On Thu, Nov 12, 2026 at 12:49:09PM +0100, Niklas Cassel wrote:
> This patch series adds a new extension - zoned format - to the
> qcow2 driver, allowing full zoned storage emulation on a qcow2
> image file. A user can attach such an image to a guest and have
> it appear as a host-managed zoned block device.
> 
> The zoned format is opt-in through a new qcow2 header extension
> that pins the zone geometry. Behind the extension is a dedicated
> zoned metadata region that stores one 8-byte write pointer (WP)
> per zone. The extension is gated by an incompatible bit, so an
> older qcow2 implementation cannot accidentally open the image.

[...]


Hi Kevin, Hanna, Stefan,


It would be great to get some guidance on what is needed to get this
series accepted. It is now at v17. Is there anything I can do to help
get it merged?

Almost all of the new code is only used by images created with the new
zoned feature, so the risk for existing users should be very low.

The only changes in code that every qcow2 image goes through are:

- Writes now go through qcow2_co_pwv_part(), but outside the zoned
  branches the code is the same as the old qcow2_co_pwritev_part().

- The additions to qcow2_refresh_limits(), qcow2_inactivate() and
  qcow2_do_close() only have an effect when the image has zoned
  metadata.

- qcow2_cache_set_dependency() now returns early when the dependency is
  already set, which avoids a redundant flush, without changing the
  write ordering. I could move this to a separate patch if you prefer.

Apart from one more entry in the feature name table of newly written
images, images that don't use the new zoned feature are unaffected.


Thanks,
Niklas

Reply via email to