#     v2: Build TF v2
#
# Feature and Third party library support options:
#     xla:          Build TF with XLA
#     tpu:          Build TF with TPU support
#     cuda:         Build with CUDA support.
#     cuda_clang    Build with CUDA Clang support.
#     rocm:         Build with AMD GPU support (rocm)
#     mkl:          Enable full mkl support.
#     tensorrt:     Enable Tensorrt support.
#     noaws:        Disable AWS S3 storage support

        write_to_bazelrc(
            'build --copt="-DEIGEN_ALTIVEC_ENABLE_MMA_DYNAMIC_DISPATCH=1"')

  with_xla_support = environ_cp.get('TF_ENABLE_XLA', None)
  if with_xla_support is not None:
    write_to_bazelrc('build --define=with_xla_support=%s' %
                     ('true' if int(with_xla_support) else 'false'))

  set_action_env_var(
      environ_cp, 'TF_NEED_ROCM', 'ROCm', False, bazel_config_name='rocm')

* <NOTES SHOULD BE GROUPED PER AREA>

* `tf.lite`
    * Added support for `stablehlo.gather`.
    * Added support for `stablehlo.add`.
    * Added support for `stablehlo.multiply`.
    * Added support for `stablehlo.maximum`.
    * Added support for `stablehlo.minimum`.
    * Added boolean parameter support for `tfl.gather_nd`.
    * C API:
        * New API functions:

// Caller must call TF_DeleteTensor() over the returned tensor. If the queue is
// empty, this call is blocked.
//
// Tensors are enqueued via the corresponding TF enqueue op.
// TODO(hongm): Add support for `timeout_ms`.
TF_CAPI_EXPORT extern TF_Tensor* TF_DequeueNamedTensor(TF_Session* session,
                                                       int tensor_id,

    std::unique_ptr<RandomAccessFile>* result) {
  if (ops_->new_random_access_file == nullptr)
    return errors::Unimplemented(tensorflow::strings::StrCat(
        "Filesystem for ", fname, " does not support NewRandomAccessFile()"));

  UniquePtrTo_TF_Status plugin_status(TF_NewStatus(), TF_DeleteStatus);
  auto file = std::make_unique<TF_RandomAccessFile>();
  std::string translated_name = TranslateName(fname);

       * Refer [linux setup guide](https://www.tensorflow.org/install/gpu#linux_setup).
     * If error still persists then, apparently your CPU model does not support AVX instruction sets.
       * Refer [hardware requirements](https://www.tensorflow.org/install/pip#hardware-requirements).
   
   -----------------------------------------------------------------------------------------------

config_setting(
    name = "stackdriver_support",
    define_values = {"stackdriver_support": "true"},
    visibility = ["//visibility:public"],
)

config_setting(
    name = "with_xla_support",
    define_values = {"with_xla_support": "true"},
    visibility = ["//visibility:public"],
)

# By default, XLA GPU is compiled into tensorflow when building with
# --config=cuda even when `with_xla_support` is false. The config setting

 	{
@@ -332,6 +333,7 @@ unsetenv (const char *name)
 	}
       else
 	++ep;
+  }
 
   UNLOCK;
 
diff --git a/support/Makefile b/support/Makefile
index a253698..2f4e2a9 100644
--- a/support/Makefile
+++ b/support/Makefile
@@ -167,13 +167,6 @@ CFLAGS-support_paths.c = \
 		-DINSTDIR_PATH=\"$(prefix)\" \
 		-DLIBDIR_PATH=\"$(libdir)\"
 
-ifeq (,$(CXX))
-LINKS_DSO_PROGRAM = links-dso-program-c

- Model produces wrong results and/or has lesser accuracy.
- Model produces correct results, but it is slower than expected.

### 4. (optional) RNN conversion support
If converting TF RNN to TFLite fused RNN ops, please prefix [RNN] in the title.

### 5. (optional) Any other info / logs

    srcs = ["modular_filesystem_test.cc"],
    linkopts = ["-ldl"],
    tags = [
        "manual",  # Requires DSOs as arguments, eventual setup
        "notap",  # b/139060984, requires implementing modular support for Google filesystem
    ],
    deps = [
        ":modular_filesystem",
        "//tensorflow/core:framework_internal",
        "//tensorflow/core/lib/io:path",
        "//tensorflow/core/platform:env",