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

6 files changed, 69 insertions, 71 deletions

diff --git a/include/llvm/MC/MCAsmInfo.h b/include/llvm/MC/MCAsmInfo.h
index 16964c3c9d..ae0dad2fd1 100644
--- a/include/llvm/MC/MCAsmInfo.h
+++ b/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;
     }
diff --git a/include/llvm/MC/MCAssembler.h b/include/llvm/MC/MCAssembler.h
index d139173c3e..4ab7f91f72 100644
--- a/include/llvm/MC/MCAssembler.h
+++ b/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 @@ public:
     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 @@ private:
 
   std::vector<IndirectSymbolData> IndirectSymbols;
 
+  std::vector<DataRegionData> DataRegions;
   /// The set of function symbols for which a .thumb_func directive has
   /// been seen.
   //
@@ -884,6 +899,33 @@ public:
   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
   /// @{
 
diff --git a/include/llvm/MC/MCContext.h b/include/llvm/MC/MCContext.h
index b586319193..59545d31a6 100644
--- a/include/llvm/MC/MCContext.h
+++ b/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);
diff --git a/include/llvm/MC/MCDirectives.h b/include/llvm/MC/MCDirectives.h
index 9180d1b369..0461766c2f 100644
--- a/include/llvm/MC/MCDirectives.h
+++ b/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
diff --git a/include/llvm/MC/MCMachObjectWriter.h b/include/llvm/MC/MCMachObjectWriter.h
index 9bb598f546..949d90700e 100644
--- a/include/llvm/MC/MCMachObjectWriter.h
+++ b/include/llvm/MC/MCMachObjectWriter.h
@@ -179,6 +179,9 @@ public:
 
   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
diff --git a/include/llvm/MC/MCStreamer.h b/include/llvm/MC/MCStreamer.h
index 25956008e0..44c5fefaa1 100644
--- a/include/llvm/MC/MCStreamer.h
+++ b/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;