Subject: Re: Some not working packages use Lesstif or XAw3d
To: Peter Teichmann <teich-p@Rcs1.urz.tu-dresden.de>
From: Richard Earnshaw <rearnsha@arm.com>
List: port-arm32
Date: 10/22/1998 13:42:07
> > : Ah, how to detect it? What problems do they cause? Yes I know there was a
> > : discussion about that a while ago, but it seemed that those people had a
> > : lot of problems, and as I have normally no problems I thought that this is
> > : unimportant to me.
> > : 
> > : RiscBSD says "Strongarm rev. 2, bugged STM^ instruction".
> > 
> > AFAIK, that means you _do_ have a buggy chip.
> 
> Hm. But I thought the buffed STM^ instruction is only important for the OS,
> that means the kernel. And there was also an easy fix for that, if I
> remember right.
> 
> So why should it have only effect with programs using Lesstif or Xaw3d?
> Of course, the effect is very strange, as it not only crashes the
> program, but also the complete X session. Also strange is that if the
> programs once started up ok, they work without problems (ok except that
> mozilla is very buggy) but they do not crash.
> 
> Was there a proper explaination for the problems this bug causes? I
> remember some weeks/months ago people were discussing whether and how
> to change their processors, but no one explained what the bug is exactly.
> 

The same silicon also had a bug handling LDM at the end of a page of 
memory.

See http://www.research.digital.com/src/iag/info/dnaext/docs/sa110-errata.p
df
for details.

Richard