2 * Author......: Jens Steube <jens.steube@gmail.com>
9 #include "include/constants.h"
10 #include "include/kernel_vendor.h"
25 #include "include/kernel_functions.c"
27 #include "common_nv.c"
30 #define VECT_COMPARE_S "check_single_vect1_comp4_warp.c"
31 #define VECT_COMPARE_M "check_multi_vect1_comp4_warp.c"
35 #define VECT_COMPARE_S "check_single_vect2_comp4_warp.c"
36 #define VECT_COMPARE_M "check_multi_vect2_comp4_warp.c"
39 __device__ __constant__ u32 lotus_magic_table[256] =
41 0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
42 0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
43 0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
44 0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
45 0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
46 0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
47 0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
48 0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
49 0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
50 0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
51 0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
52 0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
53 0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
54 0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
55 0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
56 0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
57 0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
58 0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
59 0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
60 0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
61 0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
62 0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
63 0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
64 0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
65 0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
66 0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
67 0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
68 0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
69 0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
70 0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
71 0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
72 0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
76 #define BOX(S,i) u32x ((S)[(i)])
80 #define BOX(S,i) u32x ((S)[(i).x], (S)[(i).y])
84 #define uint_to_hex_upper8(i) l_bin2asc[(i)]
88 #define uint_to_hex_upper8(i) u32x (l_bin2asc[(i).x], l_bin2asc[(i).y])
91 __device__ __constant__ char c_bin2asc[16] = { '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', 'A', 'B', 'C', 'D', 'E', 'F' };
93 __device__ __shared__ short l_bin2asc[256];
95 __device__ static void lotus_mix (u32x *in, u32 s_lotus_magic_table[256])
99 for (int i = 0; i < 18; i++)
104 for (int j = 0; j < 12; j++)
109 p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 0;
110 p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 8;
111 p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 16;
112 p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX (s_lotus_magic_table, p); tmp_out |= p << 24;
119 __device__ static void lotus_transform_password (u32x *in, u32x *out, u32 s_lotus_magic_table[256])
121 u32x t = out[3] >> 24;
125 for (int i = 0; i < 4; i++)
127 t ^= (in[i] >> 0) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 0; t = ((out[i] >> 0) & 0xff);
128 t ^= (in[i] >> 8) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 8; t = ((out[i] >> 8) & 0xff);
129 t ^= (in[i] >> 16) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 16; t = ((out[i] >> 16) & 0xff);
130 t ^= (in[i] >> 24) & 0xff; c = BOX (s_lotus_magic_table, t); out[i] ^= c << 24; t = ((out[i] >> 24) & 0xff);
134 __device__ static void pad (u32 w[4], const u32 len)
136 const u32 val = 16 - len;
138 const u32 mask1 = val << 24;
140 const u32 mask2 = val << 16
143 const u32 mask3 = val << 8
147 const u32 mask4 = val << 0
154 case 0: w[0] = mask4;
159 case 1: w[0] |= mask3;
164 case 2: w[0] |= mask2;
169 case 3: w[0] |= mask1;
174 case 4: w[1] = mask4;
178 case 5: w[1] |= mask3;
182 case 6: w[1] |= mask2;
186 case 7: w[1] |= mask1;
190 case 8: w[2] = mask4;
193 case 9: w[2] |= mask3;
196 case 10: w[2] |= mask2;
199 case 11: w[2] |= mask1;
202 case 12: w[3] = mask4;
204 case 13: w[3] |= mask3;
206 case 14: w[3] |= mask2;
208 case 15: w[3] |= mask1;
213 __device__ static void mdtransform_norecalc (u32x state[4], u32x block[4], u32 s_lotus_magic_table[256])
225 x[ 8] = state[0] ^ block[0];
226 x[ 9] = state[1] ^ block[1];
227 x[10] = state[2] ^ block[2];
228 x[11] = state[3] ^ block[3];
230 lotus_mix (x, s_lotus_magic_table);
238 __device__ static void mdtransform (u32x state[4], u32x checksum[4], u32x block[4], u32 s_lotus_magic_table[256])
240 mdtransform_norecalc (state, block, s_lotus_magic_table);
242 lotus_transform_password (block, checksum, s_lotus_magic_table);
245 __device__ static void domino_big_md (const u32x saved_key[16], const u32 size, u32x state[4], u32 s_lotus_magic_table[256])
264 for (curpos = 0, idx = 0; curpos + 16 < size; curpos += 16, idx += 4)
266 block[0] = saved_key[idx + 0];
267 block[1] = saved_key[idx + 1];
268 block[2] = saved_key[idx + 2];
269 block[3] = saved_key[idx + 3];
271 mdtransform (state, checksum, block, s_lotus_magic_table);
274 u32 left = size - curpos;
276 block[0] = saved_key[idx + 0];
277 block[1] = saved_key[idx + 1];
278 block[2] = saved_key[idx + 2];
279 block[3] = saved_key[idx + 3];
281 mdtransform (state, checksum, block, s_lotus_magic_table);
283 mdtransform_norecalc (state, checksum, s_lotus_magic_table);
286 __device__ __constant__ u32x c_bfs[1024];
288 __device__ static void m08700m (u32 s_lotus_magic_table[256], u32 w[16], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
294 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
295 const u32 lid = threadIdx.x;
303 pad (&w[ 0], pw_len & 0xf);
305 else if (pw_len < 32)
307 pad (&w[ 4], pw_len & 0xf);
309 else if (pw_len < 48)
311 pad (&w[ 8], pw_len & 0xf);
313 else if (pw_len < 64)
315 pad (&w[12], pw_len & 0xf);
322 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
323 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
329 const u32 bf_loops = ceil ((float) bfs_cnt / VECT_DIV);
333 for (u32 il_pos = 0; il_pos < bf_loops; il_pos++)
335 const u32x w0r = c_bfs[il_pos];
337 const u32x w0 = w0l | w0r;
365 domino_big_md (w_tmp, pw_len, state, s_lotus_magic_table);
367 const u32x w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
368 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
369 const u32x w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
370 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
371 const u32x w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
372 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
373 const u32x w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
374 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
375 const u32x w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
376 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
377 const u32x w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
378 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
379 const u32x w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
380 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
381 //const u32x w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
382 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
384 const u32 pade = 0x0e0e0e0e;
387 w_tmp[ 1] = salt1 | w0_t << 16;
388 w_tmp[ 2] = w0_t >> 16 | w1_t << 16;
389 w_tmp[ 3] = w1_t >> 16 | w2_t << 16;
390 w_tmp[ 4] = w2_t >> 16 | w3_t << 16;
391 w_tmp[ 5] = w3_t >> 16 | w4_t << 16;
392 w_tmp[ 6] = w4_t >> 16 | w5_t << 16;
393 w_tmp[ 7] = w5_t >> 16 | w6_t << 16;
394 w_tmp[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
408 domino_big_md (w_tmp, 34, state, s_lotus_magic_table);
410 u32x a = state[0] & 0xffffffff;
411 u32x b = state[1] & 0xffffffff;
412 u32x c = state[2] & 0x000000ff;
413 u32x d = state[3] & 0x00000000;
420 #include VECT_COMPARE_M
424 __device__ static void m08700s (u32 s_lotus_magic_table[256], u32 w[16], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
430 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
431 const u32 lid = threadIdx.x;
439 pad (&w[ 0], pw_len & 0xf);
441 else if (pw_len < 32)
443 pad (&w[ 4], pw_len & 0xf);
445 else if (pw_len < 48)
447 pad (&w[ 8], pw_len & 0xf);
449 else if (pw_len < 64)
451 pad (&w[12], pw_len & 0xf);
459 const u32 salt0 = salt_bufs[salt_pos].salt_buf[0];
460 const u32 salt1 = salt_bufs[salt_pos].salt_buf[1] & 0xff | '(' << 8;
466 const u32 search[4] =
468 digests_buf[digests_offset].digest_buf[DGST_R0],
469 digests_buf[digests_offset].digest_buf[DGST_R1],
470 digests_buf[digests_offset].digest_buf[DGST_R2],
471 digests_buf[digests_offset].digest_buf[DGST_R3]
478 const u32 bf_loops = ceil ((float) bfs_cnt / VECT_DIV);
482 for (u32 il_pos = 0; il_pos < bf_loops; il_pos++)
484 const u32x w0r = c_bfs[il_pos];
486 const u32x w0 = w0l | w0r;
514 domino_big_md (w_tmp, pw_len, state, s_lotus_magic_table);
516 const u32x w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
517 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
518 const u32x w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
519 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
520 const u32x w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
521 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
522 const u32x w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
523 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
524 const u32x w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
525 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
526 const u32x w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
527 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
528 const u32x w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
529 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
530 //const u32x w7_t = uint_to_hex_upper8 ((state[3] >> 16) & 255) << 0
531 // | uint_to_hex_upper8 ((state[3] >> 24) & 255) << 16;
533 const u32 pade = 0x0e0e0e0e;
536 w_tmp[ 1] = salt1 | w0_t << 16;
537 w_tmp[ 2] = w0_t >> 16 | w1_t << 16;
538 w_tmp[ 3] = w1_t >> 16 | w2_t << 16;
539 w_tmp[ 4] = w2_t >> 16 | w3_t << 16;
540 w_tmp[ 5] = w3_t >> 16 | w4_t << 16;
541 w_tmp[ 6] = w4_t >> 16 | w5_t << 16;
542 w_tmp[ 7] = w5_t >> 16 | w6_t << 16;
543 w_tmp[ 8] = w6_t >> 16 | pade << 16; // | w7_t << 8;
557 domino_big_md (w_tmp, 34, state, s_lotus_magic_table);
559 u32x a = state[0] & 0xffffffff;
560 u32x b = state[1] & 0xffffffff;
561 u32x c = state[2] & 0x000000ff;
562 u32x d = state[3] & 0x00000000;
569 #include VECT_COMPARE_S
573 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 u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
579 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
580 const u32 lid = threadIdx.x;
584 w[ 0] = pws[gid].i[ 0];
585 w[ 1] = pws[gid].i[ 1];
586 w[ 2] = pws[gid].i[ 2];
587 w[ 3] = pws[gid].i[ 3];
601 const u32 pw_len = pws[gid].pw_len;
607 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
608 | c_bin2asc[(lid >> 4) & 15] << 0;
616 __shared__ u32 s_lotus_magic_table[256];
618 s_lotus_magic_table[lid] = lotus_magic_table[lid];
622 if (gid >= gid_max) return;
628 m08700m (s_lotus_magic_table, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
631 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 u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
637 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
638 const u32 lid = threadIdx.x;
642 w[ 0] = pws[gid].i[ 0];
643 w[ 1] = pws[gid].i[ 1];
644 w[ 2] = pws[gid].i[ 2];
645 w[ 3] = pws[gid].i[ 3];
646 w[ 4] = pws[gid].i[ 4];
647 w[ 5] = pws[gid].i[ 5];
648 w[ 6] = pws[gid].i[ 6];
649 w[ 7] = pws[gid].i[ 7];
659 const u32 pw_len = pws[gid].pw_len;
665 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
666 | c_bin2asc[(lid >> 4) & 15] << 0;
674 __shared__ u32 s_lotus_magic_table[256];
676 s_lotus_magic_table[lid] = lotus_magic_table[lid];
680 if (gid >= gid_max) return;
686 m08700m (s_lotus_magic_table, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
689 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 u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
695 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
696 const u32 lid = threadIdx.x;
700 w[ 0] = pws[gid].i[ 0];
701 w[ 1] = pws[gid].i[ 1];
702 w[ 2] = pws[gid].i[ 2];
703 w[ 3] = pws[gid].i[ 3];
704 w[ 4] = pws[gid].i[ 4];
705 w[ 5] = pws[gid].i[ 5];
706 w[ 6] = pws[gid].i[ 6];
707 w[ 7] = pws[gid].i[ 7];
708 w[ 8] = pws[gid].i[ 8];
709 w[ 9] = pws[gid].i[ 9];
710 w[10] = pws[gid].i[10];
711 w[11] = pws[gid].i[11];
712 w[12] = pws[gid].i[12];
713 w[13] = pws[gid].i[13];
714 w[14] = pws[gid].i[14];
715 w[15] = pws[gid].i[15];
717 const u32 pw_len = pws[gid].pw_len;
723 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
724 | c_bin2asc[(lid >> 4) & 15] << 0;
732 __shared__ u32 s_lotus_magic_table[256];
734 s_lotus_magic_table[lid] = lotus_magic_table[lid];
738 if (gid >= gid_max) return;
744 m08700m (s_lotus_magic_table, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
747 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 u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
753 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
754 const u32 lid = threadIdx.x;
758 w[ 0] = pws[gid].i[ 0];
759 w[ 1] = pws[gid].i[ 1];
760 w[ 2] = pws[gid].i[ 2];
761 w[ 3] = pws[gid].i[ 3];
775 const u32 pw_len = pws[gid].pw_len;
781 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
782 | c_bin2asc[(lid >> 4) & 15] << 0;
790 __shared__ u32 s_lotus_magic_table[256];
792 s_lotus_magic_table[lid] = lotus_magic_table[lid];
796 if (gid >= gid_max) return;
802 m08700s (s_lotus_magic_table, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
805 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 u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
811 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
812 const u32 lid = threadIdx.x;
816 w[ 0] = pws[gid].i[ 0];
817 w[ 1] = pws[gid].i[ 1];
818 w[ 2] = pws[gid].i[ 2];
819 w[ 3] = pws[gid].i[ 3];
820 w[ 4] = pws[gid].i[ 4];
821 w[ 5] = pws[gid].i[ 5];
822 w[ 6] = pws[gid].i[ 6];
823 w[ 7] = pws[gid].i[ 7];
833 const u32 pw_len = pws[gid].pw_len;
839 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
840 | c_bin2asc[(lid >> 4) & 15] << 0;
848 __shared__ u32 s_lotus_magic_table[256];
850 s_lotus_magic_table[lid] = lotus_magic_table[lid];
854 if (gid >= gid_max) return;
860 m08700s (s_lotus_magic_table, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
863 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 u32x * words_buf_r, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
869 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
870 const u32 lid = threadIdx.x;
874 w[ 0] = pws[gid].i[ 0];
875 w[ 1] = pws[gid].i[ 1];
876 w[ 2] = pws[gid].i[ 2];
877 w[ 3] = pws[gid].i[ 3];
878 w[ 4] = pws[gid].i[ 4];
879 w[ 5] = pws[gid].i[ 5];
880 w[ 6] = pws[gid].i[ 6];
881 w[ 7] = pws[gid].i[ 7];
882 w[ 8] = pws[gid].i[ 8];
883 w[ 9] = pws[gid].i[ 9];
884 w[10] = pws[gid].i[10];
885 w[11] = pws[gid].i[11];
886 w[12] = pws[gid].i[12];
887 w[13] = pws[gid].i[13];
888 w[14] = pws[gid].i[14];
889 w[15] = pws[gid].i[15];
891 const u32 pw_len = pws[gid].pw_len;
897 l_bin2asc[lid] = c_bin2asc[(lid >> 0) & 15] << 8
898 | c_bin2asc[(lid >> 4) & 15] << 0;
906 __shared__ u32 s_lotus_magic_table[256];
908 s_lotus_magic_table[lid] = lotus_magic_table[lid];
912 if (gid >= gid_max) return;
918 m08700s (s_lotus_magic_table, w, pw_len, pws, rules_buf, combs_buf, words_buf_r, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, esalt_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);