Hello everybody,

On my fairly recent KDE desktop in wayland mode, I experience the behavior by Kasts to ocassionally not terminate properly and running into a loop, which causes the process attributed to Kasts to consume full load of a single CPU core. The behavior can be triggered by using Kasts briefly and quitting the program, using the controls provided in the program. The only solution in such situations is to terminate the process using kill -9 kasts or ksysguard. Running journalctl -e | grep kasts in such situations yields the following output: qt.qpa.wayland: eglSwapBuffers failed with 0x300d, surface: 0x0
This issue seems vaguely connected to the kirigami library used to implement the visuals for Kasts. But since journalctl attributes it to the Kasts process, I report it here.
If any additional need for tracing arises, I may try to contribute accordingly.

Thanks a lot for your work on the otherwise very useful and well-implemented program

Kind regards

Thomas

SUMMARY
Kasts not leaving properly under some random circumstances and causing high CPU load. Probably caused by the implementation of the framework used for the graphical user interface.

STEPS TO REPRODUCE
1. Login to KDE 6 session
2. Open Kasts
3. Scroll feeds
4. Wait some time
5. Exit Kasts using the visual buttons

OBSERVED RESULT
High CPU load attributed to Kasts without any user interaction. 

EXPECTED RESULT
Kasts simply quits without causing the CPU load

SOFTWARE/OS VERSIONS
Windows: -
macOS: -
Linux/KDE Plasma: OpenSUSE Tumbleweed 20240520, Plasma 
(available in About System)
KDE Plasma Version: 6.0.4-1.1
KDE Frameworks Version: 6.2.0
Qt Version: 6.7.0

ADDITIONAL INFORMATION