[llvm-objdump] [debuginfod] Fetch for very-stripped binaries.

[mysterymath]

Summary

When a binary is missing section headers or symbols, objdump can't
provide as good of a disassembly. This change makes objdump try to fetch
a better verion of the binary by its build ID.

[mysterymath]

Integrate.

[mysterymath]

Correct clang-format formatting.

[jhenderson]

llvm/tools/llvm-objdump/llvm-objdump.cpp

1865

There's a lot of if clauses in this new area of code, so I'd expect to see a corresponding number of new test cases, but there's only one. Please make sure there's a test case that exercises each of the different possible code paths.

1984–1986

Might want to hold off on this usage here: there's an active discussion on whether we a) should allow it (due to potential footguns), and b) what the style of the C++17 if with initializer should be. See https://discourse.llvm.org/t/rfc-code-style-of-initializer-in-if-statement/65357/1.

[mysterymath]

Add tests; address review comments.

[jhenderson]

LGTM, with small nit.

llvm/test/tools/llvm-objdump/debuginfod.test

56–58

Small nit: I'd reorder these to match the order they're used in.

[mysterymath]

Reorder checks.

[mysterymath]

Update the no-section dump syntax.

llvm/test/tools/llvm-objdump/debuginfod.test

 RUN:   llvm-objdump -d --source --no-debuginfod %t/stripped | \
 RUN:   FileCheck %s --check-prefix=NOTFOUND
 
 # Use debuginfod without URLs if requested.
 RUN: env DEBUGINFOD_CACHE_PATH=%t llvm-objdump -d --source --debuginfod \
 RUN:   %t/stripped | \
 RUN:   FileCheck %s --check-prefix=FOUND
 
+# Use debuginfod to replace very-stripped binaries entirely.
+RUN: llvm-strip --strip-sections %t/stripped
+RUN: env DEBUGINFOD_CACHE_PATH=%t llvm-objdump -d --debuginfod \
+RUN:   %t/stripped | \
+RUN:   FileCheck %s --check-prefix=SYMBOLS
+
+SYMBOLS: <main>:
+
 FOUND: int main(int argc, char *argv[]) {
 NOTFOUND-NOT: int main(int argc, char *argv[]) {

llvm/tools/llvm-objdump/llvm-objdump.cpp

static void createFakeELFSections(ObjectFile &Obj) {
   else if (auto *Elf32BEObj = dyn_cast<ELF32BEObjectFile>(&Obj))
     Elf32BEObj->createFakeSections();
   else if (auto *Elf64BEObj = cast<ELF64BEObjectFile>(&Obj))
     Elf64BEObj->createFakeSections();
   else
     llvm_unreachable("Unsupported binary format");
 }
 
-// Try to fetch a debug file for the given object file using its Build ID.
-// Returns none if no additional debug information was found.
-static Optional<OwningBinary<Binary>> fetchDebugFile(const ObjectFile &Obj) {
+// Tries to fetch a more complete version of the given object file using its
+// Build ID. Returns none if nothing was found.
+static Optional<OwningBinary<Binary>>
+fetchBinaryByBuildID(const ObjectFile &Obj) {
   Optional<object::BuildIDRef> BuildID = getBuildID(&Obj);
   if (!BuildID)
     return None;
   Optional<std::string> Path = BIDFetcher->fetch(*BuildID);
   if (!Path)
     return None;
   Expected<OwningBinary<Binary>> DebugBinary = createBinary(*Path);
   if (!DebugBinary) {
     consumeError(DebugBinary.takeError());
     return None;
   }
-  if (ObjectFile *DbgObj = dyn_cast<ObjectFile>(DebugBinary->getBinary());
-      DbgObj->hasDebugInfo())
-    return std::move(*DebugBinary);
-  return None;
 }
 
 static void disassembleObject(const Target *TheTarget, ObjectFile &Obj,
                               MCContext &Ctx, MCDisassembler *PrimaryDisAsm,
                               MCDisassembler *SecondaryDisAsm,
                               const MCInstrAnalysis *MIA, MCInstPrinter *IP,
                               const MCSubtargetInfo *PrimarySTI,
                               const MCSubtargetInfo *SecondarySTI,
static void disassembleObject(const Target *TheTarget, ObjectFile &Obj,
   }
   StringSet<> MissingDisasmSymbolSet =
       set_difference(DisasmSymbolSet, FoundDisasmSymbolSet);
   for (StringRef Sym : MissingDisasmSymbolSet.keys())
     reportWarning("failed to disassemble missing symbol " + Sym, FileName);
 }
 
 static void disassembleObject(ObjectFile *Obj, bool InlineRelocs) {
+  // If information useful for showing the disassembly is missing, try to find a
+  // more complete binary and disassemble that instead.
+  OwningBinary<Binary> FetchedBinary;
+  if (Obj->sections().empty() || Obj->symbols().empty()) {
+    if (Optional<OwningBinary<Binary>> FetchedBinaryOpt =
+            fetchBinaryByBuildID(*Obj)) {
+      if (auto *O = dyn_cast<ObjectFile>(FetchedBinaryOpt->getBinary());
+          (!O->sections().empty() || !O->symbols().empty()) &&
+          (!O->sections().empty() || Obj->sections().empty()) &&
+          (!O->symbols().empty() || Obj->symbols().empty())) {
+        FetchedBinary = std::move(*FetchedBinaryOpt);
+        Obj = O;
+      }
+    }
+  }
+
   const Target *TheTarget = getTarget(Obj);
 
   // Package up features to be passed to target/subtarget
   SubtargetFeatures Features = Obj->getFeatures();
   if (!MAttrs.empty()) {
     for (unsigned I = 0; I != MAttrs.size(); ++I)
       Features.AddFeature(MAttrs[I]);
   } else if (MCPU.empty() && Obj->getArch() == llvm::Triple::aarch64) {
reportError(Obj->getFileName(),
                 "no instruction printer for target " + TripleName);
   IP->setPrintImmHex(PrintImmHex);
   IP->setPrintBranchImmAsAddress(true);
   IP->setSymbolizeOperands(SymbolizeOperands);
   IP->setMCInstrAnalysis(MIA.get());
 
   PrettyPrinter &PIP = selectPrettyPrinter(Triple(TripleName));
   ObjectFile *DbgObj = Obj;
-  OwningBinary<Binary> DebugBinary;
-  if (!Obj->hasDebugInfo()) {
-    Optional<OwningBinary<Binary>> DebugBinaryOpt = fetchDebugFile(*Obj);
-    if (DebugBinaryOpt) {
-      DebugBinary = std::move(*DebugBinaryOpt);
-      DbgObj = cast<ObjectFile>(DebugBinary.getBinary());
+  if (!FetchedBinary.getBinary() && !Obj->hasDebugInfo()) {
+    if (Optional<OwningBinary<Binary>> DebugBinaryOpt =
+            fetchBinaryByBuildID(*Obj)) {
+      if (ObjectFile *FetchedObj =
+              dyn_cast<ObjectFile>(DebugBinaryOpt->getBinary());
+          FetchedObj->hasDebugInfo()) {
+        FetchedBinary = std::move(*DebugBinaryOpt);
+        DbgObj = FetchedObj;
+      }
     }
   }
   SourcePrinter SP(DbgObj, TheTarget->getName());
 
   for (StringRef Opt : DisassemblerOptions)
     if (!IP->applyTargetSpecificCLOption(Opt))
       reportError(Obj->getFileName(),
                   "Unrecognized disassembler option: " + Opt);