On Thu, 10 Jul 1997 09:08:48 -0400 (EDT),
ADAMGOOD@delphi.com wrote:
> 
> So I tried to hunt those down.  Unfortunately, they're no where to be
> found.  John's Web page does tell us that we don't need his patches
> anymore because all of his code is now a part of the official source
> tree and we can get it all by just supping -current.

Instead of supping, you can ftp tar balls from ftp.netbsd.org.  The
directory is /pub/NetBSD-current/tar_files/src.  To compile a kernel,
you should at least get sys.tar.gz and config.tar.gz, but you may have
to update /usr/bin/make if your make is old enough.  You might want to
grab only the source files from /pub/NetBSD-current/src/usr.bin/make
instead of grabbing the entire usr.bin.tar.gz (which is huge!).

> So many people are putting so much hard work into this.  It's a real
> shame that it seems so disorganized.

I suppose it does to you because you're chasing a moving target (or, a
few moving targets).  I like to think that things you're chasing are
meant to merge into the main source tree eventually.  They are in
constant flux as the authors make improvements.  IOW, they are in
transitional stages, and usually aren't meant for everybody.

If you really want to follow the current, you should subscribe to
current-users.  You'll see a lot of "sup on YYMMDD doesn't build" type
of complaints there.  Things are really in constant flux with the
current.

> I'd like to help, but I can't even compile a kernel that works on my
> machine.

I do understand your frustration...

> Is there any way that everyone's work could be centralized (like on
> ftp.macbsd.com), and all info on different projects be put together
> under the Web or FAQ pages?

Do you have any suggestion how we go about it?

> In any case, Scott Reynolds, could you please incorporate Michael
> Zucca's work into the -current source tree?

I'm sure he's working on that direction.  It will happen.  However,
the "it works for me" approach doesn't always yield better results in
the long run.  I think that's why it takes so long for some things to
go into the tree (e.g. ISA DMA bounce buffer).

> I really care a lot about this project.  It's hard not to be frustrated
> when all the hard work to get your machine running has already been
> done, and all the pieces are right there and there's nothing you can
> do to put things together and make them work.
> 
> Sorry if I'm complaining too much.

Not at all.  I'm sorry for your frustrating situation, and I'm sorry I
can't explain better (hopefully others will).  But I think your
frustration may be resolved with a few tips such as ftp'ing source
tar balls instead of supping, etc...

Ken