%1 = stablehlo.constant dense<[4.0, 6.0]> : tensor<2xf32>
  %2 = stablehlo.divide %0, %1 : tensor<2xf32>
  return %2 : tensor<2xf32>
}

// CHECK-LABEL: foldDivLHSSplat
// CHECK: stablehlo.constant dense<[5.000000e-01, 0.333333343]> : tensor<2xf32>

// -----

func.func @foldDivRHSSplat() -> tensor<2xf32> {
  %0 = stablehlo.constant dense<[4.0, 6.0]> : tensor<2xf32>
  %1 = stablehlo.constant dense<2.0> : tensor<2xf32>

  func.return %1 : tensor<4x4x4xf32>
}

func.func @testMultipleLstms(%arg0: tensor<4x4xf32>, %arg1: tensor<4xf32>, %arg2: tensor<4x4x4xf32>) -> tensor<4x4x4xf32> {

// CHECK: (%[[ARG:.*]]: tensor<4xf32>) -> (tensor<4xf32>, tensor<4xf32>, tensor<4xf32>)
func.func @main(%arg0: tensor<4xf32>) -> tensor<4xf32> attributes {tf.entry_function = {inputs = "mul"}} {
  %0 = "tfl.pseudo_const" () {value = dense<1.0> : tensor<4xf32>} : () -> tensor<4xf32> loc("Const")
  %1 = "tfl.squared_difference"(%arg0, %0) {fused_activation_function = "NONE"} : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32> loc("squared_difference")

  // CHECK: %[[DIV:.*]] = tfl.div
  %3 = "tfl.div"(%2, %1) {fused_activation_function = "NONE"} : (tensor<4xf32>, tensor<4xf32>) -> tensor<4xf32> loc("div")
  // CHECK: %[[EXP:.*]] = "tfl.exp"
  %4 = "tfl.exp"(%3) : (tensor<4xf32>) -> tensor<4xf32> loc("exp")
  // tfl.neg should be pruned
  // CHECK-NOT: "tfl.neg"
  %5 = "tfl.neg"(%4) : (tensor<4xf32>) -> tensor<4xf32> loc("neg")

    %4 = "quantfork.stats"(%3) {layerStats = dense<[0.000000e+00, 0.18486841]> : tensor<2xf32>} : (tensor<*xf32>) -> tensor<*xf32>
    %5 = "tf.Identity"(%4) {device = ""} : (tensor<*xf32>) -> tensor<*xf32>
    %6 = "quantfork.stats"(%5) {layerStats = dense<[0.000000e+00, 0.18486841]> : tensor<2xf32>} : (tensor<*xf32>) -> tensor<*xf32>

    // expected-remark@+1 {{Cost: 10}}
    %44 = "tf.ConcatV2"(%36, %37, %38, %39, %40, %41, %42, %43, %10) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<?x1xf32>, tensor<?x1xf32>, tensor<?x1xf32>, tensor<?x1xf32>, tensor<?x1xf32>, tensor<?x1xf32>, tensor<?x1xf32>, tensor<?x1xf32>, tensor<i32>) -> tensor<?x8xf32>

// CHECK-NEXT:  } ]
// CHECK-NEXT:  signature_defs: [ ]
// CHECK-NEXT:}

func.func @main(%arg0: tensor<4xf32>, %arg1: tensor<4x!quant.uniform<u8:f32, 0.1>>) -> tensor<4xf32> {
  "tfl.NumericVerify"(%arg1, %arg0) {tolerance = 0.1 : f32} : (tensor<4x!quant.uniform<u8:f32, 0.1>>, tensor<4xf32>) -> (tensor<4xf32>)
  func.return %arg0 : tensor<4xf32>

module{
  func.func @main(%arg0: tensor<10xf32>) -> tensor<10xf32> {
    return %arg0 : tensor<10xf32>
  }
  func.func @func1(%arg0: tensor<10xf32>, %arg1: tensor<10xf32>) -> tensor<10xf32> {
    %0 = stablehlo.add %arg0, %arg1 : tensor<10xf32>
    %1 = stablehlo.add %arg0, %arg1 : tensor<10xf32>
    return %0 : tensor<10xf32>
  }
})mlir";

    func.func @main() -> tensor<1xi32> {
      %0 = "tf.BadValue"() {value = dense<1000> : tensor<1xi32>} : () -> tensor<1xi32>
      %1 = "tf.AlsoBad"() {value = dense<10> : tensor<1xi32>} : () -> tensor<1xi32>
      func.return %0 : tensor<1xi32>
    }
  })";
  CellReader<int64_t> error(kFailedLegalizationStreamz);
  CreateModule(kMultipleFailures);

    tf_executor.graph {
      %out, %ctl = tf_executor.island wraps "tf.VarHandleOp"() {shared_name = "var_0"} : () -> tensor<!tf_type.resource<tensor<2xf32>>>
      %out_0, %ctl_0 = tf_executor.island wraps "tf.ReadVariableOp"(%out) : (tensor<!tf_type.resource<tensor<2xf32>>>) -> tensor<2xf32>
      %out_1, %ctl_1 = tf_executor.island wraps "tf.Const"() {value = dense<"var_0"> : tensor<1x!tf_type.string>} : () -> tensor<1x!tf_type.string>