%cst = "tf.Const"() {value = dense<0.000000e+00> : tensor<10xf32>} : () -> tensor<10xf32>
    %cst_0 = "tf.Const"() {value = dense<[-1, 10]> : tensor<2xi32>} : () -> tensor<2xi32>
    %1 = "tf.MatMul"(%arg0, %arg1) {
      transpose_a = false, transpose_b = false

// CHECK-NOT: call

// -----

func.func @handles_mhlo_regions() {
  "glue.someop"() : () -> ()
  %4 = mhlo.constant dense<0.0> : tensor<200x10xf32>
  %13 = mhlo.constant dense<-0.0> : tensor<f32>
  %29 = mhlo.reduce(%4 init: %13) applies mhlo.add across dimensions = [1] : (tensor<200x10xf32>, tensor<f32>) -> tensor<200xf32>
  return
}

// CHECK: func @handles_mhlo_regions
// CHECK: call [[name:@[^(]*]](

// -----

module {
  func.func @same_scale_ptq_test(%arg0: tensor<*xf32>) -> tensor<*xf32> {
    %cst = arith.constant dense<[-1, 144]> : tensor<2xi32>
    %cst_1 = arith.constant dense<1.0> : tensor<144x10xf32>
    %cst_2 = arith.constant dense<0.1> : tensor<10xf32>
    %0 = "quantfork.stats"(%arg0) {
      layerStats = dense<[-1.28, 1.28]> : tensor<2xf32>
    } : (tensor<*xf32>) -> tensor<*xf32>
    %1 = "tf.MaxPool"(%0) {

// CHECK: return %[[identity]] : tensor<1x3x2x2xf32>

func.func @cast_bf16_matmul_to_fp32(%arg0: tensor<1x10xf32>) -> (tensor<1x2xf32>) {
  %cst = "tf.Const"() {device = "", value = dense<1.000000e+01> : tensor<10x2xbf16>} : () -> tensor<10x2xbf16>
  %0 = "tf.Cast"(%arg0) {Truncate = false, device = ""} : (tensor<1x10xf32>) -> tensor<1x10xbf16>

  %11 = "mhlo.dot"(%10, %5) : (tensor<32x3920xf32>, tensor<3920x10xf32>) -> tensor<32x10xf32>
  %12 = "mhlo.broadcast_in_dim"(%2) <{broadcast_dimensions = dense<1> : tensor<1xi64>}> : (tensor<10xf32>) -> tensor<32x10xf32>
  %13 = mhlo.add %11, %12 : tensor<32x10xf32>
  %14 = mhlo.maximum %13, %0 : tensor<32x10xf32>
  return %14 : tensor<32x10xf32>
}
// CHECK: ReadVariableOp
// CHECK: mhlo.convolution
// CHECK: mhlo.add