8 Jan
2009
8 Jan
'09
3:25 p.m.
Pete Turnbull (pete at dunnington.plus.com) wrote:
a 64KB (32KW) MSV11-D with a base address of 000000(8) responds to all
addresses from 000000(8) to 177777(8). It has a jumper to disable the
top page. Maybe you need to do that?
It's certainly possible that could be a problem, but I've been reading the
MSV11-D user's manual (it's on vt100.net) and it doesn't have a jumper to
disable the top page. Actually what it has is a jumper to _enable_ the
bottom 2K of the I/O page, so you can have a tiny bit of extra RAM in
systems without many I/O devices.
The MSV11 manual says only "factory configured modules will not respond to
bank 7 addresses..." and it then goes on to say how you can enable the lower
2K of bank 7 with the aforementioned jumper, but it's quiet about how it
decides what a bank 7 address is (i.e. does it monitor BBS7 or the address
bits?).
The only other jumpers on the MSV11-D are for battery backup and to select
the starting address.
Thanks for the suggestions, but I'm afraid I'm still puzzled...
Bob
8 Jan
8 Jan
8:42 p.m.
On 08/01/2009 15:25, Bob Armstrong wrote:
> Pete Turnbull (pete at dunnington.plus.com)
wrote:
> a 64KB (32KW) MSV11-D with a base address of 000000(8) responds to all
> addresses from 000000(8) to 177777(8).
It's certainly possible that could be a problem,
but I've been reading the
MSV11-D user's manual (it's on vt100.net) and it doesn't have a jumper to
disable the top page. Actually what it has is a jumper to _enable_ the
bottom 2K of the I/O page, so you can have a tiny bit of extra RAM in
systems without many I/O devices.
Ah yes, I mis-remembered - sorry. All the diagrams and labels in the
MSV11-D,E manuals and the various handbooks refer to that jumper as
"enable/disable bank 7" and I tend to forget exactly what that means!
The MSV11 manual says only "factory configured
modules will not respond to
bank 7 addresses..." and it then goes on to say how you can enable the lower
2K of bank 7 with the aforementioned jumper, but it's quiet about how it
decides what a bank 7 address is (i.e. does it monitor BBS7 or the address
bits?).
It monitors BBS7 to determine if access is to the I/O page, and
optionally combines that with BDAL12 to see if it's the lower 2K or
upper 2K.
If you jumper the CPU to enter ODT instead of executing the bootstrap,
what happens? Or if you power things up with with the front panel set
to HALT? If you attempt to execute the BDV11 bootstrap, what happens
when it halts? Does the RUN light ever light, even briefly? What do
the lights on the BDV11 show?
Do these same boards, in exactly this combination, work properly in the
BA11-M? Exactly which version of the M7264 do you have?
--
Pete Peter Turnbull
Network Manager
University of York
9 Jan
9 Jan
10:51 a.m.
Peter Turnbull wrote:
If you jumper the CPU to enter ODT instead of executing the bootstrap,
what happens?
You never get the ODT prompt (the "@"). Never get anything on the
console, FWIW.
Or if you power things up with with the front panel
set to HALT?
Ditto.
If you attempt to execute the BDV11 bootstrap, what
happens
when it halts? Does the RUN light ever light, even briefly? What do
the lights on the BDV11 show?
The BDV11 shows all four LEDs on, and the RUN LED does blink momentarily
but then goes off.
So it's trapping immediately on startup. Either the CPU card is bad
(which I know isn't true) or it can't find any addressable RAM, or it can't
access the console SLU registers, or something is inhibiting the LSI11 from
running at all.
Do these same boards, in exactly this combination, work
properly in the
BA11-M?
Yep. And the same boards except the CPU, work in the same combination in
the BA11-N with a 11/23 CPU card. [It's one of the half sized 11/23 cards -
KDF11-A - without the onboard SLU and ROMs]
Exactly which version of the M7264 do you have?
If you mean the rev level, that I don't know off hand. It's one of the
ones with no actual onboard RAM, though (it has the space for it; the RAM
chips just aren't populated).
Bob
10 Jan
10 Jan
4:57 p.m.
On 09/01/2009 10:51, Bob Armstrong wrote:
Hmm... could mean the console isn't accessible to the CPU (perhaps due
to a memory conflict) but unlikely. Could mean the CPU is stuck in a
"sunset" loop (an endless microcode loop that can't be broken out of, it
continues "until the sun sets", ie until the power is cycled, or at
least until BPOK is cycled).
Then it's not executing the bootstrap; either it's HALTed or stuck in a
microcode loop. The latter might happen if it can't access memory
between 000000 and 000377, I think.
What happens if you move everything down one slot, with the CPU in slot
2, and everything else below that? The RUN light won't work, but
everything else should work just as if it were configured normally. At
least, that would be true with any QBus processor except possibly yours.
You might need to bridge CL1 to CK1 and DK1 to DL1 on slot 2 (or on
the card), to make it work. I'm not sure if your M7264 actually needs
those, I think they're something to do with RAM bias voltage, which
shouldn't matter on a RAMless card.
I believe that makes it a KD11-H, M7264-YA. However, I've seen very few
M7264 LSI-11s, most of the 11/03s I've seen and all the ones I have had
here, have had the later dual-height M7270 KD11-HA LSI-11/2. Although
they share exactly the same microcode, I know the backplane connections
aren't identical, but I don't know exactly what the differences are or
how that might affect things. I'm afraid I'm at a bit of a loss to
explain what's going on here.
--
Pete Peter Turnbull
Network Manager
University of York
Peter Turnbull
wrote:
If you jumper the CPU to enter ODT instead of executing the bootstrap,
what happens?
You never get the ODT prompt (the "@"). Never get anything on the
console, FWIW. Or if you
power things up with with the front panel set to HALT?
Ditto.
If you attempt to execute the BDV11 bootstrap,
what happens
when it halts? Does the RUN light ever light, even briefly? What do
the lights on the BDV11 show?
The BDV11 shows all four LEDs on, and the RUN LED does blink momentarily
but then goes off. So it's trapping immediately on startup. Either
the CPU card is bad
(which I know isn't true) or it can't find any addressable RAM, or it can't
access the console SLU registers, or something is inhibiting the LSI11 from
running at all.
I agree. I think you'll have to look at some of the backplane signals
with a scope or logic probe to see what might be wrong. I'd check BREF
and BDAL16,17 first, because these behave very differently between 11/03
and 11/23. They should all be high (inactive, logic 0) in this system.
If BREF is low, this won't affect an 11/23 or LSI-11/2 but it might
affect a quad LSI-11. Check BEVENT as well; while you're
troubleshooting it wants to be off. Note that if you disable it by
using the switch on the BDV11, that forces it on, which means the only
interrupt line the KD11-F has is also always active. I know it's
edge-triggered, but still...
Do these same
boards, in exactly this combination, work properly in the
BA11-M?
Yep. And the same boards except the CPU, work in the same combination in
the BA11-N with a 11/23 CPU card. [It's one of the half sized 11/23 cards -
KDF11-A - without the onboard SLU and ROMs]
Exactly which version of the M7264 do you have?
If you mean the rev level, that I don't know off hand. It's one of the
ones with no actual onboard RAM, though (it has the space for it; the RAM
chips just aren't populated).
5832
days inactive
5834
days old
3 comments
2 participants
participants (2)
-
bob＠jfcl.com -
pete＠dunnington.plus.com