search

vk496 / mfoc

Mifare Classic Offline Cracker with Hardnested support
GNU General Public License v2.0
88 stars 32 forks source link

Speed up cracking of hardened card by checking for key reuse #5

Closed earthnuker closed 5 years ago

earthnuker commented 5 years ago

Hi,

would it be possible to check the recovered key against all (uncracked) sectors after running a hardnested attack? This would speed up the recovery in case of key-reuse.

Best regards,

Earthnuker

vk496 commented 5 years ago

Hello,

I think this is something that mfoc already have and is used no matter of a nested or hardnested attack.

Please, correct me if I'm wrong.

BR

El 10 de agosto de 2019 17:39:08 CEST, Earthnuker notifications@github.com escribió:

Hi,

would it be possible to check the recovered key against all (uncracked) sectors after running a hardnested attack? This would speed up the recovery in case of key-reuse.

Best regards,

Earthnuker

-- You are receiving this because you are subscribed to this thread. Reply to this email directly or view it on GitHub: https://github.com/vk496/mfoc/issues/5

-- Enviado desde mi dispositivo Android con K-9 Mail. Por favor, disculpa mi brevedad.

earthnuker commented 5 years ago 
       0 |  0? |       0 | Brute force benchmark: 359 million (2^28.4) keys/s      | 140737488355328 |    5d
       0 |  0? |       0 | Using 235 precalculated bitflip state tables            | 140737488355328 |    5d
     348 |  3A |    1358 | Apply bit flip properties                               |     76356050944 |  4min
     354 |  3A |    1359 | Apply Sum property. Sum(a0) = 128                       |     34363326464 |  2min
     357 |  3A |    1362 | Apply bit flip properties                               |     34363326464 |  2min
     380 |  3A |    1362 | Starting brute force...                                 |     34363326464 |  2min
     607 |  3A |    1362 | Brute force phase completed. Key found: aabbccddeeff    |               0 |    0s          

Using AVX2 SIMD core.

 time    | trg | #nonces | Activity                                                | expected to brute force
         |     |         |                                                         | #states         | time
-------------------------------------------------------------------------------------------------------------
       0 |  0? |       0 | Start using 8 threads and AVX2 SIMD core                |                 |
       0 |  0? |       0 | Brute force benchmark: 291 million (2^28.1) keys/s      | 140737488355328 |    6d
       0 |  0? |       0 | Using 235 precalculated bitflip state tables            | 140737488355328 |    6d
     343 |  4A |    1310 | Apply bit flip properties                               |     76356050944 |  4min
     350 |  4A |    1311 | Apply Sum property. Sum(a0) = 128                       |     34363326464 |  2min
     355 |  4A |    1314 | Apply bit flip properties                               |     34363326464 |  2min
     355 |  4A |    1314 | (Ignoring Sum(a8) properties)                           |     34363326464 |  2min
     379 |  4A |    1314 | Starting brute force...                                 |     34363326464 |  2min
     613 |  4A |    1314 | Brute force phase completed. Key found: aabbccddeeff    |               0 |    0s

From the output it looks like it ran hardnested twice without checking for key reuse, so it recovered the same key twice for two different sectors.

When i have time i can implement the improvement and submit a PR, i already have a patch that does something similar but runs libnfc_crypto1_crack instead of the hardnested code from PM3

Best regards,

Earthnuker