// TFE_NewTensorHandleFromDeviceMemory.
void DeallocatorWrapperFunc(void* data, size_t len, void* dlmt_vptr) {
  TFE_CallDLManagedTensorDeleter(dlmt_vptr);
}

// Checks whether the stride array matches the layout of compact, row-majored
// data.
bool IsValidStrideCompactRowMajorData(int64_t* shape_arr, int64_t* stride_arr,
                                      int ndim) {
  bool valid = true;
  int64_t expected_stride = 1;

        //
        // TODO(allenl): There may be smarter ways to do this copy in some
        // cases, i.e. with a collective broadcast. We'll need to be careful
        // about things that are taken as inputs on the host or on their
        // existing device (for multi-device functions).
        std::unique_ptr<ParallelTensor> parallel_tensor(
            parallel_device.CopyToParallelDevice(