Fix pdb-diff test.

A test was checked in on Friday that worked by checking in an
object file and PDB generated locally by MSVC, and then having
the test run lld-link on the object file and diffing LLD's PDB
against the checked in PDB.

This failed because part of the diffing algorithm involves
determining if two modules are the same, and if so drilling into
the module and diffing individual fields of the module.  The
only thing we can use to make this determination though is the
"name" of the module, which is a path to where the module (obj
file) was read from on the machine where it was linked.  This
fails for obvious reasons when comparing a PDB generated on one
machine to a PDB on another machine.

The fix employed here is to add two command line options to the
diff subcommand, which allow the user to specify a "binary root
path".  The bin root path, if specified, is stripped from the
beginning of any embedded PDB paths.  The test is updated to
specify the user's local test output directory for the left
PDB, and is hardcoded to the location where the original PDB
was created for the right PDB.  This way all the equivalence
comparisons should succeed.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@307555 91177308-0d34-0410-b5e6-96231b3b80d8

7 files changed, 289 insertions, 147 deletions

diff --git a/tools/llvm-pdbutil/Diff.cpp b/tools/llvm-pdbutil/Diff.cpp
index 9e520470790d..75e8f34bda06 100644
--- a/tools/llvm-pdbutil/Diff.cpp
+++ b/tools/llvm-pdbutil/Diff.cpp
@@ -23,13 +23,148 @@
 #include "llvm/DebugInfo/PDB/Native/PDBStringTable.h"
 #include "llvm/DebugInfo/PDB/Native/RawConstants.h"
 
+#include "llvm/Support/FileSystem.h"
 #include "llvm/Support/FormatAdapters.h"
 #include "llvm/Support/FormatProviders.h"
 #include "llvm/Support/FormatVariadic.h"
+#include "llvm/Support/Path.h"
 
 using namespace llvm;
 using namespace llvm::pdb;
 
+namespace {
+// Compare and format two stream numbers.  Stream numbers are considered
+// identical if they contain the same value, equivalent if they are both
+// the invalid stream or neither is the invalid stream, and different if
+// one is the invalid stream and another isn't.
+struct StreamNumberProvider {
+  static DiffResult compare(uint16_t L, uint16_t R) {
+    if (L == R)
+      return DiffResult::IDENTICAL;
+    bool LP = L != kInvalidStreamIndex;
+    bool RP = R != kInvalidStreamIndex;
+    if (LP != RP)
+      return DiffResult::DIFFERENT;
+    return DiffResult::EQUIVALENT;
+  }
+
+  static std::string format(uint16_t SN, bool Right) {
+    if (SN == kInvalidStreamIndex)
+      return "(not present)";
+    return formatv("{0}", SN).str();
+  }
+};
+
+// Compares and formats two module indices.  Modis are considered identical
+// if they are identical, equivalent if they either both contain a value or
+// both don't contain a value, and different if one contains a value and the
+// other doesn't.
+struct ModiProvider {
+  DiffResult compare(Optional<uint32_t> L, Optional<uint32_t> R) {
+    if (L == R)
+      return DiffResult::IDENTICAL;
+    if (L.hasValue() != R.hasValue())
+      return DiffResult::DIFFERENT;
+    return DiffResult::EQUIVALENT;
+  }
+
+  std::string format(Optional<uint32_t> Modi, bool Right) {
+    if (!Modi.hasValue())
+      return "(not present)";
+    return formatv("{0}", *Modi).str();
+  }
+};
+
+// Compares and formats two paths embedded in the PDB, ignoring the beginning
+// of the path if the user specified it as a "root path" on the command line.
+struct BinaryPathProvider {
+  explicit BinaryPathProvider(uint32_t MaxLen) : MaxLen(MaxLen) {}
+
+  DiffResult compare(StringRef L, StringRef R) {
+    if (L == R)
+      return DiffResult::IDENTICAL;
+
+    SmallString<64> LN = removeRoot(L, false);
+    SmallString<64> RN = removeRoot(R, true);
+
+    return (LN.equals_lower(RN)) ? DiffResult::EQUIVALENT
+                                 : DiffResult::DIFFERENT;
+  }
+
+  std::string format(StringRef S, bool Right) {
+    if (S.empty())
+      return "(empty)";
+
+    SmallString<64> Native = removeRoot(S, Right);
+    return truncateStringFront(Native.str(), MaxLen);
+  }
+
+  SmallString<64> removeRoot(StringRef Path, bool IsRight) const {
+    SmallString<64> Native(Path);
+    SmallString<64> Root =
+        IsRight ? opts::diff::RightRoot : opts::diff::LeftRoot;
+    // pdb paths always use windows syntax, convert slashes to backslashes.
+    sys::path::native(Root, sys::path::Style::windows);
+    if (sys::path::has_stem(Root, sys::path::Style::windows))
+      sys::path::append(Root,
+                        sys::path::get_separator(sys::path::Style::windows));
+
+    sys::path::replace_path_prefix(Native, Root, "");
+    return Native;
+  }
+  uint32_t MaxLen;
+};
+
+// Compare and format two stream purposes.  For general streams, this just
+// compares the description.  For module streams it uses the path comparison
+// algorithm taking into consideration the binary root, described above.
+// Formatting stream purposes just prints the stream purpose, except for
+// module streams and named streams, where it prefixes the name / module
+// with an identifier.  Example:
+//
+//   Named Stream "\names"
+//   Module Stream "foo.obj"
+//
+// If a named stream is too long to fit in a column, it is truncated at the
+// end, and if a module is too long to fit in a column, it is truncated at the
+// beginning.  Example:
+//
+//  Named Stream "\Really Long Str..."
+//  Module Stream "...puts\foo.obj"
+//
+struct StreamPurposeProvider {
+  explicit StreamPurposeProvider(uint32_t MaxLen) : MaxLen(MaxLen) {}
+
+  DiffResult compare(const std::pair<StreamPurpose, std::string> &L,
+                     const std::pair<StreamPurpose, std::string> &R) {
+    if (L.first != R.first)
+      return DiffResult::DIFFERENT;
+    if (L.first == StreamPurpose::ModuleStream) {
+      BinaryPathProvider PathProvider(MaxLen);
+      return PathProvider.compare(L.second, R.second);
+    }
+    return (L.second == R.second) ? DiffResult::IDENTICAL
+                                  : DiffResult::DIFFERENT;
+  }
+
+  std::string format(const std::pair<StreamPurpose, std::string> &P,
+                     bool Right) {
+    if (P.first == StreamPurpose::Other)
+      return truncateStringBack(P.second, MaxLen);
+    if (P.first == StreamPurpose::NamedStream)
+      return truncateQuotedNameBack("Named Stream", P.second, MaxLen);
+
+    assert(P.first == StreamPurpose::ModuleStream);
+    uint32_t ExtraChars = strlen("Module \"\"");
+    BinaryPathProvider PathProvider(MaxLen - ExtraChars);
+    std::string Result = PathProvider.format(P.second, Right);
+    return formatv("Module \"{0}\"", Result);
+  }
+
+  uint32_t MaxLen;
+};
+}
+
 namespace llvm {
 template <> struct format_provider<PdbRaw_FeatureSig> {
   static void format(const PdbRaw_FeatureSig &Sig, raw_ostream &Stream,
@@ -100,19 +235,12 @@ Error DiffStyle::dump() {
   return Error::success();
 }
 
-static std::string shortFilePath(StringRef Path, uint32_t Width) {
-  if (Path.size() <= Width)
-    return Path;
-  Path = Path.take_back(Width - 3);
-  return std::string("...") + Path.str();
-}
-
 Error DiffStyle::diffSuperBlock() {
   DiffPrinter D(2, "MSF Super Block", 16, 20, opts::diff::PrintResultColumn,
                 opts::diff::PrintValueColumns, outs());
   D.printExplicit("File", DiffResult::UNSPECIFIED,
-                  shortFilePath(File1.getFilePath(), 18),
-                  shortFilePath(File2.getFilePath(), 18));
+                  truncateStringFront(File1.getFilePath(), 18),
+                  truncateStringFront(File2.getFilePath(), 18));
   D.print("Block Size", File1.getBlockSize(), File2.getBlockSize());
   D.print("Block Count", File1.getBlockCount(), File2.getBlockCount());
   D.print("Unknown 1", File1.getUnknown1(), File2.getUnknown1());
@@ -125,13 +253,13 @@ Error DiffStyle::diffStreamDirectory() {
   DiffPrinter D(2, "Stream Directory", 30, 20, opts::diff::PrintResultColumn,
                 opts::diff::PrintValueColumns, outs());
   D.printExplicit("File", DiffResult::UNSPECIFIED,
-                  shortFilePath(File1.getFilePath(), 18),
-                  shortFilePath(File2.getFilePath(), 18));
+                  truncateStringFront(File1.getFilePath(), 18),
+                  truncateStringFront(File2.getFilePath(), 18));
 
-  SmallVector<std::string, 32> P;
-  SmallVector<std::string, 32> Q;
-  discoverStreamPurposes(File1, P, 28);
-  discoverStreamPurposes(File2, Q, 28);
+  SmallVector<std::pair<StreamPurpose, std::string>, 32> P;
+  SmallVector<std::pair<StreamPurpose, std::string>, 32> Q;
+  discoverStreamPurposes(File1, P);
+  discoverStreamPurposes(File2, Q);
   D.print("Stream Count", File1.getNumStreams(), File2.getNumStreams());
   auto PI = to_vector<32>(enumerate(P));
   auto QI = to_vector<32>(enumerate(Q));
@@ -139,26 +267,31 @@ Error DiffStyle::diffStreamDirectory() {
   // Scan all streams in the left hand side, looking for ones that are also
   // in the right.  Each time we find one, remove it.  When we're done, Q
   // should contain all the streams that are in the right but not in the left.
+  StreamPurposeProvider StreamProvider(28);
   for (const auto &P : PI) {
     typedef decltype(PI) ContainerType;
     typedef typename ContainerType::value_type value_type;
 
-    auto Iter = llvm::find_if(
-        QI, [P](const value_type &V) { return V.value() == P.value(); });
+    auto Iter = llvm::find_if(QI, [P, &StreamProvider](const value_type &V) {
+      DiffResult Result = StreamProvider.compare(P.value(), V.value());
+      return Result == DiffResult::EQUIVALENT ||
+             Result == DiffResult::IDENTICAL;
+    });
 
     if (Iter == QI.end()) {
-      D.printExplicit(P.value(), DiffResult::DIFFERENT, P.index(),
-                      "(not present)");
+      D.printExplicit(StreamProvider.format(P.value(), false),
+                      DiffResult::DIFFERENT, P.index(), "(not present)");
       continue;
     }
 
-    D.print<EquivalentDiffProvider>(P.value(), P.index(), Iter->index());
+    D.print<EquivalentDiffProvider>(StreamProvider.format(P.value(), false),
+                                    P.index(), Iter->index());
     QI.erase(Iter);
   }
 
   for (const auto &Q : QI) {
-    D.printExplicit(Q.value(), DiffResult::DIFFERENT, "(not present)",
-                    Q.index());
+    D.printExplicit(StreamProvider.format(Q.value(), true),
+                    DiffResult::DIFFERENT, "(not present)", Q.index());
   }
 
   return Error::success();
@@ -168,8 +301,8 @@ Error DiffStyle::diffStringTable() {
   DiffPrinter D(2, "String Table", 30, 20, opts::diff::PrintResultColumn,
                 opts::diff::PrintValueColumns, outs());
   D.printExplicit("File", DiffResult::UNSPECIFIED,
-                  shortFilePath(File1.getFilePath(), 18),
-                  shortFilePath(File2.getFilePath(), 18));
+                  truncateStringFront(File1.getFilePath(), 18),
+                  truncateStringFront(File2.getFilePath(), 18));
 
   auto ExpectedST1 = File1.getStringTable();
   auto ExpectedST2 = File2.getStringTable();
@@ -257,8 +390,8 @@ Error DiffStyle::diffInfoStream() {
   DiffPrinter D(2, "PDB Stream", 22, 40, opts::diff::PrintResultColumn,
                 opts::diff::PrintValueColumns, outs());
   D.printExplicit("File", DiffResult::UNSPECIFIED,
-                  shortFilePath(File1.getFilePath(), 38),
-                  shortFilePath(File2.getFilePath(), 38));
+                  truncateStringFront(File1.getFilePath(), 38),
+                  truncateStringFront(File2.getFilePath(), 38));
 
   auto ExpectedInfo1 = File1.getPDBInfoStream();
   auto ExpectedInfo2 = File2.getPDBInfoStream();
@@ -292,53 +425,6 @@ Error DiffStyle::diffInfoStream() {
   return Error::success();
 }
 
-struct StreamNumberProvider {
-  static DiffResult compare(uint16_t L, uint16_t R) {
-    if (L == R)
-      return DiffResult::IDENTICAL;
-    bool LP = L != kInvalidStreamIndex;
-    bool RP = R != kInvalidStreamIndex;
-    if (LP != RP)
-      return DiffResult::DIFFERENT;
-    return DiffResult::EQUIVALENT;
-  }
-
-  static std::string format(uint16_t SN) {
-    if (SN == kInvalidStreamIndex)
-      return "(not present)";
-    return formatv("{0}", SN).str();
-  }
-};
-
-struct ModiProvider {
-  DiffResult compare(Optional<uint32_t> L, Optional<uint32_t> R) {
-    if (L == R)
-      return DiffResult::IDENTICAL;
-    if (L.hasValue() != R.hasValue())
-      return DiffResult::DIFFERENT;
-    return DiffResult::EQUIVALENT;
-  }
-
-  std::string format(Optional<uint32_t> Modi) {
-    if (!Modi.hasValue())
-      return "(not present)";
-    return formatv("{0}", *Modi).str();
-  }
-};
-
-struct StringProvider {
-  DiffResult compare(StringRef L, StringRef R) {
-    IdenticalDiffProvider I;
-    return I.compare(L, R);
-  }
-
-  std::string format(StringRef S) {
-    if (S.empty())
-      return "(empty)";
-    return S;
-  }
-};
-
 static std::vector<std::pair<uint32_t, DbiModuleDescriptor>>
 getModuleDescriptors(const DbiModuleList &ML) {
   std::vector<std::pair<uint32_t, DbiModuleDescriptor>> List;
@@ -348,16 +434,31 @@ getModuleDescriptors(const DbiModuleList &ML) {
   return List;
 }
 
-static void diffOneModule(
-    DiffPrinter &D, const std::pair<uint32_t, DbiModuleDescriptor> Item,
-    std::vector<std::pair<uint32_t, DbiModuleDescriptor>> &Other, bool Invert) {
-  D.printFullRow(
-      truncateQuotedNameFront("Module", Item.second.getModuleName(), 70));
+static void
+diffOneModule(DiffPrinter &D,
+              const std::pair<uint32_t, DbiModuleDescriptor> Item,
+              std::vector<std::pair<uint32_t, DbiModuleDescriptor>> &Other,
+              bool ItemIsRight) {
+  StreamPurposeProvider HeaderProvider(70);
+  std::pair<StreamPurpose, std::string> Header;
+  Header.first = StreamPurpose::ModuleStream;
+  Header.second = Item.second.getModuleName();
+  D.printFullRow(HeaderProvider.format(Header, ItemIsRight));
 
+  const auto *L = &Item;
+
+  BinaryPathProvider PathProvider(28);
   auto Iter = llvm::find_if(
-      Other, [&Item](const std::pair<uint32_t, DbiModuleDescriptor> &Other) {
-        return Other.second.getModuleName().equals_lower(
-            Item.second.getModuleName());
+      Other, [&Item, &PathProvider, ItemIsRight,
+              L](const std::pair<uint32_t, DbiModuleDescriptor> &Other) {
+        const auto *Left = L;
+        const auto *Right = &Other;
+        if (ItemIsRight)
+          std::swap(Left, Right);
+        DiffResult Result = PathProvider.compare(Left->second.getModuleName(),
+                                                 Right->second.getModuleName());
+        return Result == DiffResult::EQUIVALENT ||
+               Result == DiffResult::IDENTICAL;
       });
   if (Iter == Other.end()) {
     // We didn't find this module at all on the other side.  Just print one row
@@ -367,15 +468,13 @@ static void diffOneModule(
   }
 
   // We did find this module.  Go through and compare each field.
-  const auto *L = &Item;
   const auto *R = &*Iter;
-  if (Invert)
+  if (ItemIsRight)
     std::swap(L, R);
 
   D.print<ModiProvider>("- Modi", L->first, R->first);
-  D.print<StringProvider>("- Obj File Name",
-                          shortFilePath(L->second.getObjFileName(), 28),
-                          shortFilePath(R->second.getObjFileName(), 28));
+  D.print<BinaryPathProvider>("- Obj File Name", L->second.getObjFileName(),
+                              R->second.getObjFileName(), PathProvider);
   D.print<StreamNumberProvider>("- Debug Stream",
                                 L->second.getModuleStreamIndex(),
                                 R->second.getModuleStreamIndex());
@@ -398,8 +497,8 @@ Error DiffStyle::diffDbiStream() {
   DiffPrinter D(2, "DBI Stream", 40, 30, opts::diff::PrintResultColumn,
                 opts::diff::PrintValueColumns, outs());
   D.printExplicit("File", DiffResult::UNSPECIFIED,
-                  shortFilePath(File1.getFilePath(), 28),
-                  shortFilePath(File2.getFilePath(), 28));
+                  truncateStringFront(File1.getFilePath(), 28),
+                  truncateStringFront(File2.getFilePath(), 28));
 
   auto ExpectedDbi1 = File1.getPDBDbiStream();
   auto ExpectedDbi2 = File2.getPDBDbiStream();
diff --git a/tools/llvm-pdbutil/DiffPrinter.cpp b/tools/llvm-pdbutil/DiffPrinter.cpp
index 45641e2e4b08..dd61cc182593 100644
--- a/tools/llvm-pdbutil/DiffPrinter.cpp
+++ b/tools/llvm-pdbutil/DiffPrinter.cpp
@@ -6,18 +6,31 @@
 using namespace llvm;
 using namespace llvm::pdb;
 
-static void setColor(llvm::raw_ostream &OS, DiffResult Result) {
-  switch (Result) {
-  case DiffResult::IDENTICAL:
-    OS.changeColor(raw_ostream::Colors::GREEN, false);
-    break;
-  case DiffResult::EQUIVALENT:
-    OS.changeColor(raw_ostream::Colors::YELLOW, true);
-    break;
-  default:
-    OS.changeColor(raw_ostream::Colors::RED, false);
-    break;
+namespace {
+struct Colorize {
+  Colorize(raw_ostream &OS, DiffResult Result) : OS(OS) {
+    if (!OS.has_colors())
+      return;
+    switch (Result) {
+    case DiffResult::IDENTICAL:
+      OS.changeColor(raw_ostream::Colors::GREEN, false);
+      break;
+    case DiffResult::EQUIVALENT:
+      OS.changeColor(raw_ostream::Colors::YELLOW, true);
+      break;
+    default:
+      OS.changeColor(raw_ostream::Colors::RED, false);
+      break;
+    }
+  }
+
+  ~Colorize() {
+    if (OS.has_colors())
+      OS.resetColor();
   }
+
+  raw_ostream &OS;
+};
 }
 
 DiffPrinter::DiffPrinter(uint32_t Indent, StringRef Header,
@@ -124,9 +137,8 @@ void DiffPrinter::printValue(StringRef Value, DiffResult C, AlignStyle Style,
   std::string FormattedItem =
       formatv("{0}", fmt_align(Value, Style, Width)).str();
   if (C != DiffResult::UNSPECIFIED) {
-    setColor(OS, C);
+    Colorize Color(OS, C);
     OS << FormattedItem;
-    OS.resetColor();
   } else
     OS << FormattedItem;
   if (Style == AlignStyle::Right)
diff --git a/tools/llvm-pdbutil/DiffPrinter.h b/tools/llvm-pdbutil/DiffPrinter.h
index eeda92b072c0..475747d8dc11 100644
--- a/tools/llvm-pdbutil/DiffPrinter.h
+++ b/tools/llvm-pdbutil/DiffPrinter.h
@@ -43,7 +43,7 @@ struct IdenticalDiffProvider {
     return (Left == Right) ? DiffResult::IDENTICAL : DiffResult::DIFFERENT;
   }
 
-  template <typename T> std::string format(const T &Item) {
+  template <typename T> std::string format(const T &Item, bool Right) {
     return formatv("{0}", Item).str();
   }
 };
@@ -54,7 +54,7 @@ struct EquivalentDiffProvider {
     return (Left == Right) ? DiffResult::IDENTICAL : DiffResult::EQUIVALENT;
   }
 
-  template <typename T> std::string format(const T &Item) {
+  template <typename T> std::string format(const T &Item, bool Right) {
     return formatv("{0}", Item).str();
   }
 };
@@ -71,8 +71,8 @@ public:
   template <typename Provider = IdenticalDiffProvider, typename T, typename U>
   void print(StringRef Property, const T &Left, const U &Right,
              Provider P = Provider()) {
-    std::string L = P.format(Left);
-    std::string R = P.format(Right);
+    std::string L = P.format(Left, false);
+    std::string R = P.format(Right, true);
 
     DiffResult Result = P.compare(Left, Right);
     printExplicit(Property, Result, L, R);
diff --git a/tools/llvm-pdbutil/StreamUtil.cpp b/tools/llvm-pdbutil/StreamUtil.cpp
index e5778170ef88..4d352004dec3 100644
--- a/tools/llvm-pdbutil/StreamUtil.cpp
+++ b/tools/llvm-pdbutil/StreamUtil.cpp
@@ -22,10 +22,9 @@
 using namespace llvm;
 using namespace llvm::pdb;
 
-void llvm::pdb::discoverStreamPurposes(PDBFile &File,
-                                       SmallVectorImpl<std::string> &Purposes,
-                                       uint32_t MaxLen) {
-
+void llvm::pdb::discoverStreamPurposes(
+    PDBFile &File,
+    SmallVectorImpl<std::pair<StreamPurpose, std::string>> &Purposes) {
   // It's OK if we fail to load some of these streams, we still attempt to print
   // what we can.
   auto Dbi = File.getPDBDbiStream();
@@ -55,71 +54,72 @@ void llvm::pdb::discoverStreamPurposes(PDBFile &File,
 
   Purposes.resize(StreamCount);
   for (uint16_t StreamIdx = 0; StreamIdx < StreamCount; ++StreamIdx) {
-    std::string Value;
+    std::pair<StreamPurpose, std::string> Value;
     if (StreamIdx == OldMSFDirectory)
-      Value = truncateStringBack("Old MSF Directory", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Old MSF Directory");
     else if (StreamIdx == StreamPDB)
-      Value = truncateStringBack("PDB Stream", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "PDB Stream");
     else if (StreamIdx == StreamDBI)
-      Value = truncateStringBack("DBI Stream", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "DBI Stream");
     else if (StreamIdx == StreamTPI)
-      Value = truncateStringBack("TPI Stream", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "TPI Stream");
     else if (StreamIdx == StreamIPI)
-      Value = truncateStringBack("IPI Stream", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "IPI Stream");
     else if (Dbi && StreamIdx == Dbi->getGlobalSymbolStreamIndex())
-      Value = truncateStringBack("Global Symbol Hash", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Global Symbol Hash");
     else if (Dbi && StreamIdx == Dbi->getPublicSymbolStreamIndex())
-      Value = truncateStringBack("Public Symbol Hash", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Public Symbol Hash");
     else if (Dbi && StreamIdx == Dbi->getSymRecordStreamIndex())
-      Value = truncateStringBack("Public Symbol Records", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Public Symbol Records");
     else if (Tpi && StreamIdx == Tpi->getTypeHashStreamIndex())
-      Value = truncateStringBack("TPI Hash", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "TPI Hash");
     else if (Tpi && StreamIdx == Tpi->getTypeHashStreamAuxIndex())
-      Value = truncateStringBack("TPI Aux Hash", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "TPI Aux Hash");
     else if (Ipi && StreamIdx == Ipi->getTypeHashStreamIndex())
-      Value = truncateStringBack("IPI Hash", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "IPI Hash");
     else if (Ipi && StreamIdx == Ipi->getTypeHashStreamAuxIndex())
-      Value = truncateStringBack("IPI Aux Hash", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "IPI Aux Hash");
     else if (Dbi &&
              StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::Exception))
-      Value = truncateStringBack("Exception Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Exception Data");
     else if (Dbi && StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::Fixup))
-      Value = truncateStringBack("Fixup Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Fixup Data");
     else if (Dbi && StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::FPO))
-      Value = truncateStringBack("FPO Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "FPO Data");
     else if (Dbi &&
              StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::NewFPO))
-      Value = truncateStringBack("New FPO Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "New FPO Data");
     else if (Dbi &&
              StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::OmapFromSrc))
-      Value = truncateStringBack("Omap From Source Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Omap From Source Data");
     else if (Dbi &&
              StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::OmapToSrc))
-      Value = truncateStringBack("Omap To Source Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Omap To Source Data");
     else if (Dbi && StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::Pdata))
-      Value = truncateStringBack("Pdata", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Pdata");
     else if (Dbi &&
              StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::SectionHdr))
-      Value = truncateStringBack("Section Header Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Section Header Data");
     else if (Dbi &&
              StreamIdx ==
                  Dbi->getDebugStreamIndex(DbgHeaderType::SectionHdrOrig))
-      Value = truncateStringBack("Section Header Original Data", MaxLen);
+      Value =
+          std::make_pair(StreamPurpose::Other, "Section Header Original Data");
     else if (Dbi &&
              StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::TokenRidMap))
-      Value = truncateStringBack("Token Rid Data", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Token Rid Data");
     else if (Dbi && StreamIdx == Dbi->getDebugStreamIndex(DbgHeaderType::Xdata))
-      Value = truncateStringBack("Xdata", MaxLen);
+      Value = std::make_pair(StreamPurpose::Other, "Xdata");
     else {
       auto ModIter = ModStreams.find(StreamIdx);
       auto NSIter = NamedStreams.find(StreamIdx);
       if (ModIter != ModStreams.end()) {
-        Value = truncateQuotedNameFront(
-            "Module", ModIter->second.getModuleName(), MaxLen);
+        Value = std::make_pair(StreamPurpose::ModuleStream,
+                               ModIter->second.getModuleName());
       } else if (NSIter != NamedStreams.end()) {
-        Value = truncateQuotedNameBack("Named Stream", NSIter->second, MaxLen);
+        Value = std::make_pair(StreamPurpose::NamedStream, NSIter->second);
       } else {
-        Value = "???";
+        Value = std::make_pair(StreamPurpose::Other, "???");
       }
     }
     Purposes[StreamIdx] = Value;
@@ -135,3 +135,18 @@ void llvm::pdb::discoverStreamPurposes(PDBFile &File,
   if (!Info)
     consumeError(Info.takeError());
 }
+
+void llvm::pdb::discoverStreamPurposes(PDBFile &File,
+                                       SmallVectorImpl<std::string> &Purposes) {
+  SmallVector<std::pair<StreamPurpose, std::string>, 24> SP;
+  discoverStreamPurposes(File, SP);
+  Purposes.reserve(SP.size());
+  for (const auto &P : SP) {
+    if (P.first == StreamPurpose::NamedStream)
+      Purposes.push_back(formatv("Named Stream \"{0}\"", P.second));
+    else if (P.first == StreamPurpose::ModuleStream)
+      Purposes.push_back(formatv("Module \"{0}\"", P.second));
+    else
+      Purposes.push_back(P.second);
+  }
+}
diff --git a/tools/llvm-pdbutil/StreamUtil.h b/tools/llvm-pdbutil/StreamUtil.h
index cc143f95f6dd..f49c0a0eceb6 100644
--- a/tools/llvm-pdbutil/StreamUtil.h
+++ b/tools/llvm-pdbutil/StreamUtil.h
@@ -17,9 +17,13 @@
 namespace llvm {
 namespace pdb {
 class PDBFile;
+enum class StreamPurpose { NamedStream, ModuleStream, Other };
+
 void discoverStreamPurposes(PDBFile &File,
-                            SmallVectorImpl<std::string> &Purposes,
-                            uint32_t MaxLen = 0);
+                            SmallVectorImpl<std::string> &Purposes);
+void discoverStreamPurposes(
+    PDBFile &File,
+    SmallVectorImpl<std::pair<StreamPurpose, std::string>> &Purposes);
 }
 }
 
diff --git a/tools/llvm-pdbutil/llvm-pdbutil.cpp b/tools/llvm-pdbutil/llvm-pdbutil.cpp
index 6432b9ce86c0..6aa08ff3cd87 100644
--- a/tools/llvm-pdbutil/llvm-pdbutil.cpp
+++ b/tools/llvm-pdbutil/llvm-pdbutil.cpp
@@ -293,9 +293,23 @@ cl::opt<bool>
                       cl::desc("Print a column with the result status"),
                       cl::Optional, cl::sub(DiffSubcommand));
 
-cl::list<std::string> InputFilenames(cl::Positional,
-                                     cl::desc("<first> <second>"),
-                                     cl::OneOrMore, cl::sub(DiffSubcommand));
+cl::opt<std::string> LeftRoot(
+    "left-bin-root", cl::Optional,
+    cl::desc("Treats the specified path as the root of the tree containing "
+             "binaries referenced by the left PDB.  The root is stripped from "
+             "embedded paths when doing equality comparisons."),
+    cl::sub(DiffSubcommand));
+cl::opt<std::string> RightRoot(
+    "right-bin-root", cl::Optional,
+    cl::desc("Treats the specified path as the root of the tree containing "
+             "binaries referenced by the right PDB.  The root is stripped from "
+             "embedded paths when doing equality comparisons"),
+    cl::sub(DiffSubcommand));
+
+cl::opt<std::string> Left(cl::Positional, cl::desc("<left>"),
+                          cl::sub(DiffSubcommand));
+cl::opt<std::string> Right(cl::Positional, cl::desc("<right>"),
+                           cl::sub(DiffSubcommand));
 }
 
 cl::OptionCategory FileOptions("Module & File Options");
@@ -1151,11 +1165,7 @@ int main(int argc_, const char *argv_[]) {
     std::for_each(opts::bytes::InputFilenames.begin(),
                   opts::bytes::InputFilenames.end(), dumpBytes);
   } else if (opts::DiffSubcommand) {
-    if (opts::diff::InputFilenames.size() != 2) {
-      errs() << "diff subcommand expects exactly 2 arguments.\n";
-      exit(1);
-    }
-    diff(opts::diff::InputFilenames[0], opts::diff::InputFilenames[1]);
+    diff(opts::diff::Left, opts::diff::Right);
   } else if (opts::MergeSubcommand) {
     if (opts::merge::InputFilenames.size() < 2) {
       errs() << "merge subcommand requires at least 2 input files.\n";
diff --git a/tools/llvm-pdbutil/llvm-pdbutil.h b/tools/llvm-pdbutil/llvm-pdbutil.h
index b8e1cf4ed15c..4e92e639a127 100644
--- a/tools/llvm-pdbutil/llvm-pdbutil.h
+++ b/tools/llvm-pdbutil/llvm-pdbutil.h
@@ -172,6 +172,8 @@ extern llvm::cl::opt<bool> DumpModuleSyms;
 namespace diff {
 extern llvm::cl::opt<bool> PrintValueColumns;
 extern llvm::cl::opt<bool> PrintResultColumn;
+extern llvm::cl::opt<std::string> LeftRoot;
+extern llvm::cl::opt<std::string> RightRoot;
 } // namespace diff
 }