%0:1 = "tfl.while"(%arg0, %arg1) ({
  ^bb0(%arg2: tensor<*xi32>, %arg3: tensor<*xf32>):
    %1 = func.call @WhileOp_cond(%arg2, %arg3) : (tensor<*xi32>, tensor<*xf32>) -> tensor<i1>
    "tfl.yield"(%1) : (tensor<i1>) -> ()
  },  {
  ^bb0(%arg2: tensor<*xi32>, %arg3: tensor<*xf32>):
    %1:2 = func.call @WhileOp_body(%arg2, %arg3) : (tensor<*xi32>, tensor<*xf32>) -> (tensor<*xi32>, tensor<*xf32>)

  func.return %8 : tensor<*xf32>
}

func.func @lstmcellsimple(%arg0: tensor<1x?xf32>, %arg1: tensor<3x4xf32>, %arg2: tensor<2xf32>, %arg3: tensor<1x3xf32>, %arg4: tensor<?xf32>) -> tensor<1x?xf32> attributes  {tf._implements = "LSTMCellSimple", tf._reference = "mlir"} {
    %0 = "tf.BatchMatMulV2"(%arg3, %arg1) {adj_x = false, adj_y = false} : (tensor<1x3xf32>, tensor<3x4xf32>) -> tensor<1x4xf32>

func.func @OneHot(%arg0: tensor<3xi32>, %arg1: tensor<i32>, %arg2: tensor<f32>, %arg3: tensor<f32>) -> tensor<*xf32> {
  %0 = "tf.OneHot"(%arg0, %arg1, %arg2, %arg3) {axis = -1 : i64} : (tensor<3xi32>, tensor<i32>, tensor<f32>, tensor<f32>) -> tensor<*xf32>
  func.return %0: tensor<*xf32>

// CHECK-LABEL: OneHot

		{name: "SHRXLloadidx4", argLength: 4, reg: gp21shxloadidx, asm: "SHRXL", scale: 4, aux: "SymOff", typ: "Uint32", faultOnNilArg0: true, symEffect: "Read"}, // unsigned *(arg0+4*arg1+auxint+aux) >> arg2, arg3=mem, shift amount is mod 32

        %assign_1_control = tf_executor.island(%chain_1_src) wraps "tf.AssignVariableOp"(%arg1, %arg3) : (!tf_res, tensor<f32>) -> ()
        %add_out, %add_control = tf_executor.island wraps "tf.Add"(%arg2, %arg3) : (tensor<f32>, tensor<f32>) -> tensor<f32>
        %mul_out, %mul_control = tf_executor.island wraps "tf.Mul"(%arg2, %arg3) : (tensor<f32>, tensor<f32>) -> tensor<f32>

func.func @testWhileRegionMultiplePassThrough(%arg0 : tensor<*xf32>, %arg1 : tensor<*xf32>, %arg2 : tensor<*xf32>, %arg3 : tensor<i32>) -> tensor<*xf32> {
  // Verify that first 3 operands are elimiinated.
  // CHECK: "tf.WhileRegion"(%arg3)
  %0:4 = "tf.WhileRegion"(%arg0, %arg1, %arg2, %arg3) (
    {
      // condition, check if count has reached 0

    // CHECK-NOT:  "tf.Shape"(%[[ARG_1]])
    // CHECK-NOT:  "tf.Shape"(%[[ARG_3]])
    // CHECK:      %[[ARG_0_SHAPE:[0-9]*]] = "tf.Shape"(%[[ARG_0]])
    // CHECK:      %[[ARG_2_SHAPE:[0-9]*]] = "tf.Shape"(%[[ARG_2]])

  func.func @while_shape_invariant_body_func_propagate(%arg0: tensor<*xf32>, %arg1: tensor<*x!tf_type.resource>, %arg2: tensor<!tf_type.resource>, %arg3: tensor<?xi32>) -> (tensor<*xf32>, tensor<*x!tf_type.resource>, tensor<!tf_type.resource>, tensor<?xi32>) {
    %0 = "tf.SomeOp"(%arg3) : (tensor<?xi32>) -> tensor<?xi32>
    func.return %arg0, %arg1, %arg2, %0 : tensor<*xf32>, tensor<*x!tf_type.resource>, tensor<!tf_type.resource>, tensor<?xi32>

  func.return %0 : tensor<3xi64>
}

// -----

// CHECK-LABEL: func @testReshape(%arg0: tensor<*xf32>, %arg1: tensor<*xf32>, %arg2: tensor<10000xf32>, %arg3: tensor<*xi32>)
func.func @testReshape(%arg0: tensor<*xf32>, %arg1: tensor<*xf32>, %arg2: tensor<10000xf32>, %arg3: tensor<*xi32>) -> (tensor<100x100xf32>, tensor<*xf32>, tensor<100x100xf32>, tensor<100x100xf32>, tensor<*xf32>, tensor<*xf32>) {

	var x1 uint64
	p521CmovznzU64(&x1, arg1, arg2[0], arg3[0])
	var x2 uint64
	p521CmovznzU64(&x2, arg1, arg2[1], arg3[1])
	var x3 uint64
	p521CmovznzU64(&x3, arg1, arg2[2], arg3[2])
	var x4 uint64
	p521CmovznzU64(&x4, arg1, arg2[3], arg3[3])
	var x5 uint64
	p521CmovznzU64(&x5, arg1, arg2[4], arg3[4])
	var x6 uint64