Date   : Tue, 01 Dec 2009 20:28:57 +0000
From   : kevin@... (Kevin Bracey)
Subject: Tube - I/O processor memory questions

John Kortink wrote:
> On Tue, 01 Dec 2009 03:34:55 +0000, Kevin Bracey
> <kevin@...> wrote:
>
>   
>> [...]
>>
>> Well, it could have used the necessary OSWORD 5/6s. Didn't seem /that/ 
>> far-fetched to me having just spent ages faffing around using them 
>> trying to get my command-line across the Tube. (Or are you going to tell 
>> me that the command-line is automatically copied across in some way I've 
>> missed?)
>>     
>
> If you mean a regular OSCLI call : well, yeah. 
Sorry, I meant from the point of view of a language trying to access its 
launching command-line. You have to make OSWORD 5/6 calls to access 
&FFFF00F2 and &FFFF00F3 directly, then go on to make more OSWORD 5/6 
calls to find what they point to. Bleh. Can't see a neater solution, and 
indeed that's what EDIT does (I checked just to reassure myself I wasn't 
missing something).

OSARGS 1,0 is no use, as it only works if you're run from a filing 
system, so only works for a HIxxxx or xxxx language loaded from disc on 
a copro, not for a HIxxxx or xxxx resident in ROM, and such images can 
normally be used in either way. I do have a definitely-disc-based 
xxxx128 build of what I'm doing that could use it, but not sure it's 
worth the conditional assembly grief to save a few bytes on that version.

Really annoying. I guess languages that needed to parse their 
command-line were never that common, so not much attention was paid to 
making it easy.

Kevin