Minor update in cuda_pathfinder/DESCRIPTION.rst

[rwgk]

Previously overlooked, noticed in passing. This should have been included in PR #956.

[copy-pr-bot]

Auto-sync is disabled for ready for review pull requests in this repository. Workflows must be run manually.

Contributors can view more details about this message here.

[rwgk]

/ok to test

[rwgk]

Tracking a flake for easy future reference:

Test win-64 / py3.13, 13.0.1, wheels, GPU t4

============================= test session starts =============================
platform win32 -- Python 3.13.7, pytest-8.4.2, pluggy-1.6.0 -- C:\actions-runner\_work\_tool\Python\3.13.7\x64\python.exe
cachedir: .pytest_cache
benchmark: 5.1.0 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: C:\actions-runner\_work\cuda-python\cuda-python\cuda_core\tests
configfile: pytest.ini
plugins: benchmark-5.1.0
collecting ... collected 1 item

tests\cython\test_cython.py::test_get_cuda_native_handle FAILED          [100%]

================================== FAILURES ===================================
_________________________ test_get_cuda_native_handle _________________________

>   ???

test_get_cuda_native_handle.pyx:25: 
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
..\..\..\_tool\Python\3.13.7\x64\Lib\site-packages\cuda\core\experimental\_device.py:1207: in set_current
    ctx = self._get_primary_context()
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^
..\..\..\_tool\Python\3.13.7\x64\Lib\site-packages\cuda\core\experimental\_device.py:1033: in _get_primary_context
    ctx = handle_return(driver.cuDevicePrimaryCtxRetain(self._id))
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
cuda\core\experimental\_utils\cuda_utils.pyx:118: in cuda.core.experimental._utils.cuda_utils.handle_return
    _check_error(result[0], handle=handle)
cuda\core\experimental\_utils\cuda_utils.pyx:108: in cuda.core.experimental._utils.cuda_utils._check_error
    return _check_driver_error(error)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

>   raise CUDAError(f"{name}: {expl}")
E   cuda.core.experimental._utils.cuda_utils.CUDAError: CUDA_ERROR_LAUNCH_FAILED: An exception occurred on the device while executing a kernel. Common causes include dereferencing an invalid device pointer and accessing out of bounds shared memory. Less common cases can be system specific - more information about these cases can be found in the system specific user guide. This leaves the process in an inconsistent state and any further CUDA work will return the same error. To continue using CUDA, the process must be terminated and relaunched.

cuda\core\experimental\_utils\cuda_utils.pyx:69: CUDAError
======================== 1 failed in 171.89s (0:02:51) ========================

[rwgk]

For comparison, the rerun of the test that flaked before:

Test win-64 / py3.13, 13.0.1, wheels, GPU t4

============================= test session starts =============================
platform win32 -- Python 3.13.7, pytest-8.4.2, pluggy-1.6.0 -- C:\actions-runner\_work\_tool\Python\3.13.7\x64\python.exe
cachedir: .pytest_cache
benchmark: 5.1.0 (defaults: timer=time.perf_counter disable_gc=False min_rounds=5 min_time=0.000005 max_time=1.0 calibration_precision=10 warmup=False warmup_iterations=100000)
rootdir: C:\actions-runner\_work\cuda-python\cuda-python\cuda_core\tests
configfile: pytest.ini
plugins: benchmark-5.1.0
collecting ... collected 1 item
tests\cython\test_cython.py::test_get_cuda_native_handle PASSED          [100%]
============================== 1 passed in 0.60s ==============================

[github-actions]

Doc Preview CI
Preview removed because the pull request was closed or merged.