SUMMARY Plasma 6.3.0 is slower to start compared to 6.2.5 in Debian (unstable repository) STEPS TO REPRODUCE 1. Use Debian with the 'unstable' (Sid) repository 2. Go into a virtual terminal with CTRL+ALT+ F3 3. Run: sudo apt update 4. Run: sudo apt full-upgrade 5. Run: sudo systemctl reboot OBSERVED RESULT The new Plasma 6.3.0 boots slower than Plasma 6.2.5 It stays a while at the KDE / Plasma logo with the black background. Then it stays a while too before the panel is shown. I also deleted the .cache folder and reboot a few times, but it didn't help. EXPECTED RESULT Plasma 6.3 should boot at least as fast as 6.2.5. SOFTWARE/OS VERSIONS Linux/KDE Plasma: KDE Plasma Version: 6.3.0 KDE Frameworks Version: 6.10.0 Qt Version: 6.7.2 Kernel Version: 6.12.13-amd64 (64-bit) Graphics Platform: Wayland HARDWARE SPECIFICATIONS Hardware: Laptop Dell Inspiron 5770 (17" 1080p@60Hz screen) CPU: Intel® Core™ i5-8250U CPU @ 1.60GHz GPU 1: Mesa Intel® UHD Graphics 620 (main) GPU 2: AMD Radeon R5 M465 Series RAM: 8 GiB (7.7 GiB usable) ADDITIONAL INFORMATION Debian weekly build downloaded from here: https://cdimage.debian.org/cdimage/weekly-builds/amd64/iso-dvd/ Was installed and then switched from its 'testing' repository to the 'unstable' repository in this file: /etc/apt/sources.list Then Plasma was upgraded from version 6.2.5 to 6.3.0 from a virtual terminal with the 'sudo apt full-upgrade' command.
Another report of this in Bug 500180. Does the issue also happen on X11, or is it Wayland-specific? Can you use `journalctl` to get a log of what was going on during that time? Hopefully that will shed some light on the situation.
Doesn't seem to happen at all on X11, from what I can tell. But now I'm not sure if it's still happening on Wayland too as it's definitely not as noticeable as it was the first time and a very few restarts after it; I installed all the updates that came in Debian unstable in the meantime, which maybe had some effect on it. Anyway, would the last 500 lines of journal suffice? For Wayland: https://pastebin.com/R22zjLuD And for X11: https://pastebin.com/qjQ3pVcM I've put them on PasteBin as here I encountered the maximum characters limit and I'm not sure how much I need to delete from them to work. Sorry for not putting any highlight that PasteBin offers as I didn't know which one would work for these. Please let me know if you need more lines or I should use other journalctl command, thanks!
systemd-analyze shows this: Wayland: Startup finished in 6.240s (firmware) + 8.473s (loader) + 7.787s (kernel) + 4.148s (userspace) = 26.650s graphical.target reached after 3.873s in userspace. X11: Startup finished in 6.132s (firmware) + 8.512s (loader) + 7.815s (kernel) + 3.863s (userspace) = 26.323s graphical.target reached after 3.863s in userspace. Which I think it's normal, as it was always up to 30 seconds in total. So I'm not sure how it fixed itself. Or maybe it's just less bad that I cannot notice it anymore. Maybe also because it was not as bad as what the other user reported.
Thanks. Are you saying it's gotten faster since you opened this bug report, but it's still not quite as fast as you remember it being in Plasma 6.2.5? Or is it back to being as fast as 6.2.5 now?
Thanks to you also! Since i opened the bug report it got better, in speed. It's not either exactly as the same speed that 6.2.5 was or just very little bit slower. After the kernel and systemd processes load with all that output, I can see Plasma's spinning wheel for a few seconds (1-3) then it enters and the panel appears or moves a bit up. So I'm not sure anymore, it's too little to notice now. In the meantime, I installed more updates, from which the Linux kernel is now 6.12.15 and KDE Frameworks is now 6.11.0 Now sytemd-analyze says (for Wayland session): Startup finished in 7.245s (firmware) + 8.407s (loader) + 7.865s (kernel) + 3.585s (userspace) = 27.104s graphical.target reached after 3.447s in userspace. So total time increased a bit, but the userspace decreased a bit, from which I assume Plasma has a big portion of it. Even here it seems that the changes are small. If the ~3.5 seconds in userspace seem long compared to other laptops, mine even though it has a M.2 SSD, it uses it in SATA mode for some reason that I forgot why. Anyway, it's fine if you want to close the bug report as it doesn't bother me anymore as it seems pretty much like before.
Edit: It's not -> It's now...
Yeah, it seems like the login time is volatile depending on the system's status, which means any actual regressions on Plasma's side are getting lost in the noise, unfortunately. Thanks anyway!