# RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-shape-inference-on-import=false %s -tf-input-arrays=x -tf-input-data-types=DT_INT32 -tf-input-shapes=10 -tf-output-arrays=func_call -o - | FileCheck %s

node {
  name: "x"
  op: "Const"
  attr {
    key: "dtype"
    value {
      type: DT_INT32
    }
  }
  attr {
    key: "value"
    value {
      tensor {
        dtype: DT_INT32
        tensor_shape {
          dim {

    // CHECK-NEXT:       %[[FUNC_CALL:[0-9]*]]:2 = func.call @__reduce_func_3(%[[ARG_11]], %[[ARG_12]], %[[GET_VALUE]]#0, %[[GET_VALUE]]#1, %[[ARG_13]], %[[ARG_14]])
    // CHECK-SAME:       _xla_compile_device_type = "TPU"
    // CHECK:            "tf.Yield"(%[[FUNC_CALL]]#0, %[[FUNC_CALL]]#1)
    // CHECK:            "tf.Yield"(%[[ARG_11]], %[[ARG_12]])

feed {
  id { node_name: "x_const" }
  shape {
    dim { size: 1 }
  }
}
feed {
  id { node_name: "y_const" }
  shape {
    dim { size: 1 }
  }
}
fetch {
  id { node_name: "func_call" }

# RUN: tf-mlir-translate -graphdef-to-mlir -tf-enable-shape-inference-on-import=false %s -tf-output-arrays=func_call -o - | FileCheck %s

node {
  name: "x"
  op: "VarHandleOp"
  device: "/CPU:0"
  attr {
    key: "container"
    value {
      s: "a"
    }
  }
  attr {
    key: "dtype"
    value {
      type: DT_INT64
    }
  }
  attr {
    key: "shape"
    value {
      shape {
      }
    }
  }

// RUN: tf-mlir-translate -mlir-to-graphdef %s -o - | FileCheck %s

func.func @main() -> tensor<*x!tf_type.resource> attributes {tf.entry_function = {inputs = "", outputs = "func_call"}} {
  %0 = tf_executor.graph {
    %outputs, %control = tf_executor.island wraps "tf.VarHandleOp"() {container = "a", device = "/CPU:0", dtype = i64, shape = "tfshape$", shared_name = "x"} : () -> tensor<!tf_type.resource<tensor<i64>>> loc("x")

    // CHECK: tf_executor.island wraps "tf._Retval"(%outputs_0) {T = !tf_type.resource, _mlir_name = "func_call", index = 0 : i64} : (tensor<*x!tf_type.resource>) -> ()
    %ctl_1 = _Retval(%test_func_name0) name("func_call") {T = !tf_type.resource, index = 0 : i64} : tensor<*x!tf_type.resource>
    // CHECK: tf_executor.fetch
  }

  @function.Defun(dtypes.int32, dtypes.int32)
  def test_func(a, b):
    return a + b

  x = constant_op.constant([1], name='x_const')
  y = constant_op.constant([2], name='y_const')
  test_func(x, y, name='func_call')  # pylint: disable=unexpected-keyword-arg


def tfsplits(_):
  """A more complex graph, including splits."""
  x = array_ops.placeholder(dtypes.float32, shape=[2, 2], name='x')

  FunctionDef call_identity = FunctionDefHelper::Create(
      "CallIdentity",
      /*in_def=*/{"x:float"},
      /*out_def=*/{"z:float"}, /*attr_def=*/{},
      /*node_def=*/
      {{{"func_call"},
        "PartitionedCall",
        {"x"},
        {{"Tin", DataTypeSlice({DT_FLOAT})},
         {"Tout", DataTypeSlice({DT_FLOAT})},
         {"f",

// errorcheck

// Copyright 2014 The Go Authors. All rights reserved.
// Use of this source code is governed by a BSD-style
// license that can be found in the LICENSE file.

// Check for cycles in a function call.

package funccall

func fx() int {
	_ = x
	return 0
}

	switch {
	case e.ListExpr != nil:
		return e.ListExpr.aggregateRow(r, tableAlias)
	case e.SubExpression != nil:
		return e.SubExpression.aggregateRow(r, tableAlias)
	case e.FuncCall != nil:
		return e.FuncCall.aggregateRow(r, tableAlias)
	}
	return nil
}

func (e *FuncExpr) aggregateRow(r Record, tableAlias string) error {
	switch e.getFunctionName() {