Skip to content

Commit

Permalink
Skip float8 dtype UT cases and workaround (#1211)
Browse files Browse the repository at this point in the history
1. Skip float8 dtype UT cases
2. workaround: skip some cases because UT random failure

---------

Signed-off-by: Cheng, Penghui <[email protected]>
Co-authored-by: Deng, Daisy <[email protected]>
  • Loading branch information
PenghuiCheng and daisyden authored Dec 27, 2024
1 parent 802ea31 commit 826389a
Showing 1 changed file with 15 additions and 0 deletions.
15 changes: 15 additions & 0 deletions test/xpu/skip_list_common.py
Original file line number Diff line number Diff line change
Expand Up @@ -2,6 +2,21 @@
"test_ops_xpu.py": (
# Skip list of base line

# XPU implementation doesn't claimn FP8 now
# https://github.com/intel/torch-xpu-ops/issues/461
"float8",

# workarounds for the following tests
# https://github.com/intel/torch-xpu-ops/issues/1214
"test_python_ref__refs_exp_xpu_complex128",
"test_python_ref__refs_sigmoid_xpu_complex128",
"test_python_ref_executor__refs_log2_executor_aten_xpu_complex128",
"test_python_ref_executor__refs_exp_executor_aten_xpu_complex128",
"test_python_ref_torch_fallback__refs_log2_xpu_complex128",
"test_python_ref_torch_fallback__refs_log10_xpu_complex128",
"test_python_ref_torch_fallback__refs_sigmoid_xpu_complex128",
"test_python_ref_executor__refs_log10_executor_aten_xpu_complex128",

# To be removed from this file.
# CUDA and XPU both XFAIL now.
"test_out_narrow_copy_xpu_float32",
Expand Down

0 comments on commit 826389a

Please sign in to comment.