%1 = "tf.Conv2D"(%0, %cst) {data_format = "NHWC", device = "", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "SAME", strides = [1, 1, 2, 1], use_cudnn_on_gpu = true} : (tensor<1x3x4x3xf32>, tensor<2x3x3x2xf32>) -> tensor<1x3x2x2xf32>
    %2 = "tf.Relu"(%1) {device = ""} : (tensor<1x3x2x2xf32>) -> tensor<1x3x2x2xf32>

  %conv = "tf.Conv2D"(%dq_input, %dq_weight) {attr_map = "0:strides,1:use_cudnn_on_gpu,2:padding,3:explicit_paddings,4:dilations", data_format = "NHWC", device = "", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "VALID", strides = [1, 1, 2, 1], use_cudnn_on_gpu = true} : (tensor<1x3x4x3xf32>, tensor<2x3x3x2xf32>) -> tensor<*xf32>

  // CHECK-DAG: %[[RESULT:.+]] = mhlo.convolution(%[[INPUT]], %[[NEW_FILTER]]) dim_numbers = [b, 0, 1, f]x[0, 1, i, o]->[b, 0, 1, f], window = {stride = [1, 1], pad = {{\[\[}}0, 0], [0, 0]], rhs_dilate = [1, 1]} {batch_group_count = 1 : i64, feature_group_count = 1 : i64} : (tensor<1x256x256x3xf32>, tensor<1x1x3x2xf32>) -> tensor<1x256x256x2xf32>

    %0 = "tf.Cast"(%arg0) {Truncate = false, device = ""} : (tensor<1x3x4x3xf32>) -> tensor<1x3x4x3xf32>
    %1 = "tf.Conv2D"(%0, %cst) {data_format = "NHWC", dilations = [1, 1, 1, 1], explicit_paddings = [], padding = "SAME", strides = [1, 1, 2, 1], use_cudnn_on_gpu = true}
        : (tensor<1x3x4x3xf32>, tensor<2x3x3x2xf32>) -> tensor<1x3x2x2xf32> loc(fused["Conv2D:", "Model/conv2d"])

  func.return %0 : tensor<256x30x30x16xf32>
}

// -----

func.func @testConv2D(%arg0: tensor<256x32x32x3xf32>, %arg1: tensor<3x3x3x16xf32>) -> tensor<256x30x30x16xf32> {
  // expected-error @+1 {{requires positive strides}}

      CompileTimeConstantOperand<TF::MaxOp, 1>,  // $reduction_indices
      // $ksize, $strides
      CompileTimeConstantOperand<TF::MaxPoolGradGradV2Op, 3, 4>,
      // $ksize, $strides
      CompileTimeConstantOperand<TF::MaxPoolGradV2Op, 2, 3>,
      CompileTimeConstantOperand<TF::MaxPoolV2Op, 1, 2>,   // $ksize, $strides
      CompileTimeConstantOperand<TF::MeanOp, 1>,           // $reduction_indices

			} else if nrgba != nil {
				stride, pix = nrgba.Stride, nrgba.Pix
			}
			if stride != 0 {
				j0 := (y - b.Min.Y) * stride
				j1 := j0 + b.Dx()*4
				for j := j0; j < j1; j += 4 {
					cr0[i+0] = pix[j+0]
					cr0[i+1] = pix[j+1]
					cr0[i+2] = pix[j+2]
					i += 3
				}
			} else {
				for x := b.Min.X; x < b.Max.X; x++ {
					r, g, b, _ := m.At(x, y).RGBA()

		b[unzig[zig]] *= qt[zig]
	}
	idct(b)
	dst, stride := []byte(nil), 0
	if d.nComp == 1 {
		dst, stride = d.img1.Pix[8*(by*d.img1.Stride+bx):], d.img1.Stride
	} else {
		switch compIndex {
		case 0:
			dst, stride = d.img3.Y[8*(by*d.img3.YStride+bx):], d.img3.YStride
		case 1:
			dst, stride = d.img3.Cb[8*(by*d.img3.CStride+bx):], d.img3.CStride
		case 2:
			dst, stride = d.img3.Cr[8*(by*d.img3.CStride+bx):], d.img3.CStride

  %rst = "tf.Conv2D"(%arg, %fq) {T = "tfdtype$DT_FLOAT", data_format = "NHWC", dilations = [1, 2, 3, 1], padding = "SAME", strides = [1, 4, 5, 1]} : (tensor<256x32x32x3xf32>, tensor<3x3x3x16xf32>) -> tensor<256x8x7x16xf32>
  func.return %rst : tensor<256x8x7x16xf32>

  %cst2 = arith.constant dense<[1.0, 2.0]> : tensor<2xf32>
  %0 = "tf.Conv2D"(%arg0, %cst0) {T = "tfdtype$DT_FLOAT", data_format = "NHWC", dilations = [1, 2, 3, 1], padding = "SAME", strides = [1, 4, 5, 1]} : (tensor<1x112x112x3xf32>, tensor<1x3x3x2xf32>) -> tensor<1x28x23x2xf32>
  %1 = "tf.Mul"(%0, %cst2) : (tensor<1x28x23x2xf32>, tensor<2xf32>) -> tensor<1x28x23x2xf32>

  func.return %1 : tensor<1x28x23x2xf32>