Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

MotorOS: a Rust-first operating system for x64 VMs

github.com

Post reply on HN
Print view View original

111–120 of 125 posts

Re: MotorOS: a Rust-first operating system for x64 VMs

#111

Post by toast0 »

prmoustache wrote: post #88
gamache wrote: post #66

Earlier quoted context omitted.

macOS and iOS are BSDs. Pretty good niche!

This is pretty much a myth. Both run a different kernel. macOS used be known to have a network stack (and maybe some stuff like a virtual file system) from freebsd but I am pretty sure most of the code has been replaced by now. Having some BSD userland binaries doesn't make your OS a BSD. Otherwise Windows is just a fork of curl.

Last I checked, Darwin sources for tcp still look a lot like FreeBSD circa 2000 plus some Apple patches (MPTCP). No syncookies in 2024, because FreeBSD added those months after Apple forked the stack.

Re: MotorOS: a Rust-first operating system for x64 VMs

#112

Post by Ericson2314 »

emmelaich wrote: post #56
nicce wrote: post #43

Earlier quoted context omitted.

> because of Linux's complexity Linux is not complex but there are some design issues with C ABI and glibc stability and symlinks.

Indeed I would say Linux is as simple as possible given various goals ... however, 'simple as possible' can still be complex!

Linux is definitely complex. Linux's ABI stability (good!) also means we cannot really blame C / glibc so much, it has internalized that problem.

I agree the goal posts make much of this complexity inevitable, but the solution is to stop trying to intersect everyone's objectives into a much narrower and daunting goal.

The big galaxy-brained goal should be:

1. Make more of Linux Rust.

2. Leverage Rust's superior expressive power to switch from a "configuration" mindset (Linux's current gazillion options on single giant code base) to a "composition mindset": we want to use types to combine various subsystems in various flavors.

3. Something like this and Linux proper can share lots of library code / subsystems. We can blur the line between "separate kernels" and "separate 'distrobutions' mixing and matching stand-alone Linux components".

I think this is the only sustainable way to reign in the accidental complexity.

Re: MotorOS: a Rust-first operating system for x64 VMs

#113

Post by Narishma »

aredox wrote: post #76
seanw444 wrote: post #34

Earlier quoted context omitted.

Replacing Linux is hard even for the BSDs, and those are very well established and considered basically next-in-line for similar tasks.

Maybe the BSD are too close to Linux to displace it.

It's not like the myriads of these hobby OSes (written in Rust or otherwise) are any different. They're all Unix clones to various degrees.

Re: MotorOS: a Rust-first operating system for x64 VMs

#114

Post by gamache »

prmoustache wrote: post #88
gamache wrote: post #66

Earlier quoted context omitted.

macOS and iOS are BSDs. Pretty good niche!

This is pretty much a myth. Both run a different kernel. macOS used be known to have a network stack (and maybe some stuff like a virtual file system) from freebsd but I am pretty sure most of the code has been replaced by now. Having some BSD userland binaries doesn't make your OS a BSD. Otherwise Windows is just a fork of curl.

I argue that having a BSD license (in Darwin), BSD heritage (NeXTSTEP, FreeBSD, briefly NetBSD), and a mostly BSD userland 20+ years into the project makes this OS a BSD.

Re: MotorOS: a Rust-first operating system for x64 VMs

#115

Post by ianbicking »

gpm wrote: post #44
ianbicking wrote: post #41

It could be cool if there was a WASM container built in Rust that could run in this OS. I don't really have a sense of how complicated a WASM container is, though it seems non-trivial to even decide what such a thing is, so it would be nice not to treat the WASM container itself as the OS (i.e., more room to safely experiment). WASM also seems like it dodges the ABI issue by being more explicitly about composition in…

When you say wasm container, you mean something like wasmtime that provides a non-browser wasm runtime? https://github.com/bytecodealliance/wasmtime

Yeah, just that sort of thing. I mean, heck, can you put #!wasmtime at the top of a wasm file? Would such a file literally run right now (assuming #! is supported, and wasmtime doesn't hit any MotorOS Rust limitations)?

Re: MotorOS: a Rust-first operating system for x64 VMs

#116

Post by weinzierl »

crabbone wrote: post #105
weinzierl wrote: post #33

Sounds interesting, but it also reminds me of what Linus once said when asked about fearing competition. From my memory his answer was something like: I really like writing device drivers. Few people like that and until someone young and hungry comes along who likes that I'm not afraid of competition.

I don't think drivers are what warrants Linux its popularity / indispensability. I worked in a few places where people, well, at least moderately, liked to write drivers. Also, consider that it would've been definitely possible to write an OS with a mechanism for module loading compatible with Linux drivers. My take on this question would be the ease of use for application developers. This ease of use consists of ple…

"Also, consider that it would've been definitely possible to write an OS with a mechanism for module loading compatible with Linux drivers."

Sure, you can do that and it will work exactly until the next change in the driver interface, which is not very long.

So you end up in maintenance hell and will still constantly be behind the curve, because Linux gets all the updates first. You will be less secure too, because you'll get the security patches later.

That being said, someone in a similar thread from a few days ago said one of the BSDs does that. Maybe it's not that bad.

Re: MotorOS: a Rust-first operating system for x64 VMs

#117

Post by prmoustache »

gamache wrote: post #114
prmoustache wrote: post #88

Earlier quoted context omitted.

This is pretty much a myth. Both run a different kernel. macOS used be known to have a network stack (and maybe some stuff like a virtual file system) from freebsd but I am pretty sure most of the code has been replaced by now. Having some BSD userland binaries doesn't make your OS a BSD. Otherwise Windows is just a fork of curl.

I argue that having a BSD license (in Darwin), BSD heritage (NeXTSTEP, FreeBSD, briefly NetBSD), and a mostly BSD userland 20+ years into the project makes this OS a BSD.

Darwin is not Mac OS X and vice-versa.

Re: MotorOS: a Rust-first operating system for x64 VMs

#118

Post by jacquesm »

xyst wrote: post #104
jacquesm wrote: post #57

It isn't hard to start writing a new operating system. It is very hard to support that operating system for the next 5 decades.

Linus Torvalds resolve is unbreakable

Yes, it's most impressive. I couldn't do it. That kind of long term focus is an incredible asset.

Re: MotorOS: a Rust-first operating system for x64 VMs

#119

Post by chlorion »

rrr_oh_man wrote: post #2

[flagged]

There are heaps and heaps of security issues in Linux (the kernel).

Simply mounting a file system can result in your system being compromised if the file system was maliciously crafted. This isn't something that is actively protected against.

Another issue I can think of OTTOMH, is unprivileged user namespaces causing lots of security issues to surface because certain parts of the kernel's code was written with the assumption that only "root" would be able to invoke it, so if you already were "root" securing this didn't matter.

There are many, many more issues similar to these, too many for me to write about, and probably many that are way beyond my comprehension. I don't think it's wrong to say that Linux has not been very secure.

Re: MotorOS: a Rust-first operating system for x64 VMs

#120

Post by nine_k »

jbverschoor wrote: post #73
weinzierl wrote: post #33

Sounds interesting, but it also reminds me of what Linus once said when asked about fearing competition. From my memory his answer was something like: I really like writing device drivers. Few people like that and until someone young and hungry comes along who likes that I'm not afraid of competition.

So Linux will essentially just become a HAL / "bios"

No, Linux remains where it is.

AFAICT this could run on top of Xen, VMWare, or the hypervisor which runs Win 10/11.

Post reply on HN