Subject: Re: The StrongArm Bug
To: William Gallafent <William.Gallafent@comlab.ox.ac.uk>
From: Richard Earnshaw <rearnsha@arm.com>
List: port-arm32
Date: 05/15/1998 13:31:19
> On Fri, 15 May 1998, Stephen Potts wrote:
>
> > Could someone please tell me if 233MHz StrongArms are affected by the
> > problem?
> >
> > I have no idea what revision I have, or how to find out. All I know is
> > that it was one of the early 233MHz which I got with my Machine in
> > September 1997, direct from Acorn. I only know this because I had
> > ordered a 202MHz but they ran out, so I had to wait until they received
> > the 233MHz from Digital :)
>
> Well, take the lid off the machine and have a look at the chip! In the
> middle it says 'SA110#', where the # should be replaced with a letterm,
> which represents the revision of chip that you're running. Mine says
> 'SA110K', for example!
>
> Apparently, anything later than K does not suffer from this particularly
> nasty bug (AFAICR)
>
> Cheers,
>
> Bill
>
You can work it out without opening the lid -- since it is in the chip
identification number (NetBSD prints it at boot time when it prints out
the name of the CPU -- "SA110 rev X ..."). If it says rev 2 you have a
bad one, if it says rev 3 (there may be later ones as well, but I've not
heard of any) then you are safe.
If you don't have netbsd installed, you can either do a dummy boot from an
install floppy, or I think some of the speed index utilities on RISC OS
report the same information. There's probably even a RISC OS SWI that
will look it up for you.
Richard.