Re: Apple Rev.C SCSI Jumpers

[dempson@actrix.gen.nz (David Empson)]

Stingray <stingray@webzone.net> wrote:

> David Empson wrote:
> 
> >
> >
> > Nope.  There is at least one digit too many in the address on the first
> > line (it should be six digits, not seven).  There is also only one
> > equals symbol after the "X" in the second line, and a space between the
> > "BRK" and "00" on the first line.  The rest might be correct.
> >
> 
> OK here is acuurately:
> 
> 00/017D: 00 00                  BRK 00
>  A=FF00 X=00C9 Y=0000 S=0171 D=0000 P=B0 K=00 M=0C Q=A8 L=1 m=1 x=1 e=1

Ouch!  That is a crash in the middle of the stack.  The computer should
almost never be executing code there.  It might have done a random leap
into some junk, and happened to end up there.

And from your followup:

> With the card installed in a Rom 01 (previous was a Rom 3) I get:
> 
> 00/0100: BA                      TSX
>  A=FFB0 X=00FF Y=0010 S=0102 D=0000 P=B0 B=00 K=00 M=0C Q=88 L=1 m=1 x=1 e=1

The CPU hit a BRK instruction at 0100 but the instruction was
overwritten before the monitor got to see it, due to a stack overflow.

Interesting - in both cases, the CPU appears to be attempting to execute
code in the neighbourhood of the stack pointer.  This might be a "push
some code onto the stack and execute it" trick.

If so, this might be a compatibility issue between the Rev B firmware
and the IIgs, though I'm not aware of any general fault of this nature.
(I used the rev A firmware in a ROM 0 IIgs for a while, and it seemed to
be fine, apart from missing features and lack of compatibility with
GS/OS.)

The other possibility is corrupted firmware or a hardware fault on the
card, as I suggested before.

> Once again, my own stupidity catching up with me. I compared the card to a
> picture of a Rev C card that was sent to me, ignoring the number on the
> chip. It is in fact a Rev. B. card.
> 
> Where can I buy a Rev. C. EPROM?

You might be able to get one from Alltech Electronics.  Failing that, if
you know anyone near you with access to an EPROM programmer, I'm sure a
copy of the firwmare image could be forwarded to you.  It needs to be
burned into a 27C128 EPROM (16Kx8).  You can't reuse the revision B
EPROM, because rev A/B use a smaller EPROM (same number of pins but half
the capacity).

-- 
David Empson
dempson@actrix.gen.nz
Snail mail: P O Box 27-103, Wellington, New Zealand