func.return
}

// CHECK:      tf_device.replicate
// CHECK-SAME: [%[[ARG_0]], %[[ARG_0]]] as %{{[a-z0-9]*}}
// CHECK-SAME: [%[[ARG_2]], %[[ARG_2]]] as %{{[a-z0-9]*}}
// CHECK-SAME: [%[[ARG_4]], %[[ARG_4]]] as %{{[a-z0-9]*}}
// CHECK-SAME: %[[ARG_1]] as %{{[a-z0-9]*}}
// CHECK-SAME: %[[ARG_3]] as %{{[a-z0-9]*}}
// CHECK-SAME: %[[ARG_5]] as %{{[a-z0-9]*}}

}

// CHECK-LABEL: @conv_3_fn
// CHECK: tf.XlaCallModule
// CHECK-SAME: _entry_function = @composite_conv_fn_6, _original_entry_function = "composite_conv_fn_6"
// CHECK: tf.XlaCallModule
// CHECK-SAME: _entry_function = @composite_conv_fn_5, _original_entry_function = "composite_conv_fn_5"

// CHECK-LABEL: @conv_1_fn
// CHECK: tf.XlaCallModule

  // CHECK: %[[RESULT:.*]] = "tf.StridedSlice"(%arg0, %[[CST]], %[[CST0]], %[[CST1]])
  // CHECK-SAME: begin_mask = 7 : i64
  // CHECK-SAME: ellipsis_mask = 0 : i64
  // CHECK-SAME: end_mask = 14 : i64
  // CHECK-SAME: new_axis_mask = 0 : i64
  // CHECK-SAME: shrink_axis_mask = 0 : i64

// CHECK-SAME: %[[none]], %[[input_10]], %[[input_11]], %[[none]], %[[input_13]], %[[input_14]], %[[input_15]], %[[input_16]], %[[input_17]], %[[input_18]], %[[input_19]],
// CHECK-SAME: %[[none]], %[[input_21]], %[[input_22]], %[[input_23]])
// CHECK-SAME: effective_hidden_scale_intermediate = tensor<!quant.uniform<i8:f32, 0.0039215686274509803:-1>>

!tf_res = tensor<!tf_type.resource<tensor<f32>>>

// Tests independent chains of two resources.

// CHECK-LABEL: func @simple_independent_chains_while_body

  // CHECK-SAME: value = dense<true> : tensor<i1>
  %0 = "tf.VarIsInitializedOp"(%arg0) : (!tf_res) -> tensor<i1>
  // CHECK-NEXT: return [[TRUE]] :
  func.return %0 : tensor<i1>
}

// CHECK-LABEL: func @while_body
// CHECK-SAME: ({{.+}}: tensor<i1>)
func.func @while_body(%arg0: !tf_res, %arg1: tensor<i1>) -> (!tf_res, tensor<i1>) {
  // CHECK: [[TRUE:%.+]] = "tf.Const"

    // CHECK:      %[[ONE:.*]] = "tf.Const"() <{value = dense<1> : tensor<i32>}>
    // CHECK:      %[[RES_READ_VAL:[0-9]*]] = "tf.ReadVariableOp"
    // CHECK-SAME: (tensor<*x!tf_type.resource<tensor<2x8xi32>>>) -> tensor<2x8xi32>
    // CHECK:      "tf.AddV2"(%[[RES_READ_VAL]], %[[ONE]])
    // CHECK-SAME: (tensor<2x8xi32>, tensor<i32>) -> tensor<2x8xi32>
    // CHECK:      "tf.AssignVariableOp"

  // CHECK: "tf.StatefulPartitionedCall"(%[[VAR]], %[[GVAR1]], %[[GVAR2]])
  // CHECK-SAME: f = @callee_tensorarray_decomposed
  %call = "tf.StatefulPartitionedCall"(%ta#0) {f = @callee, config = "", config_proto = "", executor_type = ""}
    : (tensor<!tf_type.resource>) -> tensor<!tf_type.resource>
  // CHECK: "tf.PartitionedCall"(%[[VAR]], %[[GVAR1]], %[[GVAR2]])
  // CHECK-SAME: f = @callee_tensorarray_decomposed

// back to a pessimistic analysis.  The optimistic analysis assigns the same
// symbolic predicate to all the merge nodes whose preceding enter nodes have
// the same frame name on the first iteration.  On the second iteration, if all
// the merge nodes are pattern matched into the same AndRecurrence predicate
// instance, the optimistic assignment of the same symbolic predicate is correct
// and the analyzed result is taken.
//

	if add2Err != nil {
		t.Fatalf("AddVolumeNode failed. Expected: <no error> Actual: <%v>", add2Err)
	}

	if generatedVolumeName != generatedVolumeName2 {
		t.Fatalf(
			"Generated volume names for the same volume should be the same but they are not: %q and %q",
			generatedVolumeName,
			generatedVolumeName2)
	}

	volumeNodeComboState = asw.GetAttachState(generatedVolumeName, nodeName)