return true;
}

// Copies `src_graph` into `dst_graph`. Any node in `src_graph` with input
// `src_inputs[i]` will have that input replaced with `dst_inputs[i]`.  `prefix`
// will be prepended to copied node names. `control_deps` are nodes in
// `dst_graph` that the copied `src_graph` nodes will have control dependencies
// on. `return_nodes` are nodes in `src_graph`, and the new corresponding nodes

  // TODO(b/113217601): This is needed for EagerContext::runner_ to use a
  // threadpool, so that we avoid the possibility of running the runner_ in the
  // threadpool of GPU event mgr, as that can trigger more callbacks to be
  // scheduled on that same threadpool, causing a deadlock in cases where the
  // caller of event_mgr->ThenExecute() blocks on the completion of the callback

    for (int i = 0; i < grad_outputs.size(); i++) {
      auto grad_input = grad_outputs[i];
      // TODO(srbs): Should we add a copy contructor to AbstractTensorHandle
      // that takes care of this similar to `Tensor`?
      if (grad_input) {
        grad_input->Ref();
      }
      grad_inputs[i] = grad_input;
    }
    return absl::OkStatus();
  }

                       const char* dst, TF_Status* status) {
  // If target is a directory return TF_FAILED_PRECONDITION.
  // Target might be missing, so don't error in that case.
  struct stat st;
  if (stat(dst, &st) != 0) {
    if (errno != ENOENT) {
      TF_SetStatusFromIOError(status, errno, dst);
      return;
    }
  } else if (S_ISDIR(st.st_mode)) {

                                    fn_name, "'");
    output_tensors->emplace_back(node, idx);
  }
  return absl::OkStatus();
}

// Populates `body_nodes` with the nodes that will become function's body.
// Performs various checks.
Status ComputeBodyNodes(
    const TF_Graph* fn_body, const char* fn_name, int num_opers,
    const TF_Operation* const* opers,

  }

 public:
  bool UseMlir() const { return strcmp(std::get<0>(GetParam()), "mlir") == 0; }
  bool UseFunction() const { return std::get<2>(GetParam()); }
};

// Checks that inputs[0] is a scalar.
Status TestScalarShape(AbstractContext* ctx,
                       absl::Span<AbstractTensorHandle* const> inputs,
                       absl::Span<AbstractTensorHandle*> outputs) {

  (*sess_options.config.mutable_device_count())["CPU"] = num_cpus;
  status->status =
      tensorflow::DeviceFactory::AddCpuDevices(sess_options, prefix, &devices);

  // Remove the device that has the host device name since host device is alreay
  // in an initialized context.
  for (auto d = devices.begin(); d != devices.end();) {
    if (absl::StrContains(d->get()->name(), "CPU:0")) {
      d = devices.erase(d);

  buf->Unref();
  return tensorflow::wrap(tensorflow::TensorHandle::CreateLocalHandle(
      std::move(t), device, device, context));
}

// This function will block till the operation that produces `h` has
// completed. This is only valid on local TFE_TensorHandles. Returns the size in
// bytes of the memory pointed to by the device pointer returned above.
size_t TFE_TensorHandleDeviceMemorySize(TFE_TensorHandle* h,

  // Fetch and see that the new value is returned.
  TF_GraphGetTensorShape(graph, feed_out_0, returned_dims, num_dims, s);
  ASSERT_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
  EXPECT_EQ(dims[0], returned_dims[0]);
  EXPECT_EQ(dims[1], returned_dims[1]);

  // Try to set 'unknown' with unknown rank on the shape and see that
  // it doesn't change.

  TF_RETURN_IF_ERROR(ops::SparseSoftmaxCrossEntropyWithLogits(
      ctx, inputs[0], inputs[1], &loss, &backprop,
      "SparseSoftmaxCrossEntropyWithLogits"));
  // `gradient_checker` only works with model that returns only 1 tensor.
  // Although, `ops::SparseSoftmaxCrossEntropyWithLogits` returns 2 tensors, the
  // second tensor isn't needed for computing gradient so we could safely drop
  // it.
  outputs[0] = loss;