-   x86 and arm64
-   CPU-only, or with NVIDIA CUDA support (Linux only), or with TPUs

## How to Test Your Changes to TensorFlow

You may check how your changes will affect TensorFlow by:

1. Creating a PR and observing the presubmit test results
2. Running the CI scripts locally, as explained below
3. **Google employees only**: Google employees can use an internal-only tool

                                                    TF_Status* status);

// Get configuration key and value using coordination service.
// The config key must be set before getting its value. Getting value of
// non-existing config keys will result in errors.
// If `timeout_in_ms=0`, this call will block until the key-value is set or the
// worker shuts down.

just getting started, GitHub has a
[how-to](https://help.github.com/articles/using-pull-requests/).

TensorFlow team members will be assigned to review your pull requests. Once the
pull requests are approved and pass continuous integration checks, a TensorFlow
team member will apply `ready to pull` label to your change. This means we are

                           TF_Status* status) {
#if defined(IS_MOBILE_PLATFORM) || defined(IS_SLIM_BUILD)
  status->status = tensorflow::errors::Unimplemented(
      "Creating while loops is not supported on mobile. File a bug at "
      "https://github.com/tensorflow/tensorflow/issues if this feature is "
      "important to you");
  return EmptyWhileParams();
#else

    forward_op_->outputs.push_back(retvals[i]);
  }
  // TODO(b/166669239): This is needed to support AttrBuilder::Get for string
  // attributes. Number type attrs and DataType attrs work fine without this.
  // Consider getting rid of this and making the behavior between number types
  // and string consistent.
  forward_op_->attrs.BuildNodeDef();
  std::unique_ptr<GradientFunction> gradient_fn;

    TF_Status* status);

// ----------------------------  PJRT  -----------------------------------------
// Passes the pointer to a vector of PJRT_NamedValue and number of options to
// set options for creating a PJRT client. Passes nullptr for create_options and
// 0 for num_options if no options need to be set. You can use
// ConvertToPjRtNamedValueList in
// tensorflow/compiler/xla/pjrt/c/pjrt_c_api_helpers.h to generate the options.

    *   Add 5D support to `SLICE` op.
    *   TFLite Supports SingatureDef:
        *   TFLiteConverter exports models with SignatureDef
        *   Interpreter supports getting a list of signatures and getting
            callable function for a given signaturedef.
    *   Add int8 support for `ReshapeV2`.
    *   Add experimental support for optimization with sparsity.

// GraphDef type. These functions must be deleted by calling TF_DeleteFunction.
//
// If `mutate_proto_func` is non-NULL, run it over each FunctionDef proto,
// before creating a TF_Function out of the possibly mutated proto.
static std::vector<UniqueFuncPtr> CreateFunctionsFromTextProto(
    const char* text_proto,
    std::function<void(FunctionDef*)>* mutate_proto_func, TF_Status* status) {

// Operations the tape needs to perform on tensors to do backpropagation. Named
// "vspace" because a subset of these are related to a vector space, such as
// adding gradients, getting zeroes, etc. Currently cannot be implemented
// without using tensorflow python code, hence left unspecified here.
//
// Gradient is the type returned by gradient functions. In Python TF it's either

  args = parser.parse_args()

  _TF_WORKSPACE_ROOT = args.workspace
  _TF_BAZELRC = os.path.join(_TF_WORKSPACE_ROOT, _TF_BAZELRC_FILENAME)

  # Make a copy of os.environ to be clear when functions and getting and setting
  # environment variables.
  environ_cp = dict(os.environ)

  try:
    current_bazel_version = retrieve_bazel_version()
  except subprocess.CalledProcessError as e: