#define _MD5_
-#define NEW_SIMD_CODE
-
#include "include/constants.h"
#include "include/kernel_vendor.h"
#include "include/kernel_functions.c"
#include "OpenCL/types_ocl.c"
#include "OpenCL/common.c"
-#include "OpenCL/simd.c"
+
+#define COMPARE_S "OpenCL/check_single_comp4.c"
+#define COMPARE_M "OpenCL/check_multi_comp4.c"
__kernel void m02400_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
{
if (gid >= gid_max) return;
- u32 pws0[4] = { 0 };
- u32 pws1[4] = { 0 };
+ u32 wordl0[4];
+
+ wordl0[0] = pws[gid].i[ 0];
+ wordl0[1] = pws[gid].i[ 1];
+ wordl0[2] = pws[gid].i[ 2];
+ wordl0[3] = pws[gid].i[ 3];
+
+ u32 wordl1[4];
+
+ wordl1[0] = pws[gid].i[ 4];
+ wordl1[1] = pws[gid].i[ 5];
+ wordl1[2] = pws[gid].i[ 6];
+ wordl1[3] = pws[gid].i[ 7];
- pws0[0] = pws[gid].i[0];
- pws0[1] = pws[gid].i[1];
- pws0[2] = pws[gid].i[2];
- pws0[3] = pws[gid].i[3];
- pws1[0] = pws[gid].i[4];
- pws1[1] = pws[gid].i[5];
- pws1[2] = pws[gid].i[6];
- pws1[3] = pws[gid].i[7];
+ u32 wordl2[4];
+
+ wordl2[0] = 0;
+ wordl2[1] = 0;
+ wordl2[2] = 0;
+ wordl2[3] = 0;
+
+ u32 wordl3[4];
+
+ wordl3[0] = 0;
+ wordl3[1] = 0;
+ wordl3[2] = 0;
+ wordl3[3] = 0;
const u32 pw_l_len = pws[gid].pw_len;
+ if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
+ {
+ switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
+ }
+
/**
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
{
- const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+ const u32 pw_r_len = combs_buf[il_pos].pw_len;
+
+ const u32 pw_len = pw_l_len + pw_r_len;
+
+ u32 wordr0[4];
+
+ wordr0[0] = combs_buf[il_pos].i[0];
+ wordr0[1] = combs_buf[il_pos].i[1];
+ wordr0[2] = combs_buf[il_pos].i[2];
+ wordr0[3] = combs_buf[il_pos].i[3];
- const u32x pw_len = pw_l_len + pw_r_len;
+ u32 wordr1[4];
- u32x wordr0[4] = { 0 };
- u32x wordr1[4] = { 0 };
- u32x wordr2[4] = { 0 };
- u32x wordr3[4] = { 0 };
+ wordr1[0] = combs_buf[il_pos].i[4];
+ wordr1[1] = combs_buf[il_pos].i[5];
+ wordr1[2] = combs_buf[il_pos].i[6];
+ wordr1[3] = combs_buf[il_pos].i[7];
- wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
- wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
- wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
- wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
- wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
- wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
- wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
- wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
+ u32 wordr2[4];
+
+ wordr2[0] = 0;
+ wordr2[1] = 0;
+ wordr2[2] = 0;
+ wordr2[3] = 0;
+
+ u32 wordr3[4];
+
+ wordr3[0] = 0;
+ wordr3[1] = 0;
+ wordr3[2] = 0;
+ wordr3[3] = 0;
if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
{
switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
}
- u32x w0[4];
+ u32 w0[4];
w0[0] = wordl0[0] | wordr0[0];
w0[1] = wordl0[1] | wordr0[1];
w0[2] = wordl0[2] | wordr0[2];
w0[3] = wordl0[3] | wordr0[3];
- u32x w1[4];
+ u32 w1[4];
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
- u32x w2[4];
+ u32 w2[4];
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
- u32x w3[4];
+ u32 w3[4];
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
w1[0] = 0x80;
w3[2] = 16 * 8;
- u32x a = MD5M_A;
- u32x b = MD5M_B;
- u32x c = MD5M_C;
- u32x d = MD5M_D;
+ u32 a = MD5M_A;
+ u32 b = MD5M_B;
+ u32 c = MD5M_C;
+ u32 d = MD5M_D;
MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
d &= 0x00ffffff;
c &= 0x00ffffff;
b &= 0x00ffffff;
- COMPARE_M_SIMD (a, d, c, b);
+
+ const u32 r0 = a;
+ const u32 r1 = d;
+ const u32 r2 = c;
+ const u32 r3 = b;
+
+ #include COMPARE_M
}
}
if (gid >= gid_max) return;
- u32 pws0[4] = { 0 };
- u32 pws1[4] = { 0 };
+ u32 wordl0[4];
+
+ wordl0[0] = pws[gid].i[ 0];
+ wordl0[1] = pws[gid].i[ 1];
+ wordl0[2] = pws[gid].i[ 2];
+ wordl0[3] = pws[gid].i[ 3];
+
+ u32 wordl1[4];
+
+ wordl1[0] = pws[gid].i[ 4];
+ wordl1[1] = pws[gid].i[ 5];
+ wordl1[2] = pws[gid].i[ 6];
+ wordl1[3] = pws[gid].i[ 7];
+
+ u32 wordl2[4];
- pws0[0] = pws[gid].i[0];
- pws0[1] = pws[gid].i[1];
- pws0[2] = pws[gid].i[2];
- pws0[3] = pws[gid].i[3];
- pws1[0] = pws[gid].i[4];
- pws1[1] = pws[gid].i[5];
- pws1[2] = pws[gid].i[6];
- pws1[3] = pws[gid].i[7];
+ wordl2[0] = 0;
+ wordl2[1] = 0;
+ wordl2[2] = 0;
+ wordl2[3] = 0;
+
+ u32 wordl3[4];
+
+ wordl3[0] = 0;
+ wordl3[1] = 0;
+ wordl3[2] = 0;
+ wordl3[3] = 0;
const u32 pw_l_len = pws[gid].pw_len;
+ if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
+ {
+ switch_buffer_by_offset_le (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
+ }
+
/**
* digest
*/
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
{
- const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
+ const u32 pw_r_len = combs_buf[il_pos].pw_len;
+
+ const u32 pw_len = pw_l_len + pw_r_len;
+
+ u32 wordr0[4];
+
+ wordr0[0] = combs_buf[il_pos].i[0];
+ wordr0[1] = combs_buf[il_pos].i[1];
+ wordr0[2] = combs_buf[il_pos].i[2];
+ wordr0[3] = combs_buf[il_pos].i[3];
+
+ u32 wordr1[4];
- const u32x pw_len = pw_l_len + pw_r_len;
+ wordr1[0] = combs_buf[il_pos].i[4];
+ wordr1[1] = combs_buf[il_pos].i[5];
+ wordr1[2] = combs_buf[il_pos].i[6];
+ wordr1[3] = combs_buf[il_pos].i[7];
- u32x wordr0[4] = { 0 };
- u32x wordr1[4] = { 0 };
- u32x wordr2[4] = { 0 };
- u32x wordr3[4] = { 0 };
+ u32 wordr2[4];
- wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
- wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
- wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
- wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
- wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
- wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
- wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
- wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
+ wordr2[0] = 0;
+ wordr2[1] = 0;
+ wordr2[2] = 0;
+ wordr2[3] = 0;
+
+ u32 wordr3[4];
+
+ wordr3[0] = 0;
+ wordr3[1] = 0;
+ wordr3[2] = 0;
+ wordr3[3] = 0;
if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
{
switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
}
- u32x w0[4];
+ u32 w0[4];
w0[0] = wordl0[0] | wordr0[0];
w0[1] = wordl0[1] | wordr0[1];
w0[2] = wordl0[2] | wordr0[2];
w0[3] = wordl0[3] | wordr0[3];
- u32x w1[4];
+ u32 w1[4];
w1[0] = wordl1[0] | wordr1[0];
w1[1] = wordl1[1] | wordr1[1];
w1[2] = wordl1[2] | wordr1[2];
w1[3] = wordl1[3] | wordr1[3];
- u32x w2[4];
+ u32 w2[4];
w2[0] = wordl2[0] | wordr2[0];
w2[1] = wordl2[1] | wordr2[1];
w2[2] = wordl2[2] | wordr2[2];
w2[3] = wordl2[3] | wordr2[3];
- u32x w3[4];
+ u32 w3[4];
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
w1[0] = 0x80;
w3[2] = 16 * 8;
- u32x a = MD5M_A;
- u32x b = MD5M_B;
- u32x c = MD5M_C;
- u32x d = MD5M_D;
+ u32 a = MD5M_A;
+ u32 b = MD5M_B;
+ u32 c = MD5M_C;
+ u32 d = MD5M_D;
MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
d &= 0x00ffffff;
c &= 0x00ffffff;
b &= 0x00ffffff;
- COMPARE_S_SIMD (a, d, c, b);
+
+ const u32 r0 = a;
+ const u32 r1 = d;
+ const u32 r2 = c;
+ const u32 r3 = b;
+
+ #include COMPARE_S
}
}
projects / hashcat.git / blobdiff