[Simh] macro11

[Don North]

On 2020-03-06 04:59 PM, Johnny Billquist wrote:
> On 2020-03-07 01:52, Don North wrote:
>> On 2020-03-06 02:53 PM, Robert Armstrong wrote:
>>>    Another macro11 question - do the apostrophes in the listing indicate
>>> relocatable references, as they do in the DEC version?
>>>
>>>    If so, then I don't think it's assembling this code correctly -
>>>
>>>
>>>         1                                        .TITLE  TEST RELOCATABLE
>>> REFERENCES
>>>         2 000000                                 .ASECT
>>>         3        001000                          .=1000
>>>         4
>>>         5        001234                  X == 1234
>>>         6
>>>         7 001000 005067  001234'                 CLR     X
>>>         8
>>>
>>>    X should be an absolute address, not relocatable.
>>>
>>> Bob
>>>
>>> _______________________________________________
>>> Simh mailing list
>>> Simh at trailing-edge.com
>>> http://mailman.trailing-edge.com/mailman/listinfo/simh
>>
>> The apostrophe trailing on the octal word indicates that it is a computed target
>> address. It is not what you would actually see in the .obj or .bin files.
>>
>> If you want to force all references to be absolute instead of relocatable you
>> can add an ' .ENABL AMA'  pseudo-op before the .ASECT pseudo-op. This will
>> force all relative mode instruction references to become absolute mode. Or
>> of course you could just do ' CLR @#X' for that instruction.
>
> In the end, that is a different thing. Absolute addressing is not what Bob was 
> thinking about here...
>
> Here is what MACRO-11 does:
>
>       1 000000                                  .ASECT
>       2         001000                          .=1000
>       3         001234                  X=1234
>       4
>       5 001000  005067  000230                  CLR     X
>       6
>
> Note that the addressing is still relative. But the actual offset required can 
> be computed by MACRO-11 at compile time.
>
>   Johnny
>
Correct. I provided an answer to a slightly different question, I believe after 
rereading my response.

The .ASECT says put this code/data starting at the given absolute address, but 
it says nothing about how
addresses are computed in instructions. They can still be done using relative 67 
mode or absolute (37) mode.

The tick on the address in the listing does indicate that this address is being 
computed using a relative mode.
The octal word shown is (always) the target address, not the inline values after 
the instruction word.

As I noted above the way to force all instruction address computations to be 
absolute vs relative is to add
the .ENABL AMA  (ISTR that means Address Mode Absolute or some such) pseudo-op 
at the beginning.

Don