windows errors

20220308-tests-windows-errors.txt

$ python ../../../test/test_jit_cuda_fuser.py -v
test__softmax_function (__main__.TestCudaFuser) ... ok
test__softmax_function_half_to_float (__main__.TestCudaFuser) ... ok
test_addcmul_ops (__main__.TestCudaFuser) ... ok
test_alias_pass_fix (__main__.TestCudaFuser) ... ERROR
test_autocast_1 (__main__.TestCudaFuser) ... ok
test_autocast_1_bfloat (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_autocast_2 (__main__.TestCudaFuser) ... ok
test_autocast_2_bfloat (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_backward_type (__main__.TestCudaFuser) ... ok
test_batch_norm_half (__main__.TestCudaFuser) ... C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\nn\modules\module.py:1383: UserWarning: positional arguments a
nd argument "destination" are deprecated. nn.Module.state_dict will not accept them in the future. Refer to https://pytorch.org/docs/master/generated/torch.nn.Module.html#torch.n
n.Module.state_dict for details.
  warnings.warn(
ERROR
test_batch_norm_impl_index_correctness (__main__.TestCudaFuser) ... ERROR
test_bfloat (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_binary_bitwise (__main__.TestCudaFuser) ... ok
test_binary_ops (__main__.TestCudaFuser) ... ok
test_binary_ops_channels_last_with_bcast (__main__.TestCudaFuser) ... ok
test_binary_ops_permutation (__main__.TestCudaFuser) ... ERROR
test_branches (__main__.TestCudaFuser) ... ok
test_broadcasting_0 (__main__.TestCudaFuser) ... ok
test_broadcasting_1 (__main__.TestCudaFuser) ... ok
test_broadcasting_2 (__main__.TestCudaFuser) ... ERROR
test_broadcasting_3 (__main__.TestCudaFuser) ... ok
test_broadcasting_multiple_output (__main__.TestCudaFuser) ... skipped "broadcast on branches can't be resolved yet"
test_broadcasting_multiple_output_shape (__main__.TestCudaFuser) ... skipped 'Broadcast with different output not supported yet'
test_broadcasting_partition_logic_0 (__main__.TestCudaFuser) ... ERROR
test_broadcasting_partition_logic_1 (__main__.TestCudaFuser) ... ERROR
test_build_shape_expression_native_dropout (__main__.TestCudaFuser) ... ok
test_category_rule (__main__.TestCudaFuser) ... expected failure
test_channels_last_with_broadcast (__main__.TestCudaFuser) ... ERROR
test_chunk (__main__.TestCudaFuser) ... ok
test_clean_profile_ivalue (__main__.TestCudaFuser) ... ok
test_const (__main__.TestCudaFuser) ... ok
test_conv2d_bias (__main__.TestCudaFuser) ... ERROR
test_cpu_scalar (__main__.TestCudaFuser) ... ok
test_disable_sibling_fuse (__main__.TestCudaFuser) ... ok
test_dropout_inference_fusion (__main__.TestCudaFuser) ... ok
test_dropout_train_nograd_fusion (__main__.TestCudaFuser) ... ok
test_dropout_train_nograd_prob_check (__main__.TestCudaFuser) ... ok
test_dropout_training_fusion (__main__.TestCudaFuser) ... ok
test_dropout_training_prob_check (__main__.TestCudaFuser) ... ok
test_dynamic_size (__main__.TestCudaFuser) ... [W C:\actions-runner\_work\pytorch\pytorch\torch\csrc\jit\python\init.cpp:654] Warning: Use _jit_set_fusion_strategy, bailout depth
 is deprecated. Setting to (STATIC, 20) (function operator ())
ok
test_fix_shape_expression_bn (__main__.TestCudaFuser) ... ok
test_gelu (__main__.TestCudaFuser) ... ok
test_grad_sum_to_size (__main__.TestCudaFuser) ... ok
test_graph_for_with_missing_optimized_engine (__main__.TestCudaFuser) ... ok
test_graph_rng (__main__.TestCudaFuser) ... ok
test_half (__main__.TestCudaFuser) ... ok
test_inplace_removal (__main__.TestCudaFuser) ... ok
test_layer_norm_autodiff (__main__.TestCudaFuser) ... ok
test_layer_norm_parser (__main__.TestCudaFuser) ... ok
test_linear (__main__.TestCudaFuser) ... ok
test_linear_1d_weight_mismatch_bias_dtype (__main__.TestCudaFuser) ... ok
test_multiple_device_pw (__main__.TestCudaFuser) ... skipped 'requires multiple CUDA device'
test_native_layer_norm (__main__.TestCudaFuser) ... ok
test_native_layer_norm_bfloat (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_native_layer_norm_half (__main__.TestCudaFuser) ... ok
test_norm (__main__.TestCudaFuser) ... ok
test_norm_bfloat (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_norm_channels_last (__main__.TestCudaFuser) ... ok
test_norm_half (__main__.TestCudaFuser) ... ok
test_norm_large (__main__.TestCudaFuser) ... ERROR
test_normalization_partition (__main__.TestCudaFuser) ... ok
test_permutation_preservation (__main__.TestCudaFuser) ... ok
test_profile_ivalue (__main__.TestCudaFuser) ... ok
test_profiling_node (__main__.TestCudaFuser) ... ok
test_pw_single_reduction_partition (__main__.TestCudaFuser) ... ok
test_random_topo (__main__.TestCudaFuser) ... ok
test_reduction (__main__.TestCudaFuser) ... ok
test_reduction_dtypes_axis (__main__.TestCudaFuser) ... ok
test_reduction_multiple_output (__main__.TestCudaFuser) ... [W C:\actions-runner\_work\pytorch\pytorch\torch\csrc\jit\python\init.cpp:654] Warning: Use _jit_set_fusion_strategy,
bailout depth is deprecated. Setting to (STATIC, 20) (function operator ())
ok
test_reduction_permutation (__main__.TestCudaFuser) ... ok
test_reduction_sizes_op (__main__.TestCudaFuser) ... ok
test_remove_output_used_only_in_dtype (__main__.TestCudaFuser) ... ok
test_rsub (__main__.TestCudaFuser) ... ok
test_scalar_input (__main__.TestCudaFuser) ... ok
test_scalar_tensor (__main__.TestCudaFuser) ... ok
test_scalar_tensor_permuted (__main__.TestCudaFuser) ... ok
test_shape_expression (__main__.TestCudaFuser) ... ok
test_sibling_fusion (__main__.TestCudaFuser) ... ok
test_sibling_fusion_no_scalar_inputs (__main__.TestCudaFuser) ... ok
test_single_reduction_broadcast (__main__.TestCudaFuser) ... ok
test_singleton_fusion (__main__.TestCudaFuser) ... ok
test_softmax (__main__.TestCudaFuser) ... ERROR
test_softmax_bfloat (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_softmax_dtype (__main__.TestCudaFuser) ... ok
test_softmax_half (__main__.TestCudaFuser) ... ERROR
test_softplus_fuser (__main__.TestCudaFuser) ... ok
test_squeeze (__main__.TestCudaFuser) ... ok
test_squeeze_negative_dim (__main__.TestCudaFuser) ... ok
test_sum_to_one (__main__.TestCudaFuser) ... ok
test_sum_to_size (__main__.TestCudaFuser) ... ok
test_ternary_ops (__main__.TestCudaFuser) ... ok
test_ternary_ops_integer_compatibility (__main__.TestCudaFuser) ... ok
test_ternary_ops_type_promotion (__main__.TestCudaFuser) ... ok
test_to_dtype_bf16_to_bf16 (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_to_dtype_bf16_to_fp32 (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_to_dtype_fp16_to_fp16 (__main__.TestCudaFuser) ... ok
test_to_dtype_fp16_to_fp32 (__main__.TestCudaFuser) ... ok
test_to_dtype_fp32_to_bf16 (__main__.TestCudaFuser) ... skipped 'device does not support BFloat16'
test_to_dtype_fp32_to_fp16 (__main__.TestCudaFuser) ... ok
test_trivial_reduction (__main__.TestCudaFuser) ... ok
test_type_as_op (__main__.TestCudaFuser) ... ok
test_unary_bitwise (__main__.TestCudaFuser) ... ok
test_unary_ops (__main__.TestCudaFuser) ... ../../../test/test_jit_cuda_fuser.py:489: UserWarning: Specified kernel cache directory could not be created! This disables kernel cac
hing. Specified directory is C:\Users\runneruser\AppData\Local\Temp/torch/kernels. This warning will appear only once per process. (Triggered internally at  C:\actions-runner\_wo
rk\pytorch\pytorch\aten\src\ATen\native\cuda\jit_utils.cpp:860.)
  o = operation(o)
ok
test_unsqueeze (__main__.TestCudaFuser) ... ok
test_view (__main__.TestCudaFuser) ... ok
test_autodiff_fallback (jit.test_fuser_common.TestFuserCommon) ... ok
test_context_manager_test (__main__.TestPassManagerCudaFuser) ... ok
test_register_fuser (__main__.TestPassManagerCudaFuser) ... ok

======================================================================
ERROR: test_alias_pass_fix (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 3488, in test_alias_pass_fix
    self._run_helper(t_jit, t, x, w, b)
  File "../../../test/test_jit_cuda_fuser.py", line 149, in _run_helper
    jit_o = jit_op(*args)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV8 dim: 0 id: iS127{T0.size[1]}


======================================================================
ERROR: test_batch_norm_half (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 2881, in test_batch_norm_half
    self._test_batch_norm_impl_index_helper(4, 8, 5, affine, track_running_stats, training, torch.half)
  File "../../../test/test_jit_cuda_fuser.py", line 2799, in _test_batch_norm_impl_index_helper
    jit_o = t_jit(x)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\nn\modules\module.py", line 1111, in _call_impl
    return forward_call(*input, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 431, in prof_meth_call
    return prof_callable(meth_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV26 dim: 0 id: iS260{T0.size[1]}



======================================================================
ERROR: test_batch_norm_impl_index_correctness (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 2906, in test_batch_norm_impl_index_correctness
    self._test_batch_norm_impl_index_helper(b, c, hw, affine, track_running_stats, training)
  File "../../../test/test_jit_cuda_fuser.py", line 2799, in _test_batch_norm_impl_index_helper
    jit_o = t_jit(x)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\nn\modules\module.py", line 1111, in _call_impl
    return forward_call(*input, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 431, in prof_meth_call
    return prof_callable(meth_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV18 dim: 0 id: iS214{T0.size[1]}



======================================================================
ERROR: test_binary_ops_permutation (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 1118, in test_binary_ops_permutation
    self._permutation_helper(x, b_axis, torch.float32, "cuda", perm0, perm1)
  File "../../../test/test_jit_cuda_fuser.py", line 1095, in _permutation_helper
    jit_o = t_jit(x, y)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV6 dim: 1 id: iS105{T0.size[0]}


======================================================================
ERROR: test_broadcasting_2 (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 364, in test_broadcasting_2
    jit_o = t_jit(x, y, 2.0)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV5 dim: 2 id: iS102{T0.size[2]}


======================================================================
ERROR: test_broadcasting_partition_logic_0 (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 408, in test_broadcasting_partition_logic_0
    jit_o = t_jit(x, y, z)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV9 dim: 0 id: iS179{T0.size[0]}


======================================================================
ERROR: test_broadcasting_partition_logic_1 (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 430, in test_broadcasting_partition_logic_1
    jit_o = t_jit(x, y, z)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV9 dim: 2 id: iS186{T0.size[2]}


======================================================================
ERROR: test_channels_last_with_broadcast (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 1733, in test_channels_last_with_broadcast
    jit_o = t_jit(x, y)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV5 dim: 0 id: iS111{T0.size[0]}


======================================================================
ERROR: test_conv2d_bias (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 2968, in test_conv2d_bias
    jit_o = jitted(inp, weight, bias)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 428, in prof_func_call
    return prof_callable(func_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV6 dim: 0 id: iS113{T0.size[1]}


======================================================================
ERROR: test_norm_large (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 1453, in test_norm_large
    self._norm_helper(x, torch.float32, "cuda", 1e-4, is_batch_norm_else_instance_norm)
  File "../../../test/test_jit_cuda_fuser.py", line 1399, in _norm_helper
    jit_o = t_jit(x, jit_running_mean, jit_running_var)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\nn\modules\module.py", line 1111, in _call_impl
    return forward_call(*input, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 431, in prof_meth_call
    return prof_callable(meth_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV22 dim: 0 id: iS146{T0.size[1]}


======================================================================
ERROR: test_softmax (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 1621, in test_softmax
    self._softmax_helper(x, reduction_dim, torch.float32, "cuda", 1e-4)
  File "../../../test/test_jit_cuda_fuser.py", line 1509, in _softmax_helper
    jit_o = t_jit(x, y)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\nn\modules\module.py", line 1111, in _call_impl
    return forward_call(*input, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 431, in prof_meth_call
    return prof_callable(meth_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV12 dim: 0 id: iS126{T6.size[0]}


======================================================================
ERROR: test_softmax_half (__main__.TestCudaFuser)
----------------------------------------------------------------------
Traceback (most recent call last):
  File "../../../test/test_jit_cuda_fuser.py", line 1637, in test_softmax_half
    self._softmax_helper(x, reduction_dim, torch.float16, "cuda", 5e-3)
  File "../../../test/test_jit_cuda_fuser.py", line 1509, in _softmax_helper
    jit_o = t_jit(x, y)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\nn\modules\module.py", line 1111, in _call_impl
    return forward_call(*input, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 431, in prof_meth_call
    return prof_callable(meth_call, *args, **kwargs)
  File "C:\actions-runner\_work\pytorch\pytorch\build\win_tmp\build\torch\testing\_internal\common_utils.py", line 425, in prof_callable
    return callable(*args, **kwargs)
RuntimeError: The following operation failed in the TorchScript interpreter.
Traceback of TorchScript (most recent call last):
RuntimeError: index_map.find(root_dom[i]) != index_map.end() INTERNAL ASSERT FAILED at "C:\\actions-runner\\_work\\pytorch\\pytorch\\torch\\csrc\\jit\\codegen\\cuda\\index_comput
e.cpp":1980, please report a bug to PyTorch. Couldn't find root mapping for TV22 dim: 0 id: iS208{T14.size[0]}


----------------------------------------------------------------------
Ran 108 tests in 836.359s

FAILED (errors=12, skipped=12, expected failures=1)