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_vect4_comp4.c"
35 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
46 __device__ static void swap (RC4_KEY *rc4_key, const u32 i, const u32 j)
51 rc4_key->S[i] = rc4_key->S[j];
55 __device__ static void rc4_init_16 (RC4_KEY *rc4_key, const u32 data[4])
60 u32 *ptr = (u32 *) rc4_key->S;
63 for (u32 i = 0; i < 64; i++)
70 for (u32 i = 0; i < 16; i++)
78 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
79 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
80 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
81 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
85 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
86 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
87 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
88 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
92 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
93 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
94 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
95 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
99 j += rc4_key->S[idx] + (v >> 0); swap (rc4_key, idx, j & 0xff); idx++;
100 j += rc4_key->S[idx] + (v >> 8); swap (rc4_key, idx, j & 0xff); idx++;
101 j += rc4_key->S[idx] + (v >> 16); swap (rc4_key, idx, j & 0xff); idx++;
102 j += rc4_key->S[idx] + (v >> 24); swap (rc4_key, idx, j & 0xff); idx++;
106 __device__ static u8 rc4_next_16 (RC4_KEY *rc4_key, u8 i, u8 j, const u32 in[4], u32 out[4])
108 for (u32 k = 0; k < 4; k++)
117 swap (rc4_key, i, j);
119 idx = rc4_key->S[i] + rc4_key->S[j];
121 xor4 |= rc4_key->S[idx] << 0;
126 swap (rc4_key, i, j);
128 idx = rc4_key->S[i] + rc4_key->S[j];
130 xor4 |= rc4_key->S[idx] << 8;
135 swap (rc4_key, i, j);
137 idx = rc4_key->S[i] + rc4_key->S[j];
139 xor4 |= rc4_key->S[idx] << 16;
144 swap (rc4_key, i, j);
146 idx = rc4_key->S[i] + rc4_key->S[j];
148 xor4 |= rc4_key->S[idx] << 24;
150 out[k] = in[k] ^ xor4;
156 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
184 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
185 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
186 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
187 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
188 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
189 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
190 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
191 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
192 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
193 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
194 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
195 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
196 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
197 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
198 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
199 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
200 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
201 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
202 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
203 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
208 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
209 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
210 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
211 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
212 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
213 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
214 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
215 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
216 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
217 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
218 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
219 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
220 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
221 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
222 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
223 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
224 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
225 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
226 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
227 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
232 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
233 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
234 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
235 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
236 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
237 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
238 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
239 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
240 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
241 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
242 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
243 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
244 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
245 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
246 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
247 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
248 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
249 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
250 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
251 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
256 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
257 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
258 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
259 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
260 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
261 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
262 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
263 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
264 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
265 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
266 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
267 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
268 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
269 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
270 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
271 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
272 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
273 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
274 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
275 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
284 __device__ __constant__ bf_t c_bfs[1024];
286 __device__ static void m09800m (RC4_KEY rc4_keys[64], u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
292 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
293 const u32 lid = threadIdx.x;
295 RC4_KEY *rc4_key = &rc4_keys[lid];
303 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
304 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
305 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
306 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
308 const u32 salt_len = 16;
310 const u32 pw_salt_len = pw_len + salt_len;
316 const u32 version = oldoffice34_bufs[salt_pos].version;
318 u32 encryptedVerifier[4];
320 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
321 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
322 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
323 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
331 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
333 const u32 w0r = c_bfs[il_pos].i;
342 w0_t[0] = salt_buf[0];
343 w0_t[1] = salt_buf[1];
344 w0_t[2] = salt_buf[2];
345 w0_t[3] = salt_buf[3];
357 w3_t[3] = pw_salt_len * 8;
367 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
375 w1_t[2] = 0x80000000;
384 w3_t[3] = (20 + 4) * 8;
392 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
396 key[0] = swap_workaround (digest[0]);
397 key[1] = swap_workaround (digest[1]);
398 key[2] = swap_workaround (digest[2]);
399 key[3] = swap_workaround (digest[3]);
408 rc4_init_16 (rc4_key, key);
412 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
414 w0_t[0] = swap_workaround (out[0]);
415 w0_t[1] = swap_workaround (out[1]);
416 w0_t[2] = swap_workaround (out[2]);
417 w0_t[3] = swap_workaround (out[3]);
418 w1_t[0] = 0x80000000;
437 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
439 digest[0] = swap_workaround (digest[0]);
440 digest[1] = swap_workaround (digest[1]);
441 digest[2] = swap_workaround (digest[2]);
442 digest[3] = swap_workaround (digest[3]);
444 rc4_next_16 (rc4_key, 16, j, digest, out);
446 const u32x r0 = out[0];
447 const u32x r1 = out[1];
448 const u32x r2 = out[2];
449 const u32x r3 = out[3];
451 #include VECT_COMPARE_M
455 __device__ static void m09800s (RC4_KEY rc4_keys[64], u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], const u32 pw_len, const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset)
461 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
462 const u32 lid = threadIdx.x;
464 RC4_KEY *rc4_key = &rc4_keys[lid];
470 const u32 search[4] =
472 digests_buf[digests_offset].digest_buf[DGST_R0],
473 digests_buf[digests_offset].digest_buf[DGST_R1],
474 digests_buf[digests_offset].digest_buf[DGST_R2],
475 digests_buf[digests_offset].digest_buf[DGST_R3]
484 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
485 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
486 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
487 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
489 const u32 salt_len = 16;
491 const u32 pw_salt_len = pw_len + salt_len;
497 const u32 version = oldoffice34_bufs[salt_pos].version;
499 u32 encryptedVerifier[4];
501 encryptedVerifier[0] = oldoffice34_bufs[salt_pos].encryptedVerifier[0];
502 encryptedVerifier[1] = oldoffice34_bufs[salt_pos].encryptedVerifier[1];
503 encryptedVerifier[2] = oldoffice34_bufs[salt_pos].encryptedVerifier[2];
504 encryptedVerifier[3] = oldoffice34_bufs[salt_pos].encryptedVerifier[3];
512 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
514 const u32 w0r = c_bfs[il_pos].i;
523 w0_t[0] = salt_buf[0];
524 w0_t[1] = salt_buf[1];
525 w0_t[2] = salt_buf[2];
526 w0_t[3] = salt_buf[3];
538 w3_t[3] = pw_salt_len * 8;
548 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
556 w1_t[2] = 0x80000000;
565 w3_t[3] = (20 + 4) * 8;
573 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
577 key[0] = swap_workaround (digest[0]);
578 key[1] = swap_workaround (digest[1]);
579 key[2] = swap_workaround (digest[2]);
580 key[3] = swap_workaround (digest[3]);
589 rc4_init_16 (rc4_key, key);
593 u8 j = rc4_next_16 (rc4_key, 0, 0, encryptedVerifier, out);
595 w0_t[0] = swap_workaround (out[0]);
596 w0_t[1] = swap_workaround (out[1]);
597 w0_t[2] = swap_workaround (out[2]);
598 w0_t[3] = swap_workaround (out[3]);
599 w1_t[0] = 0x80000000;
618 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
620 digest[0] = swap_workaround (digest[0]);
621 digest[1] = swap_workaround (digest[1]);
622 digest[2] = swap_workaround (digest[2]);
623 digest[3] = swap_workaround (digest[3]);
625 rc4_next_16 (rc4_key, 16, j, digest, out);
627 const u32x r0 = out[0];
628 const u32x r1 = out[1];
629 const u32x r2 = out[2];
630 const u32x r3 = out[3];
632 #include VECT_COMPARE_S
636 extern "C" __global__ void __launch_bounds__ (64, 1) m09800_m04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
642 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
644 if (gid >= gid_max) return;
648 w0[0] = pws[gid].i[ 0];
649 w0[1] = pws[gid].i[ 1];
650 w0[2] = pws[gid].i[ 2];
651 w0[3] = pws[gid].i[ 3];
674 const u32 pw_len = pws[gid].pw_len;
680 __shared__ RC4_KEY rc4_keys[64];
682 m09800m (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, oldoffice34_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
685 extern "C" __global__ void __launch_bounds__ (64, 1) m09800_m08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
691 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
693 if (gid >= gid_max) return;
697 w0[0] = pws[gid].i[ 0];
698 w0[1] = pws[gid].i[ 1];
699 w0[2] = pws[gid].i[ 2];
700 w0[3] = pws[gid].i[ 3];
704 w1[0] = pws[gid].i[ 4];
705 w1[1] = pws[gid].i[ 5];
706 w1[2] = pws[gid].i[ 6];
707 w1[3] = pws[gid].i[ 7];
723 const u32 pw_len = pws[gid].pw_len;
729 __shared__ RC4_KEY rc4_keys[64];
731 m09800m (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, oldoffice34_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
734 extern "C" __global__ void __launch_bounds__ (64, 1) m09800_m16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
740 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
742 if (gid >= gid_max) return;
746 w0[0] = pws[gid].i[ 0];
747 w0[1] = pws[gid].i[ 1];
748 w0[2] = pws[gid].i[ 2];
749 w0[3] = pws[gid].i[ 3];
753 w1[0] = pws[gid].i[ 4];
754 w1[1] = pws[gid].i[ 5];
755 w1[2] = pws[gid].i[ 6];
756 w1[3] = pws[gid].i[ 7];
760 w2[0] = pws[gid].i[ 8];
761 w2[1] = pws[gid].i[ 9];
762 w2[2] = pws[gid].i[10];
763 w2[3] = pws[gid].i[11];
767 w3[0] = pws[gid].i[12];
768 w3[1] = pws[gid].i[13];
772 const u32 pw_len = pws[gid].pw_len;
778 __shared__ RC4_KEY rc4_keys[64];
780 m09800m (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, oldoffice34_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
783 extern "C" __global__ void __launch_bounds__ (64, 1) m09800_s04 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
789 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
791 if (gid >= gid_max) return;
795 w0[0] = pws[gid].i[ 0];
796 w0[1] = pws[gid].i[ 1];
797 w0[2] = pws[gid].i[ 2];
798 w0[3] = pws[gid].i[ 3];
821 const u32 pw_len = pws[gid].pw_len;
827 __shared__ RC4_KEY rc4_keys[64];
829 m09800s (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, oldoffice34_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
832 extern "C" __global__ void __launch_bounds__ (64, 1) m09800_s08 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
838 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
840 if (gid >= gid_max) return;
844 w0[0] = pws[gid].i[ 0];
845 w0[1] = pws[gid].i[ 1];
846 w0[2] = pws[gid].i[ 2];
847 w0[3] = pws[gid].i[ 3];
851 w1[0] = pws[gid].i[ 4];
852 w1[1] = pws[gid].i[ 5];
853 w1[2] = pws[gid].i[ 6];
854 w1[3] = pws[gid].i[ 7];
870 const u32 pw_len = pws[gid].pw_len;
876 __shared__ RC4_KEY rc4_keys[64];
878 m09800s (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, oldoffice34_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);
881 extern "C" __global__ void __launch_bounds__ (64, 1) m09800_s16 (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const u32x *bfs_buf, 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 bfs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
887 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
889 if (gid >= gid_max) return;
893 w0[0] = pws[gid].i[ 0];
894 w0[1] = pws[gid].i[ 1];
895 w0[2] = pws[gid].i[ 2];
896 w0[3] = pws[gid].i[ 3];
900 w1[0] = pws[gid].i[ 4];
901 w1[1] = pws[gid].i[ 5];
902 w1[2] = pws[gid].i[ 6];
903 w1[3] = pws[gid].i[ 7];
907 w2[0] = pws[gid].i[ 8];
908 w2[1] = pws[gid].i[ 9];
909 w2[2] = pws[gid].i[10];
910 w2[3] = pws[gid].i[11];
914 w3[0] = pws[gid].i[12];
915 w3[1] = pws[gid].i[13];
919 const u32 pw_len = pws[gid].pw_len;
925 __shared__ RC4_KEY rc4_keys[64];
927 m09800s (rc4_keys, w0, w1, w2, w3, pw_len, pws, rules_buf, combs_buf, bfs_buf, tmps, hooks, bitmaps_buf_s1_a, bitmaps_buf_s1_b, bitmaps_buf_s1_c, bitmaps_buf_s1_d, bitmaps_buf_s2_a, bitmaps_buf_s2_b, bitmaps_buf_s2_c, bitmaps_buf_s2_d, plains_buf, digests_buf, hashes_shown, salt_bufs, oldoffice34_bufs, d_return_buf, d_scryptV_buf, bitmap_mask, bitmap_shift1, bitmap_shift2, salt_pos, loop_pos, loop_cnt, bfs_cnt, digests_cnt, digests_offset);