func.func @transpose_simple_2d() -> tensor<3x2xf32> {
  %0 = stablehlo.constant dense<[[0.000000e+0, 1.000000e+0, 2.000000e+0], [3.000000e+0, 4.000000e+0, 5.000000e+0]]> : tensor<2x3xf32>
  %1 = stablehlo.transpose %0, dims = [1, 0] : (tensor<2x3xf32>) -> tensor<3x2xf32>
  return %1 : tensor<3x2xf32>
}

  func.return %0: tensor<2x3x7xf32>
}

// CHECK-LABEL: testDynamicBatchMatMulToV2
func.func @testDynamicBatchMatMulToV2(%arg0: tensor<2x3x5xf32>, %arg1: tensor<?x5x7xf32>) -> tensor<2x3x7xf32> {
  // CHECK: "tf.BatchMatMul"(%arg0, %arg1) <{adj_x = false, adj_y = false}> {device = "/job:localhost/replica:0/task:0/device:GPU:0"}

    return %1 : tensor<1x3xf32>
  }

  func.func private @composite_dot_general_fn(%arg0: tensor<1x2xf32>, %arg1: tensor<2x3xf32>) -> tensor<1x3xf32> attributes {_from_xla_call_module} {
    %0 = stablehlo.dot_general %arg0, %arg1, contracting_dims = [1] x [0] : (tensor<1x2xf32>, tensor<2x3xf32>) -> tensor<1x3xf32>
    return %0 : tensor<1x3xf32>
  }
}

  %3 = "tfl.dequantize"(%2) : (tensor<1x3x3x!quant.uniform<i8:f32, 0.004:-128>>) -> tensor<1x3x3xf32>
  %4 = "tfl.div"(%arg0, %3) {fused_activation_function = "NONE"} : (tensor<1x3x3xf32>, tensor<1x3x3xf32>) -> tensor<1x3x3xf32>
  func.return %4 : tensor<1x3x3xf32>
//  CHECK: %[[logistic:.*]] = "tfl.logistic"
//  CHECK: %[[dq:.*]] = "tfl.dequantize"(%[[logistic]])

    // CHECK: %[[Q1:.*]] = "quantfork.qcast"(%[[ARG0]]) {volatile} : (tensor<1x2xf32>) -> tensor<1x2x!quant.uniform<i8:f32, 5.000000e-03>>
    // CHECK: %[[Q2:.*]] = "quantfork.qcast"(%[[ARG1]]) {volatile} : (tensor<2x3xf32>) -> tensor<2x3x!quant.uniform<i8<-127:127>:f32:1, {6.000000e-03,6.000000e-03,6.000000e-03}>>
    // CHECK: %[[CALL:.*]] = call @quantized_dot_general_fn_1(%[[Q1]], %[[Q2]])

  module {
    func.func @main(%arg0: tensor<2x2x2xf32>, %arg1: tensor<2x2x2xf32>) -> tensor<2x2x2xf32> attributes {_from_xla_call_module} {
      %0 = stablehlo.dot_general %arg0, %arg1,
        batching_dims = [0] x [0],
        contracting_dims = [2] x [1],
        precision = [DEFAULT, DEFAULT]
      : (tensor<2x2x2xf32>, tensor<2x2x2xf32>) -> tensor<2x2x2xf32>
        return %0 : tensor<2x2x2xf32>
    }
  }
)mlir";

   %0 = stablehlo.constant dense<1.000000e+00> : tensor<2x3xf32>
   %1 = "quantfork.qcast"(%0) {volatile} : (tensor<2x3xf32>) -> tensor<2x3x!quant.uniform<i8<-127:127>:f32, 5.000000e-03>>
   %2 = "quantfork.dcast"(%1) : (tensor<2x3x!quant.uniform<i8<-127:127>:f32, 5.000000e-03>>) -> tensor<2x3xf32>

^bb0(%arg0: tensor<2x3x5x!quant.uniform<u8:f32, 0.1>>, %arg1: tensor<3xi32>, %arg2: tensor<3xi32>):
  %0 = "tfl.dequantize"(%arg0) : (tensor<2x3x5x!quant.uniform<u8:f32, 0.1>>) -> tensor<2x3x5xf32>
  %1 = "tfl.slice"(%0, %arg1, %arg2) : (tensor<2x3x5xf32>, tensor<3xi32>, tensor<3xi32>) -> tensor<?x3x5xf32>
  func.return %1 : tensor<?x3x5xf32>

// CHECK: %0 = "tfl.dequantize"(%arg0)

      return %1 : tensor<1x3xf32>
    }

    func.func private @composite_dot_general_fn(%arg0: tensor<1x2xf32>, %arg1: tensor<2x3xf32>) -> tensor<1x3xf32> {
      %0 = stablehlo.dot_general %arg0, %arg1, contracting_dims = [1] x [0] : (tensor<1x2xf32>, tensor<2x3xf32>) -> tensor<1x3xf32>
      return %0 : tensor<1x3xf32>
    }
  )mlir";

// CHECK-NEXT:      %22 = mhlo.dynamic_reshape %2, %21 : (tensor<2x?x3x4xf32>, tensor<4xi32>) -> tensor<2x?x3x4xf32>
// CHECK-NEXT:      %23 = "tfl.batch_matmul"(%12, %22) <{adj_x = false, adj_y = false, asymmetric_quantize_inputs = false}> : (tensor<2x?x2x3xf32>, tensor<2x?x3x4xf32>) -> tensor<2x?x2x4xf32>
// CHECK-NEXT:      %24 = "tfl.shape"(%arg0) : (tensor<2x?x2x3xf32>) -> tensor<4xi32>