%2 = "tf.Conv2D"(%1, %0) {data_format = "NHWC", device = "", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "SAME", strides = [1, 1, 2, 1], use_cudnn_on_gpu = true} : (tensor<*xf32>, tensor<*xf32>) -> tensor<*xf32>
  %3 = "tf.BiasAdd"(%2, %cst) {data_format = "NHWC", device = ""} : (tensor<*xf32>, tensor<2xf32>) -> tensor<*xf32>

  %5 = "tf.Conv2D"(%4, %cst) {_tpu_replicate = "cluster", data_format = "NHWC", device = "", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "SAME", strides = [1, 1, 2, 1], use_cudnn_on_gpu = true} : (tensor<1x3x4x3xbf16>, tensor<2x3x3x2xbf16>) -> tensor<1x3x2x2xbf16>
  %6 = "tf.TPUReplicatedOutput"(%5) {device = ""} : (tensor<1x3x2x2xbf16>) -> tensor<1x3x2x2xbf16>

	for i := 3; i > 6; i++ {
		a[i] = 0
	}
	for i := 7; i < 3; i-- {
		a[i] = 1
	}
}

func stride1(x *[7]int) int {
	s := 0
	for i := 0; i <= 8; i += 3 { // ERROR "Induction variable: limits \[0,6\], increment 3"
		s += x[i] // ERROR "Proved IsInBounds"
	}
	return s
}

func stride2(x *[7]int) int {
	s := 0
	for i := 0; i < 9; i += 3 { // ERROR "Induction variable: limits \[0,6\], increment 3"

// CHECK:         }
func.func @convert_slice(%arg0: tensor<1x4672xf32>) -> tensor<1x519xf32> {
  %0 = "mhlo.slice"(%arg0) <{limit_indices = dense<[1, 4672]> : tensor<2xi64>, start_indices = dense<[0, 4153]> : tensor<2xi64>, strides = dense<1> : tensor<2xi64>}> : (tensor<1x4672xf32>) -> tensor<1x519xf32>
  func.return %0 : tensor<1x519xf32>
}

// CHECK-LABEL:   func @reshape(

	}
	return
}

func testReadLine(t *testing.T, input []byte) {
	//for stride := 1; stride < len(input); stride++ {
	for stride := 1; stride < 2; stride++ {
		done := 0
		reader := testReader{input, stride}
		l := NewReaderSize(&reader, len(input)+1)
		for {
			line, isPrefix, err := l.ReadLine()
			if len(line) > 0 && err != nil {
				t.Errorf("ReadLine returned both data and error: %s", err)
			}

        offset = [0.5] * self.out_channel_size
        mean, variance = scale, offset
        out = nn_ops.conv2d(
            input_tensor,
            self.filters,
            strides=strides,
            dilations=dilations,
            padding=padding,
            data_format='NHWC',
        )
        if has_bias:
          out = nn_ops.bias_add(out, self.bias, data_format='NHWC')

    filter_shape = (2, 3, 3, 2)
    strides = (1, 1, 1, 1)
    model = self._create_conv2d_model(
        input_shape,
        filter_shape,
        self._input_saved_model_path,
        bias_fn,
        activation_fn,
        has_batch_norm,
        strides,
        dilations,
        'SAME',
        has_func_alias,
    )

  %0 = "mhlo.slice"(%arg0) <{limit_indices = dense<[1, 1, 512]> : tensor<3xi64>, start_indices = dense<[0, 0, 0]> : tensor<3xi64>, strides = dense<1> : tensor<3xi64>}> : (tensor<3x1x512xf32>) -> tensor<1x1x512xf32>
  %1 = "mhlo.slice"(%arg0) <{limit_indices = dense<[2, 1, 512]> : tensor<3xi64>, start_indices = dense<[1, 0, 0]> : tensor<3xi64>, strides = dense<1> : tensor<3xi64>}> : (tensor<3x1x512xf32>) -> tensor<1x1x512xf32>

		for iBase, y := 0, bounds.Min.Y; y < bounds.Max.Y; iBase, y = iBase+img.Stride, y+1 {
			for i, x := iBase+3, bounds.Min.X; x < bounds.Max.X; i, x = i+4, x+1 {
				img.Pix[i] = 255 - d.blackPix[(y-bounds.Min.Y)*d.blackStride+(x-bounds.Min.X)]
			}
		}
		return &image.CMYK{
			Pix:    img.Pix,
			Stride: img.Stride,
			Rect:   img.Rect,
		}, nil
	}

  %3 = stablehlo.convolution(%arg0, %0) dim_numbers = [b, 0, 1, f]x[0, 1, i, o]->[b, 0, 1, f], window = {stride = [1, 1], pad = [[1, 1], [1, 1]], rhs_dilate = [1, 1]}...