Merge "merge main into amd-staging" into amd-staging

.github/workflows/release-doxygen.yml

@@ -56,12 +56,12 @@ jobs:
           pip3 install --user -r ./llvm/docs/requirements.txt
 
       - name: Build Doxygen
-        env:
-          GITHUB_TOKEN: ${{ github.token }}
         run: |
           ./llvm/utils/release/build-docs.sh -release "${{ inputs.release-version }}" -no-sphinx
 
       - name: Upload Doxygen
         if: env.upload
+        env:
+          GITHUB_TOKEN: ${{ github.token }}
         run: |
           ./llvm/utils/release/github-upload-release.py --token "$GITHUB_TOKEN" --release "${{ inputs.release-version }}" --user "${{ github.actor }}" upload --files ./*doxygen*.tar.xz

.github/workflows/release-tasks.yml

@@ -1,7 +1,7 @@
 name: Release Task
 
 permissions:
-  contents: write
+  contents: read
 
 on:
   push:
@@ -27,6 +27,8 @@ jobs:
   release-create:
     name: Create a New Release
     runs-on: ubuntu-latest
+    permissions:
+      contents: write # For creating the release.
     needs: validate-tag
 
     steps:
@@ -55,6 +57,8 @@ jobs:
 
   release-doxygen:
     name: Build and Upload Release Doxygen
+    permissions:
+      contents: write
     needs:
       - validate-tag
       - release-create
@@ -72,6 +76,8 @@ jobs:
 
   release-binaries:
     name: Build Release Binaries
+    permissions:
+      contents: write
     needs:
       - validate-tag
       - release-create

bolt/include/bolt/Passes/IndirectCallPromotion.h

@@ -104,7 +104,7 @@ class IndirectCallPromotion : public BinaryFunctionPass {
   struct Location {
     MCSymbol *Sym{nullptr};
     uint64_t Addr{0};
-    bool isValid() const { return Sym || (!Sym && Addr != 0); }
+    bool isValid() const { return Sym || Addr != 0; }
     Location() {}
     explicit Location(MCSymbol *Sym) : Sym(Sym) {}
     explicit Location(uint64_t Addr) : Addr(Addr) {}

bolt/lib/Passes/BinaryPasses.cpp

@@ -910,6 +910,11 @@ uint64_t SimplifyConditionalTailCalls::fixTailCalls(BinaryFunction &BF) {
       auto &CTCAnnotation =
           MIB->getOrCreateAnnotationAs<uint64_t>(*CondBranch, "CTCTakenCount");
       CTCAnnotation = CTCTakenFreq;
+      // Preserve Offset annotation, used in BAT.
+      // Instr is a direct tail call instruction that was created when CTCs are
+      // first expanded, and has the original CTC offset set.
+      if (std::optional<uint32_t> Offset = MIB->getOffset(*Instr))
+        MIB->setOffset(*CondBranch, *Offset);
 
       // Remove the unused successor which may be eliminated later
       // if there are no other users.

bolt/lib/Profile/DataAggregator.cpp

@@ -23,6 +23,7 @@
 #include "llvm/ADT/STLExtras.h"
 #include "llvm/ADT/ScopeExit.h"
 #include "llvm/Support/CommandLine.h"
+#include "llvm/Support/Compiler.h"
 #include "llvm/Support/Debug.h"
 #include "llvm/Support/Errc.h"
 #include "llvm/Support/FileSystem.h"
@@ -1999,7 +2000,7 @@ std::error_code DataAggregator::parseMMapEvents() {
     std::pair<StringRef, MMapInfo> FileMMapInfo = FileMMapInfoRes.get();
     if (FileMMapInfo.second.PID == -1)
       continue;
-    if (FileMMapInfo.first.equals("(deleted)"))
+    if (FileMMapInfo.first == "(deleted)")
       continue;
 
     // Consider only the first mapping of the file for any given PID
@@ -2339,7 +2340,7 @@ std::error_code DataAggregator::writeBATYAML(BinaryContext &BC,
         continue;
       BinaryFunction *BF = BC.getBinaryFunctionAtAddress(FuncAddress);
       assert(BF);
-      YamlBF.Name = FuncName.str();
+      YamlBF.Name = getLocationName(*BF);
       YamlBF.Id = BF->getFunctionNumber();
       YamlBF.Hash = BAT->getBFHash(FuncAddress);
       YamlBF.ExecCount = BF->getKnownExecutionCount();
@@ -2378,21 +2379,27 @@ std::error_code DataAggregator::writeBATYAML(BinaryContext &BC,
         return CSI;
       };
 
+      // Lookup containing basic block offset and index
+      auto getBlock = [&BlockMap](uint32_t Offset) {
+        auto BlockIt = BlockMap.upper_bound(Offset);
+        if (LLVM_UNLIKELY(BlockIt == BlockMap.begin())) {
+          errs() << "BOLT-ERROR: invalid BAT section\n";
+          exit(1);
+        }
+        --BlockIt;
+        return std::pair(BlockIt->first, BlockIt->second.getBBIndex());
+      };
+
       for (const auto &[FromOffset, SuccKV] : Branches.IntraIndex) {
-        if (!BlockMap.isInputBlock(FromOffset))
-          continue;
-        const unsigned Index = BlockMap.getBBIndex(FromOffset);
+        const auto &[_, Index] = getBlock(FromOffset);
         yaml::bolt::BinaryBasicBlockProfile &YamlBB = YamlBF.Blocks[Index];
         for (const auto &[SuccOffset, SuccDataIdx] : SuccKV)
           if (BlockMap.isInputBlock(SuccOffset))
             YamlBB.Successors.emplace_back(
                 getSuccessorInfo(SuccOffset, SuccDataIdx));
       }
       for (const auto &[FromOffset, CallTo] : Branches.InterIndex) {
-        auto BlockIt = BlockMap.upper_bound(FromOffset);
-        --BlockIt;
-        const unsigned BlockOffset = BlockIt->first;
-        const unsigned BlockIndex = BlockIt->second.getBBIndex();
+        const auto &[BlockOffset, BlockIndex] = getBlock(FromOffset);
         yaml::bolt::BinaryBasicBlockProfile &YamlBB = YamlBF.Blocks[BlockIndex];
         const uint32_t Offset = FromOffset - BlockOffset;
         for (const auto &[CallToLoc, CallToIdx] : CallTo)
@@ -2403,6 +2410,17 @@ std::error_code DataAggregator::writeBATYAML(BinaryContext &BC,
           return A.Offset < B.Offset;
         });
       }
+      // Set entry counts, similar to DataReader::readProfile.
+      for (const llvm::bolt::BranchInfo &BI : Branches.EntryData) {
+        if (!BlockMap.isInputBlock(BI.To.Offset)) {
+          if (opts::Verbosity >= 1)
+            errs() << "BOLT-WARNING: Unexpected EntryData in " << FuncName
+                   << " at 0x" << Twine::utohexstr(BI.To.Offset) << '\n';
+          continue;
+        }
+        const unsigned BlockIndex = BlockMap.getBBIndex(BI.To.Offset);
+        YamlBF.Blocks[BlockIndex].ExecCount += BI.Branches;
+      }
       // Drop blocks without a hash, won't be useful for stale matching.
       llvm::erase_if(YamlBF.Blocks,
                      [](const yaml::bolt::BinaryBasicBlockProfile &YamlBB) {

bolt/lib/Profile/DataReader.cpp

@@ -1205,8 +1205,7 @@ std::error_code DataReader::parse() {
 
     // Add entry data for branches to another function or branches
     // to entry points (including recursive calls)
-    if (BI.To.IsSymbol &&
-        (!BI.From.Name.equals(BI.To.Name) || BI.To.Offset == 0)) {
+    if (BI.To.IsSymbol && (BI.From.Name != BI.To.Name || BI.To.Offset == 0)) {
       I = GetOrCreateFuncEntry(BI.To.Name);
       I->second.EntryData.emplace_back(std::move(BI));
     }

bolt/lib/Rewrite/DWARFRewriter.cpp

@@ -1550,7 +1550,7 @@ CUOffsetMap DWARFRewriter::finalizeTypeSections(DIEBuilder &DIEBlder,
   for (const SectionRef &Section : Obj->sections()) {
     StringRef Contents = cantFail(Section.getContents());
     StringRef Name = cantFail(Section.getName());
-    if (Name.equals(".debug_types"))
+    if (Name == ".debug_types")
       BC.registerOrUpdateNoteSection(".debug_types", copyByteArray(Contents),
                                      Contents.size());
   }
@@ -1633,10 +1633,10 @@ void DWARFRewriter::finalizeDebugSections(
   for (const SectionRef &Secs : Obj->sections()) {
     StringRef Contents = cantFail(Secs.getContents());
     StringRef Name = cantFail(Secs.getName());
-    if (Name.equals(".debug_abbrev")) {
+    if (Name == ".debug_abbrev") {
       BC.registerOrUpdateNoteSection(".debug_abbrev", copyByteArray(Contents),
                                      Contents.size());
-    } else if (Name.equals(".debug_info")) {
+    } else if (Name == ".debug_info") {
       BC.registerOrUpdateNoteSection(".debug_info", copyByteArray(Contents),
                                      Contents.size());
     }
@@ -1771,7 +1771,7 @@ std::optional<StringRef> updateDebugData(
   };
   switch (SectionIter->second.second) {
   default: {
-    if (!SectionName.equals("debug_str.dwo"))
+    if (SectionName != "debug_str.dwo")
       errs() << "BOLT-WARNING: unsupported debug section: " << SectionName
              << "\n";
     return SectionContents;
@@ -1959,7 +1959,7 @@ void DWARFRewriter::updateDWP(DWARFUnit &CU,
       continue;
     }
 
-    if (SectionName.equals("debug_str.dwo")) {
+    if (SectionName == "debug_str.dwo") {
       CurStrSection = OutData;
     } else {
       // Since handleDebugDataPatching returned true, we already know this is

bolt/lib/Rewrite/SDTRewriter.cpp

@@ -87,7 +87,7 @@ void SDTRewriter::readSection() {
 
     StringRef Name = DE.getCStr(&Offset);
 
-    if (!Name.equals("stapsdt"))
+    if (Name != "stapsdt")
       errs() << "BOLT-WARNING: SDT note name \"" << Name
              << "\" is not expected\n";
 

bolt/test/X86/Inputs/blarge_new_bat_branchentry.preagg.txt

@@ -0,0 +1 @@
+B 80010c 800194 1 0

bolt/test/X86/bolt-address-translation-yaml.test

@@ -5,6 +5,17 @@ RUN: llvm-bolt %t.exe -o %t.out --pa -p %p/Inputs/blarge_new.preagg.txt \
 RUN:   --reorder-blocks=ext-tsp --split-functions --split-strategy=cdsplit \
 RUN:   --reorder-functions=cdsort --enable-bat --dyno-stats --skip-funcs=main \
 RUN:   2>&1 | FileCheck --check-prefix WRITE-BAT-CHECK %s
+# Check that branch with entry in BAT is accounted for.
+RUN: perf2bolt %t.out --pa -p %p/Inputs/blarge_new_bat_branchentry.preagg.txt \
+RUN:   -w %t.yaml -o %t.fdata
+RUN: llvm-bolt %t.exe -data %t.fdata -w %t.yaml-fdata -o %t.null
+RUN: FileCheck --input-file %t.yaml --check-prefix BRANCHENTRY-YAML-CHECK %s
+RUN: FileCheck --input-file %t.yaml-fdata --check-prefix BRANCHENTRY-YAML-CHECK %s
+BRANCHENTRY-YAML-CHECK:    - name: SolveCubic
+BRANCHENTRY-YAML-CHECK:      bid: 0
+BRANCHENTRY-YAML-CHECK:      hash: 0x700F19D24600000
+BRANCHENTRY-YAML-CHECK-NEXT: succ: [ { bid: 7, cnt: 1 }
+# Large profile test
 RUN: perf2bolt %t.out --pa -p %p/Inputs/blarge_new_bat.preagg.txt -w %t.yaml -o %t.fdata \
 RUN:   2>&1 | FileCheck --check-prefix READ-BAT-CHECK %s
 RUN: FileCheck --input-file %t.yaml --check-prefix YAML-BAT-CHECK %s
@@ -48,6 +59,10 @@ YAML-BAT-CHECK-NEXT:   hash:    0x6AF7E61EA3966722
 YAML-BAT-CHECK-NEXT:   exec:    25
 YAML-BAT-CHECK-NEXT:   nblocks: 15
 YAML-BAT-CHECK-NEXT:   blocks:
+YAML-BAT-CHECK-NEXT:   - bid:   0
+YAML-BAT-CHECK-NEXT:     insns: [[#]]
+YAML-BAT-CHECK-NEXT:     hash:  0x700F19D24600000
+YAML-BAT-CHECK-NEXT:     exec:  25
 YAML-BAT-CHECK:        - bid:   3
 YAML-BAT-CHECK-NEXT:     insns: [[#]]
 YAML-BAT-CHECK-NEXT:     hash:  0xDDA1DC5F69F900AC

bolt/test/X86/register-fragments-bolt-symbols.s

@@ -15,6 +15,8 @@
 # PREAGG: B X:0 #chain.cold.0# 1 0
 # RUN: perf2bolt %t.bolt -p %t.preagg --pa -o %t.bat.fdata -w %t.bat.yaml -v=1 \
 # RUN:   | FileCheck %s --check-prefix=CHECK-REGISTER
+# RUN: FileCheck --input-file %t.bat.fdata --check-prefix=CHECK-FDATA %s
+# RUN: FileCheck --input-file %t.bat.yaml --check-prefix=CHECK-YAML %s
 
 # CHECK-SYMS: l df *ABS*          [[#]] chain.s
 # CHECK-SYMS: l  F .bolt.org.text [[#]] chain
@@ -24,6 +26,9 @@
 
 # CHECK-REGISTER: BOLT-INFO: marking chain.cold.0/1(*2) as a fragment of chain/2(*2)
 
+# CHECK-FDATA: 0 [unknown] 0 1 chain/chain.s/2 10 0 1
+# CHECK-YAML: - name: 'chain/chain.s/2'
+
 .file "chain.s"
         .text
         .type   chain, @function

bolt/test/X86/sctc-bug4.test

@@ -1,25 +1,29 @@
-# Check that fallthrough blocks are handled properly.
+# Check that fallthrough blocks are handled properly and Offset annotation is
+# set for conditional tail calls.
 
 RUN: %clang %cflags %S/Inputs/sctc_bug4.s -o %t
-RUN: llvm-bolt %t -o %t.null \
+RUN: llvm-bolt %t -o %t.null --enable-bat \
 RUN:   -funcs=test_func -print-sctc -sequential-disassembly 2>&1 | FileCheck %s
 
 CHECK:      .Ltmp2 (3 instructions, align : 1)
 CHECK-NEXT:   CFI State : 0
+CHECK-NEXT:   Input offset: 0x24
 CHECK-NEXT:   Predecessors: .LFT1
 CHECK-NEXT:     00000024: 	cmpq	$0x20, %rsi
-CHECK-NEXT:     00000028: 	ja	dummy # TAILCALL {{.*}}# CTCTakenCount: 0
+CHECK-NEXT:     00000028: 	ja	dummy # TAILCALL # Offset: 53 # CTCTakenCount: 0
 CHECK-NEXT:     0000002a: 	jmp .Ltmp4
 CHECK-NEXT:   Successors: .Ltmp4
 CHECK-NEXT:   CFI State: 0
 
 CHECK:      .Ltmp1 (2 instructions, align : 1)
 CHECK-NEXT:   CFI State : 0
+CHECK-NEXT:   Input offset: 0x2c
 CHECK-NEXT:   Predecessors: .LFT0
 CHECK-NEXT:     0000002c: 	xorq	%r11, %rax
 CHECK-NEXT:     0000002f: 	retq
 CHECK-NEXT:   CFI State: 0
 
 CHECK:      .Ltmp4 (4 instructions, align : 1)
 CHECK-NEXT:  CFI State : 0
+CHECK-NEXT:  Input offset: 0x3a
 CHECK-NEXT:  Predecessors: .Ltmp2

bolt/test/runtime/bolt-reserved.cpp

@@ -16,8 +16,8 @@
  * not enough for allocating new sections.
  */
 
-// RUN: %clang %s -o %t.exe -Wl,--no-eh-frame-hdr -Wl,-q -DTINY
-// RUN: not llvm-bolt %t.exe -o %t.bolt.exe 2>&1 | \
+// RUN: %clang %s -o %t.tiny.exe -Wl,--no-eh-frame-hdr -Wl,-q -DTINY
+// RUN: not llvm-bolt %t.tiny.exe -o %t.tiny.bolt.exe 2>&1 | \
 // RUN:   FileCheck %s --check-prefix=CHECK-TINY
 
 // CHECK-TINY: BOLT-ERROR: reserved space (1 byte) is smaller than required

clang-tools-extra/clang-tidy/ClangTidy.cpp

@@ -373,11 +373,11 @@ static CheckersList getAnalyzerCheckersAndPackages(ClangTidyContext &Context,
 
   const auto &RegisteredCheckers =
       AnalyzerOptions::getRegisteredCheckers(IncludeExperimental);
-  bool AnalyzerChecksEnabled = false;
-  for (StringRef CheckName : RegisteredCheckers) {
-    std::string ClangTidyCheckName((AnalyzerCheckNamePrefix + CheckName).str());
-    AnalyzerChecksEnabled |= Context.isCheckEnabled(ClangTidyCheckName);
-  }
+  const bool AnalyzerChecksEnabled =
+      llvm::any_of(RegisteredCheckers, [&](StringRef CheckName) -> bool {
+        return Context.isCheckEnabled(
+            (AnalyzerCheckNamePrefix + CheckName).str());
+      });
 
   if (!AnalyzerChecksEnabled)
     return List;

clang-tools-extra/clang-tidy/ClangTidyCheck.cpp

@@ -171,7 +171,7 @@ std::optional<int64_t> ClangTidyCheck::OptionsView::getEnumInt(
     if (IgnoreCase) {
       if (Value.equals_insensitive(NameAndEnum.second))
         return NameAndEnum.first;
-    } else if (Value.equals(NameAndEnum.second)) {
+    } else if (Value == NameAndEnum.second) {
       return NameAndEnum.first;
     } else if (Value.equals_insensitive(NameAndEnum.second)) {
       Closest = NameAndEnum.second;

clang-tools-extra/clang-tidy/bugprone/ForwardingReferenceOverloadCheck.cpp

@@ -25,8 +25,8 @@ AST_MATCHER(QualType, isEnableIf) {
     const NamedDecl *TypeDecl =
         Spec->getTemplateName().getAsTemplateDecl()->getTemplatedDecl();
     return TypeDecl->isInStdNamespace() &&
-           (TypeDecl->getName().equals("enable_if") ||
-            TypeDecl->getName().equals("enable_if_t"));
+           (TypeDecl->getName() == "enable_if" ||
+            TypeDecl->getName() == "enable_if_t");
   };
   const Type *BaseType = Node.getTypePtr();
   // Case: pointer or reference to enable_if.

clang-tools-extra/clang-tidy/modernize/LoopConvertCheck.cpp

@@ -421,7 +421,7 @@ getContainerFromBeginEndCall(const Expr *Init, bool IsBegin, bool *IsArrow,
     return {};
   if (IsReverse && !Call->Name.consume_back("r"))
     return {};
-  if (!Call->Name.empty() && !Call->Name.equals("c"))
+  if (!Call->Name.empty() && Call->Name != "c")
     return {};
   return std::make_pair(Call->Container, Call->CallKind);
 }

clang-tools-extra/clang-tidy/readability/IdentifierNamingCheck.cpp

@@ -1358,7 +1358,7 @@ IdentifierNamingCheck::getFailureInfo(
   std::replace(KindName.begin(), KindName.end(), '_', ' ');
 
   std::string Fixup = fixupWithStyle(Type, Name, Style, HNOption, ND);
-  if (StringRef(Fixup).equals(Name)) {
+  if (StringRef(Fixup) == Name) {
     if (!IgnoreFailedSplit) {
       LLVM_DEBUG(Location.print(llvm::dbgs(), SM);
                  llvm::dbgs()