func.func @padSliceTo4D(%arg0: tensor<4x384x32xf32>) -> tensor<1x384x32xf32> {
  %0 = "tfl.pseudo_const"() {value = dense<0> : tensor<3xi32>} : () -> tensor<3xi32>
  %1 = "tfl.pseudo_const"() {value = dense<[1, 384, 32]> : tensor<3xi32>} : () -> tensor<3xi32>
  %2 = "tfl.slice"(%arg0, %0, %1) : (tensor<4x384x32xf32>, tensor<3xi32>, tensor<3xi32>) -> tensor<1x384x32xf32>
  func.return %2 : tensor<1x384x32xf32>
}

// CHECK: %[[v5:.*]] = "tf.Gather"(%3, %[[cst_1]]) <{validate_indices = true}> : (tensor<3xi32>, tensor<2xi32>) -> tensor<2xi32>
// CHECK: %[[v6:.*]] = "tf.Gather"(%4, %[[cst_2]]) <{validate_indices = true}> : (tensor<4xi32>, tensor<1xi32>) -> tensor<1xi32>
// CHECK: %[[v7:.*]] = "tf.Concat"(%[[cst_3]], %5, %6) : (tensor<i32>, tensor<2xi32>, tensor<1xi32>) -> tensor<3xi32>

// CHECK-DAG:       %[[VAL_5:.*]] = arith.constant dense<[1, 384, 128]> : tensor<3xi32>
// CHECK-DAG:       %[[VAL_6:.*]] = "tfl.pseudo_const"(){{.*}}dense<[1, 1, 384, 512]> : tensor<4xi32>
// CHECK-DAG:       %[[VAL_7:.*]] = "tfl.pseudo_const"(){{.*}}dense<[128, 1, 1, 512]> : tensor<4xi32>
// CHECK-DAG:       %[[VAL_8:.*]] = "tfl.pseudo_const"(){{.*}}dense<[384, 128]> : tensor<2xi32>

  %1 = "tfl.pseudo_const"() {value = dense<0> : tensor<3xi32>} : () -> tensor<3xi32>
  %2 = "tfl.pseudo_const"() {value = dense<[0, 2, 0]> : tensor<3xi32>} : () -> tensor<3xi32>
  %3 = "tfl.pseudo_const"() {value = dense<1> : tensor<3xi32>} : () -> tensor<3xi32>

  %4 = "tfl.var_handle"() {container = "", shared_name = "read_assign/states0"} : () -> tensor<!tf_type.resource>

func.func @testReshapeShapeInference(%arg0: tensor<3x4xi32>) -> tensor<*xi32> {
  %cst = arith.constant dense<[1, 6, 2]> : tensor<3xi32>
  // CHECK: "tfl.reshape"(%arg0, %cst) : (tensor<3x4xi32>, tensor<3xi32>) -> tensor<1x6x2xi32>
  %0 = "tfl.reshape"(%arg0, %cst) : (tensor<3x4xi32>, tensor<3xi32>) -> tensor<*xi32>
  func.return %0 : tensor<*xi32>
}
}

// -----

// CHECK-LABEL: testReshapeShapeInferenceUnknownDim

// CHECK-LABEL: func @testBroadcastGradientArgs
func.func @testBroadcastGradientArgs(%s0: tensor<4xi32>, %s1: tensor<4xi32>) -> (tensor<1xi32>, tensor<0xi32>) {
  // CHECK:     tf.BroadcastGradientArgs
  %r0, %r1 = "tf.BroadcastGradientArgs"(%s0, %s1) : (tensor<4xi32>, tensor<4xi32>) -> (tensor<1xi32>, tensor<0xi32>)
  func.return %r0, %r1 : tensor<1xi32>, tensor<0xi32>
}

// -----

// CHECK-LABEL: @acos

#loc3])\0Amodule {\0A  func.func @host_func(%arg0: tensor<3360x?xi32> loc(fused[#loc1, #loc2, #loc3]), %arg1: tensor<3xi32> loc(fused[#loc1, #loc2, #loc3]), %arg2: tensor<i32> loc(fused[#loc1, #loc2, #loc3]), %arg3: tensor<2xi32> loc(fused[#loc1, #loc2, #loc3])) -> (tensor<1x1120x?xi32>, tensor<1x1120x?xi32>, tensor<1120x?xi32>, tensor<2xi32>) {\0A    %0 = \22tf.Reshape\22(%arg0, %arg1) {_xla_outside_compilation = \220\22} : (tensor<3360x?xi32>, tensor<3xi32>) -> tensor<3x1120x?xi32> loc(#loc9)\0A    %1:3 = ...

    func.func @main(%arg0: tensor<3x3x10xbf16>, %arg1: tensor<3xi32>) -> tensor<1x?x4xbf16> attributes {allow_soft_placement = false, tf.entry_function = {control_outputs = "", inputs = "_arg0,_arg1,_arg2", outputs = "_retval0"}} {
      %cst = "tf.Const"() {value = dense<[1, -1, 4]> : tensor<3xi32>} : () -> tensor<3xi32>

  // CHECK-DAG: %[[X_SHAPE:.+]] = shape.shape_of %[[X]] : tensor<3x4x256x6xf32> -> tensor<4xindex>
  // CHECK-DAG: %[[EPS:.+]] = mhlo.constant dense<1.000000e+00> : tensor<256xf32>
  // CHECK-DAG: %[[MEAN:.+]] = "tf.Mean"(%arg0, %[[CST_AXIS]]) <{keep_dims = false}> : (tensor<3x4x256x6xf32>, tensor<3xi32>) -> tensor<256xf32>

    %1 = stablehlo.constant dense<1> : tensor<1x1xi8>  // Input zero point.
    %2 = stablehlo.constant dense<5> : tensor<2x3xi32>  // Quantized filter - the pattern expects i8 but i32 is given.
    %3 = stablehlo.constant dense<4> : tensor<1x3xi32>  // Precalculated z1 * q2.
    %4 = stablehlo.constant dense<3.000000e+03> : tensor<1x3xf32>  // Merged scale: s1 * s2.