2 * Authors.....: Jens Steube <jens.steube@gmail.com>
3 * Gabriele Gristina <matrix@hashcat.net>
4 * magnum <john.magnum@hushmail.com>
12 //#define NEW_SIMD_CODE
14 #include "inc_vendor.cl"
15 #include "inc_hash_constants.h"
16 #include "inc_hash_functions.cl"
17 #include "inc_types.cl"
18 #include "inc_common.cl"
21 #include "inc_simd.cl"
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 uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i)])
62 #define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
64 #define uint_to_hex_upper8(i) (u32x) (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
66 #define uint_to_hex_upper8(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])
68 #define uint_to_hex_upper8(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], l_bin2asc[(i).s8], l_bin2asc[(i).s9], l_bin2asc[(i).sa], l_bin2asc[(i).sb], l_bin2asc[(i).sc], l_bin2asc[(i).sd], l_bin2asc[(i).se], l_bin2asc[(i).sf])
72 #define BOX1(S,i) (S)[(i)]
74 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1])
76 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3])
78 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7])
80 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7], (S)[(i).s8], (S)[(i).s9], (S)[(i).sa], (S)[(i).sb], (S)[(i).sc], (S)[(i).sd], (S)[(i).se], (S)[(i).sf])
83 void lotus_mix (u32x *in, __local u32 *s_lotus_magic_table)
87 for (int i = 0; i < 18; i++)
91 for (int j = 0; j < 12; j++)
96 p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
97 p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
98 p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
99 p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
106 void lotus_transform_password (u32x in[4], u32x out[4], __local u32 *s_lotus_magic_table)
108 u32x t = out[3] >> 24;
115 for (int i = 0; i < 4; i++)
117 t ^= (in[i] >> 0) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 0; t = ((out[i] >> 0) & 0xff);
118 t ^= (in[i] >> 8) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 8; t = ((out[i] >> 8) & 0xff);
119 t ^= (in[i] >> 16) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 16; t = ((out[i] >> 16) & 0xff);
120 t ^= (in[i] >> 24) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 24; t = ((out[i] >> 24) & 0xff);
124 void pad (u32 w[4], const u32 len)
126 const u32 val = 16 - len;
128 const u32 mask1 = val << 24;
130 const u32 mask2 = val << 16
133 const u32 mask3 = val << 8
137 const u32 mask4 = val << 0
144 case 0: w[0] = mask4;
149 case 1: w[0] |= mask3;
154 case 2: w[0] |= mask2;
159 case 3: w[0] |= mask1;
164 case 4: w[1] = mask4;
168 case 5: w[1] |= mask3;
172 case 6: w[1] |= mask2;
176 case 7: w[1] |= mask1;
180 case 8: w[2] = mask4;
183 case 9: w[2] |= mask3;
186 case 10: w[2] |= mask2;
189 case 11: w[2] |= mask1;
192 case 12: w[3] = mask4;
194 case 13: w[3] |= mask3;
196 case 14: w[3] |= mask2;
198 case 15: w[3] |= mask1;
203 void mdtransform_norecalc (u32x state[4], u32x block[4], __local u32 *s_lotus_magic_table)
215 x[ 8] = state[0] ^ block[0];
216 x[ 9] = state[1] ^ block[1];
217 x[10] = state[2] ^ block[2];
218 x[11] = state[3] ^ block[3];
220 lotus_mix (x, s_lotus_magic_table);
228 void mdtransform (u32x state[4], u32x checksum[4], u32x block[4], __local u32 *s_lotus_magic_table)
230 mdtransform_norecalc (state, block, s_lotus_magic_table);
232 lotus_transform_password (block, checksum, s_lotus_magic_table);
235 void domino_big_md (const u32x saved_key[16], const u32 size, u32x state[4], __local u32 *s_lotus_magic_table)
254 for (curpos = 0, idx = 0; curpos + 16 < size; curpos += 16, idx += 4)
256 block[0] = saved_key[idx + 0];
257 block[1] = saved_key[idx + 1];
258 block[2] = saved_key[idx + 2];
259 block[3] = saved_key[idx + 3];
261 mdtransform (state, checksum, block, s_lotus_magic_table);
264 block[0] = saved_key[idx + 0];
265 block[1] = saved_key[idx + 1];
266 block[2] = saved_key[idx + 2];
267 block[3] = saved_key[idx + 3];
269 mdtransform (state, checksum, block, s_lotus_magic_table);
271 mdtransform_norecalc (state, checksum, s_lotus_magic_table);
274 __kernel void m08700_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
280 const u32 gid = get_global_id (0);
281 const u32 lid = get_local_id (0);
282 const u32 lsz = get_local_size (0);
288 __local u32 s_lotus_magic_table[256];
290 for (u32 i = lid; i < 256; i += lsz)
292 s_lotus_magic_table[i] = lotus_magic_table[i];
295 __local u32 l_bin2asc[256];
297 for (u32 i = lid; i < 256; i += lsz)
299 const u32 i0 = (i >> 0) & 15;
300 const u32 i1 = (i >> 4) & 15;
302 l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'A' - 10 + i0) << 8
303 | ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
306 barrier (CLK_LOCAL_MEM_FENCE);
308 if (gid >= gid_max) return;
317 pw_buf0[0] = pws[gid].i[ 0];
318 pw_buf0[1] = pws[gid].i[ 1];
319 pw_buf0[2] = pws[gid].i[ 2];
320 pw_buf0[3] = pws[gid].i[ 3];
321 pw_buf1[0] = pws[gid].i[ 4];
322 pw_buf1[1] = pws[gid].i[ 5];
323 pw_buf1[2] = pws[gid].i[ 6];
324 pw_buf1[3] = pws[gid].i[ 7];
326 const u32 pw_len = pws[gid].pw_len;
332 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
333 const u32 salt1 = (salt_bufs[salt_pos].salt_buf[1] & 0xff) | '(' << 8;
339 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
346 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
384 pad (&w[ 0], out_len & 0xf);
386 else if (out_len < 32)
388 pad (&w[ 4], out_len & 0xf);
390 else if (out_len < 48)
392 pad (&w[ 8], out_len & 0xf);
394 else if (out_len < 64)
396 pad (&w[12], out_len & 0xf);
399 domino_big_md (w, out_len, state, s_lotus_magic_table);
401 const u32x w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
402 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
403 const u32x w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
404 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
405 const u32x w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
406 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
407 const u32x w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
408 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
409 const u32x w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
410 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
411 const u32x w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
412 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
413 const u32x w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
414 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
415 //const u32x w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
416 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
418 const u32 pade = 0x0e0e0e0e;
421 w[ 1] = salt1 | w0_t << 16;
422 w[ 2] = w0_t >> 16 | w1_t << 16;
423 w[ 3] = w1_t >> 16 | w2_t << 16;
424 w[ 4] = w2_t >> 16 | w3_t << 16;
425 w[ 5] = w3_t >> 16 | w4_t << 16;
426 w[ 6] = w4_t >> 16 | w5_t << 16;
427 w[ 7] = w5_t >> 16 | w6_t << 16;
428 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
442 domino_big_md (w, 34, state, s_lotus_magic_table);
444 u32x a = state[0] & 0xffffffff;
445 u32x b = state[1] & 0xffffffff;
446 u32x c = state[2] & 0x000000ff;
447 u32x d = state[3] & 0x00000000;
449 COMPARE_M_SIMD (a, b, c, d);
453 __kernel void m08700_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
457 __kernel void m08700_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
461 __kernel void m08700_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
467 const u32 gid = get_global_id (0);
468 const u32 lid = get_local_id (0);
469 const u32 lsz = get_local_size (0);
475 __local u32 s_lotus_magic_table[256];
477 for (u32 i = lid; i < 256; i += lsz)
479 s_lotus_magic_table[i] = lotus_magic_table[i];
482 __local u32 l_bin2asc[256];
484 for (u32 i = lid; i < 256; i += lsz)
486 const u32 i0 = (i >> 0) & 15;
487 const u32 i1 = (i >> 4) & 15;
489 l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'A' - 10 + i0) << 8
490 | ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
493 barrier (CLK_LOCAL_MEM_FENCE);
495 if (gid >= gid_max) return;
504 pw_buf0[0] = pws[gid].i[ 0];
505 pw_buf0[1] = pws[gid].i[ 1];
506 pw_buf0[2] = pws[gid].i[ 2];
507 pw_buf0[3] = pws[gid].i[ 3];
508 pw_buf1[0] = pws[gid].i[ 4];
509 pw_buf1[1] = pws[gid].i[ 5];
510 pw_buf1[2] = pws[gid].i[ 6];
511 pw_buf1[3] = pws[gid].i[ 7];
513 const u32 pw_len = pws[gid].pw_len;
519 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
520 const u32 salt1 = (salt_bufs[salt_pos].salt_buf[1] & 0xff) | '(' << 8;
526 const u32 search[4] =
528 digests_buf[digests_offset].digest_buf[DGST_R0],
529 digests_buf[digests_offset].digest_buf[DGST_R1],
530 digests_buf[digests_offset].digest_buf[DGST_R2],
531 digests_buf[digests_offset].digest_buf[DGST_R3]
538 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
545 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
583 pad (&w[ 0], out_len & 0xf);
585 else if (out_len < 32)
587 pad (&w[ 4], out_len & 0xf);
589 else if (out_len < 48)
591 pad (&w[ 8], out_len & 0xf);
593 else if (out_len < 64)
595 pad (&w[12], out_len & 0xf);
598 domino_big_md (w, out_len, state, s_lotus_magic_table);
600 const u32x w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
601 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
602 const u32x w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
603 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
604 const u32x w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
605 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
606 const u32x w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
607 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
608 const u32x w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
609 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
610 const u32x w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
611 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
612 const u32x w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
613 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
614 //const u32x w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
615 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
617 const u32 pade = 0x0e0e0e0e;
620 w[ 1] = salt1 | w0_t << 16;
621 w[ 2] = w0_t >> 16 | w1_t << 16;
622 w[ 3] = w1_t >> 16 | w2_t << 16;
623 w[ 4] = w2_t >> 16 | w3_t << 16;
624 w[ 5] = w3_t >> 16 | w4_t << 16;
625 w[ 6] = w4_t >> 16 | w5_t << 16;
626 w[ 7] = w5_t >> 16 | w6_t << 16;
627 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
641 domino_big_md (w, 34, state, s_lotus_magic_table);
643 u32x a = state[0] & 0xffffffff;
644 u32x b = state[1] & 0xffffffff;
645 u32x c = state[2] & 0x000000ff;
646 u32x d = state[3] & 0x00000000;
648 COMPARE_S_SIMD (a, b, c, d);
652 __kernel void m08700_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
656 __kernel void m08700_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)