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"
18 #include "OpenCL/types_ocl.c"
19 #include "OpenCL/common.c"
21 #define COMPARE_S "OpenCL/check_single_comp4.c"
22 #define COMPARE_M "OpenCL/check_multi_comp4.c"
24 void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
52 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
53 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
54 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
55 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
56 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
57 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
58 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
59 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
60 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
61 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
62 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
63 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
64 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
65 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
66 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
67 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
68 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
69 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
70 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
71 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
76 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
77 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
78 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
79 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
80 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
81 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
82 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
83 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
84 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
85 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
86 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
87 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
88 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
89 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
90 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
91 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
92 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
93 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
94 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
95 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
100 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
101 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
102 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
103 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
104 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
105 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
106 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
107 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
108 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
109 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
110 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
111 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
112 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
113 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
114 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
115 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
116 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
117 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
118 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
119 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
124 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
125 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
126 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
127 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
128 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
129 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
130 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
131 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
132 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
133 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
134 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
135 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
136 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
137 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
138 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
139 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
140 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
141 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
142 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
143 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
152 void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
154 w0[0] = w0[0] ^ 0x36363636;
155 w0[1] = w0[1] ^ 0x36363636;
156 w0[2] = w0[2] ^ 0x36363636;
157 w0[3] = w0[3] ^ 0x36363636;
158 w1[0] = w1[0] ^ 0x36363636;
159 w1[1] = w1[1] ^ 0x36363636;
160 w1[2] = w1[2] ^ 0x36363636;
161 w1[3] = w1[3] ^ 0x36363636;
162 w2[0] = w2[0] ^ 0x36363636;
163 w2[1] = w2[1] ^ 0x36363636;
164 w2[2] = w2[2] ^ 0x36363636;
165 w2[3] = w2[3] ^ 0x36363636;
166 w3[0] = w3[0] ^ 0x36363636;
167 w3[1] = w3[1] ^ 0x36363636;
168 w3[2] = w3[2] ^ 0x36363636;
169 w3[3] = w3[3] ^ 0x36363636;
177 sha1_transform (w0, w1, w2, w3, ipad);
179 w0[0] = w0[0] ^ 0x6a6a6a6a;
180 w0[1] = w0[1] ^ 0x6a6a6a6a;
181 w0[2] = w0[2] ^ 0x6a6a6a6a;
182 w0[3] = w0[3] ^ 0x6a6a6a6a;
183 w1[0] = w1[0] ^ 0x6a6a6a6a;
184 w1[1] = w1[1] ^ 0x6a6a6a6a;
185 w1[2] = w1[2] ^ 0x6a6a6a6a;
186 w1[3] = w1[3] ^ 0x6a6a6a6a;
187 w2[0] = w2[0] ^ 0x6a6a6a6a;
188 w2[1] = w2[1] ^ 0x6a6a6a6a;
189 w2[2] = w2[2] ^ 0x6a6a6a6a;
190 w2[3] = w2[3] ^ 0x6a6a6a6a;
191 w3[0] = w3[0] ^ 0x6a6a6a6a;
192 w3[1] = w3[1] ^ 0x6a6a6a6a;
193 w3[2] = w3[2] ^ 0x6a6a6a6a;
194 w3[3] = w3[3] ^ 0x6a6a6a6a;
202 sha1_transform (w0, w1, w2, w3, opad);
205 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])
213 sha1_transform (w0, w1, w2, w3, digest);
230 w3[3] = (64 + 20) * 8;
238 sha1_transform (w0, w1, w2, w3, digest);
241 __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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
247 const u32 gid = get_global_id (0);
249 if (gid >= gid_max) return;
253 w0[0] = swap32 (pws[gid].i[ 0]);
254 w0[1] = swap32 (pws[gid].i[ 1]);
255 w0[2] = swap32 (pws[gid].i[ 2]);
256 w0[3] = swap32 (pws[gid].i[ 3]);
260 w1[0] = swap32 (pws[gid].i[ 4]);
261 w1[1] = swap32 (pws[gid].i[ 5]);
262 w1[2] = swap32 (pws[gid].i[ 6]);
263 w1[3] = swap32 (pws[gid].i[ 7]);
267 w2[0] = swap32 (pws[gid].i[ 8]);
268 w2[1] = swap32 (pws[gid].i[ 9]);
269 w2[2] = swap32 (pws[gid].i[10]);
270 w2[3] = swap32 (pws[gid].i[11]);
274 w3[0] = swap32 (pws[gid].i[12]);
275 w3[1] = swap32 (pws[gid].i[13]);
276 w3[2] = swap32 (pws[gid].i[14]);
277 w3[3] = swap32 (pws[gid].i[15]);
283 const u32 salt_len = esalt_bufs[salt_pos].salt_len;
288 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
290 tmps[gid].ipad[0] = ipad[0];
291 tmps[gid].ipad[1] = ipad[1];
292 tmps[gid].ipad[2] = ipad[2];
293 tmps[gid].ipad[3] = ipad[3];
294 tmps[gid].ipad[4] = ipad[4];
296 tmps[gid].opad[0] = opad[0];
297 tmps[gid].opad[1] = opad[1];
298 tmps[gid].opad[2] = opad[2];
299 tmps[gid].opad[3] = opad[3];
300 tmps[gid].opad[4] = opad[4];
302 const u32 mode = esalt_bufs[salt_pos].mode;
310 case 1: iter_start = 0;
314 case 2: iter_start = 1;
318 case 3: iter_start = 1;
324 for (u32 i = iter_start, j = count_start; i < iter_stop; i++, j++)
326 const u32 i5 = i * 5;
330 esalt_buf[ 0] = swap32 (esalt_bufs[salt_pos].salt_buf[0]);
331 esalt_buf[ 1] = swap32 (esalt_bufs[salt_pos].salt_buf[1]);
332 esalt_buf[ 2] = swap32 (esalt_bufs[salt_pos].salt_buf[2]);
333 esalt_buf[ 3] = swap32 (esalt_bufs[salt_pos].salt_buf[3]);
345 esalt_buf[15] = (64 + salt_len + 4) * 8;
349 case 1: esalt_buf[2] = j;
350 esalt_buf[3] = 0x80000000;
352 case 2: esalt_buf[3] = j;
353 esalt_buf[4] = 0x80000000;
355 case 3: esalt_buf[4] = j;
356 esalt_buf[5] = 0x80000000;
362 hmac_sha1_run (esalt_buf + 0, esalt_buf + 4, esalt_buf + 8, esalt_buf + 12, ipad, opad, dgst);
364 tmps[gid].dgst[i5 + 0] = dgst[0];
365 tmps[gid].dgst[i5 + 1] = dgst[1];
366 tmps[gid].dgst[i5 + 2] = dgst[2];
367 tmps[gid].dgst[i5 + 3] = dgst[3];
368 tmps[gid].dgst[i5 + 4] = dgst[4];
370 tmps[gid].out[i5 + 0] = dgst[0];
371 tmps[gid].out[i5 + 1] = dgst[1];
372 tmps[gid].out[i5 + 2] = dgst[2];
373 tmps[gid].out[i5 + 3] = dgst[3];
374 tmps[gid].out[i5 + 4] = dgst[4];
378 __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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
380 const u32 gid = get_global_id (0);
382 if (gid >= gid_max) return;
386 ipad[0] = tmps[gid].ipad[0];
387 ipad[1] = tmps[gid].ipad[1];
388 ipad[2] = tmps[gid].ipad[2];
389 ipad[3] = tmps[gid].ipad[3];
390 ipad[4] = tmps[gid].ipad[4];
394 opad[0] = tmps[gid].opad[0];
395 opad[1] = tmps[gid].opad[1];
396 opad[2] = tmps[gid].opad[2];
397 opad[3] = tmps[gid].opad[3];
398 opad[4] = tmps[gid].opad[4];
400 const u32 mode = esalt_bufs[salt_pos].mode;
408 case 1: iter_start = 0;
412 case 2: iter_start = 1;
416 case 3: iter_start = 1;
422 for (u32 i = iter_start, j = count_start; i < iter_stop; i++, j++)
424 const u32 i5 = i * 5;
428 dgst[0] = tmps[gid].dgst[i5 + 0];
429 dgst[1] = tmps[gid].dgst[i5 + 1];
430 dgst[2] = tmps[gid].dgst[i5 + 2];
431 dgst[3] = tmps[gid].dgst[i5 + 3];
432 dgst[4] = tmps[gid].dgst[i5 + 4];
436 out[0] = tmps[gid].out[i5 + 0];
437 out[1] = tmps[gid].out[i5 + 1];
438 out[2] = tmps[gid].out[i5 + 2];
439 out[3] = tmps[gid].out[i5 + 3];
440 out[4] = tmps[gid].out[i5 + 4];
442 for (u32 k = 0; k < loop_cnt; k++)
464 w3[3] = (64 + 20) * 8;
466 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
475 tmps[gid].dgst[i5 + 0] = dgst[0];
476 tmps[gid].dgst[i5 + 1] = dgst[1];
477 tmps[gid].dgst[i5 + 2] = dgst[2];
478 tmps[gid].dgst[i5 + 3] = dgst[3];
479 tmps[gid].dgst[i5 + 4] = dgst[4];
481 tmps[gid].out[i5 + 0] = out[0];
482 tmps[gid].out[i5 + 1] = out[1];
483 tmps[gid].out[i5 + 2] = out[2];
484 tmps[gid].out[i5 + 3] = out[3];
485 tmps[gid].out[i5 + 4] = out[4];
489 __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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
495 const u32 gid = get_global_id (0);
497 if (gid >= gid_max) return;
499 const u32 lid = get_local_id (0);
503 const u32 mode = esalt_bufs[salt_pos].mode;
510 case 1: iter_start = 4;
513 case 2: iter_start = 6;
516 case 3: iter_start = 8;
521 for (int i = iter_start, j = 0; i < iter_stop; i++, j++)
523 key[j] = tmps[gid].out[i];
551 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
553 int data_len = esalt_bufs[salt_pos].data_len;
558 for (data_left = data_len, data_off = 0; data_left >= 56; data_left -= 64, data_off += 16)
560 w0[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 0]);
561 w0[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 1]);
562 w0[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 2]);
563 w0[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 3]);
564 w1[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 4]);
565 w1[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 5]);
566 w1[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 6]);
567 w1[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 7]);
568 w2[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 8]);
569 w2[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 9]);
570 w2[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 10]);
571 w2[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 11]);
572 w3[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 12]);
573 w3[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 13]);
574 w3[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 14]);
575 w3[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 15]);
577 sha1_transform (w0, w1, w2, w3, ipad);
580 w0[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 0]);
581 w0[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 1]);
582 w0[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 2]);
583 w0[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 3]);
584 w1[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 4]);
585 w1[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 5]);
586 w1[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 6]);
587 w1[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 7]);
588 w2[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 8]);
589 w2[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 9]);
590 w2[2] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 10]);
591 w2[3] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 11]);
592 w3[0] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 12]);
593 w3[1] = swap32 (esalt_bufs[salt_pos].data_buf[data_off + 13]);
595 w3[3] = (64 + data_len) * 8;
599 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
601 const u32 r0 = swap32 (digest[0] & 0xffffffff);
602 const u32 r1 = swap32 (digest[1] & 0xffffffff);
603 const u32 r2 = swap32 (digest[2] & 0xffff0000);
604 const u32 r3 = swap32 (digest[3] & 0x00000000);