jnc at mercury.lcs.mit.edu
Fri Jun 3 08:39:59 CDT 2016
> From: Rod Smallwood
Ah, sorry about the previous message - hadn't gotten to this one yet! :-)
> 2. This is then followed by a request to type a letter B for boot , L
> for list and so on.
> 3. You can type any letter you like and it always goes to the same
I wonder if there's a case issue - have you tried upper and lower?
> 7. The Unibus adapter appears to pass all of its tests.
> I have two and they both perform the same.
> 8. The next slot to the UBA is an RX211 .. It is set to the standard
> addresses. Again I have two and the result is the same in both cases.
Now that's _very_ interesting indeed. That cuts out a lot of the simple
explanations (broken RX211 or KTJ11 - unless both of one of them are busted).
One thing to check (sorry if you've already checked this) - did the address
switches somehow get inverted (i.e. 'off' where it should be 'on', etc)? I'm
too lazy to work my way through all the inversions in the address recognition
logic (the print set is available online, Google "MP00626") to see whether
'off' means '1' or '0'; I couldn't find a manual for the RX211, which would
say explicitly (the RK02 manual doesn't seem to cover configuration of the
If that's not it, it may be time to break out the 'scope / logic analyzer.
Luckily we do have the RX211 prints. I can whip you up a short program to try
and read the CSR which you can 'toggle' in through ODT, if you've got a
'scope, not a LA. First thing to check would be to see if there are even
cycles getting to the UNIBUS.
> 9. The next three slots have bus grant cards
In the right connector, and in the right way around? (I know, I know, but
at this point, we're grasping at the proverbial...)
> 13. I have tried a couple of other controllers - same result.
As in, other kinds of boards on the UNIBUS? That's also interesting.
> 13. ... I'd like to get both monitor and boot proms back to standard.
I suppose it's _possible_ there's an error in the non-standard PROMs that
somehow buggers up the KTJ11.
According to the "PDP-11/94 System User and Maintainence Guide"
(EK-PDP94-MG-001, available online), pg. 5.49, the KTJ11 has some
maintainence registers which can be set to cause it to do strange things (I'm
just trying to absorb them all now).
One thing to note: 'Diagnostic Mode' is set at power-on, and when set, it
"the UNIBUS is disabled". So powering the machine on with it set to 'Halt'
(so the boot isn't run), and using ODT to poke UNIBUS registers won't work.
You'd have to let the startup run (which presumably leaves the UNIBUS turned
on - or maybe not, something worth checking) and then halt it to use ODT.
Anway, it would be interesting to see what the KTJ11's DCSR (at 17777730)
More information about the cctech