arno/hashcat
1
0
Fork 0
mirror of https://github.com/hashcat/hashcat.git synced 2024-12-21 06:08:09 +00:00
Code Issues Releases Wiki Activity
0891e39b62
hashcat/OpenCL/m02710_a0.cl

762 lines
34 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 _MD5_
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +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 2710 -a 0
2016-02-05 20:20:42 +00:00
#include "OpenCL/simd.c"
Initial commit
2015-12-04 14:47:52 +00:00
Prepare _a0 kernel for SIMD
2016-02-01 20:06:56 +00:00
#if VECT_SIZE == 1
#define uint_to_hex_lower8(i) (u32x) (l_bin2asc[(i)])
#elif VECT_SIZE == 2
#define uint_to_hex_lower8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
#elif VECT_SIZE == 4
#define uint_to_hex_lower8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
#elif VECT_SIZE == 8
#define uint_to_hex_lower8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3], l_bin2asc[(i).s4], l_bin2asc[(i).s5], l_bin2asc[(i).s6], l_bin2asc[(i).s7])
#endif
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 m02710_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
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 gid = get_global_id (0);
Initial commit
2015-12-04 14:47:52 +00:00
const u32 lid = get_local_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
* base
*/
- 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 s[8];
s[0] = salt_bufs[salt_pos].salt_buf[0];
s[1] = salt_bufs[salt_pos].salt_buf[1];
s[2] = salt_bufs[salt_pos].salt_buf[2];
s[3] = salt_bufs[salt_pos].salt_buf[3];
s[4] = salt_bufs[salt_pos].salt_buf[4];
s[5] = salt_bufs[salt_pos].salt_buf[5];
s[6] = salt_bufs[salt_pos].salt_buf[6];
s[7] = salt_bufs[salt_pos].salt_buf[7];
const u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 r_14 = (32 + salt_len) * 8;
/**
* bin2asc table
*/
__local u32 l_bin2asc[256];
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
for (u32 i = lid; i < 256; i += lsz)
{
const u32 i0 = (i >> 0) & 15;
const u32 i1 = (i >> 4) & 15;
l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8
| ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0;
}
Initial commit
2015-12-04 14:47:52 +00:00
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* loop
*/
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +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 2710 -a 0
2016-02-05 20:20:42 +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 2710 -a 0
2016-02-05 20:20:42 +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
w3[2] = out_len * 8;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +00:00
u32x a = MD5M_A;
u32x b = MD5M_B;
u32x c = MD5M_C;
u32x d = MD5M_D;
Initial commit
2015-12-04 14:47:52 +00:00
MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
a += MD5M_A;
b += MD5M_B;
c += MD5M_C;
d += MD5M_D;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +00:00
w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0
| uint_to_hex_lower8 ((a >> 8) & 255) << 16;
w0[1] = uint_to_hex_lower8 ((a >> 16) & 255) << 0
| uint_to_hex_lower8 ((a >> 24) & 255) << 16;
w0[2] = uint_to_hex_lower8 ((b >> 0) & 255) << 0
| uint_to_hex_lower8 ((b >> 8) & 255) << 16;
w0[3] = uint_to_hex_lower8 ((b >> 16) & 255) << 0
| uint_to_hex_lower8 ((b >> 24) & 255) << 16;
w1[0] = uint_to_hex_lower8 ((c >> 0) & 255) << 0
| uint_to_hex_lower8 ((c >> 8) & 255) << 16;
w1[1] = uint_to_hex_lower8 ((c >> 16) & 255) << 0
| uint_to_hex_lower8 ((c >> 24) & 255) << 16;
w1[2] = uint_to_hex_lower8 ((d >> 0) & 255) << 0
| uint_to_hex_lower8 ((d >> 8) & 255) << 16;
w1[3] = uint_to_hex_lower8 ((d >> 16) & 255) << 0
| uint_to_hex_lower8 ((d >> 24) & 255) << 16;
w2[0] = s[0];
w2[1] = s[1];
w2[2] = s[2];
w2[3] = s[3];
w3[0] = s[4];
w3[1] = s[5];
w3[2] = s[6];
w3[3] = s[7];
Initial commit
2015-12-04 14:47:52 +00:00
a = MD5M_A;
b = MD5M_B;
c = MD5M_C;
d = MD5M_D;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +00:00
MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
const u32x r_a = a + MD5M_A;
const u32x r_b = b + MD5M_B;
const u32x r_c = c + MD5M_C;
const u32x r_d = d + MD5M_D;
Initial commit
2015-12-04 14:47:52 +00:00
a = r_a;
b = r_b;
c = r_c;
d = r_d;
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C00, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C01, MD5S01);
MD5_STEP0(MD5_Fo, c, d, a, b, MD5C02, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C03, MD5S03);
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C04, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C05, MD5S01);
MD5_STEP0(MD5_Fo, c, d, a, b, MD5C06, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C07, MD5S03);
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C08, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C09, MD5S01);
MD5_STEP0(MD5_Fo, c, d, a, b, MD5C0a, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C0b, MD5S03);
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C0c, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, r_14, MD5C0e, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C0f, MD5S03);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C10, MD5S10);
MD5_STEP0(MD5_Go, d, a, b, c, MD5C11, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C12, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C13, MD5S13);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C14, MD5S10);
MD5_STEP0(MD5_Go, d, a, b, c, MD5C15, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C16, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C17, MD5S13);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, r_14, MD5C19, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C1a, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C1b, MD5S13);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C1c, MD5S10);
MD5_STEP0(MD5_Go, d, a, b, c, MD5C1d, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C1e, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C1f, MD5S13);
MD5_STEP0(MD5_H , a, b, c, d, MD5C20, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C21, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, r_14, MD5C23, MD5S23);
MD5_STEP0(MD5_H , a, b, c, d, MD5C24, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C25, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C26, MD5S22);
MD5_STEP0(MD5_H , b, c, d, a, MD5C27, MD5S23);
MD5_STEP0(MD5_H , a, b, c, d, MD5C28, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C29, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C2a, MD5S22);
MD5_STEP0(MD5_H , b, c, d, a, MD5C2b, MD5S23);
MD5_STEP0(MD5_H , a, b, c, d, MD5C2c, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C2d, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C2e, MD5S22);
MD5_STEP0(MD5_H , b, c, d, a, MD5C2f, MD5S23);
MD5_STEP0(MD5_I , a, b, c, d, MD5C30, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, r_14, MD5C32, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C33, MD5S33);
MD5_STEP0(MD5_I , a, b, c, d, MD5C34, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C35, MD5S31);
MD5_STEP0(MD5_I , c, d, a, b, MD5C36, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C37, MD5S33);
MD5_STEP0(MD5_I , a, b, c, d, MD5C38, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C39, MD5S31);
MD5_STEP0(MD5_I , c, d, a, b, MD5C3a, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C3b, MD5S33);
MD5_STEP0(MD5_I , a, b, c, d, MD5C3c, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C3d, MD5S31);
MD5_STEP0(MD5_I , c, d, a, b, MD5C3e, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C3f, MD5S33);
a += r_a;
b += r_b;
c += r_c;
d += r_d;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +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 m02710_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 m02710_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 m02710_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
*/
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 gid = get_global_id (0);
Initial commit
2015-12-04 14:47:52 +00:00
const u32 lid = get_local_id (0);
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
const u32 lsz = get_local_size (0);
Initial commit
2015-12-04 14:47:52 +00:00
/**
* base
*/
- 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 s[8];
s[0] = salt_bufs[salt_pos].salt_buf[0];
s[1] = salt_bufs[salt_pos].salt_buf[1];
s[2] = salt_bufs[salt_pos].salt_buf[2];
s[3] = salt_bufs[salt_pos].salt_buf[3];
s[4] = salt_bufs[salt_pos].salt_buf[4];
s[5] = salt_bufs[salt_pos].salt_buf[5];
s[6] = salt_bufs[salt_pos].salt_buf[6];
s[7] = salt_bufs[salt_pos].salt_buf[7];
u32 salt_len = salt_bufs[salt_pos].salt_len;
const u32 r_14 = (32 + salt_len) * 8;
/**
* 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]
};
/**
* bin2asc table
*/
__local u32 l_bin2asc[256];
Upgrade kernel to support dynamic local work sizes
2016-01-19 15:06:03 +00:00
for (u32 i = lid; i < 256; i += lsz)
{
const u32 i0 = (i >> 0) & 15;
const u32 i1 = (i >> 4) & 15;
l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8
| ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0;
}
Initial commit
2015-12-04 14:47:52 +00:00
barrier (CLK_LOCAL_MEM_FENCE);
if (gid >= gid_max) return;
/**
* loop
*/
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +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 2710 -a 0
2016-02-05 20:20:42 +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 2710 -a 0
2016-02-05 20:20:42 +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
w3[2] = out_len * 8;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +00:00
u32x a = MD5M_A;
u32x b = MD5M_B;
u32x c = MD5M_C;
u32x d = MD5M_D;
Initial commit
2015-12-04 14:47:52 +00:00
MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
a += MD5M_A;
b += MD5M_B;
c += MD5M_C;
d += MD5M_D;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +00:00
w0[0] = uint_to_hex_lower8 ((a >> 0) & 255) << 0
| uint_to_hex_lower8 ((a >> 8) & 255) << 16;
w0[1] = uint_to_hex_lower8 ((a >> 16) & 255) << 0
| uint_to_hex_lower8 ((a >> 24) & 255) << 16;
w0[2] = uint_to_hex_lower8 ((b >> 0) & 255) << 0
| uint_to_hex_lower8 ((b >> 8) & 255) << 16;
w0[3] = uint_to_hex_lower8 ((b >> 16) & 255) << 0
| uint_to_hex_lower8 ((b >> 24) & 255) << 16;
w1[0] = uint_to_hex_lower8 ((c >> 0) & 255) << 0
| uint_to_hex_lower8 ((c >> 8) & 255) << 16;
w1[1] = uint_to_hex_lower8 ((c >> 16) & 255) << 0
| uint_to_hex_lower8 ((c >> 24) & 255) << 16;
w1[2] = uint_to_hex_lower8 ((d >> 0) & 255) << 0
| uint_to_hex_lower8 ((d >> 8) & 255) << 16;
w1[3] = uint_to_hex_lower8 ((d >> 16) & 255) << 0
| uint_to_hex_lower8 ((d >> 24) & 255) << 16;
w2[0] = s[0];
w2[1] = s[1];
w2[2] = s[2];
w2[3] = s[3];
w3[0] = s[4];
w3[1] = s[5];
w3[2] = s[6];
w3[3] = s[7];
Initial commit
2015-12-04 14:47:52 +00:00
a = MD5M_A;
b = MD5M_B;
c = MD5M_C;
d = MD5M_D;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +00:00
MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
const u32x r_a = a + MD5M_A;
const u32x r_b = b + MD5M_B;
const u32x r_c = c + MD5M_C;
const u32x r_d = d + MD5M_D;
Initial commit
2015-12-04 14:47:52 +00:00
a = r_a;
b = r_b;
c = r_c;
d = r_d;
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C00, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C01, MD5S01);
MD5_STEP0(MD5_Fo, c, d, a, b, MD5C02, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C03, MD5S03);
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C04, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C05, MD5S01);
MD5_STEP0(MD5_Fo, c, d, a, b, MD5C06, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C07, MD5S03);
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C08, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C09, MD5S01);
MD5_STEP0(MD5_Fo, c, d, a, b, MD5C0a, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C0b, MD5S03);
MD5_STEP0(MD5_Fo, a, b, c, d, MD5C0c, MD5S00);
MD5_STEP0(MD5_Fo, d, a, b, c, MD5C0d, MD5S01);
MD5_STEP (MD5_Fo, c, d, a, b, r_14, MD5C0e, MD5S02);
MD5_STEP0(MD5_Fo, b, c, d, a, MD5C0f, MD5S03);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C10, MD5S10);
MD5_STEP0(MD5_Go, d, a, b, c, MD5C11, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C12, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C13, MD5S13);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C14, MD5S10);
MD5_STEP0(MD5_Go, d, a, b, c, MD5C15, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C16, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C17, MD5S13);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C18, MD5S10);
MD5_STEP (MD5_Go, d, a, b, c, r_14, MD5C19, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C1a, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C1b, MD5S13);
MD5_STEP0(MD5_Go, a, b, c, d, MD5C1c, MD5S10);
MD5_STEP0(MD5_Go, d, a, b, c, MD5C1d, MD5S11);
MD5_STEP0(MD5_Go, c, d, a, b, MD5C1e, MD5S12);
MD5_STEP0(MD5_Go, b, c, d, a, MD5C1f, MD5S13);
MD5_STEP0(MD5_H , a, b, c, d, MD5C20, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C21, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C22, MD5S22);
MD5_STEP (MD5_H , b, c, d, a, r_14, MD5C23, MD5S23);
MD5_STEP0(MD5_H , a, b, c, d, MD5C24, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C25, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C26, MD5S22);
MD5_STEP0(MD5_H , b, c, d, a, MD5C27, MD5S23);
MD5_STEP0(MD5_H , a, b, c, d, MD5C28, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C29, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C2a, MD5S22);
MD5_STEP0(MD5_H , b, c, d, a, MD5C2b, MD5S23);
MD5_STEP0(MD5_H , a, b, c, d, MD5C2c, MD5S20);
MD5_STEP0(MD5_H , d, a, b, c, MD5C2d, MD5S21);
MD5_STEP0(MD5_H , c, d, a, b, MD5C2e, MD5S22);
MD5_STEP0(MD5_H , b, c, d, a, MD5C2f, MD5S23);
MD5_STEP0(MD5_I , a, b, c, d, MD5C30, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C31, MD5S31);
MD5_STEP (MD5_I , c, d, a, b, r_14, MD5C32, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C33, MD5S33);
MD5_STEP0(MD5_I , a, b, c, d, MD5C34, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C35, MD5S31);
MD5_STEP0(MD5_I , c, d, a, b, MD5C36, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C37, MD5S33);
MD5_STEP0(MD5_I , a, b, c, d, MD5C38, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C39, MD5S31);
MD5_STEP0(MD5_I , c, d, a, b, MD5C3a, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C3b, MD5S33);
MD5_STEP0(MD5_I , a, b, c, d, MD5C3c, MD5S30);
MD5_STEP0(MD5_I , d, a, b, c, MD5C3d, MD5S31);
MD5_STEP0(MD5_I , c, d, a, b, MD5C3e, MD5S32);
MD5_STEP0(MD5_I , b, c, d, a, MD5C3f, MD5S33);
a += r_a;
b += r_b;
c += r_c;
d += r_d;
Converted to new SIMD: -m 2710 -a 0
2016-02-05 20:20:42 +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 m02710_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 m02710_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