Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

2.5 UT bug triage #663

Closed
daisyden opened this issue Jul 30, 2024 · 3 comments · Fixed by #597
Closed

2.5 UT bug triage #663

daisyden opened this issue Jul 30, 2024 · 3 comments · Fixed by #597
Assignees
@majing921201 @daisyden @fengyuan14 @PenghuiCheng @xytintel @yuchengliu1 @huaiyuzh @chunhuanMeng
Labels
ut_triaged
Milestone
PT2.5

Comments

@daisyden
Copy link
Contributor

daisyden commented Jul 30, 2024 •
edited
Loading

Failures Issues Platform Comments Owner Status Comments
test_compare_cpu_nn_functional_batch_norm_xpu_float16 Case regression due to PyTorch uplift · Issue #614 · intel/torch-xpu-ops (github.com) PVC Acc Yutao Done  
test_reductions_xpu.py::TestReductionsXPU::test_median_nan_values_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Nan Penghui passed  
test_reductions_xpu.py::TestReductionsXPU::test_median_real_values_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Nan Penghui passed  
test_reductions_xpu.py::TestReductionsXPU::test_quantile_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Nan Penghui passed  
test_nn_xpu.py::TestNNDeviceTypeXPU::test_grid_sample_large_xpu MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Nan Jing JIRA:https://jira.devtools.intel.com/browse/GSD-9622  
TestCommonXPU.test_compare_cpu_index_put_xpu_bool MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Daisy Passed  
TestCommonXPU.test_compare_cpu_index_put_xpu_uint8 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Daisy Passed  
TestCommonXPU.test_python_ref__refs_all_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Yucheng Done  
TestCommonXPU::test_compare_cpu_all_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Yucheng Done  
test_reductions_xpu.py::TestReductionsXPU::test_min_xpu_bool MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Penghui->Yutao Ready for Review #692
TestCommonXPU.test_python_ref__refs_all_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Yucheng Done  
TestCommonXPU.test_python_ref_torch_fallback__refs_all_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Yucheng Done  
TestCommonXPU.test_python_ref_executor__refs_all_executor_aten_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Inf Yucheng Done  
test_ops_xpu.py::TestMathBitsXPU::test_neg_view_nn_functional_embedding_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Daisy Passed  
test_indexing_xpu.py::TestIndexingXPU::test_cuda_broadcast_index_use_deterministic_algorithms_xpu MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Huaiyu Passed  
test_tensor_creation_ops_xpu.py::TestTensorCreationXPU::test_float_to_int_conversion_finite_xpu_int64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Huaiyu GSD-9643  
test_indexing_xpu.py::TestIndexingXPU::test_index_put_accumulate_large_tensor_xpu MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Daisy OOM  
test_index_and_index_put.py::TestTorchMethod::test_index_and_index_put MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Daisy Passed  
test_ops_xpu.py::TestCommonXPU::test_compare_cpu_all_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Yucheng Done  
test_ops_xpu.py::TestCommonXPU::test_compare_cpu_index_put_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Daisy Passed  
test_reductions_xpu.py::TestReductionsXPU::test_ref_duplicate_values_all_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Jing->Yutao Ready for Review #692
TestCommonXPU.test_python_ref__refs_nn_functional_group_norm_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Penghui->Yutao    
TestCommonXPU.test_numpy_ref_searchsorted_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Penghui passed  
TestCommonXPU.test_numpy_ref_searchsorted_xpu_int64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Penghui passed  
test_reductions_xpu.py::TestReductionsXPU::test_all_any_vs_numpy_xpu_complex128 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Yu->Yutao Ready for Review #692
test_reductions_xpu.py::TestReductionsXPU::test_all_any_vs_numpy_xpu_complex64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Yu->Yutao Ready for Review #692
test_reductions_xpu.py::TestReductionsXPU::test_all_any_vs_numpy_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Yu->Yutao Ready for Review #692
TestMathBitsXPU.test_neg_view_nn_functional_embedding_xpu_float64 MTL specific issues · Issue #611 · intel/torch-xpu-ops (github.com) MTL Acc Daisy Passed  
test_quick__batch_norm_with_update_xpu_bfloat16 "AssertionError: Tensor-likes are not equal!" in test_decomp · Issue #589 · intel/torch-xpu-ops (github.com) PVC Acc Yucheng Passed  
test_quick__batch_norm_with_update_xpu_float16 "AssertionError: Tensor-likes are not equal!" in test_decomp · Issue #589 · intel/torch-xpu-ops (github.com) PVC Acc Yucheng Passed  
test_fn_grad_nn_functional_rrelu_xpu_float64 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Acc Chunhuan Done CPU Fallback fails
test_fn_grad_norm_inf_xpu_complex128 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Acc Chunhuan Passed  
test_fn_gradgrad_nn_functional_rrelu_xpu_float64 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Acc Chunhuan Done CPU Fallback fails
test_inplace_grad_nn_functional_rrelu_xpu_float64 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Acc Chunhuan Done Not Implemented
test_inplace_gradgrad_nn_functional_rrelu_xpu_float64 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Acc Chunhuan Done Not Implemented
test_fn_gradgrad_norm_inf_xpu_complex128 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Acc Chunhuan Passed  
test_fn_grad_nn_functional_group_norm_xpu_float64 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Runtime Chunhuan Passed  
test_fn_gradgrad_nn_functional_group_norm_xpu_float64 [Evaluated] Issues in test_ops_gradients.py · Issue #271 · intel/torch-xpu-ops (github.com) PVC Runtime Chunhuan Passed  
test_cow_input [Evaluated] Failures in test_ops::TestCompositeCompliance · Issue #233 · intel/torch-xpu-ops (github.com) PVC COW Huaiyu/Chunhuan    
test_compare_cpu_tanh_complex64 [To Evaluate] accuracy issue in fine-grained test · Issue #184 · intel/torch-xpu-ops (github.com) PVC Inf Daisy Done Compiler issue
test_reference_numerics_normal_polygamma_polygamma_n_1_xpu_float17 Polygamma: UT failure · Issue #622 · intel/torch-xpu-ops (github.com) PVC Inf Daisy Done CPU issue， pytorch/pytorch#132386
test_reference_numerics_normal_polygamma_polygamma_n_2_xpu_float17 Polygamma: UT failure · Issue #622 · intel/torch-xpu-ops (github.com) PVC Inf Daisy Done CPU issue， pytorch/pytorch#132386
test_reference_numerics_normal_polygamma_polygamma_n_3_xpu_float18 Polygamma: UT failure · Issue #622 · intel/torch-xpu-ops (github.com) PVC Inf Daisy Done CPU issue， pytorch/pytorch#132386
test_reference_numerics_normal_polygamma_polygamma_n_4_xpu_float19 Polygamma: UT failure · Issue #622 · intel/torch-xpu-ops (github.com) PVC Inf Daisy Done CPU issue， pytorch/pytorch#132386
test_dispatch_meta_outplace__foreach_norm_xpu_bfloat16 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed high priority to triage
test_dispatch_meta_outplace__foreach_norm_xpu_float [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed  
test_dispatch_symbolic_meta_outplace__foreach_norm_xpu_bfloat16 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed  
test_dispatch_symbolic_meta_outplace__foreach_norm_xpu_float [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed  
test_dispatch_symbolic_meta_outplace_all_strides__foreach_norm_xpu_float32 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed  
test_meta_outplace__foreach_norm_xpu_bfloat16 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed  
test_meta_outplace__foreach_norm_xpu_float [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan Passed  
test_dispatch_meta_outplace__foreach_pow_xpu_int [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan passed  
test_dispatch_symbolic_meta_outplace__foreach_pow_xpu_int [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan passed  
test_meta_outplace__foreach_pow_xpu_int [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan passed  
test_dispatch_meta_outplace__foreach_pow_xpu_uint8 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan passed  
test_dispatch_symbolic_meta_outplace__foreach_pow_xpu_uint8 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan passed  
test_meta_outplace__foreach_pow_xpu_uint8 [Evaluated] Issues in test_meta · Issue #601 · intel/torch-xpu-ops (github.com) PVC Meta Chunhuan passed  
test_cross_entropy_loss_2d_out_of_bounds_class_index_xpu_float16 #653 PVC Runtime Jianhang Ready for Review CUDA bias case
test_cross_entropy_loss_2d_out_of_bounds_class_index_xpu_float32 #653 PVC Runtime Jianhang Ready for Review CUDA bias case
test_GroupNorm_memory_format_xpu #653 PVC Runtime Yutao Done #677
test_upsamplingNearest2d_launch_fail_xpu #653 PVC Unexpected success Chunhuan Done Especially for cuda，keep skip
test_compare_cpu__refs_rsub_xpu_complex128 #645 ARC Required aspect fp64 is not supported on the device Daisy XFAIL fp64 i s not supported on ARC
test_nn_xpu.py::TestNNDeviceTypeXPU::test_variable_sequence_xpu_float16 #645 ARC Acc Jianhang Done Issue of oneDNN matmul. When fallbacking ATen matrix multiple operators, the case passes.
xpu/ test_dataloader.py #645 ARC   Penghui passed  
TestCompositeComplianceXPU.test_backward_var_mean_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
TestCompositeComplianceXPU.test_backward_var_mean_unbiased_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
TestCompositeComplianceXPU.test_operator_expand_as_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
test_backward_t_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
test_backward_fft_ihfft2_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
test_backward_lu_unpack_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
test_backward_nn_functional_max_unpool1d_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
test_backward_t_xpu_float32 #233 PVC COW Huaiyu/Chunhuan    
test_comprehensive_nn_functional_nll_loss_xpu_float16 #227 PVC ACC Huaiyu Done Not ARC specific issue. IGC optimization on data type promotion
@chuanqi129 chuanqi129 added this to the PT2.5 milestone Jul 30, 2024
@min-jean-cho min-jean-cho mentioned this issue Jul 31, 2024
Closed
81 tasks
@chuanqi129
Copy link
Contributor

@daisyden Create a standalone issue for COW issue and delay to pytorch 2.6.

@chuanqi129
Copy link
Contributor

@chunhuanMeng will double check Meta related issues

@hjhee
Copy link
Contributor

hjhee commented Aug 6, 2024

A PR (#693) is created for case test_nn_xpu.py::TestNNDeviceTypeXPU::test_variable_sequence_xpu_float16

@PenghuiCheng PenghuiCheng linked a pull request Aug 8, 2024 that will close this issue
Fixed regression issue and revert the change of checkIndexTensorTypes usage #597
Merged
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@majing921201 majing921201

@daisyden daisyden

@fengyuan14 fengyuan14

@PenghuiCheng PenghuiCheng

@xytintel xytintel

@yuchengliu1 yuchengliu1

@huaiyuzh huaiyuzh

@chunhuanMeng chunhuanMeng

Labels
ut_triaged
Projects
None yet
Milestone
PT2.5
Development

Successfully merging a pull request may close this issue.

Fixed regression issue and revert the change of checkIndexTensorTypes usage intel/torch-xpu-ops
10 participants
@hjhee @chuanqi129 @majing921201 @daisyden @fengyuan14 @PenghuiCheng @xytintel @yuchengliu1 @huaiyuzh @chunhuanMeng