Right Let's Try This Again: PS3 Hypervisor Hacked

Alec said:
I get this impression too...that he tried via software but hit a brick wall. It was 5 weeks from when he had everything he needed to accomplish the goal. I still consider that 5 weeks, though. =)
I wish my boss would think that way about my deadlines :\
"(month later) But boss, I only actively worked on this project for 2 days!"
 
Lord Error said:
It's pretty astounding that SCE went from the least protected console ever built (PSP, which in its 1.0 Japanese incarnation has even failed to do any kind of checks to see if it's running homebrew code or official code) to something like PS3.

I see now also that his claim to have hacked it in 5 weeks is also only half truth. It seems like he gave up on it at first and only much later figured something else out, and hacked something after two more weeks of active work.

Not really. They learned their lesson. MS also had a huge problem with Xbox and as result the 360 is very secure (except those DVD drives). Though Nintendo is rather slow, with their experience this generation I'd expect them to hire a few more security professionals to look at their next system too. New piracy is mostly done by taking the system and dumping all your ROMs to a hard drive and running it on the hardware itself. It used to be just disc copying and emulation. It took a while for hardware manufacturers to catch on and start protecting their consoles from the inside.
 
Have we had Hello World yet?

I mean it seems like a lot is just riding on this guy's credentials.

Obviously he's gotten somewhere or he wouldn't be posting anything at all, but I'd be surprised if this gets anywhere.
 
ymmv said:
The actual size of PS3 games is a lot smaller. Since just about all multiplatform games have to fit on a Xbox 360 DVD as well (7GB usuable space), this is true of most PS3 games as well. http://orlydb.com/s/ps3 lists disc sizes for a number of PS3 games.
It's also worth bearing in mind that they can put padding on a disc to optimise access speeds, among other reasons, that only bloats an ISO without affecting performance. I routinely rip my PSP games to the Memory Stick and I've seen ISOs go from ~1.5GB to 500MB or even less once you've stripped out the padding, junk files and on-disc firmware updates. Also a lot of PS3 games use the extra space for, say, including the voice acting for all regions, so if it's like a PSP ISO you can strip out or relink the non-English audio and cut-scenes to remove those.
 
Lord Error said:
At this point there's been so many PS3 games that couldn't have been pirated (and that sold as much as they would ever sell) that even if the full hack surfaces tomorrow, all the developers making those games should be pretty happy with how the system is secured - if they think piracy would have ruined their sales.

Yes, but it does affect them going forward, regardless of the past.
 
Somnid said:
Not really. They learned their lesson. MS also had a huge problem with Xbox and as result the 360 is very secure (except those DVD drives).
X360 was busted in a worst possible way (until recently when it was hacked completely again with jtag) - piracy only, but no homebrew. I'm sure MS would have preferred if it was the other way around, and I'd call that a pretty big security snafu.

lupinko said:
Yes, but it does affect them going forward, regardless of the past.
Well, it hasn't been for naught, which is what I was replying to. They got at least 3+ years of piracy free sales out of it, which is rarely (if ever?) seen.

NekoFever said:
It's also worth bearing in mind that they can put padding on a disc to optimise access speeds, among other reasons, that only bloats an ISO without affecting performance. I routinely rip my PSP games to the Memory Stick and I've seen ISOs go from ~1.5GB to 500MB or even less once you've stripped out the padding, junk files and on-disc firmware updates. Also a lot of PS3 games use the extra space for, say, including the voice acting for all regions, so if it's like a PSP ISO you can strip out or relink the non-English audio and cut-scenes to remove those.
Paddings are always ripped out of scene ISOs. The reason some PS3 exclusive games take so much space is that they actually put a lot of content on them that would otherwise have to be split to multiple discs, and they use decent quality video bitrate (as opposed to garbage quality that we usually get on multiplat releases). Multiple language voices usually wouldn't take that much space in comparison.
 
While there is a bigger discussion behind which I would love to discuss (regarding regional elements like copyright, watchdog ratings, censorship, etc), I believe the news is interesting because of the time it actually took to hack: If the 5 weeks are true, it means no group really tried to crack the PS3.
 
ReyBrujo said:
While there is a bigger discussion behind which I would love to discuss (regarding regional elements like copyright, watchdog ratings, censorship, etc), I believe the news is interesting because of the time it actually took to hack: If the 5 weeks are true, it means no group really tried to crack the PS3.

I wouldn't stop Geohot's hack clock until he shows he has something running. :)
 
ReyBrujo said:
While there is a bigger discussion behind which I would love to discuss (regarding regional elements like copyright, watchdog ratings, censorship, etc), I believe the news is interesting because of the time it actually took to hack: If the 5 weeks are true, it means no group really tried to crack the PS3.
Lets not be naive - there was most definately a bounty out there by the commercialized piracy industry for a PS3 exploit that they could profit off of.

He's a pretty smart guy when it comes to cracking hardware, I'll give him that.

However, I still think cracking a lowly iPhone is orders of magnitude different from being able to fully bust open the CELL, as securely designed as it is.
 
Lord Error said:
Paddings are always ripped out of scene ISOs. The reason some PS3 exclusive games take so much space is that they actually put a lot of content on them that would otherwise have to be split to multiple discs, and they use decent quality video bitrate (as opposed to garbage quality that we usually get on multiplat releases). Multiple language voices usually wouldn't take that much space in comparison.

I forget which game it was where the PS3 cutscenes were beautiful high bitrate HD and the X360 cutscene videos were compressed and sub HD to fit on 1 disc. It was about 1 year ago, a Japanese game. But the review had a side by side of the videos and the difference was jarring.

Sure its cutscenes, but the point is compression and space saving.

And as to extra languages, a couple of developers said that their PS3 voice work was huge in size and had to cut English out in other regions to fit it all. So voice work can take a significant chunk if its multi region (think including 5+ languages in Europe).
 
AndyD said:
I forget which game it was where the PS3 cutscenes were beautiful high bitrate HD and the X360 cutscene videos were compressed and sub HD to fit on 1 disc. It was about 1 year ago, a Japanese game. But the review had a side by side of the videos and the difference was jarring.

Sure its cutscenes, but the point is compression and space saving.

And as to extra languages, a couple of developers said that their PS3 voice work was huge in size and had to cut English out in other regions to fit it all. So voice work can take a significant chunk if its multi region (think including 5+ languages in Europe).

I'm guessing it could be Tales of Vesperia.
 
Is this thread killing NeoGAF?

Anyways, even if we agree that cracking an Apple is different from cracking a PS3, you don't usually sign a half-hearted crack. Once your reputation goes down, nobody will take you seriously again. Sure, it may not be 100% reliable, but again, 3 years and this is the most relevant PS3 cracking news?
 
Lord Error said:
X360 was busted in a worst possible way (until recently when it was hacked completely again with jtag) - piracy only, but no homebrew. I'm sure MS would have preferred if it was the other way around, and I'd call that a pretty big security snafu.

Homebrew and piracy are very much entangled. Usually it starts with homebrew and then the pirates start building on that. This is how it happened with Wii, early Wiis required mod chips to pirate but it got much worse once homebrew USB loaders appeared. As a hardware manufacturer you don't want either because full reign over the system let's you do naughty things as well.

Particularly in Xbox's case people just ripped games to the hard drive, which in many ways makes it easier (and cheaper) than disc copying. Obviously, MS built around what they knew best. Next time the disc drive probably won't be a trusted component.
 
Mad_Ban said:
I'm curious as to what he means by saying it would let users run PS2 games on the console. Either he was hacking a PS3 with BC capabilities, or he knows something we don't.

I think he doesn't know what he's talking about in that regard.
 
Alec said:
Those people have confronted Geohot about this in his blog's comments section and he has said that it's not a problem for his exploit.

And that's what some of us are waiting on. Hell, maybe he's right, but I'd like more than just taking his word for it.
 
Based on the IBM whitepapers I've read regarding CELL's security design, I agree that it's a timing exploit (IIRC he even said it was on one of his blog comments), based on messing with the hardware slightly so he can work his software magic to gain hypervisor control - no more, no less.

He still has no access to the isolated SPE responsible for all decryption, because it was designed to be fully isolated from hypervisor. He has the processes that go in and out, but he can't actually manipulate it in any meaningful manner.

Therefore, he still does not have the "root key", the key that forbears all keys - the so-called "holy grail". That was the source of his success with the iPhone, and without it, any exploits are piecemeal and temporary at best.

And the problem is, the CELL isn't an iPhone - unless someone at IBM really screwed up, there's no way to get to that root key - it's hardware embedded with no method of access short of sending it to a hardware decryption facility.

So while he may have SOMETHING, the barriers that still stand before him are pretty vast

1) The massive amounts of reverse-engineering that has yet to be done at this point - for a one-man operation who isn't giving away his secrets, this will take a long, long time.

2) Stabilizing the timing exploit so there's some reliability of success.

3) Circumventing any "killswitch" methods Sony undoubtedly will implement

4) Running stable unsigned code all while intercepting any processes that would touch the sensitive isolated SPE from going terror warning level black and shutting the thing down
 
hauton said:
Based on the IBM whitepapers I've read regarding CELL's security design, I agree that it's a timing exploit (IIRC he even said it was on one of his blog comments), based on messing with the hardware slightly so he can work his software magic to gain hypervisor control - no more, no less.

He still has no access to the isolated SPE responsible for all decryption, because it was designed to be fully isolated from hypervisor. He has the processes that go in and out, but he can't actually manipulate it in any meaningful manner.

Therefore, he still does not have the "root key", the key that forbears all keys - the so-called "holy grail". That was the source of his success with the iPhone, and without it, any exploits are piecemeal and temporary at best.

And the problem is, the CELL isn't an iPhone - unless someone at IBM really screwed up, there's no way to get to that root key - it's hardware embedded with no method of access short of sending it to a hardware decryption facility.

So while he may have SOMETHING, the barriers that still stand before him are pretty vast

1) The massive amounts of reverse-engineering that has yet to be done at this point - for a one-man operation who isn't giving away his secrets, this will take a long, long time.

2) Stabilizing the timing exploit so there's some reliability of success.

3) Circumventing any "killswitch" methods Sony and IBM undoubtedly will implement

4) Running stable unsigned code all while intercepting any processes that would touch the sensitive isolated SPE from going terror warning level black and shutting the thing down
You gotta start somewhere.
 
Lord Error said:
At this point there's been so many PS3 games that couldn't have been pirated (and that sold as much as they would ever sell) that even if the full hack surfaces tomorrow, all the developers making those games should be pretty happy with how the system is secured - if they think piracy would have ruined their sales.

Yeah. The best time for a system to get hacked is 4+ years into the generation, when the emergence of piracy is going to be late enough that it's extremely unlikely it'll meaningfully impact the software-buying habits of the install base.

Lord Error said:
I'm sure MS would have preferred if it was the other way around, and I'd call that a pretty big security snafu.

Yes, although the security of the stuff they put together themselves really was pretty tight -- the problem was letting in hardware from an external vendor (the off-the-shelf disc drives) and thereby ensuring that their system could never be more secure than those drives were on their own.
 
Alec said:
If he attempts to execute unsigned code, then it would send an invalid signature to the hypervisor. Since he has full hypervisor access, he can just tell it to ignore the broken signature. I'm thinking he doesn't know how to code anything that the PS3 would see as an executable. The way I understand it, he's saying "I don't have any code to test, but if I did, then it would run."


Nah, that's BS. Dude could write a simple a line of code to just print something to a console or to the screen. And I'm sure he could write much more than that if he's able to hack systems like this. It's not at all a question of him not knowing how to write code that'll run - it's a question of getting it to run.

From his interviews et al he's claiming to be able to do whatever he wants, so a little hello world demo wouldn't go amiss at this point...why it is reasonable to ask for that.

On your first point, though, about the integrity checks - it's supposed to happen entirely in hardware. As it's meant to work, the hypervisor doesn't have any say in whether code is valid or not. Precisely so if the hypervisor is hacked it doesn't offer everything up to the hacker.

That's how it's supposed to be anyway. How things actually are, and how well these things are applied may differ, and may let a hacker through.
 
Will have to see what the exploit is, and if it really does work before the interesting stuff begins. Until then, it's just speculation (like all the other claims).

Seeing how Dark-Alex has been known to have been working on cracking the PlayStation 3 for a while and is very familiar with Sony hardware/software, I'm surprised to see this iPhone hacker do in 5 weeks, what nobody could do in 3 years.
 
charlequin said:
Yeah. The best time for a system to get hacked is 4+ years into the generation, when the emergence of piracy is going to be late enough that it's extremely unlikely it'll meaningfully impact the software-buying habits of the install base.

Right.

But it can create a spike in console sales, if the homebrew/piracy/custom market is very good. Because the console gets cheaper and in addition to the legal uses, you also can do all the gray area stuff.

I can totally see more people buying a PS3 if you can get the equivalent of Boxee/XBMC quality stuff on there. If nothing else as a very high quality centerpiece for a home media center.
 
gofreak said:
Nah, that's BS. Dude could write a simple a line of code to just print something to a console or to the screen. And I'm sure he could write much more than that if he's able to hack systems like this. It's not at all a question of him not knowing how to write code that'll run - it's a question of getting it to run.

From his interviews et al he's claiming to be able to do whatever he wants, so a little hello world demo wouldn't go amiss at this point...why it is reasonable to ask for that.

On your first point, though, about the integrity checks - it's supposed to happen entirely in hardware. As it's meant to work, the hypervisor doesn't have any say in whether code is valid or not. Precisely so if the hypervisor is hacked it doesn't offer everything up to the hacker.

That's how it's supposed to be anyway. How things actually are, and how well these things are applied may differ, and may let a hacker through.

Yeah, I guess it is a little silly for me to think that he can't write a simple Hello World.
 
NekoFever said:
It's also worth bearing in mind that they can put padding on a disc to optimise access speeds, among other reasons, that only bloats an ISO without affecting performance.

No they won't. Blu-ray drives are designed to have a constant read speed regardless of where that data are stored on the disc. So there's no need to add any padding at all.
 
This could be awesome for XBMC like functionalities. BTW, doesn't anyone else find it an odd coincidence that both the xbox360 and the ps3 have been hacked to run unsigned code at about the same time?
 
gofreak said:
Nah, that's BS. Dude could write a simple a line of code to just print something to a console or to the screen. And I'm sure he could write much more than that if he's able to hack systems like this. It's not at all a question of him not knowing how to write code that'll run - it's a question of getting it to run.

From his interviews et al he's claiming to be able to do whatever he wants, so a little hello world demo wouldn't go amiss at this point...why it is reasonable to ask for that.

On your first point, though, about the integrity checks - it's supposed to happen entirely in hardware. As it's meant to work, the hypervisor doesn't have any say in whether code is valid or not. Precisely so if the hypervisor is hacked it doesn't offer everything up to the hacker.

That's how it's supposed to be anyway. How things actually are, and how well these things are applied may differ, and may let a hacker through.

He's got that. Problem is all the software he's shown is running in linux. :) He's written a kernel module, from which he's presumably running hypervisor calls to test whether his hardware glitch works. From that, apparently he has full control over the hypervisor... in OtherOS. I'm not sure what's next.

http://3.bp.blogspot.com/_NJ4JFBfr1tY/S1d3ZuG38gI/AAAAAAAAAbI/YLBQefLdIwI/s1600-h/iglitch.JPG
 
expy said:
Seeing how Dark-Alex has been known to have been working on cracking the PlayStation 3 for a while and is very familiar with Sony hardware/software, I'm surprised to see this iPhone hacker do in 5 weeks, what nobody could do in 3 years.

I am pretty sure this is completely urban myth. DA has just left the scene for whatever reason, be it Sony threat/payoff, but it certainly isnt to abandon the PSP and crack the PS3.
 
gofreak said:
Nah, that's BS. Dude could write a simple a line of code to just print something to a console or to the screen. And I'm sure he could write much more than that if he's able to hack systems like this. It's not at all a question of him not knowing how to write code that'll run - it's a question of getting it to run.
while i would not venture to guess what it takes to print something on the ps3 screen (you clearly need to have the entry points of the fw routines to be able to do that), i can only agree that ockham's edge here says that if he has not shown any, even the most minimalistic, of unsigned code running on the ps3, then he does not have one at this stage. so again, he may have all the cred in the word, but his announcement was premature.

That's how it's supposed to be anyway. How things actually are, and how well these things are applied may differ, and may let a hacker through.
yes, foolproof designs are one thing, their implementation - another. the human error factor - that's how 360 was hacked originally.
 
His latest comment:

"Everyone can already run unsigned code, it's called OtherOS"

Responding to questions if he has unsigned code running.

A bit facetious, no?

I don't think he has unsigned code running yet, but he doesn't want to say anything explicit on it until he DOES have it running and can say he has it running. He doesn't want to say 'no, I don't'.
 
gofreak said:
His latest comment:

"Everyone can already run unsigned code, it's called OtherOS"

Responding to questions if he has unsigned code running.

A bit facetious, no?

I don't think he has unsigned code running yet, but he doesn't want to say anything explicit on it until he DOES have it running and can say he has it running. He doesn't want to say 'no, I don't'.

That combined with my post... basically, he's hacked OtherOS. :lol

I'm not saying he can't do anything with that, but I'm becoming more optimistic that this hack isn't going anywhere any time soon.

For the people wanting XBMC, you'll help me out this summer on the linux version, right? Anybody good at writing multi-core h264 codecs?
 
gofreak said:
His latest comment:

"Everyone can already run unsigned code, it's called OtherOS"

Responding to questions if he has unsigned code running.

A bit facetious, no?

I don't think he has unsigned code running yet, but he doesn't want to say anything explicit on it until he DOES have it running and can say he has it running. He doesn't want to say 'no, I don't'.

a 'bit' facetious? that's akin to saying 'everybody can run unsigned code, it's called java script on a web page'.

running unsinged code means running code that the authority in the system would not allow you to run.
 
Psychotext said:
Seek times.

Maybe I'm missing something but how would padding blank data help with seek times? Storing the commonly used data in several places might help (say storing item and NPC textures together with each level's data), but those kinds of real data padding can't really be trimmed in the ISO process.
 
gofreak said:
His latest comment:

"Everyone can already run unsigned code, it's called OtherOS"

Responding to questions if he has unsigned code running.

A bit facetious, no?

I don't think he has unsigned code running yet, but he doesn't want to say anything explicit on it until he DOES have it running and can say he has it running. He doesn't want to say 'no, I don't'.
Confirmed it's a fat PS3 then.
 
I dunno. Maybe his comment is a double-speak reference to some role otheros plays in allowing his own code to run somewhere else with full privledges.

On the other hand though, he's avoided answering if his function patches have worked or not.

Well, anyway, if he has got unsigned code running I think now would be the time to show it. Or to just say he's still working on it. If he is still working on it, though, then conservatively speaking it's probably been premature to claim a system hack at this point and to talk the way he has to the register et al. Maybe it is just a matter of time, but without this bit of the puzzle he's not quite there yet.
 
linsivvi said:
Maybe I'm missing something but how would padding blank data help with seek times? Storing the commonly used data in several places might help (say storing item and NPC textures together with each level's data), but those kinds of real data padding can't really be trimmed in the ISO process.
I have no knowledge of how data may or may not be "trimmed in the iso process", I'm just commenting on why a developer would choose to replicate data over the disc.
 
hauton said:
Have we had Hello World yet?

I mean it seems like a lot is just riding on this guy's credentials.

Obviously he's gotten somewhere or he wouldn't be posting anything at all, but I'd be surprised if this gets anywhere.

And therein lies the problem. He hasn't produced the elusive "Hello World" program yet. This is probably one of the first things you do when you claim to have 'owned' a system. Now, I'm going to disregard the lame comments that have suggested he just isn't familiar with programming the system because if he's talented enough to 'own' the system then he really should have no problem with this.

He's probably realizing right about now that the PS3 isn't an iPhone.
 
Psychotext said:
I have no knowledge of how data may or may not be "trimmed in the iso process", I'm just commenting on why a developer would choose to replicate data over the disc.

Well, DVD drives read data faster on the outer portion of the disc, and that's why some developers would pad empty data to keep the actual content on the outer rings. Since those blank data are never needed or even read, they can be easily trimmed when you convert the disc to an iso file.

If a blu-ray disc uses duplicated data to lower the seek time, such data cannot be trimmed because those are actual data that the game would use.

I have never worked with blu-ray discs though so that's just my understanding from what I've read previously.
 
Alec said:
If he attempts to execute unsigned code, then it would send an invalid signature to the hypervisor. Since he has full hypervisor access, he can just tell it to ignore the broken signature. I'm thinking he doesn't know how to code anything that the PS3 would see as an executable. The way I understand it, he's saying "I don't have any code to test, but if I did, then it would run."

The Hypervisor isn't part of the chain of trust. Did you really think IBM would entrust a weak link like a software Hypervisor? The decryption keys are only exposed in the isolated SPU vault. And once an application is executing in the vault absolutely nothing can monitor or tamper with what's happening inside. If his claim to fame is taking control of the Hypervisor then he's set himself up for failure.
 
Truespeed said:
And therein lies the problem. He hasn't produced the elusive "Hello World" program yet. This is probably one of the first things you do when you claim to have 'owned' a system. Now, I'm going to disregard the lame comments that have suggested he just isn't familiar with programming the system because if he's talented enough to 'own' the system then he really should have no problem with this.

He's probably realizing right about now that the PS3 isn't an iPhone.


I'm not sure if this quote was posted already...

George Hotz said...

the stupid hypervisor is PPC and C++

if it were C and ARM, maybe i'd have a public sw exploit already.
 
Truespeed said:
He hasn't produced the elusive "Hello World" program yet.

Just a "Hello World" wouldn't actually prove anything, though, because you can already run unsigned code on a PS3. In order to produce real proof that he's accomplished anything, he needs to produce a demonstration that does something that can't be done in OtherOS.

I mean, I agree, he doesn't have a "complete" hack working yet, but just due to the nature of what specifically he's working on, a concrete demonstration is a higher standard than it is on a system that has no OtherOS-style openings.
 
charlequin said:
Just a "Hello World" wouldn't actually prove anything, though, because you can already run unsigned code on a PS3. In order to produce real proof that he's accomplished anything, he needs to produce a demonstration that does something that can't be done in OtherOS.

I mean, I agree, he doesn't have a "complete" hack working yet, but just due to the nature of what specifically he's working on, a concrete demonstration is a higher standard than it is on a system that has no OtherOS-style openings.

I thought it was implied, but the "Hello World" application would have to be executed from the GameOS (where you have access to all of the resources). The OtherOS is meaningless in this context.
 
from his comments replies in his blog & beyond3d,
I'm less opposed to piracy on the PS3 as I am on the iPhone. Obviously, it must not hurt the game manufacturers that bad, or they wouldn't continue to release PC versions of games. And if a modchip is required, that will eliminate a huge chunk of would be pirates. If you are willing to open up your system, learn some electronics, and solder, perhaps you deserve free games. I hate the tools who download blackra1n then ask me where their free apps are, and wish Apple had better DRM, which none of the top guys in the iPhone scene would touch.

Who cares about the strength of the encryption? Systems don't get hacked because the designers chose 1024-RSA instead of 2048-RSA, or 128-AES instead of 256-AES. If the system can decrypt it, you can decrypt it.

And yes, your understanding of the hypervisor is correct. If it's working properly, it shouldn't give me access to the resources I want...but thats what the hardware I add is for, to make the system not work so properly at exactly the right time.
January 21, 2010 10:14 AM

Losses due to piracy are incredibly hard to measure. For example, I have 3 Miley Cyrus songs in my iTunes library, but I really don't think she lost any money because of me...

Piracy in the iPhone scene bothers me for a different reason. The people who want cracked apps seem to be the biggest leeches around, who'd never give anything back to the scene and don't appreciate the legit uses for jailbreaks. Also theres a big difference between a $1 app and a $60 game, which is why I think the people are like this...too cheap to spend a dollar.

Thinking about piracy in television, I wouldn't be watching LOST if I couldn't pirate the first two seasons and catch up. So they gained a viewer.

The real reason I'm against piracy on this blog is the DMCA and lawyers though. It's not a moral issue.
January 21, 2010 3:22 PM

It seems like his hardware will require a physical mod. If so, I think the Sony is letting out a big sigh.

The fact that he listens to Miley Cyrus will wipe out any reputation gains and respect he'll get from hacking the ps3. :)
 
Top Bottom