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))) m08300_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 void *esalt_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);
230 const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
234 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
235 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
236 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
237 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
241 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
242 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
243 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
244 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
246 const u32 salt_len = salt_bufs[salt_pos].salt_len;
250 domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
251 domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
252 domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
253 domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
257 domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
258 domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
259 domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
262 const u32 domain_len = salt_bufs[salt_pos].salt_buf_pc[ 7];
268 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
270 const u32 pw_r_len = combs_buf[il_pos].pw_len;
272 const u32 pw_len = pw_l_len + pw_r_len;
279 wordr0[0] = combs_buf[il_pos].i[0];
280 wordr0[1] = combs_buf[il_pos].i[1];
281 wordr0[2] = combs_buf[il_pos].i[2];
282 wordr0[3] = combs_buf[il_pos].i[3];
283 wordr1[0] = combs_buf[il_pos].i[4];
284 wordr1[1] = combs_buf[il_pos].i[5];
285 wordr1[2] = combs_buf[il_pos].i[6];
286 wordr1[3] = combs_buf[il_pos].i[7];
296 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
298 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
306 w0[0] = wordl0[0] | wordr0[0];
307 w0[1] = wordl0[1] | wordr0[1];
308 w0[2] = wordl0[2] | wordr0[2];
309 w0[3] = wordl0[3] | wordr0[3];
310 w1[0] = wordl1[0] | wordr1[0];
311 w1[1] = wordl1[1] | wordr1[1];
312 w1[2] = wordl1[2] | wordr1[2];
313 w1[3] = wordl1[3] | wordr1[3];
314 w2[0] = wordl2[0] | wordr2[0];
315 w2[1] = wordl2[1] | wordr2[1];
316 w2[2] = wordl2[2] | wordr2[2];
317 w2[3] = wordl2[3] | wordr2[3];
318 w3[0] = wordl3[0] | wordr3[0];
319 w3[1] = wordl3[1] | wordr3[1];
320 w3[2] = wordl3[2] | wordr3[2];
321 w3[3] = wordl3[3] | wordr3[3];
351 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, 1);
353 w0_t[0] |= pw_len & 0xff;
361 s0[0] = salt_buf0[0];
362 s0[1] = salt_buf0[1];
363 s0[2] = salt_buf0[2];
364 s0[3] = salt_buf0[3];
368 s1[0] = salt_buf1[0];
369 s1[1] = salt_buf1[1];
370 s1[2] = salt_buf1[2];
371 s1[3] = salt_buf1[3];
387 switch_buffer_by_offset (s0, s1, s2, s3, 1 + pw_len + domain_len + 1);
391 d0[0] = domain_buf0[0];
392 d0[1] = domain_buf0[1];
393 d0[2] = domain_buf0[2];
394 d0[3] = domain_buf0[3];
398 d1[0] = domain_buf1[0];
399 d1[1] = domain_buf1[1];
400 d1[2] = domain_buf1[2];
417 switch_buffer_by_offset (d0, d1, d2, d3, 1 + pw_len);
425 w0_t2[0] = swap_workaround (w0_t[0] | d0[0] | s0[0]);
426 w0_t2[1] = swap_workaround (w0_t[1] | d0[1] | s0[1]);
427 w0_t2[2] = swap_workaround (w0_t[2] | d0[2] | s0[2]);
428 w0_t2[3] = swap_workaround (w0_t[3] | d0[3] | s0[3]);
432 w1_t2[0] = swap_workaround (w1_t[0] | d1[0] | s1[0]);
433 w1_t2[1] = swap_workaround (w1_t[1] | d1[1] | s1[1]);
434 w1_t2[2] = swap_workaround (w1_t[2] | d1[2] | s1[2]);
435 w1_t2[3] = swap_workaround (w1_t[3] | d1[3] | s1[3]);
439 w2_t2[0] = swap_workaround (w2_t[0] | d2[0] | s2[0]);
440 w2_t2[1] = swap_workaround (w2_t[1] | d2[1] | s2[1]);
441 w2_t2[2] = swap_workaround (w2_t[2] | d2[2] | s2[2]);
442 w2_t2[3] = swap_workaround (w2_t[3] | d2[3] | s2[3]);
446 w3_t2[0] = swap_workaround (w3_t[0] | d3[0] | s3[0]);
447 w3_t2[1] = swap_workaround (w3_t[1] | d3[1] | s3[1]);
449 w3_t2[3] = (1 + pw_len + domain_len + 1 + salt_len) * 8;
459 sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
463 for (u32 i = 0; i < salt_iter; i++)
467 w0_t3[0] = digest[0];
468 w0_t3[1] = digest[1];
469 w0_t3[2] = digest[2];
470 w0_t3[3] = digest[3];
474 w1_t3[0] = digest[4];
475 w1_t3[1] = swap_workaround (salt_buf0[0]);
476 w1_t3[2] = swap_workaround (salt_buf0[1]);
477 w1_t3[3] = swap_workaround (salt_buf0[2]);
481 w2_t3[0] = swap_workaround (salt_buf0[3]);
482 w2_t3[1] = swap_workaround (salt_buf1[0]);
483 w2_t3[2] = swap_workaround (salt_buf1[1]);
484 w2_t3[3] = swap_workaround (salt_buf1[2]);
488 w3_t3[0] = swap_workaround (salt_buf1[3]);
491 w3_t3[3] = (20 + salt_len) * 8;
499 sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
502 const u32x r0 = digest[3];
503 const u32x r1 = digest[4];
504 const u32x r2 = digest[2];
505 const u32x r3 = digest[1];
507 #include VECT_COMPARE_M
511 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08300_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 void *esalt_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)
515 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08300_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 void *esalt_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)
519 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08300_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 void *esalt_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)
525 const u32 lid = get_local_id (0);
531 const u32 gid = get_global_id (0);
533 if (gid >= gid_max) return;
537 wordl0[0] = pws[gid].i[ 0];
538 wordl0[1] = pws[gid].i[ 1];
539 wordl0[2] = pws[gid].i[ 2];
540 wordl0[3] = pws[gid].i[ 3];
544 wordl1[0] = pws[gid].i[ 4];
545 wordl1[1] = pws[gid].i[ 5];
546 wordl1[2] = pws[gid].i[ 6];
547 wordl1[3] = pws[gid].i[ 7];
563 const u32 pw_l_len = pws[gid].pw_len;
565 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
567 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
574 const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
578 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
579 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
580 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
581 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
585 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
586 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
587 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
588 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
590 const u32 salt_len = salt_bufs[salt_pos].salt_len;
594 domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
595 domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
596 domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
597 domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
601 domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
602 domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
603 domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
606 const u32 domain_len = salt_bufs[salt_pos].salt_buf_pc[ 7];
612 const u32 search[4] =
614 digests_buf[digests_offset].digest_buf[DGST_R0],
615 digests_buf[digests_offset].digest_buf[DGST_R1],
616 digests_buf[digests_offset].digest_buf[DGST_R2],
617 digests_buf[digests_offset].digest_buf[DGST_R3]
624 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
626 const u32 pw_r_len = combs_buf[il_pos].pw_len;
628 const u32 pw_len = pw_l_len + pw_r_len;
635 wordr0[0] = combs_buf[il_pos].i[0];
636 wordr0[1] = combs_buf[il_pos].i[1];
637 wordr0[2] = combs_buf[il_pos].i[2];
638 wordr0[3] = combs_buf[il_pos].i[3];
639 wordr1[0] = combs_buf[il_pos].i[4];
640 wordr1[1] = combs_buf[il_pos].i[5];
641 wordr1[2] = combs_buf[il_pos].i[6];
642 wordr1[3] = combs_buf[il_pos].i[7];
652 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
654 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
662 w0[0] = wordl0[0] | wordr0[0];
663 w0[1] = wordl0[1] | wordr0[1];
664 w0[2] = wordl0[2] | wordr0[2];
665 w0[3] = wordl0[3] | wordr0[3];
666 w1[0] = wordl1[0] | wordr1[0];
667 w1[1] = wordl1[1] | wordr1[1];
668 w1[2] = wordl1[2] | wordr1[2];
669 w1[3] = wordl1[3] | wordr1[3];
670 w2[0] = wordl2[0] | wordr2[0];
671 w2[1] = wordl2[1] | wordr2[1];
672 w2[2] = wordl2[2] | wordr2[2];
673 w2[3] = wordl2[3] | wordr2[3];
674 w3[0] = wordl3[0] | wordr3[0];
675 w3[1] = wordl3[1] | wordr3[1];
676 w3[2] = wordl3[2] | wordr3[2];
677 w3[3] = wordl3[3] | wordr3[3];
707 switch_buffer_by_offset (w0_t, w1_t, w2_t, w3_t, 1);
709 w0_t[0] |= pw_len & 0xff;
717 s0[0] = salt_buf0[0];
718 s0[1] = salt_buf0[1];
719 s0[2] = salt_buf0[2];
720 s0[3] = salt_buf0[3];
724 s1[0] = salt_buf1[0];
725 s1[1] = salt_buf1[1];
726 s1[2] = salt_buf1[2];
727 s1[3] = salt_buf1[3];
743 switch_buffer_by_offset (s0, s1, s2, s3, 1 + pw_len + domain_len + 1);
747 d0[0] = domain_buf0[0];
748 d0[1] = domain_buf0[1];
749 d0[2] = domain_buf0[2];
750 d0[3] = domain_buf0[3];
754 d1[0] = domain_buf1[0];
755 d1[1] = domain_buf1[1];
756 d1[2] = domain_buf1[2];
773 switch_buffer_by_offset (d0, d1, d2, d3, 1 + pw_len);
781 w0_t2[0] = swap_workaround (w0_t[0] | d0[0] | s0[0]);
782 w0_t2[1] = swap_workaround (w0_t[1] | d0[1] | s0[1]);
783 w0_t2[2] = swap_workaround (w0_t[2] | d0[2] | s0[2]);
784 w0_t2[3] = swap_workaround (w0_t[3] | d0[3] | s0[3]);
788 w1_t2[0] = swap_workaround (w1_t[0] | d1[0] | s1[0]);
789 w1_t2[1] = swap_workaround (w1_t[1] | d1[1] | s1[1]);
790 w1_t2[2] = swap_workaround (w1_t[2] | d1[2] | s1[2]);
791 w1_t2[3] = swap_workaround (w1_t[3] | d1[3] | s1[3]);
795 w2_t2[0] = swap_workaround (w2_t[0] | d2[0] | s2[0]);
796 w2_t2[1] = swap_workaround (w2_t[1] | d2[1] | s2[1]);
797 w2_t2[2] = swap_workaround (w2_t[2] | d2[2] | s2[2]);
798 w2_t2[3] = swap_workaround (w2_t[3] | d2[3] | s2[3]);
802 w3_t2[0] = swap_workaround (w3_t[0] | d3[0] | s3[0]);
803 w3_t2[1] = swap_workaround (w3_t[1] | d3[1] | s3[1]);
805 w3_t2[3] = (1 + pw_len + domain_len + 1 + salt_len) * 8;
815 sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
819 for (u32 i = 0; i < salt_iter; i++)
823 w0_t3[0] = digest[0];
824 w0_t3[1] = digest[1];
825 w0_t3[2] = digest[2];
826 w0_t3[3] = digest[3];
830 w1_t3[0] = digest[4];
831 w1_t3[1] = swap_workaround (salt_buf0[0]);
832 w1_t3[2] = swap_workaround (salt_buf0[1]);
833 w1_t3[3] = swap_workaround (salt_buf0[2]);
837 w2_t3[0] = swap_workaround (salt_buf0[3]);
838 w2_t3[1] = swap_workaround (salt_buf1[0]);
839 w2_t3[2] = swap_workaround (salt_buf1[1]);
840 w2_t3[3] = swap_workaround (salt_buf1[2]);
844 w3_t3[0] = swap_workaround (salt_buf1[3]);
847 w3_t3[3] = (20 + salt_len) * 8;
855 sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
858 const u32x r0 = digest[3];
859 const u32x r1 = digest[4];
860 const u32x r2 = digest[2];
861 const u32x r3 = digest[1];
863 #include VECT_COMPARE_S
867 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08300_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 void *esalt_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)
871 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08300_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 void *esalt_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)