←back to thread

John Carmack's arguments against building a custom XR OS at Meta

(twitter.com)
520 points OlympicMarmoto | 3 comments | 29 Aug 25 16:45 UTC | HN request time: 0.821s | source
Show context
jnwatson ◴[29 Aug 25 17:46 UTC] No.45067216[source]▶
I've written a lot of low level software, BSPs, and most of an OS, and the main reason to not write your own OS these days is silicon vendors. Back in the day, they would provide you a spec detailed enough that you could feasibly write your own drivers.

These days, you get a medium-level description and a Linux driver of questionable quality. Part of this is just laziness, but mostly this is a function of complexity. Modern hardware is just so complicated it would take a long time to completely document, and even longer to write a driver for.

replies(13): >>45067491 #>>45069282 #>>45069287 #>>45069349 #>>45069690 #>>45070345 #>>45071036 #>>45071086 #>>45072259 #>>45072391 #>>45073789 #>>45075476 #>>45081942 #
mbac32768 ◴[29 Aug 25 20:54 UTC] No.45069287[source]▶
Yeah this. I tried to modify a hobby OS recently so it would process the "soft reboot" button (to speed up being rebooted in GCP) and it was so unbelievably hard to figure out how to support it. I tried following the instructions on the OS Dev Wiki and straight up reading what both Linux and FreeBSD do and still couldn't make progress. Yes. The thing that happens when you tell Windows or Linux to "restart". Gave up on this after spending days on it.

The people who develop OSes are cut from a different cloth and are not under the usual economic pressures.

replies(3): >>45069466 #>>45069733 #>>45070466 #
sitkack ◴[29 Aug 25 21:42 UTC] No.45069733[source]▶
The VMM on GCP has only really been tested with Linux. You are kinda wasting your time, the only way to make it work is to make the hobby OS Linux.
replies(1): >>45070064 #
toast0 ◴[29 Aug 25 22:19 UTC] No.45070064[source]▶
> You are kinda wasting your time, the only way to make it work is to make the hobby OS Linux.

Not the parent, but of course they're wasting their time... That's the point of a hobby OS.

I'm working on a hobby OS, and I have no illusions that it's most likely fewer than 10 people will ever run it, and less than 100 will hear about it, but it lets me explore some interesting (to me) ideas, and forces me to learn a little more about random pieces of computing. If I ran on GCP, I'd want the reboot button to work. That sounds useful.

On the topic, I don't see why anyone would want to build a general purpose OS. There's enough already and even with the shrinking of hardware variety, there's a lot of stuff to support to make a general purpose OS work on enough hardware for people to consider using it. You can take Linux or a BSD and hack it up pretty good to explore a lot of OS ideas. Chances are you're going to borrow some of their drivers anyway, and then you'll end up with at least some similarity... may as well start there and save a lot of time. (My hobby OS has a custom kernel and custom drivers, but I only support a bare minimum of devices... (pc) console i/o, one real NIC, and virtio-net... that's all I need; I might add support for more NICs and more consoles later)

replies(1): >>45070231 #
sitkack ◴[29 Aug 25 22:41 UTC] No.45070231[source]▶
I didn't say they were wasting their time on their hobby OS, they are wasting their time trying to get it to do very esoteric thing on GCP.

They aren't trying to get reboot to work, they are trying to get their version of kexec to work so their hobby os reboots faster.

https://wiki.archlinux.org/title/Kexec

The biggest scam in the OS world is drivers, we should demand more out of our hardware. Drivers shouldn't be necessary.

replies(4): >>45070352 #>>45072184 #>>45073077 #>>45075278 #
1. baq ◴[30 Aug 25 08:53 UTC] No.45073077[source]▶
Hardware is so broken that getting useful functionality basically amounts to casting magic spells and drivers are supposed to be master wizards who know all the points where the spell book is wrong or incomplete. If you think drivers are bad, don’t look at the hardware, you’ll get depressed.
replies(1): >>45075820 #
2. sitkack ◴[30 Aug 25 16:16 UTC] No.45075820[source]▶
This is fundamentally the problem. Just like being able to send OTA updates has enshittified all software, having this magic shim layer that fixes hardware problems has enabled shit hardware, and then foisted all this complexity into the OS. Many abstractions are like bondo, they just cover rot.

I am addressing your comment and eru's question about drivers.

The hardware that would normally need drivers should present itself over a fixed, well documented protocol. Think virtio, or usb device classes but more comprehensive. This would also allow for said hardware to rigorously tested before it ever sees an OS. As it is now, because the hardware is shit and requires a driver, you can't really test the hardware in a way that an OS would expect because it requires the OS driver to even start to function. The job of the OS is now to repair broken hardware.

https://docs.oasis-open.org/virtio/virtio/v1.3/virtio-v1.3.h...

https://wiki.osdev.org/Virtio

https://en.wikipedia.org/wiki/USB_communications_device_clas... (the only good thing to come out of usb)

replies(1): >>45081874 #
3. rcxdude ◴[31 Aug 25 09:46 UTC] No.45081874[source]▶
The problem with your wish is that's you're kinda getting it in some cases, and it's a bit of a monkey paw. Hardware vendors are increasingly creating systems that abstract away the underlying hardware from the OS (usually by writing their own software on some other core that really drives the hardware), but the problem is they're generally closed, buggy, and leaky, and so the OS stops really being the OS of the system and instead you have a collection of barely-related subsystems that it's really difficult to get to work together effectively, and way more security holes than you can shake a stick at. (Oh, and they're usually only ever tested against one particular OS and so they're not actually particularly portable)