Also, regarding startup times. Here are our benchmarks, measured with Linux
perf (https://perf.wiki.kernel.org/index.php/Main_Page)
* Importing all the standard library without PEP 657:
Performance counter stats for './python import_all_stdlib.py' (500 runs):
303,78 msec task-clock # 0,999 CPUs utilized
( +- 0,05% )
4 context-switches # 0,014 K/sec
( +- 6,33% )
0 cpu-migrations # 0,000 K/sec
7.602 page-faults # 0,025 M/sec
( +- 0,05% )
1.146.643.923 cycles # 3,775 GHz
( +- 0,03% )
1.613.882.354 instructions # 1,41 insn per cycle
( +- 0,01% )
391.399.735 branches # 1288,440 M/sec
( +- 0,01% )
6.501.180 branch-misses # 1,66% of all
branches ( +- 0,02% )
0,304092 +- 0,000147 seconds time elapsed ( +- 0,05% )
* Importing all the standard library with PEP 657:
Performance counter stats for './python import_all_stdlib.py' (500 runs):
310,56 msec task-clock # 0,999 CPUs utilized
( +- 0,06% )
5 context-switches # 0,017 K/sec
( +- 9,39% )
0 cpu-migrations # 0,000 K/sec
8.359 page-faults # 0,027 M/sec
( +- 0,05% )
1.170.505.335 cycles # 3,769 GHz
( +- 0,03% )
1.637.448.123 instructions # 1,40 insn per cycle
( +- 0,01% )
396.921.300 branches # 1278,097 M/sec
( +- 0,01% )
6.568.450 branch-misses # 1,65% of all
branches ( +- 0,02% )
0,310904 +- 0,000179 seconds time elapsed ( +- 0,06% )
The overhead is a 2% increase in startup time, which is 0.00681200000000004
seconds of difference.
On Tue, 29 Jun 2021 at 16:39, Pablo Galindo Salgado <pablog...@gmail.com>
wrote:
> > I was expected the announcement of a BDFL delegate for PEP 657, as the
> author is a steering council member.
>
> Just to clarify this: as I was the author I didn't get to vote on the
> approval or rejection of the PEP. Also, there is nowhere that I know where
> this situation requires a BDFL delegate like you claim.
>
> > It seems that a PEP submitted by a SC member has been accepted by the SC
> with seemingly no external review.
>
> That is very not true. We have added ton of beedback for at least 2
> python-dev threads and a discourse thread:
>
> *
> https://discuss.python.org/t/pep-657-include-fine-grained-error-locations-in-tracebacks/8629
> *
> https://mail.python.org/archives/list/python-dev@python.org/thread/DB3RTYBF2BXTY6ZHP3Z4DXCRWPJIQUFD/
> *
> https://mail.python.org/archives/list/python-dev@python.org/thread/KR7ACFCUNMHT4M7R4XNHGRFV27HZBDFD/#KR7ACFCUNMHT4M7R4XNHGRFV27HZBDFD
>
> The design has been iterated many times and we have draft implementations
> from some of the alternative ideas as the ones proposed by Nathaniel.
>
> > interact poorly with debugging and profiling.
>
> "interact poorly with debugging and profiling." is making unfounded
> claims, which is especially bad for a feature that is supposed to aid
> debugging. You mentioned your worries on
> https://mail.python.org/archives/list/python-dev@python.org/thread/KR7ACFCUNMHT4M7R4XNHGRFV27HZBDFD/#KR7ACFCUNMHT4M7R4XNHGRFV27HZBDFD
> and we answered them.
>
> Kind regards,
> Pablo
>
>
>
>
> On Tue, 29 Jun 2021 at 16:30, Mark Shannon <m...@hotpy.org> wrote:
>
>> Hi,
>>
>> I was expected the announcement of a BDFL delegate for PEP 657, as the
>> author is a steering council member.
>>
>> It seems that a PEP submitted by a SC member has been accepted by the SC
>> with seemingly no external review.
>>
>> PEP 657 is likely to cause significant worsening of start up time and
>> interact poorly with debugging and profiling.
>>
>>
>> Cheers,
>> Mark.
>> _______________________________________________
>> Python-Dev mailing list -- python-dev@python.org
>> To unsubscribe send an email to python-dev-le...@python.org
>> https://mail.python.org/mailman3/lists/python-dev.python.org/
>> Message archived at
>> https://mail.python.org/archives/list/python-dev@python.org/message/OPTIO7SWLTEZCWMAHVDKF7WUX2LSQ46X/
>> Code of Conduct: http://python.org/psf/codeofconduct/
>>
>
_______________________________________________
Python-Dev mailing list -- python-dev@python.org
To unsubscribe send an email to python-dev-le...@python.org
https://mail.python.org/mailman3/lists/python-dev.python.org/
Message archived at
https://mail.python.org/archives/list/python-dev@python.org/message/TBDLJ5ZKN4Q7MUKOLTPNPHBQV5GUTVJO/
Code of Conduct: http://python.org/psf/codeofconduct/
