News for hackers. Stuff that matters.

A comment on Modern Python Performance Considerations. Parent comment

Re:Modern Python Performance Considerations by chrisseaton on Thursday May 05, 2022 @01:36PM (#31276293)
> This kills most potential optimizations, or even hard-code generation.

It doesn’t - this has been a basically solved problem since Self and deoptimisation were invented.

[ Parent | on HN ]
Re:Modern Python Performance Considerations by Animats on Thursday May 05, 2022 @03:15PM (#31277351)
In theory, yes. In CPython, apparently not. In PyPy, yes.[1] PyPy has to do a lot of extra work to permit some unlikely events.

[1] https://carolchen.me/blog/jits-impls/

[ Parent | on HN ]
Re:Modern Python Performance Considerations by chrisseaton on Thursday May 05, 2022 @06:25PM (#31279052)
You’re trying to correct me by posting my own mentee’s blog post at me.
[ Parent | on HN ]