B@Kyle Gospo @M2 @Robert (p5) (others too, but i know these people are engaged on F41 akmods)
BI mentioned to @Kyle Gospo , I noticed there are some kmods in COPR/ublue-os/akmods but which have source in his personal github
I want to get those moved over to ublue-os so any of us can help fixup/maintain
I want to get those moved over to ublue-os so any of us can help fixup/maintain
Bhttps://github.com/KyleGospo/openrgb-dkms is the first
Kthis one we can drop
Kbuilt into fsync
Bwell, that's easy... drop it even for F39/F40?
Kyep
MRpmrebuild is throwing some errors on F39/F40.
Looking messy. Might need to just copy the /lib/modules tree a few times instead
Looking messy. Might need to just copy the /lib/modules tree a few times instead
Byes, i saw that too, when i was building locla... was hoping you would notice in CI 
MThe symlink works on a single rpm. Then probably goes haywire trying to reuse the build root
BBthere's a lot of kmods for bazzite
ayn-platform-kmod is failing to build @Kyle Gospo
and i don't think i want this one in ublue-os
ayn-platform-kmod is failing to build @Kyle Gospo
and i don't think i want this one in ublue-os
Bit CAN be, if that's the right answer
Mopenrazer is the second kmod so likely its the buildroot getting busted
Bi haven't seen openrazor break yet... link?
MIt's not openrazor breaking, but the rebuild after dual sign
MThink I fixed it by using a new buildroot each for each rebuild. Builds locally at least for me
Bthank you
Mthe symlink works for a single rebuild.... then it gets trashed
Mand looks like that worked.
MFor ayaneo is it just a typo in the code?
Mimplicit declaration of a function... but there is one named the same thing without an s
MYepp, looks like it was a typo that was somehow snagged. Upstream repo has it without an s in it
Myepp. That's it completely. Not sure why it's not triggering on the previous ones as well. But the PR that fixed it was for exclusively that line
Mhttps://github.com/ublue-os/akmods/actions/runs/11227835145/job/31210727708#step:12:1360
It appears that the F41 version is built with the previous version. While F39 and F40 have the more up to date package
It appears that the F41 version is built with the previous version. While F39 and F40 have the more up to date package
F:dispair:
Not built into
Not built into
-mainBI’ll hit rebuild on COPR
MLooks like copr rebuilt but the run used the old package?
Bhmm
MMaybe try pushing a blank commit to the pr
MWe've had some screwy stuff in the past with PR being out of date to changes
Bi wonder if the PR' cache is maintained somehow
Bi could push a new change or merge changes to generate a fresh build
Bupdated with current main branch changes and now a new build is starting
B@M2 random but can you approve? https://github.com/ublue-os/akmods/pull/247#pullrequestreview-2355777626
MCheck the sha hash in the package for 41 extras. It should have Kyle's release like 40/39
Bhmm
Bi mean, this is what built, and it looks to be that newer version: https://download.copr.fedorainfracloud.org/results/ublue-os/akmods/fedora-41-x86_64/08120690-ayn-platform-kmod/
Bso weird, its sill pulling that older build
Boh i found it
BBthis kind of problem with not finding the "kmod-common" for a given kmod build seems new
i thought it was somehow tied to the negativo17 change, but that doesn't make sense in the context of this build
i thought it was somehow tied to the negativo17 change, but that doesn't make sense in the context of this build
Kthat one is one of ours, entirely in copr
Boh! in this case, the previous build didn't include 41
Kcommon should be a separate package
Kyea that'll do it
