[A86] Re: ROM / RAM equates database


[Prev][Next][Index][Thread]

[A86] Re: ROM / RAM equates database




On Mon, 24 Sep 2001 19:10:08 -0700 "David Phillips" <david@acz.org>
writes:
> If you want a database with a web interface, let me know and I'll 
> make one.
> Josh's site has hundreds (thousands?) of calls documented, and you 
> could
> enter them if you like.  Let me know what fields are necessary.  

I've got ~700 calls documented, but not *well* documented...
I started recording more than one-line info, (at
http://rabidcow2.cjb.net/programming/ti86 under "ROM Calls in Depth") I
tried breaking it down into:

_CALL_NAME - short description
 input:
 output:
 modified/side effects:
sample code
extra notes (charts for importat data sets, structures, etc)

Calls that use _ASAP_IND have separate sections for each subfunction

For a cross-calc database, you would want to have a section detailing
differences between calculators and probably addresses.  Maybe have links
to include files sorted by calc & function.  (eventually)

---
> > Then you could also make include files in which the ROM-entrypoints 
 > > are
> > defined with the same names as they are used on the different 
> > platforms.
> > (plotSScreen = gbuf = VIDEO_MEM = ...)
> Due to the similarity of the roms from
> different calcs, it should be
> possible to automate this somewhat. 

It works!
_plotSScreen =
88b8h on ti82
8641h on ti85
8e29h on ti83
9737h on ti73

What I do is look for references to the address in the ti86 rom (or
whichever), take note of values that will probably change between calcs
and search the other rom for a similar pattern.  In some cases it works
great, in others it still requires human intervention.  Still quite handy
if you want to know ram locations across calcs.

I guess the next step is to take an include file as input and spit
another out...

A similar, but different approach might work for identifying similar
code.

-josh