How do I restart plasmashell after it crashes?

Yes, please report these crashes. Ideally using the automatic crash reporter, but if you’re intentionally not using that, then please do it manually using https://bugs.kde.org.

I mean… not to whine about it, and I do report all the crashes… but plasmashell crashes a lot… Like I’ll literally be sitting there, on Firefox, not even touching my computer, reading a page, and I’ll see plasmashell crash… sometimes just once… other times several times in a row. I’ve even seen it happen at login before I get a chance to do anything at all… and it’s not consistent at all either. Probably about 1 crash per day, at least.

It’s kinda painful. I used to think explorer.exe crashed a lot, especially compared to SystemUIServer/Finder/Dock on the Mac, which almost never crashes… like ever… But plasmashell makes Windows look… stable…

…and this is on a fresh install, with no extensions or plugins or whatnot added by me.

there might be a hardware issue with your machine…. this level of crashing is not at all normal for plasma in my experience.

in fact it’s been MORE stable than windows ever was.

what is your output for kinfo?

It’s definitely not hardware.

Not only is it a brand new machine, but I’ve also done extensive hardware testing to make sure things like the memory, GPU, and CPU, are all operating properly.

Other applications simply don’t crash at all, or don’t crash as often as plasmashell crashes.

Also, when booted into Windows, these problems don’t occur.

At best, it could be a driver issue, but there’s no way it’s a hardware one.

$ kinfo
Operating System: Fedora Linux 43
KDE Plasma Version: 6.6.1
KDE Frameworks Version: 6.23.0
Qt Version: 6.10.2
Kernel Version: 6.18.13-200.fc43.x86_64 (64-bit)
Graphics Platform: Wayland
Processors: 24 × AMD Ryzen AI 9 HX PRO 370 w/ Radeon 890M
Memory: 64 GiB of RAM (54.6 GiB usable)
Graphics Processor: AMD Radeon 890M Graphics

This doesn’t sound good, but unfortunately there’s nothing any of us can do about it here and now unless you paste backtraces of the crashes.

It’s good that you submit the automatic reports, but these get aggregated along with everything else and put in a queue, requiring patience. If you want complaints about crashes to be actionable in a non-automatic-reporting context (e.g. here) you need to paste backtraces of the crashes so people capable of interpreting them have any information to go on. Without that, it’s just frustration and random guessing.