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"
38 __device__ static void md4_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
62 MD4_STEP (MD4_Fo, a, b, c, d, w0_t, MD4C00, MD4S00);
63 MD4_STEP (MD4_Fo, d, a, b, c, w1_t, MD4C00, MD4S01);
64 MD4_STEP (MD4_Fo, c, d, a, b, w2_t, MD4C00, MD4S02);
65 MD4_STEP (MD4_Fo, b, c, d, a, w3_t, MD4C00, MD4S03);
66 MD4_STEP (MD4_Fo, a, b, c, d, w4_t, MD4C00, MD4S00);
67 MD4_STEP (MD4_Fo, d, a, b, c, w5_t, MD4C00, MD4S01);
68 MD4_STEP (MD4_Fo, c, d, a, b, w6_t, MD4C00, MD4S02);
69 MD4_STEP (MD4_Fo, b, c, d, a, w7_t, MD4C00, MD4S03);
70 MD4_STEP (MD4_Fo, a, b, c, d, w8_t, MD4C00, MD4S00);
71 MD4_STEP (MD4_Fo, d, a, b, c, w9_t, MD4C00, MD4S01);
72 MD4_STEP (MD4_Fo, c, d, a, b, wa_t, MD4C00, MD4S02);
73 MD4_STEP (MD4_Fo, b, c, d, a, wb_t, MD4C00, MD4S03);
74 MD4_STEP (MD4_Fo, a, b, c, d, wc_t, MD4C00, MD4S00);
75 MD4_STEP (MD4_Fo, d, a, b, c, wd_t, MD4C00, MD4S01);
76 MD4_STEP (MD4_Fo, c, d, a, b, we_t, MD4C00, MD4S02);
77 MD4_STEP (MD4_Fo, b, c, d, a, wf_t, MD4C00, MD4S03);
79 MD4_STEP (MD4_Go, a, b, c, d, w0_t, MD4C01, MD4S10);
80 MD4_STEP (MD4_Go, d, a, b, c, w4_t, MD4C01, MD4S11);
81 MD4_STEP (MD4_Go, c, d, a, b, w8_t, MD4C01, MD4S12);
82 MD4_STEP (MD4_Go, b, c, d, a, wc_t, MD4C01, MD4S13);
83 MD4_STEP (MD4_Go, a, b, c, d, w1_t, MD4C01, MD4S10);
84 MD4_STEP (MD4_Go, d, a, b, c, w5_t, MD4C01, MD4S11);
85 MD4_STEP (MD4_Go, c, d, a, b, w9_t, MD4C01, MD4S12);
86 MD4_STEP (MD4_Go, b, c, d, a, wd_t, MD4C01, MD4S13);
87 MD4_STEP (MD4_Go, a, b, c, d, w2_t, MD4C01, MD4S10);
88 MD4_STEP (MD4_Go, d, a, b, c, w6_t, MD4C01, MD4S11);
89 MD4_STEP (MD4_Go, c, d, a, b, wa_t, MD4C01, MD4S12);
90 MD4_STEP (MD4_Go, b, c, d, a, we_t, MD4C01, MD4S13);
91 MD4_STEP (MD4_Go, a, b, c, d, w3_t, MD4C01, MD4S10);
92 MD4_STEP (MD4_Go, d, a, b, c, w7_t, MD4C01, MD4S11);
93 MD4_STEP (MD4_Go, c, d, a, b, wb_t, MD4C01, MD4S12);
94 MD4_STEP (MD4_Go, b, c, d, a, wf_t, MD4C01, MD4S13);
96 MD4_STEP (MD4_H , a, b, c, d, w0_t, MD4C02, MD4S20);
97 MD4_STEP (MD4_H , d, a, b, c, w8_t, MD4C02, MD4S21);
98 MD4_STEP (MD4_H , c, d, a, b, w4_t, MD4C02, MD4S22);
99 MD4_STEP (MD4_H , b, c, d, a, wc_t, MD4C02, MD4S23);
100 MD4_STEP (MD4_H , a, b, c, d, w2_t, MD4C02, MD4S20);
101 MD4_STEP (MD4_H , d, a, b, c, wa_t, MD4C02, MD4S21);
102 MD4_STEP (MD4_H , c, d, a, b, w6_t, MD4C02, MD4S22);
103 MD4_STEP (MD4_H , b, c, d, a, we_t, MD4C02, MD4S23);
104 MD4_STEP (MD4_H , a, b, c, d, w1_t, MD4C02, MD4S20);
105 MD4_STEP (MD4_H , d, a, b, c, w9_t, MD4C02, MD4S21);
106 MD4_STEP (MD4_H , c, d, a, b, w5_t, MD4C02, MD4S22);
107 MD4_STEP (MD4_H , b, c, d, a, wd_t, MD4C02, MD4S23);
108 MD4_STEP (MD4_H , a, b, c, d, w3_t, MD4C02, MD4S20);
109 MD4_STEP (MD4_H , d, a, b, c, wb_t, MD4C02, MD4S21);
110 MD4_STEP (MD4_H , c, d, a, b, w7_t, MD4C02, MD4S22);
111 MD4_STEP (MD4_H , b, c, d, a, wf_t, MD4C02, MD4S23);
119 __device__ static void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
143 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
144 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
145 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
146 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
147 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
148 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
149 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
150 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
151 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
152 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
153 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
154 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
155 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
156 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
157 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
158 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
160 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
161 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
162 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
163 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
164 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
165 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
166 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
167 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
168 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
169 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
170 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
171 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
172 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
173 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
174 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
175 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
177 MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
178 MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
179 MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
180 MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
181 MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
182 MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
183 MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
184 MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
185 MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
186 MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
187 MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
188 MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
189 MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
190 MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
191 MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
192 MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
194 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
195 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
196 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
197 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
198 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
199 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
200 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
201 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
202 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
203 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
204 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
205 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
206 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
207 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
208 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
209 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
217 __device__ static void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
219 w0[0] = w0[0] ^ 0x36363636;
220 w0[1] = w0[1] ^ 0x36363636;
221 w0[2] = w0[2] ^ 0x36363636;
222 w0[3] = w0[3] ^ 0x36363636;
223 w1[0] = w1[0] ^ 0x36363636;
224 w1[1] = w1[1] ^ 0x36363636;
225 w1[2] = w1[2] ^ 0x36363636;
226 w1[3] = w1[3] ^ 0x36363636;
227 w2[0] = w2[0] ^ 0x36363636;
228 w2[1] = w2[1] ^ 0x36363636;
229 w2[2] = w2[2] ^ 0x36363636;
230 w2[3] = w2[3] ^ 0x36363636;
231 w3[0] = w3[0] ^ 0x36363636;
232 w3[1] = w3[1] ^ 0x36363636;
233 w3[2] = w3[2] ^ 0x36363636;
234 w3[3] = w3[3] ^ 0x36363636;
241 md5_transform (w0, w1, w2, w3, ipad);
243 w0[0] = w0[0] ^ 0x6a6a6a6a;
244 w0[1] = w0[1] ^ 0x6a6a6a6a;
245 w0[2] = w0[2] ^ 0x6a6a6a6a;
246 w0[3] = w0[3] ^ 0x6a6a6a6a;
247 w1[0] = w1[0] ^ 0x6a6a6a6a;
248 w1[1] = w1[1] ^ 0x6a6a6a6a;
249 w1[2] = w1[2] ^ 0x6a6a6a6a;
250 w1[3] = w1[3] ^ 0x6a6a6a6a;
251 w2[0] = w2[0] ^ 0x6a6a6a6a;
252 w2[1] = w2[1] ^ 0x6a6a6a6a;
253 w2[2] = w2[2] ^ 0x6a6a6a6a;
254 w2[3] = w2[3] ^ 0x6a6a6a6a;
255 w3[0] = w3[0] ^ 0x6a6a6a6a;
256 w3[1] = w3[1] ^ 0x6a6a6a6a;
257 w3[2] = w3[2] ^ 0x6a6a6a6a;
258 w3[3] = w3[3] ^ 0x6a6a6a6a;
265 md5_transform (w0, w1, w2, w3, opad);
268 __device__ static void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
275 md5_transform (w0, w1, w2, w3, digest);
291 w3[2] = (64 + 16) * 8;
299 md5_transform (w0, w1, w2, w3, digest);
302 __device__ __constant__ bf_t c_bfs[1024];
304 __device__ static void m05600m (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 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 netntlm_t *netntlm_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, u32 s_userdomain_buf[64], u32 s_chall_buf[256])
310 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
311 const u32 lid = threadIdx.x;
317 const u32 userdomain_len = netntlm_bufs[salt_pos].user_len
318 + netntlm_bufs[salt_pos].domain_len;
320 const u32 chall_len = netntlm_bufs[salt_pos].srvchall_len
321 + netntlm_bufs[salt_pos].clichall_len;
329 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
331 const u32 w0r = c_bfs[il_pos].i;
342 md4_transform (w0, w1, w2, w3, digest);
374 hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
379 for (left = userdomain_len, off = 0; left >= 56; left -= 64, off += 16)
381 w0_t[0] = s_userdomain_buf[off + 0];
382 w0_t[1] = s_userdomain_buf[off + 1];
383 w0_t[2] = s_userdomain_buf[off + 2];
384 w0_t[3] = s_userdomain_buf[off + 3];
385 w1_t[0] = s_userdomain_buf[off + 4];
386 w1_t[1] = s_userdomain_buf[off + 5];
387 w1_t[2] = s_userdomain_buf[off + 6];
388 w1_t[3] = s_userdomain_buf[off + 7];
389 w2_t[0] = s_userdomain_buf[off + 8];
390 w2_t[1] = s_userdomain_buf[off + 9];
391 w2_t[2] = s_userdomain_buf[off + 10];
392 w2_t[3] = s_userdomain_buf[off + 11];
393 w3_t[0] = s_userdomain_buf[off + 12];
394 w3_t[1] = s_userdomain_buf[off + 13];
395 w3_t[2] = s_userdomain_buf[off + 14];
396 w3_t[3] = s_userdomain_buf[off + 15];
398 md5_transform (w0_t, w1_t, w2_t, w3_t, ipad);
401 w0_t[0] = s_userdomain_buf[off + 0];
402 w0_t[1] = s_userdomain_buf[off + 1];
403 w0_t[2] = s_userdomain_buf[off + 2];
404 w0_t[3] = s_userdomain_buf[off + 3];
405 w1_t[0] = s_userdomain_buf[off + 4];
406 w1_t[1] = s_userdomain_buf[off + 5];
407 w1_t[2] = s_userdomain_buf[off + 6];
408 w1_t[3] = s_userdomain_buf[off + 7];
409 w2_t[0] = s_userdomain_buf[off + 8];
410 w2_t[1] = s_userdomain_buf[off + 9];
411 w2_t[2] = s_userdomain_buf[off + 10];
412 w2_t[3] = s_userdomain_buf[off + 11];
413 w3_t[0] = s_userdomain_buf[off + 12];
414 w3_t[1] = s_userdomain_buf[off + 13];
415 w3_t[2] = (64 + userdomain_len) * 8;
418 hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
442 hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
444 for (left = chall_len, off = 0; left >= 56; left -= 64, off += 16)
446 w0_t[0] = s_chall_buf[off + 0];
447 w0_t[1] = s_chall_buf[off + 1];
448 w0_t[2] = s_chall_buf[off + 2];
449 w0_t[3] = s_chall_buf[off + 3];
450 w1_t[0] = s_chall_buf[off + 4];
451 w1_t[1] = s_chall_buf[off + 5];
452 w1_t[2] = s_chall_buf[off + 6];
453 w1_t[3] = s_chall_buf[off + 7];
454 w2_t[0] = s_chall_buf[off + 8];
455 w2_t[1] = s_chall_buf[off + 9];
456 w2_t[2] = s_chall_buf[off + 10];
457 w2_t[3] = s_chall_buf[off + 11];
458 w3_t[0] = s_chall_buf[off + 12];
459 w3_t[1] = s_chall_buf[off + 13];
460 w3_t[2] = s_chall_buf[off + 14];
461 w3_t[3] = s_chall_buf[off + 15];
463 md5_transform (w0_t, w1_t, w2_t, w3_t, ipad);
466 w0_t[0] = s_chall_buf[off + 0];
467 w0_t[1] = s_chall_buf[off + 1];
468 w0_t[2] = s_chall_buf[off + 2];
469 w0_t[3] = s_chall_buf[off + 3];
470 w1_t[0] = s_chall_buf[off + 4];
471 w1_t[1] = s_chall_buf[off + 5];
472 w1_t[2] = s_chall_buf[off + 6];
473 w1_t[3] = s_chall_buf[off + 7];
474 w2_t[0] = s_chall_buf[off + 8];
475 w2_t[1] = s_chall_buf[off + 9];
476 w2_t[2] = s_chall_buf[off + 10];
477 w2_t[3] = s_chall_buf[off + 11];
478 w3_t[0] = s_chall_buf[off + 12];
479 w3_t[1] = s_chall_buf[off + 13];
480 w3_t[2] = (64 + chall_len) * 8;
483 hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
485 const u32x r0 = digest[0];
486 const u32x r1 = digest[3];
487 const u32x r2 = digest[2];
488 const u32x r3 = digest[1];
490 #include VECT_COMPARE_M
494 __device__ static void m05600s (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 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 netntlm_t *netntlm_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, u32 s_userdomain_buf[64], u32 s_chall_buf[256])
500 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
501 const u32 lid = threadIdx.x;
507 const u32 search[4] =
509 digests_buf[digests_offset].digest_buf[DGST_R0],
510 digests_buf[digests_offset].digest_buf[DGST_R1],
511 digests_buf[digests_offset].digest_buf[DGST_R2],
512 digests_buf[digests_offset].digest_buf[DGST_R3]
519 const u32 userdomain_len = netntlm_bufs[salt_pos].user_len
520 + netntlm_bufs[salt_pos].domain_len;
522 const u32 chall_len = netntlm_bufs[salt_pos].srvchall_len
523 + netntlm_bufs[salt_pos].clichall_len;
531 for (u32 il_pos = 0; il_pos < bfs_cnt; il_pos++)
533 const u32 w0r = c_bfs[il_pos].i;
544 md4_transform (w0, w1, w2, w3, digest);
576 hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
581 for (left = userdomain_len, off = 0; left >= 56; left -= 64, off += 16)
583 w0_t[0] = s_userdomain_buf[off + 0];
584 w0_t[1] = s_userdomain_buf[off + 1];
585 w0_t[2] = s_userdomain_buf[off + 2];
586 w0_t[3] = s_userdomain_buf[off + 3];
587 w1_t[0] = s_userdomain_buf[off + 4];
588 w1_t[1] = s_userdomain_buf[off + 5];
589 w1_t[2] = s_userdomain_buf[off + 6];
590 w1_t[3] = s_userdomain_buf[off + 7];
591 w2_t[0] = s_userdomain_buf[off + 8];
592 w2_t[1] = s_userdomain_buf[off + 9];
593 w2_t[2] = s_userdomain_buf[off + 10];
594 w2_t[3] = s_userdomain_buf[off + 11];
595 w3_t[0] = s_userdomain_buf[off + 12];
596 w3_t[1] = s_userdomain_buf[off + 13];
597 w3_t[2] = s_userdomain_buf[off + 14];
598 w3_t[3] = s_userdomain_buf[off + 15];
600 md5_transform (w0_t, w1_t, w2_t, w3_t, ipad);
603 w0_t[0] = s_userdomain_buf[off + 0];
604 w0_t[1] = s_userdomain_buf[off + 1];
605 w0_t[2] = s_userdomain_buf[off + 2];
606 w0_t[3] = s_userdomain_buf[off + 3];
607 w1_t[0] = s_userdomain_buf[off + 4];
608 w1_t[1] = s_userdomain_buf[off + 5];
609 w1_t[2] = s_userdomain_buf[off + 6];
610 w1_t[3] = s_userdomain_buf[off + 7];
611 w2_t[0] = s_userdomain_buf[off + 8];
612 w2_t[1] = s_userdomain_buf[off + 9];
613 w2_t[2] = s_userdomain_buf[off + 10];
614 w2_t[3] = s_userdomain_buf[off + 11];
615 w3_t[0] = s_userdomain_buf[off + 12];
616 w3_t[1] = s_userdomain_buf[off + 13];
617 w3_t[2] = (64 + userdomain_len) * 8;
620 hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
644 hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
646 for (left = chall_len, off = 0; left >= 56; left -= 64, off += 16)
648 w0_t[0] = s_chall_buf[off + 0];
649 w0_t[1] = s_chall_buf[off + 1];
650 w0_t[2] = s_chall_buf[off + 2];
651 w0_t[3] = s_chall_buf[off + 3];
652 w1_t[0] = s_chall_buf[off + 4];
653 w1_t[1] = s_chall_buf[off + 5];
654 w1_t[2] = s_chall_buf[off + 6];
655 w1_t[3] = s_chall_buf[off + 7];
656 w2_t[0] = s_chall_buf[off + 8];
657 w2_t[1] = s_chall_buf[off + 9];
658 w2_t[2] = s_chall_buf[off + 10];
659 w2_t[3] = s_chall_buf[off + 11];
660 w3_t[0] = s_chall_buf[off + 12];
661 w3_t[1] = s_chall_buf[off + 13];
662 w3_t[2] = s_chall_buf[off + 14];
663 w3_t[3] = s_chall_buf[off + 15];
665 md5_transform (w0_t, w1_t, w2_t, w3_t, ipad);
668 w0_t[0] = s_chall_buf[off + 0];
669 w0_t[1] = s_chall_buf[off + 1];
670 w0_t[2] = s_chall_buf[off + 2];
671 w0_t[3] = s_chall_buf[off + 3];
672 w1_t[0] = s_chall_buf[off + 4];
673 w1_t[1] = s_chall_buf[off + 5];
674 w1_t[2] = s_chall_buf[off + 6];
675 w1_t[3] = s_chall_buf[off + 7];
676 w2_t[0] = s_chall_buf[off + 8];
677 w2_t[1] = s_chall_buf[off + 9];
678 w2_t[2] = s_chall_buf[off + 10];
679 w2_t[3] = s_chall_buf[off + 11];
680 w3_t[0] = s_chall_buf[off + 12];
681 w3_t[1] = s_chall_buf[off + 13];
682 w3_t[2] = (64 + chall_len) * 8;
685 hmac_md5_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
687 const u32x r0 = digest[0];
688 const u32x r1 = digest[3];
689 const u32x r2 = digest[2];
690 const u32x r3 = digest[1];
692 #include VECT_COMPARE_S
696 extern "C" __global__ void __launch_bounds__ (256, 1) m05600_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 netntlm_t *netntlm_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)
702 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
703 const u32 lid = threadIdx.x;
707 w0[0] = pws[gid].i[ 0];
708 w0[1] = pws[gid].i[ 1];
709 w0[2] = pws[gid].i[ 2];
710 w0[3] = pws[gid].i[ 3];
730 w3[2] = pws[gid].i[14];
733 const u32 pw_len = pws[gid].pw_len;
739 __shared__ u32 s_userdomain_buf[64];
740 __shared__ u32 s_chall_buf[256];
744 s_userdomain_buf[lid] = netntlm_bufs[salt_pos].userdomain_buf[lid];
747 s_chall_buf[lid] = netntlm_bufs[salt_pos].chall_buf[lid];
751 if (gid >= gid_max) return;
757 m05600m (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, netntlm_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, s_userdomain_buf, s_chall_buf);
760 extern "C" __global__ void __launch_bounds__ (256, 1) m05600_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 netntlm_t *netntlm_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)
766 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
767 const u32 lid = threadIdx.x;
771 w0[0] = pws[gid].i[ 0];
772 w0[1] = pws[gid].i[ 1];
773 w0[2] = pws[gid].i[ 2];
774 w0[3] = pws[gid].i[ 3];
778 w1[0] = pws[gid].i[ 4];
779 w1[1] = pws[gid].i[ 5];
780 w1[2] = pws[gid].i[ 6];
781 w1[3] = pws[gid].i[ 7];
794 w3[2] = pws[gid].i[14];
797 const u32 pw_len = pws[gid].pw_len;
803 __shared__ u32 s_userdomain_buf[64];
804 __shared__ u32 s_chall_buf[256];
808 s_userdomain_buf[lid] = netntlm_bufs[salt_pos].userdomain_buf[lid];
811 s_chall_buf[lid] = netntlm_bufs[salt_pos].chall_buf[lid];
815 if (gid >= gid_max) return;
821 m05600m (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, netntlm_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, s_userdomain_buf, s_chall_buf);
824 extern "C" __global__ void __launch_bounds__ (256, 1) m05600_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 netntlm_t *netntlm_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)
828 extern "C" __global__ void __launch_bounds__ (256, 1) m05600_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 netntlm_t *netntlm_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)
834 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
835 const u32 lid = threadIdx.x;
839 w0[0] = pws[gid].i[ 0];
840 w0[1] = pws[gid].i[ 1];
841 w0[2] = pws[gid].i[ 2];
842 w0[3] = pws[gid].i[ 3];
862 w3[2] = pws[gid].i[14];
865 const u32 pw_len = pws[gid].pw_len;
871 __shared__ u32 s_userdomain_buf[64];
872 __shared__ u32 s_chall_buf[256];
876 s_userdomain_buf[lid] = netntlm_bufs[salt_pos].userdomain_buf[lid];
879 s_chall_buf[lid] = netntlm_bufs[salt_pos].chall_buf[lid];
883 if (gid >= gid_max) return;
889 m05600s (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, netntlm_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, s_userdomain_buf, s_chall_buf);
892 extern "C" __global__ void __launch_bounds__ (256, 1) m05600_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 netntlm_t *netntlm_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)
898 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
899 const u32 lid = threadIdx.x;
903 w0[0] = pws[gid].i[ 0];
904 w0[1] = pws[gid].i[ 1];
905 w0[2] = pws[gid].i[ 2];
906 w0[3] = pws[gid].i[ 3];
910 w1[0] = pws[gid].i[ 4];
911 w1[1] = pws[gid].i[ 5];
912 w1[2] = pws[gid].i[ 6];
913 w1[3] = pws[gid].i[ 7];
926 w3[2] = pws[gid].i[14];
929 const u32 pw_len = pws[gid].pw_len;
935 __shared__ u32 s_userdomain_buf[64];
936 __shared__ u32 s_chall_buf[256];
940 s_userdomain_buf[lid] = netntlm_bufs[salt_pos].userdomain_buf[lid];
943 s_chall_buf[lid] = netntlm_bufs[salt_pos].chall_buf[lid];
947 if (gid >= gid_max) return;
953 m05600s (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, netntlm_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, s_userdomain_buf, s_chall_buf);
956 extern "C" __global__ void __launch_bounds__ (256, 1) m05600_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 netntlm_t *netntlm_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)