Hacker News Reader

Comment by anta40 | Hacker News Reader

What would be the challenge to write OS kernel in Lisp? Just X86, for the sake of umm simplicity, something like xv6: https://github.com/mit-pdos/xv6-public

Yes, I've read about Lisp machines. Sounds like a super expensive machine only accessible by big-budgeted research companies.

Most of Lisp folks on the net talk about how homoiconic language is elegant, or how REPL makes development faster, or how useful Lisp is for automatic theorem prover etc etc but very rarely I see Lisp folks talk about system programming, unlike C/C++/Rust folks.

View on HN

Mezzano is a recent Common Lisp OS: https://github.com/froggey/Mezzano

GUI-based. Cool. But unlike xv6, I guess skimming the code wouldn't be that easy/straightfoward.

And the build system looks unfamiliar, unlike xv6 which only requires 1 Makefile.

Let me try it, anyway.

>OS kernel in Lisp

GC is the problem there. Look up Carp-lisp

> GC is the problem there.

It is? That didn't stop Genera, Interlisp, Movitz or Mezzano to name a few. Outside Lisp there's also Smalltalk, Oberon and Singularity to name a few operating systems with GC.

If an OS has a GC-based kernel, it is automatically inferior to one lacking GC.

That's why the cases you brought up are not used in industrial/production environments in modern times.

But why?

Random pauses and inferior performance

Compared to the other pauses kernels make? Biscuit (POSIX in Go [0]) introduced at worst a pause of 115 microseconds and a slowdown due to GC and safety checks of 15%. But the forementioned systems don't have a kernel/userspace split; Microsoft measured the split made a server 25 to 33% _slower_ [1].

[0] https://pdos.csail.mit.edu/papers/biscuit:login19.pdf

[1] https://dl.acm.org/doi/10.1145/1178597.1178599

Thanks for the link. Didn't know about this.

Lisp machines are famous, but the complexity of a modern system means that even with a perfect kernel, you couldn't get any peripherals to work.

Reply on news.ycombinator.com