Hi all,

I agree with John that most of what kept the colorkit kernels from
working on the quadras has been actually fixed in -current sources.

Another thing of note is that the colorkit kernels were built during
a time when nubus ethernet support was being problematic.  So, currently,
if I want to use color X on my MacIIx I have to yank the ethernet card.
If we updated the colorkit kernel patches to -current, this problem would
also disappear.

As to John's question about Taras' patches being rolled into the source
tree, I don't think they are in there yet.  But, I haven't tried any
-current kernel on my MacIIx yet.  I guess I'll try that tonight and
see how it goes(I'm not sure what changes to look for in the source,
so I guess testing a -current kernel is the quickest way to know).

Steve Brown,
sbrown@best.com

p.s.  Are Taras' patches publicly available for ftp??? If so, could
        I have the URL??

>> I'm not perfectly sure what you mean here, but I'd say no, it
>> wouldn't: if none of the color kernels works on Q700, it would make
>> sense for a kernel, which you build out of the current source tree
>> with the same color code Taras' kernels got, *not* to work.
>>
>> ken
>
>Here's my reasoning...
>
>The kernels on ftp.macbsd.org are somewhat old. That means they fail on
>a Q700 for reasons such as no SCSI support, etc. Current kernels support
>the Q700 almost fully except for video. By using mv.c in a current
>kernel, you would fix all the SCSI problems, etc. and also have the code
>to drive a color video card. (I'm going to see if anybody's tried the
>kernel on a Q700 w/video card and if they remember where it failed. If
>it never got past the video initialization, then making a new kernel
>woudn't work. If it says "ARGH!! No Boot device!" although there's a
>NetBSD partition present, then it's a SCSI problem and making a new
>kernel might fix it.
>
>P.S. Do current kernels (i.e. GENERIC) have Taras' code in them?
>
>Thanks for your help!
>
>John Plotecher
>jplotech@diisd.k12.mi.us