2 * Author......: Jens Steube <jens.steube@gmail.com>
6 #define _PBKDF2_SHA512_
10 #include "inc_hash_constants.h"
11 #include "inc_vendor.cl"
18 #include "inc_hash_functions.cl"
19 #include "inc_types.cl"
20 #include "inc_common.cl"
21 #include "inc_simd.cl"
23 #define COMPARE_S "inc_comp_single.cl"
24 #define COMPARE_M "inc_comp_multi.cl"
26 __constant u64 k_sha512[80] =
28 SHA512C00, SHA512C01, SHA512C02, SHA512C03,
29 SHA512C04, SHA512C05, SHA512C06, SHA512C07,
30 SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
31 SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
32 SHA512C10, SHA512C11, SHA512C12, SHA512C13,
33 SHA512C14, SHA512C15, SHA512C16, SHA512C17,
34 SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
35 SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
36 SHA512C20, SHA512C21, SHA512C22, SHA512C23,
37 SHA512C24, SHA512C25, SHA512C26, SHA512C27,
38 SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
39 SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
40 SHA512C30, SHA512C31, SHA512C32, SHA512C33,
41 SHA512C34, SHA512C35, SHA512C36, SHA512C37,
42 SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
43 SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
44 SHA512C40, SHA512C41, SHA512C42, SHA512C43,
45 SHA512C44, SHA512C45, SHA512C46, SHA512C47,
46 SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
47 SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
50 void sha512_transform_S (const u64 w[16], u64 dgst[8])
78 #define ROUND_EXPAND_S() \
80 w0_t = SHA512_EXPAND_S (we_t, w9_t, w1_t, w0_t); \
81 w1_t = SHA512_EXPAND_S (wf_t, wa_t, w2_t, w1_t); \
82 w2_t = SHA512_EXPAND_S (w0_t, wb_t, w3_t, w2_t); \
83 w3_t = SHA512_EXPAND_S (w1_t, wc_t, w4_t, w3_t); \
84 w4_t = SHA512_EXPAND_S (w2_t, wd_t, w5_t, w4_t); \
85 w5_t = SHA512_EXPAND_S (w3_t, we_t, w6_t, w5_t); \
86 w6_t = SHA512_EXPAND_S (w4_t, wf_t, w7_t, w6_t); \
87 w7_t = SHA512_EXPAND_S (w5_t, w0_t, w8_t, w7_t); \
88 w8_t = SHA512_EXPAND_S (w6_t, w1_t, w9_t, w8_t); \
89 w9_t = SHA512_EXPAND_S (w7_t, w2_t, wa_t, w9_t); \
90 wa_t = SHA512_EXPAND_S (w8_t, w3_t, wb_t, wa_t); \
91 wb_t = SHA512_EXPAND_S (w9_t, w4_t, wc_t, wb_t); \
92 wc_t = SHA512_EXPAND_S (wa_t, w5_t, wd_t, wc_t); \
93 wd_t = SHA512_EXPAND_S (wb_t, w6_t, we_t, wd_t); \
94 we_t = SHA512_EXPAND_S (wc_t, w7_t, wf_t, we_t); \
95 wf_t = SHA512_EXPAND_S (wd_t, w8_t, w0_t, wf_t); \
98 #define ROUND_STEP_S(i) \
100 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
101 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
102 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
103 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
104 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
105 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
106 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
107 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
108 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
109 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
110 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
111 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
112 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
113 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
114 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
115 SHA512_STEP_S (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
123 for (int i = 16; i < 80; i += 16)
125 ROUND_EXPAND_S (); ROUND_STEP_S (i);
138 void hmac_sha512_run_S (const u64 w1[16], const u64 ipad[8], const u64 opad[8], u64 dgst[8])
149 sha512_transform_S (w1, dgst);
161 w[ 8] = 0x8000000000000000;
168 w[15] = (128 + 64) * 8;
179 sha512_transform_S (w, dgst);
182 void hmac_sha512_init_S (u64 w[16], u64 ipad[8], u64 opad[8])
184 w[ 0] ^= 0x3636363636363636;
185 w[ 1] ^= 0x3636363636363636;
186 w[ 2] ^= 0x3636363636363636;
187 w[ 3] ^= 0x3636363636363636;
188 w[ 4] ^= 0x3636363636363636;
189 w[ 5] ^= 0x3636363636363636;
190 w[ 6] ^= 0x3636363636363636;
191 w[ 7] ^= 0x3636363636363636;
192 w[ 8] ^= 0x3636363636363636;
193 w[ 9] ^= 0x3636363636363636;
194 w[10] ^= 0x3636363636363636;
195 w[11] ^= 0x3636363636363636;
196 w[12] ^= 0x3636363636363636;
197 w[13] ^= 0x3636363636363636;
198 w[14] ^= 0x3636363636363636;
199 w[15] ^= 0x3636363636363636;
210 sha512_transform_S (w, ipad);
212 w[ 0] ^= 0x6a6a6a6a6a6a6a6a;
213 w[ 1] ^= 0x6a6a6a6a6a6a6a6a;
214 w[ 2] ^= 0x6a6a6a6a6a6a6a6a;
215 w[ 3] ^= 0x6a6a6a6a6a6a6a6a;
216 w[ 4] ^= 0x6a6a6a6a6a6a6a6a;
217 w[ 5] ^= 0x6a6a6a6a6a6a6a6a;
218 w[ 6] ^= 0x6a6a6a6a6a6a6a6a;
219 w[ 7] ^= 0x6a6a6a6a6a6a6a6a;
220 w[ 8] ^= 0x6a6a6a6a6a6a6a6a;
221 w[ 9] ^= 0x6a6a6a6a6a6a6a6a;
222 w[10] ^= 0x6a6a6a6a6a6a6a6a;
223 w[11] ^= 0x6a6a6a6a6a6a6a6a;
224 w[12] ^= 0x6a6a6a6a6a6a6a6a;
225 w[13] ^= 0x6a6a6a6a6a6a6a6a;
226 w[14] ^= 0x6a6a6a6a6a6a6a6a;
227 w[15] ^= 0x6a6a6a6a6a6a6a6a;
238 sha512_transform_S (w, opad);
241 void sha512_transform_V (const u64x w[16], u64x dgst[8])
269 #define ROUND_EXPAND() \
271 w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
272 w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
273 w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
274 w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
275 w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
276 w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
277 w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
278 w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
279 w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
280 w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
281 wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
282 wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
283 wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
284 wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
285 we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
286 wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
289 #define ROUND_STEP(i) \
291 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
292 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
293 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
294 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
295 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
296 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
297 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
298 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
299 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
300 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
301 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
302 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
303 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
304 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
305 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
306 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
314 for (int i = 16; i < 80; i += 16)
316 ROUND_EXPAND (); ROUND_STEP (i);
329 void hmac_sha512_run_V (const u64x w1[16], const u64x ipad[8], const u64x opad[8], u64x dgst[8])
340 sha512_transform_V (w1, dgst);
352 w[ 8] = 0x8000000000000000;
359 w[15] = (128 + 64) * 8;
370 sha512_transform_V (w, dgst);
373 void hmac_sha512_run_V_x (const u64x ipad[8], const u64x opad[8], u64x dgst[8])
385 w[ 8] = 0x8000000000000000;
392 w[15] = (128 + 64) * 8;
403 sha512_transform_V (w, dgst);
423 sha512_transform_V (w, dgst);
426 void hmac_sha512_init_V (u64x w[16], u64x ipad[8], u64x opad[8])
428 w[ 0] ^= 0x3636363636363636;
429 w[ 1] ^= 0x3636363636363636;
430 w[ 2] ^= 0x3636363636363636;
431 w[ 3] ^= 0x3636363636363636;
432 w[ 4] ^= 0x3636363636363636;
433 w[ 5] ^= 0x3636363636363636;
434 w[ 6] ^= 0x3636363636363636;
435 w[ 7] ^= 0x3636363636363636;
436 w[ 8] ^= 0x3636363636363636;
437 w[ 9] ^= 0x3636363636363636;
438 w[10] ^= 0x3636363636363636;
439 w[11] ^= 0x3636363636363636;
440 w[12] ^= 0x3636363636363636;
441 w[13] ^= 0x3636363636363636;
442 w[14] ^= 0x3636363636363636;
443 w[15] ^= 0x3636363636363636;
454 sha512_transform_V (w, ipad);
456 w[ 0] ^= 0x6a6a6a6a6a6a6a6a;
457 w[ 1] ^= 0x6a6a6a6a6a6a6a6a;
458 w[ 2] ^= 0x6a6a6a6a6a6a6a6a;
459 w[ 3] ^= 0x6a6a6a6a6a6a6a6a;
460 w[ 4] ^= 0x6a6a6a6a6a6a6a6a;
461 w[ 5] ^= 0x6a6a6a6a6a6a6a6a;
462 w[ 6] ^= 0x6a6a6a6a6a6a6a6a;
463 w[ 7] ^= 0x6a6a6a6a6a6a6a6a;
464 w[ 8] ^= 0x6a6a6a6a6a6a6a6a;
465 w[ 9] ^= 0x6a6a6a6a6a6a6a6a;
466 w[10] ^= 0x6a6a6a6a6a6a6a6a;
467 w[11] ^= 0x6a6a6a6a6a6a6a6a;
468 w[12] ^= 0x6a6a6a6a6a6a6a6a;
469 w[13] ^= 0x6a6a6a6a6a6a6a6a;
470 w[14] ^= 0x6a6a6a6a6a6a6a6a;
471 w[15] ^= 0x6a6a6a6a6a6a6a6a;
482 sha512_transform_V (w, opad);
485 __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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
491 const u32 gid = get_global_id (0);
493 if (gid >= gid_max) return;
497 w0[0] = swap32_S (pws[gid].i[ 0]);
498 w0[1] = swap32_S (pws[gid].i[ 1]);
499 w0[2] = swap32_S (pws[gid].i[ 2]);
500 w0[3] = swap32_S (pws[gid].i[ 3]);
504 w1[0] = swap32_S (pws[gid].i[ 4]);
505 w1[1] = swap32_S (pws[gid].i[ 5]);
506 w1[2] = swap32_S (pws[gid].i[ 6]);
507 w1[3] = swap32_S (pws[gid].i[ 7]);
511 w2[0] = swap32_S (pws[gid].i[ 8]);
512 w2[1] = swap32_S (pws[gid].i[ 9]);
513 w2[2] = swap32_S (pws[gid].i[10]);
514 w2[3] = swap32_S (pws[gid].i[11]);
518 w3[0] = swap32_S (pws[gid].i[12]);
519 w3[1] = swap32_S (pws[gid].i[13]);
520 w3[2] = swap32_S (pws[gid].i[14]);
521 w3[3] = swap32_S (pws[gid].i[15]);
529 u32 salt_len = salt_bufs[salt_pos].salt_len;
531 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]));
532 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]));
533 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]));
534 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]));
535 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]));
536 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]));
537 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]));
538 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]));
539 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]));
540 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]));
541 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]));
542 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]));
543 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]));
544 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]));
546 esalt_buf[15] = (128 + salt_len + 4) * 8;
550 w[ 0] = hl32_to_64_S (w0[0], w0[1]);
551 w[ 1] = hl32_to_64_S (w0[2], w0[3]);
552 w[ 2] = hl32_to_64_S (w1[0], w1[1]);
553 w[ 3] = hl32_to_64_S (w1[2], w1[3]);
554 w[ 4] = hl32_to_64_S (w2[0], w2[1]);
555 w[ 5] = hl32_to_64_S (w2[2], w2[3]);
556 w[ 6] = hl32_to_64_S (w3[0], w3[1]);
557 w[ 7] = hl32_to_64_S (w3[2], w3[3]);
570 hmac_sha512_init_S (w, ipad, opad);
572 tmps[gid].ipad[0] = ipad[0];
573 tmps[gid].ipad[1] = ipad[1];
574 tmps[gid].ipad[2] = ipad[2];
575 tmps[gid].ipad[3] = ipad[3];
576 tmps[gid].ipad[4] = ipad[4];
577 tmps[gid].ipad[5] = ipad[5];
578 tmps[gid].ipad[6] = ipad[6];
579 tmps[gid].ipad[7] = ipad[7];
581 tmps[gid].opad[0] = opad[0];
582 tmps[gid].opad[1] = opad[1];
583 tmps[gid].opad[2] = opad[2];
584 tmps[gid].opad[3] = opad[3];
585 tmps[gid].opad[4] = opad[4];
586 tmps[gid].opad[5] = opad[5];
587 tmps[gid].opad[6] = opad[6];
588 tmps[gid].opad[7] = opad[7];
590 for (u32 i = 0, j = 1; i < 8; i += 8, j += 1)
594 hmac_sha512_run_S (esalt_buf, ipad, opad, dgst);
596 tmps[gid].dgst[i + 0] = dgst[0];
597 tmps[gid].dgst[i + 1] = dgst[1];
598 tmps[gid].dgst[i + 2] = dgst[2];
599 tmps[gid].dgst[i + 3] = dgst[3];
600 tmps[gid].dgst[i + 4] = dgst[4];
601 tmps[gid].dgst[i + 5] = dgst[5];
602 tmps[gid].dgst[i + 6] = dgst[6];
603 tmps[gid].dgst[i + 7] = dgst[7];
605 tmps[gid].out[i + 0] = dgst[0];
606 tmps[gid].out[i + 1] = dgst[1];
607 tmps[gid].out[i + 2] = dgst[2];
608 tmps[gid].out[i + 3] = dgst[3];
609 tmps[gid].out[i + 4] = dgst[4];
610 tmps[gid].out[i + 5] = dgst[5];
611 tmps[gid].out[i + 6] = dgst[6];
612 tmps[gid].out[i + 7] = dgst[7];
616 __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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
618 const u32 gid = get_global_id (0);
620 if (gid >= gid_max) return;
625 ipad[0] = pack64v (tmps, ipad, gid, 0);
626 ipad[1] = pack64v (tmps, ipad, gid, 1);
627 ipad[2] = pack64v (tmps, ipad, gid, 2);
628 ipad[3] = pack64v (tmps, ipad, gid, 3);
629 ipad[4] = pack64v (tmps, ipad, gid, 4);
630 ipad[5] = pack64v (tmps, ipad, gid, 5);
631 ipad[6] = pack64v (tmps, ipad, gid, 6);
632 ipad[7] = pack64v (tmps, ipad, gid, 7);
634 opad[0] = pack64v (tmps, opad, gid, 0);
635 opad[1] = pack64v (tmps, opad, gid, 1);
636 opad[2] = pack64v (tmps, opad, gid, 2);
637 opad[3] = pack64v (tmps, opad, gid, 3);
638 opad[4] = pack64v (tmps, opad, gid, 4);
639 opad[5] = pack64v (tmps, opad, gid, 5);
640 opad[6] = pack64v (tmps, opad, gid, 6);
641 opad[7] = pack64v (tmps, opad, gid, 7);
643 for (u32 i = 0; i < 8; i += 8)
647 dgst[0] = pack64v (tmps, dgst, gid, 0);
648 dgst[1] = pack64v (tmps, dgst, gid, 1);
649 dgst[2] = pack64v (tmps, dgst, gid, 2);
650 dgst[3] = pack64v (tmps, dgst, gid, 3);
651 dgst[4] = pack64v (tmps, dgst, gid, 4);
652 dgst[5] = pack64v (tmps, dgst, gid, 5);
653 dgst[6] = pack64v (tmps, dgst, gid, 6);
654 dgst[7] = pack64v (tmps, dgst, gid, 7);
656 for (u32 j = 0; j < loop_cnt; j++)
658 hmac_sha512_run_V_x (ipad, opad, dgst);
660 unpackv_xor (tmps, out, gid, 0, dgst[0]);
661 unpackv_xor (tmps, out, gid, 1, dgst[1]);
662 unpackv_xor (tmps, out, gid, 2, dgst[2]);
663 unpackv_xor (tmps, out, gid, 3, dgst[3]);
664 unpackv_xor (tmps, out, gid, 4, dgst[4]);
665 unpackv_xor (tmps, out, gid, 5, dgst[5]);
666 unpackv_xor (tmps, out, gid, 6, dgst[6]);
667 unpackv_xor (tmps, out, gid, 7, dgst[7]);
670 unpackv (tmps, dgst, gid, 0, dgst[0]);
671 unpackv (tmps, dgst, gid, 1, dgst[1]);
672 unpackv (tmps, dgst, gid, 2, dgst[2]);
673 unpackv (tmps, dgst, gid, 3, dgst[3]);
674 unpackv (tmps, dgst, gid, 4, dgst[4]);
675 unpackv (tmps, dgst, gid, 5, dgst[5]);
676 unpackv (tmps, dgst, gid, 6, dgst[6]);
677 unpackv (tmps, dgst, gid, 7, dgst[7]);
681 __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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
687 const u32 gid = get_global_id (0);
689 if (gid >= gid_max) return;
691 const u32 lid = get_local_id (0);
693 const u64 a = tmps[gid].out[0];
694 const u64 b = tmps[gid].out[1];
696 const u32 r0 = l32_from_64_S (a);
697 const u32 r1 = h32_from_64_S (a);
698 const u32 r2 = l32_from_64_S (b);
699 const u32 r3 = h32_from_64_S (b);