}

// CHECK-LABEL: @splat_scalar_no_change
func.func @splat_scalar_no_change() -> (tensor<f32>, tensor<i32>) {
  // CHECK-NOT: mhlo.broadcast_in_dim
  %cst0 = mhlo.constant dense<0.000000e+00> : tensor<f32>
  %cst1 = mhlo.constant dense<0> : tensor<i32>
  func.return %cst0, %cst1 : tensor<f32>, tensor<i32>

  %cst0 = mhlo.constant dense<[1.0, 2.0, 3.0, 4.0]> : tensor<4xf32>
  %cst1 = mhlo.constant dense<[[[[1.0, 2.0, 3.0, 4.0], [5.0, 6.0, 7.0, 8.0]]]]> : tensor<1x1x2x4xf32>
  %0 = "mhlo.broadcast_in_dim"(%cst0) <{broadcast_dimensions = dense<3> : tensor<1xi64>}> : (tensor<4xf32>) -> tensor<1x1x2x4xf32>
  %1 = mhlo.multiply %0, %cst1 : tensor<1x1x2x4xf32>
  // CHECK:      return %[[RES]] : tensor<1x1x2x4xf32>

  %cst0 = "tfl.pseudo_const" () {value = dense<0.0> : tensor<1x4xf32>} : () -> tensor<1x4xf32> loc("Const")
  %cst1 = "tfl.pseudo_const" () {value = dense<0.0> : tensor<1x4xf32>} : () -> tensor<1x4xf32> loc("Const")

    attributes {tf.entry_function = {outputs = "result"}} {
  %cst0 = arith.constant dense<5> : tensor<i32> loc("N")
  %cst1 = arith.constant dense<3.0> : tensor<1xf32> loc("val")

  %0:2 = "tfl.while"(%cst0, %cst1) ({
    ^bb0(%arg2: tensor<*xi32>, %arg3: tensor<*xf32>):
      %cst_0 = arith.constant dense<0> : tensor<i32>
      %1 = "tfl.greater"(%arg2, %cst_0) : (tensor<*xi32>, tensor<i32>) -> tensor<i1>

    // CHECK: %[[CST0:.*]] = stablehlo.constant dense<0xFF800000> : tensor<f32>
    // CHECK: %[[CST1:.*]] = stablehlo.constant dense<0xFF80000E> : tensor<2x3x1024x3xf32>
    // CHECK: %[[Q0:.*]] = "quantfork.qcast"(%[[CST0]])
    // CHECK: %[[Q1:.*]] = "quantfork.qcast"(%[[CST1]])
    // CHECK: %[[Q2:.*]] = "quantfork.qcast"(%[[ARG0]])

// CHECK-DAG:  %[[CST:.*]] = arith.constant dense<-2> : tensor<i32>
// CHECK-DAG:  %[[CST0:.*]] = arith.constant dense<1> : tensor<1xi32>
// CHECK-DAG:  %[[CST1:.*]] = arith.constant dense<0> : tensor<i32>
// CHECK:  %[[EXP:.*]] = "tf.ExpandDims"(%arg0, %[[CST]]) : (tensor<10x20x30xf32>, tensor<i32>) -> tensor<10x20x1x30xf32>

  // CHECK:      %[[CST0:.+]] = arith.constant dense<0.000000e+00>
  // CHECK-NEXT: %[[CST1:.+]] = arith.constant dense<0.000000e+00>
  // CHECK-NEXT: "vhlo.add_v1"(%[[CST0]], %[[CST1]]) : (tensor<f32>, tensor<f32>) -> tensor<f32>
  %cst_0 = arith.constant dense<0.000000e+00> : tensor<f32>
  %cst_1 = arith.constant dense<0.000000e+00> : tensor<f32>
  %0 = stablehlo.add %cst_0, %cst_1 : tensor<f32>
  return %0 : tensor<f32>

  %cst0 = arith.constant dense<[[1.0, 2.0], [3.0, 4.0]]> : tensor<2x2xf32>
  %cst1 = arith.constant dense<2.0> : tensor<2xf32>
  %cst2 = arith.constant dense<[1.0, 2.0]> : tensor<2xf32>
  %cst3 = arith.constant dense<[1, 8]> : tensor<2xi32>

// CHECK: %[[cst0:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<32x!quant.uniform<i32:f32, 7.812500e-04>>, value = dense<-1583> : tensor<32xi32>}>
// CHECK: %[[cst1:.*]] = "tfl.pseudo_qconst"() <{qtype = tensor<32x3x3x3x!quant.uniform<u8<1:255>:f32, 1.000000e-01>>, value = dense<1> : tensor<32x3x3x3xi8>}>
// CHECK: %[[conv:.*]] = "tfl.conv_2d"(%arg0, %[[cst1]], %[[cst0]])

  module attributes {tf.versions = {bad_consumers = [], min_consumer = 0 : i32, producer = 268 : i32}} {
    func.func @main() -> () {
      %cst0 = "tf.Const"(){ value = dense<0> : tensor<3x5xi1>} : () -> tensor<3x5xi1>
      %0 = "tf.Where"(%cst0) : (tensor<3x5xi1>) -> tensor<?x2xi64>
      func.return
    }
  })";

  CellReader<int64_t> dynamism_counter(kDynamismOpCounterStreamzName);