status_ = StatusFromTF_Status(status.get());
    ASSERT_EQ(errors::OK, status_.code()) << status_.message();

    {
      AbstractContext* ctx_raw = nullptr;
      status_ =
          BuildImmediateExecutionContext(std::get<1>(GetParam()), &ctx_raw);
      ASSERT_EQ(errors::OK, status_.code()) << status_.message();
      immediate_execution_ctx_.reset(ctx_raw);
    }

  ASSERT_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  const DLTensor* dltensor_out = &dlm_out->dl_tensor;
  EXPECT_EQ(dltensor_out->device.device_type, dltensor_in->device.device_type);
  EXPECT_EQ(dltensor_out->device.device_id, dltensor_in->device.device_id);
  EXPECT_EQ(dltensor_out->ndim, dltensor_in->ndim);
  EXPECT_EQ(dltensor_out->dtype.code, dltensor_in->dtype.code);

    ASSERT_EQ(errors::OK, s.code()) << s.message();
    s = dyn_cast<tracing::TracingContext>(ctx.get())->AddParameter(
        DT_FLOAT, shape, &x_raw);
    ASSERT_EQ(errors::OK, s.code()) << s.message();
    x.reset(x_raw);
  }

  PartialTensorShape shape;
  Status s = x->Shape(&shape);
  ASSERT_EQ(errors::OK, s.code()) << s.message();
  ASSERT_FALSE(shape.unknown_rank());

    status_ = StatusFromTF_Status(status.get());
    ASSERT_EQ(errors::OK, status_.code()) << status_.message();

    {
      AbstractContext* ctx_raw = nullptr;
      status_ =
          BuildImmediateExecutionContext(std::get<1>(GetParam()), &ctx_raw);
      ASSERT_EQ(errors::OK, status_.code()) << status_.message();
      immediate_execution_ctx_.reset(ctx_raw);
    }

                          &numerical_grad_raw);
    ASSERT_EQ(errors::OK, s.code()) << s.message();
    numerical_grad.reset(numerical_grad_raw);
  }

  TF_Tensor* numerical_tensor;
  s = GetValue(numerical_grad.get(), &numerical_tensor);
  ASSERT_EQ(errors::OK, s.code()) << s.message();
  auto num_elem_numerical = TF_TensorElementCount(numerical_tensor);
  ASSERT_EQ(num_elem_numerical, num_grad);

  auto s = GetValue(t, &analytical_tensor);
  ASSERT_EQ(errors::OK, s.code()) << s.message();

  int64_t num_elem_analytical = 1;
  auto num_dims_analytical = TF_NumDims(analytical_tensor);
  ASSERT_EQ(dims.size(), num_dims_analytical);
  for (int j = 0; j < num_dims_analytical; j++) {
    auto dim_analytical = TF_Dim(analytical_tensor, j);
    ASSERT_EQ(dims[j], dim_analytical);
    num_elem_analytical *= dim_analytical;
  }

    CHECK_EQ(errors::OK, s.code()) << s.message();
  }
};

class PassThroughGradientFunction : public GradientFunction {
 public:
  Status Compute(AbstractContext* ctx,
                 absl::Span<AbstractTensorHandle* const> grad_outputs,
                 absl::Span<AbstractTensorHandle*> grad_inputs) override {
    CHECK_EQ(grad_outputs.size(), 1);
    CHECK_EQ(grad_inputs.size(), 1);

  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);

                  .ok());
  ASSERT_TRUE(worker_server2->Start().ok());

  TFE_ContextSetServerDef(context.get(), 0, serialized.data(),
                          serialized.size(), status.get());
  EXPECT_EQ(TF_OK, TF_GetCode(status.get())) << TF_Message(status.get());

  BasicTestsForTwoDevices(context.get(),
                          "/job:worker/replica:0/task:1/device:CPU:0",

  ASSERT_EQ(errors::OK, s.code()) << s.message();
  if (isa<TracingOperation>(check_numerics_op.get())) {
    s = dyn_cast<TracingOperation>(check_numerics_op.get())
            ->SetOpName("check_numerics");
    ASSERT_EQ(errors::OK, s.code()) << s.message();
  }
  s = AddInput(check_numerics_op.get(), t.get(), &forward_op);
  ASSERT_EQ(errors::OK, s.code()) << s.message();