> On Feb 19, 2022, at 10:51 AM, Noel Chiappa via cctalk <cctalk@classiccmp.org>
> wrote:
> The -11/34 (not the /34A) has something unusual for grant timeouts, but I
> forget the details. I'll look it up.
> On Feb 19, 2022, at 12:11 PM, Fritz Mueller <fri...@fritzm.org> wrote:
> I just did an 11/34 restoration last year, so this is fresh in mind — I think
> you are thinking of the M9302, Noel: a far-side terminator card with
> integrated SACK turnaround?
A few additional details, in case it is helpful:
IIRC, the 11/34 doesn’t have SACK timeout implemented in the CPU cards (the
/34A did add this, however.) So without an M9302 on the far end of the bus,
the CPU could hang in a situation with an unacknowledged grant (which I think
would be somewhat rare, but possible). I believe it is possible to boot an
11/34 at least as far as a ROM boot monitor with a non-SACK (M930 or such)
terminator on the far side.
An 11/34 with a M9302 SACK terminator _will_ hang, though, with a broken grant
chain — the floating grant will trigger the M9302 to continuously assert SACK
locking the bus. “Better” CPUs, which used SACK timeout instead of
turn-around, were more robust this way.
Incidentally, I think the 11/34 was contemporaneous with, and usually found
with, the slightly earlier M9301 ROM terminator on the near side, rather than
the later M9312, though both should work fine.
cheers,
—FritzM.
