On Fri, Mar 23, 2018 at 11:37:17 +0000, Daniel Berrange wrote:
> On Fri, Mar 23, 2018 at 12:32:29PM +0100, Peter Krempa wrote:
> > There's no tests for this format and nothing seems to specifically care
> > about this format. QEMU does not even recognize it. Remove it completely.
>
> This format is from the User Mode Linux driver, not QEMU.
Hmmm. I wanted to delete the UML driver too, but that was slightly more
work and it also looked like it might be possible to build it even now,
but I did not manage to do it.
>
> > Signed-off-by: Peter Krempa <pkrempa@xxxxxxxxxx>
> > ---
> > docs/schemas/storagecommon.rng | 1 -
> > src/util/virstoragefile.c | 31 +------------------------------
> > src/util/virstoragefile.h | 3 +--
> > 3 files changed, 2 insertions(+), 33 deletions(-)
[...]
> > diff --git a/src/util/virstoragefile.h b/src/util/virstoragefile.h
> > index 596746ccb7..c84e013f46 100644
> > --- a/src/util/virstoragefile.h
> > +++ b/src/util/virstoragefile.h
> > @@ -82,8 +82,7 @@ typedef enum {
> > * libvirt support for following a backing chain */
> > VIR_STORAGE_FILE_BACKING,
> >
> > - VIR_STORAGE_FILE_COW = VIR_STORAGE_FILE_BACKING,
> > - VIR_STORAGE_FILE_QCOW,
> > + VIR_STORAGE_FILE_QCOW = VIR_STORAGE_FILE_BACKING,
> > VIR_STORAGE_FILE_QCOW2,
> > VIR_STORAGE_FILE_QED,
> > VIR_STORAGE_FILE_VMDK,
>
> Removing this from the XML schema, is akin to deleting an enum
> entry from the public API. This is not something we can do, even if
> we think it is unlikely to be used.
Yes I know. But given the chance to delete unused code I had to try it.
Especially since it was easy and I doubt that anyone would ever care.
> It is of course fine if we generate an error in drivers when trying to use
> it at runtime though.
Yes. This is the patch I was doing after sending this.
Attachment:
signature.asc
Description: PGP signature
-- libvir-list mailing list libvir-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/libvir-list
