On Sat, 31 Jan 1998 18:13:45 EST,
Michael R Zucca rearranged the electrons to say:
>
>Then I moved the card to my IIvx. Lo and behold the card worked with the
>stock 1.3 Distribution! My cpudelay went from 80 to 240! This was without
>the Daystar extension installed. I then installed the extension and booted
>with full 040 caching on and the cpudelay went to 425! 

I've had a Daystar 040 working in a IIvx for over a year.  I haven't
tried it in a IIci for a long time, the original problem was that the
IIci has non-contiguous memory mapping and the code that probes that
only worked with the MMU in the 030.  Has this changed?

I wasn't aware that enabling the DayStar extension made a difference
(I've never actually tested it w/o the extension).  Here's the beginning
of my boot process:

NetBSD 1.3B (MCM-SBC) #0: Fri Jan 23 00:21:36 EST 1998
    root@marauders.ibl.org:/usr/src/sys/arch/mac68k/compile/MCM-SBC
Apple Macintosh IIvx  (68040)
cpu: delay factor 426
real mem = 20971520
avail mem = 16740352

Looks like my custom kernel is a little bit faster!  :-)  Anyone know
if there's a faster netbsd-mac68k system out there?

>I tried two different ae type ethernet cards and both work quite speedily.
>However, I haven't tried any serial tests and I wonder if there will be
>any weird impact. ADB seems to work OK (At least MRG. I'll compile a
>-current kernel soon).

You must use MRG ADB with the IIvx and the Daystar.  All kernels
I've ever compiled using the hardware direct ADB hang during boot
(my most recent try was with source supped around 1/15).  The biggest
drag is that I've never been able to get the machine to reboot
properly (reboot command just messes up the video and hangs), requiring
a manual power cycle instead.

Sean

---
Sean Sweda
sweda@ibl.org                     http://www.ibl.org/~sweda
sweda@netcommandos.com            http://advantage.netcommandos.com