*   [Google Objective-C Style Guide](https://google.github.io/styleguide/objcguide.html)

#### Running sanity check

If you have Docker installed on your system, you can perform a sanity check on
your changes by running the command:

```bash
tensorflow/tools/ci_build/ci_build.sh CPU tensorflow/tools/ci_build/ci_sanity.sh
```

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
called "MLCI" that makes testing more convenient: it can execute any full CI job
against a pending change. Search for "MLCI" internally to find it.

// maximum number of in flight async nodes. Enqueuing of additional async ops
// after the limit is reached blocks until some inflight nodes finishes.
// The effect is bounding the memory held by inflight TensorHandles that are
// referenced by the inflight nodes.
// A recommended value has not been established.
// A value of 0 removes the limit, which is the behavior of TensorFlow 2.11.

# Find the location of this exact file.
_current_file_location = _inspect.getfile(_inspect.currentframe())

def _running_from_pip_package():
  return any(
      _current_file_location.startswith(dir_) for dir_ in _site_packages_dirs)

if _running_from_pip_package():
  # TODO(gunan): Add sanity checks to loaded modules here.

  # Load first party dynamic kernels.

unless indicated otherwise.

1) Add the new version to the `VERSIONS` variable inside
   `tensorflow/tools/toolchains/python/python_repo.bzl`. \
   While this isn't necessary for running the updater, it is required for
   actually using the new version with Tensorflow.

2) In the `WORKSPACE` file, add the new version to the `python_versions`
   parameter of the `python_register_multi_toolchains` function.

  TF_TriState implementation_selector;
  TF_TriState function_optimization;
  TF_TriState common_subgraph_elimination;
  TF_TriState arithmetic_optimization;
  TF_TriState debug_stripper;
  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;

  CONFIG_TOGGLE(function_optimization);
  CONFIG_TOGGLE(common_subgraph_elimination);
  CONFIG_TOGGLE(arithmetic_optimization);
  CONFIG_TOGGLE(debug_stripper);
  CONFIG_TOGGLE(constant_folding);
  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);

*   Fixes an undefined behavior arising from reference binding to nullptr in
    `RaggedTensorToSparse`
    ([CVE-2021-37656](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-37656))
*   Fixes an undefined behavior arising from reference binding to nullptr in
    `MatrixDiagV*` ops
    ([CVE-2021-37657](https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-37657))

research](https://scholar.google.com/scholar?q=gpu+side+channel) shows that it
might be possible to use side channel attacks on the GPU to leak data from other
running models or processes in the same system. GPUs can also have
implementation bugs that might allow attackers to leave malicious code running
and leak or tamper with applications from other users. Please report
vulnerabilities to the vendor of the affected hardware accelerator.

unsigned char TF_GetXlaConstantFoldingDisabled() {
  return static_cast<unsigned char>(
      tensorflow::GetBuildXlaOpsPassFlags()->tf_xla_disable_constant_folding);
}

void TF_SetXlaConstantFoldingDisabled(unsigned char should_enable) {
  tensorflow::GetBuildXlaOpsPassFlags()->tf_xla_disable_constant_folding =
      static_cast<bool>(should_enable);
}

void TF_SetXlaMinClusterSize(int size) {
  tensorflow::MarkForCompilationPassFlags* flags =