I'm not sure if this is the same problem, but I've been testing some kernels
with the interrupt fixes that Allen has been working on, and notice that
a lot of scsi errors are being reported that I hadn't seen before, ie.:

sd0(ncrscsi0:0:0): command aborted, data = 00 00 06 00 00 00 00 ff ff f
f 00 00 00 00 00 00 00 00 00 00 2a 00 00 33

I thought that this might be a case of some sort of error reporting being 
turned on that wasn't before, but they may actually be new errors.

Brad
> Bas news. Looks like the April 1 sup without any mods causes the SCSI phase
> problems.
> 
> This is the straight kernel without any of my intvid mods. Is this mysterious
> phase/memory problem justa IIvx thing or are other people having this
> problem?
> 
> In addition, it seems that while the standard 1.2 has no problem recogizing
> the fact that slot D has a non-configured NB-PRL card, the new kernels
> don't see it. This is likely a related problem.
> 
> My cards run like this:
> 
> slot c:	AE Shadow 4 port card (No driver)
> slot d: National Inst. NB-PRL Card (No driver, disappears under new)
> slot e: Interlan Ethernet card (recognized under old or new)
> 
> Someday I'd like to write an LKM for the NB-PRL but it needs to keep showing
> up in the card probe :)
>