• Just a reminder that providing specifics on, sharing links to, or naming websites where ROMs can be accessed is against the rules. If your post has any of this information it will be removed.
  • Ever thought it'd be cool to have your art, writing, or challenge runs featured on PokéCommunity? Click here for info - we'd love to spotlight your work!
  • PokéCommunity's Easter Egg Hunt is now on! If you're interested in looking for some eggs and earning badges, click here!
  • Our weekly protagonist poll is now up! Vote for your favorite Conquest protagonist in the poll by clicking here.
  • Welcome to PokéCommunity! Register now and join one of the best fan communities on the 'net to talk Pokémon and more! We are not affiliated with The Pokémon Company or Nintendo.
    Register

Help Thread: ASM & Disassembly

Status
Not open for further replies.
Sagiri said:
What's the most idiomatic way to clear all but the least significant byte from a register?

I currently have two thoughts:

Code: 
lsl r1, r1, #0x18
lsr r1, r1, #0x18

Code: 
mov r0, #0xFF
and r1, r0

I like the first because I don't have to use a second register, but the second is a bit clearer, and I'm guessing, faster. Which one should I go with? Or perhaps even a different alternative I haven't considered?
Click to expand...

For the first one, you're losing the significant byte.
You've got the shifts backwards. Clearing the last byte would be done like this:

Code: 
lsr r0, r0, #0x8
lsl r0, r0, #0x8

Say you have FFFF. Shifting it right 8 bits (1 byte) gives you 0FFF, then you shift it left 8 more bits to get FFF0 :)
 
FBI agent said:
For the first one, you're losing the significant byte.
You've got the shifts backwards. Clearing the last byte would be done like this:

Code: 
lsr r0, r0, #0x8
lsl r0, r0, #0x8

Say you have FFFF. Shifting it right 8 bits (1 byte) gives you 0FFF, then you shift it left 8 more bits to get FFF0 :)
Click to expand...

Sorry, I may have been a little unclear. I'm trying to get clear all but the least significant byte. As in, if you have FFFF, you shift up to F000, and then back down to 000F.
 
Sagiri said:
Sorry, I may have been a little unclear. I'm trying to get clear all but the least significant byte. As in, if you have FFFF, you shift up to F000, and then back down to 000F.
Click to expand...
Oh, looks like I've read incorrectly. Yeah, shifting would be faster and better.
 
A friend shared with me this script to check a pokemon's ability:
ldrb r0, [08XXXXXX] (=$02023D6B) offsetof the Pokémon attacking
mov r1, #0x58
mul r0, r1
ldr r1, [08XXXXXX] (=$02023BE4)Offset for all of the 4 Pokémon's datas each Pokémon takes 58 bytes of data according to bulbapedia
add r1, r1, r0
add r1, r1, #0x20
ldrb r1, [r1]
cmp r1, #0x? with the ? as the ability I want to check
Now I do various things based on what the ability is correct? How and where do I place this within the ROM? I am just having trouble in general finding where I am supposed to put my scripts in general so that the game checks for it without replacing existing code
 
Last edited:
GOLDstandard said:
A friend shared with me this script to check a pokemon's ability:
ldrb r0, [08XXXXXX] (=$02023D6B) offsetof the Pokémon attacking
mov r1, #0x58
mul r0, r1
ldr r1, [08XXXXXX] (=$02023BE4)Offset for all of the 4 Pokémon's datas each Pokémon takes 58 bytes of data according to bulbapedia
add r1, r1, r0
add r1, r1, #0x20
ldrb r1, [r1]
cmp r1, #0x? with the ? as the ability I want to check
Now I do various things based on what the ability is correct? How and where do I place this within the ROM? I am just having trouble in general finding where I am supposed to put my scripts in general so that the game checks for it without replacing existing code
Click to expand...

This is probably not what you're looking for. Those offsets are for battle data. In the battle it records various things about each combatant and one of the things happens to be ability.

If you just want to check the Pokemon's ability in overworld it would be different. You'd use the Pokemon decrypter to figure out which ability bit is turned on and depending on that, you would need to use the Pokemon Base stats table to search for which ability it currently has. Then, finally link the ability number to a string.

I just wrote a routine that happens to do all that for you: https://www.pokecommunity.com/posts/8533783/
 
Knight of Duty said:
How do i make my routine read from a table? I think it would be different when hard coded or calling via script.
Click to expand...

You need to know the format of the table before you do anything. Most of the time tables in ROM are very easy, while tables in RAM require some sort of field for Table length. This is because

In ROM, it's rather simple to read pointer and data it look like this:
Code: 
main:
	mov r2, #0x0
loop:
	ldr r0, .TableStart
	lsl r1, r2, #0x2 @data length (I'm assuming it's a pointer, so multiply by 4)
	add r0, r0, r1
	ldr r0, [r0] @new pointer in table
	lsr r1, r0, #0x18
	cmp r1, #0xFF @end of table, you may want to use 00 00 00 00 as the ending bytes, however.
	beq end
	@right now r0 contains a valid pointer in the table
	@you can bl to any other sub routines that utilize it here
	
next:
	add r2, r2, #0x1 @increment table counter
	b loop

In RAM, depending on the contents of the table you may need a table counter. For example, in my Roaming Pokemon data structure, it was not always 100% chance that the last X bytes would be "00" and on top of that I want to be able to see if an Xth one is there (I don't want to be loading garbage data). So a table counter is very nice. However, if you're just writing something like a short table of strings, you can determine the end of the table because 0xFF is the string terminating byte.

Code: 
main:
	ldr r3, .TableLength
	mov r2, #0x0
loop:
	cmp r3, r2 @check if we're finished the table
	beq end
	ldr r0, .TableStart
	lsl r1, r2, #0x2 @data length (I'm assuming it's a pointer, so multiply by 4)
	add r0, r0, r1
	@right now, r0 contains a pointer to the start of the current data
	@you can bl to any other sub routines that utilize it here
	
next:
	add r2, r2, #0x1 @increment table counter
	b loop

You'll notice that I didn't need to check if it was the end of the table because I already know the length of the table. 1 extra byte in RAM space is saving up operation time. Worth? Yes.


Finally, the last and probably the best way is if you already know which index you want.

Code: 
main:
	ldr r0, .TableStart
	mov r1, #0x[data length]
	mov r2, #0x[date index]
	mul r1, r1, r2
	add r0, r0, r1 @start of the new data

Hope this helps. If you have questions, feel free to ask here :)
 
Last edited:
Mickey` said:
Hi,

Is it possible to call a XSE script from an ASM script ?
Click to expand...

Yes. You would make a routine which passes the script to the script executor. That will run your script. A few posts before, someone was attempting something similar, take a look.
 
I think I'm wrong {XD}

I have to simple XSE scripts. I want to call the second script in the first.

XSE script #1 (at 0x900000) :
Spoiler:


XSE script #2 (at 0x910000) :
Spoiler:


And my ASM code (inserted at 0xB00000) :
Spoiler:


Note that the offset 69AE4 in an English Fire Red ROM corresponds with the offset 69B94 in a French Fire Red ROM (I use French ROMs).

The result of the first XSE script should be two messages, "Script 1" then "Script 2", right ? Instead of it, when I talk to a person with the first XSE script, I can hear the Bulbasaur's cry, and there isn't any message.

What's the matter ?
 
Mickey` said:
I think I'm wrong {XD}

I have to simple XSE scripts. I want to call the second script in the first.

XSE script #1 (at 0x900000) :
Spoiler:


XSE script #2 (at 0x910000) :
Spoiler:


And my ASM code (inserted at 0xB00000) :
Spoiler:


Note that the offset 69AE4 in an English Fire Red ROM corresponds with the offset 69B94 in a French Fire Red ROM (I use French ROMs).

The result of the first XSE script should be two messages, "Script 1" then "Script 2", right ? Instead of it, when I talk to a person with the first XSE script, I can hear the Bulbasaur's cry, and there isn't any message.

What's the matter ?
Click to expand...

Oh I see what you mean now. Well, I'm not sure how the limitations work for calling the script from ASM within a script. I'm pretty sure you're going to eventually overwrite parts of the stack. I would rather just use the scripting command "call" and have a return at the end of the second script. On another note, a quick peek at your routine tells me it doesn't look wrong :P
 
FBI agent said:
Oh I see what you mean now. Well, I'm not sure how the limitations work for calling the script from ASM within a script. I'm pretty sure you're going to eventually overwrite parts of the stack. I would rather just use the scripting command "call" and have a return at the end of the second script. On another note, a quick peek at your routine tells me it doesn't look wrong :P
Click to expand...

Yes, of course, the command "call" is really easier to use, it was just to try the ASM code.

So, if I understand, I can call a XSE script only when I'm in an ASM script not called from another XSE script ? x)
 
Mickey` said:
Yes, of course, the command "call" is really easier to use, it was just to try the ASM code.

So, if I understand, I can call a XSE script only when I'm in an ASM script not called from another XSE script ? x)
Click to expand...

Yeah, I don't think you can load and run two instances of the script execution routine at once.
 
Okay ! Another question related to the previous one : is it possible to run an XSE script from an item, changing the "Field Usage Code Pointer" in Item Editor ?
 
Mickey` said:
Okay ! Another question related to the previous one : is it possible to run an XSE script from an item, changing the "Field Usage Code Pointer" in Item Editor ?
Click to expand...

Yes, doing this has been documented in two different ways I know of, by davidjcobb (who uses JPAN's engine) and Darthatron who uses the script executor.
 
Mickey` said:
I think I'm wrong {XD}

I have to simple XSE scripts. I want to call the second script in the first.

XSE script #1 (at 0x900000) :
Spoiler:


XSE script #2 (at 0x910000) :
Spoiler:


And my ASM code (inserted at 0xB00000) :
Spoiler:


Note that the offset 69AE4 in an English Fire Red ROM corresponds with the offset 69B94 in a French Fire Red ROM (I use French ROMs).

The result of the first XSE script should be two messages, "Script 1" then "Script 2", right ? Instead of it, when I talk to a person with the first XSE script, I can hear the Bulbasaur's cry, and there isn't any message.

What's the matter ?
Click to expand...

In your routine, you don't need the +1 at the script address. And I expect "Script 2" to be shown before "Script 1", and maybe just Script 2.
 
Another problem :(

I used the Darthatron's code, and it works, but only when I use my item pressing Select. If I click on "Use" in the bag, nothing happens, except the bag closes (but no crash).

Here's the ASM code with French Offsets :
Spoiler:

My XSE script is compiled at 0x900000. In Item Editor, I put C00001 at "Field Usage Code Pointer", because my ASM code is compiled to 0xC00000.


EDIT : I didn't see your post !

daniilS said:
In your routine, you don't need the +1 at the script address. And I expect "Script 2" to be shown before "Script 1", and maybe just Script 2.
Click to expand...

You're right, it works ! And effectively, just "Script 2" is shown.

But I can't use XSE scripts with items if I don't register them...
 
Last edited:
FBI agent said:
Oh, I've done some research on this before :D

The Map's name is based on the map value (I think that's what it's called), it's a byte in the map header.
You will need to change this to an unused byte value and then do some editing\expansion on the routine at 080C4D78. This routine takes the following parameters:
r0: Location to store string
r1: Map value byte (I was just talking about)
r2: ??? I just use 0x0, though for your purposes it doesn't matter.

Basically it's the routine that will read Map names. There's some calculations and stuff that go on in there which you may want to edit slightly for it to return a map name of your liking. The easiest way is to hook from the start, make an exception case for your specific map value.

Code: 
Main:
	push {r0-r3}
	cmp r1, #0x[your custom map value]
	bne default

forceName:
	ldr r1, =(0x8[pointer to your map's name in hex])
	ldr r2, =(0x8008D84 +1) @str_cpy
	bl linker
	b end

linker:
	bx r2

default:
	pop {r0-r3}
	@restore overwritten instructions
	@bx to appropriate location to continue safely
	@note these comments are for you to fill in with the appropriate assembly instructions

end:
	pop {r0-r3}
	ldr r4, =(0x80C4DF2 +1)
	bx r4


Hopefully that helps (it should, as I practically spoon fed you the solution :P). If you have ASM questions ask in the Assembly Help thread :D
Click to expand...

I was wondering if somebody can explain me more of this. For example, I know it's ASM but how do I insert it in my rom? Do I need to disassemble it and go to the right 'memory location' or what? Sorry if I sound a bit noobish but I'm not an expert.
 
JoshTheTrainer said:
I was wondering if somebody can explain me more of this. For example, I know it's ASM but how do I insert it in my rom? Do I need to disassemble it and go to the right 'memory location' or what? Sorry if I sound a bit noobish but I'm not an expert.
Click to expand...

You would insert it right after the push for the function, which would be at 080C4D7A. Though you may obviously note that this is not a word aligned offset so you'd pad it with 2 bytes of zeros to reach 080C4D7C. From there you insert to that offset in a hex editor, the hex you get after compiling:
Code: 
ldr r6, =(0x8[Custom snip +1])
bx r6

Custom snip would be the pointer to map name assigning routine from last post +1. You may notice we've overwritten some instructions for our little jump. You have to edit
Code: 
default:
	pop {r0-r3}
	@restore overwritten instructions
	@bx to appropriate location to continue safely
	@note these comments are for you to fill in with the appropriate assembly instructions
to include those instructions, and use a safe register to return to some non-junk code which would run subsequently after and at the same time not mess up existing register values.
 
FBI agent said:
You would insert it right after the push for the function, which would be at 080C4D7A. Though you may obviously note that this is not a word aligned offset so you'd pad it with 2 bytes of zeros to reach 080C4D7C. From there you insert to that offset in a hex editor, the hex you get after compiling:
Code: 
ldr r6, =(0x8[Custom snip +1])
bx r6

Custom snip would be the pointer to map name assigning routine from last post +1. You may notice we've overwritten some instructions for our little jump. You have to edit
Code: 
default:
	pop {r0-r3}
	@restore overwritten instructions
	@bx to appropriate location to continue safely
	@note these comments are for you to fill in with the appropriate assembly instructions
to include those instructions, and use a safe register to return to some non-junk code which would run subsequently after and at the same time not mess up existing register values.
Click to expand...

And that's in the assembly of my rom, right? So I just need a disassembler and assembler or is it just basic script editing?
 
Status
Not open for further replies.
Back
Top