Merge branch 'main' into extract_vec2

Author: abhishek-kaushik22

bolt/include/bolt/Core/BinaryFunction.h

@@ -2407,6 +2407,19 @@ inline raw_ostream &operator<<(raw_ostream &OS,
   return OS;
 }
 
+/// Compare function by index if it is valid, fall back to the original address
+/// otherwise.
+inline bool compareBinaryFunctionByIndex(const BinaryFunction *A,
+                                         const BinaryFunction *B) {
+  if (A->hasValidIndex() && B->hasValidIndex())
+    return A->getIndex() < B->getIndex();
+  if (A->hasValidIndex() && !B->hasValidIndex())
+    return true;
+  if (!A->hasValidIndex() && B->hasValidIndex())
+    return false;
+  return A->getAddress() < B->getAddress();
+}
+
 } // namespace bolt
 
 // GraphTraits specializations for function basic block graphs (CFGs)

bolt/include/bolt/Profile/DataAggregator.h

@@ -170,9 +170,6 @@ class DataAggregator : public DataReader {
   std::string BuildIDBinaryName;
 
   /// Memory map info for a single file as recorded in perf.data
-  /// When a binary has multiple text segments, the Size is computed as the
-  /// difference of the last address of these segments from the BaseAddress.
-  /// The base addresses of all text segments must be the same.
   struct MMapInfo {
     uint64_t BaseAddress{0}; /// Base address of the mapped binary.
     uint64_t MMapAddress{0}; /// Address of the executable segment.
@@ -496,11 +493,6 @@ class DataAggregator : public DataReader {
   /// and return a file name matching a given \p FileBuildID.
   std::optional<StringRef> getFileNameForBuildID(StringRef FileBuildID);
 
-  /// Get a constant reference to the parsed binary mmap entries.
-  const std::unordered_map<uint64_t, MMapInfo> &getBinaryMMapInfo() {
-    return BinaryMMapInfo;
-  }
-
   friend class YAMLProfileWriter;
 };
 } // namespace bolt

bolt/lib/Core/BinaryContext.cpp

@@ -1607,13 +1607,7 @@ std::vector<BinaryFunction *> BinaryContext::getSortedFunctions() {
                   SortedFunctions.begin(),
                   [](BinaryFunction &BF) { return &BF; });
 
-  llvm::stable_sort(SortedFunctions,
-                    [](const BinaryFunction *A, const BinaryFunction *B) {
-                      if (A->hasValidIndex() && B->hasValidIndex()) {
-                        return A->getIndex() < B->getIndex();
-                      }
-                      return A->hasValidIndex();
-                    });
+  llvm::stable_sort(SortedFunctions, compareBinaryFunctionByIndex);
   return SortedFunctions;
 }
 

bolt/lib/Core/BinaryFunction.cpp

@@ -385,16 +385,7 @@ bool BinaryFunction::isForwardCall(const MCSymbol *CalleeSymbol) const {
   if (CalleeBF) {
     if (CalleeBF->isInjected())
       return true;
-
-    if (hasValidIndex() && CalleeBF->hasValidIndex()) {
-      return getIndex() < CalleeBF->getIndex();
-    } else if (hasValidIndex() && !CalleeBF->hasValidIndex()) {
-      return true;
-    } else if (!hasValidIndex() && CalleeBF->hasValidIndex()) {
-      return false;
-    } else {
-      return getAddress() < CalleeBF->getAddress();
-    }
+    return compareBinaryFunctionByIndex(this, CalleeBF);
   } else {
     // Absolute symbol.
     ErrorOr<uint64_t> CalleeAddressOrError = BC.getSymbolValue(*CalleeSymbol);

bolt/lib/Passes/ReorderFunctions.cpp

@@ -473,16 +473,7 @@ Error ReorderFunctions::runOnFunctions(BinaryContext &BC) {
                     [](BinaryFunction &BF) { return &BF; });
 
     // Sort functions by index.
-    llvm::stable_sort(SortedFunctions,
-                      [](const BinaryFunction *A, const BinaryFunction *B) {
-                        if (A->hasValidIndex() && B->hasValidIndex())
-                          return A->getIndex() < B->getIndex();
-                        if (A->hasValidIndex() && !B->hasValidIndex())
-                          return true;
-                        if (!A->hasValidIndex() && B->hasValidIndex())
-                          return false;
-                        return A->getAddress() < B->getAddress();
-                      });
+    llvm::stable_sort(SortedFunctions, compareBinaryFunctionByIndex);
 
     for (const BinaryFunction *Func : SortedFunctions) {
       if (!Func->hasValidIndex())

bolt/lib/Profile/DataAggregator.cpp

@@ -95,12 +95,6 @@ cl::opt<bool> ReadPreAggregated(
     "pa", cl::desc("skip perf and read data from a pre-aggregated file format"),
     cl::cat(AggregatorCategory));
 
-cl::opt<std::string>
-    ReadPerfEvents("perf-script-events",
-                   cl::desc("skip perf event collection by supplying a "
-                            "perf-script output in a textual format"),
-                   cl::ReallyHidden, cl::init(""), cl::cat(AggregatorCategory));
-
 static cl::opt<bool>
 TimeAggregator("time-aggr",
   cl::desc("time BOLT aggregator"),
@@ -173,9 +167,8 @@ void DataAggregator::findPerfExecutable() {
 void DataAggregator::start() {
   outs() << "PERF2BOLT: Starting data aggregation job for " << Filename << "\n";
 
-  // Don't launch perf for pre-aggregated files or when perf input is specified
-  // by the user.
-  if (opts::ReadPreAggregated || !opts::ReadPerfEvents.empty())
+  // Don't launch perf for pre-aggregated files
+  if (opts::ReadPreAggregated)
     return;
 
   findPerfExecutable();
@@ -471,13 +464,6 @@ void DataAggregator::filterBinaryMMapInfo() {
 
 int DataAggregator::prepareToParse(StringRef Name, PerfProcessInfo &Process,
                                    PerfProcessErrorCallbackTy Callback) {
-  if (!opts::ReadPerfEvents.empty()) {
-    outs() << "PERF2BOLT: using pre-processed perf events for '" << Name
-           << "' (perf-script-events)\n";
-    ParsingBuf = opts::ReadPerfEvents;
-    return 0;
-  }
-
   std::string Error;
   outs() << "PERF2BOLT: waiting for perf " << Name
          << " collection to finish...\n";
@@ -2070,6 +2056,15 @@ std::error_code DataAggregator::parseMMapEvents() {
     if (FileMMapInfo.first == "(deleted)")
       continue;
 
+    // Consider only the first mapping of the file for any given PID
+    auto Range = GlobalMMapInfo.equal_range(FileMMapInfo.first);
+    bool PIDExists = llvm::any_of(make_range(Range), [&](const auto &MI) {
+      return MI.second.PID == FileMMapInfo.second.PID;
+    });
+
+    if (PIDExists)
+      continue;
+
     GlobalMMapInfo.insert(FileMMapInfo);
   }
 
@@ -2121,22 +2116,12 @@ std::error_code DataAggregator::parseMMapEvents() {
                << " using file offset 0x" << Twine::utohexstr(MMapInfo.Offset)
                << ". Ignoring profile data for this mapping\n";
         continue;
+      } else {
+        MMapInfo.BaseAddress = *BaseAddress;
       }
-      MMapInfo.BaseAddress = *BaseAddress;
     }
 
-    // Try to add MMapInfo to the map and update its size. Large binaries may
-    // span to multiple text segments, so the mapping is inserted only on the
-    // first occurrence.
-    if (!BinaryMMapInfo.insert(std::make_pair(MMapInfo.PID, MMapInfo)).second)
-      assert(MMapInfo.BaseAddress == BinaryMMapInfo[MMapInfo.PID].BaseAddress &&
-             "Base address on multiple segment mappings should match");
-
-    // Update mapping size.
-    const uint64_t EndAddress = MMapInfo.MMapAddress + MMapInfo.Size;
-    const uint64_t Size = EndAddress - BinaryMMapInfo[MMapInfo.PID].BaseAddress;
-    if (Size > BinaryMMapInfo[MMapInfo.PID].Size)
-      BinaryMMapInfo[MMapInfo.PID].Size = Size;
+    BinaryMMapInfo.insert(std::make_pair(MMapInfo.PID, MMapInfo));
   }
 
   if (BinaryMMapInfo.empty()) {

bolt/test/AArch64/data-at-0-offset.c

@@ -1,4 +1,4 @@
-// RUN: %clang %cflags -O2 -fPIE -Wl,-q -pie  %s -o %t.exe
+// RUN: %clang %cflags -O2 -fPIE -std=gnu99 -Wl,-q -pie  %s -o %t.exe
 // RUN: llvm-bolt %t.exe -o %t.bolt 2>&1 | FileCheck %s
 // CHECK-NOT: BOLT-WARNING: unable to disassemble instruction at offset
 

bolt/test/AArch64/double_jump.cpp

@@ -1,6 +1,6 @@
 // A contrived example to test the double jump removal peephole.
 
-// RUN: %clang %cflags -O0 %s -o %t.exe
+// RUN: %clangxx %cxxflags -O0 %s -o %t.exe
 // RUN: llvm-bolt %t.exe -o %t.bolt --peepholes=double-jumps | \
 // RUN:   FileCheck %s -check-prefix=CHECKBOLT
 // RUN: llvm-objdump --no-print-imm-hex -d %t.bolt | FileCheck %s

bolt/test/R_ABS.pic.lld.cpp

@@ -6,7 +6,7 @@
  * with libc available.
  * REQUIRES: system-linux
  *
- * RUN: %clang %cflags -fPIC -shared %s -o %t.so -Wl,-q -fuse-ld=lld
+ * RUN: %clangxx %cxxflags -fPIC -shared %s -o %t.so -Wl,-q -fuse-ld=lld
  * RUN: llvm-bolt %t.so -o %t.so.bolt --relocs
  */
 

bolt/test/X86/double-jump.test

@@ -4,7 +4,7 @@
 ## correctly on Windows e.g. subshell execution
 REQUIRES: shell
 
-RUN: %clang %cflags %p/Inputs/double_jump.cpp -o %t.exe
+RUN: %clangxx %cxxflags %p/Inputs/double_jump.cpp -o %t.exe
 RUN: (llvm-bolt %t.exe --peepholes=double-jumps \
 RUN:   --eliminate-unreachable -o %t 2>&1 \
 RUN:   && llvm-objdump -d %t --print-imm-hex --no-show-raw-insn) | FileCheck %s