1
0
mirror of https://github.com/hashcat/hashcat.git synced 2024-12-27 08:58:10 +00:00
hashcat/nv/m00200_a0.cu
2015-12-04 15:47:52 +01:00

362 lines
11 KiB
Plaintext

/**
* Author......: Jens Steube <jens.steube@gmail.com>
* License.....: MIT
*/
#define _MYSQL323_
#include "include/constants.h"
#include "include/kernel_vendor.h"
#ifdef VLIW1
#define VECT_SIZE1
#endif
#ifdef VLIW2
#define VECT_SIZE1
#endif
#define DGST_R0 0
#define DGST_R1 1
#define DGST_R2 2
#define DGST_R3 3
#include "include/kernel_functions.c"
#include "types_nv.c"
#include "common_nv.c"
#include "include/rp_gpu.h"
#include "rp_nv.c"
#ifdef VECT_SIZE1
#define VECT_COMPARE_S "check_single_vect1_comp4.c"
#define VECT_COMPARE_M "check_multi_vect1_comp4.c"
#endif
#ifdef VECT_SIZE2
#define VECT_COMPARE_S "check_single_vect2_comp4.c"
#define VECT_COMPARE_M "check_multi_vect2_comp4.c"
#endif
#ifdef VECT_SIZE4
#define VECT_COMPARE_S "check_single_vect4_comp4.c"
#define VECT_COMPARE_M "check_multi_vect4_comp4.c"
#endif
__device__ __constant__ gpu_rule_t c_rules[1024];
extern "C" __global__ void __launch_bounds__ (256, 1) m00200_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
/**
* modifier
*/
const u32 lid = threadIdx.x;
/**
* base
*/
const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
if (gid >= gid_max) return;
u32x pw_buf0[4];
pw_buf0[0] = pws[gid].i[ 0];
pw_buf0[1] = pws[gid].i[ 1];
pw_buf0[2] = pws[gid].i[ 2];
pw_buf0[3] = pws[gid].i[ 3];
u32x pw_buf1[4];
pw_buf1[0] = pws[gid].i[ 4];
pw_buf1[1] = pws[gid].i[ 5];
pw_buf1[2] = pws[gid].i[ 6];
pw_buf1[3] = pws[gid].i[ 7];
const u32 pw_len = pws[gid].pw_len;
/**
* loop
*/
for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
{
u32x w0[4];
w0[0] = pw_buf0[0];
w0[1] = pw_buf0[1];
w0[2] = pw_buf0[2];
w0[3] = pw_buf0[3];
u32x w1[4];
w1[0] = pw_buf1[0];
w1[1] = pw_buf1[1];
w1[2] = pw_buf1[2];
w1[3] = pw_buf1[3];
u32x w2[4];
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
u32x w3[4];
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
const u32 out_len = apply_rules (c_rules[il_pos].cmds, w0, w1, pw_len);
u32x w_t[16];
w_t[ 0] = w0[0];
w_t[ 1] = w0[1];
w_t[ 2] = w0[2];
w_t[ 3] = w0[3];
w_t[ 4] = w1[0];
w_t[ 5] = w1[1];
w_t[ 6] = w1[2];
w_t[ 7] = w1[3];
w_t[ 8] = 0;
w_t[ 9] = 0;
w_t[10] = 0;
w_t[11] = 0;
w_t[12] = 0;
w_t[13] = 0;
w_t[14] = 0;
w_t[15] = 0;
u32x a = MYSQL323_A;
u32x b = MYSQL323_B;
u32x add = 7;
#define ROUND(v) \
{ \
a ^= (((a & 0x3f) + add) * (v)) + (a << 8); \
b += (b << 8) ^ a; \
add += v; \
}
int i;
int j;
for (i = 0, j = 0; i <= (int) out_len - 4; i += 4, j += 1)
{
const u32x wj = w_t[j];
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND ((wj >> 24) & 0xff);
}
const u32x wj = w_t[j];
const u32 left = out_len - i;
if (left == 3)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
}
else if (left == 2)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
}
else if (left == 1)
{
ROUND ((wj >> 0) & 0xff);
}
a &= 0x7fffffff;
b &= 0x7fffffff;
const u32x r0 = a;
const u32x r1 = b;
const u32x r2 = 0;
const u32x r3 = 0;
#include VECT_COMPARE_M
}
}
extern "C" __global__ void __launch_bounds__ (256, 1) m00200_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
}
extern "C" __global__ void __launch_bounds__ (256, 1) m00200_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
}
extern "C" __global__ void __launch_bounds__ (256, 1) m00200_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
/**
* modifier
*/
const u32 lid = threadIdx.x;
/**
* base
*/
const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
if (gid >= gid_max) return;
u32x pw_buf0[4];
pw_buf0[0] = pws[gid].i[ 0];
pw_buf0[1] = pws[gid].i[ 1];
pw_buf0[2] = pws[gid].i[ 2];
pw_buf0[3] = pws[gid].i[ 3];
u32x pw_buf1[4];
pw_buf1[0] = pws[gid].i[ 4];
pw_buf1[1] = pws[gid].i[ 5];
pw_buf1[2] = pws[gid].i[ 6];
pw_buf1[3] = pws[gid].i[ 7];
const u32 pw_len = pws[gid].pw_len;
/**
* 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]
};
/**
* loop
*/
for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
{
u32x w0[4];
w0[0] = pw_buf0[0];
w0[1] = pw_buf0[1];
w0[2] = pw_buf0[2];
w0[3] = pw_buf0[3];
u32x w1[4];
w1[0] = pw_buf1[0];
w1[1] = pw_buf1[1];
w1[2] = pw_buf1[2];
w1[3] = pw_buf1[3];
u32x w2[4];
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
u32x w3[4];
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
const u32 out_len = apply_rules (c_rules[il_pos].cmds, w0, w1, pw_len);
u32x w_t[16];
w_t[ 0] = w0[0];
w_t[ 1] = w0[1];
w_t[ 2] = w0[2];
w_t[ 3] = w0[3];
w_t[ 4] = w1[0];
w_t[ 5] = w1[1];
w_t[ 6] = w1[2];
w_t[ 7] = w1[3];
w_t[ 8] = 0;
w_t[ 9] = 0;
w_t[10] = 0;
w_t[11] = 0;
w_t[12] = 0;
w_t[13] = 0;
w_t[14] = 0;
w_t[15] = 0;
u32x a = MYSQL323_A;
u32x b = MYSQL323_B;
u32x add = 7;
#define ROUND(v) \
{ \
a ^= (((a & 0x3f) + add) * (v)) + (a << 8); \
b += (b << 8) ^ a; \
add += v; \
}
int i;
int j;
for (i = 0, j = 0; i <= (int) out_len - 4; i += 4, j += 1)
{
const u32x wj = w_t[j];
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
ROUND ((wj >> 24) & 0xff);
}
const u32x wj = w_t[j];
const u32 left = out_len - i;
if (left == 3)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
ROUND ((wj >> 16) & 0xff);
}
else if (left == 2)
{
ROUND ((wj >> 0) & 0xff);
ROUND ((wj >> 8) & 0xff);
}
else if (left == 1)
{
ROUND ((wj >> 0) & 0xff);
}
a &= 0x7fffffff;
b &= 0x7fffffff;
const u32x r0 = a;
const u32x r1 = b;
const u32x r2 = 0;
const u32x r3 = 0;
#include VECT_COMPARE_S
}
}
extern "C" __global__ void __launch_bounds__ (256, 1) m00200_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
}
extern "C" __global__ void __launch_bounds__ (256, 1) m00200_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, u32 *d_scryptV_buf, const u32 bitmap_mask, const u32 bitmap_shift1, const u32 bitmap_shift2, const u32 salt_pos, const u32 loop_pos, const u32 loop_cnt, const u32 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
}