2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
24 #include "include/kernel_functions.c"
26 #include "common_nv.c"
29 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
30 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
34 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
35 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
39 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
40 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
51 __device__ static void swap (RC4_KEY *rc4_key, const u8 i, const u8 j)
56 rc4_key->S[i] = rc4_key->S[j];
60 __device__ static void rc4_init_16 (RC4_KEY *rc4_key, const u32 data[4])
65 u32 *ptr = (u32 *) rc4_key->S;
68 for (u32 i = 0; i < 64; i++)
75 for (u32 i = 0; i < 16; i++)
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 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
91 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
92 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
93 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
97 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
98 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
99 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
100 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
104 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
105 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
106 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
107 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
111 __device__ static u8 rc4_next_16 (RC4_KEY *rc4_key, u8 i, u8 j, const u32 in[4], u32 out[4])
113 for (u32 k = 0; k < 4; k++)
122 swap (rc4_key, i, j);
124 idx = rc4_key->S[i] + rc4_key->S[j];
126 xor4 |= rc4_key->S[idx] << 0;
131 swap (rc4_key, i, j);
133 idx = rc4_key->S[i] + rc4_key->S[j];
135 xor4 |= rc4_key->S[idx] << 8;
140 swap (rc4_key, i, j);
142 idx = rc4_key->S[i] + rc4_key->S[j];
144 xor4 |= rc4_key->S[idx] << 16;
149 swap (rc4_key, i, j);
151 idx = rc4_key->S[i] + rc4_key->S[j];
153 xor4 |= rc4_key->S[idx] << 24;
155 out[k] = in[k] ^ xor4;
161 __device__ static void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
185 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
186 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
187 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
188 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
189 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
190 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
191 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
192 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
193 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
194 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
195 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
196 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
197 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
198 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
199 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
200 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
202 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
203 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
204 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
205 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
206 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
207 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
208 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
209 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
210 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
211 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
212 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
213 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
214 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
215 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
216 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
217 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
219 MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
220 MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
221 MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
222 MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
223 MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
224 MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
225 MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
226 MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
227 MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
228 MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
229 MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
230 MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
231 MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
232 MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
233 MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
234 MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
236 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
237 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
238 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
239 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
240 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
241 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
242 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
243 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
244 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
245 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
246 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
247 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
248 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
249 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
250 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
251 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
259 __device__ static void gen336 (u32x digest_pre[4], u32 salt_buf[4], u32x digest[4])
266 digest_t0[0] = digest_pre[0];
267 digest_t0[1] = digest_pre[1] & 0xff;
269 digest_t1[0] = digest_pre[0] << 8;
270 digest_t1[1] = digest_pre[0] >> 24 | digest_pre[1] << 8;
272 digest_t2[0] = digest_pre[0] << 16;
273 digest_t2[1] = digest_pre[0] >> 16 | digest_pre[1] << 16;
275 digest_t3[0] = digest_pre[0] << 24;
276 digest_t3[1] = digest_pre[0] >> 8 | digest_pre[1] << 24;
283 salt_buf_t0[0] = salt_buf[0];
284 salt_buf_t0[1] = salt_buf[1];
285 salt_buf_t0[2] = salt_buf[2];
286 salt_buf_t0[3] = salt_buf[3];
288 salt_buf_t1[0] = salt_buf[0] << 8;
289 salt_buf_t1[1] = salt_buf[0] >> 24 | salt_buf[1] << 8;
290 salt_buf_t1[2] = salt_buf[1] >> 24 | salt_buf[2] << 8;
291 salt_buf_t1[3] = salt_buf[2] >> 24 | salt_buf[3] << 8;
292 salt_buf_t1[4] = salt_buf[3] >> 24;
294 salt_buf_t2[0] = salt_buf[0] << 16;
295 salt_buf_t2[1] = salt_buf[0] >> 16 | salt_buf[1] << 16;
296 salt_buf_t2[2] = salt_buf[1] >> 16 | salt_buf[2] << 16;
297 salt_buf_t2[3] = salt_buf[2] >> 16 | salt_buf[3] << 16;
298 salt_buf_t2[4] = salt_buf[3] >> 16;
300 salt_buf_t3[0] = salt_buf[0] << 24;
301 salt_buf_t3[1] = salt_buf[0] >> 8 | salt_buf[1] << 24;
302 salt_buf_t3[2] = salt_buf[1] >> 8 | salt_buf[2] << 24;
303 salt_buf_t3[3] = salt_buf[2] >> 8 | salt_buf[3] << 24;
304 salt_buf_t3[4] = salt_buf[3] >> 8;
311 // generate the 16 * 21 buffer
331 w0_t[0] = digest_t0[0];
332 w0_t[1] = digest_t0[1];
335 w0_t[1] |= salt_buf_t1[0];
336 w0_t[2] = salt_buf_t1[1];
337 w0_t[3] = salt_buf_t1[2];
338 w1_t[0] = salt_buf_t1[3];
339 w1_t[1] = salt_buf_t1[4];
342 w1_t[1] |= digest_t1[0];
343 w1_t[2] = digest_t1[1];
346 w1_t[2] |= salt_buf_t2[0];
347 w1_t[3] = salt_buf_t2[1];
348 w2_t[0] = salt_buf_t2[2];
349 w2_t[1] = salt_buf_t2[3];
350 w2_t[2] = salt_buf_t2[4];
353 w2_t[2] |= digest_t2[0];
354 w2_t[3] = digest_t2[1];
357 w2_t[3] |= salt_buf_t3[0];
358 w3_t[0] = salt_buf_t3[1];
359 w3_t[1] = salt_buf_t3[2];
360 w3_t[2] = salt_buf_t3[3];
361 w3_t[3] = salt_buf_t3[4];
365 w3_t[3] |= digest_t3[0];
367 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
387 w0_t[0] = digest_t3[1];
390 w0_t[1] = salt_buf_t0[0];
391 w0_t[2] = salt_buf_t0[1];
392 w0_t[3] = salt_buf_t0[2];
393 w1_t[0] = salt_buf_t0[3];
396 w1_t[1] = digest_t0[0];
397 w1_t[2] = digest_t0[1];
400 w1_t[2] |= salt_buf_t1[0];
401 w1_t[3] = salt_buf_t1[1];
402 w2_t[0] = salt_buf_t1[2];
403 w2_t[1] = salt_buf_t1[3];
404 w2_t[2] = salt_buf_t1[4];
407 w2_t[2] |= digest_t1[0];
408 w2_t[3] = digest_t1[1];
411 w2_t[3] |= salt_buf_t2[0];
412 w3_t[0] = salt_buf_t2[1];
413 w3_t[1] = salt_buf_t2[2];
414 w3_t[2] = salt_buf_t2[3];
415 w3_t[3] = salt_buf_t2[4];
418 w3_t[3] |= digest_t2[0];
420 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
440 w0_t[0] = digest_t2[1];
443 w0_t[0] |= salt_buf_t3[0];
444 w0_t[1] = salt_buf_t3[1];
445 w0_t[2] = salt_buf_t3[2];
446 w0_t[3] = salt_buf_t3[3];
447 w1_t[0] = salt_buf_t3[4];
450 w1_t[0] |= digest_t3[0];
451 w1_t[1] = digest_t3[1];
454 w1_t[2] = salt_buf_t0[0];
455 w1_t[3] = salt_buf_t0[1];
456 w2_t[0] = salt_buf_t0[2];
457 w2_t[1] = salt_buf_t0[3];
460 w2_t[2] = digest_t0[0];
461 w2_t[3] = digest_t0[1];
464 w2_t[3] |= salt_buf_t1[0];
465 w3_t[0] = salt_buf_t1[1];
466 w3_t[1] = salt_buf_t1[2];
467 w3_t[2] = salt_buf_t1[3];
468 w3_t[3] = salt_buf_t1[4];
471 w3_t[3] |= digest_t1[0];
473 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
493 w0_t[0] = digest_t1[1];
496 w0_t[0] |= salt_buf_t2[0];
497 w0_t[1] = salt_buf_t2[1];
498 w0_t[2] = salt_buf_t2[2];
499 w0_t[3] = salt_buf_t2[3];
500 w1_t[0] = salt_buf_t2[4];
503 w1_t[0] |= digest_t2[0];
504 w1_t[1] = digest_t2[1];
507 w1_t[1] |= salt_buf_t3[0];
508 w1_t[2] = salt_buf_t3[1];
509 w1_t[3] = salt_buf_t3[2];
510 w2_t[0] = salt_buf_t3[3];
511 w2_t[1] = salt_buf_t3[4];
514 w2_t[1] |= digest_t3[0];
515 w2_t[2] = digest_t3[1];
518 w2_t[3] = salt_buf_t0[0];
519 w3_t[0] = salt_buf_t0[1];
520 w3_t[1] = salt_buf_t0[2];
521 w3_t[2] = salt_buf_t0[3];
524 w3_t[3] = digest_t0[0];
526 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
546 w0_t[0] = digest_t0[1];
549 w0_t[0] |= salt_buf_t1[0];
550 w0_t[1] = salt_buf_t1[1];
551 w0_t[2] = salt_buf_t1[2];
552 w0_t[3] = salt_buf_t1[3];
553 w1_t[0] = salt_buf_t1[4];
556 w1_t[0] |= digest_t1[0];
557 w1_t[1] = digest_t1[1];
560 w1_t[1] |= salt_buf_t2[0];
561 w1_t[2] = salt_buf_t2[1];
562 w1_t[3] = salt_buf_t2[2];
563 w2_t[0] = salt_buf_t2[3];
564 w2_t[1] = salt_buf_t2[4];
567 w2_t[1] |= digest_t2[0];
568 w2_t[2] = digest_t2[1];
571 w2_t[2] |= salt_buf_t3[0];
572 w2_t[3] = salt_buf_t3[1];
573 w3_t[0] = salt_buf_t3[2];
574 w3_t[1] = salt_buf_t3[3];
575 w3_t[2] = salt_buf_t3[4];
578 w3_t[2] |= digest_t3[0];
579 w3_t[3] = digest_t3[1];
581 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
583 w0_t[0] = salt_buf_t0[0];
584 w0_t[1] = salt_buf_t0[1];
585 w0_t[2] = salt_buf_t0[2];
586 w0_t[3] = salt_buf_t0[3];
597 w3_t[2] = 21 * 16 * 8;
600 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
603 __device__ __constant__ comb_t c_combs[1024];
605 extern "C" __global__ void __launch_bounds__ (64, 1) m09700_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const oldoffice01_t *oldoffice01_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
611 const u32 lid = threadIdx.x;
613 __shared__ RC4_KEY rc4_keys[64];
615 RC4_KEY *rc4_key = &rc4_keys[lid];
621 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
623 if (gid >= gid_max) return;
627 wordl0[0] = pws[gid].i[ 0];
628 wordl0[1] = pws[gid].i[ 1];
629 wordl0[2] = pws[gid].i[ 2];
630 wordl0[3] = pws[gid].i[ 3];
634 wordl1[0] = pws[gid].i[ 4];
635 wordl1[1] = pws[gid].i[ 5];
636 wordl1[2] = pws[gid].i[ 6];
637 wordl1[3] = pws[gid].i[ 7];
653 const u32 pw_l_len = pws[gid].pw_len;
655 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
657 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
664 const u32 search[4] =
666 digests_buf[digests_offset].digest_buf[DGST_R0],
667 digests_buf[digests_offset].digest_buf[DGST_R1],
668 digests_buf[digests_offset].digest_buf[DGST_R2],
669 digests_buf[digests_offset].digest_buf[DGST_R3]
678 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
679 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
680 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
681 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
687 const u32 version = oldoffice01_bufs[salt_pos].version;
689 u32 encryptedVerifier[4];
691 encryptedVerifier[0] = oldoffice01_bufs[salt_pos].encryptedVerifier[0];
692 encryptedVerifier[1] = oldoffice01_bufs[salt_pos].encryptedVerifier[1];
693 encryptedVerifier[2] = oldoffice01_bufs[salt_pos].encryptedVerifier[2];
694 encryptedVerifier[3] = oldoffice01_bufs[salt_pos].encryptedVerifier[3];
700 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
702 const u32 pw_r_len = c_combs[il_pos].pw_len;
704 const u32 pw_len = pw_l_len + pw_r_len;
708 wordr0[0] = c_combs[il_pos].i[0];
709 wordr0[1] = c_combs[il_pos].i[1];
710 wordr0[2] = c_combs[il_pos].i[2];
711 wordr0[3] = c_combs[il_pos].i[3];
715 wordr1[0] = c_combs[il_pos].i[4];
716 wordr1[1] = c_combs[il_pos].i[5];
717 wordr1[2] = c_combs[il_pos].i[6];
718 wordr1[3] = c_combs[il_pos].i[7];
734 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
736 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
741 w0[0] = wordl0[0] | wordr0[0];
742 w0[1] = wordl0[1] | wordr0[1];
743 w0[2] = wordl0[2] | wordr0[2];
744 w0[3] = wordl0[3] | wordr0[3];
748 w1[0] = wordl1[0] | wordr1[0];
749 w1[1] = wordl1[1] | wordr1[1];
750 w1[2] = wordl1[2] | wordr1[2];
751 w1[3] = wordl1[3] | wordr1[3];
755 w2[0] = wordl2[0] | wordr2[0];
756 w2[1] = wordl2[1] | wordr2[1];
757 w2[2] = wordl2[2] | wordr2[2];
758 w2[3] = wordl2[3] | wordr2[3];
762 w3[0] = wordl3[0] | wordr3[0];
763 w3[1] = wordl3[1] | wordr3[1];
767 append_0x80_2 (w0, w1, pw_len);
774 make_unicode (w0, w0_t, w1_t);
775 make_unicode (w1, w2_t, w3_t);
777 w3_t[2] = pw_len * 8 * 2;
781 digest_pre[0] = MD5M_A;
782 digest_pre[1] = MD5M_B;
783 digest_pre[2] = MD5M_C;
784 digest_pre[3] = MD5M_D;
786 md5_transform (w0_t, w1_t, w2_t, w3_t, digest_pre);
788 digest_pre[0] &= 0xffffffff;
789 digest_pre[1] &= 0x000000ff;
790 digest_pre[2] &= 0x00000000;
791 digest_pre[3] &= 0x00000000;
800 gen336 (digest_pre, salt_buf, digest);
802 // now the 40 bit input for the MD5 which then will generate the RC4 key, so it's precomputable!
805 w0_t[1] = digest[1] & 0xff;
826 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
837 rc4_init_16 (rc4_key, key);
841 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
865 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
867 rc4_next_16 (rc4_key, 16, j, digest, out);
869 const u32x r0 = out[0];
870 const u32x r1 = out[1];
871 const u32x r2 = out[2];
872 const u32x r3 = out[3];
874 #include VECT_COMPARE_M
878 extern "C" __global__ void __launch_bounds__ (64, 1) m09700_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const oldoffice01_t *oldoffice01_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
882 extern "C" __global__ void __launch_bounds__ (64, 1) m09700_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const oldoffice01_t *oldoffice01_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
886 extern "C" __global__ void __launch_bounds__ (64, 1) m09700_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const oldoffice01_t *oldoffice01_bufs, u32 *d_return_buf, 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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
892 const u32 lid = threadIdx.x;
894 __shared__ RC4_KEY rc4_keys[64];
896 RC4_KEY *rc4_key = &rc4_keys[lid];
902 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
904 if (gid >= gid_max) return;
908 wordl0[0] = pws[gid].i[ 0];
909 wordl0[1] = pws[gid].i[ 1];
910 wordl0[2] = pws[gid].i[ 2];
911 wordl0[3] = pws[gid].i[ 3];
915 wordl1[0] = pws[gid].i[ 4];
916 wordl1[1] = pws[gid].i[ 5];
917 wordl1[2] = pws[gid].i[ 6];
918 wordl1[3] = pws[gid].i[ 7];
934 const u32 pw_l_len = pws[gid].pw_len;
936 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
938 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
945 const u32 search[4] =
947 digests_buf[digests_offset].digest_buf[DGST_R0],
948 digests_buf[digests_offset].digest_buf[DGST_R1],
949 digests_buf[digests_offset].digest_buf[DGST_R2],
950 digests_buf[digests_offset].digest_buf[DGST_R3]
959 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
960 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
961 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
962 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
968 const u32 version = oldoffice01_bufs[salt_pos].version;
970 u32 encryptedVerifier[4];
972 encryptedVerifier[0] = oldoffice01_bufs[salt_pos].encryptedVerifier[0];
973 encryptedVerifier[1] = oldoffice01_bufs[salt_pos].encryptedVerifier[1];
974 encryptedVerifier[2] = oldoffice01_bufs[salt_pos].encryptedVerifier[2];
975 encryptedVerifier[3] = oldoffice01_bufs[salt_pos].encryptedVerifier[3];
981 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
983 const u32 pw_r_len = c_combs[il_pos].pw_len;
985 const u32 pw_len = pw_l_len + pw_r_len;
989 wordr0[0] = c_combs[il_pos].i[0];
990 wordr0[1] = c_combs[il_pos].i[1];
991 wordr0[2] = c_combs[il_pos].i[2];
992 wordr0[3] = c_combs[il_pos].i[3];
996 wordr1[0] = c_combs[il_pos].i[4];
997 wordr1[1] = c_combs[il_pos].i[5];
998 wordr1[2] = c_combs[il_pos].i[6];
999 wordr1[3] = c_combs[il_pos].i[7];
1015 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
1017 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
1022 w0[0] = wordl0[0] | wordr0[0];
1023 w0[1] = wordl0[1] | wordr0[1];
1024 w0[2] = wordl0[2] | wordr0[2];
1025 w0[3] = wordl0[3] | wordr0[3];
1029 w1[0] = wordl1[0] | wordr1[0];
1030 w1[1] = wordl1[1] | wordr1[1];
1031 w1[2] = wordl1[2] | wordr1[2];
1032 w1[3] = wordl1[3] | wordr1[3];
1036 w2[0] = wordl2[0] | wordr2[0];
1037 w2[1] = wordl2[1] | wordr2[1];
1038 w2[2] = wordl2[2] | wordr2[2];
1039 w2[3] = wordl2[3] | wordr2[3];
1043 w3[0] = wordl3[0] | wordr3[0];
1044 w3[1] = wordl3[1] | wordr3[1];
1048 append_0x80_2 (w0, w1, pw_len);
1055 make_unicode (w0, w0_t, w1_t);
1056 make_unicode (w1, w2_t, w3_t);
1058 w3_t[2] = pw_len * 8 * 2;
1062 digest_pre[0] = MD5M_A;
1063 digest_pre[1] = MD5M_B;
1064 digest_pre[2] = MD5M_C;
1065 digest_pre[3] = MD5M_D;
1067 md5_transform (w0_t, w1_t, w2_t, w3_t, digest_pre);
1069 digest_pre[0] &= 0xffffffff;
1070 digest_pre[1] &= 0x000000ff;
1071 digest_pre[2] &= 0x00000000;
1072 digest_pre[3] &= 0x00000000;
1081 gen336 (digest_pre, salt_buf, digest);
1083 // now the 40 bit input for the MD5 which then will generate the RC4 key, so it's precomputable!
1085 w0_t[0] = digest[0];
1086 w0_t[1] = digest[1] & 0xff;
1107 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
1118 rc4_init_16 (rc4_key, key);
1122 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
1146 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
1148 rc4_next_16 (rc4_key, 16, j, digest, out);
1150 const u32x r0 = out[0];
1151 const u32x r1 = out[1];
1152 const u32x r2 = out[2];
1153 const u32x r3 = out[3];
1155 #include VECT_COMPARE_S
1159 extern "C" __global__ void __launch_bounds__ (64, 1) m09700_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const oldoffice01_t *oldoffice01_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
1163 extern "C" __global__ void __launch_bounds__ (64, 1) m09700_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, const void *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const oldoffice01_t *oldoffice01_bufs, u32 *d_return_buf, 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)