Refactor data-in-code annotations.

Use a dedicated MachO load command to annotate data-in-code regions.
This is the same format the linker produces for final executable images,
allowing consistency of representation and use of introspection tools
for both object and executable files.

Data-in-code regions are annotated via ".data_region"/".end_data_region"
directive pairs, with an optional region type.

data_region_directive := ".data_region" { region_type }
region_type := "jt8" | "jt16" | "jt32" | "jta32"
end_data_region_directive := ".end_data_region"

The previous handling of ARM-style "$d.*" labels was broken and has
been removed. Specifically, it didn't handle ARM vs. Thumb mode when
marking the end of the section.

rdar://11459456

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@157062 91177308-0d34-0410-b5e6-96231b3b80d8

include/llvm/MC/MCAsmInfo.h

@@ -147,6 +147,11 @@ namespace llvm {
     // FIXME: Make this a more general encoding setting?
     bool AllowUTF8;
 
+    /// UseDataRegionDirectives - This is true if data region markers should
+    /// be printed as ".data_region/.end_data_region" directives. If false,
+    /// use "$d/$a" labels instead.
+    bool UseDataRegionDirectives;
+
     //===--- Data Emission Directives -------------------------------------===//
 
     /// ZeroDirective - this should be set to the directive used to get some
@@ -172,18 +177,6 @@ namespace llvm {
     const char *Data32bitsDirective;         // Defaults to "\t.long\t"
     const char *Data64bitsDirective;         // Defaults to "\t.quad\t"
 
-    /// [Data|Code]Begin - These magic labels are used to marked a region as
-    /// data or code, and are used to provide additional information for
-    /// correct disassembly on targets that like to mix data and code within
-    /// a segment.  These labels will be implicitly suffixed by the streamer
-    /// to give them unique names.
-    const char *DataBegin;                   // Defaults to "$d."
-    const char *CodeBegin;                   // Defaults to "$a."
-    const char *JT8Begin;                    // Defaults to "$a."
-    const char *JT16Begin;                   // Defaults to "$a."
-    const char *JT32Begin;                   // Defaults to "$a."
-    bool SupportsDataRegions;
-
     /// GPRel64Directive - if non-null, a directive that is used to emit a word
     /// which should be relocated as a 64-bit GP-relative offset, e.g. .gpdword
     /// on Mips.
@@ -384,14 +377,6 @@ namespace llvm {
     const char *getGPRel64Directive() const { return GPRel64Directive; }
     const char *getGPRel32Directive() const { return GPRel32Directive; }
 
-    /// [Code|Data]Begin label name accessors.
-    const char *getCodeBeginLabelName() const { return CodeBegin; }
-    const char *getDataBeginLabelName() const { return DataBegin; }
-    const char *getJumpTable8BeginLabelName() const { return JT8Begin; }
-    const char *getJumpTable16BeginLabelName() const { return JT16Begin; }
-    const char *getJumpTable32BeginLabelName() const { return JT32Begin; }
-    bool getSupportsDataRegions() const { return SupportsDataRegions; }
-
     /// getNonexecutableStackSection - Targets can implement this method to
     /// specify a section to switch to if the translation unit doesn't have any
     /// trampolines that require an executable stack.
@@ -488,6 +473,9 @@ namespace llvm {
     bool doesAllowUTF8() const {
       return AllowUTF8;
     }
+    bool doesSupportDataRegionDirectives() const {
+      return UseDataRegionDirectives;
+    }
     const char *getZeroDirective() const {
       return ZeroDirective;
     }

include/llvm/MC/MCAssembler.h

@@ -651,6 +651,16 @@ struct IndirectSymbolData {
   MCSectionData *SectionData;
 };
 
+// FIXME: Ditto this. Purely so the Streamer and the ObjectWriter can talk
+// to one another.
+struct DataRegionData {
+  // This enum should be kept in sync w/ the mach-o definition in
+  // llvm/Object/MachOFormat.h.
+  enum KindTy { Data = 1, JumpTable8, JumpTable16, JumpTable32 } Kind;
+  MCSymbol *Start;
+  MCSymbol *End;
+};
+
 class MCAssembler {
   friend class MCAsmLayout;
 
@@ -668,6 +678,10 @@ class MCAssembler {
     const_indirect_symbol_iterator;
   typedef std::vector<IndirectSymbolData>::iterator indirect_symbol_iterator;
 
+  typedef std::vector<DataRegionData>::const_iterator
+    const_data_region_iterator;
+  typedef std::vector<DataRegionData>::iterator data_region_iterator;
+
 private:
   MCAssembler(const MCAssembler&);    // DO NOT IMPLEMENT
   void operator=(const MCAssembler&); // DO NOT IMPLEMENT
@@ -698,6 +712,7 @@ class MCAssembler {
 
   std::vector<IndirectSymbolData> IndirectSymbols;
 
+  std::vector<DataRegionData> DataRegions;
   /// The set of function symbols for which a .thumb_func directive has
   /// been seen.
   //
@@ -883,6 +898,33 @@ class MCAssembler {
 
   size_t indirect_symbol_size() const { return IndirectSymbols.size(); }
 
+  /// @}
+  /// @name Data Region List Access
+  /// @{
+
+  // FIXME: This is a total hack, this should not be here. Once things are
+  // factored so that the streamer has direct access to the .o writer, it can
+  // disappear.
+  std::vector<DataRegionData> &getDataRegions() {
+    return DataRegions;
+  }
+
+  data_region_iterator data_region_begin() {
+    return DataRegions.begin();
+  }
+  const_data_region_iterator data_region_begin() const {
+    return DataRegions.begin();
+  }
+
+  data_region_iterator data_region_end() {
+    return DataRegions.end();
+  }
+  const_data_region_iterator data_region_end() const {
+    return DataRegions.end();
+  }
+
+  size_t data_region_size() const { return DataRegions.size(); }
+
   /// @}
   /// @name Backend Data Access
   /// @{

include/llvm/MC/MCContext.h

@@ -161,6 +161,10 @@ namespace llvm {
     /// with a unique but unspecified name.
     MCSymbol *CreateTempSymbol();
 
+    /// getUniqueSymbolID() - Return a unique identifier for use in constructing
+    /// symbol names.
+    unsigned getUniqueSymbolID() { return NextUniqueID++; }
+
     /// CreateDirectionalLocalSymbol - Create the definition of a directional
     /// local symbol for numbered label (used for "1:" definitions).
     MCSymbol *CreateDirectionalLocalSymbol(int64_t LocalLabelVal);

include/llvm/MC/MCDirectives.h

@@ -52,6 +52,14 @@ enum MCAssemblerFlag {
   MCAF_Code64                 ///< .code64 (X86)
 };
 
+enum MCDataRegionType {
+  MCDR_DataRegion,            ///< .data_region
+  MCDR_DataRegionJT8,         ///< .data_region jt8
+  MCDR_DataRegionJT16,        ///< .data_region jt16
+  MCDR_DataRegionJT32,        ///< .data_region jt32
+  MCDR_DataRegionEnd          ///< .end_data_region
+};
+
 } // end namespace llvm
 
 #endif

include/llvm/MC/MCMachObjectWriter.h

@@ -179,6 +179,9 @@ class MachObjectWriter : public MCObjectWriter {
 
   void WriteNlist(MachSymbolData &MSD, const MCAsmLayout &Layout);
 
+  void WriteLinkeditLoadCommand(uint32_t Type, uint32_t DataOffset,
+                                uint32_t DataSize);
+
   // FIXME: We really need to improve the relocation validation. Basically, we
   // want to implement a separate computation which evaluates the relocation
   // entry as the linker would, and verifies that the resultant fixup value is

include/llvm/MC/MCStreamer.h

@@ -69,22 +69,7 @@ namespace llvm {
     SmallVector<std::pair<const MCSection *,
                 const MCSection *>, 4> SectionStack;
 
-    unsigned UniqueCodeBeginSuffix;
-    unsigned UniqueDataBeginSuffix;
-
   protected:
-    /// Indicator of whether the previous data-or-code indicator was for
-    /// code or not.  Used to determine when we need to emit a new indicator.
-    enum DataType {
-      Data,
-      Code,
-      JumpTable8,
-      JumpTable16,
-      JumpTable32
-    };
-    DataType RegionIndicator;
-
-
     MCStreamer(MCContext &Ctx);
 
     const MCExpr *BuildSymbolDiff(MCContext &Context, const MCSymbol *A,
@@ -241,47 +226,15 @@ namespace llvm {
     /// used in an assignment.
     virtual void EmitLabel(MCSymbol *Symbol);
 
-    /// EmitDataRegion - Emit a label that marks the beginning of a data
-    /// region.
-    /// On ELF targets, this corresponds to an assembler statement such as:
-    ///   $d.1:
-    virtual void EmitDataRegion();
-
-    /// EmitJumpTable8Region - Emit a label that marks the beginning of a
-    /// jump table composed of 8-bit offsets.
-    /// On ELF targets, this corresponds to an assembler statement such as:
-    ///   $d.1:
-    virtual void EmitJumpTable8Region();
-
-    /// EmitJumpTable16Region - Emit a label that marks the beginning of a
-    /// jump table composed of 16-bit offsets.
-    /// On ELF targets, this corresponds to an assembler statement such as:
-    ///   $d.1:
-    virtual void EmitJumpTable16Region();
-
-    /// EmitJumpTable32Region - Emit a label that marks the beginning of a
-    /// jump table composed of 32-bit offsets.
-    /// On ELF targets, this corresponds to an assembler statement such as:
-    ///   $d.1:
-    virtual void EmitJumpTable32Region();
-
-    /// EmitCodeRegion - Emit a label that marks the beginning of a code
-    /// region.
-    /// On ELF targets, this corresponds to an assembler statement such as:
-    ///   $a.1:
-    virtual void EmitCodeRegion();
-
-    /// ForceCodeRegion - Forcibly sets the current region mode to code.  Used
-    /// at function entry points.
-    void ForceCodeRegion() { RegionIndicator = Code; }
-
-
     virtual void EmitEHSymAttributes(const MCSymbol *Symbol,
                                      MCSymbol *EHSymbol);
 
-    /// EmitAssemblerFlag - Note in the output the specified @p Flag
+    /// EmitAssemblerFlag - Note in the output the specified @p Flag.
     virtual void EmitAssemblerFlag(MCAssemblerFlag Flag) = 0;
 
+    /// EmitDataRegion - Note in the output the specified region @p Kind.
+    virtual void EmitDataRegion(MCDataRegionType Kind) {}
+
     /// EmitThumbFunc - Note in the output that the specified @p Func is
     /// a Thumb mode function (ARM target only).
     virtual void EmitThumbFunc(MCSymbol *Func) = 0;

include/llvm/Object/MachOFormat.h

@@ -97,7 +97,8 @@ namespace macho {
     DysymtabLoadCommandSize = 80,
     Nlist32Size = 12,
     Nlist64Size = 16,
-    RelocationInfoSize = 8
+    RelocationInfoSize = 8,
+    LinkeditLoadCommandSize = 16
   };
 
   /// \brief Constants for header magic field.
@@ -140,7 +141,8 @@ namespace macho {
     LCT_UUID = 0x1b,
     LCT_CodeSignature = 0x1d,
     LCT_SegmentSplitInfo = 0x1e,
-    LCT_FunctionStarts = 0x26
+    LCT_FunctionStarts = 0x26,
+    LCT_DataInCode = 0x29
   };
 
   /// \brief Load command structure.
@@ -279,6 +281,18 @@ namespace macho {
     uint64_t Value;
   };
 
+  /// @}
+  /// @name Data-in-code Table Entry
+  /// @{
+
+  // See <mach-o/loader.h>.
+  enum DataRegionType { Data = 1, JumpTable8, JumpTable16, JumpTable32 };
+  struct DataInCodeTableEntry {
+    uint32_t Offset;  /* from mach_header to start of data region */
+    uint16_t Length;  /* number of bytes in data region */
+    uint16_t Kind;    /* a DataRegionType value  */
+  };
+
   /// @}
   /// @name Indirect Symbol Table
   /// @{

include/llvm/Object/MachOObject.h

@@ -174,6 +174,9 @@ class MachOObject {
   void ReadSymbol64TableEntry(
     uint64_t SymbolTableOffset, unsigned Index,
     InMemoryStruct<macho::Symbol64TableEntry> &Res) const;
+  void ReadDataInCodeTableEntry(
+    uint64_t TableOffset, unsigned Index,
+    InMemoryStruct<macho::DataInCodeTableEntry> &Res) const;
   void ReadULEB128s(uint64_t Index, SmallVectorImpl<uint64_t> &Out) const;
 
   /// @}

lib/CodeGen/AsmPrinter/AsmPrinter.cpp

@@ -475,10 +475,8 @@ void AsmPrinter::EmitFunctionHeader() {
 void AsmPrinter::EmitFunctionEntryLabel() {
   // The function label could have already been emitted if two symbols end up
   // conflicting due to asm renaming.  Detect this and emit an error.
-  if (CurrentFnSym->isUndefined()) {
-    OutStreamer.ForceCodeRegion();
+  if (CurrentFnSym->isUndefined())
     return OutStreamer.EmitLabel(CurrentFnSym);
-  }
 
   report_fatal_error("'" + Twine(CurrentFnSym->getName()) +
                      "' label emitted multiple times to assembly file");
@@ -1085,15 +1083,6 @@ void AsmPrinter::EmitJumpTableInfo() {
 
   EmitAlignment(Log2_32(MJTI->getEntryAlignment(*TM.getTargetData())));
 
-  // If we know the form of the jump table, go ahead and tag it as such.
-  if (!JTInDiffSection) {
-    if (MJTI->getEntryKind() == MachineJumpTableInfo::EK_LabelDifference32) {
-      OutStreamer.EmitJumpTable32Region();
-    } else {
-      OutStreamer.EmitDataRegion();
-    }
-  }
-
   for (unsigned JTI = 0, e = JT.size(); JTI != e; ++JTI) {
     const std::vector<MachineBasicBlock*> &JTBBs = JT[JTI].MBBs;
 

lib/MC/MCAsmInfo.cpp

@@ -50,19 +50,14 @@ MCAsmInfo::MCAsmInfo() {
   AllowNameToStartWithDigit = false;
   AllowPeriodsInName = true;
   AllowUTF8 = true;
+  UseDataRegionDirectives = false;
   ZeroDirective = "\t.zero\t";
   AsciiDirective = "\t.ascii\t";
   AscizDirective = "\t.asciz\t";
   Data8bitsDirective = "\t.byte\t";
   Data16bitsDirective = "\t.short\t";
   Data32bitsDirective = "\t.long\t";
   Data64bitsDirective = "\t.quad\t";
-  DataBegin = "$d.";
-  CodeBegin = "$a.";
-  JT8Begin = "$d.";
-  JT16Begin = "$d.";
-  JT32Begin = "$d.";
-  SupportsDataRegions = false;
   SunStyleELFSectionSwitchSyntax = false;
   UsesELFSectionDirectiveForBSS = false;
   AlignDirective = "\t.align\t";

lib/MC/MCAsmInfoCOFF.cpp

@@ -36,8 +36,6 @@ MCAsmInfoCOFF::MCAsmInfoCOFF() {
   SupportsDebugInformation = true;
   DwarfSectionOffsetDirective = "\t.secrel32\t";
   HasMicrosoftFastStdCallMangling = true;
-
-  SupportsDataRegions = false;
 }
 
 void MCAsmInfoMicrosoft::anchor() { }

lib/MC/MCAsmInfoDarwin.cpp

@@ -43,13 +43,6 @@ MCAsmInfoDarwin::MCAsmInfoDarwin() {
   HasMachoTBSSDirective = true; // Uses .tbss
   HasStaticCtorDtorReferenceInStaticMode = true;
 
-  CodeBegin = "L$start$code$";
-  DataBegin = "L$start$data$";
-  JT8Begin  = "L$start$jt8$";
-  JT16Begin = "L$start$jt16$";
-  JT32Begin = "L$start$jt32$";
-  SupportsDataRegions = true;
-
   // FIXME: Darwin 10 and newer don't need this.
   LinkerRequiresNonEmptyDwarfLines = true;
 

lib/MC/MCAsmStreamer.cpp

@@ -138,6 +138,7 @@ class MCAsmStreamer : public MCStreamer {
   virtual void EmitEHSymAttributes(const MCSymbol *Symbol,
                                    MCSymbol *EHSymbol);
   virtual void EmitAssemblerFlag(MCAssemblerFlag Flag);
+  virtual void EmitDataRegion(MCDataRegionType Kind);
   virtual void EmitThumbFunc(MCSymbol *Func);
 
   virtual void EmitAssignment(MCSymbol *Symbol, const MCExpr *Value);
@@ -352,6 +353,21 @@ void MCAsmStreamer::EmitAssemblerFlag(MCAssemblerFlag Flag) {
   EmitEOL();
 }
 
+void MCAsmStreamer::EmitDataRegion(MCDataRegionType Kind) {
+  MCContext &Ctx = getContext();
+  const MCAsmInfo &MAI = Ctx.getAsmInfo();
+  if (!MAI.doesSupportDataRegionDirectives())
+    return;
+  switch (Kind) {
+  case MCDR_DataRegion:            OS << "\t.data_region"; break;
+  case MCDR_DataRegionJT8:         OS << "\t.data_region jt8"; break;
+  case MCDR_DataRegionJT16:        OS << "\t.data_region jt16"; break;
+  case MCDR_DataRegionJT32:        OS << "\t.data_region jt32"; break;
+  case MCDR_DataRegionEnd:         OS << "\t.end_data_region"; break;
+  }
+  EmitEOL();
+}
+
 void MCAsmStreamer::EmitThumbFunc(MCSymbol *Func) {
   // This needs to emit to a temporary string to get properly quoted
   // MCSymbols when they have spaces in them.

lib/MC/MCELFStreamer.cpp

@@ -13,6 +13,8 @@
 
 #include "MCELF.h"
 #include "llvm/ADT/SmallPtrSet.h"
+#include "llvm/ADT/StringExtras.h"
+#include "llvm/ADT/Twine.h"
 #include "llvm/MC/MCAssembler.h"
 #include "llvm/MC/MCCodeEmitter.h"
 #include "llvm/MC/MCContext.h"