oclHashcat-plus cracks TrueCrypt

[atom]

Hey Guys,

I've some good news! I've justed finished a working oclHashcat-plus version to support cracking TrueCrypt

Everything is written 100% on GPU.

There is no copy overhead to the host, thus your CPU will stay cool at 0% and you can do whatever you want to do with it.

Current implementation is able to crack all the supported hashes:

• RipeMD160
  
• SHA512
  
• Whirlpool
  

So far for the hashes (and thats what makes cracking TrueCrypt slow).

For the ciphers, I'm currently doing only AES and of course the XTS block-cipher.

Serpent and Twofish and there cascaded modes will be added next.

root@sf:~/oclHashcat# ./oclHashcat-plus64.bin -m 6211 test.tc -a 3 ?l?l?l?l?lat -n 32 -u 2000
oclHashcat-plus v0.15 by atom starting...

Hashes: 1 total, 1 unique salts, 1 unique digests
Bitmaps: 8 bits, 256 entries, 0x000000ff mask, 1024 bytes
Workload: 1999 loops, 32 accel
Watchdog: Temperature abort trigger set to 90c
Watchdog: Temperature retain trigger set to 80c
Device #1: Cayman, 1024MB, 830Mhz, 24MCU
Device #2: Cayman, 1024MB, 830Mhz, 24MCU
Device #3: Cayman, 1024MB, 830Mhz, 24MCU
Device #4: Cayman, 1024MB, 830Mhz, 24MCU
Device #1: Kernel ./kernels/4098/m6211.Cayman_1124.2_1124.2.kernel (480220 bytes)
Device #1: Kernel ./kernels/4098/markov_le_plus.Cayman_1124.2_1124.2.kernel (145332 bytes)
Device #2: Kernel ./kernels/4098/m6211.Cayman_1124.2_1124.2.kernel (480220 bytes)
Device #2: Kernel ./kernels/4098/markov_le_plus.Cayman_1124.2_1124.2.kernel (145332 bytes)
Device #3: Kernel ./kernels/4098/m6211.Cayman_1124.2_1124.2.kernel (480220 bytes)
Device #3: Kernel ./kernels/4098/markov_le_plus.Cayman_1124.2_1124.2.kernel (145332 bytes)
Device #4: Kernel ./kernels/4098/m6211.Cayman_1124.2_1124.2.kernel (480220 bytes)
Device #4: Kernel ./kernels/4098/markov_le_plus.Cayman_1124.2_1124.2.kernel (145332 bytes)

test.tc:hashcat

Session.Name...: oclHashcat-plus
Status.........: Cracked
Input.Mode.....: Mask (?l?l?l?l?lat)
Hash.Target....: File (test.tc)
Hash.Type......: TrueCrypt 5.0+ / PBKDF2-HMAC-RipeMD160 / AES
Time.Started...: Fri May 17 14:27:40 2013 (40 secs)
Speed.GPU.#1...: 55788 H/s
Speed.GPU.#2...: 55775 H/s
Speed.GPU.#3...: 55795 H/s
Speed.GPU.#4...: 55797 H/s
Speed.GPU.#*...: 223.2 kH/s
Recovered......: 1/1 (100.00%) Digests, 1/1 (100.00%) Salts
Progress.......: 8749056/11881376 (73.64%)
Rejected.......: 0/8749056 (0.00%)
HWMon.GPU.#1...: 99% Util, 55c Temp, 29% Fan
HWMon.GPU.#2...: 99% Util, 57c Temp, N/A Fan
HWMon.GPU.#3...: 99% Util, 55c Temp, 29% Fan
HWMon.GPU.#4...: 99% Util, 57c Temp, N/A Fan

Started: Fri May 17 14:27:40 2013
Stopped: Fri May 17 14:28:23 2013

Here are some speeds from 2x hd6990:

• PBKDF2-HMAC-RipeMD160 / AES: 223 kHash/s
  
• PBKDF2-HMAC-SHA512 / AES: 95 kHash/s
  
• PBKDF2-HMAC-Whirlpool / AES: 49 kHash/s *updated*
  
• PBKDF2-HMAC-RipeMD160 boot-mode / AES: 451 kHash/s
  

These tests show oclHashcat-plus is new worlds fastest TrueCrypt cracker

While I was implementing the Whirlpool hash I found out how this hash can be optimized by over 50% in raw mode. However, some of these technique can also be used within an PBKDF2-HMAC construct.

PS: I'll explain the technique at my talk on passwordscon 2013 in Las Vegas.

Stay tuned for v0.15 release

--
atom

[unix-ninja]

Nicely done! And congrats!
Is your talk going to be taped?

[blandyuk]

Totally awesome work atom congrats!

[atom]

Thanks guys. I'm not sure if it's taped, question for epixoip

[atom]

The speeds on RipeMD160 will improve again once AMD fixes this confirmed bug:

http://devgurus.amd.com/thread/166396

Currently I have to workaround this which makes the code a little bit slower.

[Hash-IT]

Thank you

[Kuci]

Great ! Well, now is time for release of new oclHashcat, I guess

[atom]

No, I'm working on > length 15 support. If I'm done with that, I will release.

[Kuci]

Oh, that's even better ! Keep up the good work

[epixoip]

I'm not sure if it's taped, question for epixoip

The best way to know is to actually come to the conference, and see if we are videotaping the talks