2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
16 #include "include/kernel_functions.c"
17 #include "types_ocl.c"
20 #define COMPARE_S "check_single_comp4.c"
21 #define COMPARE_M "check_multi_comp4.c"
31 static void swap (__local RC4_KEY *rc4_key, const u8 i, const u8 j)
36 rc4_key->S[i] = rc4_key->S[j];
40 static void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4])
45 __local u32 *ptr = (__local u32 *) rc4_key->S;
48 for (u32 i = 0; i < 64; i++)
56 for (u32 i = 0; i < 16; i++)
64 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
65 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
66 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
67 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
71 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
72 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
73 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
74 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
78 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
79 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
80 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
81 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
85 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j); idx++;
86 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j); idx++;
87 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j); idx++;
88 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j); idx++;
92 static 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 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
171 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
172 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
173 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
174 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
175 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
176 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
177 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
178 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
179 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
180 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
181 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
182 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
183 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
184 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
185 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
186 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
187 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
188 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
189 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
190 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
195 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
196 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
197 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
198 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
199 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
200 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
201 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
202 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
203 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
204 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
205 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
206 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
207 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
208 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
209 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
210 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
211 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
212 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
213 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
214 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
219 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
220 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
221 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
222 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
223 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
224 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
225 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
226 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
227 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
228 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
229 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
230 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
231 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
232 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
233 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
234 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
235 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
236 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
237 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
238 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
243 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
244 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
245 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
246 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
247 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
248 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
249 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
250 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
251 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
252 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
253 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
254 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
255 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
256 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
257 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
258 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
259 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
260 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
261 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
262 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
271 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09810_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)
277 const u32 lid = get_local_id (0);
279 __local RC4_KEY rc4_keys[64];
281 __local RC4_KEY *rc4_key = &rc4_keys[lid];
287 const u32 gid = get_global_id (0);
289 if (gid >= gid_max) return;
293 wordl0[0] = pws[gid].i[ 0];
294 wordl0[1] = pws[gid].i[ 1];
295 wordl0[2] = pws[gid].i[ 2];
296 wordl0[3] = pws[gid].i[ 3];
300 wordl1[0] = pws[gid].i[ 4];
301 wordl1[1] = pws[gid].i[ 5];
302 wordl1[2] = pws[gid].i[ 6];
303 wordl1[3] = pws[gid].i[ 7];
319 const u32 pw_l_len = pws[gid].pw_len;
321 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
323 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
330 const u32 version = oldoffice34_bufs[salt_pos].version;
332 u32 encryptedVerifier[4];
334 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
335 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
336 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
337 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
343 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
345 const u32 pw_r_len = combs_buf[il_pos].pw_len;
347 const u32 pw_len = pw_l_len + pw_r_len;
351 wordr0[0] = combs_buf[il_pos].i[0];
352 wordr0[1] = combs_buf[il_pos].i[1];
353 wordr0[2] = combs_buf[il_pos].i[2];
354 wordr0[3] = combs_buf[il_pos].i[3];
358 wordr1[0] = combs_buf[il_pos].i[4];
359 wordr1[1] = combs_buf[il_pos].i[5];
360 wordr1[2] = combs_buf[il_pos].i[6];
361 wordr1[3] = combs_buf[il_pos].i[7];
377 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
379 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
386 key[0] = wordl0[0] | wordr0[0];
387 key[1] = (wordl0[1] | wordr0[1]) & 0xff;
391 rc4_init_16 (rc4_key, key);
395 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
402 w0_t[0] = swap_workaround (out[0]);
403 w0_t[1] = swap_workaround (out[1]);
404 w0_t[2] = swap_workaround (out[2]);
405 w0_t[3] = swap_workaround (out[3]);
406 w1_t[0] = 0x80000000;
427 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
429 digest[0] = swap_workaround (digest[0]);
430 digest[1] = swap_workaround (digest[1]);
431 digest[2] = swap_workaround (digest[2]);
432 digest[3] = swap_workaround (digest[3]);
434 rc4_next_16 (rc4_key, 16, j, digest, out);
436 const u32 r0 = out[0];
437 const u32 r1 = out[1];
438 const u32 r2 = out[2];
439 const u32 r3 = out[3];
445 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09810_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)
449 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09810_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)
453 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09810_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)
459 const u32 lid = get_local_id (0);
461 __local RC4_KEY rc4_keys[64];
463 __local RC4_KEY *rc4_key = &rc4_keys[lid];
469 const u32 gid = get_global_id (0);
471 if (gid >= gid_max) return;
475 wordl0[0] = pws[gid].i[ 0];
476 wordl0[1] = pws[gid].i[ 1];
477 wordl0[2] = pws[gid].i[ 2];
478 wordl0[3] = pws[gid].i[ 3];
482 wordl1[0] = pws[gid].i[ 4];
483 wordl1[1] = pws[gid].i[ 5];
484 wordl1[2] = pws[gid].i[ 6];
485 wordl1[3] = pws[gid].i[ 7];
501 const u32 pw_l_len = pws[gid].pw_len;
503 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
505 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
512 const u32 search[4] =
514 digests_buf[digests_offset].digest_buf[DGST_R0],
515 digests_buf[digests_offset].digest_buf[DGST_R1],
516 digests_buf[digests_offset].digest_buf[DGST_R2],
517 digests_buf[digests_offset].digest_buf[DGST_R3]
524 const u32 version = oldoffice34_bufs[salt_pos].version;
526 u32 encryptedVerifier[4];
528 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
529 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
530 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
531 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
537 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
539 const u32 pw_r_len = combs_buf[il_pos].pw_len;
541 const u32 pw_len = pw_l_len + pw_r_len;
545 wordr0[0] = combs_buf[il_pos].i[0];
546 wordr0[1] = combs_buf[il_pos].i[1];
547 wordr0[2] = combs_buf[il_pos].i[2];
548 wordr0[3] = combs_buf[il_pos].i[3];
552 wordr1[0] = combs_buf[il_pos].i[4];
553 wordr1[1] = combs_buf[il_pos].i[5];
554 wordr1[2] = combs_buf[il_pos].i[6];
555 wordr1[3] = combs_buf[il_pos].i[7];
571 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
573 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
580 key[0] = wordl0[0] | wordr0[0];
581 key[1] = (wordl0[1] | wordr0[1]) & 0xff;
585 rc4_init_16 (rc4_key, key);
589 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
596 w0_t[0] = swap_workaround (out[0]);
597 w0_t[1] = swap_workaround (out[1]);
598 w0_t[2] = swap_workaround (out[2]);
599 w0_t[3] = swap_workaround (out[3]);
600 w1_t[0] = 0x80000000;
621 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
623 digest[0] = swap_workaround (digest[0]);
624 digest[1] = swap_workaround (digest[1]);
625 digest[2] = swap_workaround (digest[2]);
626 digest[3] = swap_workaround (digest[3]);
628 rc4_next_16 (rc4_key, 16, j, digest, out);
630 const u32 r0 = out[0];
631 const u32 r1 = out[1];
632 const u32 r2 = out[2];
633 const u32 r3 = out[3];
639 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09810_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)
643 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09810_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)