On Fri, Dec 05, 2025 at 12:48:25PM +0000, Alex Bennée wrote: > Thomas Huth <[email protected]> writes: > > > On 04/12/2025 20.48, Alex Bennée wrote: > >> Technically we don't need a crypto library to do the base build but > >> I couldn't see an easy way to skip iotests which would otherwise fail. > >> Besides libnettle is a fairly small library even if its not the > >> fastest crypto implementation. > > > > I think we added checks to some of the iotests in the past already to > > skip if the crypto stuff is not available? So I think it might be > > better to fix the related iotests instead.
We have _require_working_luks called by "_supported_fmt luks" in the shell tests, and iotests.verify_working_luks called in the python tests. Possibly we didn't call them in all the required places ? Or there is a non-luks related test that uses crypto we missed ? > > > > Which test is failing for you? > > I did have 049 qcow failing and two others but I can't remember exactly > which combo it was. The error messages where all about missing crypto > functions stopping the run. > > Where are the crypto library tests done? I couldn't find them. > > But I'm happy to drop this patch until we find out: > > https://gitlab.com/stsquad/qemu/-/pipelines/2197896632 Is that the right pipeline ? I'm not seeing crypto failures in any of the 8 failed jobs in that pipeline, unless I missed it in the noise of one of them. With regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
