> I just wanted to drop you a line and ask you if there was any chance
> that int-vid could be officially supported in the 1.3 release . . .

Well, here's the deal. The old intvid 0.x patches for the powerbooks
encompassed a few different things. One was getting the location and
offset of the frame buffer correct. That was the only strictly *video*
related thing in the 0.x patches. The remainder of the patches actually
had to do with the way memory got mapped and if a NuBus probe was supported.

As far as I can tell, Scott has managed to put the first part in a table
and that obviously works since you can see the "Preserving x bytes"
message. So that means we've got "working" video.

The two other parts were basically hacks. Part 2 fakes the machine into
thinking it was doing a serial boot because NetBSD does some range faking
and some other bizzarities when it does a serial boot. Part 3 simply
skipped the NuBus space memory probe in machdep.c.

Since those last two parts are hacks I really can't recommend that they
go into our production kernel. The better alternative is for somebody
to take those hacks and see what they did and do a *proper* fix to the
kernel.

Unfortunately, while I might be able to do this I lack both the time and
a PowerBook to really dig into the project. The fact that I was able to
even get the hack working in the first place given the
"compile->email->pray->return" development cycle is quite a mircle in
itself. :)

> It would be really nice to be able to keep up with -current from
> 1.3 onward, and to know that the video would be part of the 'mainstream'
> source tree.

Definately. I'm also looking to support the GSC video controller chip in the
future but since I lack a Powerbook that could take a long time. :)

My goals with intvid2 are to concentrate on getting real video to folks
with interal video. This is different than my goal for intvid0 which was
simply to get everybody working on B&W internal video who, at the time,
were not able to boot on internal video at all.

I'm hoping that we can all look forward to a day when you can boot on
internal video and treat it just like MacOS where you can change modes,
colors, resolutions, etc. Obviously this is limited by what your hardware
can do but most of the hardware out there can do alot.

> P.S.  If this problem involves anything more than the video portion of
> the kernel (I don't think it does), then whoever knows what needs to be
> done, please let me know and I'll help out any way that I can. 

Since we're dealing with memory/NuBus problems I would suggest you talk to
Scott about it and see what he can do. We *can* boot intvid on a powerbook
as we've done it in the past. The trick is to find out which magical
incantation was getting done that isn't getting done now. :)