[libvirt] [PATCH 4/6] qemu: Implement startupPolicy for USB passed through devices

Daniel P. Berrange berrange at redhat.com
Thu Oct 11 09:13:02 UTC 2012


On Wed, Oct 10, 2012 at 10:39:53PM +0200, Jiri Denemark wrote:
> On Wed, Oct 10, 2012 at 16:39:06 +0100, Daniel P. Berrange wrote:
> > On Tue, Oct 09, 2012 at 02:13:25PM +0200, Jiri Denemark wrote:
> > > 
> > > +    virBufferAddLit(&buf, "usb-host");
> > > +    if (!dev->missing) {
> > > +        virBufferAsprintf(&buf, ",hostbus=%d,hostaddr=%d",
> > > +                          dev->source.subsys.u.usb.bus,
> > > +                          dev->source.subsys.u.usb.device);
> > > +    }
> > > +    virBufferAsprintf(&buf, ",id=%s", dev->info->alias);
> > 
> > I'm curious about this - it will result in
> > 
> >    -device  usb-host,id=XXXXX
> > 
> > shouldn't we actually just leave out the entire arg. I'm not sure
> > what QEMU would do with such a device specification.
> 
> According to Gerd (and my testing) it's the desired command line. QEMU needs
> to know about the usb-host device during incoming migration to avoid aborting
> on unknown section in migration data. It's not necessary when starting a
> domain from scratch but it doesn't seem to cause any harm.

Ok, that makes sense now. ACK

Daniel
-- 
|: http://berrange.com      -o-    http://www.flickr.com/photos/dberrange/ :|
|: http://libvirt.org              -o-             http://virt-manager.org :|
|: http://autobuild.org       -o-         http://search.cpan.org/~danberr/ :|
|: http://entangle-photo.org       -o-       http://live.gnome.org/gtk-vnc :|




More information about the libvir-list mailing list