2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
16 #include "include/kernel_functions.c"
17 #include "types_ocl.c"
20 #define COMPARE_S "check_single_comp4.c"
21 #define COMPARE_M "check_multi_comp4.c"
23 __constant u64 k_sha512[80] =
25 SHA512C00, SHA512C01, SHA512C02, SHA512C03,
26 SHA512C04, SHA512C05, SHA512C06, SHA512C07,
27 SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
28 SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
29 SHA512C10, SHA512C11, SHA512C12, SHA512C13,
30 SHA512C14, SHA512C15, SHA512C16, SHA512C17,
31 SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
32 SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
33 SHA512C20, SHA512C21, SHA512C22, SHA512C23,
34 SHA512C24, SHA512C25, SHA512C26, SHA512C27,
35 SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
36 SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
37 SHA512C30, SHA512C31, SHA512C32, SHA512C33,
38 SHA512C34, SHA512C35, SHA512C36, SHA512C37,
39 SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
40 SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
41 SHA512C40, SHA512C41, SHA512C42, SHA512C43,
42 SHA512C44, SHA512C45, SHA512C46, SHA512C47,
43 SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
44 SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
47 static void sha512_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u64 digest[8])
49 u64 w0_t = hl32_to_64 (w0[0], w0[1]);
50 u64 w1_t = hl32_to_64 (w0[2], w0[3]);
51 u64 w2_t = hl32_to_64 (w1[0], w1[1]);
52 u64 w3_t = hl32_to_64 (w1[2], w1[3]);
53 u64 w4_t = hl32_to_64 (w2[0], w2[1]);
54 u64 w5_t = hl32_to_64 (w2[2], w2[3]);
55 u64 w6_t = hl32_to_64 (w3[0], w3[1]);
64 u64 wf_t = hl32_to_64 (w3[2], w3[3]);
75 #define ROUND_EXPAND() \
77 w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
78 w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
79 w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
80 w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
81 w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
82 w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
83 w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
84 w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
85 w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
86 w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
87 wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
88 wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
89 wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
90 wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
91 we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
92 wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
95 #define ROUND_STEP(i) \
97 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
98 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
99 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
100 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
101 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
102 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
103 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
104 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
105 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
106 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
107 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
108 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
109 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
110 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
111 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
112 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
118 for (int i = 16; i < 80; i += 16)
120 ROUND_EXPAND (); ROUND_STEP (i);
144 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01720_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)
150 const u32 lid = get_local_id (0);
156 const u32 gid = get_global_id (0);
158 if (gid >= gid_max) return;
162 wordl0[0] = pws[gid].i[ 0];
163 wordl0[1] = pws[gid].i[ 1];
164 wordl0[2] = pws[gid].i[ 2];
165 wordl0[3] = pws[gid].i[ 3];
169 wordl1[0] = pws[gid].i[ 4];
170 wordl1[1] = pws[gid].i[ 5];
171 wordl1[2] = pws[gid].i[ 6];
172 wordl1[3] = pws[gid].i[ 7];
188 const u32 pw_l_len = pws[gid].pw_len;
190 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
192 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
201 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
202 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
203 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
204 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[3];
208 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[4];
209 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[5];
210 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[6];
211 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[7];
213 const u32 salt_len = salt_bufs[salt_pos].salt_len;
219 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
221 const u32 pw_r_len = combs_buf[il_pos].pw_len;
223 const u32 pw_len = pw_l_len + pw_r_len;
230 wordr0[0] = combs_buf[il_pos].i[0];
231 wordr0[1] = combs_buf[il_pos].i[1];
232 wordr0[2] = combs_buf[il_pos].i[2];
233 wordr0[3] = combs_buf[il_pos].i[3];
234 wordr1[0] = combs_buf[il_pos].i[4];
235 wordr1[1] = combs_buf[il_pos].i[5];
236 wordr1[2] = combs_buf[il_pos].i[6];
237 wordr1[3] = combs_buf[il_pos].i[7];
247 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
249 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
257 w0[0] = wordl0[0] | wordr0[0];
258 w0[1] = wordl0[1] | wordr0[1];
259 w0[2] = wordl0[2] | wordr0[2];
260 w0[3] = wordl0[3] | wordr0[3];
261 w1[0] = wordl1[0] | wordr1[0];
262 w1[1] = wordl1[1] | wordr1[1];
263 w1[2] = wordl1[2] | wordr1[2];
264 w1[3] = wordl1[3] | wordr1[3];
265 w2[0] = wordl2[0] | wordr2[0];
266 w2[1] = wordl2[1] | wordr2[1];
267 w2[2] = wordl2[2] | wordr2[2];
268 w2[3] = wordl2[3] | wordr2[3];
269 w3[0] = wordl3[0] | wordr3[0];
270 w3[1] = wordl3[1] | wordr3[1];
271 w3[2] = wordl3[2] | wordr3[2];
272 w3[3] = wordl3[3] | wordr3[3];
278 const u32 pw_salt_len = pw_len + salt_len;
280 switch_buffer_by_offset (w0, w1, w2, w3, salt_len);
282 w0[0] |= salt_buf0[0];
283 w0[1] |= salt_buf0[1];
284 w0[2] |= salt_buf0[2];
285 w0[3] |= salt_buf0[3];
286 w1[0] |= salt_buf1[0];
287 w1[1] |= salt_buf1[1];
288 w1[2] |= salt_buf1[2];
289 w1[3] |= salt_buf1[3];
291 append_0x80_4 (w0, w1, w2, w3, pw_salt_len);
302 w0_t[0] = swap_workaround (w0[0]);
303 w0_t[1] = swap_workaround (w0[1]);
304 w0_t[2] = swap_workaround (w0[2]);
305 w0_t[3] = swap_workaround (w0[3]);
306 w1_t[0] = swap_workaround (w1[0]);
307 w1_t[1] = swap_workaround (w1[1]);
308 w1_t[2] = swap_workaround (w1[2]);
309 w1_t[3] = swap_workaround (w1[3]);
310 w2_t[0] = swap_workaround (w2[0]);
311 w2_t[1] = swap_workaround (w2[1]);
312 w2_t[2] = swap_workaround (w2[2]);
313 w2_t[3] = swap_workaround (w2[3]);
314 w3_t[0] = swap_workaround (w3[0]);
315 w3_t[1] = swap_workaround (w3[1]);
317 w3_t[3] = pw_salt_len * 8;
321 digest[0] = SHA512M_A;
322 digest[1] = SHA512M_B;
323 digest[2] = SHA512M_C;
324 digest[3] = SHA512M_D;
325 digest[4] = SHA512M_E;
326 digest[5] = SHA512M_F;
327 digest[6] = SHA512M_G;
328 digest[7] = SHA512M_H;
330 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
333 const u32 r0 = l32_from_64 (digest[7]);
334 const u32 r1 = h32_from_64 (digest[7]);
335 const u32 r2 = l32_from_64 (digest[3]);
336 const u32 r3 = h32_from_64 (digest[3]);
342 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01720_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)
346 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01720_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)
350 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01720_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)
356 const u32 lid = get_local_id (0);
362 const u32 gid = get_global_id (0);
364 if (gid >= gid_max) return;
368 wordl0[0] = pws[gid].i[ 0];
369 wordl0[1] = pws[gid].i[ 1];
370 wordl0[2] = pws[gid].i[ 2];
371 wordl0[3] = pws[gid].i[ 3];
375 wordl1[0] = pws[gid].i[ 4];
376 wordl1[1] = pws[gid].i[ 5];
377 wordl1[2] = pws[gid].i[ 6];
378 wordl1[3] = pws[gid].i[ 7];
394 const u32 pw_l_len = pws[gid].pw_len;
396 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
398 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
407 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
408 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
409 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
410 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[3];
414 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[4];
415 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[5];
416 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[6];
417 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[7];
419 const u32 salt_len = salt_bufs[salt_pos].salt_len;
425 const u32 search[4] =
427 digests_buf[digests_offset].digest_buf[DGST_R0],
428 digests_buf[digests_offset].digest_buf[DGST_R1],
429 digests_buf[digests_offset].digest_buf[DGST_R2],
430 digests_buf[digests_offset].digest_buf[DGST_R3]
437 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
439 const u32 pw_r_len = combs_buf[il_pos].pw_len;
441 const u32 pw_len = pw_l_len + pw_r_len;
448 wordr0[0] = combs_buf[il_pos].i[0];
449 wordr0[1] = combs_buf[il_pos].i[1];
450 wordr0[2] = combs_buf[il_pos].i[2];
451 wordr0[3] = combs_buf[il_pos].i[3];
452 wordr1[0] = combs_buf[il_pos].i[4];
453 wordr1[1] = combs_buf[il_pos].i[5];
454 wordr1[2] = combs_buf[il_pos].i[6];
455 wordr1[3] = combs_buf[il_pos].i[7];
465 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
467 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
475 w0[0] = wordl0[0] | wordr0[0];
476 w0[1] = wordl0[1] | wordr0[1];
477 w0[2] = wordl0[2] | wordr0[2];
478 w0[3] = wordl0[3] | wordr0[3];
479 w1[0] = wordl1[0] | wordr1[0];
480 w1[1] = wordl1[1] | wordr1[1];
481 w1[2] = wordl1[2] | wordr1[2];
482 w1[3] = wordl1[3] | wordr1[3];
483 w2[0] = wordl2[0] | wordr2[0];
484 w2[1] = wordl2[1] | wordr2[1];
485 w2[2] = wordl2[2] | wordr2[2];
486 w2[3] = wordl2[3] | wordr2[3];
487 w3[0] = wordl3[0] | wordr3[0];
488 w3[1] = wordl3[1] | wordr3[1];
489 w3[2] = wordl3[2] | wordr3[2];
490 w3[3] = wordl3[3] | wordr3[3];
496 const u32 pw_salt_len = pw_len + salt_len;
498 switch_buffer_by_offset (w0, w1, w2, w3, salt_len);
500 w0[0] |= salt_buf0[0];
501 w0[1] |= salt_buf0[1];
502 w0[2] |= salt_buf0[2];
503 w0[3] |= salt_buf0[3];
504 w1[0] |= salt_buf1[0];
505 w1[1] |= salt_buf1[1];
506 w1[2] |= salt_buf1[2];
507 w1[3] |= salt_buf1[3];
509 append_0x80_4 (w0, w1, w2, w3, pw_salt_len);
520 w0_t[0] = swap_workaround (w0[0]);
521 w0_t[1] = swap_workaround (w0[1]);
522 w0_t[2] = swap_workaround (w0[2]);
523 w0_t[3] = swap_workaround (w0[3]);
524 w1_t[0] = swap_workaround (w1[0]);
525 w1_t[1] = swap_workaround (w1[1]);
526 w1_t[2] = swap_workaround (w1[2]);
527 w1_t[3] = swap_workaround (w1[3]);
528 w2_t[0] = swap_workaround (w2[0]);
529 w2_t[1] = swap_workaround (w2[1]);
530 w2_t[2] = swap_workaround (w2[2]);
531 w2_t[3] = swap_workaround (w2[3]);
532 w3_t[0] = swap_workaround (w3[0]);
533 w3_t[1] = swap_workaround (w3[1]);
535 w3_t[3] = pw_salt_len * 8;
539 digest[0] = SHA512M_A;
540 digest[1] = SHA512M_B;
541 digest[2] = SHA512M_C;
542 digest[3] = SHA512M_D;
543 digest[4] = SHA512M_E;
544 digest[5] = SHA512M_F;
545 digest[6] = SHA512M_G;
546 digest[7] = SHA512M_H;
548 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
551 const u32 r0 = l32_from_64 (digest[7]);
552 const u32 r1 = h32_from_64 (digest[7]);
553 const u32 r2 = l32_from_64 (digest[3]);
554 const u32 r3 = h32_from_64 (digest[3]);
560 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01720_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)
564 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01720_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)