2 * Author......: Jens Steube <jens.steube@gmail.com>
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 __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 void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
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]); \
120 for (int i = 16; i < 80; i += 16)
122 ROUND_EXPAND (); ROUND_STEP (i);
135 void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
142 w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x3636363636363636;
143 w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x3636363636363636;
144 w0_t[2] = hl32_to_64 (w1[0], w1[1]) ^ (u64x) 0x3636363636363636;
145 w0_t[3] = hl32_to_64 (w1[2], w1[3]) ^ (u64x) 0x3636363636363636;
146 w1_t[0] = hl32_to_64 (w2[0], w2[1]) ^ (u64x) 0x3636363636363636;
147 w1_t[1] = hl32_to_64 (w2[2], w2[3]) ^ (u64x) 0x3636363636363636;
148 w1_t[2] = hl32_to_64 (w3[0], w3[1]) ^ (u64x) 0x3636363636363636;
149 w1_t[3] = hl32_to_64 (w3[2], w3[3]) ^ (u64x) 0x3636363636363636;
150 w2_t[0] = (u64x) 0x3636363636363636;
151 w2_t[1] = (u64x) 0x3636363636363636;
152 w2_t[2] = (u64x) 0x3636363636363636;
153 w2_t[3] = (u64x) 0x3636363636363636;
154 w3_t[0] = (u64x) 0x3636363636363636;
155 w3_t[1] = (u64x) 0x3636363636363636;
156 w3_t[2] = (u64x) 0x3636363636363636;
157 w3_t[3] = (u64x) 0x3636363636363636;
168 sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
170 w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
171 w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
172 w0_t[2] = hl32_to_64 (w1[0], w1[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
173 w0_t[3] = hl32_to_64 (w1[2], w1[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
174 w1_t[0] = hl32_to_64 (w2[0], w2[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
175 w1_t[1] = hl32_to_64 (w2[2], w2[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
176 w1_t[2] = hl32_to_64 (w3[0], w3[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
177 w1_t[3] = hl32_to_64 (w3[2], w3[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
178 w2_t[0] = (u64x) 0x5c5c5c5c5c5c5c5c;
179 w2_t[1] = (u64x) 0x5c5c5c5c5c5c5c5c;
180 w2_t[2] = (u64x) 0x5c5c5c5c5c5c5c5c;
181 w2_t[3] = (u64x) 0x5c5c5c5c5c5c5c5c;
182 w3_t[0] = (u64x) 0x5c5c5c5c5c5c5c5c;
183 w3_t[1] = (u64x) 0x5c5c5c5c5c5c5c5c;
184 w3_t[2] = (u64x) 0x5c5c5c5c5c5c5c5c;
185 w3_t[3] = (u64x) 0x5c5c5c5c5c5c5c5c;
196 sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
199 void hmac_sha512_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8], u64x digest[8])
206 w0_t[0] = hl32_to_64 (w0[0], w0[1]);
207 w0_t[1] = hl32_to_64 (w0[2], w0[3]);
208 w0_t[2] = hl32_to_64 (w1[0], w1[1]);
209 w0_t[3] = hl32_to_64 (w1[2], w1[3]);
210 w1_t[0] = hl32_to_64 (w2[0], w2[1]);
211 w1_t[1] = hl32_to_64 (w2[2], w2[3]);
212 w1_t[2] = hl32_to_64 (w3[0], w3[1]);
221 w3_t[3] = hl32_to_64 (w3[2], w3[3]);
232 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
242 w2_t[0] = 0x8000000000000000;
249 w3_t[3] = (128 + 64) * 8;
260 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
263 __kernel void m01750_m04 (__global pw_t *pws, __global kernel_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
269 const u32 lid = get_local_id (0);
275 const u32 gid = get_global_id (0);
277 if (gid >= gid_max) return;
282 pw_buf0[0] = pws[gid].i[0];
283 pw_buf0[1] = pws[gid].i[1];
284 pw_buf0[2] = pws[gid].i[2];
285 pw_buf0[3] = pws[gid].i[3];
286 pw_buf1[0] = pws[gid].i[4];
287 pw_buf1[1] = pws[gid].i[5];
288 pw_buf1[2] = pws[gid].i[6];
289 pw_buf1[3] = pws[gid].i[7];
291 const u32 pw_l_len = pws[gid].pw_len;
302 salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
303 salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
304 salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
305 salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
306 salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
307 salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
308 salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
309 salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
310 salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
311 salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
312 salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
313 salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
314 salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
315 salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
316 salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
317 salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
319 const u32 salt_len = salt_bufs[salt_pos].salt_len;
325 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
327 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
329 const u32x pw_len = pw_l_len + pw_r_len;
332 * concat password candidate
335 u32x wordl0[4] = { 0 };
336 u32x wordl1[4] = { 0 };
337 u32x wordl2[4] = { 0 };
338 u32x wordl3[4] = { 0 };
340 wordl0[0] = pw_buf0[0];
341 wordl0[1] = pw_buf0[1];
342 wordl0[2] = pw_buf0[2];
343 wordl0[3] = pw_buf0[3];
344 wordl1[0] = pw_buf1[0];
345 wordl1[1] = pw_buf1[1];
346 wordl1[2] = pw_buf1[2];
347 wordl1[3] = pw_buf1[3];
349 u32x wordr0[4] = { 0 };
350 u32x wordr1[4] = { 0 };
351 u32x wordr2[4] = { 0 };
352 u32x wordr3[4] = { 0 };
354 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
355 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
356 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
357 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
358 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
359 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
360 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
361 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
363 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
365 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
369 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
377 w0[0] = wordl0[0] | wordr0[0];
378 w0[1] = wordl0[1] | wordr0[1];
379 w0[2] = wordl0[2] | wordr0[2];
380 w0[3] = wordl0[3] | wordr0[3];
381 w1[0] = wordl1[0] | wordr1[0];
382 w1[1] = wordl1[1] | wordr1[1];
383 w1[2] = wordl1[2] | wordr1[2];
384 w1[3] = wordl1[3] | wordr1[3];
385 w2[0] = wordl2[0] | wordr2[0];
386 w2[1] = wordl2[1] | wordr2[1];
387 w2[2] = wordl2[2] | wordr2[2];
388 w2[3] = wordl2[3] | wordr2[3];
389 w3[0] = wordl3[0] | wordr3[0];
390 w3[1] = wordl3[1] | wordr3[1];
391 w3[2] = wordl3[2] | wordr3[2];
392 w3[3] = wordl3[3] | wordr3[3];
394 w0[0] = swap32 (w0[0]);
395 w0[1] = swap32 (w0[1]);
396 w0[2] = swap32 (w0[2]);
397 w0[3] = swap32 (w0[3]);
398 w1[0] = swap32 (w1[0]);
399 w1[1] = swap32 (w1[1]);
400 w1[2] = swap32 (w1[2]);
401 w1[3] = swap32 (w1[3]);
402 w2[0] = swap32 (w2[0]);
403 w2[1] = swap32 (w2[1]);
404 w2[2] = swap32 (w2[2]);
405 w2[3] = swap32 (w2[3]);
406 w3[0] = swap32 (w3[0]);
407 w3[1] = swap32 (w3[1]);
408 w3[2] = swap32 (w3[2]);
409 w3[3] = swap32 (w3[3]);
418 hmac_sha512_pad (w0, w1, w2, w3, ipad, opad);
420 w0[0] = salt_buf0[0];
421 w0[1] = salt_buf0[1];
422 w0[2] = salt_buf0[2];
423 w0[3] = salt_buf0[3];
424 w1[0] = salt_buf1[0];
425 w1[1] = salt_buf1[1];
426 w1[2] = salt_buf1[2];
427 w1[3] = salt_buf1[3];
428 w2[0] = salt_buf2[0];
429 w2[1] = salt_buf2[1];
430 w2[2] = salt_buf2[2];
431 w2[3] = salt_buf2[3];
432 w3[0] = salt_buf3[0];
433 w3[1] = salt_buf3[1];
435 w3[3] = (128 + salt_len) * 8;
439 hmac_sha512_run (w0, w1, w2, w3, ipad, opad, digest);
441 const u32x r0 = l32_from_64 (digest[7]);
442 const u32x r1 = h32_from_64 (digest[7]);
443 const u32x r2 = l32_from_64 (digest[3]);
444 const u32x r3 = h32_from_64 (digest[3]);
446 COMPARE_M_SIMD (r0, r1, r2, r3);
450 __kernel void m01750_m08 (__global pw_t *pws, __global kernel_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
454 __kernel void m01750_m16 (__global pw_t *pws, __global kernel_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
458 __kernel void m01750_s04 (__global pw_t *pws, __global kernel_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
464 const u32 lid = get_local_id (0);
470 const u32 gid = get_global_id (0);
472 if (gid >= gid_max) return;
477 pw_buf0[0] = pws[gid].i[0];
478 pw_buf0[1] = pws[gid].i[1];
479 pw_buf0[2] = pws[gid].i[2];
480 pw_buf0[3] = pws[gid].i[3];
481 pw_buf1[0] = pws[gid].i[4];
482 pw_buf1[1] = pws[gid].i[5];
483 pw_buf1[2] = pws[gid].i[6];
484 pw_buf1[3] = pws[gid].i[7];
486 const u32 pw_l_len = pws[gid].pw_len;
497 salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
498 salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
499 salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
500 salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
501 salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
502 salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
503 salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
504 salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
505 salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
506 salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
507 salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
508 salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
509 salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
510 salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
511 salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
512 salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
514 const u32 salt_len = salt_bufs[salt_pos].salt_len;
520 const u32 search[4] =
522 digests_buf[digests_offset].digest_buf[DGST_R0],
523 digests_buf[digests_offset].digest_buf[DGST_R1],
524 digests_buf[digests_offset].digest_buf[DGST_R2],
525 digests_buf[digests_offset].digest_buf[DGST_R3]
532 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
534 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
536 const u32x pw_len = pw_l_len + pw_r_len;
539 * concat password candidate
542 u32x wordl0[4] = { 0 };
543 u32x wordl1[4] = { 0 };
544 u32x wordl2[4] = { 0 };
545 u32x wordl3[4] = { 0 };
547 wordl0[0] = pw_buf0[0];
548 wordl0[1] = pw_buf0[1];
549 wordl0[2] = pw_buf0[2];
550 wordl0[3] = pw_buf0[3];
551 wordl1[0] = pw_buf1[0];
552 wordl1[1] = pw_buf1[1];
553 wordl1[2] = pw_buf1[2];
554 wordl1[3] = pw_buf1[3];
556 u32x wordr0[4] = { 0 };
557 u32x wordr1[4] = { 0 };
558 u32x wordr2[4] = { 0 };
559 u32x wordr3[4] = { 0 };
561 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
562 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
563 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
564 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
565 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
566 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
567 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
568 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
570 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
572 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
576 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
584 w0[0] = wordl0[0] | wordr0[0];
585 w0[1] = wordl0[1] | wordr0[1];
586 w0[2] = wordl0[2] | wordr0[2];
587 w0[3] = wordl0[3] | wordr0[3];
588 w1[0] = wordl1[0] | wordr1[0];
589 w1[1] = wordl1[1] | wordr1[1];
590 w1[2] = wordl1[2] | wordr1[2];
591 w1[3] = wordl1[3] | wordr1[3];
592 w2[0] = wordl2[0] | wordr2[0];
593 w2[1] = wordl2[1] | wordr2[1];
594 w2[2] = wordl2[2] | wordr2[2];
595 w2[3] = wordl2[3] | wordr2[3];
596 w3[0] = wordl3[0] | wordr3[0];
597 w3[1] = wordl3[1] | wordr3[1];
598 w3[2] = wordl3[2] | wordr3[2];
599 w3[3] = wordl3[3] | wordr3[3];
601 w0[0] = swap32 (w0[0]);
602 w0[1] = swap32 (w0[1]);
603 w0[2] = swap32 (w0[2]);
604 w0[3] = swap32 (w0[3]);
605 w1[0] = swap32 (w1[0]);
606 w1[1] = swap32 (w1[1]);
607 w1[2] = swap32 (w1[2]);
608 w1[3] = swap32 (w1[3]);
609 w2[0] = swap32 (w2[0]);
610 w2[1] = swap32 (w2[1]);
611 w2[2] = swap32 (w2[2]);
612 w2[3] = swap32 (w2[3]);
613 w3[0] = swap32 (w3[0]);
614 w3[1] = swap32 (w3[1]);
615 w3[2] = swap32 (w3[2]);
616 w3[3] = swap32 (w3[3]);
625 hmac_sha512_pad (w0, w1, w2, w3, ipad, opad);
627 w0[0] = salt_buf0[0];
628 w0[1] = salt_buf0[1];
629 w0[2] = salt_buf0[2];
630 w0[3] = salt_buf0[3];
631 w1[0] = salt_buf1[0];
632 w1[1] = salt_buf1[1];
633 w1[2] = salt_buf1[2];
634 w1[3] = salt_buf1[3];
635 w2[0] = salt_buf2[0];
636 w2[1] = salt_buf2[1];
637 w2[2] = salt_buf2[2];
638 w2[3] = salt_buf2[3];
639 w3[0] = salt_buf3[0];
640 w3[1] = salt_buf3[1];
642 w3[3] = (128 + salt_len) * 8;
646 hmac_sha512_run (w0, w1, w2, w3, ipad, opad, digest);
648 const u32x r0 = l32_from_64 (digest[7]);
649 const u32x r1 = h32_from_64 (digest[7]);
650 const u32x r2 = l32_from_64 (digest[3]);
651 const u32x r3 = h32_from_64 (digest[3]);
653 COMPARE_S_SIMD (r0, r1, r2, r3);
657 __kernel void m01750_s08 (__global pw_t *pws, __global kernel_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
661 __kernel void m01750_s16 (__global pw_t *pws, __global kernel_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)