A83: Re: weird ROM...
[Prev][Next][Index][Thread]
A83: Re: weird ROM...
I think you've got a couple different issues here. First, on the seven
pointers plus backups, it sounds like you're talking about the system
monitor vectors. Actually, there are only six of these vectors. The last
two bytes in the table are the page of the current application, and the
context number of the current application. These vectors are used by the
TIOS to pass information to the currently running application. Check my
"A83: Re: ROMcall reference" from the yesterday (the 27th) for more
information on these vectors.
Your second issue seems to be about ROM page swapping. The TI-83 has 16
different ROM pages that can be loaded into the 4000h-7FFFh memory space.
When an asm program is run, the page that contains the ROM entry points is
loaded. All this page contains are the entry points (plus a little sample
prog), so as you noticed, the majority of it is blank. Try stepping through
a call to FPMult or some other TIOS routine and you'll see that it
temporarily switches the ROM page to another page, runs the routine you
called, and then resets the page back to the entry point page.
Hope this helps,
-Dan Englender
----- Original Message -----
From: "yhean" <yhean@lycos.com>
To: "Ti83 Help List" <assembly-83@lists.ticalc.org>
Sent: Sunday, January 28, 2001 7:16 PM
Subject: A83: weird ROM...
>
>
> I discovered this while working on how the TI-OS works with BASIC
> programs, how it runs them, where are stored pointers etc. but over all,
what
> stuff is happening when the TI-OS is running and that the user does
> something (this is vague, I know). All this work is for ZeS, if someone
cares...
> I came out with a 14 bytes handle, where are stored 7 pointers, plus 7
"copies".
> Some of these pointers are addresses where the TI-OS is jumping if
something
> special is done (like pressing the [graph] key). They are like a scheme
that
> the TI-OS must fallow, and they change if you are in the MEM menu or
editing
> a program. So I started searching where in the ROM were stored the
"originals"
> pointers... I found them, but not directly... the weird stuff begins here.
> These pointers were accessible from the VTI debugger, but when I tried to
get
> them from an ASM program, wow!?
> >From $52E9 to almost the end of the ROM, only little $FF's ! What the!?
And
> this phenomenon only occurs when an ASM program is running...
> Anybody has an idea of what's going on here? Is TI hidding some special
parts
> of the ROM? I know that ROM pages are swapped some times, but I don't
> understand this thing...
>
>
> All comments are welcome,
> Jean
>
>
>
>
> Get your small business started at Lycos Small Business at
http://www.lycos.com/business/mail.html
>
References: