frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

Remove old ROCm version check in tests (#164245)

Browse Source 
This PR removes ROCm<6 version checks.

Pull Request resolved: https://github.com/pytorch/pytorch/pull/164245
Approved by: https://github.com/jeffdaily
This commit is contained in:
Yuanyuan Chen
2025-10-06 22:42:01 +00:00
committed by PyTorch MergeBot
parent 3912ba3e94
commit b63bbe1661
4 changed files with 2 additions and 15 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
test/test_matmul_cuda.py
View File

@ -36,7 +36,6 @@ from torch.testing._internal.common_utils import (
parametrize,
run_tests,
skipIfRocm,
skipIfRocmVersionLessThan,
TEST_CUDA,
TEST_WITH_ROCM,
TestCase,
@ -144,7 +143,6 @@ class TestMatmulCuda(InductorTestCase):
torch.backends.cuda.matmul.allow_fp16_accumulation = orig_fp16_accumulate
@onlyCUDA
@skipIfRocmVersionLessThan((5, 2))
# imported 'tol' as 'xtol' to avoid aliasing in code above
@toleranceOverride({torch.float16: xtol(atol=1e-1, rtol=1e-1),
torch.bfloat16: xtol(atol=1e-1, rtol=1e-1),
@ -158,7 +156,6 @@ class TestMatmulCuda(InductorTestCase):
@onlyCUDA
@xfailIfSM100OrLaterAndCondition(lambda params: params.get('dtype') == torch.bfloat16 and params.get('size') == 10000)
@skipIfRocmVersionLessThan((5, 2))
# imported 'tol' as 'xtol' to avoid aliasing in code above
@toleranceOverride({torch.float16: xtol(atol=7e-1, rtol=2e-1),
torch.bfloat16: xtol(atol=1e1, rtol=2e-1)})
@ -170,7 +167,6 @@ class TestMatmulCuda(InductorTestCase):
self.cublas_addmm(size, dtype, True)
@onlyCUDA
@skipIfRocmVersionLessThan((5, 2))
@dtypes(torch.float16)
# m == 4 chooses OUTPUT_TYPE reduction on H200
# m == 8 chooses OUTPUT_TYPE reduction on A100
@ -191,7 +187,6 @@ class TestMatmulCuda(InductorTestCase):
torch.backends.cuda.matmul.allow_fp16_reduced_precision_reduction = orig_precision
@onlyCUDA
@skipIfRocmVersionLessThan((5, 2))
# imported 'tol' as 'xtol' to avoid aliasing in code above
@toleranceOverride({torch.float16: xtol(atol=7e-1, rtol=2e-1),
torch.bfloat16: xtol(atol=1e1, rtol=2e-1)})