%input_scale = "tf.Const"() {value = dense<1.0> : tensor<f32>} : () -> tensor<f32>
  %side_input_scale = "tf.Const"() {value = dense<2.0> : tensor<f32>} : () -> tensor<f32>
  %conv2d = "tf._FusedConv2D"(%input, %filter, %bias, %act, %input_scale, %side_input_scale) {

                    <conf name="conf2" />
                </configurations>
                <publications defaultconf="conf2">
                    <artifact/>
                </publications>
            </ivy-module>
        """

        when:
        parse(parseContext, file)

        then:
        artifact().configurations == ["conf2"] as Set
    }

    %clamp_min = "tf.Minimum"(%clamp_max, %clip_max) : (tensor<*xf32>, tensor<*xf32>) -> tensor<*xf32>

    func.return %clamp_min : tensor<*xf32>
  }

  // Conv2D with int32 accumulation.
  func.func private @internal_conv2d_fn(
                         %input : tensor<*xi8>, %filter : tensor<*xi8>,
                         %input_scale : tensor<*xf32>, %input_zp : tensor<*xi32>,

// issue 7205

var i9 interface{}

func f9() bool {
	g8()
	x := i9
	y := interface{}(g18()) // ERROR "live at call to convT: x.data$" "live at call to g18: x.data$" "stack object .autotmp_[0-9]+ \[2\]string$"
	i9 = y                  // make y escape so the line above has to call convT
	return x != y
}

// liveness formerly confused by UNDEF followed by RET,
// leading to "live at entry to f10: ~r1" (unnamed result).

// func.func func_name_${key1}_fn (...) {
//   ...${key2}...
// }
// ```
// The above template with generate two functions by substituting `key1` and
// `key2` with given values.

module {

  for main_op in ["Conv2D", "DepthwiseConv2D", "MatMul"] {
    parameters[
      {"quantized_ops": ["${main_op}", "BiasAdd"], "act_func": "internal_requantize_no_activation_fn", "output_type": "!tf_type.qint8"},

  // Canonicalization includes const folding, which is utilized here to optimize
  // away ops that can't get constant folded after PrepareTF pass. For example,
  // tf.Conv2D is split into tf.Transpose and tfl.Conv2D.
  pass_manager->addNestedPass<mlir::func::FuncOp>(
      mlir::createCanonicalizerPass());
  pass_manager->addNestedPass<mlir::func::FuncOp>(mlir::createCSEPass());

// functions fail on a nil input.

// convT converts a value of type t, which is pointed to by v, to a pointer that can
// be used as the second word of an interface value.
func convT(t *_type, v unsafe.Pointer) unsafe.Pointer {
	if raceenabled {
		raceReadObjectPC(t, v, getcallerpc(), abi.FuncPCABIInternal(convT))
	}
	if msanenabled {
		msanread(v, t.Size_)
	}
	if asanenabled {

// CHECK: %[[DEQUANTIZE:.*]] = "tfl.dequantize"(%[[QUANTIZE]])
// CHECK: %[[CONV:.*]] = "tfl.conv_2d"(%arg0, %[[DEQUANTIZE]], %[[CONSTANT]])
// CHECK: return %[[CONV]]
}

// CHECK-LABEL: perChannelFakeQuantWithConv2D
func.func @perChannelFakeQuantWithConv2D(tensor<256x32x32x3xf32>) -> (tensor<256x8x7x16xf32>) {

  %0 = "tf.Conv2D"(%output, %cst) <{data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "SAME", strides = [1, 2, 2, 1], use_cudnn_on_gpu = true}> {attr_map = "0:strides,1:use_cudnn_on_gpu,2:padding,3:explicit_paddings,4:dilations", device...

  %result = "tf.Select"(%cond, %out_matrix, %matrix) : (tensor<i1>, tensor<3x3xf32>, tensor<3x3xf32>) -> tensor<3x3xf32>
  %cond2 = "tf.Identity"(%cond) : (tensor<i1>) -> tensor<i1>
  %result2 = "tf.Select"(%cond2, %out_matrix2, %matrix_2) : (tensor<i1>, tensor<3x3xf32>, tensor<3x3xf32>) -> tensor<3x3xf32>