Hi Dave,

Dave McGuire wrote:
> On Jul 1, 2007, at 10:59 PM, Kim Hawtin wrote:
>>>> so i issued a boot, but i get a fail and retry ...
>>>> need to figure out whats going on with the disk drive now =)
>>>
>>>   Ok.  The drive is most certainly SMD.
>>
>> the emulex controller has three sockets on it, one about 50 pin the
>> other about 14-16 pin. cabled running off to the drive is one 50 and
>> on 14
> 
>   SMD uses a 60-pin daisy-chained & terminated control cable, and a
> 26-pin radially-connected data cable, FYI.

you probably right, I'm running on last nights fuzzy memory here.

>> the drive has a terminator in the second 50 pin socket on the back of
>> the drive, tied to ground.
>   Ok.
>>> What's the make and model of the controller?
>>
>> the card is an Emulex, I didn't find a model number on it first
>> inspection last night.
>>
>> will have another look and hit it with some contact cleaner tonight.
> 
>   Look at the larger (28- or 40-pin) DIP packages on the board...Emulex
> usually put a white sticker on one of them which will contain the model
> number, which will be the first few characters of the long number.  It
> might be something like QD32, SC03, etc. 

this rings a bell.

> We'll need to know that so we
> can figure out what type of DEC controller it emulates, which in turn
> will determine how we'll try to boot the machine.

is this why they are called "Emulex" ?

>>>> do earlier KA630's not know; 'show dev' or 'help'
>>>> in their firmware a the >>> prompt?
>>>
>>>   Nope...In fact I've never seen a KA630 that had such advanced commands
>>> in the ROMs; earlier or otherwise.  The oldest CPU board I've seen with
>>> those capabilities is the KA650.
>>
>> ah, maybe I'm thinking of the ka41 microvax 3100 I have ...?
>> does not matter. just need to learn how to use it then =)
> 
>   Yup.  The KA630 has no such functionality.  You can manually query the
> disk controller's registers, but that's a bit painful for quick
> diagnostic checking.

well i'm probably going to have to learn how to poke each of these things.

>> also has a tk50, four port serial card, deqna (sp?) ethernet
>> and a grant continuity card... (not sure how that works yet)
> 
>   A grant continuity card is required for unused slots, to maintain the
> continuity of the bus grant lines.

it just passes through various signals?

>> do all cards need to be bunched up together?
>> no gaps between cards, ie no empty slots?
>> is there any specific order of cards in the qbus?
>> must the CPU always be on the right? (when looking from the back)
> 
>   Be careful here.  The ordering of the cards in Qbus is (in many cases)
> important, and with the BA23 in particular, the bus is straight-through
> A-B-C-D for the first three slots (for CPU & memory), 

C & D are for CPU and memory info only?

> then changes to a serpentine A-B-A-B layout for the rest of the bus.


an article i found mentioned serpentine Q22/Q22 bus. On the ba23, the
first three slots are Q22/CD, then after that Q22/Q22.

  BA23:
        Q <- Q    Q <- Q <- Q <- Q
        |    ^    |
        V    |    V
    <-  Q    Q <- Q   CD <-CD <-CD

  BA123:
        Q <- Q    Q <- Q <- Q <- Q <- Q
        |    ^    |
        V    |    V
    <-  Q    Q <- Q   CD <-CD <-CD <-CD

copied from here;
  http://h71000.www7.hp.com/wizard/wiz_1149.html

> Google for something like "ba23 slot ordering" or something like that
> for a diagram.  If you can't find one, let me know and I'll dig it up.
> 
>   If you give me a complete list of all the cards in the system, I can
> cook up a functional card placement for you.
> 
>   Now, however, I must sleep, for the caffeine is wearing off.

Thanks Dave, I'll type up my notes when I get home.

cheers,

Kim
-- 
Operating Systems, Services and Operations
Information Technology Services, The University of Adelaide
kim.hawtin@adelaide.edu.au