CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
    TFE_Execute(op, &h, &num_retvals, status);
    CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
    CHECK_EQ(1, num_retvals);
    CHECK(h);
    CHECK_EQ(TF_FLOAT, TFE_TensorHandleDataType(h));
    CHECK_EQ(0, TFE_TensorHandleNumDims(h, status));
    CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
    h = nullptr;
    TFE_DeleteOp(op);
  }

  TFE_ContextOptions* opts = TFE_NewContextOptions();
  TFE_Context* ctx = TFE_NewContext(opts, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TFE_DeleteContextOptions(opts);

  TFE_TensorHandle* h = TestScalarTensorHandle(ctx, 1.0f);
  TFE_TensorDebugInfo* debug_info = TFE_TensorHandleTensorDebugInfo(h, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);

  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TFE_OpAddInput(op, a, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TF_DeleteStatus(status);
  TFE_OpSetAttrType(op, "T", TFE_TensorHandleDataType(a));

  return op;
}

TFE_Op* ShapeOp(TFE_Context* ctx, TFE_TensorHandle* a) {
  TF_Status* status = TF_NewStatus();

  TFE_Op* op = TFE_NewOp(ctx, "Shape", status);

  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  int num_devices = TF_DeviceListCount(devices);
  for (int d = 0; d < num_devices; ++d) {
    const char* name = TF_DeviceListName(devices, d, status);
    CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
    TFE_TensorHandle* copy = TFE_TensorHandleCopyToDevice(m, ctx, name, status);
    CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);

  // Prepare some tensors for shape.
  TF_Tensor* tensor_1X6 = Int32Tensor({1, 6});
  CHECK_EQ(TF_OK, TF_GetCode(status_)) << TF_Message(status_);
  TF_Tensor* tensor_1X1X6 = Int32Tensor({1, 1, 6});
  CHECK_EQ(TF_OK, TF_GetCode(status_)) << TF_Message(status_);

  TFE_Op* reshape_op = TFE_NewOp(tfe_context_, "Reshape", status_);
  CHECK_EQ(TF_OK, TF_GetCode(status_)) << TF_Message(status_);
  TFE_OpSetAttrType(reshape_op, "T", TF_FLOAT);

  auto* counter = CreateCounter0(counter_name);
  auto* reader = TFE_MonitoringNewCounterReader(counter_name);
  IncrementCounter0(counter);

  int64_t actual = TFE_MonitoringReadCounter0(reader);

  CHECK_EQ(actual, 1);
}

TEST(CAPI, MonitoringCellReader1) {
  auto counter_name = "test/counter1";
  auto label_name = "test/label";
  auto* counter = CreateCounter1(counter_name, label_name);

  TF_Tensor* t = TFE_AllocateHostTensor(ctx, datatype, dims, num_dims, status);
  memcpy(TF_TensorData(t), data, TF_TensorByteSize(t));
  TFE_TensorHandle* th = TFE_NewTensorHandleFromTensor(ctx, t, status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TF_DeleteTensor(t);
  TF_DeleteStatus(status);
  return th;
}

// Return a scalar tensor handle with given values.
template <class T, TF_DataType datatype>

 protected:
  void SetUp() override {
    TF_StatusPtr status(TF_NewStatus());
    TF_SetTracingImplementation(std::get<0>(GetParam()), status.get());
    absl::Status s = StatusFromTF_Status(status.get());
    CHECK_EQ(errors::OK, s.code()) << s.message();
  }
};

absl::Status RegisterGradients(GradientRegistry* registry) {
  TF_RETURN_IF_ERROR(RegisterNotDifferentiable(registry, "CheckNumerics"));
  return absl::OkStatus();
}

    string function_def = MatMulFunction(matmul_device);
    TFE_ContextAddFunctionDef(ctx, function_def.data(), function_def.size(),
                              status);
    CHECK_EQ(TF_GetCode(status), TF_OK) << TF_Message(status);

    matmul = TFE_NewOp(ctx, "MatMulFunction", status);
    ASSERT_EQ(TF_GetCode(status), TF_OK) << TF_Message(status);
    TFE_OpAddInput(matmul, h0_task0, status);

  void SetUp() override {
    TF_StatusPtr status(TF_NewStatus());
    TF_SetTracingImplementation(std::get<0>(GetParam()), status.get());

    {
      absl::Status s = StatusFromTF_Status(status.get());
      CHECK_EQ(errors::OK, s.code()) << s.message();
    }

    {
      AbstractContext* ctx_raw = nullptr;
      absl::Status s =
          BuildImmediateExecutionContext(std::get<1>(GetParam()), &ctx_raw);