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"
21 #define COMPARE_M "check_multi_vect1_comp4.c"
25 #define COMPARE_M "check_multi_vect2_comp4.c"
29 #define COMPARE_M "check_multi_vect4_comp4.c"
32 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
60 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
61 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
62 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
63 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
64 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
65 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
66 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
67 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
68 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
69 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
70 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
71 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
72 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
73 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
74 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
75 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
76 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
77 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
78 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
79 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
84 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
85 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
86 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
87 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
88 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
89 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
90 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
91 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
92 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
93 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
94 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
95 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
96 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
97 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
98 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
99 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
100 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
101 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
102 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
103 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
108 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
109 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
110 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
111 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
112 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
113 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
114 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
115 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
116 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
117 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
118 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
119 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
120 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
121 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
122 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
123 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
124 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
125 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
126 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
127 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
132 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
133 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
134 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
135 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
136 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
137 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
138 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
139 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
140 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
141 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
142 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
143 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
144 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
145 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
146 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
147 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
148 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
149 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
150 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
151 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
160 static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
162 w0[0] = w0[0] ^ 0x36363636;
163 w0[1] = w0[1] ^ 0x36363636;
164 w0[2] = w0[2] ^ 0x36363636;
165 w0[3] = w0[3] ^ 0x36363636;
166 w1[0] = w1[0] ^ 0x36363636;
167 w1[1] = w1[1] ^ 0x36363636;
168 w1[2] = w1[2] ^ 0x36363636;
169 w1[3] = w1[3] ^ 0x36363636;
170 w2[0] = w2[0] ^ 0x36363636;
171 w2[1] = w2[1] ^ 0x36363636;
172 w2[2] = w2[2] ^ 0x36363636;
173 w2[3] = w2[3] ^ 0x36363636;
174 w3[0] = w3[0] ^ 0x36363636;
175 w3[1] = w3[1] ^ 0x36363636;
176 w3[2] = w3[2] ^ 0x36363636;
177 w3[3] = w3[3] ^ 0x36363636;
185 sha1_transform (w0, w1, w2, w3, ipad);
187 w0[0] = w0[0] ^ 0x6a6a6a6a;
188 w0[1] = w0[1] ^ 0x6a6a6a6a;
189 w0[2] = w0[2] ^ 0x6a6a6a6a;
190 w0[3] = w0[3] ^ 0x6a6a6a6a;
191 w1[0] = w1[0] ^ 0x6a6a6a6a;
192 w1[1] = w1[1] ^ 0x6a6a6a6a;
193 w1[2] = w1[2] ^ 0x6a6a6a6a;
194 w1[3] = w1[3] ^ 0x6a6a6a6a;
195 w2[0] = w2[0] ^ 0x6a6a6a6a;
196 w2[1] = w2[1] ^ 0x6a6a6a6a;
197 w2[2] = w2[2] ^ 0x6a6a6a6a;
198 w2[3] = w2[3] ^ 0x6a6a6a6a;
199 w3[0] = w3[0] ^ 0x6a6a6a6a;
200 w3[1] = w3[1] ^ 0x6a6a6a6a;
201 w3[2] = w3[2] ^ 0x6a6a6a6a;
202 w3[3] = w3[3] ^ 0x6a6a6a6a;
210 sha1_transform (w0, w1, w2, w3, opad);
213 static void hmac_sha1_run (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5], u32 digest[5])
221 sha1_transform (w0, w1, w2, w3, digest);
238 w3[3] = (64 + 20) * 8;
246 sha1_transform (w0, w1, w2, w3, digest);
249 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m06700_init (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global sha1aix_tmp_t *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 wpa_t *wpa_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)
255 const u32 gid = get_global_id (0);
257 if (gid >= gid_max) return;
261 w0[0] = pws[gid].i[ 0];
262 w0[1] = pws[gid].i[ 1];
263 w0[2] = pws[gid].i[ 2];
264 w0[3] = pws[gid].i[ 3];
268 w1[0] = pws[gid].i[ 4];
269 w1[1] = pws[gid].i[ 5];
270 w1[2] = pws[gid].i[ 6];
271 w1[3] = pws[gid].i[ 7];
275 w2[0] = pws[gid].i[ 8];
276 w2[1] = pws[gid].i[ 9];
277 w2[2] = pws[gid].i[10];
278 w2[3] = pws[gid].i[11];
282 w3[0] = pws[gid].i[12];
283 w3[1] = pws[gid].i[13];
284 w3[2] = pws[gid].i[14];
285 w3[3] = pws[gid].i[15];
291 u32 salt_len = salt_bufs[salt_pos].salt_len;
295 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
296 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
297 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
298 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
302 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
303 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
304 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
305 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
309 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
310 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
311 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
312 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
321 append_0x01_4 (salt_buf0, salt_buf1, salt_buf2, salt_buf3, salt_len + 3);
323 append_0x80_4 (salt_buf0, salt_buf1, salt_buf2, salt_buf3, salt_len + 4);
329 w0[0] = swap_workaround (w0[0]);
330 w0[1] = swap_workaround (w0[1]);
331 w0[2] = swap_workaround (w0[2]);
332 w0[3] = swap_workaround (w0[3]);
333 w1[0] = swap_workaround (w1[0]);
334 w1[1] = swap_workaround (w1[1]);
335 w1[2] = swap_workaround (w1[2]);
336 w1[3] = swap_workaround (w1[3]);
337 w2[0] = swap_workaround (w2[0]);
338 w2[1] = swap_workaround (w2[1]);
339 w2[2] = swap_workaround (w2[2]);
340 w2[3] = swap_workaround (w2[3]);
341 w3[0] = swap_workaround (w3[0]);
342 w3[1] = swap_workaround (w3[1]);
343 w3[2] = swap_workaround (w3[2]);
344 w3[3] = swap_workaround (w3[3]);
349 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
351 tmps[gid].ipad[0] = ipad[0];
352 tmps[gid].ipad[1] = ipad[1];
353 tmps[gid].ipad[2] = ipad[2];
354 tmps[gid].ipad[3] = ipad[3];
355 tmps[gid].ipad[4] = ipad[4];
357 tmps[gid].opad[0] = opad[0];
358 tmps[gid].opad[1] = opad[1];
359 tmps[gid].opad[2] = opad[2];
360 tmps[gid].opad[3] = opad[3];
361 tmps[gid].opad[4] = opad[4];
363 w0[0] = salt_buf0[0];
364 w0[1] = salt_buf0[1];
365 w0[2] = salt_buf0[2];
366 w0[3] = salt_buf0[3];
367 w1[0] = salt_buf1[0];
368 w1[1] = salt_buf1[1];
369 w1[2] = salt_buf1[2];
370 w1[3] = salt_buf1[3];
371 w2[0] = salt_buf2[0];
372 w2[1] = salt_buf2[1];
373 w2[2] = salt_buf2[2];
374 w2[3] = salt_buf2[3];
375 w3[0] = salt_buf3[0];
376 w3[1] = salt_buf3[1];
377 w3[2] = salt_buf3[2];
378 //w3[3] = salt_buf3[3];
380 w0[0] = swap_workaround (w0[0]);
381 w0[1] = swap_workaround (w0[1]);
382 w0[2] = swap_workaround (w0[2]);
383 w0[3] = swap_workaround (w0[3]);
384 w1[0] = swap_workaround (w1[0]);
385 w1[1] = swap_workaround (w1[1]);
386 w1[2] = swap_workaround (w1[2]);
387 w1[3] = swap_workaround (w1[3]);
388 w2[0] = swap_workaround (w2[0]);
389 w2[1] = swap_workaround (w2[1]);
390 w2[2] = swap_workaround (w2[2]);
391 w2[3] = swap_workaround (w2[3]);
392 w3[0] = swap_workaround (w3[0]);
393 w3[1] = swap_workaround (w3[1]);
394 w3[2] = swap_workaround (w3[2]);
395 w3[3] = (64 + salt_len + 4) * 8;
399 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
401 tmps[gid].dgst[0] = dgst[0];
402 tmps[gid].dgst[1] = dgst[1];
403 tmps[gid].dgst[2] = dgst[2];
404 tmps[gid].dgst[3] = dgst[3];
405 tmps[gid].dgst[4] = dgst[4];
407 tmps[gid].out[0] = dgst[0];
408 tmps[gid].out[1] = dgst[1];
409 tmps[gid].out[2] = dgst[2];
410 tmps[gid].out[3] = dgst[3];
411 tmps[gid].out[4] = dgst[4];
414 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m06700_loop (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global sha1aix_tmp_t *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 wpa_t *wpa_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)
416 const u32 gid = get_global_id (0);
418 if (gid >= gid_max) return;
423 ipad[0] = tmps[gid].ipad[0];
424 ipad[1] = tmps[gid].ipad[1];
425 ipad[2] = tmps[gid].ipad[2];
426 ipad[3] = tmps[gid].ipad[3];
427 ipad[4] = tmps[gid].ipad[4];
429 opad[0] = tmps[gid].opad[0];
430 opad[1] = tmps[gid].opad[1];
431 opad[2] = tmps[gid].opad[2];
432 opad[3] = tmps[gid].opad[3];
433 opad[4] = tmps[gid].opad[4];
438 dgst[0] = tmps[gid].dgst[0];
439 dgst[1] = tmps[gid].dgst[1];
440 dgst[2] = tmps[gid].dgst[2];
441 dgst[3] = tmps[gid].dgst[3];
442 dgst[4] = tmps[gid].dgst[4];
444 out[0] = tmps[gid].out[0];
445 out[1] = tmps[gid].out[1];
446 out[2] = tmps[gid].out[2];
447 out[3] = tmps[gid].out[3];
448 out[4] = tmps[gid].out[4];
450 for (u32 j = 0; j < loop_cnt; j++)
472 w3[3] = (64 + 20) * 8;
474 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
483 tmps[gid].dgst[0] = dgst[0];
484 tmps[gid].dgst[1] = dgst[1];
485 tmps[gid].dgst[2] = dgst[2];
486 tmps[gid].dgst[3] = dgst[3];
487 tmps[gid].dgst[4] = dgst[4];
489 tmps[gid].out[0] = out[0];
490 tmps[gid].out[1] = out[1];
491 tmps[gid].out[2] = out[2];
492 tmps[gid].out[3] = out[3];
493 tmps[gid].out[4] = out[4];
496 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m06700_comp (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global sha1aix_tmp_t *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 wpa_t *wpa_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)
502 const u32 gid = get_global_id (0);
504 if (gid >= gid_max) return;
506 const u32 lid = get_local_id (0);
513 u32 a = tmps[gid].out[0];
514 u32 b = tmps[gid].out[1];
515 u32 c = tmps[gid].out[2];
516 u32 d = tmps[gid].out[3];
517 u32 e = tmps[gid].out[4] & 0xffff03ff;
520 const u32 r0 = tmps[gid].out[DGST_R0];
521 const u32 r1 = tmps[gid].out[DGST_R1];
522 const u32 r2 = tmps[gid].out[DGST_R2];
523 const u32 r3 = tmps[gid].out[DGST_R3];