| Author |
Message |
|
fougere
Advanced Newbie
Joined: 07 Aug 2009
Posts: 56
|
 Posted: 07 Aug 2009 03:19:08 pm Post subject: |
|
|
Hello,
Where does the execution of a TI-83+ restart at if a battery is removed and replaced? $0000, or some other location that the boot code selects like $0053, or what?
Thanks
Last edited by Guest on 07 Aug 2009 03:19:39 pm; edited 1 time in total |
|
| Back to top |
|
|
Igrek
Member
Joined: 23 Aug 2007
Posts: 151
|
| Posted: 07 Aug 2009 05:37:58 pm Post subject: |
|
|
| I believe it normally is $0000 and and $0053 when you booted while holding clear, as you probably know this is used by TI-OS to reset ram. |
|
| Back to top |
|
|
FloppusMaximus
Advanced Member
Joined: 22 Aug 2008
Posts: 472
|
| Posted: 07 Aug 2009 05:49:39 pm Post subject: |
|
|
My best guess (there's no way to be sure without some hardware hacking) is that the PC is initialized to 8000h, with ports 4 and 6 initialized to 77 and 1F respectively.
This seems odd, since all of the non-Flash calculators reset at 0000h - that is the reason page 0 is mapped there, after all! - so perhaps there is still more memory-mapping weirdness that nobody has discovered yet. |
|
| Back to top |
|
|
Igrek
Member
Joined: 23 Aug 2007
Posts: 151
|
| Posted: 07 Aug 2009 06:13:01 pm Post subject: |
|
|
| That's odd, very odd it starts in mapping mode 1? Oo (And I shouldn't believe everything that's on the internet, but that's where I got my false (?) information) |
|
| Back to top |
|
|
fougere
Advanced Newbie
Joined: 07 Aug 2009
Posts: 56
|
| Posted: 07 Aug 2009 09:37:06 pm Post subject: |
|
|
hmmm...
considering this section of code from a very basic operating system: (Benjamin, most of this code is from PongOS)
Code:
.org $0000
jp Boot
;...
.org $0053
jp Boot
;...
Boot:
di
ld sp,$0000 ;initialize stack
xor a
out (0),a ;reset link
dec a
out (1),a ;reset keyboard
ld a,$76
out (4),a ;standard clock interrupt
ld a,$41
out (7),a ;RAM page 1 in bank 3 (basic edition)
in a,(2)
and $80
jr z,BootBE
ld a,$81
out (7),a ;RAM page 1 in bank 3 (silver edition)
dec a
out (8),a ;reset data bus
xor a
out (8),a ;enable data bus
out (5),a ;RAM page 0 in bank 4
out ($20),a ;reset cpu clock
ld a,$14
out ($29),a ;reset LCD clock
xor a
out ($30),a ;clear timers
out ($33),a
out ($36),a
BootBE:
ld a,$18 ;no test mode
call LCDReadyOut
ld a,19 ;*****contrast setting here*****
ld (contrast),a
add a,$D8
call LCDReadyOut
ld a,$40 ;0 vertical shift
call LCDReadyOut
ld a,$05 ;x auto-increment mode
call LCDReadyOut
ld a,$03 ;connect power supply
call LCDReadyOut
call ClrLCDBuffer
call DispLCDBuffer
ld a,%0001001 ;set interrupts (on-key)
out (3),a
ld a,1
ld (calcOn),a
im 1
ei
Hang:
jp Hang
It appears that the execution does not return to $0000 or $0053 after replacing a battery because the contrast darkens from the original setting and does not go back.
If this is the case, then where in the OS code DOES execution return after being rebooted?
I mean, it can't stay in up in the $8000's on page $1F forever. Sometime it MUST re-enter the OS code ...right? so, where?
Maybe I'm missing something about the contrast....?
Any ideas?
Thanks
Last edited by Guest on 07 Aug 2009 09:46:54 pm; edited 1 time in total |
|
| Back to top |
|
|
Igrek
Member
Joined: 23 Aug 2007
Posts: 151
|
| Posted: 08 Aug 2009 04:26:06 am Post subject: |
|
|
| As far as I know page $1F is a page of the OS (in fact it's one of the privileged pages on TI-83+) |
|
| Back to top |
|
|
fougere
Advanced Newbie
Joined: 07 Aug 2009
Posts: 56
|
| Posted: 08 Aug 2009 07:42:54 am Post subject: |
|
|
| From what I gather at WikiTI (here) page $1F is boot code built into the calculator, and it is not a part of the OS considering that it can function even with no OS installed. |
|
| Back to top |
|
|
Igrek
Member
Joined: 23 Aug 2007
Posts: 151
|
| Posted: 08 Aug 2009 07:59:41 am Post subject: |
|
|
Yes, you're right, my bad. I assume this page eventually jumps somewhere into the OS, possibly to page 0. It actually makes sens now  . It jumps to the boot page to check whether there is an OS (an also check whether the user is holding DEL or CLEAR). |
|
| Back to top |
|
|
fougere
Advanced Newbie
Joined: 07 Aug 2009
Posts: 56
|
| Posted: 08 Aug 2009 03:34:21 pm Post subject: |
|
|
| Well, that part does seem to make sense, but it still fails to explain darkening of the contrast. Do you think it's just one of those weird unexplainable things about the calc's hardware and that there's nothing I can do about it??? |
|
| Back to top |
|
|
Igrek
Member
Joined: 23 Aug 2007
Posts: 151
|
| Posted: 08 Aug 2009 04:22:10 pm Post subject: |
|
|
| Does it stop at "hang: jp hang", you might want to add some test code that actually displays something. |
|
| Back to top |
|
|
darkstone knight
Advanced Member
Joined: 07 Sep 2008
Posts: 438
|
| Posted: 08 Aug 2009 04:32:25 pm Post subject: |
|
|
| probaly he's waiting for an interrupt |
|
| Back to top |
|
|
fougere
Advanced Newbie
Joined: 07 Aug 2009
Posts: 56
|
| Posted: 08 Aug 2009 04:56:58 pm Post subject: |
|
|
you're both right.
It does stop at the jp Hang command, but my interrupt routine allows the calc to be turned off and on via the On-key. (As I said, it is a very basic OS)
Here, with this updated code, pressing [clear] manually jumps to $0000, but still the contrast does not return to original setting after battery removal:
Code:
;from 'Hang:' forward
Hang:
ld bc,$FDBF ;group $FD, key $BF [CLEAR]
call AskKey ;returns 1 in A if key requested in BC is pressed
cp 1
jp nz,Hang
jp $0000
Last edited by Guest on 08 Aug 2009 05:19:32 pm; edited 1 time in total |
|
| Back to top |
|
|
|
Z80 & 68k Assembly => z80 & ez80 Assembly
