Add MCSymbolizer for symbolic/annotated disassembly.

This is a basic first step towards symbolization of disassembled
instructions. This used to be done using externally provided (C API)
callbacks. This patch introduces:
- the MCSymbolizer class, that mimics the same functions that were used
  in the X86 and ARM disassemblers to symbolize immediate operands and
  to annotate loads based off PC (for things like c string literals).
- the MCExternalSymbolizer class, which implements the old C API.
- the MCRelocationInfo class, which provides a way for targets to
  translate relocations (either object::RelocationRef, or disassembler
  C API VariantKinds) to MCExprs.
- the MCObjectSymbolizer class, which does symbolization using what it
  finds in an object::ObjectFile. This makes simple symbolization (with
  no fancy relocation stuff) work for all object formats!
- x86-64 Mach-O and ELF MCRelocationInfos.
- A basic ARM Mach-O MCRelocationInfo, that provides just enough to
  support the C API VariantKinds.

Most of what works in otool (the only user of the old symbolization API
that I know of) for x86-64 symbolic disassembly (-tvV) works, namely:
- symbol references: call _foo; jmp 15 <_foo+50>
- relocations:       call _foo-_bar; call _foo-4
- __cf?string:       leaq 193(%rip), %rax ## literal pool for "hello"
Stub support is the main missing part (because libObject doesn't know,
among other things, about mach-o indirect symbols).

As for the MCSymbolizer API, instead of relying on the disassemblers
to call the tryAdding* methods, maybe this could be done automagically
using InstrInfo? For instance, even though PC-relative LEAs are used
to get the address of string literals in a typical Mach-O file, a MOV
would be used in an ELF file. And right now, the explicit symbolization
only recognizes PC-relative LEAs. InstrInfo should have already have
most of what is needed to know what to symbolize, so this can
definitely be improved.

I'd also like to remove object::RelocationRef::getValueString (it seems
only used by relocation printing in objdump), as simply printing the
created MCExpr is definitely enough (and cleaner than string concats).



git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@182625 91177308-0d34-0410-b5e6-96231b3b80d8

7 files changed, 430 insertions, 1 deletions

diff --git a/lib/MC/CMakeLists.txt b/lib/MC/CMakeLists.txt
index db882c0..5377c5c 100644
--- a/lib/MC/CMakeLists.txt
+++ b/lib/MC/CMakeLists.txt
@@ -16,6 +16,7 @@ add_llvm_library(LLVMMC
   MCELFObjectTargetWriter.cpp
   MCELFStreamer.cpp
   MCExpr.cpp
+  MCExternalSymbolizer.cpp
   MCInst.cpp
   MCInstPrinter.cpp
   MCInstrAnalysis.cpp
@@ -26,9 +27,11 @@ add_llvm_library(LLVMMC
   MCNullStreamer.cpp
   MCObjectFileInfo.cpp
   MCObjectStreamer.cpp
+  MCObjectSymbolizer.cpp
   MCObjectWriter.cpp
   MCPureStreamer.cpp
   MCRegisterInfo.cpp
+  MCRelocationInfo.cpp
   MCSection.cpp
   MCSectionCOFF.cpp
   MCSectionELF.cpp
@@ -36,6 +39,7 @@ add_llvm_library(LLVMMC
   MCStreamer.cpp
   MCSubtargetInfo.cpp
   MCSymbol.cpp
+  MCSymbolizer.cpp
   MCValue.cpp
   MCWin64EH.cpp
   MachObjectWriter.cpp
diff --git a/lib/MC/MCDisassembler.cpp b/lib/MC/MCDisassembler.cpp
index 0809690..c4cdbba 100644
--- a/lib/MC/MCDisassembler.cpp
+++ b/lib/MC/MCDisassembler.cpp
@@ -8,7 +8,44 @@
 //===----------------------------------------------------------------------===//
 
 #include "llvm/MC/MCDisassembler.h"
+#include "llvm/MC/MCExternalSymbolizer.h"
+#include "llvm/Support/raw_ostream.h"
+
 using namespace llvm;
 
 MCDisassembler::~MCDisassembler() {
 }
+
+void
+MCDisassembler::setupForSymbolicDisassembly(
+    LLVMOpInfoCallback GetOpInfo,
+    LLVMSymbolLookupCallback SymbolLookUp,
+    void *DisInfo,
+    MCContext *Ctx,
+    OwningPtr<MCRelocationInfo> &RelInfo) {
+  assert(Ctx != 0 && "No MCContext given for symbolic disassembly");
+  Symbolizer.reset(new MCExternalSymbolizer(*Ctx, RelInfo, GetOpInfo,
+                                            SymbolLookUp, DisInfo));
+}
+
+bool MCDisassembler::tryAddingSymbolicOperand(MCInst &Inst, int64_t Value,
+                                              uint64_t Address, bool IsBranch,
+                                              uint64_t Offset,
+                                              uint64_t InstSize) const {
+  raw_ostream &cStream = CommentStream ? *CommentStream : nulls();
+  if (Symbolizer)
+    return Symbolizer->tryAddingSymbolicOperand(Inst, cStream, Value, Address,
+                                                IsBranch, Offset, InstSize);
+  return false;
+}
+
+void MCDisassembler::tryAddingPcLoadReferenceComment(int64_t Value,
+                                                     uint64_t Address) const {
+  raw_ostream &cStream = CommentStream ? *CommentStream : nulls();
+  if (Symbolizer)
+    Symbolizer->tryAddingPcLoadReferenceComment(cStream, Value, Address);
+}
+
+void MCDisassembler::setSymbolizer(OwningPtr<MCSymbolizer> &Symzer) {
+  Symbolizer.reset(Symzer.take());
+}
diff --git a/lib/MC/MCDisassembler/Disassembler.cpp b/lib/MC/MCDisassembler/Disassembler.cpp
index c80a167..ba76936 100644
--- a/lib/MC/MCDisassembler/Disassembler.cpp
+++ b/lib/MC/MCDisassembler/Disassembler.cpp
@@ -16,6 +16,7 @@
 #include "llvm/MC/MCInstPrinter.h"
 #include "llvm/MC/MCInstrInfo.h"
 #include "llvm/MC/MCRegisterInfo.h"
+#include "llvm/MC/MCRelocationInfo.h"
 #include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/Support/ErrorHandling.h"
 #include "llvm/Support/MemoryObject.h"
@@ -73,7 +74,14 @@ LLVMDisasmContextRef LLVMCreateDisasmCPU(const char *Triple, const char *CPU,
   MCDisassembler *DisAsm = TheTarget->createMCDisassembler(*STI);
   if (!DisAsm)
     return 0;
-  DisAsm->setupForSymbolicDisassembly(GetOpInfo, SymbolLookUp, DisInfo, Ctx);
+
+  OwningPtr<MCRelocationInfo> RelInfo(
+    TheTarget->createMCRelocationInfo(Triple, *Ctx));
+  if (!RelInfo)
+    return 0;
+
+  DisAsm->setupForSymbolicDisassembly(GetOpInfo, SymbolLookUp, DisInfo,
+                                      Ctx, RelInfo);
 
   // Set up the instruction printer.
   int AsmPrinterVariant = MAI->getAssemblerDialect();
diff --git a/lib/MC/MCExternalSymbolizer.cpp b/lib/MC/MCExternalSymbolizer.cpp
new file mode 100644
index 0000000..5fb52b3
--- /dev/null
+++ b/lib/MC/MCExternalSymbolizer.cpp
@@ -0,0 +1,146 @@
+//===-- lib/MC/MCExternalSymbolizer.cpp - External symbolizer ---*- C++ -*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/MC/MCExternalSymbolizer.h"
+#include "llvm/MC/MCContext.h"
+#include "llvm/MC/MCExpr.h"
+#include "llvm/MC/MCInst.h"
+#include "llvm/Support/raw_ostream.h"
+#include <cstring>
+
+using namespace llvm;
+
+// This function tries to add a symbolic operand in place of the immediate
+// Value in the MCInst. The immediate Value has had any PC adjustment made by
+// the caller. If the instruction is a branch instruction then IsBranch is true,
+// else false. If the getOpInfo() function was set as part of the
+// setupForSymbolicDisassembly() call then that function is called to get any
+// symbolic information at the Address for this instruction. If that returns
+// non-zero then the symbolic information it returns is used to create an MCExpr
+// and that is added as an operand to the MCInst. If getOpInfo() returns zero
+// and IsBranch is true then a symbol look up for Value is done and if a symbol
+// is found an MCExpr is created with that, else an MCExpr with Value is
+// created. This function returns true if it adds an operand to the MCInst and
+// false otherwise.
+bool MCExternalSymbolizer::tryAddingSymbolicOperand(MCInst &MI,
+                                                    raw_ostream &cStream,
+                                                    int64_t Value,
+                                                    uint64_t Address,
+                                                    bool IsBranch,
+                                                    uint64_t Offset,
+                                                    uint64_t InstSize) {
+  struct LLVMOpInfo1 SymbolicOp;
+  std::memset(&SymbolicOp, '\0', sizeof(struct LLVMOpInfo1));
+  SymbolicOp.Value = Value;
+
+  if (!GetOpInfo ||
+      !GetOpInfo(DisInfo, Address, Offset, InstSize, 1, &SymbolicOp)) {
+    // Clear SymbolicOp.Value from above and also all other fields.
+    std::memset(&SymbolicOp, '\0', sizeof(struct LLVMOpInfo1));
+    if (!SymbolLookUp)
+      return false;
+    uint64_t ReferenceType;
+    if (IsBranch)
+       ReferenceType = LLVMDisassembler_ReferenceType_In_Branch;
+    else
+       ReferenceType = LLVMDisassembler_ReferenceType_InOut_None;
+    const char *ReferenceName;
+    const char *Name = SymbolLookUp(DisInfo, Value, &ReferenceType, Address,
+                                    &ReferenceName);
+    if (Name) {
+      SymbolicOp.AddSymbol.Name = Name;
+      SymbolicOp.AddSymbol.Present = true;
+    }
+    // For branches always create an MCExpr so it gets printed as hex address.
+    else if (IsBranch) {
+      SymbolicOp.Value = Value;
+    }
+    if(ReferenceType == LLVMDisassembler_ReferenceType_Out_SymbolStub)
+      cStream << "symbol stub for: " << ReferenceName;
+    if (!Name && !IsBranch)
+      return false;
+  }
+
+  const MCExpr *Add = NULL;
+  if (SymbolicOp.AddSymbol.Present) {
+    if (SymbolicOp.AddSymbol.Name) {
+      StringRef Name(SymbolicOp.AddSymbol.Name);
+      MCSymbol *Sym = Ctx.GetOrCreateSymbol(Name);
+      Add = MCSymbolRefExpr::Create(Sym, Ctx);
+    } else {
+      Add = MCConstantExpr::Create((int)SymbolicOp.AddSymbol.Value, Ctx);
+    }
+  }
+
+  const MCExpr *Sub = NULL;
+  if (SymbolicOp.SubtractSymbol.Present) {
+      if (SymbolicOp.SubtractSymbol.Name) {
+      StringRef Name(SymbolicOp.SubtractSymbol.Name);
+      MCSymbol *Sym = Ctx.GetOrCreateSymbol(Name);
+      Sub = MCSymbolRefExpr::Create(Sym, Ctx);
+    } else {
+      Sub = MCConstantExpr::Create((int)SymbolicOp.SubtractSymbol.Value, Ctx);
+    }
+  }
+
+  const MCExpr *Off = NULL;
+  if (SymbolicOp.Value != 0)
+    Off = MCConstantExpr::Create(SymbolicOp.Value, Ctx);
+
+  const MCExpr *Expr;
+  if (Sub) {
+    const MCExpr *LHS;
+    if (Add)
+      LHS = MCBinaryExpr::CreateSub(Add, Sub, Ctx);
+    else
+      LHS = MCUnaryExpr::CreateMinus(Sub, Ctx);
+    if (Off != 0)
+      Expr = MCBinaryExpr::CreateAdd(LHS, Off, Ctx);
+    else
+      Expr = LHS;
+  } else if (Add) {
+    if (Off != 0)
+      Expr = MCBinaryExpr::CreateAdd(Add, Off, Ctx);
+    else
+      Expr = Add;
+  } else {
+    if (Off != 0)
+      Expr = Off;
+    else
+      Expr = MCConstantExpr::Create(0, Ctx);
+  }
+
+  Expr = RelInfo->createExprForCAPIVariantKind(Expr, SymbolicOp.VariantKind);
+  if (!Expr)
+    return false;
+
+  MI.addOperand(MCOperand::CreateExpr(Expr));
+  return true;
+}
+
+// This function tries to add a comment as to what is being referenced by a load
+// instruction with the base register that is the Pc.  These can often be values
+// in a literal pool near the Address of the instruction. The Address of the
+// instruction and its immediate Value are used as a possible literal pool entry.
+// The SymbolLookUp call back will return the name of a symbol referenced by the
+// literal pool's entry if the referenced address is that of a symbol. Or it
+// will return a pointer to a literal 'C' string if the referenced address of
+// the literal pool's entry is an address into a section with C string literals.
+void MCExternalSymbolizer::tryAddingPcLoadReferenceComment(raw_ostream &cStream,
+                                                           int64_t Value,
+                                                           uint64_t Address) {
+  if (SymbolLookUp) {
+    uint64_t ReferenceType = LLVMDisassembler_ReferenceType_In_PCrel_Load;
+    const char *ReferenceName;
+    (void)SymbolLookUp(DisInfo, Value, &ReferenceType, Address, &ReferenceName);
+    if(ReferenceType == LLVMDisassembler_ReferenceType_Out_LitPool_SymAddr ||
+       ReferenceType == LLVMDisassembler_ReferenceType_Out_LitPool_CstrAddr)
+      cStream << "literal pool for: " << ReferenceName;
+  }
+}
diff --git a/lib/MC/MCObjectSymbolizer.cpp b/lib/MC/MCObjectSymbolizer.cpp
new file mode 100644
index 0000000..e1d504e
--- /dev/null
+++ b/lib/MC/MCObjectSymbolizer.cpp
@@ -0,0 +1,175 @@
+//===-- lib/MC/MCObjectSymbolizer.cpp -------------------------------------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/MC/MCObjectSymbolizer.h"
+#include "llvm/ADT/SmallString.h"
+#include "llvm/MC/MCContext.h"
+#include "llvm/MC/MCExpr.h"
+#include "llvm/MC/MCInst.h"
+#include "llvm/MC/MCRelocationInfo.h"
+#include "llvm/MC/MCSymbol.h"
+#include "llvm/Object/MachO.h"
+#include "llvm/Object/ELF.h"
+#include "llvm/Support/raw_ostream.h"
+
+using namespace llvm;
+using namespace object;
+
+//===- MCMachObjectSymbolizer ---------------------------------------------===//
+
+namespace {
+class MCMachObjectSymbolizer : public MCObjectSymbolizer {
+public:
+  MCMachObjectSymbolizer(MCContext &Ctx, OwningPtr<MCRelocationInfo> &RelInfo,
+                         const object::MachOObjectFile *MachOOF)
+    : MCObjectSymbolizer(Ctx, RelInfo, MachOOF)
+  {}
+
+  void tryAddingPcLoadReferenceComment(raw_ostream &cStream,
+                                       int64_t Value, uint64_t Address) {
+    AddrToRelocMap::iterator RI = AddrToReloc.find(Address);
+    if (RI != AddrToReloc.end()) {
+      const MCExpr *RelExpr = RelInfo->createExprForRelocation(RI->second);
+      if (!RelExpr || RelExpr->EvaluateAsAbsolute(Value) == false)
+        return;
+    }
+    uint64_t Addr = Value;
+    AddrToSectionMap::const_iterator SI = AddrToSection.find(Addr);
+    if (SI.valid()) {
+      DataRefImpl DRI; DRI.p = *SI;
+      SectionRef S(DRI, Obj);
+      StringRef Name; S.getName(Name);
+      if (Name == "__cstring") {
+        StringRef Contents;
+        S.getContents(Contents);
+        Contents = Contents.substr(Addr - SI.start());
+        cStream << " ## literal pool for: "
+                << Contents.substr(0, Contents.find_first_of(0));
+      }
+    }
+  }
+};
+} // End unnamed namespace
+
+//===- MCObjectSymbolizer -------------------------------------------------===//
+
+MCObjectSymbolizer::MCObjectSymbolizer(MCContext &Ctx,
+                                       OwningPtr<MCRelocationInfo> &RelInfo,
+                                       const ObjectFile *Obj)
+    : MCSymbolizer(Ctx, RelInfo), Obj(Obj),
+      AddrToSectionAllocator(), AddrToSection(AddrToSectionAllocator),
+      AddrToReloc() {
+  error_code ec;
+  for (section_iterator SI = Obj->begin_sections(),
+                        SE = Obj->end_sections();
+                        SI != SE;
+                        SI.increment(ec)) {
+    if (ec) break;
+    uint64_t StartAddr; SI->getAddress(StartAddr);
+    uint64_t Size; SI->getSize(Size);
+    StringRef SecName; SI->getName(SecName);
+    bool RequiredForExec; SI->isRequiredForExecution(RequiredForExec);
+    if (RequiredForExec == false || Size == 0)
+      continue;
+    AddrToSection.insert(StartAddr, StartAddr + Size - 1,
+                         SI->getRawDataRefImpl().p);
+    for (relocation_iterator RI = SI->begin_relocations(),
+                             RE = SI->end_relocations();
+                             RI != RE;
+                             RI.increment(ec)) {
+      if (ec) break;
+      // FIXME: libObject is inconsistent regarding error handling. The
+      // overwhelming majority of methods always return object_error::success,
+      // and assert for simple errors.. Here, ELFObjectFile::getRelocationOffset
+      // asserts when the file type isn't ET_REL.
+      // This workaround handles x86-64 elf, the only one that has a relocinfo.
+      uint64_t Offset;
+      if (Obj->isELF()) {
+        const ELF64LEObjectFile *ELFObj = dyn_cast<ELF64LEObjectFile>(Obj);
+        if (ELFObj == 0)
+          break;
+        if (ELFObj->getElfHeader()->e_type == ELF::ET_REL) {
+          RI->getOffset(Offset);
+          Offset += StartAddr;
+        } else {
+          RI->getAddress(Offset);
+        }
+      } else {
+        RI->getOffset(Offset);
+        Offset += StartAddr;
+      }
+      // At a specific address, only keep the first relocation.
+      if (AddrToReloc.find(Offset) == AddrToReloc.end())
+        AddrToReloc[Offset] = *RI;
+    }
+  }
+}
+
+bool MCObjectSymbolizer::
+tryAddingSymbolicOperand(MCInst &MI, raw_ostream &cStream,
+                         int64_t Value, uint64_t Address, bool IsBranch,
+                         uint64_t Offset, uint64_t InstSize) {
+  AddrToRelocMap::iterator RI = AddrToReloc.find(Address + Offset);
+  if (RI != AddrToReloc.end()) {
+    if (const MCExpr *RelExpr = RelInfo->createExprForRelocation(RI->second)) {
+      MI.addOperand(MCOperand::CreateExpr(RelExpr));
+      return true;
+    }
+    // Only try to create a symbol+offset expression if there is no relocation.
+    return false;
+  }
+
+  // Interpret Value as a branch target.
+  if (IsBranch == false)
+    return false;
+  uint64_t UValue = Value;
+  // FIXME: map instead of looping each time?
+  error_code ec;
+  for (symbol_iterator SI = Obj->begin_symbols(),
+       SE = Obj->end_symbols();
+       SI != SE;
+       SI.increment(ec)) {
+    if (ec) break;
+    uint64_t SymAddr; SI->getAddress(SymAddr);
+    uint64_t SymSize; SI->getSize(SymSize);
+    StringRef SymName; SI->getName(SymName);
+    SymbolRef::Type SymType; SI->getType(SymType);
+    if (SymAddr == UnknownAddressOrSize || SymSize == UnknownAddressOrSize
+        || SymName.empty() || SymType != SymbolRef::ST_Function)
+      continue;
+
+    if ( SymAddr == UValue ||
+        (SymAddr <= UValue && SymAddr + SymSize > UValue)) {
+      MCSymbol *Sym = Ctx.GetOrCreateSymbol(SymName);
+      const MCExpr *Expr = MCSymbolRefExpr::Create(Sym, Ctx);
+      if (SymAddr != UValue) {
+        const MCExpr *Off = MCConstantExpr::Create(UValue - SymAddr, Ctx);
+        Expr = MCBinaryExpr::CreateAdd(Expr, Off, Ctx);
+      }
+      MI.addOperand(MCOperand::CreateExpr(Expr));
+      return true;
+    }
+  }
+  return false;
+}
+
+void MCObjectSymbolizer::
+tryAddingPcLoadReferenceComment(raw_ostream &cStream,
+                                int64_t Value, uint64_t Address) {
+}
+
+MCObjectSymbolizer *
+MCObjectSymbolizer::createObjectSymbolizer(MCContext &Ctx,
+                                           OwningPtr<MCRelocationInfo> &RelInfo,
+                                           const ObjectFile *Obj) {
+  if (const MachOObjectFile *MachOOF = dyn_cast<MachOObjectFile>(Obj)) {
+    return new MCMachObjectSymbolizer(Ctx, RelInfo, MachOOF);
+  }
+  return new MCObjectSymbolizer(Ctx, RelInfo, Obj);
+}
diff --git a/lib/MC/MCRelocationInfo.cpp b/lib/MC/MCRelocationInfo.cpp
new file mode 100644
index 0000000..a2a3174
--- /dev/null
+++ b/lib/MC/MCRelocationInfo.cpp
@@ -0,0 +1,39 @@
+//==-- lib/MC/MCRelocationInfo.cpp -------------------------------*- C++ -*-==//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/MC/MCRelocationInfo.h"
+#include "llvm/Object/ObjectFile.h"
+#include "llvm/Support/TargetRegistry.h"
+#include "llvm-c/Disassembler.h"
+
+using namespace llvm;
+
+MCRelocationInfo::MCRelocationInfo(MCContext &Ctx)
+  : Ctx(Ctx) {
+}
+
+MCRelocationInfo::~MCRelocationInfo() {
+}
+
+const MCExpr *
+MCRelocationInfo::createExprForRelocation(object::RelocationRef Rel) {
+  return 0;
+}
+
+const MCExpr *
+MCRelocationInfo::createExprForCAPIVariantKind(const MCExpr *SubExpr,
+                                               unsigned VariantKind) {
+  if (VariantKind != LLVMDisassembler_VariantKind_None)
+    return 0;
+  return SubExpr;
+}
+
+MCRelocationInfo *llvm::createMCRelocationInfo(MCContext &Ctx) {
+  return new MCRelocationInfo(Ctx);
+}
diff --git a/lib/MC/MCSymbolizer.cpp b/lib/MC/MCSymbolizer.cpp
new file mode 100644
index 0000000..1020b74
--- /dev/null
+++ b/lib/MC/MCSymbolizer.cpp
@@ -0,0 +1,20 @@
+//===-- llvm/MC/MCSymbolizer.cpp - MCSymbolizer class -----------*- C++ -*-===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/MC/MCSymbolizer.h"
+#include "llvm/MC/MCRelocationInfo.h"
+
+using namespace llvm;
+
+MCSymbolizer::MCSymbolizer(MCContext &Ctx, OwningPtr<MCRelocationInfo> &RelInfo)
+  : Ctx(Ctx), RelInfo(RelInfo.take()) {
+}
+
+MCSymbolizer::~MCSymbolizer() {
+}