On Wed, Jan 12, 2005 at 10:25:21AM -0500, David Maxwell wrote:
> > Good.  hopefully we can ignore memory issues.
> 
> I have no reason to highlight memory issues here - I just want to point
> out that a single pass of even the extensive tests isn't enough to rule
> out memory issues. Some of them are very situation specific, because a
> flaw on one signal line may only show up under certain data
> transitions...

Or the problem may not be permanent. I've let memtest86+ run for more
than 2 days a suspect systems, it showed only a few errors in the
undred of pass than ran (but all at the same address, and for the same
pattern).

-- 
Manuel Bouyer <bouyer@antioche.eu.org>
     NetBSD: 26 ans d'experience feront toujours la difference
--