OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 12:20:52 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

[AOTI] Skip test_simple_multi_arch_embed_kernel_binary_True_cuda (#157301)

Browse Source 
Summary: For https://github.com/pytorch/pytorch/issues/156930, still no clue on what went wrong as it is not reproducible locally, but somehow the problem seems only exists when embed_kernel_binary is True. Let's skip it for now.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/157301
Approved by: https://github.com/yushangdi
This commit is contained in:
Bin Bao 2025-07-07 18:43:03 -07:00 committed by PyTorch MergeBot
parent 75f489d37f
commit e5edd013ab
1 changed files with 3 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
test/inductor/test_aot_inductor.py
View File

@ -181,7 +181,9 @@ class AOTInductorTestsTemplate:
"toolchain doesn't support ptx to fatbin",
)
@skipIfRocm
@common_utils.parametrize("embed_kernel_binary", [True, False])
# Skip embed_kernel_binary == True for now as it shows random
# failure on CI
@common_utils.parametrize("embed_kernel_binary", [False])
def test_simple_multi_arch(self, embed_kernel_binary):
if self.device != GPU_TYPE:
raise unittest.SkipTest("requires GPU_TYPE")