#define _MD5_
+#define NEW_SIMD_CODE
+
#include "include/constants.h"
#include "include/kernel_vendor.h"
#define DGST_R3 1
#include "include/kernel_functions.c"
-#include "types_ocl.c"
-#include "common.c"
-
-#define COMPARE_S "check_single_comp4.c"
-#define COMPARE_M "check_multi_comp4.c"
+#include "OpenCL/types_ocl.c"
+#include "OpenCL/common.c"
+#include "OpenCL/simd.c"
#define uint_to_hex_lower8(i) l_bin2asc[(i)]
const u32 mod = block_len & 3;
const u32 div = block_len / 4;
- #ifdef IS_AMD
+ #if defined IS_AMD || defined IS_GENERIC
const int offset_minus_4 = 4 - mod;
u32 append0_t[4];
return new_len;
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11400_m04 (__global pw_t *pws, __global gpu_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 sip_t *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)
+__kernel void m11400_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 sip_t *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)
{
/**
* modifier
*/
+ const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
+ const u32 lsz = get_local_size (0);
/**
- * base
+ * bin2asc table
*/
- const u32 gid = get_global_id (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];
+ __local u32 l_bin2asc[256];
- u32 wordl2[4];
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0;
+ }
- u32 wordl3[4];
+ barrier (CLK_LOCAL_MEM_FENCE);
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ if (gid >= gid_max) return;
/**
- * bin2asc table
+ * base
*/
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 8
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 0;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 8
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 0;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 8
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 0;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 8
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 0;
+ u32 pws0[4] = { 0 };
+ u32 pws1[4] = { 0 };
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
+ 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];
const u32 pw_l_len = pws[gid].pw_len;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- append_0x80_2x4 (wordl0, wordl1, pw_l_len);
-
- switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* salt
*/
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- const u32 pw_len = pw_l_len + pw_r_len;
+ const u32x pw_len = pw_l_len + pw_r_len;
- u32 wordr0[4];
+ u32x wordr0[4] = { 0 };
+ u32x wordr1[4] = { 0 };
+ u32x wordr2[4] = { 0 };
+ u32x wordr3[4] = { 0 };
- 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];
-
- 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];
-
- 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;
+ 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);
if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
{
- switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
}
- u32 w0[4];
+ u32x 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];
- u32 w1[4];
+ u32x 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];
- u32 w2[4];
+ u32x 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];
- u32 w3[4];
+ u32x w3[4];
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
block_len = memcat32 (block0, block1, salt_len, w0, w1, w2, w3, pw_len);
- u32 w0_t[4];
+ u32x w0_t[4];
w0_t[0] = block0[ 0];
w0_t[1] = block0[ 1];
w0_t[2] = block0[ 2];
w0_t[3] = block0[ 3];
- u32 w1_t[4];
+ u32x w1_t[4];
w1_t[0] = block0[ 4];
w1_t[1] = block0[ 5];
w1_t[2] = block0[ 6];
w1_t[3] = block0[ 7];
- u32 w2_t[4];
+ u32x w2_t[4];
w2_t[0] = block0[ 8];
w2_t[1] = block0[ 9];
w2_t[2] = block0[10];
w2_t[3] = block0[11];
- u32 w3_t[4];
+ u32x w3_t[4];
w3_t[0] = block0[12];
w3_t[1] = block0[13];
// md5
- u32 tmp2;
-
- u32 a = MD5M_A;
- u32 b = MD5M_B;
- u32 c = MD5M_C;
- u32 d = MD5M_D;
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x d = MD5M_D;
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_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
b += r_b;
c += r_c;
d += r_d;
-
- const u32 r0 = a;
- const u32 r1 = d;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_M
+ COMPARE_M_SIMD (a, d, c, b);
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11400_m08 (__global pw_t *pws, __global gpu_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 sip_t *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)
+__kernel void m11400_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 sip_t *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)
{
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11400_m16 (__global pw_t *pws, __global gpu_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 sip_t *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)
+__kernel void m11400_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 sip_t *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)
{
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11400_s04 (__global pw_t *pws, __global gpu_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 sip_t *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)
+__kernel void m11400_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 sip_t *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)
{
/**
* modifier
*/
+ const u32 gid = get_global_id (0);
const u32 lid = get_local_id (0);
+ const u32 lsz = get_local_size (0);
/**
- * base
+ * bin2asc table
*/
- const u32 gid = get_global_id (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];
+ __local u32 l_bin2asc[256];
- u32 wordl2[4];
+ for (u32 i = lid; i < 256; i += lsz)
+ {
+ const u32 i0 = (i >> 0) & 15;
+ const u32 i1 = (i >> 4) & 15;
- wordl2[0] = 0;
- wordl2[1] = 0;
- wordl2[2] = 0;
- wordl2[3] = 0;
+ l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'a' - 10 + i0) << 8
+ | ((i1 < 10) ? '0' + i1 : 'a' - 10 + i1) << 0;
+ }
- u32 wordl3[4];
+ barrier (CLK_LOCAL_MEM_FENCE);
- wordl3[0] = 0;
- wordl3[1] = 0;
- wordl3[2] = 0;
- wordl3[3] = 0;
+ if (gid >= gid_max) return;
/**
- * bin2asc table
+ * base
*/
- __local u32 l_bin2asc[256];
-
- const u32 lid4 = lid * 4;
-
- const u32 lid40 = lid4 + 0;
- const u32 lid41 = lid4 + 1;
- const u32 lid42 = lid4 + 2;
- const u32 lid43 = lid4 + 3;
-
- const u32 v400 = (lid40 >> 0) & 15;
- const u32 v401 = (lid40 >> 4) & 15;
- const u32 v410 = (lid41 >> 0) & 15;
- const u32 v411 = (lid41 >> 4) & 15;
- const u32 v420 = (lid42 >> 0) & 15;
- const u32 v421 = (lid42 >> 4) & 15;
- const u32 v430 = (lid43 >> 0) & 15;
- const u32 v431 = (lid43 >> 4) & 15;
-
- l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 8
- | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 0;
- l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 8
- | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 0;
- l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 8
- | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 0;
- l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 8
- | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 0;
+ u32 pws0[4] = { 0 };
+ u32 pws1[4] = { 0 };
- barrier (CLK_LOCAL_MEM_FENCE);
-
- if (gid >= gid_max) return;
+ 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];
const u32 pw_l_len = pws[gid].pw_len;
- if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
- {
- append_0x80_2x4 (wordl0, wordl1, pw_l_len);
-
- switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
- }
-
/**
* salt
*/
* loop
*/
- for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
+ for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
{
- const u32 pw_r_len = combs_buf[il_pos].pw_len;
-
- const u32 pw_len = pw_l_len + pw_r_len;
+ const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
- u32 wordr0[4];
+ const u32x pw_len = pw_l_len + pw_r_len;
- 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];
+ u32x wordr0[4] = { 0 };
+ u32x wordr1[4] = { 0 };
+ u32x wordr2[4] = { 0 };
+ u32x wordr3[4] = { 0 };
- u32 wordr1[4];
-
- 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];
-
- 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;
+ 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);
if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
{
- switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
+ switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
}
- u32 w0[4];
+ u32x 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];
- u32 w1[4];
+ u32x 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];
- u32 w2[4];
+ u32x 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];
- u32 w3[4];
+ u32x w3[4];
w3[0] = wordl3[0] | wordr3[0];
w3[1] = wordl3[1] | wordr3[1];
block_len = memcat32 (block0, block1, salt_len, w0, w1, w2, w3, pw_len);
- u32 w0_t[4];
+ u32x w0_t[4];
w0_t[0] = block0[ 0];
w0_t[1] = block0[ 1];
w0_t[2] = block0[ 2];
w0_t[3] = block0[ 3];
- u32 w1_t[4];
+ u32x w1_t[4];
w1_t[0] = block0[ 4];
w1_t[1] = block0[ 5];
w1_t[2] = block0[ 6];
w1_t[3] = block0[ 7];
- u32 w2_t[4];
+ u32x w2_t[4];
w2_t[0] = block0[ 8];
w2_t[1] = block0[ 9];
w2_t[2] = block0[10];
w2_t[3] = block0[11];
- u32 w3_t[4];
+ u32x w3_t[4];
w3_t[0] = block0[12];
w3_t[1] = block0[13];
// md5
- u32 tmp2;
-
- u32 a = MD5M_A;
- u32 b = MD5M_B;
- u32 c = MD5M_C;
- u32 d = MD5M_D;
+ u32x a = MD5M_A;
+ u32x b = MD5M_B;
+ u32x c = MD5M_C;
+ u32x d = MD5M_D;
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_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
MD5_STEP (MD5_Go, c, d, a, b, w1_t[3], MD5C1e, MD5S12);
MD5_STEP (MD5_Go, b, c, d, a, w3_t[0], MD5C1f, MD5S13);
- MD5_STEP (MD5_H1, a, b, c, d, w1_t[1], MD5C20, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w2_t[0], MD5C21, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w2_t[3], MD5C22, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w3_t[2], MD5C23, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w0_t[1], MD5C24, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w1_t[0], MD5C25, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w1_t[3], MD5C26, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w2_t[2], MD5C27, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w3_t[1], MD5C28, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w0_t[0], MD5C29, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w0_t[3], MD5C2a, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w1_t[2], MD5C2b, MD5S23);
- MD5_STEP (MD5_H1, a, b, c, d, w2_t[1], MD5C2c, MD5S20);
- MD5_STEP (MD5_H2, d, a, b, c, w3_t[0], MD5C2d, MD5S21);
- MD5_STEP (MD5_H1, c, d, a, b, w3_t[3], MD5C2e, MD5S22);
- MD5_STEP (MD5_H2, b, c, d, a, w0_t[2], MD5C2f, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w1_t[1], MD5C20, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w2_t[0], MD5C21, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w2_t[3], MD5C22, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w3_t[2], MD5C23, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w0_t[1], MD5C24, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w1_t[0], MD5C25, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w1_t[3], MD5C26, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w2_t[2], MD5C27, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w3_t[1], MD5C28, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w0_t[0], MD5C29, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w0_t[3], MD5C2a, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w1_t[2], MD5C2b, MD5S23);
+ MD5_STEP (MD5_H , a, b, c, d, w2_t[1], MD5C2c, MD5S20);
+ MD5_STEP (MD5_H , d, a, b, c, w3_t[0], MD5C2d, MD5S21);
+ MD5_STEP (MD5_H , c, d, a, b, w3_t[3], MD5C2e, MD5S22);
+ MD5_STEP (MD5_H , b, c, d, a, w0_t[2], MD5C2f, MD5S23);
MD5_STEP (MD5_I , a, b, c, d, w0_t[0], MD5C30, MD5S30);
MD5_STEP (MD5_I , d, a, b, c, w1_t[3], MD5C31, MD5S31);
b += r_b;
c += r_c;
d += r_d;
-
- const u32 r0 = a;
- const u32 r1 = d;
- const u32 r2 = c;
- const u32 r3 = b;
-
- #include COMPARE_S
+ COMPARE_S_SIMD (a, d, c, b);
}
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11400_s08 (__global pw_t *pws, __global gpu_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 sip_t *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)
+__kernel void m11400_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 sip_t *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)
{
}
-__kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11400_s16 (__global pw_t *pws, __global gpu_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 sip_t *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)
+__kernel void m11400_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 sip_t *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)
{
}