rpm-ostree currently does not support DKMS, and it’s unlikely that’ll be implemented anytime soon, if ever. It does support akmods, though, which is the preferred way to build Kernel modules on Fedora. You could ask if the packager can build that way to support the Fedora Atomic editions.

If you need these Kernel modules now, I think your only option would be to build manually from source, but that has the downside of requiring a manual step every time the Kernel is updated.

Edit: there are a few issue reports already: https://github.com/pop-os/system76-dkms/issues/58 https://github.com/pop-os/system76-acpi-dkms/issues/16

The last commit on that repository was 7 years ago, seems unmaintained. I wouldn’t recommend using this.

If someone decides to use it anyway, don’t use their udev rules. Just install steam-devices or game-devices-udev instead. If you don’t have those packages available in your distro, all rules can be found in the git repo https://codeberg.org/fabiscafe/game-devices-udev

The controller itself is insecure, it doesn’t exactly conform to Bluetooth standard. There’s no indication Sony ever planned cross-compatibility, the DualShock 3 was made to be used only on the PS3 console, where the lack of authorization supposedly wouldn’t be a problem.

Of course, you can still use it on a system where you can accept the risk, as well as on the PS3, or wired. The controllers are not e-waste yet.

Hi, I can answer about the PS3 controller issue. I thought about making a public announcement about this, but I forgot. I’ll work on that now and then link to it here, but to sum up the situation: Support for insecure legacy devices is now disabled due to CVE-2023-45866, and that includes the PS3 controller. You can re-enable support, but that will make your PC vulnerable when Bluetooth is in discoverable mode — that’s when you’re pairing a device; in GNOME that’s when you just have the Bluetooth settings open; easy to have on by accident.

I’ll explain how to re-enable support in the PSA post. It’s a one-liner, but I won’t put it here because I think people should be well-informed of the risks before considering it.

Edit: PSA posted at https://lemmy.world/post/11498269

You should be able to type ç the way I described for all apps, so you could just remove your custom layout. I highly recommend the English (intl., with AltGr dead keys) layout, it’s perfect for coding and writing in English. It’s a bit more work to write in Portuguese, though, so it took me a while to get used to it, but it’s worth it if coding is what you’re doing most of the time. In this layout, you must hold AltGr to get the dead keys, otherwise it’s a normal English layout.

You can also use two layouts — one for English/coding, one for Portuguese — and the keyboard shortcut Super+Space to switch between them. I always have two layouts setup like this, but I never switch anymore because I just learned to love the English (intl., with AltGr dead keys) layout — and I don’t write much Portuguese nowadays.

I’m using GNOME Wayland on Fedora 39 and I don’t have the problem you describe. I just go to settings and select my keyboard layouts:

• English (US, intl., with dead keys)
• English (intl., with AltGr dead keys)

And everything just works. I specially like the second one because it doesn’t interfere with keybindings in games, which can be a problem in GNOME Wayland.

Oh, I think I get the issue you’re having, you can’t find the Çç character on the Linux layout 😅 I always have to explain this to people migrating from Windows, it’s AltGr+, (right Alt key plus Comma). I like this shortcut better than the Windows layout, but I understand some people might not like it. Unfortunately, I can’t answer your question, as I too don’t know how to customize the keyboard layout. I just got used to the Linux layout.

Although that link exists, that’s not what is being used by default. [[ is a shell builtin in ash/busybox, so that takes precedence.

On Alpine:

❯ which [[
/usr/bin/[[

❯ command -V [[
[[ is a shell builtin

gog galaxy through wine is not an option…

That’s the primary way I install and play GOG games. It’s easy to set up using Bottles. Galaxy used to run horribly on Wine, but it has improved recently. I help maintain the Galaxy installer in Bottles, and earlier this year we increased its grade from silver to gold, meaning all functions work with minimal glitches now.

It isn’t perfect yet, it lags for about a minute right after launch, and I’d recommend going into settings and disabling the “Cloud saves” and “Overlay” features as these can cause crashes sometimes. Other than that, everything works well and performance is good.

If you have read it, you might have noticed that the theme of the article is a company called Chainguard. Enterprises can pay them and get a secure software supply chain all the way down to the container image. More than that, their container distro is actually free and open-source, anyone can use it for free, it’s a one line change in your build script to go from Alpine to Wolfi. Enterprises can also buy a secure OS for bare-metal from Red Hat, SUSE, etc…

This article lacks focus and mixes unrelated security concepts in questionable ways. It ends like just an ad for Wolfi. Don’t get me wrong, Wolfi is neat, it’s probably deserving of being talked up. But it doesn’t solve the supply-chain issues pointed out by the article (it doesn’t even try). Supply-chain attacks are currently not a major issue in Linux distributions, and enterprises are already tackling the issue of provenance elsewhere, and the article itself notes that. Dependency management for enterprise software is NOT the responsibility of Linux distros. So what is the point of the article? To me, this article is security mumble jumbo.

Thanks for the report. This issue was supposed to have been fixed in the Flatpak package, but you just brought to my attention that part of the fix was accidentally reverted. I’m sending a new PR right now to try to fix the issue again.

I’ve created a tool for similar of use-cases: https://codeberg.org/contr/contr
You could run your workload inside, say, an alpine container:

cd path/to/evil/dir
contr alpine
❯ # inside container, run dangerous program
❯ ./dangerous_program

If the program needs extra dependencies, you’ll have to write a Containerfile and build an image with the dependencies installed – there’s an example in the repository. Just installing the dependencies at runtime inside the container is also an option, but all changes inside the container are lost on exit.

This is the right way to solve the issue, because this is actually a security issue, and I think most people don’t realize that.

I’m aiming to make this easier, so everybody can benefit from this approach. I haven’t publicly announced it yet because I’m still missing better documentation. But it’s pretty advanced already, and I could use feedback on how it performs on your use-cases: https://codeberg.org/contr/contr

cc @knfrmity@lemmygrad.ml