noTick()

		// Test that len sees an immediate tick arrive
		// for Reset of timer in heap.
		tim.Reset(1)
		assertLen()
		assertTick()

		// Test that len sees an immediate tick arrive
		// for Reset of timer NOT in heap.
		tim.Stop()
		drainAsync()
		tim.Reset(1)
		assertLen()
		assertTick()
	}

	// Sleep long enough that a second tick must happen if this is a ticker.

    TF_EXPECT_OK(Env::Default()->FileExists(asset_filepath));

    std::vector<Tensor> path_outputs;
    TF_ASSERT_OK(
        bundle.session->Run({}, {"filename_tensor:0"}, {}, &path_outputs));
    ASSERT_EQ(1, path_outputs.size());

    test::ExpectTensorEqual<tstring>(
        test::AsTensor<tstring>({"foo.txt"}, TensorShape({})), path_outputs[0]);
  }

  void CheckSavedModelBundle(const string& export_dir,

TF_CAPI_EXPORT extern int TF_ShapeInferenceContextRankKnown(
    TF_ShapeInferenceContext* ctx, TF_ShapeHandle* handle);

// If <handle> has rank <rank>, or its rank is unknown, return OK and return the
// shape with asserted rank in <*result>. Otherwise an error is placed into
// `status`.
TF_CAPI_EXPORT extern void TF_ShapeInferenceContextWithRank(
    TF_ShapeInferenceContext* ctx, TF_ShapeHandle* handle, int64_t rank,

  ASSERT_TRUE(muladd_shape != nullptr);
  ASSERT_EQ(muladd_shape->parameters_size(), 2);
  EXPECT_TRUE(
      ShapeUtil::Compatible(xla::Shape(muladd_shape->parameters(0)), f32_2x2));
  EXPECT_TRUE(
      ShapeUtil::Compatible(xla::Shape(muladd_shape->parameters(1)), f32_2x2));

  const xla::Shape muladd_result(muladd_shape->result());
  ASSERT_EQ(muladd_result.element_type(), xla::TUPLE);

  TF_ASSERT_OK(BuildXlaOps(root, fdef_lib, &graph));

  Node* stateful_partitioned_call_op = nullptr;
  for (Node* n : graph->op_nodes()) {
    if (n->type_string() == "StatefulPartitionedCall") {
      ASSERT_EQ(stateful_partitioned_call_op, nullptr);
      stateful_partitioned_call_op = n;
    }
  }

  ASSERT_NE(stateful_partitioned_call_op, nullptr);
  auto xla_compile = NodeWith(Op("_XlaCompile"));

  EXPECT_TRUE(mlir::isa<mlir::DenseStringElementsAttr>(attr));
  auto string_attr = mlir::cast<mlir::DenseStringElementsAttr>(attr);
  auto string_values = string_attr.getRawStringData();
  ASSERT_EQ(string_values.size(), 4);
  EXPECT_EQ(string_values[0], mlir::StringRef("one"));
  EXPECT_EQ(string_values[1], mlir::StringRef("two"));
  EXPECT_EQ(string_values[2], mlir::StringRef("three"));

//   ...
//
// where pred_format is a function or functor that takes n (in the
// case of ASSERT_PRED_FORMATn) values and their source expression
// text, and returns a testing::AssertionResult.  See the definition
// of ASSERT_EQ in gtest.h for an example.
//
// If you don't care about formatting, you can use the more
// restrictive version:
//
//   ASSERT_PRED1(pred, v1)
//   ASSERT_PRED2(pred, v1, v2)
//   ...
//

      of any other Contributor, and only if You agree to indemnify,
      defend, and hold each Contributor harmless for any liability
      incurred by, or claims asserted against, such Contributor by reason
      of your accepting any such warranty or additional liability.

      of any other Contributor, and only if You agree to indemnify,
      defend, and hold each Contributor harmless for any liability
      incurred by, or claims asserted against, such Contributor by reason

//   ...
//
// where pred_format is a function or functor that takes n (in the
// case of ASSERT_PRED_FORMATn) values and their source expression
// text, and returns a testing::AssertionResult.  See the definition
// of ASSERT_EQ in gtest.h for an example.
//
// If you don't care about formatting, you can use the more
// restrictive version:
//
//   ASSERT_PRED1(pred, v1)
//   ASSERT_PRED2(pred, v1, v2)
//   ...
//