On 01/05/2012 08:49 PM, Jan Kiszka wrote: > To me this still sounds like a cirrus-only xen workaround that > nevertheless spreads widely.
It is. > Again, what speaks against migrating the information Xen needs before > creating the machine or a single device? That would only introduce a > generic concept of an (optional) "early", let's call it > "accelerator-related" vmstate and would allow Xen to deal with all the > specifics behind the curtain. > Adding more concepts, just to work around a bug (and this is really a bug in the qemu/xen interface) makes it harder to refactor things later on. -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.
