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 u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u64x digest[8])
43 u64x w0_t = hl32_to_64 (w0[0], w0[1]);
44 u64x w1_t = hl32_to_64 (w0[2], w0[3]);
45 u64x w2_t = hl32_to_64 (w1[0], w1[1]);
46 u64x w3_t = hl32_to_64 (w1[2], w1[3]);
47 u64x w4_t = hl32_to_64 (w2[0], w2[1]);
48 u64x w5_t = hl32_to_64 (w2[2], w2[3]);
49 u64x w6_t = hl32_to_64 (w3[0], w3[1]);
58 u64x wf_t = hl32_to_64 (w3[2], w3[3]);
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);
140 __kernel void m01720_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)
146 const u32 lid = get_local_id (0);
152 const u32 gid = get_global_id (0);
154 if (gid >= gid_max) return;
159 pw_buf0[0] = pws[gid].i[0];
160 pw_buf0[1] = pws[gid].i[1];
161 pw_buf0[2] = pws[gid].i[2];
162 pw_buf0[3] = pws[gid].i[3];
163 pw_buf1[0] = pws[gid].i[4];
164 pw_buf1[1] = pws[gid].i[5];
165 pw_buf1[2] = pws[gid].i[6];
166 pw_buf1[3] = pws[gid].i[7];
168 const u32 pw_l_len = pws[gid].pw_len;
179 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
180 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
181 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
182 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
183 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
184 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
185 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
186 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
187 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
188 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
189 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
190 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
191 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
192 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
193 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
194 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
196 const u32 salt_len = salt_bufs[salt_pos].salt_len;
202 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
204 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
206 const u32x pw_len = pw_l_len + pw_r_len;
209 * concat password candidate
212 u32x wordl0[4] = { 0 };
213 u32x wordl1[4] = { 0 };
214 u32x wordl2[4] = { 0 };
215 u32x wordl3[4] = { 0 };
217 wordl0[0] = pw_buf0[0];
218 wordl0[1] = pw_buf0[1];
219 wordl0[2] = pw_buf0[2];
220 wordl0[3] = pw_buf0[3];
221 wordl1[0] = pw_buf1[0];
222 wordl1[1] = pw_buf1[1];
223 wordl1[2] = pw_buf1[2];
224 wordl1[3] = pw_buf1[3];
226 u32x wordr0[4] = { 0 };
227 u32x wordr1[4] = { 0 };
228 u32x wordr2[4] = { 0 };
229 u32x wordr3[4] = { 0 };
231 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
232 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
233 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
234 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
235 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
236 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
237 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
238 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
240 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
242 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
246 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
254 w0[0] = wordl0[0] | wordr0[0];
255 w0[1] = wordl0[1] | wordr0[1];
256 w0[2] = wordl0[2] | wordr0[2];
257 w0[3] = wordl0[3] | wordr0[3];
258 w1[0] = wordl1[0] | wordr1[0];
259 w1[1] = wordl1[1] | wordr1[1];
260 w1[2] = wordl1[2] | wordr1[2];
261 w1[3] = wordl1[3] | wordr1[3];
262 w2[0] = wordl2[0] | wordr2[0];
263 w2[1] = wordl2[1] | wordr2[1];
264 w2[2] = wordl2[2] | wordr2[2];
265 w2[3] = wordl2[3] | wordr2[3];
266 w3[0] = wordl3[0] | wordr3[0];
267 w3[1] = wordl3[1] | wordr3[1];
268 w3[2] = wordl3[2] | wordr3[2];
269 w3[3] = wordl3[3] | wordr3[3];
275 switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
277 const u32x pw_salt_len = pw_len + salt_len;
279 w0[0] |= salt_buf0[0];
280 w0[1] |= salt_buf0[1];
281 w0[2] |= salt_buf0[2];
282 w0[3] |= salt_buf0[3];
283 w1[0] |= salt_buf1[0];
284 w1[1] |= salt_buf1[1];
285 w1[2] |= salt_buf1[2];
286 w1[3] |= salt_buf1[3];
287 w2[0] |= salt_buf2[0];
288 w2[1] |= salt_buf2[1];
289 w2[2] |= salt_buf2[2];
290 w2[3] |= salt_buf2[3];
291 w3[0] |= salt_buf3[0];
292 w3[1] |= salt_buf3[1];
293 w3[2] |= salt_buf3[2];
294 w3[3] |= salt_buf3[3];
305 w0_t[0] = swap32 (w0[0]);
306 w0_t[1] = swap32 (w0[1]);
307 w0_t[2] = swap32 (w0[2]);
308 w0_t[3] = swap32 (w0[3]);
309 w1_t[0] = swap32 (w1[0]);
310 w1_t[1] = swap32 (w1[1]);
311 w1_t[2] = swap32 (w1[2]);
312 w1_t[3] = swap32 (w1[3]);
313 w2_t[0] = swap32 (w2[0]);
314 w2_t[1] = swap32 (w2[1]);
315 w2_t[2] = swap32 (w2[2]);
316 w2_t[3] = swap32 (w2[3]);
317 w3_t[0] = swap32 (w3[0]);
318 w3_t[1] = swap32 (w3[1]);
320 w3_t[3] = pw_salt_len * 8;
324 digest[0] = SHA512M_A;
325 digest[1] = SHA512M_B;
326 digest[2] = SHA512M_C;
327 digest[3] = SHA512M_D;
328 digest[4] = SHA512M_E;
329 digest[5] = SHA512M_F;
330 digest[6] = SHA512M_G;
331 digest[7] = SHA512M_H;
333 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
335 const u32x r0 = l32_from_64 (digest[7]);
336 const u32x r1 = h32_from_64 (digest[7]);
337 const u32x r2 = l32_from_64 (digest[3]);
338 const u32x r3 = h32_from_64 (digest[3]);
340 COMPARE_M_SIMD (r0, r1, r2, r3);
344 __kernel void m01720_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)
348 __kernel void m01720_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)
352 __kernel void m01720_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)
358 const u32 lid = get_local_id (0);
364 const u32 gid = get_global_id (0);
366 if (gid >= gid_max) return;
371 pw_buf0[0] = pws[gid].i[0];
372 pw_buf0[1] = pws[gid].i[1];
373 pw_buf0[2] = pws[gid].i[2];
374 pw_buf0[3] = pws[gid].i[3];
375 pw_buf1[0] = pws[gid].i[4];
376 pw_buf1[1] = pws[gid].i[5];
377 pw_buf1[2] = pws[gid].i[6];
378 pw_buf1[3] = pws[gid].i[7];
380 const u32 pw_l_len = pws[gid].pw_len;
391 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
392 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
393 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
394 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
395 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
396 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
397 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
398 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
399 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
400 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
401 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
402 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
403 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
404 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
405 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
406 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
408 const u32 salt_len = salt_bufs[salt_pos].salt_len;
414 const u32 search[4] =
416 digests_buf[digests_offset].digest_buf[DGST_R0],
417 digests_buf[digests_offset].digest_buf[DGST_R1],
418 digests_buf[digests_offset].digest_buf[DGST_R2],
419 digests_buf[digests_offset].digest_buf[DGST_R3]
426 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
428 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
430 const u32x pw_len = pw_l_len + pw_r_len;
433 * concat password candidate
436 u32x wordl0[4] = { 0 };
437 u32x wordl1[4] = { 0 };
438 u32x wordl2[4] = { 0 };
439 u32x wordl3[4] = { 0 };
441 wordl0[0] = pw_buf0[0];
442 wordl0[1] = pw_buf0[1];
443 wordl0[2] = pw_buf0[2];
444 wordl0[3] = pw_buf0[3];
445 wordl1[0] = pw_buf1[0];
446 wordl1[1] = pw_buf1[1];
447 wordl1[2] = pw_buf1[2];
448 wordl1[3] = pw_buf1[3];
450 u32x wordr0[4] = { 0 };
451 u32x wordr1[4] = { 0 };
452 u32x wordr2[4] = { 0 };
453 u32x wordr3[4] = { 0 };
455 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
456 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
457 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
458 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
459 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
460 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
461 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
462 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
464 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
466 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
470 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
478 w0[0] = wordl0[0] | wordr0[0];
479 w0[1] = wordl0[1] | wordr0[1];
480 w0[2] = wordl0[2] | wordr0[2];
481 w0[3] = wordl0[3] | wordr0[3];
482 w1[0] = wordl1[0] | wordr1[0];
483 w1[1] = wordl1[1] | wordr1[1];
484 w1[2] = wordl1[2] | wordr1[2];
485 w1[3] = wordl1[3] | wordr1[3];
486 w2[0] = wordl2[0] | wordr2[0];
487 w2[1] = wordl2[1] | wordr2[1];
488 w2[2] = wordl2[2] | wordr2[2];
489 w2[3] = wordl2[3] | wordr2[3];
490 w3[0] = wordl3[0] | wordr3[0];
491 w3[1] = wordl3[1] | wordr3[1];
492 w3[2] = wordl3[2] | wordr3[2];
493 w3[3] = wordl3[3] | wordr3[3];
499 switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
501 const u32x pw_salt_len = pw_len + salt_len;
503 w0[0] |= salt_buf0[0];
504 w0[1] |= salt_buf0[1];
505 w0[2] |= salt_buf0[2];
506 w0[3] |= salt_buf0[3];
507 w1[0] |= salt_buf1[0];
508 w1[1] |= salt_buf1[1];
509 w1[2] |= salt_buf1[2];
510 w1[3] |= salt_buf1[3];
511 w2[0] |= salt_buf2[0];
512 w2[1] |= salt_buf2[1];
513 w2[2] |= salt_buf2[2];
514 w2[3] |= salt_buf2[3];
515 w3[0] |= salt_buf3[0];
516 w3[1] |= salt_buf3[1];
517 w3[2] |= salt_buf3[2];
518 w3[3] |= salt_buf3[3];
529 w0_t[0] = swap32 (w0[0]);
530 w0_t[1] = swap32 (w0[1]);
531 w0_t[2] = swap32 (w0[2]);
532 w0_t[3] = swap32 (w0[3]);
533 w1_t[0] = swap32 (w1[0]);
534 w1_t[1] = swap32 (w1[1]);
535 w1_t[2] = swap32 (w1[2]);
536 w1_t[3] = swap32 (w1[3]);
537 w2_t[0] = swap32 (w2[0]);
538 w2_t[1] = swap32 (w2[1]);
539 w2_t[2] = swap32 (w2[2]);
540 w2_t[3] = swap32 (w2[3]);
541 w3_t[0] = swap32 (w3[0]);
542 w3_t[1] = swap32 (w3[1]);
544 w3_t[3] = pw_salt_len * 8;
548 digest[0] = SHA512M_A;
549 digest[1] = SHA512M_B;
550 digest[2] = SHA512M_C;
551 digest[3] = SHA512M_D;
552 digest[4] = SHA512M_E;
553 digest[5] = SHA512M_F;
554 digest[6] = SHA512M_G;
555 digest[7] = SHA512M_H;
557 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
559 const u32x r0 = l32_from_64 (digest[7]);
560 const u32x r1 = h32_from_64 (digest[7]);
561 const u32x r2 = l32_from_64 (digest[3]);
562 const u32x r3 = h32_from_64 (digest[3]);
564 COMPARE_S_SIMD (r0, r1, r2, r3);
568 __kernel void m01720_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)
572 __kernel void m01720_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)