// on the given device.
TFE_TensorHandle* TestVariable(TFE_Context* ctx, float value,
                               const tensorflow::string& device_name = "");

// Return an add op multiplying `a` by `b`.
TFE_Op* AddOp(TFE_Context* ctx, TFE_TensorHandle* a, TFE_TensorHandle* b);

// Return a matmul op multiplying `a` by `b`.
TFE_Op* MatMulOp(TFE_Context* ctx, TFE_TensorHandle* a, TFE_TensorHandle* b);

// Return an identity op.

// A handle to a tensor on a device.
//
// Like a TF_Tensor, a TFE_TensorHandle refers to a tensor with a value, shape,
// type etc. Unlike a TF_Tensor, a TFE_TensorHandle may refer to such tensors
// placed in the memory of different devices or remote address spaces.
typedef struct TFE_TensorHandle TFE_TensorHandle;

TF_CAPI_EXPORT extern TFE_TensorHandle* TFE_NewTensorHandle(const TF_Tensor* t,

                                         status);
}

TFE_TensorHandle* CopyToLoggingDevice(TFE_Context* context,
                                      TFE_TensorHandle* tensor,
                                      TF_Status* status, void* device_info) {
  LoggingDevice* dev = reinterpret_cast<LoggingDevice*>(device_info);
  TFE_TensorHandle* t = TFE_TensorHandleCopyToDevice(

  TFE_DeleteContextOptions(opts);

  TFE_TensorHandle* input1 = TestMatrixTensorHandle(ctx);
  TFE_TensorHandle* input2 = TestMatrixTensorHandle(ctx);
  TFE_TensorHandle* dim = TestScalarTensorHandle(ctx, 0);
  TFE_Op* concatOp = TFE_NewOp(ctx, "Concat", status);
  CHECK_EQ(TF_OK, TF_GetCode(status)) << TF_Message(status);
  TFE_TensorHandle* inputs[] = {input1, input2};
  TFE_OpAddInput(concatOp, dim, status);

using tensorflow::string;
using tensorflow::tstring;

TFE_TensorHandle* TestScalarTensorHandle(TFE_Context* ctx, float value) {
  float data[] = {value};
  TF_Status* status = TF_NewStatus();
  TF_Tensor* t = TFE_AllocateHostTensor(ctx, TF_FLOAT, nullptr, 0, status);
  memcpy(TF_TensorData(t), &data[0], TF_TensorByteSize(t));
  TFE_TensorHandle* th = TFE_NewTensorHandleFromTensor(ctx, t, status);

}

TF_CAPI_EXPORT extern TFE_TensorHandle* TFE_TensorHandleCopySharingTensor(
    TFE_TensorHandle* h, TF_Status* status) {
  if (h == nullptr) {
    status->status = tensorflow::errors::InvalidArgument("Invalid handle");
    return nullptr;
  }

  tensorflow::unwrap(h)->Ref();
  return h;
}

TF_Tensor* TFE_TensorHandleResolve(TFE_TensorHandle* h, TF_Status* status) {
  if (h == nullptr) {

std::unique_ptr<ParallelTensor> ParallelDevice::CopyToParallelDevice(
    TFE_Context* context, TFE_TensorHandle* tensor, TF_Status* status) const {
  std::vector<TensorHandlePtr> components;
  components.reserve(underlying_devices_.size());
  for (const std::string& underlying_device_name : underlying_devices_) {
    TFE_TensorHandle* t = TFE_TensorHandleCopyToDevice(
        tensor, context, underlying_device_name.c_str(), status);

                        &arrived, &executed, status.get());
  ASSERT_TRUE(TF_GetCode(status.get()) == TF_OK) << TF_Message(status.get());
  TFE_TensorHandle* hcpu = TestMatrixTensorHandle(context);
  ASSERT_FALSE(arrived);
  TFE_TensorHandle* hdevice =
      TFE_TensorHandleCopyToDevice(hcpu, context, name, status.get());
  ASSERT_TRUE(arrived);
  ASSERT_FALSE(executed);

//
// Since this function is used to satisfy the TFE_CustomDevice C API,
// device_info is passed in using a C-style generic. It must always be a
// ParallelDevice.
TFE_TensorHandle* CopyToParallelDevice(TFE_Context* context,
                                       TFE_TensorHandle* tensor,
                                       TF_Status* status, void* device_info) {
  TF_SetStatus(
      status, TF_UNIMPLEMENTED,

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

  TFE_TensorHandle* h0_task0 = TestMatrixTensorHandle(ctx);
  TFE_TensorHandle* h1_task0 = TestMatrixTensorHandle(ctx);
  std::vector<TFE_TensorHandle*> handles_task0;
  if (heavy_load_on_streaming_rpc) {
    // Send 50 tensor copy requests to simulate that there have been some RPC