Hacking Coding vWii 3-core support - everything you need to know.

[comex]

Everyone has skills. Mine are basketball, Texas Hold'em, Psychology, working with developmentally disabled (mostly children but some adults), and martial arts. I also am well versed in Physics. Unfortunately, while this helps my self esteem, and I doubt Marcan has any of those skills, this is not applicable here. I dabbled a LITTLE in Java but that is laughable to most who code and this was one semester to see if I wanted to go that route. Building, repairing, upgrading networks and computers has been my passion--and soldering. I should have kept contacts from college who were heavily into all of this stuff. Advocating for the poor and disadvantaged for public services and legal rights is something else I am known for. Some decent mathematical abilities as well.

As someone who, as previously mentioned, can't solder worth anything, would like to learn how (along with other, deeper aspects of hardware, and some aspects of CS and math and Haskell and whatnot), is not learning how, and is thus somewhat hypocritical, I still recommend giving learning programming on your own a try if you're interested. Mathematical abilities help!

 

[the_randomizer]

As someone who, as previously mentioned, can't solder worth anything, would like to learn how (along with other, deeper aspects of hardware, and some aspects of CS and math and Haskell and whatnot), is not learning how, and is thus somewhat hypocritical, I still recommend giving learning programming on your own a try if you're interested. Mathematical abilities help!

Which I struggle with like no tomorrow, but I don't want to elaborate. I'm not unappreciative of peoples' help, quite the contrary; it's one thing to have the desire to learn programming, it's quite another to put for the effort. Yes, I do want to learn programming and eventually, be able to contribute to these discussions in the near future, but again, that's where I run into an issue. By the time I actually learn something useful at my current rate of learning, Nintendo would have phased the Wii U out., but I digress. Pathetic I know.

 

[uyjulian]

I only know most of Java :/ no C** However, I attempted to learn C# at one point [forgot when] So yeah, I would try to help if I could...

 

[Maxternal]

I've said this before in other threads but it seems relevant to this discussion, too :

Someone needs to make a java virtual machine for Wii. It would be SO useful.

 

[Slimmmmmm]

Regarding "the wad part", if you install bootmii-ios into an unused slot it shouldn't cause a brick, and I vaguely think it would run correctly, but I'm not sure.

On the Wii (not vWii) I installed an edited bootmii ios to an alternate slot and it loaded as normal (using load title). (this was part of DELUXE for Sorg). No idea about the vWii though, however an unused slot *should* be safe afaik and the files removed with any wad manager unless bootmii ios writes extra files I'm unaware of.

*EDIT*
btw, thanks Maxternal for undertaking this. I myself have no knowledge of ASM worth noting.

Thanks Marcan for the opportunity and information so that someone with the know-how to begin using more cores for maths etc can make a POC and maybe get your wisdom and input on building a community out of Homebrew on WiiU (the hardware IS interesting, to some).

Maxternal, maybe there are other devs that are working on adding 3 core support into their apps and your time would be better spent helping them, as their 3 core app would perhaps be POC for Marcan & Co... ? Just thinking out aloud. I wont name names on a forum based on my speculation, but I'm sure a few spring to your mind and some others that popped up for Wii, did 1 or 2 things and went quiet are still around.

Anyway, good luck.

There is 1 dev that's in love with Nintendo from before Wii, a big kid. Another that has multiple, popular apps on Wii....another that used to post lessons on his site and said many a time he would love to do more (if he had time). Their are a few devs out there that worked alone on some interesting apps and hombrew games, a couple of devs that made loaders that never cloned most of the code from existing loaders and were very talented etc etc etc.

If you can find someone then I'm sure people can help them, even if it's typing in a wiki as long as it benefits both sides.

 

[naddel81]

@ Ray Lewis: I have been a quiet listener/reader here for weeks/months and I have to admit that your effort is admirable!
Even if there are bigger stones on your way you don't become tired of organizing someone who can help. Astonishing!
As many others deep down we are all willing to contribute, but most of us do not hage the time and skill. I did have a few lessons in basic coding in school, but that's it
But one thing is for sure: I would pull a geohot or mathielh act if I had discovered something that many others are waiting for (for whatever reason).
Keep it up!

 

[Cyan]

I think Maxternal should still learn things by himself, instead of only help another team who already know what to do and who will do most of the work.
It will be more profitable for him to learn and understand what he is doing, even if it will takes two years. He may not be the one who will provide access to the 3 WiiU cores, but he will help on other and future project later.

I wish I could help and understand more things, but I'll only follow from far away.
I started learning ASM in 1995 at school, but it was for a very small chipset (I don't even remember the name/ref) for an automated device we built from scratch, making our own printed circuit board, electronics part choice, etc.
I liked it and always wanted to learn programming since that time, but I never took the time to correctly learn, I only watched other's code and try to modify things blindly to see how it worked to learn by myself. I can modify, but can't create.

 

[Cyan]

no, I was only replying to a post above mine saying "if there's another team, maxternal should work with them" (instead of having two different groups working on the same thing).

Edit: I'm just encouraging Maxternal to continue learning instead of depend on others.

 

[Crewman]

So guys, where we at?

Two sites of conversation without any news about the project.

Whats next Maxternal, are there any unsolved problems atm.?

 

[Ninja_Carver]

So guys, where we at?

Two sites of conversation without any news about the project.

Whats next Maxternal, are there any unsolved problems atm.?

I know, its kind of weird not having any activity for an hour on either thread

 

[Slimmmmmm]

Cyan, I agree with you that it would be best in his interest IF learning is his agenda.
If getting results is the agenda then what I was saying would apply. Obviously there is a grey area in-between where he can do both to some degree and it maybe the perfect solution. Who knows.

I don't even nearly have the time to learn ASM to the point of being useful, so my post count in this topic will stay low. I have a massive interest though, as this seems to be a stepping stone to things I can participate in (gamepad libs for unique homebrew games, redirecting of read/writes to solve N's problems and customs).


What can benefit from 3 core support.... this is a good question !

- Emulators
- Media Players
- Homebrew games (judging by the Wii ones, I doubt any will take advantage really)
- Applications that would do several tasks at the same time (web browser, file manager etc)

Perhaps asking people that might be doing some of these things would be a good place to start. Remember though, it's not C etc...

 

[djbubba2002]

I PM him 14 mins ago ..

 

[Maxternal]

Sorry, they're cracking down more about Internet usage at work so I'll slip in when I can. (before that, I was on the bus and typing on my cell phone is kinda slow.)

If you want to know what there is stil to do, I've put some spoilers on the OP that outline some of the more immediate things I need to know, plan to do, etc.

Someone (yes, djbubba2002) is currently risking their vWii to test just if ARM code can be run the old fashioned way on vWii so we'll see how that goes.


Learning is great but so our results, of course. If anyone more experienced is working on this separately and wants to compare notes ... or isn't working on it completely but wants to occasionally just look at what I've got and throw in a suggestion or even section of code, I'm perfectly willing to learn by example, too.

 

[pezed]

I have a sleeping wiiU there if needed for some tests

 

[Maxternal]

I have a sleeping wiiU there if needed for some tests

I'll add you to the test PM-group-thingy then (there's not much to test yet but there will be more)

 

[cory1492]

These are the three barriers I'm looking at for just sitting down and trying stuff with this exploit:
1 - I have no existing backup/restore method if something goes wrong
2 - I am not familiar enough with the ARM processes in the wii to be arrogant and assume I won't need #1
3 - debug output, I've got nothing but clues
Added together, and having very limited funds/resources (including time), pretty much means it's a non-starter (for now) here.

GOOD NEWS, it looks like the masking and casting isn't really necessary.

It may not be necessary, but it's always best to be clear to the compiler what it is you are doing so you don't have to later debug it at an asm level just because the toolchain got an update and changed it's mind how it does something specific (I think you noted the basis for this comment by having to manipulate the code so it'd work on that paste site at all.) The mask is only important really if you aren't paying attention to the inputs. That said, you are absolutely free to do behaviour tests on the actual ARM/PPC compilers being used to see if simpler code produces the desired result and is better optimized by the compiler if you are willing to risk possibly random breaks.

(In this case you would also have to understand Linux enough to know how to patch the kernel and pthreads to deal with the quirk mentioned in the blog post.)

This longstanding PPC atomicity bug, and whatever solution Fail came up with, is actually what has had my attention. We tried all the different methods we could find for PPC errata on the 360, even the old apple PPC ones, and in the end only two methods showed any real promise - one was kind of ridiculous (though effective) which involved splitting the lock function so it was across 2 or more memory pages; the other was resetting the timebase register on lock attempts (it apparently was not sufficient to just reset it at a point before overflow, outside the lock function.)

 

[Quincy]

I am able to test as well

 

[Quincy]

I'll add you to the test PM-group-thingy then (there's not much to test yet but there will be more)

A little quote alert for you

 

[Maxternal]

I am able to test as well

sure, not much to see at the moment, though.

 

[Maxternal]

These are the three barriers I'm looking at for just sitting down and trying stuff with this exploit:
1 - I have no existing backup/restore method if something goes wrong
2 - I am not familiar enough with the ARM processes in the wii to be arrogant and assume I won't need #1
3 - debug output, I've got nothing but clues
Added together, and having very limited funds/resources (including time), pretty much means it's a non-starter (for now) here.

#1 and #2 ) We shouldn't be doing anything that will make any permanent changes so nothing we can break that won't be fixed by a power cycle.
#3 ) This could pose a hurtle. One idea I'd thought of was to init core0 completely with more traditional methods to get access to the screen before even touching the other cores. Might give us SOME method to get debug output ... but that's the only idea I've had so far with my presently limited skilz and not nearly that far yet.