2 * Author......: Jens Steube <jens.steube@gmail.com>
8 //too much register pressure
9 //#define NEW_SIMD_CODE
11 #include "inc_vendor.cl"
12 #include "inc_hash_constants.h"
13 #include "inc_hash_functions.cl"
14 #include "inc_types.cl"
15 #include "inc_common.cl"
18 #include "inc_simd.cl"
28 void swap (__local RC4_KEY *rc4_key, const u8 i, const u8 j)
33 rc4_key->S[i] = rc4_key->S[j];
37 void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4])
42 __local u32 *ptr = (__local u32 *) rc4_key->S;
47 for (u32 i = 0; i < 64; i++)
54 for (u32 i = 0; i < 16; i++)
62 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
63 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
64 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
65 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
69 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
70 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
71 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
72 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
76 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
77 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
78 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
79 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
83 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
84 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
85 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
86 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
90 u8 rc4_next_16 (__local RC4_KEY *rc4_key, u8 i, u8 j, const u32 in[4], u32 out[4])
95 for (u32 k = 0; k < 4; k++)
104 swap (rc4_key, i, j);
106 idx = rc4_key->S[i] + rc4_key->S[j];
108 xor4 |= rc4_key->S[idx] << 0;
113 swap (rc4_key, i, j);
115 idx = rc4_key->S[i] + rc4_key->S[j];
117 xor4 |= rc4_key->S[idx] << 8;
122 swap (rc4_key, i, j);
124 idx = rc4_key->S[i] + rc4_key->S[j];
126 xor4 |= rc4_key->S[idx] << 16;
131 swap (rc4_key, i, j);
133 idx = rc4_key->S[i] + rc4_key->S[j];
135 xor4 |= rc4_key->S[idx] << 24;
137 out[k] = in[k] ^ xor4;
143 void md5_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4])
167 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
168 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
169 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
170 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
171 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
172 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
173 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
174 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
175 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
176 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
177 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
178 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
179 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
180 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
181 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
182 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
184 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
185 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
186 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
187 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
188 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
189 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
190 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
191 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
192 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
193 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
194 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
195 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
196 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
197 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
198 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
199 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
201 MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
202 MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
203 MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
204 MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
205 MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
206 MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
207 MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
208 MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
209 MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
210 MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
211 MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
212 MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
213 MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
214 MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
215 MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
216 MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
218 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
219 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
220 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
221 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
222 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
223 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
224 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
225 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
226 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
227 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
228 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
229 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
230 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
231 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
232 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
233 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
241 void gen336 (u32 digest_pre[4], u32 salt_buf[4], u32 digest[4])
248 digest_t0[0] = digest_pre[0];
249 digest_t0[1] = digest_pre[1] & 0xff;
251 digest_t1[0] = digest_pre[0] << 8;
252 digest_t1[1] = digest_pre[0] >> 24 | digest_pre[1] << 8;
254 digest_t2[0] = digest_pre[0] << 16;
255 digest_t2[1] = digest_pre[0] >> 16 | digest_pre[1] << 16;
257 digest_t3[0] = digest_pre[0] << 24;
258 digest_t3[1] = digest_pre[0] >> 8 | digest_pre[1] << 24;
265 salt_buf_t0[0] = salt_buf[0];
266 salt_buf_t0[1] = salt_buf[1];
267 salt_buf_t0[2] = salt_buf[2];
268 salt_buf_t0[3] = salt_buf[3];
270 salt_buf_t1[0] = salt_buf[0] << 8;
271 salt_buf_t1[1] = salt_buf[0] >> 24 | salt_buf[1] << 8;
272 salt_buf_t1[2] = salt_buf[1] >> 24 | salt_buf[2] << 8;
273 salt_buf_t1[3] = salt_buf[2] >> 24 | salt_buf[3] << 8;
274 salt_buf_t1[4] = salt_buf[3] >> 24;
276 salt_buf_t2[0] = salt_buf[0] << 16;
277 salt_buf_t2[1] = salt_buf[0] >> 16 | salt_buf[1] << 16;
278 salt_buf_t2[2] = salt_buf[1] >> 16 | salt_buf[2] << 16;
279 salt_buf_t2[3] = salt_buf[2] >> 16 | salt_buf[3] << 16;
280 salt_buf_t2[4] = salt_buf[3] >> 16;
282 salt_buf_t3[0] = salt_buf[0] << 24;
283 salt_buf_t3[1] = salt_buf[0] >> 8 | salt_buf[1] << 24;
284 salt_buf_t3[2] = salt_buf[1] >> 8 | salt_buf[2] << 24;
285 salt_buf_t3[3] = salt_buf[2] >> 8 | salt_buf[3] << 24;
286 salt_buf_t3[4] = salt_buf[3] >> 8;
293 // generate the 16 * 21 buffer
313 w0_t[0] = digest_t0[0];
314 w0_t[1] = digest_t0[1];
317 w0_t[1] |= salt_buf_t1[0];
318 w0_t[2] = salt_buf_t1[1];
319 w0_t[3] = salt_buf_t1[2];
320 w1_t[0] = salt_buf_t1[3];
321 w1_t[1] = salt_buf_t1[4];
324 w1_t[1] |= digest_t1[0];
325 w1_t[2] = digest_t1[1];
328 w1_t[2] |= salt_buf_t2[0];
329 w1_t[3] = salt_buf_t2[1];
330 w2_t[0] = salt_buf_t2[2];
331 w2_t[1] = salt_buf_t2[3];
332 w2_t[2] = salt_buf_t2[4];
335 w2_t[2] |= digest_t2[0];
336 w2_t[3] = digest_t2[1];
339 w2_t[3] |= salt_buf_t3[0];
340 w3_t[0] = salt_buf_t3[1];
341 w3_t[1] = salt_buf_t3[2];
342 w3_t[2] = salt_buf_t3[3];
343 w3_t[3] = salt_buf_t3[4];
347 w3_t[3] |= digest_t3[0];
349 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
369 w0_t[0] = digest_t3[1];
372 w0_t[1] = salt_buf_t0[0];
373 w0_t[2] = salt_buf_t0[1];
374 w0_t[3] = salt_buf_t0[2];
375 w1_t[0] = salt_buf_t0[3];
378 w1_t[1] = digest_t0[0];
379 w1_t[2] = digest_t0[1];
382 w1_t[2] |= salt_buf_t1[0];
383 w1_t[3] = salt_buf_t1[1];
384 w2_t[0] = salt_buf_t1[2];
385 w2_t[1] = salt_buf_t1[3];
386 w2_t[2] = salt_buf_t1[4];
389 w2_t[2] |= digest_t1[0];
390 w2_t[3] = digest_t1[1];
393 w2_t[3] |= salt_buf_t2[0];
394 w3_t[0] = salt_buf_t2[1];
395 w3_t[1] = salt_buf_t2[2];
396 w3_t[2] = salt_buf_t2[3];
397 w3_t[3] = salt_buf_t2[4];
400 w3_t[3] |= digest_t2[0];
402 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
422 w0_t[0] = digest_t2[1];
425 w0_t[0] |= salt_buf_t3[0];
426 w0_t[1] = salt_buf_t3[1];
427 w0_t[2] = salt_buf_t3[2];
428 w0_t[3] = salt_buf_t3[3];
429 w1_t[0] = salt_buf_t3[4];
432 w1_t[0] |= digest_t3[0];
433 w1_t[1] = digest_t3[1];
436 w1_t[2] = salt_buf_t0[0];
437 w1_t[3] = salt_buf_t0[1];
438 w2_t[0] = salt_buf_t0[2];
439 w2_t[1] = salt_buf_t0[3];
442 w2_t[2] = digest_t0[0];
443 w2_t[3] = digest_t0[1];
446 w2_t[3] |= salt_buf_t1[0];
447 w3_t[0] = salt_buf_t1[1];
448 w3_t[1] = salt_buf_t1[2];
449 w3_t[2] = salt_buf_t1[3];
450 w3_t[3] = salt_buf_t1[4];
453 w3_t[3] |= digest_t1[0];
455 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
475 w0_t[0] = digest_t1[1];
478 w0_t[0] |= salt_buf_t2[0];
479 w0_t[1] = salt_buf_t2[1];
480 w0_t[2] = salt_buf_t2[2];
481 w0_t[3] = salt_buf_t2[3];
482 w1_t[0] = salt_buf_t2[4];
485 w1_t[0] |= digest_t2[0];
486 w1_t[1] = digest_t2[1];
489 w1_t[1] |= salt_buf_t3[0];
490 w1_t[2] = salt_buf_t3[1];
491 w1_t[3] = salt_buf_t3[2];
492 w2_t[0] = salt_buf_t3[3];
493 w2_t[1] = salt_buf_t3[4];
496 w2_t[1] |= digest_t3[0];
497 w2_t[2] = digest_t3[1];
500 w2_t[3] = salt_buf_t0[0];
501 w3_t[0] = salt_buf_t0[1];
502 w3_t[1] = salt_buf_t0[2];
503 w3_t[2] = salt_buf_t0[3];
506 w3_t[3] = digest_t0[0];
508 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
528 w0_t[0] = digest_t0[1];
531 w0_t[0] |= salt_buf_t1[0];
532 w0_t[1] = salt_buf_t1[1];
533 w0_t[2] = salt_buf_t1[2];
534 w0_t[3] = salt_buf_t1[3];
535 w1_t[0] = salt_buf_t1[4];
538 w1_t[0] |= digest_t1[0];
539 w1_t[1] = digest_t1[1];
542 w1_t[1] |= salt_buf_t2[0];
543 w1_t[2] = salt_buf_t2[1];
544 w1_t[3] = salt_buf_t2[2];
545 w2_t[0] = salt_buf_t2[3];
546 w2_t[1] = salt_buf_t2[4];
549 w2_t[1] |= digest_t2[0];
550 w2_t[2] = digest_t2[1];
553 w2_t[2] |= salt_buf_t3[0];
554 w2_t[3] = salt_buf_t3[1];
555 w3_t[0] = salt_buf_t3[2];
556 w3_t[1] = salt_buf_t3[3];
557 w3_t[2] = salt_buf_t3[4];
560 w3_t[2] |= digest_t3[0];
561 w3_t[3] = digest_t3[1];
563 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
565 w0_t[0] = salt_buf_t0[0];
566 w0_t[1] = salt_buf_t0[1];
567 w0_t[2] = salt_buf_t0[2];
568 w0_t[3] = salt_buf_t0[3];
579 w3_t[2] = 21 * 16 * 8;
582 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
585 __kernel void m09700_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 oldoffice01_t *oldoffice01_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)
591 const u32 lid = get_local_id (0);
597 const u32 gid = get_global_id (0);
599 if (gid >= gid_max) return;
604 pw_buf0[0] = pws[gid].i[ 0];
605 pw_buf0[1] = pws[gid].i[ 1];
606 pw_buf0[2] = pws[gid].i[ 2];
607 pw_buf0[3] = pws[gid].i[ 3];
608 pw_buf1[0] = pws[gid].i[ 4];
609 pw_buf1[1] = pws[gid].i[ 5];
610 pw_buf1[2] = pws[gid].i[ 6];
611 pw_buf1[3] = pws[gid].i[ 7];
613 const u32 pw_len = pws[gid].pw_len;
619 __local RC4_KEY rc4_keys[64];
621 __local RC4_KEY *rc4_key = &rc4_keys[lid];
629 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
630 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
631 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
632 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
638 const u32 version = oldoffice01_bufs[salt_pos].version;
640 u32 encryptedVerifier[4];
642 encryptedVerifier[0] = oldoffice01_bufs[salt_pos].encryptedVerifier[0];
643 encryptedVerifier[1] = oldoffice01_bufs[salt_pos].encryptedVerifier[1];
644 encryptedVerifier[2] = oldoffice01_bufs[salt_pos].encryptedVerifier[2];
645 encryptedVerifier[3] = oldoffice01_bufs[salt_pos].encryptedVerifier[3];
651 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
658 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
660 append_0x80_2x4_VV (w0, w1, out_len);
666 make_unicode (w1, w2, w3);
667 make_unicode (w0, w0, w1);
669 w3[2] = out_len * 8 * 2;
674 digest_pre[0] = MD5M_A;
675 digest_pre[1] = MD5M_B;
676 digest_pre[2] = MD5M_C;
677 digest_pre[3] = MD5M_D;
679 md5_transform (w0, w1, w2, w3, digest_pre);
681 digest_pre[0] &= 0xffffffff;
682 digest_pre[1] &= 0x000000ff;
683 digest_pre[2] &= 0x00000000;
684 digest_pre[3] &= 0x00000000;
693 gen336 (digest_pre, salt_buf, digest);
695 // now the 40 bit input for the MD5 which then will generate the RC4 key, so it's precomputable!
698 w0[1] = digest[1] & 0xff;
719 md5_transform (w0, w1, w2, w3, digest);
730 rc4_init_16 (rc4_key, key);
734 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
758 md5_transform (w0, w1, w2, w3, digest);
760 rc4_next_16 (rc4_key, 16, j, digest, out);
762 COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
766 __kernel void m09700_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 oldoffice01_t *oldoffice01_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)
770 __kernel void m09700_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 oldoffice01_t *oldoffice01_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)
774 __kernel void m09700_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 oldoffice01_t *oldoffice01_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)
780 const u32 lid = get_local_id (0);
786 const u32 gid = get_global_id (0);
788 if (gid >= gid_max) return;
793 pw_buf0[0] = pws[gid].i[ 0];
794 pw_buf0[1] = pws[gid].i[ 1];
795 pw_buf0[2] = pws[gid].i[ 2];
796 pw_buf0[3] = pws[gid].i[ 3];
797 pw_buf1[0] = pws[gid].i[ 4];
798 pw_buf1[1] = pws[gid].i[ 5];
799 pw_buf1[2] = pws[gid].i[ 6];
800 pw_buf1[3] = pws[gid].i[ 7];
802 const u32 pw_len = pws[gid].pw_len;
808 __local RC4_KEY rc4_keys[64];
810 __local RC4_KEY *rc4_key = &rc4_keys[lid];
818 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
819 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
820 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
821 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
827 const u32 version = oldoffice01_bufs[salt_pos].version;
829 u32 encryptedVerifier[4];
831 encryptedVerifier[0] = oldoffice01_bufs[salt_pos].encryptedVerifier[0];
832 encryptedVerifier[1] = oldoffice01_bufs[salt_pos].encryptedVerifier[1];
833 encryptedVerifier[2] = oldoffice01_bufs[salt_pos].encryptedVerifier[2];
834 encryptedVerifier[3] = oldoffice01_bufs[salt_pos].encryptedVerifier[3];
840 const u32 search[4] =
842 digests_buf[digests_offset].digest_buf[DGST_R0],
843 digests_buf[digests_offset].digest_buf[DGST_R1],
844 digests_buf[digests_offset].digest_buf[DGST_R2],
845 digests_buf[digests_offset].digest_buf[DGST_R3]
852 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
859 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
861 append_0x80_2x4_VV (w0, w1, out_len);
867 make_unicode (w1, w2, w3);
868 make_unicode (w0, w0, w1);
870 w3[2] = out_len * 8 * 2;
875 digest_pre[0] = MD5M_A;
876 digest_pre[1] = MD5M_B;
877 digest_pre[2] = MD5M_C;
878 digest_pre[3] = MD5M_D;
880 md5_transform (w0, w1, w2, w3, digest_pre);
882 digest_pre[0] &= 0xffffffff;
883 digest_pre[1] &= 0x000000ff;
884 digest_pre[2] &= 0x00000000;
885 digest_pre[3] &= 0x00000000;
894 gen336 (digest_pre, salt_buf, digest);
896 // now the 40 bit input for the MD5 which then will generate the RC4 key, so it's precomputable!
899 w0[1] = digest[1] & 0xff;
920 md5_transform (w0, w1, w2, w3, digest);
931 rc4_init_16 (rc4_key, key);
935 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
959 md5_transform (w0, w1, w2, w3, digest);
961 rc4_next_16 (rc4_key, 16, j, digest, out);
963 COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
967 __kernel void m09700_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 oldoffice01_t *oldoffice01_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)
971 __kernel void m09700_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 oldoffice01_t *oldoffice01_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)