Broken xserver: amdgpu: amdgpu_device_initialize failed.

Something of the I/O doesn’t work.

amdgpu_device_initialize: amdgpu_query_info(ACCEL_WORKING) failed (-13)
amdgpu: amdgpu_device_initialize failed. 

The name of my machine is:

lspci -k | grep -EA3 'VGA'
00:01.0 **VGA** compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Kabini [Radeon HD 8330]
        Subsystem: ASRock Incorporation QC5000-ITX/PH
        Kernel driver in use: amdgpu
        Kernel modules: radeon, amdgpu 

Further information is:

lshw -c video
description: VGA compatible controller
product: Kabini [Radeon HD 8330]
vendor: Advanced Micro Devices, Inc. [AMD/ATI]
physical id: 1
bus info: pci@0000:00:01.0
logical name: /dev/fb0
version: 00
width: 64 bits
clock: 33MHz
capabilities: pm pciexpress msi vga_controller bus_master cap_list rom fb
configuration: depth=32 driver=amdgpu latency=0 resolution=1280,1024
resources: irq:40 memory:c0000000-cfffffff memory:d0000000-d07fffff ioport:f000(size=256) memory:ffb00000-ffb3ffff mem

How do I get this fixed?

Let us see what’s happening. Paste output from the following here using code tags:

inxi -Gayz
cat /var/log/Xorg.0.log | susepaste

You may need to install susepaste and inxi first. Output from the first can be redirected to a file on USB, and the file content copied and pasted. More important is the log if inxi is too much trouble.

Here is the output:

inxi -Gayz
  **Device-1:** AMD Kabini [Radeon HD 8330] **vendor:** ASRock QC5000-ITX/PH
  **driver:** amdgpu **v:** kernel **alternate:** radeon **bus-ID:** 00:01.0
  **chip-ID:** 1002:9832 **class-ID:** 0300
  **Display:****server:** X.Org **compositor:** kwin_x11 **driver:**
  **loaded:** modesetting **display-ID:** :0 **screens:** 1
  **Screen-1:** 0 **s-res:** 1280x1024 **s-dpi:** 96 **s-size:** 338x270mm (13.3x10.6")
  **s-diag:** 433mm (17")
  **Monitor-1:** VGA-1 **res:** 1280x1024 **hz:** 60 **dpi:** 96 **size:** 340x270mm (13.4x10.6")
  **diag:** 434mm (17.1")
  **OpenGL:****renderer:** AMD KABINI (LLVM 13.0.1 DRM 3.44 5.17.1-1-default)
  **v:** 4.6 Mesa 22.0.1 **direct render:** Yes 


How is the “failure” manifesting, other than log messages? (EE) modeset(0): Failed to make import prime FD as pixmap: 22 is the only log content I recognize as indicating any problem. It appears early, and the rest of the log looks normal, except for its length. Everything in inxi output looks kosher.

Try removing amdgpu.si_support=1 from the line beginning linu after striking the E key at the Grub menu, then proceeding with boot. If that’s no help, try in same manner replacing it with radeon.cik_support=0 amdgpu.cik_support=1. If that doesn’t help, try radeon.cik_support=1 amdgpu.cik_support=0 instead. If any of those temporary changes work, they need to be incorporated in /etc/default/grub, then grub2-mkconfig -o /boot/grub2/grub.cfg needs to be run, or Grub updated via YaST.

I have a Mullins (Same Island series as Kabani) they are Sea Island (cik), not Southern Island (si) cards and with Leap 15.3 to get amdgpu working I use;

amdgpu.cik_support=1 amdgpu.si_support=0

Graphics:  Device-1: Advanced Micro Devices [AMD/ATI] Mullins [Radeon R3 Graphics] vendor: Hewlett-Packard driver: amdgpu 
           v: kernel bus ID: 00:01.0 chip ID: 1002:9850 
           Device-2: Chicony HP TrueVision HD type: USB driver: uvcvideo bus ID: 1-1.4:3 chip ID: 04f2:b56c serial: <filter> 
           Display: x11 server: X.Org 1.20.3 compositor: gnome-shell v: 3.34.5 driver: amdgpu,ati 
           unloaded: fbdev,modesetting,vesa display ID: :0 screens: 1 
           Screen-1: 0 s-res: 1920x1080 s-dpi: 96 s-size: 508x285mm (20.0x11.2") s-diag: 582mm (22.9") 
           Monitor-1: eDP res: 1920x1080 hz: 60 dpi: 158 size: 308x173mm (12.1x6.8") diag: 353mm (13.9") 
           OpenGL: renderer: AMD KABINI (DRM 3.39.0 5.3.18-150300.59.60-default LLVM 11.0.1) v: 4.6 Mesa 20.2.4 
           direct render: Yes 

The problem is, the machine is freezing from time to time. Quality of graphics is good.

You’re right, the error comes early. It feels like even grub suffers from that. But no, radeon is no solution.

radeon: Failed to get PCI ID, error number -13
EGL client extensions string:

It’s the same as:

amdgpu_device_initialize: amdgpu_query_info(ACCEL_WORKING) failed (-13)amdgpu: amdgpu_device_initialize failed.
EGL client extensions string:

With radeon driver it’s even worse. The amdgpu is much more stable. The key is the PCI ID, error number -13. Before graphics is loaded something goes wrong. Either from systemctl some other PCI handler.