[issue23530] os and multiprocessing.cpu_count do not respect cpuset/affinity

[Davin Potts]

Davin Potts added the comment:

Adding an option does sound like a better possibility.  Still, when I start looking through the examples that psutil provides, it reminds me how this is but one small piece of a much larger picture which psutil has done a nice, focused job of working to address.

If the patch you create were to depend upon gnu coreutils, I do not think it can be accepted for licensing reasons.  Interestingly psutil does not appear to depend upon that library.

Regarding the docs, what text would you propose instead of what's currently there for describing cpu_count?

----------

_______________________________________
Python tracker <report at bugs.python.org>
<http://bugs.python.org/issue23530>
_______________________________________