2 * Author......: Jens Steube <jens.steube@gmail.com>
6 #define _PBKDF2_SHA512_
10 #include "inc_vendor.cl"
11 #include "inc_hash_constants.h"
12 #include "inc_hash_functions.cl"
13 #include "inc_types.cl"
14 #include "inc_common.cl"
15 #include "inc_simd.cl"
17 #define COMPARE_S "inc_comp_single.cl"
18 #define COMPARE_M "inc_comp_multi.cl"
20 __constant u64 k_sha512[80] =
22 SHA512C00, SHA512C01, SHA512C02, SHA512C03,
23 SHA512C04, SHA512C05, SHA512C06, SHA512C07,
24 SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
25 SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
26 SHA512C10, SHA512C11, SHA512C12, SHA512C13,
27 SHA512C14, SHA512C15, SHA512C16, SHA512C17,
28 SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
29 SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
30 SHA512C20, SHA512C21, SHA512C22, SHA512C23,
31 SHA512C24, SHA512C25, SHA512C26, SHA512C27,
32 SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
33 SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
34 SHA512C30, SHA512C31, SHA512C32, SHA512C33,
35 SHA512C34, SHA512C35, SHA512C36, SHA512C37,
36 SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
37 SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
38 SHA512C40, SHA512C41, SHA512C42, SHA512C43,
39 SHA512C44, SHA512C45, SHA512C46, SHA512C47,
40 SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
41 SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
44 void sha512_transform_S (const u64 w[16], u64 dgst[8])
72 #define ROUND_EXPAND_S() \
74 w0_t = SHA512_EXPAND_S (we_t, w9_t, w1_t, w0_t); \
75 w1_t = SHA512_EXPAND_S (wf_t, wa_t, w2_t, w1_t); \
76 w2_t = SHA512_EXPAND_S (w0_t, wb_t, w3_t, w2_t); \
77 w3_t = SHA512_EXPAND_S (w1_t, wc_t, w4_t, w3_t); \
78 w4_t = SHA512_EXPAND_S (w2_t, wd_t, w5_t, w4_t); \
79 w5_t = SHA512_EXPAND_S (w3_t, we_t, w6_t, w5_t); \
80 w6_t = SHA512_EXPAND_S (w4_t, wf_t, w7_t, w6_t); \
81 w7_t = SHA512_EXPAND_S (w5_t, w0_t, w8_t, w7_t); \
82 w8_t = SHA512_EXPAND_S (w6_t, w1_t, w9_t, w8_t); \
83 w9_t = SHA512_EXPAND_S (w7_t, w2_t, wa_t, w9_t); \
84 wa_t = SHA512_EXPAND_S (w8_t, w3_t, wb_t, wa_t); \
85 wb_t = SHA512_EXPAND_S (w9_t, w4_t, wc_t, wb_t); \
86 wc_t = SHA512_EXPAND_S (wa_t, w5_t, wd_t, wc_t); \
87 wd_t = SHA512_EXPAND_S (wb_t, w6_t, we_t, wd_t); \
88 we_t = SHA512_EXPAND_S (wc_t, w7_t, wf_t, we_t); \
89 wf_t = SHA512_EXPAND_S (wd_t, w8_t, w0_t, wf_t); \
92 #define ROUND_STEP_S(i) \
94 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
95 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
96 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
97 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
98 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
99 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
100 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
101 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
102 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
103 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
104 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
105 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
106 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
107 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
108 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
109 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
117 for (int i = 16; i < 80; i += 16)
119 ROUND_EXPAND_S (); ROUND_STEP_S (i);
132 void hmac_sha512_run_S (const u64 w1[16], const u64 ipad[8], const u64 opad[8], u64 dgst[8])
143 sha512_transform_S (w1, dgst);
155 w[ 8] = 0x8000000000000000;
162 w[15] = (128 + 64) * 8;
173 sha512_transform_S (w, dgst);
176 void hmac_sha512_init_S (u64 w[16], u64 ipad[8], u64 opad[8])
178 w[ 0] ^= 0x3636363636363636;
179 w[ 1] ^= 0x3636363636363636;
180 w[ 2] ^= 0x3636363636363636;
181 w[ 3] ^= 0x3636363636363636;
182 w[ 4] ^= 0x3636363636363636;
183 w[ 5] ^= 0x3636363636363636;
184 w[ 6] ^= 0x3636363636363636;
185 w[ 7] ^= 0x3636363636363636;
186 w[ 8] ^= 0x3636363636363636;
187 w[ 9] ^= 0x3636363636363636;
188 w[10] ^= 0x3636363636363636;
189 w[11] ^= 0x3636363636363636;
190 w[12] ^= 0x3636363636363636;
191 w[13] ^= 0x3636363636363636;
192 w[14] ^= 0x3636363636363636;
193 w[15] ^= 0x3636363636363636;
204 sha512_transform_S (w, ipad);
206 w[ 0] ^= 0x6a6a6a6a6a6a6a6a;
207 w[ 1] ^= 0x6a6a6a6a6a6a6a6a;
208 w[ 2] ^= 0x6a6a6a6a6a6a6a6a;
209 w[ 3] ^= 0x6a6a6a6a6a6a6a6a;
210 w[ 4] ^= 0x6a6a6a6a6a6a6a6a;
211 w[ 5] ^= 0x6a6a6a6a6a6a6a6a;
212 w[ 6] ^= 0x6a6a6a6a6a6a6a6a;
213 w[ 7] ^= 0x6a6a6a6a6a6a6a6a;
214 w[ 8] ^= 0x6a6a6a6a6a6a6a6a;
215 w[ 9] ^= 0x6a6a6a6a6a6a6a6a;
216 w[10] ^= 0x6a6a6a6a6a6a6a6a;
217 w[11] ^= 0x6a6a6a6a6a6a6a6a;
218 w[12] ^= 0x6a6a6a6a6a6a6a6a;
219 w[13] ^= 0x6a6a6a6a6a6a6a6a;
220 w[14] ^= 0x6a6a6a6a6a6a6a6a;
221 w[15] ^= 0x6a6a6a6a6a6a6a6a;
232 sha512_transform_S (w, opad);
235 void sha512_transform_V (const u64x w[16], u64x dgst[8])
263 #define ROUND_EXPAND() \
265 w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
266 w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
267 w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
268 w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
269 w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
270 w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
271 w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
272 w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
273 w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
274 w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
275 wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
276 wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
277 wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
278 wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
279 we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
280 wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
283 #define ROUND_STEP(i) \
285 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
286 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
287 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
288 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
289 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
290 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
291 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
292 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
293 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
294 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
295 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
296 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
297 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
298 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
299 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
300 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
308 for (int i = 16; i < 80; i += 16)
310 ROUND_EXPAND (); ROUND_STEP (i);
323 void hmac_sha512_run_V (const u64x w1[16], const u64x ipad[8], const u64x opad[8], u64x dgst[8])
334 sha512_transform_V (w1, dgst);
346 w[ 8] = 0x8000000000000000;
353 w[15] = (128 + 64) * 8;
364 sha512_transform_V (w, dgst);
367 void hmac_sha512_run_V_x (const u64x ipad[8], const u64x opad[8], u64x dgst[8])
379 w[ 8] = 0x8000000000000000;
386 w[15] = (128 + 64) * 8;
397 sha512_transform_V (w, dgst);
417 sha512_transform_V (w, dgst);
420 void hmac_sha512_init_V (u64x w[16], u64x ipad[8], u64x opad[8])
422 w[ 0] ^= 0x3636363636363636;
423 w[ 1] ^= 0x3636363636363636;
424 w[ 2] ^= 0x3636363636363636;
425 w[ 3] ^= 0x3636363636363636;
426 w[ 4] ^= 0x3636363636363636;
427 w[ 5] ^= 0x3636363636363636;
428 w[ 6] ^= 0x3636363636363636;
429 w[ 7] ^= 0x3636363636363636;
430 w[ 8] ^= 0x3636363636363636;
431 w[ 9] ^= 0x3636363636363636;
432 w[10] ^= 0x3636363636363636;
433 w[11] ^= 0x3636363636363636;
434 w[12] ^= 0x3636363636363636;
435 w[13] ^= 0x3636363636363636;
436 w[14] ^= 0x3636363636363636;
437 w[15] ^= 0x3636363636363636;
448 sha512_transform_V (w, ipad);
450 w[ 0] ^= 0x6a6a6a6a6a6a6a6a;
451 w[ 1] ^= 0x6a6a6a6a6a6a6a6a;
452 w[ 2] ^= 0x6a6a6a6a6a6a6a6a;
453 w[ 3] ^= 0x6a6a6a6a6a6a6a6a;
454 w[ 4] ^= 0x6a6a6a6a6a6a6a6a;
455 w[ 5] ^= 0x6a6a6a6a6a6a6a6a;
456 w[ 6] ^= 0x6a6a6a6a6a6a6a6a;
457 w[ 7] ^= 0x6a6a6a6a6a6a6a6a;
458 w[ 8] ^= 0x6a6a6a6a6a6a6a6a;
459 w[ 9] ^= 0x6a6a6a6a6a6a6a6a;
460 w[10] ^= 0x6a6a6a6a6a6a6a6a;
461 w[11] ^= 0x6a6a6a6a6a6a6a6a;
462 w[12] ^= 0x6a6a6a6a6a6a6a6a;
463 w[13] ^= 0x6a6a6a6a6a6a6a6a;
464 w[14] ^= 0x6a6a6a6a6a6a6a6a;
465 w[15] ^= 0x6a6a6a6a6a6a6a6a;
476 sha512_transform_V (w, opad);
479 __kernel void m07100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global pbkdf2_sha512_tmp_t *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 pbkdf2_sha512_t *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
485 const u32 gid = get_global_id (0);
487 if (gid >= gid_max) return;
491 w0[0] = swap32_S (pws[gid].i[ 0]);
492 w0[1] = swap32_S (pws[gid].i[ 1]);
493 w0[2] = swap32_S (pws[gid].i[ 2]);
494 w0[3] = swap32_S (pws[gid].i[ 3]);
498 w1[0] = swap32_S (pws[gid].i[ 4]);
499 w1[1] = swap32_S (pws[gid].i[ 5]);
500 w1[2] = swap32_S (pws[gid].i[ 6]);
501 w1[3] = swap32_S (pws[gid].i[ 7]);
505 w2[0] = swap32_S (pws[gid].i[ 8]);
506 w2[1] = swap32_S (pws[gid].i[ 9]);
507 w2[2] = swap32_S (pws[gid].i[10]);
508 w2[3] = swap32_S (pws[gid].i[11]);
512 w3[0] = swap32_S (pws[gid].i[12]);
513 w3[1] = swap32_S (pws[gid].i[13]);
514 w3[2] = swap32_S (pws[gid].i[14]);
515 w3[3] = swap32_S (pws[gid].i[15]);
523 u32 salt_len = salt_bufs[salt_pos].salt_len;
525 esalt_buf[ 0] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[ 0]), swap32_S (esalt_bufs[salt_pos].salt_buf[ 1]));
526 esalt_buf[ 1] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[ 2]), swap32_S (esalt_bufs[salt_pos].salt_buf[ 3]));
527 esalt_buf[ 2] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[ 4]), swap32_S (esalt_bufs[salt_pos].salt_buf[ 5]));
528 esalt_buf[ 3] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[ 6]), swap32_S (esalt_bufs[salt_pos].salt_buf[ 7]));
529 esalt_buf[ 4] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[ 8]), swap32_S (esalt_bufs[salt_pos].salt_buf[ 9]));
530 esalt_buf[ 5] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[10]), swap32_S (esalt_bufs[salt_pos].salt_buf[11]));
531 esalt_buf[ 6] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[12]), swap32_S (esalt_bufs[salt_pos].salt_buf[13]));
532 esalt_buf[ 7] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[14]), swap32_S (esalt_bufs[salt_pos].salt_buf[15]));
533 esalt_buf[ 8] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[16]), swap32_S (esalt_bufs[salt_pos].salt_buf[17]));
534 esalt_buf[ 9] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[18]), swap32_S (esalt_bufs[salt_pos].salt_buf[19]));
535 esalt_buf[10] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[20]), swap32_S (esalt_bufs[salt_pos].salt_buf[21]));
536 esalt_buf[11] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[22]), swap32_S (esalt_bufs[salt_pos].salt_buf[23]));
537 esalt_buf[12] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[24]), swap32_S (esalt_bufs[salt_pos].salt_buf[25]));
538 esalt_buf[13] = hl32_to_64_S (swap32_S (esalt_bufs[salt_pos].salt_buf[26]), swap32_S (esalt_bufs[salt_pos].salt_buf[27]));
540 esalt_buf[15] = (128 + salt_len + 4) * 8;
544 w[ 0] = hl32_to_64_S (w0[0], w0[1]);
545 w[ 1] = hl32_to_64_S (w0[2], w0[3]);
546 w[ 2] = hl32_to_64_S (w1[0], w1[1]);
547 w[ 3] = hl32_to_64_S (w1[2], w1[3]);
548 w[ 4] = hl32_to_64_S (w2[0], w2[1]);
549 w[ 5] = hl32_to_64_S (w2[2], w2[3]);
550 w[ 6] = hl32_to_64_S (w3[0], w3[1]);
551 w[ 7] = hl32_to_64_S (w3[2], w3[3]);
564 hmac_sha512_init_S (w, ipad, opad);
566 tmps[gid].ipad[0] = ipad[0];
567 tmps[gid].ipad[1] = ipad[1];
568 tmps[gid].ipad[2] = ipad[2];
569 tmps[gid].ipad[3] = ipad[3];
570 tmps[gid].ipad[4] = ipad[4];
571 tmps[gid].ipad[5] = ipad[5];
572 tmps[gid].ipad[6] = ipad[6];
573 tmps[gid].ipad[7] = ipad[7];
575 tmps[gid].opad[0] = opad[0];
576 tmps[gid].opad[1] = opad[1];
577 tmps[gid].opad[2] = opad[2];
578 tmps[gid].opad[3] = opad[3];
579 tmps[gid].opad[4] = opad[4];
580 tmps[gid].opad[5] = opad[5];
581 tmps[gid].opad[6] = opad[6];
582 tmps[gid].opad[7] = opad[7];
584 for (u32 i = 0, j = 1; i < 8; i += 8, j += 1)
588 hmac_sha512_run_S (esalt_buf, ipad, opad, dgst);
590 tmps[gid].dgst[i + 0] = dgst[0];
591 tmps[gid].dgst[i + 1] = dgst[1];
592 tmps[gid].dgst[i + 2] = dgst[2];
593 tmps[gid].dgst[i + 3] = dgst[3];
594 tmps[gid].dgst[i + 4] = dgst[4];
595 tmps[gid].dgst[i + 5] = dgst[5];
596 tmps[gid].dgst[i + 6] = dgst[6];
597 tmps[gid].dgst[i + 7] = dgst[7];
599 tmps[gid].out[i + 0] = dgst[0];
600 tmps[gid].out[i + 1] = dgst[1];
601 tmps[gid].out[i + 2] = dgst[2];
602 tmps[gid].out[i + 3] = dgst[3];
603 tmps[gid].out[i + 4] = dgst[4];
604 tmps[gid].out[i + 5] = dgst[5];
605 tmps[gid].out[i + 6] = dgst[6];
606 tmps[gid].out[i + 7] = dgst[7];
610 __kernel void m07100_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global pbkdf2_sha512_tmp_t *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 pbkdf2_sha512_t *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
612 const u32 gid = get_global_id (0);
614 if (gid >= gid_max) return;
619 ipad[0] = pack64v (tmps, ipad, gid, 0);
620 ipad[1] = pack64v (tmps, ipad, gid, 1);
621 ipad[2] = pack64v (tmps, ipad, gid, 2);
622 ipad[3] = pack64v (tmps, ipad, gid, 3);
623 ipad[4] = pack64v (tmps, ipad, gid, 4);
624 ipad[5] = pack64v (tmps, ipad, gid, 5);
625 ipad[6] = pack64v (tmps, ipad, gid, 6);
626 ipad[7] = pack64v (tmps, ipad, gid, 7);
628 opad[0] = pack64v (tmps, opad, gid, 0);
629 opad[1] = pack64v (tmps, opad, gid, 1);
630 opad[2] = pack64v (tmps, opad, gid, 2);
631 opad[3] = pack64v (tmps, opad, gid, 3);
632 opad[4] = pack64v (tmps, opad, gid, 4);
633 opad[5] = pack64v (tmps, opad, gid, 5);
634 opad[6] = pack64v (tmps, opad, gid, 6);
635 opad[7] = pack64v (tmps, opad, gid, 7);
637 for (u32 i = 0; i < 8; i += 8)
641 dgst[0] = pack64v (tmps, dgst, gid, 0);
642 dgst[1] = pack64v (tmps, dgst, gid, 1);
643 dgst[2] = pack64v (tmps, dgst, gid, 2);
644 dgst[3] = pack64v (tmps, dgst, gid, 3);
645 dgst[4] = pack64v (tmps, dgst, gid, 4);
646 dgst[5] = pack64v (tmps, dgst, gid, 5);
647 dgst[6] = pack64v (tmps, dgst, gid, 6);
648 dgst[7] = pack64v (tmps, dgst, gid, 7);
650 for (u32 j = 0; j < loop_cnt; j++)
652 hmac_sha512_run_V_x (ipad, opad, dgst);
654 unpackv_xor (tmps, out, gid, 0, dgst[0]);
655 unpackv_xor (tmps, out, gid, 1, dgst[1]);
656 unpackv_xor (tmps, out, gid, 2, dgst[2]);
657 unpackv_xor (tmps, out, gid, 3, dgst[3]);
658 unpackv_xor (tmps, out, gid, 4, dgst[4]);
659 unpackv_xor (tmps, out, gid, 5, dgst[5]);
660 unpackv_xor (tmps, out, gid, 6, dgst[6]);
661 unpackv_xor (tmps, out, gid, 7, dgst[7]);
664 unpackv (tmps, dgst, gid, 0, dgst[0]);
665 unpackv (tmps, dgst, gid, 1, dgst[1]);
666 unpackv (tmps, dgst, gid, 2, dgst[2]);
667 unpackv (tmps, dgst, gid, 3, dgst[3]);
668 unpackv (tmps, dgst, gid, 4, dgst[4]);
669 unpackv (tmps, dgst, gid, 5, dgst[5]);
670 unpackv (tmps, dgst, gid, 6, dgst[6]);
671 unpackv (tmps, dgst, gid, 7, dgst[7]);
675 __kernel void m07100_comp (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global pbkdf2_sha512_tmp_t *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 pbkdf2_sha512_t *esalt_bufs, __global u32 *d_return_buf, __global u32 *d_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
681 const u32 gid = get_global_id (0);
683 if (gid >= gid_max) return;
685 const u32 lid = get_local_id (0);
687 const u64 a = tmps[gid].out[0];
688 const u64 b = tmps[gid].out[1];
690 const u32 r0 = l32_from_64_S (a);
691 const u32 r1 = h32_from_64_S (a);
692 const u32 r2 = l32_from_64_S (b);
693 const u32 r3 = h32_from_64_S (b);