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__ __constant__ comb_t c_combs[1024];
40 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
68 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
69 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
70 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
71 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
72 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
73 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
74 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
75 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
76 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
77 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
78 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
79 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
80 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
81 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
82 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
83 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
84 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
85 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
86 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
87 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
92 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
93 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
94 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
95 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
96 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
97 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
98 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
99 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
100 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
101 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
102 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
103 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
104 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
105 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
106 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
107 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
108 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
109 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
110 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
111 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
116 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
117 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
118 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
119 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
120 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
121 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
122 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
123 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
124 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
125 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
126 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
127 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
128 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
129 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
130 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
131 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
132 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
133 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
134 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
135 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
140 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
141 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
142 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
143 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
144 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
145 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
146 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
147 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
148 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
149 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
150 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
151 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
152 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
153 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
154 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
155 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
156 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
157 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
158 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
159 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
168 extern "C" __global__ void __launch_bounds__ (256, 1) m08300_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 void *esalt_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)
174 const u32 lid = threadIdx.x;
180 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
182 if (gid >= gid_max) return;
186 wordl0[0] = pws[gid].i[ 0];
187 wordl0[1] = pws[gid].i[ 1];
188 wordl0[2] = pws[gid].i[ 2];
189 wordl0[3] = pws[gid].i[ 3];
193 wordl1[0] = pws[gid].i[ 4];
194 wordl1[1] = pws[gid].i[ 5];
195 wordl1[2] = pws[gid].i[ 6];
196 wordl1[3] = pws[gid].i[ 7];
212 const u32 pw_l_len = pws[gid].pw_len;
214 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
216 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
223 const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
227 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
228 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
229 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
230 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
234 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
235 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
236 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
237 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
239 const u32 salt_len = salt_bufs[salt_pos].salt_len;
243 domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
244 domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
245 domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
246 domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
250 domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
251 domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
252 domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
255 const u32 domain_len = salt_bufs[salt_pos].salt_buf_pc[ 7];
261 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
263 const u32 pw_r_len = c_combs[il_pos].pw_len;
265 const u32 pw_len = pw_l_len + pw_r_len;
272 wordr0[0] = c_combs[il_pos].i[0];
273 wordr0[1] = c_combs[il_pos].i[1];
274 wordr0[2] = c_combs[il_pos].i[2];
275 wordr0[3] = c_combs[il_pos].i[3];
276 wordr1[0] = c_combs[il_pos].i[4];
277 wordr1[1] = c_combs[il_pos].i[5];
278 wordr1[2] = c_combs[il_pos].i[6];
279 wordr1[3] = c_combs[il_pos].i[7];
289 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
291 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
299 w0[0] = wordl0[0] | wordr0[0];
300 w0[1] = wordl0[1] | wordr0[1];
301 w0[2] = wordl0[2] | wordr0[2];
302 w0[3] = wordl0[3] | wordr0[3];
303 w1[0] = wordl1[0] | wordr1[0];
304 w1[1] = wordl1[1] | wordr1[1];
305 w1[2] = wordl1[2] | wordr1[2];
306 w1[3] = wordl1[3] | wordr1[3];
307 w2[0] = wordl2[0] | wordr2[0];
308 w2[1] = wordl2[1] | wordr2[1];
309 w2[2] = wordl2[2] | wordr2[2];
310 w2[3] = wordl2[3] | wordr2[3];
311 w3[0] = wordl3[0] | wordr3[0];
312 w3[1] = wordl3[1] | wordr3[1];
313 w3[2] = wordl3[2] | wordr3[2];
314 w3[3] = wordl3[3] | wordr3[3];
344 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, 1);
346 w0_t[0] |= pw_len & 0xff;
354 s0[0] = salt_buf0[0];
355 s0[1] = salt_buf0[1];
356 s0[2] = salt_buf0[2];
357 s0[3] = salt_buf0[3];
361 s1[0] = salt_buf1[0];
362 s1[1] = salt_buf1[1];
363 s1[2] = salt_buf1[2];
364 s1[3] = salt_buf1[3];
380 switch_buffer_by_offset (s0, s1, s2, s3, 1 + pw_len + domain_len + 1);
384 d0[0] = domain_buf0[0];
385 d0[1] = domain_buf0[1];
386 d0[2] = domain_buf0[2];
387 d0[3] = domain_buf0[3];
391 d1[0] = domain_buf1[0];
392 d1[1] = domain_buf1[1];
393 d1[2] = domain_buf1[2];
410 switch_buffer_by_offset (d0, d1, d2, d3, 1 + pw_len);
418 w0_t2[0] = swap_workaround (w0_t[0] | d0[0] | s0[0]);
419 w0_t2[1] = swap_workaround (w0_t[1] | d0[1] | s0[1]);
420 w0_t2[2] = swap_workaround (w0_t[2] | d0[2] | s0[2]);
421 w0_t2[3] = swap_workaround (w0_t[3] | d0[3] | s0[3]);
425 w1_t2[0] = swap_workaround (w1_t[0] | d1[0] | s1[0]);
426 w1_t2[1] = swap_workaround (w1_t[1] | d1[1] | s1[1]);
427 w1_t2[2] = swap_workaround (w1_t[2] | d1[2] | s1[2]);
428 w1_t2[3] = swap_workaround (w1_t[3] | d1[3] | s1[3]);
432 w2_t2[0] = swap_workaround (w2_t[0] | d2[0] | s2[0]);
433 w2_t2[1] = swap_workaround (w2_t[1] | d2[1] | s2[1]);
434 w2_t2[2] = swap_workaround (w2_t[2] | d2[2] | s2[2]);
435 w2_t2[3] = swap_workaround (w2_t[3] | d2[3] | s2[3]);
439 w3_t2[0] = swap_workaround (w3_t[0] | d3[0] | s3[0]);
440 w3_t2[1] = swap_workaround (w3_t[1] | d3[1] | s3[1]);
442 w3_t2[3] = (1 + pw_len + domain_len + 1 + salt_len) * 8;
452 sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
456 for (u32 i = 0; i < salt_iter; i++)
460 w0_t3[0] = digest[0];
461 w0_t3[1] = digest[1];
462 w0_t3[2] = digest[2];
463 w0_t3[3] = digest[3];
467 w1_t3[0] = digest[4];
468 w1_t3[1] = swap_workaround (salt_buf0[0]);
469 w1_t3[2] = swap_workaround (salt_buf0[1]);
470 w1_t3[3] = swap_workaround (salt_buf0[2]);
474 w2_t3[0] = swap_workaround (salt_buf0[3]);
475 w2_t3[1] = swap_workaround (salt_buf1[0]);
476 w2_t3[2] = swap_workaround (salt_buf1[1]);
477 w2_t3[3] = swap_workaround (salt_buf1[2]);
481 w3_t3[0] = swap_workaround (salt_buf1[3]);
484 w3_t3[3] = (20 + salt_len) * 8;
492 sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
495 const u32x r0 = digest[3];
496 const u32x r1 = digest[4];
497 const u32x r2 = digest[2];
498 const u32x r3 = digest[1];
500 #include VECT_COMPARE_M
504 extern "C" __global__ void __launch_bounds__ (256, 1) m08300_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 void *esalt_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)
508 extern "C" __global__ void __launch_bounds__ (256, 1) m08300_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 void *esalt_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)
512 extern "C" __global__ void __launch_bounds__ (256, 1) m08300_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 void *esalt_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)
518 const u32 lid = threadIdx.x;
524 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
526 if (gid >= gid_max) return;
530 wordl0[0] = pws[gid].i[ 0];
531 wordl0[1] = pws[gid].i[ 1];
532 wordl0[2] = pws[gid].i[ 2];
533 wordl0[3] = pws[gid].i[ 3];
537 wordl1[0] = pws[gid].i[ 4];
538 wordl1[1] = pws[gid].i[ 5];
539 wordl1[2] = pws[gid].i[ 6];
540 wordl1[3] = pws[gid].i[ 7];
556 const u32 pw_l_len = pws[gid].pw_len;
558 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
560 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
567 const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
571 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
572 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
573 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
574 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
578 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
579 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
580 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
581 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
583 const u32 salt_len = salt_bufs[salt_pos].salt_len;
587 domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
588 domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
589 domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
590 domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
594 domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
595 domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
596 domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
599 const u32 domain_len = salt_bufs[salt_pos].salt_buf_pc[ 7];
605 const u32 search[4] =
607 digests_buf[digests_offset].digest_buf[DGST_R0],
608 digests_buf[digests_offset].digest_buf[DGST_R1],
609 digests_buf[digests_offset].digest_buf[DGST_R2],
610 digests_buf[digests_offset].digest_buf[DGST_R3]
617 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
619 const u32 pw_r_len = c_combs[il_pos].pw_len;
621 const u32 pw_len = pw_l_len + pw_r_len;
628 wordr0[0] = c_combs[il_pos].i[0];
629 wordr0[1] = c_combs[il_pos].i[1];
630 wordr0[2] = c_combs[il_pos].i[2];
631 wordr0[3] = c_combs[il_pos].i[3];
632 wordr1[0] = c_combs[il_pos].i[4];
633 wordr1[1] = c_combs[il_pos].i[5];
634 wordr1[2] = c_combs[il_pos].i[6];
635 wordr1[3] = c_combs[il_pos].i[7];
645 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
647 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
655 w0[0] = wordl0[0] | wordr0[0];
656 w0[1] = wordl0[1] | wordr0[1];
657 w0[2] = wordl0[2] | wordr0[2];
658 w0[3] = wordl0[3] | wordr0[3];
659 w1[0] = wordl1[0] | wordr1[0];
660 w1[1] = wordl1[1] | wordr1[1];
661 w1[2] = wordl1[2] | wordr1[2];
662 w1[3] = wordl1[3] | wordr1[3];
663 w2[0] = wordl2[0] | wordr2[0];
664 w2[1] = wordl2[1] | wordr2[1];
665 w2[2] = wordl2[2] | wordr2[2];
666 w2[3] = wordl2[3] | wordr2[3];
667 w3[0] = wordl3[0] | wordr3[0];
668 w3[1] = wordl3[1] | wordr3[1];
669 w3[2] = wordl3[2] | wordr3[2];
670 w3[3] = wordl3[3] | wordr3[3];
700 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, 1);
702 w0_t[0] |= pw_len & 0xff;
710 s0[0] = salt_buf0[0];
711 s0[1] = salt_buf0[1];
712 s0[2] = salt_buf0[2];
713 s0[3] = salt_buf0[3];
717 s1[0] = salt_buf1[0];
718 s1[1] = salt_buf1[1];
719 s1[2] = salt_buf1[2];
720 s1[3] = salt_buf1[3];
736 switch_buffer_by_offset (s0, s1, s2, s3, 1 + pw_len + domain_len + 1);
740 d0[0] = domain_buf0[0];
741 d0[1] = domain_buf0[1];
742 d0[2] = domain_buf0[2];
743 d0[3] = domain_buf0[3];
747 d1[0] = domain_buf1[0];
748 d1[1] = domain_buf1[1];
749 d1[2] = domain_buf1[2];
766 switch_buffer_by_offset (d0, d1, d2, d3, 1 + pw_len);
774 w0_t2[0] = swap_workaround (w0_t[0] | d0[0] | s0[0]);
775 w0_t2[1] = swap_workaround (w0_t[1] | d0[1] | s0[1]);
776 w0_t2[2] = swap_workaround (w0_t[2] | d0[2] | s0[2]);
777 w0_t2[3] = swap_workaround (w0_t[3] | d0[3] | s0[3]);
781 w1_t2[0] = swap_workaround (w1_t[0] | d1[0] | s1[0]);
782 w1_t2[1] = swap_workaround (w1_t[1] | d1[1] | s1[1]);
783 w1_t2[2] = swap_workaround (w1_t[2] | d1[2] | s1[2]);
784 w1_t2[3] = swap_workaround (w1_t[3] | d1[3] | s1[3]);
788 w2_t2[0] = swap_workaround (w2_t[0] | d2[0] | s2[0]);
789 w2_t2[1] = swap_workaround (w2_t[1] | d2[1] | s2[1]);
790 w2_t2[2] = swap_workaround (w2_t[2] | d2[2] | s2[2]);
791 w2_t2[3] = swap_workaround (w2_t[3] | d2[3] | s2[3]);
795 w3_t2[0] = swap_workaround (w3_t[0] | d3[0] | s3[0]);
796 w3_t2[1] = swap_workaround (w3_t[1] | d3[1] | s3[1]);
798 w3_t2[3] = (1 + pw_len + domain_len + 1 + salt_len) * 8;
808 sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
812 for (u32 i = 0; i < salt_iter; i++)
816 w0_t3[0] = digest[0];
817 w0_t3[1] = digest[1];
818 w0_t3[2] = digest[2];
819 w0_t3[3] = digest[3];
823 w1_t3[0] = digest[4];
824 w1_t3[1] = swap_workaround (salt_buf0[0]);
825 w1_t3[2] = swap_workaround (salt_buf0[1]);
826 w1_t3[3] = swap_workaround (salt_buf0[2]);
830 w2_t3[0] = swap_workaround (salt_buf0[3]);
831 w2_t3[1] = swap_workaround (salt_buf1[0]);
832 w2_t3[2] = swap_workaround (salt_buf1[1]);
833 w2_t3[3] = swap_workaround (salt_buf1[2]);
837 w3_t3[0] = swap_workaround (salt_buf1[3]);
840 w3_t3[3] = (20 + salt_len) * 8;
848 sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
851 const u32x r0 = digest[3];
852 const u32x r1 = digest[4];
853 const u32x r2 = digest[2];
854 const u32x r3 = digest[1];
856 #include VECT_COMPARE_S
860 extern "C" __global__ void __launch_bounds__ (256, 1) m08300_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 void *esalt_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)
864 extern "C" __global__ void __launch_bounds__ (256, 1) m08300_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 void *esalt_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)