The Experimental GCN 1.0 GPU Support Might Be Dropped From AMDGPU Linux Driver

[MagicMyth]

Well this sucks. I have a perfectly well functioning Radeon R7 370 that is able to play everything available on Linux just fine using the AMDGPU driver, including games like Rise of the Tomb Raider and now thanks to Proton far more games just effortlessly. We are not simply talking about being not able to run the latest and greatest here but loosing the existing ability to play games going back the last couple of years. Hopefully this can be forked and some skilled enough developers can get things working well enough (wish I had that low level skill).

[ssokolow]

I do understand the frustration ....but what do you want to buy instead? Nvidia is even worse and Intel has no dGPU card yet.

I've never had an nVidia card outlast its Linux driver support window and Canonical does a good job of picking kernel-driver combinations that Just Work™.

...and I don't mean that the hardware died young. Since 2007, I went from a GeForce 7600 to a GT430 to my current GTX750 and both previous cards died in the ~5-6 year window I've come to expect as a reasonable lifespan for a video card in a machine that runs 24/7/365. Heck, the 7600's usable life was ended by having some capacitors go bad, so it could probably be revived.

[jason.oliveira]

I just put an R7 240 in my camera server at work, so this is a huge bummer. Now I have to keep my kernel/xorg/mesa/libdrm frozen just to maintain AMDGPU working (which was not fun to set up).

[r1348]

Ok, and what about GCN 1.0 fixes that were implemented in AMDGPU only? https://www.phoronix.com/scan.php?pa...eri-2018&num=1

[mzs.112000]

All of my GPUs are GCN 1.0, and yet they manage to play modern games just fine(my 8970M plays ARK at 30fps medium settings). Why would we drop support for cards that are still usable in 2020.

Same thing that happened with FGLRX when it was dropped, now we have to use a slower alternative driver(Radeon).

[Nille_kungen]

GCN1 is so different from later GCN so it makes sense, if i remember correctly it was someone in phoronix forum that added the initial support.
I never used amdgpu with my GCN1 i only used radeonsi, i didn't have vulkan but i didn't have much use for it but now with many wine games using vulkan i can see why one would want vulkan support.
I didn't get the expression from that post that anything has been decided yet, if it wasn't for vulkan support i don't think anyone would mind if it was dropped from amdgpu but to be honest that hardware wasn't sold to me with vulkan support because there wasn't any vulkan API then.
About the firmware and UVD i wouldn't be surprised if it needed to go thru a detailed review by legal and we all knows how long that could take and i don't think that cost is worth it.
Try selling that to economics.

How well does vulkan work and how fast is it on old GCN1.0 hardware?
I hope someone that use vulkan on GCN1.0 hardware could enlighten me about current state of vulkan support.

[mzs.112000]

Nille_kungen Vulkan seems like it's working nicely on GCN 1.0. ARK: Survival Evolved, a notoriously under-optimized game, runs at 30fps on medium settings, on my 8970M, which is a laptop GPU.

Right now the only thing limiting my laptop performance is the throttling (It's either the battery or the charger, causing the CPU to stay at 798MHz all the time.)

[Nille_kungen]

Nille_kungen Vulkan seems like it's working nicely on GCN 1.0. ARK: Survival Evolved, a notoriously under-optimized game, runs at 30fps on medium settings, on my 8970M, which is a laptop GPU.

Right now the only thing limiting my laptop performance is the throttling (It's either the battery or the charger, causing the CPU to stay at 798MHz all the time.)

Thanks for the feedback that it works well with your hardware.
Is that also true for vulkan games in general? wine?
What CPU/APU does your laptop have?
My old laptop had an APU with NI graphics (A10-5750M/HD 8650G) + an 8670M so i didn't think amdgpu was an option for me, but maybe that combo works now?

[bridgman]

Here we go again. AMD doesn't provide any official driver for my RadeonSI GPU since 2015, and I have been using Mesa drivers for 4 years. Now they are going to drop also the firmware completely from the kernel! They don't provide any support their own GPUs, and they don't want anyone else to support them! WTF is wrong with them?!!

Huh ? If you are using Mesa then you *are* using the official drivers (ignoring Vulkan for a moment).

When you say "official" are you thinking about the packaged drivers ? They use the same open source driver code as upstream for non-PRO, and the PRO option (for workstation aka RadeonPRO) replaces the 3D drivers with closed source OpenGL and Vulkan code.

The radeonsi driver supports everything from SI through Navi - are you talking about an SI (GFX6) GPU ? If so then it is supported by the packaged drivers even today.

2015... maybe you are talking about fglrx ? If so then we replaced that with the amdgpu/amdgpu-pro package... it wasn't just dropped for older parts.

The point you may be missing is that we replaced fglrx with an upstream-based driver, where AMD developers continue to be the primary developers of the upstream code, and we added the fglrx kernel/X team to the upstream effort as well.

[bridgman]

bridgman Umpf...mh usually I'm advocating the way AMD is approaching their Open Source Drivers. But this decision is premature.

What decision are you talking about ?

The only "decisions" I am aware of are (a) Christian deciding to make a casual comment on a mailing list about the *possibility* of retiring amdgpu support for SI parts if we couldn't get the video working and (b) Michael deciding to turn that comment into an article.

No decisions have been made about amdgpu... everyone seems to be reacting to other people's comments rather than reacting to the article.

I guess that's the "internet echo chamber" thing everyone talks about

BTW the article talks about "AMD is unlikely to release" firmware to let UVD work with amdgpu, but that is kind of misleading since it implies that the firmware exists to release. The reality is that the firmware does not exist and does not seem to be feasible due to code size limitations.