Rename IdentifierTranslation to IdentifierHashes

Summary:
The field "identifierTranslations" is used to store the hashes
of identifiers. It was renamed from "hashes" to "translations" in
D14457157 and now we basically need to rename it back, including
related fields, variables and comments.

Reviewed By: tmikov

Differential Revision: D21336820

fbshipit-source-id: fea83d649d08093daca4c9ded41def5f01481078

doc/Strings.md

@@ -20,7 +20,7 @@ The distinction between these is important during bytecode initialisation, parti
 In the Hermes bytecode format, the string table is split into the following consecutive sections of the file:
 
  1. String Kinds.  A sequence which describes the kinds of the strings in the table (as described in "Hermes Bytecode String Kinds").  Abstractly, the i-th element in the sequence is the kind of the i-th string in the table.  Represented as a run-length encoding.
- 2. Identifier Translations.  This sequence has an element corresponding to each identifier in the table (This includes Predefined strings).  The i-th element of this sequence corresponds to the i-th identifier in the table.  If the string is Predefined, the element contains the index in the Identifier Table where the that string is expected to be.  If the string is an Identifier but not Predefined, the element contains a hash of the identifier's string representation.
+ 2. Identifier Hashes.  This sequence has an element corresponding to each identifier in the table.  The i-th element of this sequence corresponds to the i-th identifier in the table.  If the string is an Identifier, the element contains a hash of the identifier's string representation.
  3. Small String Table.  The primary index into the string table.  When a bytecode instruction refers to a string by its index, it uses the offset into this data structure.  Represents the string with three pieces of information:
   - Whether it is UTF16 or not.
   - Its offset into the character storage.

include/hermes/BCGen/HBC/Bytecode.h

@@ -187,9 +187,9 @@ class BytecodeModule {
   /// Run-length encoding representing the kinds of strings in the table.
   std::vector<StringKind::Entry> stringKinds_;
 
-  /// The list of identifier translations, corresponding to the string entries
+  /// The list of identifier hashes, corresponding to the string entries
   /// marked as identifiers, in order.
-  std::vector<uint32_t> identifierTranslations_;
+  std::vector<uint32_t> identifierHashes_;
 
   /// The global string table, a list of <offset, length> pair to represent
   /// each string in the string storage.
@@ -251,7 +251,7 @@ class BytecodeModule {
   explicit BytecodeModule(
       uint32_t functionCount,
       std::vector<StringKind::Entry> &&stringKinds,
-      std::vector<uint32_t> &&identifierTranslations,
+      std::vector<uint32_t> &&identifierHashes,
       std::vector<StringTableEntry> &&stringTable,
       std::vector<unsigned char> &&stringStorage,
       std::vector<RegExpTableEntry> &&regExpTable,
@@ -266,7 +266,7 @@ class BytecodeModule {
       BytecodeOptions options)
       : globalFunctionIndex_(globalFunctionIndex),
         stringKinds_(std::move(stringKinds)),
-        identifierTranslations_(std::move(identifierTranslations)),
+        identifierHashes_(std::move(identifierHashes)),
         stringTable_(std::move(stringTable)),
         stringStorage_(std::move(stringStorage)),
         regExpStorage_(std::move(regExpStorage)),
@@ -312,11 +312,11 @@ class BytecodeModule {
 
   /// \return the number of identifiers.
   uint32_t getIdentifierCount() const {
-    return identifierTranslations_.size();
+    return identifierHashes_.size();
   }
 
-  llvm::ArrayRef<uint32_t> getIdentifierTranslations() const {
-    return identifierTranslations_;
+  llvm::ArrayRef<uint32_t> getIdentifierHashes() const {
+    return identifierHashes_;
   }
 
   uint32_t getStringTableSize() const {

include/hermes/BCGen/HBC/BytecodeDataProvider.h

@@ -116,7 +116,7 @@ class BCProviderBase {
   /// Below are all the different global data tables/storage.
   uint32_t stringCount_{};
   llvm::ArrayRef<StringKind::Entry> stringKinds_{};
-  llvm::ArrayRef<uint32_t> identifierTranslations_{};
+  llvm::ArrayRef<uint32_t> identifierHashes_{};
   llvm::ArrayRef<unsigned char> stringStorage_{};
 
   llvm::ArrayRef<unsigned char> arrayBuffer_{};
@@ -166,8 +166,8 @@ class BCProviderBase {
   llvm::ArrayRef<StringKind::Entry> getStringKinds() const {
     return stringKinds_;
   }
-  llvm::ArrayRef<uint32_t> getIdentifierTranslations() const {
-    return identifierTranslations_;
+  llvm::ArrayRef<uint32_t> getIdentifierHashes() const {
+    return identifierHashes_;
   }
   llvm::ArrayRef<unsigned char> getStringStorage() const {
     return stringStorage_;
@@ -274,9 +274,9 @@ class BCProviderBase {
   /// soon.  Only forwards this information to the OS for buffers.
   virtual void willNeedStringTable() {}
 
-  /// Advise the provider that identifier translations are no longer needed.
+  /// Advise the provider that identifier hashes are no longer needed.
   /// Only forwards this information to the OS for buffers.
-  virtual void dontNeedIdentifierTranslations() {}
+  virtual void dontNeedIdentifierHashes() {}
 
   /// Start tracking I/O (only implemented for buffers). Any access before this
   /// call (e.g. reading header to construct the provider) will not be recorded.
@@ -484,7 +484,7 @@ class BCProviderFromBuffer final : public BCProviderBase {
   void adviseStringTableSequential() override;
   void adviseStringTableRandom() override;
   void willNeedStringTable() override;
-  void dontNeedIdentifierTranslations() override;
+  void dontNeedIdentifierHashes() override;
 
   void startPageAccessTracker() override;
 

include/hermes/BCGen/HBC/BytecodeFileFormat.h

@@ -380,7 +380,7 @@ template <typename Visitor>
 void visitBytecodeSegmentsInOrder(Visitor &visitor) {
   visitor.visitFunctionHeaders();
   visitor.visitStringKinds();
-  visitor.visitIdentifierTranslations();
+  visitor.visitIdentifierHashes();
   visitor.visitSmallStringTable();
   visitor.visitOverflowStringTable();
   visitor.visitStringStorage();
@@ -420,8 +420,8 @@ struct BytecodeFileFields {
   /// Run-length encoding representing the kinds of strings in the table.
   Array<StringKind::Entry> stringKinds{};
 
-  /// The list of identifier translations.
-  Array<uint32_t> identifierTranslations{};
+  /// The list of identifier hashes.
+  Array<uint32_t> identifierHashes{};
 
   /// The list of overflowed string table entries.
   Array<hbc::OverflowStringTableEntry> stringTableOverflowEntries{};

include/hermes/BCGen/HBC/BytecodeStream.h

@@ -105,7 +105,7 @@ class BytecodeSerializer {
 
   void visitFunctionHeaders();
   void visitStringKinds();
-  void visitIdentifierTranslations();
+  void visitIdentifierHashes();
   void visitSmallStringTable();
   void visitOverflowStringTable();
   void visitStringStorage();

include/hermes/BCGen/HBC/UniquingStringLiteralTable.h

@@ -64,9 +64,9 @@ struct StringLiteralTable final : public StringLiteralIDMapping {
   inline std::vector<StringTableEntry> acquireStringTable();
   inline std::vector<unsigned char> acquireStringStorage();
 
-  /// \returns a list of translations corresponding to the strings marked as
+  /// \returns a list of hashes corresponding to the strings marked as
   /// identifiers, in their order in the underlying storage.
-  std::vector<uint32_t> getIdentifierTranslations() const;
+  std::vector<uint32_t> getIdentifierHashes() const;
 
   /// \return a sequence of string kinds represented by a run-length encoding.
   /// The i'th kind in the abstract sequence (i.e. not the i'th entry in the

lib/BCGen/HBC/BytecodeDataProvider.cpp

@@ -203,10 +203,9 @@ bool BytecodeFileFields<Mutable>::populateFromBuffer(
           castArrayRef<StringKind::Entry>(buf, h->stringKindCount, end);
     }
 
-    void visitIdentifierTranslations() {
+    void visitIdentifierHashes() {
       align(buf);
-      f.identifierTranslations =
-          castArrayRef<uint32_t>(buf, h->identifierCount, end);
+      f.identifierHashes = castArrayRef<uint32_t>(buf, h->identifierCount, end);
     }
 
     void visitSmallStringTable() {
@@ -418,14 +417,14 @@ void BCProviderFromBuffer::adviseStringTableSequential() {
   size_t adviceLength = end - start;
 
   ASSERT_BOUNDED(start, stringKinds_, end);
-  ASSERT_BOUNDED(start, identifierTranslations_, end);
+  ASSERT_BOUNDED(start, identifierHashes_, end);
   ASSERT_BOUNDED(start, smallStringTableEntries, end);
   ASSERT_BOUNDED(start, overflowStringTableEntries_, end);
 
   ASSERT_TOTAL_ARRAY_LEN(
       adviceLength,
       stringKinds_,
-      identifierTranslations_,
+      identifierHashes_,
       smallStringTableEntries,
       overflowStringTableEntries_);
 
@@ -439,7 +438,7 @@ void BCProviderFromBuffer::adviseStringTableRandom() {
 
   // We only advise the small string table entries, overflow string table
   // entries and storage.  We do not give advice about the identifier
-  // translations or string kinds because they are not referred to after
+  // hashes or string kinds because they are not referred to after
   // initialisation.
 
   auto *tableStart = rawptr_cast(stringTableEntries_);
@@ -470,24 +469,24 @@ void BCProviderFromBuffer::willNeedStringTable() {
   size_t prefetchLength = end - start;
 
   ASSERT_BOUNDED(start, stringKinds_, end);
-  ASSERT_BOUNDED(start, identifierTranslations_, end);
+  ASSERT_BOUNDED(start, identifierHashes_, end);
   ASSERT_BOUNDED(start, smallStringTableEntries, end);
   ASSERT_BOUNDED(start, overflowStringTableEntries_, end);
 
   ASSERT_TOTAL_ARRAY_LEN(
       prefetchLength,
       stringKinds_,
-      identifierTranslations_,
+      identifierHashes_,
       smallStringTableEntries,
       overflowStringTableEntries_);
 
   pageAlignDown(&start, &prefetchLength);
   oscompat::vm_prefetch(start, prefetchLength);
 }
 
-void BCProviderFromBuffer::dontNeedIdentifierTranslations() {
-  auto start = reinterpret_cast<uintptr_t>(identifierTranslations_.begin());
-  auto end = reinterpret_cast<uintptr_t>(identifierTranslations_.end());
+void BCProviderFromBuffer::dontNeedIdentifierHashes() {
+  auto start = reinterpret_cast<uintptr_t>(identifierHashes_.begin());
+  auto end = reinterpret_cast<uintptr_t>(identifierHashes_.end());
   const size_t PS = oscompat::page_size();
   start = llvm::alignTo(start, PS);
   end = llvm::alignDown(end, PS);
@@ -525,7 +524,7 @@ BCProviderFromBuffer::BCProviderFromBuffer(
   debugInfoOffset_ = fileHeader->debugInfoOffset;
   functionHeaders_ = fields.functionHeaders.data();
   stringKinds_ = fields.stringKinds;
-  identifierTranslations_ = fields.identifierTranslations;
+  identifierHashes_ = fields.identifierHashes;
   stringCount_ = fileHeader->stringCount;
   stringTableEntries_ = fields.stringTableEntries.data();
   overflowStringTableEntries_ = fields.stringTableOverflowEntries;

lib/BCGen/HBC/BytecodeDisassembler.cpp

@@ -139,18 +139,18 @@ void BytecodeDisassembler::disassembleBytecodeFileHeader(raw_ostream &OS) {
 
 void BytecodeDisassembler::disassembleStringStorage(raw_ostream &OS) {
   auto strStorage = bcProvider_->getStringStorage();
-  auto translations = bcProvider_->getIdentifierTranslations();
+  auto hashes = bcProvider_->getIdentifierHashes();
 
   const auto strCount = bcProvider_->getStringCount();
-  const auto trnCount = translations.size();
+  const auto hashCount = hashes.size();
 
   if (strCount == 0)
     return;
 
   auto kinds = bcProvider_->getStringKinds();
 
   uint32_t strID = 0;
-  uint32_t trnID = 0;
+  uint32_t hashID = 0;
 
   OS << "Global String Table:\n";
   const std::locale loc("C");
@@ -175,7 +175,7 @@ void BytecodeDisassembler::disassembleStringStorage(raw_ostream &OS) {
         case StringKind::Identifier:
           OS << " #"
              << llvm::format_hex_no_prefix(
-                    translations[trnID++], 8, /* Upper */ true);
+                    hashes[hashID++], 8, /* Upper */ true);
           break;
 
         default:
@@ -198,8 +198,8 @@ void BytecodeDisassembler::disassembleStringStorage(raw_ostream &OS) {
 
   assert(strID == strCount && "Visited all strings.");
   (void)strCount;
-  assert(trnID == trnCount && "Visited all translations.");
-  (void)trnCount;
+  assert(hashID == hashCount && "Visited all hashes.");
+  (void)hashCount;
 }
 
 /// NOTE: The output might not show the value of every literal used
@@ -757,9 +757,9 @@ BytecodeSectionWalker::BytecodeSectionWalker(
       bcProvider->getStringKinds().begin(),
       bcProvider->getStringKinds().end());
   addSection(
-      "Identifier translations",
-      bcProvider->getIdentifierTranslations().begin(),
-      bcProvider->getIdentifierTranslations().end());
+      "Identifier hashes",
+      bcProvider->getIdentifierHashes().begin(),
+      bcProvider->getIdentifierHashes().end());
   addSection(
       "String table",
       bcProvider->getSmallStringTableEntries().begin(),

lib/BCGen/HBC/BytecodeGenerator.cpp

@@ -213,7 +213,7 @@ std::unique_ptr<BytecodeModule> BytecodeModuleGenerator::generate() {
       "Missing functions.");
 
   auto kinds = stringTable_.getStringKinds();
-  auto hashes = stringTable_.getIdentifierTranslations();
+  auto hashes = stringTable_.getIdentifierHashes();
 
   BytecodeOptions bytecodeOptions;
   bytecodeOptions.staticBuiltins = options_.staticBuiltinsEnabled;

lib/BCGen/HBC/BytecodeProviderFromSrc.cpp

@@ -61,7 +61,7 @@ BCProviderFromSrc::BCProviderFromSrc(
   globalFunctionIndex_ = module_->getGlobalFunctionIndex();
 
   stringKinds_ = module_->getStringKinds();
-  identifierTranslations_ = module_->getIdentifierTranslations();
+  identifierHashes_ = module_->getIdentifierHashes();
   stringCount_ = module_->getStringTable().size();
   stringStorage_ = module_->getStringStorage();
 

lib/BCGen/HBC/BytecodeStream.cpp

@@ -236,9 +236,9 @@ void BytecodeSerializer::visitStringKinds() {
   writeBinaryArray(bytecodeModule_->getStringKinds());
 }
 
-void BytecodeSerializer::visitIdentifierTranslations() {
+void BytecodeSerializer::visitIdentifierHashes() {
   pad(BYTECODE_ALIGNMENT);
-  writeBinaryArray(bytecodeModule_->getIdentifierTranslations());
+  writeBinaryArray(bytecodeModule_->getIdentifierHashes());
 }
 
 void BytecodeSerializer::visitSmallStringTable() {

lib/BCGen/HBC/UniquingStringLiteralTable.cpp

@@ -41,7 +41,7 @@ StringLiteralIDMapping::StringLiteralIDMapping(
   }
 }
 
-std::vector<uint32_t> StringLiteralTable::getIdentifierTranslations() const {
+std::vector<uint32_t> StringLiteralTable::getIdentifierHashes() const {
   std::vector<uint32_t> result;
   assert(strings_.size() == isIdentifier_.size());
   for (size_t i = 0; i < strings_.size(); ++i) {

lib/VM/RuntimeModule.cpp

@@ -236,21 +236,21 @@ void RuntimeModule::importStringIDMapMayAllocate() {
     bcProvider_->willNeedStringTable();
   }
 
-  // Get the array of pre-computed translations from identifiers in the bytecode
+  // Get the array of pre-computed hashes from identifiers in the bytecode
   // to their runtime representation as SymbolIDs.
   auto kinds = bcProvider_->getStringKinds();
-  auto translations = bcProvider_->getIdentifierTranslations();
+  auto hashes = bcProvider_->getIdentifierHashes();
   assert(
-      translations.size() <= strTableSize &&
+      hashes.size() <= strTableSize &&
       "Should not have more strings than identifiers");
 
   // Preallocate enough space to store all identifiers to prevent
   // unnecessary allocations. NOTE: If this module is not the first module,
   // then this is an underestimate.
-  runtime_->getIdentifierTable().reserve(translations.size());
+  runtime_->getIdentifierTable().reserve(hashes.size());
   {
     StringID strID = 0;
-    uint32_t trnID = 0;
+    uint32_t hashID = 0;
 
     for (auto entry : kinds) {
       switch (entry.kind()) {
@@ -259,18 +259,16 @@ void RuntimeModule::importStringIDMapMayAllocate() {
           break;
 
         case StringKind::Identifier:
-          for (uint32_t i = 0; i < entry.count(); ++i, ++strID, ++trnID) {
+          for (uint32_t i = 0; i < entry.count(); ++i, ++strID, ++hashID) {
             createSymbolFromStringIDMayAllocate(
-                strID,
-                bcProvider_->getStringTableEntry(strID),
-                translations[trnID]);
+                strID, bcProvider_->getStringTableEntry(strID), hashes[hashID]);
           }
           break;
       }
     }
 
     assert(strID == strTableSize && "Should map every string in the bytecode.");
-    assert(trnID == translations.size() && "Should translate all identifiers.");
+    assert(hashID == hashes.size() && "Should hash all identifiers.");
   }
 
   if (runtime_->getVMExperimentFlags() & experiments::MAdviseStringsRandom) {
@@ -292,8 +290,8 @@ void RuntimeModule::importStringIDMapMayAllocate() {
     mapStringMayAllocate(s, 0, hashString(s));
   }
 
-  // Done with translations, so advise them out if possible.
-  bcProvider_->dontNeedIdentifierTranslations();
+  // Done with hashes, so advise them out if possible.
+  bcProvider_->dontNeedIdentifierHashes();
 }
 
 void RuntimeModule::initializeFunctionMap() {

tools/hbc-read-trace/test.sh

@@ -36,7 +36,7 @@ cat <<EOF > "$TMP/expected"
 <unknown>
 function-table
 string-kinds
-identifier-translations
+identifier-hashes
 string-table
 overflow-string-table
 string-storage