Pull in Trunk

.git-blame-ignore-revs

@@ -43,3 +43,6 @@ d8f0e6caa91e230a486c948ab643174e40bdf215
 
 # Use C++11 default member initializers in LLDB. NFC.
 9494c510af56d9c8593ab69017dcaa232210b235
+
+# [libc++][NFC] clang-format <__config>
+ac251726f84d5b7e6533a2e3712920184435b61b

.github/workflows/closed-issues.yml

@@ -10,4 +10,4 @@ jobs:
     steps:
       - uses: andymckay/labeler@1.0.4
         with:
-          remove-labels: "awaiting-review"
+          remove-labels: 'awaiting-review'

.github/workflows/issue-release-workflow.yml

@@ -4,7 +4,7 @@
 # /cherry-pick <commit> <...>
 #
 # This comment will attempt to cherry-pick the given commits to the latest
-# release branch (release/Y.x) and if successful push the result to a branch
+# release branch (release/Y.x) and if successful, push the result to a branch
 # on github.
 #
 # /branch <owner>/<repo>/<branch>
@@ -31,9 +31,9 @@ jobs:
     name: Backport Commits
     runs-on: ubuntu-20.04
     if: >-
-        (github.repository == 'llvm/llvm-project') &&
-        !startswith(github.event.comment.body, '<!--IGNORE-->') &&
-        contains(github.event.action == 'opened' && github.event.issue.body || github.event.comment.body, '/cherry-pick')
+      (github.repository == 'llvm/llvm-project') &&
+      !startswith(github.event.comment.body, '<!--IGNORE-->') &&
+      contains(github.event.action == 'opened' && github.event.issue.body || github.event.comment.body, '/cherry-pick')
     steps:
       - name: Fetch LLVM sources
         uses: actions/checkout@v2
@@ -64,16 +64,16 @@ jobs:
     name: Create Pull Request
     runs-on: ubuntu-20.04
     if: >-
-        (github.repository == 'llvm/llvm-project') &&
-        !startswith(github.event.comment.body, '<!--IGNORE-->') &&
-        contains(github.event.comment.body, '/branch')
+      (github.repository == 'llvm/llvm-project') &&
+      !startswith(github.event.comment.body, '<!--IGNORE-->') &&
+      contains(github.event.comment.body, '/branch')
 
     steps:
       - name: Fetch LLVM sources
         uses: actions/checkout@v2
 
       - name: Setup Environment
-        run:  |
+        run: |
           pip install -r ./llvm/utils/git/requirements.txt
 
       - name: Create Pull Request

.github/workflows/issue-subscriber.yml

@@ -10,20 +10,20 @@ jobs:
     runs-on: ubuntu-latest
     if: github.repository == 'llvm/llvm-project'
     steps:
-    - name: Setup Automation Script
-      run:  |
-        curl -O -L https://raw.githubusercontent.com/$GITHUB_REPOSITORY/$GITHUB_SHA/llvm/utils/git/github-automation.py
-        curl -O -L https://raw.githubusercontent.com/$GITHUB_REPOSITORY/$GITHUB_SHA/llvm/utils/git/requirements.txt
-        chmod a+x github-automation.py
-        pip install -r requirements.txt
+      - name: Setup Automation Script
+        run: |
+          curl -O -L https://raw.githubusercontent.com/$GITHUB_REPOSITORY/$GITHUB_SHA/llvm/utils/git/github-automation.py
+          curl -O -L https://raw.githubusercontent.com/$GITHUB_REPOSITORY/$GITHUB_SHA/llvm/utils/git/requirements.txt
+          chmod a+x github-automation.py
+          pip install -r requirements.txt
 
-    - name: Update watchers
-      # https://docs.github.com/en/actions/security-guides/security-hardening-for-github-actions#using-an-intermediate-environment-variable
-      env:
-        LABEL_NAME: ${{ github.event.label.name }}
-      run: |
-        ./github-automation.py \
-          --token '${{ secrets.ISSUE_SUBSCRIBER_TOKEN }}' \
-          issue-subscriber \
-          --issue-number '${{ github.event.issue.number }}' \
-          --label-name "$LABEL_NAME"
+      - name: Update watchers
+        # https://docs.github.com/en/actions/security-guides/security-hardening-for-github-actions#using-an-intermediate-environment-variable
+        env:
+          LABEL_NAME: ${{ github.event.label.name }}
+        run: |
+          ./github-automation.py \
+            --token '${{ secrets.ISSUE_SUBSCRIBER_TOKEN }}' \
+            issue-subscriber \
+            --issue-number '${{ github.event.issue.number }}' \
+            --label-name "$LABEL_NAME"

.github/workflows/llvm-bugs.yml

@@ -19,39 +19,39 @@ jobs:
           MAILGUN_API_KEY: ${{ secrets.LLVM_BUGS_KEY }}
         with:
           script: |
-            const Mailgun = require("mailgun.js");
+            const Mailgun = require('mailgun.js');
             const formData = require('form-data');
-            const mailgun = new Mailgun(formData);
 
-            const DOMAIN = "email.llvm.org";
+            const mailgun = new Mailgun(formData);
+            const DOMAIN = 'email.llvm.org';
 
-            const mg = mailgun.client({username: 'api', key: process.env.MAILGUN_API_KEY});
+            const mg = mailgun.client({ username: 'api', key: process.env.MAILGUN_API_KEY });
 
             github.rest.issues.get({
               issue_number: context.issue.number,
               owner: context.repo.owner,
-              repo: context.repo.repo,
+              repo: context.repo.repo
             })
-            .then(function(issue) {
+            .then((issue) => {
               const payload = {
                 author : issue.data.user.login,
                 issue  : issue.data.number,
                 title  : issue.data.title,
                 url    : issue.data.html_url,
-                labels : issue.data.labels.map(label => { return label.name }),
-                assignee : issue.data.assignees.map(assignee => { return assignee.login }),
+                labels : issue.data.labels.map((label) => label.name),
+                assignee : issue.data.assignees.map((assignee) => assignee.login),
                 body   : issue.data.body
               };
-            
+
               const data = {
-                from: "LLVM Bugs <llvm-bugs@email.llvm.org>",
-                to: "llvm-bugs@lists.llvm.org",
+                from: 'LLVM Bugs <llvm-bugs@email.llvm.org>',
+                to: 'llvm-bugs@lists.llvm.org',
                 subject: `[Bug ${issue.data.number}] ${issue.data.title}`,
-                template: "new-github-issue",
+                template: 'new-github-issue',
                 'o:tracking-clicks': 'no',
                 'h:X-Mailgun-Variables': JSON.stringify(payload)
               };
 
-              return mg.messages.create(DOMAIN, data)
+              return mg.messages.create(DOMAIN, data);
             })
-            .then(msg => console.log(msg));
+            .then((msg) => console.log(msg));

.github/workflows/new-issues.yml

@@ -10,5 +10,5 @@ jobs:
     steps:
       - uses: andymckay/labeler@1.0.4
         with:
-          add-labels: "new issue"
+          add-labels: 'new issue'
           ignore-if-labeled: true

bolt/README.md

@@ -180,7 +180,7 @@ Once you have `perf.fdata` ready, you can use it for optimizations with
 BOLT. Assuming your environment is setup to include the right path, execute
 `llvm-bolt`:
 ```
-$ llvm-bolt <executable> -o <executable>.bolt -data=perf.fdata -reorder-blocks=cache+ -reorder-functions=hfsort -split-functions=2 -split-all-cold -split-eh -dyno-stats
+$ llvm-bolt <executable> -o <executable>.bolt -data=perf.fdata -reorder-blocks=ext-tsp -reorder-functions=hfsort -split-functions=2 -split-all-cold -split-eh -dyno-stats
 ```
 
 If you do need an updated debug info, then add `-update-debug-sections` option

bolt/docs/OptimizingClang.md

@@ -64,7 +64,7 @@ Notice that we are passing `clang-7` to `perf2bolt` which is the real binary tha
 the generated profile:
 ```bash
 $ llvm-bolt $CPATH/clang-7 -o $CPATH/clang-7.bolt -b clang-7.yaml \
-    -reorder-blocks=cache+ -reorder-functions=hfsort+ -split-functions=3 \
+    -reorder-blocks=ext-tsp -reorder-functions=hfsort+ -split-functions=3 \
     -split-all-cold -dyno-stats -icf=1 -use-gnu-stack
 ```
 The output will look similar to the one below:

bolt/docs/doxygen.cfg.in

@@ -38,7 +38,7 @@ PROJECT_NAME           = "BOLT"
 # could be handy for archiving the generated documentation or if some version
 # control system is used.
 
-PROJECT_NUMBER         =
+PROJECT_NUMBER         = @PACKAGE_VERSION@
 
 # Using the PROJECT_BRIEF tag one can provide an optional one line description
 # for a project that appears at the top of each page and should give viewer a

bolt/include/bolt/Core/BinaryContext.h

@@ -82,6 +82,13 @@ inline raw_ostream &operator<<(raw_ostream &OS, const SegmentInfo &SegInfo) {
   return OS;
 }
 
+// AArch64-specific symbol markers used to delimit code/data in .text.
+enum class MarkerSymType : char {
+  NONE = 0,
+  CODE,
+  DATA,
+};
+
 enum class MemoryContentsType : char {
   UNKNOWN = 0,             /// Unknown contents.
   POSSIBLE_JUMP_TABLE,     /// Possibly a non-PIC jump table.
@@ -232,7 +239,7 @@ class BinaryContext {
   Optional<DWARFUnit *> getDWOCU(uint64_t DWOId);
 
   /// Returns DWOContext if it exists.
-  DWARFContext *getDWOContext();
+  DWARFContext *getDWOContext() const;
 
   /// Get Number of DWOCUs in a map.
   uint32_t getNumDWOCUs() { return DWOCUs.size(); }
@@ -549,6 +556,9 @@ class BinaryContext {
 
   std::unique_ptr<MCDisassembler> DisAsm;
 
+  /// Symbolic disassembler.
+  std::unique_ptr<MCDisassembler> SymbolicDisAsm;
+
   std::unique_ptr<MCAsmBackend> MAB;
 
   /// Indicates if relocations are available for usage.
@@ -662,6 +672,11 @@ class BinaryContext {
            TheTriple->getArch() == llvm::Triple::x86_64;
   }
 
+  // AArch64-specific functions to check if symbol is used to delimit
+  // code/data in .text. Code is marked by $x, data by $d.
+  MarkerSymType getMarkerType(const SymbolRef &Symbol) const;
+  bool isMarker(const SymbolRef &Symbol) const;
+
   /// Iterate over all BinaryData.
   iterator_range<binary_data_const_iterator> getBinaryData() const {
     return make_range(BinaryDataMap.begin(), BinaryDataMap.end());

bolt/include/bolt/Core/BinaryFunction.h

@@ -833,6 +833,29 @@ class BinaryFunction {
     return make_range(JumpTables.begin(), JumpTables.end());
   }
 
+  /// Return relocation associated with a given \p Offset in the function,
+  /// or nullptr if no such relocation exists.
+  const Relocation *getRelocationAt(uint64_t Offset) const {
+    assert(CurrentState == State::Empty &&
+           "Relocations unavailable in the current function state.");
+    auto RI = Relocations.find(Offset);
+    return (RI == Relocations.end()) ? nullptr : &RI->second;
+  }
+
+  /// Return the first relocation in the function that starts at an address in
+  /// the [StartOffset, EndOffset) range. Return nullptr if no such relocation
+  /// exists.
+  const Relocation *getRelocationInRange(uint64_t StartOffset,
+                                         uint64_t EndOffset) const {
+    assert(CurrentState == State::Empty &&
+           "Relocations unavailable in the current function state.");
+    auto RI = Relocations.lower_bound(StartOffset);
+    if (RI != Relocations.end() && RI->first < EndOffset)
+      return &RI->second;
+
+    return nullptr;
+  }
+
   /// Returns the raw binary encoding of this function.
   ErrorOr<ArrayRef<uint8_t>> getData() const;
 
@@ -959,6 +982,15 @@ class BinaryFunction {
     return const_cast<BinaryFunction *>(this)->getInstructionAtOffset(Offset);
   }
 
+  /// Return offset for the first instruction. If there is data at the
+  /// beginning of a function then offset of the first instruction could
+  /// be different from 0
+  uint64_t getFirstInstructionOffset() const {
+    if (Instructions.empty())
+      return 0;
+    return Instructions.begin()->first;
+  }
+
   /// Return jump table that covers a given \p Address in memory.
   JumpTable *getJumpTableContainingAddress(uint64_t Address) {
     auto JTI = JumpTables.upper_bound(Address);
@@ -1305,11 +1337,11 @@ class BinaryFunction {
     case ELF::R_X86_64_PC8:
     case ELF::R_X86_64_PC32:
     case ELF::R_X86_64_PC64:
+    case ELF::R_X86_64_GOTPCRELX:
+    case ELF::R_X86_64_REX_GOTPCRELX:
       Relocations[Offset] = Relocation{Offset, Symbol, RelType, Addend, Value};
       return;
     case ELF::R_X86_64_PLT32:
-    case ELF::R_X86_64_GOTPCRELX:
-    case ELF::R_X86_64_REX_GOTPCRELX:
     case ELF::R_X86_64_GOTPCREL:
     case ELF::R_X86_64_TPOFF32:
     case ELF::R_X86_64_GOTTPOFF:
@@ -1948,11 +1980,6 @@ class BinaryFunction {
     return ColdLSDASymbol;
   }
 
-  /// True if the symbol is a mapping symbol used in AArch64 to delimit
-  /// data inside code section.
-  bool isDataMarker(const SymbolRef &Symbol, uint64_t SymbolSize) const;
-  bool isCodeMarker(const SymbolRef &Symbol, uint64_t SymbolSize) const;
-
   void setOutputDataAddress(uint64_t Address) { OutputDataOffset = Address; }
 
   uint64_t getOutputDataAddress() const { return OutputDataOffset; }