2 * Author......: Jens Steube <jens.steube@gmail.com>
8 //incompatible data-dependant code
9 //#define NEW_SIMD_CODE
11 #include "inc_vendor.cl"
12 #include "inc_hash_constants.h"
13 #include "inc_hash_functions.cl"
14 #include "inc_types.cl"
15 #include "inc_common.cl"
16 #include "inc_simd.cl"
18 #define GETSHIFTEDINT(a,n) amd_bytealign ((a)[((n)/4)+1], (a)[((n)/4)+0], (n))
20 #define SETSHIFTEDINT(a,n,v) \
22 const u32 s = ((n) & 3) * 8; \
23 const u64 x = (u64) (v) << s; \
24 (a)[((n)/4)+0] |= x; \
25 (a)[((n)/4)+1] = x >> 32; \
28 __constant u32 theMagicArray[64] =
30 0x1451ac91,0x4354679f,0xe03be724,0xc27b7428,0xeb133386,0x5ccb4f5a,0x37730a08,0x2f1c5d0e,
31 0xe5e68f33,0xddae9bf8,0x8d4bf216,0xdcd4e12c,0x9ddfcbb0,0x176d70d4,0x3f424df9,0x94111b9b,
32 0x9bc15b9f,0x039d0506,0x8a135e9d,0xe86a9a1e,0x17147cd9,0xf62ac758,0x0a6399a1,0xc370fdd7,
33 0x13745ef6,0x040bc903,0x26f79826,0x2593928a,0x230da2b0,0x6d7963ed,0x3cfa3213,0xa39a0235,
34 0x0a8eddb3,0xc351bf24,0x9f55cd7c,0x4c94af37,0x82520829,0x374e3bb2,0x9107179f,0xcdfd3b11,
35 0, 0, 0, 0, 0, 0, 0, 0,
36 0, 0, 0, 0, 0, 0, 0, 0,
37 0, 0, 0, 0, 0, 0, 0, 0
40 void swap_buffer (u32 final[16])
42 final[ 0] = swap32 (final[ 0]);
43 final[ 1] = swap32 (final[ 1]);
44 final[ 2] = swap32 (final[ 2]);
45 final[ 3] = swap32 (final[ 3]);
46 final[ 4] = swap32 (final[ 4]);
47 final[ 5] = swap32 (final[ 5]);
48 final[ 6] = swap32 (final[ 6]);
49 final[ 7] = swap32 (final[ 7]);
50 final[ 8] = swap32 (final[ 8]);
51 final[ 9] = swap32 (final[ 9]);
52 final[10] = swap32 (final[10]);
53 final[11] = swap32 (final[11]);
54 final[12] = swap32 (final[12]);
55 final[13] = swap32 (final[13]);
56 final[14] = swap32 (final[14]);
57 final[15] = swap32 (final[15]);
60 void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
88 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
89 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
90 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
91 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
92 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
93 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
94 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
95 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
96 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
97 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
98 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
99 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
100 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
101 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
102 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
103 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
104 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
105 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
106 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
107 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
112 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
113 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
114 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
115 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
116 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
117 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
118 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
119 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
120 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
121 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
122 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
123 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
124 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
125 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
126 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
127 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
128 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
129 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
130 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
131 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
136 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
137 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
138 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
139 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
140 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
141 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
142 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
143 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
144 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
145 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
146 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
147 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
148 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
149 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
150 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
151 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
152 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
153 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
154 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
155 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
160 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
161 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
162 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
163 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
164 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
165 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
166 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
167 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
168 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
169 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
170 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
171 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
172 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
173 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
174 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
175 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
176 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
177 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
178 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
179 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
188 __kernel void m07800_m04 (__global pw_t *pws, __global kernel_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
194 const u32 lid = get_local_id (0);
200 const u32 gid = get_global_id (0);
202 if (gid >= gid_max) return;
207 pw_buf0[0] = pws[gid].i[0];
208 pw_buf0[1] = pws[gid].i[1];
209 pw_buf0[2] = pws[gid].i[2];
210 pw_buf0[3] = pws[gid].i[3];
211 pw_buf1[0] = pws[gid].i[4];
212 pw_buf1[1] = pws[gid].i[5];
213 pw_buf1[2] = pws[gid].i[6];
214 pw_buf1[3] = pws[gid].i[7];
216 const u32 pw_l_len = pws[gid].pw_len;
224 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
225 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
226 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
227 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
228 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
229 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
230 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
231 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
233 const u32 salt_len = salt_bufs[salt_pos].salt_len;
239 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
241 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
243 const u32x pw_len = pw_l_len + pw_r_len;
246 * concat password candidate
249 u32x wordl0[4] = { 0 };
250 u32x wordl1[4] = { 0 };
251 u32x wordl2[4] = { 0 };
252 u32x wordl3[4] = { 0 };
254 wordl0[0] = pw_buf0[0];
255 wordl0[1] = pw_buf0[1];
256 wordl0[2] = pw_buf0[2];
257 wordl0[3] = pw_buf0[3];
258 wordl1[0] = pw_buf1[0];
259 wordl1[1] = pw_buf1[1];
260 wordl1[2] = pw_buf1[2];
261 wordl1[3] = pw_buf1[3];
263 u32x wordr0[4] = { 0 };
264 u32x wordr1[4] = { 0 };
265 u32x wordr2[4] = { 0 };
266 u32x wordr3[4] = { 0 };
268 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
269 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
270 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
271 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
272 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
273 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
274 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
275 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
277 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
279 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
283 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
291 w0[0] = wordl0[0] | wordr0[0];
292 w0[1] = wordl0[1] | wordr0[1];
293 w0[2] = wordl0[2] | wordr0[2];
294 w0[3] = wordl0[3] | wordr0[3];
295 w1[0] = wordl1[0] | wordr1[0];
296 w1[1] = wordl1[1] | wordr1[1];
297 w1[2] = wordl1[2] | wordr1[2];
298 w1[3] = wordl1[3] | wordr1[3];
299 w2[0] = wordl2[0] | wordr2[0];
300 w2[1] = wordl2[1] | wordr2[1];
301 w2[2] = wordl2[2] | wordr2[2];
302 w2[3] = wordl2[3] | wordr2[3];
303 w3[0] = wordl3[0] | wordr3[0];
304 w3[1] = wordl3[1] | wordr3[1];
305 w3[2] = wordl3[2] | wordr3[2];
306 w3[3] = wordl3[3] | wordr3[3];
334 switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
336 const u32x pw_salt_len = pw_len + salt_len;
344 final[ 0] = swap32 (w0[0] | s0[0]);
345 final[ 1] = swap32 (w0[1] | s0[1]);
346 final[ 2] = swap32 (w0[2] | s0[2]);
347 final[ 3] = swap32 (w0[3] | s0[3]);
348 final[ 4] = swap32 (w1[0] | s1[0]);
349 final[ 5] = swap32 (w1[1] | s1[1]);
350 final[ 6] = swap32 (w1[2] | s1[2]);
351 final[ 7] = swap32 (w1[3] | s1[3]);
352 final[ 8] = swap32 (w2[0] | s2[0]);
353 final[ 9] = swap32 (w2[1] | s2[1]);
354 final[10] = swap32 (w2[2] | s2[2]);
355 final[11] = swap32 (w2[3] | s2[3]);
356 final[12] = swap32 (w3[0] | s3[0]);
357 final[13] = swap32 (w3[1] | s3[1]);
359 final[15] = pw_salt_len * 8;
369 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
371 // prepare magic array range
373 u32 lengthMagicArray = 0x20;
374 u32 offsetMagicArray = 0;
376 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
377 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
378 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
379 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
380 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
381 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
382 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
383 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
384 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
385 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
386 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
387 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
388 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
389 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
390 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
391 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
392 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
393 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
394 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
395 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
408 for (int i = 0; i < 32; i++) final[i] = 0;
419 u32 final_len = pw_len;
425 for (i = 0; i < lengthMagicArray - 4; i += 4)
427 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
429 SETSHIFTEDINT (final, final_len + i, tmp);
432 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
434 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
436 SETSHIFTEDINT (final, final_len + i, tmp);
438 final_len += lengthMagicArray;
442 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
444 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
446 SETSHIFTEDINT (final, final_len + i, tmp);
449 final_len += salt_len;
456 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
458 swap_buffer (&final[off]);
460 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
463 swap_buffer (&final[off]);
466 final[off + 15] = final_len * 8;
468 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
470 COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
474 __kernel void m07800_m08 (__global pw_t *pws, __global kernel_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
478 __kernel void m07800_m16 (__global pw_t *pws, __global kernel_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
482 __kernel void m07800_s04 (__global pw_t *pws, __global kernel_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
488 const u32 lid = get_local_id (0);
494 const u32 gid = get_global_id (0);
496 if (gid >= gid_max) return;
501 pw_buf0[0] = pws[gid].i[0];
502 pw_buf0[1] = pws[gid].i[1];
503 pw_buf0[2] = pws[gid].i[2];
504 pw_buf0[3] = pws[gid].i[3];
505 pw_buf1[0] = pws[gid].i[4];
506 pw_buf1[1] = pws[gid].i[5];
507 pw_buf1[2] = pws[gid].i[6];
508 pw_buf1[3] = pws[gid].i[7];
510 const u32 pw_l_len = pws[gid].pw_len;
518 salt_buf[0] = salt_bufs[salt_pos].salt_buf[0];
519 salt_buf[1] = salt_bufs[salt_pos].salt_buf[1];
520 salt_buf[2] = salt_bufs[salt_pos].salt_buf[2];
521 salt_buf[3] = salt_bufs[salt_pos].salt_buf[3];
522 salt_buf[4] = salt_bufs[salt_pos].salt_buf[4];
523 salt_buf[5] = salt_bufs[salt_pos].salt_buf[5];
524 salt_buf[6] = salt_bufs[salt_pos].salt_buf[6];
525 salt_buf[7] = salt_bufs[salt_pos].salt_buf[7];
527 const u32 salt_len = salt_bufs[salt_pos].salt_len;
533 const u32 search[4] =
535 digests_buf[digests_offset].digest_buf[DGST_R0],
536 digests_buf[digests_offset].digest_buf[DGST_R1],
537 digests_buf[digests_offset].digest_buf[DGST_R2],
538 digests_buf[digests_offset].digest_buf[DGST_R3]
545 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
547 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
549 const u32x pw_len = pw_l_len + pw_r_len;
552 * concat password candidate
555 u32x wordl0[4] = { 0 };
556 u32x wordl1[4] = { 0 };
557 u32x wordl2[4] = { 0 };
558 u32x wordl3[4] = { 0 };
560 wordl0[0] = pw_buf0[0];
561 wordl0[1] = pw_buf0[1];
562 wordl0[2] = pw_buf0[2];
563 wordl0[3] = pw_buf0[3];
564 wordl1[0] = pw_buf1[0];
565 wordl1[1] = pw_buf1[1];
566 wordl1[2] = pw_buf1[2];
567 wordl1[3] = pw_buf1[3];
569 u32x wordr0[4] = { 0 };
570 u32x wordr1[4] = { 0 };
571 u32x wordr2[4] = { 0 };
572 u32x wordr3[4] = { 0 };
574 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
575 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
576 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
577 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
578 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
579 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
580 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
581 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
583 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
585 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
589 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
597 w0[0] = wordl0[0] | wordr0[0];
598 w0[1] = wordl0[1] | wordr0[1];
599 w0[2] = wordl0[2] | wordr0[2];
600 w0[3] = wordl0[3] | wordr0[3];
601 w1[0] = wordl1[0] | wordr1[0];
602 w1[1] = wordl1[1] | wordr1[1];
603 w1[2] = wordl1[2] | wordr1[2];
604 w1[3] = wordl1[3] | wordr1[3];
605 w2[0] = wordl2[0] | wordr2[0];
606 w2[1] = wordl2[1] | wordr2[1];
607 w2[2] = wordl2[2] | wordr2[2];
608 w2[3] = wordl2[3] | wordr2[3];
609 w3[0] = wordl3[0] | wordr3[0];
610 w3[1] = wordl3[1] | wordr3[1];
611 w3[2] = wordl3[2] | wordr3[2];
612 w3[3] = wordl3[3] | wordr3[3];
640 switch_buffer_by_offset_le_VV (s0, s1, s2, s3, pw_len);
642 const u32x pw_salt_len = pw_len + salt_len;
650 final[ 0] = swap32 (w0[0] | s0[0]);
651 final[ 1] = swap32 (w0[1] | s0[1]);
652 final[ 2] = swap32 (w0[2] | s0[2]);
653 final[ 3] = swap32 (w0[3] | s0[3]);
654 final[ 4] = swap32 (w1[0] | s1[0]);
655 final[ 5] = swap32 (w1[1] | s1[1]);
656 final[ 6] = swap32 (w1[2] | s1[2]);
657 final[ 7] = swap32 (w1[3] | s1[3]);
658 final[ 8] = swap32 (w2[0] | s2[0]);
659 final[ 9] = swap32 (w2[1] | s2[1]);
660 final[10] = swap32 (w2[2] | s2[2]);
661 final[11] = swap32 (w2[3] | s2[3]);
662 final[12] = swap32 (w3[0] | s3[0]);
663 final[13] = swap32 (w3[1] | s3[1]);
665 final[15] = pw_salt_len * 8;
675 sha1_transform (&final[0], &final[4], &final[8], &final[12], digest);
677 // prepare magic array range
679 u32 lengthMagicArray = 0x20;
680 u32 offsetMagicArray = 0;
682 lengthMagicArray += ((digest[0] >> 24) & 0xff) % 6;
683 lengthMagicArray += ((digest[0] >> 16) & 0xff) % 6;
684 lengthMagicArray += ((digest[0] >> 8) & 0xff) % 6;
685 lengthMagicArray += ((digest[0] >> 0) & 0xff) % 6;
686 lengthMagicArray += ((digest[1] >> 24) & 0xff) % 6;
687 lengthMagicArray += ((digest[1] >> 16) & 0xff) % 6;
688 lengthMagicArray += ((digest[1] >> 8) & 0xff) % 6;
689 lengthMagicArray += ((digest[1] >> 0) & 0xff) % 6;
690 lengthMagicArray += ((digest[2] >> 24) & 0xff) % 6;
691 lengthMagicArray += ((digest[2] >> 16) & 0xff) % 6;
692 offsetMagicArray += ((digest[2] >> 8) & 0xff) % 8;
693 offsetMagicArray += ((digest[2] >> 0) & 0xff) % 8;
694 offsetMagicArray += ((digest[3] >> 24) & 0xff) % 8;
695 offsetMagicArray += ((digest[3] >> 16) & 0xff) % 8;
696 offsetMagicArray += ((digest[3] >> 8) & 0xff) % 8;
697 offsetMagicArray += ((digest[3] >> 0) & 0xff) % 8;
698 offsetMagicArray += ((digest[4] >> 24) & 0xff) % 8;
699 offsetMagicArray += ((digest[4] >> 16) & 0xff) % 8;
700 offsetMagicArray += ((digest[4] >> 8) & 0xff) % 8;
701 offsetMagicArray += ((digest[4] >> 0) & 0xff) % 8;
714 for (int i = 0; i < 32; i++) final[i] = 0;
725 u32 final_len = pw_len;
731 for (i = 0; i < lengthMagicArray - 4; i += 4)
733 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i);
735 SETSHIFTEDINT (final, final_len + i, tmp);
738 const u32 mask = 0xffffffff >> (((i - lengthMagicArray) & 3) * 8);
740 const u32 tmp = GETSHIFTEDINT (theMagicArray, offsetMagicArray + i) & mask;
742 SETSHIFTEDINT (final, final_len + i, tmp);
744 final_len += lengthMagicArray;
748 for (i = 0; i < salt_len + 1; i += 4) // +1 for the 0x80
750 const u32 tmp = salt_buf[i / 4]; // attention, int[] not char[]
752 SETSHIFTEDINT (final, final_len + i, tmp);
755 final_len += salt_len;
762 for (left = final_len, off = 0; left >= 56; left -= 64, off += 16)
764 swap_buffer (&final[off]);
766 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
769 swap_buffer (&final[off]);
772 final[off + 15] = final_len * 8;
774 sha1_transform (&final[off + 0], &final[off + 4], &final[off + 8], &final[off + 12], digest);
776 COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
780 __kernel void m07800_s08 (__global pw_t *pws, __global kernel_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
784 __kernel void m07800_s16 (__global pw_t *pws, __global kernel_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)