log get flodded with amdgpu [drm] Cannot find any crtc or sizes

[corgan]

Hello,

 

after I updated today from 6.9.2 to 6.12.2 my syslog gets extremly flodded with these messaged:

 

Quote

 

Feb 14 20:40:30 serva4 kernel: amdgpu 0000:2c:00.0: [drm] Cannot find any crtc or sizes

Feb 14 20:40:40 serva4 kernel: [drm] PCIE GART of 256M enabled (table at 0x000000F400000000).

Feb 14 20:40:40 serva4 kernel: [drm] UVD and UVD ENC initialized successfully.

Feb 14 20:40:40 serva4 kernel: [drm] VCE initialized successfully.

Feb 14 20:40:40 serva4 kernel: amdgpu 0000:2c:00.0: [drm] Cannot find any crtc or sizes

Feb 14 20:40:50 serva4 kernel: [drm] PCIE GART of 256M enabled (table at 0x000000F400000000).

Feb 14 20:40:50 serva4 kernel: [drm] UVD and UVD ENC initialized successfully.

Feb 14 20:40:50 serva4 kernel: [drm] VCE initialized successfully.

Feb 14 20:40:50 serva4 kernel: amdgpu 0000:2c:00.0: [drm] Cannot find any crtc or sizes

Feb 14 20:41:00 serva4 kernel: [drm] PCIE GART of 256M enabled (table at 0x000000F400000000).

Feb 14 20:41:00 serva4 kernel: [drm] UVD and UVD ENC initialized successfully.

Feb 14 20:41:00 serva4 kernel: [drm] VCE initialized successfully.

Feb 14 20:41:00 serva4 kernel: amdgpu 0000:2c:00.0: [drm] Cannot find any crtc or sizes

Feb 14 20:41:10 serva4 kernel: [drm] PCIE GART of 256M enabled (table at 0x000000F400000000).

Feb 14 20:41:10 serva4 kernel: [drm] UVD and UVD ENC initialized successfully.

 

serva4-diagnostics-20240214-2049.zip

Ruffly every 10 second 3 lines.

 

Hardware:

ASRockRack X470D4U2-2T with onboard GPU

AMD Ryzen 7 3700X

1x NVIDIA GeForce RTX 2060

1x Radeon RX 470/480/570/570X/580/580X/590

 

 

System Drivers:

 

 

 

 

What I already tried:

- I updated all outdated plugins

- deinstall/install radeonTop
- deinstall/install gpu statistic

- deinstall/install AMD Vendor Reset

 

The Radoen RX is not used in any way. No VM/Docker is using it.

I saw [1], [2], [3], [4], [5] Post here in the Forum with the same problem, but no solution.

 

Any Ideas?

Edited February 14 by corgan

[trurl]

2 minutes ago, corgan said:

Radoen RX is not used in any way

Can you remove it?

[corgan]

1 minute ago, trurl said:

Can you remove it?

 

Not without a bigger operation and shutting down my homelab and all services. (not wife approved)

So sadly not really.

If everything runs fine some day, I want to use the AMD with path trough in a window WM.
Which worked till the update.
 

Thanks, Stefan

[trurl]

3 minutes ago, corgan said:

some day, I want to use the AMD with path trough

What happens if you check the box next to it in System Devices?

[corgan]

2 minutes ago, trurl said:

What happens if you check the box next to it in System Devices?

 

you mean this one?

 

 

[trurl]

Yes, the checkbox will reserve it for passthru and Unraid won't see it.

[corgan]

2 minutes ago, trurl said:

Yes, the checkbox will reserve it for passthru and Unraid won't see it.

Ok, is checked. But have to reboot.
brb

[corgan]

This looks good. Not one error in 13min uptime.

But if I understand this correct, with this check we have restricted the access from unraid to the amd, but If I want to use it in a VM/Docker, this will work?

 

Anyway, thanks a lot!!!

[JorgeB]

You can also try backlisting the driver:

 

https://docs.unraid.net/unraid-os/release-notes/6.10.0#linux-kernel

 

You will still be able to use the GPU with a VM, not sure about docker.