Re: A85: ROM Browser

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

Re: A85: ROM Browser


On Thu, 24 Sep 1998, Dines Christy Justesen wrote:
> The program you suggest would probably find most functions, but I am not
> sure whether it would be very useful. The problem is that only a small
> part of what is stored in the system mem has been descirbed. A lot of the
> function which has not been desriebed yet uses many of these addresses,
> which makes it very hard to understand what they are doing. Most of the
> Basic functions consist of almost nothing but calls to unknown functions,
> which makes it even harder to understand what they are doing. So unless
> you have an idea of what the function might be doing before you start
> disassembling it, it is almosty impossible to find out what it does.
> 
Couldn't you then just convert all calls in a function to source?  Putting
a comment on the side like "from $123F".  The code would then be more
readable.  Since, as you say, most routines just make calls, you could
find and several functions at once by examining each function one by one.

The program would definitly have some sort of database to keep track of
routine addresses and the names you assign them.

> Using a program like the one you are suggesting would return a lot of
> functions, and it would be very hard to figure out which of them to have a
> look at. Some of the functions the program would find would probably be
> disassembled data, which makes no sense. 
> 
The way I imagine using a program like this is browsing though routines
until you find something that makes sense.  You give it a name and little
by little other routines make more sense.

So maybe the best way would be to define the short routines first, give
them names(labels whatever) and then the bigger routines will be easier to
read.

As far as interpreting data as code, ROM can't be that messy, I mean when
we write asm programs we put data in one place, routines in another..

> My suggestion to anyone who wants to find functions in the ROM is
> therefore to use other methods to finds the functions to disassemble. If
> you know what kind of functions you are looking for it is normally not
> hard to find the area where they are placed.
> 
> Dines
> 
It would also have some sort of search, where you could search for
instructions, like FIND: "call GET_KEY" or some other function you've
named.

> On Thu, 24 Sep 1998, Humberto Yeverino wrote:

> > My previous method seemed a little too simple.  So my current method would
> > be scan for absolute rets and jps, if a conditional jr or jp is
> > encountered, follow it only if it jumps forward, continue searching for
> > absolute jps and rets...
> > I don't think I overlooked anything this time.

I'm sure this question has been asked before but doesn't TI have extensive
documentation on 85 ROM routines and thier locations? I would think
that they would have to.  Has anyone tried asking TI to make that
information public?  It seems like were putting a lot of effort into
finding out what they already know.

-Humberto Yeverino Jr.

"I kick ass for the Lord."

***********************************************************
Home Page:                                               
  http://www.engr.csufresno.edu/~humberto/Home.html      

Ti Page:                                                 
  http://www.engr.csufresno.edu/~humberto/tex.html       

z80 Source Page:                                         
  http://www.engr.csufresno.edu/~humberto/z80source.html 

Official Tyrant Home Page:                              
  http://www.engr.csufresno.edu/~humberto/tyrant.html    

E-mail:                                                  
  humberto@engr.csufresno.edu                            
***********************************************************
Follow-Ups: References: