Damiano et al-
	It seems to be the case that all LC 475s are broken.  Aside from
that, the general consensus seems to be that if ls -l fails most of the
time, it's probably a bad mask revision.  There is a mask revision number
(2E71M) after which all LC040s are supposed to work, but that's been sort
of questionable to date.  Unfortunately, it seems like even non-broken
LC040s disagree with some of the instructions used.
	There's some information on it in the archives here:
http://mail-index.netbsd.org/port-mac68k/2003/11/
It starts on 11/13/2003, with my question about "Current level of LC040
support?"  There's more located at:
http://www.macbsd.com/macbsd/LC040-and-BSD.html
although the information there is partially obsoleted by recent
developments (primarily the -msoft-float build).
	The best answer seems to be the softfloat sets, so I second that
thanks!
	By the way, since this could become a frequently-asked question,
I'd be happy to write some brief documentation and/or update FAQ 6.9 with
information about the softfload build.

	-Seth

On Mon, 1 Mar 2004, Damiano Giorgi wrote:

> Bruce ONeel wrote:
>
> > I tried to figure out why my (theoritically) not
> > broken LC040 system seemed to get a randomally trashed stack
> > when it hit an F instruction but I was never sucessful.
>
> How do you tell if your LC040 is broken or not? I've got an LC 475, and
> I've always been wondering whether its LC040 is a "good" or a "bad" one. Is
> there some kind of test you can run? Or maybe looking at some serial number
> on the chip?
>
> Anyway, since I installed the softfloat sets on your site, the old pizza
> box has never ever segfaulted again. Thanks for compiling those sets, and
> for making them available!
>
> Cheers,
>
> Damiano Giorgi
>