Homebrew Python Tools for 3DS

[Kane49]

the problem is that nobody really knows what they're doing.

Well yeah it says that in the readme

 

[zecoxao]

nice to see some homebrew showing up keep it coming guys

 

[windwakr]

the problem is that nobody really knows what they're doing. it's an armv5te arm946e-s, not an armv6 mpcore

Thanks, that fixes all of the issues I've been having.


But I'm a bit confused. Isn't that the same as what's used to compile for the original DS?

 

[Drenn]

Thanks, that fixes all of the issues I've been having.


But I'm a bit confused. Isn't that the same as what's used to compile for the original DS?

This code is running on the arm9 processor which is mostly the same as the DS. There's also the arm11 processor.

 

[Bond697]

you're not running code on the mpcore cpu.

 

[tomiga]

For anybody using GNU/Linux, I made a modified build.py file. Be sure to chmod +x it.
Also to get this to compile, I had to add the following to 3dsploit.py in lib/p3ds:

#!/usr/bin/python2

Then set 3dsploit.py to have execute permissions using chmod +x. The compiler I'm using is arm-eabi-gcc instead of arm-none-eabi-gcc. I'm not sure what the difference between the two is, but arm-eabi-gcc seems to work fine.

 

[Dr.Razor]

Can someone make a quick guide for Windows ?
Are the addresses for the inputs referenced somewhere? Thanks

 

[Kane49]

you're not running code on the mpcore cpu.

I can confirm that you are right as well, thanks

 

[aliak11]

I have created a new thread for homebrew development discussion. http://gbatemp.net/threads/homebrew-development.360646/

 

[escherbach]

I can confirm that you are right as well, thanks

Funny then that the build otions in build.py have

CC+" -g source/*.c -c -mcpu=mpcore -march=armv6k -mlittle-endian

You guys are posting about stuff you don't understand. Spreading fog and misinformation. Please keep your dumb wrong technical guesswork to yourself and leave it to the pros to discuss the advanced stuff.

http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0301h/apbs02s02.html

 

[windwakr]

Are the addresses for the inputs referenced somewhere? Thanks

http://3dbrew.org/wiki/IO

 

[Kane49]

Funny then that the build otions in build.py have

CC+" -g source/*.c -c -mcpu=mpcore -march=armv6k -mlittle-endian

You guys are posting about stuff you don't understand. Spreading fog and misinformation. Please keep your dumb wrong technical guesswork to yourself and leave it to the pros to discuss the advanced stuff.

http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0301h/apbs02s02.html

What the fuck are you talking about, you know who wrote that build.py right ?

 

[escherbach]

What the fuck are you talking about, you know who wrote that build.py right ?

Not you, that's for sure. It's from user tomiga - he knows what he's doing - you and your mates don't

 

[escherbach]

edit - ok maybe you did the original.

I still don't think the code is running on arm9 - where is the fucking arm9 processor? - source of this information please?

 

[Kane49]

edit - ok maybe you did the original.

I still don't think the code is running on arm9 - where is the fucking arm9 processor? - source of this information please?

http://3dbrew.org/wiki/FIRM

 

[escherbach]

http://3dbrew.org/wiki/FIRM

Ok, well that page doesn't say anything about a physical arm9 cpu - it would surely be visible on the circuit board anyway. The arm11 processor is backward compatible with arm9 code so I assume they just switch the chip into a special DSi compatible mode by lowering the speed and disabling dual-core. Similarly arm11 is backwards compatible with arm7 code (but arm9 wasn't backward compatible with arm7 - which is why the DSi had a separate arm7 cpu) - so for DS mode even lower speed single core processor mode needs to be engaged.

http://www.arm.com/products/processors/classic/arm11/index.php

Maybe your compiler options are causing arm11 simd instructions to be generated and executed in a mode for which the arm11 can't execute these codes (because dual-core disabled??)

lol - now I'd doing what i accused you guys of, just guessing.

Anyway - I was wrong to act like a dick - but I still think there is no physical arm9 chip as Bond697 posted - just various compatibility modes of the dual-core arm11 cpu which you switch between via register pokes

 

[Bond697]

and yet there is and you're still completely wrong.

 

[escherbach]

and yet there is and you're still completely wrong.

Why would there be? - the arm11 is backwards compatible with arm9 (and arm7)

Where is your source? - eg a picture of this extra cpu would be good.

 

[escherbach]

Also - I have now solved the confusion of how they execute arm7 DS code - the code is executed on an arm11 (in single core mode) NOT on an arm9 cpu?

 

[profi200]