Problem 1: package mesa-vulkan-drivers-22.0.1-1.fc36.x86_64 obsoletes mesa-vulkan-devel < 22.0.1-1.fc36 provided by mesa-vulkan-devel-21.3.8-2.fc36.i686
- cannot install the best update candidate for package mesa-vulkan-devel-21.3.8-2.fc36.i686
- cannot install the best update candidate for package mesa-vulkan-devel-21.3.8-2.fc36.x86_64
Problem 2: mesa-vulkan-drivers-21.3.8-2.fc36.i686 has inferior architecture
- cannot install both mesa-vulkan-drivers-22.0.1-1.fc36.x86_64 and mesa-vulkan-drivers-21.3.8-2.fc36.x86_64
- cannot install the best update candidate for package mesa-vulkan-drivers-21.3.8-2.fc36.i686
- cannot install the best update candidate for package mesa-vulkan-drivers-21.3.8-2.fc36.x86_64
and "sudo dnf update --best --allowerasing" wants to remove steam, lutris, pipewire, etc
Bodhi is disabling automatic push to stable due to negative karma. The maintainer may push manually if they determine that the issue is not severe.
Looks like mesa-vulkan-drivers from the new version is no longer recognized as "multilib", and breaks the upgrade to this version if you have both the .i686 and the .x86_64 version of the package installed (as the .i686 version is gone).
I believe this multilib issue with mesa-vulkan-drivers should go away as soon as this package is pushed to stable. The reason for that is that lutris which needs the i686 version of mesa-vulkan-drivers and pulls it in to the multilib x86_64 repo is only available in the "base" (stable) repo but not in updates-testing. Because of this there is nothing that would cause i686 mesa-vulkan-drivers to appear in x86_64 updates-testing compose, but as soon as it gets pushed to stable the mechanism should properly kick in.
Could you guys give it +1 karma please so that we can test this out properly before the final freeze starts and have time to fix it in some other way if it doesn't work the way I assume? Thanks!
This update has been submitted for testing by pwalter.
This update's test gating status has been changed to 'waiting'.
This update's test gating status has been changed to 'failed'.
This update's test gating status has been changed to 'passed'.
This update has been pushed to testing.
Problem 1: package mesa-vulkan-drivers-22.0.1-1.fc36.x86_64 obsoletes mesa-vulkan-devel < 22.0.1-1.fc36 provided by mesa-vulkan-devel-21.3.8-2.fc36.i686 - cannot install the best update candidate for package mesa-vulkan-devel-21.3.8-2.fc36.i686 - cannot install the best update candidate for package mesa-vulkan-devel-21.3.8-2.fc36.x86_64 Problem 2: mesa-vulkan-drivers-21.3.8-2.fc36.i686 has inferior architecture - cannot install both mesa-vulkan-drivers-22.0.1-1.fc36.x86_64 and mesa-vulkan-drivers-21.3.8-2.fc36.x86_64 - cannot install the best update candidate for package mesa-vulkan-drivers-21.3.8-2.fc36.i686 - cannot install the best update candidate for package mesa-vulkan-drivers-21.3.8-2.fc36.x86_64
and "sudo dnf update --best --allowerasing" wants to remove steam, lutris, pipewire, etc
Bodhi is disabling automatic push to stable due to negative karma. The maintainer may push manually if they determine that the issue is not severe.
Looks like mesa-vulkan-drivers from the new version is no longer recognized as "multilib", and breaks the upgrade to this version if you have both the .i686 and the .x86_64 version of the package installed (as the .i686 version is gone).
I do not have installed any i686 mesa package.
It fixed the very annoying bug: https://bugzilla.redhat.com/show_bug.cgi?id=2031410 For me it works flawlessly.
I believe this multilib issue with mesa-vulkan-drivers should go away as soon as this package is pushed to stable. The reason for that is that lutris which needs the i686 version of mesa-vulkan-drivers and pulls it in to the multilib x86_64 repo is only available in the "base" (stable) repo but not in updates-testing. Because of this there is nothing that would cause i686 mesa-vulkan-drivers to appear in x86_64 updates-testing compose, but as soon as it gets pushed to stable the mechanism should properly kick in.
Could you guys give it +1 karma please so that we can test this out properly before the final freeze starts and have time to fix it in some other way if it doesn't work the way I assume? Thanks!
Works nicely on Intel UHD 620 and RX 580.
Giving +1 karma for testing purposes as pwalter suggested.
This update can be pushed to stable now if the maintainer wishes
tested on a 6800XT. no issue to report
This update has been submitted for stable by pwalter.
This update has been pushed to stable.