All of lore.kernel.org
 help / color / mirror / Atom feed
* Why device-list-properties?
@ 2026-11-19  9:17 Markus Armbruster
  2026-11-19 10:38 ` Michal Prívozník
  2026-11-19 11:47 ` Daniel P. Berrangé
  0 siblings, 2 replies; 4+ messages in thread
From: Markus Armbruster @ 2026-11-19  9:17 UTC (permalink / raw)
  To: qemu-devel; +Cc: devel, Paolo Bonzini, Daniel P. Berrangé

As far as I can tell, device-list-properties can do nothing you can't do
with qom-list-properties.

Both take a QOM type name.

device-list-properties rejects types that aren't concrete subtypes of
"device".

device-list-properties filters out properties named "type", "realized",
"hotpluggable", "hotplugged", and "parent_bus".  These are precisely the
properties every device inherits from "device".

Why is that useful?


^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Why device-list-properties?
  2026-11-19  9:17 Why device-list-properties? Markus Armbruster
@ 2026-11-19 10:38 ` Michal Prívozník
  2026-11-19 14:41   ` Markus Armbruster
  2026-11-19 11:47 ` Daniel P. Berrangé
  1 sibling, 1 reply; 4+ messages in thread
From: Michal Prívozník @ 2026-11-19 10:38 UTC (permalink / raw)
  To: Markus Armbruster, qemu-devel
  Cc: devel, Paolo Bonzini, Daniel P. Berrangé

On 11/19/26 10:17, Markus Armbruster wrote:
> As far as I can tell, device-list-properties can do nothing you can't do
> with qom-list-properties.
> 
> Both take a QOM type name.
> 
> device-list-properties rejects types that aren't concrete subtypes of
> "device".
> 
> device-list-properties filters out properties named "type", "realized",
> "hotpluggable", "hotplugged", and "parent_bus".  These are precisely the
> properties every device inherits from "device".
> 
> Why is that useful?

I think it's only by inertia (from libvirt's POV). I mean,
device-list-properties was introduced to libvirt in 2012
(v1.0.0-rc1~216) and qom-list-properties in 2018 (v4.4.0-rc1~209). Are
you looking at obsoleting the former? If so, libvirt should adapt.

Michal



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Why device-list-properties?
  2026-11-19  9:17 Why device-list-properties? Markus Armbruster
  2026-11-19 10:38 ` Michal Prívozník
@ 2026-11-19 11:47 ` Daniel P. Berrangé
  1 sibling, 0 replies; 4+ messages in thread
From: Daniel P. Berrangé @ 2026-11-19 11:47 UTC (permalink / raw)
  To: Markus Armbruster; +Cc: qemu-devel, devel, Paolo Bonzini

On Thu, Nov 19, 2026 at 10:17:27AM +0100, Markus Armbruster wrote:
> As far as I can tell, device-list-properties can do nothing you can't do
> with qom-list-properties.
> 
> Both take a QOM type name.
> 
> device-list-properties rejects types that aren't concrete subtypes of
> "device".
> 
> device-list-properties filters out properties named "type", "realized",
> "hotpluggable", "hotplugged", and "parent_bus".  These are precisely the
> properties every device inherits from "device".
> 
> Why is that useful?

I guess someone considered those to be internal propperties not
relevant to report to client apps.

With UserCreatable Objects, we have QAPI schema to define what are
public properties. One day we should extend that to other objects.
so public vs internal is clear.

None the less, I'm not seeing an obvious reason for device-list-properties
to be kept, if we intend qom-list-properties to stick around.

With regards,
Daniel
-- 
|: https://berrange.com       ~~        https://hachyderm.io/@berrange :|
|: https://libvirt.org          ~~          https://entangle-photo.org :|
|: https://pixelfed.art/berrange   ~~    https://fstop138.berrange.com :|



^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: Why device-list-properties?
  2026-11-19 10:38 ` Michal Prívozník
@ 2026-11-19 14:41   ` Markus Armbruster
  0 siblings, 0 replies; 4+ messages in thread
From: Markus Armbruster @ 2026-11-19 14:41 UTC (permalink / raw)
  To: Michal Prívozník
  Cc: qemu-devel, devel, Paolo Bonzini, Daniel P. Berrangé

Michal Prívozník <mprivozn@redhat.com> writes:

> On 11/19/26 10:17, Markus Armbruster wrote:
>> As far as I can tell, device-list-properties can do nothing you can't do
>> with qom-list-properties.
>> 
>> Both take a QOM type name.
>> 
>> device-list-properties rejects types that aren't concrete subtypes of
>> "device".
>> 
>> device-list-properties filters out properties named "type", "realized",
>> "hotpluggable", "hotplugged", and "parent_bus".  These are precisely the
>> properties every device inherits from "device".
>> 
>> Why is that useful?
>
> I think it's only by inertia (from libvirt's POV). I mean,
> device-list-properties was introduced to libvirt in 2012
> (v1.0.0-rc1~216) and qom-list-properties in 2018 (v4.4.0-rc1~209). Are
> you looking at obsoleting the former? If so, libvirt should adapt.

I indeed wonder whether we could deprecate device-list-properties in
favor of qom-list-properties.


^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2026-11-19 14:41 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-11-19  9:17 Why device-list-properties? Markus Armbruster
2026-11-19 10:38 ` Michal Prívozník
2026-11-19 14:41   ` Markus Armbruster
2026-11-19 11:47 ` Daniel P. Berrangé

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.