Showing content from http://mail.python.org/pipermail/python-dev/attachments/20180711/0ce1913d/attachment.html below:
<div dir="ltr">May be is something obvious but I find myself forgetting often about<div>the fact that most modern CPUs can change speed (and energy consumption)</div><div>depending on a moving average of CPU load.</div><div><br></div><div>If you don't disable this "green" feature and the benchmarks are quick then the</div><div>result can have huge variations depending on exactly when and if the CPU</div><div>switches to fast mode.</div></div><br><div class="gmail_quote"><div dir="ltr">On Wed, Jul 11, 2018 at 12:53 AM Victor Stinner <<a href="mailto:vstinner@redhat.com">vstinner@redhat.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">The pyperformance benchmark suite had micro benchmarks on function<br>
calls, but I removed them because they were sending the wrong signal.<br>
A function call by itself doesn't matter to compare two versions of<br>
CPython, or CPython to PyPy. It's also very hard to measure the cost<br>
of a function call when you are using a JIT compiler which is able to<br>
inline the code into the caller... So I removed all these stupid<br>
"micro benchmarks" to a dedicated Git repository:<br>
<a href="https://github.com/vstinner/pymicrobench" rel="noreferrer" target="_blank">https://github.com/vstinner/pymicrobench</a><br>
<br>
Sometimes, I add new micro benchmarks when I work on one specific<br>
micro optimization.<br>
<br>
But more generally, I suggest you to not run micro benchmarks and<br>
avoid micro optimizations :-)<br>
<br>
Victor<br>
<br>
2018-07-10 0:20 GMT+02:00 Jeroen Demeyer <<a href="mailto:J.Demeyer@ugent.be" target="_blank">J.Demeyer@ugent.be</a>>:<br>
> Here is an initial version of a micro-benchmark for C function calling:<br>
><br>
> <a href="https://github.com/jdemeyer/callbench" rel="noreferrer" target="_blank">https://github.com/jdemeyer/callbench</a><br>
><br>
> I don't have results yet, since I'm struggling to find the right options to<br>
> "perf timeit" to get a stable result. If somebody knows how to do this, help<br>
> is welcome.<br>
><br>
><br>
> Jeroen.<br>
> _______________________________________________<br>
> Python-Dev mailing list<br>
> <a href="mailto:Python-Dev@python.org" target="_blank">Python-Dev@python.org</a><br>
> <a href="https://mail.python.org/mailman/listinfo/python-dev" rel="noreferrer" target="_blank">https://mail.python.org/mailman/listinfo/python-dev</a><br>
> Unsubscribe:<br>
> <a href="https://mail.python.org/mailman/options/python-dev/vstinner%40redhat.com" rel="noreferrer" target="_blank">https://mail.python.org/mailman/options/python-dev/vstinner%40redhat.com</a><br>
_______________________________________________<br>
Python-Dev mailing list<br>
<a href="mailto:Python-Dev@python.org" target="_blank">Python-Dev@python.org</a><br>
<a href="https://mail.python.org/mailman/listinfo/python-dev" rel="noreferrer" target="_blank">https://mail.python.org/mailman/listinfo/python-dev</a><br>
Unsubscribe: <a href="https://mail.python.org/mailman/options/python-dev/agriff%40tin.it" rel="noreferrer" target="_blank">https://mail.python.org/mailman/options/python-dev/agriff%40tin.it</a><br>
</blockquote></div>
RetroSearch is an open source project built by @garambo
| Open a GitHub Issue
Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo
HTML:
3.2
| Encoding:
UTF-8
| Version:
0.7.4