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"
19 #include "include/rp_gpu.h"
22 #define COMPARE_S "check_single_comp4.c"
23 #define COMPARE_M "check_multi_comp4.c"
25 __constant u32 lotus_magic_table[256] =
27 0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
28 0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
29 0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
30 0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
31 0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
32 0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
33 0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
34 0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
35 0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
36 0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
37 0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
38 0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
39 0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
40 0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
41 0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
42 0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
43 0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
44 0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
45 0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
46 0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
47 0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
48 0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
49 0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
50 0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
51 0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
52 0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
53 0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
54 0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
55 0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
56 0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
57 0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
58 0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
61 #define BOX(S,i) (S)[(i)]
63 #define uint_to_hex_upper8(i) l_bin2asc[(i)]
65 static void lotus_mix (u32 *in, __local u32 s_lotus_magic_table[256])
69 for (int i = 0; i < 18; i++)
74 for (int j = 0; j < 12; j++)
79 p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 0;
80 p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 8;
81 p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 16;
82 p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 24;
89 static void lotus_transform_password (u32 in[4], u32 out[4], __local u32 s_lotus_magic_table[256])
95 //#pragma unroll // kernel fails if used
96 for (int i = 0; i < 4; i++)
98 t ^= (in[i] >> 0) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 0; t = ((out[i] >> 0) & 0xff);
99 t ^= (in[i] >> 8) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 8; t = ((out[i] >> 8) & 0xff);
100 t ^= (in[i] >> 16) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 16; t = ((out[i] >> 16) & 0xff);
101 t ^= (in[i] >> 24) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 24; t = ((out[i] >> 24) & 0xff);
105 static void pad (u32 w[4], const u32 len)
107 const u32 val = 16 - len;
109 const u32 mask1 = val << 24;
111 const u32 mask2 = val << 16
114 const u32 mask3 = val << 8
118 const u32 mask4 = val << 0
125 case 0: w[0] = mask4;
130 case 1: w[0] |= mask3;
135 case 2: w[0] |= mask2;
140 case 3: w[0] |= mask1;
145 case 4: w[1] = mask4;
149 case 5: w[1] |= mask3;
153 case 6: w[1] |= mask2;
157 case 7: w[1] |= mask1;
161 case 8: w[2] = mask4;
164 case 9: w[2] |= mask3;
167 case 10: w[2] |= mask2;
170 case 11: w[2] |= mask1;
173 case 12: w[3] = mask4;
175 case 13: w[3] |= mask3;
177 case 14: w[3] |= mask2;
179 case 15: w[3] |= mask1;
184 static void mdtransform_norecalc (u32 state[4], u32 block[4], __local u32 s_lotus_magic_table[256])
196 x[ 8] = state[0] ^ block[0];
197 x[ 9] = state[1] ^ block[1];
198 x[10] = state[2] ^ block[2];
199 x[11] = state[3] ^ block[3];
201 lotus_mix (x, s_lotus_magic_table);
209 static void mdtransform (u32 state[4], u32 checksum[4], u32 block[4], __local u32 s_lotus_magic_table[256])
211 mdtransform_norecalc (state, block, s_lotus_magic_table);
213 lotus_transform_password (block, checksum, s_lotus_magic_table);
216 static void domino_big_md (const u32 saved_key[16], const u32 size, u32 state[4], __local u32 s_lotus_magic_table[256])
235 for (curpos = 0, idx = 0; curpos + 16 < size; curpos += 16, idx += 4)
237 block[0] = saved_key[idx + 0];
238 block[1] = saved_key[idx + 1];
239 block[2] = saved_key[idx + 2];
240 block[3] = saved_key[idx + 3];
242 mdtransform (state, checksum, block, s_lotus_magic_table);
245 block[0] = saved_key[idx + 0];
246 block[1] = saved_key[idx + 1];
247 block[2] = saved_key[idx + 2];
248 block[3] = saved_key[idx + 3];
250 mdtransform (state, checksum, block, s_lotus_magic_table);
252 mdtransform_norecalc (state, checksum, s_lotus_magic_table);
255 __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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
261 const u32 lid = get_local_id (0);
267 const u32 gid = get_global_id (0);
273 __local u32 l_bin2asc[256];
275 const u32 lid4 = lid * 4;
277 const u32 lid40 = lid4 + 0;
278 const u32 lid41 = lid4 + 1;
279 const u32 lid42 = lid4 + 2;
280 const u32 lid43 = lid4 + 3;
282 const u32 v400 = (lid40 >> 0) & 15;
283 const u32 v401 = (lid40 >> 4) & 15;
284 const u32 v410 = (lid41 >> 0) & 15;
285 const u32 v411 = (lid41 >> 4) & 15;
286 const u32 v420 = (lid42 >> 0) & 15;
287 const u32 v421 = (lid42 >> 4) & 15;
288 const u32 v430 = (lid43 >> 0) & 15;
289 const u32 v431 = (lid43 >> 4) & 15;
291 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'A' - 10 + v400) << 8
292 | ((v401 < 10) ? '0' + v401 : 'A' - 10 + v401) << 0;
293 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'A' - 10 + v410) << 8
294 | ((v411 < 10) ? '0' + v411 : 'A' - 10 + v411) << 0;
295 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'A' - 10 + v420) << 8
296 | ((v421 < 10) ? '0' + v421 : 'A' - 10 + v421) << 0;
297 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'A' - 10 + v430) << 8
298 | ((v431 < 10) ? '0' + v431 : 'A' - 10 + v431) << 0;
300 barrier (CLK_LOCAL_MEM_FENCE);
306 __local u32 s_lotus_magic_table[256];
308 s_lotus_magic_table[lid4 + 0] = lotus_magic_table[lid4 + 0];
309 s_lotus_magic_table[lid4 + 1] = lotus_magic_table[lid4 + 1];
310 s_lotus_magic_table[lid4 + 2] = lotus_magic_table[lid4 + 2];
311 s_lotus_magic_table[lid4 + 3] = lotus_magic_table[lid4 + 3];
313 barrier (CLK_LOCAL_MEM_FENCE);
315 if (gid >= gid_max) return;
319 pw_buf0[0] = pws[gid].i[ 0];
320 pw_buf0[1] = pws[gid].i[ 1];
321 pw_buf0[2] = pws[gid].i[ 2];
322 pw_buf0[3] = pws[gid].i[ 3];
326 pw_buf1[0] = pws[gid].i[ 4];
327 pw_buf1[1] = pws[gid].i[ 5];
328 pw_buf1[2] = pws[gid].i[ 6];
329 pw_buf1[3] = pws[gid].i[ 7];
331 const u32 pw_len = pws[gid].pw_len;
337 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
338 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
344 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
374 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
408 pad (&w[ 0], pw_len & 0xf);
410 else if (pw_len < 32)
412 pad (&w[ 4], pw_len & 0xf);
414 else if (pw_len < 48)
416 pad (&w[ 8], pw_len & 0xf);
418 else if (pw_len < 64)
420 pad (&w[12], pw_len & 0xf);
423 domino_big_md (w, pw_len, state, s_lotus_magic_table);
425 const u32 w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
426 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
427 const u32 w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
428 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
429 const u32 w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
430 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
431 const u32 w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
432 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
433 const u32 w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
434 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
435 const u32 w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
436 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
437 const u32 w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
438 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
439 //const u32 w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
440 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
442 const u32 pade = 0x0e0e0e0e;
445 w[ 1] = salt1 | w0_t << 16;
446 w[ 2] = w0_t >> 16 | w1_t << 16;
447 w[ 3] = w1_t >> 16 | w2_t << 16;
448 w[ 4] = w2_t >> 16 | w3_t << 16;
449 w[ 5] = w3_t >> 16 | w4_t << 16;
450 w[ 6] = w4_t >> 16 | w5_t << 16;
451 w[ 7] = w5_t >> 16 | w6_t << 16;
452 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
466 domino_big_md (w, 34, state, s_lotus_magic_table);
468 u32 a = state[0] & 0xffffffff;
469 u32 b = state[1] & 0xffffffff;
470 u32 c = state[2] & 0x000000ff;
471 u32 d = state[3] & 0x00000000;
482 __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)
486 __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)
490 __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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
496 const u32 lid = get_local_id (0);
502 const u32 gid = get_global_id (0);
508 __local u32 l_bin2asc[256];
510 const u32 lid4 = lid * 4;
512 const u32 lid40 = lid4 + 0;
513 const u32 lid41 = lid4 + 1;
514 const u32 lid42 = lid4 + 2;
515 const u32 lid43 = lid4 + 3;
517 const u32 v400 = (lid40 >> 0) & 15;
518 const u32 v401 = (lid40 >> 4) & 15;
519 const u32 v410 = (lid41 >> 0) & 15;
520 const u32 v411 = (lid41 >> 4) & 15;
521 const u32 v420 = (lid42 >> 0) & 15;
522 const u32 v421 = (lid42 >> 4) & 15;
523 const u32 v430 = (lid43 >> 0) & 15;
524 const u32 v431 = (lid43 >> 4) & 15;
526 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'A' - 10 + v400) << 8
527 | ((v401 < 10) ? '0' + v401 : 'A' - 10 + v401) << 0;
528 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'A' - 10 + v410) << 8
529 | ((v411 < 10) ? '0' + v411 : 'A' - 10 + v411) << 0;
530 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'A' - 10 + v420) << 8
531 | ((v421 < 10) ? '0' + v421 : 'A' - 10 + v421) << 0;
532 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'A' - 10 + v430) << 8
533 | ((v431 < 10) ? '0' + v431 : 'A' - 10 + v431) << 0;
535 barrier (CLK_LOCAL_MEM_FENCE);
541 __local u32 s_lotus_magic_table[256];
543 s_lotus_magic_table[lid4 + 0] = lotus_magic_table[lid4 + 0];
544 s_lotus_magic_table[lid4 + 1] = lotus_magic_table[lid4 + 1];
545 s_lotus_magic_table[lid4 + 2] = lotus_magic_table[lid4 + 2];
546 s_lotus_magic_table[lid4 + 3] = lotus_magic_table[lid4 + 3];
548 barrier (CLK_LOCAL_MEM_FENCE);
550 if (gid >= gid_max) return;
554 pw_buf0[0] = pws[gid].i[ 0];
555 pw_buf0[1] = pws[gid].i[ 1];
556 pw_buf0[2] = pws[gid].i[ 2];
557 pw_buf0[3] = pws[gid].i[ 3];
561 pw_buf1[0] = pws[gid].i[ 4];
562 pw_buf1[1] = pws[gid].i[ 5];
563 pw_buf1[2] = pws[gid].i[ 6];
564 pw_buf1[3] = pws[gid].i[ 7];
566 const u32 pw_len = pws[gid].pw_len;
572 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
573 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
579 const u32 search[4] =
581 digests_buf[digests_offset].digest_buf[DGST_R0],
582 digests_buf[digests_offset].digest_buf[DGST_R1],
583 digests_buf[digests_offset].digest_buf[DGST_R2],
584 digests_buf[digests_offset].digest_buf[DGST_R3]
591 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
621 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
655 pad (&w[ 0], pw_len & 0xf);
657 else if (pw_len < 32)
659 pad (&w[ 4], pw_len & 0xf);
661 else if (pw_len < 48)
663 pad (&w[ 8], pw_len & 0xf);
665 else if (pw_len < 64)
667 pad (&w[12], pw_len & 0xf);
670 domino_big_md (w, pw_len, state, s_lotus_magic_table);
672 const u32 w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
673 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
674 const u32 w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
675 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
676 const u32 w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
677 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
678 const u32 w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
679 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
680 const u32 w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
681 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
682 const u32 w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
683 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
684 const u32 w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
685 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
686 //const u32 w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
687 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
689 const u32 pade = 0x0e0e0e0e;
692 w[ 1] = salt1 | w0_t << 16;
693 w[ 2] = w0_t >> 16 | w1_t << 16;
694 w[ 3] = w1_t >> 16 | w2_t << 16;
695 w[ 4] = w2_t >> 16 | w3_t << 16;
696 w[ 5] = w3_t >> 16 | w4_t << 16;
697 w[ 6] = w4_t >> 16 | w5_t << 16;
698 w[ 7] = w5_t >> 16 | w6_t << 16;
699 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
713 domino_big_md (w, 34, state, s_lotus_magic_table);
715 u32 a = state[0] & 0xffffffff;
716 u32 b = state[1] & 0xffffffff;
717 u32 c = state[2] & 0x000000ff;
718 u32 d = state[3] & 0x00000000;
729 __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)
733 __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)