Starting the kdbus discussions

[Lennart Poettering]

On Thu, 02.01.14 13:56, Kevin Krammer (krammer at kde.org) wrote:

> On Thursday, 2014-01-02, 02:18:49, Lennart Poettering wrote:
> > On Wed, 01.01.14 22:38, Kevin Krammer (krammer at kde.org) wrote:
> 
> > > When you write "kdbus as system bus" does that imply it is also being used
> > > for session busses (since those are usually "easier") or is kdbus only
> > > used for system bus and session busses are still backed by dbus-daemon?
> > 
> > Our primary focus for now is the system bus, and dbus-daemon can
> > continue to be used for the session bus, even on a system that uses
> > kdbus for the system bus.
> 
> I see, thanks.
> I was primarly asking because there was no explicit reference to the session 
> bus, so I was wondering whether session bus was an implied usage or still 
> handled by the traditional daemon.
> 
> My understanding is now that the latter is to expected situation short and 
> medium term.
> 
> Please correct me if I am wrong: that implies that client libraries need to 
> support the traditional marshalling format anyway and can, additionally, 
> support the new one for use with kdbus backed system bus.
> I.e. it is currently not viable to just talk the new format.

That is correct. The new bus library we created for systemd and other
low-level tools speaks both marshallings and protocols and hides the
differences away. I'd expect all other bus libraries that want to
support kdbus, to do the same and keep dbus1 compat in.

This is particularly the case as we have not defined gvariant
marshalling for usage on socket transports so far, which means that by
only supporting gvariant marshallings you lose network transparency as
kdbus is a strictly local transport.

Lennart

-- 
Lennart Poettering, Red Hat