[Coco] NitrOS-9 Level2 and 6309 language label

[Gene Heskett]

On Saturday 08 June 2019 09:21:06 pm Robert Gault wrote:

> Gene,
>
> Sorry to read about your drivewire and lawnmower problems. :(
>
> I strongly agree with your comment "changing anything to Obj6309
> should only be done if the executable actually contains some 6309
> codes." That is exactly what I was doing. I did a search of the Level
> 1&2 commands and modules for H6309. If found and there was actual 6309
> code then I added the equivalent of:
>   ifne H6309
> lg set Obj6309
>   else
> lg set Objct
>   endc
>
> Nothing I've yet found permits me to do that with grfdrv.

Is there in fact, any 6309 code in grfdrv? I do not personally know, 
never haveing waded around in it. Theres also a small chance a 
conditional is wrong and any 6309 code erroniously assembles to a 
forever loop, or even a stack cleanup is wrong when the Obj6309 is in 
effect. That would probably take a carefull audit, krn4, and a ream of 
paper to keep notes on. I don't even know if the src would fit in a #56k 
tsedit buffer.  rbf.mn won't if you leave enough comments in it to tell 
you whats its doing.  There were times, when I was putting the cluster 
stuff back in that K.D. took out & had the audacity to call it a 
Christmas present. He took it out so we couldn't go beyond a 132 
megabyte disk, when I'd have to go back and take out 200 bytes of 
comments so I could add enough more comments to keep track of what I was 
doing.  The lack of comments in rbf.mn today wasn't because I wanted to 
keep it a secret, but because I was down to 20 or fewer bytes of working 
room in "vi"s 56k buffer. I did not have anything but the coco3 to do 
that work with back then.

Later I had a big box amiga, and ced could handle bigger files, but it 
wasn't stable so I didn't ever trust it to work on nitros9. It wasn't 
until I made the jump to linux that editors with gigabyte buffers were 
available. But that was a decade and change later when memory was down 
to 5% of its former cost.

Do we even have the srcs for grfdrv?  Or are they from disasm?
Looks  like original.  Or does in my 2016 copy.

Observations; lots of 6309 codes but maybe there are some ELSE's missing?

IFNE H6309's	256
ELSE's		247
ENDC's		256

So theoretically, there are  9 places where there is no 6809 code. That 
seems odd.

Bears rechecking for missing else's?  Since the count of IFNE H6309's 
matches the count of ENDC's, it stands to reason that there should also 
be an equal number of ELSE's in order to have the 6809 stuff segregated.  
The code is obviously going to change according to the H6309 setting. 
And right at the top of the file, there is nothing to set ty/la to 
Obj6309.  Shouldn't that be decided in the Makefile and fixed by a 
conditional in the .asm file?

Now I could also be totally wrong because the newest version I have here 
is now 2 years old.

I have a script that will download lwtools, toolshed and nitros9, 
building them from fresh pulls of the repo's. Has anything been changed 
like repo's moved etc, that would prevent me from making a fresh copy of 
everything today? 

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page <http://geneslinuxbox.net:6309/gene>