@pothos the change seems to work as expected, tested with the github artifacts produced by the github action.

BEFORE

AFTER:

Thanks for testing! Please move the change to the commonconfig file to also have it in amd64. We also need a changelog entry like - Enabled the virtio GPU driver ([scripts#830](https://github.com/flatcar/scripts/pull/830)) in a new file changelog/changes/2023-05-22-virtio-gpu.md

Thanks for testing! Please move the change to the commonconfig file to also have it in amd64. We also need a changelog entry like - Enabled the virtio GPU driver ([scripts#830](https://github.com/flatcar/scripts/pull/830)) in a new file changelog/changes/2023-05-22-virtio-gpu.md

Sure, need to verify first if the issue is actually present on the amd64 image.

Thanks for testing! Please move the change to the commonconfig file to also have it in amd64. We also need a changelog entry like - Enabled the virtio GPU driver ([scripts#830](https://github.com/flatcar/scripts/pull/830)) in a new file changelog/changes/2023-05-22-virtio-gpu.md

Hello, I have tested on amd64, and the issue is not reproducing wihout this fix. should I leave the amd64 config unchanged in this case?

I think it's a good idea to still have it on amd64 even if by default another driver is used.

I think it's a good idea to still have it on amd64 even if by default another driver is used.

Hello @pothos, I have updated the PR with the changes requested.

Thank you.

Thank, looks good - I'll start a test job now.

/build-image

Hello @pothos, I have updated the PR and @gabriel-samfira triggered a PR test suite run here https://github.com/flatcar/scripts/actions/runs/5111437226, which partially passed (kolla still broken).

Anything else required for the PR to be merged?

Thank you.

Thank, looks good - I'll start a test job now.

The test job you triggered seems to have been cancelled by another higher priority run: https://github.com/flatcar/scripts/actions/runs/5090846591 -> Canceling since a higher priority waiting request for 'PR command build dispatcher-pr-command-main' exists.

All good, there are some kernel config changes compared to 5.15 and I don't know if this change or the 6.1 kernel update caused them. Anyway something for a new PR.

thank you, @pothos. Because CONFIG_DRM is now set, the make old|defconfig should add some more diffs in the resulted config. Possibly this is the difference you are seeing.