2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
28 #include "include/kernel_functions.c"
29 #include "types_amd.c"
30 #include "common_amd.c"
33 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
34 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
38 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
39 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
43 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
44 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
55 static void swap (__local RC4_KEY *rc4_key, const u8 i, const u8 j)
60 rc4_key->S[i] = rc4_key->S[j];
64 static void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4])
69 __local u32 *ptr = (__local u32 *) rc4_key->S;
72 for (u32 i = 0; i < 64; i++)
80 for (u32 i = 0; i < 16; i++)
88 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
89 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
90 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
91 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
95 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
96 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
97 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
98 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
102 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
103 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
104 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
105 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
109 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
110 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
111 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
112 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
116 static u8 rc4_next_16 (__local RC4_KEY *rc4_key, u8 i, u8 j, const u32 in[4], u32 out[4])
119 for (u32 k = 0; k < 4; k++)
128 swap (rc4_key, i, j);
130 idx = rc4_key->S[i] + rc4_key->S[j];
132 xor4 |= rc4_key->S[idx] << 0;
137 swap (rc4_key, i, j);
139 idx = rc4_key->S[i] + rc4_key->S[j];
141 xor4 |= rc4_key->S[idx] << 8;
146 swap (rc4_key, i, j);
148 idx = rc4_key->S[i] + rc4_key->S[j];
150 xor4 |= rc4_key->S[idx] << 16;
155 swap (rc4_key, i, j);
157 idx = rc4_key->S[i] + rc4_key->S[j];
159 xor4 |= rc4_key->S[idx] << 24;
161 out[k] = in[k] ^ xor4;
167 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
195 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
196 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
197 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
198 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
199 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
200 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
201 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
202 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
203 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
204 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
205 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
206 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
207 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
208 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
209 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
210 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
211 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
212 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
213 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
214 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
219 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
220 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
221 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
222 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
223 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
224 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
225 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
226 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
227 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
228 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
229 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
230 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
231 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
232 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
233 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
234 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
235 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
236 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
237 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
238 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
243 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
244 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
245 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
246 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
247 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
248 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
249 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
250 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
251 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
252 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
253 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
254 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
255 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
256 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
257 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
258 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
259 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
260 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
261 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
262 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
267 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
268 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
269 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
270 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
271 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
272 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
273 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
274 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
275 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
276 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
277 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
278 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
279 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
280 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
281 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
282 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
283 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
284 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
285 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
286 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
295 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09800_m04 (__global pw_t *pws, __global gpu_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
301 const u32 lid = get_local_id (0);
303 __local RC4_KEY rc4_keys[64];
305 __local RC4_KEY *rc4_key = &rc4_keys[lid];
311 const u32 gid = get_global_id (0);
313 if (gid >= gid_max) return;
317 wordl0[0] = pws[gid].i[ 0];
318 wordl0[1] = pws[gid].i[ 1];
319 wordl0[2] = pws[gid].i[ 2];
320 wordl0[3] = pws[gid].i[ 3];
324 wordl1[0] = pws[gid].i[ 4];
325 wordl1[1] = pws[gid].i[ 5];
326 wordl1[2] = pws[gid].i[ 6];
327 wordl1[3] = pws[gid].i[ 7];
343 const u32 pw_l_len = pws[gid].pw_len;
345 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
347 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
356 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
357 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
358 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
359 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
361 const u32 salt_len = 16;
367 const u32 version = oldoffice34_bufs[salt_pos].version;
369 u32 encryptedVerifier[4];
371 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
372 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
373 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
374 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
380 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
382 const u32 pw_r_len = combs_buf[il_pos].pw_len;
384 const u32 pw_len = pw_l_len + pw_r_len;
386 const u32 pw_salt_len = (pw_len * 2) + salt_len;
390 wordr0[0] = combs_buf[il_pos].i[0];
391 wordr0[1] = combs_buf[il_pos].i[1];
392 wordr0[2] = combs_buf[il_pos].i[2];
393 wordr0[3] = combs_buf[il_pos].i[3];
397 wordr1[0] = combs_buf[il_pos].i[4];
398 wordr1[1] = combs_buf[il_pos].i[5];
399 wordr1[2] = combs_buf[il_pos].i[6];
400 wordr1[3] = combs_buf[il_pos].i[7];
416 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
418 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
423 w0[0] = wordl0[0] | wordr0[0];
424 w0[1] = wordl0[1] | wordr0[1];
425 w0[2] = wordl0[2] | wordr0[2];
426 w0[3] = wordl0[3] | wordr0[3];
430 w1[0] = wordl1[0] | wordr1[0];
431 w1[1] = wordl1[1] | wordr1[1];
432 w1[2] = wordl1[2] | wordr1[2];
433 w1[3] = wordl1[3] | wordr1[3];
437 w2[0] = wordl2[0] | wordr2[0];
438 w2[1] = wordl2[1] | wordr2[1];
439 w2[2] = wordl2[2] | wordr2[2];
440 w2[3] = wordl2[3] | wordr2[3];
444 w3[0] = wordl3[0] | wordr3[0];
445 w3[1] = wordl3[1] | wordr3[1];
449 append_0x80_2 (w0, w1, pw_len);
456 make_unicode (w0, w0_t, w1_t);
457 make_unicode (w1, w2_t, w3_t);
459 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
461 w0_t[0] = salt_buf[0];
462 w0_t[1] = salt_buf[1];
463 w0_t[2] = salt_buf[2];
464 w0_t[3] = salt_buf[3];
465 w1_t[0] = swap_workaround (w1_t[0]);
466 w1_t[1] = swap_workaround (w1_t[1]);
467 w1_t[2] = swap_workaround (w1_t[2]);
468 w1_t[3] = swap_workaround (w1_t[3]);
469 w2_t[0] = swap_workaround (w2_t[0]);
470 w2_t[1] = swap_workaround (w2_t[1]);
471 w2_t[2] = swap_workaround (w2_t[2]);
472 w2_t[3] = swap_workaround (w2_t[3]);
473 w3_t[0] = swap_workaround (w3_t[0]);
474 w3_t[1] = swap_workaround (w3_t[1]);
476 w3_t[3] = pw_salt_len * 8;
486 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
494 w1_t[2] = 0x80000000;
503 w3_t[3] = (20 + 4) * 8;
511 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
515 key[0] = swap_workaround (digest[0]);
516 key[1] = swap_workaround (digest[1]);
517 key[2] = swap_workaround (digest[2]);
518 key[3] = swap_workaround (digest[3]);
527 rc4_init_16 (rc4_key, key);
531 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
533 w0_t[0] = swap_workaround (out[0]);
534 w0_t[1] = swap_workaround (out[1]);
535 w0_t[2] = swap_workaround (out[2]);
536 w0_t[3] = swap_workaround (out[3]);
537 w1_t[0] = 0x80000000;
556 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
558 digest[0] = swap_workaround (digest[0]);
559 digest[1] = swap_workaround (digest[1]);
560 digest[2] = swap_workaround (digest[2]);
561 digest[3] = swap_workaround (digest[3]);
563 rc4_next_16 (rc4_key, 16, j, digest, out);
565 const u32x r0 = out[0];
566 const u32x r1 = out[1];
567 const u32x r2 = out[2];
568 const u32x r3 = out[3];
570 #include VECT_COMPARE_M
574 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09800_m08 (__global pw_t *pws, __global gpu_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
578 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09800_m16 (__global pw_t *pws, __global gpu_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
582 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09800_s04 (__global pw_t *pws, __global gpu_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
588 const u32 lid = get_local_id (0);
590 __local RC4_KEY rc4_keys[64];
592 __local RC4_KEY *rc4_key = &rc4_keys[lid];
598 const u32 gid = get_global_id (0);
600 if (gid >= gid_max) return;
604 wordl0[0] = pws[gid].i[ 0];
605 wordl0[1] = pws[gid].i[ 1];
606 wordl0[2] = pws[gid].i[ 2];
607 wordl0[3] = pws[gid].i[ 3];
611 wordl1[0] = pws[gid].i[ 4];
612 wordl1[1] = pws[gid].i[ 5];
613 wordl1[2] = pws[gid].i[ 6];
614 wordl1[3] = pws[gid].i[ 7];
630 const u32 pw_l_len = pws[gid].pw_len;
632 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
634 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
641 const u32 search[4] =
643 digests_buf[digests_offset].digest_buf[DGST_R0],
644 digests_buf[digests_offset].digest_buf[DGST_R1],
645 digests_buf[digests_offset].digest_buf[DGST_R2],
646 digests_buf[digests_offset].digest_buf[DGST_R3]
655 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
656 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
657 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
658 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
660 const u32 salt_len = 16;
666 const u32 version = oldoffice34_bufs[salt_pos].version;
668 u32 encryptedVerifier[4];
670 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
671 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
672 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
673 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
679 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
681 const u32 pw_r_len = combs_buf[il_pos].pw_len;
683 const u32 pw_len = pw_l_len + pw_r_len;
685 const u32 pw_salt_len = (pw_len * 2) + salt_len;
689 wordr0[0] = combs_buf[il_pos].i[0];
690 wordr0[1] = combs_buf[il_pos].i[1];
691 wordr0[2] = combs_buf[il_pos].i[2];
692 wordr0[3] = combs_buf[il_pos].i[3];
696 wordr1[0] = combs_buf[il_pos].i[4];
697 wordr1[1] = combs_buf[il_pos].i[5];
698 wordr1[2] = combs_buf[il_pos].i[6];
699 wordr1[3] = combs_buf[il_pos].i[7];
715 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
717 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
722 w0[0] = wordl0[0] | wordr0[0];
723 w0[1] = wordl0[1] | wordr0[1];
724 w0[2] = wordl0[2] | wordr0[2];
725 w0[3] = wordl0[3] | wordr0[3];
729 w1[0] = wordl1[0] | wordr1[0];
730 w1[1] = wordl1[1] | wordr1[1];
731 w1[2] = wordl1[2] | wordr1[2];
732 w1[3] = wordl1[3] | wordr1[3];
736 w2[0] = wordl2[0] | wordr2[0];
737 w2[1] = wordl2[1] | wordr2[1];
738 w2[2] = wordl2[2] | wordr2[2];
739 w2[3] = wordl2[3] | wordr2[3];
743 w3[0] = wordl3[0] | wordr3[0];
744 w3[1] = wordl3[1] | wordr3[1];
748 append_0x80_2 (w0, w1, pw_len);
755 make_unicode (w0, w0_t, w1_t);
756 make_unicode (w1, w2_t, w3_t);
758 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
760 w0_t[0] = salt_buf[0];
761 w0_t[1] = salt_buf[1];
762 w0_t[2] = salt_buf[2];
763 w0_t[3] = salt_buf[3];
764 w1_t[0] = swap_workaround (w1_t[0]);
765 w1_t[1] = swap_workaround (w1_t[1]);
766 w1_t[2] = swap_workaround (w1_t[2]);
767 w1_t[3] = swap_workaround (w1_t[3]);
768 w2_t[0] = swap_workaround (w2_t[0]);
769 w2_t[1] = swap_workaround (w2_t[1]);
770 w2_t[2] = swap_workaround (w2_t[2]);
771 w2_t[3] = swap_workaround (w2_t[3]);
772 w3_t[0] = swap_workaround (w3_t[0]);
773 w3_t[1] = swap_workaround (w3_t[1]);
775 w3_t[3] = pw_salt_len * 8;
785 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
793 w1_t[2] = 0x80000000;
802 w3_t[3] = (20 + 4) * 8;
810 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
814 key[0] = swap_workaround (digest[0]);
815 key[1] = swap_workaround (digest[1]);
816 key[2] = swap_workaround (digest[2]);
817 key[3] = swap_workaround (digest[3]);
826 rc4_init_16 (rc4_key, key);
830 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
832 w0_t[0] = swap_workaround (out[0]);
833 w0_t[1] = swap_workaround (out[1]);
834 w0_t[2] = swap_workaround (out[2]);
835 w0_t[3] = swap_workaround (out[3]);
836 w1_t[0] = 0x80000000;
855 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
857 digest[0] = swap_workaround (digest[0]);
858 digest[1] = swap_workaround (digest[1]);
859 digest[2] = swap_workaround (digest[2]);
860 digest[3] = swap_workaround (digest[3]);
862 rc4_next_16 (rc4_key, 16, j, digest, out);
864 const u32x r0 = out[0];
865 const u32x r1 = out[1];
866 const u32x r2 = out[2];
867 const u32x r3 = out[3];
869 #include VECT_COMPARE_S
873 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09800_s08 (__global pw_t *pws, __global gpu_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
877 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09800_s16 (__global pw_t *pws, __global gpu_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 oldoffice34_t *oldoffice34_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)