merge main into amd-staging

Author: z1.cciauto

.github/workflows/release-binaries.yml

@@ -49,7 +49,7 @@ jobs:
   prepare:
     name: Prepare to build binaries
     runs-on: ${{ inputs.runs-on }}
-    if: github.repository == 'llvm/llvm-project'
+    if: github.repository_owner == 'llvm'
     outputs:
       release-version: ${{ steps.vars.outputs.release-version }}
       ref: ${{ steps.vars.outputs.ref }}
@@ -177,7 +177,7 @@ jobs:
   build-release-package:
     name: "Build Release Package"
     needs: prepare
-    if: github.repository == 'llvm/llvm-project'
+    if: github.repository_owner == 'llvm'
     runs-on: ${{ needs.prepare.outputs.build-runs-on }}
     steps:
 
@@ -327,7 +327,7 @@ jobs:
       - prepare
       - build-release-package
     if: >-
-      github.repository == 'llvm/llvm-project'
+      github.repository_owner == 'llvm'
     runs-on: ${{ needs.prepare.outputs.test-runs-on }}
     steps:
     - name: Checkout Actions

clang/docs/BoundsSafety.rst

@@ -996,4 +996,11 @@ and the soundness of the type system. This may incur significant code size
 overhead in unoptimized builds and leaving some of the adoption mistakes to be
 caught only at run time. This is not a fundamental limitation, however, because
 incrementally adding necessary static analysis will allow us to catch issues
-early on and remove unnecessary bounds checks in unoptimized builds.
+early on and remove unnecessary bounds checks in unoptimized builds.
+
+Try it out
+==========
+
+Your feedback on the programming model is valuable. You may want to follow the
+instruction in :doc:`BoundsSafetyAdoptionGuide` to play with ``-fbounds-safety``
+and please send your feedback to `Yeoul Na <mailto:[email protected]>`_.

clang/docs/BoundsSafetyAdoptionGuide.rst

@@ -0,0 +1,90 @@
+======================================
+Adoption Guide for ``-fbounds-safety``
+======================================
+
+.. contents::
+   :local:
+
+Where to get ``-fbounds-safety``
+================================
+
+The open sourcing to llvm.org's ``llvm-project`` is still on going and the
+feature is not available yet. In the mean time, the preview implementation is
+available
+`here <https://github.com/swiftlang/llvm-project/tree/stable/20240723>`_ in a
+fork of ``llvm-project``. Please follow
+`Building LLVM with CMake <https://llvm.org/docs/CMake.html>`_ to build the
+compiler.
+
+Feature flag
+============
+
+Pass ``-fbounds-safety`` as a Clang compilation flag for the C file that you
+want to adopt. We recommend adopting the model file by file, because adoption
+requires some effort to add bounds annotations and fix compiler diagnostics.
+
+Include ``ptrcheck.h``
+======================
+
+``ptrcheck.h`` is a Clang toolchain header to provide definition of the bounds
+annotations such as ``__counted_by``, ``__counted_by_or_null``, ``__sized_by``,
+etc. In the LLVM source tree, the header is located in
+``llvm-project/clang/lib/Headers/ptrcheck.h``.
+
+
+Add bounds annotations on pointers as necessary
+===============================================
+
+Annotate pointers on struct fields and function parameters if they are pointing
+to an array of object, with appropriate bounds annotations. Please see
+:doc:`BoundsSafety` to learn what kind of bounds annotations are available and
+their semantics. Note that local pointer variables typically don't need bounds
+annotations because they are implicitely a wide pointer (``__bidi_indexable``)
+that automatically carries the bounds information.
+
+Address compiler diagnostics
+============================
+
+Once you pass ``-fbounds-safety`` to compiler a C file, you will see some new
+compiler warnings and errors, which guide adoption of ``-fbounds-safety``.
+Consider the following example:
+
+.. code-block:: c
+
+   #include <ptrcheck.h>
+
+   void init_buf(int *p, int n) {
+      for (int i = 0; i < n; ++i)
+         p[i] = 0; // error: array subscript on single pointer 'p' must use a constant index of 0 to be in bounds
+   }
+
+The parameter ``int *p`` doesn't have a bounds annotation, so the compiler will
+complain about the code indexing into it (``p[i]``) as it assumes that ``p`` is
+pointing to a single ``int`` object or null. To address the diagnostics, you
+should add a bounds annotation on ``int *p`` so that the compiler can reason
+about the safety of the array subscript. In the following example, ``p`` is now
+``int *__counted_by(n)``, so the compiler will allow the array subscript with
+additional run-time checks as necessary.
+
+.. code-block:: c
+
+   #include <ptrcheck.h>
+
+   void init_buf(int *__counted_by(n) p, int n) {
+      for (int i = 0; i < n; ++i)
+         p[i] = 0; // ok; `p` is now has a type with bounds annotation.
+   }
+
+Run test suites to fix new run-time traps
+=========================================
+
+Adopting ``-fbounds-safety`` may cause your program to trap if it violates
+bounds safety or it has incorrect adoption. Thus, it is necessary to perform
+run-time testing of your program to gain confidence that it won't trap at
+run time.
+
+Repeat the process for each remaining file
+==========================================
+
+Once you've done with adopting a single C file, please repeat the same process
+for each remaining C file that you want to adopt.

clang/docs/ReleaseNotes.rst

@@ -371,6 +371,9 @@ Resolutions to C++ Defect Reports
 - Clang now allows trailing requires clause on explicit deduction guides.
   (`CWG2707: Deduction guides cannot have a trailing requires-clause <https://cplusplus.github.io/CWG/issues/2707.html>`_).
 
+- Respect constructor constraints during CTAD.
+  (`CWG2628: Implicit deduction guides should propagate constraints <https://cplusplus.github.io/CWG/issues/2628.html>`_).
+
 - Clang now diagnoses a space in the first production of a ``literal-operator-id``
   by default.
   (`CWG2521: User-defined literals and reserved identifiers <https://cplusplus.github.io/CWG/issues/2521.html>`_).
@@ -1166,6 +1169,20 @@ Windows Support
 LoongArch Support
 ^^^^^^^^^^^^^^^^^
 
+- Types of parameters and return value of ``__builtin_lsx_vorn_v`` and ``__builtin_lasx_xvorn_v``
+  are changed from ``signed char`` to ``unsigned char``. (#GH114514)
+
+- ``-mrelax`` and ``-mno-relax`` are supported now on LoongArch that can be used
+  to enable / disable the linker relaxation optimization. (#GH123587)
+
+- Fine-grained la64v1.1 options are added including ``-m{no-,}frecipe``, ``-m{no-,}lam-bh``,
+  ``-m{no-,}ld-seq-sa``, ``-m{no-,}div32``, ``-m{no-,}lamcas`` and ``-m{no-,}scq``.
+
+- Two options ``-m{no-,}annotate-tablejump`` are added to enable / disable
+  annotating table jump instruction to correlate it with the jump table. (#GH102411)
+
+- FreeBSD support is added for LoongArch64 and has been tested by building kernel-toolchain. (#GH119191)
+
 RISC-V Support
 ^^^^^^^^^^^^^^
 

clang/docs/index.rst

@@ -40,6 +40,7 @@ Using Clang as a Compiler
    SanitizerStats
    SanitizerSpecialCaseList
    BoundsSafety
+   BoundsSafetyAdoptionGuide
    BoundsSafetyImplPlans
    ControlFlowIntegrity
    LTOVisibility

clang/include/clang/AST/UnresolvedSet.h

@@ -71,9 +71,8 @@ class UnresolvedSetImpl {
   UnresolvedSetImpl(const UnresolvedSetImpl &) = default;
   UnresolvedSetImpl &operator=(const UnresolvedSetImpl &) = default;
 
-  // FIXME: Switch these to "= default" once MSVC supports generating move ops
-  UnresolvedSetImpl(UnresolvedSetImpl &&) {}
-  UnresolvedSetImpl &operator=(UnresolvedSetImpl &&) { return *this; }
+  UnresolvedSetImpl(UnresolvedSetImpl &&) = default;
+  UnresolvedSetImpl &operator=(UnresolvedSetImpl &&) = default;
 
 public:
   // We don't currently support assignment through this iterator, so we might

clang/include/clang/Basic/BuiltinsSPIRV.td

@@ -19,9 +19,3 @@ def SPIRVLength : Builtin {
   let Attributes = [NoThrow, Const];
   let Prototype = "void(...)";
 }
-
-def SPIRVReflect : Builtin {
-  let Spellings = ["__builtin_spirv_reflect"];
-  let Attributes = [NoThrow, Const];
-  let Prototype = "void(...)";
-}

clang/include/clang/Basic/DiagnosticDriverKinds.td

@@ -822,6 +822,9 @@ def err_drv_loongarch_invalid_simd_option_combination : Error<
 def err_drv_loongarch_invalid_msimd_EQ : Error<
   "invalid argument '%0' to -msimd=; must be one of: none, lsx, lasx">;
 
+def err_drv_loongarch_unsupported_with_linker_relaxation : Error<
+  "%0 is unsupported with LoongArch linker relaxation (-mrelax)">;
+
 def err_drv_expand_response_file : Error<
   "failed to expand response file: %0">;
 

clang/include/clang/CodeGen/BackendUtil.h

@@ -39,8 +39,8 @@ enum BackendAction {
   Backend_EmitObj       ///< Emit native object files
 };
 
-void emitBackendOutput(CompilerInstance &CI, StringRef TDesc, llvm::Module *M,
-                       BackendAction Action,
+void emitBackendOutput(CompilerInstance &CI, CodeGenOptions &CGOpts,
+                       StringRef TDesc, llvm::Module *M, BackendAction Action,
                        llvm::IntrusiveRefCntPtr<llvm::vfs::FileSystem> VFS,
                        std::unique_ptr<raw_pwrite_stream> OS,
                        BackendConsumer *BC = nullptr);

clang/include/clang/Driver/Options.td

@@ -5602,6 +5602,10 @@ def mdiv32 : Flag<["-"], "mdiv32">, Group<m_loongarch_Features_Group>,
   HelpText<"Use div.w[u] and mod.w[u] instructions with input not sign-extended.">;
 def mno_div32 : Flag<["-"], "mno-div32">, Group<m_loongarch_Features_Group>,
   HelpText<"Do not use div.w[u] and mod.w[u] instructions with input not sign-extended.">;
+def mscq : Flag<["-"], "mscq">, Group<m_loongarch_Features_Group>,
+  HelpText<"Enable sc.q instruction.">;
+def mno_scq : Flag<["-"], "mno-scq">, Group<m_loongarch_Features_Group>,
+  HelpText<"Disable sc.q instruction.">;
 def mannotate_tablejump : Flag<["-"], "mannotate-tablejump">, Group<m_loongarch_Features_Group>,
   HelpText<"Enable annotate table jump instruction to correlate it with the jump table.">;
 def mno_annotate_tablejump : Flag<["-"], "mno-annotate-tablejump">, Group<m_loongarch_Features_Group>,

clang/lib/Basic/Targets/LoongArch.cpp

@@ -206,7 +206,7 @@ void LoongArchTargetInfo::getTargetDefines(const LangOptions &Opts,
       // arch feature set will be used to include all sub-features belonging to
       // the V1.1 ISA version.
       if (HasFeatureFrecipe && HasFeatureLAM_BH && HasFeatureLAMCAS &&
-          HasFeatureLD_SEQ_SA && HasFeatureDiv32)
+          HasFeatureLD_SEQ_SA && HasFeatureDiv32 && HasFeatureSCQ)
         Builder.defineMacro("__loongarch_arch",
                             Twine('"') + "la64v1.1" + Twine('"'));
       else
@@ -249,6 +249,9 @@ void LoongArchTargetInfo::getTargetDefines(const LangOptions &Opts,
   if (HasFeatureDiv32)
     Builder.defineMacro("__loongarch_div32", Twine(1));
 
+  if (HasFeatureSCQ)
+    Builder.defineMacro("__loongarch_scq", Twine(1));
+
   StringRef ABI = getABI();
   if (ABI == "lp64d" || ABI == "lp64f" || ABI == "lp64s")
     Builder.defineMacro("__loongarch_lp64");
@@ -333,6 +336,8 @@ bool LoongArchTargetInfo::handleTargetFeatures(
       HasFeatureLD_SEQ_SA = true;
     else if (Feature == "+div32")
       HasFeatureDiv32 = true;
+    else if (Feature == "+scq")
+      HasFeatureSCQ = true;
   }
   return true;
 }

clang/lib/Basic/Targets/LoongArch.h

@@ -34,6 +34,7 @@ class LLVM_LIBRARY_VISIBILITY LoongArchTargetInfo : public TargetInfo {
   bool HasFeatureLAMCAS;
   bool HasFeatureLD_SEQ_SA;
   bool HasFeatureDiv32;
+  bool HasFeatureSCQ;
 
 public:
   LoongArchTargetInfo(const llvm::Triple &Triple, const TargetOptions &)
@@ -47,6 +48,7 @@ class LLVM_LIBRARY_VISIBILITY LoongArchTargetInfo : public TargetInfo {
     HasFeatureLAMCAS = false;
     HasFeatureLD_SEQ_SA = false;
     HasFeatureDiv32 = false;
+    HasFeatureSCQ = false;
     LongDoubleWidth = 128;
     LongDoubleAlign = 128;
     LongDoubleFormat = &llvm::APFloat::IEEEquad();

clang/lib/CodeGen/BackendUtil.cpp

@@ -206,9 +206,10 @@ class EmitAssemblyHelper {
   }
 
 public:
-  EmitAssemblyHelper(CompilerInstance &CI, llvm::Module *M,
+  EmitAssemblyHelper(CompilerInstance &CI, CodeGenOptions &CGOpts,
+                     llvm::Module *M,
                      IntrusiveRefCntPtr<llvm::vfs::FileSystem> VFS)
-      : CI(CI), Diags(CI.getDiagnostics()), CodeGenOpts(CI.getCodeGenOpts()),
+      : CI(CI), Diags(CI.getDiagnostics()), CodeGenOpts(CGOpts),
         TargetOpts(CI.getTargetOpts()), LangOpts(CI.getLangOpts()),
         TheModule(M), VFS(std::move(VFS)),
         TargetTriple(TheModule->getTargetTriple()) {}
@@ -1364,14 +1365,14 @@ runThinLTOBackend(CompilerInstance &CI, ModuleSummaryIndex *CombinedIndex,
   }
 }
 
-void clang::emitBackendOutput(CompilerInstance &CI, StringRef TDesc,
-                              llvm::Module *M, BackendAction Action,
+void clang::emitBackendOutput(CompilerInstance &CI, CodeGenOptions &CGOpts,
+                              StringRef TDesc, llvm::Module *M,
+                              BackendAction Action,
                               IntrusiveRefCntPtr<llvm::vfs::FileSystem> VFS,
                               std::unique_ptr<raw_pwrite_stream> OS,
                               BackendConsumer *BC) {
   llvm::TimeTraceScope TimeScope("Backend");
   DiagnosticsEngine &Diags = CI.getDiagnostics();
-  const auto &CGOpts = CI.getCodeGenOpts();
 
   std::unique_ptr<llvm::Module> EmptyModule;
   if (!CGOpts.ThinLTOIndexFile.empty()) {
@@ -1411,7 +1412,7 @@ void clang::emitBackendOutput(CompilerInstance &CI, StringRef TDesc,
     }
   }
 
-  EmitAssemblyHelper AsmHelper(CI, M, VFS);
+  EmitAssemblyHelper AsmHelper(CI, CGOpts, M, VFS);
   AsmHelper.emitAssembly(Action, std::move(OS), BC);
 
   // Verify clang's TargetInfo DataLayout against the LLVM TargetMachine's

clang/lib/CodeGen/CGBuiltin.cpp

@@ -20430,19 +20430,6 @@ Value *CodeGenFunction::EmitSPIRVBuiltinExpr(unsigned BuiltinID,
         /*ReturnType=*/X->getType()->getScalarType(), Intrinsic::spv_length,
         ArrayRef<Value *>{X}, nullptr, "spv.length");
   }
-  case SPIRV::BI__builtin_spirv_reflect: {
-    Value *I = EmitScalarExpr(E->getArg(0));
-    Value *N = EmitScalarExpr(E->getArg(1));
-    assert(E->getArg(0)->getType()->hasFloatingRepresentation() &&
-           E->getArg(1)->getType()->hasFloatingRepresentation() &&
-           "Reflect operands must have a float representation");
-    assert(E->getArg(0)->getType()->isVectorType() &&
-           E->getArg(1)->getType()->isVectorType() &&
-           "Reflect operands must be a vector");
-    return Builder.CreateIntrinsic(
-        /*ReturnType=*/I->getType(), Intrinsic::spv_reflect,
-        ArrayRef<Value *>{I, N}, nullptr, "spv.reflect");
-  }
   }
   return nullptr;
 }

clang/lib/CodeGen/CodeGenAction.cpp

@@ -313,7 +313,8 @@ void BackendConsumer::HandleTranslationUnit(ASTContext &C) {
 
   EmbedBitcode(getModule(), CodeGenOpts, llvm::MemoryBufferRef());
 
-  emitBackendOutput(CI, C.getTargetInfo().getDataLayoutString(), getModule(),
+  emitBackendOutput(CI, CI.getCodeGenOpts(),
+                    C.getTargetInfo().getDataLayoutString(), getModule(),
                     Action, FS, std::move(AsmOutStream), this);
 
   Ctx.setDiagnosticHandler(std::move(OldDiagnosticHandler));
@@ -1245,8 +1246,9 @@ void CodeGenAction::ExecuteAction() {
   std::unique_ptr<llvm::ToolOutputFile> OptRecordFile =
       std::move(*OptRecordFileOrErr);
 
-  emitBackendOutput(CI, CI.getTarget().getDataLayoutString(), TheModule.get(),
-                    BA, CI.getFileManager().getVirtualFileSystemPtr(),
+  emitBackendOutput(CI, CI.getCodeGenOpts(),
+                    CI.getTarget().getDataLayoutString(), TheModule.get(), BA,
+                    CI.getFileManager().getVirtualFileSystemPtr(),
                     std::move(OS));
   if (OptRecordFile)
     OptRecordFile->keep();

clang/lib/CodeGen/ObjectFilePCHContainerWriter.cpp

@@ -322,16 +322,16 @@ class PCHContainerGenerator : public ASTConsumer {
       // Print the IR for the PCH container to the debug output.
       llvm::SmallString<0> Buffer;
       clang::emitBackendOutput(
-          CI, Ctx.getTargetInfo().getDataLayoutString(), M.get(),
+          CI, CodeGenOpts, Ctx.getTargetInfo().getDataLayoutString(), M.get(),
           BackendAction::Backend_EmitLL, FS,
           std::make_unique<llvm::raw_svector_ostream>(Buffer));
       llvm::dbgs() << Buffer;
     });
 
     // Use the LLVM backend to emit the pch container.
-    clang::emitBackendOutput(CI, Ctx.getTargetInfo().getDataLayoutString(),
-                             M.get(), BackendAction::Backend_EmitObj, FS,
-                             std::move(OS));
+    clang::emitBackendOutput(CI, CodeGenOpts,
+                             Ctx.getTargetInfo().getDataLayoutString(), M.get(),
+                             BackendAction::Backend_EmitObj, FS, std::move(OS));
 
     // Free the memory for the temporary buffer.
     llvm::SmallVector<char, 0> Empty;