Add support for inline asm symbols to IRObjectFile.

This also enables it in llvm-nm so that it can be tested.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@212282 91177308-0d34-0410-b5e6-96231b3b80d8

include/llvm/Object/IRObjectFile.h

@@ -25,6 +25,7 @@ namespace object {
 class IRObjectFile : public SymbolicFile {
   std::unique_ptr<Module> M;
   std::unique_ptr<Mangler> Mang;
+  std::vector<std::pair<std::string, uint32_t>> AsmSymbols;
 
 public:
   IRObjectFile(std::unique_ptr<MemoryBuffer> Object, std::error_code &EC,
@@ -34,7 +35,7 @@ public:
   std::error_code printSymbolName(raw_ostream &OS,
                                   DataRefImpl Symb) const override;
   uint32_t getSymbolFlags(DataRefImpl Symb) const override;
-  const GlobalValue &getSymbolGV(DataRefImpl Symb) const;
+  const GlobalValue *getSymbolGV(DataRefImpl Symb) const;
   basic_symbol_iterator symbol_begin_impl() const override;
   basic_symbol_iterator symbol_end_impl() const override;
 

include/llvm/Object/RecordStreamer.h

@@ -0,0 +1,42 @@
+//===-- RecordStreamer.h - Record asm defined and used symbols ---*- C++ -*===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_OBJECT_RECORD_STREAMER
+#define LLVM_OBJECT_RECORD_STREAMER
+
+#include "llvm/MC/MCStreamer.h"
+
+namespace llvm {
+class RecordStreamer : public MCStreamer {
+public:
+  enum State { NeverSeen, Global, Defined, DefinedGlobal, Used };
+
+private:
+  StringMap<State> Symbols;
+  void markDefined(const MCSymbol &Symbol);
+  void markGlobal(const MCSymbol &Symbol);
+  void markUsed(const MCSymbol &Symbol);
+  void visitUsedSymbol(const MCSymbol &Sym) override;
+
+public:
+  typedef StringMap<State>::const_iterator const_iterator;
+  const_iterator begin();
+  const_iterator end();
+  RecordStreamer(MCContext &Context);
+  void EmitInstruction(const MCInst &Inst, const MCSubtargetInfo &STI) override;
+  void EmitLabel(MCSymbol *Symbol) override;
+  void EmitAssignment(MCSymbol *Symbol, const MCExpr *Value) override;
+  bool EmitSymbolAttribute(MCSymbol *Symbol, MCSymbolAttr Attribute) override;
+  void EmitZerofill(const MCSection *Section, MCSymbol *Symbol, uint64_t Size,
+                    unsigned ByteAlignment) override;
+  void EmitCommonSymbol(MCSymbol *Symbol, uint64_t Size,
+                        unsigned ByteAlignment) override;
+};
+}
+#endif

lib/LTO/LTOModule.cpp

@@ -24,11 +24,11 @@
 #include "llvm/MC/MCInstrInfo.h"
 #include "llvm/MC/MCParser/MCAsmParser.h"
 #include "llvm/MC/MCSection.h"
-#include "llvm/MC/MCStreamer.h"
 #include "llvm/MC/MCSubtargetInfo.h"
 #include "llvm/MC/MCSymbol.h"
 #include "llvm/MC/MCTargetAsmParser.h"
 #include "llvm/MC/SubtargetFeature.h"
+#include "llvm/Object/RecordStreamer.h"
 #include "llvm/Support/CommandLine.h"
 #include "llvm/Support/FileSystem.h"
 #include "llvm/Support/Host.h"
@@ -552,105 +552,6 @@ LTOModule::addPotentialUndefinedSymbol(const GlobalValue *decl, bool isFunc) {
   entry.setValue(info);
 }
 
-namespace {
-
-  class RecordStreamer : public MCStreamer {
-  public:
-    enum State { NeverSeen, Global, Defined, DefinedGlobal, Used };
-
-  private:
-    StringMap<State> Symbols;
-
-    void markDefined(const MCSymbol &Symbol) {
-      State &S = Symbols[Symbol.getName()];
-      switch (S) {
-      case DefinedGlobal:
-      case Global:
-        S = DefinedGlobal;
-        break;
-      case NeverSeen:
-      case Defined:
-      case Used:
-        S = Defined;
-        break;
-      }
-    }
-    void markGlobal(const MCSymbol &Symbol) {
-      State &S = Symbols[Symbol.getName()];
-      switch (S) {
-      case DefinedGlobal:
-      case Defined:
-        S = DefinedGlobal;
-        break;
-
-      case NeverSeen:
-      case Global:
-      case Used:
-        S = Global;
-        break;
-      }
-    }
-    void markUsed(const MCSymbol &Symbol) {
-      State &S = Symbols[Symbol.getName()];
-      switch (S) {
-      case DefinedGlobal:
-      case Defined:
-      case Global:
-        break;
-
-      case NeverSeen:
-      case Used:
-        S = Used;
-        break;
-      }
-    }
-
-    void visitUsedSymbol(const MCSymbol &Sym) override {
-      markUsed(Sym);
-    }
-
-  public:
-    typedef StringMap<State>::const_iterator const_iterator;
-
-    const_iterator begin() {
-      return Symbols.begin();
-    }
-
-    const_iterator end() {
-      return Symbols.end();
-    }
-
-    RecordStreamer(MCContext &Context) : MCStreamer(Context) {}
-
-    void EmitInstruction(const MCInst &Inst,
-                         const MCSubtargetInfo &STI) override {
-      MCStreamer::EmitInstruction(Inst, STI);
-    }
-    void EmitLabel(MCSymbol *Symbol) override {
-      MCStreamer::EmitLabel(Symbol);
-      markDefined(*Symbol);
-    }
-    void EmitAssignment(MCSymbol *Symbol, const MCExpr *Value) override {
-      markDefined(*Symbol);
-      MCStreamer::EmitAssignment(Symbol, Value);
-    }
-    bool EmitSymbolAttribute(MCSymbol *Symbol,
-                             MCSymbolAttr Attribute) override {
-      if (Attribute == MCSA_Global)
-        markGlobal(*Symbol);
-      return true;
-    }
-    void EmitZerofill(const MCSection *Section, MCSymbol *Symbol,
-                      uint64_t Size , unsigned ByteAlignment) override {
-      markDefined(*Symbol);
-    }
-    void EmitCommonSymbol(MCSymbol *Symbol, uint64_t Size,
-                          unsigned ByteAlignment) override {
-      markDefined(*Symbol);
-    }
-  };
-} // end anonymous namespace
-
 /// addAsmGlobalSymbols - Add global symbols from module-level ASM to the
 /// defined or undefined lists.
 bool LTOModule::addAsmGlobalSymbols(std::string &errMsg) {

lib/Object/CMakeLists.txt

@@ -12,5 +12,6 @@ add_llvm_library(LLVMObject
   MachOUniversal.cpp
   Object.cpp
   ObjectFile.cpp
+  RecordStreamer.cpp
   SymbolicFile.cpp
   )

lib/Object/IRObjectFile.cpp

@@ -17,7 +17,17 @@
 #include "llvm/IR/Mangler.h"
 #include "llvm/IR/Module.h"
 #include "llvm/Object/IRObjectFile.h"
+#include "llvm/Object/RecordStreamer.h"
+#include "llvm/MC/MCRegisterInfo.h"
+#include "llvm/MC/MCAsmInfo.h"
+#include "llvm/MC/MCContext.h"
+#include "llvm/MC/MCInstrInfo.h"
+#include "llvm/MC/MCObjectFileInfo.h"
+#include "llvm/MC/MCTargetAsmParser.h"
+#include "llvm/MC/MCParser/MCAsmParser.h"
 #include "llvm/Support/MemoryBuffer.h"
+#include "llvm/Support/SourceMgr.h"
+#include "llvm/Support/TargetRegistry.h"
 #include "llvm/Support/raw_ostream.h"
 using namespace llvm;
 using namespace object;
@@ -37,12 +47,85 @@ IRObjectFile::IRObjectFile(std::unique_ptr<MemoryBuffer> Object,
     return;
 
   Mang.reset(new Mangler(DL));
+
+  const std::string &InlineAsm = M->getModuleInlineAsm();
+  if (InlineAsm.empty())
+    return;
+
+  StringRef Triple = M->getTargetTriple();
+  std::string Err;
+  const Target *T = TargetRegistry::lookupTarget(Triple, Err);
+  if (!T)
+    return;
+
+  std::unique_ptr<MCRegisterInfo> MRI(T->createMCRegInfo(Triple));
+  if (!MRI)
+    return;
+
+  std::unique_ptr<MCAsmInfo> MAI(T->createMCAsmInfo(*MRI, Triple));
+  if (!MAI)
+    return;
+
+  std::unique_ptr<MCSubtargetInfo> STI(
+      T->createMCSubtargetInfo(Triple, "", ""));
+  if (!STI)
+    return;
+
+  std::unique_ptr<MCInstrInfo> MCII(T->createMCInstrInfo());
+  if (!MCII)
+    return;
+
+  MCObjectFileInfo MOFI;
+  MCContext MCCtx(MAI.get(), MRI.get(), &MOFI);
+  MOFI.InitMCObjectFileInfo(Triple, Reloc::Default, CodeModel::Default, MCCtx);
+  std::unique_ptr<RecordStreamer> Streamer(new RecordStreamer(MCCtx));
+
+  std::unique_ptr<MemoryBuffer> Buffer(MemoryBuffer::getMemBuffer(InlineAsm));
+  SourceMgr SrcMgr;
+  SrcMgr.AddNewSourceBuffer(Buffer.release(), SMLoc());
+  std::unique_ptr<MCAsmParser> Parser(
+      createMCAsmParser(SrcMgr, MCCtx, *Streamer, *MAI));
+
+  MCTargetOptions MCOptions;
+  std::unique_ptr<MCTargetAsmParser> TAP(
+      T->createMCAsmParser(*STI, *Parser, *MCII, MCOptions));
+  if (!TAP)
+    return;
+
+  Parser->setTargetParser(*TAP);
+  if (Parser->Run(false))
+    return;
+
+  for (auto &KV : *Streamer) {
+    StringRef Key = KV.first();
+    RecordStreamer::State Value = KV.second;
+    uint32_t Res = BasicSymbolRef::SF_None;
+    switch (Value) {
+    case RecordStreamer::NeverSeen:
+      llvm_unreachable("foo");
+    case RecordStreamer::DefinedGlobal:
+      Res |= BasicSymbolRef::SF_Global;
+      break;
+    case RecordStreamer::Defined:
+      break;
+    case RecordStreamer::Global:
+    case RecordStreamer::Used:
+      Res |= BasicSymbolRef::SF_Undefined;
+      Res |= BasicSymbolRef::SF_Global;
+      break;
+    }
+    AsmSymbols.push_back(
+        std::make_pair<std::string, uint32_t>(Key, std::move(Res)));
+  }
 }
 
 IRObjectFile::~IRObjectFile() { M->getMaterializer()->releaseBuffer(); }
 
-static const GlobalValue &getGV(DataRefImpl &Symb) {
+static const GlobalValue *getGV(DataRefImpl &Symb) {
-  return *reinterpret_cast<GlobalValue*>(Symb.p & ~uintptr_t(3));
+  if ((Symb.p & 3) == 3)
+    return nullptr;
+
+  return reinterpret_cast<GlobalValue*>(Symb.p & ~uintptr_t(3));
 }
 
 static uintptr_t skipEmpty(Module::const_alias_iterator I, const Module &M) {
@@ -66,31 +149,43 @@ static uintptr_t skipEmpty(Module::const_iterator I, const Module &M) {
   return reinterpret_cast<uintptr_t>(GV) | 0;
 }
 
+static unsigned getAsmSymIndex(DataRefImpl Symb) {
+  assert((Symb.p & uintptr_t(3)) == 3);
+  uintptr_t Index = Symb.p & ~uintptr_t(3);
+  Index >>= 2;
+  return Index;
+}
+
 void IRObjectFile::moveSymbolNext(DataRefImpl &Symb) const {
-  const GlobalValue *GV = &getGV(Symb);
+  const GlobalValue *GV = getGV(Symb);
-  const Module &M = *GV->getParent();
   uintptr_t Res;
+
   switch (Symb.p & 3) {
   case 0: {
     Module::const_iterator Iter(static_cast<const Function*>(GV));
     ++Iter;
-    Res = skipEmpty(Iter, M);
+    Res = skipEmpty(Iter, *M);
     break;
   }
   case 1: {
     Module::const_global_iterator Iter(static_cast<const GlobalVariable*>(GV));
     ++Iter;
-    Res = skipEmpty(Iter, M);
+    Res = skipEmpty(Iter, *M);
     break;
   }
   case 2: {
     Module::const_alias_iterator Iter(static_cast<const GlobalAlias*>(GV));
     ++Iter;
-    Res = skipEmpty(Iter, M);
+    Res = skipEmpty(Iter, *M);
+    break;
+  }
+  case 3: {
+    unsigned Index = getAsmSymIndex(Symb);
+    assert(Index < AsmSymbols.size());
+    ++Index;
+    Res = (Index << 2) | 3;
     break;
   }
-  case 3:
-    llvm_unreachable("Invalid symbol reference");
   }
 
   Symb.p = Res;
@@ -98,12 +193,18 @@ void IRObjectFile::moveSymbolNext(DataRefImpl &Symb) const {
 
 std::error_code IRObjectFile::printSymbolName(raw_ostream &OS,
                                               DataRefImpl Symb) const {
-  const GlobalValue &GV = getGV(Symb);
+  const GlobalValue *GV = getGV(Symb);
+  if (!GV) {
+    unsigned Index = getAsmSymIndex(Symb);
+    assert(Index <= AsmSymbols.size());
+    OS << AsmSymbols[Index].first;
+    return object_error::success;;
+  }
 
   if (Mang)
-    Mang->getNameWithPrefix(OS, &GV, false);
+    Mang->getNameWithPrefix(OS, GV, false);
   else
-    OS << GV.getName();
+    OS << GV->getName();
 
   return object_error::success;
 }
@@ -119,25 +220,31 @@ static bool isDeclaration(const GlobalValue &V) {
 }
 
 uint32_t IRObjectFile::getSymbolFlags(DataRefImpl Symb) const {
-  const GlobalValue &GV = getGV(Symb);
+  const GlobalValue *GV = getGV(Symb);
+
+  if (!GV) {
+    unsigned Index = getAsmSymIndex(Symb);
+    assert(Index <= AsmSymbols.size());
+    return AsmSymbols[Index].second;
+  }
 
   uint32_t Res = BasicSymbolRef::SF_None;
-  if (isDeclaration(GV))
+  if (isDeclaration(*GV))
     Res |= BasicSymbolRef::SF_Undefined;
-  if (GV.hasPrivateLinkage())
+  if (GV->hasPrivateLinkage())
     Res |= BasicSymbolRef::SF_FormatSpecific;
-  if (!GV.hasLocalLinkage())
+  if (!GV->hasLocalLinkage())
     Res |= BasicSymbolRef::SF_Global;
-  if (GV.hasCommonLinkage())
+  if (GV->hasCommonLinkage())
     Res |= BasicSymbolRef::SF_Common;
-  if (GV.hasLinkOnceLinkage() || GV.hasWeakLinkage())
+  if (GV->hasLinkOnceLinkage() || GV->hasWeakLinkage())
     Res |= BasicSymbolRef::SF_Weak;
 
   return Res;
 }
 
-const GlobalValue &IRObjectFile::getSymbolGV(DataRefImpl Symb) const {
+const GlobalValue *IRObjectFile::getSymbolGV(DataRefImpl Symb) const {
-  const GlobalValue &GV = getGV(Symb);
+  const GlobalValue *GV = getGV(Symb);
   return GV;
 }
 
@@ -150,7 +257,9 @@ basic_symbol_iterator IRObjectFile::symbol_begin_impl() const {
 
 basic_symbol_iterator IRObjectFile::symbol_end_impl() const {
   DataRefImpl Ret;
-  Ret.p = 3;
+  uint64_t NumAsm = AsmSymbols.size();
+  NumAsm <<= 2;
+  Ret.p = 3 | NumAsm;
   return basic_symbol_iterator(BasicSymbolRef(Ret, this));
 }
 

lib/Object/LLVMBuild.txt

@@ -19,4 +19,4 @@
 type = Library
 name = Object
 parent = Libraries
-required_libraries = BitReader Core Support MC
+required_libraries = BitReader Core Support MC MCParser

lib/Object/RecordStreamer.cpp

@@ -0,0 +1,100 @@
+//===-- RecordStreamer.cpp - Record asm definde and used symbols ----------===//
+//
+//                     The LLVM Compiler Infrastructure
+//
+// This file is distributed under the University of Illinois Open Source
+// License. See LICENSE.TXT for details.
+//
+//===----------------------------------------------------------------------===//
+
+#include "llvm/Object/RecordStreamer.h"
+#include "llvm/MC/MCSymbol.h"
+using namespace llvm;
+
+void RecordStreamer::markDefined(const MCSymbol &Symbol) {
+  State &S = Symbols[Symbol.getName()];
+  switch (S) {
+  case DefinedGlobal:
+  case Global:
+    S = DefinedGlobal;
+    break;
+  case NeverSeen:
+  case Defined:
+  case Used:
+    S = Defined;
+    break;
+  }
+}
+
+void RecordStreamer::markGlobal(const MCSymbol &Symbol) {
+  State &S = Symbols[Symbol.getName()];
+  switch (S) {
+  case DefinedGlobal:
+  case Defined:
+    S = DefinedGlobal;
+    break;
+
+  case NeverSeen:
+  case Global:
+  case Used:
+    S = Global;
+    break;
+  }
+}
+
+void RecordStreamer::markUsed(const MCSymbol &Symbol) {
+  State &S = Symbols[Symbol.getName()];
+  switch (S) {
+  case DefinedGlobal:
+  case Defined:
+  case Global:
+    break;
+
+  case NeverSeen:
+  case Used:
+    S = Used;
+    break;
+  }
+}
+
+void RecordStreamer::visitUsedSymbol(const MCSymbol &Sym) { markUsed(Sym); }
+
+RecordStreamer::const_iterator RecordStreamer::begin() {
+  return Symbols.begin();
+}
+
+RecordStreamer::const_iterator RecordStreamer::end() { return Symbols.end(); }
+
+RecordStreamer::RecordStreamer(MCContext &Context) : MCStreamer(Context) {}
+
+void RecordStreamer::EmitInstruction(const MCInst &Inst,
+                                     const MCSubtargetInfo &STI) {
+  MCStreamer::EmitInstruction(Inst, STI);
+}
+
+void RecordStreamer::EmitLabel(MCSymbol *Symbol) {
+  MCStreamer::EmitLabel(Symbol);
+  markDefined(*Symbol);
+}
+
+void RecordStreamer::EmitAssignment(MCSymbol *Symbol, const MCExpr *Value) {
+  markDefined(*Symbol);
+  MCStreamer::EmitAssignment(Symbol, Value);
+}
+
+bool RecordStreamer::EmitSymbolAttribute(MCSymbol *Symbol,
+                                         MCSymbolAttr Attribute) {
+  if (Attribute == MCSA_Global)
+    markGlobal(*Symbol);
+  return true;
+}
+
+void RecordStreamer::EmitZerofill(const MCSection *Section, MCSymbol *Symbol,
+                                  uint64_t Size, unsigned ByteAlignment) {
+  markDefined(*Symbol);
+}
+
+void RecordStreamer::EmitCommonSymbol(MCSymbol *Symbol, uint64_t Size,
+                                      unsigned ByteAlignment) {
+  markDefined(*Symbol);
+}

test/Object/nm-ir.ll

@@ -10,6 +10,17 @@
 ; CHECK-NEXT: d g2
 ; CHECK-NEXT: C g3
 ; CHECK-NOT: g4
+; CHECK-NEXT: T global_asm_sym
+; CHECK-NEXT: t local_asm_sym
+; CHECK-NEXT: U undef_asm_sy
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+module asm ".global global_asm_sym"
+module asm "global_asm_sym:"
+module asm "local_asm_sym:"
+module asm ".long undef_asm_sym"
 
 @g1 = global i32 42
 @g2 = internal global i32 42

tools/llvm-nm/CMakeLists.txt

@@ -1,4 +1,5 @@
 set(LLVM_LINK_COMPONENTS
+  ${LLVM_TARGETS_TO_BUILD}
   Object
   Support
   )

tools/llvm-nm/llvm-nm.cpp

@@ -37,6 +37,7 @@
 #include "llvm/Support/Program.h"
 #include "llvm/Support/Signals.h"
 #include "llvm/Support/raw_ostream.h"
+#include "llvm/Support/TargetSelect.h"
 #include <algorithm>
 #include <cctype>
 #include <cerrno>
@@ -637,8 +638,10 @@ static char getSymbolNMTypeChar(const GlobalValue &GV) {
 }
 
 static char getSymbolNMTypeChar(IRObjectFile &Obj, basic_symbol_iterator I) {
-  const GlobalValue &GV = Obj.getSymbolGV(I->getRawDataRefImpl());
+  const GlobalValue *GV = Obj.getSymbolGV(I->getRawDataRefImpl());
-  return getSymbolNMTypeChar(GV);
+  if (!GV)
+    return 't';
+  return getSymbolNMTypeChar(*GV);
 }
 
 template <class ELFT>
@@ -723,8 +726,8 @@ static void dumpSymbolNamesFromObject(SymbolicFile *Obj, bool printName) {
       continue;
     if (WithoutAliases) {
       if (IRObjectFile *IR = dyn_cast<IRObjectFile>(Obj)) {
-        const GlobalValue &GV = IR->getSymbolGV(I->getRawDataRefImpl());
+        const GlobalValue *GV = IR->getSymbolGV(I->getRawDataRefImpl());
-        if (isa<GlobalAlias>(GV))
+        if (GV && isa<GlobalAlias>(GV))
           continue;
       }
     }
@@ -993,6 +996,10 @@ int main(int argc, char **argv) {
   if (error(sys::ChangeStdinToBinary()))
     return 1;
 
+  llvm::InitializeAllTargetInfos();
+  llvm::InitializeAllTargetMCs();
+  llvm::InitializeAllAsmParsers();
+
   ToolName = argv[0];
   if (BSDFormat)
     OutputFormat = bsd;