Skip to content

tests/internal_bench: Benchmarks for descriptor-related features. #16825

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Open
wants to merge 2 commits into
base: master
Choose a base branch
from
Open

tests/internal_bench: Benchmarks for descriptor-related features. #16825

wants to merge 2 commits into from
+300 −0

Conversation

AJMansfield
Copy link
Contributor

@AJMansfield AJMansfield commented Feb 26, 2025 •
edited
Loading

Summary

This includes new internal_bench benchmarks for how quickly micropython can process class definitions, created in order to gauge the performance penalty for implementing __set_name__ in the different ways proposed in #15503, #16806, and #16816.

Testing

I've tested these benchmarks using the unix port inside WSL on my development machine, with the following results:

 . . .
internal_bench/class_create:
    0.351s (+00.00%) internal_bench/class_create-0-empty.py
    0.476s (+35.91%) internal_bench/class_create-1-slots.py
    0.484s (+38.01%) internal_bench/class_create-1.1-slots5.py
    0.427s (+21.87%) internal_bench/class_create-2-classattr.py
    0.773s (+120.54%) internal_bench/class_create-2.1-classattr5.py
    0.458s (+30.62%) internal_bench/class_create-3-instancemethod.py
    0.500s (+42.76%) internal_bench/class_create-4-classmethod.py
    0.462s (+31.65%) internal_bench/class_create-4.1-classmethod_implicit.py
    0.498s (+41.93%) internal_bench/class_create-5-staticmethod.py
    0.456s (+29.97%) internal_bench/class_create-6-getattribute.py
    0.467s (+33.13%) internal_bench/class_create-6.1-getattr.py
    0.388s (+10.58%) internal_bench/class_create-6.2-descriptor.py
    0.529s (+50.89%) internal_bench/class_create-6.3-descriptor_setname.py
    0.425s (+21.22%) internal_bench/class_create-6.4-property.py
    0.368s (+05.02%) internal_bench/class_create-7-inherit.py
    0.359s (+02.35%) internal_bench/class_create-7.1-inherit_initsubclass.py
 . . .
internal_bench/var:
    0.459s (+00.00%) internal_bench/var-1-constant.py
    0.620s (+35.05%) internal_bench/var-2-global.py
    0.443s (-03.42%) internal_bench/var-3-local.py
    0.441s (-03.96%) internal_bench/var-4-arg.py
    1.131s (+146.39%) internal_bench/var-5-class-attr.py
    0.539s (+17.34%) internal_bench/var-6-instance-attr.py
    0.555s (+20.85%) internal_bench/var-6.1-instance-attr-5.py
    0.537s (+17.04%) internal_bench/var-6.2-instance-speciallookup.py
    2.436s (+430.54%) internal_bench/var-6.3-instance-property.py
    3.102s (+575.63%) internal_bench/var-6.4-instance-descriptor.py
    3.144s (+584.85%) internal_bench/var-6.5-instance-getattr.py
    2.422s (+427.62%) internal_bench/var-7-instance-meth.py
    0.865s (+88.46%) internal_bench/var-8-namedtuple-1st.py
    0.924s (+101.30%) internal_bench/var-8.1-namedtuple-5th.py
 . . .

Trade-offs and Alternatives

Due to the overall slow speed of class creation in general, the class_create tests reduce the number of iterations from 20,000,000 to just 500,000; but this should still be wide enough to give the results adequate statistical significance (especially on non-multitasking microcontrollers).

@AJMansfield
tests/internal_bench/class_create: Benchmark class creation.
589c488 
Signed-off-by: Anson Mansfield <amansfield@mantaro.com>
@codecov Codecov
Copy link

codecov bot commented Feb 26, 2025 •
edited
Loading

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 98.53%. Comparing base (1034b17) to head (2409a66).
Report is 61 commits behind head on master.

Additional details and impacted files 
@@           Coverage Diff           @@
##           master   #16825   +/-   ##
=======================================
  Coverage   98.53%   98.53%           
=======================================
  Files         169      169           
  Lines       21822    21822           
=======================================
  Hits        21502    21502           
  Misses        320      320

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:
  • ❄ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
  • 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

@AJMansfield
tests/internal_bench/var: Benchmark descriptor access.
2409a66 
Signed-off-by: Anson Mansfield <amansfield@mantaro.com>
This was referenced Feb 26, 2025
Open
Open
Open
@AJMansfield
Copy link
Contributor Author

It looks like the unix port / settrace_stackless test failure is spurious? It failed in thread/thread_gc1.py, which should have nothing to do with anything this PR does.

@AJMansfield AJMansfield changed the title tests/internal_bench: Benchmarks for descriptor-related things. tests/internal_bench: Benchmarks for descriptor-related features. Feb 26, 2025
@dpgeorge dpgeorge added the tests Relates to tests/ directory in source label Mar 13, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
tests Relates to tests/ directory in source
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
2 participants
@AJMansfield @dpgeorge
pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy