Comments

21 Comments

@thl, no, that's a bug and it should be solved by: https://src.fedoraproject.org/rpms/mesa/pull-request/52

If you see the error in other subpackage, please report it or open a MR.

Hey @agurenko,

Since Fedora 41 will be released soon, no, I didn't plan to build mesa 24.2.x for Fedora 40. Instead, I'd like to stick to the stable version there.

@leigh123linux I'm not sure if you need me to disable auto-push on Beta versions so you can sync with RPM Fusion. Pushes are freezed anyways, but it'd be good to know for future releases.

Hey @leigh123linux, that's good to know, I wasn't aware of the RPM Fusion schedule.

Since this release doesn't fix any urgent bug reported to the Fedora bug tracker, I disabled karma auto-push so you can coordinate both releases next week.

Lately, I'm handling the Fedora mesa updates. It'd be nice to coordinate and avoid issues to RPM Fusion users. Let me know if this works for you.

It'd be nice if we could work together to avoid issues to the RPM Fusion users. At least when there are no urgent bugs that require a mesa release as soon as possible.

Hi @leigh123linux, thanks for the heads up. Do you know if the version mismatch could cause any problems? And also, do you known their reason for waiting a week? I'm not familiar with RPM fusion's release calendar, so I'm not sure if this is the usual schedule or an exception .

@injiniero I'd recommend to comment in the upstream issue: https://gitlab.freedesktop.org/mesa/mesa/-/issues/11352

Or open a new one if you don't consider that they are similar. Hopefully the developers would be able to point into the right direction.

@veilink I think it is caused by the patch backported from: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29876

Adding a comment in that issue clarifying the problem. If you could test mesa-git + that MR it'd be super useful.

I only backported 1 patch from the MR and it'd be good to know if the other patches fix the issue or the MR will introduce a regression.

I think it is related to the backport this version of mesa includes from: https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/29876

As a workaround, you can downgrade mesa to the previous version (mesa-24.1.2-2.fc40 if I remember correctly).

@veilink @dm-zharov I think it is the same issue as: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10927

The commit that is described as the culprit ("ac, radeonsi: remove has_syncobj, has_fence_to_handle") is already addressed in this mesa release, so as mentioned in the last comment, there might be another error: https://gitlab.freedesktop.org/mesa/mesa/-/issues/10927#note_2439112

If you could comment on that bug report or open a new one it'd be helpful.

It looks like an AMD-only issue. I tested it on Intel and the problem is not present.

@nixuser thanks for testing! I submitted this same version of mesa to CentOS Stream 9. I should be available in a few days. If you have problems with it, please open a bug report and I'll look into it.

Hi @de-clan-c

Unfortunately, this update has been already pushed to stable. Do you mind reporting the issues upstream? https://gitlab.freedesktop.org/mesa/mesa/-/issues

Copy here the link to the issue and I'll apply the fix in Fedora as soon as it is available.

Unpushing since @frantisekz reported that this update breaks the X.Org session.

Frantisek, have you reported the issue upstream? https://gitlab.freedesktop.org/xorg/lib/libx11/-/issues

This update has been unpushed.

This update included 2 regressions: https://lists.x.org/archives/xorg-announce/2024-April/003496.html

Lets wait for v1.8.9 to fix them.

This update has been unpushed.

No changes, only adding a missing dependency (vulkan-utility-libraries-devel) of this package.

BZ#2251637 vulkan-validation-layers needs to be updated to not give false positives that waste a huge amount of time for end users