V47 bugs AMD
#11
I am not sure. My testhashes look like this:

echo -n 1306 | ./oclHashcat-plus64.bin -m 141 '$episerver$*0*Mzc2NzY1NTU0MTM3NDIxOA==*ITBVvl4lw3w7UGLk/soqInuRkos='
echo -n 56038 | ./oclHashcat-plus64.bin -m 141 '$episerver$*0*NzI0NjAwMzYyNTUxMjU4Mw==*/3ZbqTedR8MRWeNe5I6kg5G330s='
#12
(08-21-2012, 08:38 PM)atom Wrote: I am not sure. My testhashes look like this:

echo -n 1306 | ./oclHashcat-plus64.bin -m 141 '$episerver$*0*Mzc2NzY1NTU0MTM3NDIxOA==*ITBVvl4lw3w7UGLk/soqInuRkos='
echo -n 56038 | ./oclHashcat-plus64.bin -m 141 '$episerver$*0*NzI0NjAwMzYyNTUxMjU4Mw==*/3ZbqTedR8MRWeNe5I6kg5G330s='
Ah, I see. Here's what seems to be the problem. I don't know how the algorithm is supposed to work but it looks like the first part after the $episerver$*0* is the salt in base64 and the last part after the ==* is the hashed password in base64. For my hashes, the last part is the same length as yours. The problem is with the salt. Mine are of various length and never as long as yours. So either the algo is supposed to use variable length salt and you need to implement the acceptation of different total hash length by OCLhashcat-plus (that's the error I was getting "length exception error") or if it is supposed to be fixed, then maybe I have installed the wrong perl module for that algo and got wrong hashes generated. In the latter case, can you hash for me the following passwords with that algo so that I can test properly:

Code:
mc
hash
ha$hcat6
HashcatSuite
Tough4Pas$word!
qz
ToDo
C4t&Best
CatIsTheB3st
hashcat!sf0ryou
%*
P0$7
aaaabbbb
aaaabbbbcccc
aaaabbbbccccddd
q!
T0Be
AtomRule
AtomIsOnFire
AtomHelpedUsAll
$D
($$)
BestOfIT
zzzzYouSnore
1234ThereYouGo@
#13
Dynamic salt length should work now (New version uploaded). Here are the requested hashes:

Quote:$episerver$*0*NzYyNDg4MDUxMTEx*IdPKaPmk3qyjeth39wskfa3Uei4=
$episerver$*0*MTAwMzA2NzU2MzA2NA==*AH5j3UVOXw1xZsNMe+YxpZ3ZVfk=
$episerver$*0*NDYzMTYzNDgxMjE3*Y/GHMjUSnVNifV4v8QOUijUzAKQ=
$episerver$*0*ODY3MDA2MjEzNjY4MTc=*mXmdSWzR01V+XLo1x6RDeFCdAMI=
$episerver$*0*NjU2*+OX/ZidYAcptLXP9XUl157uDMio=
$episerver$*0*NzU4MzQxNTc1*ZEhhtn62EceaYv0p5ulg7/vuhx4=
$episerver$*0*ODEyMDg4MTQxMzMx*nUcdDm4bgFllgVtw+gx2i9OZ/hk=
$episerver$*0*MzE3MjY0NDEzMw==*n3Oi4HuNhqDrWeQJBUWoqJWg/fg=
$episerver$*0*ODc0MDU4Ng==*EOVy9uNJf2GJGYY2LOOuSetxq7I=
$episerver$*0*NDEwNzcyNDQ1MzQwMjY=*EOEVB8Kvf5MJOvrhGzjVunvxyiw=
$episerver$*0*MjUzNzUxNjgxNzc2*q6bXs10ryElV6hFku6JOyPW2wKQ=
$episerver$*0*NTYz*zkM1dk4WKShJu1l5U6v5IycRle0=
$episerver$*0*MDExNzgzMQ==*tdbjqyFyy9OzqxgaB8rIKdTrALc=
$episerver$*0*NzU4NTI=*efIn4MDfa1J1NubXHNb7jzbAi8w=
$episerver$*0*NDQ2NTI=*qrvIWVjU3AEV2I3O39KVDlrtO5s=
$episerver$*0*Mjc2OA==*uZOBqA6auwA1UUHZEz86WccLziw=
$episerver$*0*NTE3NTI0*SiEcTzOqfnScsdbbaofpXAYN9so=
$episerver$*0*Mg==*wKSK0RxtfALG/VhieBMcnMCMslA=
$episerver$*0*MA==*2cAArafch1BzPOTg2RtR40E8DRE=
$episerver$*0*NTg0MjY2NjczMQ==*U7fPNvz/f+VMjgH3oqhwpPkBIi4=
$episerver$*0*MjcwMTE2*dLPMIvc6aI1SadArNFnY/wpzxg4=
$episerver$*0*MDE0NTA4MTI=*fow9wJg02OOd5yyxanZ9ZtsynKw=
$episerver$*0*NDUzNzc0*k8dCR98er7FtFb8T7UdszZPksx8=
$episerver$*0*Mzc4NTI0NA==*YLDCKTp7WQ/mFXUvRTD43oQ4ksA=
$episerver$*0*NjcxNzQyNjYyODA4NDg=*W4njCqHDw3HQdHaMMMOz2ECCyCM=
#14
Yep, now -m 141 is fully functional. I used my original hashes to test and everything was fine.
#15
I tried to fix the bugs with 2711 and 2811 but they are driver related. Next thing is this:

Quote:- Progress and rejected counters wrong.

@mastercracker: How can I reproduce?
#16
There is a problem with long salts at m1710. If the salts are small it works. Is this also driver related?
#17
It depends on what you define a long salt. On -m 1710 supported are salts up to length 16.
#18
The salt lenght is 31
#19
yeah ok, that wont work
#20
Quote:- For -m 1710 sha512($pass.$salt), Bruteforce on length 4 fails. Everything else works.
- For -m 1800 sha512crypt, SHA512(Unix), -a 1 fails for length 15.

Should be fixed with b55, please test.