On Wed, Feb 06, 2013 at 12:36:37PM +0100, Ján Tomko wrote:
On 02/05/13 18:14, Daniel P. Berrange wrote:
>
> I can understand why you're doing this, but gives me a really bad
> feeling. After 5 years we've finally stopped parsing -help from
> the QEMU binary...only to start doing the same for qemu-img.
>
>> ...
>
> IMHO we should just invoke qemu-img with the feature list we have, and
> rely on it to report any unsupported features itself, rather than trying
> to pre-detect them ourselves.
>
> Daniel
I only did it because of the 'compat' feature, which might be needed to
create qcow2 images in the future, if the default of qemu-img gets
changed to qcow3.
Well IMHO that would be a mistake on QMEU's part. IMHO they ought to just
allow the user to specify an explicit '-f qcow3' to get all the latests
qcow3 format features enabled and leave '-f qcow2' on its current behaviour
forever.
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 :|