Re: 5.0beta won't boot on a Dell Dimension 8400

["Steven M. Bellovin" <smb%cs.columbia.edu@localhost>]

On Sat, 29 Nov 2008 20:58:50 +0100
Manuel Bouyer <bouyer%antioche.eu.org@localhost> wrote:

> On Fri, Nov 28, 2008 at 08:55:45PM -0500, Steven M. Bellovin wrote:
> > I have two almost-identical Dell 8400s.  On one of them, I can't
> > boot a 5.0beta installation CD from the other day -- I get an NMI
> > panic.  (The exact version doesn't matter; I haven't been able to
> > boot any -current for a while.) However, 4.99.72 from August 9 runs
> > just fine.  Oddly enough, the other one has no trouble booting.
> > 
> > If I go into the BIOS and change the disks from AHCI to ATA,
> > however, I can boot.  The last line printed before the panic is
> > 
> >     ahcisata0 port 0: device present, speed: 1.5Gb/s
> 
> Can you get a stack trace from here ?
> 
No.  When I boot a 5.0beta CD, the keyboard acts like a serial line
with a speed mismatch: I type 'b' and see three 'y's echoed.  A 't'
produces a different 3-character sequence: 'yhy'.  I tried booting with
a USB keyboard connected instead of the usual PS/2; it wouldn't talk at
all at that point, though it could talk to the boot loader (not
surprising, I suspect).  When I boot a -current kernel, the PS/2
keyboard isn't responsive at all.

The last few lines on the screen (with -current failing) are:

IPsec: Initialized Security Association Processing
ahcisata0 port 0: device present, speed: 1.5GB/s
NMI ... going to debugger
NStopped in pid 0.2 (system) at  netbsd:sse2_idlezero_page+0x18: jnz
netbsd:sse2_idlezero_page+0x46

Is there some kernel option or patch I could try to make the kernel go
to the backtrace function immediately?  

Someone else suggested that I might have different BIOS versions
between the two machines.  That's entirely possible, but I won't have
physical access to the working machine until Monday.  I'm also
reluctant to update the BIOS on the non-working one pending a
diagnosis, since it had been working well on NetBSD for >3.5 years.


                --Steve Bellovin, http://www.cs.columbia.edu/~smb