2 * Author......: Jens Steube <jens.steube@gmail.com>
8 //too much register pressure
9 //#define NEW_SIMD_CODE
11 #include "include/constants.h"
12 #include "include/kernel_vendor.h"
19 #include "include/kernel_functions.c"
20 #include "OpenCL/types_ocl.c"
21 #include "OpenCL/common.c"
22 #include "include/rp_kernel.h"
23 #include "OpenCL/rp.c"
24 #include "OpenCL/simd.c"
34 static void swap (__local RC4_KEY *rc4_key, const u8 i, const u8 j)
39 rc4_key->S[i] = rc4_key->S[j];
43 static void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4])
48 __local u32 *ptr = (__local u32 *) rc4_key->S;
51 for (u32 i = 0; i < 64; i++)
58 for (u32 i = 0; i < 16; i++)
66 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
67 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
68 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
69 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
73 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
74 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
75 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
76 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
80 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
81 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
82 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
83 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
87 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
88 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
89 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
90 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
94 static u8 rc4_next_16 (__local RC4_KEY *rc4_key, u8 i, u8 j, const u32 in[4], u32 out[4])
97 for (u32 k = 0; k < 4; k++)
106 swap (rc4_key, i, j);
108 idx = rc4_key->S[i] + rc4_key->S[j];
110 xor4 |= rc4_key->S[idx] << 0;
115 swap (rc4_key, i, j);
117 idx = rc4_key->S[i] + rc4_key->S[j];
119 xor4 |= rc4_key->S[idx] << 8;
124 swap (rc4_key, i, j);
126 idx = rc4_key->S[i] + rc4_key->S[j];
128 xor4 |= rc4_key->S[idx] << 16;
133 swap (rc4_key, i, j);
135 idx = rc4_key->S[i] + rc4_key->S[j];
137 xor4 |= rc4_key->S[idx] << 24;
139 out[k] = in[k] ^ xor4;
145 static void md5_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4])
169 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
170 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
171 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
172 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
173 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
174 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
175 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
176 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
177 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
178 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
179 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
180 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
181 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
182 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
183 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
184 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
186 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
187 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
188 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
189 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
190 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
191 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
192 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
193 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
194 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
195 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
196 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
197 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
198 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
199 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
200 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
201 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
203 MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
204 MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
205 MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
206 MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
207 MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
208 MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
209 MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
210 MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
211 MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
212 MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
213 MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
214 MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
215 MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
216 MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
217 MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
218 MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
220 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
221 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
222 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
223 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
224 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
225 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
226 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
227 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
228 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
229 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
230 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
231 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
232 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
233 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
234 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
235 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
243 static void gen336 (u32 digest_pre[4], u32 salt_buf[4], u32 digest[4])
250 digest_t0[0] = digest_pre[0];
251 digest_t0[1] = digest_pre[1] & 0xff;
253 digest_t1[0] = digest_pre[0] << 8;
254 digest_t1[1] = digest_pre[0] >> 24 | digest_pre[1] << 8;
256 digest_t2[0] = digest_pre[0] << 16;
257 digest_t2[1] = digest_pre[0] >> 16 | digest_pre[1] << 16;
259 digest_t3[0] = digest_pre[0] << 24;
260 digest_t3[1] = digest_pre[0] >> 8 | digest_pre[1] << 24;
267 salt_buf_t0[0] = salt_buf[0];
268 salt_buf_t0[1] = salt_buf[1];
269 salt_buf_t0[2] = salt_buf[2];
270 salt_buf_t0[3] = salt_buf[3];
272 salt_buf_t1[0] = salt_buf[0] << 8;
273 salt_buf_t1[1] = salt_buf[0] >> 24 | salt_buf[1] << 8;
274 salt_buf_t1[2] = salt_buf[1] >> 24 | salt_buf[2] << 8;
275 salt_buf_t1[3] = salt_buf[2] >> 24 | salt_buf[3] << 8;
276 salt_buf_t1[4] = salt_buf[3] >> 24;
278 salt_buf_t2[0] = salt_buf[0] << 16;
279 salt_buf_t2[1] = salt_buf[0] >> 16 | salt_buf[1] << 16;
280 salt_buf_t2[2] = salt_buf[1] >> 16 | salt_buf[2] << 16;
281 salt_buf_t2[3] = salt_buf[2] >> 16 | salt_buf[3] << 16;
282 salt_buf_t2[4] = salt_buf[3] >> 16;
284 salt_buf_t3[0] = salt_buf[0] << 24;
285 salt_buf_t3[1] = salt_buf[0] >> 8 | salt_buf[1] << 24;
286 salt_buf_t3[2] = salt_buf[1] >> 8 | salt_buf[2] << 24;
287 salt_buf_t3[3] = salt_buf[2] >> 8 | salt_buf[3] << 24;
288 salt_buf_t3[4] = salt_buf[3] >> 8;
295 // generate the 16 * 21 buffer
315 w0_t[0] = digest_t0[0];
316 w0_t[1] = digest_t0[1];
319 w0_t[1] |= salt_buf_t1[0];
320 w0_t[2] = salt_buf_t1[1];
321 w0_t[3] = salt_buf_t1[2];
322 w1_t[0] = salt_buf_t1[3];
323 w1_t[1] = salt_buf_t1[4];
326 w1_t[1] |= digest_t1[0];
327 w1_t[2] = digest_t1[1];
330 w1_t[2] |= salt_buf_t2[0];
331 w1_t[3] = salt_buf_t2[1];
332 w2_t[0] = salt_buf_t2[2];
333 w2_t[1] = salt_buf_t2[3];
334 w2_t[2] = salt_buf_t2[4];
337 w2_t[2] |= digest_t2[0];
338 w2_t[3] = digest_t2[1];
341 w2_t[3] |= salt_buf_t3[0];
342 w3_t[0] = salt_buf_t3[1];
343 w3_t[1] = salt_buf_t3[2];
344 w3_t[2] = salt_buf_t3[3];
345 w3_t[3] = salt_buf_t3[4];
349 w3_t[3] |= digest_t3[0];
351 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
371 w0_t[0] = digest_t3[1];
374 w0_t[1] = salt_buf_t0[0];
375 w0_t[2] = salt_buf_t0[1];
376 w0_t[3] = salt_buf_t0[2];
377 w1_t[0] = salt_buf_t0[3];
380 w1_t[1] = digest_t0[0];
381 w1_t[2] = digest_t0[1];
384 w1_t[2] |= salt_buf_t1[0];
385 w1_t[3] = salt_buf_t1[1];
386 w2_t[0] = salt_buf_t1[2];
387 w2_t[1] = salt_buf_t1[3];
388 w2_t[2] = salt_buf_t1[4];
391 w2_t[2] |= digest_t1[0];
392 w2_t[3] = digest_t1[1];
395 w2_t[3] |= salt_buf_t2[0];
396 w3_t[0] = salt_buf_t2[1];
397 w3_t[1] = salt_buf_t2[2];
398 w3_t[2] = salt_buf_t2[3];
399 w3_t[3] = salt_buf_t2[4];
402 w3_t[3] |= digest_t2[0];
404 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
424 w0_t[0] = digest_t2[1];
427 w0_t[0] |= salt_buf_t3[0];
428 w0_t[1] = salt_buf_t3[1];
429 w0_t[2] = salt_buf_t3[2];
430 w0_t[3] = salt_buf_t3[3];
431 w1_t[0] = salt_buf_t3[4];
434 w1_t[0] |= digest_t3[0];
435 w1_t[1] = digest_t3[1];
438 w1_t[2] = salt_buf_t0[0];
439 w1_t[3] = salt_buf_t0[1];
440 w2_t[0] = salt_buf_t0[2];
441 w2_t[1] = salt_buf_t0[3];
444 w2_t[2] = digest_t0[0];
445 w2_t[3] = digest_t0[1];
448 w2_t[3] |= salt_buf_t1[0];
449 w3_t[0] = salt_buf_t1[1];
450 w3_t[1] = salt_buf_t1[2];
451 w3_t[2] = salt_buf_t1[3];
452 w3_t[3] = salt_buf_t1[4];
455 w3_t[3] |= digest_t1[0];
457 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
477 w0_t[0] = digest_t1[1];
480 w0_t[0] |= salt_buf_t2[0];
481 w0_t[1] = salt_buf_t2[1];
482 w0_t[2] = salt_buf_t2[2];
483 w0_t[3] = salt_buf_t2[3];
484 w1_t[0] = salt_buf_t2[4];
487 w1_t[0] |= digest_t2[0];
488 w1_t[1] = digest_t2[1];
491 w1_t[1] |= salt_buf_t3[0];
492 w1_t[2] = salt_buf_t3[1];
493 w1_t[3] = salt_buf_t3[2];
494 w2_t[0] = salt_buf_t3[3];
495 w2_t[1] = salt_buf_t3[4];
498 w2_t[1] |= digest_t3[0];
499 w2_t[2] = digest_t3[1];
502 w2_t[3] = salt_buf_t0[0];
503 w3_t[0] = salt_buf_t0[1];
504 w3_t[1] = salt_buf_t0[2];
505 w3_t[2] = salt_buf_t0[3];
508 w3_t[3] = digest_t0[0];
510 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
530 w0_t[0] = digest_t0[1];
533 w0_t[0] |= salt_buf_t1[0];
534 w0_t[1] = salt_buf_t1[1];
535 w0_t[2] = salt_buf_t1[2];
536 w0_t[3] = salt_buf_t1[3];
537 w1_t[0] = salt_buf_t1[4];
540 w1_t[0] |= digest_t1[0];
541 w1_t[1] = digest_t1[1];
544 w1_t[1] |= salt_buf_t2[0];
545 w1_t[2] = salt_buf_t2[1];
546 w1_t[3] = salt_buf_t2[2];
547 w2_t[0] = salt_buf_t2[3];
548 w2_t[1] = salt_buf_t2[4];
551 w2_t[1] |= digest_t2[0];
552 w2_t[2] = digest_t2[1];
555 w2_t[2] |= salt_buf_t3[0];
556 w2_t[3] = salt_buf_t3[1];
557 w3_t[0] = salt_buf_t3[2];
558 w3_t[1] = salt_buf_t3[3];
559 w3_t[2] = salt_buf_t3[4];
562 w3_t[2] |= digest_t3[0];
563 w3_t[3] = digest_t3[1];
565 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
567 w0_t[0] = salt_buf_t0[0];
568 w0_t[1] = salt_buf_t0[1];
569 w0_t[2] = salt_buf_t0[2];
570 w0_t[3] = salt_buf_t0[3];
581 w3_t[2] = 21 * 16 * 8;
584 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
587 __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_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)
593 const u32 lid = get_local_id (0);
599 const u32 gid = get_global_id (0);
601 if (gid >= gid_max) return;
606 pw_buf0[0] = pws[gid].i[ 0];
607 pw_buf0[1] = pws[gid].i[ 1];
608 pw_buf0[2] = pws[gid].i[ 2];
609 pw_buf0[3] = pws[gid].i[ 3];
610 pw_buf1[0] = pws[gid].i[ 4];
611 pw_buf1[1] = pws[gid].i[ 5];
612 pw_buf1[2] = pws[gid].i[ 6];
613 pw_buf1[3] = pws[gid].i[ 7];
615 const u32 pw_len = pws[gid].pw_len;
621 __local RC4_KEY rc4_keys[64];
623 __local RC4_KEY *rc4_key = &rc4_keys[lid];
631 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
632 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
633 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
634 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
640 const u32 version = oldoffice01_bufs[salt_pos].version;
642 u32 encryptedVerifier[4];
644 encryptedVerifier[0] = oldoffice01_bufs[salt_pos].encryptedVerifier[0];
645 encryptedVerifier[1] = oldoffice01_bufs[salt_pos].encryptedVerifier[1];
646 encryptedVerifier[2] = oldoffice01_bufs[salt_pos].encryptedVerifier[2];
647 encryptedVerifier[3] = oldoffice01_bufs[salt_pos].encryptedVerifier[3];
653 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
660 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
662 append_0x80_2x4_VV (w0, w1, out_len);
668 make_unicode (w1, w2, w3);
669 make_unicode (w0, w0, w1);
671 w3[2] = out_len * 8 * 2;
676 digest_pre[0] = MD5M_A;
677 digest_pre[1] = MD5M_B;
678 digest_pre[2] = MD5M_C;
679 digest_pre[3] = MD5M_D;
681 md5_transform (w0, w1, w2, w3, digest_pre);
683 digest_pre[0] &= 0xffffffff;
684 digest_pre[1] &= 0x000000ff;
685 digest_pre[2] &= 0x00000000;
686 digest_pre[3] &= 0x00000000;
695 gen336 (digest_pre, salt_buf, digest);
697 // now the 40 bit input for the MD5 which then will generate the RC4 key, so it's precomputable!
700 w0[1] = digest[1] & 0xff;
721 md5_transform (w0, w1, w2, w3, digest);
732 rc4_init_16 (rc4_key, key);
736 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
760 md5_transform (w0, w1, w2, w3, digest);
762 rc4_next_16 (rc4_key, 16, j, digest, out);
764 COMPARE_M_SIMD (out[0], out[1], out[2], out[3]);
768 __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_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)
772 __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_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)
776 __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_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)
782 const u32 lid = get_local_id (0);
788 const u32 gid = get_global_id (0);
790 if (gid >= gid_max) return;
795 pw_buf0[0] = pws[gid].i[ 0];
796 pw_buf0[1] = pws[gid].i[ 1];
797 pw_buf0[2] = pws[gid].i[ 2];
798 pw_buf0[3] = pws[gid].i[ 3];
799 pw_buf1[0] = pws[gid].i[ 4];
800 pw_buf1[1] = pws[gid].i[ 5];
801 pw_buf1[2] = pws[gid].i[ 6];
802 pw_buf1[3] = pws[gid].i[ 7];
804 const u32 pw_len = pws[gid].pw_len;
810 __local RC4_KEY rc4_keys[64];
812 __local RC4_KEY *rc4_key = &rc4_keys[lid];
820 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
821 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
822 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
823 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
829 const u32 version = oldoffice01_bufs[salt_pos].version;
831 u32 encryptedVerifier[4];
833 encryptedVerifier[0] = oldoffice01_bufs[salt_pos].encryptedVerifier[0];
834 encryptedVerifier[1] = oldoffice01_bufs[salt_pos].encryptedVerifier[1];
835 encryptedVerifier[2] = oldoffice01_bufs[salt_pos].encryptedVerifier[2];
836 encryptedVerifier[3] = oldoffice01_bufs[salt_pos].encryptedVerifier[3];
842 const u32 search[4] =
844 digests_buf[digests_offset].digest_buf[DGST_R0],
845 digests_buf[digests_offset].digest_buf[DGST_R1],
846 digests_buf[digests_offset].digest_buf[DGST_R2],
847 digests_buf[digests_offset].digest_buf[DGST_R3]
854 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
861 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
863 append_0x80_2x4_VV (w0, w1, out_len);
869 make_unicode (w1, w2, w3);
870 make_unicode (w0, w0, w1);
872 w3[2] = out_len * 8 * 2;
877 digest_pre[0] = MD5M_A;
878 digest_pre[1] = MD5M_B;
879 digest_pre[2] = MD5M_C;
880 digest_pre[3] = MD5M_D;
882 md5_transform (w0, w1, w2, w3, digest_pre);
884 digest_pre[0] &= 0xffffffff;
885 digest_pre[1] &= 0x000000ff;
886 digest_pre[2] &= 0x00000000;
887 digest_pre[3] &= 0x00000000;
896 gen336 (digest_pre, salt_buf, digest);
898 // now the 40 bit input for the MD5 which then will generate the RC4 key, so it's precomputable!
901 w0[1] = digest[1] & 0xff;
922 md5_transform (w0, w1, w2, w3, digest);
933 rc4_init_16 (rc4_key, key);
937 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
961 md5_transform (w0, w1, w2, w3, digest);
963 rc4_next_16 (rc4_key, 16, j, digest, out);
965 COMPARE_S_SIMD (out[0], out[1], out[2], out[3]);
969 __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_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)
973 __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_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)