Arch Linux

kieran1701

Member

Registered: 2020-07-15

Posts: 6

AMDGPU Driver Not Working Post-6.12.9

In brief, after returning from traveling, I updated my kernel from 6.12.8 to 6.13.3, and I got a green screen and then nothing.  The system boots fine, and I can get into a shell via ssh.  When digging through journalctl, I have the following lines:

kernel: [drm:amdgpu_discovery_set_ip_blocks [amdgpu]] *ERROR* amdgpu_discovery_init failed
kernel: amdgpu 0000:03:00.0: amdgpu: Fatal error during GPU init
kernel: amdgpu 0000:03:00.0: amdgpu: amdgpu: finishing device.
kernel: amdgpu 0000:03:00.0: probe with driver amdgpu failed with error -22

inxi -G
Graphics:
  Device-1: Advanced Micro Devices [AMD/ATI] Navi 23 [Radeon RX 6600/6600
    XT/6600M] driver: amdgpu v: kernel
  Display: x11 server: X.Org v: 21.1.15 with: Xwayland v: 24.1.4 driver: X:
    loaded: amdgpu unloaded: modesetting,vesa dri: radeonsi gpu: amdgpu
    resolution: 2560x1440~60Hz
  API: EGL v: 1.5 drivers: kms_swrast,radeonsi,swrast
    platforms: gbm,x11,surfaceless,device
  API: OpenGL v: 4.6 compat-v: 4.5 vendor: amd mesa v: 24.3.3-arch1.2
    renderer: AMD Radeon RX 6600 (radeonsi navi23 LLVM 19.1.6 DRM 3.59
    6.12.9-arch1-1)
  API: Vulkan v: 1.4.303 drivers: N/A surfaces: xcb,xlib
  Info: Tools: api: eglinfo, glxinfo, vulkaninfo x11: xdriinfo, xdpyinfo,
    xprop, xrandr

I've since been able to trace the problem back to the upgrade from kernel 6.12.9 to 6.12.10.  I've tried manually adding amdgpu to the mkinitcpio.conf modules to no avail.  I've been through forums and other places to see where the issue is, but I am not familiar enough with drivers to really know where to go from here.  Any help would be appreciated.  Thanks!

loqs

Member

Registered: 2014-03-06

Posts: 18,859

Re: AMDGPU Driver Not Working Post-6.12.9

Please post the rest of the kernel messages for a boot with the issue.  Have you considered bisecting between 6.12.9 to 6.12.10?

kieran1701

Member

Registered: 2020-07-15

Posts: 6

Re: AMDGPU Driver Not Working Post-6.12.9

I was planning on doing a bisect if the problem is more than just changing a setting or a package.

Full boot log

jhenson

Member

Registered: 2015-11-01

Posts: 3

Re: AMDGPU Driver Not Working Post-6.12.9

I'm seeing the same thing on the 6.13.x kernels with a Radeon RX 5600/5700 card. The GPU works when booting into the LTS kernel (currently 6.12.16).

gromit

Administrator

From: Germany

Registered: 2024-02-10

Posts: 1,519

Website

Re: AMDGPU Driver Not Working Post-6.12.9

@kieran1701 if you want some prebuild kernel images I can also provide them to you

kieran1701

Member

Registered: 2020-07-15

Posts: 6

Re: AMDGPU Driver Not Working Post-6.12.9

Thanks everyone!  It took a couple of days (work, etc.), but I finished the bisect.  The output is below.

23af4d2c49d8b270be71029af5e0cb5595604321 is the first bad commit
commit 23af4d2c49d8b270be71029af5e0cb5595604321
Author: Alex Hung <alex.hung@amd.com>
Date:   Tue Dec 17 14:03:50 2024 -0700

    drm/amd/display: Remove unnecessary amdgpu_irq_get/put

    commit 5009628d8509dbb90e1b88e01eda00430fa24b4b upstream.

    [WHY & HOW]
    commit 7fb363c57522 ("drm/amd/display: Let drm_crtc_vblank_on/off manage interrupts")
    lets drm_crtc_vblank_* to manage interrupts in amdgpu_dm_crtc_set_vblank,
    and amdgpu_irq_get/put do not need to be called here.  Part of that
    patch got lost somehow, so fix it up.

    Fixes: 7fb363c57522 ("drm/amd/display: Let drm_crtc_vblank_on/off manage interrupts")
    Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
    Reviewed-by: Leo Li <sunpeng.li@amd.com>
    Signed-off-by: Alex Hung <alex.hung@amd.com>
    Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
    (cherry picked from commit 3782305ce5807c18fbf092124b9e8303cf1723ae)
    Cc: stable@vger.kernel.org
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm.c | 31 -------------------------------
 1 file changed, 31 deletions(-)

It looks like something got deleted that shouldn't have been.  I added the deleted lines back in place in a known bad kernel version, and everything started to work.  I opened a bug report on Bugzilla (219808).

Last edited by kieran1701 (2025-02-23 18:26:16)

loqs

Member

Registered: 2014-03-06

Posts: 18,859

Re: AMDGPU Driver Not Working Post-6.12.9

amdgpu bugs are handled on https://gitlab.freedesktop.org/drm/amd/-/issues I would suggest opening an issue there.
Edit:
https://gitlab.freedesktop.org/drm/amd/-/issues/3994

Last edited by loqs (2025-02-24 11:49:28)

jhenson

Member

Registered: 2015-11-01

Posts: 3

Re: AMDGPU Driver Not Working Post-6.12.9

Thanks for tracking it down, @kieran1701!