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"
27 #include "include/rp_gpu.h"
31 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
32 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
36 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
37 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
41 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
42 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
45 __device__ static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
73 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
74 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
75 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
76 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
77 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
78 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
79 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
80 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
81 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
82 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
83 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
84 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
85 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
86 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
87 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
88 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
89 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
90 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
91 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
92 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
97 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
98 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
99 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
100 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
101 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
102 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
103 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
104 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
105 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
106 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
107 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
108 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
109 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
110 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
111 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
112 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
113 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
114 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
115 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
116 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
121 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
122 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
123 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
124 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
125 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
126 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
127 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
128 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
129 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
130 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
131 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
132 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
133 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
134 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
135 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
136 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
137 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
138 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
139 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
140 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
145 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
146 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
147 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
148 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
149 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
150 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
151 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
152 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
153 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
154 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
155 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
156 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
157 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
158 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
159 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
160 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
161 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
162 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
163 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
164 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
173 __device__ __constant__ gpu_rule_t c_rules[1024];
175 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
181 const u32 lid = threadIdx.x;
187 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
189 if (gid >= gid_max) return;
193 pw_buf0[0] = pws[gid].i[ 0];
194 pw_buf0[1] = pws[gid].i[ 1];
195 pw_buf0[2] = pws[gid].i[ 2];
196 pw_buf0[3] = pws[gid].i[ 3];
200 pw_buf1[0] = pws[gid].i[ 4];
201 pw_buf1[1] = pws[gid].i[ 5];
202 pw_buf1[2] = pws[gid].i[ 6];
203 pw_buf1[3] = pws[gid].i[ 7];
205 const u32 pw_len = pws[gid].pw_len;
213 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
214 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
215 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
216 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
218 const u32 salt_len = 16;
224 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
254 const u32 out_len = apply_rules (c_rules[il_pos].cmds, w0, w1, pw_len);
256 const u32 pw_salt_len = (out_len * 2) + salt_len;
258 append_0x80_2 (w0, w1, out_len);
265 make_unicode (w0, w0_t, w1_t);
266 make_unicode (w1, w2_t, w3_t);
268 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
270 w0_t[0] = salt_buf[0];
271 w0_t[1] = salt_buf[1];
272 w0_t[2] = salt_buf[2];
273 w0_t[3] = salt_buf[3];
274 w1_t[0] = swap_workaround (w1_t[0]);
275 w1_t[1] = swap_workaround (w1_t[1]);
276 w1_t[2] = swap_workaround (w1_t[2]);
277 w1_t[3] = swap_workaround (w1_t[3]);
278 w2_t[0] = swap_workaround (w2_t[0]);
279 w2_t[1] = swap_workaround (w2_t[1]);
280 w2_t[2] = swap_workaround (w2_t[2]);
281 w2_t[3] = swap_workaround (w2_t[3]);
282 w3_t[0] = swap_workaround (w3_t[0]);
283 w3_t[1] = swap_workaround (w3_t[1]);
285 w3_t[3] = pw_salt_len * 8;
295 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
303 w1_t[2] = 0x80000000;
312 w3_t[3] = (20 + 4) * 8;
320 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
322 u32x a = swap_workaround (digest[0]);
323 u32x b = swap_workaround (digest[1]) & 0xff;
330 #include VECT_COMPARE_M
334 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)
338 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)
342 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
348 const u32 lid = threadIdx.x;
354 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
356 if (gid >= gid_max) return;
360 pw_buf0[0] = pws[gid].i[ 0];
361 pw_buf0[1] = pws[gid].i[ 1];
362 pw_buf0[2] = pws[gid].i[ 2];
363 pw_buf0[3] = pws[gid].i[ 3];
367 pw_buf1[0] = pws[gid].i[ 4];
368 pw_buf1[1] = pws[gid].i[ 5];
369 pw_buf1[2] = pws[gid].i[ 6];
370 pw_buf1[3] = pws[gid].i[ 7];
372 const u32 pw_len = pws[gid].pw_len;
378 const u32 search[4] =
380 digests_buf[digests_offset].digest_buf[DGST_R0],
381 digests_buf[digests_offset].digest_buf[DGST_R1],
382 digests_buf[digests_offset].digest_buf[DGST_R2],
383 digests_buf[digests_offset].digest_buf[DGST_R3]
392 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
393 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
394 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
395 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
397 const u32 salt_len = 16;
403 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
433 const u32 out_len = apply_rules (c_rules[il_pos].cmds, w0, w1, pw_len);
435 const u32 pw_salt_len = (out_len * 2) + salt_len;
437 append_0x80_2 (w0, w1, out_len);
444 make_unicode (w0, w0_t, w1_t);
445 make_unicode (w1, w2_t, w3_t);
447 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
449 w0_t[0] = salt_buf[0];
450 w0_t[1] = salt_buf[1];
451 w0_t[2] = salt_buf[2];
452 w0_t[3] = salt_buf[3];
453 w1_t[0] = swap_workaround (w1_t[0]);
454 w1_t[1] = swap_workaround (w1_t[1]);
455 w1_t[2] = swap_workaround (w1_t[2]);
456 w1_t[3] = swap_workaround (w1_t[3]);
457 w2_t[0] = swap_workaround (w2_t[0]);
458 w2_t[1] = swap_workaround (w2_t[1]);
459 w2_t[2] = swap_workaround (w2_t[2]);
460 w2_t[3] = swap_workaround (w2_t[3]);
461 w3_t[0] = swap_workaround (w3_t[0]);
462 w3_t[1] = swap_workaround (w3_t[1]);
464 w3_t[3] = pw_salt_len * 8;
474 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
482 w1_t[2] = 0x80000000;
491 w3_t[3] = (20 + 4) * 8;
499 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
501 u32x a = swap_workaround (digest[0]);
502 u32x b = swap_workaround (digest[1]) & 0xff;
509 #include VECT_COMPARE_S
513 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)
517 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)