Atari Open-Source Linux DRM Graphics Driver Being Worked On In 2022

[phoronix]

Phoronix: Atari Open-Source Linux DRM Graphics Driver Being Worked On In 2022

In addition to the OpenChrome DRM/KMS driver hoping to be finally mainlined in 2022 for supporting aging VIA graphics hardware from the long-ago days of their x86 chipsets, separately there is a DRM/KMS kernel driver in the works for something even older... A Linux DRM graphics driver for the Atari Falcon from the early 90's...

https://www.phoronix.com/scan.php?page=news_item&px=Atari-DRM-Driver-2022

[Vorpal]

Cool!

(Cue person complaining about people spending time and effort on linux for retro computers).

[timofonic]

Despite this may not be useful for most, the DRM low-color framebuffer support might be useful for certain hardware such as embedded.

[skeevy420]

Cool!

(Cue person complaining about people spending time and effort on linux for retro computers).

Retro!? Puh-leez. This has only been tested on a custom Atari VM for modern hardware. Much Outrage.

/s

[Quackdoc]

I wonder if we can find this a viable use for hardware this old outside of leisure.

[Developer12]

There's no vintage hardware involved at all. This seems to be entirely targeted at running Linux in a modern VM on modern computers.

Sure, the VM is heavily inspired-by/compatible-with some old hardware, but it's definitely modern software with modern features.

It's reminiscent of Pico8, which is a modern game engine that emulates a 1980's-style computer for which no real hardware ever existed. Pico8 only exists for game developers to make new retro-style games. eg. https://freds72.itch.io/poom/devlog/...ourney-to-poom

[skeevy420]

I wonder if we can find this a viable use for hardware this old outside of leisure.

From the ARAnyM page:

Maybe some hardware vendor could even sell preconfigured computers that would boot directly to TOS desktop... That could help some of the remaining Atari users that wish to upgrade their Atari machines.

We could start a non-profit on Kickstarter to source some hardware and make some money. Since there aren't very many Atari users, I assume, that means we'd also have to sell FreeDOS, RetroArch, Haiku, AROS, and other niche OS setups. The RetrOS NPC (Non Profit Company). We're allowed to make more than a living wage at a non-profit. We'd just need transparency in regards to our numbers, cap our pay and employee pay at something reasonable like $400K a piece (I figure if it's good enough for the President, it's good enough for me), and then donate all our profits after that to open source projects that we depend on. Everybody wins. Is that a viable enough use?

Spec wise -- any AM4 Zen or newer APU with at least 4GBx2 of DDR4-2400 is good enough so we'd be able to source and sell older and cheaper chips that are more than powerful enough for most all retro needs. If demand rises we could start making higher end models. 4GBx2 DDR4-2400 would be the minimum ram spec for any machine; preferably faster, but in the beginning these would need to be lower cost retro-like machines to get people excited and do an impulse buy.

We won't be one of those companies that sells an APU with one stick of ram. There's a special place in Hell for the people who sign off on that.

[skeevy420]

It's reminiscent of Pico8, which is a modern game engine that emulates a 1980's-style computer for which no real hardware ever existed. Pico8 only exists for game developers to make new retro-style games. eg. https://freds72.itch.io/poom/devlog/...ourney-to-poom

That was a very interesting read. Even though I grew up with those graphics, those screenshots and gifs make me appreciate the PS2 era and newer games even more now. Nostalgia and actually seeing it aren't the same things.

[_ReD_]

Useless—and delightfully so!
This quirky stuff is what makes Linux so much more than just.

[yoshi314]

i wonder if such a driver should be included in the kernel mainline.

to be fair, we do have drivers for specific virtualization solutions. but this seems like they are really pushing it with something that seems like a very obscure use case.

at some point the distributions will likely draw the line between useful and optional drivers, because this is clearly starting to get out of hand for a typical desktop/server user. Arch Linux already split the firmware into two packages since it got so huge and most desktop users didn't really need most of it.

Could be that at some point we'll have the same thing with drivers - common desktop drivers, common server drivers, exotic drivers.