Re: libvirt backport to Bookworm

On Sat, Jun 29, 2024 at 08:26:16PM +0200, Guido Günther wrote:
> On Wed, Jun 19, 2024 at 05:48:33PM +0200, Thomas Goirand wrote:
> > Therefore, we've installed the latest qemu from bookworm-backports. However,
> > libvirt is still too old, and doesn't know about this CPU. Therefore, nova
> > thinks it's model is EPYC-Milan, and not EPYC-Genoa.
> 
> Did you check if updating the cpu map in /usr/share/libvirt/cpu_map/ is sufficient?

Yeah, you could try cherry-picking just [1] and seeing whether that
is enough. Since we might be looking into a stable upload (see [2])
soon, it could make sense to go that route instead of involving
bookworm-backports and a full rebase.

But I'm not at all familiar with the stable updates workflow, so I
could be suggesting nonsense O:-)


[1] https://gitlab.com/libvirt/libvirt/-/commit/bfe53e9145cd5996a791c5caff0686572b850f82
[2] https://salsa.debian.org/libvirt-team/libvirt/-/merge_requests/224
-- 
Andrea Bolognani <[email protected]>
Resistance is futile, you will be garbage collected.

-----BEGIN PGP SIGNATURE-----
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=tpjC
-----END PGP SIGNATURE-----