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"
50 #define SBOG_LPSti64 \
51 s_sbob_sl64[0][(t[0] >> (i * 8)) & 0xff] ^ \
52 s_sbob_sl64[1][(t[1] >> (i * 8)) & 0xff] ^ \
53 s_sbob_sl64[2][(t[2] >> (i * 8)) & 0xff] ^ \
54 s_sbob_sl64[3][(t[3] >> (i * 8)) & 0xff] ^ \
55 s_sbob_sl64[4][(t[4] >> (i * 8)) & 0xff] ^ \
56 s_sbob_sl64[5][(t[5] >> (i * 8)) & 0xff] ^ \
57 s_sbob_sl64[6][(t[6] >> (i * 8)) & 0xff] ^ \
58 s_sbob_sl64[7][(t[7] >> (i * 8)) & 0xff]
62 __constant u64 sbob_sl64[8][256] =
2130 __constant u64 sbob_rc64[12][8] =
2254 static void streebog_g (u64 h[8], const u64 m[8], __local u64 s_sbob_sl64[8][256])
2261 for (int i = 0; i < 8; i++)
2266 for (int i = 0; i < 8; i++)
2268 k[i] = SBOG_LPSti64;
2272 for (int i = 0; i < 8; i++)
2277 for (int r = 0; r < 12; r++)
2280 for (int i = 0; i < 8; i++)
2286 for (int i = 0; i < 8; i++)
2288 s[i] = SBOG_LPSti64;
2291 for (int i = 0; i < 8; i++)
2293 t[i] = k[i] ^ sbob_rc64[r][i];
2297 for (int i = 0; i < 8; i++)
2299 k[i] = SBOG_LPSti64;
2304 for (int i = 0; i < 8; i++)
2306 h[i] ^= s[i] ^ k[i] ^ m[i];
2310 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11800_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)
2316 const u32 lid = get_local_id (0);
2319 * shared lookup table
2322 const u32 lid4 = lid * 4;
2324 __local u64 s_sbob_sl64[8][256];
2326 s_sbob_sl64[0][lid4 + 0] = sbob_sl64[0][lid4 + 0];
2327 s_sbob_sl64[0][lid4 + 1] = sbob_sl64[0][lid4 + 1];
2328 s_sbob_sl64[0][lid4 + 2] = sbob_sl64[0][lid4 + 2];
2329 s_sbob_sl64[0][lid4 + 3] = sbob_sl64[0][lid4 + 3];
2330 s_sbob_sl64[1][lid4 + 0] = sbob_sl64[1][lid4 + 0];
2331 s_sbob_sl64[1][lid4 + 1] = sbob_sl64[1][lid4 + 1];
2332 s_sbob_sl64[1][lid4 + 2] = sbob_sl64[1][lid4 + 2];
2333 s_sbob_sl64[1][lid4 + 3] = sbob_sl64[1][lid4 + 3];
2334 s_sbob_sl64[2][lid4 + 0] = sbob_sl64[2][lid4 + 0];
2335 s_sbob_sl64[2][lid4 + 1] = sbob_sl64[2][lid4 + 1];
2336 s_sbob_sl64[2][lid4 + 2] = sbob_sl64[2][lid4 + 2];
2337 s_sbob_sl64[2][lid4 + 3] = sbob_sl64[2][lid4 + 3];
2338 s_sbob_sl64[3][lid4 + 0] = sbob_sl64[3][lid4 + 0];
2339 s_sbob_sl64[3][lid4 + 1] = sbob_sl64[3][lid4 + 1];
2340 s_sbob_sl64[3][lid4 + 2] = sbob_sl64[3][lid4 + 2];
2341 s_sbob_sl64[3][lid4 + 3] = sbob_sl64[3][lid4 + 3];
2342 s_sbob_sl64[4][lid4 + 0] = sbob_sl64[4][lid4 + 0];
2343 s_sbob_sl64[4][lid4 + 1] = sbob_sl64[4][lid4 + 1];
2344 s_sbob_sl64[4][lid4 + 2] = sbob_sl64[4][lid4 + 2];
2345 s_sbob_sl64[4][lid4 + 3] = sbob_sl64[4][lid4 + 3];
2346 s_sbob_sl64[5][lid4 + 0] = sbob_sl64[5][lid4 + 0];
2347 s_sbob_sl64[5][lid4 + 1] = sbob_sl64[5][lid4 + 1];
2348 s_sbob_sl64[5][lid4 + 2] = sbob_sl64[5][lid4 + 2];
2349 s_sbob_sl64[5][lid4 + 3] = sbob_sl64[5][lid4 + 3];
2350 s_sbob_sl64[6][lid4 + 0] = sbob_sl64[6][lid4 + 0];
2351 s_sbob_sl64[6][lid4 + 1] = sbob_sl64[6][lid4 + 1];
2352 s_sbob_sl64[6][lid4 + 2] = sbob_sl64[6][lid4 + 2];
2353 s_sbob_sl64[6][lid4 + 3] = sbob_sl64[6][lid4 + 3];
2354 s_sbob_sl64[7][lid4 + 0] = sbob_sl64[7][lid4 + 0];
2355 s_sbob_sl64[7][lid4 + 1] = sbob_sl64[7][lid4 + 1];
2356 s_sbob_sl64[7][lid4 + 2] = sbob_sl64[7][lid4 + 2];
2357 s_sbob_sl64[7][lid4 + 3] = sbob_sl64[7][lid4 + 3];
2359 barrier (CLK_LOCAL_MEM_FENCE);
2365 const u32 gid = get_global_id (0);
2367 if (gid >= gid_max) return;
2371 wordl0[0] = pws[gid].i[ 0];
2372 wordl0[1] = pws[gid].i[ 1];
2373 wordl0[2] = pws[gid].i[ 2];
2374 wordl0[3] = pws[gid].i[ 3];
2378 wordl1[0] = pws[gid].i[ 4];
2379 wordl1[1] = pws[gid].i[ 5];
2380 wordl1[2] = pws[gid].i[ 6];
2381 wordl1[3] = pws[gid].i[ 7];
2397 const u32 pw_l_len = pws[gid].pw_len;
2399 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
2401 append_0x80_2 (wordl0, wordl1, pw_l_len);
2403 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
2410 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
2412 const u32 pw_r_len = combs_buf[il_pos].pw_len;
2414 const u32 pw_len = pw_l_len + pw_r_len;
2418 wordr0[0] = combs_buf[il_pos].i[0];
2419 wordr0[1] = combs_buf[il_pos].i[1];
2420 wordr0[2] = combs_buf[il_pos].i[2];
2421 wordr0[3] = combs_buf[il_pos].i[3];
2425 wordr1[0] = combs_buf[il_pos].i[4];
2426 wordr1[1] = combs_buf[il_pos].i[5];
2427 wordr1[2] = combs_buf[il_pos].i[6];
2428 wordr1[3] = combs_buf[il_pos].i[7];
2444 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
2446 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
2451 w[ 0] = wordl0[0] | wordr0[0];
2452 w[ 1] = wordl0[1] | wordr0[1];
2453 w[ 2] = wordl0[2] | wordr0[2];
2454 w[ 3] = wordl0[3] | wordr0[3];
2455 w[ 4] = wordl1[0] | wordr1[0];
2456 w[ 5] = wordl1[1] | wordr1[1];
2457 w[ 6] = wordl1[2] | wordr1[2];
2458 w[ 7] = wordl1[3] | wordr1[3];
2459 w[ 8] = wordl2[0] | wordr2[0];
2460 w[ 9] = wordl2[1] | wordr2[1];
2461 w[10] = wordl2[2] | wordr2[2];
2462 w[11] = wordl2[3] | wordr2[3];
2463 w[12] = wordl3[0] | wordr3[0];
2464 w[13] = wordl3[1] | wordr3[1];
2465 w[14] = wordl3[1] | wordr3[1];
2466 w[15] = wordl3[1] | wordr3[1];
2468 append_0x01_4 (&w[0], &w[1], &w[2], &w[3], pw_len);
2471 * reverse message block
2476 m[0] = hl32_to_64 (w[15], w[14]);
2477 m[1] = hl32_to_64 (w[13], w[12]);
2478 m[2] = hl32_to_64 (w[11], w[10]);
2479 m[3] = hl32_to_64 (w[ 9], w[ 8]);
2480 m[4] = hl32_to_64 (w[ 7], w[ 6]);
2481 m[5] = hl32_to_64 (w[ 5], w[ 4]);
2482 m[6] = hl32_to_64 (w[ 3], w[ 2]);
2483 m[7] = hl32_to_64 (w[ 1], w[ 0]);
2485 m[0] = swap_workaround (m[0]);
2486 m[1] = swap_workaround (m[1]);
2487 m[2] = swap_workaround (m[2]);
2488 m[3] = swap_workaround (m[3]);
2489 m[4] = swap_workaround (m[4]);
2490 m[5] = swap_workaround (m[5]);
2491 m[6] = swap_workaround (m[6]);
2492 m[7] = swap_workaround (m[7]);
2494 // state buffer (hash)
2507 streebog_g (h, m, s_sbob_sl64);
2518 z[7] = swap_workaround ((u64) (pw_len * 8));
2520 streebog_g (h, z, s_sbob_sl64);
2521 streebog_g (h, m, s_sbob_sl64);
2523 const u32 r0 = l32_from_64 (h[0]);
2524 const u32 r1 = h32_from_64 (h[0]);
2525 const u32 r2 = l32_from_64 (h[1]);
2526 const u32 r3 = h32_from_64 (h[1]);
2528 #include VECT_COMPARE_M
2532 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11800_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)
2536 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11800_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)
2540 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11800_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)
2546 const u32 lid = get_local_id (0);
2550 * shared lookup table
2553 const u32 lid4 = lid * 4;
2555 __local u64 s_sbob_sl64[8][256];
2557 s_sbob_sl64[0][lid4 + 0] = sbob_sl64[0][lid4 + 0];
2558 s_sbob_sl64[0][lid4 + 1] = sbob_sl64[0][lid4 + 1];
2559 s_sbob_sl64[0][lid4 + 2] = sbob_sl64[0][lid4 + 2];
2560 s_sbob_sl64[0][lid4 + 3] = sbob_sl64[0][lid4 + 3];
2561 s_sbob_sl64[1][lid4 + 0] = sbob_sl64[1][lid4 + 0];
2562 s_sbob_sl64[1][lid4 + 1] = sbob_sl64[1][lid4 + 1];
2563 s_sbob_sl64[1][lid4 + 2] = sbob_sl64[1][lid4 + 2];
2564 s_sbob_sl64[1][lid4 + 3] = sbob_sl64[1][lid4 + 3];
2565 s_sbob_sl64[2][lid4 + 0] = sbob_sl64[2][lid4 + 0];
2566 s_sbob_sl64[2][lid4 + 1] = sbob_sl64[2][lid4 + 1];
2567 s_sbob_sl64[2][lid4 + 2] = sbob_sl64[2][lid4 + 2];
2568 s_sbob_sl64[2][lid4 + 3] = sbob_sl64[2][lid4 + 3];
2569 s_sbob_sl64[3][lid4 + 0] = sbob_sl64[3][lid4 + 0];
2570 s_sbob_sl64[3][lid4 + 1] = sbob_sl64[3][lid4 + 1];
2571 s_sbob_sl64[3][lid4 + 2] = sbob_sl64[3][lid4 + 2];
2572 s_sbob_sl64[3][lid4 + 3] = sbob_sl64[3][lid4 + 3];
2573 s_sbob_sl64[4][lid4 + 0] = sbob_sl64[4][lid4 + 0];
2574 s_sbob_sl64[4][lid4 + 1] = sbob_sl64[4][lid4 + 1];
2575 s_sbob_sl64[4][lid4 + 2] = sbob_sl64[4][lid4 + 2];
2576 s_sbob_sl64[4][lid4 + 3] = sbob_sl64[4][lid4 + 3];
2577 s_sbob_sl64[5][lid4 + 0] = sbob_sl64[5][lid4 + 0];
2578 s_sbob_sl64[5][lid4 + 1] = sbob_sl64[5][lid4 + 1];
2579 s_sbob_sl64[5][lid4 + 2] = sbob_sl64[5][lid4 + 2];
2580 s_sbob_sl64[5][lid4 + 3] = sbob_sl64[5][lid4 + 3];
2581 s_sbob_sl64[6][lid4 + 0] = sbob_sl64[6][lid4 + 0];
2582 s_sbob_sl64[6][lid4 + 1] = sbob_sl64[6][lid4 + 1];
2583 s_sbob_sl64[6][lid4 + 2] = sbob_sl64[6][lid4 + 2];
2584 s_sbob_sl64[6][lid4 + 3] = sbob_sl64[6][lid4 + 3];
2585 s_sbob_sl64[7][lid4 + 0] = sbob_sl64[7][lid4 + 0];
2586 s_sbob_sl64[7][lid4 + 1] = sbob_sl64[7][lid4 + 1];
2587 s_sbob_sl64[7][lid4 + 2] = sbob_sl64[7][lid4 + 2];
2588 s_sbob_sl64[7][lid4 + 3] = sbob_sl64[7][lid4 + 3];
2590 barrier (CLK_LOCAL_MEM_FENCE);
2596 const u32 search[4] =
2598 digests_buf[digests_offset].digest_buf[DGST_R0],
2599 digests_buf[digests_offset].digest_buf[DGST_R1],
2600 digests_buf[digests_offset].digest_buf[DGST_R2],
2601 digests_buf[digests_offset].digest_buf[DGST_R3]
2608 const u32 gid = get_global_id (0);
2610 if (gid >= gid_max) return;
2614 wordl0[0] = pws[gid].i[ 0];
2615 wordl0[1] = pws[gid].i[ 1];
2616 wordl0[2] = pws[gid].i[ 2];
2617 wordl0[3] = pws[gid].i[ 3];
2621 wordl1[0] = pws[gid].i[ 4];
2622 wordl1[1] = pws[gid].i[ 5];
2623 wordl1[2] = pws[gid].i[ 6];
2624 wordl1[3] = pws[gid].i[ 7];
2640 const u32 pw_l_len = pws[gid].pw_len;
2642 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
2644 append_0x80_2 (wordl0, wordl1, pw_l_len);
2646 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
2653 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
2655 const u32 pw_r_len = combs_buf[il_pos].pw_len;
2657 const u32 pw_len = pw_l_len + pw_r_len;
2661 wordr0[0] = combs_buf[il_pos].i[0];
2662 wordr0[1] = combs_buf[il_pos].i[1];
2663 wordr0[2] = combs_buf[il_pos].i[2];
2664 wordr0[3] = combs_buf[il_pos].i[3];
2668 wordr1[0] = combs_buf[il_pos].i[4];
2669 wordr1[1] = combs_buf[il_pos].i[5];
2670 wordr1[2] = combs_buf[il_pos].i[6];
2671 wordr1[3] = combs_buf[il_pos].i[7];
2687 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
2689 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
2694 w[ 0] = wordl0[0] | wordr0[0];
2695 w[ 1] = wordl0[1] | wordr0[1];
2696 w[ 2] = wordl0[2] | wordr0[2];
2697 w[ 3] = wordl0[3] | wordr0[3];
2698 w[ 4] = wordl1[0] | wordr1[0];
2699 w[ 5] = wordl1[1] | wordr1[1];
2700 w[ 6] = wordl1[2] | wordr1[2];
2701 w[ 7] = wordl1[3] | wordr1[3];
2702 w[ 8] = wordl2[0] | wordr2[0];
2703 w[ 9] = wordl2[1] | wordr2[1];
2704 w[10] = wordl2[2] | wordr2[2];
2705 w[11] = wordl2[3] | wordr2[3];
2706 w[12] = wordl3[0] | wordr3[0];
2707 w[13] = wordl3[1] | wordr3[1];
2708 w[14] = wordl3[1] | wordr3[1];
2709 w[15] = wordl3[1] | wordr3[1];
2711 append_0x01_4 (&w[0], &w[1], &w[2], &w[3], pw_len);
2714 * reverse message block
2719 m[0] = hl32_to_64 (w[15], w[14]);
2720 m[1] = hl32_to_64 (w[13], w[12]);
2721 m[2] = hl32_to_64 (w[11], w[10]);
2722 m[3] = hl32_to_64 (w[ 9], w[ 8]);
2723 m[4] = hl32_to_64 (w[ 7], w[ 6]);
2724 m[5] = hl32_to_64 (w[ 5], w[ 4]);
2725 m[6] = hl32_to_64 (w[ 3], w[ 2]);
2726 m[7] = hl32_to_64 (w[ 1], w[ 0]);
2728 m[0] = swap_workaround (m[0]);
2729 m[1] = swap_workaround (m[1]);
2730 m[2] = swap_workaround (m[2]);
2731 m[3] = swap_workaround (m[3]);
2732 m[4] = swap_workaround (m[4]);
2733 m[5] = swap_workaround (m[5]);
2734 m[6] = swap_workaround (m[6]);
2735 m[7] = swap_workaround (m[7]);
2737 // state buffer (hash)
2750 streebog_g (h, m, s_sbob_sl64);
2761 z[7] = swap_workaround ((u64) (pw_len * 8));
2763 streebog_g (h, z, s_sbob_sl64);
2764 streebog_g (h, m, s_sbob_sl64);
2766 const u32 r0 = l32_from_64 (h[0]);
2767 const u32 r1 = h32_from_64 (h[0]);
2768 const u32 r2 = l32_from_64 (h[1]);
2769 const u32 r3 = h32_from_64 (h[1]);
2771 #include VECT_COMPARE_S
2775 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11800_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)
2779 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m11800_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)