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 sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
66 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
67 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
68 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
69 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
70 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
71 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
72 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
73 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
74 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
75 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
76 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
77 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
78 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
79 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
80 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
81 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
82 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
83 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
84 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
85 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
90 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
91 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
92 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
93 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
94 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
95 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
96 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
97 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
98 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
99 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
100 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
101 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
102 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
103 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
104 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
105 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
106 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
107 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
108 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
109 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
114 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
115 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
116 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
117 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
118 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
119 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
120 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
121 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
122 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
123 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
124 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
125 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
126 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
127 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
128 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
129 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
130 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
131 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
132 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
133 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
138 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
139 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
140 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
141 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
142 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
143 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
144 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
145 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
146 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
147 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
148 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
149 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
150 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
151 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
152 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
153 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
154 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
155 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
156 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
157 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
166 __device__ static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
168 w0[0] = w0[0] ^ 0x36363636;
169 w0[1] = w0[1] ^ 0x36363636;
170 w0[2] = w0[2] ^ 0x36363636;
171 w0[3] = w0[3] ^ 0x36363636;
172 w1[0] = w1[0] ^ 0x36363636;
173 w1[1] = w1[1] ^ 0x36363636;
174 w1[2] = w1[2] ^ 0x36363636;
175 w1[3] = w1[3] ^ 0x36363636;
176 w2[0] = w2[0] ^ 0x36363636;
177 w2[1] = w2[1] ^ 0x36363636;
178 w2[2] = w2[2] ^ 0x36363636;
179 w2[3] = w2[3] ^ 0x36363636;
180 w3[0] = w3[0] ^ 0x36363636;
181 w3[1] = w3[1] ^ 0x36363636;
182 w3[2] = w3[2] ^ 0x36363636;
183 w3[3] = w3[3] ^ 0x36363636;
191 sha1_transform (w0, w1, w2, w3, ipad);
193 w0[0] = w0[0] ^ 0x6a6a6a6a;
194 w0[1] = w0[1] ^ 0x6a6a6a6a;
195 w0[2] = w0[2] ^ 0x6a6a6a6a;
196 w0[3] = w0[3] ^ 0x6a6a6a6a;
197 w1[0] = w1[0] ^ 0x6a6a6a6a;
198 w1[1] = w1[1] ^ 0x6a6a6a6a;
199 w1[2] = w1[2] ^ 0x6a6a6a6a;
200 w1[3] = w1[3] ^ 0x6a6a6a6a;
201 w2[0] = w2[0] ^ 0x6a6a6a6a;
202 w2[1] = w2[1] ^ 0x6a6a6a6a;
203 w2[2] = w2[2] ^ 0x6a6a6a6a;
204 w2[3] = w2[3] ^ 0x6a6a6a6a;
205 w3[0] = w3[0] ^ 0x6a6a6a6a;
206 w3[1] = w3[1] ^ 0x6a6a6a6a;
207 w3[2] = w3[2] ^ 0x6a6a6a6a;
208 w3[3] = w3[3] ^ 0x6a6a6a6a;
216 sha1_transform (w0, w1, w2, w3, opad);
219 __device__ static void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
227 sha1_transform (w0, w1, w2, w3, digest);
244 w3[3] = (64 + 20) * 8;
252 sha1_transform (w0, w1, w2, w3, digest);
255 __device__ __constant__ comb_t c_combs[1024];
257 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
263 const u32 lid = threadIdx.x;
269 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
271 if (gid >= gid_max) return;
275 wordl0[0] = pws[gid].i[ 0];
276 wordl0[1] = pws[gid].i[ 1];
277 wordl0[2] = pws[gid].i[ 2];
278 wordl0[3] = pws[gid].i[ 3];
282 wordl1[0] = pws[gid].i[ 4];
283 wordl1[1] = pws[gid].i[ 5];
284 wordl1[2] = pws[gid].i[ 6];
285 wordl1[3] = pws[gid].i[ 7];
301 const u32 pw_l_len = pws[gid].pw_len;
303 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
305 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
312 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
318 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
320 const u32 pw_r_len = c_combs[il_pos].pw_len;
322 const u32 pw_len = pw_l_len + pw_r_len;
326 wordr0[0] = c_combs[il_pos].i[0];
327 wordr0[1] = c_combs[il_pos].i[1];
328 wordr0[2] = c_combs[il_pos].i[2];
329 wordr0[3] = c_combs[il_pos].i[3];
333 wordr1[0] = c_combs[il_pos].i[4];
334 wordr1[1] = c_combs[il_pos].i[5];
335 wordr1[2] = c_combs[il_pos].i[6];
336 wordr1[3] = c_combs[il_pos].i[7];
352 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
354 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
359 w0[0] = wordl0[0] | wordr0[0];
360 w0[1] = wordl0[1] | wordr0[1];
361 w0[2] = wordl0[2] | wordr0[2];
362 w0[3] = wordl0[3] | wordr0[3];
366 w1[0] = wordl1[0] | wordr1[0];
367 w1[1] = wordl1[1] | wordr1[1];
368 w1[2] = wordl1[2] | wordr1[2];
369 w1[3] = wordl1[3] | wordr1[3];
373 w2[0] = wordl2[0] | wordr2[0];
374 w2[1] = wordl2[1] | wordr2[1];
375 w2[2] = wordl2[2] | wordr2[2];
376 w2[3] = wordl2[3] | wordr2[3];
380 w3[0] = wordl3[0] | wordr3[0];
381 w3[1] = wordl3[1] | wordr3[1];
382 w3[2] = wordl3[2] | wordr3[2];
383 w3[3] = wordl3[3] | wordr3[3];
391 w0_t[0] = swap_workaround (w0[0]);
392 w0_t[1] = swap_workaround (w0[1]);
393 w0_t[2] = swap_workaround (w0[2]);
394 w0_t[3] = swap_workaround (w0[3]);
398 w1_t[0] = swap_workaround (w1[0]);
399 w1_t[1] = swap_workaround (w1[1]);
400 w1_t[2] = swap_workaround (w1[2]);
401 w1_t[3] = swap_workaround (w1[3]);
420 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
422 int esalt_size = esalt_len;
427 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
429 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
430 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
431 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
432 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
433 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
434 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
435 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
436 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
437 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
438 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
439 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
440 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
441 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
442 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
443 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
444 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
446 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
449 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
450 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
451 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
452 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
453 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
454 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
455 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
456 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
457 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
458 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
459 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
460 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
461 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
462 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
464 w3_t[3] = (64 + esalt_size) * 8;
468 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
470 const u32x r0 = digest[3];
471 const u32x r1 = digest[4];
472 const u32x r2 = digest[2];
473 const u32x r3 = digest[1];
475 #include VECT_COMPARE_M
479 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
483 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
487 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
493 const u32 lid = threadIdx.x;
499 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
501 if (gid >= gid_max) return;
505 wordl0[0] = pws[gid].i[ 0];
506 wordl0[1] = pws[gid].i[ 1];
507 wordl0[2] = pws[gid].i[ 2];
508 wordl0[3] = pws[gid].i[ 3];
512 wordl1[0] = pws[gid].i[ 4];
513 wordl1[1] = pws[gid].i[ 5];
514 wordl1[2] = pws[gid].i[ 6];
515 wordl1[3] = pws[gid].i[ 7];
531 const u32 pw_l_len = pws[gid].pw_len;
533 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
535 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
542 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
548 const u32 search[4] =
550 digests_buf[digests_offset].digest_buf[DGST_R0],
551 digests_buf[digests_offset].digest_buf[DGST_R1],
552 digests_buf[digests_offset].digest_buf[DGST_R2],
553 digests_buf[digests_offset].digest_buf[DGST_R3]
560 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
562 const u32 pw_r_len = c_combs[il_pos].pw_len;
564 const u32 pw_len = pw_l_len + pw_r_len;
568 wordr0[0] = c_combs[il_pos].i[0];
569 wordr0[1] = c_combs[il_pos].i[1];
570 wordr0[2] = c_combs[il_pos].i[2];
571 wordr0[3] = c_combs[il_pos].i[3];
575 wordr1[0] = c_combs[il_pos].i[4];
576 wordr1[1] = c_combs[il_pos].i[5];
577 wordr1[2] = c_combs[il_pos].i[6];
578 wordr1[3] = c_combs[il_pos].i[7];
594 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
596 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
601 w0[0] = wordl0[0] | wordr0[0];
602 w0[1] = wordl0[1] | wordr0[1];
603 w0[2] = wordl0[2] | wordr0[2];
604 w0[3] = wordl0[3] | wordr0[3];
608 w1[0] = wordl1[0] | wordr1[0];
609 w1[1] = wordl1[1] | wordr1[1];
610 w1[2] = wordl1[2] | wordr1[2];
611 w1[3] = wordl1[3] | wordr1[3];
615 w2[0] = wordl2[0] | wordr2[0];
616 w2[1] = wordl2[1] | wordr2[1];
617 w2[2] = wordl2[2] | wordr2[2];
618 w2[3] = wordl2[3] | wordr2[3];
622 w3[0] = wordl3[0] | wordr3[0];
623 w3[1] = wordl3[1] | wordr3[1];
624 w3[2] = wordl3[2] | wordr3[2];
625 w3[3] = wordl3[3] | wordr3[3];
633 w0_t[0] = swap_workaround (w0[0]);
634 w0_t[1] = swap_workaround (w0[1]);
635 w0_t[2] = swap_workaround (w0[2]);
636 w0_t[3] = swap_workaround (w0[3]);
640 w1_t[0] = swap_workaround (w1[0]);
641 w1_t[1] = swap_workaround (w1[1]);
642 w1_t[2] = swap_workaround (w1[2]);
643 w1_t[3] = swap_workaround (w1[3]);
662 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
664 int esalt_size = esalt_len;
669 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
671 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
672 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
673 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
674 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
675 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
676 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
677 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
678 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
679 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
680 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
681 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
682 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
683 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
684 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
685 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
686 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
688 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
691 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
692 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
693 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
694 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
695 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
696 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
697 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
698 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
699 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
700 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
701 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
702 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
703 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
704 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
706 w3_t[3] = (64 + esalt_size) * 8;
710 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
712 const u32x r0 = digest[3];
713 const u32x r1 = digest[4];
714 const u32x r2 = digest[2];
715 const u32x r3 = digest[1];
717 #include VECT_COMPARE_S
721 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)
725 extern "C" __global__ void __launch_bounds__ (256, 1) m07300_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 rakp_t *rakp_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)