[rescue] Axil-255?

[Jeremy RJ Towers]

Why are we assuming that this board follows the 1+ convention?  (As you 
were heading in the direction of)

If we assume it follows the 2 convention, things make a lot more sense.

http://sunsolve.sun.com/handbook_pub/Systems/Sun4c/CPU_Station_2.html

Try following the SS2 convention and see what happens.

When the six  + six are in , what is the memory reported?

I am fairly sure these are 4mb sims so I think you said earlier 32mb was 
shown????  That would make sense if SS" was being followed.

Also, those look like 72 pin Dimm sockets. Maybe try a pair of those, if 
you have them. They will be  72 pin fast page mode parity.

It looks to me like memconf is quite wrong about this board.

HTH

Jeremy

Katherine Strojny wrote:

>Thanks James and Jeremy for the help.
>
>James Lockwood wrote:
>  
>
>>A picture of the motherboard, even a blurry one, would be a terrific
>>    
>>
>help.
>
>Here:
>http://kstrojny.home.att.net/axil255.jpg (725 KB)
>
>James Lockwood wrote:
>  
>
>>What configuration are the memory slots (type, labels, locations)?  I
>>haven't worked with a 255 but the 245 (which was a close relative) used
>>pairs of 72-pin parity SIMMs.  The SIMM pairings may not be physically
>>contiguous on the board.
>>    
>>
>
>Memory appears to have a label on the system board for every two slots.
>You can almost read the labels in the jpg link above if you zoom in.  In
>case you can't:
>
>   Front edge
>J0403        J0404
>J0303        J0304
>J0401        J0402
>J0301        J0302
>   Back edge
>
>Here's an extract of what Jeremy sent from memconf for comparison:
>        $simmbanks = 8;
>        $simmsperbank = 1;
>        @simmsizes = (8,32);
>        @socketstr =
>("J0300","J0301","J0302","J0303","J0400","J0401","J0402","J0403");
>
>(Yes, a little different...)
>
>I was going by the SS1+ board layout and assuming J0403/J0303 was bank 0,
>J0404/J0304 was bank 1, etc.  Maybe that's wrong.  Anyway, with 6 SIMMs on
>each side, it powers up to the PROM monitor.
>
>The SIMMs are 30-pin, 9 ICs per SIMM (can I assume that means parity?).
>Various mfrs.  ICs are labeled as follows:
>
>Toshiba TC514100ASJ-70 JAPAN 9219HCK  (4 of these)
>(V-like logo) 9203 V53C400K70  (3 like this)
>OKI JAPAN M514100A-70SJ 2492277 9A9Z  (2 of these)
>(logo) KOREA 218Y KM41C4000AJ-7  (one left, the other one was the
>overheating culprit)
>OKI JAPAN 9A9Z M514100-80J 91105578  (4 of these)
>
>The first four above are of the form shown on top in the following link.
>The last type has a taller form shown on the bottom. Here's the pic:
>http://kstrojny.home.att.net/simms.jpg (598 KB)
>
>Jeremy RJ Towers wrote:
>
>  
>
>>You should simplify the board as much as possible, the memory is added
>>    
>>
>one at a time. Start with only the working 32mB module.
>
>Sounds like good advice.  I tried to follow it, but unfortunately, I'm
>still trying to figure out how to get the monitor to start with any fewer
>than 12 SIMMs installed (6 on each side).  If I power up with fewer than 12
>(like 10 or 8 or 1 or 2), I get memory errors (indicated by Compose light
>on kbd, and the monitor never comes on).  Like I said, I don't know if I
>can assume the memory banks corresponds to SS1+.  Starting to think they
>don't, or they do but something got fried by the overheating SIMM.  Lemme
>keep working on it.  Not sure if I need to be adding/removing in pairs or
>what.
>
>Any further insight is much appreciated!
>
>-k
>_______________________________________________
>rescue list - http://www.sunhelp.org/mailman/listinfo/rescue