[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: [libvirt] [PATCH 09/14] conf: Add VIR_DOMAIN_CHR_SERIAL_TARGET_TYPE_PL011

On Thursday, 16 November 2017 16:44:24 CET Andrea Bolognani wrote:
> On Thu, 2017-11-16 at 16:17 +0100, Pino Toscano wrote:
> > On Thursday, 16 November 2017 14:56:32 CET Pavel Hrdina wrote:
> > > How about we introduce another attribute/element that would specify the
> > > exact model of the serial device and it would be optional, libvirt would
> > > be able to choose the model if none is specified.
> > 
> > Can you please provide an example of how it would look like?
>   <serial type='pty'>
>     <target type='xxx-serial>
>       <model name='splc'/>
>     </target>
>   </serial>

IMHO that looks a bit more convoluted than necessary.

> > > I was no able to find anything about s390 and its sclp/sclplm consoles
> > > but it's the same case.  I would expect that both devices are connected
> > > to the same BUS, we just need to find the BUS name to use it $BUS-serial
> > > and have sclp/sclplm as models.
> > 
> > There is no real bus for sclp/sclplm -- there is only an internal "bus"
> > in QEMU, but it should not be exposed in upper layers (as it is only an
> > implementation detail).
> There's probably some name, in some spec, somewhere :)


> Or we could use 's390-serial', but that would be suboptimal.

That is way too generic, and representing something which does not
exist (unlike the current -serial ones).

I still do not see the advantage of this, though.

Pino Toscano

Attachment: signature.asc
Description: This is a digitally signed message part.

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]