|
From: | Anthony Liguori |
Subject: | Re: [Qemu-devel] [PATCH v2 00/18] qom: dynamic properties and composition tree (v2) |
Date: | Mon, 05 Dec 2011 10:13:02 -0600 |
User-agent: | Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.2.21) Gecko/20110831 Lightning/1.0b2 Thunderbird/3.1.13 |
On 12/05/2011 09:47 AM, Paolo Bonzini wrote:
On 12/05/2011 03:36 PM, Anthony Liguori wrote:No. A device is-a interface. Hopefully the above example will make it more clear.No, but I'm confident that there will be a sane way to access the list of interfaces that you embed in the Object type. :)
We definitely need to think through introspection. There are a few things we need to address with introspection:
0) What are all of the classes available 1) What are all of the classes/interfaces implemented by a type 2) What are all of the properties available in a class (0) and (1) just need QMP interfaces. It's very straight forward.(2) is a bit more tricky. You can instantiate a dummy object and introspect on the live object. Dynamic properties make that a bit challenging through.
Maybe now is the right time to rename the legacy properties to all be prefixed with qdev-? That way we don't need to introduce two different types for a single property.Why do you need such a prefix?
To avoid an all-at-once conversion. I don't want to break -device and maintaining string properties make it quite a bit easier to support -device.
I'd like to leave the string properties in place until we have a good way to create objects via QMP.
My rough thinking is that we keep -device around until 2.0. Regards, Anthony Liguori
Paolo
[Prev in Thread] | Current Thread | [Next in Thread] |