Merge branch 'main' into compiler-rt-env_vars

.git-blame-ignore-revs

@@ -87,3 +87,8 @@ e2c2ffbe7a1b5d9e32a2ce64279475b50c4cba5b
 
 # [lldb][nfc] Deindent ProcessGDBRemote::SetThreadStopInfo by two levels
 b32931c5b32eb0d2cf37d688b34f8548c9674c19
+
+# [libc++] Various consistency fixes to the CSV files that we use for tracking Standards Conformance
+64946fdaf9864d8279da1c30e4d7214fe13d1427
+b6262880b34629e9d7a72b5a42f315a3c9ed8139
+39c7dc7207e76e72da21cf4fedda21b5311bf62d

.github/CODEOWNERS

@@ -150,3 +150,6 @@ clang/test/AST/Interp/ @tbaederr
 /llvm/**/DWARFLinker/ @JDevlieghere
 /llvm/**/dsymutil/ @JDevlieghere
 /llvm/**/llvm-dwarfutil/ @JDevlieghere
+
+# libclang/Python bindings
+/clang/bindings/python @DeinAlptraum

.github/workflows/get-llvm-version/action.yml

@@ -0,0 +1,26 @@
+name: Get LLVM Version
+description: >-
+  Get the LLVM version from the llvm-project source tree.  This action assumes
+  the llvm-project sources have already been checked out into GITHUB_WORKSPACE.
+
+outputs:
+  major:
+    description: LLVM major version
+    value: ${{ steps.version.outputs.major }}
+  minor:
+    description: LLVM minor version
+    value: ${{ steps.version.outputs.minor }}
+  patch:
+    description: LLVM patch version
+    value: ${{ steps.version.outputs.patch }}
+
+runs:
+  using: "composite"
+  steps:
+    - name: Get Version
+      shell: bash
+      id: version
+      run: |
+        for v in major minor patch; do
+          echo "$v=`llvm/utils/release/get-llvm-version.sh --$v`" >> $GITHUB_OUTPUT
+        done

.github/workflows/libclang-abi-tests.yml

@@ -33,9 +33,9 @@ jobs:
       ABI_HEADERS: ${{ steps.vars.outputs.ABI_HEADERS }}
       ABI_LIBS: ${{ steps.vars.outputs.ABI_LIBS }}
       BASELINE_VERSION_MAJOR: ${{ steps.vars.outputs.BASELINE_VERSION_MAJOR }}
-      LLVM_VERSION_MAJOR: ${{ steps.version.outputs.LLVM_VERSION_MAJOR }}
-      LLVM_VERSION_MINOR: ${{ steps.version.outputs.LLVM_VERSION_MINOR }}
-      LLVM_VERSION_PATCH: ${{ steps.version.outputs.LLVM_VERSION_PATCH }}
+      LLVM_VERSION_MAJOR: ${{ steps.version.outputs.major }}
+      LLVM_VERSION_MINOR: ${{ steps.version.outputs.minor }}
+      LLVM_VERSION_PATCH: ${{ steps.version.outputs.patch }}
     steps:
       - name: Checkout source
         uses: actions/checkout@v4
@@ -44,14 +44,14 @@ jobs:
 
       - name: Get LLVM version
         id: version
-        uses: llvm/actions/get-llvm-version@main
+        uses: ./.github/workflows/get-llvm-version
 
       - name: Setup Variables
         id: vars
         run: |
           remote_repo='https://github.com/llvm/llvm-project'
-          if [ ${{ steps.version.outputs.LLVM_VERSION_PATCH }} -eq 0 ]; then
-            major_version=$(( ${{ steps.version.outputs.LLVM_VERSION_MAJOR }} - 1))
+          if [ ${{ steps.version.outputs.patch }} -eq 0 ]; then
+            major_version=$(( ${{ steps.version.outputs.major }} - 1))
             baseline_ref="llvmorg-$major_version.1.0"
 
             # If there is a minor release, we want to use that as the base line.
@@ -73,8 +73,8 @@ jobs:
             } >> "$GITHUB_OUTPUT"
           else
             {
-              echo "BASELINE_VERSION_MAJOR=${{ steps.version.outputs.LLVM_VERSION_MAJOR }}"
-              echo "BASELINE_REF=llvmorg-${{ steps.version.outputs.LLVM_VERSION_MAJOR }}.1.0"
+              echo "BASELINE_VERSION_MAJOR=${{ steps.version.outputs.major }}"
+              echo "BASELINE_REF=llvmorg-${{ steps.version.outputs.major }}.1.0"
               echo "ABI_HEADERS=."
               echo "ABI_LIBS=libclang.so libclang-cpp.so"
             } >> "$GITHUB_OUTPUT"

.github/workflows/llvm-tests.yml

@@ -43,9 +43,9 @@ jobs:
       ABI_HEADERS: ${{ steps.vars.outputs.ABI_HEADERS }}
       BASELINE_VERSION_MAJOR: ${{ steps.vars.outputs.BASELINE_VERSION_MAJOR }}
       BASELINE_VERSION_MINOR: ${{ steps.vars.outputs.BASELINE_VERSION_MINOR }}
-      LLVM_VERSION_MAJOR: ${{ steps.version.outputs.LLVM_VERSION_MAJOR }}
-      LLVM_VERSION_MINOR: ${{ steps.version.outputs.LLVM_VERSION_MINOR }}
-      LLVM_VERSION_PATCH: ${{ steps.version.outputs.LLVM_VERSION_PATCH }}
+      LLVM_VERSION_MAJOR: ${{ steps.version.outputs.major }}
+      LLVM_VERSION_MINOR: ${{ steps.version.outputs.minor }}
+      LLVM_VERSION_PATCH: ${{ steps.version.outputs.patch }}
     steps:
       - name: Checkout source
         uses: actions/checkout@v4
@@ -54,7 +54,7 @@ jobs:
 
       - name: Get LLVM version
         id: version
-        uses: llvm/actions/get-llvm-version@main
+        uses: ./.github/workflows/get-llvm-version
 
       - name: Setup Variables
         id: vars
@@ -66,14 +66,14 @@ jobs:
           # 18.1.0 We want to check 17.0.x
           # 18.1.1 We want to check 18.1.0
           echo "BASELINE_VERSION_MINOR=1" >> "$GITHUB_OUTPUT"
-          if [ ${{ steps.version.outputs.LLVM_VERSION_PATCH }} -eq 0 ]; then
+          if [ ${{ steps.version.outputs.patch }} -eq 0 ]; then
             {
-              echo "BASELINE_VERSION_MAJOR=$(( ${{ steps.version.outputs.LLVM_VERSION_MAJOR }} - 1))"
+              echo "BASELINE_VERSION_MAJOR=$(( ${{ steps.version.outputs.major }} - 1))"
               echo "ABI_HEADERS=llvm-c"
             } >> "$GITHUB_OUTPUT"
           else
             {
-              echo "BASELINE_VERSION_MAJOR=${{ steps.version.outputs.LLVM_VERSION_MAJOR }}"
+              echo "BASELINE_VERSION_MAJOR=${{ steps.version.outputs.major }}"
               echo "ABI_HEADERS=."
             } >> "$GITHUB_OUTPUT"
           fi

.github/workflows/release-binaries-all.yml

@@ -91,4 +91,8 @@ jobs:
       release-version: "${{ needs.setup-variables.outputs.release-version }}"
       upload: ${{ needs.setup-variables.outputs.upload == 'true'}}
       runs-on: "${{ matrix.runs-on }}"
-
+    secrets:
+      # This will be empty for pull_request events, but that's fine, because
+      # the release-binaries workflow does not use this secret for the
+      # pull_request event.
+      RELEASE_TASKS_USER_TOKEN: ${{ secrets.RELEASE_TASKS_USER_TOKEN }}

.github/workflows/release-binaries.yml

@@ -37,6 +37,11 @@ on:
         description: "Runner to use for the build"
         required: true
         type: string
+    secrets:
+      RELEASE_TASKS_USER_TOKEN:
+        description: "Secret used to check user permissions."
+        required: false
+
 
 permissions:
   contents: read # Default everything to read-only
@@ -57,6 +62,12 @@ jobs:
       release-binary-filename: ${{ steps.vars.outputs.release-binary-filename }}
 
     steps:
+    # It's good practice to use setup-python, but this is also required on macos-14
+    # due to https://github.com/actions/runner-images/issues/10385
+    - uses: actions/setup-python@39cd14951b08e74b54015e9e001cdefcf80e669f
+      with:
+        python-version: '3.12'
+
     - name: Checkout LLVM
       uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
 
@@ -436,7 +447,7 @@ jobs:
     - name: Upload Build Provenance
       uses: actions/upload-artifact@65462800fd760344b1a7b4382951275a0abb4808 #v4.3.3
       with:
-        name: ${{ runner.os }}-${{ runner.arch }}-release-binary-attestation
+        name: ${{ needs.prepare.outputs.release-binary-filename }}-attestation
         path: ${{ needs.prepare.outputs.release-binary-filename }}.jsonl
 
     - name: Upload Release

.github/workflows/release-tasks.yml

@@ -78,6 +78,8 @@ jobs:
     name: Build Release Binaries
     permissions:
       contents: write
+      id-token: write
+      attestations: write
     needs:
       - validate-tag
       - release-create

bolt/CMakeLists.txt

@@ -82,7 +82,7 @@ endforeach()
 
 set(BOLT_ENABLE_RUNTIME_default OFF)
 if ((CMAKE_SYSTEM_PROCESSOR STREQUAL "x86_64"
-    OR CMAKE_SYSTEM_PROCESSOR STREQUAL "aarch64")
+    OR CMAKE_SYSTEM_PROCESSOR MATCHES "^(arm64|aarch64)$")
     AND (CMAKE_SYSTEM_NAME STREQUAL "Linux"
       OR CMAKE_SYSTEM_NAME STREQUAL "Darwin")
     AND (NOT CMAKE_CROSSCOMPILING))

bolt/include/bolt/Core/BinaryContext.h

@@ -911,7 +911,8 @@ class BinaryContext {
   /// of \p Flags.
   MCSymbol *registerNameAtAddress(StringRef Name, uint64_t Address,
                                   uint64_t Size, uint16_t Alignment,
-                                  unsigned Flags = 0);
+                                  unsigned Flags = 0,
+                                  BinarySection *Section = NULL);
 
   /// Return BinaryData registered at a given \p Address or nullptr if no
   /// global symbol was registered at the location.

bolt/lib/Core/BinaryContext.cpp

@@ -1056,18 +1056,28 @@ void BinaryContext::adjustCodePadding() {
 MCSymbol *BinaryContext::registerNameAtAddress(StringRef Name, uint64_t Address,
                                                uint64_t Size,
                                                uint16_t Alignment,
-                                               unsigned Flags) {
+                                               unsigned Flags,
+                                               BinarySection *Section) {
   // Register the name with MCContext.
   MCSymbol *Symbol = Ctx->getOrCreateSymbol(Name);
+  BinaryData *BD;
+
+  // Register out of section symbols only in GlobalSymbols map
+  if (Section && Section->getEndAddress() == Address) {
+    BD = new BinaryData(*Symbol, Address, Size, Alignment ? Alignment : 1,
+                        *Section, Flags);
+    GlobalSymbols[Name] = BD;
+    return Symbol;
+  }
 
   auto GAI = BinaryDataMap.find(Address);
-  BinaryData *BD;
   if (GAI == BinaryDataMap.end()) {
     ErrorOr<BinarySection &> SectionOrErr = getSectionForAddress(Address);
-    BinarySection &Section =
-        SectionOrErr ? SectionOrErr.get() : absoluteSection();
+    BinarySection &SectionRef = Section        ? *Section
+                                : SectionOrErr ? SectionOrErr.get()
+                                               : absoluteSection();
     BD = new BinaryData(*Symbol, Address, Size, Alignment ? Alignment : 1,
-                        Section, Flags);
+                        SectionRef, Flags);
     GAI = BinaryDataMap.emplace(Address, BD).first;
     GlobalSymbols[Name] = BD;
     updateObjectNesting(GAI);
@@ -1402,7 +1412,7 @@ void BinaryContext::postProcessSymbolTable() {
     if ((BD->getName().starts_with("SYMBOLat") ||
          BD->getName().starts_with("DATAat")) &&
         !BD->getParent() && !BD->getSize() && !BD->isAbsolute() &&
-        BD->getSection()) {
+        BD->getSection().getSize()) {
       this->errs() << "BOLT-WARNING: zero-sized top level symbol: " << *BD
                    << "\n";
       Valid = false;

bolt/lib/Core/BinaryFunction.cpp

@@ -2502,7 +2502,10 @@ void BinaryFunction::annotateCFIState() {
     }
   }
 
-  assert(StateStack.empty() && "corrupt CFI stack");
+  if (!StateStack.empty()) {
+    BC.errs() << "BOLT-WARNING: non-empty CFI stack at the end of " << *this
+              << '\n';
+  }
 }
 
 namespace {

bolt/lib/Rewrite/BinaryPassManager.cpp

@@ -357,6 +357,8 @@ Error BinaryFunctionPassManager::runAllPasses(BinaryContext &BC) {
 
     Manager.registerPass(
         std::make_unique<VeneerElimination>(PrintVeneerElimination));
+
+    Manager.registerPass(std::make_unique<ADRRelaxationPass>());
   }
 
   if (BC.isRISCV()) {
@@ -490,8 +492,6 @@ Error BinaryFunctionPassManager::runAllPasses(BinaryContext &BC) {
   Manager.registerPass(std::make_unique<ReorderData>());
 
   if (BC.isAArch64()) {
-    Manager.registerPass(std::make_unique<ADRRelaxationPass>());
-
     // Tighten branches according to offset differences between branch and
     // targets. No extra instructions after this pass, otherwise we may have
     // relocations out of range and crash during linking.

bolt/lib/Rewrite/RewriteInstance.cpp

@@ -955,13 +955,13 @@ void RewriteInstance::discoverFileObjects() {
     uint64_t SymbolSize = ELFSymbolRef(Symbol).getSize();
     uint64_t SymbolAlignment = Symbol.getAlignment();
 
-    auto registerName = [&](uint64_t FinalSize) {
+    auto registerName = [&](uint64_t FinalSize, BinarySection *Section = NULL) {
       // Register names even if it's not a function, e.g. for an entry point.
       BC->registerNameAtAddress(UniqueName, SymbolAddress, FinalSize,
-                                SymbolAlignment, SymbolFlags);
+                                SymbolAlignment, SymbolFlags, Section);
       if (!AlternativeName.empty())
         BC->registerNameAtAddress(AlternativeName, SymbolAddress, FinalSize,
-                                  SymbolAlignment, SymbolFlags);
+                                  SymbolAlignment, SymbolFlags, Section);
     };
 
     section_iterator Section =
@@ -986,12 +986,25 @@ void RewriteInstance::discoverFileObjects() {
                       << " for function\n");
 
     if (SymbolAddress == Section->getAddress() + Section->getSize()) {
+      ErrorOr<BinarySection &> SectionOrError =
+          BC->getSectionForAddress(Section->getAddress());
+
+      // Skip symbols from invalid sections
+      if (!SectionOrError) {
+        BC->errs() << "BOLT-WARNING: " << UniqueName << " (0x"
+                   << Twine::utohexstr(SymbolAddress)
+                   << ") does not have any section\n";
+        continue;
+      }
+
       assert(SymbolSize == 0 &&
              "unexpect non-zero sized symbol at end of section");
-      LLVM_DEBUG(
-          dbgs()
-          << "BOLT-DEBUG: rejecting as symbol points to end of its section\n");
-      registerName(SymbolSize);
+      LLVM_DEBUG({
+        dbgs() << "BOLT-DEBUG: rejecting as symbol " << UniqueName
+               << " points to end of " << SectionOrError->getName()
+               << " section\n";
+      });
+      registerName(SymbolSize, &SectionOrError.get());
       continue;
     }
 
@@ -2143,6 +2156,14 @@ bool RewriteInstance::analyzeRelocation(
   if (!Relocation::isSupported(RType))
     return false;
 
+  auto IsWeakReference = [](const SymbolRef &Symbol) {
+    Expected<uint32_t> SymFlagsOrErr = Symbol.getFlags();
+    if (!SymFlagsOrErr)
+      return false;
+    return (*SymFlagsOrErr & SymbolRef::SF_Undefined) &&
+           (*SymFlagsOrErr & SymbolRef::SF_Weak);
+  };
+
   const bool IsAArch64 = BC->isAArch64();
 
   const size_t RelSize = Relocation::getSizeForType(RType);
@@ -2174,7 +2195,8 @@ bool RewriteInstance::analyzeRelocation(
     // Section symbols are marked as ST_Debug.
     IsSectionRelocation = (cantFail(Symbol.getType()) == SymbolRef::ST_Debug);
     // Check for PLT entry registered with symbol name
-    if (!SymbolAddress && (IsAArch64 || BC->isRISCV())) {
+    if (!SymbolAddress && !IsWeakReference(Symbol) &&
+        (IsAArch64 || BC->isRISCV())) {
       const BinaryData *BD = BC->getPLTBinaryDataByName(SymbolName);
       SymbolAddress = BD ? BD->getAddress() : 0;
     }
@@ -2603,7 +2625,7 @@ void RewriteInstance::handleRelocation(const SectionRef &RelocatedSection,
       Expected<StringRef> SectionName = Section->getName();
       if (SectionName && !SectionName->empty())
         ReferencedSection = BC->getUniqueSectionByName(*SectionName);
-    } else if (ReferencedSymbol && ContainingBF &&
+    } else if (BC->isRISCV() && ReferencedSymbol && ContainingBF &&
                (cantFail(Symbol.getFlags()) & SymbolRef::SF_Absolute)) {
       // This might be a relocation for an ABS symbols like __global_pointer$ on
       // RISC-V
@@ -2614,6 +2636,30 @@ void RewriteInstance::handleRelocation(const SectionRef &RelocatedSection,
     }
   }
 
+  if (Relocation::isGOT(RType) && !Relocation::isTLS(RType)) {
+    auto exitOnGotEndSymol = [&](StringRef Name) {
+      BC->errs() << "BOLT-ERROR: GOT table contains currently unsupported "
+                    "section end symbol "
+                 << Name << "\n";
+      exit(1);
+    };
+
+    if (SymbolIter != InputFile->symbol_end() && ReferencedSection) {
+      if (cantFail(SymbolIter->getAddress()) ==
+          ReferencedSection->getEndAddress())
+        exitOnGotEndSymol(cantFail(SymbolIter->getName()));
+    } else {
+      // If no section and symbol are provided by relocation, try to find the
+      // symbol by its name, including the possibility that the symbol is local.
+      BinaryData *BD = BC->getBinaryDataByName(SymbolName);
+      if (!BD && NR.getUniquifiedNameCount(SymbolName) == 1)
+        BD = BC->getBinaryDataByName(NR.getUniqueName(SymbolName, 1));
+
+      if ((BD && BD->getAddress() == BD->getSection().getEndAddress()))
+        exitOnGotEndSymol(BD->getName());
+    }
+  }
+
   if (!ReferencedSection)
     ReferencedSection = BC->getSectionForAddress(SymbolAddress);
 
@@ -5509,6 +5555,14 @@ uint64_t RewriteInstance::getNewFunctionOrDataAddress(uint64_t OldAddress) {
   if (const BinaryFunction *BF =
           BC->getBinaryFunctionContainingAddress(OldAddress)) {
     if (BF->isEmitted()) {
+      // If OldAddress is the another entry point of
+      // the function, then BOLT could get the new address.
+      if (BF->isMultiEntry()) {
+        for (const BinaryBasicBlock &BB : *BF)
+          if (BB.isEntryPoint() &&
+              (BF->getAddress() + BB.getOffset()) == OldAddress)
+            return BF->getOutputAddress() + BB.getOffset();
+      }
       BC->errs() << "BOLT-ERROR: unable to get new address corresponding to "
                     "input address 0x"
                  << Twine::utohexstr(OldAddress) << " in function " << *BF