LuaJIT: New Garbage Collector (2012) | Hacker News Reader

LuaJIT: New Garbage Collector (2012)

wiki.luajit.org

135 pointstosh22 comments

Open articleView on HN

Previous discussion, with comments from Mike Pall and Patrick Walton:

https://news.ycombinator.com/item?id=4053969

Github issue, which Mike said "is up for takers":

https://github.com/LuaJIT/LuaJIT/issues/38

FYI: from 4.5 years ago

Mike Pall is a super hero. Not sure what his secret identity is, but the origin story is http://www.pagetable.com/?p=656

IIRC this project was abandoned after Mike Pall stepped down from LuaJIT maintenance.

There are a couple attempts made by other people to get this implemented:

- https://github.com/fsfod/LuaJIT/commits/gcarena - https://github.com/corsix/LuaJIT/commits/newgc

I'm not sure what the status is on either, though; they're obviously completely different implementations so it's not clear which one of these is preferable (also both authors seem to be busy with other stuff recently).

"Clone Mike Pall": https://github.com/LuaJIT/LuaJIT/issues/45

Fixes still being committed by Mike Pall, https://github.com/LuaJIT/LuaJIT/commits/master

What does he do right now?

Legend has it that after completing his mission, he returned to a future where performant JIT compilers have ensured world peace and the prosperity of mankind for thousands of years.

Still committing regularly to the 2.1 branch.

https://github.com/LuaJIT/LuaJIT/commits/v2.1

What's his day job? Internet is dark about this guy.

I've met Mike Pall in person. I've worked with him some since I helped facilitate a sponsorship from Google (https://opensource.googleblog.com/2010/01/love-for-luajit.ht...). A lot of his work on LuaJIT has been supported by various sponsorships (see: https://luajit.org/sponsors.html). This gives him a lot of freedom to work the way he wants to, instead of working as part of a larger organization.

Also note that the sponsorship page says that Mike is currently working on "unrelated projects", so apparently things besides LuaJIT are keeping him busy now.

What did he work on before LuaJIT? Seems like he could have been involved in CPU design? He thinks more like a hardware guy than a C programmer.

I don't know that much about his background. I didn't get much of a chance to talk shop with him (which is too bad, I would have loved to pick his brain about all sorts of things). I also assume from the lack of info about him on the Internet that he's a somewhat private person, so I definitely wanted/want to respect that.

This won't do all that much to fill in any gaps in his CV, but here's something else he wrote: http://www.pagetable.com/?p=656

Project updated as recently as June. http://wiki.luajit.org/latest_changes

Is there something about lua that makes it amenable to the performance of luaJIT? In priciple, would it be possible to get comparable performance out of something like pypy?

In other words: is getting pypy to perform like luajit "just" a matter of time/money/expertise, or are there technical considerations?

Mike Pall said,

Well, nothing really. It's just a matter of engineering, i.e. man-years put into the VM and the compiler.

He has a few posts on Reddit regarding this subject: https://www.reddit.com/user/mikemike

That quote was in context of

What in your opinion keeps LuaJITs performance away from C/C++, i.e. what other "hard" things need to be solved?

TBH that was the spirit of my question. I was wondering if there was something special about Lua (the language), and it seems like the answer is a resounding "no".

So basically, I should donate to pypy...

I'm not sure I could call that "resounding" - for one a PyPy dev disagreed with him in a sibling thread.

> Yes, you can go a long way without extra APIs etc, but sometimes there is info that's available, that's simply not there, because it's only in a programmers head.

https://www.reddit.com/r/programming/comments/19gv4c/why_pyt...

The new garbage collector should be based on well-researched and proven algorithms, together with a couple of thoroughly evaluated innovations, where appropriate. The real innovation should be in the specific mix of techniques, forming a coherent and well-balanced system, with meticulous attention to detail and relentless optimization for performance.

Sometimes I feel there's an inverse correlation between the strength of claims for the future and the end result.

Reply on news.ycombinator.com