Merge branch 'main' into hidden_visibility

Author: s-perron

bolt/lib/Profile/DataAggregator.cpp

@@ -61,6 +61,12 @@ FilterMemProfile("filter-mem-profile",
   cl::init(true),
   cl::cat(AggregatorCategory));
 
+static cl::opt<bool> ParseMemProfile(
+    "parse-mem-profile",
+    cl::desc("enable memory profile parsing if it's present in the input data, "
+             "on by default unless `--itrace` is set."),
+    cl::init(true), cl::cat(AggregatorCategory));
+
 static cl::opt<unsigned long long>
 FilterPID("pid",
   cl::desc("only use samples from process with specified PID"),
@@ -181,6 +187,10 @@ void DataAggregator::start() {
                       "script -F pid,event,ip",
                       /*Wait = */false);
   } else if (!opts::ITraceAggregation.empty()) {
+    // Disable parsing memory profile from trace data, unless requested by user.
+    if (!opts::ParseMemProfile.getNumOccurrences())
+      opts::ParseMemProfile = false;
+
     std::string ItracePerfScriptArgs = llvm::formatv(
         "script -F pid,brstack --itrace={0}", opts::ITraceAggregation);
     launchPerfProcess("branch events with itrace", MainEventsPPI,
@@ -191,12 +201,9 @@ void DataAggregator::start() {
                       /*Wait = */ false);
   }
 
-  // Note: we launch script for mem events regardless of the option, as the
-  //       command fails fairly fast if mem events were not collected.
-  launchPerfProcess("mem events",
-                    MemEventsPPI,
-                    "script -F pid,event,addr,ip",
-                    /*Wait = */false);
+  if (opts::ParseMemProfile)
+    launchPerfProcess("mem events", MemEventsPPI, "script -F pid,event,addr,ip",
+                      /*Wait = */ false);
 
   launchPerfProcess("process events", MMapEventsPPI,
                     "script --show-mmap-events --no-itrace",
@@ -217,7 +224,8 @@ void DataAggregator::abort() {
   sys::Wait(TaskEventsPPI.PI, 1, &Error);
   sys::Wait(MMapEventsPPI.PI, 1, &Error);
   sys::Wait(MainEventsPPI.PI, 1, &Error);
-  sys::Wait(MemEventsPPI.PI, 1, &Error);
+  if (opts::ParseMemProfile)
+    sys::Wait(MemEventsPPI.PI, 1, &Error);
 
   deleteTempFiles();
 
@@ -506,7 +514,8 @@ Error DataAggregator::preprocessProfile(BinaryContext &BC) {
     errs() << "PERF2BOLT: failed to parse samples\n";
 
   // Special handling for memory events
-  if (!prepareToParse("mem events", MemEventsPPI, MemEventsErrorCallback))
+  if (opts::ParseMemProfile &&
+      !prepareToParse("mem events", MemEventsPPI, MemEventsErrorCallback))
     if (const std::error_code EC = parseMemEvents())
       errs() << "PERF2BOLT: failed to parse memory events: " << EC.message()
              << '\n';

clang-tools-extra/clangd/unittests/HoverTests.cpp

@@ -974,7 +974,7 @@ class Foo final {})cpp";
          HI.Name = "abc";
          HI.Kind = index::SymbolKind::Variable;
          HI.NamespaceScope = "";
-         HI.Definition = "int abc = <recovery - expr>()";
+         HI.Definition = "int abc";
          HI.Type = "int";
          HI.AccessSpecifier = "public";
        }},

clang-tools-extra/clangd/unittests/XRefsTests.cpp

@@ -2311,6 +2311,14 @@ TEST(FindReferences, WithinAST) {
             $(S::deleteObject)[[de^lete]] S;
           }
         };
+      )cpp",
+      // Array designators
+      R"cpp(
+        const int $def[[F^oo]] = 0;
+        int Bar[] = {
+          [$(Bar)[[F^oo]]...$(Bar)[[Fo^o]] + 1] = 0,
+          [$(Bar)[[^Foo]] + 2] = 1
+        };
       )cpp"};
   for (const char *Test : Tests)
     checkFindRefs(Test);

clang/docs/HIPSupport.rst

@@ -17,7 +17,7 @@
 HIP Support
 =============
 
-HIP (Heterogeneous-Compute Interface for Portability) `<https://github.com/ROCm-Developer-Tools/HIP>`_ is
+HIP (Heterogeneous-Compute Interface for Portability) `<https://github.com/ROCm/HIP>`_ is
 a C++ Runtime API and Kernel Language. It enables developers to create portable applications for
 offloading computation to different hardware platforms from a single source code.
 
@@ -41,9 +41,9 @@ backend or the out-of-tree LLVM-SPIRV translator. The SPIR-V is then bundled and
 .. note::
    While Clang does not directly provide HIP support for NVIDIA GPUs and CPUs, these platforms are supported via other means:
 
-   - NVIDIA GPUs: HIP support is offered through the HIP project `<https://github.com/ROCm-Developer-Tools/HIP>`_, which provides a header-only library for translating HIP runtime APIs into CUDA runtime APIs. The code is subsequently compiled using NVIDIA's `nvcc`.
+   - NVIDIA GPUs: HIP support is offered through the HIP project `<https://github.com/ROCm/HIP>`_, which provides a header-only library for translating HIP runtime APIs into CUDA runtime APIs. The code is subsequently compiled using NVIDIA's `nvcc`.
 
-   - CPUs: HIP support is available through the HIP-CPU runtime library `<https://github.com/ROCm-Developer-Tools/HIP-CPU>`_. This header-only library enables CPUs to execute unmodified HIP code.
+   - CPUs: HIP support is available through the HIP-CPU runtime library `<https://github.com/ROCm/HIP-CPU>`_. This header-only library enables CPUs to execute unmodified HIP code.
 
 
 Example Usage
@@ -328,7 +328,7 @@ The `parallel_unsequenced_policy <https://en.cppreference.com/w/cpp/algorithm/ex
 maps relatively well to the execution model of AMD GPUs. This, coupled with the
 the availability and maturity of GPU accelerated algorithm libraries that
 implement most / all corresponding algorithms in the standard library
-(e.g. `rocThrust <https://github.com/ROCmSoftwarePlatform/rocThrust>`__), makes
+(e.g. `rocThrust <https://github.com/ROCm/rocm-libraries/tree/develop/projects/rocthrust>`__), makes
 it feasible to provide seamless accelerator offload for supported algorithms,
 when an accelerated version exists. Thus, it becomes possible to easily access
 the computational resources of an AMD accelerator, via a well specified,
@@ -483,7 +483,7 @@ such as GPUs, work.
      allocation / deallocation functions with accelerator-aware equivalents,
      based on a pre-established table; the list of functions that can be
      interposed is available
-     `here <https://github.com/ROCmSoftwarePlatform/roc-stdpar#allocation--deallocation-interposition-status>`__;
+     `here <https://github.com/ROCm/roc-stdpar#allocation--deallocation-interposition-status>`__;
    - This is only run when compiling for the host.
 
 The second pass is optional.
@@ -627,7 +627,7 @@ Linux operating system. Support is synthesised in the following table:
 The minimum Linux kernel version for running in HMM mode is 6.4.
 
 The forwarding header can be obtained from
-`its GitHub repository <https://github.com/ROCmSoftwarePlatform/roc-stdpar>`_.
+`its GitHub repository <https://github.com/ROCm/roc-stdpar>`_.
 It will be packaged with a future `ROCm <https://rocm.docs.amd.com/en/latest/>`_
 release. Because accelerated algorithms are provided via
 `rocThrust <https://rocm.docs.amd.com/projects/rocThrust/en/latest/>`_, a
@@ -636,7 +636,7 @@ transitive dependency on
 can be obtained either by installing their associated components of the
 `ROCm <https://rocm.docs.amd.com/en/latest/>`_ stack, or from their respective
 repositories. The list algorithms that can be offloaded is available
-`here <https://github.com/ROCmSoftwarePlatform/roc-stdpar#algorithm-support-status>`_.
+`here <https://github.com/ROCm/roc-stdpar#algorithm-support-status>`_.
 
 HIP Specific Elements
 ---------------------

clang/docs/ReleaseNotes.rst

@@ -622,6 +622,14 @@ Improvements to Clang's diagnostics
 
 - Improved the FixIts for unused lambda captures.
 
+- Delayed typo correction was removed from the compiler; immediate typo
+  correction behavior remains the same. Delayed typo correction facilities were
+  fragile and unmaintained, and the removal closed the following issues:
+  #GH142457, #GH139913, #GH138850, #GH137867, #GH137860, #GH107840, #GH93308,
+  #GH69470, #GH59391, #GH58172, #GH46215, #GH45915, #GH45891, #GH44490,
+  #GH36703, #GH32903, #GH23312, #GH69874.
+
+
 Improvements to Clang's time-trace
 ----------------------------------
 

clang/include/clang/AST/Expr.h

@@ -240,8 +240,7 @@ class Expr : public ValueStmt {
     return static_cast<bool>(getDependence() & ExprDependence::UnexpandedPack);
   }
 
-  /// Whether this expression contains subexpressions which had errors, e.g. a
-  /// TypoExpr.
+  /// Whether this expression contains subexpressions which had errors.
   bool containsErrors() const {
     return static_cast<bool>(getDependence() & ExprDependence::Error);
   }
@@ -6965,36 +6964,6 @@ class AtomicExpr : public Expr {
   }
 };
 
-/// TypoExpr - Internal placeholder for expressions where typo correction
-/// still needs to be performed and/or an error diagnostic emitted.
-class TypoExpr : public Expr {
-  // The location for the typo name.
-  SourceLocation TypoLoc;
-
-public:
-  TypoExpr(QualType T, SourceLocation TypoLoc)
-      : Expr(TypoExprClass, T, VK_LValue, OK_Ordinary), TypoLoc(TypoLoc) {
-    assert(T->isDependentType() && "TypoExpr given a non-dependent type");
-    setDependence(ExprDependence::TypeValueInstantiation |
-                  ExprDependence::Error);
-  }
-
-  child_range children() {
-    return child_range(child_iterator(), child_iterator());
-  }
-  const_child_range children() const {
-    return const_child_range(const_child_iterator(), const_child_iterator());
-  }
-
-  SourceLocation getBeginLoc() const LLVM_READONLY { return TypoLoc; }
-  SourceLocation getEndLoc() const LLVM_READONLY { return TypoLoc; }
-
-  static bool classof(const Stmt *T) {
-    return T->getStmtClass() == TypoExprClass;
-  }
-
-};
-
 /// This class represents BOTH the OpenMP Array Section and OpenACC 'subarray',
 /// with a boolean differentiator.
 /// OpenMP 5.0 [2.1.5, Array Sections].

clang/include/clang/AST/RecursiveASTVisitor.h

@@ -2956,7 +2956,6 @@ DEF_TRAVERSE_STMT(CXXRewrittenBinaryOperator, {
   }
 })
 DEF_TRAVERSE_STMT(OpaqueValueExpr, {})
-DEF_TRAVERSE_STMT(TypoExpr, {})
 DEF_TRAVERSE_STMT(RecoveryExpr, {})
 DEF_TRAVERSE_STMT(CUDAKernelCallExpr, {})
 

clang/include/clang/Basic/DiagnosticGroups.td

@@ -806,7 +806,9 @@ def UniqueObjectDuplication : DiagGroup<"unique-object-duplication"> {
 Warns when objects which are supposed to be globally unique might get duplicated
 when built into a shared library.
 
-If an object with hidden visibility is built into a shared library, each instance
+This can occur to objects which are hidden from the dynamic linker, due to
+having hidden visibility (on posix) or lacking a dllimport/dllexport attribute
+(on windows). If such an object is built into a shared library, each instance
 of the library will get its own copy. This can cause very subtle bugs if there was
 only supposed to be one copy of the object in question: singletons aren't single,
 changes to one object won't affect the others, the object's initializer will run
@@ -815,7 +817,7 @@ once per copy, etc.
 Specifically, this warning fires when it detects an object which:
 1. Is defined as ``inline`` in a header file (so it might get compiled into multiple libaries), and
 2. Has external linkage (otherwise it's supposed to be duplicated), and
-3. Has hidden visibility.
+3. Has hidden visibility (posix) or lacks a dllimport/dllexport attribute (windows).
 
 As well as one of the following:
 1. The object is mutable, or
@@ -825,13 +827,15 @@ The warning can be resolved by removing one of the conditions above. In rough
 order of preference, this may be done by:
 1. Marking the object ``const`` (if possible)
 2. Moving the object's definition to a source file
-3. Giving the object non-hidden visibility, e.g. using ``__attribute((visibility("default")))``.
+3. Making the object visible using ``__attribute((visibility("default")))``,
+   ``__declspec(dllimport)``, or ``__declspec(dllexport)``.
+
+When annotating an object with ``__declspec(dllimport)`` or ``__declspec(dllexport)``,
+take care to ensure that the object is only exported from one dll, and is imported
+everywhere else.
 
 Note that for (2), all levels of a pointer variable must be constant;
 ``const int*`` will trigger the warning because the pointer itself is mutable.
-
-This warning is not yet implemented for Windows, since Windows uses
-import/export rules instead of visibility.
 }];
 }
 

clang/include/clang/Basic/DiagnosticSemaKinds.td

@@ -1767,7 +1767,8 @@ def note_unsatisfied_trait
     : Note<"%0 is not %enum_select<TraitName>{"
            "%TriviallyRelocatable{trivially relocatable}|"
            "%Replaceable{replaceable}|"
-           "%TriviallyCopyable{trivially copyable}"
+           "%TriviallyCopyable{trivially copyable}|"
+           "%Constructible{constructible with provided types}"
            "}1">;
 
 def note_unsatisfied_trait_reason
@@ -1797,7 +1798,10 @@ def note_unsatisfied_trait_reason
            "%DeletedAssign{has a deleted %select{copy|move}1 "
            "assignment operator}|"
            "%UnionWithUserDeclaredSMF{is a union with a user-declared "
-           "%sub{select_special_member_kind}1}"
+           "%sub{select_special_member_kind}1}|"
+           "%FunctionType{is a function type}|"
+           "%CVVoidType{is a cv void type}|"
+           "%IncompleteArrayType{is an incomplete array type}"
            "}0">;
 
 def warn_consteval_if_always_true : Warning<
@@ -6263,14 +6267,19 @@ def warn_static_local_in_extern_inline : Warning<
 def note_convert_inline_to_static : Note<
   "use 'static' to give inline function %0 internal linkage">;
 
-def warn_possible_object_duplication_mutable : Warning<
-  "%0 may be duplicated when built into a shared library: "
-  "it is mutable, has hidden visibility, and external linkage">,
-  InGroup<UniqueObjectDuplication>, DefaultIgnore;
-def warn_possible_object_duplication_init : Warning<
-  "initialization of %0 may run twice when built into a shared library: "
-  "it has hidden visibility and external linkage">,
-  InGroup<UniqueObjectDuplication>, DefaultIgnore;
+def warn_possible_object_duplication_mutable
+    : Warning<"%0 may be duplicated when built into a shared library: "
+              "it is mutable, with external linkage and "
+              "%select{hidden visibility|no import/export annotation}1">,
+      InGroup<UniqueObjectDuplication>,
+      DefaultIgnore;
+def warn_possible_object_duplication_init
+    : Warning<"initialization of %0 may run twice when built into a shared "
+              "library: "
+              "it has external linkage and "
+              "%select{hidden visibility|no import/export annotation}1">,
+      InGroup<UniqueObjectDuplication>,
+      DefaultIgnore;
 
 def ext_redefinition_of_typedef : ExtWarn<
   "redefinition of typedef %0 is a C11 feature">,

clang/include/clang/Basic/StmtNodes.td

@@ -202,7 +202,6 @@ def ShuffleVectorExpr : StmtNode<Expr>;
 def ConvertVectorExpr : StmtNode<Expr>;
 def BlockExpr : StmtNode<Expr>;
 def OpaqueValueExpr : StmtNode<Expr>;
-def TypoExpr : StmtNode<Expr>;
 def RecoveryExpr : StmtNode<Expr>;
 def BuiltinBitCastExpr : StmtNode<ExplicitCastExpr>;
 def EmbedExpr : StmtNode<Expr>;