On Thu, Jul 27, 2017 at 15:08:11 +0200, Andrea Bolognani wrote: > On Thu, 2017-07-27 at 14:36 +0200, Peter Krempa wrote: > > > Would you still be against it if the enumeration was extended > > > to include every NIC model ever supported by QEMU and Xen? As > > > mentioned, other drivers already perform their own validation > > > and only accept a very limited number of models, so there's > > > no risk of regression there. > > > > I don't think that will help. It will prevent missing any working > > configuration, but it will still make any invalid configuration vanish. > > Do we really need to care that much about guests that never > had a chance to even start? I honestly don't see the point. But they were present in the list of VMs and could be edited. If they fail to parse and thus vanish, users have to go into the file with a big "DO NOT EDIT THIS FILE" sign and edit it. Even if you don't see the point we still try to not make existing configs vanish. Validation callback is your friend in this case.
Description: Digital signature