TF_EXPECT_OK(ReadCache(&cache, "", block_size, 1, &out));
  TF_EXPECT_OK(ReadCache(&cache, "", block_size, 1, &out));
  calls.push_back(2 * block_size);
  // Cache miss followed by cache hit.  Causes eviction of LRU element.
  TF_EXPECT_OK(ReadCache(&cache, "", 2 * block_size, 1, &out));
  TF_EXPECT_OK(ReadCache(&cache, "", 2 * block_size, 1, &out));
  // LRU element was at offset 0.  Cache miss.
  calls.push_back(0);

// If assume_valid_feeds is true, it can help infer shapes in the fanout of fed
// nodes. This may cause incorrectness in graph analyses, but is useful for
// simulation or scheduling.
// If aggressive_shape_inference is true, nodes are executed on the host to
// identify output values when possible and does other aggressive strategies.
// This may cause incorrectness in graph analyses, but is useful for simulation
// or scheduling.

// make sure all nodes are finished.
TF_CAPI_EXPORT extern void TFE_DeleteExecutor(TFE_Executor*);

// Returns true if the executor is in async mode.
TF_CAPI_EXPORT extern bool TFE_ExecutorIsAsync(TFE_Executor*);

// Causes the calling thread to block till all ops dispatched in this executor
// have been executed. Note that "execution" here refers to kernel execution /
// scheduling of copies, etc. Similar to sync execution, it doesn't guarantee

    **Note:** You can add `--test_sharding_strategy=disabled` to the `flags` to
    disable the sharding so that all the test outputs are in one file. However,
    it may slow down the tests for not running in parallel and may cause the
    test to timeout but it could be useful when you need to execute a single
    test or more in general your filtered/selected tests have a very low
    execution time and the sharding

   ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
   (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
   LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
   ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
   (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
   SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

      "/job:localhost/replica:0/task:1/device:CPU:0";
  CheckRemoteMatMulExecutesOK(ctx, remote_device_name, local_device_name);

  // Adding a non-existent remote worker to cluster def. This should cause the
  // UpdateServerDef call to fail.
  tensorflow::ClusterDef* cluster_def = server_def.mutable_cluster();
  tensorflow::JobDef* job_def = cluster_def->mutable_job(0);
  int port = tensorflow::testing::PickUnusedPortOrDie();

      // ones_like for variants.
      aid = output_tensor.ZerosLike();
    } else {
      // TODO(allenl): Figure out why using zeros_like everywhere causes issues
      // for some gradient functions and if there's another way to work around
      // it (e.g. conds instead of ifs). The value shouldn't really matter.

                        const int in_flight_nodes_limit)
      : status_(TF_NewStatus()),
        // If the context's default exector is set to async, re-using that in
        // each thread would cause collectives to deadlock. For consistency we
        // create a new sync executor for every thread.
        //
        // TODO(allenl): We should have an async API that works with the
        // parallel device.