How to use the taichi.all_archs function in taichi

@ti.all_archs
def test_linear_nested_aos():
  x = ti.var(ti.i32)
  y = ti.var(ti.i32)

  n = 128

  @ti.layout
  def place():
    ti.root.dense(ti.i, n // 16).dense(ti.i, 16).place(x, y)

  for i in range(n):
    x[i] = i
    y[i] = i + 123

  for i in range(n):
    assert x[i] == i

@ti.all_archs
def test_f64():
  val = ti.var(ti.f64)

  n = 4
  m = 7

  @ti.layout
  def values():
    ti.root.dense(ti.ij, (n, m)).place(val)

  for i in range(n):
    for j in range(m):
      val[i, j] = (i + j * 3) * 1e100

  val.from_numpy(val.to_numpy() * 2)

@ti.all_archs
def test_abs():

  @ti.kernel
  def func():
    print(1)

  func()

@ti.all_archs
def test_loop_grad_complex():
  return  # This case is not supported yet
  x = ti.var(ti.f32)

  n = 16
  m = 8

  @ti.layout
  def place():
    ti.root.dense(ti.ij, (n, m)).place(x)
    ti.root.lazy_grad()

  @ti.kernel
  def func():
    for k in range(n):
      t = k * k

@ti.all_archs
def test_pointer():
  if ti.get_os_name() == 'win':
    # This test not supported on Windows due to the VirtualAlloc issue #251
    return
  x = ti.var(ti.f32)
  s = ti.var(ti.i32)

  n = 128

  @ti.layout
  def place():
    ti.root.dense(ti.i, n).pointer().dense(ti.i, n).place(x)
    ti.root.place(s)

  @ti.kernel
  def activate():

@ti.all_archs
def test_io():
  ti.cfg.arch = ti.cuda
  x = ti.var(ti.i32)

  n = 128

  @ti.layout
  def place():
    ti.root.dense(ti.i, n).place(x)

  x[3] = 123
  x[4] = 456
  assert x[3] == 123
  assert x[4] == 456

@ti.all_archs
def test_numpy_2d_transpose():
  val = ti.var(ti.i32)

  n = 8
  m = 8

  @ti.layout
  def values():
    ti.root.dense(ti.ij, (n, m)).place(val)

  @ti.kernel
  def test_numpy(arr: ti.ext_arr()):
    for i in ti.grouped(val):
      val[i] = arr[i]

  a = np.empty(shape=(n, m), dtype=np.int32)

@ti.all_archs
def test_ad_reduce():
  x = ti.var(ti.f32)
  loss = ti.var(ti.f32)

  N = 16

  @ti.layout
  def place():
    ti.root.place(loss, loss.grad).dense(ti.i, N).place(x, x.grad)

  @ti.kernel
  def func():
    for i in x:
      loss.atomic_add(ti.sqr(x[i]))

  total_loss = 0

@ti.all_archs
def test_simple():
  x = ti.var(ti.i32)

  n = 128

  @ti.layout
  def place():
    ti.root.dense(ti.i, n).place(x)

  @ti.kernel
  def func():
    x[7] = 120

  func()

  for i in range(n):

@ti.all_archs
def test_kernel_template_mapper_numpy():
  x = ti.var(ti.i32)
  y = ti.var(ti.f32)

  @ti.layout
  def layout():
    ti.root.place(x, y)

  annotations = (ti.template(), ti.template(), ti.ext_arr())

  import numpy as np

  mapper = ti.KernelTemplateMapper(annotations, (0, 1, 2))
  assert mapper.lookup((0, 0, np.ones(shape=(1, 2, 3), dtype=np.float32))) == 0
  assert mapper.lookup((0, 0, np.ones(shape=(1, 2, 4), dtype=np.float32))) == 0
  assert mapper.lookup((0, 0, np.ones(shape=(1, 2, 1), dtype=np.int32))) == 1