[Evaluated] Issues in test_ops_gradients.py

[daisyden]

gradchecker failed: first compare with cpu, if align with cpu it is low priority.

• "test_fn_grad_linalg_norm_xpu_complex128", report oneDNN issue Double and complex datatype matmul is not supported in oneDNN
• "test_fn_grad_linalg_vector_norm_xpu_complex128",
• "test_fn_grad_nn_functional_rrelu_xpu_float64" : rrelu is fallback case and op is not targeted in 2.5, so marked it. However the fallback mechanism of it has problem, the saved variable noise because 0 in backward. Need further investigation
• "test_fn_grad_norm_inf_xpu_complex128",
• "test_fn_gradgrad_nn_functional_rrelu_xpu_float64",
• "test_inplace_grad_nn_functional_rrelu_xpu_float64",
• "test_inplace_gradgrad_nn_functional_rrelu_xpu_float64",
• test_fn_gradgrad_norm_inf_xpu_complex128

Runtime error

• test_fn_grad_nn_functional_group_norm_xpu_float64",
• "test_fn_gradgrad_nn_functional_group_norm_xpu_float64",

grad reentrant issue - Low priority @guangyey please help to check.

• "test_fn_grad_nn_functional_max_pool2d_xpu_float64",
• "test_fn_gradgrad_index_reduce_mean_xpu_float64",
• "test_fn_gradgrad_index_reduce_prod_xpu_float64",
• "test_inplace_gradgrad_index_reduce_mean_xpu_float64",
• "test_inplace_gradgrad_index_reduce_prod_xpu_float64",

🐛 Describe the bug

Error #2 in TestBwdGradientsXPU , totally 8 , torch.autograd.gradcheck.GradcheckError: Jacobian mismatch for output 0 with respect to input 0,

"test_fn_grad_bernoulli_xpu_float64", - Fixed
"test_fn_grad_linalg_norm_xpu_complex128",
"test_fn_grad_linalg_vector_norm_xpu_complex128",
"test_fn_grad_nn_functional_rrelu_xpu_float64",
"test_fn_grad_norm_inf_xpu_complex128",
"test_fn_gradgrad_nn_functional_rrelu_xpu_float64",
"test_inplace_grad_nn_functional_rrelu_xpu_float64",
"test_inplace_gradgrad_nn_functional_rrelu_xpu_float64",

__ TestBwdGradientsXPU.test_inplace_gradgrad_nn_functional_rrelu_xpu_float64 ___
Traceback (most recent call last):
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 971, in test_wrapper
    return test(*args, **kwargs)
  File "/home/daisyden/workspace/skiplist/pytorch4/third_party/torch-xpu-ops/test/xpu/../../../../test/test_ops_gradients.py", line 100, in test_inplace_gradgrad
    self._check_helper(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 4966, in _check_helper
    self.assertTrue(gradgradcheck(fn, gradcheck_args, **kwargs))
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 4513, in gradgradcheck
    return torch.autograd.gradgradcheck(fn, inputs, grad_outputs, **kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 2251, in gradgradcheck
    return gradcheck(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 2049, in gradcheck
    return _gradcheck_helper(**args)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 2078, in _gradcheck_helper
    _gradcheck_real_imag(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 1488, in _gradcheck_real_imag
    gradcheck_fn(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 1922, in _fast_gradcheck
    _check_analytical_numerical_equal(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 1851, in _check_analytical_numerical_equal
    raise GradcheckError(
torch.autograd.gradcheck.GradcheckError: Jacobian mismatch for output 0 with respect to input 0,
numerical:tensor(-36129.4310, device='xpu:0', dtype=torch.float64)
analytical:tensor(0., device='xpu:0', dtype=torch.float64)

The above quantities relating the numerical and analytical jacobians are computed 
in fast mode. See: https://github.com/pytorch/pytorch/issues/53876 for more background 
about fast mode. Below, we recompute numerical and analytical jacobians in slow mode:

Numerical:
 tensor([[ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  4.5690e+05,  4.1017e+04, -6.2109e+04,  1.4156e+05,
         -6.7045e+04,  4.3481e+04, -9.1453e+05, -1.7943e+03,  1.9339e+05,
         -3.7344e+04,  1.4382e+05, -1.0356e+04, -3.2299e+04, -3.3910e+05,
          3.6282e+05,  2.3169e+05, -2.7192e+05,  1.6888e+05, -2.6284e+03],
        [ 1.1349e+09,  9.2627e+03,  1.5689e+04, -5.4139e+04, -7.0075e+04,
         -2.1482e+04,  8.0559e+04,  2.3670e+03, -2.7632e+05,  9.7094e+04,
          7.5590e+01,  1.0317e+05,  4.7310e+02, -3.4251e+05,  5.2503e+02,
          1.9780e+02,  6.5617e+04, -2.6653e+05, -5.4435e+02,  6.8815e+00],
        [-7.1105e+07, -4.1032e+05, -1.5262e+05,  2.4873e+05, -4.8516e+05,
          3.6793e+05, -8.4320e+05, -5.5769e+05,  8.6143e+05, -9.7033e+05,
          2.1987e+05, -6.8245e+05,  5.2499e+03,  2.2227e+05,  1.2335e+06,
         -6.6694e+05, -4.6186e+05,  3.0343e+06, -6.6249e+05,  2.1468e+04],
        [-4.4799e+06, -2.4603e+05,  1.5622e+04, -5.3404e+04, -1.0327e+05,
         -1.9264e+04,  7.6337e+04,  5.7548e+05, -2.7504e+05,  7.9826e+04,
          7.5327e+03,  9.2431e+04,  4.9740e+02, -3.1177e+05,  1.6536e+05,
         -3.5223e+04,  1.1443e+04, -6.6247e+04, -1.0484e+04,  6.9758e+00],
        [-1.9987e+05,  7.5717e+04,  1.5364e+05, -2.4786e+05,  5.8765e+05,
         -3.5965e+05,  8.3165e+05, -1.6815e+05, -1.0941e+06,  9.9911e+05,
         -2.3659e+05,  6.9356e+05, -5.7745e+03, -5.2491e+05, -1.4703e+06,
          5.5363e+05,  5.4217e+05, -3.0297e+06,  6.2139e+05, -2.1484e+04],
        [ 1.2549e+04, -4.4499e+04,  5.6097e+03,  2.0426e+04,  3.0208e+05,
          6.0800e+04,  2.4187e+05, -1.3350e+05,  4.9460e+05,  2.7821e+05,
         -1.0574e+05,  1.6904e+05, -3.0153e+02, -1.4804e+05, -1.0252e+06,
         -1.7290e+05,  2.4578e+05,  2.1751e+06, -1.5092e+05, -1.2872e+02],
        [ 1.2549e+04, -4.4499e+04,  5.6097e+03,  2.0426e+04,  3.0208e+05,
          6.0800e+04,  2.4187e+05, -1.3350e+05,  4.9460e+05,  2.7821e+05,
         -1.0574e+05,  1.6904e+05, -3.0153e+02, -1.4804e+05, -1.0252e+06,
         -1.7290e+05,  2.4578e+05,  2.1751e+06, -1.5092e+05, -1.2872e+02]],
       device='xpu:0', dtype=torch.float64)
Analytical:
tensor([[-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.],
        [-0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0., -0.]],
       device='xpu:0', dtype=torch.float64)

The max per-element difference (slow mode) is: 1134945589.3745055.


The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/unittest/case.py", line 59, in testPartExecutor
    yield
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/unittest/case.py", line 591, in run
    self._callTestMethod(testMethod)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/unittest/case.py", line 549, in _callTestMethod
    method()
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 2756, in wrapper
    method(*args, **kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 419, in instantiated_test
    result = test(self, **param_kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 1361, in wrapper
    fn(*args, **kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 977, in test_wrapper
    raise Exception(  # noqa: TRY002
Exception: Caused by sample input at index 0: SampleInput(input=Tensor[size=(20,), device="xpu:0", dtype=torch.float64], args=(), kwargs={'lower': '0.0', 'upper': '1.0', 'training': 'True'}, broadcasts_input=False, name='')

To execute this test, run the following from the base repo dir:
    PYTORCH_TEST_WITH_SLOW=1 python test/test_ops_gradients.py -k TestBwdGradientsXPU.test_inplace_gradgrad_nn_functional_rrelu_xpu_float64

Error #3 in TestBwdGradientsXPU , totally 8 , torch.autograd.gradcheck.GradcheckError: While considering the imaginary part of complex outputs only, Jacobian mismatch for output 0 with respect to input 0, - Only test_fn_gradgrad_norm_inf_xpu_complex128 failed.

"test_fn_grad_masked_normalize_xpu_complex128",
"test_fn_grad_renorm_xpu_complex128",
"test_fn_gradgrad_linalg_vector_norm_xpu_complex128",
"test_fn_gradgrad_masked_normalize_xpu_complex128",
"test_fn_gradgrad_norm_inf_xpu_complex128", - Failed
"test_fn_gradgrad_renorm_xpu_complex128",
"test_inplace_grad_renorm_xpu_complex128",
"test_inplace_gradgrad_renorm_xpu_complex128",

_______ TestBwdGradientsXPU.test_inplace_gradgrad_renorm_xpu_complex128 ________
Traceback (most recent call last):
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 971, in test_wrapper
    return test(*args, **kwargs)
  File "/home/daisyden/workspace/skiplist/pytorch4/third_party/torch-xpu-ops/test/xpu/../../../../test/test_ops_gradients.py", line 100, in test_inplace_gradgrad
    self._check_helper(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 4966, in _check_helper
    self.assertTrue(gradgradcheck(fn, gradcheck_args, **kwargs))
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 4513, in gradgradcheck
    return torch.autograd.gradgradcheck(fn, inputs, grad_outputs, **kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 2251, in gradgradcheck
    return gradcheck(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 2049, in gradcheck
    return _gradcheck_helper(**args)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 2078, in _gradcheck_helper
    _gradcheck_real_imag(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 1459, in _gradcheck_real_imag
    gradcheck_fn(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 1922, in _fast_gradcheck
    _check_analytical_numerical_equal(
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/autograd/gradcheck.py", line 1851, in _check_analytical_numerical_equal
    raise GradcheckError(
torch.autograd.gradcheck.GradcheckError: While considering the imaginary part of complex outputs only, Jacobian mismatch for output 0 with respect to input 0,
numerical:tensor(nan+nanj, device='xpu:0', dtype=torch.complex128)
analytical:tensor(nan+nanj, device='xpu:0', dtype=torch.complex128)

The above quantities relating the numerical and analytical jacobians are computed 
in fast mode. See: https://github.com/pytorch/pytorch/issues/53876 for more background 
about fast mode. Below, we recompute numerical and analytical jacobians in slow mode:

Numerical:
 tensor([[nan+nanj, 0.+0.j, nan+nanj,  ..., nan+nanj, 0.+0.j, 0.+0.j],
        [nan+nanj, 0.+0.j, nan+nanj,  ..., nan+nanj, 0.+0.j, 0.+0.j],
        [nan+nanj, 0.+0.j, nan+nanj,  ..., nan+nanj, 0.+0.j, 0.+0.j],
        ...,
        [nan+nanj, 0.+0.j, nan+nanj,  ..., nan+nanj, 0.+0.j, 0.+0.j],
        [nan+nanj, 0.+0.j, nan+nanj,  ..., nan+nanj, 0.+0.j, 0.+0.j],
        [nan+nanj, 0.+0.j, nan+nanj,  ..., nan+nanj, 0.+0.j, 0.+0.j]],
       device='xpu:0', dtype=torch.complex128)
Analytical:
tensor([[nan+nanj, nan+nanj, nan+nanj,  ..., nan+nanj, nan+nanj, nan+nanj],
        [0.+0.j, 0.+0.j, 0.+0.j,  ..., 0.+0.j, 0.+0.j, 0.+0.j],
        [nan+nanj, nan+nanj, nan+nanj,  ..., nan+nanj, nan+nanj, nan+nanj],
        ...,
        [nan+nanj, nan+nanj, nan+nanj,  ..., nan+nanj, nan+nanj, nan+nanj],
        [0.+0.j, 0.+0.j, 0.+0.j,  ..., 0.+0.j, 0.+0.j, 0.+0.j],
        [0.+0.j, 0.+0.j, 0.+0.j,  ..., 0.+0.j, 0.+0.j, 0.+0.j]],
       device='xpu:0', dtype=torch.complex128)

The max per-element difference (slow mode) is: nan.


The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/unittest/case.py", line 59, in testPartExecutor
    yield
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/unittest/case.py", line 591, in run
    self._callTestMethod(testMethod)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/unittest/case.py", line 549, in _callTestMethod
    method()
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 2756, in wrapper
    method(*args, **kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 419, in instantiated_test
    result = test(self, **param_kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_utils.py", line 1361, in wrapper
    fn(*args, **kwargs)
  File "/home/daisyden/miniconda3/envs/xpupatch2/lib/python3.10/site-packages/torch/testing/_internal/common_device_type.py", line 977, in test_wrapper
    raise Exception(  # noqa: TRY002
Exception: Caused by sample input at index 3: SampleInput(input=Tensor[size=(5, 5, 5), device="xpu:0", dtype=torch.complex128], args=(inf,2,0.5), kwargs={}, broadcasts_input=False, name='')

To execute this test, run the following from the base repo dir:
    PYTORCH_TEST_WITH_SLOW=1 python test/test_ops_gradients.py -k TestBwdGradientsXPU.test_inplace_gradgrad_renorm_xpu_complex128

Error #5 in TestBwdGradientsXPU , totally 2 , RuntimeError: input tensor must have at least one element, but got input_sizes = [1, 0, 1], see also #249

"test_fn_grad_nn_functional_group_norm_xpu_float64",
"test_fn_gradgrad_nn_functional_group_norm_xpu_float64",

Error #6 in TestBwdGradientsXPU , totally 5 , torch.autograd.gradcheck.GradcheckError: Backward is not reentrant, i.e., running backward with same input and grad_output multiple times gives different values, although analytical gradient matches numerical gradient.The tolerance for nondeterminism was 0.0.

"test_fn_grad_nn_functional_max_pool2d_xpu_float64",
"test_fn_gradgrad_index_reduce_mean_xpu_float64",
"test_fn_gradgrad_index_reduce_prod_xpu_float64",
"test_inplace_gradgrad_index_reduce_mean_xpu_float64",
"test_inplace_gradgrad_index_reduce_prod_xpu_float64",

Error #7 in TestBwdGradientsXPU , totally 2 , NotImplementedError: Could not run 'aten::_sparse_coo_tensor_with_dims_and_tensors' with arguments from the 'SparseXPU' backend. This could be because the operator doesn't exist for this backend, or was omitted during the selective/custom build process (if using custom build). If you are a Facebook employee using PyTorch on mobile, please visit https://fburl.com/ptmfixes for possible resolutions. 'aten::_sparse_coo_tensor_with_dims_and_tensors' is only available for these backends: [XPU, Meta, SparseCPU, SparseMeta, BackendSelect, Python, FuncTorchDynamicLayerBackMode, Functionalize, Named, Conjugate, Negative, ZeroTensor, ADInplaceOrView, AutogradOther, AutogradCPU, AutogradCUDA, AutogradHIP, AutogradXLA, AutogradMPS, AutogradIPU, AutogradXPU, AutogradHPU, AutogradVE, AutogradLazy, AutogradMTIA, AutogradPrivateUse1, AutogradPrivateUse2, AutogradPrivateUse3, AutogradMeta, AutogradNestedTensor, Tracer, AutocastCPU, AutocastXPU, AutocastCUDA, FuncTorchBatched, BatchedNestedTensor, FuncTorchVmapMode, Batched, VmapMode, FuncTorchGradWrapper, PythonTLSSnapshot, FuncTorchDynamicLayerFrontMode, PreDispatch, PythonDispatcher].

see also #240 - Move to #357

"test_fn_grad_to_sparse_xpu_float64",
"test_fn_gradgrad_to_sparse_xpu_float64",

Error #8 in TestBwdGradientsXPU , totally 2 , RuntimeError: DispatchStub: unsupported device typexpu, see also #249 - Fixed

"test_inplace_grad_conj_physical_xpu_complex128",
"test_inplace_gradgrad_conj_physical_xpu_complex128",

Versions

Collecting environment information...
PyTorch version: N/A
Is debug build: N/A
CUDA used to build PyTorch: N/A
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: 14.0.0-1ubuntu1.1
CMake version: version 3.26.4
Libc version: glibc-2.35

Python version: 3.10.14 (main, May 6 2024, 19:42:50) [GCC 11.2.0] (64-bit runtime)
Python platform: Linux-5.15.0-106-generic-x86_64-with-glibc2.35
Is CUDA available: N/A
CUDA runtime version: Could not collect
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: Could not collect
Nvidia driver version: Could not collect
cuDNN version: Could not collect
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: N/A

CPU:
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Address sizes: 52 bits physical, 57 bits virtual
Byte Order: Little Endian
CPU(s): 224
On-line CPU(s) list: 0-223
Vendor ID: GenuineIntel
Model name: Intel(R) Xeon(R) Platinum 8480+
CPU family: 6
Model: 143
Thread(s) per core: 2
Core(s) per socket: 56
Socket(s): 2
Stepping: 8
CPU max MHz: 3800.0000
CPU min MHz: 800.0000
BogoMIPS: 4000.00
Flags: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf tsc_known_freq pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cat_l2 cdp_l3 invpcid_single intel_ppin cdp_l2 ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect avx_vnni avx512_bf16 wbnoinvd dtherm ida arat pln pts hwp hwp_act_window hwp_epp hwp_pkg_req avx512vbmi umip pku ospke waitpkg avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid bus_lock_detect cldemote movdiri movdir64b enqcmd fsrm md_clear serialize tsxldtrk pconfig arch_lbr amx_bf16 avx512_fp16 amx_tile amx_int8 flush_l1d arch_capabilities
Virtualization: VT-x
L1d cache: 5.3 MiB (112 instances)
L1i cache: 3.5 MiB (112 instances)
L2 cache: 224 MiB (112 instances)
L3 cache: 210 MiB (2 instances)
NUMA node(s): 2
NUMA node0 CPU(s): 0-55,112-167
NUMA node1 CPU(s): 56-111,168-223
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit: Not affected
Vulnerability L1tf: Not affected
Vulnerability Mds: Not affected
Vulnerability Meltdown: Not affected
Vulnerability Mmio stale data: Not affected
Vulnerability Retbleed: Not affected
Vulnerability Spec rstack overflow: Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1: Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2: Mitigation; Enhanced IBRS; IBPB conditional; RSB filling; PBRSB-eIBRS SW sequence; BHI BHI_DIS_S
Vulnerability Srbds: Not affected
Vulnerability Tsx async abort: Not affected

Versions of relevant libraries:
[pip3] numpy==1.26.4
[pip3] optree==0.11.0
[pip3] torch==2.4.0a0+git5fb11cd
[conda] mkl-include 2024.1.0 intel_691 intel
[conda] mkl-static 2024.1.0 intel_691 intel
[conda] numpy 1.26.4 pypi_0 pypi
[conda] optree 0.11.0 pypi_0 pypi
[conda] torch 2.4.0a0+git5fb11cd pypi_0 pypi

[daisyden]

Fixed some issues in index_put, #514

test_fn_grad_bernoulli_xpu_float64: The analytical grad is right, numerical grad is not. The issue could in forward pass. could depending on the fix of uniform issue that is owned by @xytintel .
TestTorchDeviceType.test_discontiguous_out_cumsum: According to @xytintel cumsum implementation should align with cuda, or wait for the structure element feature.

[fengyuan14]

Fixed some issues in index_put, #514

test_fn_grad_bernoulli_xpu_float64: The analytical grad is right, numerical grad is not. The issue could in forward pass. could depending on the fix of uniform issue that is owned by @xytintel . TestTorchDeviceType.test_discontiguous_out_cumsum: According to @xytintel cumsum implementation should align with cuda, or wait for the structure element feature.

Assignee updated.

[daisyden]

test_ops_gradients_xpu.py::TestBwdGradientsXPU::test_fn_grad_masked_normalize_xpu_complex128 PASSED

[daisyden]

@ZhiweiYan-96 please also help to check test_fn_grad_addbmm_xpu_complex128. when you work on #436.