On Thu, 29 Jun 1995, Brian J. Fox wrote:

>    On Wed, 28 Jun 1995, 'Most Excellent' Eric Mehlhaff wrote:
> 
>    > Herb Peyerl <hpeyerl@beer.org> recently wrote:
>    > 
>    > 
>    > >"Eduardo E. Horvath eeh@btr.com" <eeh@btr.btr.com>  wrote:
>    > > > 1) I have been trying to build a new kernel for an hp-9000/345.  The new 
>    > > > kernel boots, but as soon as I go into multi-user mode I get a panic 
>    > >
>    > >There are a lot of panic() calls in the kernel.  Unfortunately, it's
>    > >difficult to diagnose your problem without knowing where the kernel
>    > >panic'd... 
>    > 
>    > I just want to say I have similar problems trying to boot up a
>    > hp-9000/350 on kernels compiled from current tarballs.  I've tracked it
>    > down to right where /etc/rc does the mount -nonfs (just after fsck).
> 
>    While this is rather interesting, I'm trying to get a 1.0 kernel to work, 
>    not a -current.  I finally have some info on the panic (I usually get two 
>    of them so the first scrolls off the screen):
> 
>    trap type=2 code=0 v=0
>    pid=24 pc=000051ea ps=2000
>    sfc = 0001 dfc=0001
> 
>    followed, of course, by the manditory register dump and stack dump.
> 
>    I was hoping that someone would tell me that this is a well known problem 
>    with a simple fix.
> 
>    I'm going to try to compile the INSTALL kernel and see how that fares.
> 
>    Eduardo
> 
> Turn off all forms of debugging in the config.  Rebuild from the 1.0
> sources.
> 
> Don't ask me why, but it worked for me.

Well, I re-built and re-installed the patched version of config (the 
original was not core dumping though) and then cleaned and re-built the 
INSTALL kernel, and now it seems to run just fine.

Eduardo