> From: "Perry E. Metzger" <perry@piermont.com>
:::
> Are you running -current or something reasonably close to it? Ross
> *did* check in a bunch of fixes a while back...

Sadly, I didn't do enough. Here is the status on that thing:

	* the 1016 snapshot does have all the fixes, so his problem
	  won't be fixed by an update

	* i did a giant re-integration with freebsd; but it didn't help
	  at all, and so I just checked in the integrated driver on a branch
	  and left it there. (It's also up for ftp, but I do NOT recommend
	  that anyone waste time trying it, it won't help.) FreeBSD has
	  since diverged again in a different direction, and they haven't
	  fixed ncr.c, so I think we are on our own here.

	* i've collected a huge pile of docs on the ncr hardware and
	  the script engine, as well as a fairly large assortment of
	  ncr HW, and I'm trying to find a day to beat on that driver
	  again.

	* i've been in touch with one of the authors and a SCSI project
	  goning on down here...there is some chance that I can arrange
	  a funded project to beef up SCSI or at least get some resources
	  contributed to our project as a kind of compensation for
	  our developing more elaborate SCSI support.

	* there is a netbsd'er who has started writing a replacement

To fix the recently appearing problems, we just need a little bit of
error recovery added...that is...it's going to be an easy fix, it won't
take a lot of work, it's just that you can't work on that thing casually,
you need to dedicate some time given the intermittent nature of the error
and the slow compile--reboot--test cycle.

By the way, speaking of controllers, does anyone want to sell me some old
EISA hardware for my two recently acquiered (but completely bare) Jensen
boards?  (:-) Preferably, HW that SRM can grok?

  --Ross Harvey