Created attachment 188546 [details] Journald log Sometimes, when I boot my laptop (cold boot), all I can see is a black screen after plymouth disappears. It is responsive, I can go to tty2, tty3, but tty1 is just empty. I can even login to tty2 and try to start plasma from there, but there will be small weirdnesses like no app.slice in systemd. The sddm-helper-start-wayland prints these messages: (split into multiple lines by me) No backend specified, automatically choosing drm Accepting client connections on sockets: QList(\"wayland-0\") kwin_core: Failed to open drm node: \"/dev/dri/card1\" kwin_wayland_drm: No suitable DRM devices have been found QThreadStorage: entry 6 destroyed before end of thread 0x5591069f70a0 QThreadStorage: entry 1 destroyed before end of thread 0x5591069f70a0 QThreadStorage: entry 0 destroyed before end of thread 0x5591069f70a0 Because of that output, I presume it is something racy. Whatever it is, that open() somehow syscall fails. STEPS TO REPRODUCE 1. Cold boot I’m sorry, it is very inconsistent and hard to reproduce even for me. OBSERVED RESULT Black screen. EXPECTED RESULT Login screen (sddm in my case) SOFTWARE/OS VERSIONS Operating System: Fedora Linux 43 KDE Plasma Version: 6.5.4 KDE Frameworks Version: 6.22.0 Qt Version: 6.10.1 Kernel Version: 6.17.12-300.fc43.x86_64 (64-bit) Graphics Platform: Wayland Processors: 12 × AMD Ryzen 5 7535HS with Radeon Graphics Memory: 16 GiB of RAM (12.7 GiB usable) Graphics Processor: AMD Radeon 660M ADDITIONAL INFORMATION I’ve attached the relevant log (basically the complete life of the sddm user — it’s pretty short).
This sounds very similar to me as 511118 so marking as duplicate. (I might be wrong however) *** This bug has been marked as a duplicate of bug 511118 ***
Might be, I am not sure of the internals of KWin. The workaround for that other bug is much better however than for this one, because in my case, KWin isn't running anymore and I can only reboot or start it manually from tty.
Hm, I'll reopen this then and show this to KWin folk, they will know better. Thanks for clarifying!
I'm pretty sure this is https://github.com/sddm/sddm/issues/1917 We already have some retry logic in KWin, maybe we can extend it more.
A possibly relevant merge request was started @ https://invent.kde.org/plasma/kwin/-/merge_requests/8673
Btw, it's not easy to reproduce that way, but it's kinda consistent, when you take the "cold boot" literally. I encountered this issue most often, when I was outside with my laptop in my backpack while it was about -4°C (or colder). When I then turned it on too fast after going inside again, this happens. I assume now that the graphics aren't initialized fast enough and KWin fails. It is interesting however, that plymouth and everything up to it works.
Okay, so apparently SDDM and PLM both are missing a workaround that should help with this. At some point we'll be able to implement a full fix (instead of just a workaround) in KWin, until then, the workaround in the display manager will have to suffice.