func.func @main(%arg0: tensor<3x3x10xbf16>, %arg1: tensor<3xi32>) -> tensor<1x?x4xbf16> attributes {allow_soft_placement = false, tf.entry_function = {control_outputs = "", inputs = "_arg0,_arg1,_arg2", outputs = "_retval0"}} {
      %cst = "tf.Const"() {value = dense<[1, -1, 4]> : tensor<3xi32>} : () -> tensor<3xi32>

// CHECK-SAME: %[[STATE:.*]]: tensor<2xui64>
func.func @xla_rng_bit_generator(%arg0: tensor<2xui64>) -> (tensor<2xui64>, tensor<10x12xui32>) attributes {tf.entry_function = {control_outputs = "", inputs = "_arg0,_arg1,_arg2", outputs = "_retval0,_retval1"}} {
  // CHECK-NEXT: %0 = mhlo.constant dense<[10, 12]> : tensor<2xi32>
  %cst = "tf.Const"() {value = dense<[10, 12]> : tensor<2xi32>} : () -> tensor<2xi32>
  // CHECK-NEXT: %1 = mhlo.constant dense<3> : tensor<i32>

  might become the name of Retval nodes as well (with an index suffix if there
  are multiple output tensors from one node). Since Retval nodes are not used in
  SavedModel, this function removes them and restore the names to the actual
  output tensors.

  Args:
    graph_def: the converted GraphDef.

  Returns:
    The GraphDef with Retval nodes removed and output tensor names restored.

  for (auto sharding_and_retval :
       llvm::zip(sharding_for_rets, terminator->getOpOperands())) {
    const auto& sharding = std::get<0>(sharding_and_retval);
    OpOperand& retval = std::get<1>(sharding_and_retval);
    if (failed(VerifySharding(retval.get().getType(), sharding)))
      return mlir::failure();
  }
  return mlir::success();
}

// Assign the logical device if an op has an attribute `TPU_REPLICATED_CORE:n`,

// CHECK-NEXT:    "tf.IfrtCall"(%arg0, [[ARRAYKEY]]) <{program_id = 6515870160938153680 : i64, variable_arg_indices = [1 : i32]}> {__tpu_compile_metadata_text = "retvals { sharding { } }"} : (tensor<1x3xf32>, tensor<!tf_type.string>) -> tensor<1x1xf32>
// CHECK-NEXT:    return
//
 func.func @serving_default(%arg0: tensor<1x3xf32>) -> tensor<1x1xf32> {

				}
				if retVal, err = util.MergeAnyWithAny(filter.GetTypedConfig(), userFilter.GetTypedConfig()); err != nil {
					retVal = filter.GetTypedConfig()
				}
			}
			filter.Name = filterName
			if retVal != nil {
				filter.ConfigType = &listener.Filter_TypedConfig{TypedConfig: retVal}
			}
		}
	}
	if filter.Name == wellknown.HTTPConnectionManager {

	if err != nil {
		return nil, err
	}
	retval := make([]*v1alpha1.IstioOperator, 0)
	for _, un := range ul.Items {
		fixTimestampRelatedUnmarshalIssues(&un)
		by := util.ToYAML(un.Object)
		iop, err := operator_istio.UnmarshalIstioOperator(by, true)
		if err != nil {
			return nil, err
		}
		retval = append(retval, iop)
	}
	return retval, nil
}

                             int* num_retvals) override {
    std::vector<TFE_TensorHandle*> outputs(*num_retvals);
    TF_Status status;
    device_.execute(tensorflow::wrap(op), num_retvals, outputs.data(), &status,
                    info_);
    if (status.status.ok()) {
      for (int i = 0; i < *num_retvals; ++i) {
        retvals[i] = tensorflow::unwrap(outputs[i]);

}

Status MlirAbstractOp::Execute(absl::Span<AbstractTensorHandle*> retvals,
                               int* num_retvals) {
  OperationState* state;
  TF_RETURN_IF_ERROR(Create(operands_, &state));
  Operation* op = function_context_->CreateOperationFromState(*state);
  *num_retvals = op->getNumResults();
  for (int i = 0; i < *num_retvals; i++)
    retvals[i] = new MlirTensor(op->getResult(i));
  return absl::OkStatus();
}

// Note: this list must be updated if ever kubelet wants to allow mutations to other fields.
func pickFieldsToHash(container *v1.Container) map[string]string {
	retval := map[string]string{
		"name":  container.Name,
		"image": container.Image,
	}
	return retval
}

// envVarsToMap constructs a map of environment name to value from a slice
// of env vars.
func envVarsToMap(envs []EnvVar) map[string]string {