// CHECK-DAG: %[[CAST_0:.*]] = "tf.Cast"(%[[ARG_0]]) <{Truncate = false}> : (tensor<2xbf16>) -> tensor<2xf32>
// CHECK-DAG: %[[CAST_1:.*]] = "tf.Cast"(%[[ARG_1]]) <{Truncate = false}> : (tensor<2xbf16>) -> tensor<2xf32>
// CHECK: %[[ADD:.*]] = "tf.AddV2"(%[[CAST_0]], %[[CAST_1]]) : (tensor<2xf32>, tensor<2xf32>) -> tensor<2xf32>

}

// -----

// CHECK-LABEL: @acos
func.func @acos(%arg0: tensor<2xf32>) -> tensor<2xf32> {
  // CHECK-NOT:  tf.Acos
  %0 = "tf.Acos"(%arg0) : (tensor<2xf32>) -> tensor<2xf32>
  func.return %0 : tensor<2xf32>
}

// -----

// NOFALLBACK-LABEL: @xla_svd
func.func @xla_svd(%arg0: tensor<1x1xf32>) -> (tensor<1xf32>, tensor<1x1xf32>, tensor<1x1xf32>) {
  // NOFALLBACK: XlaSvd

# CHECK:  tf.TensorListReserve{{.*}}(tensor<2xi32>, tensor<i32>) -> tensor<!tf_type.variant<tensor<*x!tf_type.variant>>>

# CHECK:  tf.TensorListSetItem{{.*}}(tensor<!tf_type.variant<tensor<*xf32>>>, tensor<i32>, tensor<2x2xf32>) -> tensor<*x!tf_type.variant>

// CHECK-DAG: %[[input_18:.*]] = "quantfork.stats"({{.*}}) <{layerStats = dense<[-8.000000e-01, 1.600000e+00]> : tensor<2xf32>}> : (tensor<1x4xf32>) -> tensor<1x4xf32>
// CHECK-DAG: %[[input_19:.*]] = "quantfork.stats"({{.*}}) <{layerStats = dense<[-2.000000e+00, 4.000000e+00]> : tensor<2xf32>}> : (tensor<1x2xf32>) -> tensor<1x2xf32>

    }) {is_stateless = false} : (tensor<i32>, tensor<!tf_type.variant<tensor<?x1xf32>>>) -> (tensor<i32>, tensor<!tf_type.variant<tensor<?x1xf32>>>)
    %elem_1 = "tf._SomeOtherOp"() : () -> tensor<8x1xf32>
    %tl_set_item = "tf.TensorListSetItem"(%while#1, %one, %elem_1) : (tensor<!tf_type.variant<tensor<?x1xf32>>>, tensor<i32>, tensor<8x1xf32>) -> tensor<!tf_type.variant<tensor<?x1xf32>>>
    func.return
  }

  // CHECK: %[[OLD_SLICE2:.*]] = "tf.Slice"(%[[READ2]],
  // CHECK: %[[RESHAPE2:.*]] = "tf.Reshape"(%[[VALUE]],
  // CHECK: %[[ADD2:.*]] = "tf.AddV2"(%[[RESHAPE2]], %[[OLD_SLICE2]]) : (tensor<1x3xf32>, tensor<1x3xf32>) -> tensor<1x3xf32>
  // CHECK: %[[UPDATE2:.*]] = "tf.XlaDynamicUpdateSlice"(%[[READ2]], %[[ADD2]],

        # CHECK: value = dense<{{\[\[}}1.000000e+00, 2.000000e+00, 3.000000e+00], [4.000000e+00, 5.000000e+00, 6.000000e+00]]> : tensor<2x3xf32>
      }
    }
  }
}
node {
  name: "y"
  op: "Const"
  device: "/device:CPU:0"
  attr {
    key: "dtype"
    value {
      type: DT_INT64
    }
  }
  attr {

  %1 = "tf.BiasAdd"(%0, %cst_0) {data_format = "NHWC"} : (tensor<?x?x?x256xf32>, tensor<256xf32>) -> tensor<?x?x?x256xf32>
  %2 = "tf.Mul"(%1, %cst_1) : (tensor<?x?x?x256xf32>, tensor<256xf32>) -> tensor<?x?x?x256xf32>
  func.return %2 : tensor<?x?x?x256xf32>
}
// CHECK: func @conv2d_with_large_weight_and_mul

  %outputs =  "tf.Const"() {device = "/job:localhost/replica:0/task:0/device:CPU:0", value = dense<0> : tensor<i32>} : () -> tensor<i32>...

  %one = "glue.constant"() { value = 1: i32 } : () -> i32
  %done = "glue.compare" (%one, %one) { predicate = #glue<"compare LTE"> } : (i32, i32) -> i1
  %2 = mhlo.constant dense<[[1.1]]> : tensor<1x1xf32>
  %3 = mhlo.multiply %2, %2 : tensor<1x1xf32>
  %cst = "tf.Const"() {value = dense<0.0> : tensor<f32>} : () -> tensor<f32>
  %0 = "tf.AddV2"(%arg0, %cst) {device = "/device:CPU:0"} : (tensor<f32>, tensor<f32>) -> tensor<f32>