%cst = arith.constant dense<[false, true]> : tensor<2xi1>
  %0 = "tfl.cast"(%cst) : (tensor<2xi1>) -> tensor<2xi8>
  func.return %0 : tensor<2xi8>

// CHECK: %[[CST:.*]] = arith.constant dense<[0, 1]> : tensor<2xi8>
// CHECK:  return %[[CST]]
}

// CHECK-LABEL: @cast_i1_to_ui8
func.func @cast_i1_to_ui8() -> tensor<2xui8> {
  %cst = arith.constant dense<[false, true]> : tensor<2xi1>

module {
  func.func @not_propagate_matmul(%arg0: tensor<1x2x2x2xf32>) -> tensor<*xf32> {
    %cst = "tf.Const"() {value = dense<127> : tensor<2x1024xi8>} : () -> tensor<2x1024xi8>
    %cst_0 = "tf.Const"() {value = dense<0.0157480314> : tensor<f32>} : () -> tensor<f32>
    %0 = "tf.Identity"(%cst) : (tensor<2x1024xi8>) -> tensor<2x1024xi8>
    %1 = "tf.Cast"(%0) {Truncate = false} : (tensor<2x1024xi8>) -> tensor<2x1024xf32>

  %cst = "tf.Const"() {device = "", value = dense<[1]> : tensor<1xi32>} : () -> tensor<1xi32>
  // idx 2 should be a compile time constant
  %0 = "tf.GatherV2"(%arg0, %arg1, %cst) {batch_dims = 1: i64} : (tensor<16x4xf32>, tensor<16xi32>, tensor<1xi32>) -> tensor<16xf32>

  // Just to introduce an extra use for %cst.
  %1 = "tf.AddV2"(%cst, %cst) {device = ""} : (tensor<1xi32>, tensor<1xi32>) -> tensor<1xi32>

  func.return %0 : tensor<3x5x4xf32>
  // CHECK-LABEL: unary_einsum_reduce_sum_transpose
  // CHECK-DAG: %[[cst:.*]] = arith.constant dense<3> : tensor<1xi32>
  // CHECK-DAG: %[[cst_1:.*]] = arith.constant dense<[0, 2, 1]> : tensor<3xi32>
  // CHECK: %[[v0:.*]] = "tf.Sum"(%arg0, %[[cst]]) <{keep_dims = false}> : (tensor<3x4x5x6xf32>, tensor<1xi32>) -> tensor<3x4x5xf32>

  %0 = "tfl.transpose"(%arg0, %perm) : (tensor<2x3x4x5xf32>, tensor<4xi32>) -> tensor<5x2x3x4xf32>
  %cst = arith.constant dense<1.0> : tensor<5x2x3x4xf32>
  %1 = tfl.add %0, %cst { fused_activation_function = "NONE" } : tensor<5x2x3x4xf32>
  func.return %1 : tensor<5x2x3x4xf32>
}

// CHECK: %cst = arith.constant dense<[3, 0, 1, 2]> : tensor<4xi32>
// CHECK: %cst_0 = arith.constant dense<1.000000e+00> : tensor<2x3x4x5xf32>

func.func @lift_float_batch_matmul(%arg0: tensor<4x4x3xf32>) -> (tensor<4x4x3xf32>) {
  %cst = "tf.Const"() {device = "", value = dense<1.0> : tensor<4x3x3xf32>} : () -> tensor<4x3x3xf32>
  %0 = "tf.BatchMatMulV2"(%arg0, %cst) {adj_x = false, adj_y = false, device = ""} : (tensor<4x4x3xf32>, tensor<4x3x3xf32>) -> tensor<4x4x3xf32>
  return %0 : tensor<4x4x3xf32>

// CHECK-DAG: %[[CST:.*]] = "tf.Const"() {{.*}} : () -> tensor<4x3x3xf32>

// CHECK-LABEL: func.func @consecutive_add_add

// CHECK: %[[CST:.*]] = "tf.Const"() <{value = dense<-30> : tensor<i32>}> : () -> tensor<i32>
// CHECK: %[[ADD:.*]] = "tf.AddV2"(%arg0, %[[CST]]) : (tensor<i32>, tensor<i32>) -> tensor<i32>
// CHECK: return %[[ADD]] : tensor<i32>
}

func.func @consecutive_add_sub(%arg0: tensor<i32>) -> (tensor<i32>) {
  %cst = "tf.Const"() {value = dense<-18> : tensor<i32>} : () -> tensor<i32>

// CHECK-DAG: %[[cst:.*]] = "tf.Const"() <{value = dense<{{.*}}> : tensor<10x2xf32>}> : () -> tensor<10x2xf32>
// CHECK: %[[matmul:.*]] = "tf.MatMul"(%arg0, %[[cst]])
// CHECK: %[[identity:.*]] = "tf.IdentityN"(%[[matmul]])
// CHECK: return %[[identity]] : tensor<1x2xf32>

func.func @cast_bf16_depthwise_conv_to_fp32(%arg0: tensor<1x3x4x3xf32>) -> (tensor<1x2x2x6xf32>) {

func.func @NotQuantizeNoneType() -> none {
  %cst = "tfl.no_value"() {value = unit} : () -> none
  func.return %cst : none

// CHECK-NEXT:  %[[cst:.*]] = "tfl.no_value"() <{value}> : () -> none
// CHECK-NEXT:  return %[[cst]]
}

// CHECK-LABEL: QuantizeZeroSplat
func.func @QuantizeZeroSplat() -> tensor<2x3xf32> {
  %cst = arith.constant dense<0.0> : tensor<2x3xf32>
  func.return %cst : tensor<2x3xf32>

func.func @init_all_tables() {
  %cst = arith.constant dense<"%FILE_PLACEHOLDER"> : tensor<!tf_type.string>
  %0 = "tf.HashTableV2"() {container = "", device = "", key_dtype = !tf_type.string, shared_name = "hash_table_/tmp/vocab.txt_-2_-1", use_node_name_sharing = false, value_dtype = i64} : () -> tensor<!tf_type.resource>