2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
28 #include "include/kernel_functions.c"
29 #include "types_amd.c"
30 #include "common_amd.c"
33 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
34 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
38 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
39 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
43 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
44 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
47 #define INITVAL 0x0101010101010101
49 #define SBOG_LPSti64 \
50 s_sbob_sl64[0][(t[0] >> (i * 8)) & 0xff] ^ \
51 s_sbob_sl64[1][(t[1] >> (i * 8)) & 0xff] ^ \
52 s_sbob_sl64[2][(t[2] >> (i * 8)) & 0xff] ^ \
53 s_sbob_sl64[3][(t[3] >> (i * 8)) & 0xff] ^ \
54 s_sbob_sl64[4][(t[4] >> (i * 8)) & 0xff] ^ \
55 s_sbob_sl64[5][(t[5] >> (i * 8)) & 0xff] ^ \
56 s_sbob_sl64[6][(t[6] >> (i * 8)) & 0xff] ^ \
57 s_sbob_sl64[7][(t[7] >> (i * 8)) & 0xff]
61 __constant u64 sbob_sl64[8][256] =
2129 __constant u64 sbob_rc64[12][8] =
2253 static void streebog_g (u64 h[8], const u64 m[8], __local u64 s_sbob_sl64[8][256])
2260 for (int i = 0; i < 8; i++)
2265 for (int i = 0; i < 8; i++)
2267 k[i] = SBOG_LPSti64;
2271 for (int i = 0; i < 8; i++)
2276 for (int r = 0; r < 12; r++)
2279 for (int i = 0; i < 8; i++)
2285 for (int i = 0; i < 8; i++)
2287 s[i] = SBOG_LPSti64;
2290 for (int i = 0; i < 8; i++)
2292 t[i] = k[i] ^ sbob_rc64[r][i];
2296 for (int i = 0; i < 8; i++)
2298 k[i] = SBOG_LPSti64;
2303 for (int i = 0; i < 8; i++)
2305 h[i] ^= s[i] ^ k[i] ^ m[i];
2309 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11700_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)
2315 const u32 lid = get_local_id (0);
2318 * shared lookup table
2321 const u32 lid4 = lid * 4;
2323 __local u64 s_sbob_sl64[8][256];
2325 s_sbob_sl64[0][lid4 + 0] = sbob_sl64[0][lid4 + 0];
2326 s_sbob_sl64[0][lid4 + 1] = sbob_sl64[0][lid4 + 1];
2327 s_sbob_sl64[0][lid4 + 2] = sbob_sl64[0][lid4 + 2];
2328 s_sbob_sl64[0][lid4 + 3] = sbob_sl64[0][lid4 + 3];
2329 s_sbob_sl64[1][lid4 + 0] = sbob_sl64[1][lid4 + 0];
2330 s_sbob_sl64[1][lid4 + 1] = sbob_sl64[1][lid4 + 1];
2331 s_sbob_sl64[1][lid4 + 2] = sbob_sl64[1][lid4 + 2];
2332 s_sbob_sl64[1][lid4 + 3] = sbob_sl64[1][lid4 + 3];
2333 s_sbob_sl64[2][lid4 + 0] = sbob_sl64[2][lid4 + 0];
2334 s_sbob_sl64[2][lid4 + 1] = sbob_sl64[2][lid4 + 1];
2335 s_sbob_sl64[2][lid4 + 2] = sbob_sl64[2][lid4 + 2];
2336 s_sbob_sl64[2][lid4 + 3] = sbob_sl64[2][lid4 + 3];
2337 s_sbob_sl64[3][lid4 + 0] = sbob_sl64[3][lid4 + 0];
2338 s_sbob_sl64[3][lid4 + 1] = sbob_sl64[3][lid4 + 1];
2339 s_sbob_sl64[3][lid4 + 2] = sbob_sl64[3][lid4 + 2];
2340 s_sbob_sl64[3][lid4 + 3] = sbob_sl64[3][lid4 + 3];
2341 s_sbob_sl64[4][lid4 + 0] = sbob_sl64[4][lid4 + 0];
2342 s_sbob_sl64[4][lid4 + 1] = sbob_sl64[4][lid4 + 1];
2343 s_sbob_sl64[4][lid4 + 2] = sbob_sl64[4][lid4 + 2];
2344 s_sbob_sl64[4][lid4 + 3] = sbob_sl64[4][lid4 + 3];
2345 s_sbob_sl64[5][lid4 + 0] = sbob_sl64[5][lid4 + 0];
2346 s_sbob_sl64[5][lid4 + 1] = sbob_sl64[5][lid4 + 1];
2347 s_sbob_sl64[5][lid4 + 2] = sbob_sl64[5][lid4 + 2];
2348 s_sbob_sl64[5][lid4 + 3] = sbob_sl64[5][lid4 + 3];
2349 s_sbob_sl64[6][lid4 + 0] = sbob_sl64[6][lid4 + 0];
2350 s_sbob_sl64[6][lid4 + 1] = sbob_sl64[6][lid4 + 1];
2351 s_sbob_sl64[6][lid4 + 2] = sbob_sl64[6][lid4 + 2];
2352 s_sbob_sl64[6][lid4 + 3] = sbob_sl64[6][lid4 + 3];
2353 s_sbob_sl64[7][lid4 + 0] = sbob_sl64[7][lid4 + 0];
2354 s_sbob_sl64[7][lid4 + 1] = sbob_sl64[7][lid4 + 1];
2355 s_sbob_sl64[7][lid4 + 2] = sbob_sl64[7][lid4 + 2];
2356 s_sbob_sl64[7][lid4 + 3] = sbob_sl64[7][lid4 + 3];
2358 barrier (CLK_LOCAL_MEM_FENCE);
2364 const u32 gid = get_global_id (0);
2366 if (gid >= gid_max) return;
2370 wordl0[0] = pws[gid].i[ 0];
2371 wordl0[1] = pws[gid].i[ 1];
2372 wordl0[2] = pws[gid].i[ 2];
2373 wordl0[3] = pws[gid].i[ 3];
2377 wordl1[0] = pws[gid].i[ 4];
2378 wordl1[1] = pws[gid].i[ 5];
2379 wordl1[2] = pws[gid].i[ 6];
2380 wordl1[3] = pws[gid].i[ 7];
2396 const u32 pw_l_len = pws[gid].pw_len;
2398 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
2400 append_0x80_2 (wordl0, wordl1, pw_l_len);
2402 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
2409 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
2411 const u32 pw_r_len = combs_buf[il_pos].pw_len;
2413 const u32 pw_len = pw_l_len + pw_r_len;
2417 wordr0[0] = combs_buf[il_pos].i[0];
2418 wordr0[1] = combs_buf[il_pos].i[1];
2419 wordr0[2] = combs_buf[il_pos].i[2];
2420 wordr0[3] = combs_buf[il_pos].i[3];
2424 wordr1[0] = combs_buf[il_pos].i[4];
2425 wordr1[1] = combs_buf[il_pos].i[5];
2426 wordr1[2] = combs_buf[il_pos].i[6];
2427 wordr1[3] = combs_buf[il_pos].i[7];
2443 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
2445 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
2450 w[ 0] = wordl0[0] | wordr0[0];
2451 w[ 1] = wordl0[1] | wordr0[1];
2452 w[ 2] = wordl0[2] | wordr0[2];
2453 w[ 3] = wordl0[3] | wordr0[3];
2454 w[ 4] = wordl1[0] | wordr1[0];
2455 w[ 5] = wordl1[1] | wordr1[1];
2456 w[ 6] = wordl1[2] | wordr1[2];
2457 w[ 7] = wordl1[3] | wordr1[3];
2458 w[ 8] = wordl2[0] | wordr2[0];
2459 w[ 9] = wordl2[1] | wordr2[1];
2460 w[10] = wordl2[2] | wordr2[2];
2461 w[11] = wordl2[3] | wordr2[3];
2462 w[12] = wordl3[0] | wordr3[0];
2463 w[13] = wordl3[1] | wordr3[1];
2464 w[14] = wordl3[1] | wordr3[1];
2465 w[15] = wordl3[1] | wordr3[1];
2467 append_0x01_4 (&w[0], &w[1], &w[2], &w[3], pw_len);
2470 * reverse message block
2475 m[0] = hl32_to_64 (w[15], w[14]);
2476 m[1] = hl32_to_64 (w[13], w[12]);
2477 m[2] = hl32_to_64 (w[11], w[10]);
2478 m[3] = hl32_to_64 (w[ 9], w[ 8]);
2479 m[4] = hl32_to_64 (w[ 7], w[ 6]);
2480 m[5] = hl32_to_64 (w[ 5], w[ 4]);
2481 m[6] = hl32_to_64 (w[ 3], w[ 2]);
2482 m[7] = hl32_to_64 (w[ 1], w[ 0]);
2484 m[0] = swap_workaround (m[0]);
2485 m[1] = swap_workaround (m[1]);
2486 m[2] = swap_workaround (m[2]);
2487 m[3] = swap_workaround (m[3]);
2488 m[4] = swap_workaround (m[4]);
2489 m[5] = swap_workaround (m[5]);
2490 m[6] = swap_workaround (m[6]);
2491 m[7] = swap_workaround (m[7]);
2493 // state buffer (hash)
2506 streebog_g (h, m, s_sbob_sl64);
2517 z[7] = swap_workaround ((u64) (pw_len * 8));
2519 streebog_g (h, z, s_sbob_sl64);
2520 streebog_g (h, m, s_sbob_sl64);
2522 const u32 r0 = l32_from_64 (h[0]);
2523 const u32 r1 = h32_from_64 (h[0]);
2524 const u32 r2 = l32_from_64 (h[1]);
2525 const u32 r3 = h32_from_64 (h[1]);
2527 #include VECT_COMPARE_M
2531 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11700_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)
2535 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11700_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)
2539 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11700_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)
2545 const u32 lid = get_local_id (0);
2549 * shared lookup table
2552 const u32 lid4 = lid * 4;
2554 __local u64 s_sbob_sl64[8][256];
2556 s_sbob_sl64[0][lid4 + 0] = sbob_sl64[0][lid4 + 0];
2557 s_sbob_sl64[0][lid4 + 1] = sbob_sl64[0][lid4 + 1];
2558 s_sbob_sl64[0][lid4 + 2] = sbob_sl64[0][lid4 + 2];
2559 s_sbob_sl64[0][lid4 + 3] = sbob_sl64[0][lid4 + 3];
2560 s_sbob_sl64[1][lid4 + 0] = sbob_sl64[1][lid4 + 0];
2561 s_sbob_sl64[1][lid4 + 1] = sbob_sl64[1][lid4 + 1];
2562 s_sbob_sl64[1][lid4 + 2] = sbob_sl64[1][lid4 + 2];
2563 s_sbob_sl64[1][lid4 + 3] = sbob_sl64[1][lid4 + 3];
2564 s_sbob_sl64[2][lid4 + 0] = sbob_sl64[2][lid4 + 0];
2565 s_sbob_sl64[2][lid4 + 1] = sbob_sl64[2][lid4 + 1];
2566 s_sbob_sl64[2][lid4 + 2] = sbob_sl64[2][lid4 + 2];
2567 s_sbob_sl64[2][lid4 + 3] = sbob_sl64[2][lid4 + 3];
2568 s_sbob_sl64[3][lid4 + 0] = sbob_sl64[3][lid4 + 0];
2569 s_sbob_sl64[3][lid4 + 1] = sbob_sl64[3][lid4 + 1];
2570 s_sbob_sl64[3][lid4 + 2] = sbob_sl64[3][lid4 + 2];
2571 s_sbob_sl64[3][lid4 + 3] = sbob_sl64[3][lid4 + 3];
2572 s_sbob_sl64[4][lid4 + 0] = sbob_sl64[4][lid4 + 0];
2573 s_sbob_sl64[4][lid4 + 1] = sbob_sl64[4][lid4 + 1];
2574 s_sbob_sl64[4][lid4 + 2] = sbob_sl64[4][lid4 + 2];
2575 s_sbob_sl64[4][lid4 + 3] = sbob_sl64[4][lid4 + 3];
2576 s_sbob_sl64[5][lid4 + 0] = sbob_sl64[5][lid4 + 0];
2577 s_sbob_sl64[5][lid4 + 1] = sbob_sl64[5][lid4 + 1];
2578 s_sbob_sl64[5][lid4 + 2] = sbob_sl64[5][lid4 + 2];
2579 s_sbob_sl64[5][lid4 + 3] = sbob_sl64[5][lid4 + 3];
2580 s_sbob_sl64[6][lid4 + 0] = sbob_sl64[6][lid4 + 0];
2581 s_sbob_sl64[6][lid4 + 1] = sbob_sl64[6][lid4 + 1];
2582 s_sbob_sl64[6][lid4 + 2] = sbob_sl64[6][lid4 + 2];
2583 s_sbob_sl64[6][lid4 + 3] = sbob_sl64[6][lid4 + 3];
2584 s_sbob_sl64[7][lid4 + 0] = sbob_sl64[7][lid4 + 0];
2585 s_sbob_sl64[7][lid4 + 1] = sbob_sl64[7][lid4 + 1];
2586 s_sbob_sl64[7][lid4 + 2] = sbob_sl64[7][lid4 + 2];
2587 s_sbob_sl64[7][lid4 + 3] = sbob_sl64[7][lid4 + 3];
2589 barrier (CLK_LOCAL_MEM_FENCE);
2595 const u32 search[4] =
2597 digests_buf[digests_offset].digest_buf[DGST_R0],
2598 digests_buf[digests_offset].digest_buf[DGST_R1],
2599 digests_buf[digests_offset].digest_buf[DGST_R2],
2600 digests_buf[digests_offset].digest_buf[DGST_R3]
2607 const u32 gid = get_global_id (0);
2609 if (gid >= gid_max) return;
2613 wordl0[0] = pws[gid].i[ 0];
2614 wordl0[1] = pws[gid].i[ 1];
2615 wordl0[2] = pws[gid].i[ 2];
2616 wordl0[3] = pws[gid].i[ 3];
2620 wordl1[0] = pws[gid].i[ 4];
2621 wordl1[1] = pws[gid].i[ 5];
2622 wordl1[2] = pws[gid].i[ 6];
2623 wordl1[3] = pws[gid].i[ 7];
2639 const u32 pw_l_len = pws[gid].pw_len;
2641 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
2643 append_0x80_2 (wordl0, wordl1, pw_l_len);
2645 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
2652 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
2654 const u32 pw_r_len = combs_buf[il_pos].pw_len;
2656 const u32 pw_len = pw_l_len + pw_r_len;
2660 wordr0[0] = combs_buf[il_pos].i[0];
2661 wordr0[1] = combs_buf[il_pos].i[1];
2662 wordr0[2] = combs_buf[il_pos].i[2];
2663 wordr0[3] = combs_buf[il_pos].i[3];
2667 wordr1[0] = combs_buf[il_pos].i[4];
2668 wordr1[1] = combs_buf[il_pos].i[5];
2669 wordr1[2] = combs_buf[il_pos].i[6];
2670 wordr1[3] = combs_buf[il_pos].i[7];
2686 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
2688 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
2693 w[ 0] = wordl0[0] | wordr0[0];
2694 w[ 1] = wordl0[1] | wordr0[1];
2695 w[ 2] = wordl0[2] | wordr0[2];
2696 w[ 3] = wordl0[3] | wordr0[3];
2697 w[ 4] = wordl1[0] | wordr1[0];
2698 w[ 5] = wordl1[1] | wordr1[1];
2699 w[ 6] = wordl1[2] | wordr1[2];
2700 w[ 7] = wordl1[3] | wordr1[3];
2701 w[ 8] = wordl2[0] | wordr2[0];
2702 w[ 9] = wordl2[1] | wordr2[1];
2703 w[10] = wordl2[2] | wordr2[2];
2704 w[11] = wordl2[3] | wordr2[3];
2705 w[12] = wordl3[0] | wordr3[0];
2706 w[13] = wordl3[1] | wordr3[1];
2707 w[14] = wordl3[1] | wordr3[1];
2708 w[15] = wordl3[1] | wordr3[1];
2710 append_0x01_4 (&w[0], &w[1], &w[2], &w[3], pw_len);
2713 * reverse message block
2718 m[0] = hl32_to_64 (w[15], w[14]);
2719 m[1] = hl32_to_64 (w[13], w[12]);
2720 m[2] = hl32_to_64 (w[11], w[10]);
2721 m[3] = hl32_to_64 (w[ 9], w[ 8]);
2722 m[4] = hl32_to_64 (w[ 7], w[ 6]);
2723 m[5] = hl32_to_64 (w[ 5], w[ 4]);
2724 m[6] = hl32_to_64 (w[ 3], w[ 2]);
2725 m[7] = hl32_to_64 (w[ 1], w[ 0]);
2727 m[0] = swap_workaround (m[0]);
2728 m[1] = swap_workaround (m[1]);
2729 m[2] = swap_workaround (m[2]);
2730 m[3] = swap_workaround (m[3]);
2731 m[4] = swap_workaround (m[4]);
2732 m[5] = swap_workaround (m[5]);
2733 m[6] = swap_workaround (m[6]);
2734 m[7] = swap_workaround (m[7]);
2736 // state buffer (hash)
2749 streebog_g (h, m, s_sbob_sl64);
2760 z[7] = swap_workaround ((u64) (pw_len * 8));
2762 streebog_g (h, z, s_sbob_sl64);
2763 streebog_g (h, m, s_sbob_sl64);
2765 const u32 r0 = l32_from_64 (h[0]);
2766 const u32 r1 = h32_from_64 (h[0]);
2767 const u32 r2 = l32_from_64 (h[1]);
2768 const u32 r3 = h32_from_64 (h[1]);
2770 #include VECT_COMPARE_S
2774 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11700_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)
2778 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11700_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)