Subject: Re: exec: short read
To: Rob Windsor <windsor@warthog.com>
From: Mason Loring Bliss <mason@acheron.middleboro.ma.us>
List: port-i386
Date: 10/07/1997 18:44:33
On 10/7/97 at 1:12 PM -0500, you wrote:

> I can consistently compile a kernel that will not boot.  I haven't narrowed
> down which specific option is causing the problem, but I can tell you that
> if I compile a GENERIC kernel, it works just fine.

Hm... I got a similar error on a Cyrix 486 DX2/66. The cause ended up being
flaky cache. If you can boot normally with a kernel someone else has built,
then maybe this *isn't* your problem, but you might try disabling your
cache (both in BIOS and on the motherboard - doing it in the BIOS alone was
insufficient on my system) and see if that does anything. You might also do
a careful check with fsck to see if you have any disk corruption... If
you're getting a short read while booting, you might also be having
problems at other, less dramatic and noticeable times.

Take my advice with a grain of salt! I haven't done a lot with NetBSD/i386,
and I've never managed to successfully run -current. (My 486 is running
1.2.1 right now.) I'm simply responding because I've had vaguely similar
errors.

Later...

--
Mason Loring Bliss   mason@acheron.middleboro.ma.us   www.webtrek.com/mason
awake ? sleep() : dream();  If a cow laughed, would milk come out her nose?