dbus 1.2.16

[Marcel Holtmann]

Hi Thiago,

> >Definitely; I think they're probably right but haven't had time to go
> >through them myself.  This release was to get out some new stuff that
> >Upstart depended on, at the same time I went through bugzilla and took
> >some easy pickings, along with the dbus-send race.
> >
> >Just chatted with Thiago who did a review on fd passing, we'll get it in
> > soon
> 
> So, update on the fd-passing stuff:
>  - I did a full review of the patch at GCDS, when Lennart was right next 
> to me to explain stuff
>  - Everything looked fine
>  - I had a compilation error with the __sync stuff, but I think it's 
> because of my distro upgrading gcc to 4.4 but not all libraries. I need to 
> investigate that.
>  - the Qt DBus tests passed flawlessly
>  - I didn't get to run "make check"

sounds great. I was asking, because I had no issues with Lennart's
patches and couldn't see what is holding them up.

> I'd also like to point out that the fd-passing stuff should not go into 
> 1.2. It should go into master, which will end up in the 1.4.0 release[*]. 
> So we should start creating a timeline for 1.4: if OSX and Windows 
> features want to be in 1.4, they'll have to be merged into master before a 
> deadline. Otherwise, they'll be on hold until 1.6.
> 
> [*] As soon as we merge the fd-passing, I propose we release 1.3.0 (marked 
> unstable).

I agree that fd-passing support should come with 1.4 release, but I
don't like for this to take another 3 month or even longer. The patches
have been waiting for inclusion for some time now. So can we go ahead
and create 1.3.0 with the fd-passing. I would even go that far that a
1.4 release where the only new feature is the fd-passing is acceptable.

Regards

Marcel