On Tue, 2020-04-07 at 09:57 -0400, Liang Yan wrote: > On 4/7/20 9:24 AM, Andrea Bolognani wrote: > > On Tue, 2020-04-07 at 08:01 -0400, Liang Yan wrote: > > > I agree we could document this situation at least. It does block our > > > aarch64 tests while it is ok for x86_64. > > > > Can't you just use the default? That works on all architectures. > > Yes, that is what I suggest our QA people to do. To me, I just thought > if we could setup "--cpu host-model" and "--cpu host-passthrough" > manually, does it mean we can remove "--cpu host"? Of course, I may > misunderstood the meaning of "--cpu host". --cpu host is just a poorly named alias for --cpu host-model, so you can definitely avoid using it but I'd still recommend relying on the virt-install default if at all possible. -- Andrea Bolognani / Red Hat / Virtualization
