Heya folks! It's Friday, and that means it's time for the weekly Solus roundup! We have a bunch of desktop updates this time around, so let's dive in.

Note: After next week's (Week 50) sync, Solus will no longer run on kernels older than 5.15. This is because, with the release of kernel 6.12, we are bumping the glibc syscall baseline to 5.15, which will disable all code paths for older kernels. This only affects people who are compiling their own kernel, running Solus in a container, or otherwise using an old kernel, including chrooting into Solus using a very old ISO. Our previous baseline was 5.10. This baseline is updated regularly, about once a year.

GNOME 47.2 has landed in Solus this week. This is primarily a bugfix release. At the time of writing, I have been unable to locate any centralized release notes or announcement, so I have nothing to link you this time.

Our Qt stack was updated to 6.8.1. Qt 6.8 contains many new features, several new modules, and many bug fixes. The svgtoqml tool, used to generate QML code from SVG documents, is now out of tech preview. The Qt Graphs, Qt HTTP Server, Qt Protobuf, and Qt GRPC modules are all promoted out of tech preview state. A new Qt Quick Vector Image module has been added for displaying SVG documents as scalable vector graphics in a Qt Quick scene. Read all about it here!

A fix has been backported for Qt for rendering emojis. This means that emoji should now work correctly in Qt applications, including Plasma. This has been a significant pain point for a lot of users. Qt apps should also look better in general on GNOME Wayland.

Major Qt updates often break applications, and we now have more Qt6 applications than ever before. So, please open a Github issue if a Qt6 application either doesn't launch, or has major issues of some kind.

In addition to GNOME and Qt, KDE Plasma has been updated to 6.2.4. This is primarily a bugfix release, including:

  • libkscreen Doctor: clarify the meaning of max. brightness zero.
  • Plasma Workspace: Battery Icon, Add headphone icon.
  • Plasma Audio Volume Control: Fix speaker test layout for Pro-Audio profile.

The full announcement can be found here.

GStreamer has been updated to 1.24.10. This version contains a bunch of bug and security fixes. The changelog for this version can be found here.

The NVIDIA Beta driver has been updated to 565.77, and the NVIDIA Developer driver has been updated to 550.40.81. These releases have many bugfixes. You can read the update information here for the Beta driver.

The Nim programming language has been updated to 2.2.0. The last version in the Solus repository was 1.6.20, so this is likely to be a significant jump.

The Minetest project has been renamed to Luanti. Check out their blog post on the subject to find out their reason. Our package has also been updated to 5.10.0, the changelog for which can be found here.

Our mps-youtube package has been replaced by yewtube, a fork of the original project with a new name. It has also been updated to 2.12.0.

Node.js 18 has been removed from the repository, as no other package now uses it.

Security updates

We have some security updates this week. Make sure to install these updates, as well as the GStreamer updates, for the latest security protections!

General updates

The full list of updated packages can be found here.

For the list of currently known issues, see the dedicated thread for it.

That’s all for this week, folks! We'll be here same time, same place next week for another roundup of the news!

How did the sync go for you?

This poll has ended.

@brent This one is on me. I merged the changes but did not publish the youtube-search-python package. I'll cherry-pick it as soon as the builder is free.

In the meantime, if you want to update, remove mps-youtube first. Then try an update. When the cherry-pick is complete, you would install yewtube

    Two plasma machines updated ok. Will test apps tomorrow off to bed..

    222 packages updated on my Plasma laptop, but afterwards I couldn't reboot, neither with a reboot command in the terminal, or a restart command from the menu. Finally did a hard reset and restarted. After that I could not do a mount -a to remount my NAS shares. Dunno what's up with that.

    EDIT: I rebooted again from the terminal, and this time it worked normally. Then I mounted my NAS shares, and that worked this time, too. So my issue somehow resolved itself.

    Works great. I haven't encountered any issues. (Plasma)

    Hi all,
    Small issue on this update. When launched, protonmail-bridge give me the following message :

    Could not load qml component

    Then protonmail-bridge stops working.
    I tried uninstall then reinstall, but no success.
    I will file a bug if necessary.
    The rest of the update is fine, as always.
    Budgie desktop by the way.

      I updated my two machines with Solus KDE, and Solus xfce, and so far so good. 🙂

      all good, had to make a minor change in my dunst config though

      Aw, fsck. When I restored from hibernation this morning, things didn't look so good. The panel that works as a dock has gone the other way from the condition it's been lately. Instead of squeezing everything into a tiny square, it now displays full-size, but showing only the two left icons.

      Strangely enough, when I clicked on the location where my Terminology icon used to display (far right), Terminology is what started, so I was able to get things started there.

      So, I clicked where Vivaldi should be, but just got the tired old error message about "No internet..." At that point, I did the screencap you see above, getting Spectacle from the menu instead of the dock. Also, the hibernation problem that I thought was solved by changing Vivaldi's setting to save a week of browsing data instead of the 3 months it defaulted to was back.

      C'mon KDE. 222 packages updated, and this is what I get? Gonna try one more reboot, and see if it sorts itself out once more.

      EDIT: I did a full shutdown instead of a reboot, as I wanted to make sure I didn't leave any debris anywhere. Here's the dock now, so I'm hopeful that things are once again working as they should. So far, so good.

      Next, I'll populate the workspaces in the upper panel, and do a brief hibernation, to see if the Internet is still connected when the laptop wakes up. I suspect it will be fine.

        WetGeek 222 packages updated

        Most were rebuilt for Qt 6.8.1, not updated in the sense you are thinking.

        Brace for rant.

        ACPI (Advanced Configuration and Power Interface) is the standard that defines these power states. Every single piece of hardware in your computer needs to support these states properly.

        In 2003, Linus Torvald said:

        Modern PCs are horrible. ACPI is a complete design disaster in every way. But we're kind of stuck with it. If any Intel people are listening to this and you had anything to do with ACPI, shoot yourself now, before you reproduce.

        Source: https://www.linuxjournal.com/article/7279

        For context Intel helped develop this standard and he is not only talking about ACPI for sleep states.

        I stopped using these states over a decade ago because they are not reliable. Some hardware will never support it properly and even if you have hardware that does have decent support for it, every driver/kernel update or hardware addition can break it. I have had the simple addition of a usb dongle break it.

        This is not a Solus problem. It is not even a Linux problem. Microsoft has a hardware certification program and part of their testing makes sure these states work and its still a cluster fuck (Microsoft share the blame for why ACPI sucks). They can only certify that it works at that point in time with that hardware. If you update the system or add new hardware it can break it.

        24th January 1999, Bill Gates

        One thing I find myself wondering about is whether we shouldn't try and make "ACPI" extensions somehow Windows specific.
        It seems unfortunate if we do all this work and get our partners to do the work and the result is that Linux works great having to do the work.

        Maybe there is no way to avoid this problem but it does bother me.

        Maybe we could define the APIs so that they work well with NT and not the others even if they are open.

        Or maybe we could patent something related to this.

        Their efforts ensured it sucked for everyone instead.

        The only thing about sleep states I can guarantee is on a long enough time line it will break. If you haven't encountered this sort of breakage before with these states you are extremely lucky. You are rolling a d100 every time you try use it and the DC is 80.

            Harvey Brace for rant.

            Thanks for that summary of the ACPI issue. I guess it's just something we need to put up with, if we want to use sleep states. The last time I fixed everything, it lasted for quite a while, so that's just a decision I need to make. The alternative for me is just to lock my screen at night and unlock in the morning, which is not so terrible. I just hate leaving the computer running for that many hours when I'm not using it.

            My computers are all DELLs, so all Intel inside. I imagine things could be worse. Now it's time to forget all about this for now, and watch the Abu Dhabi grand prix - the last one of the F1 season. And the last one for Louis Hamilton to be driving a Mercedes.