Monday meeting notes for 22/5/2023

[Marco Martin]

Meven
* Plasma related I have been trying to fix the kcm touchpad Hi bug
https://invent.kde.org/plasma/plasma-desktop/-/merge_requests/1526
* On the build side, I made a couple patch for kdesrc-build to help
building baloo-widgets and dolphin-plugins
* there is a patch
https://invent.kde.org/sdk/kdesrc-build/-/merge_requests/232 to help
with the appstream runtime errors, that I'd encourage to review
* Other than that I have been trying to fix an important regression in
dolphin, by due to KActivities architecture I think I will have to
reintroduce the bug that the regression fixed in the first place
* Bug regression was introduced by
https://invent.kde.org/system/dolphin/-/merge_requests/529

Nate
* I have a few Plasma MRs that could use review:
    https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/2931
    https://invent.kde.org/plasma/plasma-desktop/-/merge_requests/1524
    https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/2936
* been working on miscellaneous Plasma porting and UI improvement topics

Jonathan
* I just switched over KDE neon unstable edition to Qt6 builds of
frameworks and Plasma


Vlad
The biggest update from my side is: the panel has been ported to the
layer-shell protocol

We're not done with the panel yet though.

There are two more things that must be addressed:

    improve the panel configuration ui to play nicer with layer-shell
    the panel configuration should be ported away from the plasma-shell protocol

the second is still unclear



Xaver
atm I'm mostly working on HDR stuff and color management:
    I want to make brightness control go through KWin:
https://invent.kde.org/plasma/kwin/-/issues/156
    I'm researching tonemapping and gamut mapping, so that we can show
HDR stuff on SDR screens appropriately
    I'm thinking about how to handle lookup tables, which can improve
performance a lot
        generating the lookup tables can be quite resource intensive,
a 64^3 3D lookup table for example has more than 260000 entries. So
for that I think we should use a shader to generate a texture that can
then be used in other shaders
        Generating large LUTs for KMS will probably need to use a
thread to not cause any stutter, and some careful handling to fall
back to a shader until the LUT is ready
I'm also looking into fractional scaling bugs but my patches for that
are super broken atm, so there's no real news there yet


Fushan
* 3 🍒 for 5.27 that is waiting for someone to pick.
* Reload holiday events without restarting plasmashell:
https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/2927
* Fix blurry SVG wallpaper if it's from a KPackage:
https://invent.kde.org/plasma/plasma-workspace/-/merge_requests/2904
* Revert button padding hack to make Firefox CSD window buttons look
nice again: https://invent.kde.org/plasma/breeze-gtk/-/merge_requests/83

Marco
ok
on my part last thing.
i have a bit on hold the ksvg stuff atm, concentrated almost only on
the big refactor
those are the mrs so far:
https://invent.kde.org/frameworks/plasma-framework/-/issues/15



There has been discussion on how to do the move of p-f to workspace,
not conclusive

-- 
Marco Martin