On 19. 11. 19 21:55, Fabio Valentini wrote:
On Tue, Nov 19, 2019 at 8:32 PM Miro Hrončok
<mhroncok(a)redhat.com> wrote:
>
> On 19. 11. 19 20:20, Adam Jackson wrote:
>> In the spirit of positivity and collaboration, I spent a few minutes
>> looking at the results given to try to find some easy wins. Here's what
>> I found:
>>
>> python3-libs ships multiple copies of its pyc files, corresponding to
>> different optimization levels. I don't know what a good packaging
>> solution to that would look like, but if we only shipped the -O2 kind
>> (which seems appropriate for minimization, as they're smallest) we
>> could drop about 13M out of 32M, which seems pretty great.
>
> Note that it would make more sense to drop the optimized ones, as most of our
> packages invoke Python without -O or -OO.
>
> It would certainly need more planning and figuring out what shall happen if the
> optimized bytecode is not there and users actually run Python with -O or -OO,
> but we can discusses that if you want.
I wonder, does running python programs with -O2 actually yield
noticeably better performance?
If that's the case, maybe we could discuss setting -O2 as a
distro-wide default setting?
It drops asserts and docstrings. Some code require both to function, hence
defaulting to this would be dangerous.
--
Miro Hrončok
--
Phone: +420777974800
IRC: mhroncok