diff --git a/test/cctest/compiler/test-representation-change.cc b/test/cctest/compiler/test-representation-change.cc
new file mode 100644
index 0000000..6c9026b
--- /dev/null
+++ b/test/cctest/compiler/test-representation-change.cc
@@ -0,0 +1,305 @@
+// Copyright 2014 the V8 project authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include <limits>
+
+#include "src/v8.h"
+#include "test/cctest/cctest.h"
+#include "test/cctest/compiler/graph-builder-tester.h"
+
+#include "src/compiler/node-matchers.h"
+#include "src/compiler/representation-change.h"
+#include "src/compiler/typer.h"
+
+using namespace v8::internal;
+using namespace v8::internal::compiler;
+
+namespace v8 {  // for friendiness.
+namespace internal {
+namespace compiler {
+
+class RepresentationChangerTester : public HandleAndZoneScope,
+                                    public GraphAndBuilders {
+ public:
+  explicit RepresentationChangerTester(int num_parameters = 0)
+      : GraphAndBuilders(main_zone()),
+        typer_(main_zone()),
+        javascript_(main_zone()),
+        jsgraph_(main_graph_, &main_common_, &javascript_, &typer_,
+                 &main_machine_),
+        changer_(&jsgraph_, &main_simplified_, main_isolate()) {
+    Node* s = graph()->NewNode(common()->Start(num_parameters));
+    graph()->SetStart(s);
+  }
+
+  Typer typer_;
+  JSOperatorBuilder javascript_;
+  JSGraph jsgraph_;
+  RepresentationChanger changer_;
+
+  Isolate* isolate() { return main_isolate(); }
+  Graph* graph() { return main_graph_; }
+  CommonOperatorBuilder* common() { return &main_common_; }
+  JSGraph* jsgraph() { return &jsgraph_; }
+  RepresentationChanger* changer() { return &changer_; }
+
+  // TODO(titzer): use ValueChecker / ValueUtil
+  void CheckInt32Constant(Node* n, int32_t expected) {
+    Int32Matcher m(n);
+    CHECK(m.HasValue());
+    CHECK_EQ(expected, m.Value());
+  }
+
+  void CheckHeapConstant(Node* n, HeapObject* expected) {
+    HeapObjectMatcher<HeapObject> m(n);
+    CHECK(m.HasValue());
+    CHECK_EQ(expected, *m.Value().handle());
+  }
+
+  void CheckNumberConstant(Node* n, double expected) {
+    NumberMatcher m(n);
+    CHECK_EQ(IrOpcode::kNumberConstant, n->opcode());
+    CHECK(m.HasValue());
+    CHECK_EQ(expected, m.Value());
+  }
+
+  Node* Parameter(int index = 0) {
+    return graph()->NewNode(common()->Parameter(index), graph()->start());
+  }
+
+  void CheckTypeError(MachineTypeUnion from, MachineTypeUnion to) {
+    changer()->testing_type_errors_ = true;
+    changer()->type_error_ = false;
+    Node* n = Parameter(0);
+    Node* c = changer()->GetRepresentationFor(n, from, to);
+    CHECK(changer()->type_error_);
+    CHECK_EQ(n, c);
+  }
+
+  void CheckNop(MachineTypeUnion from, MachineTypeUnion to) {
+    Node* n = Parameter(0);
+    Node* c = changer()->GetRepresentationFor(n, from, to);
+    CHECK_EQ(n, c);
+  }
+};
+}
+}
+}  // namespace v8::internal::compiler
+
+
+// TODO(titzer): add kRepFloat32 when fully supported.
+static const MachineType all_reps[] = {kRepBit, kRepWord32, kRepWord64,
+                                       kRepFloat64, kRepTagged};
+
+
+// TODO(titzer): lift this to ValueHelper
+static const double double_inputs[] = {
+    0.0,   -0.0,    1.0,    -1.0,        0.1,         1.4,    -1.7,
+    2,     5,       6,      982983,      888,         -999.8, 3.1e7,
+    -2e66, 2.3e124, -12e73, V8_INFINITY, -V8_INFINITY};
+
+
+static const int32_t int32_inputs[] = {
+    0,      1,                                -1,
+    2,      5,                                6,
+    982983, 888,                              -999,
+    65535,  static_cast<int32_t>(0xFFFFFFFF), static_cast<int32_t>(0x80000000)};
+
+
+static const uint32_t uint32_inputs[] = {
+    0,      1,   static_cast<uint32_t>(-1),   2,     5,          6,
+    982983, 888, static_cast<uint32_t>(-999), 65535, 0xFFFFFFFF, 0x80000000};
+
+
+TEST(BoolToBit_constant) {
+  RepresentationChangerTester r;
+
+  Node* true_node = r.jsgraph()->TrueConstant();
+  Node* true_bit =
+      r.changer()->GetRepresentationFor(true_node, kRepTagged, kRepBit);
+  r.CheckInt32Constant(true_bit, 1);
+
+  Node* false_node = r.jsgraph()->FalseConstant();
+  Node* false_bit =
+      r.changer()->GetRepresentationFor(false_node, kRepTagged, kRepBit);
+  r.CheckInt32Constant(false_bit, 0);
+}
+
+
+TEST(BitToBool_constant) {
+  RepresentationChangerTester r;
+
+  for (int i = -5; i < 5; i++) {
+    Node* node = r.jsgraph()->Int32Constant(i);
+    Node* val = r.changer()->GetRepresentationFor(node, kRepBit, kRepTagged);
+    r.CheckHeapConstant(val, i == 0 ? r.isolate()->heap()->false_value()
+                                    : r.isolate()->heap()->true_value());
+  }
+}
+
+
+TEST(ToTagged_constant) {
+  RepresentationChangerTester r;
+
+  for (size_t i = 0; i < arraysize(double_inputs); i++) {
+    Node* n = r.jsgraph()->Float64Constant(double_inputs[i]);
+    Node* c = r.changer()->GetRepresentationFor(n, kRepFloat64, kRepTagged);
+    r.CheckNumberConstant(c, double_inputs[i]);
+  }
+
+  for (size_t i = 0; i < arraysize(int32_inputs); i++) {
+    Node* n = r.jsgraph()->Int32Constant(int32_inputs[i]);
+    Node* c = r.changer()->GetRepresentationFor(n, kRepWord32 | kTypeInt32,
+                                                kRepTagged);
+    r.CheckNumberConstant(c, static_cast<double>(int32_inputs[i]));
+  }
+
+  for (size_t i = 0; i < arraysize(uint32_inputs); i++) {
+    Node* n = r.jsgraph()->Int32Constant(uint32_inputs[i]);
+    Node* c = r.changer()->GetRepresentationFor(n, kRepWord32 | kTypeUint32,
+                                                kRepTagged);
+    r.CheckNumberConstant(c, static_cast<double>(uint32_inputs[i]));
+  }
+}
+
+
+static void CheckChange(IrOpcode::Value expected, MachineTypeUnion from,
+                        MachineTypeUnion to) {
+  RepresentationChangerTester r;
+
+  Node* n = r.Parameter();
+  Node* c = r.changer()->GetRepresentationFor(n, from, to);
+
+  CHECK_NE(c, n);
+  CHECK_EQ(expected, c->opcode());
+  CHECK_EQ(n, c->InputAt(0));
+}
+
+
+TEST(SingleChanges) {
+  CheckChange(IrOpcode::kChangeBoolToBit, kRepTagged, kRepBit);
+  CheckChange(IrOpcode::kChangeBitToBool, kRepBit, kRepTagged);
+
+  CheckChange(IrOpcode::kChangeInt32ToTagged, kRepWord32 | kTypeInt32,
+              kRepTagged);
+  CheckChange(IrOpcode::kChangeUint32ToTagged, kRepWord32 | kTypeUint32,
+              kRepTagged);
+  CheckChange(IrOpcode::kChangeFloat64ToTagged, kRepFloat64, kRepTagged);
+
+  CheckChange(IrOpcode::kChangeTaggedToInt32, kRepTagged | kTypeInt32,
+              kRepWord32);
+  CheckChange(IrOpcode::kChangeTaggedToUint32, kRepTagged | kTypeUint32,
+              kRepWord32);
+  CheckChange(IrOpcode::kChangeTaggedToFloat64, kRepTagged, kRepFloat64);
+
+  // Int32,Uint32 <-> Float64 are actually machine conversions.
+  CheckChange(IrOpcode::kChangeInt32ToFloat64, kRepWord32 | kTypeInt32,
+              kRepFloat64);
+  CheckChange(IrOpcode::kChangeUint32ToFloat64, kRepWord32 | kTypeUint32,
+              kRepFloat64);
+  CheckChange(IrOpcode::kChangeFloat64ToInt32, kRepFloat64 | kTypeInt32,
+              kRepWord32);
+  CheckChange(IrOpcode::kChangeFloat64ToUint32, kRepFloat64 | kTypeUint32,
+              kRepWord32);
+}
+
+
+TEST(SignednessInWord32) {
+  RepresentationChangerTester r;
+
+  // TODO(titzer): assume that uses of a word32 without a sign mean kTypeInt32.
+  CheckChange(IrOpcode::kChangeTaggedToInt32, kRepTagged,
+              kRepWord32 | kTypeInt32);
+  CheckChange(IrOpcode::kChangeTaggedToUint32, kRepTagged,
+              kRepWord32 | kTypeUint32);
+  CheckChange(IrOpcode::kChangeInt32ToFloat64, kRepWord32, kRepFloat64);
+  CheckChange(IrOpcode::kChangeFloat64ToInt32, kRepFloat64, kRepWord32);
+}
+
+
+TEST(Nops) {
+  RepresentationChangerTester r;
+
+  // X -> X is always a nop for any single representation X.
+  for (size_t i = 0; i < arraysize(all_reps); i++) {
+    r.CheckNop(all_reps[i], all_reps[i]);
+  }
+
+  // 32-bit floats.
+  r.CheckNop(kRepFloat32, kRepFloat32);
+  r.CheckNop(kRepFloat32 | kTypeNumber, kRepFloat32);
+  r.CheckNop(kRepFloat32, kRepFloat32 | kTypeNumber);
+
+  // 32-bit or 64-bit words can be used as branch conditions (kRepBit).
+  r.CheckNop(kRepWord32, kRepBit);
+  r.CheckNop(kRepWord32, kRepBit | kTypeBool);
+  r.CheckNop(kRepWord64, kRepBit);
+  r.CheckNop(kRepWord64, kRepBit | kTypeBool);
+
+  // 32-bit words can be used as smaller word sizes and vice versa, because
+  // loads from memory implicitly sign or zero extend the value to the
+  // full machine word size, and stores implicitly truncate.
+  r.CheckNop(kRepWord32, kRepWord8);
+  r.CheckNop(kRepWord32, kRepWord16);
+  r.CheckNop(kRepWord32, kRepWord32);
+  r.CheckNop(kRepWord8, kRepWord32);
+  r.CheckNop(kRepWord16, kRepWord32);
+
+  // kRepBit (result of comparison) is implicitly a wordish thing.
+  r.CheckNop(kRepBit, kRepWord8);
+  r.CheckNop(kRepBit | kTypeBool, kRepWord8);
+  r.CheckNop(kRepBit, kRepWord16);
+  r.CheckNop(kRepBit | kTypeBool, kRepWord16);
+  r.CheckNop(kRepBit, kRepWord32);
+  r.CheckNop(kRepBit | kTypeBool, kRepWord32);
+  r.CheckNop(kRepBit, kRepWord64);
+  r.CheckNop(kRepBit | kTypeBool, kRepWord64);
+}
+
+
+TEST(TypeErrors) {
+  RepresentationChangerTester r;
+
+  // Floats cannot be implicitly converted to/from comparison conditions.
+  r.CheckTypeError(kRepFloat64, kRepBit);
+  r.CheckTypeError(kRepFloat64, kRepBit | kTypeBool);
+  r.CheckTypeError(kRepBit, kRepFloat64);
+  r.CheckTypeError(kRepBit | kTypeBool, kRepFloat64);
+
+  // Floats cannot be implicitly converted to/from comparison conditions.
+  r.CheckTypeError(kRepFloat32, kRepBit);
+  r.CheckTypeError(kRepFloat32, kRepBit | kTypeBool);
+  r.CheckTypeError(kRepBit, kRepFloat32);
+  r.CheckTypeError(kRepBit | kTypeBool, kRepFloat32);
+
+  // Word64 is internal and shouldn't be implicitly converted.
+  r.CheckTypeError(kRepWord64, kRepTagged | kTypeBool);
+  r.CheckTypeError(kRepWord64, kRepTagged);
+  r.CheckTypeError(kRepWord64, kRepTagged | kTypeBool);
+  r.CheckTypeError(kRepTagged, kRepWord64);
+  r.CheckTypeError(kRepTagged | kTypeBool, kRepWord64);
+
+  // Word64 / Word32 shouldn't be implicitly converted.
+  r.CheckTypeError(kRepWord64, kRepWord32);
+  r.CheckTypeError(kRepWord32, kRepWord64);
+  r.CheckTypeError(kRepWord64, kRepWord32 | kTypeInt32);
+  r.CheckTypeError(kRepWord32 | kTypeInt32, kRepWord64);
+  r.CheckTypeError(kRepWord64, kRepWord32 | kTypeUint32);
+  r.CheckTypeError(kRepWord32 | kTypeUint32, kRepWord64);
+
+  for (size_t i = 0; i < arraysize(all_reps); i++) {
+    for (size_t j = 0; j < arraysize(all_reps); j++) {
+      if (i == j) continue;
+      // Only a single from representation is allowed.
+      r.CheckTypeError(all_reps[i] | all_reps[j], kRepTagged);
+    }
+  }
+
+  // TODO(titzer): Float32 representation changes trigger type errors now.
+  // Enforce current behavior to test all paths through representation changer.
+  for (size_t i = 0; i < arraysize(all_reps); i++) {
+    r.CheckTypeError(all_reps[i], kRepFloat32);
+    r.CheckTypeError(kRepFloat32, all_reps[i]);
+  }
+}