Re: [Qemu-devel] [PATCH 3/6] virtio-console: Add a virtio-console bus, support for multiple ports

[Amit Shah]

On (Wed) Oct 07 2009 [15:06:08], Gerd Hoffmann wrote:
> On 10/07/09 13:42, Amit Shah wrote:
>> On (Wed) Oct 07 2009 [13:33:57], Gerd Hoffmann wrote:
>>> Second, the bus= argument is optional.  If not specified, qdev will pick
>>> the first bus of a matching type it finds.  So as long you have a single
>>> port/scsi/usb/... bus only you don't need bus= at all.  You can do:
>>
>> The problem with this is that the management solution needs to know then
>> what is the default bus name (which could change if the code gets
>> updated).
>
> No problem.  Just don't use -virtioconsole.  Go with -device  
> virtio-serial-bus,id=... + -device virtport,bus=.. then and explicitly  
> name your devices (and thereby the buses too).
>
> -virtioconsole should *really* be a pure backward compatibility thing.  
> Use case:  You have a script starting qemu using -virtioconsole.  After  
> upgrading qemu it should continue to work, i.e. create a device which  
> the guest can use as before the upgrade and which is linked up to a  
> chardev as it was before.
>
> Anything which wants to use the new features can (and should) completely  
> ignore -virtioconsole.  I just wanted to point out that mixing old and  
> new style is *possible*.  It wasn't my intention to imply that I  
> *recommend* doing that.

There should be some way of deprecating commands in qemu. Maybe in 1-2
release cycles.

>> And also there's the other problem of a console port spawning a bus
>> (which could end up spawning another console port at #0...)
>
> parse error.
> I don't understand what problem you are trying to point out.

Oh; I was stuck at the earlier suggestion made by you:

-device virtio-serial-pci,chardev=...

which implies -device virtioconsole (with a port at id 0)

and then someone also doing

-virtioconsole ...

which would end up creating another bus for this current backward compat
reason.

                Amit