inlined")
+  : AttributeKey("instrument-function-exit");
 
   StringRef EntryFunc = F.getFnAttribute(EntryAttr).getValueAsString();
   StringRef ExitFunc = F.getFnAttribute(ExitAttr).getValueAsString();
Index: llvm/lib/Transforms/Utils/CodeExtractor.cpp
===
--- llvm/lib/Transforms/Utils/CodeExtractor.cpp
+++ llvm/lib/Transforms/Utils/CodeExtractor.cpp
@@ -888,7 +888,7 @@
   //   attribute can not be inherited.
   for (const auto  : oldFunction->getAttributes().getFnAttrs()) {
 if (Attr.isStringAttribute()) {
-  if (Attr.getKindAsString() == "thunk")
+  if (Attr.getKindAsKey() == "thunk")
 continue;
 } else
   switch (Attr.getKindAsEnum()) {
Index: llvm/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp
===
--- llvm/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp
+++ llvm/lib/Transforms/Scalar/RewriteStatepointsForGC.cpp
@@ -1528,7 +1528,7 @@
 } // end anonymous namespace
 
 static StringRef getDeoptLowering(CallBase *Call) {
-  const char *DeoptLowering = "deopt-lowering";
+  const char DeoptLowering[] = "deopt-lowering";
   if (Call->hasFnAttr(DeoptLowering)) {
 // FIXME: Calls have a *really* confusing interface around attributes
 // with values.
Index: llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
===
--- llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
+++ llvm/lib/Transforms/Instrumentation/MemorySanitizer.cpp
@@ -4194,14 +4194,9 @@
 MemorySanitizerVisitor )
   : F(F), MS(MS), MSV(MSV) {
 AMD64FpEndOffset = AMD64FpEndOffsetSSE;
-for (const auto  : F.getAttributes().getFnAttrs()) {
-  if (Attr.isStringAttribute() &&
-  (Attr.getKindAsString() == "target-features")) {
-if (Attr.getValueAsString().contains("-sse"))
-  AMD64FpEndOffset = AMD64FpEndOffsetNoSSE;
-break;
-  }
-}
+if (F.hasFnAttribute("target-features") &&
+F.getFnAttribute("target-features").getValueAsString().contains("-sse"))
+  AMD64FpEndOffset = AMD64FpEndOffsetNoSSE;
   }
 
   ArgKind classifyArgument(Value* arg) {
Index: llvm/lib/Transforms/IPO/OpenMPOpt.cpp
===
--- llvm/lib/Transforms/IPO/OpenMPOpt.cpp
+++ 

[PATCH] D113753: [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

[Gabor Marton via Phabricator via cfe-commits]

martong updated this revision to Diff 389186.
martong marked an inline comment as done.
martong added a comment.

- Add new test case


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113753/new/

https://reviews.llvm.org/D113753

Files:
  clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
  clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp


Index: clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
===
--- /dev/null
+++ clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
@@ -0,0 +1,30 @@
+// RUN: %clang_analyze_cc1 %s \
+// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=debug.ExprInspection \
+// RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -verify
+
+// Here we test whether the SValBuilder is capable to simplify existing
+// SVals based on a newly added constraints when evaluating a BinOp.
+
+void clang_analyzer_eval(bool);
+
+void test_evalBinOp_simplifies_lhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the LHS being simplified.
+  clang_analyzer_eval(x == 1); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
+
+void test_evalBinOp_simplifies_rhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the RHS being simplified.
+  clang_analyzer_eval(1 == x); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
Index: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
===
--- clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -372,6 +372,15 @@
   NonLoc InputLHS = lhs;
   NonLoc InputRHS = rhs;
 
+  // Constraints may have changed since the creation of a bound SVal. Check if
+  // the values can be simplified based on those new constraints.
+  SVal simplifiedLhs = simplifySVal(state, lhs);
+  SVal simplifiedRhs = simplifySVal(state, rhs);
+  if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs())
+lhs = *simplifiedLhsAsNonLoc;
+  if (auto simplifiedRhsAsNonLoc = simplifiedRhs.getAs())
+rhs = *simplifiedRhsAsNonLoc;
+
   // Handle trivial case where left-side and right-side are the same.
   if (lhs == rhs)
 switch (op) {
@@ -619,16 +628,6 @@
 }
   }
 
-  // Does the symbolic expression simplify to a constant?
-  // If so, "fold" the constant by setting 'lhs' to a ConcreteInt
-  // and try again.
-  SVal simplifiedLhs = simplifySVal(state, lhs);
-  if (simplifiedLhs != lhs)
-if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs()) {
-  lhs = *simplifiedLhsAsNonLoc;
-  continue;
-}
-
   // Is the RHS a constant?
   if (const llvm::APSInt *RHSValue = getKnownValue(state, rhs))
 return MakeSymIntVal(Sym, op, *RHSValue, resultTy);


Index: clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
===
--- /dev/null
+++ clang/test/Analysis/svalbuilder-simplify-in-evalbinop.cpp
@@ -0,0 +1,30 @@
+// RUN: %clang_analyze_cc1 %s \
+// RUN:   -analyzer-checker=core \
+// RUN:   -analyzer-checker=debug.ExprInspection \
+// RUN:   -analyzer-config eagerly-assume=false \
+// RUN:   -verify
+
+// Here we test whether the SValBuilder is capable to simplify existing
+// SVals based on a newly added constraints when evaluating a BinOp.
+
+void clang_analyzer_eval(bool);
+
+void test_evalBinOp_simplifies_lhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the LHS being simplified.
+  clang_analyzer_eval(x == 1); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
+
+void test_evalBinOp_simplifies_rhs(int y) {
+  int x = y / 77;
+  if (y != 77)
+return;
+
+  // Below `x` is the RHS being simplified.
+  clang_analyzer_eval(1 == x); // expected-warning{{TRUE}}
+  (void)(x * y);
+}
Index: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
===
--- clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
+++ clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp
@@ -372,6 +372,15 @@
   NonLoc InputLHS = lhs;
   NonLoc InputRHS = rhs;
 
+  // Constraints may have changed since the creation of a bound SVal. Check if
+  // the values can be simplified based on those new constraints.
+  SVal simplifiedLhs = simplifySVal(state, lhs);
+  SVal simplifiedRhs = simplifySVal(state, rhs);
+  if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs())
+lhs = *simplifiedLhsAsNonLoc;
+  if (auto simplifiedRhsAsNonLoc = simplifiedRhs.getAs())
+rhs = *simplifiedRhsAsNonLoc;
+
   // Handle trivial case where left-side and right-side are the same.
   if (lhs == rhs)
 switch (op) {
@@ -619,16 +628,6 @@
 }
   }
 
-  // Does the symbolic expression simplify to a constant?
-  // If so, "fold" the constant by setting 'lhs' to a ConcreteInt
-  // and try 

[PATCH] D113753: [Analyzer][Core] Better simplification in SimpleSValBuilder::evalBinOpNN

[Gabor Marton via Phabricator via cfe-commits]

martong marked 2 inline comments as done.
martong added a comment.

I'm attaching the coverage of the new test file for the related change:

  375 :   // Constraints may have changed since the creation of a 
bound SVal. Check if
  376 :   // the values can be simplified based on those new 
constraints.
  377  12 :   SVal simplifiedLhs = simplifySVal(state, lhs);
  378  12 :   SVal simplifiedRhs = simplifySVal(state, rhs);
  379  12 :   if (auto simplifiedLhsAsNonLoc = 
simplifiedLhs.getAs())
  380  12 : lhs = *simplifiedLhsAsNonLoc;
  381  12 :   if (auto simplifiedRhsAsNonLoc = 
simplifiedRhs.getAs())
  382  12 : rhs = *simplifiedRhsAsNonLoc;
  383 :




Comment at: clang/lib/StaticAnalyzer/Core/SimpleSValBuilder.cpp:378
+  SVal simplifiedLhs = simplifySVal(state, lhs);
+  SVal simplifiedRhs = simplifySVal(state, rhs);
+  if (auto simplifiedLhsAsNonLoc = simplifiedLhs.getAs())

steakhal wrote:
> It seems like you simplify the `rhs` as well. Could we have a test for that?
Ok, I've added a new test case and reworked the existing a bit.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113753/new/

https://reviews.llvm.org/D113753

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

[PATCH] D114249: [clang-tidy] performance-unnecessary-copy-initialization: Fix false negative.

[Clement Courbet via Phabricator via cfe-commits]

courbet updated this revision to Diff 389184.
courbet added a comment.

Canonicalize more types and add more container tests.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114249/new/

https://reviews.llvm.org/D114249

Files:
  clang-tools-extra/clang-tidy/performance/UnnecessaryCopyInitialization.cpp
  
clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp

Index: clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
===
--- clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
+++ clang-tools-extra/test/clang-tidy/checkers/performance-unnecessary-copy-initialization.cpp
@@ -3,11 +3,19 @@
 template 
 struct Iterator {
   void operator++();
-  const T *() const;
+  T *() const;
   bool operator!=(const Iterator &) const;
   typedef const T _reference;
 };
 
+template 
+struct ConstIterator {
+  void operator++();
+  const T *() const;
+  bool operator!=(const ConstIterator &) const;
+  typedef const T _reference;
+};
+
 struct ExpensiveToCopyType {
   ExpensiveToCopyType();
   virtual ~ExpensiveToCopyType();
@@ -15,8 +23,6 @@
   using ConstRef = const ExpensiveToCopyType &;
   ConstRef referenceWithAlias() const;
   const ExpensiveToCopyType *pointer() const;
-  Iterator begin() const;
-  Iterator end() const;
   void nonConstMethod();
   bool constMethod() const;
   template 
@@ -24,6 +30,21 @@
   operator int() const; // Implicit conversion to int.
 };
 
+template 
+struct Container {
+  bool empty() const;
+  const T& operator[](int) const;
+  const T& operator[](int);
+  Iterator begin();
+  Iterator end();
+  ConstIterator begin() const;
+  ConstIterator end() const;
+  void nonConstMethod();
+  bool constMethod() const;
+};
+
+using ExpensiveToCopyContainerAlias = Container;
+
 struct TrivialToCopyType {
   const TrivialToCopyType () const;
 };
@@ -138,6 +159,94 @@
   VarCopyConstructed.constMethod();
 }
 
+void PositiveOperatorCallConstReferenceParam(const Container ) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarCopyConstructed'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarCopyConstructed(C[42]);
+  VarCopyConstructed.constMethod();
+}
+
+void PositiveOperatorCallConstValueParam(const Container C) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarCopyConstructed'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarCopyConstructed(C[42]);
+  VarCopyConstructed.constMethod();
+}
+
+void PositiveOperatorCallConstValueParamAlias(const ExpensiveToCopyContainerAlias C) {
+  const auto AutoAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoAssigned'
+  // CHECK-FIXES: const auto& AutoAssigned = C[42];
+  AutoAssigned.constMethod();
+
+  const auto AutoCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:14: warning: the const qualified variable 'AutoCopyConstructed'
+  // CHECK-FIXES: const auto& AutoCopyConstructed(C[42]);
+  AutoCopyConstructed.constMethod();
+
+  const ExpensiveToCopyType VarAssigned = C[42];
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: the const qualified variable 'VarAssigned'
+  // CHECK-FIXES: const ExpensiveToCopyType& VarAssigned = C[42];
+  VarAssigned.constMethod();
+
+  const ExpensiveToCopyType VarCopyConstructed(C[42]);
+  // CHECK-MESSAGES: [[@LINE-1]]:29: warning: 

[PATCH] D114370: [clangd] IncludeCleaner: Attribute symbols from non self-contained headers to their parents

[Kirill Bobyrev via Phabricator via cfe-commits]

kbobyrev added inline comments.



Comment at: clang-tools-extra/clangd/IncludeCleaner.cpp:262
+ ID != SM.getMainFileID() && FE &&
+ !isSelfContainedHeader(PP, ID, FE);) {
+  ID = SM.getFileID(SM.getIncludeLoc(ID));

sammccall wrote:
> it seems like we'd be better off storing the "is-self-contained" in the 
> IncludeStructure and looking up the HeaderID here rather than asking the 
> preprocessor. That way we rely on info that's better obtained at preamble 
> build time.
I am slightly confused: we don't really have the `IncludeStructure` here and it 
is logically detached from the `IncludeStructure` processing. We'd still have 
to unroll the chain of FIDs in here, so the only difference would be querying 
`IncludeStructure` data for the cached `isSelfContainedHeader` value, is that 
right? Why is obtaining that info at preamble build time better?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114370/new/

https://reviews.llvm.org/D114370

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

[PATCH] D114370: [clangd] IncludeCleaner: Attribute symbols from non self-contained headers to their parents

[Kirill Bobyrev via Phabricator via cfe-commits]

kbobyrev updated this revision to Diff 389183.
kbobyrev marked 2 inline comments as done.
kbobyrev added a comment.

Resolve most comments.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114370/new/

https://reviews.llvm.org/D114370

Files:
  clang-tools-extra/clangd/IncludeCleaner.cpp
  clang-tools-extra/clangd/IncludeCleaner.h
  clang-tools-extra/clangd/SourceCode.cpp
  clang-tools-extra/clangd/SourceCode.h
  clang-tools-extra/clangd/index/SymbolCollector.cpp
  clang-tools-extra/clangd/unittests/IncludeCleanerTests.cpp

Index: clang-tools-extra/clangd/unittests/IncludeCleanerTests.cpp
===
--- clang-tools-extra/clangd/unittests/IncludeCleanerTests.cpp
+++ clang-tools-extra/clangd/unittests/IncludeCleanerTests.cpp
@@ -281,7 +281,8 @@
   auto  = AST.getSourceManager();
   auto  = AST.getIncludeStructure();
 
-  auto ReferencedFiles = findReferencedFiles(findReferencedLocations(AST), SM);
+  auto ReferencedFiles =
+  findReferencedFiles(findReferencedLocations(AST), AST.getPreprocessor());
   llvm::StringSet<> ReferencedFileNames;
   for (FileID FID : ReferencedFiles)
 ReferencedFileNames.insert(
@@ -303,6 +304,52 @@
   EXPECT_THAT(getUnused(AST, ReferencedHeaders), ::testing::IsEmpty());
 }
 
+TEST(IncludeCleaner, NonSelfContainedHeaders) {
+  TestTU TU;
+  TU.Code = R"cpp(
+#include "bar.h"
+#include "foo.h"
+
+int LocalFoo = foo::Variable,
+LocalBar = bar::Variable;
+)cpp";
+  TU.AdditionalFiles["bar.h"] = R"cpp(
+#pragma once
+namespace bar {
+#include "indirection.h"
+}
+)cpp";
+  TU.AdditionalFiles["foo.h"] = R"cpp(
+#pragma once
+namespace foo {
+#include "unguarded.h"
+}
+)cpp";
+  TU.AdditionalFiles["indirection.h"] = R"cpp(
+#include "unguarded.h"
+)cpp";
+  TU.AdditionalFiles["unguarded.h"] = R"cpp(
+constexpr int Variable = 42;
+)cpp";
+
+  ParsedAST AST = TU.build();
+
+  auto ReferencedFiles =
+  findReferencedFiles(findReferencedLocations(AST), AST.getPreprocessor());
+  llvm::StringSet<> ReferencedFileNames;
+  auto  = AST.getSourceManager();
+  for (FileID FID : ReferencedFiles)
+ReferencedFileNames.insert(
+SM.getPresumedLoc(SM.getLocForStartOfFile(FID)).getFilename());
+  // Note that we have uplifted the referenced files from non self-contained
+  // headers to header-guarded ones.
+  for (const auto  : ReferencedFileNames.keys()) {
+llvm::outs() << "Key: " << Key << '\n';
+  }
+  EXPECT_THAT(ReferencedFileNames.keys(),
+  UnorderedElementsAre(testPath("bar.h"), testPath("foo.h")));
+}
+
 } // namespace
 } // namespace clangd
 } // namespace clang
Index: clang-tools-extra/clangd/index/SymbolCollector.cpp
===
--- clang-tools-extra/clangd/index/SymbolCollector.cpp
+++ clang-tools-extra/clangd/index/SymbolCollector.cpp
@@ -266,7 +266,7 @@
 return toURI(Canonical);
   }
 }
-if (!isSelfContainedHeader(FID, FE)) {
+if (!isSelfContainedHeader(*PP, FID, FE, /*ExpensiveCheck=*/true)) {
   // A .inc or .def file is often included into a real header to define
   // symbols (e.g. LLVM tablegen files).
   if (Filename.endswith(".inc") || Filename.endswith(".def"))
@@ -279,20 +279,6 @@
 return toURI(FE);
   }
 
-  bool isSelfContainedHeader(FileID FID, const FileEntry *FE) {
-// FIXME: Should files that have been #import'd be considered
-// self-contained? That's really a property of the includer,
-// not of the file.
-if (!PP->getHeaderSearchInfo().isFileMultipleIncludeGuarded(FE) &&
-!PP->getHeaderSearchInfo().hasFileBeenImported(FE))
-  return false;
-// This pattern indicates that a header can't be used without
-// particular preprocessor state, usually set up by another header.
-if (isDontIncludeMeHeader(SM.getBufferData(FID)))
-  return false;
-return true;
-  }
-
   // Is Line an #if or #ifdef directive?
   static bool isIf(llvm::StringRef Line) {
 Line = Line.ltrim();
Index: clang-tools-extra/clangd/SourceCode.h
===
--- clang-tools-extra/clangd/SourceCode.h
+++ clang-tools-extra/clangd/SourceCode.h
@@ -324,6 +324,13 @@
 /// Returns true if the given location is in a generated protobuf file.
 bool isProtoFile(SourceLocation Loc, const SourceManager );
 
+/// Use heuristics to check whether the header is self-contained (has header
+/// guard, does not rely on preprocessor state).
+/// When \p ExpensiveCheck is true, this will read a portion of the file which
+/// be discouraged in performance-sensitive context.
+bool isSelfContainedHeader(const Preprocessor , FileID FID,
+   const FileEntry *FE, bool ExpensiveCheck = false);
+
 } // namespace clangd
 } // namespace clang
 #endif
Index: clang-tools-extra/clangd/SourceCode.cpp

[PATCH] D110663: [Driver] Support Debian multiarch style lib/clang/14.0.0/x86_64-linux-gnu runtime path and include/x86_64-linux-gnu/c++/v1 libc++ path

[Sylvestre Ledru via Phabricator via cfe-commits]

sylvestre.ledru added a comment.

I am not sure why you are pinging this review often?! Maybe chat with people 
who gave you feedback before directly?!

Myself I wasted so much time because of D107799 
 that I am very reluctant to enable 
LLVM_ENABLE_PER_TARGET_RUNTIME_DIR again.
The mismatch of the triple / quadruple between the system triple and what have 
been decided in llvm ( x86_64-linux-gnu vs x86_64-pc-linux-gnu)

> Moreover, llvm-project has always had the attitude: "downstream projects are 
> on their own".

I am not sure it is a correct statement. llvm project is a bunch of people with 
different focus. 
But in general, we are trying to be nice with the ecosystem and people who rely 
on this software.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110663/new/

https://reviews.llvm.org/D110663

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

[PATCH] D113372: [Driver] Add CLANG_DEFAULT_PIE to emulate GCC --enable-default-pie

[Sylvestre Ledru via Phabricator via cfe-commits]

sylvestre.ledru added a comment.

Maybe add it to the release notes too?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D113372/new/

https://reviews.llvm.org/D113372

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

[PATCH] D110216: [clang] retain type sugar in auto / template argument deduction

[Matheus Izvekov via Phabricator via cfe-commits]

mizvekov added a comment.

Fix pushed to https://reviews.llvm.org/D114207


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110216/new/

https://reviews.llvm.org/D110216

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

[PATCH] D110216: [clang] retain type sugar in auto / template argument deduction

[Matheus Izvekov via Phabricator via cfe-commits]

mizvekov added a comment.

In D110216#3139129 , @lkail wrote:

> Hi we found regression in our internal tests. It compiles with clang-13.0.0 
> https://godbolt.org/z/3abGrcf7o and gcc https://godbolt.org/z/K9oj3Grs1, but 
> fails with clang-trunk https://godbolt.org/z/1Tehxa1x9. Is it an intended 
> change? If so, do we have to document this?

Thanks for the report!

No it was not intended change, this patch should not affect partial ordering 
like that, modulo any bugs we might have accidentally fixed.
What happened here is that we lost a very small piece of code in the 
refactoring, which was introduced more than 10 years ago, but had no test 
coverage.
You just provided it though, and I am working on a patch to restore it shortly.

In D110216#3140678 , @thakis wrote:

> Haha :) Thanks for taking a look. aka for typedefs sounds like a nice 
> approach to me.

Thanks, low on bandwidth right now but I will get to it :-)


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D110216/new/

https://reviews.llvm.org/D110216

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

[PATCH] D114099: Enable `_Float16` type support on X86 without the avx512fp16 flag

[Zahira Ammarguellat via Phabricator via cfe-commits]

zahiraam added a comment.

In D114099#3148631 , @sylvestre.ledru 
wrote:

> Actually, it breaks on all Debian.
> Could you please revert it?

Done.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114099/new/

https://reviews.llvm.org/D114099

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

[clang] fd759d4 - Revert "The _Float16 type is supported on x86 systems with SSE2 enabled."

[Zahira Ammarguellat via cfe-commits]

Author: Zahira Ammarguellat
Date: 2021-11-23T08:00:57-05:00
New Revision: fd759d42c9f84d16efa99a59620cbb3e6836fed4

URL: 
https://github.com/llvm/llvm-project/commit/fd759d42c9f84d16efa99a59620cbb3e6836fed4
DIFF: 
https://github.com/llvm/llvm-project/commit/fd759d42c9f84d16efa99a59620cbb3e6836fed4.diff

LOG: Revert "The _Float16 type is supported on x86 systems with SSE2 enabled."

This reverts commit 6623c02d70c3732dbea59c6d79c69501baf9627b.
The change seems to be breaking build of compiler-rt on Debian.

Added: 
clang/test/CodeGen/X86/avx512fp16-abi.c
clang/test/CodeGen/X86/avx512fp16-complex.c

Modified: 
clang/docs/LanguageExtensions.rst
clang/docs/ReleaseNotes.rst
clang/lib/Basic/Targets/X86.cpp
clang/test/Sema/Float16.c
clang/test/Sema/conversion-target-dep.c
clang/test/SemaCXX/Float16.cpp

Removed: 
clang/test/CodeGen/X86/Float16-arithmetic.c
clang/test/CodeGen/X86/fp16-abi.c
clang/test/CodeGen/X86/fp16-complex.c



diff  --git a/clang/docs/LanguageExtensions.rst 
b/clang/docs/LanguageExtensions.rst
index dfdb01b8ff542..60b1ed383a1ff 100644
--- a/clang/docs/LanguageExtensions.rst
+++ b/clang/docs/LanguageExtensions.rst
@@ -673,7 +673,7 @@ targets pending ABI standardization:
 * 64-bit ARM (AArch64)
 * AMDGPU
 * SPIR
-* X86 (Available with feature SSE2 and up enabled)
+* X86 (Only available under feature AVX512-FP16)
 
 ``_Float16`` will be supported on more targets as they define ABIs for it.
 

diff  --git a/clang/docs/ReleaseNotes.rst b/clang/docs/ReleaseNotes.rst
index d2fa7ff05a160..104d2e908d809 100644
--- a/clang/docs/ReleaseNotes.rst
+++ b/clang/docs/ReleaseNotes.rst
@@ -187,7 +187,6 @@ X86 Support in Clang
 
 
 - Support for ``AVX512-FP16`` instructions has been added.
-- Support for ``_Float16`` type has been added.
 
 Arm and AArch64 Support in Clang
 

diff  --git a/clang/lib/Basic/Targets/X86.cpp b/clang/lib/Basic/Targets/X86.cpp
index 5e36868937194..454a7743dded3 100644
--- a/clang/lib/Basic/Targets/X86.cpp
+++ b/clang/lib/Basic/Targets/X86.cpp
@@ -239,9 +239,9 @@ bool 
X86TargetInfo::handleTargetFeatures(std::vector ,
   HasAVX512ER = true;
 } else if (Feature == "+avx512fp16") {
   HasAVX512FP16 = true;
+  HasFloat16 = true;
 } else if (Feature == "+avx512pf") {
   HasAVX512PF = true;
-  HasLegalHalfType = true;
 } else if (Feature == "+avx512dq") {
   HasAVX512DQ = true;
 } else if (Feature == "+avx512bitalg") {
@@ -369,8 +369,6 @@ bool 
X86TargetInfo::handleTargetFeatures(std::vector ,
  .Default(NoXOP);
 XOPLevel = std::max(XOPLevel, XLevel);
   }
-  // Turn on _float16 for x86 (feature sse2)
-  HasFloat16 = SSELevel >= SSE2;
 
   // LLVM doesn't have a separate switch for fpmath, so only accept it if it
   // matches the selected sse level.

diff  --git a/clang/test/CodeGen/X86/Float16-arithmetic.c 
b/clang/test/CodeGen/X86/Float16-arithmetic.c
deleted file mode 100644
index 7f7b3ff594424..0
--- a/clang/test/CodeGen/X86/Float16-arithmetic.c
+++ /dev/null
@@ -1,73 +0,0 @@
-// RUN: %clang_cc1 -triple  x86_64-unknown-unknown -emit-llvm  \
-// RUN: < %s  | FileCheck %s --check-prefixes=CHECK
-
-_Float16 add1(_Float16 a, _Float16 b) {
-  // CHECK-LABEL: define{{.*}} half @add1
-  // CHECK: alloca half
-  // CHECK: alloca half
-  // CHECK: store half {{.*}}, half*
-  // CHECK: store half {{.*}}, half*
-  // CHECK: load half, half*
-  // CHECK: load half, half* {{.*}}
-  // CHECK: fadd half {{.*}}, {{.*}}
-  // CHECK: ret half
-  return a + b;
-}
-
-_Float16 add2(_Float16 a, _Float16 b, _Float16 c) {
-  // CHECK-LABEL: define{{.*}} half @add2
-  // CHECK: alloca half
-  // CHECK: alloca half
-  // CHECK: alloca half
-  // CHECK: store half {{.*}}, half*
-  // CHECK: store half {{.*}}, half*
-  // CHECK: store half {{.*}}, half*
-  // CHECK: load half, half* {{.*}}
-  // CHECK: load half, half* {{.*}}
-  // CHECK: fadd half {{.*}}, {{.*}}
-  // CHECK: load half, half* {{.*}}
-  // CHECK: fadd half {{.*}}, {{.*}}
-  // CHECK: ret half
-return a + b + c;
-}
-
-_Float16 sub(_Float16 a, _Float16 b) {
-  // CHECK-LABEL: define{{.*}} half @sub
-  // CHECK: alloca half
-  // CHECK: alloca half
-  // CHECK: store half {{.*}}, half*
-  // CHECK: store half {{.*}}, half*
-  // CHECK: load half, half*
-  // CHECK: load half, half* {{.*}}
-  // CHECK: fsub half {{.*}}, {{.*}}
-  // CHECK: ret half
-  return a - b;
-}
-
-_Float16 div(_Float16 a, _Float16 b) {
-  // CHECK-LABEL: define{{.*}} half @div
-  // CHECK: alloca half
-  // CHECK: alloca half
-  // CHECK: store half {{.*}}, half*
-  // CHECK: store half {{.*}}, half*
-  // CHECK: load half, half* {{.*}}
-  // CHECK: load half, half* {{.*}}
-  // CHECK: fdiv half {{.*}}, {{.*}}
-  // CHECK: ret half
-  return a / b;
-}
-
-_Float16 mul(_Float16 a, _Float16 b) {
-  // CHECK-LABEL: 

[PATCH] D114099: Enable `_Float16` type support on X86 without the avx512fp16 flag

[Sylvestre Ledru via Phabricator via cfe-commits]

sylvestre.ledru added a comment.

Actually, it breaks on all Debian.
Could you please revert it?


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114099/new/

https://reviews.llvm.org/D114099

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

[PATCH] D114099: Enable `_Float16` type support on X86 without the avx512fp16 flag

[Sylvestre Ledru via Phabricator via cfe-commits]

sylvestre.ledru added a comment.

Full log:
https://llvm-jenkins.debian.net/view/Debian%20sid/job/llvm-toolchain-binaries/architecture=amd64,distribution=unstable,label=amd64/104/consoleFull


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114099/new/

https://reviews.llvm.org/D114099

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

[PATCH] D114099: Enable `_Float16` type support on X86 without the avx512fp16 flag

[Sylvestre Ledru via Phabricator via cfe-commits]

sylvestre.ledru added a comment.

As mentioned in 
https://github.com/llvm/llvm-project/commit/6623c02d70c3732dbea59c6d79c69501baf9627b#commitcomment-60741407

This change is breaking build of compiler-rt on Ubuntu bionic and others on 
amd64:

  
"/build/llvm-toolchain-snapshot-14~++2029100719+d729f4c38fca/build-llvm/./bin/clang"
 --target=x86_64-pc-linux-gnu -DVISIBILITY_HIDDEN  -fstack-protector-strong 
-Wformat -Werror=format-security -Wno-unused-command-line-argument -Wdate-time 
-D_FORTIFY_SOURCE=2 -O3 -DNDEBUG  -m32 -DCOMPILER_RT_HAS_FLOAT16 -std=c11 -fPIC 
-fno-builtin -fvisibility=hidden -fomit-frame-pointer -MD -MT 
CMakeFiles/clang_rt.builtins-i386.dir/extendhfsf2.c.o -MF 
CMakeFiles/clang_rt.builtins-i386.dir/extendhfsf2.c.o.d -o 
CMakeFiles/clang_rt.builtins-i386.dir/extendhfsf2.c.o -c 
'/build/llvm-toolchain-snapshot-14~++2029100719+d729f4c38fca/compiler-rt/lib/builtins/extendhfsf2.c'
  In file included from 
/build/llvm-toolchain-snapshot-14~++2029100719+d729f4c38fca/compiler-rt/lib/builtins/extendhfsf2.c:11:
  In file included from 
/build/llvm-toolchain-snapshot-14~++2029100719+d729f4c38fca/compiler-rt/lib/builtins/fp_extend_impl.inc:38:
  
/build/llvm-toolchain-snapshot-14~++2029100719+d729f4c38fca/compiler-rt/lib/builtins/fp_extend.h:44:9:
 error: _Float16 is not supported on this target
  typedef _Float16 src_t;
  ^
  1 error generated.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114099/new/

https://reviews.llvm.org/D114099

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

[PATCH] D114234: [clang][dataflow] Add base types for building dataflow analyses

[Stanislav Gatev via Phabricator via cfe-commits]

sgatev updated this revision to Diff 389166.
sgatev added a comment.

Remove unnecessary constructor.


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D114234/new/

https://reviews.llvm.org/D114234

Files:
  clang/include/clang/Analysis/FlowSensitive/DataflowAnalysis.h
  clang/include/clang/Analysis/FlowSensitive/Environment.h
  clang/lib/Analysis/CMakeLists.txt
  clang/lib/Analysis/FlowSensitive/CMakeLists.txt
  clang/lib/Analysis/FlowSensitive/DataflowAnalysis.cpp

Index: clang/lib/Analysis/FlowSensitive/DataflowAnalysis.cpp
===
--- /dev/null
+++ clang/lib/Analysis/FlowSensitive/DataflowAnalysis.cpp
@@ -0,0 +1,30 @@
+//===- DataflowAnalysis.cpp ---===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+//
+//  This file defines base types for building dataflow analyses that run over
+//  Control-Flow Graphs (CFGs).
+//
+//===--===//
+
+#include 
+
+#include "clang/Analysis/CFG.h"
+#include "clang/Analysis/FlowSensitive/DataflowAnalysis.h"
+#include "clang/Analysis/FlowSensitive/Environment.h"
+#include "llvm/ADT/Optional.h"
+
+using namespace clang;
+using namespace dataflow;
+
+std::vector>
+runDataflowAnalysis(const CFG , DataflowAnalysisDynamic ,
+const Environment ) {
+  // FIXME: Implement work list-based algorithm to compute the fixed
+  // point of `Analysis::transform` for every basic block in `Cfg`.
+  return {};
+}
Index: clang/lib/Analysis/FlowSensitive/CMakeLists.txt
===
--- /dev/null
+++ clang/lib/Analysis/FlowSensitive/CMakeLists.txt
@@ -0,0 +1,7 @@
+add_clang_library(clangAnalysisFlowSensitive
+  DataflowAnalysis.cpp
+
+  LINK_LIBS
+  clangAnalysis
+  clangAST
+  )
Index: clang/lib/Analysis/CMakeLists.txt
===
--- clang/lib/Analysis/CMakeLists.txt
+++ clang/lib/Analysis/CMakeLists.txt
@@ -44,3 +44,4 @@
   )
 
 add_subdirectory(plugins)
+add_subdirectory(FlowSensitive)
Index: clang/include/clang/Analysis/FlowSensitive/Environment.h
===
--- /dev/null
+++ clang/include/clang/Analysis/FlowSensitive/Environment.h
@@ -0,0 +1,27 @@
+//===-- Environment.h ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+//
+//  This file defines an Environment class that is used by dataflow analyses
+//  that run over Control-Flow Graphs (CFGs) to keep track of the state of the
+//  program at given program points.
+//
+//===--===//
+
+#ifndef LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_ENVIRONMENT_H
+#define LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_ENVIRONMENT_H
+
+namespace clang {
+namespace dataflow {
+
+/// Holds the state of the program (store and heap) at a given program point.
+class Environment {};
+
+} // namespace dataflow
+} // namespace clang
+
+#endif // LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_ENVIRONMENT_H
Index: clang/include/clang/Analysis/FlowSensitive/DataflowAnalysis.h
===
--- /dev/null
+++ clang/include/clang/Analysis/FlowSensitive/DataflowAnalysis.h
@@ -0,0 +1,160 @@
+//===- DataflowAnalysis.h ---*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===--===//
+//
+//  This file defines base types for building dataflow analyses that run over
+//  Control-Flow Graphs (CFGs).
+//
+//===--===//
+
+#ifndef LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_DATAFLOWANALYSIS_H
+#define LLVM_CLANG_ANALYSIS_FLOWSENSITIVE_DATAFLOWANALYSIS_H
+
+#include 
+#include 
+
+#include "clang/AST/ASTContext.h"
+#include "clang/AST/Stmt.h"
+#include "clang/Analysis/CFG.h"
+#include "clang/Analysis/FlowSensitive/Environment.h"
+#include "llvm/ADT/Any.h"
+#include "llvm/ADT/Optional.h"
+
+namespace clang {
+namespace dataflow {
+
+/// Type-erased lattice element container.
+///
+/// Requirements:
+///
+///  The type