[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH for-2.5 2/2] input: Promote 'input-send-event' t
From: |
Markus Armbruster |
Subject: |
Re: [Qemu-devel] [PATCH for-2.5 2/2] input: Promote 'input-send-event' to stable API |
Date: |
Thu, 12 Nov 2015 12:10:12 +0100 |
User-agent: |
Gnus/5.13 (Gnus v5.13) Emacs/24.5 (gnu/linux) |
Gerd Hoffmann <address@hidden> writes:
> On Do, 2015-11-12 at 09:23 +0100, Markus Armbruster wrote:
>> Eric Blake <address@hidden> writes:
>>
>> > We've had 'x-input-send-event' since 2.3, with no further
>> > changes to the interface other than tweaks in the previous patch
>> > to the spelling of the enum constants ('X' and 'WheelUp' changed
>> > to 'x' and 'wheel-up').
>> >
>> > What's more, changing the spelling of enum constants is not easy
>> > to introspect prior to 2.5; so a client that was relying on the
>> > experimental command can't easily tell which spelling is expected.
>> > But 'query-commands' works in all qemu versions that supported
>> > the command, so renaming the command now makes it an easy thing
>> > to determine which spelling of the enum values to use.
>> >
>> > Thus, it's time to promote this interface to stable.
>>
>> The x- goes back to commit df5b2ad:
>>
>> input: move input-send-event into experimental namespace
>>
>> Ongoing discussions on how we are going to specify the console,
>> so tag the command as experiental so we can refine things in
>> the 2.3 development cycle.
>>
>> Have we settled "how we are going to specify the console"? If yes,
>> commit, please. If no, I'm afraid the command should stay experimental.
>
> Good question. I don't think so.
>
> IIRC the question was whenever we'll leave it as-is (console=<index>),
> or whenever we'll do something like display=<id>,head=<nr> instead.
>
> The latter would be consistent with how we are doing input routing, i.e.
> grouping display and input devices to a seat for multiseat setups (see
> docs/multiseat.txt for more details).
>
> The consoles are already present in the qom tree
> as /backend/console[<index>] nodes, and they have device + head
> children. So qom users can map console=<index> to
> display=<id>,head=<nr> and visa versa already. So from a functionality
> point of view it doesn't really matter, it is largely a matter of
> taste ...
The thread leading to the x-:
http://lists.nongnu.org/archive/html/qemu-devel/2014-11/msg03197.html
I'm not sure the console numbers were visible in QOM back then.
We also discussed use of qdev ID.
Let's keep the x- until we've figured this out.