Replace rsmi_init with amdsmi_init (via dlsym) in intra_node_comm#3299
Open
adam360x wants to merge 62 commits into
Open
Replace rsmi_init with amdsmi_init (via dlsym) in intra_node_comm#3299adam360x wants to merge 62 commits into
adam360x wants to merge 62 commits into
ROCm Repo Management API / Tests / Tests / Test Inductor / Run pytorch_inductor_1
failed
Jun 15, 2026 in 0s
GPUTests.test_var_mean_tile_reduction_True_cuda failed
GPUTests.test_var_mean_tile_reduction_True_cuda failed
Details
GPUTests.test_var_mean_tile_reduction_True_cuda
AssertionError: Tensor-likes are not close!
Mismatched elements: 4 / 4 (100.0%)
Greatest absolute difference: 0.58544921875 at index (0, 3) (up to 1e-05 allowed)
Greatest relative difference: 0.57568359375 at index (0, 1) (up to 0.001 allowed)
The failure occurred for item [2]
To execute this test, run the following from the base repo dir:
PYTORCH_TEST_WITH_ROCM=1 python test/inductor/test_torchinductor.py GPUTests.test_var_mean_tile_reduction_True_cuda
This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0
Stack trace
Traceback (most recent call last):
File "/var/lib/jenkins/pytorch/test/inductor/test_torchinductor.py", line 5997, in test_var_mean
self.common(
File "/opt/conda/envs/py_3.12/lib/python3.12/contextlib.py", line 81, in inner
return func(*args, **kwds)
^^^^^^^^^^^^^^^^^^^
File "/var/lib/jenkins/pytorch/test/inductor/test_torchinductor.py", line 781, in check_model_gpu
check_model(
File "/var/lib/jenkins/pytorch/test/inductor/test_torchinductor.py", line 600, in check_model
assert_equal_fn(
File "/opt/conda/envs/py_3.12/lib/python3.12/site-packages/torch/_dynamo/test_case.py", line 113, in assertEqual
return super().assertEqual(x, y, *args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/py_3.12/lib/python3.12/site-packages/torch/testing/_internal/common_utils.py", line 4357, in assertEqual
raise error_metas.pop()[0].to_error( # type: ignore[index]
AssertionError: Tensor-likes are not close!
Mismatched elements: 4 / 4 (100.0%)
Greatest absolute difference: 0.58544921875 at index (0, 3) (up to 1e-05 allowed)
Greatest relative difference: 0.57568359375 at index (0, 1) (up to 0.001 allowed)
The failure occurred for item [2]
To execute this test, run the following from the base repo dir:
PYTORCH_TEST_WITH_ROCM=1 python test/inductor/test_torchinductor.py GPUTests.test_var_mean_tile_reduction_True_cuda
This message can be suppressed by setting PYTORCH_PRINT_REPRO_ON_FAILURE=0
Loading