2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "inc_vendor.cl"
9 #include "inc_hash_constants.h"
10 #include "inc_hash_functions.cl"
12 #include "inc_types.cl"
13 #include "inc_common.cl"
15 #define COMPARE_S "inc_comp_single.cl"
16 #define COMPARE_M "inc_comp_multi.cl"
18 void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
46 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
47 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
48 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
49 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
50 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
51 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
52 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
53 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
54 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
55 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
56 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
57 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
58 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
59 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
60 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
61 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
62 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
63 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
64 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
65 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
70 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
71 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
72 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
73 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
74 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
75 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
76 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
77 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
78 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
79 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
80 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
81 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
82 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
83 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
84 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
85 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
86 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
87 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
88 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
89 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
94 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
95 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
96 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
97 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
98 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
99 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
100 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
101 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
102 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
103 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
104 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
105 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
106 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
107 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
108 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
109 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
110 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
111 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
112 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
113 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
118 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
119 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
120 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
121 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
122 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
123 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
124 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
125 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
126 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
127 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
128 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
129 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
130 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
131 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
132 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
133 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
134 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
135 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
136 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
137 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
146 void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
148 w0[0] = w0[0] ^ 0x36363636;
149 w0[1] = w0[1] ^ 0x36363636;
150 w0[2] = w0[2] ^ 0x36363636;
151 w0[3] = w0[3] ^ 0x36363636;
152 w1[0] = w1[0] ^ 0x36363636;
153 w1[1] = w1[1] ^ 0x36363636;
154 w1[2] = w1[2] ^ 0x36363636;
155 w1[3] = w1[3] ^ 0x36363636;
156 w2[0] = w2[0] ^ 0x36363636;
157 w2[1] = w2[1] ^ 0x36363636;
158 w2[2] = w2[2] ^ 0x36363636;
159 w2[3] = w2[3] ^ 0x36363636;
160 w3[0] = w3[0] ^ 0x36363636;
161 w3[1] = w3[1] ^ 0x36363636;
162 w3[2] = w3[2] ^ 0x36363636;
163 w3[3] = w3[3] ^ 0x36363636;
171 sha1_transform (w0, w1, w2, w3, ipad);
173 w0[0] = w0[0] ^ 0x6a6a6a6a;
174 w0[1] = w0[1] ^ 0x6a6a6a6a;
175 w0[2] = w0[2] ^ 0x6a6a6a6a;
176 w0[3] = w0[3] ^ 0x6a6a6a6a;
177 w1[0] = w1[0] ^ 0x6a6a6a6a;
178 w1[1] = w1[1] ^ 0x6a6a6a6a;
179 w1[2] = w1[2] ^ 0x6a6a6a6a;
180 w1[3] = w1[3] ^ 0x6a6a6a6a;
181 w2[0] = w2[0] ^ 0x6a6a6a6a;
182 w2[1] = w2[1] ^ 0x6a6a6a6a;
183 w2[2] = w2[2] ^ 0x6a6a6a6a;
184 w2[3] = w2[3] ^ 0x6a6a6a6a;
185 w3[0] = w3[0] ^ 0x6a6a6a6a;
186 w3[1] = w3[1] ^ 0x6a6a6a6a;
187 w3[2] = w3[2] ^ 0x6a6a6a6a;
188 w3[3] = w3[3] ^ 0x6a6a6a6a;
196 sha1_transform (w0, w1, w2, w3, opad);
199 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])
207 sha1_transform (w0, w1, w2, w3, digest);
224 w3[3] = (64 + 20) * 8;
232 sha1_transform (w0, w1, w2, w3, digest);
235 __kernel void m13600_init (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global pbkdf2_sha1_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 zip2_t *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)
241 const u32 gid = get_global_id (0);
243 if (gid >= gid_max) return;
247 w0[0] = swap32 (pws[gid].i[ 0]);
248 w0[1] = swap32 (pws[gid].i[ 1]);
249 w0[2] = swap32 (pws[gid].i[ 2]);
250 w0[3] = swap32 (pws[gid].i[ 3]);
254 w1[0] = swap32 (pws[gid].i[ 4]);
255 w1[1] = swap32 (pws[gid].i[ 5]);
256 w1[2] = swap32 (pws[gid].i[ 6]);
257 w1[3] = swap32 (pws[gid].i[ 7]);
261 w2[0] = swap32 (pws[gid].i[ 8]);
262 w2[1] = swap32 (pws[gid].i[ 9]);
263 w2[2] = swap32 (pws[gid].i[10]);
264 w2[3] = swap32 (pws[gid].i[11]);
268 w3[0] = swap32 (pws[gid].i[12]);
269 w3[1] = swap32 (pws[gid].i[13]);
270 w3[2] = swap32 (pws[gid].i[14]);
271 w3[3] = swap32 (pws[gid].i[15]);
277 const u32 salt_len = esalt_bufs[salt_pos].salt_len;
282 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
284 tmps[gid].ipad[0] = ipad[0];
285 tmps[gid].ipad[1] = ipad[1];
286 tmps[gid].ipad[2] = ipad[2];
287 tmps[gid].ipad[3] = ipad[3];
288 tmps[gid].ipad[4] = ipad[4];
290 tmps[gid].opad[0] = opad[0];
291 tmps[gid].opad[1] = opad[1];
292 tmps[gid].opad[2] = opad[2];
293 tmps[gid].opad[3] = opad[3];
294 tmps[gid].opad[4] = opad[4];
296 const u32 mode = esalt_bufs[salt_pos].mode;
304 case 1: iter_start = 0;
308 case 2: iter_start = 1;
312 case 3: iter_start = 1;
318 for (u32 i = iter_start, j = count_start; i < iter_stop; i++, j++)
320 const u32 i5 = i * 5;
324 esalt_buf[ 0] = swap32 (esalt_bufs[salt_pos].salt_buf[0]);
325 esalt_buf[ 1] = swap32 (esalt_bufs[salt_pos].salt_buf[1]);
326 esalt_buf[ 2] = swap32 (esalt_bufs[salt_pos].salt_buf[2]);
327 esalt_buf[ 3] = swap32 (esalt_bufs[salt_pos].salt_buf[3]);
339 esalt_buf[15] = (64 + salt_len + 4) * 8;
343 case 1: esalt_buf[2] = j;
344 esalt_buf[3] = 0x80000000;
346 case 2: esalt_buf[3] = j;
347 esalt_buf[4] = 0x80000000;
349 case 3: esalt_buf[4] = j;
350 esalt_buf[5] = 0x80000000;
356 hmac_sha1_run (esalt_buf + 0, esalt_buf + 4, esalt_buf + 8, esalt_buf + 12, ipad, opad, dgst);
358 tmps[gid].dgst[i5 + 0] = dgst[0];
359 tmps[gid].dgst[i5 + 1] = dgst[1];
360 tmps[gid].dgst[i5 + 2] = dgst[2];
361 tmps[gid].dgst[i5 + 3] = dgst[3];
362 tmps[gid].dgst[i5 + 4] = dgst[4];
364 tmps[gid].out[i5 + 0] = dgst[0];
365 tmps[gid].out[i5 + 1] = dgst[1];
366 tmps[gid].out[i5 + 2] = dgst[2];
367 tmps[gid].out[i5 + 3] = dgst[3];
368 tmps[gid].out[i5 + 4] = dgst[4];
372 __kernel void m13600_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global pbkdf2_sha1_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 zip2_t *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)
374 const u32 gid = get_global_id (0);
376 if (gid >= gid_max) return;
380 ipad[0] = tmps[gid].ipad[0];
381 ipad[1] = tmps[gid].ipad[1];
382 ipad[2] = tmps[gid].ipad[2];
383 ipad[3] = tmps[gid].ipad[3];
384 ipad[4] = tmps[gid].ipad[4];
388 opad[0] = tmps[gid].opad[0];
389 opad[1] = tmps[gid].opad[1];
390 opad[2] = tmps[gid].opad[2];
391 opad[3] = tmps[gid].opad[3];
392 opad[4] = tmps[gid].opad[4];
394 const u32 mode = esalt_bufs[salt_pos].mode;
402 case 1: iter_start = 0;
406 case 2: iter_start = 1;
410 case 3: iter_start = 1;
416 for (u32 i = iter_start, j = count_start; i < iter_stop; i++, j++)
418 const u32 i5 = i * 5;
422 dgst[0] = tmps[gid].dgst[i5 + 0];
423 dgst[1] = tmps[gid].dgst[i5 + 1];
424 dgst[2] = tmps[gid].dgst[i5 + 2];
425 dgst[3] = tmps[gid].dgst[i5 + 3];
426 dgst[4] = tmps[gid].dgst[i5 + 4];
430 out[0] = tmps[gid].out[i5 + 0];
431 out[1] = tmps[gid].out[i5 + 1];
432 out[2] = tmps[gid].out[i5 + 2];
433 out[3] = tmps[gid].out[i5 + 3];
434 out[4] = tmps[gid].out[i5 + 4];
436 for (u32 k = 0; k < loop_cnt; k++)
458 w3[3] = (64 + 20) * 8;
460 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
469 tmps[gid].dgst[i5 + 0] = dgst[0];
470 tmps[gid].dgst[i5 + 1] = dgst[1];
471 tmps[gid].dgst[i5 + 2] = dgst[2];
472 tmps[gid].dgst[i5 + 3] = dgst[3];
473 tmps[gid].dgst[i5 + 4] = dgst[4];
475 tmps[gid].out[i5 + 0] = out[0];
476 tmps[gid].out[i5 + 1] = out[1];
477 tmps[gid].out[i5 + 2] = out[2];
478 tmps[gid].out[i5 + 3] = out[3];
479 tmps[gid].out[i5 + 4] = out[4];
483 __kernel void m13600_comp (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global pbkdf2_sha1_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 zip2_t *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)
489 const u32 gid = get_global_id (0);
491 if (gid >= gid_max) return;
493 const u32 lid = get_local_id (0);
497 const u32 mode = esalt_bufs[salt_pos].mode;
504 case 1: iter_start = 4;
507 case 2: iter_start = 6;
510 case 3: iter_start = 8;
515 for (int i = iter_start, j = 0; i < iter_stop; i++, j++)
517 key[j] = tmps[gid].out[i];
545 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
547 int data_len = esalt_bufs[salt_pos].data_len;
552 for (data_left = data_len, data_off = 0; data_left >= 56; data_left -= 64, data_off += 16)
554 w0[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 0]);
555 w0[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 1]);
556 w0[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 2]);
557 w0[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 3]);
558 w1[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 4]);
559 w1[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 5]);
560 w1[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 6]);
561 w1[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 7]);
562 w2[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 8]);
563 w2[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 9]);
564 w2[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 10]);
565 w2[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 11]);
566 w3[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 12]);
567 w3[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 13]);
568 w3[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 14]);
569 w3[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 15]);
571 sha1_transform (w0, w1, w2, w3, ipad);
574 w0[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 0]);
575 w0[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 1]);
576 w0[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 2]);
577 w0[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 3]);
578 w1[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 4]);
579 w1[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 5]);
580 w1[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 6]);
581 w1[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 7]);
582 w2[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 8]);
583 w2[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 9]);
584 w2[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 10]);
585 w2[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 11]);
586 w3[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 12]);
587 w3[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 13]);
589 w3[3] = (64 + data_len) * 8;
593 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
595 const u32 r0 = swap32 (digest[0] & 0xffffffff);
596 const u32 r1 = swap32 (digest[1] & 0xffffffff);
597 const u32 r2 = swap32 (digest[2] & 0xffff0000);
598 const u32 r3 = swap32 (digest[3] & 0x00000000);