cvs commit: src/sbin/kldunload kldunload.8 kldunload.c

[Poul-Henning Kamp]

In message <40F43FB3.1030607 at freebsd.org>, Scott Long writes:
>Poul-Henning Kamp wrote:
>
>> In message <40F43DFE.3000402 at freebsd.org>, Scott Long writes:
>> 
>> 
>>>It typically won't be duplicated.  This particular work is good for
>>>things like GEOM modules that don't fit into newbus.  If at some point
>>>in the future newbus becomes aware of pseudo-drivers (drivers not
>>>attached by a hardware bus) then this should indeed be revisited.
>> 
>> 
>> I wish I could grasp from where this "newbus fits all" attiude
>> comes, I certainly see nothing anywhere that could explain it.
>
>NetBSD has most things under newbus.

yes, and ?

I can't help to notice that only the words "NetBSD" and "most" are
present in your reply whereas the words "FreeBSD" and "all" are
significantly absent.

It is not realistic to expect us to ever tie netgraph modules, tty
linedisciplines, network protocols, filesystems, vm objects, vnodes
and geom classes together with newbus.

Not now, not in one year and not in ten years time.

And since all of these things are involved in loadable modules,
I think we should stop wasting time dreaming about the day the
world will be one bit tangle of newbus.

If we had a big company we could throw a load of money after random
acts of "top down architecture" like that, but with a volunteer
project it is just not going to happen.

-- 
Poul-Henning Kamp       | UNIX since Zilog Zeus 3.20
phk at FreeBSD.ORG         | TCP/IP since RFC 956
FreeBSD committer       | BSD since 4.3-tahoe    
Never attribute to malice what can adequately be explained by incompetence.