>From: Stephen Degler SYSTEMS-ADMIN Admin NY x9476 <sdegler@swaps-comm.ml.com>
>Received: (sdegler@localhost) by table.ndswaps.ml.com (8.6.9/8.6.9) id SAA20805 for port-pmax@NetBSD.ORG; Wed, 8 Mar 1995 18:10:16 -0500
>Message-Id: <199503082310.SAA20805@table.ndswaps.ml.com>
>Subject: which binaries ??
>To: port-pmax@NetBSD.ORG


>Hi,

>I was able to boot a DS3100 using the valentines day kernel, but it
>seemed that I had to use binaries from stanford, not netbsd.org.  When
>I tried to boot with them, the exec of init would fail.  Can anyone
>shed some light on this?

I've been working too hard and my memory is fuzzy, but:
perhaps it's an a.out vs. ELF problem.  I still use `native' NetBSD
a.out binaries. 

It's possible that the kernel on Gregorio.Stanford.EDU was built
*before* Ted Lemon's changes to support ELF binaries were completely
merged into the NetBSD kernel.  This seems unlikely, though: more
likely the kernel on Gregorio doesn't support 4.4bsd a.out binaries.

Personally, I've had no problems with the a.out binaries, except of
course for the  FP bugs biting gawk and pow().
I guess I should just bite the bullet and update my toolset to ELF.

If anyone who had problems with the bootblocks on Gregorio
could  try yesterday's bootblocks, they'll get much more informative
error messages. The bootblocks should work with either set of binaries.

--Jonathan