2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
28 #include "include/kernel_functions.c"
29 #include "types_amd.c"
30 #include "common_amd.c"
33 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
34 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
38 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
39 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
43 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
44 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
47 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
75 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
76 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
77 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
78 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
79 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
80 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
81 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
82 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
83 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
84 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
85 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
86 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
87 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
88 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
89 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
90 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
91 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
92 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
93 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
94 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
99 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
100 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
101 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
102 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
103 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
104 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
105 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
106 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
107 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
108 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
109 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
110 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
111 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
112 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
113 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
114 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
115 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
116 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
117 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
118 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
123 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
124 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
125 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
126 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
127 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
128 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
129 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
130 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
131 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
132 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
133 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
134 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
135 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
136 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
137 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
138 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
139 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
140 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
141 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
142 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
147 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
148 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
149 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
150 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
151 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
152 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
153 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
154 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
155 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
156 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
157 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
158 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
159 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
160 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
161 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
162 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
163 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
164 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
165 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
166 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
175 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_m04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global 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)
181 const u32 lid = get_local_id (0);
187 const u32 gid = get_global_id (0);
189 if (gid >= gid_max) return;
193 wordl0[0] = pws[gid].i[ 0];
194 wordl0[1] = pws[gid].i[ 1];
195 wordl0[2] = pws[gid].i[ 2];
196 wordl0[3] = pws[gid].i[ 3];
200 wordl1[0] = pws[gid].i[ 4];
201 wordl1[1] = pws[gid].i[ 5];
202 wordl1[2] = pws[gid].i[ 6];
203 wordl1[3] = pws[gid].i[ 7];
219 const u32 pw_l_len = pws[gid].pw_len;
221 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
223 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
232 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
233 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
234 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
235 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
237 const u32 salt_len = 16;
243 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
245 const u32 pw_r_len = combs_buf[il_pos].pw_len;
247 const u32 pw_len = pw_l_len + pw_r_len;
249 const u32 pw_salt_len = (pw_len * 2) + salt_len;
253 wordr0[0] = combs_buf[il_pos].i[0];
254 wordr0[1] = combs_buf[il_pos].i[1];
255 wordr0[2] = combs_buf[il_pos].i[2];
256 wordr0[3] = combs_buf[il_pos].i[3];
260 wordr1[0] = combs_buf[il_pos].i[4];
261 wordr1[1] = combs_buf[il_pos].i[5];
262 wordr1[2] = combs_buf[il_pos].i[6];
263 wordr1[3] = combs_buf[il_pos].i[7];
279 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
281 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
286 w0[0] = wordl0[0] | wordr0[0];
287 w0[1] = wordl0[1] | wordr0[1];
288 w0[2] = wordl0[2] | wordr0[2];
289 w0[3] = wordl0[3] | wordr0[3];
293 w1[0] = wordl1[0] | wordr1[0];
294 w1[1] = wordl1[1] | wordr1[1];
295 w1[2] = wordl1[2] | wordr1[2];
296 w1[3] = wordl1[3] | wordr1[3];
300 w2[0] = wordl2[0] | wordr2[0];
301 w2[1] = wordl2[1] | wordr2[1];
302 w2[2] = wordl2[2] | wordr2[2];
303 w2[3] = wordl2[3] | wordr2[3];
307 w3[0] = wordl3[0] | wordr3[0];
308 w3[1] = wordl3[1] | wordr3[1];
312 append_0x80_2 (w0, w1, pw_len);
319 make_unicode (w0, w0_t, w1_t);
320 make_unicode (w1, w2_t, w3_t);
322 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
324 w0_t[0] = salt_buf[0];
325 w0_t[1] = salt_buf[1];
326 w0_t[2] = salt_buf[2];
327 w0_t[3] = salt_buf[3];
328 w1_t[0] = swap_workaround (w1_t[0]);
329 w1_t[1] = swap_workaround (w1_t[1]);
330 w1_t[2] = swap_workaround (w1_t[2]);
331 w1_t[3] = swap_workaround (w1_t[3]);
332 w2_t[0] = swap_workaround (w2_t[0]);
333 w2_t[1] = swap_workaround (w2_t[1]);
334 w2_t[2] = swap_workaround (w2_t[2]);
335 w2_t[3] = swap_workaround (w2_t[3]);
336 w3_t[0] = swap_workaround (w3_t[0]);
337 w3_t[1] = swap_workaround (w3_t[1]);
339 w3_t[3] = pw_salt_len * 8;
349 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
357 w1_t[2] = 0x80000000;
366 w3_t[3] = (20 + 4) * 8;
374 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
376 u32x a = swap_workaround (digest[0]);
377 u32x b = swap_workaround (digest[1]) & 0xff;
384 #include VECT_COMPARE_M
388 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_m08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global 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)
392 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_m16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global 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)
396 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_s04 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global 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)
402 const u32 lid = get_local_id (0);
408 const u32 gid = get_global_id (0);
410 if (gid >= gid_max) return;
414 wordl0[0] = pws[gid].i[ 0];
415 wordl0[1] = pws[gid].i[ 1];
416 wordl0[2] = pws[gid].i[ 2];
417 wordl0[3] = pws[gid].i[ 3];
421 wordl1[0] = pws[gid].i[ 4];
422 wordl1[1] = pws[gid].i[ 5];
423 wordl1[2] = pws[gid].i[ 6];
424 wordl1[3] = pws[gid].i[ 7];
440 const u32 pw_l_len = pws[gid].pw_len;
442 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
444 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
451 const u32 search[4] =
453 digests_buf[digests_offset].digest_buf[DGST_R0],
454 digests_buf[digests_offset].digest_buf[DGST_R1],
455 digests_buf[digests_offset].digest_buf[DGST_R2],
456 digests_buf[digests_offset].digest_buf[DGST_R3]
465 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
466 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
467 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
468 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
470 const u32 salt_len = 16;
476 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
478 const u32 pw_r_len = combs_buf[il_pos].pw_len;
480 const u32 pw_len = pw_l_len + pw_r_len;
482 const u32 pw_salt_len = (pw_len * 2) + salt_len;
486 wordr0[0] = combs_buf[il_pos].i[0];
487 wordr0[1] = combs_buf[il_pos].i[1];
488 wordr0[2] = combs_buf[il_pos].i[2];
489 wordr0[3] = combs_buf[il_pos].i[3];
493 wordr1[0] = combs_buf[il_pos].i[4];
494 wordr1[1] = combs_buf[il_pos].i[5];
495 wordr1[2] = combs_buf[il_pos].i[6];
496 wordr1[3] = combs_buf[il_pos].i[7];
512 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
514 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
519 w0[0] = wordl0[0] | wordr0[0];
520 w0[1] = wordl0[1] | wordr0[1];
521 w0[2] = wordl0[2] | wordr0[2];
522 w0[3] = wordl0[3] | wordr0[3];
526 w1[0] = wordl1[0] | wordr1[0];
527 w1[1] = wordl1[1] | wordr1[1];
528 w1[2] = wordl1[2] | wordr1[2];
529 w1[3] = wordl1[3] | wordr1[3];
533 w2[0] = wordl2[0] | wordr2[0];
534 w2[1] = wordl2[1] | wordr2[1];
535 w2[2] = wordl2[2] | wordr2[2];
536 w2[3] = wordl2[3] | wordr2[3];
540 w3[0] = wordl3[0] | wordr3[0];
541 w3[1] = wordl3[1] | wordr3[1];
545 append_0x80_2 (w0, w1, pw_len);
552 make_unicode (w0, w0_t, w1_t);
553 make_unicode (w1, w2_t, w3_t);
555 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, salt_len);
557 w0_t[0] = salt_buf[0];
558 w0_t[1] = salt_buf[1];
559 w0_t[2] = salt_buf[2];
560 w0_t[3] = salt_buf[3];
561 w1_t[0] = swap_workaround (w1_t[0]);
562 w1_t[1] = swap_workaround (w1_t[1]);
563 w1_t[2] = swap_workaround (w1_t[2]);
564 w1_t[3] = swap_workaround (w1_t[3]);
565 w2_t[0] = swap_workaround (w2_t[0]);
566 w2_t[1] = swap_workaround (w2_t[1]);
567 w2_t[2] = swap_workaround (w2_t[2]);
568 w2_t[3] = swap_workaround (w2_t[3]);
569 w3_t[0] = swap_workaround (w3_t[0]);
570 w3_t[1] = swap_workaround (w3_t[1]);
572 w3_t[3] = pw_salt_len * 8;
582 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
590 w1_t[2] = 0x80000000;
599 w3_t[3] = (20 + 4) * 8;
607 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
609 u32x a = swap_workaround (digest[0]);
610 u32x b = swap_workaround (digest[1]) & 0xff;
617 #include VECT_COMPARE_S
621 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_s08 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global 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)
625 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m09820_s16 (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global void *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global oldoffice34_t *oldoffice34_bufs, __global u32 *d_return_buf, __global 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)