SUMMARY Attempting to use the network manager results in screen and panel corruption. This is a rather elderly openSUSE Tumblweed machine. TW is normally kept up to date although the machine is only used as a "spare", I can't be precise as to when this issue first appeared. This problem can be reproduced with a new user and the screen shots attached are those from a new, default, user. STEPS TO REPRODUCE 1. Launch Network Manager 2. Select a network connection 3. OBSERVED RESULT 1. At initial launch all is OK: Screenshot 1 2. Network Manager and the Plasma Panel now become corrupt: Screenshot 2 EXPECTED RESULT Normal operation... SOFTWARE/OS VERSIONS Operating System: openSUSE Tumbleweed 20231129 KDE Plasma Version: 5.27.9 KDE Frameworks Version: 5.112.0 Qt Version: 5.15.11 Kernel Version: 6.6.2-1-default (64-bit) Graphics Platform: X11 Processors: 2 × AMD Athlon(tm) 64 X2 Dual Core Processor 5600+ Memory: 3.8 GiB of RAM Graphics Processor: NV84 Manufacturer: Gigabyte Technology Co., Ltd. Product Name: GA-MA770-DS3 ADDITIONAL INFORMATION
Created attachment 163755 [details] Screen Shot 1 - Initial state
Created attachment 163756 [details] Screen Shot 2 - Corrupted state
I guess I must be the only one seeing this issue :( Network Manager is the only Plasma Applet that triggers this screen/video corruption, I don't see the issue with, for example, Digital Clock / Calendar, Disks & Devices, Printers, Notifications, etc... I appreciate I'm running Plasma on very elderly HW, but, I would like to resolve this issue if possible.
Bulk transfer as requested in T17796
Thanks for the bug report, and I'm sorry nobody looked at it. This "black squares" issue is characteristic of graphical corruption caused by faulty graphics drivers. Is the "NV84" an NVIDIA G84 GPU from 2007? And if you're still using the affected hardware with the latest supported NVIDIA drivers and Plasma 6.3.5 or later, can you mention whether the issue is still happening? If it is, does the issue go away if you switch to the open-source Nouveau drivers? Thanks again!
In the intervening year and a half since the initial report this hardware has now been retired... :)
OK thanks! Then probably there's nothing we can do to debug this, unfortunately.