arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-11-27 02:18:21 +00:00
Code Issues Releases Wiki Activity
2ea295171c
hashcat/OpenCL/m01100_a0.cl

499 lines
22 KiB
Common Lisp
Raw Normal View History

Initial commit
2015-12-04 14:47:52 +00:00
/**
* Author......: Jens Steube <jens.steube@gmail.com>
* License.....: MIT
*/
#define _MD4_
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
#define NEW_SIMD_CODE
Initial commit
2015-12-04 14:47:52 +00:00
#include "include/constants.h"
#include "include/kernel_vendor.h"
#define DGST_R0 0
#define DGST_R1 3
#define DGST_R2 2
#define DGST_R3 1
#include "include/kernel_functions.c"
Fix path to includes
2016-01-03 00:48:05 +00:00
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
Replace the substring GPU to a more appropriate "device" or "kernel" substring depending on the context
2016-01-05 07:26:44 +00:00
#include "include/rp_kernel.h"
Fix path to includes
2016-01-03 00:56:41 +00:00
#include "OpenCL/rp.c"
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
#include "OpenCL/simd.c"
Initial commit
2015-12-04 14:47:52 +00:00
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__kernel void m01100_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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
* modifier
*/
const u32 lid = get_local_id (0);
/**
* base
*/
const u32 gid = get_global_id (0);
if (gid >= gid_max) return;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 pw_buf0[4];
Initial commit
2015-12-04 14:47:52 +00:00
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];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 pw_buf1[4];
Initial commit
2015-12-04 14:47:52 +00:00
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;
/**
* salt
*/
u32 salt_buf0[4];
salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
u32 salt_buf1[4];
salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
u32 salt_buf2[4];
salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
salt_buf2[2] = 0;
salt_buf2[3] = 0;
const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* loop
*/
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
for (u32 il_pos = 0; il_pos < rules_cnt; il_pos += VECT_SIZE)
Initial commit
2015-12-04 14:47:52 +00:00
{
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
u32x w0[4] = { 0 };
u32x w1[4] = { 0 };
u32x w2[4] = { 0 };
u32x w3[4] = { 0 };
Initial commit
2015-12-04 14:47:52 +00:00
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
Initial commit
2015-12-04 14:47:52 +00:00
Fix append_* function calls
2015-12-15 12:42:37 +00:00
append_0x80_2x4 (w0, w1, out_len);
Initial commit
2015-12-04 14:47:52 +00:00
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
u32x w0_t[4];
u32x w1_t[4];
u32x w2_t[4];
u32x w3_t[4];
Initial commit
2015-12-04 14:47:52 +00:00
make_unicode (w0, w0_t, w1_t);
make_unicode (w1, w2_t, w3_t);
w3_t[2] = out_len * 8 * 2;
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
u32x a = MD4M_A;
u32x b = MD4M_B;
u32x c = MD4M_C;
u32x d = MD4M_D;
Initial commit
2015-12-04 14:47:52 +00:00
MD4_STEP (MD4_Fo, a, b, c, d, w0_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w0_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w0_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w0_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w1_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w1_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w1_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w1_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w2_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w2_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w2_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w2_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w3_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w3_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w3_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w3_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[0], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[0], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[0], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[0], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[1], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[1], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[1], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[1], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[2], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[2], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[2], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[2], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[3], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[3], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[3], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[3], MD4C01, MD4S13);
MD4_STEP (MD4_H , a, b, c, d, w0_t[0], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[0], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[0], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[0], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[2], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[2], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[2], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[2], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[1], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[1], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[1], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[1], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23);
a += MD4M_A;
b += MD4M_B;
c += MD4M_C;
d += MD4M_D;
w0_t[0] = a;
w0_t[1] = b;
w0_t[2] = c;
w0_t[3] = d;
w1_t[0] = salt_buf0[0];
w1_t[1] = salt_buf0[1];
w1_t[2] = salt_buf0[2];
w1_t[3] = salt_buf0[3];
w2_t[0] = salt_buf1[0];
w2_t[1] = salt_buf1[1];
w2_t[2] = salt_buf1[2];
w2_t[3] = salt_buf1[3];
w3_t[0] = salt_buf2[0];
w3_t[1] = salt_buf2[1];
w3_t[2] = (16 + salt_len) * 8;
w3_t[3] = 0;
a = MD4M_A;
b = MD4M_B;
c = MD4M_C;
d = MD4M_D;
MD4_STEP (MD4_Fo, a, b, c, d, w0_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w0_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w0_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w0_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w1_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w1_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w1_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w1_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w2_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w2_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w2_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w2_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w3_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w3_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w3_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w3_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[0], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[0], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[0], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[0], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[1], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[1], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[1], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[1], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[2], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[2], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[2], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[2], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[3], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[3], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[3], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[3], MD4C01, MD4S13);
MD4_STEP (MD4_H , a, b, c, d, w0_t[0], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[0], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[0], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[0], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[2], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[2], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[2], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[2], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[1], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[1], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[1], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[1], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23);
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
COMPARE_M_SIMD (a, d, c, b);
Initial commit
2015-12-04 14:47:52 +00:00
}
}
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__kernel void m01100_m08 (__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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
Initial commit
2015-12-04 14:47:52 +00:00
{
}
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__kernel void m01100_m16 (__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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
Initial commit
2015-12-04 14:47:52 +00:00
{
}
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__kernel void m01100_s04 (__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)
Initial commit
2015-12-04 14:47:52 +00:00
{
/**
* modifier
*/
const u32 lid = get_local_id (0);
/**
* base
*/
const u32 gid = get_global_id (0);
if (gid >= gid_max) return;
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 pw_buf0[4];
Initial commit
2015-12-04 14:47:52 +00:00
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];
- Dropped all vector code since new GPU's are all scalar, makes the code much easier - Some performance on low-end GPU may drop because of that, but only for a few hash-modes - Dropped scalar code (aka warp) since we do not have any vector datatypes anymore - Renamed C++ overloading functions memcat32_9 -> memcat_c32_w4x4_a3x4 - Still need to fix kernels to new function names, needs to be done manually - Temperature Management needs to be rewritten partially because of conflicting datatypes names - Added code to create different codepaths for NV on AMD in runtime in host (see data.vendor_id) - Added code to create different codepaths for NV on AMD in runtime in kernels (see IS_NV and IS_AMD) - First tests working for -m 0, for example - Great performance increases in general for NV so far - Tested amp_* and markov_* kernel - Migrated special NV optimizations for rule processor
2015-12-15 11:04:22 +00:00
u32 pw_buf1[4];
Initial commit
2015-12-04 14:47:52 +00:00
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]
};
/**
* salt
*/
u32 salt_buf0[4];
salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
u32 salt_buf1[4];
salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
u32 salt_buf2[4];
salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
salt_buf2[2] = 0;
salt_buf2[3] = 0;
const u32 salt_len = salt_bufs[salt_pos].salt_len;
/**
* loop
*/
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
for (u32 il_pos = 0; il_pos < rules_cnt; il_pos += VECT_SIZE)
Initial commit
2015-12-04 14:47:52 +00:00
{
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
u32x w0[4] = { 0 };
u32x w1[4] = { 0 };
u32x w2[4] = { 0 };
u32x w3[4] = { 0 };
Initial commit
2015-12-04 14:47:52 +00:00
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
const u32 out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
Initial commit
2015-12-04 14:47:52 +00:00
Fix append_* function calls
2015-12-15 12:42:37 +00:00
append_0x80_2x4 (w0, w1, out_len);
Initial commit
2015-12-04 14:47:52 +00:00
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
u32x w0_t[4];
u32x w1_t[4];
u32x w2_t[4];
u32x w3_t[4];
Initial commit
2015-12-04 14:47:52 +00:00
make_unicode (w0, w0_t, w1_t);
make_unicode (w1, w2_t, w3_t);
w3_t[2] = out_len * 8 * 2;
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
u32x a = MD4M_A;
u32x b = MD4M_B;
u32x c = MD4M_C;
u32x d = MD4M_D;
Initial commit
2015-12-04 14:47:52 +00:00
MD4_STEP (MD4_Fo, a, b, c, d, w0_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w0_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w0_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w0_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w1_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w1_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w1_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w1_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w2_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w2_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w2_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w2_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w3_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w3_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w3_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w3_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[0], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[0], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[0], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[0], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[1], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[1], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[1], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[1], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[2], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[2], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[2], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[2], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[3], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[3], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[3], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[3], MD4C01, MD4S13);
MD4_STEP (MD4_H , a, b, c, d, w0_t[0], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[0], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[0], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[0], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[2], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[2], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[2], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[2], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[1], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[1], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[1], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[1], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23);
a += MD4M_A;
b += MD4M_B;
c += MD4M_C;
d += MD4M_D;
w0_t[0] = a;
w0_t[1] = b;
w0_t[2] = c;
w0_t[3] = d;
w1_t[0] = salt_buf0[0];
w1_t[1] = salt_buf0[1];
w1_t[2] = salt_buf0[2];
w1_t[3] = salt_buf0[3];
w2_t[0] = salt_buf1[0];
w2_t[1] = salt_buf1[1];
w2_t[2] = salt_buf1[2];
w2_t[3] = salt_buf1[3];
w3_t[0] = salt_buf2[0];
w3_t[1] = salt_buf2[1];
w3_t[2] = (16 + salt_len) * 8;
w3_t[3] = 0;
a = MD4M_A;
b = MD4M_B;
c = MD4M_C;
d = MD4M_D;
MD4_STEP (MD4_Fo, a, b, c, d, w0_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w0_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w0_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w0_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w1_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w1_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w1_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w1_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w2_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w2_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w2_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w2_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Fo, a, b, c, d, w3_t[0], MD4C00, MD4S00);
MD4_STEP (MD4_Fo, d, a, b, c, w3_t[1], MD4C00, MD4S01);
MD4_STEP (MD4_Fo, c, d, a, b, w3_t[2], MD4C00, MD4S02);
MD4_STEP (MD4_Fo, b, c, d, a, w3_t[3], MD4C00, MD4S03);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[0], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[0], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[0], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[0], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[1], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[1], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[1], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[1], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[2], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[2], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[2], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[2], MD4C01, MD4S13);
MD4_STEP (MD4_Go, a, b, c, d, w0_t[3], MD4C01, MD4S10);
MD4_STEP (MD4_Go, d, a, b, c, w1_t[3], MD4C01, MD4S11);
MD4_STEP (MD4_Go, c, d, a, b, w2_t[3], MD4C01, MD4S12);
MD4_STEP (MD4_Go, b, c, d, a, w3_t[3], MD4C01, MD4S13);
MD4_STEP (MD4_H , a, b, c, d, w0_t[0], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[0], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[0], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[0], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[2], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[2], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[2], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[2], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[1], MD4C02, MD4S20);
MD4_STEP (MD4_H , d, a, b, c, w2_t[1], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[1], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[1], MD4C02, MD4S23);
MD4_STEP (MD4_H , a, b, c, d, w0_t[3], MD4C02, MD4S20);
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
if (MATCHES_NONE_VS (a, search[0])) continue;
Initial commit
2015-12-04 14:47:52 +00:00
MD4_STEP (MD4_H , d, a, b, c, w2_t[3], MD4C02, MD4S21);
MD4_STEP (MD4_H , c, d, a, b, w1_t[3], MD4C02, MD4S22);
MD4_STEP (MD4_H , b, c, d, a, w3_t[3], MD4C02, MD4S23);
Converted to new SIMD: -m 1100 -a 0
2016-02-05 16:43:10 +00:00
COMPARE_S_SIMD (a, d, c, b);
Initial commit
2015-12-04 14:47:52 +00:00
}
}
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__kernel void m01100_s08 (__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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
Initial commit
2015-12-04 14:47:52 +00:00
{
}
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
__kernel void m01100_s16 (__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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
Initial commit
2015-12-04 14:47:52 +00:00
{
}
Reference in New Issue Copy Permalink