[LispM] Outstanding Meroko bugs and a question of my own

[nyef at lisphacker.com]

Brad Parker writes: 

> Wow  
> 
> I guess I should have subscribed to lispm sooner (I just did).   
> 
> Maybe I'm confused - if it's like a CADR, the instruction stream gets advanced
> if bit 14 is set in the popped return address.

It's not quite like a CADR. There's no byte mode, and there's a 
macroinstruction loop chain enable bit in the MCR. 
http://paste.lisp.org/display/12411 shows the details of how Raven does it. 

> I was going to mumble about Meroko and Nevermore not having any
> 'dispatch' handling :-) I've been adding it to Meroko.  Have you already
> added it to nevermore?

Yeah, a while ago. I haven't released a new version in ages, though. 

> (maybe my nevermore sources are hopelessly out of date) 
> 
> I'm guessing that the instruction stream h/w is the same as the CADR, but
> that's a total guess.

As I said above, it's not -quite- the same. And one of the cases is 
sufficiently different that it threw me for some large number of months. 

> I've got Meroko running with SDL and it's trying to get through the
> dispatch diags but I've still got some debugging to do...

SDL? Maybe I'll be able to run it soon then (coLinux doesn't yet have a 
graphical console). Nevermore will almost certainly sprout an X11 interface 
once it gets past the whole set of EXPT tests and extended board diagnostics 
(the X11 interface was given me as a patch, I just haven't installed it 
yet). 

> -brad

For the curious, I just finished checking Nevermore out on the EXPT suite. 
The results are as follows: 

EXPT test status, 2005/Oct/10: 

+---+------------------------------------+------+
| 0 | ALU/OBUS                           | pass |
| 1 | BYTE/BARREL SHIFTER                | pass |
| 2 | I-MEMORY                           |      | (expected to fail)
| 3 | PUSHDOWN LIST (PDL)                | pass |
| 4 | M-MEMORY                           | pass |
| 5 | A-MEMORY                           | pass |
| 6 | MICRO-PROGRAM COUNTER STACK (UPCS) | pass |
| 7 | TAG CLASSIFIER                     | pass | (somewhat spammy)
| 8 | UNMAPPED ACCESSING                 | pass | (somewhat spammy)
| 9 | VMA                                | fail |
| A | GCVF LOGIC                         | pass |
| B | DISPATCH                           | pass |
| C | MACROCODE DECODE                   | pass |
| D | EVENT POSTING                      | fail |
| E | INSTRUCTION STREAM HARDWARE        | pass |
| F | MEMORY CYCLE ABORT                 | fail |
+---+------------------------------------+------+ 

 --Alastair Bridgewater