On 19/06/20 14:02, Dr. David Alan Gilbert wrote: >> Paolo, does this break migration of encrypted memory assumptions? >> >> Cc: Dr. David Alan Gilbert <dgilbert@xxxxxxxxxx> >> Supersedes: <20200617130800.26355-1-philmd@xxxxxxxxxx> > I'm not seeing the relevance to migration. I told Philippe that there could be a possibility in the future of having more than one KVMState to support SEV live migration, so removing all KVMState* arguments is not a good idea. But for kvm_check_extension calls it should be safe, as those should be the same for all VMs. Paolo
