%x = "tf.TensorArrayReadV3"(%handle_0, %index, %flow_0) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<2x!tf_type.resource<tensor<?x100xf32>>>, tensor<i32>, tensor<f32>) -> tensor<?x100xf32>
  %y = "tf.MatMul"(%x, %cst) {device = "/job:localhost/replica:0/task:0/device:CPU:0"} : (tensor<?x100xf32>, tensor<100x512xf32>) -> (tensor<?x512xf32>)

// CHECK:         }
func.func @broadcast_add(%arg0: tensor<1x1xf32>, %arg1: tensor<1x1000xf32>) -> (tensor<1x1000xf32>, tensor<1x1000xf32>) {
  %0 = "mhlo.broadcast_in_dim"(%arg0) <{broadcast_dimensions = dense<[0, 1]> : tensor<2xi64>}> : (tensor<1x1xf32>) -> tensor<1x1000xf32>
  %1 = mhlo.add %0, %arg1 : tensor<1x1000xf32>
  %2 = mhlo.add %arg1, %0 : tensor<1x1000xf32>

  %166 = "tf.MatMul"(%arg0, %arg1) {T = "tfdtype$DT_FLOAT", _output_shapes = ["tfshape$dim { size = 1} dim { size = 1000}"], device = "", name = "matmul", transpose_a = false, transpose_b = false} : (tensor<1x1280xf32>, tensor<1280x1000xf32>) -> tensor<1x1000xf32>
  func.return %166 : tensor<1x1000xf32>

  // CHECK-LABEL: matmulNoTransposeAOrB

  func.return %1 : tensor<2x1024xf32>

  // CHECK: tfl.mul
}

// CHECK-LABEL: @FuseMulWithFullyConnectedNoBias
func.func @FuseMulWithFullyConnectedNoBias(%arg: tensor<2x512xf32>) -> tensor<2x1024xf32> {
  %cst_mul = arith.constant dense<2.0> : tensor<512xf32>

  func.return %0 : tensor<2x1024xf32>
  // CHECK-DAG: %[[CST:.*]] = arith.constant dense<1.000000e+00> : tensor<2x512xf32>
  // CHECK-DAG: %[[CST_0:.*]] = arith.constant dense<2.000000e+00> : tensor<1024x512xf32>

// CHECK: %[[v0:.*]] = "tf.Reshape"(%arg0, %[[cst]]) : (tensor<2x1x1x11xf32>, tensor<3xi64>) -> tensor<2x1x11xf32>
// CHECK: %[[v1:.*]] = "tf.BatchMatMulV2"(%[[v0]], %arg1) <{adj_x = false, adj_y = false}> : (tensor<2x1x11xf32>, tensor<2x11x2xf32>) -> tensor<2x1x2xf32>
// CHECK: %[[v2:.*]] = "tf.Reshape"(%[[v1]], %[[cst_1]]) : (tensor<2x1x2xf32>, tensor<4xi64>) -> tensor<2x1x1x2xf32>

func.func @infeed_dequeue_tuple() -> (tensor<1x8x4x4xi32>, tensor<1x100x1xf32>) {
// CHECK: [[TOKEN:%.*]] = mhlo.create_token  : !mhlo.token
// CHECK: [[INFEED:%.*]]:3 = "mhlo.infeed"([[TOKEN]]) <{infeed_config = ""{{.*}}}> : (!mhlo.token) -> (tensor<1x8x4x4xi32>, tensor<1x100x1xf32>, !mhlo.token)
// CHECK: return [[INFEED]]#0, [[INFEED]]#1
  %0:2 = "tf.InfeedDequeueTuple"() : () -> (tensor<1x8x4x4xi32>, tensor<1x100x1xf32>)

}

// -----

func.func @testBatchMatMulV2InvalidOutputBatchDimension(%lhs: tensor<10x2x5x10xf32>, %rhs: tensor<2x10x10xf32>) {
  // expected-error @+1 {{has mismatching input batch dimension 2 and output batch dimension 3}}
  %0 = "tf.BatchMatMulV2"(%lhs, %rhs) : (tensor<10x2x5x10xf32>, tensor<2x10x10xf32>) -> tensor<10x3x10x10xf32>
}

// -----