2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
24 #include "include/kernel_functions.c"
26 #include "common_nv.c"
29 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
30 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
34 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
35 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
38 __device__ __constant__ u32 lotus_magic_table[256] =
40 0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
41 0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
42 0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
43 0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
44 0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
45 0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
46 0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
47 0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
48 0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
49 0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
50 0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
51 0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
52 0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
53 0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
54 0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
55 0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
56 0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
57 0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
58 0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
59 0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
60 0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
61 0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
62 0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
63 0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
64 0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
65 0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
66 0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
67 0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
68 0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
69 0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
70 0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
71 0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
75 #define BOX(S,i) u32x ((S)[(i)])
79 #define BOX(S,i) u32x ((S)[(i).x], (S)[(i).y])
83 #define uint_to_hex_upper8(i) l_bin2asc[(i)]
87 #define uint_to_hex_upper8(i) u32x (l_bin2asc[(i).x], l_bin2asc[(i).y])
90 __device__ __constant__ char c_bin2asc[16] = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' };
92 __device__ __shared__ short l_bin2asc[256];
94 __device__ static void lotus_mix (u32x *in, u32 s_lotus_magic_table[256])
98 for (int i = 0; i < 18; i++)
103 for (int j = 0; j < 12; j++)
108 p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 0;
109 p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 8;
110 p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 16;
111 p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 24;
118 __device__ static void lotus_transform_password (u32x *in, u32x *out, u32 s_lotus_magic_table[256])
120 u32x t = out[3] >> 24;
124 for (int i = 0; i < 4; i++)
126 t ^= (in[i] >> 0) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 0; t = ((out[i] >> 0) & 0xff);
127 t ^= (in[i] >> 8) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 8; t = ((out[i] >> 8) & 0xff);
128 t ^= (in[i] >> 16) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 16; t = ((out[i] >> 16) & 0xff);
129 t ^= (in[i] >> 24) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 24; t = ((out[i] >> 24) & 0xff);
133 __device__ static void pad (u32x w[4], const u32 len)
135 const u32 val = 16 - len;
137 const u32 mask1 = val << 24;
139 const u32 mask2 = val << 16
142 const u32 mask3 = val << 8
146 const u32 mask4 = val << 0
153 case 0: w[0] = mask4;
158 case 1: w[0] |= mask3;
163 case 2: w[0] |= mask2;
168 case 3: w[0] |= mask1;
173 case 4: w[1] = mask4;
177 case 5: w[1] |= mask3;
181 case 6: w[1] |= mask2;
185 case 7: w[1] |= mask1;
189 case 8: w[2] = mask4;
192 case 9: w[2] |= mask3;
195 case 10: w[2] |= mask2;
198 case 11: w[2] |= mask1;
201 case 12: w[3] = mask4;
203 case 13: w[3] |= mask3;
205 case 14: w[3] |= mask2;
207 case 15: w[3] |= mask1;
212 __device__ static void mdtransform_norecalc (u32x state[4], u32x block[4], u32 s_lotus_magic_table[256])
224 x[ 8] = state[0] ^ block[0];
225 x[ 9] = state[1] ^ block[1];
226 x[10] = state[2] ^ block[2];
227 x[11] = state[3] ^ block[3];
229 lotus_mix (x, s_lotus_magic_table);
237 __device__ static void mdtransform (u32x state[4], u32x checksum[4], u32x block[4], u32 s_lotus_magic_table[256])
239 mdtransform_norecalc (state, block, s_lotus_magic_table);
241 lotus_transform_password (block, checksum, s_lotus_magic_table);
244 __device__ static void domino_big_md (const u32x saved_key[16], const u32 size, u32x state[4], u32 s_lotus_magic_table[256])
263 for (curpos = 0, idx = 0; curpos + 16 < size; curpos += 16, idx += 4)
265 block[0] = saved_key[idx + 0];
266 block[1] = saved_key[idx + 1];
267 block[2] = saved_key[idx + 2];
268 block[3] = saved_key[idx + 3];
270 mdtransform (state, checksum, block, s_lotus_magic_table);
273 u32 left = size - curpos;
275 block[0] = saved_key[idx + 0];
276 block[1] = saved_key[idx + 1];
277 block[2] = saved_key[idx + 2];
278 block[3] = saved_key[idx + 3];
280 mdtransform (state, checksum, block, s_lotus_magic_table);
282 mdtransform_norecalc (state, checksum, s_lotus_magic_table);
285 __device__ __constant__ comb_t c_combs[1024];
287 extern "C" __global__ void __launch_bounds__ (256, 1) m08700_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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)
293 const u32 lid = threadIdx.x;
299 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
305 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
306 | c_bin2asc[(lid >> 4) & 15] << 0;
314 __shared__ u32 s_lotus_magic_table[256];
316 s_lotus_magic_table[lid] = lotus_magic_table[lid];
320 if (gid >= gid_max) return;
328 wordl0[0] = pws[gid].i[ 0];
329 wordl0[1] = pws[gid].i[ 1];
330 wordl0[2] = pws[gid].i[ 2];
331 wordl0[3] = pws[gid].i[ 3];
335 wordl1[0] = pws[gid].i[ 4];
336 wordl1[1] = pws[gid].i[ 5];
337 wordl1[2] = pws[gid].i[ 6];
338 wordl1[3] = pws[gid].i[ 7];
354 const u32 pw_l_len = pws[gid].pw_len;
356 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
358 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
365 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
366 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
372 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
374 const u32 pw_r_len = c_combs[il_pos].pw_len;
376 const u32 pw_len = pw_l_len + pw_r_len;
380 wordr0[0] = c_combs[il_pos].i[0];
381 wordr0[1] = c_combs[il_pos].i[1];
382 wordr0[2] = c_combs[il_pos].i[2];
383 wordr0[3] = c_combs[il_pos].i[3];
387 wordr1[0] = c_combs[il_pos].i[4];
388 wordr1[1] = c_combs[il_pos].i[5];
389 wordr1[2] = c_combs[il_pos].i[6];
390 wordr1[3] = c_combs[il_pos].i[7];
406 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
408 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
413 w[ 0] = wordl0[0] | wordr0[0];
414 w[ 1] = wordl0[1] | wordr0[1];
415 w[ 2] = wordl0[2] | wordr0[2];
416 w[ 3] = wordl0[3] | wordr0[3];
417 w[ 4] = wordl1[0] | wordr1[0];
418 w[ 5] = wordl1[1] | wordr1[1];
419 w[ 6] = wordl1[2] | wordr1[2];
420 w[ 7] = wordl1[3] | wordr1[3];
421 w[ 8] = wordl2[0] | wordr2[0];
422 w[ 9] = wordl2[1] | wordr2[1];
423 w[10] = wordl2[2] | wordr2[2];
424 w[11] = wordl2[3] | wordr2[3];
425 w[12] = wordl3[0] | wordr3[0];
426 w[13] = wordl3[1] | wordr3[1];
427 w[14] = wordl3[2] | wordr3[2];
428 w[15] = wordl3[3] | wordr3[3];
443 pad (&w[ 0], pw_len & 0xf);
445 else if (pw_len < 32)
447 pad (&w[ 4], pw_len & 0xf);
449 else if (pw_len < 48)
451 pad (&w[ 8], pw_len & 0xf);
453 else if (pw_len < 64)
455 pad (&w[12], pw_len & 0xf);
458 domino_big_md (w, pw_len, state, s_lotus_magic_table);
460 const u32x w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
461 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
462 const u32x w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
463 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
464 const u32x w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
465 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
466 const u32x w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
467 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
468 const u32x w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
469 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
470 const u32x w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
471 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
472 const u32x w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
473 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
474 //const u32x w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
475 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
477 const u32 pade = 0x0e0e0e0e;
480 w[ 1] = salt1 | w0_t << 16;
481 w[ 2] = w0_t >> 16 | w1_t << 16;
482 w[ 3] = w1_t >> 16 | w2_t << 16;
483 w[ 4] = w2_t >> 16 | w3_t << 16;
484 w[ 5] = w3_t >> 16 | w4_t << 16;
485 w[ 6] = w4_t >> 16 | w5_t << 16;
486 w[ 7] = w5_t >> 16 | w6_t << 16;
487 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
501 domino_big_md (w, 34, state, s_lotus_magic_table);
503 u32x a = state[0] & 0xffffffff;
504 u32x b = state[1] & 0xffffffff;
505 u32x c = state[2] & 0x000000ff;
506 u32x d = state[3] & 0x00000000;
513 #include VECT_COMPARE_M
517 extern "C" __global__ void __launch_bounds__ (256, 1) m08700_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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)
521 extern "C" __global__ void __launch_bounds__ (256, 1) m08700_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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)
525 extern "C" __global__ void __launch_bounds__ (256, 1) m08700_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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 const u32 lid = threadIdx.x;
537 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
543 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
544 | c_bin2asc[(lid >> 4) & 15] << 0;
552 __shared__ u32 s_lotus_magic_table[256];
554 s_lotus_magic_table[lid] = lotus_magic_table[lid];
558 if (gid >= gid_max) return;
566 wordl0[0] = pws[gid].i[ 0];
567 wordl0[1] = pws[gid].i[ 1];
568 wordl0[2] = pws[gid].i[ 2];
569 wordl0[3] = pws[gid].i[ 3];
573 wordl1[0] = pws[gid].i[ 4];
574 wordl1[1] = pws[gid].i[ 5];
575 wordl1[2] = pws[gid].i[ 6];
576 wordl1[3] = pws[gid].i[ 7];
593 const u32 pw_l_len = pws[gid].pw_len;
595 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
597 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
604 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
605 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
611 const u32 search[4] =
613 digests_buf[digests_offset].digest_buf[DGST_R0],
614 digests_buf[digests_offset].digest_buf[DGST_R1],
615 digests_buf[digests_offset].digest_buf[DGST_R2],
616 digests_buf[digests_offset].digest_buf[DGST_R3]
623 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
625 const u32 pw_r_len = c_combs[il_pos].pw_len;
627 const u32 pw_len = pw_l_len + pw_r_len;
631 wordr0[0] = c_combs[il_pos].i[0];
632 wordr0[1] = c_combs[il_pos].i[1];
633 wordr0[2] = c_combs[il_pos].i[2];
634 wordr0[3] = c_combs[il_pos].i[3];
638 wordr1[0] = c_combs[il_pos].i[4];
639 wordr1[1] = c_combs[il_pos].i[5];
640 wordr1[2] = c_combs[il_pos].i[6];
641 wordr1[3] = c_combs[il_pos].i[7];
657 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
659 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
664 w[ 0] = wordl0[0] | wordr0[0];
665 w[ 1] = wordl0[1] | wordr0[1];
666 w[ 2] = wordl0[2] | wordr0[2];
667 w[ 3] = wordl0[3] | wordr0[3];
668 w[ 4] = wordl1[0] | wordr1[0];
669 w[ 5] = wordl1[1] | wordr1[1];
670 w[ 6] = wordl1[2] | wordr1[2];
671 w[ 7] = wordl1[3] | wordr1[3];
672 w[ 8] = wordl2[0] | wordr2[0];
673 w[ 9] = wordl2[1] | wordr2[1];
674 w[10] = wordl2[2] | wordr2[2];
675 w[11] = wordl2[3] | wordr2[3];
676 w[12] = wordl3[0] | wordr3[0];
677 w[13] = wordl3[1] | wordr3[1];
678 w[14] = wordl3[2] | wordr3[2];
679 w[15] = wordl3[3] | wordr3[3];
694 pad (&w[ 0], pw_len & 0xf);
696 else if (pw_len < 32)
698 pad (&w[ 4], pw_len & 0xf);
700 else if (pw_len < 48)
702 pad (&w[ 8], pw_len & 0xf);
704 else if (pw_len < 64)
706 pad (&w[12], pw_len & 0xf);
709 domino_big_md (w, pw_len, state, s_lotus_magic_table);
711 const u32x w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
712 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
713 const u32x w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
714 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
715 const u32x w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
716 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
717 const u32x w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
718 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
719 const u32x w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
720 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
721 const u32x w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
722 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
723 const u32x w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
724 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
725 //const u32x w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
726 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
728 const u32 pade = 0x0e0e0e0e;
731 w[ 1] = salt1 | w0_t << 16;
732 w[ 2] = w0_t >> 16 | w1_t << 16;
733 w[ 3] = w1_t >> 16 | w2_t << 16;
734 w[ 4] = w2_t >> 16 | w3_t << 16;
735 w[ 5] = w3_t >> 16 | w4_t << 16;
736 w[ 6] = w4_t >> 16 | w5_t << 16;
737 w[ 7] = w5_t >> 16 | w6_t << 16;
738 w[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
752 domino_big_md (w, 34, state, s_lotus_magic_table);
754 u32x a = state[0] & 0xffffffff;
755 u32x b = state[1] & 0xffffffff;
756 u32x c = state[2] & 0x000000ff;
757 u32x d = state[3] & 0x00000000;
764 #include VECT_COMPARE_S
768 extern "C" __global__ void __launch_bounds__ (256, 1) m08700_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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)
772 extern "C" __global__ void __launch_bounds__ (256, 1) m08700_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const void *esalt_bufs, u32 *d_return_buf, 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)