On Thu, Jul 13, 2017 at 10:52:41AM +0200, Kevin Wolf wrote: > Am 13.07.2017 um 10:45 hat Peter Lieven geschrieben: > > Am 10.07.2017 um 15:30 schrieb Kevin Wolf: > > >Am 10.07.2017 um 15:24 hat Peter Lieven geschrieben: > > >>Am 10.07.2017 um 15:10 schrieb Kevin Wolf: > > >>>Am 29.06.2017 um 12:57 hat Peter Lieven geschrieben: > > >>>>Signed-off-by: Peter Lieven <p...@kamp.de> > > >>>>--- > > >>>> qapi/block-core.json | 44 +++++++++++++++++++++++++++++++++++++++++++- > > >>>> 1 file changed, 43 insertions(+), 1 deletion(-) > > >>>How does this make sense as a runtime option? What would happen if the > > >>>image contains one compression format and I specify another one on the > > >>>command line or in blockdev-add? > > >>> > > >>>Shouldn't it just be a create-time option and when you run qemu, it uses > > >>>whatever format that image has? > > >>I was asked to add it here. It is indeed only a create option and has > > >>no other effect. > > >But if it's only a create option, why is it added to blockdev-add? > > > > Is there already a separate schemata for the create options? > > If you mean a separate schema file, no. Putting it in the same file > isn't a problem either, but hooking it up in structs used by > blockdev-add (or other QMP commands) probably is. > > So if you want to follow Dan's suggestion, I think you just need to put > the new type into the schema, without referencing it anywhere except in > qcow2's .bdrv_create C implementation.
Yep, that's exactly what I intended by my suggestion Regards, Daniel -- |: https://berrange.com -o- https://www.flickr.com/photos/dberrange :| |: https://libvirt.org -o- https://fstop138.berrange.com :| |: https://entangle-photo.org -o- https://www.instagram.com/dberrange :|
