Subject: Re: Tracking down memory corruption.
To: Stephen Borrill <netbsd@precedence.co.uk>
From: Peter Seebach <seebs@plethora.net>
List: tech-kern
Date: 09/20/2006 08:35:42
In message <Pine.NEB.4.64.0609201304070.22902@ugly.precedence.co.uk>, Stephen B
orrill writes:
>As before, I'm willing to test out, but any US-only hardwired things will 
>be a prob.

You can change the one line in the driver, or possibly even fix it.

>Which driver(s) are you converting? Just the TDM400P ones or the ISDN ones 
>too?

The only one I can test is wcfxs (the TDM400P), but I'd be happy to work on
the others, if someone else is using them.

I have probably a bug in mine, although it may have gone away.  This is
absolutely some of the worst code I have ever seen.  There's a long and
elaborate loop with reads from hardware registers... And DURING the test for
each read, so far as I can tell, it ignores the result on certain models of
card.

Why is the test there?  I have no clue.

-s