Path: utzoo!utgpu!cunews!bnrgate!brtph3!brchh104!brchs1!bnr.ca!rice.edu!sun-spots-request
From: kriso@lombard.dartmouth.edu (Kris Olander)
Newsgroups: comp.sys.sun
Subject: Selftest-#megs=## EEPROM attribute bug???
Keywords: Miscellaneous
Message-ID: <1953@brchh104.bnr.ca>
Date: 18 Mar 91 17:28:00 GMT
Sender: news@brchh104.bnr.ca
Organization: Sun-Spots
Lines: 17
Approved: Sun-Spots@rice.edu
X-Original-Date: Wed, 6 Mar 1991 14:12:55 GMT
X-Sun-Spots-Digest: Volume 10, Issue 57, message 10
X-Note: Submissions: sun-spots@rice.edu, Admin: sun-spots-request@rice.edu

We keep getting some kernel panics on one of our Sparcstation1s running
4.0.3 with some reference to memory errors.  When I look at the EEPROM
parameters, I see 'selftest-#megs=1'.  We have 16 meg of RAM in all our
Sparc1s so I set this EEPROM parameter to 16 and run the monitor level
memory test.  The test always fails with a 'memory alignment error'.  

It turns out, that any other valid RAM value except 'selftest-#megs=1'
fails with the same error message.

I tested this on our other Sun4/80s and came up with the same results.
Can some kind soul explain what's going on here?  Does the Sun4/80 memory
test only test 1 meg of RAM regardless of how much memory is installed?

Is there a better way to test for bad RAM?

Kris Olander
--------------kris.olander@dartmouth.edu----------------