[AutoBump] Merge with fe0dee4 (Jun 10)

.ci/generate-buildkite-pipeline-premerge

@@ -153,7 +153,6 @@ function exclude-linux() {
   for project in ${projects}; do
     case ${project} in
     cross-project-tests) ;; # tests failing
-    lldb)                ;; # tests failing
     openmp)              ;; # https://github.com/google/llvm-premerge-checks/issues/410
     *)
       echo "${project}"
@@ -170,7 +169,7 @@ function exclude-windows() {
     compiler-rt)         ;; # tests taking too long
     openmp)              ;; # TODO: having trouble with the Perl installation
     libc)                ;; # no Windows support
-    lldb)                ;; # tests failing
+    lldb)                ;; # custom environment requirements (https://github.com/llvm/llvm-project/pull/94208#issuecomment-2146256857)
     bolt)                ;; # tests are not supported yet
     *)
       echo "${project}"
@@ -213,7 +212,7 @@ function check-targets() {
       echo "check-unwind"
     ;;
     lldb)
-      echo "check-all" # TODO: check-lldb may not include all the LLDB tests?
+      echo "check-lldb"
     ;;
     pstl)
       echo "check-all"

.ci/monolithic-linux.sh

@@ -39,6 +39,7 @@ targets="${2}"
 
 echo "--- cmake"
 pip install -q -r "${MONOREPO_ROOT}"/mlir/python/requirements.txt
+pip install -q -r "${MONOREPO_ROOT}"/lldb/test/requirements.txt
 cmake -S "${MONOREPO_ROOT}"/llvm -B "${BUILD_DIR}" \
       -D LLVM_ENABLE_PROJECTS="${projects}" \
       -G Ninja \

.github/workflows/ci-post-commit-analyzer-run.py

@@ -0,0 +1,34 @@
+import json
+import multiprocessing
+import os
+import re
+import subprocess
+import sys
+
+
+def run_analyzer(data):
+    os.chdir(data["directory"])
+    command = (
+        data["command"]
+        + f" --analyze --analyzer-output html -o analyzer-results -Xclang -analyzer-config -Xclang max-nodes=75000"
+    )
+    print(command)
+    subprocess.run(command, shell=True, check=True)
+
+
+def pool_error(e):
+    print("Error analyzing file:", e)
+
+
+def main():
+    db_path = sys.argv[1]
+    database = json.load(open(db_path))
+
+    with multiprocessing.Pool() as pool:
+        pool.map_async(run_analyzer, [k for k in database], error_callback=pool_error)
+        pool.close()
+        pool.join()
+
+
+if __name__ == "__main__":
+    main()

.github/workflows/ci-post-commit-analyzer.yml

@@ -0,0 +1,95 @@
+name: Post-Commit Static Analyzer
+
+permissions:
+  contents: read
+
+on:
+  push:
+    branches:
+      - 'release/**'
+    paths:
+      - 'clang/**'
+      - 'llvm/**'
+      - '.github/workflows/ci-post-commit-analyzer.yml'
+  pull_request:
+    types:
+      - opened
+      - synchronize
+      - reopened
+      - closed
+    paths:
+      - '.github/workflows/ci-post-commit-analyzer.yml'
+      - '.github/workflows/ci-post-commit-analyzer-run.py'
+  schedule:
+    - cron: '30 0 * * *'
+
+concurrency:
+  group: >-
+    llvm-project-${{ github.workflow }}-${{ github.event_name == 'pull_request' &&
+      ( github.event.pull_request.number || github.ref) }}
+  cancel-in-progress: ${{ startsWith(github.ref, 'refs/pull/') }}
+
+jobs:
+  post-commit-analyzer:
+    if: >-
+      github.repository_owner == 'llvm' &&
+      github.event.action != 'closed'
+    runs-on: ubuntu-22.04
+    container:
+      image: 'ghcr.io/llvm/ci-ubuntu-22.04:latest'
+    env:
+      LLVM_VERSION: 18
+    steps:
+      - name: Checkout Source
+        uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
+
+      - name: Setup ccache
+        uses: hendrikmuhs/ccache-action@v1
+        with:
+          # A full build of llvm, clang, lld, and lldb takes about 250MB
+          # of ccache space. There's not much reason to have more than this,
+          # because we usually won't need to save cache entries from older
+          # builds.  Also, there is an overall 10GB cache limit, and each
+          # run creates a new cache entry so we want to ensure that we have
+          # enough cache space for all the tests to run at once and still
+          # fit under the 10 GB limit.
+          # Default to 2G to workaround: https://github.com/hendrikmuhs/ccache-action/issues/174
+          max-size: 2G
+          key: post-commit-analyzer
+          variant: sccache
+
+      - name: Configure
+        run: |
+              cmake -B build -S llvm -G Ninja \
+                  -DLLVM_ENABLE_ASSERTIONS=ON \
+                  -DLLVM_ENABLE_PROJECTS=clang \
+                  -DLLVM_BUILD_LLVM_DYLIB=ON \
+                  -DLLVM_LINK_LLVM_DYLIB=ON \
+                  -DCMAKE_CXX_COMPILER=clang++ \
+                  -DCMAKE_C_COMPILER=clang \
+                  -DCMAKE_CXX_COMPILER_LAUNCHER=sccache \
+                  -DCMAKE_C_COMPILER_LAUNCHER=sccache \
+                  -DCMAKE_EXPORT_COMPILE_COMMANDS=ON \
+                  -DLLVM_INCLUDE_TESTS=OFF \
+                  -DCLANG_INCLUDE_TESTS=OFF \
+                  -DCMAKE_BUILD_TYPE=Release
+
+      - name: Build
+        run: |
+          # FIXME: We need to build all the generated header files in order to be able to run
+          # the analyzer on every file.  Building libLLVM and libclang is probably overkill for
+          # this, but it's better than building every target.
+          ninja -v -C build libLLVM.so libclang.so
+
+          # Run the analyzer.
+          python3 .github/workflows/ci-post-commit-analyzer-run.py build/compile_commands.json
+
+          scan-build --generate-index-only build/analyzer-results
+
+      - name: Upload Results
+        uses: actions/upload-artifact@26f96dfa697d77e81fd5907df203aa23a56210a8 #v4.3.0
+        if: always()
+        with:
+          name: analyzer-results
+          path: 'build/analyzer-results/*'
+

.github/workflows/containers/github-action-ci/stage1.Dockerfile

@@ -37,7 +37,7 @@ RUN cmake -B ./build -G Ninja ./llvm \
   -DLLVM_ENABLE_RUNTIMES="compiler-rt" \
   -DCMAKE_INSTALL_PREFIX="$LLVM_SYSROOT" \
   -DLLVM_ENABLE_PROJECTS="bolt;clang;lld;clang-tools-extra" \
-  -DLLVM_DISTRIBUTION_COMPONENTS="lld;compiler-rt;clang-format" \
+  -DLLVM_DISTRIBUTION_COMPONENTS="lld;compiler-rt;clang-format;scan-build" \
   -DCLANG_DEFAULT_LINKER="lld" \
   -DBOOTSTRAP_CLANG_PGO_TRAINING_DATA_SOURCE_DIR=/llvm-project-llvmorg-$LLVM_VERSION/llvm
 

.github/workflows/release-binaries.yml

@@ -156,6 +156,8 @@ jobs:
         rm build.tar.zst
 
     - name: Build Stage 2
+      # Re-enable once PGO builds are supported.
+      if: false
       run: |
         ninja -C /mnt/build stage2-instrumented
 
@@ -214,7 +216,7 @@ jobs:
 
     - id: package-info
       run: |
-        filename="LLVM-${{ needs.prepare.outputs.release-version }}-Linux.tar.gz"
+        filename="LLVM-${{ needs.prepare.outputs.release-version }}-Linux.tar.xz"
         echo "filename=$filename" >> $GITHUB_OUTPUT
         echo "path=/mnt/build/tools/clang/stage2-bins/$filename" >> $GITHUB_OUTPUT
 

bolt/include/bolt/Core/DIEBuilder.h

@@ -215,10 +215,9 @@ class DIEBuilder {
   /// Along with current CU, and DIE being processed and the new DIE offset to
   /// be updated, it takes in Parents vector that can be empty if this DIE has
   /// no parents.
-  uint32_t
-  finalizeDIEs(DWARFUnit &CU, DIE &Die,
-               std::vector<std::optional<BOLTDWARF5AccelTableData *>> &Parents,
-               uint32_t &CurOffset);
+  uint32_t finalizeDIEs(DWARFUnit &CU, DIE &Die,
+                        std::optional<BOLTDWARF5AccelTableData *> Parent,
+                        uint32_t NumberParentsInChain, uint32_t &CurOffset);
 
   void registerUnit(DWARFUnit &DU, bool NeedSort);
 

bolt/include/bolt/Core/DebugNames.h

@@ -24,23 +24,25 @@ class BOLTDWARF5AccelTableData : public DWARF5AccelTableData {
   BOLTDWARF5AccelTableData(const uint64_t DieOffset,
                            const std::optional<uint64_t> DefiningParentOffset,
                            const unsigned DieTag, const unsigned UnitID,
-                           const bool IsTU,
+                           const bool IsParentRoot, const bool IsTU,
                            const std::optional<unsigned> SecondUnitID)
       : DWARF5AccelTableData(DieOffset, DefiningParentOffset, DieTag, UnitID,
                              IsTU),
-        SecondUnitID(SecondUnitID) {}
+        SecondUnitID(SecondUnitID), IsParentRoot(IsParentRoot) {}
 
   uint64_t getDieOffset() const { return DWARF5AccelTableData::getDieOffset(); }
   unsigned getDieTag() const { return DWARF5AccelTableData::getDieTag(); }
   unsigned getUnitID() const { return DWARF5AccelTableData::getUnitID(); }
   bool isTU() const { return DWARF5AccelTableData::isTU(); }
+  bool isParentRoot() const { return IsParentRoot; }
   std::optional<unsigned> getSecondUnitID() const { return SecondUnitID; }
 
   void setPatchOffset(uint64_t PatchOffset) { OffsetVal = PatchOffset; }
   uint64_t getPatchOffset() const { return std::get<uint64_t>(OffsetVal); }
 
 private:
   std::optional<unsigned> SecondUnitID;
+  bool IsParentRoot;
 };
 
 class DWARF5AcceleratorTable {
@@ -57,6 +59,7 @@ class DWARF5AcceleratorTable {
   std::optional<BOLTDWARF5AccelTableData *>
   addAccelTableEntry(DWARFUnit &Unit, const DIE &Die,
                      const std::optional<uint64_t> &DWOID,
+                     const uint32_t NumberParentsInChain,
                      std::optional<BOLTDWARF5AccelTableData *> &Parent);
   /// Set current unit being processed.
   void setCurrentUnit(DWARFUnit &Unit, const uint64_t UnitStartOffset);

bolt/include/bolt/Core/GDBIndex.h

@@ -0,0 +1,61 @@
+//===-- bolt/Core/GDBIndex.h - GDB Index support ----------------*- 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 contains declaration of classes required for generation of
+/// .gdb_index section.
+///
+//===----------------------------------------------------------------------===//
+
+#ifndef BOLT_CORE_GDB_INDEX_H
+#define BOLT_CORE_GDB_INDEX_H
+
+#include "bolt/Core/BinaryContext.h"
+#include <vector>
+
+namespace llvm {
+namespace bolt {
+
+class GDBIndex {
+public:
+  /// Contains information about TU so we can write out correct entries in GDB
+  /// index.
+  struct GDBIndexTUEntry {
+    uint64_t UnitOffset;
+    uint64_t TypeHash;
+    uint64_t TypeDIERelativeOffset;
+  };
+
+private:
+  BinaryContext &BC;
+
+  /// Entries for GDB Index Types CU List.
+  using GDBIndexTUEntryType = std::vector<GDBIndexTUEntry>;
+  GDBIndexTUEntryType GDBIndexTUEntryVector;
+
+public:
+  GDBIndex(BinaryContext &BC) : BC(BC) {}
+
+  std::mutex GDBIndexMutex;
+
+  /// Adds an GDBIndexTUEntry if .gdb_index section exists.
+  void addGDBTypeUnitEntry(const GDBIndexTUEntry &&Entry);
+
+  /// Rewrite .gdb_index section if present.
+  void updateGdbIndexSection(const CUOffsetMap &CUMap, const uint32_t NumCUs,
+                             DebugARangesSectionWriter &ARangesSectionWriter);
+
+  /// Returns all entries needed for Types CU list.
+  const GDBIndexTUEntryType &getGDBIndexTUEntryVector() const {
+    return GDBIndexTUEntryVector;
+  }
+};
+
+} // namespace bolt
+} // namespace llvm
+
+#endif

bolt/include/bolt/Core/MCPlusBuilder.h

@@ -1706,12 +1706,9 @@ class MCPlusBuilder {
   }
 
   /// Reverses the branch condition in Inst and update its taken target to TBB.
-  ///
-  /// Returns true on success.
-  virtual bool reverseBranchCondition(MCInst &Inst, const MCSymbol *TBB,
+  virtual void reverseBranchCondition(MCInst &Inst, const MCSymbol *TBB,
                                       MCContext *Ctx) const {
     llvm_unreachable("not implemented");
-    return false;
   }
 
   virtual bool replaceBranchCondition(MCInst &Inst, const MCSymbol *TBB,
@@ -1751,12 +1748,9 @@ class MCPlusBuilder {
   }
 
   /// Sets the taken target of the branch instruction to Target.
-  ///
-  /// Returns true on success.
-  virtual bool replaceBranchTarget(MCInst &Inst, const MCSymbol *TBB,
+  virtual void replaceBranchTarget(MCInst &Inst, const MCSymbol *TBB,
                                    MCContext *Ctx) const {
     llvm_unreachable("not implemented");
-    return false;
   }
 
   /// Extract a symbol and an addend out of the fixup value expression.

bolt/lib/Core/BinaryEmitter.cpp

@@ -194,6 +194,7 @@ class BinaryEmitter {
 
 void BinaryEmitter::emitAll(StringRef OrgSecPrefix) {
   Streamer.initSections(false, *BC.STI);
+  Streamer.setUseAssemblerInfoForParsing(false);
 
   if (opts::UpdateDebugSections && BC.isELF()) {
     // Force the emission of debug line info into allocatable section to ensure
@@ -226,6 +227,7 @@ void BinaryEmitter::emitAll(StringRef OrgSecPrefix) {
   // TODO Enable for Mach-O once BinaryContext::getDataSection supports it.
   if (BC.isELF())
     AddressMap::emit(Streamer, BC);
+  Streamer.setUseAssemblerInfoForParsing(true);
 }
 
 void BinaryEmitter::emitFunctions() {

bolt/lib/Core/CMakeLists.txt

@@ -25,6 +25,7 @@ add_llvm_library(LLVMBOLTCore
   DynoStats.cpp
   Exceptions.cpp
   FunctionLayout.cpp
+  GDBIndex.cpp
   HashUtilities.cpp
   JumpTable.cpp
   MCPlusBuilder.cpp