hashcat
advanced password recovery

siddhartha · (This post was last modified: 10-01-2021, 05:13 PM by siddhartha.)

I'm having trouble with this one. I have an ancient piece of software that takes a user password, manipulates it, then calculates the SHA1 hash. The manipulation is as follows:

Repeat the uppercase alphabet from A to Z until we have a string of length 255 bytes.
Append a null char so the string is now 256 bytes.
Overwrite the first n bytes of the string (string[0]) with the user password of length n.
Overwrite the middle n bytes starting at string[29] with the user password.
Calculate the SHA1 hash of the 256 byte string.

I've created a hex charset mask that follows everything I need, but it's the repeating password inside the string that is tripping me up. With the mask attack an 8 char password is suddenly 16 chars to hashcat. Is there a way to crack this without writing a module? Thanks!

siddhartha · 10-07-2021, 05:21 PM

I was able to write a functioning mode 0 pure in OpenCL. The module is the exact same as SHA1 (00100). It still manages 300+MH/s on my 5600XT.

Code:
/**
* Author......: See docs/credits.txt
* License.....: MIT
*/

//#define NEW_SIMD_CODE

#ifdef KERNEL_STATIC
#include "inc_vendor.h"
#include "inc_types.h"
#include "inc_platform.cl"
#include "inc_common.cl"
#include "inc_rp.h"
#include "inc_rp.cl"
#include "inc_scalar.cl"
#include "inc_hash_sha1.cl"
#endif

#define PAD_LEN 256

KERNEL_FQ void m10240_mxx (KERN_ATTR_RULES ())
{
  /**
  * modifier
  */

  const u64 lid = get_local_id (0);
  const u64 gid = get_global_id (0);

  if (gid >= gid_max) return;
  
  u8 pad[PAD_LEN] = {
    'A','B','C','D','E','F','G','H','I','J','K','L','M','N','O','P',
    'Q','R','S','T','U','V','W','X','Y','Z','A','B','C','D','E','F',
    'G','H','I','J','K','L','M','N','O','P','Q','R','S','T','U','V',
    'W','X','Y','Z','A','B','C','D','E','F','G','H','I','J','K','L',
    'M','N','O','P','Q','R','S','T','U','V','W','X','Y','Z','A','B',
    'C','D','E','F','G','H','I','J','K','L','M','N','O','P','Q','R',
    'S','T','U','V','W','X','Y','Z','A','B','C','D','E','F','G','H',
    'I','J','K','L','M','N','O','P','Q','R','S','T','U','V','W','X',
    'Y','Z','A','B','C','D','E','F','G','H','I','J','K','L','M','N',
    'O','P','Q','R','S','T','U','V','W','X','Y','Z','A','B','C','D',
    'E','F','G','H','I','J','K','L','M','N','O','P','Q','R','S','T',
    'U','V','W','X','Y','Z','A','B','C','D','E','F','G','H','I','J',
    'K','L','M','N','O','P','Q','R','S','T','U','V','W','X','Y','Z',
    'A','B','C','D','E','F','G','H','I','J','K','L','M','N','O','P',
    'Q','R','S','T','U','V','W','X','Y','Z','A','B','C','D','E','F',
    'G','H','I','J','K','L','M','N','O','P','Q','R','S','T','U',0
  };

  /**
  * base
  */

  COPY_PW (pws[gid]);

  /**
  * loop
  */

  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
  {
    pw_t tmp = PASTE_PW;

    tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len);

    sha1_ctx_t ctx;

    sha1_init (&ctx);
    
    u8 *bytes = &tmp.i;
    
    for (u32 i = 0 ; i < tmp.pw_len ; ++i) {
        pad[i] = bytes[i];
        pad[i + 39] = bytes[i];        
    }  
  
    for (u32 i = 0 ; i < 4 ; ++i) {
        u32 buf[16] = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0};
        
        buf[0] = (pad[(i*64)] << 24) + (pad[1+(i*64)] << 16) + (pad[2+(i*64)] << 8) + (pad[3+(i*64)]);
        buf[1] = (pad[4+(i*64)] << 24) + (pad[5+(i*64)] << 16) + (pad[6+(i*64)] << 8) + (pad[7+(i*64)]);  
        buf[2] = (pad[8+(i*64)] << 24) + (pad[9+(i*64)] << 16) + (pad[10+(i*64)] << 8) + (pad[11+(i*64)]);  
        buf[3] = (pad[12+(i*64)] << 24) + (pad[13+(i*64)] << 16) + (pad[14+(i*64)] << 8) + (pad[15+(i*64)]);  
        buf[4] = (pad[16+(i*64)] << 24) + (pad[17+(i*64)] << 16) + (pad[18+(i*64)] << 8) + (pad[19+(i*64)]);  
        buf[5] = (pad[20+(i*64)] << 24) + (pad[21+(i*64)] << 16) + (pad[22+(i*64)] << 8) + (pad[23+(i*64)]);  
        buf[6] = (pad[24+(i*64)] << 24) + (pad[25+(i*64)] << 16) + (pad[26+(i*64)] << 8) + (pad[27+(i*64)]);  
        buf[7] = (pad[28+(i*64)] << 24) + (pad[29+(i*64)] << 16) + (pad[30+(i*64)] << 8) + (pad[31+(i*64)]);  
        buf[8] = (pad[32+(i*64)] << 24) + (pad[33+(i*64)] << 16) + (pad[34+(i*64)] << 8) + (pad[35+(i*64)]);  
        buf[9] = (pad[36+(i*64)] << 24) + (pad[37+(i*64)] << 16) + (pad[38+(i*64)] << 8) + (pad[39+(i*64)]);  
        buf[10] = (pad[40+(i*64)] << 24) + (pad[41+(i*64)] << 16) + (pad[42+(i*64)] << 8) + (pad[43+(i*64)]);  
        buf[11] = (pad[44+(i*64)] << 24) + (pad[45+(i*64)] << 16) + (pad[46+(i*64)] << 8) + (pad[47+(i*64)]);  
        buf[12] = (pad[48+(i*64)] << 24) + (pad[49+(i*64)] << 16) + (pad[50+(i*64)] << 8) + (pad[51+(i*64)]);  
        buf[13] = (pad[52+(i*64)] << 24) + (pad[53+(i*64)] << 16) + (pad[54+(i*64)] << 8) + (pad[55+(i*64)]);  
        buf[14] = (pad[56+(i*64)] << 24) + (pad[57+(i*64)] << 16) + (pad[58+(i*64)] << 8) + (pad[59+(i*64)]);  
        buf[15] = (pad[60+(i*64)] << 24) + (pad[61+(i*64)] << 16) + (pad[62+(i*64)] << 8) + (pad[63+(i*64)]);        
        
        sha1_update(&ctx, buf, 64);
    }
    
    sha1_final (&ctx);

    const u32 r0 = ctx.h[DGST_R0];
    const u32 r1 = ctx.h[DGST_R1];
    const u32 r2 = ctx.h[DGST_R2];
    const u32 r3 = ctx.h[DGST_R3];

    COMPARE_M_SCALAR (r0, r1, r2, r3);
  }
}

KERNEL_FQ void m10240_sxx (KERN_ATTR_RULES ())
{
  /**
  * modifier
  */

  const u64 lid = get_local_id (0);
  const u64 gid = get_global_id (0);

  if (gid >= gid_max) return;
  
  u8 pad[PAD_LEN] = {
    'A','B','C','D','E','F','G','H','I','J','K','L','M','N','O','P',
    'Q','R','S','T','U','V','W','X','Y','Z','A','B','C','D','E','F',
    'G','H','I','J','K','L','M','N','O','P','Q','R','S','T','U','V',
    'W','X','Y','Z','A','B','C','D','E','F','G','H','I','J','K','L',
    'M','N','O','P','Q','R','S','T','U','V','W','X','Y','Z','A','B',
    'C','D','E','F','G','H','I','J','K','L','M','N','O','P','Q','R',
    'S','T','U','V','W','X','Y','Z','A','B','C','D','E','F','G','H',
    'I','J','K','L','M','N','O','P','Q','R','S','T','U','V','W','X',
    'Y','Z','A','B','C','D','E','F','G','H','I','J','K','L','M','N',
    'O','P','Q','R','S','T','U','V','W','X','Y','Z','A','B','C','D',
    'E','F','G','H','I','J','K','L','M','N','O','P','Q','R','S','T',
    'U','V','W','X','Y','Z','A','B','C','D','E','F','G','H','I','J',
    'K','L','M','N','O','P','Q','R','S','T','U','V','W','X','Y','Z',
    'A','B','C','D','E','F','G','H','I','J','K','L','M','N','O','P',
    'Q','R','S','T','U','V','W','X','Y','Z','A','B','C','D','E','F',
    'G','H','I','J','K','L','M','N','O','P','Q','R','S','T','U',0
  };
  
  /**
  * digest
  */

  const u32 search[4] =
  {
    digests_buf[DIGESTS_OFFSET].digest_buf[DGST_R0],
    digests_buf[DIGESTS_OFFSET].digest_buf[DGST_R1],
    digests_buf[DIGESTS_OFFSET].digest_buf[DGST_R2],
    digests_buf[DIGESTS_OFFSET].digest_buf[DGST_R3]
  };

  /**
  * base
  */

  COPY_PW (pws[gid]);

  /**
  * loop
  */

  for (u32 il_pos = 0; il_pos < il_cnt; il_pos++)
  {
    pw_t tmp = PASTE_PW;

    tmp.pw_len = apply_rules (rules_buf[il_pos].cmds, tmp.i, tmp.pw_len);

    sha1_ctx_t ctx;

    sha1_init (&ctx);
    
    u8 *bytes = &tmp.i;
    
    for (u32 i = 0 ; i < tmp.pw_len ; ++i) {
        pad[i] = bytes[i];
        pad[i + 39] = bytes[i];        
    }  
  
    for (u32 i = 0 ; i < 4 ; ++i) {
        u32 buf[16] = {0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0};
        
        buf[0] = (pad[(i*64)] << 24) + (pad[1+(i*64)] << 16) + (pad[2+(i*64)] << 8) + (pad[3+(i*64)]);
        buf[1] = (pad[4+(i*64)] << 24) + (pad[5+(i*64)] << 16) + (pad[6+(i*64)] << 8) + (pad[7+(i*64)]);  
        buf[2] = (pad[8+(i*64)] << 24) + (pad[9+(i*64)] << 16) + (pad[10+(i*64)] << 8) + (pad[11+(i*64)]);  
        buf[3] = (pad[12+(i*64)] << 24) + (pad[13+(i*64)] << 16) + (pad[14+(i*64)] << 8) + (pad[15+(i*64)]);  
        buf[4] = (pad[16+(i*64)] << 24) + (pad[17+(i*64)] << 16) + (pad[18+(i*64)] << 8) + (pad[19+(i*64)]);  
        buf[5] = (pad[20+(i*64)] << 24) + (pad[21+(i*64)] << 16) + (pad[22+(i*64)] << 8) + (pad[23+(i*64)]);  
        buf[6] = (pad[24+(i*64)] << 24) + (pad[25+(i*64)] << 16) + (pad[26+(i*64)] << 8) + (pad[27+(i*64)]);  
        buf[7] = (pad[28+(i*64)] << 24) + (pad[29+(i*64)] << 16) + (pad[30+(i*64)] << 8) + (pad[31+(i*64)]);  
        buf[8] = (pad[32+(i*64)] << 24) + (pad[33+(i*64)] << 16) + (pad[34+(i*64)] << 8) + (pad[35+(i*64)]);  
        buf[9] = (pad[36+(i*64)] << 24) + (pad[37+(i*64)] << 16) + (pad[38+(i*64)] << 8) + (pad[39+(i*64)]);  
        buf[10] = (pad[40+(i*64)] << 24) + (pad[41+(i*64)] << 16) + (pad[42+(i*64)] << 8) + (pad[43+(i*64)]);  
        buf[11] = (pad[44+(i*64)] << 24) + (pad[45+(i*64)] << 16) + (pad[46+(i*64)] << 8) + (pad[47+(i*64)]);  
        buf[12] = (pad[48+(i*64)] << 24) + (pad[49+(i*64)] << 16) + (pad[50+(i*64)] << 8) + (pad[51+(i*64)]);  
        buf[13] = (pad[52+(i*64)] << 24) + (pad[53+(i*64)] << 16) + (pad[54+(i*64)] << 8) + (pad[55+(i*64)]);  
        buf[14] = (pad[56+(i*64)] << 24) + (pad[57+(i*64)] << 16) + (pad[58+(i*64)] << 8) + (pad[59+(i*64)]);  
        buf[15] = (pad[60+(i*64)] << 24) + (pad[61+(i*64)] << 16) + (pad[62+(i*64)] << 8) + (pad[63+(i*64)]);
        
        sha1_update(&ctx, buf, 64);
    }
    
    sha1_final (&ctx);

    const u32 r0 = ctx.h[DGST_R0];
    const u32 r1 = ctx.h[DGST_R1];
    const u32 r2 = ctx.h[DGST_R2];
    const u32 r3 = ctx.h[DGST_R3];

    COMPARE_S_SCALAR (r0, r1, r2, r3);
  }
}

siddhartha · 10-15-2021, 05:09 AM

Do you think it would be possible to create an optimized kernel of this, even though the password is modified within the kernel to 256 bytes?

Login
Username/Email:
Password:	Lost Password?
	Remember me

hashcat advanced password recovery

hashcat
advanced password recovery