Subject: Re: AXPpci-33: kernel stack not valid halt...
To: None <Marc.Gutschner@Frankfurt.NetSurf.DE>
From: Jason Thorpe <thorpej@nas.nasa.gov>
List: port-alpha
Date: 10/05/1999 12:11:57
On Tue, 5 Oct 1999 21:01:11 +0200 (CEST) 
 "Marc.Gutschner" <marc@Frankfurt.NetSurf.DE> wrote:

 > On Tue, 5 Oct 1999, Matthew Jacob wrote:
 > 
 > > 
 > > This has been happening for over a  year on the 8200.
 > > 
 > 
 > Uh! This doesn't really sound encouraging :-/ In the meantime I've done
 > some - uh - experiments with FreeBSD/alpha (boots all the way to sysinst)
 > and (yuck!) Linux (boots up to the point where it tries to mount '/'). So
 > now I'm pretty sure my problem is not (at least not obviously) hardware
 > related...

The problem on the 8200 I believe to be a bug in changes that were
made to the DWLPx code.  At least, I have seen not other evidence to
suggest otherwise.

 > Currently I'm sucking down all available floppy images from different
 > releases just to see if this makes any difference. I gave the 1.4.1
 > floppies another try and here is a manual transcript of what happens:
 > 
 > 	AXPpci Common Console X4.7-1860 built on Nov 1 1996 at 06:00:01
 > 	>>> boot dva0
 > 	[mumbling about finding a valib boot block etc.pp.]
 > 	
 > 	base=110000, image_start=0, image_bytes=1e00
 > 	initializing HWRPB at 2000
 > 	initializing page table at 102000
 > 	initializing machine state
 > 	setting affinity to the primary CPU
 > 
 > 	Jumping to bootstrap code
 > 
 > 	NetBSD/alpha 1.4.1 Primary Boot +
 > 	+.
 > 	Jumping to entry point
 > 
 > 	halted CPU0
 > 	halt code = 2
 > 	kernel stack not valid halt
 > 	PC=d6af60d3ca0794c0

...if that's really what the PC is, then something is Very Very Wrong.

        -- Jason R. Thorpe <thorpej@nas.nasa.gov>