moving ALTQ out of contrib

[Gleb Smirnoff]

On Wed, Apr 15, 2015 at 03:02:03PM +0200, Luigi Rizzo wrote:
L> > L> Then if you do sys/net/altq/ do you also plan to split the current
L> > L> content of sys/net/ into separate subdirectories ?
L> > L>
L> > L> We currently have quite a few separate things in sys/net/, such as
L> > L> - various bpf files
L> > L> - generic ifnet support (including raw sockets)
L> > L> - various libraries (compression and hash functions)
L> > L> - routing code
L> > L> - bridging code
L> > L> - a ton of special ifnets, (tun, tap, epair, gif, ....)
L> > L> - bridging code
L> > L> that could benefit from a bit of partitioning
L> >
L> > I definitely agree that a) special interfaces b) lagg+lacp
L> > c) generic libraries should be separated. I don't mind if anyone does
L> > this job :)
L> >
L> > But I personally would prefer is this is done after the lifetime
L> > of the projects/ifnet branch, since if stuff is moved while I work
L> > on projects/ifnet, my merging will become a nightmare. I already have
L> > conflicts quite often.
L> >
L> 
L> sure, there is no rush.
L> I was just trying to understand why your preference is for sys/net/altq
L> instead of sys/netaltq as we have for other components.

I think that dropping everything into sys/ is historical. Perfectly,
protoctols should live in net/ as well, like net/inet and net/inet6,
but of course no one would support that move.

The most recent network related subdir in sys/ is netpfil. Frankly
speaking I didn't like the name, and that word wasn't my, but Bjoern's.
But, if I advocated for net/pfil, then the argument would be that pfil
hooks reside in sys/netinet and in sys/netinet6, not in sys/net, thus
moving that into net/pfil is wrong. At the time I really wanted to gather
or pfil consumers in one place, and move pf out of contrib, so I accepted
name suggested by Bjoern without arguing.

-- 
Totus tuus, Glebius.