SUMMARY Seems like Application Launcher widget is somewhat responsible for elevated CPU (and battery) usage on plasmashell. STEPS TO REPRODUCE 1. Run a vanilla clean Plasma session. 2. Wait a but until everything finishes launching and becomes idle. 3. Attach strace to a running plasmashell process ("sudo strace -f -p `pidof plasmashell`") OBSERVED RESULT Lots of repeated polls with timeout values less than 20ms, easily eating significant amounts of CPU (around 5% here). EXPECTED RESULT Some occasional polls, plasmashell achieving <1% CPU usage when idle. -------------------------------------- Afterwards (additional info gathered when investigating): STEPS TO REPRODUCE 1. Remove Application Launcher widget from the default panel. 2. Add it back. OBSERVED RESULT plasmashell becomes way more CPU friendly. -------------------------------------- STEPS TO REPRODUCE 1. Restart plasmashell. OBSERVED RESULT plasmashell becomes CPU hungry again. EXPECTED RESULT plasmashell should still be CPU friendly. -------------------------------------- STEPS TO REPRODUCE 1. Remove Application Launcher. 2. (optionally) Add Application Menu. 3. Restart plasmashell. OBSERVED RESULT Plasma is still CPU friendly SOFTWARE/OS VERSIONS Tested with latest (2019-05-03) Neon Developer Edition ISO. ADDITIONAL INFORMATION All the steps above suggest that something is wrong with Application Launcher widget, but only when it's loaded together with Plasma and/or panel (adding it with a default panel causes the same result as restarting whole Plasma).
Not here. A bunch from the rendering every few seconds. Nothing special. Can you paste a small snippet of your "sudo strace -f -p `pidof plasmashell`" output
Created attachment 119845 [details] strace A freshly booted Neon ISO in a virtual machine. The same thing I observed locally on my Arch with stable Plasma.
Since 15-16 ms timeouts suggest that it's likely some rendering loop, let me add that I use Intel Mesa drivers on Arch and llvmpipe in the VM, if that makes any difference.
Dear Bug Submitter, This bug has been in NEEDSINFO status with no change for at least 15 days. Please provide the requested information as soon as possible and set the bug status as REPORTED. Due to regular bug tracker maintenance, if the bug is still in NEEDSINFO status with no change in 30 days the bug will be closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging If you have already provided the requested information, please mark the bug as REPORTED so that the KDE team knows that the bug is ready to be confirmed. Thank you for helping us make KDE software even better for everyone!
Updating the status (info already provided above).
I did reproduce it on a VM. There's something animating in the hidden window. I don't know what. Possibly (yet more) fallout from b257029afc3bfcd6366d2c2b910b8d91403ab8aa ? I can't reproduce on my dev machines where I can just attach the QML debugger.
Can you see if this is fixed with the new Kickoff UI in Plasma 5.21 (currently in git master)? A lot of this code was rewritten which means there's a chance it got fixed in the process. Thanks!
Just booted neon-unstable-20210122-1501.iso in a VM (using gnome-boxes) and it's still happening straight out-of-box until I edit the panel and remove Application Launcher. It's easily reproducible and super simple to check by anyone with "sudo strace -f -p `pidof plasmashell`".
Does this happen with other launchers as well (kicker, kickerdash)?
I can reproduce this issue right after reboot. Sometimes just opening Application Lanucher fixes polling spam, sometimes it's not. Other launchers (Application Dashboard and Application Menu) does not have this issue. I'm running the release version of Plasma, this may be already fixed in https://bugs.kde.org/show_bug.cgi?id=433103#c15 Operating System: Fedora 34 KDE Plasma Version: 5.22.4 KDE Frameworks Version: 5.83.0 Qt Version: 5.15.2 Kernel Version: 5.13.10-200.fc34.x86_64 (64-bit) Graphics Platform: X11 Processors: 4 × Intel® Core™ i7-2640M CPU @ 2.80GHz Memory: 7,6 GiB of RAM Graphics Processor: Mesa DRI Intel® HD Graphics 3000
Is it fixed now for you in Plasma 5.24, 5.25, or git master?
This bug has been in NEEDSINFO status with no change for at least 30 days. The bug is now closed as RESOLVED > WORKSFORME due to lack of needed information. For more information about our bug triaging procedures please read the wiki located here: https://community.kde.org/Guidelines_and_HOWTOs/Bug_triaging Thank you for helping us make KDE software even better for everyone!