Emmanuel Fleury schrieb: > Hi, > > Bahattin TOZYILMAZ wrote: > >> sorry for disturbing. but, our code tries to "branch" (in x86 assembly >> "jump" or "call" as i understand) an illegal address(0xFE3BE444). even >> i tried to negate the address. it produced 29mb or so. I am asking >> that, because i dont know much about ARM asm. My only explanation to >> this branch is that, this address is handled by MU (memory unit, i >> think).where does it branch, anybody knows? >> > > We are almost all thinking that the dump has few 'clear text' parts > which are meaningful in ARM asm but is mainly filled up with ciphered > parts which are totally useless until we guess the code. So do not try > to execute it blindly because it won't work for sure. > > For now, we try to identify what kind of encryption scheme is used here > and were could be located the key. We highly suspect that this last line > of defense is hold tight inside the ARM processor but we have no proof > yet. Hopefully, this encryption is a weak one and can be breakable > through a brute-force attack. > > For the ones with a (very) little knowledge in cryptanalysis, you can > try to perform some statistical analysis on the ciphered parts of the > dump and try to compare it with clear text ARM asm trying to find some > bias in the scheme. > > That's all folks ! :) > > Regards > Does the ARM have some tiny internal ROM? Then probably that one holds the key and deciphering code. If not, it simply MUST be located in that dump. There should be some reset vector that copies the image to RAM, XORs it or whatever, and executes it. I don't think that the ARM is able to do that in hardware ;) What we now need is some knowledge how that ARM boots. Where is the reset vector located? I guess that this one points to the only little bit of unciphered code. Somebody worked out a likely XOR key a few messages ago. Does the decoded stuff look useful?
_______________________________________________ Linux4nano-dev mailing list [email protected] https://mail.gna.org/listinfo/linux4nano-dev http://www.linux4nano.org
