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__ __constant__ comb_t c_combs[1024];
173 extern "C" __global__ void __launch_bounds__ (256, 1) m09820_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 oldoffice34_t *oldoffice34_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)
179 const u32 lid = threadIdx.x;
185 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
187 if (gid >= gid_max) return;
191 wordl0[0] = pws[gid].i[ 0];
192 wordl0[1] = pws[gid].i[ 1];
193 wordl0[2] = pws[gid].i[ 2];
194 wordl0[3] = pws[gid].i[ 3];
198 wordl1[0] = pws[gid].i[ 4];
199 wordl1[1] = pws[gid].i[ 5];
200 wordl1[2] = pws[gid].i[ 6];
201 wordl1[3] = pws[gid].i[ 7];
217 const u32 pw_l_len = pws[gid].pw_len;
219 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
221 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
230 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
231 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
232 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
233 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
235 const u32 salt_len = 16;
241 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
243 const u32 pw_r_len = c_combs[il_pos].pw_len;
245 const u32 pw_len = pw_l_len + pw_r_len;
247 const u32 pw_salt_len = (pw_len * 2) + salt_len;
251 wordr0[0] = c_combs[il_pos].i[0];
252 wordr0[1] = c_combs[il_pos].i[1];
253 wordr0[2] = c_combs[il_pos].i[2];
254 wordr0[3] = c_combs[il_pos].i[3];
258 wordr1[0] = c_combs[il_pos].i[4];
259 wordr1[1] = c_combs[il_pos].i[5];
260 wordr1[2] = c_combs[il_pos].i[6];
261 wordr1[3] = c_combs[il_pos].i[7];
277 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
279 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
284 w0[0] = wordl0[0] | wordr0[0];
285 w0[1] = wordl0[1] | wordr0[1];
286 w0[2] = wordl0[2] | wordr0[2];
287 w0[3] = wordl0[3] | wordr0[3];
291 w1[0] = wordl1[0] | wordr1[0];
292 w1[1] = wordl1[1] | wordr1[1];
293 w1[2] = wordl1[2] | wordr1[2];
294 w1[3] = wordl1[3] | wordr1[3];
298 w2[0] = wordl2[0] | wordr2[0];
299 w2[1] = wordl2[1] | wordr2[1];
300 w2[2] = wordl2[2] | wordr2[2];
301 w2[3] = wordl2[3] | wordr2[3];
305 w3[0] = wordl3[0] | wordr3[0];
306 w3[1] = wordl3[1] | wordr3[1];
310 append_0x80_2 (w0, w1, pw_len);
317 make_unicode (w0, w0_t, w1_t);
318 make_unicode (w1, w2_t, w3_t);
320 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
322 w0_t[0] = salt_buf[0];
323 w0_t[1] = salt_buf[1];
324 w0_t[2] = salt_buf[2];
325 w0_t[3] = salt_buf[3];
326 w1_t[0] = swap_workaround (w1_t[0]);
327 w1_t[1] = swap_workaround (w1_t[1]);
328 w1_t[2] = swap_workaround (w1_t[2]);
329 w1_t[3] = swap_workaround (w1_t[3]);
330 w2_t[0] = swap_workaround (w2_t[0]);
331 w2_t[1] = swap_workaround (w2_t[1]);
332 w2_t[2] = swap_workaround (w2_t[2]);
333 w2_t[3] = swap_workaround (w2_t[3]);
334 w3_t[0] = swap_workaround (w3_t[0]);
335 w3_t[1] = swap_workaround (w3_t[1]);
337 w3_t[3] = pw_salt_len * 8;
347 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
355 w1_t[2] = 0x80000000;
364 w3_t[3] = (20 + 4) * 8;
372 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
374 u32x a = swap_workaround (digest[0]);
375 u32x b = swap_workaround (digest[1]) & 0xff;
382 #include VECT_COMPARE_M
386 extern "C" __global__ void __launch_bounds__ (256, 1) m09820_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 oldoffice34_t *oldoffice34_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)
390 extern "C" __global__ void __launch_bounds__ (256, 1) m09820_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 oldoffice34_t *oldoffice34_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)
394 extern "C" __global__ void __launch_bounds__ (256, 1) m09820_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 oldoffice34_t *oldoffice34_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)
400 const u32 lid = threadIdx.x;
406 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
408 if (gid >= gid_max) return;
412 wordl0[0] = pws[gid].i[ 0];
413 wordl0[1] = pws[gid].i[ 1];
414 wordl0[2] = pws[gid].i[ 2];
415 wordl0[3] = pws[gid].i[ 3];
419 wordl1[0] = pws[gid].i[ 4];
420 wordl1[1] = pws[gid].i[ 5];
421 wordl1[2] = pws[gid].i[ 6];
422 wordl1[3] = pws[gid].i[ 7];
438 const u32 pw_l_len = pws[gid].pw_len;
440 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
442 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, c_combs[0].pw_len);
449 const u32 search[4] =
451 digests_buf[digests_offset].digest_buf[DGST_R0],
452 digests_buf[digests_offset].digest_buf[DGST_R1],
453 digests_buf[digests_offset].digest_buf[DGST_R2],
454 digests_buf[digests_offset].digest_buf[DGST_R3]
463 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
464 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
465 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
466 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
468 const u32 salt_len = 16;
474 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
476 const u32 pw_r_len = c_combs[il_pos].pw_len;
478 const u32 pw_len = pw_l_len + pw_r_len;
480 const u32 pw_salt_len = (pw_len * 2) + salt_len;
484 wordr0[0] = c_combs[il_pos].i[0];
485 wordr0[1] = c_combs[il_pos].i[1];
486 wordr0[2] = c_combs[il_pos].i[2];
487 wordr0[3] = c_combs[il_pos].i[3];
491 wordr1[0] = c_combs[il_pos].i[4];
492 wordr1[1] = c_combs[il_pos].i[5];
493 wordr1[2] = c_combs[il_pos].i[6];
494 wordr1[3] = c_combs[il_pos].i[7];
510 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
512 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
517 w0[0] = wordl0[0] | wordr0[0];
518 w0[1] = wordl0[1] | wordr0[1];
519 w0[2] = wordl0[2] | wordr0[2];
520 w0[3] = wordl0[3] | wordr0[3];
524 w1[0] = wordl1[0] | wordr1[0];
525 w1[1] = wordl1[1] | wordr1[1];
526 w1[2] = wordl1[2] | wordr1[2];
527 w1[3] = wordl1[3] | wordr1[3];
531 w2[0] = wordl2[0] | wordr2[0];
532 w2[1] = wordl2[1] | wordr2[1];
533 w2[2] = wordl2[2] | wordr2[2];
534 w2[3] = wordl2[3] | wordr2[3];
538 w3[0] = wordl3[0] | wordr3[0];
539 w3[1] = wordl3[1] | wordr3[1];
543 append_0x80_2 (w0, w1, pw_len);
550 make_unicode (w0, w0_t, w1_t);
551 make_unicode (w1, w2_t, w3_t);
553 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
555 w0_t[0] = salt_buf[0];
556 w0_t[1] = salt_buf[1];
557 w0_t[2] = salt_buf[2];
558 w0_t[3] = salt_buf[3];
559 w1_t[0] = swap_workaround (w1_t[0]);
560 w1_t[1] = swap_workaround (w1_t[1]);
561 w1_t[2] = swap_workaround (w1_t[2]);
562 w1_t[3] = swap_workaround (w1_t[3]);
563 w2_t[0] = swap_workaround (w2_t[0]);
564 w2_t[1] = swap_workaround (w2_t[1]);
565 w2_t[2] = swap_workaround (w2_t[2]);
566 w2_t[3] = swap_workaround (w2_t[3]);
567 w3_t[0] = swap_workaround (w3_t[0]);
568 w3_t[1] = swap_workaround (w3_t[1]);
570 w3_t[3] = pw_salt_len * 8;
580 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
588 w1_t[2] = 0x80000000;
597 w3_t[3] = (20 + 4) * 8;
605 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
607 u32x a = swap_workaround (digest[0]);
608 u32x b = swap_workaround (digest[1]) & 0xff;
615 #include VECT_COMPARE_S
619 extern "C" __global__ void __launch_bounds__ (256, 1) m09820_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 oldoffice34_t *oldoffice34_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)
623 extern "C" __global__ void __launch_bounds__ (256, 1) m09820_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 oldoffice34_t *oldoffice34_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)