Bug details: ** https://bugzilla.redhat.com/show_bug.cgi?id=2272624 ** Information from BlockerBugs App:
Commented but haven't voted yet: ajax
The votes have been last counted at 2024-04-05 19:27 UTC and the last processed comment was #comment-904412
To learn how to vote, see: https://pagure.io/fedora-qa/blocker-review A quick example: BetaBlocker +1 (where the tracker name is one of BetaBlocker/FinalBlocker/BetaFE/FinalFE/0Day/PreviousRelease and the vote is one of +1/0/-1)
BetaBlocker +1
BetaBlocker
FinalBlocker
BetaFE
FinalFE
0Day
PreviousRelease
+1
0
-1
This bug likely affects more than one GPU, but you'd still get display (OpenCL might fail, and I've seen that on similar machines GPU decoding of video might suddenly stop working, with processes in read-blocked zombie states on DRM interfaces). Following the upstream discussion, there's an intel userland / kernel DRM mismatch, and that seems to me to be a bit too risky to carry into a release. Intel iGPUs drivers are probably the single most prolific DRM driver for fedora desktop!
FinalFE +1
I'd kinda like opinions on this from more domain experts, hard to judge how worried we should be as a non-expert. If it's just CL stuff it seems reasonable to fix that after release, if it can affect video output that's more concerning. Maybe @ajax or @airlied could offer an opinion?
Or @jexposit ...
Oh, @marcusmueller , I took the liberty of editing your comment to FinalFE rather than BetaFE, as I'm pretty sure that's what you meant!
This shouldn't affect the video output at all, that would be for kernel/mesa/intel-media-driver-free/intel-gmmlib depending on the context. The packages in question for this FE proposal (intel-igc and intel-compute-runtime) aren't installed by default.
The sole purpose why to get this through the freeze would be one step less for the CL part of the Intel testday as users won't need to have u-t enabled for that.
Yeah, this isn't even installed by default and it really shouldn't affect whether display works. I can't speak to what's a reasonable "freeze exception" these days, but I don't have any objection to it.
Anyhow, I stand for:
I do believe we turn off updates-testing during the final freeze, and without this bugfix release, all CL workloads are going to fail on F40. These packages aren't included in the default set, and F39 is getting the same builds as are the ones in question here.
ehh, I guess I can be:
to make the test day go smoother.
AGREED AcceptedFinalFE
The following votes have been closed:
Metadata Update from @blockerbot: - Issue status updated to: Closed (was: Open)
Release F40 is no longer tracked by BlockerBugs, closing this ticket.
Log in to comment on this ticket.