*   Include unit tests when you contribute new features, as they help to a)
    prove that your code works correctly, and b) guard against future breaking
    changes to lower the maintenance cost.
*   Bug fixes also generally require unit tests, because the presence of bugs
    usually indicates insufficient test coverage.
*   Keep API compatibility in mind when you change code in core TensorFlow,
    e.g., code in

  TF_TriState constant_folding;
  TF_TriState shape_optimization;
  TF_TriState auto_mixed_precision;
  TF_TriState auto_mixed_precision_onednn_bfloat16;
  TF_TriState auto_mixed_precision_mkl;
  TF_TriState pin_to_host_optimization;
  TF_TriState layout_optimizer;
  TF_TriState remapping;
  TF_TriState loop_optimization;
  TF_TriState dependency_optimization;
  TF_TriState auto_parallel;
  TF_TriState memory_optimization;

    int num_dims, void* data, size_t len,
    void (*deallocator)(void* data, size_t len, void* arg),
    void* deallocator_arg, TF_Status* status);

// Retrieves the address space (i.e. job, replia, task) of the local host and
// saves it in the buffer.
TF_CAPI_EXPORT extern void TFE_HostAddressSpace(TFE_Context* ctx,
                                                TF_Buffer* buf);

  absl::Mutex mu_;

  /// The cache (a map from string key to Entry).
  std::map<std::string, Entry> cache_ ABSL_GUARDED_BY(mu_);

  /// The LRU list of entries. The front of the list identifies the most
  /// recently accessed entry.
  std::list<std::string> lru_list_ ABSL_GUARDED_BY(mu_);
};

}  // namespace tf_gcs_filesystem

  CONFIG_TOGGLE(shape_optimization);
  CONFIG_TOGGLE(auto_mixed_precision);
  CONFIG_TOGGLE(auto_mixed_precision_onednn_bfloat16);
  CONFIG_TOGGLE(auto_mixed_precision_mkl);
  CONFIG_TOGGLE(pin_to_host_optimization);
  CONFIG_TOGGLE(layout_optimizer);
  CONFIG_TOGGLE(remapping);
  CONFIG_TOGGLE(loop_optimization);
  CONFIG_TOGGLE(dependency_optimization);
  CONFIG_TOGGLE(auto_parallel);

  EXPECT_EQ(TF_GraphNumFunctions(host_graph_), 1);
  EXPECT_EQ(TF_GraphGetFunctions(host_graph_, funcs, 0, s_), 0);
  ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_);
  EXPECT_EQ(TF_GraphGetFunctions(host_graph_, funcs, 1, s_), 1);
  ASSERT_EQ(TF_OK, TF_GetCode(s_)) << TF_Message(s_);
  AssertEqual(func0, funcs[0]);
  TF_DeleteFunction(funcs[0]);

  DEVICE_PLACEMENT_EXPLICIT = 0,
  // Copy the tensor to the right device but log a warning.
  DEVICE_PLACEMENT_WARN = 1,
  // Silently copy the tensor, which has a performance cost since the operation
  // will be blocked till the copy completes. This is the default policy.
  DEVICE_PLACEMENT_SILENT = 2,
  // Placement policy which silently copies int32 tensors but not other dtypes.

  echo "More than one wheel file is present: moving the oldest to"
  echo "$TFCI_OUTPUT_DIR/extra_wheels."
  # List all .whl files by their modification time (ls -t) and move anything
  # other than the most recently-modified one (the newest one).
  mkdir -p $TFCI_OUTPUT_DIR/extra_wheels
  ls -t *.whl | tail -n +2 | xargs mv -t $TFCI_OUTPUT_DIR/extra_wheels
fi

# Repair wheels with auditwheel and delete the old one.

  BlockMap block_map_ ABSL_GUARDED_BY(mu_);

  /// The LRU list of block keys. The front of the list identifies the most
  /// recently accessed block.
  std::list<Key> lru_list_ ABSL_GUARDED_BY(mu_);

  /// The LRA (least recently added) list of block keys. The front of the list
  /// identifies the most recently added block.
  ///
  /// Note: blocks are added to lra_list_ only after they have successfully been

}

void ReplaceTaskInServerDef(tensorflow::ServerDef* server_def, int task_index,
                            const string& host, int port) {
  tensorflow::JobDef* job_def = server_def->mutable_cluster()->mutable_job(0);
  job_def->mutable_tasks()->at(task_index) =
      tensorflow::strings::StrCat(host, ":", port);
}

std::vector<std::string> ListDeviceNames(TFE_Context* ctx) {
  TF_Status* status = TF_NewStatus();