Arch Linux BBS - Applications & Desktop Environments

newer builds of vivaldi break HW acceleration under Wayland/HyprlandSame for* an X11 session* a different wayland compositor (weston)* --ozone-platform=x11?

Hey all,I rented a SpyderX from a local rental to calibrate my laptop screen and external screens.I measured and profiled for a 6500K color temperature at 200cd/m² at Gamma 2.2.After measuring all three screens, two identical external screens and the built in one, I tried applying the created ICC profilein Plasma's System Settings under Display Configuration, but it results in heavy artifacts on the screen.Under XOrg, applying through color management works fine for the built in display. The external displaysdon't work under X because I'm using an eGPU (a GTX1060) setup which only works properly under Wayland.Under X it uses the laptop's GPU, a Quadro P1000 which uses the same drivers, so there shouldn't be any differences there,though only the embedded display works then.Running linux 6.19.9-arch1-1 with Nvidia drivers 580.142 (the nvidia-580xx-dkms package).Picture of two screens, the bottom one has the ICC profile applied (both should be same brightness):https://imgur.com/a/czH1yGfAll screens show the same type of artifact when their ICC profile is applied to themI should note, under X11 with the working ICC profile applied, Darktable distorts images in a similar way when"System display profile" or the imported ICC is selected for the display profile. Choosing anything elseworks fine.

seth wrote:i didnt think it would be a big dealIt's not but it's better to get to the bottom of this than to pitch anecdotal resolutions.You can check older journals to see whether you had those secrets-encryption-key related errors.sudo journalctl | grep -C3 secrets-encryption-keyi did checked journal but didn't see any messages about that, the message was very generic about dependency too, unfortunately i kept clearing the journal every time i tried new things to see the logs because the old messages threw me off, so now i don't have the exact error message it gave me, probably a horrible practice that's why i deleted that from the original post i made

Hmm, that's odd..I mean, it's not as though it totally ignores my fontconfig. As mentioned, adding the following does actually make it do what I want (though at the cost of replacing Arial everywhere):<alias> <family>Arial</family> <prefer> <family>Noto Sans</family> <family>Noto Naskh Arabic</family> </prefer> </alias>But I guess it seems that Firefox respects some things in fontconfig and not others.

I made it like this: matugen image $SELECTED_PATH -m dark --source-color-index 1So --source-color-index skip interaction and choose 2 variant

\o/Please always remember to mark resolved threads by editing your initial posts subject - so others will know that there's no task left, but maybe a solution to find.Thanks.

That fixed it, thanks!

Sounds similar to https://bbs.archlinux.org/viewtopic.php?id=312291 (chromium + kwallet, unresolved)

Last issue that this configuration creates, is that because LAlt is a modifier that now jumps to a different group on-press I now always switch from the US layout to any other as opposed to going to the next because in the moment of activating the binding I am in the US layout. I've partially resolved it by putting a 0.2s sleep before calling the command to change layouts and invoking it on key release instead of press, but it is still not quite ideal.From the perspective of how pleasant it is to switch between keyboards, the approach of creating a virtual modifier that gets sacrificed later in the keybind processing would be nicer. However, this approach is so much simpler and cleaner I think it is worth it.

There's nothing to monitor, but there's https://aur.archlinux.org/packages?O=0&K=noglycin

If not solved, I did a test on the latest Fvwm3 GitHub (fvwm3 1.1.5 (1.1.4-8-g843e7fd6b). It works fine with xterm. Check the screenshot.

Gnome ships it in the gnome shell package so Arch ships it in the gnome shell package. There's nothing inherently special about this and how this has always worked, what exactly do you expect Arch to do?

If this helps you want to figure which parameter is critical - "intel_idle.max_cstate=1" is the worst one, so rule that out first.Ie. try whether you can remove "intel_idle.max_cstate=1" w/o breaking things, then to cut down the other ones - I suspect it's "i915.enable_dc=0"?

Well for this case it’s the nvidia-390xx-utils package from AUR, and updating it fixed the issue. Thank you for your time

I have a computer with three disks: an HDD, a SATA SSD and a NVMe SSD; all three using btrfs.The whole system is installed on the NVMe SSD, except for /home and the XDG user directories.The other two disks have the following subvolumes and mount points:SSD@ → media/ssd@home → /homeHDD@ → media/hddhome/<name>/{@Documents,@Downloads,...} → /home/<name>/{Documents,Downloads,...} (apart from Desktop which is kept on the SATA SSD)The idea is to have the whole home data outside of the system disk and spread over the other two, and then be able to use the two mount points in /media for miscellaneous data storage.Dolphin, however, uses the "wrong" mount points in the Devices list for both disk entries, with no obvious options to change them: I want to have them use the two mount points /media; but instead for the SSD, it points to /home; and for the HDD, it points to /home/<name>/Documents.What I've tried so far, and doesn't work, is:Using x-gvfs-hide and x-gvfs-show in fstab (also tried with comment=x-gvfs-hide)Removing all mount points except for the two in /media from fstab and using systemd mount for the rest (https://bbs.archlinux.org/viewtopic.php?id=178734, I know this was on another file manager but it was worth a shot)I've also seen people suggesting udev rules, but I'm not sure if those would work for subvolumes.I know that I could hide the entries from the list and make entries in Places, but I'd really like to keep the separate entries for the disks in the appropriate section (as you can't move custom entries to Devices).