Hi Team,
We are planning on adding a new function called `np.show_runtime` to
display the runtime information of the machine in addition to
`np.show_config` which displays the build-related information.
Relevant PR: ENH: Use threadpoolctl in show_runtime (a new function) #21468
<https://github.com/numpy/numpy/pull/21468>
Reasons for new function:
1. Will be helpful in root-causing issuing with information from the
acceleration frameworks.
2. Build time information alone is not always helpful [1]
Sample Output of new function
>>> np.show_runtime()
[{'simd_extensions': {'baseline': ['SSE', 'SSE2', 'SSE3'],
'found': ['SSSE3',
'SSE41',
'POPCNT',
'SSE42',
'AVX',
'F16C',
'FMA3',
'AVX2'],
'not_found': ['AVX512F',
'AVX512CD',
'AVX512_KNL',
'AVX512_KNM',
'AVX512_SKX',
'AVX512_CLX',
'AVX512_CNL',
'AVX512_ICL']}},
{'architecture': 'Zen',
'filepath':
'/usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.20.so',
'internal_api': 'openblas',
'num_threads': 12,
'prefix': 'libopenblas',
'threading_layer': 'pthreads',
'user_api': 'blas',
'version': '0.3.20'}]
Under the hood
`np.show_runtime` will use `threadpoolctl` [2] to get the runtime
information of the frameworks. In addition to this, we will also display
useful SIMD extensions that are supported.
Future Scope
1. When we move towards meson, we can leverage the additional information
and revamp `np.show_config` to utilize it.
2. If needed, we can gradually merge `show_runtime` and `show_config` to
display information in sections and various other formats.
Please do let us know if you have any suggestions on improving this
function or concerns about the information provided by the new function.
Thanks,
Ganesh
[1] show_config displays useless info when installed through conda-forge
channel <https://github.com/numpy/numpy/issues/14492> (#14492)
[2] threadpoolctl <https://pypi.org/project/threadpoolctl/>
_______________________________________________
NumPy-Discussion mailing list -- numpy-discussion@python.org
To unsubscribe send an email to numpy-discussion-le...@python.org
https://mail.python.org/mailman3/lists/numpy-discussion.python.org/
Member address: arch...@mail-archive.com
