const char* device_name, void* device_info,
                              TF_Status* status) {
  // Fill in default values for optional functionality.
  if (device.pack == nullptr) {
    device.pack = &DefaultCustomDevicePack;
  }
  auto custom_device = std::make_unique<tensorflow::CustomDeviceAPI>(
      ctx, device, device_info, device_name);

				info.Name = devName
				qst, err := bfs.SysBlockDeviceQueueStats(devName)
				if err != nil { // Mostly not found error
					// Check if there is a parent device:
					//   e.g. if the mount is based on /dev/nvme0n1p1, let's calculate the
					//        real device name (nvme0n1) to get its sysfs information
					parentDevPath, e := os.Readlink("/sys/class/block/" + devName)
					if e == nil {

 * Prefer recent JDK builds, and results are temperamental to slight environment changes.
 * Different instructions and configuration may be required for other hardware devices.
 *
 * Using a yubikey device as a SSL key store.
 * https://lauri.võsandi.com/2017/03/yubikey-for-ssh-auth.html
 *
 * Using PKCS11 support in the JDK.
 * https://tersesystems.com/blog/2018/09/08/keymanagers-and-keystores/
 *

  }

  auto* gpu_options = config.mutable_gpu_options();
  gpu_options->set_allow_growth(gpu_memory_allow_growth);

  (*config.mutable_device_count())["CPU"] = num_cpu_devices;

  // 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

    CHECK_EQ(TF_OK, TF_GetCode(s)) << TF_Message(s);
    LOG(INFO) << "Device " << i << " has name " << device_name << ", type "
              << device_type;
    if (string(device_type) == DEVICE_GPU) {
      return device_name;
    }
  }
  // No GPU device found.
  return "";
}

string GPUDeviceName() {
  std::unique_ptr<TF_Status, decltype(&TF_DeleteStatus)> status(

    TF_ImportGraphDefOptions* opts, const char* prefix);

// Set the execution device for nodes in `graph_def`.
// Only applies to nodes where a device was not already explicitly specified.
// `device` is copied and has no lifetime requirements.
TF_CAPI_EXPORT extern void TF_ImportGraphDefOptionsSetDefaultDevice(
    TF_ImportGraphDefOptions* opts, const char* device);

// Set whether to uniquify imported operation names. If true, imported operation

[CUDA-enabled GPU cards](https://www.tensorflow.org/install/gpu) *(Ubuntu and
Windows)*:

```
$ pip install tensorflow
```

Other devices (DirectX and MacOS-metal) are supported using
[Device plugins](https://www.tensorflow.org/install/gpu_plugins#available_devices).

A smaller CPU-only package is also available:

```
$ pip install tensorflow-cpu
```

  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);

  DLDevice ctx;
  const char* device_name =
      tensorflow::unwrap(h)->BackingDeviceName(&status->status);
  DeviceNameUtils::ParsedName parsed_name;
  tensorflow::DeviceNameUtils::ParseFullName(device_name, &parsed_name);
  std::string device_type = parsed_name.type;
  int device_id = 0;
  if (parsed_name.has_id) {
    device_id = parsed_name.id;
  }

  ctx.device_id = device_id;
  if (device_type == "CPU") {

// C API for device. The API is under active development and eventually
// should allow registering a plugin device with TensorFlow.

#ifdef __cplusplus
extern "C" {
#endif

// TF_Device is a C wrapper to the C++ TF Device class. This is to be passed
// through TF_OpKernelContext, and is opaque to plugin.
typedef struct TF_Device TF_Device;

typedef struct TF_VariableInfo TF_VariableInfo;