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"
27 #include "common_nv.c"
30 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
33 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
61 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
62 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
63 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
64 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
65 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
66 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
67 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
68 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
69 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
70 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
71 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
72 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
73 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
74 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
75 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
76 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
77 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
78 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
79 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
80 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
85 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
86 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
87 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
88 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
89 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
90 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
91 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
92 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
93 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
94 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
95 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
96 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
97 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
98 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
99 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
100 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
101 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
102 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
103 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
104 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
109 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
110 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
111 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
112 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
113 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
114 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
115 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
116 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
117 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
118 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
119 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
120 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
121 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
122 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
123 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
124 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
125 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
126 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
127 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
128 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
133 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
134 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
135 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
136 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
137 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
138 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
139 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
140 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
141 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
142 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
143 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
144 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
145 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
146 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
147 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
148 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
149 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
150 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
151 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
152 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
161 __device__ static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
163 w0[0] = w0[0] ^ 0x36363636;
164 w0[1] = w0[1] ^ 0x36363636;
165 w0[2] = w0[2] ^ 0x36363636;
166 w0[3] = w0[3] ^ 0x36363636;
167 w1[0] = w1[0] ^ 0x36363636;
168 w1[1] = w1[1] ^ 0x36363636;
169 w1[2] = w1[2] ^ 0x36363636;
170 w1[3] = w1[3] ^ 0x36363636;
171 w2[0] = w2[0] ^ 0x36363636;
172 w2[1] = w2[1] ^ 0x36363636;
173 w2[2] = w2[2] ^ 0x36363636;
174 w2[3] = w2[3] ^ 0x36363636;
175 w3[0] = w3[0] ^ 0x36363636;
176 w3[1] = w3[1] ^ 0x36363636;
177 w3[2] = w3[2] ^ 0x36363636;
178 w3[3] = w3[3] ^ 0x36363636;
186 sha1_transform (w0, w1, w2, w3, ipad);
188 w0[0] = w0[0] ^ 0x6a6a6a6a;
189 w0[1] = w0[1] ^ 0x6a6a6a6a;
190 w0[2] = w0[2] ^ 0x6a6a6a6a;
191 w0[3] = w0[3] ^ 0x6a6a6a6a;
192 w1[0] = w1[0] ^ 0x6a6a6a6a;
193 w1[1] = w1[1] ^ 0x6a6a6a6a;
194 w1[2] = w1[2] ^ 0x6a6a6a6a;
195 w1[3] = w1[3] ^ 0x6a6a6a6a;
196 w2[0] = w2[0] ^ 0x6a6a6a6a;
197 w2[1] = w2[1] ^ 0x6a6a6a6a;
198 w2[2] = w2[2] ^ 0x6a6a6a6a;
199 w2[3] = w2[3] ^ 0x6a6a6a6a;
200 w3[0] = w3[0] ^ 0x6a6a6a6a;
201 w3[1] = w3[1] ^ 0x6a6a6a6a;
202 w3[2] = w3[2] ^ 0x6a6a6a6a;
203 w3[3] = w3[3] ^ 0x6a6a6a6a;
211 sha1_transform (w0, w1, w2, w3, opad);
214 __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])
222 sha1_transform (w0, w1, w2, w3, digest);
239 w3[3] = (64 + 20) * 8;
247 sha1_transform (w0, w1, w2, w3, digest);
250 extern "C" __global__ void __launch_bounds__ (256, 1) m12000_init (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, pbkdf2_sha1_tmp_t *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 pbkdf2_sha1_t *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)
256 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
258 if (gid >= gid_max) return;
262 w0[0] = swap_workaround (pws[gid].i[ 0]);
263 w0[1] = swap_workaround (pws[gid].i[ 1]);
264 w0[2] = swap_workaround (pws[gid].i[ 2]);
265 w0[3] = swap_workaround (pws[gid].i[ 3]);
269 w1[0] = swap_workaround (pws[gid].i[ 4]);
270 w1[1] = swap_workaround (pws[gid].i[ 5]);
271 w1[2] = swap_workaround (pws[gid].i[ 6]);
272 w1[3] = swap_workaround (pws[gid].i[ 7]);
276 w2[0] = swap_workaround (pws[gid].i[ 8]);
277 w2[1] = swap_workaround (pws[gid].i[ 9]);
278 w2[2] = swap_workaround (pws[gid].i[10]);
279 w2[3] = swap_workaround (pws[gid].i[11]);
283 w3[0] = swap_workaround (pws[gid].i[12]);
284 w3[1] = swap_workaround (pws[gid].i[13]);
285 w3[2] = swap_workaround (pws[gid].i[14]);
286 w3[3] = swap_workaround (pws[gid].i[15]);
292 const u32 salt_len = salt_bufs[salt_pos].salt_len;
299 esalt_buf0[0] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 0]);
300 esalt_buf0[1] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 1]);
301 esalt_buf0[2] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 2]);
302 esalt_buf0[3] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 3]);
303 esalt_buf1[0] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 4]);
304 esalt_buf1[1] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 5]);
305 esalt_buf1[2] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 6]);
306 esalt_buf1[3] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 7]);
307 esalt_buf2[0] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 8]);
308 esalt_buf2[1] = swap_workaround (esalt_bufs[salt_pos].salt_buf[ 9]);
309 esalt_buf2[2] = swap_workaround (esalt_bufs[salt_pos].salt_buf[10]);
310 esalt_buf2[3] = swap_workaround (esalt_bufs[salt_pos].salt_buf[11]);
311 esalt_buf3[0] = swap_workaround (esalt_bufs[salt_pos].salt_buf[12]);
312 esalt_buf3[1] = swap_workaround (esalt_bufs[salt_pos].salt_buf[13]);
314 esalt_buf3[3] = (64 + salt_len + 4) * 8;
319 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
321 tmps[gid].ipad[0] = ipad[0];
322 tmps[gid].ipad[1] = ipad[1];
323 tmps[gid].ipad[2] = ipad[2];
324 tmps[gid].ipad[3] = ipad[3];
325 tmps[gid].ipad[4] = ipad[4];
327 tmps[gid].opad[0] = opad[0];
328 tmps[gid].opad[1] = opad[1];
329 tmps[gid].opad[2] = opad[2];
330 tmps[gid].opad[3] = opad[3];
331 tmps[gid].opad[4] = opad[4];
333 for (u32 i = 0, j = 1; i < 5; i += 5, j += 1)
337 hmac_sha1_run (esalt_buf0, esalt_buf1, esalt_buf2, esalt_buf3, ipad, opad, dgst);
339 tmps[gid].dgst[i + 0] = dgst[0];
340 tmps[gid].dgst[i + 1] = dgst[1];
341 tmps[gid].dgst[i + 2] = dgst[2];
342 tmps[gid].dgst[i + 3] = dgst[3];
343 tmps[gid].dgst[i + 4] = dgst[4];
345 tmps[gid].out[i + 0] = dgst[0];
346 tmps[gid].out[i + 1] = dgst[1];
347 tmps[gid].out[i + 2] = dgst[2];
348 tmps[gid].out[i + 3] = dgst[3];
349 tmps[gid].out[i + 4] = dgst[4];
353 extern "C" __global__ void __launch_bounds__ (256, 1) m12000_loop (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, pbkdf2_sha1_tmp_t *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 pbkdf2_sha1_t *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)
355 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
357 if (gid >= gid_max) return;
361 ipad[0] = tmps[gid].ipad[0];
362 ipad[1] = tmps[gid].ipad[1];
363 ipad[2] = tmps[gid].ipad[2];
364 ipad[3] = tmps[gid].ipad[3];
365 ipad[4] = tmps[gid].ipad[4];
369 opad[0] = tmps[gid].opad[0];
370 opad[1] = tmps[gid].opad[1];
371 opad[2] = tmps[gid].opad[2];
372 opad[3] = tmps[gid].opad[3];
373 opad[4] = tmps[gid].opad[4];
375 for (u32 i = 0; i < 5; i += 5)
379 dgst[0] = tmps[gid].dgst[i + 0];
380 dgst[1] = tmps[gid].dgst[i + 1];
381 dgst[2] = tmps[gid].dgst[i + 2];
382 dgst[3] = tmps[gid].dgst[i + 3];
383 dgst[4] = tmps[gid].dgst[i + 4];
387 out[0] = tmps[gid].out[i + 0];
388 out[1] = tmps[gid].out[i + 1];
389 out[2] = tmps[gid].out[i + 2];
390 out[3] = tmps[gid].out[i + 3];
391 out[4] = tmps[gid].out[i + 4];
393 for (u32 j = 0; j < loop_cnt; j++)
415 w3[3] = (64 + 20) * 8;
417 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
426 tmps[gid].dgst[i + 0] = dgst[0];
427 tmps[gid].dgst[i + 1] = dgst[1];
428 tmps[gid].dgst[i + 2] = dgst[2];
429 tmps[gid].dgst[i + 3] = dgst[3];
430 tmps[gid].dgst[i + 4] = dgst[4];
432 tmps[gid].out[i + 0] = out[0];
433 tmps[gid].out[i + 1] = out[1];
434 tmps[gid].out[i + 2] = out[2];
435 tmps[gid].out[i + 3] = out[3];
436 tmps[gid].out[i + 4] = out[4];
440 extern "C" __global__ void __launch_bounds__ (256, 1) m12000_comp (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, pbkdf2_sha1_tmp_t *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 pbkdf2_sha1_t *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)
446 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
448 if (gid >= gid_max) return;
450 const u32 lid = threadIdx.x;
452 const u32x r0 = tmps[gid].out[DGST_R0];
453 const u32x r1 = tmps[gid].out[DGST_R1];
454 const u32x r2 = tmps[gid].out[DGST_R2];
455 const u32x r3 = tmps[gid].out[DGST_R3];
459 #include VECT_COMPARE_M