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"
31 #include "include/rp_gpu.h"
35 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
36 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
40 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
41 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
45 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
46 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
50 #define uint_to_hex_lower8_le(i) l_bin2asc[(i)]
54 #define uint_to_hex_lower8_le(i) u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
58 #define uint_to_hex_lower8_le(i) u32x (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
61 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
89 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
90 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
91 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
92 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
93 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
94 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
95 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
96 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
97 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
98 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
99 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
100 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
101 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
102 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
103 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
104 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
105 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
106 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
107 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
108 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
113 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
114 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
115 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
116 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
117 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
118 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
119 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
120 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
121 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
122 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
123 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
124 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
125 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
126 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
127 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
128 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
129 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
130 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
131 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
132 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
137 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
138 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
139 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
140 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
141 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
142 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
143 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
144 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
145 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
146 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
147 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
148 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
149 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
150 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
151 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
152 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
153 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
154 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
155 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
156 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
161 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
162 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
163 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
164 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
165 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
166 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
167 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
168 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
169 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
170 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
171 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
172 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
173 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
174 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
175 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
176 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
177 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
178 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
179 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
180 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
189 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
195 const u32 lid = get_local_id (0);
201 const u32 gid = get_global_id (0);
205 pw_buf0[0] = pws[gid].i[ 0];
206 pw_buf0[1] = pws[gid].i[ 1];
207 pw_buf0[2] = pws[gid].i[ 2];
208 pw_buf0[3] = pws[gid].i[ 3];
212 pw_buf1[0] = pws[gid].i[ 4];
213 pw_buf1[1] = pws[gid].i[ 5];
214 pw_buf1[2] = pws[gid].i[ 6];
215 pw_buf1[3] = pws[gid].i[ 7];
217 const u32 pw_len = pws[gid].pw_len;
223 __local u32 l_bin2asc[256];
225 const u32 lid4 = lid * 4;
227 const u32 lid40 = lid4 + 0;
228 const u32 lid41 = lid4 + 1;
229 const u32 lid42 = lid4 + 2;
230 const u32 lid43 = lid4 + 3;
232 const u32 v400 = (lid40 >> 0) & 15;
233 const u32 v401 = (lid40 >> 4) & 15;
234 const u32 v410 = (lid41 >> 0) & 15;
235 const u32 v411 = (lid41 >> 4) & 15;
236 const u32 v420 = (lid42 >> 0) & 15;
237 const u32 v421 = (lid42 >> 4) & 15;
238 const u32 v430 = (lid43 >> 0) & 15;
239 const u32 v431 = (lid43 >> 4) & 15;
241 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
242 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
243 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
244 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
245 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
246 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
247 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
248 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
250 barrier (CLK_LOCAL_MEM_FENCE);
252 if (gid >= gid_max) return;
260 salt_buf0[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 0]);
261 salt_buf0[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 1]);
262 salt_buf0[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 2]);
263 salt_buf0[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 3]);
267 salt_buf1[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 4]);
268 salt_buf1[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 5]);
269 salt_buf1[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 6]);
270 salt_buf1[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 7]);
274 salt_buf2[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 8]);
275 salt_buf2[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 9]);
279 const u32 salt_len = salt_bufs[salt_pos].salt_len;
285 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
315 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
317 append_0x80_2 (w0, w1, out_len);
321 w0_t[0] = swap_workaround (w0[0]);
322 w0_t[1] = swap_workaround (w0[1]);
323 w0_t[2] = swap_workaround (w0[2]);
324 w0_t[3] = swap_workaround (w0[3]);
328 w1_t[0] = swap_workaround (w1[0]);
329 w1_t[1] = swap_workaround (w1[1]);
330 w1_t[2] = swap_workaround (w1[2]);
331 w1_t[3] = swap_workaround (w1[3]);
335 w2_t[0] = swap_workaround (w2[0]);
336 w2_t[1] = swap_workaround (w2[1]);
337 w2_t[2] = swap_workaround (w2[2]);
338 w2_t[3] = swap_workaround (w2[3]);
342 w3_t[0] = swap_workaround (w3[0]);
343 w3_t[1] = swap_workaround (w3[1]);
345 w3_t[3] = pw_len * 8;
355 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
369 w0_t[0] = salt_buf0[0];
370 w0_t[1] = salt_buf0[1];
371 w0_t[2] = salt_buf0[2];
372 w0_t[3] = salt_buf0[3];
373 w1_t[0] = salt_buf1[0];
374 w1_t[1] = salt_buf1[1];
375 w1_t[2] = salt_buf1[2];
376 w1_t[3] = salt_buf1[3];
377 w2_t[0] = salt_buf2[0];
378 w2_t[1] = salt_buf2[1];
379 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
380 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
381 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
382 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
383 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
384 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
385 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
386 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
387 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
388 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
389 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
390 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
398 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
400 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
401 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
402 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
403 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
404 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
405 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
406 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
407 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
408 w1_t[0] = 0x80000000;
421 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
429 w0_t[0] = salt_buf0[0];
430 w0_t[1] = salt_buf0[1];
431 w0_t[2] = salt_buf0[2];
432 w0_t[3] = salt_buf0[3];
433 w1_t[0] = salt_buf1[0];
434 w1_t[1] = salt_buf1[1];
435 w1_t[2] = salt_buf1[2];
436 w1_t[3] = salt_buf1[3];
437 w2_t[0] = salt_buf2[0];
438 w2_t[1] = salt_buf2[1];
439 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
440 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
441 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
442 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
443 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
444 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
445 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
446 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
447 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
448 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
449 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
450 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
458 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
460 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
461 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
462 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
463 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
464 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
465 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
466 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
467 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
468 w1_t[0] = 0x80000000;
481 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
483 const u32x r0 = digest[3];
484 const u32x r1 = digest[4];
485 const u32x r2 = digest[2];
486 const u32x r3 = digest[1];
488 #include VECT_COMPARE_M
492 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
496 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
500 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
506 const u32 lid = get_local_id (0);
512 const u32 gid = get_global_id (0);
516 pw_buf0[0] = pws[gid].i[ 0];
517 pw_buf0[1] = pws[gid].i[ 1];
518 pw_buf0[2] = pws[gid].i[ 2];
519 pw_buf0[3] = pws[gid].i[ 3];
523 pw_buf1[0] = pws[gid].i[ 4];
524 pw_buf1[1] = pws[gid].i[ 5];
525 pw_buf1[2] = pws[gid].i[ 6];
526 pw_buf1[3] = pws[gid].i[ 7];
528 const u32 pw_len = pws[gid].pw_len;
534 __local u32 l_bin2asc[256];
536 const u32 lid4 = lid * 4;
538 const u32 lid40 = lid4 + 0;
539 const u32 lid41 = lid4 + 1;
540 const u32 lid42 = lid4 + 2;
541 const u32 lid43 = lid4 + 3;
543 const u32 v400 = (lid40 >> 0) & 15;
544 const u32 v401 = (lid40 >> 4) & 15;
545 const u32 v410 = (lid41 >> 0) & 15;
546 const u32 v411 = (lid41 >> 4) & 15;
547 const u32 v420 = (lid42 >> 0) & 15;
548 const u32 v421 = (lid42 >> 4) & 15;
549 const u32 v430 = (lid43 >> 0) & 15;
550 const u32 v431 = (lid43 >> 4) & 15;
552 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
553 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
554 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
555 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
556 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
557 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
558 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
559 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
561 barrier (CLK_LOCAL_MEM_FENCE);
563 if (gid >= gid_max) return;
571 salt_buf0[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 0]);
572 salt_buf0[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 1]);
573 salt_buf0[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 2]);
574 salt_buf0[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 3]);
578 salt_buf1[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 4]);
579 salt_buf1[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 5]);
580 salt_buf1[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 6]);
581 salt_buf1[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 7]);
585 salt_buf2[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 8]);
586 salt_buf2[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 9]);
590 const u32 salt_len = salt_bufs[salt_pos].salt_len;
596 const u32 search[4] =
598 digests_buf[digests_offset].digest_buf[DGST_R0],
599 digests_buf[digests_offset].digest_buf[DGST_R1],
600 digests_buf[digests_offset].digest_buf[DGST_R2],
601 digests_buf[digests_offset].digest_buf[DGST_R3]
608 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
638 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
640 append_0x80_2 (w0, w1, out_len);
644 w0_t[0] = swap_workaround (w0[0]);
645 w0_t[1] = swap_workaround (w0[1]);
646 w0_t[2] = swap_workaround (w0[2]);
647 w0_t[3] = swap_workaround (w0[3]);
651 w1_t[0] = swap_workaround (w1[0]);
652 w1_t[1] = swap_workaround (w1[1]);
653 w1_t[2] = swap_workaround (w1[2]);
654 w1_t[3] = swap_workaround (w1[3]);
658 w2_t[0] = swap_workaround (w2[0]);
659 w2_t[1] = swap_workaround (w2[1]);
660 w2_t[2] = swap_workaround (w2[2]);
661 w2_t[3] = swap_workaround (w2[3]);
665 w3_t[0] = swap_workaround (w3[0]);
666 w3_t[1] = swap_workaround (w3[1]);
668 w3_t[3] = pw_len * 8;
678 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
692 w0_t[0] = salt_buf0[0];
693 w0_t[1] = salt_buf0[1];
694 w0_t[2] = salt_buf0[2];
695 w0_t[3] = salt_buf0[3];
696 w1_t[0] = salt_buf1[0];
697 w1_t[1] = salt_buf1[1];
698 w1_t[2] = salt_buf1[2];
699 w1_t[3] = salt_buf1[3];
700 w2_t[0] = salt_buf2[0];
701 w2_t[1] = salt_buf2[1];
702 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
703 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
704 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
705 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
706 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
707 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
708 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
709 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
710 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
711 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
712 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
713 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
721 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
723 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
724 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
725 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
726 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
727 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
728 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
729 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
730 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
731 w1_t[0] = 0x80000000;
742 w3_t[3] = (salt_len + 40) * 8;
744 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
752 w0_t[0] = salt_buf0[0];
753 w0_t[1] = salt_buf0[1];
754 w0_t[2] = salt_buf0[2];
755 w0_t[3] = salt_buf0[3];
756 w1_t[0] = salt_buf1[0];
757 w1_t[1] = salt_buf1[1];
758 w1_t[2] = salt_buf1[2];
759 w1_t[3] = salt_buf1[3];
760 w2_t[0] = salt_buf2[0];
761 w2_t[1] = salt_buf2[1];
762 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
763 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
764 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
765 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
766 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
767 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
768 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
769 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
770 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
771 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
772 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
773 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
781 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
783 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
784 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
785 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
786 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
787 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
788 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
789 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
790 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
791 w1_t[0] = 0x80000000;
802 w3_t[3] = (salt_len + 40) * 8;
804 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
806 const u32x r0 = digest[3];
807 const u32x r1 = digest[4];
808 const u32x r2 = digest[2];
809 const u32x r3 = digest[1];
811 #include VECT_COMPARE_S
815 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
819 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)