Subject: Re: Danger: Hardware Recommendation Request
To: Jordan K. Hubbard <jkh@time.cdrom.com>
From: Perry E. Metzger <perry@piermont.com>
List: port-i386
Date: 04/16/1997 11:19:16
"Jordan K. Hubbard" writes:
> > As I said, I'm stuck with the thing (and sadly maintain it for us),
> > but the only thing that will fix the NCR driver is to nuke it from
> > orbit. Its the only way to be sure.
> 
> Let us know when you're done with your rewrite. :-)

I doubt it will happen. The usual way with such things is that the
"good enough" driver ends up being used "forever", and it isn't clear
how it could really be fixed in the first place.

> P.S. I've had only good luck with the NCR driver, and certainly far
> better luck with the cards than I ever had with [the old] Buslogic,
> for example.  I qualify that because I think Buslogic has finally
> gotten their act together again, but still...  It seems a mite unfair
> to single out the NCR driver for this kind of negative attention, and
> the responsiveness of Stefan Esser to any and all problems has been
> more than good.

He's done an nice job. However, its in some sense beyond the
point. The fact that the driver must of necessity contain self
modifying script code -- the horror... the horror...

> At least it's actively supported, which is more than
> I can say for a lot of other drivers in the *BSD universe.

True enough. None the less, a percentage of setups simply lose, which
is not a good thing.

Perry