Possible Ways to Crack the Bootloader

[czerdrill]

You have thoroughly made your point regarding its low probability.. or you would say infinitesimal probability. Have you considered the fact that my research might lead to new discoveries, not even regarding the encryption but regarding the bootloader itself; that my data will actually reveal something unforeseen; and that by focusing my attention towards a very narrow goal will actually be the source of discovery. What if my test shows widely different results than those that were predicted. We both agree that theory is different than practice, than practice can potentially reveal something that theory overlooked, or vice versa. Most of the world's greatest discoveries were the product of an unintentional result during unrelated experimentation.

{{ WugFresh }}

I've never once denied the possibility. I have and still do deny the plausibility and probability though...

 

[Snow02]

I concur. Again. Not going to happen. Best to devote your time to something achievable.

 

[czerdrill]

I concur. Again. Not going to happen. Best to devote your time to something achievable.

Exactly, I feel as if this would be a detriment to getting around the bootloader rather then a benefit. Too much time will be wasted trying to do something that is essentially "impossible".

 

[13th angel]

I concur. Again. Not going to happen. Best to devote your time to something achievable.

atleast by trying there's a chance though (I know its a very very small chance but still). If everyone just gave up then it deffinately wont get cracked.

Sent from my Liberated D2G

 

[czerdrill]

I concur. Again. Not going to happen. Best to devote your time to something achievable.

atleast by trying there's a chance though (I know its a very very small chance but still). If everyone just gave up then it deffinately wont get cracked.

Sent from my Liberated D2G

Well, you don't have to give up, but in this case, there has to be a more feasible approach.

It's like saying you have the opportunity to reach a star that is 20 light years away. You have two options, build a ship that can travel at the speed of light and get there in 20 years, or try to get there in a ship that you already have that travels at a fast speed but will still take a long unimaginable amount of time to get there. Sure, we can try to build a ship that travels at the speed of light, and if no one tried to do so it would definitely never happen, but it's not a feasible approach. To be honest, neither way is feasible (hash collisions or some workaround), but attempting to travel at light speed (hash collisions) is definitely not the way to plan this out, and using our existing ship is more "feasible" in the sense that light speed travel is impossible.

I think i confused the issue even more with my dumb analogy...lol

 

[Snow02]

No, you're right. The bigger problem with a hash collision approach is that just because you find a collision does not inherently make it useful. You need a collision between the item you're attempting to fake and a replacement that is not gibberish. Just finding a collision is difficult enough, finding one you can use is just not going to happen.

 

[WugFresh]

I understand what you're saying. And I very well might agree with you and look back on my suggestion and come to the same conclusion. But talking with experts never hurt. In addition to talking to my math professor I plan on talking to a quantum physicist.. because they always have a way of making the impossible seem possible.. lol.. again, thats all in theory though.. not in practice. My hope is that through my research I may be presented with an alternative approach. I do appreciate you giving me a reality check, but my goal now remains the common goal of this thread, to generate possible solution procedures to a near impossible scenario. The more I understand about what is involved the more I will be willing to give my concrete opinion on it's potential outcome. But from what you have said, I am more aware that hash collisions may not be the path to take. But I remain determined to find a solution regardless. I am curious why aliasxerog approach was halted and why reverse engineering the radio baseband drivers hasn't been achieved. Maybe a mathematical or physics related approach can be utilised to tackle that frontier. I firmly believe that if things can be built they can be reversed engineered. I will remain determined to find a solution until I hit every brick wall.. and when I do, you can say I told you so.. I don't care. If you have zero intention on solving a problem then there is a 100% chance it won't be solved. I hold to my original statement, I can offer my brain for math and science to tackle this problem. If hash collisions are out of the question then I want to reorient the approach. I will rule hash collisions out of the question when I speak to a few people, then I will shift my juxtaposition towards tackling this problem using an alternative approach. I don't think what I am doing is illogical or a waste of time, I fundamentally disagree with that statement.

{{ WugFresh }}

 

[droydhead]

That's the spirit! i love u (no homo)

 

[WugFresh]

Lol. I agree that it is the spirit. For those who have been adamantly trying to tell me no, thats it's not going to happen, if you haven't figured it out already.. convincing me of that is also not going to happen. I will remain determined to find a solution until I truly believe it can't be done. As of today I still say, maybe.

{{ WugFresh }}

 

[kptphalkon]

Assuming 2^52 possibilities and assuming a computer can calculate ~100/second, then I figure with ~1000 computers working on this, it will take ~1428 years. With a little bit of money, I can get access to ~1000 computers. The ~1428 years part, though, is a little more difficult.

Do better. ;-)

So by your math, all we'd need are roughly 10 million computers to get this done in 5-6 weeks lol.

Your numbers assume two constants though, and possibly a third. The first are the order of magnitude of possibilities. The second would be how many hash ops per second a computer can handle, and the third would be assuming that every computer over the distributed network has a similar-speed bandwidth connection.

With the monstrous availability of low to mid to high end parts what would you say the 'average' PC is capable of, and how exactly do you determine the average number of operations a computer is capable of? For some it might be far less than 100, for other it could be well above.

/edit/ bad nomenclature removed /edit/

 

[Snow02]

Clearly you have no clue what you're talking about. Hash collisions per second? Smh. :facepalm:

 

[kptphalkon]

Im just going by what this guy posted, with his ~100/second comment. Granted i was mistaken in calling it collisions per second lol.

Google'd article says a 700 mhz pentium 3 can run an 8kbyte sha1 hash per 130 microseconds, which is on the order of 7,500 a second.

Disclaimer: I dont claim to know what I am talking about. Just posting my views of what I've been reading.

 

[WugFresh]

I read some interesting stuff on ways to set up machines differently for the purpose of hash collisions, the order of magnitude still obviously makes this whole idea highly unlikely, but I found this book that goes into great detail about ways to improve the likelihood of performing successful hash collisions by modifying the system to utilise its cpu cache differently/more effectively... still though... they weren't using this method for SHA1 and the book seems to be a little outdated.. regardless, it was very interesting. You can literally set up a hash collision machine so to speak..

{{ WugFresh }}

 

[lithiasales.js]

I believe the most logical solution would be to do a SETI@home solution. Use the mass amounts of computers to run hash collisions. I believe this is possible from reading articles about the Chinese that accomplished this. With the right type of determination I think it would be possible.

P.S Or just create a virus style SETI@home so people don't even know there running it bhaahha!

 

[ahC_hED]

I'm gonna take the hardware route and try to pull code off the omap, but I need some help on finding a data sheet for the omap. I have access to a reflow station and the knowledge to do it but I need a pinout.

And also, what is the backup boot, if booting the rom fails, could it be scripted boot using usb or does it just sit in the bootloader?

Sent from my DROID2 of shame.