}

  // CHECK-LABEL: func @get_shape
  func.func @get_shape(%arg0 : tensor<*xi64>) -> tensor<?xi64> {
    %0 = "tf.Shape"(%arg0) : (tensor<*xi64>) -> tensor<?xi64>
    func.return %0 : tensor<?xi64>
  }

      : (tensor<?x!tf_type.string>, tensor<0x!tf_type.string>, tensor<0x!tf_type.string>, tensor<1x!tf_type.string>, tensor<0x!tf_type.string>, tensor<0xi64>)
      -> tensor<?xi64>

  func.return %dense_value : tensor<?xi64>

func.func @broadcast_both_operand(%arg0: tensor<7xf32>, %arg1: tensor<5x1xf32>) -> tensor<5x7xf32> {
  %cst = arith.constant dense<[5, 7]> : tensor<2xi64>
  %0 = "tf.BroadcastTo"(%arg0, %cst) : (tensor<7xf32>, tensor<2xi64>) -> tensor<5x7xf32>
  %1 = "tf.BroadcastTo"(%arg1, %cst) : (tensor<5x1xf32>, tensor<2xi64>) -> tensor<5x7xf32>
  %2 = "tf.Add"(%0, %1) : (tensor<5x7xf32>, tensor<5x7xf32>) -> tensor<5x7xf32>
  func.return %2 : tensor<5x7xf32>

}

func.func @scatter_nd_add(%arg0: tensor<7xi64>, %arg1: tensor<1x1xi32>, %arg2: tensor<1xi64>) -> tensor<7xi64> {
  %0 = "tf.TensorScatterAdd"(%arg0, %arg1, %arg2) : (tensor<7xi64>, tensor<1x1xi32>, tensor<1xi64>) -> tensor<7xi64>
  func.return %0 : tensor<7xi64>

  // CHECK-LABEL: scatter_nd_add
  // CHECK:  %[[GATHER:.*]] = "tf.GatherNd"(%arg0, %arg1) <{bad_indices_policy = ""}> : (tensor<7xi64>, tensor<1x1xi32>) -> tensor<1xi64>

  %1 = "tfl.shape"(%conv_s) : (tensor<?x26x26x26x16xf32>) -> tensor<5xi64>
  %2 = "tfl.broadcast_args"(%1, %cst) : (tensor<5xi64>, tensor<1xi64>) -> tensor<5xi64>
  %broad1 = "tfl.broadcast_to"(%conv_s, %2) : (tensor<?x26x26x26x16xf32>, tensor<5xi64>) -> tensor<?x26x26x26x16xf32>
  %broad2 = "tfl.broadcast_to"(%b, %2) : (tensor<16xf32>, tensor<5xi64>) -> tensor<?x26x26x26x16xf32>

// CHECK-DAG:       [[VAL_35:%.*]] = arith.constant dense<0> : tensor<1xi64>
// CHECK-DAG:       [[VAL_36:%.*]] = arith.constant dense<1> : tensor<1xi64>
// CHECK:           [[VAL_37:%.*]] = "tf.Slice"([[VAL_3]], [[VAL_35]], [[VAL_36]]) : (tensor<2xf32>, tensor<1xi64>, tensor<1xi64>) -> tensor<1xf32>
// CHECK-DAG:       [[VAL_38:%.*]] = arith.constant dense<1> : tensor<1xi64>

  // CHECK-LABEL: func @init_all_tables
  // CHECK-DAG: [[CST:%.*]]  = arith.constant dense<["apple", "banana", "grape"]> : tensor<3x!tf_type.string>
  // CHECK-DAG: [[CST_0:%.*]]  = arith.constant dense<[0, 1, 2]> : tensor<3xi64>
  // CHECK: [[VAL:%.*]] = "tf.HashTableV2"()
  // CHECK: "tf.LookupTableImportV2"([[VAL]], [[CST]], [[CST_0]])
}

// Tests that the tf.InitializeTableFromTextFileV2 op with explicit vocab size.

  // CHECK-DAG: %[[X_NORMED:.+]] = mhlo.multiply %[[X]], %[[MULTIPLIER_BCAST]] : tensor<4x256xf32>
  // CHECK-DAG: %[[RHS_BCAST:.+]] = "mhlo.broadcast_in_dim"(%[[RHS]]) <{broadcast_dimensions = dense<1> : tensor<1xi64>}> : (tensor<256xf32>) -> tensor<4x256xf32>

  // CHECK-NEXT:   %[[LHS_BCAST:.+]] = "mhlo.dynamic_broadcast_in_dim"(%arg0, %[[RESULT_EXTENTS]]) <{broadcast_dimensions = dense<1> : tensor<1xi64>}>
  // CHECK-NEXT:   %[[RHS_BCAST:.+]] = "mhlo.dynamic_broadcast_in_dim"(%arg1, %[[RESULT_EXTENTS]]) <{broadcast_dimensions = dense<[0, 1]> : tensor<2xi64>}>
  // CHECK-NEXT:   %[[RESULT:.+]] = mhlo.add %[[LHS_BCAST]], %[[RHS_BCAST]] : tensor<?x?xi32>
  // CHECK-NEXT:   shape.assuming_yield %[[RESULT]]

// CHECK:           [[LHSSHAPEEXTENTS:%.*]] = shape.to_extent_tensor [[LHSBCASTSHAPE]]
// CHECK:           [[LHSBCAST:%.*]] = "mhlo.dynamic_broadcast_in_dim"([[LHS]], [[LHSSHAPEEXTENTS]]) <{broadcast_dimensions = dense<[0, 1, 2]> : tensor<3xi64>}> : (tensor<1x4x2xf32>, tensor<3xindex>) -> tensor<3x4x2xf32>
// CHECK:           [[RHSBCASTSHAPE:%.*]] = shape.concat [[BCASTHEAD]], [[RHSTAIL]]
// CHECK:           [[RESULT:%.*]] = "mhlo.dot_general"([[LHSBCAST]], [[RHS]])