once you get into the running container so `bazel` can find the `tensorflow`
    workspace).

    you can do this by using the following command. As an example-

    ```bash
    docker run -it --rm -v $PWD:/tmp -w /tmp tensorflow/build:2.15-python3.10
    ```

    Once you have the packages installed, you can run a specific unit test in
    bazel by doing as follows:

    ```bash

// EagerService to maintain consistency between client and worker. The
// context_id is initialized with a dummy value and is later set when the worker
// is initialized (either locally or remotely). The context_id can change during
// the process lifetime although this should cause the worker to be
// reinitialized (e.g. cleared caches) as well.
TF_CAPI_EXPORT extern uint64_t TFE_GetContextId(TFE_Context* ctx);

        "//conditions:default": [],
    }),
    tags = [
        "no_cuda_asan",  # TODO(b/181771536)
        "no_windows",  # TODO(b/155444728)
    ],
    # We must ensure that the dependencies can be dynamically linked since
    # the shared library must be able to use core:framework.
    deps = [
        ":c_api",
        ":c_api_internal",
        ":c_test_util",
        ":test_op_kernel",

  int64_t expected_stride = 1;
  for (int i = ndim - 1; i >= 0; --i) {
    // Empty tensors are always compact regardless of strides.
    if (shape_arr[i] == 0) return true;
    // Note that dimensions with size=1 can have any stride.
    if (shape_arr[i] != 1 && stride_arr[i] != expected_stride) {
      valid = false;
    }
    expected_stride *= shape_arr[i];
  }
  return valid;
}
}  // namespace

// gradients available initially).
//
// Then we repeatedly pop an entry from the stack, run its backprop, and update
// the gradients of its inputs. Once we have computed all gradients for a single
// input we can mark this input as done, and this can trigger adding an entry to
// the stack if all outputs of that entry are now done.
//
// When the stack is empty we have gradients for all tensors we're interested
// in.

namespace {

# checks for cross-file issues like bad includes or missing BUILD definitions.
#
# We can't test on the windows toolchains because they're using a legacy
# toolchain format (or something) that specifies the toolchain directly instead
# of as a "repository". They can't be valid on Linux because Linux can't do
# anything with a Windows-only toolchain, and bazel errors if trying to build
# that directory.

  // 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
  // threadpool of GPU event mgr, as that can trigger more callbacks to be
  // scheduled on that same threadpool, causing a deadlock in cases where the
  // caller of event_mgr->ThenExecute() blocks on the completion of the callback

  CancellationManager* cancellation_manager_ TF_GUARDED_BY(execution_mutex_);
  //   Outputs:
  std::vector<TensorHandlePtr> op_outputs_ TF_GUARDED_BY(execution_mutex_);
  // TF_Status is an incomplete type and so can't be stack allocated. To avoid
  // unnecessary allocations each Execute call, we keep one heap-allocated
  // version for the thread.
  StatusPtr status_ TF_GUARDED_BY(execution_mutex_);

  const std::string device_;

    }
    return tensorflow::errors::Internal("Graph pass runs for more than once!");
  }

 private:
  bool first_call_ = true;
};

// After the graph pass is registered, it takes effect globally and can affect
// other test cases. Define a static variable to switch it on and off.
bool GraphErrorInjectionPass::enabled_ = false;

// Test to ensure that a registered graph optimization pass is only executed