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 __constant u64 k_sha512[80] =
49 SHA512C00, SHA512C01, SHA512C02, SHA512C03,
50 SHA512C04, SHA512C05, SHA512C06, SHA512C07,
51 SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
52 SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
53 SHA512C10, SHA512C11, SHA512C12, SHA512C13,
54 SHA512C14, SHA512C15, SHA512C16, SHA512C17,
55 SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
56 SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
57 SHA512C20, SHA512C21, SHA512C22, SHA512C23,
58 SHA512C24, SHA512C25, SHA512C26, SHA512C27,
59 SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
60 SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
61 SHA512C30, SHA512C31, SHA512C32, SHA512C33,
62 SHA512C34, SHA512C35, SHA512C36, SHA512C37,
63 SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
64 SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
65 SHA512C40, SHA512C41, SHA512C42, SHA512C43,
66 SHA512C44, SHA512C45, SHA512C46, SHA512C47,
67 SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
68 SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
71 static void sha512_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u64x digest[8])
73 u64x w0_t = hl32_to_64 (w0[0], w0[1]);
74 u64x w1_t = hl32_to_64 (w0[2], w0[3]);
75 u64x w2_t = hl32_to_64 (w1[0], w1[1]);
76 u64x w3_t = hl32_to_64 (w1[2], w1[3]);
77 u64x w4_t = hl32_to_64 (w2[0], w2[1]);
78 u64x w5_t = hl32_to_64 (w2[2], w2[3]);
79 u64x w6_t = hl32_to_64 (w3[0], w3[1]);
88 u64x wf_t = hl32_to_64 (w3[2], w3[3]);
99 #define ROUND_EXPAND() \
101 w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
102 w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
103 w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
104 w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
105 w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
106 w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
107 w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
108 w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
109 w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
110 w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
111 wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
112 wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
113 wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
114 wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
115 we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
116 wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
119 #define ROUND_STEP(i) \
121 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
122 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
123 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
124 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
125 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
126 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
127 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
128 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
129 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
130 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
131 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
132 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
133 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
134 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
135 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
136 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
142 for (int i = 16; i < 80; i += 16)
144 ROUND_EXPAND (); ROUND_STEP (i);
168 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01740_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)
174 const u32 lid = get_local_id (0);
180 const u32 gid = get_global_id (0);
182 if (gid >= gid_max) return;
186 wordl0[0] = pws[gid].i[ 0];
187 wordl0[1] = pws[gid].i[ 1];
188 wordl0[2] = pws[gid].i[ 2];
189 wordl0[3] = pws[gid].i[ 3];
193 wordl1[0] = pws[gid].i[ 4];
194 wordl1[1] = pws[gid].i[ 5];
195 wordl1[2] = pws[gid].i[ 6];
196 wordl1[3] = pws[gid].i[ 7];
212 const u32 pw_l_len = pws[gid].pw_len;
214 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
216 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
225 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
226 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
227 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
228 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[3];
232 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[4];
233 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[5];
234 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[6];
235 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[7];
237 const u32 salt_len = salt_bufs[salt_pos].salt_len;
243 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
245 const u32 pw_r_len = combs_buf[il_pos].pw_len;
247 const u32 pw_len = pw_l_len + pw_r_len;
254 wordr0[0] = combs_buf[il_pos].i[0];
255 wordr0[1] = combs_buf[il_pos].i[1];
256 wordr0[2] = combs_buf[il_pos].i[2];
257 wordr0[3] = combs_buf[il_pos].i[3];
258 wordr1[0] = combs_buf[il_pos].i[4];
259 wordr1[1] = combs_buf[il_pos].i[5];
260 wordr1[2] = combs_buf[il_pos].i[6];
261 wordr1[3] = combs_buf[il_pos].i[7];
271 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
273 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
281 w0[0] = wordl0[0] | wordr0[0];
282 w0[1] = wordl0[1] | wordr0[1];
283 w0[2] = wordl0[2] | wordr0[2];
284 w0[3] = wordl0[3] | wordr0[3];
285 w1[0] = wordl1[0] | wordr1[0];
286 w1[1] = wordl1[1] | wordr1[1];
287 w1[2] = wordl1[2] | wordr1[2];
288 w1[3] = wordl1[3] | wordr1[3];
289 w2[0] = wordl2[0] | wordr2[0];
290 w2[1] = wordl2[1] | wordr2[1];
291 w2[2] = wordl2[2] | wordr2[2];
292 w2[3] = wordl2[3] | wordr2[3];
293 w3[0] = wordl3[0] | wordr3[0];
294 w3[1] = wordl3[1] | wordr3[1];
295 w3[2] = wordl3[2] | wordr3[2];
296 w3[3] = wordl3[3] | wordr3[3];
302 const u32 pw_salt_len = (pw_len * 2) + salt_len;
309 make_unicode (w0, w0_t, w1_t);
310 make_unicode (w1, w2_t, w3_t);
312 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
314 w0_t[0] |= salt_buf0[0];
315 w0_t[1] |= salt_buf0[1];
316 w0_t[2] |= salt_buf0[2];
317 w0_t[3] |= salt_buf0[3];
318 w1_t[0] |= salt_buf1[0];
319 w1_t[1] |= salt_buf1[1];
320 w1_t[2] |= salt_buf1[2];
321 w1_t[3] |= salt_buf1[3];
323 append_0x80_4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
329 w0_t[0] = swap_workaround (w0_t[0]);
330 w0_t[1] = swap_workaround (w0_t[1]);
331 w0_t[2] = swap_workaround (w0_t[2]);
332 w0_t[3] = swap_workaround (w0_t[3]);
333 w1_t[0] = swap_workaround (w1_t[0]);
334 w1_t[1] = swap_workaround (w1_t[1]);
335 w1_t[2] = swap_workaround (w1_t[2]);
336 w1_t[3] = swap_workaround (w1_t[3]);
337 w2_t[0] = swap_workaround (w2_t[0]);
338 w2_t[1] = swap_workaround (w2_t[1]);
339 w2_t[2] = swap_workaround (w2_t[2]);
340 w2_t[3] = swap_workaround (w2_t[3]);
341 w3_t[0] = swap_workaround (w3_t[0]);
342 w3_t[1] = swap_workaround (w3_t[1]);
344 w3_t[3] = pw_salt_len * 8;
348 digest[0] = SHA512M_A;
349 digest[1] = SHA512M_B;
350 digest[2] = SHA512M_C;
351 digest[3] = SHA512M_D;
352 digest[4] = SHA512M_E;
353 digest[5] = SHA512M_F;
354 digest[6] = SHA512M_G;
355 digest[7] = SHA512M_H;
357 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
360 const u32x r0 = l32_from_64 (digest[7]);
361 const u32x r1 = h32_from_64 (digest[7]);
362 const u32x r2 = l32_from_64 (digest[3]);
363 const u32x r3 = h32_from_64 (digest[3]);
365 #include VECT_COMPARE_M
369 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01740_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)
373 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01740_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)
377 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01740_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)
383 const u32 lid = get_local_id (0);
389 const u32 gid = get_global_id (0);
391 if (gid >= gid_max) return;
395 wordl0[0] = pws[gid].i[ 0];
396 wordl0[1] = pws[gid].i[ 1];
397 wordl0[2] = pws[gid].i[ 2];
398 wordl0[3] = pws[gid].i[ 3];
402 wordl1[0] = pws[gid].i[ 4];
403 wordl1[1] = pws[gid].i[ 5];
404 wordl1[2] = pws[gid].i[ 6];
405 wordl1[3] = pws[gid].i[ 7];
421 const u32 pw_l_len = pws[gid].pw_len;
423 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
425 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
434 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
435 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
436 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
437 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[3];
441 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[4];
442 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[5];
443 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[6];
444 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[7];
446 const u32 salt_len = salt_bufs[salt_pos].salt_len;
452 const u32 search[4] =
454 digests_buf[digests_offset].digest_buf[DGST_R0],
455 digests_buf[digests_offset].digest_buf[DGST_R1],
456 digests_buf[digests_offset].digest_buf[DGST_R2],
457 digests_buf[digests_offset].digest_buf[DGST_R3]
464 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
466 const u32 pw_r_len = combs_buf[il_pos].pw_len;
468 const u32 pw_len = pw_l_len + pw_r_len;
475 wordr0[0] = combs_buf[il_pos].i[0];
476 wordr0[1] = combs_buf[il_pos].i[1];
477 wordr0[2] = combs_buf[il_pos].i[2];
478 wordr0[3] = combs_buf[il_pos].i[3];
479 wordr1[0] = combs_buf[il_pos].i[4];
480 wordr1[1] = combs_buf[il_pos].i[5];
481 wordr1[2] = combs_buf[il_pos].i[6];
482 wordr1[3] = combs_buf[il_pos].i[7];
492 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
494 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
502 w0[0] = wordl0[0] | wordr0[0];
503 w0[1] = wordl0[1] | wordr0[1];
504 w0[2] = wordl0[2] | wordr0[2];
505 w0[3] = wordl0[3] | wordr0[3];
506 w1[0] = wordl1[0] | wordr1[0];
507 w1[1] = wordl1[1] | wordr1[1];
508 w1[2] = wordl1[2] | wordr1[2];
509 w1[3] = wordl1[3] | wordr1[3];
510 w2[0] = wordl2[0] | wordr2[0];
511 w2[1] = wordl2[1] | wordr2[1];
512 w2[2] = wordl2[2] | wordr2[2];
513 w2[3] = wordl2[3] | wordr2[3];
514 w3[0] = wordl3[0] | wordr3[0];
515 w3[1] = wordl3[1] | wordr3[1];
516 w3[2] = wordl3[2] | wordr3[2];
517 w3[3] = wordl3[3] | wordr3[3];
523 const u32 pw_salt_len = (pw_len * 2) + salt_len;
530 make_unicode (w0, w0_t, w1_t);
531 make_unicode (w1, w2_t, w3_t);
533 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
535 w0_t[0] |= salt_buf0[0];
536 w0_t[1] |= salt_buf0[1];
537 w0_t[2] |= salt_buf0[2];
538 w0_t[3] |= salt_buf0[3];
539 w1_t[0] |= salt_buf1[0];
540 w1_t[1] |= salt_buf1[1];
541 w1_t[2] |= salt_buf1[2];
542 w1_t[3] |= salt_buf1[3];
544 append_0x80_4 (w0_t, w1_t, w2_t, w3_t, pw_salt_len);
550 w0_t[0] = swap_workaround (w0_t[0]);
551 w0_t[1] = swap_workaround (w0_t[1]);
552 w0_t[2] = swap_workaround (w0_t[2]);
553 w0_t[3] = swap_workaround (w0_t[3]);
554 w1_t[0] = swap_workaround (w1_t[0]);
555 w1_t[1] = swap_workaround (w1_t[1]);
556 w1_t[2] = swap_workaround (w1_t[2]);
557 w1_t[3] = swap_workaround (w1_t[3]);
558 w2_t[0] = swap_workaround (w2_t[0]);
559 w2_t[1] = swap_workaround (w2_t[1]);
560 w2_t[2] = swap_workaround (w2_t[2]);
561 w2_t[3] = swap_workaround (w2_t[3]);
562 w3_t[0] = swap_workaround (w3_t[0]);
563 w3_t[1] = swap_workaround (w3_t[1]);
565 w3_t[3] = pw_salt_len * 8;
569 digest[0] = SHA512M_A;
570 digest[1] = SHA512M_B;
571 digest[2] = SHA512M_C;
572 digest[3] = SHA512M_D;
573 digest[4] = SHA512M_E;
574 digest[5] = SHA512M_F;
575 digest[6] = SHA512M_G;
576 digest[7] = SHA512M_H;
578 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
581 const u32x r0 = l32_from_64 (digest[7]);
582 const u32x r1 = h32_from_64 (digest[7]);
583 const u32x r2 = l32_from_64 (digest[3]);
584 const u32x r3 = h32_from_64 (digest[3]);
586 #include VECT_COMPARE_S
590 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01740_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)
594 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01740_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)