Subject: Re: machine checks on pc164
To: None <port-alpha@netbsd.org>
From: Andrei A. Dergatchev <A.Dergatchev@tn.utwente.nl>
List: port-alpha
Date: 08/23/2000 17:35:21
This is "unrecoverable memory errors" :-(

From post in redhat axp-list
(http://www.lib.uaa.alaska.edu/axp-list/archive/1999-12/0813.html)

From "Alpha Architecture Reference Manual" :

Machine checks are of four (4) types, as in the following table:

Code Reason Example or Common Cause
==== ====== =======================

620 System Correctable correctable errors in the memory subsystem,
                                  eg single bit ECC errors, detected async
to
                                  processor execution

630 Processor Correctable correctable cache and TLB errors, detected
                                  internally by the processor

660 System Uncorrectable unrecoverable memory errors

670 Processor Uncorrectable unrecoverable cache or TLB errors, or
                                  read of a non-existent I/O space location

This means that you're getting a double-bit error in your RAM.

So, it seems you'd better check (or replace ?) RAM.

HTH,

Andrei


> Folks,
>
> Twice in the last couple of days I've had my pc164 panic with a machine
> check:
>
>         unexpected machine check:
>
>             mces    = 0x1
>             vector  = 0x660
>             param   = 0xfffffc0000006060
>             pc      = 0xfffffc0000300640
>             ra      = 0xfffffc000036600c
>             curproc = 0x0
>
>         panic: machine check
>
> >From the archives, it seems that this is a hardware-type problem
> (especially the reference to the "vector = 0x660").
>
> Both times when this has happened, I've been restoring files from an
> Exabyte 8500 (attached to an Adaptec 1542) to a IBM IDE disk (I've only
> had IDE disks on this box for a week or so now).  The latest time, I got
> the machine check almost immediately after I started netscape, before
> the window even appeared.  I can't remember exactly what I was doing the
> first time, but I think the circumstances were similiar.
>
> If it is a hardware problem it seems a little too repeatable...
>
> Any ideas?
> --
> Simon Burge                            <simonb@wasabisystems.com>
> NetBSD Sales, Support and Service:  http://www.wasabisystems.com/