Homebrew [WIP] SafeSysUpdater

[cpasjuste]

Is there anyway to tell users if memchunkhax2 has worked/didnt work before safesysupdater gives you the option to downgrade? On the old sysupdater its kinda just "press Y and pray"

Yes, if the hack doesn't succeed, then it won't continue (it will either freez or you'll get a message to try again).

 

[AileenLumina]

If it's possible, could you please make a tiny homebrew off this that just checks the MD5s? That would be a good compromise at the moment.

 

[cpasjuste]

If it's possible, could you please make a tiny homebrew off this that just checks the MD5s? That would be a good compromise at the moment.

That's exactly what's added in the latest version. Don't worry it's safe, even if you press (Y) to downgrade instead (A) to check MD5, you'll have to pass the memchunk hax then you'll have a second confirmation after the MD5 checks.

 

[DKB]

That's exactly what's added in the latest version. Don't worry it's safe, even if you press (Y) to downgrade instead (A) to check MD5, you'll have to pass the memchunk hax then you'll have a second confirmation after the MD5 checks.

You're making me want to buy another 3DS right now, just to use this.

 

[reiyu]

Hi Reiy,
This is not more prone to failure than original sysupdater as it's exactly the same downgrade code. You just have a few more checks before it run the downgrade (autodetect device/region and md5 check all files).

because you added more code means you have more points of failure. it's amazing you can code for the 3ds, but you should know this as a general programmer that more code means more to fix. and as of what @Xenon Hacks experienced, it seems like your sysupdater is FAR LESS stable than the normal sysupdater.

 

[cpasjuste]

because you added more code means you have more points of failure. it's amazing you can code for the 3ds, but you should know this as a general programmer that more code means more to fix. and as of what @Xenon Hacks experienced, it seems like your sysupdater is FAR LESS stable than the normal sysupdater.

Im not sure your a coder because if so your a bad one if you looked at my sources. The downgrade code is 100% identical as original sysupdater and so are success/fail rate. If @Xenon Hacks had a brick then he would have been the same with original SysUpdater, believe me.

 

[DKB]

because you added more code means you have more points of failure. it's amazing you can code for the 3ds, but you should know this as a general programmer that more code means more to fix. and as of what @Xenon Hacks experienced, it seems like your sysupdater is FAR LESS stable than the normal sysupdater.

He didn't touch the downgrade code, all he did was add more checks so that people would have LESS points of failure. So, saying it would have more points of failure is bullshit. A perfect example would be..KTM.

 

[AileenLumina]

That's exactly what's added in the latest version. Don't worry it's safe, even if you press (Y) to downgrade instead (A) to check MD5, you'll have to pass the memchunk hax then you'll have a second confirmation after the MD5 checks.

Haha, thanks then! You're doing great work I really appreciate ^-^ However, I'd like to stay at using Steveice10's build of sysUpdater because your code somehow doesn't work as intended (because of unclear memchunkhax2 bugs probably or sth. like that, I don't really know, but your code just doesn't work well with memchunkhax2)

 

[cpasjuste]

Haha, thanks then! You're doing great work I really appreciate ^-^ However, I'd like to stay at using Steveice10's build of sysUpdater because your code somehow doesn't work as intended (because of unclear memchunkhax2 bugs probably or sth. like that, I don't really know, but your code just doesn't work well with memchunkhax2)

Hi,

What do you mean, you tried to downgrade ? In fact it's Steveice10 memchunkhax2 code
I think people have a lot of difficulties to get Steveice10 sysupdater memchunkhax2 to succeed, and so SafeSysupdater too.

 

[AileenLumina]

Hi,

What do you mean, you tried to downgrade ? In fact it's Steveice10 memchunkhax2 code
I think people have a lot of difficulties to get Steveice10 sysupdater memchunkhax2 to succeed, and so SafeSysupdater too.

I was just thinking that some algorithms don't work that well "on top of" memchunkhax2. I'm really unsure about this though

--------------------- MERGED ---------------------------

But so far I'd say it's less stable (somehow) but still safer as you didn't touch the downgrade process itself and everything else is working. Not on every try, but everything works. So, in terms of safety:
Less stability outside the actual downgrade process: ±
Same stability during the actual downgrade process: ±
More checks before the actual downgrade process starts: +

And your code is well-written as far as I can tell. So yeah, no point in using Steveice10's sysUpdater anymore then, I'd say.

 

[Clydefrosch]

because you added more code means you have more points of failure. it's amazing you can code for the 3ds, but you should know this as a general programmer that more code means more to fix. and as of what @Xenon Hacks experienced, it seems like your sysupdater is FAR LESS stable than the normal sysupdater.

technically, you're right, of course, the more code you write, the more likely it is that you make a mistake, but the downgrading code is already working and wasn't changed.
he added hash checks, to prevent you from using the program with the wrong files, which would cause a brick right there. technically, that wasn't necessary, but its for the end user.
at the end of the day, it's still going to have to go through the 3ds's issues though. sometimes, memchunkhax initiates like it should, sometimes it doesnt. there's seemingly little one can do to change that.
and sometimes, installing a cia file will fail regardless of using the right files.

as far as we all can tell, this is as save as its ever going to get. a new entrypoint with more stable kernel access may or may not give us a more stable downgrade somewhere along the line, we'll have to see.

but for now, its a bit of luck and trying over and over.

 

[Urbanshadow]

And your code is well-written as far as I can tell. So yeah, no point in using Steveice10's sysUpdater anymore then, I'd say.

So from work in progress to beta. Looks good.

--------------------- MERGED ---------------------------

As far as we all can tell, this is as save as its ever going to get. a new entrypoint with more stable kernel access may or may not give us a more stable downgrade somewhere along the line, we'll have to see.

Wouldn't an arm9 access allow direct emunand boot (no downgrade needed)?
I guess you are refering to stable arm11k access, with proper permissions to am:u (which would still require dowgrade).

 

[Damon_girl]

Thanks for the new update! Do you still need to be hardmodded in order to use it or can you use it without one?

 

[ketal]

- Try hard to get memchunkhax2 to succeed

fun fact: call consoleInit() twice and the exploit always works

 

[AileenLumina]

fun fact: call consoleInit() twice and the exploit always works

@Steveice10 @cpasjuste Thanks @ketal

 

[cpasjuste]

fun fact: call consoleInit() twice and the exploit always works

Strange, I'll try

 

[Damon_girl]

Ugh, I'm trying my hardest to get this to work, but sysUpdater never wants to get the cia info.

 

[AileenLumina]

Ugh, I'm trying my hardest to get this to work, but sysUpdater never wants to get the cia info.

Wait until the next version is released, it will cut many possible issues. That'll ease troubleshooting probably.

 

[Coleman_C18]

fun fact: call consoleInit() twice and the exploit always works

Too bad this was only now discovered, but awesome nonetheless

 

[ketal]

Please, don't take for granted that it will work. It's a pretty odd solution indeed