2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
24 #include "include/kernel_functions.c"
26 #include "common_nv.c"
29 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
30 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
34 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
35 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
39 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
40 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
43 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
71 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
72 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
73 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
74 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
75 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
76 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
77 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
78 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
79 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
80 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
81 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
82 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
83 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
84 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
85 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
86 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
87 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
88 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
89 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
90 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
95 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
96 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
97 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
98 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
99 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
100 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
101 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
102 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
103 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
104 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
105 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
106 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
107 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
108 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
109 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
110 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
111 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
112 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
113 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
114 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
119 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
120 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
121 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
122 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
123 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
124 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
125 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
126 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
127 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
128 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
129 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
130 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
131 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
132 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
133 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
134 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
135 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
136 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
137 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
138 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
143 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
144 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
145 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
146 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
147 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
148 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
149 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
150 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
151 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
152 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
153 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
154 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
155 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
156 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
157 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
158 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
159 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
160 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
161 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
162 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
171 __device__ static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
173 w0[0] = w0[0] ^ 0x36363636;
174 w0[1] = w0[1] ^ 0x36363636;
175 w0[2] = w0[2] ^ 0x36363636;
176 w0[3] = w0[3] ^ 0x36363636;
177 w1[0] = w1[0] ^ 0x36363636;
178 w1[1] = w1[1] ^ 0x36363636;
179 w1[2] = w1[2] ^ 0x36363636;
180 w1[3] = w1[3] ^ 0x36363636;
181 w2[0] = w2[0] ^ 0x36363636;
182 w2[1] = w2[1] ^ 0x36363636;
183 w2[2] = w2[2] ^ 0x36363636;
184 w2[3] = w2[3] ^ 0x36363636;
185 w3[0] = w3[0] ^ 0x36363636;
186 w3[1] = w3[1] ^ 0x36363636;
187 w3[2] = w3[2] ^ 0x36363636;
188 w3[3] = w3[3] ^ 0x36363636;
196 sha1_transform (w0, w1, w2, w3, ipad);
198 w0[0] = w0[0] ^ 0x6a6a6a6a;
199 w0[1] = w0[1] ^ 0x6a6a6a6a;
200 w0[2] = w0[2] ^ 0x6a6a6a6a;
201 w0[3] = w0[3] ^ 0x6a6a6a6a;
202 w1[0] = w1[0] ^ 0x6a6a6a6a;
203 w1[1] = w1[1] ^ 0x6a6a6a6a;
204 w1[2] = w1[2] ^ 0x6a6a6a6a;
205 w1[3] = w1[3] ^ 0x6a6a6a6a;
206 w2[0] = w2[0] ^ 0x6a6a6a6a;
207 w2[1] = w2[1] ^ 0x6a6a6a6a;
208 w2[2] = w2[2] ^ 0x6a6a6a6a;
209 w2[3] = w2[3] ^ 0x6a6a6a6a;
210 w3[0] = w3[0] ^ 0x6a6a6a6a;
211 w3[1] = w3[1] ^ 0x6a6a6a6a;
212 w3[2] = w3[2] ^ 0x6a6a6a6a;
213 w3[3] = w3[3] ^ 0x6a6a6a6a;
221 sha1_transform (w0, w1, w2, w3, opad);
224 __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])
232 sha1_transform (w0, w1, w2, w3, digest);
249 w3[3] = (64 + 20) * 8;
257 sha1_transform (w0, w1, w2, w3, digest);
260 __device__ __constant__ comb_t c_combs[1024];
262 extern "C" __global__ void __launch_bounds__ (256, 1) m00160_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)
268 const u32 lid = threadIdx.x;
274 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
276 if (gid >= gid_max) return;
280 wordl0[0] = pws[gid].i[ 0];
281 wordl0[1] = pws[gid].i[ 1];
282 wordl0[2] = pws[gid].i[ 2];
283 wordl0[3] = pws[gid].i[ 3];
287 wordl1[0] = pws[gid].i[ 4];
288 wordl1[1] = pws[gid].i[ 5];
289 wordl1[2] = pws[gid].i[ 6];
290 wordl1[3] = pws[gid].i[ 7];
306 const u32 pw_l_len = pws[gid].pw_len;
308 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
310 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
319 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
320 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
321 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
322 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
326 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
327 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
328 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
329 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
337 w0_t[0] = swap_workaround (salt_buf0[0]);
338 w0_t[1] = swap_workaround (salt_buf0[1]);
339 w0_t[2] = swap_workaround (salt_buf0[2]);
340 w0_t[3] = swap_workaround (salt_buf0[3]);
344 w1_t[0] = swap_workaround (salt_buf1[0]);
345 w1_t[1] = swap_workaround (salt_buf1[1]);
346 w1_t[2] = swap_workaround (salt_buf1[2]);
347 w1_t[3] = swap_workaround (salt_buf1[3]);
366 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
372 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
374 const u32 pw_r_len = c_combs[il_pos].pw_len;
376 const u32 pw_len = pw_l_len + pw_r_len;
380 wordr0[0] = c_combs[il_pos].i[0];
381 wordr0[1] = c_combs[il_pos].i[1];
382 wordr0[2] = c_combs[il_pos].i[2];
383 wordr0[3] = c_combs[il_pos].i[3];
387 wordr1[0] = c_combs[il_pos].i[4];
388 wordr1[1] = c_combs[il_pos].i[5];
389 wordr1[2] = c_combs[il_pos].i[6];
390 wordr1[3] = c_combs[il_pos].i[7];
406 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
408 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
413 w0[0] = wordl0[0] | wordr0[0];
414 w0[1] = wordl0[1] | wordr0[1];
415 w0[2] = wordl0[2] | wordr0[2];
416 w0[3] = wordl0[3] | wordr0[3];
420 w1[0] = wordl1[0] | wordr1[0];
421 w1[1] = wordl1[1] | wordr1[1];
422 w1[2] = wordl1[2] | wordr1[2];
423 w1[3] = wordl1[3] | wordr1[3];
427 w2[0] = wordl2[0] | wordr2[0];
428 w2[1] = wordl2[1] | wordr2[1];
429 w2[2] = wordl2[2] | wordr2[2];
430 w2[3] = wordl2[3] | wordr2[3];
434 w3[0] = wordl3[0] | wordr3[0];
435 w3[1] = wordl3[1] | wordr3[1];
439 append_0x80_4 (w0, w1, w2, w3, pw_len);
441 w0_t[0] = swap_workaround (w0[0]);
442 w0_t[1] = swap_workaround (w0[1]);
443 w0_t[2] = swap_workaround (w0[2]);
444 w0_t[3] = swap_workaround (w0[3]);
445 w1_t[0] = swap_workaround (w1[0]);
446 w1_t[1] = swap_workaround (w1[1]);
447 w1_t[2] = swap_workaround (w1[2]);
448 w1_t[3] = swap_workaround (w1[3]);
449 w2_t[0] = swap_workaround (w2[0]);
450 w2_t[1] = swap_workaround (w2[1]);
451 w2_t[2] = swap_workaround (w2[2]);
452 w2_t[3] = swap_workaround (w2[3]);
453 w3_t[0] = swap_workaround (w3[0]);
454 w3_t[1] = swap_workaround (w3[1]);
456 w3_t[3] = (64 + pw_len) * 8;
460 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
462 const u32x r0 = digest[3];
463 const u32x r1 = digest[4];
464 const u32x r2 = digest[2];
465 const u32x r3 = digest[1];
467 #include VECT_COMPARE_M
471 extern "C" __global__ void __launch_bounds__ (256, 1) m00160_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
475 extern "C" __global__ void __launch_bounds__ (256, 1) m00160_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
479 extern "C" __global__ void __launch_bounds__ (256, 1) m00160_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)
485 const u32 lid = threadIdx.x;
491 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
493 if (gid >= gid_max) return;
497 wordl0[0] = pws[gid].i[ 0];
498 wordl0[1] = pws[gid].i[ 1];
499 wordl0[2] = pws[gid].i[ 2];
500 wordl0[3] = pws[gid].i[ 3];
504 wordl1[0] = pws[gid].i[ 4];
505 wordl1[1] = pws[gid].i[ 5];
506 wordl1[2] = pws[gid].i[ 6];
507 wordl1[3] = pws[gid].i[ 7];
523 const u32 pw_l_len = pws[gid].pw_len;
525 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
527 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
536 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
537 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
538 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
539 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
543 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
544 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
545 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
546 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
554 w0_t[0] = swap_workaround (salt_buf0[0]);
555 w0_t[1] = swap_workaround (salt_buf0[1]);
556 w0_t[2] = swap_workaround (salt_buf0[2]);
557 w0_t[3] = swap_workaround (salt_buf0[3]);
561 w1_t[0] = swap_workaround (salt_buf1[0]);
562 w1_t[1] = swap_workaround (salt_buf1[1]);
563 w1_t[2] = swap_workaround (salt_buf1[2]);
564 w1_t[3] = swap_workaround (salt_buf1[3]);
583 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
589 const u32 search[4] =
591 digests_buf[digests_offset].digest_buf[DGST_R0],
592 digests_buf[digests_offset].digest_buf[DGST_R1],
593 digests_buf[digests_offset].digest_buf[DGST_R2],
594 digests_buf[digests_offset].digest_buf[DGST_R3]
601 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
603 const u32 pw_r_len = c_combs[il_pos].pw_len;
605 const u32 pw_len = pw_l_len + pw_r_len;
609 wordr0[0] = c_combs[il_pos].i[0];
610 wordr0[1] = c_combs[il_pos].i[1];
611 wordr0[2] = c_combs[il_pos].i[2];
612 wordr0[3] = c_combs[il_pos].i[3];
616 wordr1[0] = c_combs[il_pos].i[4];
617 wordr1[1] = c_combs[il_pos].i[5];
618 wordr1[2] = c_combs[il_pos].i[6];
619 wordr1[3] = c_combs[il_pos].i[7];
635 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
637 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
642 w0[0] = wordl0[0] | wordr0[0];
643 w0[1] = wordl0[1] | wordr0[1];
644 w0[2] = wordl0[2] | wordr0[2];
645 w0[3] = wordl0[3] | wordr0[3];
649 w1[0] = wordl1[0] | wordr1[0];
650 w1[1] = wordl1[1] | wordr1[1];
651 w1[2] = wordl1[2] | wordr1[2];
652 w1[3] = wordl1[3] | wordr1[3];
656 w2[0] = wordl2[0] | wordr2[0];
657 w2[1] = wordl2[1] | wordr2[1];
658 w2[2] = wordl2[2] | wordr2[2];
659 w2[3] = wordl2[3] | wordr2[3];
663 w3[0] = wordl3[0] | wordr3[0];
664 w3[1] = wordl3[1] | wordr3[1];
668 append_0x80_4 (w0, w1, w2, w3, pw_len);
670 w0_t[0] = swap_workaround (w0[0]);
671 w0_t[1] = swap_workaround (w0[1]);
672 w0_t[2] = swap_workaround (w0[2]);
673 w0_t[3] = swap_workaround (w0[3]);
674 w1_t[0] = swap_workaround (w1[0]);
675 w1_t[1] = swap_workaround (w1[1]);
676 w1_t[2] = swap_workaround (w1[2]);
677 w1_t[3] = swap_workaround (w1[3]);
678 w2_t[0] = swap_workaround (w2[0]);
679 w2_t[1] = swap_workaround (w2[1]);
680 w2_t[2] = swap_workaround (w2[2]);
681 w2_t[3] = swap_workaround (w2[3]);
682 w3_t[0] = swap_workaround (w3[0]);
683 w3_t[1] = swap_workaround (w3[1]);
685 w3_t[3] = (64 + pw_len) * 8;
689 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
691 const u32x r0 = digest[3];
692 const u32x r1 = digest[4];
693 const u32x r2 = digest[2];
694 const u32x r3 = digest[1];
696 #include VECT_COMPARE_S
700 extern "C" __global__ void __launch_bounds__ (256, 1) m00160_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
704 extern "C" __global__ void __launch_bounds__ (256, 1) m00160_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)