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

diff --git a/include/llvm/MC/MCAsmInfo.h b/include/llvm/MC/MCAsmInfo.h
index 16964c3c9d8fcf90e860a89dc2256b1913a57343..ae0dad2fd1c0578f2673e4d52b9b8bdb25dcddc8 100644 (file)
--- 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 d139173c3e13f27f373311f1d3b3add1280ffd45..4ab7f91f72a5b6325ae0029688c5eef7e44d74d6 100644 (file)
--- 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.
   //
@@ -883,6 +898,33 @@ 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 b58631919330d68b1aa06b63972753d2d3a3dca3..59545d31a6551bfecd90c77954fe142be1805e7e 100644 (file)
--- 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 9180d1b369fed0a389bd16ddb2015dbe117a03d1..0461766c2fd3c0ac71aadb37f51e1f9b0f45b914 100644 (file)
--- 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 9bb598f5468a9707dc1017d864677f7825847f95..949d90700e083e10e87990200131ae5271092b70 100644 (file)
--- 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 25956008e021f7dd6cf34327e40a618763d82c71..44c5fefaa1d9aa34128b18b7263ae4b02fb2c6a5 100644 (file)
--- 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;

diff --git a/include/llvm/Object/MachOFormat.h b/include/llvm/Object/MachOFormat.h
index 089cde92a0a368c124c635e4d4f1f29b0d0194d4..f30d431b69da61449401f5a5fbac0ec2a703c4a7 100644 (file)
--- a/include/llvm/Object/MachOFormat.h
+++ b/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
   /// @{

diff --git a/include/llvm/Object/MachOObject.h b/include/llvm/Object/MachOObject.h
index 0560402743196a8dc5a3d4ce8efb92597cdf468c..86f150a2940f1cc4211a6eb0a6f99dc50d047009 100644 (file)
--- a/include/llvm/Object/MachOObject.h
+++ b/include/llvm/Object/MachOObject.h
@@ -174,6 +174,9 @@ public:
   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;
 
   /// @}

diff --git a/lib/CodeGen/AsmPrinter/AsmPrinter.cpp b/lib/CodeGen/AsmPrinter/AsmPrinter.cpp
index b0b2ff4882afaec30a2ae4c7577fab87c1283c00..ae498dd27d356c694202ece3444e8d20ba89bce9 100644 (file)
--- a/lib/CodeGen/AsmPrinter/AsmPrinter.cpp
+++ b/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;
 

diff --git a/lib/MC/MCAsmInfo.cpp b/lib/MC/MCAsmInfo.cpp
index 2456858ded4313bfe85f4ac7e244f1d8c512bdf2..d9c1d51d185fbc451ee11da5aa2abd95df1c116f 100644 (file)
--- a/lib/MC/MCAsmInfo.cpp
+++ b/lib/MC/MCAsmInfo.cpp
@@ -50,6 +50,7 @@ MCAsmInfo::MCAsmInfo() {
   AllowNameToStartWithDigit = false;
   AllowPeriodsInName = true;
   AllowUTF8 = true;
+  UseDataRegionDirectives = false;
   ZeroDirective = "\t.zero\t";
   AsciiDirective = "\t.ascii\t";
   AscizDirective = "\t.asciz\t";
@@ -57,12 +58,6 @@ MCAsmInfo::MCAsmInfo() {
   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";

diff --git a/lib/MC/MCAsmInfoCOFF.cpp b/lib/MC/MCAsmInfoCOFF.cpp
index 89413652ef0cc814b8c3f6011435f48e6734e29d..678e75af5dab4e8c72c3a7f6c82cff164307ac62 100644 (file)
--- a/lib/MC/MCAsmInfoCOFF.cpp
+++ b/lib/MC/MCAsmInfoCOFF.cpp
@@ -36,8 +36,6 @@ MCAsmInfoCOFF::MCAsmInfoCOFF() {
   SupportsDebugInformation = true;
   DwarfSectionOffsetDirective = "\t.secrel32\t";
   HasMicrosoftFastStdCallMangling = true;
-
-  SupportsDataRegions = false;
 }
 
 void MCAsmInfoMicrosoft::anchor() { }

diff --git a/lib/MC/MCAsmInfoDarwin.cpp b/lib/MC/MCAsmInfoDarwin.cpp
index 4d675929e1ea416488071cc5305fa2583e786820..73ef7ba0604a0cfd014518363809c43fad28c8f8 100644 (file)
--- a/lib/MC/MCAsmInfoDarwin.cpp
+++ b/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;
 

diff --git a/lib/MC/MCAsmStreamer.cpp b/lib/MC/MCAsmStreamer.cpp
index 3ce84c2cd9708dc26e230b496c3ac1f27f5a0e2d..9497b6be491c66570c6a7ce9c5ce508708d35f30 100644 (file)
--- a/lib/MC/MCAsmStreamer.cpp
+++ b/lib/MC/MCAsmStreamer.cpp
@@ -138,6 +138,7 @@ public:
   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.

diff --git a/lib/MC/MCELFStreamer.cpp b/lib/MC/MCELFStreamer.cpp
index 6c4d0e33a115ecbb93a42a0ccd9a2b35e16d139d..6ac9d9d51cbc77333e6b285fb579553f1793c498 100644 (file)
--- a/lib/MC/MCELFStreamer.cpp
+++ b/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"

diff --git a/lib/MC/MCMachOStreamer.cpp b/lib/MC/MCMachOStreamer.cpp
index bc6cf773217cd0e72c228180e715e25c5e2e9fc8..970aa8baf43c094c390be5b6add172ed509d669e 100644 (file)
--- a/lib/MC/MCMachOStreamer.cpp
+++ b/lib/MC/MCMachOStreamer.cpp
@@ -1,4 +1,3 @@
-//===- lib/MC/MCMachOStreamer.cpp - Mach-O Object Output ------------===//
 //
 //                     The LLVM Compiler Infrastructure
 //
@@ -33,6 +32,8 @@ class MCMachOStreamer : public MCObjectStreamer {
 private:
   virtual void EmitInstToData(const MCInst &Inst);
 
+  void EmitDataRegion(DataRegionData::KindTy Kind);
+  void EmitDataRegionEnd();
 public:
   MCMachOStreamer(MCContext &Context, MCAsmBackend &MAB,
                   raw_ostream &OS, MCCodeEmitter *Emitter)
@@ -46,6 +47,7 @@ public:
   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);
   virtual void EmitSymbolAttribute(MCSymbol *Symbol, MCSymbolAttr Attribute);
@@ -138,6 +140,26 @@ void MCMachOStreamer::EmitLabel(MCSymbol *Symbol) {
   SD.setFlags(SD.getFlags() & ~SF_ReferenceTypeMask);
 }
 
+void MCMachOStreamer::EmitDataRegion(DataRegionData::KindTy Kind) {
+  // Create a temporary label to mark the start of the data region.
+  MCSymbol *Start = getContext().CreateTempSymbol();
+  EmitLabel(Start);
+  // Record the region for the object writer to use.
+  DataRegionData Data = { Kind, Start, NULL };
+  std::vector<DataRegionData> &Regions = getAssembler().getDataRegions();
+  Regions.push_back(Data);
+}
+
+void MCMachOStreamer::EmitDataRegionEnd() {
+  std::vector<DataRegionData> &Regions = getAssembler().getDataRegions();
+  assert(Regions.size() && "Mismatched .end_data_region!");
+  DataRegionData &Data = Regions.back();
+  assert(Data.End == NULL && "Mismatched .end_data_region!");
+  // Create a temporary label to mark the end of the data region.
+  Data.End = getContext().CreateTempSymbol();
+  EmitLabel(Data.End);
+}
+
 void MCMachOStreamer::EmitAssemblerFlag(MCAssemblerFlag Flag) {
   // Let the target do whatever target specific stuff it needs to do.
   getAssembler().getBackend().handleAssemblerFlag(Flag);
@@ -153,6 +175,26 @@ void MCMachOStreamer::EmitAssemblerFlag(MCAssemblerFlag Flag) {
   }
 }
 
+void MCMachOStreamer::EmitDataRegion(MCDataRegionType Kind) {
+  switch (Kind) {
+  case MCDR_DataRegion:
+    EmitDataRegion(DataRegionData::Data);
+    return;
+  case MCDR_DataRegionJT8:
+    EmitDataRegion(DataRegionData::JumpTable8);
+    return;
+  case MCDR_DataRegionJT16:
+    EmitDataRegion(DataRegionData::JumpTable16);
+    return;
+  case MCDR_DataRegionJT32:
+    EmitDataRegion(DataRegionData::JumpTable32);
+    return;
+  case MCDR_DataRegionEnd:
+    EmitDataRegionEnd();
+    return;
+  }
+}
+
 void MCMachOStreamer::EmitThumbFunc(MCSymbol *Symbol) {
   // Remember that the function is a thumb function. Fixup and relocation
   // values will need adjusted.

diff --git a/lib/MC/MCParser/DarwinAsmParser.cpp b/lib/MC/MCParser/DarwinAsmParser.cpp
index 6f450682cbebd46d49f13cc714b878738addfbfb..5662fea86ced2f13e99413aaaeff9fc1c84d6c47 100644 (file)
--- a/lib/MC/MCParser/DarwinAsmParser.cpp
+++ b/lib/MC/MCParser/DarwinAsmParser.cpp
@@ -14,6 +14,7 @@
 #include "llvm/MC/MCSymbol.h"
 #include "llvm/MC/MCParser/MCAsmLexer.h"
 #include "llvm/MC/MCParser/MCAsmParser.h"
+#include "llvm/ADT/StringSwitch.h"
 #include "llvm/ADT/StringRef.h"
 #include "llvm/ADT/Twine.h"
 #include "llvm/Support/MemoryBuffer.h"
@@ -56,6 +57,9 @@ public:
     AddDirectiveHandler<&DarwinAsmParser::ParseDirectiveTBSS>(".tbss");
     AddDirectiveHandler<&DarwinAsmParser::ParseDirectiveZerofill>(".zerofill");
 
+    AddDirectiveHandler<&DarwinAsmParser::ParseDirectiveDataRegion>(".data_region");
+    AddDirectiveHandler<&DarwinAsmParser::ParseDirectiveDataRegionEnd>(".end_data_region");
+
     // Special section directives.
     AddDirectiveHandler<&DarwinAsmParser::ParseSectionDirectiveConst>(".const");
     AddDirectiveHandler<&DarwinAsmParser::ParseSectionDirectiveConstData>(".const_data");
@@ -113,6 +117,8 @@ public:
   bool ParseDirectiveSubsectionsViaSymbols(StringRef, SMLoc);
   bool ParseDirectiveTBSS(StringRef, SMLoc);
   bool ParseDirectiveZerofill(StringRef, SMLoc);
+  bool ParseDirectiveDataRegion(StringRef, SMLoc);
+  bool ParseDirectiveDataRegionEnd(StringRef, SMLoc);
 
   // Named Section Directive
   bool ParseSectionDirectiveConst(StringRef, SMLoc) {
@@ -659,6 +665,42 @@ bool DarwinAsmParser::ParseDirectiveZerofill(StringRef, SMLoc) {
   return false;
 }
 
+/// ParseDirectiveDataRegion
+///  ::= .data_region [ ( jt8 | jt16 | jt32 ) ]
+bool DarwinAsmParser::ParseDirectiveDataRegion(StringRef, SMLoc) {
+  if (getLexer().is(AsmToken::EndOfStatement)) {
+    Lex();
+    getStreamer().EmitDataRegion(MCDR_DataRegion);
+    return false;
+  }
+  StringRef RegionType;
+  SMLoc Loc = getParser().getTok().getLoc();
+  if (getParser().ParseIdentifier(RegionType))
+    return TokError("expected region type after '.data_region' directive");
+  int Kind = StringSwitch<int>(RegionType)
+    .Case("jt8", MCDR_DataRegionJT8)
+    .Case("jt16", MCDR_DataRegionJT16)
+    .Case("jt32", MCDR_DataRegionJT32)
+    .Default(-1);
+  if (Kind == -1)
+    return Error(Loc, "unknown region type in '.data_region' directive");
+  Lex();
+
+  getStreamer().EmitDataRegion((MCDataRegionType)Kind);
+  return false;
+}
+
+/// ParseDirectiveDataRegionEnd
+///  ::= .end_data_region
+bool DarwinAsmParser::ParseDirectiveDataRegionEnd(StringRef, SMLoc) {
+  if (getLexer().isNot(AsmToken::EndOfStatement))
+    return TokError("unexpected token in '.end_data_region' directive");
+
+  Lex();
+  getStreamer().EmitDataRegion(MCDR_DataRegionEnd);
+  return false;
+}
+
 namespace llvm {
 
 MCAsmParserExtension *createDarwinAsmParser() {

diff --git a/lib/MC/MCStreamer.cpp b/lib/MC/MCStreamer.cpp
index 43e62ff89a0fe1eb9efd9d8b024aadfe07a56f08..e363f28f7725d7243a595e56a2713a8f609ec087 100644 (file)
--- a/lib/MC/MCStreamer.cpp
+++ b/lib/MC/MCStreamer.cpp
@@ -20,12 +20,9 @@
 #include <cstdlib>
 using namespace llvm;
 
-MCStreamer::MCStreamer(MCContext &Ctx) : Context(Ctx), EmitEHFrame(true),
-                                         EmitDebugFrame(false),
-                                         CurrentW64UnwindInfo(0),
-                                         LastSymbol(0),
-                                         UniqueCodeBeginSuffix(0),
-                                         UniqueDataBeginSuffix(0) {
+MCStreamer::MCStreamer(MCContext &Ctx)
+  : Context(Ctx), EmitEHFrame(true), EmitDebugFrame(false),
+    CurrentW64UnwindInfo(0), LastSymbol(0) {
   const MCSection *section = NULL;
   SectionStack.push_back(std::make_pair(section, section));
 }
@@ -183,85 +180,6 @@ void MCStreamer::EmitLabel(MCSymbol *Symbol) {
   LastSymbol = Symbol;
 }
 
-void MCStreamer::EmitDataRegion() {
-  if (RegionIndicator == Data) return;
-
-  MCContext &Context = getContext();
-  const MCAsmInfo &MAI = Context.getAsmInfo();
-  if (!MAI.getSupportsDataRegions()) return;
-
-  // Generate a unique symbol name.
-  MCSymbol *NewSym = Context.GetOrCreateSymbol(MAI.getDataBeginLabelName() +
-                                               Twine(UniqueDataBeginSuffix++));
-  EmitLabel(NewSym);
-
-  RegionIndicator = Data;
-}
-
-void MCStreamer::EmitCodeRegion() {
-  if (RegionIndicator == Code) return;
-
-  MCContext &Context = getContext();
-  const MCAsmInfo &MAI = Context.getAsmInfo();
-  if (!MAI.getSupportsDataRegions()) return;
-
-  // Generate a unique symbol name.
-  MCSymbol *NewSym = Context.GetOrCreateSymbol(MAI.getCodeBeginLabelName() +
-                                               Twine(UniqueCodeBeginSuffix++));
-  EmitLabel(NewSym);
-
-  RegionIndicator = Code;
-}
-
-void MCStreamer::EmitJumpTable8Region() {
-  if (RegionIndicator == JumpTable8) return;
-
-  MCContext &Context = getContext();
-  const MCAsmInfo &MAI = Context.getAsmInfo();
-  if (!MAI.getSupportsDataRegions()) return;
-
-  // Generate a unique symbol name.
-  MCSymbol *NewSym =
-    Context.GetOrCreateSymbol(MAI.getJumpTable8BeginLabelName() +
-                              Twine(UniqueDataBeginSuffix++));
-  EmitLabel(NewSym);
-
-  RegionIndicator = JumpTable8;
-}
-
-void MCStreamer::EmitJumpTable16Region() {
-  if (RegionIndicator == JumpTable16) return;
-
-  MCContext &Context = getContext();
-  const MCAsmInfo &MAI = Context.getAsmInfo();
-  if (!MAI.getSupportsDataRegions()) return;
-
-  // Generate a unique symbol name.
-  MCSymbol *NewSym =
-    Context.GetOrCreateSymbol(MAI.getJumpTable16BeginLabelName() +
-                              Twine(UniqueDataBeginSuffix++));
-  EmitLabel(NewSym);
-
-  RegionIndicator = JumpTable16;
-}
-
-
-void MCStreamer::EmitJumpTable32Region() {
-  if (RegionIndicator == JumpTable32) return;
-
-  MCContext &Context = getContext();
-  const MCAsmInfo &MAI = Context.getAsmInfo();
-  if (!MAI.getSupportsDataRegions()) return;
-
-  // Generate a unique symbol name.
-  MCSymbol *NewSym =
-    Context.GetOrCreateSymbol(MAI.getJumpTable32BeginLabelName() +
-                              Twine(UniqueDataBeginSuffix++));
-  EmitLabel(NewSym);
-
-  RegionIndicator = JumpTable32;
-}
-
 void MCStreamer::EmitCompactUnwindEncoding(uint32_t CompactUnwindEncoding) {
   EnsureValidFrame();
   MCDwarfFrameInfo *CurFrame = getCurrentFrameInfo();
@@ -283,7 +201,6 @@ void MCStreamer::EmitCFIStartProc() {
   EmitCFIStartProcImpl(Frame);
 
   FrameInfos.push_back(Frame);
-  RegionIndicator = Code;
 }
 
 void MCStreamer::EmitCFIStartProcImpl(MCDwarfFrameInfo &Frame) {

diff --git a/lib/MC/MachObjectWriter.cpp b/lib/MC/MachObjectWriter.cpp
index 8e4066c894ba68f7621f46e6cc853dd87658c015..5820a224c527b57f060d3212fc36512de3649079 100644 (file)
--- a/lib/MC/MachObjectWriter.cpp
+++ b/lib/MC/MachObjectWriter.cpp
@@ -21,6 +21,7 @@
 #include "llvm/MC/MCMachOSymbolFlags.h"
 #include "llvm/MC/MCValue.h"
 #include "llvm/Object/MachOFormat.h"
+#include "llvm/Support/Debug.h"
 #include "llvm/Support/ErrorHandling.h"
 
 #include <vector>
@@ -351,6 +352,21 @@ void MachObjectWriter::WriteNlist(MachSymbolData &MSD,
     Write32(Address);
 }
 
+void MachObjectWriter::WriteLinkeditLoadCommand(uint32_t Type,
+                                                uint32_t DataOffset,
+                                                uint32_t DataSize) {
+  uint64_t Start = OS.tell();
+  (void) Start;
+
+  Write32(Type);
+  Write32(macho::LinkeditLoadCommandSize);
+  Write32(DataOffset);
+  Write32(DataSize);
+
+  assert(OS.tell() - Start == macho::LinkeditLoadCommandSize);
+}
+
+
 void MachObjectWriter::RecordRelocation(const MCAssembler &Asm,
                                         const MCAsmLayout &Layout,
                                         const MCFragment *Fragment,
@@ -654,6 +670,13 @@ void MachObjectWriter::WriteObject(MCAssembler &Asm,
                          macho::DysymtabLoadCommandSize);
   }
 
+  // Add the data-in-code load command size, if used.
+  unsigned NumDataRegions = Asm.getDataRegions().size();
+  if (NumDataRegions) {
+    ++NumLoadCommands;
+    LoadCommandsSize += macho::LinkeditLoadCommandSize;
+  }
+
   // Compute the total size of the section data, as well as its file size and vm
   // size.
   uint64_t SectionDataStart = (is64Bit() ? macho::Header64Size :
@@ -701,6 +724,15 @@ void MachObjectWriter::WriteObject(MCAssembler &Asm,
     RelocTableEnd += NumRelocs * macho::RelocationInfoSize;
   }
 
+  // Write the data-in-code load command, if used.
+  uint64_t DataInCodeTableEnd = RelocTableEnd + NumDataRegions * 8;
+  if (NumDataRegions) {
+    uint64_t DataRegionsOffset = RelocTableEnd;
+    uint64_t DataRegionsSize = NumDataRegions * 8;
+    WriteLinkeditLoadCommand(macho::LCT_DataInCode, DataRegionsOffset,
+                             DataRegionsSize);
+  }
+
   // Write the symbol table load command, if used.
   if (NumSymbols) {
     unsigned FirstLocalSymbol = 0;
@@ -717,10 +749,10 @@ void MachObjectWriter::WriteObject(MCAssembler &Asm,
 
     // If used, the indirect symbols are written after the section data.
     if (NumIndirectSymbols)
-      IndirectSymbolOffset = RelocTableEnd;
+      IndirectSymbolOffset = DataInCodeTableEnd;
 
     // The symbol table is written after the indirect symbol data.
-    uint64_t SymbolTableOffset = RelocTableEnd + IndirectSymbolSize;
+    uint64_t SymbolTableOffset = DataInCodeTableEnd + IndirectSymbolSize;
 
     // The string table is written after symbol table.
     uint64_t StringTableOffset =
@@ -760,6 +792,23 @@ void MachObjectWriter::WriteObject(MCAssembler &Asm,
     }
   }
 
+  // Write out the data-in-code region payload, if there is one.
+  for (MCAssembler::const_data_region_iterator
+         it = Asm.data_region_begin(), ie = Asm.data_region_end();
+         it != ie; ++it) {
+    const DataRegionData *Data = &(*it);
+    uint64_t Start = getSymbolAddress(&Layout.getAssembler().getSymbolData(*Data->Start), Layout);
+    uint64_t End = getSymbolAddress(&Layout.getAssembler().getSymbolData(*Data->End), Layout);
+    DEBUG(dbgs() << "data in code region-- kind: " << Data->Kind
+                 << "  start: " << Start << "(" << Data->Start->getName() << ")"
+                 << "  end: " << End << "(" << Data->End->getName() << ")"
+                 << "  size: " << End - Start
+                 << "\n");
+    Write32(Start);
+    Write16(End - Start);
+    Write16(Data->Kind);
+  }
+
   // Write the symbol table data, if used.
   if (NumSymbols) {
     // Write the indirect symbol entries.

diff --git a/lib/Object/MachOObject.cpp b/lib/Object/MachOObject.cpp
index b7e5cdcd6b846c33d1c18bd5d6715383e18e9286..00dea3fe476973ca642795b63d77c177d6236dc8 100644 (file)
--- a/lib/Object/MachOObject.cpp
+++ b/lib/Object/MachOObject.cpp
@@ -357,6 +357,19 @@ void MachOObject::ReadSymbol64TableEntry(uint64_t SymbolTableOffset,
   ReadInMemoryStruct(*this, Buffer->getBuffer(), Offset, Res);
 }
 
+template<>
+void SwapStruct(macho::DataInCodeTableEntry &Value) {
+  SwapValue(Value.Offset);
+  SwapValue(Value.Length);
+  SwapValue(Value.Kind);
+}
+void MachOObject::ReadDataInCodeTableEntry(uint64_t TableOffset,
+                                           unsigned Index,
+                       InMemoryStruct<macho::DataInCodeTableEntry> &Res) const {
+  uint64_t Offset = (TableOffset +
+                     Index * sizeof(macho::DataInCodeTableEntry));
+  ReadInMemoryStruct(*this, Buffer->getBuffer(), Offset, Res);
+}
 
 void MachOObject::ReadULEB128s(uint64_t Index,
                                SmallVectorImpl<uint64_t> &Out) const {

diff --git a/lib/Target/ARM/ARMAsmPrinter.cpp b/lib/Target/ARM/ARMAsmPrinter.cpp
index 410790a7baa03872ea8172364475acd96f08a1f9..e781498f1dfd31bb96b3959a78b1aca69ea9974d 100644 (file)
--- a/lib/Target/ARM/ARMAsmPrinter.cpp
+++ b/lib/Target/ARM/ARMAsmPrinter.cpp
@@ -283,9 +283,16 @@ void ARMAsmPrinter::EmitDwarfRegOp(const MachineLocation &MLoc) const {
   }
 }
 
-void ARMAsmPrinter::EmitFunctionEntryLabel() {
-  OutStreamer.ForceCodeRegion();
+void ARMAsmPrinter::EmitFunctionBodyEnd() {
+  // Make sure to terminate any constant pools that were at the end
+  // of the function.
+  if (!InConstantPool)
+    return;
+  InConstantPool = false;
+  OutStreamer.EmitDataRegion(MCDR_DataRegionEnd);
+}
 
+void ARMAsmPrinter::EmitFunctionEntryLabel() {
   if (AFI->isThumbFunction()) {
     OutStreamer.EmitAssemblerFlag(MCAF_Code16);
     OutStreamer.EmitThumbFunc(CurrentFnSym);
@@ -934,13 +941,13 @@ void ARMAsmPrinter::EmitJumpTable(const MachineInstr *MI) {
   const MachineOperand &MO2 = MI->getOperand(OpNum+1); // Unique Id
   unsigned JTI = MO1.getIndex();
 
-  // Tag the jump table appropriately for precise disassembly.
-  OutStreamer.EmitJumpTable32Region();
-
   // Emit a label for the jump table.
   MCSymbol *JTISymbol = GetARMJTIPICJumpTableLabel2(JTI, MO2.getImm());
   OutStreamer.EmitLabel(JTISymbol);
 
+  // Mark the jump table as data-in-code.
+  OutStreamer.EmitDataRegion(MCDR_DataRegionJT32);
+
   // Emit each entry of the table.
   const MachineJumpTableInfo *MJTI = MF->getJumpTableInfo();
   const std::vector<MachineJumpTableEntry> &JT = MJTI->getJumpTables();
@@ -969,6 +976,8 @@ void ARMAsmPrinter::EmitJumpTable(const MachineInstr *MI) {
                                      OutContext);
     OutStreamer.EmitValue(Expr, 4);
   }
+  // Mark the end of jump table data-in-code region.
+  OutStreamer.EmitDataRegion(MCDR_DataRegionEnd);
 }
 
 void ARMAsmPrinter::EmitJump2Table(const MachineInstr *MI) {
@@ -978,15 +987,6 @@ void ARMAsmPrinter::EmitJump2Table(const MachineInstr *MI) {
   const MachineOperand &MO2 = MI->getOperand(OpNum+1); // Unique Id
   unsigned JTI = MO1.getIndex();
 
-  // Emit a label for the jump table.
-  if (MI->getOpcode() == ARM::t2TBB_JT) {
-    OutStreamer.EmitJumpTable8Region();
-  } else if (MI->getOpcode() == ARM::t2TBH_JT) {
-    OutStreamer.EmitJumpTable16Region();
-  } else {
-    OutStreamer.EmitJumpTable32Region();
-  }
-
   MCSymbol *JTISymbol = GetARMJTIPICJumpTableLabel2(JTI, MO2.getImm());
   OutStreamer.EmitLabel(JTISymbol);
 
@@ -995,10 +995,15 @@ void ARMAsmPrinter::EmitJump2Table(const MachineInstr *MI) {
   const std::vector<MachineJumpTableEntry> &JT = MJTI->getJumpTables();
   const std::vector<MachineBasicBlock*> &JTBBs = JT[JTI].MBBs;
   unsigned OffsetWidth = 4;
-  if (MI->getOpcode() == ARM::t2TBB_JT)
+  if (MI->getOpcode() == ARM::t2TBB_JT) {
     OffsetWidth = 1;
-  else if (MI->getOpcode() == ARM::t2TBH_JT)
+    // Mark the jump table as data-in-code.
+    OutStreamer.EmitDataRegion(MCDR_DataRegionJT8);
+  } else if (MI->getOpcode() == ARM::t2TBH_JT) {
     OffsetWidth = 2;
+    // Mark the jump table as data-in-code.
+    OutStreamer.EmitDataRegion(MCDR_DataRegionJT16);
+  }
 
   for (unsigned i = 0, e = JTBBs.size(); i != e; ++i) {
     MachineBasicBlock *MBB = JTBBs[i];
@@ -1031,6 +1036,8 @@ void ARMAsmPrinter::EmitJump2Table(const MachineInstr *MI) {
                                    OutContext);
     OutStreamer.EmitValue(Expr, OffsetWidth);
   }
+  // Mark the end of jump table data-in-code region.
+  OutStreamer.EmitDataRegion(MCDR_DataRegionEnd);
 }
 
 void ARMAsmPrinter::PrintDebugValueComment(const MachineInstr *MI,
@@ -1208,8 +1215,11 @@ extern cl::opt<bool> EnableARMEHABI;
 #include "ARMGenMCPseudoLowering.inc"
 
 void ARMAsmPrinter::EmitInstruction(const MachineInstr *MI) {
-  if (MI->getOpcode() != ARM::CONSTPOOL_ENTRY)
-    OutStreamer.EmitCodeRegion();
+  // If we just ended a constant pool, mark it as such.
+  if (InConstantPool && MI->getOpcode() != ARM::CONSTPOOL_ENTRY) {
+    OutStreamer.EmitDataRegion(MCDR_DataRegionEnd);
+    InConstantPool = false;
+  }
 
   // Emit unwinding stuff for frame-related instructions
   if (EnableARMEHABI && MI->getFlag(MachineInstr::FrameSetup))
@@ -1565,9 +1575,12 @@ void ARMAsmPrinter::EmitInstruction(const MachineInstr *MI) {
     unsigned LabelId = (unsigned)MI->getOperand(0).getImm();
     unsigned CPIdx   = (unsigned)MI->getOperand(1).getIndex();
 
-    // Mark the constant pool entry as data if we're not already in a data
-    // region.
-    OutStreamer.EmitDataRegion();
+    // If this is the first entry of the pool, mark it.
+    if (!InConstantPool) {
+      OutStreamer.EmitDataRegion(MCDR_DataRegion);
+      InConstantPool = true;
+    }
+
     OutStreamer.EmitLabel(GetCPISymbol(LabelId));
 
     const MachineConstantPoolEntry &MCPE = MCP->getConstants()[CPIdx];

diff --git a/lib/Target/ARM/ARMAsmPrinter.h b/lib/Target/ARM/ARMAsmPrinter.h
index af3f75a0e8925c9d798ffcbd0666f91bfbee4c2e..3555e8f50a0d35a72a77cd3526f0437ebbf975df 100644 (file)
--- a/lib/Target/ARM/ARMAsmPrinter.h
+++ b/lib/Target/ARM/ARMAsmPrinter.h
@@ -44,9 +44,12 @@ class LLVM_LIBRARY_VISIBILITY ARMAsmPrinter : public AsmPrinter {
   /// MachineFunction.
   const MachineConstantPool *MCP;
 
+  /// InConstantPool - Maintain state when emitting a sequence of constant
+  /// pool entries so we can properly mark them as data regions.
+  bool InConstantPool;
 public:
   explicit ARMAsmPrinter(TargetMachine &TM, MCStreamer &Streamer)
-    : AsmPrinter(TM, Streamer), AFI(NULL), MCP(NULL) {
+    : AsmPrinter(TM, Streamer), AFI(NULL), MCP(NULL), InConstantPool(false) {
       Subtarget = &TM.getSubtarget<ARMSubtarget>();
     }
 
@@ -70,6 +73,7 @@ public:
   bool runOnMachineFunction(MachineFunction &F);
 
   virtual void EmitConstantPool() {} // we emit constant pools customly!
+  virtual void EmitFunctionBodyEnd();
   virtual void EmitFunctionEntryLabel();
   void EmitStartOfAsmFile(Module &M);
   void EmitEndOfAsmFile(Module &M);

diff --git a/lib/Target/ARM/MCTargetDesc/ARMMCAsmInfo.cpp b/lib/Target/ARM/MCTargetDesc/ARMMCAsmInfo.cpp
index 4ac7741ba5b730026e9513630ebd91233c873b7c..d32805e522a378f96396524cb25e4b0af4ccd5ac 100644 (file)
--- a/lib/Target/ARM/MCTargetDesc/ARMMCAsmInfo.cpp
+++ b/lib/Target/ARM/MCTargetDesc/ARMMCAsmInfo.cpp
@@ -29,6 +29,7 @@ ARMMCAsmInfoDarwin::ARMMCAsmInfoDarwin() {
   CommentString = "@";
   Code16Directive = ".code\t16";
   Code32Directive = ".code\t32";
+  UseDataRegionDirectives = true;
 
   SupportsDebugInformation = true;
 

diff --git a/lib/Target/X86/X86MCInstLower.cpp b/lib/Target/X86/X86MCInstLower.cpp
index 54bfce1298cc93635a89147ef8d8fa9edb2a8349..2f368b6a5ac3eb8f5e6a7f402c51282b8e7261fa 100644 (file)
--- a/lib/Target/X86/X86MCInstLower.cpp
+++ b/lib/Target/X86/X86MCInstLower.cpp
@@ -610,8 +610,6 @@ static void LowerTlsAddr(MCStreamer &OutStreamer,
 }
 
 void X86AsmPrinter::EmitInstruction(const MachineInstr *MI) {
-  OutStreamer.EmitCodeRegion();
-
   X86MCInstLower MCInstLowering(Mang, *MF, *this);
   switch (MI->getOpcode()) {
   case TargetOpcode::DBG_VALUE:

diff --git a/test/CodeGen/ARM/data-in-code-annotations.ll b/test/CodeGen/ARM/data-in-code-annotations.ll
new file mode 100644 (file)
index 0000000..a66a9d1
--- /dev/null
+++ b/test/CodeGen/ARM/data-in-code-annotations.ll
@@ -0,0 +1,42 @@
+; RUN: llc < %s -mtriple=armv7-apple-darwin | FileCheck %s
+
+define double @f1() nounwind {
+; CHECK: f1:
+; CHECK: .data_region
+; CHECK: .long 1413754129
+; CHECK: .long 1074340347
+; CHECK: .end_data_region
+  ret double 0x400921FB54442D11
+}
+
+
+define i32 @f2()  {
+; CHECK: f2:
+; CHECK: .data_region jt32
+; CHECK: .end_data_region
+
+entry:
+  switch i32 undef, label %return [
+    i32 1, label %sw.bb
+    i32 2, label %sw.bb6
+    i32 3, label %sw.bb13
+    i32 4, label %sw.bb20
+  ]
+
+sw.bb:                                            ; preds = %entry
+  br label %return
+
+sw.bb6:                                           ; preds = %entry
+  br label %return
+
+sw.bb13:                                          ; preds = %entry
+  br label %return
+
+sw.bb20:                                          ; preds = %entry
+  %div = sdiv i32 undef, undef
+  br label %return
+
+return:                                           ; preds = %sw.bb20, %sw.bb13, %sw.bb6, %sw.bb, %entry
+  %retval.0 = phi i32 [ %div, %sw.bb20 ], [ undef, %sw.bb13 ], [ undef, %sw.bb6 ], [ undef, %sw.bb ], [ 0, %entry ]
+  ret i32 %retval.0
+}

diff --git a/test/CodeGen/Thumb2/thumb2-tbb.ll b/test/CodeGen/Thumb2/thumb2-tbb.ll
index 5dc3cc3ce70a9bd5860131e159be43d85daef617..a9d71d6bda15fe2353b7b9e29d6b700d331f71ec 100644 (file)
--- a/test/CodeGen/Thumb2/thumb2-tbb.ll
+++ b/test/CodeGen/Thumb2/thumb2-tbb.ll
@@ -5,7 +5,9 @@ define void @bar(i32 %n.u) {
 entry:
 ; CHECK: bar:
 ; CHECK: tbb
-; CHECK: .align 1
+; CHECK: .data_region jt8
+; CHECK: .end_data_region
+; CHECK-NEXT: .align 1
 
     switch i32 %n.u, label %bb12 [i32 1, label %bb i32 2, label %bb6 i32 4, label %bb7 i32 5, label %bb8 i32 6, label %bb10 i32 7, label %bb1 i32 8, label %bb3 i32 9, label %bb4 i32 10, label %bb9 i32 11, label %bb2 i32 12, label %bb5 i32 13, label %bb11 ]
 bb:

diff --git a/test/MC/ARM/elf-reloc-01.ll b/test/MC/ARM/elf-reloc-01.ll
index 6899d92b50f8c8744f49c234a4729ca0eae92c82..c98026b6a0437e712f91d5f51f33873dd01890ec 100644 (file)
--- a/test/MC/ARM/elf-reloc-01.ll
+++ b/test/MC/ARM/elf-reloc-01.ll
@@ -61,7 +61,7 @@ bb3:                                              ; preds = %bb, %entry
 declare void @exit(i32) noreturn nounwind
 
 ;; OBJ:          Relocation 1
-;; OBJ-NEXT:     'r_offset', 
+;; OBJ-NEXT:     'r_offset',
 ;; OBJ-NEXT:     'r_sym', 0x000002
 ;; OBJ-NEXT:     'r_type', 0x2b
 

diff --git a/test/MC/MachO/ARM/data-in-code.s b/test/MC/MachO/ARM/data-in-code.s
new file mode 100644 (file)
index 0000000..bbcb9aa
--- /dev/null
+++ b/test/MC/MachO/ARM/data-in-code.s
@@ -0,0 +1,33 @@
+@ RUN: llvm-mc -triple armv7-apple-darwin10 -filetype=obj -o - < %s | macho-dump | FileCheck %s
+        .text
+_foo:
+@ CHECK: # DICE 0
+@ CHECK: ('offset', 0)
+@ CHECK: ('length', 4)
+@ CHECK: ('kind', 1)
+@ CHECK: # DICE 1
+@ CHECK: ('offset', 4)
+@ CHECK: ('length', 4)
+@ CHECK: ('kind', 4)
+@ CHECK: # DICE 2
+@ CHECK: ('offset', 8)
+@ CHECK: ('length', 2)
+@ CHECK: ('kind', 3)
+@ CHECK: # DICE 3
+@ CHECK: ('offset', 10)
+@ CHECK: ('length', 1)
+@ CHECK: ('kind', 2)
+
+.data_region
+        .long 10
+.end_data_region
+.data_region jt32
+        .long 1
+.end_data_region
+.data_region jt16
+        .short 2
+.end_data_region
+.data_region jt8
+        .byte 3
+.end_data_region
+

diff --git a/tools/macho-dump/macho-dump.cpp b/tools/macho-dump/macho-dump.cpp
index f65c79595a95350c72cc502599e9dac390be0e4e..20deda9a0cbd8f1d0faf5b873554df8722de6770 100644 (file)
--- a/tools/macho-dump/macho-dump.cpp
+++ b/tools/macho-dump/macho-dump.cpp
@@ -332,6 +332,35 @@ static int DumpLinkeditDataCommand(MachOObject &Obj,
   return 0;
 }
 
+static int DumpDataInCodeDataCommand(MachOObject &Obj,
+                                     const MachOObject::LoadCommandInfo &LCI) {
+  InMemoryStruct<macho::LinkeditDataLoadCommand> LLC;
+  Obj.ReadLinkeditDataLoadCommand(LCI, LLC);
+  if (!LLC)
+    return Error("unable to read segment load command");
+
+  outs() << "  ('dataoff', " << LLC->DataOffset << ")\n"
+         << "  ('datasize', " << LLC->DataSize << ")\n"
+         << "  ('_data_regions', [\n";
+
+
+  unsigned NumRegions = LLC->DataSize / 8;
+  for (unsigned i = 0; i < NumRegions; ++i) {
+    InMemoryStruct<macho::DataInCodeTableEntry> DICE;
+    Obj.ReadDataInCodeTableEntry(LLC->DataOffset, i, DICE);
+    if (!DICE)
+      return Error("unable to read DataInCodeTableEntry");
+    outs() << "    # DICE " << i << "\n"
+           << "    ('offset', " << DICE->Offset << ")\n"
+           << "    ('length', " << DICE->Length << ")\n"
+           << "    ('kind', " << DICE->Kind << ")\n";
+  }
+
+  outs() <<"  ])\n";
+
+  return 0;
+}
+
 
 static int DumpLoadCommand(MachOObject &Obj, unsigned Index) {
   const MachOObject::LoadCommandInfo &LCI = Obj.getLoadCommandInfo(Index);
@@ -358,6 +387,9 @@ static int DumpLoadCommand(MachOObject &Obj, unsigned Index) {
   case macho::LCT_FunctionStarts:
     Res = DumpLinkeditDataCommand(Obj, LCI);
     break;
+  case macho::LCT_DataInCode:
+    Res = DumpDataInCodeDataCommand(Obj, LCI);
+    break;
   default:
     Warning("unknown load command: " + Twine(LCI.Command.Type));
     break;