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 u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u64x digest[8])
49 u64x w0_t = hl32_to_64 (w0[0], w0[1]);
50 u64x w1_t = hl32_to_64 (w0[2], w0[3]);
51 u64x w2_t = hl32_to_64 (w1[0], w1[1]);
52 u64x w3_t = hl32_to_64 (w1[2], w1[3]);
53 u64x w4_t = hl32_to_64 (w2[0], w2[1]);
54 u64x w5_t = hl32_to_64 (w2[2], w2[3]);
55 u64x w6_t = hl32_to_64 (w3[0], w3[1]);
64 u64x wf_t = hl32_to_64 (w3[2], w3[3]);
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);
146 __kernel void m01730_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)
152 const u32 lid = get_local_id (0);
158 const u32 gid = get_global_id (0);
160 if (gid >= gid_max) return;
165 pw_buf0[0] = pws[gid].i[0];
166 pw_buf0[1] = pws[gid].i[1];
167 pw_buf0[2] = pws[gid].i[2];
168 pw_buf0[3] = pws[gid].i[3];
169 pw_buf1[0] = pws[gid].i[4];
170 pw_buf1[1] = pws[gid].i[5];
171 pw_buf1[2] = pws[gid].i[6];
172 pw_buf1[3] = pws[gid].i[7];
174 const u32 pw_l_len = pws[gid].pw_len;
185 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
186 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
187 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
188 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
189 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
190 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
191 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
192 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
193 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
194 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
195 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
196 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
197 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
198 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
199 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
200 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
202 const u32 salt_len = salt_bufs[salt_pos].salt_len;
208 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
210 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
212 const u32x pw_len = pw_l_len + pw_r_len;
215 * concat password candidate
218 u32x wordl0[4] = { 0 };
219 u32x wordl1[4] = { 0 };
220 u32x wordl2[4] = { 0 };
221 u32x wordl3[4] = { 0 };
223 wordl0[0] = pw_buf0[0];
224 wordl0[1] = pw_buf0[1];
225 wordl0[2] = pw_buf0[2];
226 wordl0[3] = pw_buf0[3];
227 wordl1[0] = pw_buf1[0];
228 wordl1[1] = pw_buf1[1];
229 wordl1[2] = pw_buf1[2];
230 wordl1[3] = pw_buf1[3];
232 u32x wordr0[4] = { 0 };
233 u32x wordr1[4] = { 0 };
234 u32x wordr2[4] = { 0 };
235 u32x wordr3[4] = { 0 };
237 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
238 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
239 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
240 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
241 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
242 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
243 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
244 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
246 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
248 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
252 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
260 w0[0] = wordl0[0] | wordr0[0];
261 w0[1] = wordl0[1] | wordr0[1];
262 w0[2] = wordl0[2] | wordr0[2];
263 w0[3] = wordl0[3] | wordr0[3];
264 w1[0] = wordl1[0] | wordr1[0];
265 w1[1] = wordl1[1] | wordr1[1];
266 w1[2] = wordl1[2] | wordr1[2];
267 w1[3] = wordl1[3] | wordr1[3];
268 w2[0] = wordl2[0] | wordr2[0];
269 w2[1] = wordl2[1] | wordr2[1];
270 w2[2] = wordl2[2] | wordr2[2];
271 w2[3] = wordl2[3] | wordr2[3];
272 w3[0] = wordl3[0] | wordr3[0];
273 w3[1] = wordl3[1] | wordr3[1];
274 w3[2] = wordl3[2] | wordr3[2];
275 w3[3] = wordl3[3] | wordr3[3];
277 make_unicode (w1, w2, w3);
278 make_unicode (w0, w0, w1);
280 const u32x pw_len2 = pw_len * 2;
291 s0[0] = salt_buf0[0];
292 s0[1] = salt_buf0[1];
293 s0[2] = salt_buf0[2];
294 s0[3] = salt_buf0[3];
295 s1[0] = salt_buf1[0];
296 s1[1] = salt_buf1[1];
297 s1[2] = salt_buf1[2];
298 s1[3] = salt_buf1[3];
299 s2[0] = salt_buf2[0];
300 s2[1] = salt_buf2[1];
301 s2[2] = salt_buf2[2];
302 s2[3] = salt_buf2[3];
303 s3[0] = salt_buf3[0];
304 s3[1] = salt_buf3[1];
305 s3[2] = salt_buf3[2];
306 s3[3] = salt_buf3[3];
308 switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len2);
310 const u32x pw_salt_len = pw_len2 + salt_len;
338 w0_t[0] = swap32 (w0[0]);
339 w0_t[1] = swap32 (w0[1]);
340 w0_t[2] = swap32 (w0[2]);
341 w0_t[3] = swap32 (w0[3]);
342 w1_t[0] = swap32 (w1[0]);
343 w1_t[1] = swap32 (w1[1]);
344 w1_t[2] = swap32 (w1[2]);
345 w1_t[3] = swap32 (w1[3]);
346 w2_t[0] = swap32 (w2[0]);
347 w2_t[1] = swap32 (w2[1]);
348 w2_t[2] = swap32 (w2[2]);
349 w2_t[3] = swap32 (w2[3]);
350 w3_t[0] = swap32 (w3[0]);
351 w3_t[1] = swap32 (w3[1]);
353 w3_t[3] = pw_salt_len * 8;
357 digest[0] = SHA512M_A;
358 digest[1] = SHA512M_B;
359 digest[2] = SHA512M_C;
360 digest[3] = SHA512M_D;
361 digest[4] = SHA512M_E;
362 digest[5] = SHA512M_F;
363 digest[6] = SHA512M_G;
364 digest[7] = SHA512M_H;
366 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
368 const u32x r0 = l32_from_64 (digest[7]);
369 const u32x r1 = h32_from_64 (digest[7]);
370 const u32x r2 = l32_from_64 (digest[3]);
371 const u32x r3 = h32_from_64 (digest[3]);
373 COMPARE_M_SIMD (r0, r1, r2, r3);
377 __kernel void m01730_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)
381 __kernel void m01730_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)
385 __kernel void m01730_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)
391 const u32 lid = get_local_id (0);
397 const u32 gid = get_global_id (0);
399 if (gid >= gid_max) return;
404 pw_buf0[0] = pws[gid].i[0];
405 pw_buf0[1] = pws[gid].i[1];
406 pw_buf0[2] = pws[gid].i[2];
407 pw_buf0[3] = pws[gid].i[3];
408 pw_buf1[0] = pws[gid].i[4];
409 pw_buf1[1] = pws[gid].i[5];
410 pw_buf1[2] = pws[gid].i[6];
411 pw_buf1[3] = pws[gid].i[7];
413 const u32 pw_l_len = pws[gid].pw_len;
424 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
425 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
426 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
427 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
428 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
429 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
430 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
431 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
432 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
433 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
434 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
435 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
436 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
437 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
438 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
439 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
441 const u32 salt_len = salt_bufs[salt_pos].salt_len;
447 const u32 search[4] =
449 digests_buf[digests_offset].digest_buf[DGST_R0],
450 digests_buf[digests_offset].digest_buf[DGST_R1],
451 digests_buf[digests_offset].digest_buf[DGST_R2],
452 digests_buf[digests_offset].digest_buf[DGST_R3]
459 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
461 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
463 const u32x pw_len = pw_l_len + pw_r_len;
466 * concat password candidate
469 u32x wordl0[4] = { 0 };
470 u32x wordl1[4] = { 0 };
471 u32x wordl2[4] = { 0 };
472 u32x wordl3[4] = { 0 };
474 wordl0[0] = pw_buf0[0];
475 wordl0[1] = pw_buf0[1];
476 wordl0[2] = pw_buf0[2];
477 wordl0[3] = pw_buf0[3];
478 wordl1[0] = pw_buf1[0];
479 wordl1[1] = pw_buf1[1];
480 wordl1[2] = pw_buf1[2];
481 wordl1[3] = pw_buf1[3];
483 u32x wordr0[4] = { 0 };
484 u32x wordr1[4] = { 0 };
485 u32x wordr2[4] = { 0 };
486 u32x wordr3[4] = { 0 };
488 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
489 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
490 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
491 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
492 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
493 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
494 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
495 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
497 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
499 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
503 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
511 w0[0] = wordl0[0] | wordr0[0];
512 w0[1] = wordl0[1] | wordr0[1];
513 w0[2] = wordl0[2] | wordr0[2];
514 w0[3] = wordl0[3] | wordr0[3];
515 w1[0] = wordl1[0] | wordr1[0];
516 w1[1] = wordl1[1] | wordr1[1];
517 w1[2] = wordl1[2] | wordr1[2];
518 w1[3] = wordl1[3] | wordr1[3];
519 w2[0] = wordl2[0] | wordr2[0];
520 w2[1] = wordl2[1] | wordr2[1];
521 w2[2] = wordl2[2] | wordr2[2];
522 w2[3] = wordl2[3] | wordr2[3];
523 w3[0] = wordl3[0] | wordr3[0];
524 w3[1] = wordl3[1] | wordr3[1];
525 w3[2] = wordl3[2] | wordr3[2];
526 w3[3] = wordl3[3] | wordr3[3];
528 make_unicode (w1, w2, w3);
529 make_unicode (w0, w0, w1);
531 const u32x pw_len2 = pw_len * 2;
542 s0[0] = salt_buf0[0];
543 s0[1] = salt_buf0[1];
544 s0[2] = salt_buf0[2];
545 s0[3] = salt_buf0[3];
546 s1[0] = salt_buf1[0];
547 s1[1] = salt_buf1[1];
548 s1[2] = salt_buf1[2];
549 s1[3] = salt_buf1[3];
550 s2[0] = salt_buf2[0];
551 s2[1] = salt_buf2[1];
552 s2[2] = salt_buf2[2];
553 s2[3] = salt_buf2[3];
554 s3[0] = salt_buf3[0];
555 s3[1] = salt_buf3[1];
556 s3[2] = salt_buf3[2];
557 s3[3] = salt_buf3[3];
559 switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len2);
561 const u32x pw_salt_len = pw_len2 + salt_len;
589 w0_t[0] = swap32 (w0[0]);
590 w0_t[1] = swap32 (w0[1]);
591 w0_t[2] = swap32 (w0[2]);
592 w0_t[3] = swap32 (w0[3]);
593 w1_t[0] = swap32 (w1[0]);
594 w1_t[1] = swap32 (w1[1]);
595 w1_t[2] = swap32 (w1[2]);
596 w1_t[3] = swap32 (w1[3]);
597 w2_t[0] = swap32 (w2[0]);
598 w2_t[1] = swap32 (w2[1]);
599 w2_t[2] = swap32 (w2[2]);
600 w2_t[3] = swap32 (w2[3]);
601 w3_t[0] = swap32 (w3[0]);
602 w3_t[1] = swap32 (w3[1]);
604 w3_t[3] = pw_salt_len * 8;
608 digest[0] = SHA512M_A;
609 digest[1] = SHA512M_B;
610 digest[2] = SHA512M_C;
611 digest[3] = SHA512M_D;
612 digest[4] = SHA512M_E;
613 digest[5] = SHA512M_F;
614 digest[6] = SHA512M_G;
615 digest[7] = SHA512M_H;
617 sha512_transform (w0_t, w1_t, w2_t, w3_t, digest);
619 const u32x r0 = l32_from_64 (digest[7]);
620 const u32x r1 = h32_from_64 (digest[7]);
621 const u32x r2 = l32_from_64 (digest[3]);
622 const u32x r3 = h32_from_64 (digest[3]);
624 COMPARE_S_SIMD (r0, r1, r2, r3);
628 __kernel void m01730_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)
632 __kernel void m01730_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)