Skip to content

Bazzite 44 on ASUS ROG Xbox Ally X: black screen / relogin loop with default gamescope-session-ogui-steam #5125

Description

@lonsdaleite

Describe the bug

After updating from Bazzite 43 to Bazzite 44 testing, my ASUS ROG Xbox Ally X no longer boots into Gaming Mode.

Observed on testing-44.20260620 and on all previous Bazzite 44 testing builds that I tried.

On boot, I get a black screen / blinking cursor loop.

TTY is reachable with Ctrl+Alt+F2, but only for a few seconds at a time because the graphical session repeatedly takes over the display again.

Running:

sudo systemctl isolate multi-user.target

stops the loop and gives me a stable console.

From that console I performed the following checks:

systemctl --user reset-failed
systemctl --user start gamescope-session-plus@ogui-steam.service
echo $?

Result:

0

gamescope-session-plus@ogui-steam.service does not just silently do nothing. The system journal shows that opengamepad-ui.x86_64 aborts and dumps core.

Relevant log lines include:

Process ... (opengamepad-ui.) terminated abnormally with signal 6/ABRT
Process ... (opengamepad-ui.) dumped core
/usr/share/gamescope-session-plus/gamescope-session-plus: line 340: ... Aborted (core dumped) /usr/bin/switcherooctl launch $CLIENTCMD
ogui-steam failed

I then tested:

systemctl --user start gamescope-session-plus@steam.service

This opens a visible Gamescope / Steam UI.

However, I only verified that this UI opens. That manually started session is not fully functional in my test:

  • HHD does not work there
  • there is no internet connection there
  • switching to Desktop Mode returns me back to the console

I did not test that manually started session further.

What did you expect to happen?

The device should boot normally into Gaming Mode after updating to Bazzite 44 testing.

Session Logs

https://paste.centos.org/view/5ac1d0bf

Hardware

ASUS ROG Xbox Ally X

Extra information or context

Additional observation: the journal also contains repeated SELinux AVC denials involving unlabeled_t, including systemd-journal access to user-1000.journal, NetworkManager timestamps, and bluetooth cache files. I am not sure whether this is related to the OGUI crash, but it is present in the same boot log.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions