Summary: | Global Application shortcuts are slower than they could be | ||
---|---|---|---|
Product: | [Frameworks and Libraries] frameworks-kglobalaccel | Reporter: | Aleksey Kladov <aleksey.kladov> |
Component: | general | Assignee: | kdelibs bugs <kdelibs-bugs> |
Status: | RESOLVED DOWNSTREAM | ||
Severity: | minor | CC: | nate, nicolas.fella, pereira.alex |
Priority: | NOR | ||
Version: | 5.105.0 | ||
Target Milestone: | --- | ||
Platform: | Other | ||
OS: | Linux | ||
Latest Commit: | Version Fixed In: |
Description
Aleksey Kladov
2023-04-12 23:12:55 UTC
I haven't measured it, but https://invent.kde.org/plasma/kglobalacceld/-/merge_requests/3 could help here since it cuts out an intermediate helper process >could help here since it cuts out an intermediate helper process Aha, I think this helper process would explain 100% of the slowness I observe. On my system, spawning any plasma-related processes is very slow: ``` $ t kstart -h <SNIP> real 339.00ms cpu 171.86ms (127.87ms user + 43.99ms sys) rss 94.29mb ``` I fixed that dowstream (https://github.com/NixOS/nixpkgs/issues/225871), so I think it's fair to say that there's no "global shortcuts are slow" bug in plasma after all. Still, getting rid of an intermediate process feels like a good idea regardless! ah, that explains a lot. "so I think it's fair to say that there's no "global shortcuts are slow" bug in plasma after all." Well, after your tip on using Meta+1, Meta+2 to launch apps, I started using that, and it "seems faster" than "global application shortcuts" even on Gentoo and Arch. I say "seem faster" because timing kstart here gives around 70~80ms delay, so its hard to measure it really. So at the end this will improve things :) Thanks for reports this bug and fixing nixos side and Thanks KDE devs for improving KDE all the time! :) |