Subject: Re: what's this machine check mean?
To: None <>
From: der Mouse <mouse@Rodents.Montreal.QC.CA>
List: port-alpha
Date: 04/15/2000 14:04:15
>> I've now had three machine checks, two at 0xfffffc000050f3c8 and one
>> at 0xfffffc000050f3e4.

> One thing to remember about machine checks is that they're not
> precise traps.  Without doing a trap barrier explicitly, you may not
> get the exception until several instructions later.

Oh, the joys of pipelined architectures.

> I'm guessing you have a Multia.  As much as I hate to say it, I'm
> pretty sure your Multia is experiencing Multia Heat Death.

After reading the NetBSD/alpha Multia FAQ, I think your guess is wrong.
I looked at the "bottom" side of the board and was unable to find any
device there with more than two terminals.

What I got is motherboard-plus-CPU.  I added RAM, laid it flat on a
nonconducting surface (melamine-coated chipboard); connected power
supply, serial (after discovering a lovely pinout gotcha), and SCSI;
added a PCI Ethernet I've been hanging onto against the day I get a
PCI-capable machine; set up fans to cool the processor and the 3.3V
regulator next to it; and away I went.

The machine seems to be accurately described in the

	Digital AXPpci33
	Alpha PC Motherboard
	OEM Design Guide

	Order Number: EK-AXPCI-DG.C01

if that's any help.

I now suspect memory.  A fuzzy memory of reading the aforementioned PDF
file prompted me to check the memory.  The docs say it needs 70ns
memory, and my 72-pin memory seems to all be rated at 80ns.  I do have
4MB of 30-pin 70ns memory, and some 4x30pin->72pin adapters, but even
if that works, I don't have enough 30-pin 70ns memory to fill up two
72-pin slots...and this board takes its memory in pairs.  I'm going to
be searching for 70ns memory, and won't worry about the machine checks
unless they keep happening even with 70ns-rated parts.

					der Mouse

		     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B