Feico Dillema happened to mention:
> I more-or-less solved my problem by switching motherboards. I tried the 
> 1542CF on another '486 MB and a '386 I had access to. Both work fine,
> ....
> Well, I've swapped motherboards, which for a 386/486 is a low-cost solution
> nowadays, so basically my problem is `solved'. Still curious about why the
> original MB is problematic, but alas...

I went home last night to check my configuration.  My motherboard is a
Pentium PCI (the flagship of my gang of computers).  In particular it
is a Gigabyte board, with Intel HX chipset, and a P5-133 chip.

The story is that /var/log/messages tells me the 1542 is IRQ-11 and DRQ-5.
I went into the bios using the usual DEL key thing and found the setup
page marked PnP/PCI configuration (an award bios BTW), and in particular
IRQ-11 was marked PnP/PCI, so I changed it to "Legacy ISA" (did the same
for DRQ-5 also, tho this wasn't and isn't particularly relevant).  It turns
out that the PCI bios was handing IRQ-11 to my PCI video card.  Now it gives
IRQ-9 to the video and there is no conflict.

Now I am on to ANOTHER problem:

Namely, I cable up my scsi drive (I am booting and running from an IDE drive),
and try for a test:

	dd if=/dev/sd0d of=zzz count=8

Now I get lots of errors, including the following messages:
	
	aha0: DMA beyond end of ISA
	aha0: BAD CCB ADDR!
	aha0: timed out AGAIN!
	aha0: .... (and so on)

Did I forget to mention that I have 64M of ram and I am running netBSD 1.1 ?

The QUESTION is:

Does NetBSD 1.2 or beyond have a fix for the 16M isa limit of this kind
of hardware??  Thanks to anyone with info willing to take the time to help!

	Tom


-- 
	Tom Trebisky			Steward Observatory
	ttrebisky@as.arizona.edu	University of Arizona
	http://kofa.as.arizona.edu/	Tucson, Arizona 85721
					(520) 621-5135