diff --git a/mlir/include/mlir/IR/DialectSymbolRegistry.def b/mlir/include/mlir/IR/DialectSymbolRegistry.def
--- a/mlir/include/mlir/IR/DialectSymbolRegistry.def
+++ b/mlir/include/mlir/IR/DialectSymbolRegistry.def
@@ -25,7 +25,6 @@
 DEFINE_SYM_KIND_RANGE(SPIRV) // SPIR-V dialect
 DEFINE_SYM_KIND_RANGE(XLA_HLO) // XLA HLO dialect
 DEFINE_SYM_KIND_RANGE(SHAPE) // Shape dialect
-DEFINE_SYM_KIND_RANGE(TEST) // Test dialect
 
 // The following ranges are reserved for experimenting with MLIR dialects in a
 // private context without having to register them here.
diff --git a/mlir/lib/IR/AttributeDetail.h b/mlir/lib/IR/AttributeDetail.h
--- a/mlir/lib/IR/AttributeDetail.h
+++ b/mlir/lib/IR/AttributeDetail.h
@@ -619,34 +619,37 @@
     // If the data buffer is non-empty, we copy it into the allocator with a
     // 64-bit alignment.
     ArrayRef<StringRef> copy, data = key.data;
-    if (!data.empty()) {
-      int numEntries = key.isSplat ? 1 : data.size();
+    if (data.empty()) {
+      return new (allocator.allocate<DenseStringElementsAttributeStorage>())
+          DenseStringElementsAttributeStorage(key.type, copy, key.isSplat);
+    }
 
-      // Compute the amount data needed to store the ArrayRef and StringRef
-      // contents.
-      size_t dataSize = sizeof(ArrayRef<StringRef>) * numEntries;
-      for (int i = 0; i < numEntries; i++) {
-        dataSize += data[i].size();
-      }
+    int numEntries = key.isSplat ? 1 : data.size();
 
-      char *rawData = reinterpret_cast<char *>(
-          allocator.allocate(dataSize, alignof(uint64_t)));
+    // Compute the amount data needed to store the ArrayRef and StringRef
+    // contents.
+    size_t dataSize = sizeof(ArrayRef<StringRef>) * numEntries;
+    for (int i = 0; i < numEntries; i++) {
+      dataSize += data[i].size();
+    }
 
-      // Setup the ArrayRef
-      auto mutable_copy = MutableArrayRef<StringRef>(
-          reinterpret_cast<StringRef *>(rawData), numEntries);
-      auto stringData = rawData + numEntries * sizeof(StringRef);
+    char *rawData = reinterpret_cast<char *>(
+        allocator.allocate(dataSize, alignof(uint64_t)));
 
-      for (int i = 0; i < numEntries; i++) {
-        memcpy(stringData, data[i].data(), data[i].size());
-        mutable_copy[i] = StringRef(stringData, data[i].size());
-        stringData += data[i].size();
-      }
+    // Setup the ArrayRef
+    auto mutableCopy = MutableArrayRef<StringRef>(
+        reinterpret_cast<StringRef *>(rawData), numEntries);
+    auto stringData = rawData + numEntries * sizeof(StringRef);
 
-      copy = ArrayRef<StringRef>(reinterpret_cast<StringRef *>(rawData),
-                                 numEntries);
+    for (int i = 0; i < numEntries; i++) {
+      memcpy(stringData, data[i].data(), data[i].size());
+      mutableCopy[i] = StringRef(stringData, data[i].size());
+      stringData += data[i].size();
     }
 
+    copy =
+        ArrayRef<StringRef>(reinterpret_cast<StringRef *>(rawData), numEntries);
+
     return new (allocator.allocate<DenseStringElementsAttributeStorage>())
         DenseStringElementsAttributeStorage(key.type, copy, key.isSplat);
   }
diff --git a/mlir/lib/Parser/Parser.cpp b/mlir/lib/Parser/Parser.cpp
--- a/mlir/lib/Parser/Parser.cpp
+++ b/mlir/lib/Parser/Parser.cpp
@@ -2299,9 +2299,8 @@
 Attribute Parser::parseDenseElementsAttr(Type attrType) {
   consumeToken(Token::kw_dense);
 
-  if (parseToken(Token::less, "expected '<' after 'dense'")) {
+  if (parseToken(Token::less, "expected '<' after 'dense'"))
     return nullptr;
-  }
 
   // Parse the literal data.
   TensorLiteralParser literalParser(*this);
diff --git a/mlir/test/IR/attribute.mlir b/mlir/test/IR/attribute.mlir
--- a/mlir/test/IR/attribute.mlir
+++ b/mlir/test/IR/attribute.mlir
@@ -397,7 +397,7 @@
 func @simple_scalar_example() {
   "test.string_elements_attr"() {
     // CHECK: dense<"example">
-    scalar_string_attr = dense<"example"> : tensor<2x!test.custom_type>
+    scalar_string_attr = dense<"example"> : tensor<2x!unknown<"">>
   } : () -> ()
   return
 }
@@ -407,7 +407,7 @@
 func @escape_string_example() {
   "test.string_elements_attr"() {
     // CHECK: dense<"new\0Aline">
-    scalar_string_attr = dense<"new\nline"> : tensor<2x!test.custom_type>
+    scalar_string_attr = dense<"new\nline"> : tensor<2x!unknown<"">>
   } : () -> ()
   return
 }
@@ -417,7 +417,7 @@
 func @simple_scalar_example() {
   "test.string_elements_attr"() {
     // CHECK: dense<["example1", "example2"]>
-    scalar_string_attr = dense<["example1", "example2"]> : tensor<2x!test.custom_type>
+    scalar_string_attr = dense<["example1", "example2"]> : tensor<2x!unknown<"">>
   } : () -> ()
   return
 }
diff --git a/mlir/test/lib/Dialect/Test/TestDialect.h b/mlir/test/lib/Dialect/Test/TestDialect.h
--- a/mlir/test/lib/Dialect/Test/TestDialect.h
+++ b/mlir/test/lib/Dialect/Test/TestDialect.h
@@ -30,38 +30,8 @@
 
 namespace mlir {
 
-namespace TestTypes {
-enum Kind {
-  FIRST_USED_TEST_TYPE = Type::FIRST_TEST_TYPE,
-  CustomTestType,
-  LAST_USED_TEST_TYPE
-};
-} // namespace TestTypes
-
 #include "TestOpsDialect.h.inc"
 
-class TestType : public Type {
-public:
-  using Type::Type;
-
-  static bool classof(Type type) {
-    return type.getKind() >= TestTypes::FIRST_USED_TEST_TYPE &&
-           type.getKind() <= TestTypes::LAST_USED_TEST_TYPE;
-  }
-};
-
-class CustomTestType : public Type::TypeBase<CustomTestType, TestType> {
-public:
-  using Base::Base;
-  static CustomTestType get(MLIRContext *context) {
-    return Base::get(context, TestTypes::CustomTestType);
-  }
-
-  static bool kindof(unsigned kind) {
-    return kind == TestTypes::CustomTestType;
-  }
-};
-
 #define GET_OP_CLASSES
 #include "TestOps.h.inc"
 
diff --git a/mlir/test/lib/Dialect/Test/TestDialect.cpp b/mlir/test/lib/Dialect/Test/TestDialect.cpp
--- a/mlir/test/lib/Dialect/Test/TestDialect.cpp
+++ b/mlir/test/lib/Dialect/Test/TestDialect.cpp
@@ -8,7 +8,6 @@
 
 #include "TestDialect.h"
 #include "mlir/Dialect/StandardOps/IR/Ops.h"
-#include "mlir/IR/DialectImplementation.h"
 #include "mlir/IR/Function.h"
 #include "mlir/IR/Module.h"
 #include "mlir/IR/PatternMatch.h"
@@ -130,8 +129,6 @@
 
 TestDialect::TestDialect(MLIRContext *context)
     : Dialect(getDialectNamespace(), context) {
-  addTypes<CustomTestType>();
-
   addOperations<
 #define GET_OP_LIST
 #include "TestOps.cpp.inc"
@@ -166,27 +163,6 @@
   return success();
 }
 
-Type TestDialect::parseType(DialectAsmParser &parser) const {
-  Location loc = parser.getEncodedSourceLoc(parser.getNameLoc());
-  llvm::StringRef spec = parser.getFullSymbolSpec();
-  if (spec == "custom_type") {
-    return CustomTestType::get(getContext());
-  }
-  emitError(loc, "unknown TestDialect type:") << spec;
-
-  return Type();
-}
-
-void TestDialect::printType(Type type, DialectAsmPrinter &os) const {
-  switch (type.getKind()) {
-  case TestTypes::CustomTestType:
-    os << "custom_type";
-    break;
-  default:
-    llvm_unreachable("unhandle test dialect type");
-  }
-}
-
 //===----------------------------------------------------------------------===//
 // TestBranchOp
 //===----------------------------------------------------------------------===//
diff --git a/mlir/test/lib/Dialect/Test/TestOps.td b/mlir/test/lib/Dialect/Test/TestOps.td
--- a/mlir/test/lib/Dialect/Test/TestOps.td
+++ b/mlir/test/lib/Dialect/Test/TestOps.td
@@ -32,14 +32,6 @@
 // Test Types
 //===----------------------------------------------------------------------===//
 
-def Test_CustomTestType : DialectType<
-    Test_Dialect,
-  CPred<"$_self.isa<CustomTestType>()">, "custom_type"> {
-    let typeDescription = [{
-      Custom type example.
-    }];
-}
-
 def IntTypesOp : TEST_Op<"int_types"> {
   let results = (outs
     AnyI16:$any_i16,