> Something else going wrong could be register preservation, do you
> have more than 1 interrupt type going ? its possible that another
> interrupt is getting triggered... if you clear the mask to early...
> this is definitely a problem, you'll get register corruption.

Yeah, aside from the vbl, codewave's sound lib uses timer0 or timer1
or both, and DMA1. It's updater must be called somewhere in the vbl.
I'm also using timer2 as a general purpose timer.

Where did you get the specs from?  I remember reading that the GBA + (GBA
SP) would get a hardware upgrade (possibly) but I really thought that was
a rumor.

>
> Q:  Are there any hardware differences?
> A:  The GBA SP has exactly the same technical specifications as the GBA.
> The packaging and form factor are different, as well as the LCD screen
> and batteries. There are no hidden registers nor new coprocessors, nor
> extra buttons. From a developer's point of view, the SP is the same as
> the GBA.
>
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>

I heard that the flash cards won't be working on the SP. I don't know how
much of this is true, but maybe someone else can confirm this ?

-----Oorspronkelijk bericht-----
Van: Aaron Pendley [mailto:aaron@...]
Verzonden: woensdag 8 januari 2003 19:15
Aan: gbadev@yahoogroups.com
Onderwerp: Re: [gbadev] GBA SP



Q:  Are there any hardware differences?
A:  The GBA SP has exactly the same technical specifications as the GBA.
The packaging and form factor are different, as well as the LCD screen
and batteries. There are no hidden registers nor new coprocessors, nor
extra buttons. From a developer's point of view, the SP is the same as
the GBA.





Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

Sprites can ONLY be downloaded to OAMRAM in the vblank... you could
enable the forceblank bit in the display till the sprites are
downloaded. This though is NOT recommended ;) I would make sure your
sprite download call is the primary call.

Something else going wrong could be register preservation, do you
have more than 1 interrupt type going ? its possible that another
interrupt is getting triggered... if you clear the mask to early...
this is definitely a problem, you'll get register corruption.

andi


--- In gbadev@yahoogroups.com, "j_north2002 <johnny_north@h...>"
<johnny_north@h...> wrote:
> I'm creating a tetris game that uses a 180 entry array of ints to
> represent a 10 x 18 play area. I'm experiencing a bug where this
> array is being filled with garbage data. Until recently, I could
not
> determine what operation was causing the bug as it seemed to occur
> only very rarely, and under no consistent circumstances. I found
> though that it was happening consistently in the game's demo mode
at
> the same place and time (the demo has no random seed and performs
the
> same every time). I started stepping through the code using gdb,
but
> the error doesn't happen during debugging.
>
> I removed the vbl interrupt for drawing, instead calling Draw()
once
> per game loop. The error doesn't seem to occur now!
>
> Could this be a symptom of doing too much in the vbl, or could this
> interrupt cause corruption of a write to a 180 entry array?
>
> Thanks for the help!

Am I the only one that's worried about how uncomfortable it looks like it'll
be to play.

the bit you hold is pretty much square and the d-pad and a/b buttons seem
very close together.  Just pretended to play it using a cd case and my
wrists seemed to turn upwards (i.e index finger pointed away from me)
whereas with the current one, your hands curve round the back of the unit
more and are further apart.

Am I talking rubbish?  It could be very similar to the GB Color I guess but
haven't one to hand (or that good an imagination - it's 7pm and I've been in
the office for 10 hours already).


   -----Original Message-----
   From: Aaron Pendley [mailto:aaron@...]
   Sent: 08 January 2003 18:15
   To: gbadev@yahoogroups.com
   Subject: Re: [gbadev] GBA SP



   Q:  Are there any hardware differences?
   A:  The GBA SP has exactly the same technical specifications as the GBA.
   The packaging and form factor are different, as well as the LCD screen
   and batteries. There are no hidden registers nor new coprocessors, nor
   extra buttons. From a developer's point of view, the SP is the same as
   the GBA.



   Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

Q:  Are there any hardware differences?
A:  The GBA SP has exactly the same technical specifications as the GBA.
The packaging and form factor are different, as well as the LCD screen
and batteries. There are no hidden registers nor new coprocessors, nor
extra buttons. From a developer's point of view, the SP is the same as
the GBA.

So now that Nintendo has announced the GBA SP, and thus making the rumor
of the GBA + come true, what about the CPU upgrade?  Haven't seen anything
supporting that.

-Rory

I'm creating a tetris game that uses a 180 entry array of ints to
represent a 10 x 18 play area. I'm experiencing a bug where this
array is being filled with garbage data. Until recently, I could not
determine what operation was causing the bug as it seemed to occur
only very rarely, and under no consistent circumstances. I found
though that it was happening consistently in the game's demo mode at
the same place and time (the demo has no random seed and performs the
same every time). I started stepping through the code using gdb, but
the error doesn't happen during debugging.

I removed the vbl interrupt for drawing, instead calling Draw() once
per game loop. The error doesn't seem to occur now!

Could this be a symptom of doing too much in the vbl, or could this
interrupt cause corruption of a write to a 180 entry array?

Thanks for the help!

http://www.ncsx.com is importing them for a reasonably inflated 130$.
I've purchased many items off them before, not only is the shipping
decent, but there is always someone on the phone after a ring or two to
answer questions. I'd recommend them if you're that impatient :P

-Armen
armen@...

On Tuesday, January 7, 2003, at 01:17 PM, Kurt Mahan wrote:

> Does anybody know of any US places that will be importing the GBA SP?
>
> Thanks!
>
> Kurt
> kmahan@...
>
>
<image.tiff>
>
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.

Under the terms of the contract, the materials do belong to the
publisher and work will continue on the book--just with an
author they have more confidence in being able to crank it
out.

In a way it is a relief.  I have been trying to work on the book
as well as struggle with finding work in the wake of the dot-bomb
situations.  Fortunately some old colleagues and I have formed
a new company developing games for connected handheld devices--
currently that means cell phones, but could go into connected PDAs
in the future.  Losing the book actually gives me more time to focus
on this venture.

I'll still be here on the list, and will still be doing some stuff on the
GBA, but I've lost the pressures of having deadlines I can't meet.

- John

The book cover has been up there since the Summer as a book forthcoming.
This decision was made in just the past week and I'm sure that Premier
simply hasn't had the chance to re-render the cover and update their website.

I may be on the cover as a co-author or listed as a contributor, depending
upon how much of my work the new author uses.

My work was focussed on DevKit Advance.

- John
----- Original Message -----
From: "Steve Garcia" <pureedlizard@...>
To: <gbadev@yahoogroups.com>
Sent: Tuesday, January 07, 2003 7:45 AM
Subject: Re: [gbadev] GBA programming book


>
> Yet your name is still on the cover...If you look at the close up image of
> the cover, its there.
>
> Does the book cover devkits used today, like Devkit Advance, SDT?
>
> >From: "John Seghers" <johnse@...>
> >Reply-To: gbadev@yahoogroups.com
> >To: <gbadev@yahoogroups.com>
> >Subject: Re: [gbadev] GBA programming book
> >Date: Tue, 7 Jan 2003 05:16:03 -0800
> >
> >I was the author of that book.  Unfortunately my progress on it was too
> >slow
> >(missed too many deadlines) and they have pulled it from me and given it to
> >another author.  This has just occured in the last week and I have no idea
> >who the new author is or when the publishing date will be.
> >
> >- John
> >----- Original Message -----
> >From: "George Agnelli" <george@...>
> >To: "Gbadev" <gbadev@yahoogroups.com>
> >Sent: Tuesday, January 07, 2003 1:55 AM
> >Subject: [gbadev] GBA programming book
> >
> >
> > > Has anyone seen a copy of the Premier Press Programming the Gameboy
> >Advance
> > > book yet (http://www.amazon.co.uk/exec/obidos/isbn=1931841780)? Can
> >anyone
> > > recommend it to a relative beginner such as myself? This series of books
> > > seems to get very mixed reviews.
> > >
> > > George
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>

That's really unfortunate, John.  I was looking forward to that book...
:(

So, do you keep rights to what you wrote, or did you sign them over to
the publisher?  If they're yours still, are you thinking about
publishing what you do have?

dennis

On Tuesday, January 7, 2003, at 08:16 AM, John Seghers wrote:

> I was the author of that book.  Unfortunately my progress on it was
> too slow
> (missed too many deadlines) and they have pulled it from me and given
> it to
> another author.  This has just occured in the last week and I have no
> idea
> who the new author is or when the publishing date will be.

Does anybody know of any US places that will be importing the GBA SP?

Thanks!

Kurt
kmahan@...

well at least this time round it's true

http://www.nintendo.com/news/news_articles.jsp?articleID=7901

recased gba with frontlit screen & lithium ion battery

> -----Original Message-----
> From: Jay [mailto:jay@...]
> Sent: 07 January 2003 12:13
> To: gbadev@yahoogroups.com
> Subject: [gbadev] New GBA
>
>
> Sorry if this is old news:
>
> http://www.mcvuk.com/link.asp?newsId=490
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>

Yet your name is still on the cover...If you look at the close up image of
the cover, its there.

Does the book cover devkits used today, like Devkit Advance, SDT?

>From: "John Seghers" <johnse@...>
>Reply-To: gbadev@yahoogroups.com
>To: <gbadev@yahoogroups.com>
>Subject: Re: [gbadev] GBA programming book
>Date: Tue, 7 Jan 2003 05:16:03 -0800
>
>I was the author of that book.  Unfortunately my progress on it was too
>slow
>(missed too many deadlines) and they have pulled it from me and given it to
>another author.  This has just occured in the last week and I have no idea
>who the new author is or when the publishing date will be.
>
>- John
>----- Original Message -----
>From: "George Agnelli" <george@...>
>To: "Gbadev" <gbadev@yahoogroups.com>
>Sent: Tuesday, January 07, 2003 1:55 AM
>Subject: [gbadev] GBA programming book
>
>
> > Has anyone seen a copy of the Premier Press Programming the Gameboy
>Advance
> > book yet (http://www.amazon.co.uk/exec/obidos/isbn=1931841780)? Can
>anyone
> > recommend it to a relative beginner such as myself? This series of books
> > seems to get very mixed reviews.
> >
> > George

I was the author of that book.  Unfortunately my progress on it was too slow
(missed too many deadlines) and they have pulled it from me and given it to
another author.  This has just occured in the last week and I have no idea
who the new author is or when the publishing date will be.

- John
----- Original Message -----
From: "George Agnelli" <george@...>
To: "Gbadev" <gbadev@yahoogroups.com>
Sent: Tuesday, January 07, 2003 1:55 AM
Subject: [gbadev] GBA programming book


> Has anyone seen a copy of the Premier Press Programming the Gameboy Advance
> book yet (http://www.amazon.co.uk/exec/obidos/isbn=1931841780)? Can anyone
> recommend it to a relative beginner such as myself? This series of books
> seems to get very mixed reviews.
>
> George
>
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>

Sorry if this is old news:

http://www.mcvuk.com/link.asp?newsId=490

Has anyone seen a copy of the Premier Press Programming the Gameboy Advance
book yet (http://www.amazon.co.uk/exec/obidos/isbn=1931841780)? Can anyone
recommend it to a relative beginner such as myself? This series of books
seems to get very mixed reviews.

George

That is exactly the purpose of the device.  It seems there are still
some little bugs in the program, the device has to reset several times
before GBA will start to accept data... other than that the first
prototype works great.

I forget to mention, the RESET line on the PIC chip should be tied to
Parallel port pin 1.

Steve Garcia wrote:
> Interesting..is the device capable of sending the data from flash to the
> gba on powerup? If so I'd like to try it out.
>
>
>
>>Happy new year.
>>
>>This is a project I have always want to do, which is a Enhanced MB cable
>>capable of downloading MultiBoot code to GBA without a computer.
>>
>>This perticular design contains a 256kbyte FLASH memory wich is capable
>>of storing a single program.
>>
>>The cable is very similar to the cable from www.godsmaze.org, except the
>>external FLASH memory device.
>>
>>The cable is compatable with the software from www.godsmaze.org.  I also
>>have written a PC side software that is used to download the code to the
>>FLASH memory, I wll post it as soon as I can get it cleaned up.  The
>>software is kind of large and hard to use, if you need it, please send
>>me a note to request for it, since I want to avoid sending it though the
>>mailing list. (The program also includes a version of Multiboot
>>downloading code that does not use 32bit multiplication, could be useful
>>for people who are writing multiboot games who want to speed up the
>>download.)
>>
>>I hope someone can make good use of this software, since I'm not good
>>with webpages, so chances are, I won't put it up on a webpage. This is
>>why I'm releasing it though the mailing list.
>>
>>If anyone want to post this on their website, please feel free to do so.
>>
>>The hardware description is included in the code.  Feel free to ask any
>>question.
>>
>>I probally won't do more work on this since I'm out of time.
>>
>>
>>Tn.
>
>
> [30kb code cropped by mod to save everyones mailboxs ;) ]
>
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
>

On Wednesday 01 January 2003 23:37, Mike Schwartz wrote:
> #define OBJPaletteMem ((volatile u16 *)0x5000200)
>
> Is fastest and proper.  If you don't declare it volatile, the compiler
> may optimize out references to the memory in ways you don't want.  This
> #define doesn't use a memory location, like the third declaration below
> does, and requires one less indirect memory fetch (is faster).

while that statement is true for some cisc-cpus, it doesnt apply to most (if
not all) risc cpus.... ie, risc-cpus always do memory access indirect anyway
and thus for the speed of the generated code it doesnt make a difference
which of the former mentioned methods is used.

however, other than the rest i would really recommend using the third method
(through a -visible in C code- pointer), since this will result in your code
using _one_ pointer for _all_ these acesses (as a sideffect making it easy to
abuse that pointer for doublebuffering or sth) instead of inserting that
pointer into the code everytime its beeing used. (this will happen with gcc
atleast, no idea if ads can optimize this out).

--
           ___ ___ .___________________ .___________ _______.
     c=64 /   |   \|   \__    ___/     \ \_   _____/ \      \ [groepaz]
      gb /    '    \   | |    | /  \ /  \ |    __)_  /   |   \ gp32
     cgb \    .    /   | |    |/    '    \|        \/    |    \ psx
      gba \___|_  /|___| |____|\____|__  /_______  /\____|__  / dc
           -----\/-----'---------------\/--------\/---------\/
        http://www.hitmen-console.org    Hitmen WWW Headquarters
        http://fly.to/hitmen-groepaz     my personal playground
        http://rr.c64.org/silversurfer   home of the RR debugger
        ftp.musoftware.de/pub/groepaz    cc65 dump site

Interesting..is the device capable of sending the data from flash to the
gba on powerup? If so I'd like to try it out.


>Happy new year.
>
>This is a project I have always want to do, which is a Enhanced MB cable
>capable of downloading MultiBoot code to GBA without a computer.
>
>This perticular design contains a 256kbyte FLASH memory wich is capable
>of storing a single program.
>
>The cable is very similar to the cable from www.godsmaze.org, except the
>external FLASH memory device.
>
>The cable is compatable with the software from www.godsmaze.org.  I also
>have written a PC side software that is used to download the code to the
>FLASH memory, I wll post it as soon as I can get it cleaned up.  The
>software is kind of large and hard to use, if you need it, please send
>me a note to request for it, since I want to avoid sending it though the
>mailing list. (The program also includes a version of Multiboot
>downloading code that does not use 32bit multiplication, could be useful
>for people who are writing multiboot games who want to speed up the
>download.)
>
>I hope someone can make good use of this software, since I'm not good
>with webpages, so chances are, I won't put it up on a webpage. This is
>why I'm releasing it though the mailing list.
>
>If anyone want to post this on their website, please feel free to do so.
>
>The hardware description is included in the code.  Feel free to ask any
>question.
>
>I probally won't do more work on this since I'm out of time.
>
>
>Tn.

[30kb code cropped by mod to save everyones mailboxs ;) ]

--- In gbadev@yahoogroups.com, Groepaz <groepaz@g...> wrote:
> On Wednesday 01 January 2003 20:14, Patriek Lesparre wrote:
> > >Will the GBA display controler display the change of HOFS
> > >of a background layer while he is already in HWRITE? For
> > >example, the first 40 y values of a certain scanline will
> > >be displayed with a HOFS value of 0, the next 200 y with
> > >a value of 1?

It is possible.  Read on...

> > >I have tryed it on VisualBoyAdvance, but it doesn't work.
> >
> > I haven't tested it, but I would assume it's possible. The
> > emulator is probably using a line-based renderer

Even on the NES, where some of the games do some real nasty
raster tricks with the background and the emulators try to keep
up, most NES emulators still use line-based rendering code, and
games such as "Pirates" that bankswitch tile data during draw
time don't work.

> > instead of a (slow) pixel-based one, so you won't be able
> > to see the effect.

It's possible to make a hybrid line/pixel based renderer in
much the same way that emulators' sound engines work.  Simply
time-stamp all video register writes, and when drawing the
frame, draw portions of scanlines between register writes.
This keeps full speed for the common case of no mid-scanline
writes, but it goes down to pixel accuracy.  I think the NES
emulator 'Nintendulator' does something similar.

> i wouldnt bet on that... imagine how the display controller
> works and how much memory (ie registers) access it had to do
> if every register-change would have an immediate effect on
> per-pixel basis.

Remember... the registers are local to the display controller.
Look how much mid-scanline register writing was necessary to
get ANYTHING to display on the Atari 2600 console, which had
128 bytes of RAM and about half a scanline's worth of video RAM.

> this might work for some "simple" things, like colors
> (good old "rastersplits" :))

I've known that mid-scanline palette writes on the GBA work
for quite some time.

> but for more complicated stuff, like displaying actual
> character data (which x-offset belongs to), registers
> are probably fetched once per scanline, or atleast not
> per-pixel, but per character-cell.

It's fetched per character-cell.


The long story:

Somebody asked about writing to the GBA's scroll registers for
"text mode" backgrounds (mode 0 bg 0-3 and mode 1 bg 0-1) during
draw time (as opposed to hblank or vblank time).  I've done some
tests on my GBA, by setting up timer 3 as a faux LCD_X register
(speed: 16 MHz; initial count: 65536-1232), and then spinwaiting on
timer 3 and setting the scroll registers.  Here are the results:

Changes to BGSCROLL[0].y during draw have an effect that may not
be familiar to programmers who grew up with graphic systems such
as the NES PPU where writing 16 to the Y scroll register caused
background line 16 to be drawn on the next scanline.  The GBA,
however, considers BGSCROLL[0].y as an *offset* from (0, 0), and
while it draws the background, it *continuously* adds the value
in BGSCROLL[0].y to LCD_Y.  Thus, writing 16 to BGSCROLL[0].y on
scanline 20 will cause an immediate switch to background line 36.

Changes to BGSCROLL[0].x have similar effects, with the same
"offset" scheme.

However, a scroll write does cause an area of background
corruption about 8 pixels wide.  This could be due to a tile
data prefetch within the compositor.  (The NES PPU performs
some prefetch as well.)

One should be able to do a vertical status bar or vertically
split screen with this.  But with the large sprites, four
backgrounds, and windows that the GBA provides, why would a
fellow want to waste CPU or DMA time on this technique?

E-mail me at d_yerrick@... if you want the source or
binary that I used.

--
Damian

Happy new year.

This is a project I have always want to do, which is a Enhanced MB cable
capable of downloading MultiBoot code to GBA without a computer.

This perticular design contains a 256kbyte FLASH memory wich is capable
of storing a single program.

The cable is very similar to the cable from www.godsmaze.org, except the
external FLASH memory device.

The cable is compatable with the software from www.godsmaze.org.  I also
have written a PC side software that is used to download the code to the
FLASH memory, I wll post it as soon as I can get it cleaned up.  The
software is kind of large and hard to use, if you need it, please send
me a note to request for it, since I want to avoid sending it though the
mailing list. (The program also includes a version of Multiboot
downloading code that does not use 32bit multiplication, could be useful
for people who are writing multiboot games who want to speed up the
download.)

I hope someone can make good use of this software, since I'm not good
with webpages, so chances are, I won't put it up on a webpage. This is
why I'm releasing it though the mailing list.

If anyone want to post this on their website, please feel free to do so.

The hardware description is included in the code.  Feel free to ask any
question.

I probally won't do more work on this since I'm out of time.


Tn.

;**********************************************************************
;
; Enhanced GBA multiboot smart cable
; First release, Version 0.1.0, Jan 1, 2003 (Happy new year!)
;
; - Compiled by Tuna (by hand... )
; - Origional C code from godmaze.org
; - You can use this code for watever you wish, but I take no
;   responsability for any injury, damage to property or copywrite
;   breaches.
;
; - 100% compatable to the software from godsmaze.org
; - Extended function includes:
;  - EEprom read/write
;  - External FLASH read/write
;  - 10 bits ADC
;  - Debug serial port output, 57600 baud
; - The external dataFLASH are from Atmel, the code is tested
;   with the 2-mbit device (AT45DB021B)
;
; Connection:
;  Device tested: PIC16F872
;  Device that might work: PIC16F87X (X = 3, 4, 6, 7)
; PortB 0: Printer D0 (pin 2) through 500Ohm resistor
; PortB 1: Printer D1 (pin 3) through 500Ohm resistor
; PortB 2: Printer D2 (pin 4) through 500Ohm resistor
; PortB 3: Printer D3 (pin 5) through 500Ohm resistor
; PortB 4: Printer (pin 11) through 100Ohm resistor
; PortB 5: Printer (pin 16) through 2.2kOhm resistor
; PortB 6: Printer (pin 17) through 2.2kOhm resistor
; PortB 7: Debug port, connect to RX of serial port
; 	 though MAX232 (Optional)
; - Pulling the Debug port to Vcc though 10Kohm resistor start the
;   processor in slave mode, waiting command from computer.
; - Pulling the Debug port to GND though 10Kohm resistor start the
;   processor in mastor mode, sending the stored program to GBA
;
; PortA 0:  Analog port, accept any analog input (Optional)
; PortA 1:  GBA Pin 3
; PortA 2: GBA Pin 2
; PortA 3: Analog V+ reference port, can not be used when
; 	 communicating with GBA. (Optional)
; PortA 4: GBA Pin 5
; PortA 5: GBA Pin 4
;
; GBA Pin 1 - Vcc (3.3V) connected to the Vcc of the entire circuit
; 	 The entire system power is supplied by GBA
; GBA Pin 6 - System ground. (GND)
;
; PortC 3: to /CS pin of the dataFLASH
; PortC 4: to /SO pin of the dataFLASH (Data Input)
; PortC 5: to /SI pin of the dataFLASH (Data Output)
; PortC 6: to /CLK pin of the dataFLASH (Serial Clock)
;
; PortC 0,1,2,7 Not used, maybe someone will find a use for them
; System clock:  3.6864Mhz, same as the design from godsmaze.org
;
;**********************************************************************

	 list     p=16F872      ; list directive to define processor
	 include  <p16F872.inc>   ; processor specific variable definitions

	 __CONFIG   _CP_OFF & _WDT_OFF & _PWRTE_ON & _XT_OSC & _LVP_OFF & _BODEN_OFF

;**********************************************************************
#define _DEBUG  PORTB,7
#define _CMD    PORTB,6
#define _HSi    PORTB,5
#define _HSo    PORTB,4

; Reserv PORTA,0 and PORTA,1 for analog signals
#define _SO     PORTA,1
#define _SI     PORTA,2
#define _SC     PORTA,4
#define _SD     PORTA,5

#define _CS     PORTC,6



;********************************************************************** Variable
Command  EQU 0x20
Nibble0  EQU 0x21
Nibble1  EQU 0x22
Nibble2  EQU 0x23
Nibble3  EQU 0x24
_Data  EQU 0x25
Delay  EQU 0x26
sendBy  EQU 0x27
eeAddr  EQU 0x28
Loop  EQU 0x29
DF_ADDR_H EQU 0x2A
DF_ADDR_L EQU 0x2B
Loop2  EQU 0x2C
WordH  EQU 0x22 ; Overlap Nibble1
WordL  EQU 0x24 ; Overlap Nibble3
pAddr  EQU 0x2D
Wtemp  EQU 0x2E

Var1L  EQU 0x30
LengthH  EQU 0x31
LengthM  EQU 0x32
LengthL  EQU 0x33
Var8L  EQU 0x34
VarCH  EQU 0x35
VarCL  EQU 0x36
PosH  EQU 0x37
PosM  EQU 0x38
PosL  EQU 0x39

CRC32H  EQU 0x3b
CRC32L  EQU 0x3c
Var30  EQU 0x3f
SeedA  EQU 0x40
SeedB  EQU 0x41
SeedC  EQU 0x42
SeedD  EQU 0x43
TempA  EQU 0x44
TempB  EQU 0x45
TempC  EQU 0x46
TempD  EQU 0x47
DataH  EQU 0x48
DataL  EQU 0x49
DebugD  EQU 0x4a
DebugL  EQU 0x4b
;**********************************************************************
	 ORG     0x000           ; processor reset vector
  	 goto    main            ; go to beginning of program

	 ORG     0x004           ; interrupt vector location

;********************************************************************** Main
main:
	 clrf INTCON
	 clrf PORTB
	 bsf STATUS, RP0

	 movlw b'10101000'
	 movwf OPTION_REG
	 clrf PIE1
	 clrf PIE2

	 movlw b'10001110' ; Analog channel setup
	 movwf ADCON1
	 bsf PORTA, 0 ; Analog input
	 bsf PORTA, 3 ; Analog input
				 ; GBA signals
	 bcf _SO  ; O
	 bsf _SI  ; I
	 bsf _SD  ; I
	 bsf _SC  ; I

	 bsf _HSi  ; I
	 bsf _CMD  ; I
	 bcf _HSo  ; O
	 bsf _DEBUG  ; I Use as mode select when starting

	 bcf STATUS, RP0
	 bcf _DEBUG  ; Debug is inactive low

	 movlw b'01000000' ; Analog converter off, Fosc/8 as clock
	 movwf ADCON0
	 bsf _SO  ; Idle high

	 ;************ DEBUG here!
	 ;goto chip_id
	 ;************ END DEBUG JP PT

;************************************************************ Receive a command
from computer
get_cmd:
	 bsf _HSo
	 bsf STATUS, RP0
	 movlw 0x0f
	 iorwf TRISB, F ; Set portB[0:3] to input
	 bcf STATUS, RP0
wait_cmd:
	 btfss _DEBUG
	 goto send_mb_image
	 btfss _CMD
	 goto wait_cmd

	 movfw PORTB
	 andlw 0x0f
	 movwf Command
	 bcf _HSo

wait_cmd_done:
	 btfss _DEBUG
	 goto send_mb_image
	 btfsc _CMD
	 goto wait_cmd_done

	 bsf STATUS, RP0
	 bcf _DEBUG  ; Return to output
	 bcf STATUS, RP0

	 movfw Command
	 addwf PCL, f
	 goto get_cmd  ; CMD 0 - Reserved
	 goto cmd_xfer ; CMD 1 - Transfer
	 goto chip_id  ; CMD 2 - Standard ID
	 goto get_cmd  ; CMD 3 - Reserved
	 goto get_cmd  ; CMD 4 - Reserved
	 goto get_cmd  ; CMD 5 - Reserved
	 goto get_cmd  ; CMD 6 - Reserved
	 goto get_cmd  ; CMD 7 - Reserved
	 goto ee_read  ; CMD 8 - EEprom read
	 goto ee_write ; CMD 9 - EEprom write
	 goto spi_read ; CMD A - SPI read
	 goto spi_write ; CMD B - SPI write
	 goto adc_read ; CMD C - ADC read
	 goto get_cmd  ; CMD D - Reserved
	 goto get_cmd  ; CMD E - Reserved
	 goto EXT_id  ; CMD F - Extended ID
	 goto get_cmd

;************************************************************
; CMD 0 - Transfer
;
cmd_xfer:
	 bsf  STATUS,RP0
	 bcf  _SO 	  ; Output
	 bsf  _SI  	 ; Input
	 bsf  _SD  	 ; Input
	 bsf  _SC  	 ; Input
	 bcf  STATUS,RP0

	 call  get_nibbles    ; Get a word from the computer
waitbh:
	 btfss  _SC
	 goto  waitbh
	 btfss  _SD
	 goto  waitbh

	 bsf  STATUS,RP0
	 bcf  _SD  ; Output
	 bcf  _SC  ; Output
	 bcf  STATUS,RP0

	 call  send_bits ; send 16bits to GBA
	 call  recv_bits  ; receive 16bits from GBA
	 call  put_nibbles  ; put the 16bits to the computer

	 goto  cmd_xfer ; Loop forever
	 goto get_cmd  ; Waste catcher

;************************************************************
; CMD 1 - Chip ID - Return 0x636100
;
chip_id:
         bsf STATUS,RP0
	 movfw TRISB
	 andlw 0xf0  ;PORTB[0:3] AS OUTPUTS
	 movwf TRISB
	 bcf STATUS,RP0

	 bsf _DEBUG  ; Inactive high

	 movlw 0x06
	 call put_nibble_wH
	 movlw 0x03
	 call put_nibble_wL

	 movlw 0x06
	 call put_nibble_wH
	 movlw 0x01
	 call put_nibble_wL

	 movlw 0x00
	 call put_nibble_wH
	 movlw 0x00
	 call put_nibble_wL

	 movlw 'a'
	 call  Send_Debug
	 goto  get_cmd

;************************************************************
; CMD 8 - EEProm data read
;
ee_read:
	 bsf  STATUS, RP0
	 movlw 0xf0
	 andwf  TRISB, F ; PORTB[0:3] AS OUTPUTS
	 bcf  STATUS, RP0

	 movlw 0x40
	 movwf Loop
	 movlw 0x00
	 movwf eeAddr
read_loop:
	 bsf STATUS, RP1 ; Bank 2
	 movfw eeAddr
	 movwf EEADR
	 bsf STATUS, RP0 ; Bank 3
	 bcf EECON1, EEPGD ; Data
	 bsf EECON1, RD ; Read
	 bcf STATUS, RP0 ; Bank 2
	 movfw EEDATA
	 bcf STATUS, RP1 ; Bank 0
	 call par_send_byte
	 incf eeAddr, F
	 decfsz Loop, F
	 goto read_loop
	 goto  get_cmd

;************************************************************
; CMD A - EEProm data write
;
ee_write:
	 call par_read_byte ; Read address first
	 movwf eeAddr
	 call par_read_byte ; Read data next
	 movwf _Data

	 bsf STATUS, RP1 ; Bank 2
	 movfw eeAddr
	 movwf EEADR  ; Address
	 movfw _Data
	 movwf EEDATA  ; Data
	 bsf STATUS, RP0 ; Bank 3
	 bcf EECON1, EEPGD
	 bsf EECON1, WREN
	 bcf INTCON, GIE
	 movlw 0x55
	 movwf EECON2
	 movlw 0xAA
	 movwf EECON2
	 bsf EECON1, WR ; Start writing
	 ;bsf INTCON, GIE ; Don't need it at this point
	 bcf EECON1, WREN
	 bcf STATUS, RP0
	 bcf STATUS, RP1 ; Bank 0
Write_loop:
	 btfss PIR2, EEIF
	 goto Write_loop
	 bcf PIR2, EEIF
	 goto  get_cmd

;************************************************************
; CMD A - SPI DataFLASH read
;  Receive a 2 byte block address
;  Always send out a block of data, for 2Mbit
;  chip, it is 264 bytes
spi_read:
	 bsf STATUS, RP0
	 bcf PORTC, 3 ; SCK set to output
	 bsf PORTC, 4 ; SDI set to input
	 bcf PORTC, 5 ; SDO set to output
	 bcf _CS  ; CS set to output
	 bsf SSPSTAT, CKE ; DataFLASH Sample on falling cycle
	 bcf SSPSTAT, SMP ; PIC sample on falling cycle
	 bcf STATUS, RP0
	 movlw b'00100000' ; SPI master Fosc/4
	 movwf SSPCON  ; SPI clock Idle low
	 bsf _CS  ; CS idle high

	 call par_read_byte ; Read address first
	 movwf DF_ADDR_H
	 call par_read_byte ; Read data next
	 movwf DF_ADDR_L
	 ;***********************  Finished setting up SPI port
	 bsf  STATUS, RP0
	 movlw 0xf0
	 andwf  TRISB, F ; PORTB[0:3] AS OUTPUTS
	 bcf  STATUS, RP0

	 call read_df_setup

	 movlw 2
	 movwf Loop2
	 movlw 8
	 movwf Loop
SPI_read_loop2:
	 call Read_Next
	 call  par_send_byte
	 decfsz Loop, F
	 goto SPI_read_loop2
	 decfsz Loop2, F
	 goto SPI_read_loop2
	 bsf _CS
	 goto  get_cmd

;************************************************************
; CMD B - SPI DataFLASH write
spi_write:
	 bsf STATUS, RP0
	 bcf PORTC, 3 ; SCK set to output
	 bsf PORTC, 4 ; SDI set to input
	 bcf PORTC, 5 ; SDO set to output
	 bcf _CS  ; CS set to output
	 bsf SSPSTAT, CKE ; DataFLASH Sample on falling cycle
	 bcf SSPSTAT, SMP ; PIC sample on falling cycle
	 bcf STATUS, RP0
	 movlw b'00100000' ; SPI master Fosc/4
	 movwf SSPCON  ; SPI clock Idle low
	 bsf _CS  ; CS idle high

	 call par_read_byte ; Read address first
	 movwf DF_ADDR_H
	 call par_read_byte ; Read data next
	 movwf DF_ADDR_L
spi_test:
	 ;***********************  Finished setting up SPI port
	 bsf _DEBUG
	 bcf _DEBUG
	 bcf _CS
	 movlw 0x82  ; Buffer 1 write
	 movwf SSPBUF
	 bcf STATUS, C ; 1
	 rlf DF_ADDR_L, F ; 2
	 rlf DF_ADDR_H, F ; 3
	 nop 	 ; 4
	 nop 	 ; 5
	 nop 	 ; 6
	 nop 	 ; 7
	 nop 	 ; 8
	 nop 	 ; 9
	 movfw DF_ADDR_H ; 10
	 movwf SSPBUF  ; High Address
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop 	 ; 9
	 movfw DF_ADDR_L ; 10
	 movwf SSPBUF  ; Low Address
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 movlw 0  ; 6
	 call Send_SPI_d ; Empty address, no addr within page
	 nop
	 nop
	 movlw 2
	 movwf Loop2
	 movlw 8
	 movwf Loop
SPI_write_loop:
	 call  par_read_byte
	 movwf SSPBUF  ; Send the data out
	 decfsz Loop, F
	 goto SPI_write_loop
	 decfsz Loop2, F
	 goto SPI_write_loop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop 	 ; Last byte delay
	 bsf _CS
	 bcf  _HSo  ; Set _HSo low to indicate busy
SPI_write_wait:
	 bsf _CS
	 bcf _CS
	 movlw 0xd7
	 call Send_SPI_d
	 nop
	 movlw 0
	 call Send_SPI_d ; Read status
	 nop
	 nop
	 nop
	 nop
	 movfw SSPBUF
	 andlw 0x80  ; 7th bit is BUSY/READY
	 btfss STATUS, Z
	 goto SPI_write_wait
	 bsf _HSo  ; High indicate ready, same as get_cmd
Deadlock:
	 bsf _CS
	 goto Deadlock ; Stall the process,
	 goto  get_cmd  ; must use a hard reset to exit
				 ; Required for the computer to sense
				 ; HSo correctly
;************************************************************
; CMD C - ADC read
;
adc_read:
	 bsf  STATUS, RP0
	 movlw 0xf0
	 andwf  TRISB, F ; PORTB[0:3] AS OUTPUTS
	 bcf  STATUS, RP0

	 movlw 0xc7
	 andwf ADCON0, F ; Select channel 0
	 bsf ADCON0, ADON
	 call long_delay ; Wait at least 20uS for 5V
	 bsf ADCON0, GO
Wait_ADC:
	 btfsc ADCON0, GO
	 goto Wait_ADC
	 movf ADRESH, W
	 call par_send_byte
	 bsf STATUS, RP0
	 movf ADRESL, W
	 bcf STATUS, RP0
	 call par_send_byte
	 bcf ADCON0, ADON ; ADC off
	 goto get_cmd

;************************************************************
; CMD F - Extended Chip ID - Return 0x0001
;
EXT_id:
         bsf  STATUS,RP0
	 movfw  TRISB
	 andlw  0xf0  ;PORTB(0-3) AS OUTPUTS
	 movwf TRISB
	 bcf  STATUS,RP0

	 movlw  0x0
	 call  put_nibble_wH
	 movlw  0x0
	 call  put_nibble_wL
	 movlw  0x0
	 call  put_nibble_wH
	 movlw  0x1
	 call  put_nibble_wL
	 goto  get_cmd


;************************************************************
; Boot up mode: send MB image from FLASH to GBA
;
send_mb_image:
	 bsf STATUS, RP0 ; Setup DataFLASH first
	 bcf PORTC, 3 ; SCK set to output
	 bsf PORTC, 4 ; SDI set to input
	 bcf PORTC, 5 ; SDO set to output
	 bcf _CS  ; CS set to output
	 bsf SSPSTAT, CKE ; DataFLASH Sample on falling cycle
	 bcf SSPSTAT, SMP ; PIC sample on falling cycle
	 bcf _DEBUG  ; Return to output
	 bcf STATUS, RP0

	 movlw b'00100000' ; SPI master Fosc/4
	 movwf SSPCON  ; SPI clock Idle low
	 bsf _CS  ; CS idle high
	 bsf _DEBUG  ; Inactive high Use as serial out
	 call long_delay
	 movlw 'S'
	 call Send_Debug ; Mark Starting point
	 movlw ':'
	 call Send_Debug ; Again...

	 movlw 0x62  ; Send 0x6202, Attention
	 movwf WordH
	 movlw 0x02
	 movwf WordL  ; Get GBA attention
	 call xfer_word ; Check if returning 0x7202

	 movf WordH, W
	 call Send_Debug
	 movf WordL, W
	 call Send_Debug ; Send out debug

	 movlw 0x72
	 xorwf WordH, F
	 btfss  STATUS, Z
	 goto get_cmd  ; Error, try again
	 movlw 0x02
	 xorwf WordL, F
	 btfss STATUS, Z
	 goto get_cmd  ; Error, try again

	 movlw 0x61  ; Send 0x6100, header data
	 movwf WordH
	 clrf WordL
	 call xfer_word

	 movlw 0x60
	 movwf Loop  ; Send 60 words of data
	 clrf DF_ADDR_H
	 clrf DF_ADDR_L ; Start from 1st page
	 call read_df_setup
	 nop
	 nop
	 nop
	 nop 	 ; Wait for the first byte to be ready
	 nop
	 nop
load_loop1:
	 call Read_Next
	 movwf WordL
	 ;call Send_Debug ; Also debug
	 call Read_Next
	 movwf WordH
	 ;call Send_Debug ; Send out debug

	 call xfer_word
	 decfsz Loop, F
	 goto load_loop1

	 movlw 0x62  ; Sending command
	 movwf WordH
	 movlw 0x02
	 movwf WordL
	 call xfer_word

	 ;movf WordH, W
	 ;call Send_Debug
	 ;movf WordL, W
	 ;call Send_Debug ; Send out debug

	 movlw 0x63  ; Send encryption value
	 movwf WordH
	 movlw 0xc1
	 movwf WordL
	 call xfer_word

	 ;movf WordH, W
	 ;call Send_Debug
	 ;movf WordL, W
	 ;call Send_Debug ; Send out debug

	 movlw 0x63  ; Get encryption value
	 movwf WordH
	 movlw 0xc1
	 movwf WordL
	 call xfer_word

	 ;movf WordH, W
	 ;call Send_Debug
	 ;movf WordL, W
	 ;call Send_Debug ; Send out debug

	 movf WordL, W
	 movwf SeedC  ; Seed of encryption
	 movlw 0xc1
	 movwf SeedD
	 movlw 0xff
	 movwf SeedB  ; Setup the Encryption seed with 0xFFFFxxC1
	 movwf SeedA  ; Note, xx is given by GBA
	 movf WordL, W
	 addlw 0x0F
	 movwf Var1L
	 movlw 0x64
	 movwf WordH
	 movf Var1L, W
	 movwf WordL
	 call xfer_word ; Send back Var1L | 0x6400

	 ;*********************** Read the length of the program
	 bsf STATUS, RP1 ; Bank 2
	 movlw 0x00
	 movwf EEADR
	 bsf STATUS, RP0 ; Bank 3
	 bcf EECON1, EEPGD ; Data
	 bsf EECON1, RD ; Read
	 bcf STATUS, RP0 ; Bank 2
	 movlw LengthH
	 movwf FSR
	 movf EEDATA, W ; LengthH
	 movwf INDF
	 incf EEADR, F
	 bsf STATUS, RP0 ; Bank 3
	 bcf EECON1, EEPGD ; Data
	 bsf EECON1, RD ; Read
	 bcf STATUS, RP0 ; Bank 2
	 incf FSR, F
	 movf EEDATA, W ; LengthM
	 movwf INDF
	 incf EEADR, F
	 bsf STATUS, RP0 ; Bank 3
	 bcf EECON1, EEPGD ; Data
	 bsf EECON1, RD ; Read
	 bcf STATUS, RP0 ; Bank 2
	 incf FSR, F
	 movf EEDATA, W ; LengthL
	 movwf INDF
	 bcf STATUS, RP1 ; Bank 0


	 movf LengthL, W
	 movwf TempD
	 movf LengthM, W
	 movwf TempC
	 movf LengthH, W
	 movwf TempB

	 movlw 0xc0
	 subwf TempD, F
	 btfsc STATUS, C
	 goto SkipDecCarry
	 movf TempC, F
	 btfsc STATUS, Z
	 decf TempB, F
	 decf TempC, F ; Length - 0xc0
SkipDecCarry:
	 bcf STATUS, C
	 rrf TempB, F
	 rrf TempC, F
	 rrf TempD, F
	 bcf  STATUS, C
	 rrf TempB, F
	 rrf TempC, F
	 rrf TempD, F ; Length >> 2

	 movlw 0x34
	 subwf TempD, W
	 movwf WordL
	 movf TempC, W
	 btfss STATUS, C
	 decf TempC, W
	 movwf WordH  ; Length - 0x34
	 call xfer_word

	 movf WordL, W
	 movwf Var8L

	 movlw 0xff
	 movwf VarCH
	 movlw 0xf8
	 movwf VarCL
	 movlw 0xc0
	 movwf PosL
	 clrf PosM
	 clrf PosH
	 ;----------------------------------------------------
Send_data_loop:
	 call Read_Next
	 movwf DataL
	 movwf CRC32L
	 call Read_Next
	 movwf DataH
	 movwf CRC32H

	 ;movf DataL, W
	 ;call Send_Debug
	 ;movf DataH, W
	 ;call Send_Debug ; Send out debug

	 call Do_CRC

	 ;movf SeedA, W
	 ;call Send_Debug
	 ;movf SeedB, W
	 ;call Send_Debug ; Send out debug
	 ;movf SeedC, W
	 ;call Send_Debug
	 ;movf SeedD, W
	 ;call Send_Debug ; Send out debug

	 btfsc PosL, 1
	 goto Skip_Encrypt
				 ;Compute Encryption seed, which is
				 ; (seed * 0x6F64573) + 1
	 bcf STATUS, C
	 rlf SeedD, W
	 movwf TempD
	 rlf SeedC, W
	 movwf TempC
	 rlf SeedB, W
	 movwf TempB
	 rlf SeedA, W
	 movwf TempA
	 call AddM  ; Bit 0 and 1
	 call Shift
	 call Shift
	 call Shift
	 call AddM  ; Bit 4
	 call Shift
	 call AddM  ; Bit 5
	 call Shift
	 call AddM  ; Bit 6
	 call Shift
	 call Shift
	 call AddM1  ; Bit 8
	 call Shift
	 call Shift
	 call AddM1  ; Bit 10
	 call Shift
	 call Shift
	 call Shift
	 call AddM1  ; Bit 13
	 call Shift
	 call AddM1  ; Bit 14
	 call Shift
	 call Shift
	 call Shift
	 call Shift
	 call AddM2  ; Bit 18
	 call Shift
	 call Shift
	 call Shift
	 call AddM2  ; Bit 21
	 call Shift
	 call AddM2  ; Bit 22
	 call Shift
	 call Shift
	 call AddM2  ; Bit 24
	 call Shift
	 call AddM3  ; Bit 25
	 call Shift
	 call AddM3  ; Bit 26
	 call Shift
	 call AddM3  ; Bit 27
	 call Shift
	 call Shift
	 call AddM3  ; Bit 29
	 call Shift
	 call AddM3  ; Bit 30
	 incf SeedD, F
	 btfsc STATUS, Z
	 incf SeedC, F
	 btfsc STATUS, Z
	 incf SeedB, F
	 btfsc STATUS, Z
	 incf SeedA, F ; Increase Seed by 1

	 comf PosL, W
	 movwf WordL
	 comf PosM, W
	 movwf WordH
	 comf PosH, W  ; Use temp B & A for the client_pos
	 movwf TempB  ; of next xfer
	 movlw 0xfd  ; ~0x20
	 movwf TempA
	 incf WordL, F
	 btfsc STATUS, Z
	 incf WordH, F
	 btfsc STATUS, Z
	 incf TempB, F
	 btfsc STATUS, Z
	 incf TempA, F ; ~client_pos+1

	 movf SeedD, W
	 xorwf DataL, W
	 xorlw 0x6f
	 xorwf WordL, F
	 movf SeedC, W
	 xorwf DataH, W
	 xorlw 0x64
	 xorwf WordH, F
	 goto  Send_data_16
Skip_Encrypt:
	 movf TempB, W
	 xorwf SeedB, W
	 xorwf DataL, W
	 xorlw 0x65
	 movwf WordL
	 movf TempA, W
	 xorwf SeedA, W
	 xorwf DataH, W
	 xorlw 0x64
	 movwf WordH
Send_data_16:
	 call xfer_word ; Send out the computed data

	 movlw 0x02
	 addwf PosL, F  ; client_pos += 2
	 btfsc STATUS, C
	 incf PosM, F
	 btfsc STATUS, Z
	 incf PosH, F

	 movf PosL, W
	 xorwf LengthL, W
	 btfss STATUS, Z
	 goto Send_data_loop

	 movf PosM, W
	 xorwf LengthM, W
	 btfss STATUS, Z
	 goto Send_data_loop

	 movf PosH, W
	 xorwf LengthH, W
	 btfss STATUS, Z
	 goto Send_data_loop

	 movlw 'D'
	 call Send_Debug
	 movlw ':'
	 call Send_Debug ; Send out debug
Wait_for_ack:

	 movlw 0x65
	 movwf WordL
	 clrf WordH
	 call xfer_word ; End of transfer

	 movf WordH, W
	 call Send_Debug
	 movf WordL, W
	 call Send_Debug ; Send out debug

	 movlw 0x75
	 xorwf WordL, W
	 btfss STATUS, Z
	 goto Wait_for_ack ; Wait for GBA to ack

	 movlw 0x66
	 movwf WordL
	 clrf WordH
	 call xfer_word

	 movf VarCH, W
	 call Send_Debug
	 movf VarCL, W
	 call Send_Debug ; Send out debug


	 movf Var8L, W
	 movwf CRC32H
	 movf Var1L, W
	 movwf CRC32L
	 call Do_CRC
	 movlw 0xff
	 movwf CRC32H
	 movwf CRC32L
	 call Do_CRC
	 movf VarCL, W
	 movwf WordL
	 movf VarCH, W
	 movwf WordH
	 call xfer_word  ; Send CRC out

	 movf WordH, W
	 call Send_Debug
	 movf WordL, W
	 call Send_Debug ; Send out debug

	 movf VarCH, W
	 call Send_Debug
	 movf VarCL, W
	 call Send_Debug ; Send out debug

	 movf Var8L, W ; Final CRC value for debug
	 call Send_Debug
	 movf Var1L, W
	 call Send_Debug ; Send out debug
	 goto get_cmd


Shift:
	 bcf STATUS, C
	 rlf TempD, F
	 rlf TempC, F
	 rlf TempB, F
	 rlf TempA, F
	 return

AddM:
	 movf TempD, W
	 addwf SeedD, F
	 bcf STATUS, Z ; If incf is not executed,
	 btfsc STATUS, C ; Make sure carry does not goes on
	 incf SeedC, F
	 btfsc STATUS, Z
	 incf SeedB, F
	 btfsc STATUS, Z
	 incf SeedA, F ; Byte D
AddM1:
	 movf TempC, W
	 addwf SeedC, F
	 bcf STATUS, Z
	 btfsc STATUS, C
	 incf SeedB, F
	 btfsc STATUS, Z
	 incf SeedA, F ; Byte C
AddM2:
	 movf TempB, W
	 addwf SeedB, F
	 btfsc STATUS, C
	 incf SeedA, F
AddM3:
	 movf TempA, W
	 addwf SeedA, F
	 bcf STATUS, C
	 return

Do_CRC:
	 ;******* Compute CRC
	 movlw 0x10
	 movwf Loop  ; Process 16 bits
CRCLoop:
	 movf VarCL, W
	 xorwf CRC32L, W
	 movwf Var30  ; var_30 = var_C^CRCTemp
	 bcf STATUS, C
	 rrf VarCH, F
	 rrf VarCL, F ; Var_C = Var_C >> 1
	 bcf STATUS, C
	 rrf CRC32H, F
	 rrf CRC32L, F ; CRCTemp = CRCTemp >> 1
	 btfss Var30, 0
	 goto Skip_reset
	 movlw 0xa5
	 xorwf VarCH, F
	 movlw 0x17
	 xorwf VarCL, F
Skip_reset:
	 decfsz Loop, F
	 goto CRCLoop
	 return

;************************************************************
; subroutine: send nibble bits to GBA,
;
send_nibble_bits:
	 btfss  INDF,0    ;6    ;38
	 goto  Set0_Lo  ;7    ;39
         bsf   _SD    	 ;8    ;40 (*)
	 goto  Set1   	 ;9    ;41
Set0_Lo:
	 bcf  _SD    	 ;9   (**) <-- ;41
	 nop         	 ;10   ;42
Set1:
	 btfss  INDF,1    ;11 <-- ;43
	 goto  Set1_Lo  ;12  ;44
	 nop     	 ;13  ;45
	 nop     	 ;14  ;46
	 nop     	 ;15  ;47
	 bsf   _SD    	 ;16 (*) ;48
	 goto  Set2   	 ;17  ;49
Set1_Lo:
	 nop         	 ;14 <-- ;46
	 nop     	 ;15  ;47
	 bcf   _SD    	 ;16 (*) ;48
	 nop         	 ;17
  	 nop         	 ;18
Set2:
	 btfss  INDF,2    ;19 <--
         goto  Set2_Lo  ;20
 	 nop         	 ;21
 	 nop         	 ;22
	 nop         	 ;23
	 bsf   _SD    	 ;24 (*) ;56
	 goto  Set3   	 ;25
Set2_Lo:
	 nop         	 ;22 <--
	 nop         	 ;23
	 bcf   _SD    	 ;24 (*) ;56
	 nop     	 ;25
	 nop         	 ;26
Set3:
	 btfss  INDF,3 	 ;27 <--
	 goto  Set3_Lo  ;28
	 nop         	 ;29
	 nop         	 ;30
	 nop         	 ;31
	 bsf   _SD    	 ;32 (*) ;64
	 return      	 ;33
Set3_Lo:
	 nop         	 ;30 <--
	 nop         	 ;31
	 bcf   _SD    	 ;32 (*) ;64
	 return     	 ;33

;************************************************************
; subroutine: send a 16 bit word to GBA, stored in Nibble
;
send_bits:
	 movlw  Nibble3
	 movwf  FSR

	 movfw  PORTA
	 andlw  B'11001111'  ; raise SC & SD (Active low)
	 movwf  PORTA      ; SC- Clock, SD- Data
	 nop
	 nop
	 nop
	 nop
	 nop                  	 ;3 -2
	 call  send_nibble_bits ;4
	 decf  FSR,f      	 ;35
	 call  send_nibble_bits  ;36
	 decf  FSR,f         	 ;67
	 call  send_nibble_bits  ;68
	 decf  FSR,f        	 ;99
	 call  send_nibble_bits  ;100
		                      	 ;131
	 nop 	 ;(3)
	 nop              ;(4)  keep the last bit
	 nop              ;(5)
	 nop              ;(6)
	 nop              ;(7)
	 bsf   _SD         ;(8) STOP bit

	 return

;************************************************************
; subroutine: send and receive a 16 bit word to GBA,
; 	 stored in Word
xfer_word:
	 bsf  STATUS,RP0
	 bcf  _SO 	  ; Output
	 bsf  _SI  	 ; Input
	 bsf  _SD  	 ; Input
	 bsf  _SC  	 ; Input
	 bcf  STATUS,RP0
	 swapf Nibble3, W ; [4:7] Word L
	 movwf Nibble2
	 swapf Nibble1, W ; [8:11] Word H
	 movwf Nibble0

waitbh_W:
	 btfss  _SC
	 goto  waitbh_W
	 btfss  _SD
	 goto  waitbh_W

	 bsf  STATUS,RP0
	 bcf  _SD  ; Output
	 bcf  _SC  ; Output
	 bcf  STATUS,RP0

	 call  send_bits ; send 16bits to GBA
	 call  recv_bits  ; receive 16bits from GBA
	 return

;************************************************************
; subroutine: receive nibble bits,
;
recv_nibble_bits:
	 btfss  _SD   	 ;9    ;38
	 goto  rSet0_Lo ;7    ;39
	 bsf   INDF,0 	 ;8    ;40 (*)
	 goto  rSet1  	 ;9    ;41
rSet0_Lo:
	 bcf   INDF,0 	 ;9   (**) <-- ;41
	 nop         	 ;10  ;42
rSet1:
	 btfss  _SD   	 ;11 <-- ;43
	 goto  rSet1_Lo ;12 ;44
	 nop     	 ;13 ;45
	 nop     	 ;14 ;46
	 nop     	 ;15 ;47
	 bsf   INDF,1 	 ;16 (*) ;48
	 goto  rSet2    ;17 ;49
rSet1_Lo:
	 nop         	 ;14 <-- ;46
	 nop     	 ;15   ;47
	 bcf   INDF,1 	 ;16 (*) ;48
	 nop         	 ;17
	 nop         	 ;18
rSet2:
	 btfss  _SD   	 ;19 <--
	 goto  rSet2_Lo ;20
	 nop         	 ;21
	 nop        	 ;22
	 nop         	 ;23
	 bsf   INDF,2 	 ;24 (*) ;56
	 goto  rSet3    ;25
rSet2_Lo:
	 nop         	 ;22 <--
	 nop         	 ;23
	 bcf   INDF,2 	 ;24 (*) ;56
	 nop     	 ;25
	 nop         	 ;26
rSet3:
	 btfss  _SD   	 ;27 <--
	 goto  rSet3_Lo ;28
	 nop         	 ;29
	 nop         	 ;30
	 nop         	 ;31
	 bsf   INDF,3 	 ;32 (*) ;64
	 return      	 ;33
rSet3_Lo:
	 nop         	 ;30 <--
	 nop         	 ;31
	 bcf   INDF,3 	 ;32 (*)  ;64
	 return     	 ;33

;************************************************************
; subroutine: receive 16 bit word from GBA, stored in Nibble
;
recv_bits:
	 bsf  STATUS,RP0
	 bsf _SD  ;SD as input
	 bcf  STATUS,RP0
	 movlw  Nibble3
	 movwf  FSR
	 bcf  _SO  ;ask for 16bits
wait_for_start:
	 btfsc  _SD  ;0
	 goto  wait_for_start ;1
	 nop 	 ;2
	 nop 	 ;3
	 nop 	 ;4
	 nop 	 ;5
	 nop 	 ;6
 	 nop                     ;-------centering waveform ;P
	 nop
	 nop
	 nop
	 nop
	 call  recv_nibble_bits ;7
	 decf  FSR,f
	 call  recv_nibble_bits
	 decf  FSR,f
	 call  recv_nibble_bits
	 decf  FSR,f
	 call  recv_nibble_bits
	 movlw 0x0f  ; Steal 5 cycle to use
	 andwf Nibble0, F
	 andwf Nibble1, F
	 andwf Nibble2, F
	 andwf Nibble3, F ; stop bit -5

	 swapf Nibble2, W ; Steal 4 more cycle to assemble the 16bit word
	 iorwf Nibble3, F
	 swapf Nibble0, W
	 bsf  _SO  ; SO high as ACK
	 iorwf Nibble1, F
wait_ack:
	 btfsc  _SI
	 goto  wait_ack
wait_nack
	 btfss  _SI
	 goto  wait_nack
	 bsf  _SC  ;put _SC high ; EOT
	 return

;************************************************************
; subroutine: write a byte into parallel port, data in W
par_send_byte:
	 movwf  sendBy
	 swapf  sendBy, W
	 call  put_nibble_wH
	 movfw sendBy
	 call  put_nibble_wL
	 return

;************************************************************
; subroutine: read a byte from parallel port, place in W
par_read_byte:
	 call get_nibble_wH
	 movwf sendBy
	 call get_nibble_wL
	 movlw 0x0f
	 andwf _Data, F
	 swapf sendBy, W
	 andlw 0xf0
	 iorwf _Data, W
	 return

;************************************************************
; subroutine: Read from parallel port - 4 nibbles
;
get_nibbles:
	 bsf STATUS, RP0
	 movlw 0x0f  ; PortB[0,3] as input
	 iorwf TRISB, F
	 bcf STATUS, RP0
	 call get_nibble_wH ; [15,12]
	 movwf Nibble0
	 call get_nibble_wL ; [11, 8]
	 movwf Nibble1

	 call get_nibble_wH ; [ 7, 4]
	 movwf Nibble2
	 call get_nibble_wL ; [ 3, 0]
	 movwf Nibble3
	 return

;************************************************************
; subroutine: Write to parallel port - 4 nibbles
;
put_nibbles:
	 bsf STATUS, RP0
	 movlw 0xf0
	 andwf TRISB, F
	 bcf STATUS, RP0

	 movfw  Nibble0
	 call  put_nibble_wH
	 movfw Nibble1
	 call  put_nibble_wL

	 movfw  Nibble2
	 call  put_nibble_wH
	 movfw  Nibble3
	 call  put_nibble_wL
	 return

;************************************************************
; subroutine: Get nibble from parallel port

get_nibble_wH:
	 btfss _HSi 	 ; Wait for_HSi high
	 goto get_nibble_wH
	 Call GetBits
	 bsf _HSo   	 ; Set _HSo high
	 return

get_nibble_wL:
	 btfsc _HSi  	 ; Wait for _HSi low
	 goto get_nibble_wL
	 Call GetBits  ; both W and _Data contain the data
	 bcf _HSo    	 ; set _HSo low
	 return

;************************************************************
; subroutine: Put nibble to parallel port

put_nibble_wH:
	 movwf _Data
	 Call PutBits
	 bsf _HSo
pn_wH:
	 btfss _HSi 	 ;Wait for_HSi high
	 goto pn_wH
	 return

put_nibble_wL:
	 movwf _Data
	 Call PutBits
	 bcf _HSo
pn_wL:
	 btfsc _HSi 	 ; Wait for_HSi low
	 goto pn_wL
	 return

;************************************************************
; subroutine: Put/get data on/to the parallel port
PutBits:
	 movlw 0xf0
	 andwf PORTB, F
	 movfw _Data
	 andlw 0x0f
	 iorwf PORTB, F
	 return

GetBits:
	 movlw 0x0f
	 andwf PORTB, W
	 movwf _Data
	 return

;************************************************************
; subroutine: Delay, 255 cycles 830uS for 3.684Mhz clock
long_delay:
	 decfsz Delay, F
	 goto long_delay
	 return

;************************************************************
; subroutine: Setup DataFlash for transfer at address
;  specified in DF_ADDR.
;
read_df_setup:
	 bsf _CS  ; Reset for next transfer
	 bcf _CS
	 movlw 0xd2  ; Main memory read 0xd2
	 movwf SSPBUF
	 bcf STATUS, C ; 1
	 nop 	 ; 2
	 nop 	 ; 3
	 nop 	 ; 4
	 nop 	 ; 5
	 nop 	 ; 6
	 nop 	 ; 7
	 nop 	 ; 8
	 rlf DF_ADDR_L, W ; 9
	 rlf DF_ADDR_H, W ; 10
	 movwf SSPBUF  ; High Address
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 bcf STATUS, C
	 rlf DF_ADDR_L, W ; 9
	 nop 	 ; 10
	 movwf SSPBUF  ; Low Address
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 movlw 0  ; 6
	 call Send_SPI_d ; Empty address, no addr within page
	 nop
	 nop
	 call Send_SPI_d ; 4 don't care bytes
	 nop
	 nop
	 call Send_SPI_d
	 nop
	 nop
	 call Send_SPI_d
	 nop
	 nop
	 call Send_SPI_d
	 nop
	 movf SSPBUF, W ; EMPTY out buffer
	 call Send_SPI_d ; Init receving
	 clrf pAddr  ; Clear address pt
	 return

;************************************************************
; subroutine: Read the next byte from DataFLASH
Read_Next:
	 movf SSPBUF, W
	 clrf SSPBUF  ; Start the next transfer
	 incf pAddr, F ; Point to next byte
	 nop
	 nop
	 btfss STATUS, Z
	 return
				 ; Read the next page
	 incf DF_ADDR_L, F
	 btfsc STATUS, Z
	 incf DF_ADDR_H, F
	 movwf Wtemp
	 call  read_df_setup
	 movf Wtemp, W
	 return

;************************************************************
; subroutine: SendSPI with delay
Send_SPI_d:  ; Call need 2 cycles
	 movwf SSPBUF
	 nop  ; Total 9 cycles, leaving 1 cycle to load data
	 nop  ;
	 nop
	 nop
	 return  ; return need 2 cycles
;************************************************************
; subroutine: SendSPI with delay, roughly 57600 baud
Send_Debug:
	 movwf DebugD
	 movlw 0x08
	 movwf DebugL
	 bcf _DEBUG  ; Start bit, always low
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
DebugLoop:
	 rrf DebugD, F
	 btfss STATUS, C
	 goto BitSetLow
	 nop
	 bsf _DEBUG
	 goto BitFinish
BitSetLow:
	 bcf _DEBUG
	 nop
	 nop
BitFinish:
	 nop
	 nop
	 nop
	 nop
	 nop
	 nop
	 decfsz  DebugL, F
	 goto DebugLoop
	 nop
	 nop
	 nop
	 nop
	 nop
	 bsf _DEBUG  ; Stop bit, always high
	 return
;************************************************************ End
	 end

#define OBJPaletteMem ((volatile u16 *)0x5000200)

Is fastest and proper.  If you don't declare it volatile, the compiler
may optimize out references to the memory in ways you don't want.  This
#define doesn't use a memory location, like the third declaration below
does, and requires one less indirect memory fetch (is faster).

Happy New Year to everyone!


=> -----Original Message-----
=> From: PurpleDecay@... [mailto:PurpleDecay@...]
=> Sent: Wednesday, January 01, 2003 11:52 AM
=> To: gbadev@yahoogroups.com
=> Subject: [gbadev] accessing memory
=>
=>
=> i've been looking at various gba.h files and have a question
=> about setting up
=> memory locations. I've notived 3 main ways people are doing this....
=>
=> #define OBJPaletteMem  0x5000200
=> #define OBJPaletteMem  (u16*)0x5000200
=> u16* OBJPaletteMem = (u16*)0x5000200;
=>
=> Can someone please explain which method is best? Am i right
=> in assuming one
=> method is faster?
=>
=> thanks
=>
=>
=>
=>
=>
=>
=> Your use of Yahoo! Groups is subject to
=> http://docs.yahoo.com/info/terms/
=>
=>
=>

> i've been looking at various gba.h files and have a question about setting
up
> memory locations. I've notived 3 main ways people are doing this....
>
> #define OBJPaletteMem  0x5000200

Using this method you have an integer which still needs to be casted to a
pointer, so if you want to change color entry 0 to white, you'd do:

(u16*)OBJPaletteMem[0] = 0x7FFF;

or

*(u16*)OBJPaletteMem = 0x7FFF;

This method requires some extra typing as you can see, but you can see the
access size in your code.

> #define OBJPaletteMem  (u16*)0x5000200

This is the same as above but saves typing:

OBJPaletteMem[0] = 0x7FFF;

or

*OBJPaletteMem = 0x7FFF;

> u16* OBJPaletteMem = (u16*)0x5000200;

This method will generate the same code as the others in an optimized build
(it might give you some overhead in a non-optimized build). I would say: use
a #define so that your code will be more readable. Up to you whether you
want to do the typecast in your code or within the macro.


Jan-Lieuwe

i've been looking at various gba.h files and have a question about setting up
memory locations. I've notived 3 main ways people are doing this....

#define OBJPaletteMem  0x5000200
#define OBJPaletteMem  (u16*)0x5000200
u16* OBJPaletteMem = (u16*)0x5000200;

Can someone please explain which method is best? Am i right in assuming one
method is faster?

thanks

On Wednesday 01 January 2003 20:14, Patriek Lesparre wrote:
> >i have a question. Will the GBA display controler display the change
> >of HOFS of a background layer while he is already in HWRITE? For
> >example, the first 40 y values of a certain scanline will be displayed
> >with a HOFS value of 0, the next 200 y with a value of 1? Or is it not
> >possible to change the scanline while he is in HWRITE? I have tryed it
> >on VisualBoyAdvance, but it doesn't work. So there is possible that
> >the emulator does not work correctly, my code is wrong or that the
> >display controler don't display such a change.
>
> I haven't tested it, but I would assume it's possible. The emulator is
> probably using a line-based renderer instead of a (slow) pixel-based one,
> so you won't be able to see the effect.

i wouldnt bet on that... imagine how the display controller works and how much
memory (ie registers) access it had to do if every register-change would have
an immediate effect on per-pixel basis. this might work for some "simple"
things, like colors (good old "rastersplits" :)) but for more complicated
stuff, like displaying actual character data (which x-offset belongs to),
registers are probably fetched once per scanline, or atleast not per-pixel,
but per character-cell.

however, the babble is void :=P someone gotta clear this up with testcode eh?
:) (if it _would_ work, changing y-offset within a scanline might give
interisting results aswell hehe)

--
           ___ ___ .___________________ .___________ _______.
     c=64 /   |   \|   \__    ___/     \ \_   _____/ \      \ [groepaz]
      gb /    '    \   | |    | /  \ /  \ |    __)_  /   |   \ gp32
     cgb \    .    /   | |    |/    '    \|        \/    |    \ psx
      gba \___|_  /|___| |____|\____|__  /_______  /\____|__  / dc
           -----\/-----'---------------\/--------\/---------\/
        http://www.hitmen-console.org    Hitmen WWW Headquarters
        http://fly.to/hitmen-groepaz     my personal playground
        http://rr.c64.org/silversurfer   home of the RR debugger
        ftp.musoftware.de/pub/groepaz    cc65 dump site

>i have a question. Will the GBA display controler display the change
>of HOFS of a background layer while he is already in HWRITE? For
>example, the first 40 y values of a certain scanline will be displayed
>with a HOFS value of 0, the next 200 y with a value of 1? Or is it not
>possible to change the scanline while he is in HWRITE? I have tryed it
>on VisualBoyAdvance, but it doesn't work. So there is possible that
>the emulator does not work correctly, my code is wrong or that the
>display controler don't display such a change.

I haven't tested it, but I would assume it's possible. The emulator is
probably using a line-based renderer instead of a (slow) pixel-based one,
so you won't be able to see the effect.

Thank you, that is just what we needed!

>From: Simon B <simonb@...>
>Reply-To: gbadev@yahoogroups.com
>To: gbadev@yahoogroups.com
>Subject: [gbadev] gba dev forum (forum.gbadev.org)
>Date: Wed, 1 Jan 2003 14:30:02 +1100 (EST)
>
>Hello,
>
>Just thought I would start off the new year with a
>mail about a newly opened gba development forum.
>Its located at http://forum.gbadev.org and covers
>most parts of dev (coding,art,music,hardware etc).
>Please pay a visit and feel free to register aswell.
>
>Best wishes
>
>Simon
>gbadev.org
>


_________________________________________________________________
Add photos to your e-mail with MSN 8. Get 2 months FREE*.
http://join.msn.com/?page=features/featuredemail

Hello,

Just thought I would start off the new year with a
mail about a newly opened gba development forum.
Its located at http://forum.gbadev.org and covers
most parts of dev (coding,art,music,hardware etc).
Please pay a visit and feel free to register aswell.

Best wishes

Simon
gbadev.org