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 u32 i, const u32 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 & 0xff); idx++;
84 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
85 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
86 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
90 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
91 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
92 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
93 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
97 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
98 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
99 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
100 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
104 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
105 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
106 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
107 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); 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 sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
189 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
190 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
191 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
192 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
193 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
194 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
195 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
196 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
197 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
198 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
199 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
200 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
201 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
202 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
203 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
204 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
205 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
206 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
207 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
208 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
213 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
214 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
215 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
216 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
217 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
218 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
219 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
220 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
221 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
222 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
223 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
224 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
225 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
226 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
227 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
228 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
229 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
230 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
231 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
232 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
237 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
238 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
239 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
240 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
241 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
242 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
243 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
244 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
245 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
246 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
247 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
248 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
249 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
250 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
251 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
252 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
253 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
254 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
255 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
256 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
261 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
262 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
263 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
264 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
265 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
266 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
267 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
268 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
269 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
270 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
271 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
272 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
273 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
274 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
275 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
276 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
277 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
278 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
279 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
280 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
289 __device__ __constant__ comb_t c_combs[1024];
291 extern "C" __global__ void __launch_bounds__ (64, 1) m09810_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 oldoffice34_t *oldoffice34_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)
297 const u32 lid = threadIdx.x;
299 __shared__ RC4_KEY rc4_keys[64];
301 RC4_KEY *rc4_key = &rc4_keys[lid];
307 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
309 if (gid >= gid_max) return;
313 wordl0[0] = pws[gid].i[ 0];
314 wordl0[1] = pws[gid].i[ 1];
315 wordl0[2] = pws[gid].i[ 2];
316 wordl0[3] = pws[gid].i[ 3];
320 wordl1[0] = pws[gid].i[ 4];
321 wordl1[1] = pws[gid].i[ 5];
322 wordl1[2] = pws[gid].i[ 6];
323 wordl1[3] = pws[gid].i[ 7];
339 const u32 pw_l_len = pws[gid].pw_len;
341 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
343 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
350 const u32 version = oldoffice34_bufs[salt_pos].version;
352 u32 encryptedVerifier[4];
354 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
355 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
356 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
357 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
363 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
367 wordr0[0] = c_combs[il_pos].i[0];
368 wordr0[1] = c_combs[il_pos].i[1];
369 wordr0[2] = c_combs[il_pos].i[2];
370 wordr0[3] = c_combs[il_pos].i[3];
374 wordr1[0] = c_combs[il_pos].i[4];
375 wordr1[1] = c_combs[il_pos].i[5];
376 wordr1[2] = c_combs[il_pos].i[6];
377 wordr1[3] = c_combs[il_pos].i[7];
393 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
395 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
402 key[0] = wordl0[0] | wordr0[0];
403 key[1] = (wordl0[1] | wordr0[1]) & 0xff;
407 rc4_init_16 (rc4_key, key);
411 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
418 w0_t[0] = swap_workaround (out[0]);
419 w0_t[1] = swap_workaround (out[1]);
420 w0_t[2] = swap_workaround (out[2]);
421 w0_t[3] = swap_workaround (out[3]);
422 w1_t[0] = 0x80000000;
443 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
445 digest[0] = swap_workaround (digest[0]);
446 digest[1] = swap_workaround (digest[1]);
447 digest[2] = swap_workaround (digest[2]);
448 digest[3] = swap_workaround (digest[3]);
450 rc4_next_16 (rc4_key, 16, j, digest, out);
452 const u32x r0 = out[0];
453 const u32x r1 = out[1];
454 const u32x r2 = out[2];
455 const u32x r3 = out[3];
457 #include VECT_COMPARE_M
461 extern "C" __global__ void __launch_bounds__ (64, 1) m09810_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 oldoffice34_t *oldoffice34_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)
465 extern "C" __global__ void __launch_bounds__ (64, 1) m09810_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 oldoffice34_t *oldoffice34_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)
469 extern "C" __global__ void __launch_bounds__ (64, 1) m09810_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 oldoffice34_t *oldoffice34_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)
475 const u32 lid = threadIdx.x;
477 __shared__ RC4_KEY rc4_keys[64];
479 RC4_KEY *rc4_key = &rc4_keys[lid];
485 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
487 if (gid >= gid_max) return;
491 wordl0[0] = pws[gid].i[ 0];
492 wordl0[1] = pws[gid].i[ 1];
493 wordl0[2] = pws[gid].i[ 2];
494 wordl0[3] = pws[gid].i[ 3];
498 wordl1[0] = pws[gid].i[ 4];
499 wordl1[1] = pws[gid].i[ 5];
500 wordl1[2] = pws[gid].i[ 6];
501 wordl1[3] = pws[gid].i[ 7];
517 const u32 pw_l_len = pws[gid].pw_len;
519 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
521 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
528 const u32 search[4] =
530 digests_buf[digests_offset].digest_buf[DGST_R0],
531 digests_buf[digests_offset].digest_buf[DGST_R1],
532 digests_buf[digests_offset].digest_buf[DGST_R2],
533 digests_buf[digests_offset].digest_buf[DGST_R3]
540 const u32 version = oldoffice34_bufs[salt_pos].version;
542 u32 encryptedVerifier[4];
544 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
545 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
546 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
547 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
553 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
557 wordr0[0] = c_combs[il_pos].i[0];
558 wordr0[1] = c_combs[il_pos].i[1];
559 wordr0[2] = c_combs[il_pos].i[2];
560 wordr0[3] = c_combs[il_pos].i[3];
564 wordr1[0] = c_combs[il_pos].i[4];
565 wordr1[1] = c_combs[il_pos].i[5];
566 wordr1[2] = c_combs[il_pos].i[6];
567 wordr1[3] = c_combs[il_pos].i[7];
583 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
585 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
592 key[0] = wordl0[0] | wordr0[0];
593 key[1] = (wordl0[1] | wordr0[1]) & 0xff;
597 rc4_init_16 (rc4_key, key);
601 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
608 w0_t[0] = swap_workaround (out[0]);
609 w0_t[1] = swap_workaround (out[1]);
610 w0_t[2] = swap_workaround (out[2]);
611 w0_t[3] = swap_workaround (out[3]);
612 w1_t[0] = 0x80000000;
633 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
635 digest[0] = swap_workaround (digest[0]);
636 digest[1] = swap_workaround (digest[1]);
637 digest[2] = swap_workaround (digest[2]);
638 digest[3] = swap_workaround (digest[3]);
640 rc4_next_16 (rc4_key, 16, j, digest, out);
642 const u32x r0 = out[0];
643 const u32x r1 = out[1];
644 const u32x r2 = out[2];
645 const u32x r3 = out[3];
647 #include VECT_COMPARE_S
651 extern "C" __global__ void __launch_bounds__ (64, 1) m09810_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 oldoffice34_t *oldoffice34_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)
655 extern "C" __global__ void __launch_bounds__ (64, 1) m09810_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 oldoffice34_t *oldoffice34_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)