On Thu, Nov 30, 2023 at 13:24:18 -0600, Eric Blake wrote: > On Thu, Nov 30, 2023 at 05:06:03PM +0100, Peter Krempa wrote: > > Introduce a new flag 'backing_file_format_no_protocol' for the > > block-commit QMP command which instructs the internals to use 'raw' > > instead of the protocol driver in case when a image is used without a > > dummy 'raw' wrapper. > > > > The flag is designed such that it can be always asserted by management > > tools even when there isn't any update to backing files. > > > > The flag will be used by libvirt so that the backing images still > > reference the proper format even when libvirt will stop using the dummy > > raw driver (raw driver with no other config). Libvirt needs this so that > > the images stay compatible with older libvirt versions which didn't > > expect that a protocol driver name can appear in the backing file format > > field. > > > > Signed-off-by: Peter Krempa <pkre...@redhat.com> > > Reviewed-by: Vladimir Sementsov-Ogievskiy <vsement...@yandex-team.ru> > > --- > > > +++ b/qapi/block-core.json > > @@ -1810,6 +1810,14 @@ > > # Care should be taken when specifying the string, to specify a > > # valid filename or protocol. (Since 2.1) > > # > > +# @backing-file-format-no-protocol: If true always use a 'format' driver > > name > > +# for the 'backing file format' field if updating the image header of > > the > > +# overlay of 'top'. Otherwise the real name of the driver of the > > backing > > +# image may be used which may be a protocol driver. > > +# > > +# Can be used also when no image header will be updated. > > +# (default: false; since: 9.0)
As I've previously stated, I don't really care about a name as long as I don't have to keep re-sending, > This is a long name. What about: But is the long name really a problem? > @backing-mask-protocol: If true, replace any protocol mentioned in the > 'backing file format' with 'raw', rather than storing the protocol > name as the backing format. Can be used even when no image header > will be updated (default false; since 9.0). Sounds okay to me. In the end, nobody will really see this as libvirt will be using it internally
