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"
19 #include "include/rp_gpu.h"
22 #define COMPARE_S "check_single_comp4.c"
23 #define COMPARE_M "check_multi_comp4.c"
26 #define uint_to_hex_lower8_le(i) l_bin2asc[(i)]
30 #define uint_to_hex_lower8_le(i) u32 (l_bin2asc[(i).s0], l_bin2asc[(i).s1])
34 #define uint_to_hex_lower8_le(i) u32 (l_bin2asc[(i).s0], l_bin2asc[(i).s1], l_bin2asc[(i).s2], l_bin2asc[(i).s3])
37 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
65 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
66 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
67 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
68 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
69 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
70 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
71 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
72 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
73 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
74 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
75 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
76 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
77 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
78 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
79 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
80 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
81 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
82 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
83 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
84 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
89 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
90 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
91 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
92 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
93 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
94 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
95 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
96 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
97 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
98 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
99 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
100 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
101 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
102 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
103 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
104 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
105 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
106 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
107 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
108 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
113 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
114 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
115 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
116 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
117 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
118 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
119 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
120 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
121 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
122 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
123 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
124 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
125 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
126 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
127 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
128 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
129 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
130 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
131 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
132 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
137 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
138 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
139 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
140 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
141 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
142 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
143 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
144 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
145 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
146 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
147 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
148 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
149 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
150 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
151 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
152 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
153 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
154 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
155 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
156 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
165 __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)
171 const u32 lid = get_local_id (0);
177 const u32 gid = get_global_id (0);
181 pw_buf0[0] = pws[gid].i[ 0];
182 pw_buf0[1] = pws[gid].i[ 1];
183 pw_buf0[2] = pws[gid].i[ 2];
184 pw_buf0[3] = pws[gid].i[ 3];
188 pw_buf1[0] = pws[gid].i[ 4];
189 pw_buf1[1] = pws[gid].i[ 5];
190 pw_buf1[2] = pws[gid].i[ 6];
191 pw_buf1[3] = pws[gid].i[ 7];
193 const u32 pw_len = pws[gid].pw_len;
199 __local u32 l_bin2asc[256];
201 const u32 lid4 = lid * 4;
203 const u32 lid40 = lid4 + 0;
204 const u32 lid41 = lid4 + 1;
205 const u32 lid42 = lid4 + 2;
206 const u32 lid43 = lid4 + 3;
208 const u32 v400 = (lid40 >> 0) & 15;
209 const u32 v401 = (lid40 >> 4) & 15;
210 const u32 v410 = (lid41 >> 0) & 15;
211 const u32 v411 = (lid41 >> 4) & 15;
212 const u32 v420 = (lid42 >> 0) & 15;
213 const u32 v421 = (lid42 >> 4) & 15;
214 const u32 v430 = (lid43 >> 0) & 15;
215 const u32 v431 = (lid43 >> 4) & 15;
217 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
218 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
219 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
220 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
221 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
222 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
223 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
224 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
226 barrier (CLK_LOCAL_MEM_FENCE);
228 if (gid >= gid_max) return;
236 salt_buf0[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 0]);
237 salt_buf0[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 1]);
238 salt_buf0[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 2]);
239 salt_buf0[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 3]);
243 salt_buf1[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 4]);
244 salt_buf1[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 5]);
245 salt_buf1[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 6]);
246 salt_buf1[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 7]);
250 salt_buf2[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 8]);
251 salt_buf2[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 9]);
255 const u32 salt_len = salt_bufs[salt_pos].salt_len;
261 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
291 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
293 append_0x80_2x4 (w0, w1, out_len);
297 w0_t[0] = swap_workaround (w0[0]);
298 w0_t[1] = swap_workaround (w0[1]);
299 w0_t[2] = swap_workaround (w0[2]);
300 w0_t[3] = swap_workaround (w0[3]);
304 w1_t[0] = swap_workaround (w1[0]);
305 w1_t[1] = swap_workaround (w1[1]);
306 w1_t[2] = swap_workaround (w1[2]);
307 w1_t[3] = swap_workaround (w1[3]);
311 w2_t[0] = swap_workaround (w2[0]);
312 w2_t[1] = swap_workaround (w2[1]);
313 w2_t[2] = swap_workaround (w2[2]);
314 w2_t[3] = swap_workaround (w2[3]);
318 w3_t[0] = swap_workaround (w3[0]);
319 w3_t[1] = swap_workaround (w3[1]);
321 w3_t[3] = pw_len * 8;
331 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
345 w0_t[0] = salt_buf0[0];
346 w0_t[1] = salt_buf0[1];
347 w0_t[2] = salt_buf0[2];
348 w0_t[3] = salt_buf0[3];
349 w1_t[0] = salt_buf1[0];
350 w1_t[1] = salt_buf1[1];
351 w1_t[2] = salt_buf1[2];
352 w1_t[3] = salt_buf1[3];
353 w2_t[0] = salt_buf2[0];
354 w2_t[1] = salt_buf2[1];
355 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
356 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
357 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
358 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
359 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
360 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
361 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
362 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
363 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
364 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
365 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
366 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
374 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
376 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
377 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
378 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
379 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
380 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
381 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
382 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
383 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
384 w1_t[0] = 0x80000000;
397 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
405 w0_t[0] = salt_buf0[0];
406 w0_t[1] = salt_buf0[1];
407 w0_t[2] = salt_buf0[2];
408 w0_t[3] = salt_buf0[3];
409 w1_t[0] = salt_buf1[0];
410 w1_t[1] = salt_buf1[1];
411 w1_t[2] = salt_buf1[2];
412 w1_t[3] = salt_buf1[3];
413 w2_t[0] = salt_buf2[0];
414 w2_t[1] = salt_buf2[1];
415 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
416 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
417 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
418 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
419 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
420 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
421 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
422 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
423 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
424 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
425 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
426 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
434 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
436 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
437 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
438 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
439 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
440 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
441 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
442 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
443 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
444 w1_t[0] = 0x80000000;
457 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
459 const u32 r0 = digest[3];
460 const u32 r1 = digest[4];
461 const u32 r2 = digest[2];
462 const u32 r3 = digest[1];
468 __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)
472 __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)
476 __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)
482 const u32 lid = get_local_id (0);
488 const u32 gid = get_global_id (0);
492 pw_buf0[0] = pws[gid].i[ 0];
493 pw_buf0[1] = pws[gid].i[ 1];
494 pw_buf0[2] = pws[gid].i[ 2];
495 pw_buf0[3] = pws[gid].i[ 3];
499 pw_buf1[0] = pws[gid].i[ 4];
500 pw_buf1[1] = pws[gid].i[ 5];
501 pw_buf1[2] = pws[gid].i[ 6];
502 pw_buf1[3] = pws[gid].i[ 7];
504 const u32 pw_len = pws[gid].pw_len;
510 __local u32 l_bin2asc[256];
512 const u32 lid4 = lid * 4;
514 const u32 lid40 = lid4 + 0;
515 const u32 lid41 = lid4 + 1;
516 const u32 lid42 = lid4 + 2;
517 const u32 lid43 = lid4 + 3;
519 const u32 v400 = (lid40 >> 0) & 15;
520 const u32 v401 = (lid40 >> 4) & 15;
521 const u32 v410 = (lid41 >> 0) & 15;
522 const u32 v411 = (lid41 >> 4) & 15;
523 const u32 v420 = (lid42 >> 0) & 15;
524 const u32 v421 = (lid42 >> 4) & 15;
525 const u32 v430 = (lid43 >> 0) & 15;
526 const u32 v431 = (lid43 >> 4) & 15;
528 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
529 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
530 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
531 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
532 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
533 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
534 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
535 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
537 barrier (CLK_LOCAL_MEM_FENCE);
539 if (gid >= gid_max) return;
547 salt_buf0[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 0]);
548 salt_buf0[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 1]);
549 salt_buf0[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 2]);
550 salt_buf0[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 3]);
554 salt_buf1[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 4]);
555 salt_buf1[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 5]);
556 salt_buf1[2] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 6]);
557 salt_buf1[3] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 7]);
561 salt_buf2[0] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 8]);
562 salt_buf2[1] = swap_workaround (salt_bufs[salt_pos].salt_buf[ 9]);
566 const u32 salt_len = salt_bufs[salt_pos].salt_len;
572 const u32 search[4] =
574 digests_buf[digests_offset].digest_buf[DGST_R0],
575 digests_buf[digests_offset].digest_buf[DGST_R1],
576 digests_buf[digests_offset].digest_buf[DGST_R2],
577 digests_buf[digests_offset].digest_buf[DGST_R3]
584 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
614 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
616 append_0x80_2x4 (w0, w1, out_len);
620 w0_t[0] = swap_workaround (w0[0]);
621 w0_t[1] = swap_workaround (w0[1]);
622 w0_t[2] = swap_workaround (w0[2]);
623 w0_t[3] = swap_workaround (w0[3]);
627 w1_t[0] = swap_workaround (w1[0]);
628 w1_t[1] = swap_workaround (w1[1]);
629 w1_t[2] = swap_workaround (w1[2]);
630 w1_t[3] = swap_workaround (w1[3]);
634 w2_t[0] = swap_workaround (w2[0]);
635 w2_t[1] = swap_workaround (w2[1]);
636 w2_t[2] = swap_workaround (w2[2]);
637 w2_t[3] = swap_workaround (w2[3]);
641 w3_t[0] = swap_workaround (w3[0]);
642 w3_t[1] = swap_workaround (w3[1]);
644 w3_t[3] = pw_len * 8;
654 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
668 w0_t[0] = salt_buf0[0];
669 w0_t[1] = salt_buf0[1];
670 w0_t[2] = salt_buf0[2];
671 w0_t[3] = salt_buf0[3];
672 w1_t[0] = salt_buf1[0];
673 w1_t[1] = salt_buf1[1];
674 w1_t[2] = salt_buf1[2];
675 w1_t[3] = salt_buf1[3];
676 w2_t[0] = salt_buf2[0];
677 w2_t[1] = salt_buf2[1];
678 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
679 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
680 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
681 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
682 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
683 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
684 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
685 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
686 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
687 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
688 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
689 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
697 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
699 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
700 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
701 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
702 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
703 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
704 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
705 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
706 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
707 w1_t[0] = 0x80000000;
718 w3_t[3] = (salt_len + 40) * 8;
720 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
728 w0_t[0] = salt_buf0[0];
729 w0_t[1] = salt_buf0[1];
730 w0_t[2] = salt_buf0[2];
731 w0_t[3] = salt_buf0[3];
732 w1_t[0] = salt_buf1[0];
733 w1_t[1] = salt_buf1[1];
734 w1_t[2] = salt_buf1[2];
735 w1_t[3] = salt_buf1[3];
736 w2_t[0] = salt_buf2[0];
737 w2_t[1] = salt_buf2[1];
738 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
739 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
740 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
741 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
742 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
743 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
744 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
745 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
746 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
747 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
748 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
749 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
757 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
759 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
760 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
761 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
762 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
763 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
764 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
765 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
766 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
767 w1_t[0] = 0x80000000;
778 w3_t[3] = (salt_len + 40) * 8;
780 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
782 const u32 r0 = digest[3];
783 const u32 r1 = digest[4];
784 const u32 r2 = digest[2];
785 const u32 r3 = digest[1];
791 __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)
795 __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)