Hello, I believe that I am experiencing the same issue, or at least one
that is very closely related.

My specs:

Motherboard: ASUS CROSSHAIR VI HERO
Memory: 48.0 GiB DDR4 3200 MHz (2×16 GB + 2×8 GB)
Processor: Ryzen 7 1700X
Graphics: Sapphire Nitro Radeon R9 380 Dual-X OC 4G

Firmware Version: 8601
OS Name: Ubuntu 26.04 LTS
OS Type: 64-bit
GNOME Version: 50
Windowing System: Wayland
Kernel Version: Linux 7.0.0-28-generic

Display:

I have three displays connected: two Dell P2419H monitors (1920×1080 @
60 Hz) connected via DisplayPort and VGA, and a TV connected via HDMI.
My primary display is the monitor connected through DisplayPort.

The issue:

When I leave my PC unattended for some time, the displays turn off as
expected. However, when I try to wake them up, they never come back on,
regardless of what key I press. Looking through the `journalctl` logs, I
can see that pressing the power button is still logged, so this does not
appear to be a complete OS crash.

After troubleshooting the issue with an AI agent (Codex), comparing my
`journalctl` logs with this report and also with
https://gitlab.freedesktop.org/drm/amd/-/work_items/5123, I believe that
my issue is closely related, although not identical.

The last log entries before I have to perform a hard reset are
consistently:

2026-08-01T20:30:48+02:00 kernel: amdgpu 0000:0b:00.0: last message was failed 
ret is 0
2026-08-01T20:30:56+02:00 kernel: amdgpu 0000:0b:00.0: last message was failed 
ret is 0
2026-08-01T20:31:02+02:00 kernel: amdgpu 0000:0b:00.0: last message was failed 
ret is 0
2026-08-01T20:31:07+02:00 kernel: amdgpu 0000:0b:00.0: last message was failed 
ret is 0
2026-08-01T20:31:15+02:00 kernel: amdgpu 0000:0b:00.0: last message was failed 
ret is 0
2026-08-01T20:31:18+02:00 systemd-logind: Power key pressed short.

I have not found any `No EDID read` messages in my retained logs.
However, I do repeatedly see:

amdgpu 0000:0b:00.0: [drm] Failed to setup vendor infoframe on connector
HDMI-A-1: -22

I also found these earlier GPU reset logs:

2026-07-28T21:35:13+02:00 kernel: amdgpu 0000:0b:00.0: AMDGPU device coredump 
file has been created
2026-07-28T21:35:13+02:00 kernel: amdgpu 0000:0b:00.0: ring uvd timeout, 
signaled seq=2, emitted seq=2
2026-07-28T21:35:13+02:00 kernel: amdgpu 0000:0b:00.0: GPU reset begin!. 
Source: 1

2026-07-30T20:11:59+02:00 kernel: amdgpu 0000:0b:00.0: AMDGPU device coredump 
file has been created
2026-07-30T20:11:59+02:00 kernel: amdgpu 0000:0b:00.0: ring uvd timeout, 
signaled seq=1705, emitted seq=1705
2026-07-30T20:11:59+02:00 kernel: amdgpu 0000:0b:00.0: GPU reset begin!. 
Source: 1

I am not entirely sure that these earlier GPU resets are related to the
same issue. I have also had a couple of freezes where the screens did
not go dark. They simply froze and stayed that way until I forcefully
restarted the machine.

Initially, I suspected that my GPU might be failing. However, after
reading these reports, I am starting to think that this could be a
software issue instead. I just wanted to share this information in case
it helps narrow down the problem.

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2142389

Title:
  amdgpu (R9 380) fails to resume from suspend (deep sleep) – black
  screen, requires hard reboot

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/2142389/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to