2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
16 #include "include/kernel_functions.c"
17 #include "types_ocl.c"
20 #define COMPARE_S "check_single_comp4.c"
21 #define COMPARE_M "check_multi_comp4.c"
23 __constant u32 lotus_magic_table[256] =
25 0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
26 0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
27 0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
28 0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
29 0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
30 0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
31 0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
32 0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
33 0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
34 0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
35 0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
36 0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
37 0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
38 0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
39 0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
40 0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
41 0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
42 0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
43 0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
44 0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
45 0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
46 0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
47 0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
48 0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
49 0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
50 0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
51 0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
52 0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
53 0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
54 0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
55 0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
56 0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
60 #define BOX(S,i) (u32x) ((S)[(i)])
64 #define BOX(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1])
68 #define BOX(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3])
72 #define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i)])
76 #define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
80 #define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
83 static void lotus_mix (u32 *in, __local u32 s_lotus_magic_table[256])
87 for (int i = 0; i < 18; i++)
92 for (int j = 0; j < 12; j++)
97 p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 0;
98 p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 8;
99 p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 16;
100 p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 24;
107 static void lotus_transform_password (u32 *in, u32 *out, __local u32 s_lotus_magic_table[256])
109 u32 t = out[3] >> 24;
113 //#pragma unroll // kernel fails if used
114 for (int i = 0; i < 4; i++)
116 t ^= (in[i] >> 0) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 0; t = ((out[i] >> 0) & 0xff);
117 t ^= (in[i] >> 8) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 8; t = ((out[i] >> 8) & 0xff);
118 t ^= (in[i] >> 16) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 16; t = ((out[i] >> 16) & 0xff);
119 t ^= (in[i] >> 24) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 24; t = ((out[i] >> 24) & 0xff);
123 static void pad (u32 w[4], const u32 len)
125 const u32 val = 16 - len;
127 const u32 mask1 = val << 24;
129 const u32 mask2 = val << 16
132 const u32 mask3 = val << 8
136 const u32 mask4 = val << 0
143 case 0: w[0] = mask4;
148 case 1: w[0] |= mask3;
153 case 2: w[0] |= mask2;
158 case 3: w[0] |= mask1;
163 case 4: w[1] = mask4;
167 case 5: w[1] |= mask3;
171 case 6: w[1] |= mask2;
175 case 7: w[1] |= mask1;
179 case 8: w[2] = mask4;
182 case 9: w[2] |= mask3;
185 case 10: w[2] |= mask2;
188 case 11: w[2] |= mask1;
191 case 12: w[3] = mask4;
193 case 13: w[3] |= mask3;
195 case 14: w[3] |= mask2;
197 case 15: w[3] |= mask1;
202 static void mdtransform_norecalc (u32 state[4], u32 block[4], __local u32 s_lotus_magic_table[256])
214 x[ 8] = state[0] ^ block[0];
215 x[ 9] = state[1] ^ block[1];
216 x[10] = state[2] ^ block[2];
217 x[11] = state[3] ^ block[3];
219 lotus_mix (x, s_lotus_magic_table);
227 static void mdtransform (u32 state[4], u32 checksum[4], u32 block[4], __local u32 s_lotus_magic_table[256])
229 mdtransform_norecalc (state, block, s_lotus_magic_table);
231 lotus_transform_password (block, checksum, s_lotus_magic_table);
234 static void domino_big_md (const u32 saved_key[16], const u32 size, u32 state[4], __local u32 s_lotus_magic_table[256])
253 for (curpos = 0, idx = 0; curpos + 16 < size; curpos += 16, idx += 4)
255 block[0] = saved_key[idx + 0];
256 block[1] = saved_key[idx + 1];
257 block[2] = saved_key[idx + 2];
258 block[3] = saved_key[idx + 3];
260 mdtransform (state, checksum, block, s_lotus_magic_table);
263 block[0] = saved_key[idx + 0];
264 block[1] = saved_key[idx + 1];
265 block[2] = saved_key[idx + 2];
266 block[3] = saved_key[idx + 3];
268 mdtransform (state, checksum, block, s_lotus_magic_table);
270 mdtransform_norecalc (state, checksum, s_lotus_magic_table);
273 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08700_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 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)
279 const u32 lid = get_local_id (0);
285 const u32 gid = get_global_id (0);
291 __local u32 l_bin2asc[256];
293 const u32 lid4 = lid * 4;
295 const u32 lid40 = lid4 + 0;
296 const u32 lid41 = lid4 + 1;
297 const u32 lid42 = lid4 + 2;
298 const u32 lid43 = lid4 + 3;
300 const u32 v400 = (lid40 >> 0) & 15;
301 const u32 v401 = (lid40 >> 4) & 15;
302 const u32 v410 = (lid41 >> 0) & 15;
303 const u32 v411 = (lid41 >> 4) & 15;
304 const u32 v420 = (lid42 >> 0) & 15;
305 const u32 v421 = (lid42 >> 4) & 15;
306 const u32 v430 = (lid43 >> 0) & 15;
307 const u32 v431 = (lid43 >> 4) & 15;
309 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'A' - 10 + v400) << 8
310 | ((v401 < 10) ? '0' + v401 : 'A' - 10 + v401) << 0;
311 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'A' - 10 + v410) << 8
312 | ((v411 < 10) ? '0' + v411 : 'A' - 10 + v411) << 0;
313 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'A' - 10 + v420) << 8
314 | ((v421 < 10) ? '0' + v421 : 'A' - 10 + v421) << 0;
315 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'A' - 10 + v430) << 8
316 | ((v431 < 10) ? '0' + v431 : 'A' - 10 + v431) << 0;
318 barrier (CLK_LOCAL_MEM_FENCE);
324 __local u32 s_lotus_magic_table[256];
326 s_lotus_magic_table[lid4 + 0] = lotus_magic_table[lid4 + 0];
327 s_lotus_magic_table[lid4 + 1] = lotus_magic_table[lid4 + 1];
328 s_lotus_magic_table[lid4 + 2] = lotus_magic_table[lid4 + 2];
329 s_lotus_magic_table[lid4 + 3] = lotus_magic_table[lid4 + 3];
331 barrier (CLK_LOCAL_MEM_FENCE);
333 if (gid >= gid_max) return;
337 wordl0[0] = pws[gid].i[ 0];
338 wordl0[1] = pws[gid].i[ 1];
339 wordl0[2] = pws[gid].i[ 2];
340 wordl0[3] = pws[gid].i[ 3];
344 wordl1[0] = pws[gid].i[ 4];
345 wordl1[1] = pws[gid].i[ 5];
346 wordl1[2] = pws[gid].i[ 6];
347 wordl1[3] = pws[gid].i[ 7];
364 const u32 pw_l_len = pws[gid].pw_len;
366 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
368 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
375 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
376 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
382 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
384 const u32 pw_r_len = combs_buf[il_pos].pw_len;
386 const u32 pw_len = pw_l_len + pw_r_len;
390 wordr0[0] = combs_buf[il_pos].i[0];
391 wordr0[1] = combs_buf[il_pos].i[1];
392 wordr0[2] = combs_buf[il_pos].i[2];
393 wordr0[3] = combs_buf[il_pos].i[3];
397 wordr1[0] = combs_buf[il_pos].i[4];
398 wordr1[1] = combs_buf[il_pos].i[5];
399 wordr1[2] = combs_buf[il_pos].i[6];
400 wordr1[3] = combs_buf[il_pos].i[7];
416 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
418 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
423 w[ 0] = wordl0[0] | wordr0[0];
424 w[ 1] = wordl0[1] | wordr0[1];
425 w[ 2] = wordl0[2] | wordr0[2];
426 w[ 3] = wordl0[3] | wordr0[3];
427 w[ 4] = wordl1[0] | wordr1[0];
428 w[ 5] = wordl1[1] | wordr1[1];
429 w[ 6] = wordl1[2] | wordr1[2];
430 w[ 7] = wordl1[3] | wordr1[3];
431 w[ 8] = wordl2[0] | wordr2[0];
432 w[ 9] = wordl2[1] | wordr2[1];
433 w[10] = wordl2[2] | wordr2[2];
434 w[11] = wordl2[3] | wordr2[3];
435 w[12] = wordl3[0] | wordr3[0];
436 w[13] = wordl3[1] | wordr3[1];
437 w[14] = wordl3[2] | wordr3[2];
438 w[15] = wordl3[3] | wordr3[3];
453 pad (&w[ 0], pw_len & 0xf);
455 else if (pw_len < 32)
457 pad (&w[ 4], pw_len & 0xf);
459 else if (pw_len < 48)
461 pad (&w[ 8], pw_len & 0xf);
463 else if (pw_len < 64)
465 pad (&w[12], pw_len & 0xf);
468 domino_big_md (w, pw_len, state, s_lotus_magic_table);
470 const u32 w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
471 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
472 const u32 w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
473 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
474 const u32 w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
475 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
476 const u32 w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
477 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
478 const u32 w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
479 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
480 const u32 w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
481 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
482 const u32 w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
483 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
484 //const u32 w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
485 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
487 const u32 pade = 0x0e0e0e0e;
490 w[ 1] = salt1 | w0_t << 16;
491 w[ 2] = w0_t >> 16 | w1_t << 16;
492 w[ 3] = w1_t >> 16 | w2_t << 16;
493 w[ 4] = w2_t >> 16 | w3_t << 16;
494 w[ 5] = w3_t >> 16 | w4_t << 16;
495 w[ 6] = w4_t >> 16 | w5_t << 16;
496 w[ 7] = w5_t >> 16 | w6_t << 16;
497 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
511 domino_big_md (w, 34, state, s_lotus_magic_table);
513 u32 a = state[0] & 0xffffffff;
514 u32 b = state[1] & 0xffffffff;
515 u32 c = state[2] & 0x000000ff;
516 u32 d = state[3] & 0x00000000;
527 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08700_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 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)
531 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08700_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 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)
535 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08700_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 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)
541 const u32 lid = get_local_id (0);
547 const u32 gid = get_global_id (0);
553 __local u32 l_bin2asc[256];
555 const u32 lid4 = lid * 4;
557 const u32 lid40 = lid4 + 0;
558 const u32 lid41 = lid4 + 1;
559 const u32 lid42 = lid4 + 2;
560 const u32 lid43 = lid4 + 3;
562 const u32 v400 = (lid40 >> 0) & 15;
563 const u32 v401 = (lid40 >> 4) & 15;
564 const u32 v410 = (lid41 >> 0) & 15;
565 const u32 v411 = (lid41 >> 4) & 15;
566 const u32 v420 = (lid42 >> 0) & 15;
567 const u32 v421 = (lid42 >> 4) & 15;
568 const u32 v430 = (lid43 >> 0) & 15;
569 const u32 v431 = (lid43 >> 4) & 15;
571 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'A' - 10 + v400) << 8
572 | ((v401 < 10) ? '0' + v401 : 'A' - 10 + v401) << 0;
573 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'A' - 10 + v410) << 8
574 | ((v411 < 10) ? '0' + v411 : 'A' - 10 + v411) << 0;
575 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'A' - 10 + v420) << 8
576 | ((v421 < 10) ? '0' + v421 : 'A' - 10 + v421) << 0;
577 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'A' - 10 + v430) << 8
578 | ((v431 < 10) ? '0' + v431 : 'A' - 10 + v431) << 0;
580 barrier (CLK_LOCAL_MEM_FENCE);
586 __local u32 s_lotus_magic_table[256];
588 s_lotus_magic_table[lid4 + 0] = lotus_magic_table[lid4 + 0];
589 s_lotus_magic_table[lid4 + 1] = lotus_magic_table[lid4 + 1];
590 s_lotus_magic_table[lid4 + 2] = lotus_magic_table[lid4 + 2];
591 s_lotus_magic_table[lid4 + 3] = lotus_magic_table[lid4 + 3];
593 barrier (CLK_LOCAL_MEM_FENCE);
595 if (gid >= gid_max) return;
599 wordl0[0] = pws[gid].i[ 0];
600 wordl0[1] = pws[gid].i[ 1];
601 wordl0[2] = pws[gid].i[ 2];
602 wordl0[3] = pws[gid].i[ 3];
606 wordl1[0] = pws[gid].i[ 4];
607 wordl1[1] = pws[gid].i[ 5];
608 wordl1[2] = pws[gid].i[ 6];
609 wordl1[3] = pws[gid].i[ 7];
626 const u32 pw_l_len = pws[gid].pw_len;
628 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
630 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
637 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
638 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
644 const u32 search[4] =
646 digests_buf[digests_offset].digest_buf[DGST_R0],
647 digests_buf[digests_offset].digest_buf[DGST_R1],
648 digests_buf[digests_offset].digest_buf[DGST_R2],
649 digests_buf[digests_offset].digest_buf[DGST_R3]
656 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
658 const u32 pw_r_len = combs_buf[il_pos].pw_len;
660 const u32 pw_len = pw_l_len + pw_r_len;
664 wordr0[0] = combs_buf[il_pos].i[0];
665 wordr0[1] = combs_buf[il_pos].i[1];
666 wordr0[2] = combs_buf[il_pos].i[2];
667 wordr0[3] = combs_buf[il_pos].i[3];
671 wordr1[0] = combs_buf[il_pos].i[4];
672 wordr1[1] = combs_buf[il_pos].i[5];
673 wordr1[2] = combs_buf[il_pos].i[6];
674 wordr1[3] = combs_buf[il_pos].i[7];
690 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
692 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
697 w[ 0] = wordl0[0] | wordr0[0];
698 w[ 1] = wordl0[1] | wordr0[1];
699 w[ 2] = wordl0[2] | wordr0[2];
700 w[ 3] = wordl0[3] | wordr0[3];
701 w[ 4] = wordl1[0] | wordr1[0];
702 w[ 5] = wordl1[1] | wordr1[1];
703 w[ 6] = wordl1[2] | wordr1[2];
704 w[ 7] = wordl1[3] | wordr1[3];
705 w[ 8] = wordl2[0] | wordr2[0];
706 w[ 9] = wordl2[1] | wordr2[1];
707 w[10] = wordl2[2] | wordr2[2];
708 w[11] = wordl2[3] | wordr2[3];
709 w[12] = wordl3[0] | wordr3[0];
710 w[13] = wordl3[1] | wordr3[1];
711 w[14] = wordl3[2] | wordr3[2];
712 w[15] = wordl3[3] | wordr3[3];
727 pad (&w[ 0], pw_len & 0xf);
729 else if (pw_len < 32)
731 pad (&w[ 4], pw_len & 0xf);
733 else if (pw_len < 48)
735 pad (&w[ 8], pw_len & 0xf);
737 else if (pw_len < 64)
739 pad (&w[12], pw_len & 0xf);
742 domino_big_md (w, pw_len, state, s_lotus_magic_table);
744 const u32 w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
745 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
746 const u32 w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
747 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
748 const u32 w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
749 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
750 const u32 w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
751 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
752 const u32 w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
753 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
754 const u32 w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
755 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
756 const u32 w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
757 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
758 //const u32 w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
759 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
761 const u32 pade = 0x0e0e0e0e;
764 w[ 1] = salt1 | w0_t << 16;
765 w[ 2] = w0_t >> 16 | w1_t << 16;
766 w[ 3] = w1_t >> 16 | w2_t << 16;
767 w[ 4] = w2_t >> 16 | w3_t << 16;
768 w[ 5] = w3_t >> 16 | w4_t << 16;
769 w[ 6] = w4_t >> 16 | w5_t << 16;
770 w[ 7] = w5_t >> 16 | w6_t << 16;
771 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
785 domino_big_md (w, 34, state, s_lotus_magic_table);
787 u32 a = state[0] & 0xffffffff;
788 u32 b = state[1] & 0xffffffff;
789 u32 c = state[2] & 0x000000ff;
790 u32 d = state[3] & 0x00000000;
801 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08700_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 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)
805 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08700_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 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)