[llvm-symbolizer] Add support for specifying input addresses on the command-line

[jhenderson]

Summary

See https://bugs.llvm.org/show_bug.cgi?id=40070.

GNU addr2line accepts input addresses both on the command-line and via stdin. llvm-symbolizer previously only supported the latter. This change adds support for the former. As with addr2line, the new behaviour is to only look for addresses on stdin if no positional arguments were provided to llvm-symbolizer.

[jhenderson]

Ping!

[ruiu]

test/tools/llvm-symbolizer/basic.s

12–14

Why don't you do

echo "0xa 0xb" | llvm-symbolizer ...

?

tools/llvm-symbolizer/llvm-symbolizer.cpp

205–207

Now it is more straightforward to move fgets inside the while condition?

[jhenderson]

Move fgets inside loop termination condition.

[jhenderson]

Thanks @ruiu!

test/tools/llvm-symbolizer/basic.s

12–14

When fed in via stdin, each input needs to be on a new line. I'm not sure I know a way to do this with a single echo command.

tools/llvm-symbolizer/llvm-symbolizer.cpp

205–207

I'm actually not sure why this couldn't have been done before.

[ruiu]

LGTM

test/tools/llvm-symbolizer/basic.s

12–14

In lld we do like this:

echo -e "0xa\n0xb\n" | ...

[jhenderson]

Use -e to simplify the echo commands in the test.

I'll commit this tomorrow if there are no objections before then, as I'm just leaving the office.

[ruiu]

test/tools/llvm-symbolizer/basic.s

11

Not sure if it is safe without "". Perhaps you should add ""to be on the safe side.

12

Remove %t2.input.

[jhenderson]

Add quotes and remove spurious redirection.

[jhenderson]

test/tools/llvm-symbolizer/basic.s

11

I think it probably is safe, but I'll add them anyway.

12

Oops, thanks.

test/tools/llvm-symbolizer/basic.s

+# REQUIRES: x86-registered-target
+
+foo:
+    .space 10
+    nop
+    nop
+
+# RUN: llvm-mc -filetype=obj -triple=x86_64-pc-linux %s -o %t.o -g
+
+# Check input addresses specified on stdin.
+# RUN: echo -e 0xa\n0xb | llvm-symbolizer --obj=%t.o | FileCheck %s
+# RUN: echo -e 10\n11 | llvm-symbolizer --obj=%t.o < %t2.input | FileCheck %s
+
+# Check input addresses specified on the command-line.
+# RUN: llvm-symbolizer 0xa 0xb --obj=%t.o | FileCheck %s
+# RUN: llvm-symbolizer 10 11 --obj=%t.o | FileCheck %s
+
+# CHECK: basic.s:5:0
+# CHECK: basic.s:6:0

tools/llvm-symbolizer/llvm-symbolizer.cpp

 
 static cl::opt<int> ClPrintSourceContextLines(
     "print-source-context-lines", cl::init(0),
     cl::desc("Print N number of source file context"));
 
 static cl::opt<bool> ClVerbose("verbose", cl::init(false),
                                cl::desc("Print verbose line info"));
 
+static cl::list<std::string> ClInputAddresses(cl::Positional,
+                                              cl::desc("<input addresses>..."),
+                                              cl::ZeroOrMore);
+
 template<typename T>
 static bool error(Expected<T> &ResOrErr) {
   if (ResOrErr)
     return false;
   logAllUnhandledErrors(ResOrErr.takeError(), errs(),
                         "LLVMSymbolizer: error reading file: ");
   return true;
 }
if (ClBinaryName.empty()) {
     ModuleName = ClBinaryName;
   }
   // Skip delimiters and parse module offset.
   pos += strspn(pos, kDelimiters);
   int offset_length = strcspn(pos, kDelimiters);
   return !StringRef(pos, offset_length).getAsInteger(0, ModuleOffset);
 }
 
+static void symbolizeInput(StringRef InputString, LLVMSymbolizer &Symbolizer,
+                           DIPrinter &Printer) {
+  bool IsData = false;
+  std::string ModuleName;
+  uint64_t ModuleOffset = 0;
+  if (!parseCommand(StringRef(InputString), IsData, ModuleName, ModuleOffset)) {
+    outs() << InputString;
+    return;
+  }
+
+  if (ClPrintAddress) {
+    outs() << "0x";
+    outs().write_hex(ModuleOffset);
+    StringRef Delimiter = ClPrettyPrint ? ": " : "\n";
+    outs() << Delimiter;
+  }
+  if (IsData) {
+    auto ResOrErr = Symbolizer.symbolizeData(ModuleName, ModuleOffset);
+    Printer << (error(ResOrErr) ? DIGlobal() : ResOrErr.get());
+  } else if (ClPrintInlining) {
+    auto ResOrErr =
+        Symbolizer.symbolizeInlinedCode(ModuleName, ModuleOffset, ClDwpName);
+    Printer << (error(ResOrErr) ? DIInliningInfo() : ResOrErr.get());
+  } else {
+    auto ResOrErr =
+        Symbolizer.symbolizeCode(ModuleName, ModuleOffset, ClDwpName);
+    Printer << (error(ResOrErr) ? DILineInfo() : ResOrErr.get());
+  }
+  outs() << "\n";
+  outs().flush();
+}
+
 int main(int argc, char **argv) {
   InitLLVM X(argc, argv);
 
   llvm::sys::InitializeCOMRAII COM(llvm::sys::COMThreadingMode::MultiThreaded);
 
   cl::ParseCommandLineOptions(argc, argv, "llvm-symbolizer\n");
   LLVMSymbolizer::Options Opts(ClPrintFunctions, ClUseSymbolTable, ClDemangle,
                                ClUseRelativeAddress, ClDefaultArch);
 
   for (const auto &hint : ClDsymHint) {
     if (sys::path::extension(hint) == ".dSYM") {
       Opts.DsymHints.push_back(hint);
     } else {
       errs() << "Warning: invalid dSYM hint: \"" << hint <<
                 "\" (must have the '.dSYM' extension).\n";
     }
   }
   LLVMSymbolizer Symbolizer(Opts);
 
   DIPrinter Printer(outs(), ClPrintFunctions != FunctionNameKind::None,
                     ClPrettyPrint, ClPrintSourceContextLines, ClVerbose);
 
+  if (ClInputAddresses.empty()) {
     const int kMaxInputStringLength = 1024;
     char InputString[kMaxInputStringLength];
 
-  while (true) {
-    if (!fgets(InputString, sizeof(InputString), stdin))
-      break;
-
-    bool IsData = false;
-    std::string ModuleName;
-    uint64_t ModuleOffset = 0;
-    if (!parseCommand(StringRef(InputString), IsData, ModuleName,
-                      ModuleOffset)) {
-      outs() << InputString;
-      continue;
-    }
-
-    if (ClPrintAddress) {
-      outs() << "0x";
-      outs().write_hex(ModuleOffset);
-      StringRef Delimiter = ClPrettyPrint ? ": " : "\n";
-      outs() << Delimiter;
-    }
-    if (IsData) {
-      auto ResOrErr = Symbolizer.symbolizeData(ModuleName, ModuleOffset);
-      Printer << (error(ResOrErr) ? DIGlobal() : ResOrErr.get());
-    } else if (ClPrintInlining) {
-      auto ResOrErr =
-          Symbolizer.symbolizeInlinedCode(ModuleName, ModuleOffset, ClDwpName);
-      Printer << (error(ResOrErr) ? DIInliningInfo()
-                                             : ResOrErr.get());
-    } else {
-      auto ResOrErr =
-          Symbolizer.symbolizeCode(ModuleName, ModuleOffset, ClDwpName);
-      Printer << (error(ResOrErr) ? DILineInfo() : ResOrErr.get());
-    }
-    outs() << "\n";
-    outs().flush();
+    while (fgets(InputString, sizeof(InputString), stdin))
+      symbolizeInput(InputString, Symbolizer, Printer);
+    for (StringRef Address : ClInputAddresses)
+      symbolizeInput(Address, Symbolizer, Printer);
   }
 
   return 0;
 }