Hacker News Reader

Comment by pshirshov | Hacker News Reader

> Support your graphics cards on linux using kernel drivers that you upstream. All of them. Not just a handful - all the ones you sell from say 18 months ago till today.

All the stuff works even if it's not officially supported. It's not that hard to set a single environment variable (HSA_OVERRIDE_GFX_VERSION).

Like literally, everything works, from Vega 56/64 to ryzen 99xx iGPUs.

Also, try nixos. Everything literally works with a single config entry after recent merge of rocm 6.3. I successfully run a zoo of various radeons of different generations.

View on HN

I’m using Nix and setting HSA_OVERRIDE_GFX_VERSION. It’s not working on my GPU (Radeon RX7600)

This https://www.techpowerup.com/gpu-specs/radeon-rx-7600.c4153 says that's a gfx1102. I'd expect that to work out of the box. Linux kernel version vaguely near one of the nominally supported ones?

I mean, so would I expect it to work out of the box, which is part of why I find this situation frustrating

But why do you even have to do this fucking bullshit that you randomly stumble upon, while googling error message after error message, ending up in random github repos and issues?

And no, just because the three random cards you have work doesn't mean "everything works". Just tried an MI300A a few months ago... I just wanted to test ollama as this is one of the hottest applications for GPU acceleration now, it will surely be well supported right? First, the gfx version listed for it in the ollama docs is wrong - but OK, figured it out. Then Tried some random models with it, the only output it ever generates is GGGGGGGGGGGGG. Apparently only fp16 models work, nothing more quantized. So I pick one explicitly. Then it's slower than running on the cpu in the same system.

Thanks but no thanks; this cost me two days when Nvidia just works first try.

> But why do you even have to do this fucking bullshit

Because it's like 2-4 times cheaper than to go nvidia?..

> the three random cards you have

It's more than 3 random cards. I run 6900 XT, 7900 XTX, W7900 Pro, VII, VII Pro, Vega 56, Vega 64, 6800 XT, 5700 XT plus I've experimented with a 9950 iGPU, a 5xxx series iGPU and the only thing which didn't work was 3400g iGPU.

> Apparently only fp16 models work

fp8 works for me

> Because it's like 2-4 times cheaper than to go nvidia?..

That still makes nvidia way cheaper compared to the time wasted getting that shit to run plus debug an unknown amount of issues down the road. I told my boss "look I wasted two days on this and it's barely working, I could've done actual work here". Two other colleagues poked at it some more without success. going Nvidia was a no-brainer then.

Yes, if you can fork out 5-12K for an nvidia card, it's a no-brainer. Otherwise an equivalent 1-3K radeon could do some good job for you.

If you decide to dispose of that MI300A of yours, send it to me, I'll pay for shipping and even buy you a pizza.

Reply on news.ycombinator.com