2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
16 #include "include/kernel_functions.c"
17 #include "types_ocl.c"
20 #define COMPARE_S "check_single_comp4.c"
21 #define COMPARE_M "check_multi_comp4.c"
24 #define uint_to_hex_lower8_le(i) l_bin2asc[(i)]
28 #define uint_to_hex_lower8_le(i) u32 (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
32 #define uint_to_hex_lower8_le(i) u32 (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
35 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
63 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
64 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
65 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
66 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
67 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
68 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
69 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
70 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
71 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
72 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
73 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
74 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
75 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
76 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
77 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
78 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
79 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
80 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
81 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
82 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
87 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
88 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
89 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
90 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
91 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
92 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
93 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
94 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
95 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
96 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
97 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
98 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
99 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
100 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
101 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
102 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
103 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
104 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
105 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
106 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
111 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
112 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
113 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
114 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
115 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
116 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
117 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
118 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
119 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
120 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
121 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
122 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
123 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
124 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
125 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
126 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
127 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
128 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
129 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
130 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
135 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
136 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
137 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
138 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
139 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
140 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
141 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
142 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
143 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
144 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
145 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
146 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
147 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
148 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
149 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
150 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
151 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
152 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
153 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
154 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
163 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
169 const u32 lid = get_local_id (0);
175 const u32 gid = get_global_id (0);
179 wordl0[0] = pws[gid].i[ 0];
180 wordl0[1] = pws[gid].i[ 1];
181 wordl0[2] = pws[gid].i[ 2];
182 wordl0[3] = pws[gid].i[ 3];
186 wordl1[0] = pws[gid].i[ 4];
187 wordl1[1] = pws[gid].i[ 5];
188 wordl1[2] = pws[gid].i[ 6];
189 wordl1[3] = pws[gid].i[ 7];
205 const u32 pw_l_len = pws[gid].pw_len;
207 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
209 append_0x80_2x4 (wordl0, wordl1, pw_l_len);
211 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
218 __local u32 l_bin2asc[256];
220 const u32 lid4 = lid * 4;
222 const u32 lid40 = lid4 + 0;
223 const u32 lid41 = lid4 + 1;
224 const u32 lid42 = lid4 + 2;
225 const u32 lid43 = lid4 + 3;
227 const u32 v400 = (lid40 >> 0) & 15;
228 const u32 v401 = (lid40 >> 4) & 15;
229 const u32 v410 = (lid41 >> 0) & 15;
230 const u32 v411 = (lid41 >> 4) & 15;
231 const u32 v420 = (lid42 >> 0) & 15;
232 const u32 v421 = (lid42 >> 4) & 15;
233 const u32 v430 = (lid43 >> 0) & 15;
234 const u32 v431 = (lid43 >> 4) & 15;
236 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
237 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
238 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
239 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
240 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
241 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
242 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
243 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
245 barrier (CLK_LOCAL_MEM_FENCE);
247 if (gid >= gid_max) return;
255 salt_buf0[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
256 salt_buf0[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
257 salt_buf0[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 2]);
258 salt_buf0[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 3]);
262 salt_buf1[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 4]);
263 salt_buf1[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 5]);
264 salt_buf1[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 6]);
265 salt_buf1[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 7]);
269 salt_buf2[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 8]);
270 salt_buf2[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 9]);
274 const u32 salt_len = salt_bufs[salt_pos].salt_len;
280 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
282 const u32 pw_r_len = combs_buf[il_pos].pw_len;
284 const u32 pw_len = pw_l_len + pw_r_len;
288 wordr0[0] = combs_buf[il_pos].i[0];
289 wordr0[1] = combs_buf[il_pos].i[1];
290 wordr0[2] = combs_buf[il_pos].i[2];
291 wordr0[3] = combs_buf[il_pos].i[3];
295 wordr1[0] = combs_buf[il_pos].i[4];
296 wordr1[1] = combs_buf[il_pos].i[5];
297 wordr1[2] = combs_buf[il_pos].i[6];
298 wordr1[3] = combs_buf[il_pos].i[7];
314 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
316 append_0x80_2x4 (wordr0, wordr1, pw_r_len);
318 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
323 w0[0] = wordl0[0] | wordr0[0];
324 w0[1] = wordl0[1] | wordr0[1];
325 w0[2] = wordl0[2] | wordr0[2];
326 w0[3] = wordl0[3] | wordr0[3];
330 w1[0] = wordl1[0] | wordr1[0];
331 w1[1] = wordl1[1] | wordr1[1];
332 w1[2] = wordl1[2] | wordr1[2];
333 w1[3] = wordl1[3] | wordr1[3];
337 w2[0] = wordl2[0] | wordr2[0];
338 w2[1] = wordl2[1] | wordr2[1];
339 w2[2] = wordl2[2] | wordr2[2];
340 w2[3] = wordl2[3] | wordr2[3];
344 w3[0] = wordl3[0] | wordr3[0];
345 w3[1] = wordl3[1] | wordr3[1];
346 w3[2] = wordl3[2] | wordr3[2];
347 w3[3] = wordl3[3] | wordr3[3];
351 w0_t[0] = swap32 (w0[0]);
352 w0_t[1] = swap32 (w0[1]);
353 w0_t[2] = swap32 (w0[2]);
354 w0_t[3] = swap32 (w0[3]);
358 w1_t[0] = swap32 (w1[0]);
359 w1_t[1] = swap32 (w1[1]);
360 w1_t[2] = swap32 (w1[2]);
361 w1_t[3] = swap32 (w1[3]);
365 w2_t[0] = swap32 (w2[0]);
366 w2_t[1] = swap32 (w2[1]);
367 w2_t[2] = swap32 (w2[2]);
368 w2_t[3] = swap32 (w2[3]);
372 w3_t[0] = swap32 (w3[0]);
373 w3_t[1] = swap32 (w3[1]);
375 w3_t[3] = pw_len * 8;
385 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
399 w0_t[0] = salt_buf0[0];
400 w0_t[1] = salt_buf0[1];
401 w0_t[2] = salt_buf0[2];
402 w0_t[3] = salt_buf0[3];
403 w1_t[0] = salt_buf1[0];
404 w1_t[1] = salt_buf1[1];
405 w1_t[2] = salt_buf1[2];
406 w1_t[3] = salt_buf1[3];
407 w2_t[0] = salt_buf2[0];
408 w2_t[1] = salt_buf2[1];
409 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
410 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
411 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
412 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
413 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
414 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
415 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
416 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
417 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
418 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
419 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
420 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
428 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
430 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
431 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
432 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
433 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
434 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
435 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
436 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
437 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
438 w1_t[0] = 0x80000000;
451 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
459 w0_t[0] = salt_buf0[0];
460 w0_t[1] = salt_buf0[1];
461 w0_t[2] = salt_buf0[2];
462 w0_t[3] = salt_buf0[3];
463 w1_t[0] = salt_buf1[0];
464 w1_t[1] = salt_buf1[1];
465 w1_t[2] = salt_buf1[2];
466 w1_t[3] = salt_buf1[3];
467 w2_t[0] = salt_buf2[0];
468 w2_t[1] = salt_buf2[1];
469 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
470 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
471 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
472 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
473 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
474 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
475 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
476 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
477 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
478 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
479 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
480 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
488 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
490 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
491 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
492 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
493 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
494 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
495 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
496 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
497 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
498 w1_t[0] = 0x80000000;
511 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
513 const u32 r0 = digest[3];
514 const u32 r1 = digest[4];
515 const u32 r2 = digest[2];
516 const u32 r3 = digest[1];
522 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
526 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
530 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
536 const u32 lid = get_local_id (0);
542 const u32 gid = get_global_id (0);
546 wordl0[0] = pws[gid].i[ 0];
547 wordl0[1] = pws[gid].i[ 1];
548 wordl0[2] = pws[gid].i[ 2];
549 wordl0[3] = pws[gid].i[ 3];
553 wordl1[0] = pws[gid].i[ 4];
554 wordl1[1] = pws[gid].i[ 5];
555 wordl1[2] = pws[gid].i[ 6];
556 wordl1[3] = pws[gid].i[ 7];
572 const u32 pw_l_len = pws[gid].pw_len;
574 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
576 append_0x80_2x4 (wordl0, wordl1, pw_l_len);
578 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
585 __local u32 l_bin2asc[256];
587 const u32 lid4 = lid * 4;
589 const u32 lid40 = lid4 + 0;
590 const u32 lid41 = lid4 + 1;
591 const u32 lid42 = lid4 + 2;
592 const u32 lid43 = lid4 + 3;
594 const u32 v400 = (lid40 >> 0) & 15;
595 const u32 v401 = (lid40 >> 4) & 15;
596 const u32 v410 = (lid41 >> 0) & 15;
597 const u32 v411 = (lid41 >> 4) & 15;
598 const u32 v420 = (lid42 >> 0) & 15;
599 const u32 v421 = (lid42 >> 4) & 15;
600 const u32 v430 = (lid43 >> 0) & 15;
601 const u32 v431 = (lid43 >> 4) & 15;
603 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
604 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
605 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
606 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
607 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
608 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
609 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
610 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
612 barrier (CLK_LOCAL_MEM_FENCE);
614 if (gid >= gid_max) return;
622 salt_buf0[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
623 salt_buf0[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
624 salt_buf0[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 2]);
625 salt_buf0[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 3]);
629 salt_buf1[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 4]);
630 salt_buf1[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 5]);
631 salt_buf1[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 6]);
632 salt_buf1[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 7]);
636 salt_buf2[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 8]);
637 salt_buf2[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 9]);
641 const u32 salt_len = salt_bufs[salt_pos].salt_len;
647 const u32 search[4] =
649 digests_buf[digests_offset].digest_buf[DGST_R0],
650 digests_buf[digests_offset].digest_buf[DGST_R1],
651 digests_buf[digests_offset].digest_buf[DGST_R2],
652 digests_buf[digests_offset].digest_buf[DGST_R3]
659 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
661 const u32 pw_r_len = combs_buf[il_pos].pw_len;
663 const u32 pw_len = pw_l_len + pw_r_len;
667 wordr0[0] = combs_buf[il_pos].i[0];
668 wordr0[1] = combs_buf[il_pos].i[1];
669 wordr0[2] = combs_buf[il_pos].i[2];
670 wordr0[3] = combs_buf[il_pos].i[3];
674 wordr1[0] = combs_buf[il_pos].i[4];
675 wordr1[1] = combs_buf[il_pos].i[5];
676 wordr1[2] = combs_buf[il_pos].i[6];
677 wordr1[3] = combs_buf[il_pos].i[7];
693 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
695 append_0x80_2x4 (wordr0, wordr1, pw_r_len);
697 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
702 w0[0] = wordl0[0] | wordr0[0];
703 w0[1] = wordl0[1] | wordr0[1];
704 w0[2] = wordl0[2] | wordr0[2];
705 w0[3] = wordl0[3] | wordr0[3];
709 w1[0] = wordl1[0] | wordr1[0];
710 w1[1] = wordl1[1] | wordr1[1];
711 w1[2] = wordl1[2] | wordr1[2];
712 w1[3] = wordl1[3] | wordr1[3];
716 w2[0] = wordl2[0] | wordr2[0];
717 w2[1] = wordl2[1] | wordr2[1];
718 w2[2] = wordl2[2] | wordr2[2];
719 w2[3] = wordl2[3] | wordr2[3];
723 w3[0] = wordl3[0] | wordr3[0];
724 w3[1] = wordl3[1] | wordr3[1];
725 w3[2] = wordl3[2] | wordr3[2];
726 w3[3] = wordl3[3] | wordr3[3];
730 w0_t[0] = swap32 (w0[0]);
731 w0_t[1] = swap32 (w0[1]);
732 w0_t[2] = swap32 (w0[2]);
733 w0_t[3] = swap32 (w0[3]);
737 w1_t[0] = swap32 (w1[0]);
738 w1_t[1] = swap32 (w1[1]);
739 w1_t[2] = swap32 (w1[2]);
740 w1_t[3] = swap32 (w1[3]);
744 w2_t[0] = swap32 (w2[0]);
745 w2_t[1] = swap32 (w2[1]);
746 w2_t[2] = swap32 (w2[2]);
747 w2_t[3] = swap32 (w2[3]);
751 w3_t[0] = swap32 (w3[0]);
752 w3_t[1] = swap32 (w3[1]);
754 w3_t[3] = pw_len * 8;
764 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
778 w0_t[0] = salt_buf0[0];
779 w0_t[1] = salt_buf0[1];
780 w0_t[2] = salt_buf0[2];
781 w0_t[3] = salt_buf0[3];
782 w1_t[0] = salt_buf1[0];
783 w1_t[1] = salt_buf1[1];
784 w1_t[2] = salt_buf1[2];
785 w1_t[3] = salt_buf1[3];
786 w2_t[0] = salt_buf2[0];
787 w2_t[1] = salt_buf2[1];
788 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
789 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
790 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
791 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
792 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
793 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
794 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
795 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
796 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
797 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
798 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
799 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
807 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
809 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
810 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
811 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
812 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
813 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
814 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
815 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
816 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
817 w1_t[0] = 0x80000000;
828 w3_t[3] = (salt_len + 40) * 8;
830 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
838 w0_t[0] = salt_buf0[0];
839 w0_t[1] = salt_buf0[1];
840 w0_t[2] = salt_buf0[2];
841 w0_t[3] = salt_buf0[3];
842 w1_t[0] = salt_buf1[0];
843 w1_t[1] = salt_buf1[1];
844 w1_t[2] = salt_buf1[2];
845 w1_t[3] = salt_buf1[3];
846 w2_t[0] = salt_buf2[0];
847 w2_t[1] = salt_buf2[1];
848 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
849 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
850 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
851 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
852 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
853 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
854 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
855 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
856 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
857 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
858 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
859 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
867 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
869 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
870 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
871 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
872 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
873 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
874 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
875 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
876 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
877 w1_t[0] = 0x80000000;
888 w3_t[3] = (salt_len + 40) * 8;
890 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
892 const u32 r0 = digest[3];
893 const u32 r1 = digest[4];
894 const u32 r2 = digest[2];
895 const u32 r3 = digest[1];
901 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
905 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)