arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-22 08:08:10 +00:00
Code Issues Releases Wiki Activity

RC4: Update -m 97x0 and -m 104x0 to new RC4 crypto library code, improving performance by 20% or more

Browse Source
This commit is contained in:
Jens Steube 2021-05-30 15:40:33 +02:00
parent 58467e0869
commit 2a55fd7f33
17 changed files with 444 additions and 997 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

138
OpenCL/inc_cipher_rc4.cl
View File

@ -106,16 +106,7 @@ DECLSPEC void SET_KEY32 (LOCAL_AS u32 *S, const int k, const u32 v)
#endif
DECLSPEC void rc4_swap (LOCAL_AS u32 *S, const u8 i, const u8 j)
{
u8 tmp;
tmp = GET_KEY8 (S, i);
SET_KEY8 (S, i, GET_KEY8 (S, j));
SET_KEY8 (S, j, tmp);
}
DECLSPEC void rc4_init_16 (LOCAL_AS u32 *S, const u32 *data)
DECLSPEC void rc4_init_40 (LOCAL_AS u32 *S, const u32 *key)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
@ -128,11 +119,11 @@ DECLSPEC void rc4_init_16 (LOCAL_AS u32 *S, const u32 *data)
SET_KEY32 (S, i, v); v += a;
}
const u32 d0 = data[0] >> 0;
const u32 d1 = data[0] >> 8;
const u32 d2 = data[0] >> 16;
const u32 d3 = data[0] >> 24;
const u32 d4 = data[1] >> 0;
const u32 d0 = key[0] >> 0;
const u32 d1 = key[0] >> 8;
const u32 d2 = key[0] >> 16;
const u32 d3 = key[0] >> 24;
const u32 d4 = key[1] >> 0;
u32 j = 0;
@ -151,8 +142,71 @@ DECLSPEC void rc4_init_16 (LOCAL_AS u32 *S, const u32 *data)
j += GET_KEY8 (S, 255) + d0; rc4_swap (S, 255, j);
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS u32 *S, u8 i, u8 j, CONSTANT_AS u32a *in, u32 *out)
DECLSPEC void rc4_init_128 (LOCAL_AS u32 *S, const u32 *key)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
SET_KEY32 (S, i, v); v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = key[0];
j += GET_KEY8 (S, idx) + v8a_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8b_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8c_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8d_from_v32_S (v); rc4_swap (S, idx, j); idx++;
v = key[1];
j += GET_KEY8 (S, idx) + v8a_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8b_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8c_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8d_from_v32_S (v); rc4_swap (S, idx, j); idx++;
v = key[2];
j += GET_KEY8 (S, idx) + v8a_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8b_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8c_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8d_from_v32_S (v); rc4_swap (S, idx, j); idx++;
v = key[3];
j += GET_KEY8 (S, idx) + v8a_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8b_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8c_from_v32_S (v); rc4_swap (S, idx, j); idx++;
j += GET_KEY8 (S, idx) + v8d_from_v32_S (v); rc4_swap (S, idx, j); idx++;
}
}
DECLSPEC void rc4_swap (LOCAL_AS u32 *S, const u8 i, const u8 j)
{
u8 tmp;
tmp = GET_KEY8 (S, i);
SET_KEY8 (S, i, GET_KEY8 (S, j));
SET_KEY8 (S, j, tmp);
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS u32 *S, const u8 i, const u8 j, const u32 *in, u32 *out)
{
u8 src = i;
u8 dst = j;
#ifdef _unroll
#pragma unroll
#endif
@ -162,44 +216,54 @@ DECLSPEC u8 rc4_next_16 (LOCAL_AS u32 *S, u8 i, u8 j, CONSTANT_AS u32a *in, u32
u8 idx;
i += 1;
j += GET_KEY8 (S, i);
u32 r;
rc4_swap (S, i, j);
src += 1;
dst += GET_KEY8 (S, src);
idx = GET_KEY8 (S, i) + GET_KEY8 (S, j);
rc4_swap (S, src, dst);
xor4 |= GET_KEY8 (S, idx) << 0;
idx = GET_KEY8 (S, src) + GET_KEY8 (S, dst);
i += 1;
j += GET_KEY8 (S, i);
r = GET_KEY8 (S, idx);
rc4_swap (S, i, j);
xor4 |= r << 0;
idx = GET_KEY8 (S, i) + GET_KEY8 (S, j);
src += 1;
dst += GET_KEY8 (S, src);
xor4 |= GET_KEY8 (S, idx) << 8;
rc4_swap (S, src, dst);
i += 1;
j += GET_KEY8 (S, i);
idx = GET_KEY8 (S, src) + GET_KEY8 (S, dst);
rc4_swap (S, i, j);
r = GET_KEY8 (S, idx);
idx = GET_KEY8 (S, i) + GET_KEY8 (S, j);
xor4 |= r << 8;
xor4 |= GET_KEY8 (S, idx) << 16;
src += 1;
dst += GET_KEY8 (S, src);
i += 1;
j += GET_KEY8 (S, i);
rc4_swap (S, src, dst);
rc4_swap (S, i, j);
idx = GET_KEY8 (S, src) + GET_KEY8 (S, dst);
idx = GET_KEY8 (S, i) + GET_KEY8 (S, j);
r = GET_KEY8 (S, idx);
xor4 |= GET_KEY8 (S, idx) << 24;
xor4 |= r << 16;
src += 1;
dst += GET_KEY8 (S, src);
rc4_swap (S, src, dst);
idx = GET_KEY8 (S, src) + GET_KEY8 (S, dst);
r = GET_KEY8 (S, idx);
xor4 |= r << 24;
out[k] = in[k] ^ xor4;
}
return j;
return dst;
}

7
OpenCL/inc_cipher_rc4.h
View File

@ -10,8 +10,9 @@ DECLSPEC u8 GET_KEY8 (LOCAL_AS u32 *S, const int k);
DECLSPEC void SET_KEY8 (LOCAL_AS u32 *S, const int k, const u8 v);
DECLSPEC void SET_KEY32 (LOCAL_AS u32 *S, const int k, const u32 v);
DECLSPEC void rc4_swap (LOCAL_AS u32 *S, const u8 i, const u8 j);
DECLSPEC void rc4_init_16 (LOCAL_AS u32 *S, const u32 *data);
DECLSPEC u8 rc4_next_16 (LOCAL_AS u32 *S, u8 i, u8 j, CONSTANT_AS u32a *in, u32 *out);
DECLSPEC void rc4_init_40 (LOCAL_AS u32 *S, const u32 *key);
DECLSPEC void rc4_init_128 (LOCAL_AS u32 *S, const u32 *key);
DECLSPEC void rc4_swap (LOCAL_AS u32 *S, const u8 i, const u8 j);
DECLSPEC u8 rc4_next_16 (LOCAL_AS u32 *S, const u8 i, const u8 j, const u32 *in, u32 *out);
#endif // _INC_CIPHER_RC4_H

144
OpenCL/m09700_a0-optimized.cl
View File

@ -15,6 +15,7 @@
#include "inc_rp_optimized.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#include "inc_cipher_rc4.cl"
#endif
typedef struct oldoffice01
@ -26,129 +27,6 @@ typedef struct oldoffice01
} oldoffice01_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
DECLSPEC void gen336 (u32 *digest_pre, u32 *salt_buf, u32 *digest)
{
u32 digest_t0[2];
@ -527,9 +405,7 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* salt
@ -636,11 +512,11 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
key[2] = digest[2];
key[3] = digest[3];
rc4_init_16 (rc4_key, key);
rc4_init_128 (S, key);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -666,7 +542,7 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
}
@ -714,9 +590,7 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* salt
@ -835,11 +709,11 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
key[2] = digest[2];
key[3] = digest[3];
rc4_init_16 (rc4_key, key);
rc4_init_128 (S, key);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -865,7 +739,7 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
}

144
OpenCL/m09700_a1-optimized.cl
View File

@ -13,6 +13,7 @@
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#include "inc_cipher_rc4.cl"
#endif
typedef struct oldoffice01
@ -24,129 +25,6 @@ typedef struct oldoffice01
} oldoffice01_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
DECLSPEC void gen336 (u32 *digest_pre, u32 *salt_buf, u32 *digest)
{
u32 digest_t0[2];
@ -525,9 +403,7 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* salt
@ -692,11 +568,11 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_ESALT (oldoffice01_t))
key[2] = digest[2];
key[3] = digest[3];
rc4_init_16 (rc4_key, key);
rc4_init_128 (S, key);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -722,7 +598,7 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
}
@ -770,9 +646,7 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* salt
@ -949,11 +823,11 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_ESALT (oldoffice01_t))
key[2] = digest[2];
key[3] = digest[3];
rc4_init_16 (rc4_key, key);
rc4_init_128 (S, key);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -979,7 +853,7 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
}

168
OpenCL/m09700_a3-optimized.cl
View File

@ -10,6 +10,7 @@
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#include "inc_cipher_rc4.cl"
#endif
typedef struct oldoffice01
@ -21,130 +22,7 @@ typedef struct oldoffice01
} oldoffice01_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
DECLSPEC void m09700m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
DECLSPEC void m09700m (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
{
/**
* modifier
@ -153,12 +31,6 @@ DECLSPEC void m09700m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* shared
*/
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -514,11 +386,11 @@ DECLSPEC void m09700m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0_t[0] = out[0];
w0_t[1] = out[1];
@ -544,13 +416,13 @@ DECLSPEC void m09700m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
}
}
DECLSPEC void m09700s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
DECLSPEC void m09700s (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
{
/**
* modifier
@ -559,12 +431,6 @@ DECLSPEC void m09700s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* shared
*/
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* salt
*/
@ -932,11 +798,11 @@ DECLSPEC void m09700s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0_t[0] = out[0];
w0_t[1] = out[1];
@ -962,7 +828,7 @@ DECLSPEC void m09700s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
}
@ -1012,9 +878,9 @@ KERNEL_FQ void m09700_m04 (KERN_ATTR_ESALT (oldoffice01_t))
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
m09700m (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
m09700m (S, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
}
KERNEL_FQ void m09700_m08 (KERN_ATTR_ESALT (oldoffice01_t))
@ -1061,9 +927,9 @@ KERNEL_FQ void m09700_m08 (KERN_ATTR_ESALT (oldoffice01_t))
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
m09700m (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
m09700m (S, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
}
KERNEL_FQ void m09700_m16 (KERN_ATTR_ESALT (oldoffice01_t))
@ -1114,9 +980,9 @@ KERNEL_FQ void m09700_s04 (KERN_ATTR_ESALT (oldoffice01_t))
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
m09700s (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
m09700s (S, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
}
KERNEL_FQ void m09700_s08 (KERN_ATTR_ESALT (oldoffice01_t))
@ -1163,9 +1029,9 @@ KERNEL_FQ void m09700_s08 (KERN_ATTR_ESALT (oldoffice01_t))
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
m09700s (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
m09700s (S, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
}
KERNEL_FQ void m09700_s16 (KERN_ATTR_ESALT (oldoffice01_t))

144
OpenCL/m09710_a0-optimized.cl
View File

@ -15,6 +15,7 @@
#include "inc_rp_optimized.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#include "inc_cipher_rc4.cl"
#endif
typedef struct oldoffice01
@ -26,129 +27,6 @@ typedef struct oldoffice01
} oldoffice01_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
KERNEL_FQ void m09710_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
{
/**
@ -183,9 +61,7 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* esalt
@ -243,11 +119,11 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -273,7 +149,7 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
}
@ -321,9 +197,7 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* esalt
@ -393,11 +267,11 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -423,7 +297,7 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_RULES_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
}

144
OpenCL/m09710_a1-optimized.cl
View File

@ -13,6 +13,7 @@
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#include "inc_cipher_rc4.cl"
#endif
typedef struct oldoffice01
@ -24,129 +25,6 @@ typedef struct oldoffice01
} oldoffice01_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
KERNEL_FQ void m09710_m04 (KERN_ATTR_ESALT (oldoffice01_t))
{
/**
@ -181,9 +59,7 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* esalt
@ -287,11 +163,11 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_ESALT (oldoffice01_t))
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -317,7 +193,7 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
}
@ -365,9 +241,7 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_ESALT (oldoffice01_t))
* shared
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
/**
* esalt
@ -483,11 +357,11 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_ESALT (oldoffice01_t))
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0[0] = out[0];
w0[1] = out[1];
@ -513,7 +387,7 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_ESALT (oldoffice01_t))
md5_transform (w0, w1, w2, w3, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
}

160
OpenCL/m09710_a3-optimized.cl
View File

@ -13,6 +13,7 @@
#include "inc_common.cl"
#include "inc_simd.cl"
#include "inc_hash_md5.cl"
#include "inc_cipher_rc4.cl"
#endif
typedef struct oldoffice01
@ -24,130 +25,7 @@ typedef struct oldoffice01
} oldoffice01_t;
typedef struct
{
u8 S[256];
u32 wtf_its_faster;
} RC4_KEY;
DECLSPEC void swap (LOCAL_AS RC4_KEY *rc4_key, const u8 i, const u8 j)
{
u8 tmp;
tmp = rc4_key->S[i];
rc4_key->S[i] = rc4_key->S[j];
rc4_key->S[j] = tmp;
}
DECLSPEC void rc4_init_16 (LOCAL_AS RC4_KEY *rc4_key, const u32 *data)
{
u32 v = 0x03020100;
u32 a = 0x04040404;
LOCAL_AS u32 *ptr = (LOCAL_AS u32 *) rc4_key->S;
#ifdef _unroll
#pragma unroll
#endif
for (u32 i = 0; i < 64; i++)
{
*ptr++ = v; v += a;
}
u32 j = 0;
for (u32 i = 0; i < 16; i++)
{
u32 idx = i * 16;
u32 v;
v = data[0];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[1];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[2];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
v = data[3];
j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
}
}
DECLSPEC u8 rc4_next_16 (LOCAL_AS RC4_KEY *rc4_key, u8 i, u8 j, const u32 *in, u32 *out)
{
#ifdef _unroll
#pragma unroll
#endif
for (u32 k = 0; k < 4; k++)
{
u32 xor4 = 0;
u8 idx;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 0;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 8;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 16;
i += 1;
j += rc4_key->S[i];
swap (rc4_key, i, j);
idx = rc4_key->S[i] + rc4_key->S[j];
xor4 |= rc4_key->S[idx] << 24;
out[k] = in[k] ^ xor4;
}
return j;
}
DECLSPEC void m09710m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
DECLSPEC void m09710m (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
{
/**
* modifier
@ -156,12 +34,6 @@ DECLSPEC void m09710m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* shared
*/
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* esalt
*/
@ -220,11 +92,11 @@ DECLSPEC void m09710m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0_t[0] = out[0];
w0_t[1] = out[1];
@ -250,13 +122,13 @@ DECLSPEC void m09710m (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
}
}
DECLSPEC void m09710s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
DECLSPEC void m09710s (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, const u32 pw_len, KERN_ATTR_ESALT (oldoffice01_t))
{
/**
* modifier
@ -265,12 +137,6 @@ DECLSPEC void m09710s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* shared
*/
LOCAL_AS RC4_KEY *rc4_key = &rc4_keys[lid];
/**
* esalt
*/
@ -341,11 +207,11 @@ DECLSPEC void m09710s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
// now the RC4 part
rc4_init_16 (rc4_key, digest);
rc4_init_128 (S, digest);
u32 out[4];
u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
u8 j = rc4_next_16 (S, 0, 0, encryptedVerifier, out);
w0_t[0] = out[0];
w0_t[1] = out[1];
@ -371,7 +237,7 @@ DECLSPEC void m09710s (LOCAL_AS RC4_KEY *rc4_keys, u32 *w0, u32 *w1, u32 *w2, u3
md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
rc4_next_16 (rc4_key, 16, j, digest, out);
rc4_next_16 (S, 16, j, digest, out);
COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
}
@ -421,9 +287,9 @@ KERNEL_FQ void m09710_m04 (KERN_ATTR_ESALT (oldoffice01_t))
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
m09710m (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
m09710m (S, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
}
KERNEL_FQ void m09710_m08 (KERN_ATTR_ESALT (oldoffice01_t))
@ -478,9 +344,9 @@ KERNEL_FQ void m09710_s04 (KERN_ATTR_ESALT (oldoffice01_t))
* main
*/
LOCAL_VK RC4_KEY rc4_keys[64];
LOCAL_VK u32 S[64 * FIXED_LOCAL_SIZE];
m09710s (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
m09710s (S, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_extra0_buf, d_extra1_buf, d_extra2_buf, d_extra3_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, SALT_POS, loop_pos, loop_cnt, il_cnt, digests_cnt, DIGESTS_OFFSET, combs_mode, salt_repeat, pws_pos, gid_max);
}
KERNEL_FQ void m09710_s08 (KERN_ATTR_ESALT (oldoffice01_t))

48
OpenCL/m10400_a0-optimized.cl
View File

@ -18,18 +18,6 @@
#include "inc_cipher_rc4.cl"
#endif
CONSTANT_VK u32a padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
typedef struct pdf
{
int V;
@ -81,6 +69,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_m04 (KERN_ATTR_RULES_
const u32 pw_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -207,7 +211,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_m04 (KERN_ATTR_RULES_
digest[2] = 0;
digest[3] = 0;
rc4_init_16 (S, digest);
rc4_init_40 (S, digest);
u32 out[4];
@ -255,6 +259,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_s04 (KERN_ATTR_RULES_
const u32 pw_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -393,7 +413,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_s04 (KERN_ATTR_RULES_
digest[2] = 0;
digest[3] = 0;
rc4_init_16 (S, digest);
rc4_init_40 (S, digest);
u32 out[4];

48
OpenCL/m10400_a1-optimized.cl
View File

@ -16,18 +16,6 @@
#include "inc_cipher_rc4.cl"
#endif
CONSTANT_VK u32a padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
typedef struct pdf
{
int V;
@ -79,6 +67,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_m04 (KERN_ATTR_ESALT
const u32 pw_l_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -265,7 +269,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_m04 (KERN_ATTR_ESALT
digest[2] = 0;
digest[3] = 0;
rc4_init_16 (S, digest);
rc4_init_40 (S, digest);
u32 out[4];
@ -313,6 +317,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_s04 (KERN_ATTR_ESALT
const u32 pw_l_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -511,7 +531,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10400_s04 (KERN_ATTR_ESALT
digest[2] = 0;
digest[3] = 0;
rc4_init_16 (S, digest);
rc4_init_40 (S, digest);
u32 out[4];

48
OpenCL/m10400_a3-optimized.cl
View File

@ -16,18 +16,6 @@
#include "inc_cipher_rc4.cl"
#endif
CONSTANT_VK u32a padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
typedef struct pdf
{
int V;
@ -58,6 +46,22 @@ DECLSPEC void m10400m (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* U_buf
*/
@ -199,7 +203,7 @@ DECLSPEC void m10400m (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
digest[2] = 0;
digest[3] = 0;
rc4_init_16 (S, digest);
rc4_init_40 (S, digest);
u32 out[4];
@ -218,6 +222,22 @@ DECLSPEC void m10400s (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* U_buf
*/
@ -371,7 +391,7 @@ DECLSPEC void m10400s (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
digest[2] = 0;
digest[3] = 0;
rc4_init_16 (S, digest);
rc4_init_40 (S, digest);
u32 out[4];

48
OpenCL/m10410_a0-optimized.cl
View File

@ -18,18 +18,6 @@
#include "inc_cipher_rc4.cl"
#endif
CONSTANT_VK u32a padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
typedef struct pdf
{
int V;
@ -81,6 +69,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_m04 (KERN_ATTR_RULES_
const u32 pw_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -104,7 +108,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_m04 (KERN_ATTR_RULES_
* pdf
*/
rc4_init_16 (S, w0);
rc4_init_40 (S, w0);
u32 out[4];
@ -152,6 +156,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_s04 (KERN_ATTR_RULES_
const u32 pw_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -187,7 +207,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_s04 (KERN_ATTR_RULES_
* pdf
*/
rc4_init_16 (S, w0);
rc4_init_40 (S, w0);
u32 out[4];

48
OpenCL/m10410_a1-optimized.cl
View File

@ -16,18 +16,6 @@
#include "inc_cipher_rc4.cl"
#endif
CONSTANT_VK u32a padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
typedef struct pdf
{
int V;
@ -79,6 +67,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_m04 (KERN_ATTR_ESALT
const u32 pw_l_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -171,7 +175,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_m04 (KERN_ATTR_ESALT
* pdf
*/
rc4_init_16 (S, w0);
rc4_init_40 (S, w0);
u32 out[4];
@ -219,6 +223,22 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_s04 (KERN_ATTR_ESALT
const u32 pw_l_len = pws[gid].pw_len & 63;
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* shared
*/
@ -323,7 +343,7 @@ KERNEL_FQ void FIXED_THREAD_COUNT(FIXED_LOCAL_SIZE) m10410_s04 (KERN_ATTR_ESALT
* pdf
*/
rc4_init_16 (S, w0);
rc4_init_40 (S, w0);
u32 out[4];

48
OpenCL/m10410_a3-optimized.cl
View File

@ -16,18 +16,6 @@
#include "inc_cipher_rc4.cl"
#endif
CONSTANT_VK u32a padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
typedef struct pdf
{
int V;
@ -58,6 +46,22 @@ DECLSPEC void m10410m (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* loop
*/
@ -76,7 +80,7 @@ DECLSPEC void m10410m (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
* pdf
*/
rc4_init_16 (S, w0);
rc4_init_40 (S, w0);
u32 out[4];
@ -95,6 +99,22 @@ DECLSPEC void m10410s (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
const u64 gid = get_global_id (0);
const u64 lid = get_local_id (0);
/**
* constant
*/
const u32 padding[8] =
{
0x5e4ebf28,
0x418a754e,
0x564e0064,
0x0801faff,
0xb6002e2e,
0x803e68d0,
0xfea90c2f,
0x7a695364
};
/**
* digest
*/
@ -125,7 +145,7 @@ DECLSPEC void m10410s (LOCAL_AS u32 *S, u32 *w0, u32 *w1, u32 *w2, u32 *w3, cons
* pdf
*/
rc4_init_16 (S, w0);
rc4_init_40 (S, w0);
u32 out[4];

2
docs/changes.txt
View File

@ -30,7 +30,7 @@
- AES Crypt Plugin: Reduced max password length from 256 to 128 which improved performance by 22%
- RAR3-p (Compressed): Fix workaround in unrar library in AES constant table generation to enable multi-threading support
- CRC32: Prevent decompression of data used in CRC32 calculation on host. This leads to false negatives with TrueCrypt/VeraCrypt keyfiles
- PDF: Update -m 10400 and -m 10410 to new RC4 crypto library code, improving performance by 20% or more
- RC4: Update -m 97x0 and -m 104x0 to new RC4 crypto library code, improving performance by 20% or more
##
## Technical

51
src/modules/module_09700.c
View File

@ -57,6 +57,37 @@ static const char *SIGNATURE_OLDOFFICE = "$oldoffice$";
static const char *SIGNATURE_OLDOFFICE0 = "$oldoffice$0";
static const char *SIGNATURE_OLDOFFICE1 = "$oldoffice$1";
char *module_jit_build_options (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra, MAYBE_UNUSED const hashes_t *hashes, MAYBE_UNUSED const hc_device_param_t *device_param)
{
char *jit_build_options = NULL;
u32 native_threads = 0;
if (device_param->opencl_device_type & CL_DEVICE_TYPE_CPU)
{
native_threads = 1;
}
else if (device_param->opencl_device_type & CL_DEVICE_TYPE_GPU)
{
if (device_param->opencl_device_vendor_id == VENDOR_ID_INTEL_SDK)
{
native_threads = 8;
}
else if (device_param->opencl_device_vendor_id == VENDOR_ID_AMD)
{
native_threads = 64;
}
else
{
native_threads = 32;
}
}
hc_asprintf (&jit_build_options, "-D FIXED_LOCAL_SIZE=%u -D _unroll", native_threads);
return jit_build_options;
}
u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u64 esalt_size = (const u64) sizeof (oldoffice01_t);
@ -64,20 +95,6 @@ u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED
return esalt_size;
}
u32 module_kernel_threads_min (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u32 kernel_threads_min = 64; // RC4
return kernel_threads_min;
}
u32 module_kernel_threads_max (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u32 kernel_threads_max = 64; // RC4
return kernel_threads_max;
}
u32 module_pw_max (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u32 pw_max = 15; // https://msdn.microsoft.com/en-us/library/dd772916(v=office.12).aspx
@ -255,14 +272,14 @@ void module_init (module_ctx_t *module_ctx)
module_ctx->module_hook23 = MODULE_DEFAULT;
module_ctx->module_hook_salt_size = MODULE_DEFAULT;
module_ctx->module_hook_size = MODULE_DEFAULT;
module_ctx->module_jit_build_options = MODULE_DEFAULT;
module_ctx->module_jit_build_options = module_jit_build_options;
module_ctx->module_jit_cache_disable = MODULE_DEFAULT;
module_ctx->module_kernel_accel_max = MODULE_DEFAULT;
module_ctx->module_kernel_accel_min = MODULE_DEFAULT;
module_ctx->module_kernel_loops_max = MODULE_DEFAULT;
module_ctx->module_kernel_loops_min = MODULE_DEFAULT;
module_ctx->module_kernel_threads_max = module_kernel_threads_max;
module_ctx->module_kernel_threads_min = module_kernel_threads_min;
module_ctx->module_kernel_threads_max = MODULE_DEFAULT;
module_ctx->module_kernel_threads_min = MODULE_DEFAULT;
module_ctx->module_kern_type = module_kern_type;
module_ctx->module_kern_type_dynamic = MODULE_DEFAULT;
module_ctx->module_opti_type = module_opti_type;

51
src/modules/module_09710.c
View File

@ -57,6 +57,37 @@ static const char *SIGNATURE_OLDOFFICE = "$oldoffice$";
static const char *SIGNATURE_OLDOFFICE0 = "$oldoffice$0";
static const char *SIGNATURE_OLDOFFICE1 = "$oldoffice$1";
char *module_jit_build_options (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra, MAYBE_UNUSED const hashes_t *hashes, MAYBE_UNUSED const hc_device_param_t *device_param)
{
char *jit_build_options = NULL;
u32 native_threads = 0;
if (device_param->opencl_device_type & CL_DEVICE_TYPE_CPU)
{
native_threads = 1;
}
else if (device_param->opencl_device_type & CL_DEVICE_TYPE_GPU)
{
if (device_param->opencl_device_vendor_id == VENDOR_ID_INTEL_SDK)
{
native_threads = 8;
}
else if (device_param->opencl_device_vendor_id == VENDOR_ID_AMD)
{
native_threads = 64;
}
else
{
native_threads = 32;
}
}
hc_asprintf (&jit_build_options, "-D FIXED_LOCAL_SIZE=%u -D _unroll", native_threads);
return jit_build_options;
}
u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u64 esalt_size = (const u64) sizeof (oldoffice01_t);
@ -64,20 +95,6 @@ u64 module_esalt_size (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED
return esalt_size;
}
u32 module_kernel_threads_min (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u32 kernel_threads_min = 64; // RC4
return kernel_threads_min;
}
u32 module_kernel_threads_max (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u32 kernel_threads_max = 64; // RC4
return kernel_threads_max;
}
u32 module_pw_min (MAYBE_UNUSED const hashconfig_t *hashconfig, MAYBE_UNUSED const user_options_t *user_options, MAYBE_UNUSED const user_options_extra_t *user_options_extra)
{
const u32 pw_min = 5; // RC4-40 fixed
@ -274,14 +291,14 @@ void module_init (module_ctx_t *module_ctx)
module_ctx->module_hook23 = MODULE_DEFAULT;
module_ctx->module_hook_salt_size = MODULE_DEFAULT;
module_ctx->module_hook_size = MODULE_DEFAULT;
module_ctx->module_jit_build_options = MODULE_DEFAULT;
module_ctx->module_jit_build_options = module_jit_build_options;
module_ctx->module_jit_cache_disable = MODULE_DEFAULT;
module_ctx->module_kernel_accel_max = MODULE_DEFAULT;
module_ctx->module_kernel_accel_min = MODULE_DEFAULT;
module_ctx->module_kernel_loops_max = MODULE_DEFAULT;
module_ctx->module_kernel_loops_min = MODULE_DEFAULT;
module_ctx->module_kernel_threads_max = module_kernel_threads_max;
module_ctx->module_kernel_threads_min = module_kernel_threads_min;
module_ctx->module_kernel_threads_max = MODULE_DEFAULT;
module_ctx->module_kernel_threads_min = MODULE_DEFAULT;
module_ctx->module_kern_type = module_kern_type;
module_ctx->module_kern_type_dynamic = MODULE_DEFAULT;
module_ctx->module_opti_type = module_opti_type;