2 * Author......: Jens Steube <jens.steube@gmail.com>
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 __constant u64 k_sha512[80] =
19 SHA512C00, SHA512C01, SHA512C02, SHA512C03,
20 SHA512C04, SHA512C05, SHA512C06, SHA512C07,
21 SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
22 SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
23 SHA512C10, SHA512C11, SHA512C12, SHA512C13,
24 SHA512C14, SHA512C15, SHA512C16, SHA512C17,
25 SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
26 SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
27 SHA512C20, SHA512C21, SHA512C22, SHA512C23,
28 SHA512C24, SHA512C25, SHA512C26, SHA512C27,
29 SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
30 SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
31 SHA512C30, SHA512C31, SHA512C32, SHA512C33,
32 SHA512C34, SHA512C35, SHA512C36, SHA512C37,
33 SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
34 SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
35 SHA512C40, SHA512C41, SHA512C42, SHA512C43,
36 SHA512C44, SHA512C45, SHA512C46, SHA512C47,
37 SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
38 SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
41 void sha512_transform (const u64x w0[4], const u64x w1[4], const u64x w2[4], const u64x w3[4], u64x digest[8])
69 #define ROUND_EXPAND() \
71 w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
72 w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
73 w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
74 w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
75 w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
76 w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
77 w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
78 w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
79 w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
80 w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
81 wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
82 wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
83 wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
84 wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
85 we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
86 wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
89 #define ROUND_STEP(i) \
91 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
92 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
93 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
94 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
95 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
96 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
97 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
98 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
99 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
100 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
101 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
102 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
103 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
104 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
105 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
106 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
114 for (int i = 16; i < 80; i += 16)
116 ROUND_EXPAND (); ROUND_STEP (i);
129 void hmac_sha512_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u64x ipad[8], u64x opad[8])
136 w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x3636363636363636;
137 w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x3636363636363636;
138 w0_t[2] = hl32_to_64 (w1[0], w1[1]) ^ (u64x) 0x3636363636363636;
139 w0_t[3] = hl32_to_64 (w1[2], w1[3]) ^ (u64x) 0x3636363636363636;
140 w1_t[0] = hl32_to_64 (w2[0], w2[1]) ^ (u64x) 0x3636363636363636;
141 w1_t[1] = hl32_to_64 (w2[2], w2[3]) ^ (u64x) 0x3636363636363636;
142 w1_t[2] = hl32_to_64 (w3[0], w3[1]) ^ (u64x) 0x3636363636363636;
143 w1_t[3] = hl32_to_64 (w3[2], w3[3]) ^ (u64x) 0x3636363636363636;
144 w2_t[0] = (u64x) 0x3636363636363636;
145 w2_t[1] = (u64x) 0x3636363636363636;
146 w2_t[2] = (u64x) 0x3636363636363636;
147 w2_t[3] = (u64x) 0x3636363636363636;
148 w3_t[0] = (u64x) 0x3636363636363636;
149 w3_t[1] = (u64x) 0x3636363636363636;
150 w3_t[2] = (u64x) 0x3636363636363636;
151 w3_t[3] = (u64x) 0x3636363636363636;
162 sha512_transform (w0_t, w1_t, w2_t, w3_t, ipad);
164 w0_t[0] = hl32_to_64 (w0[0], w0[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
165 w0_t[1] = hl32_to_64 (w0[2], w0[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
166 w0_t[2] = hl32_to_64 (w1[0], w1[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
167 w0_t[3] = hl32_to_64 (w1[2], w1[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
168 w1_t[0] = hl32_to_64 (w2[0], w2[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
169 w1_t[1] = hl32_to_64 (w2[2], w2[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
170 w1_t[2] = hl32_to_64 (w3[0], w3[1]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
171 w1_t[3] = hl32_to_64 (w3[2], w3[3]) ^ (u64x) 0x5c5c5c5c5c5c5c5c;
172 w2_t[0] = (u64x) 0x5c5c5c5c5c5c5c5c;
173 w2_t[1] = (u64x) 0x5c5c5c5c5c5c5c5c;
174 w2_t[2] = (u64x) 0x5c5c5c5c5c5c5c5c;
175 w2_t[3] = (u64x) 0x5c5c5c5c5c5c5c5c;
176 w3_t[0] = (u64x) 0x5c5c5c5c5c5c5c5c;
177 w3_t[1] = (u64x) 0x5c5c5c5c5c5c5c5c;
178 w3_t[2] = (u64x) 0x5c5c5c5c5c5c5c5c;
179 w3_t[3] = (u64x) 0x5c5c5c5c5c5c5c5c;
190 sha512_transform (w0_t, w1_t, w2_t, w3_t, opad);
193 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])
200 w0_t[0] = hl32_to_64 (w0[0], w0[1]);
201 w0_t[1] = hl32_to_64 (w0[2], w0[3]);
202 w0_t[2] = hl32_to_64 (w1[0], w1[1]);
203 w0_t[3] = hl32_to_64 (w1[2], w1[3]);
204 w1_t[0] = hl32_to_64 (w2[0], w2[1]);
205 w1_t[1] = hl32_to_64 (w2[2], w2[3]);
206 w1_t[2] = hl32_to_64 (w3[0], w3[1]);
215 w3_t[3] = hl32_to_64 (w3[2], w3[3]);
226 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
236 w2_t[0] = 0x8000000000000000;
243 w3_t[3] = (128 + 64) * 8;
254 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
257 __kernel void m01760_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_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)
263 const u32 lid = get_local_id (0);
269 const u32 gid = get_global_id (0);
271 if (gid >= gid_max) return;
276 pw_buf0[0] = pws[gid].i[0];
277 pw_buf0[1] = pws[gid].i[1];
278 pw_buf0[2] = pws[gid].i[2];
279 pw_buf0[3] = pws[gid].i[3];
280 pw_buf1[0] = pws[gid].i[4];
281 pw_buf1[1] = pws[gid].i[5];
282 pw_buf1[2] = pws[gid].i[6];
283 pw_buf1[3] = pws[gid].i[7];
285 const u32 pw_l_len = pws[gid].pw_len;
296 salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
297 salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
298 salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
299 salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
300 salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
301 salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
302 salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
303 salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
304 salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
305 salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
306 salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
307 salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
308 salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
309 salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
310 salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
311 salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
313 const u32 salt_len = salt_bufs[salt_pos].salt_len;
324 w0_t[0] = salt_buf0[0];
325 w0_t[1] = salt_buf0[1];
326 w0_t[2] = salt_buf0[2];
327 w0_t[3] = salt_buf0[3];
328 w1_t[0] = salt_buf1[0];
329 w1_t[1] = salt_buf1[1];
330 w1_t[2] = salt_buf1[2];
331 w1_t[3] = salt_buf1[3];
332 w2_t[0] = salt_buf2[0];
333 w2_t[1] = salt_buf2[1];
334 w2_t[2] = salt_buf2[2];
335 w2_t[3] = salt_buf2[3];
336 w3_t[0] = salt_buf3[0];
337 w3_t[1] = salt_buf3[1];
338 w3_t[2] = salt_buf3[2];
339 w3_t[3] = salt_buf3[3];
344 hmac_sha512_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
350 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
352 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
354 const u32x pw_len = pw_l_len + pw_r_len;
357 * concat password candidate
360 u32x wordl0[4] = { 0 };
361 u32x wordl1[4] = { 0 };
362 u32x wordl2[4] = { 0 };
363 u32x wordl3[4] = { 0 };
365 wordl0[0] = pw_buf0[0];
366 wordl0[1] = pw_buf0[1];
367 wordl0[2] = pw_buf0[2];
368 wordl0[3] = pw_buf0[3];
369 wordl1[0] = pw_buf1[0];
370 wordl1[1] = pw_buf1[1];
371 wordl1[2] = pw_buf1[2];
372 wordl1[3] = pw_buf1[3];
374 u32x wordr0[4] = { 0 };
375 u32x wordr1[4] = { 0 };
376 u32x wordr2[4] = { 0 };
377 u32x wordr3[4] = { 0 };
379 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
380 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
381 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
382 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
383 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
384 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
385 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
386 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
388 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
390 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
394 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
402 w0[0] = wordl0[0] | wordr0[0];
403 w0[1] = wordl0[1] | wordr0[1];
404 w0[2] = wordl0[2] | wordr0[2];
405 w0[3] = wordl0[3] | wordr0[3];
406 w1[0] = wordl1[0] | wordr1[0];
407 w1[1] = wordl1[1] | wordr1[1];
408 w1[2] = wordl1[2] | wordr1[2];
409 w1[3] = wordl1[3] | wordr1[3];
410 w2[0] = wordl2[0] | wordr2[0];
411 w2[1] = wordl2[1] | wordr2[1];
412 w2[2] = wordl2[2] | wordr2[2];
413 w2[3] = wordl2[3] | wordr2[3];
414 w3[0] = wordl3[0] | wordr3[0];
415 w3[1] = wordl3[1] | wordr3[1];
416 w3[2] = wordl3[2] | wordr3[2];
417 w3[3] = wordl3[3] | wordr3[3];
419 append_0x80_4x4_VV (w0, w1, w2, w3, pw_len);
421 w0[0] = swap32 (w0[0]);
422 w0[1] = swap32 (w0[1]);
423 w0[2] = swap32 (w0[2]);
424 w0[3] = swap32 (w0[3]);
425 w1[0] = swap32 (w1[0]);
426 w1[1] = swap32 (w1[1]);
427 w1[2] = swap32 (w1[2]);
428 w1[3] = swap32 (w1[3]);
429 w2[0] = swap32 (w2[0]);
430 w2[1] = swap32 (w2[1]);
431 w2[2] = swap32 (w2[2]);
432 w2[3] = swap32 (w2[3]);
433 w3[0] = swap32 (w3[0]);
434 w3[1] = swap32 (w3[1]);
436 w3[3] = (128 + pw_len) * 8;
440 hmac_sha512_run (w0, w1, w2, w3, ipad, opad, digest);
442 const u32x r0 = l32_from_64 (digest[7]);
443 const u32x r1 = h32_from_64 (digest[7]);
444 const u32x r2 = l32_from_64 (digest[3]);
445 const u32x r3 = h32_from_64 (digest[3]);
447 COMPARE_M_SIMD (r0, r1, r2, r3);
451 __kernel void m01760_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_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)
455 __kernel void m01760_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_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)
459 __kernel void m01760_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_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)
465 const u32 lid = get_local_id (0);
471 const u32 gid = get_global_id (0);
473 if (gid >= gid_max) return;
478 pw_buf0[0] = pws[gid].i[0];
479 pw_buf0[1] = pws[gid].i[1];
480 pw_buf0[2] = pws[gid].i[2];
481 pw_buf0[3] = pws[gid].i[3];
482 pw_buf1[0] = pws[gid].i[4];
483 pw_buf1[1] = pws[gid].i[5];
484 pw_buf1[2] = pws[gid].i[6];
485 pw_buf1[3] = pws[gid].i[7];
487 const u32 pw_l_len = pws[gid].pw_len;
498 salt_buf0[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
499 salt_buf0[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
500 salt_buf0[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]);
501 salt_buf0[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 3]);
502 salt_buf1[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 4]);
503 salt_buf1[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 5]);
504 salt_buf1[2] = swap32_S (salt_bufs[salt_pos].salt_buf[ 6]);
505 salt_buf1[3] = swap32_S (salt_bufs[salt_pos].salt_buf[ 7]);
506 salt_buf2[0] = swap32_S (salt_bufs[salt_pos].salt_buf[ 8]);
507 salt_buf2[1] = swap32_S (salt_bufs[salt_pos].salt_buf[ 9]);
508 salt_buf2[2] = swap32_S (salt_bufs[salt_pos].salt_buf[10]);
509 salt_buf2[3] = swap32_S (salt_bufs[salt_pos].salt_buf[11]);
510 salt_buf3[0] = swap32_S (salt_bufs[salt_pos].salt_buf[12]);
511 salt_buf3[1] = swap32_S (salt_bufs[salt_pos].salt_buf[13]);
512 salt_buf3[2] = swap32_S (salt_bufs[salt_pos].salt_buf[14]);
513 salt_buf3[3] = swap32_S (salt_bufs[salt_pos].salt_buf[15]);
515 const u32 salt_len = salt_bufs[salt_pos].salt_len;
526 w0_t[0] = salt_buf0[0];
527 w0_t[1] = salt_buf0[1];
528 w0_t[2] = salt_buf0[2];
529 w0_t[3] = salt_buf0[3];
530 w1_t[0] = salt_buf1[0];
531 w1_t[1] = salt_buf1[1];
532 w1_t[2] = salt_buf1[2];
533 w1_t[3] = salt_buf1[3];
534 w2_t[0] = salt_buf2[0];
535 w2_t[1] = salt_buf2[1];
536 w2_t[2] = salt_buf2[2];
537 w2_t[3] = salt_buf2[3];
538 w3_t[0] = salt_buf3[0];
539 w3_t[1] = salt_buf3[1];
540 w3_t[2] = salt_buf3[2];
541 w3_t[3] = salt_buf3[3];
546 hmac_sha512_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
552 const u32 search[4] =
554 digests_buf[digests_offset].digest_buf[DGST_R0],
555 digests_buf[digests_offset].digest_buf[DGST_R1],
556 digests_buf[digests_offset].digest_buf[DGST_R2],
557 digests_buf[digests_offset].digest_buf[DGST_R3]
564 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
566 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
568 const u32x pw_len = pw_l_len + pw_r_len;
571 * concat password candidate
574 u32x wordl0[4] = { 0 };
575 u32x wordl1[4] = { 0 };
576 u32x wordl2[4] = { 0 };
577 u32x wordl3[4] = { 0 };
579 wordl0[0] = pw_buf0[0];
580 wordl0[1] = pw_buf0[1];
581 wordl0[2] = pw_buf0[2];
582 wordl0[3] = pw_buf0[3];
583 wordl1[0] = pw_buf1[0];
584 wordl1[1] = pw_buf1[1];
585 wordl1[2] = pw_buf1[2];
586 wordl1[3] = pw_buf1[3];
588 u32x wordr0[4] = { 0 };
589 u32x wordr1[4] = { 0 };
590 u32x wordr2[4] = { 0 };
591 u32x wordr3[4] = { 0 };
593 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
594 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
595 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
596 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
597 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
598 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
599 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
600 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
602 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
604 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
608 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
616 w0[0] = wordl0[0] | wordr0[0];
617 w0[1] = wordl0[1] | wordr0[1];
618 w0[2] = wordl0[2] | wordr0[2];
619 w0[3] = wordl0[3] | wordr0[3];
620 w1[0] = wordl1[0] | wordr1[0];
621 w1[1] = wordl1[1] | wordr1[1];
622 w1[2] = wordl1[2] | wordr1[2];
623 w1[3] = wordl1[3] | wordr1[3];
624 w2[0] = wordl2[0] | wordr2[0];
625 w2[1] = wordl2[1] | wordr2[1];
626 w2[2] = wordl2[2] | wordr2[2];
627 w2[3] = wordl2[3] | wordr2[3];
628 w3[0] = wordl3[0] | wordr3[0];
629 w3[1] = wordl3[1] | wordr3[1];
630 w3[2] = wordl3[2] | wordr3[2];
631 w3[3] = wordl3[3] | wordr3[3];
633 append_0x80_4x4_VV (w0, w1, w2, w3, pw_len);
635 w0[0] = swap32 (w0[0]);
636 w0[1] = swap32 (w0[1]);
637 w0[2] = swap32 (w0[2]);
638 w0[3] = swap32 (w0[3]);
639 w1[0] = swap32 (w1[0]);
640 w1[1] = swap32 (w1[1]);
641 w1[2] = swap32 (w1[2]);
642 w1[3] = swap32 (w1[3]);
643 w2[0] = swap32 (w2[0]);
644 w2[1] = swap32 (w2[1]);
645 w2[2] = swap32 (w2[2]);
646 w2[3] = swap32 (w2[3]);
647 w3[0] = swap32 (w3[0]);
648 w3[1] = swap32 (w3[1]);
650 w3[3] = (128 + pw_len) * 8;
654 hmac_sha512_run (w0, w1, w2, w3, ipad, opad, digest);
656 const u32x r0 = l32_from_64 (digest[7]);
657 const u32x r1 = h32_from_64 (digest[7]);
658 const u32x r2 = l32_from_64 (digest[3]);
659 const u32x r3 = h32_from_64 (digest[3]);
661 COMPARE_S_SIMD (r0, r1, r2, r3);
665 __kernel void m01760_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_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)
669 __kernel void m01760_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_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)