arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-14 03:39:09 +00:00
Code Issues Releases Wiki Activity

Converted to new SIMD: -m 3800 -a 0

Browse Source
This commit is contained in:
Jens Steube 2016-02-05 21:53:47 +01:00
parent de79ce8958
commit 53a31f840e
1 changed files with 41 additions and 100 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

141
OpenCL/m03800_a0.cl
View File

@ -5,6 +5,8 @@
#define _MD5_ #define _MD5_
#define NEW_SIMD_CODE
#include "include/constants.h" #include "include/constants.h"
#include "include/kernel_vendor.h" #include "include/kernel_vendor.h"
@ -18,9 +20,7 @@
#include "OpenCL/common.c" #include "OpenCL/common.c"
#include "include/rp_kernel.h" #include "include/rp_kernel.h"
#include "OpenCL/rp.c" #include "OpenCL/rp.c"
#include "OpenCL/simd.c"
#define COMPARE_S "OpenCL/check_single_comp4.c"
#define COMPARE_M "OpenCL/check_multi_comp4.c"
__kernel void m03800_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global 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) __kernel void m03800_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global void *esalt_bufs, __global u32 *d_return_buf, __global 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)
{ {
@ -90,67 +90,43 @@ __kernel void m03800_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
* loop * loop
*/ */
for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++) for (u32 il_pos = 0; il_pos < rules_cnt; il_pos += VECT_SIZE)
{ {
u32 w0[4]; u32x w0[4] = { 0 };
u32x w1[4] = { 0 };
u32x w2[4] = { 0 };
u32x w3[4] = { 0 };
w0[0] = pw_buf0[0]; const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
w0[1] = pw_buf0[1];
w0[2] = pw_buf0[2];
w0[3] = pw_buf0[3];
u32 w1[4]; u32x w0_t[4];
w1[0] = pw_buf1[0];
w1[1] = pw_buf1[1];
w1[2] = pw_buf1[2];
w1[3] = pw_buf1[3];
u32 w2[4];
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
u32 w3[4];
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
u32 w0_t[4];
w0_t[0] = w0[0]; w0_t[0] = w0[0];
w0_t[1] = w0[1]; w0_t[1] = w0[1];
w0_t[2] = w0[2]; w0_t[2] = w0[2];
w0_t[3] = w0[3]; w0_t[3] = w0[3];
u32 w1_t[4]; u32x w1_t[4];
w1_t[0] = w1[0]; w1_t[0] = w1[0];
w1_t[1] = w1[1]; w1_t[1] = w1[1];
w1_t[2] = w1[2]; w1_t[2] = w1[2];
w1_t[3] = w1[3]; w1_t[3] = w1[3];
u32 w2_t[4]; u32x w2_t[4];
w2_t[0] = w2[0]; w2_t[0] = w2[0];
w2_t[1] = w2[1]; w2_t[1] = w2[1];
w2_t[2] = w2[2]; w2_t[2] = w2[2];
w2_t[3] = w2[3]; w2_t[3] = w2[3];
u32 w3_t[4]; u32x w3_t[4];
w3_t[0] = w3[0]; w3_t[0] = w3[0];
w3_t[1] = w3[1]; w3_t[1] = w3[1];
w3_t[2] = w3[2]; w3_t[2] = w3[2];
w3_t[3] = w3[3]; w3_t[3] = w3[3];
/** /**
* prepend salt * prepend salt
*/ */
@ -178,28 +154,28 @@ __kernel void m03800_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
* append salt * append salt
*/ */
u32 s0[4]; u32x s0[4];
s0[0] = salt_buf0[0]; s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1]; s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2]; s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3]; s0[3] = salt_buf0[3];
u32 s1[4]; u32x s1[4];
s1[0] = salt_buf1[0]; s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1]; s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2]; s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3]; s1[3] = salt_buf1[3];
u32 s2[4]; u32x s2[4];
s2[0] = 0; s2[0] = 0;
s2[1] = 0; s2[1] = 0;
s2[2] = 0; s2[2] = 0;
s2[3] = 0; s2[3] = 0;
u32 s3[4]; u32x s3[4];
s3[0] = 0; s3[0] = 0;
s3[1] = 0; s3[1] = 0;
@ -235,10 +211,10 @@ __kernel void m03800_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
* md5 * md5
*/ */
u32 a = MD5M_A; u32x a = MD5M_A;
u32 b = MD5M_B; u32x b = MD5M_B;
u32 c = MD5M_C; u32x c = MD5M_C;
u32 d = MD5M_D; u32x d = MD5M_D;
MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00); MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01); MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01);
@ -304,17 +280,11 @@ __kernel void m03800_m04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
MD5_STEP (MD5_I , c, d, a, b, w1_t[2], MD5C3a, MD5S32); MD5_STEP (MD5_I , c, d, a, b, w1_t[2], MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w3_t[1], MD5C3b, MD5S33); MD5_STEP (MD5_I , b, c, d, a, w3_t[1], MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w1_t[0], MD5C3c, MD5S30); MD5_STEP (MD5_I , a, b, c, d, w1_t[0], MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w2_t[3], MD5C3d, MD5S31); MD5_STEP (MD5_I , d, a, b, c, w2_t[3], MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33);
const u32 r0 = a; COMPARE_M_SIMD (a, d, c, b);
const u32 r1 = d;
const u32 r2 = c;
const u32 r3 = b;
#include COMPARE_M
} }
} }
@ -406,67 +376,43 @@ __kernel void m03800_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
* loop * loop
*/ */
for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++) for (u32 il_pos = 0; il_pos < rules_cnt; il_pos += VECT_SIZE)
{ {
u32 w0[4]; u32x w0[4] = { 0 };
u32x w1[4] = { 0 };
u32x w2[4] = { 0 };
u32x w3[4] = { 0 };
w0[0] = pw_buf0[0]; const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
w0[1] = pw_buf0[1];
w0[2] = pw_buf0[2];
w0[3] = pw_buf0[3];
u32 w1[4]; u32x w0_t[4];
w1[0] = pw_buf1[0];
w1[1] = pw_buf1[1];
w1[2] = pw_buf1[2];
w1[3] = pw_buf1[3];
u32 w2[4];
w2[0] = 0;
w2[1] = 0;
w2[2] = 0;
w2[3] = 0;
u32 w3[4];
w3[0] = 0;
w3[1] = 0;
w3[2] = 0;
w3[3] = 0;
const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
u32 w0_t[4];
w0_t[0] = w0[0]; w0_t[0] = w0[0];
w0_t[1] = w0[1]; w0_t[1] = w0[1];
w0_t[2] = w0[2]; w0_t[2] = w0[2];
w0_t[3] = w0[3]; w0_t[3] = w0[3];
u32 w1_t[4]; u32x w1_t[4];
w1_t[0] = w1[0]; w1_t[0] = w1[0];
w1_t[1] = w1[1]; w1_t[1] = w1[1];
w1_t[2] = w1[2]; w1_t[2] = w1[2];
w1_t[3] = w1[3]; w1_t[3] = w1[3];
u32 w2_t[4]; u32x w2_t[4];
w2_t[0] = w2[0]; w2_t[0] = w2[0];
w2_t[1] = w2[1]; w2_t[1] = w2[1];
w2_t[2] = w2[2]; w2_t[2] = w2[2];
w2_t[3] = w2[3]; w2_t[3] = w2[3];
u32 w3_t[4]; u32x w3_t[4];
w3_t[0] = w3[0]; w3_t[0] = w3[0];
w3_t[1] = w3[1]; w3_t[1] = w3[1];
w3_t[2] = w3[2]; w3_t[2] = w3[2];
w3_t[3] = w3[3]; w3_t[3] = w3[3];
/** /**
* prepend salt * prepend salt
*/ */
@ -494,28 +440,28 @@ __kernel void m03800_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
* append salt * append salt
*/ */
u32 s0[4]; u32x s0[4];
s0[0] = salt_buf0[0]; s0[0] = salt_buf0[0];
s0[1] = salt_buf0[1]; s0[1] = salt_buf0[1];
s0[2] = salt_buf0[2]; s0[2] = salt_buf0[2];
s0[3] = salt_buf0[3]; s0[3] = salt_buf0[3];
u32 s1[4]; u32x s1[4];
s1[0] = salt_buf1[0]; s1[0] = salt_buf1[0];
s1[1] = salt_buf1[1]; s1[1] = salt_buf1[1];
s1[2] = salt_buf1[2]; s1[2] = salt_buf1[2];
s1[3] = salt_buf1[3]; s1[3] = salt_buf1[3];
u32 s2[4]; u32x s2[4];
s2[0] = 0; s2[0] = 0;
s2[1] = 0; s2[1] = 0;
s2[2] = 0; s2[2] = 0;
s2[3] = 0; s2[3] = 0;
u32 s3[4]; u32x s3[4];
s3[0] = 0; s3[0] = 0;
s3[1] = 0; s3[1] = 0;
@ -551,10 +497,10 @@ __kernel void m03800_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
* md5 * md5
*/ */
u32 a = MD5M_A; u32x a = MD5M_A;
u32 b = MD5M_B; u32x b = MD5M_B;
u32 c = MD5M_C; u32x c = MD5M_C;
u32 d = MD5M_D; u32x d = MD5M_D;
MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00); MD5_STEP (MD5_Fo, a, b, c, d, w0_t[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01); MD5_STEP (MD5_Fo, d, a, b, c, w0_t[1], MD5C01, MD5S01);
@ -624,12 +570,7 @@ __kernel void m03800_s04 (__global pw_t *pws, __global kernel_rule_t *rules_buf,
MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32); MD5_STEP (MD5_I , c, d, a, b, w0_t[2], MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33); MD5_STEP (MD5_I , b, c, d, a, w2_t[1], MD5C3f, MD5S33);
const u32 r0 = a; COMPARE_S_SIMD (a, d, c, b);
const u32 r1 = d;
const u32 r2 = c;
const u32 r3 = b;
#include COMPARE_S
} }
} }