The circle goes like this: (equery -d format) dev-qt/qtwayland-5.15.11 (=dev-qt/qtgui-5.15.11*:5[egl,libinput]) (compositor ? =dev-qt/qtgui-5.15.11*:5[vulkan]) dev-qt/qtgui-5.15.11 (wayland ? =dev-qt/qtwayland-5.15.11*) So with wayland use flag qtwayland and qtgui depend on each other. This leads to qt libraries not being able to be updated by portage. Reproducible: Always
Ah, with USE=compositor.
qtgui's dependency is in PDEPEND though (aka it allows qtwayland to be emerged after qtgui) and should not cause issues. What did portage say exactly that prevented updating?
Im currently having problems updating @world because something tries to pull in QT 5.15.10 while I have installed 5.15.11. I was unable to determine the root casue yet because the conflicts shown are not complete and mostly other Qt packages. I thought that this circle was part of the problem. But I had similar problems updating Qt before.
Please share the full output - the issue is likely something else.
Created attachment 872940 [details] last emerges
You shouldn't really use ~arch dev-qt/*, it leads to messes like this if your list is incomplete. Please fix your keywords (portage is already complaining about some).
I was already trying to fix these conflicts. I didnt look closely enough and thought they were about an qt upgrade which portage has problems withs. So I have script for that which removes all installed dev-qt packages and installs them but it got interrupted. I thought I could emerge -r from that state but looks like emerge didnt like that. So I tried to emerge them manually again which was incomplete at this point yes. But the conflicts had been there previoulsy. And as you can see from the log they disappeared after I unmasked and emerged llvm/clang/lldb 16. And some else I didnt get any conflicts when I ran USE=-wayland emerge -uDN world (I didnt merge it though) So this is the reason for this bug report.
(In reply to farmboy0 from comment #7) > So this is the reason for this bug report. Something else blocked emerge on your system and verbose output confused you into thinking Qt was the reason. Happens to many people, but $summary is still not correct.
(In reply to Andreas Sturmlechner from comment #8) > Something else blocked emerge on your system and verbose output confused you > into thinking Qt was the reason. Happens to many people, but $summary is > still not correct. Yes the output made me think it is Qt. What confirmed it for me was trying to emerge without the wayland use flag. (see last comment)
No, it just took out this component off Portage's equation: > games-util/steam-launcher-1.0.0.76-r2 (wayland ? >=x11-base/xwayland-23.2.1[libei(+)]) So Portage did not have to resolve steam-launcher[wayland]'s dependency, it was able to succeed with a depgraph to upgrade. Portage was telling you all along: > !!! The following update has been skipped due to unsatisfied dependencies: > > games-util/steam-launcher:0 > > selected: (games-util/steam-launcher-1.0.0.76-r1:0/0::steam-overlay, installed) > skipped: (games-util/steam-launcher-1.0.0.76-r2:0/0::steam-overlay, ebuild scheduled for merge) (see unsatisfied dependency below) > > !!! All ebuilds that could satisfy ">=x11-base/xwayland-23.2.1[libei(+)]" have been masked. > !!! One of the following masked packages is required to complete your request: > - x11-base/xwayland-9999::gentoo (masked by: missing keyword) > - x11-base/xwayland-23.2.1::gentoo (masked by: ~amd64 keyword)
