OSSForks/pytorch
1
0
Fork 0
mirror of https://github.com/zebrajr/pytorch.git synced 2025-12-06 00:20:18 +01:00
Code Issues Actions 38 Packages Projects Releases Wiki Activity

[ROCm][CI] adjust tf32 tolerance for test_compile_kernel_advanced (#163783)

Browse Source 
Fixes #ISSUE_NUMBER

Pull Request resolved: https://github.com/pytorch/pytorch/pull/163783
Approved by: https://github.com/jeffdaily

Co-authored-by: Jeff Daily <jeff.daily@amd.com>
This commit is contained in:
Jeff Daily 2025-09-24 19:39:12 +00:00 committed by PyTorch MergeBot
parent 71eec6a0bf
commit 0dce2afd44
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
test/test_cuda.py
View File

@ -6923,7 +6923,7 @@ class TestCompileKernel(TestCase):
with self.assertRaises(RuntimeError):
kernel.set_shared_memory_config(excessive_shared_mem)
@tf32_on_and_off(0.005)
@tf32_on_and_off(0.05 if TEST_WITH_ROCM else 0.005)
@unittest.skipIf(not TEST_CUDA, "No CUDA")
def test_compile_kernel_advanced(self):
# Test matrix multiplication