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 "OpenCL/types_ocl.c"
18 #include "OpenCL/common.c"
20 #define COMPARE_S "OpenCL/check_single_comp4.c"
21 #define COMPARE_M "OpenCL/check_multi_comp4.c"
23 #define GETSHIFTEDINT(a,n) amd_bytealign ((a)[((n)/4)+1], (a)[((n)/4)+0], (n))
25 #define SETSHIFTEDINT(a,n,v) \
27 const u32 s = ((n) & 3) * 8; \
28 const u64 x = (u64) (v) << s; \
29 (a)[((n)/4)+0] |= x; \
30 (a)[((n)/4)+1] = x >> 32; \
33 __constant u32 theMagicArray[64] =
35 0x1451ac91,0x4354679f,0xe03be724,0xc27b7428,0xeb133386,0x5ccb4f5a,0x37730a08,0x2f1c5d0e,
36 0xe5e68f33,0xddae9bf8,0x8d4bf216,0xdcd4e12c,0x9ddfcbb0,0x176d70d4,0x3f424df9,0x94111b9b,
37 0x9bc15b9f,0x039d0506,0x8a135e9d,0xe86a9a1e,0x17147cd9,0xf62ac758,0x0a6399a1,0xc370fdd7,
38 0x13745ef6,0x040bc903,0x26f79826,0x2593928a,0x230da2b0,0x6d7963ed,0x3cfa3213,0xa39a0235,
39 0x0a8eddb3,0xc351bf24,0x9f55cd7c,0x4c94af37,0x82520829,0x374e3bb2,0x9107179f,0xcdfd3b11,
40 0, 0, 0, 0, 0, 0, 0, 0,
41 0, 0, 0, 0, 0, 0, 0, 0,
42 0, 0, 0, 0, 0, 0, 0, 0
45 static void swap_buffer (u32 final[16])
47 final[ 0] = swap32 (final[ 0]);
48 final[ 1] = swap32 (final[ 1]);
49 final[ 2] = swap32 (final[ 2]);
50 final[ 3] = swap32 (final[ 3]);
51 final[ 4] = swap32 (final[ 4]);
52 final[ 5] = swap32 (final[ 5]);
53 final[ 6] = swap32 (final[ 6]);
54 final[ 7] = swap32 (final[ 7]);
55 final[ 8] = swap32 (final[ 8]);
56 final[ 9] = swap32 (final[ 9]);
57 final[10] = swap32 (final[10]);
58 final[11] = swap32 (final[11]);
59 final[12] = swap32 (final[12]);
60 final[13] = swap32 (final[13]);
61 final[14] = swap32 (final[14]);
62 final[15] = swap32 (final[15]);
65 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
93 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
94 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
95 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
96 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
97 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
98 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
99 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
100 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
101 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
102 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
103 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
104 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
105 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
106 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
107 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
108 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
109 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
110 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
111 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
112 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
117 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
118 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
119 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
120 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
121 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
122 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
123 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
124 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
125 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
126 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
127 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
128 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
129 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
130 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
131 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
132 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
133 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
134 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
135 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
136 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
141 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
142 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
143 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
144 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
145 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
146 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
147 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
148 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
149 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
150 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
151 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
152 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
153 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
154 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
155 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
156 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
157 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
158 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
159 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
160 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
165 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
166 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
167 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
168 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
169 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
170 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
171 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
172 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
173 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
174 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
175 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
176 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
177 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
178 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
179 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
180 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
181 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
182 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
183 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
184 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
193 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_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)
199 const u32 lid = get_local_id (0);
205 const u32 gid = get_global_id (0);
207 if (gid >= gid_max) return;
211 wordl0[0] = pws[gid].i[ 0];
212 wordl0[1] = pws[gid].i[ 1];
213 wordl0[2] = pws[gid].i[ 2];
214 wordl0[3] = pws[gid].i[ 3];
218 wordl1[0] = pws[gid].i[ 4];
219 wordl1[1] = pws[gid].i[ 5];
220 wordl1[2] = pws[gid].i[ 6];
221 wordl1[3] = pws[gid].i[ 7];
237 const u32 pw_l_len = pws[gid].pw_len;
239 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
241 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
250 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
251 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
252 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
253 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
254 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
255 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
256 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
257 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
259 const u32 salt_len = salt_bufs[salt_pos].salt_len;
265 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
267 const u32 pw_r_len = combs_buf[il_pos].pw_len;
269 const u32 pw_len = pw_l_len + pw_r_len;
273 wordr0[0] = combs_buf[il_pos].i[0];
274 wordr0[1] = combs_buf[il_pos].i[1];
275 wordr0[2] = combs_buf[il_pos].i[2];
276 wordr0[3] = combs_buf[il_pos].i[3];
280 wordr1[0] = combs_buf[il_pos].i[4];
281 wordr1[1] = combs_buf[il_pos].i[5];
282 wordr1[2] = combs_buf[il_pos].i[6];
283 wordr1[3] = combs_buf[il_pos].i[7];
299 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
301 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
336 switch_buffer_by_offset (s0, s1, s2, s3, pw_len);
338 const u32 pw_salt_len = pw_len + salt_len;
342 w0[0] = wordl0[0] | wordr0[0];
343 w0[1] = wordl0[1] | wordr0[1];
344 w0[2] = wordl0[2] | wordr0[2];
345 w0[3] = wordl0[3] | wordr0[3];
349 w1[0] = wordl1[0] | wordr1[0];
350 w1[1] = wordl1[1] | wordr1[1];
351 w1[2] = wordl1[2] | wordr1[2];
352 w1[3] = wordl1[3] | wordr1[3];
356 w2[0] = wordl2[0] | wordr2[0];
357 w2[1] = wordl2[1] | wordr2[1];
358 w2[2] = wordl2[2] | wordr2[2];
359 w2[3] = wordl2[3] | wordr2[3];
363 w3[0] = wordl3[0] | wordr3[0];
364 w3[1] = wordl3[1] | wordr3[1];
374 final[ 0] = swap32 (w0[0] | s0[0]);
375 final[ 1] = swap32 (w0[1] | s0[1]);
376 final[ 2] = swap32 (w0[2] | s0[2]);
377 final[ 3] = swap32 (w0[3] | s0[3]);
378 final[ 4] = swap32 (w1[0] | s1[0]);
379 final[ 5] = swap32 (w1[1] | s1[1]);
380 final[ 6] = swap32 (w1[2] | s1[2]);
381 final[ 7] = swap32 (w1[3] | s1[3]);
382 final[ 8] = swap32 (w2[0] | s2[0]);
383 final[ 9] = swap32 (w2[1] | s2[1]);
384 final[10] = swap32 (w2[2] | s2[2]);
385 final[11] = swap32 (w2[3] | s2[3]);
386 final[12] = swap32 (w3[0] | s3[0]);
387 final[13] = swap32 (w3[1] | s3[1]);
389 final[15] = pw_salt_len * 8;
399 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
401 // prepare magic array range
403 u32 lengthMagicArray = 0x20;
404 u32 offsetMagicArray = 0;
406 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
407 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
408 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
409 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
410 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
411 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
412 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
413 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
414 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
415 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
416 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
417 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
418 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
419 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
420 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
421 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
422 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
423 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
424 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
425 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
436 for (int i = 0; i < 32; i++) final[i] = 0;
447 u32 final_len = pw_len;
453 for (i = 0; i < lengthMagicArray - 4; i += 4)
455 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
457 SETSHIFTEDINT (final, final_len + i, tmp);
460 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
462 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
464 SETSHIFTEDINT (final, final_len + i, tmp);
466 final_len += lengthMagicArray;
470 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
472 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
474 SETSHIFTEDINT (final, final_len + i, tmp);
477 final_len += salt_len;
484 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
486 swap_buffer (&final[off]);
488 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
491 swap_buffer (&final[off]);
494 final[off + 15] = final_len * 8;
496 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
498 const u32 r0 = digest[3];
499 const u32 r1 = digest[4];
500 const u32 r2 = digest[2];
501 const u32 r3 = digest[1];
507 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_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)
511 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_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)
515 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_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)
521 const u32 lid = get_local_id (0);
527 const u32 gid = get_global_id (0);
529 if (gid >= gid_max) return;
533 wordl0[0] = pws[gid].i[ 0];
534 wordl0[1] = pws[gid].i[ 1];
535 wordl0[2] = pws[gid].i[ 2];
536 wordl0[3] = pws[gid].i[ 3];
540 wordl1[0] = pws[gid].i[ 4];
541 wordl1[1] = pws[gid].i[ 5];
542 wordl1[2] = pws[gid].i[ 6];
543 wordl1[3] = pws[gid].i[ 7];
559 const u32 pw_l_len = pws[gid].pw_len;
561 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
563 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
566 const u32 search[4] =
568 digests_buf[digests_offset].digest_buf[DGST_R0],
569 digests_buf[digests_offset].digest_buf[DGST_R1],
570 digests_buf[digests_offset].digest_buf[DGST_R2],
571 digests_buf[digests_offset].digest_buf[DGST_R3]
580 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
581 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
582 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
583 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
584 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
585 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
586 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
587 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
589 const u32 salt_len = salt_bufs[salt_pos].salt_len;
595 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
597 const u32 pw_r_len = combs_buf[il_pos].pw_len;
599 const u32 pw_len = pw_l_len + pw_r_len;
603 wordr0[0] = combs_buf[il_pos].i[0];
604 wordr0[1] = combs_buf[il_pos].i[1];
605 wordr0[2] = combs_buf[il_pos].i[2];
606 wordr0[3] = combs_buf[il_pos].i[3];
610 wordr1[0] = combs_buf[il_pos].i[4];
611 wordr1[1] = combs_buf[il_pos].i[5];
612 wordr1[2] = combs_buf[il_pos].i[6];
613 wordr1[3] = combs_buf[il_pos].i[7];
629 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
631 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
666 switch_buffer_by_offset (s0, s1, s2, s3, pw_len);
668 const u32 pw_salt_len = pw_len + salt_len;
672 w0[0] = wordl0[0] | wordr0[0];
673 w0[1] = wordl0[1] | wordr0[1];
674 w0[2] = wordl0[2] | wordr0[2];
675 w0[3] = wordl0[3] | wordr0[3];
679 w1[0] = wordl1[0] | wordr1[0];
680 w1[1] = wordl1[1] | wordr1[1];
681 w1[2] = wordl1[2] | wordr1[2];
682 w1[3] = wordl1[3] | wordr1[3];
686 w2[0] = wordl2[0] | wordr2[0];
687 w2[1] = wordl2[1] | wordr2[1];
688 w2[2] = wordl2[2] | wordr2[2];
689 w2[3] = wordl2[3] | wordr2[3];
693 w3[0] = wordl3[0] | wordr3[0];
694 w3[1] = wordl3[1] | wordr3[1];
704 final[ 0] = swap32 (w0[0] | s0[0]);
705 final[ 1] = swap32 (w0[1] | s0[1]);
706 final[ 2] = swap32 (w0[2] | s0[2]);
707 final[ 3] = swap32 (w0[3] | s0[3]);
708 final[ 4] = swap32 (w1[0] | s1[0]);
709 final[ 5] = swap32 (w1[1] | s1[1]);
710 final[ 6] = swap32 (w1[2] | s1[2]);
711 final[ 7] = swap32 (w1[3] | s1[3]);
712 final[ 8] = swap32 (w2[0] | s2[0]);
713 final[ 9] = swap32 (w2[1] | s2[1]);
714 final[10] = swap32 (w2[2] | s2[2]);
715 final[11] = swap32 (w2[3] | s2[3]);
716 final[12] = swap32 (w3[0] | s3[0]);
717 final[13] = swap32 (w3[1] | s3[1]);
719 final[15] = pw_salt_len * 8;
729 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
731 // prepare magic array range
733 u32 lengthMagicArray = 0x20;
734 u32 offsetMagicArray = 0;
736 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
737 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
738 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
739 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
740 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
741 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
742 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
743 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
744 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
745 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
746 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
747 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
748 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
749 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
750 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
751 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
752 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
753 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
754 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
755 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
766 for (int i = 0; i < 32; i++) final[i] = 0;
777 u32 final_len = pw_len;
783 for (i = 0; i < lengthMagicArray - 4; i += 4)
785 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
787 SETSHIFTEDINT (final, final_len + i, tmp);
790 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
792 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
794 SETSHIFTEDINT (final, final_len + i, tmp);
796 final_len += lengthMagicArray;
800 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
802 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
804 SETSHIFTEDINT (final, final_len + i, tmp);
807 final_len += salt_len;
814 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
816 swap_buffer (&final[off]);
818 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
821 swap_buffer (&final[off]);
824 final[off + 15] = final_len * 8;
826 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
828 const u32 r0 = digest[3];
829 const u32 r1 = digest[4];
830 const u32 r2 = digest[2];
831 const u32 r3 = digest[1];
837 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_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)
841 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07800_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)