Re: Linux & ECC memory

Mark H. Wood (
Fri, 15 Nov 1996 13:09:48 -0500 (EST)

On Thu, 14 Nov 1996, Kenneth Albanowski wrote:

> This is what I'm curious about. Does Linux's NMI code attempt to work
> around some memory problems, or does it just panic? Also, can the glue
> report successful error-correction, as well as failed error-correction?
> (Or is it not useful to know if your memory has/had an error that was
> correctable?)

Oh, yes, you can do strange and wonderful things with a properly
instrumented architecture. You can keep stat.s and do failure-
prediction, for example. Our VMS system here can patch out a bad section
of memory, then dial the phone, send the "syndrome" data to Field
Service, and request that they schedule a visit to replace the failing
board (which has been called out automagically) before anybody here knows

The real question is, what can we do with this sort of information in the
free-software / no-support context? It could be fun to tinker with
anyway, and maybe even profitable eventually, but somebody's gotta come
up with a practical way to use it.

Mark H. Wood, Lead System Programmer MWOOD@INDYVAX.IUPUI.EDU
Those who will not learn from history are doomed to reimplement it.