> Followup to: <34895864.7A5F2C91@linux.de>
> By author: Stefan Reinauer <stepan@linux.de>
> In newsgroup: linux.dev.kernel
> >
> > Argh? We´re already on feature freeze? Why is it so early this time?
> >
>
> Early? If anything, it's pretty late (2.0.0 was a long time ago), and
> we're already up to 2.1.71. Many kernel internals have changed,
> especially the VFS (dcache), and it's going to take some work to get
> the lowlevel stuff fixed, especially for filesystems like nfs, smbfs
> and autofs. If we want a 2.2 in early 1998 it is time to focus on
> that stuff. Hence, 2.2 will have a (hopefully cleaned-up) version of
> autofs v3, and I have suspended all work on autofs v4 (another reason
> is that I would like libc 6 to be better distributed before releasing
> autofs v4 -- it will most likely not support libc 5 at all.)
>
> The cleft between 2.0 and 2.1 has grown to the point where it is
> impeding Linux development, especially since people keep clamoring for
> their favourite features to get backported to 2.0.
>
> -hpa
Agreed, however, I beleve that 2.3 should be started shortly after the 2.2
release.. None of this doddling around, it encourages features to be added
to the stable kernel.. There is no sensiable reason that there had to be
32 stable kernel patch levels..
Maby Linus should set a goal of no more then 20 stable kernel patch
levels..