This morning, after updating at-spi2-core, the system became very slow!
I went back to the previous version and everything is back to normal.
Here's the version I'm currently using and it works fine:
jp-artix:[jp]:~$ pacman -Q at-spi2-core
at-spi2-core 2.48.0-2
$ neofetch
' jp@jp-artix
'o' -----------
'ooo' OS: Artix Linux x86_64
'ooxoo' Kernel: 6.1.27-1-lts
'ooxxxoo' Uptime: 1 min
'oookkxxoo' Packages: 1481 (pacman)
'oiioxkkxxoo' Shell: bash 5.1.16
':;:iiiioxxxoo' Resolution: 1920x1080
`'.;::ioxxoo' DE: Xfce
'-. `':;jiooo' WM: Xfwm4
'oooio-.. `'i:io' WM Theme: Daloa
'ooooxxxxoio:,. `'-;' Theme: Adwaita [GTK2/3]
'ooooxxxxxkkxoooIi:-. `' Icons: oxygen [GTK2], matefaenzadark
'ooooxxxxxkkkkxoiiiiiji' Terminal: xfce4-terminal
It just booted to a blank black screen with me, as the desktop failed to start, using Mate, downgrading at-spi2-core fixed it:
ctrl-alt-f2/3 etc. for alternate tty
$ cd /var/cache/pacman/pkg/
$ sudo pacman -U at-spi2-core-2.48.0-2-x86_64.pkg.tar.zst
at-spi2-core is related to dbus operation apparently.
Yes indeed, it seems to affect startup times. In my case, launching xfce4-terminal took a long time for whatever reason (I don't actually use xfce; just i3). There's reports on Arch (https://bugs.archlinux.org/task/78479) and upstream (https://gitlab.gnome.org/GNOME/at-spi2-core/-/issues/115) about it.
I've pushed at-spi2-core 2.48.1-2 which has the problematic commit reverted, so that should resolve this. After that, we'll just wait for the proper fix from upstream.
Edit: The package is now at 2.48.2-1, a new release, which should fix this.
After updating to 2.48.2-1 and rebooting, it all worked fine, so I can confirm it's fixed things for me here, thank you.
guess we just got "Arched"! 8)