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 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
51 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
52 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
53 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
54 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
55 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
56 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
57 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
58 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
59 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
60 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
61 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
62 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
63 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
64 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
65 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
66 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
67 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
68 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
69 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
70 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
75 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
76 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
77 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
78 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
79 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
80 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
81 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
82 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
83 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
84 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
85 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
86 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
87 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
88 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
89 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
90 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
91 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
92 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
93 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
94 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
99 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
100 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
101 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
102 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
103 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
104 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
105 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
106 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
107 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
108 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
109 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
110 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
111 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
112 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
113 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
114 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
115 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
116 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
117 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
118 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
123 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
124 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
125 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
126 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
127 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
128 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
129 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
130 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
131 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
132 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
133 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
134 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
135 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
136 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
137 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
138 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
139 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
140 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
141 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
142 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
151 static void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
153 w0[0] = w0[0] ^ 0x36363636;
154 w0[1] = w0[1] ^ 0x36363636;
155 w0[2] = w0[2] ^ 0x36363636;
156 w0[3] = w0[3] ^ 0x36363636;
157 w1[0] = w1[0] ^ 0x36363636;
158 w1[1] = w1[1] ^ 0x36363636;
159 w1[2] = w1[2] ^ 0x36363636;
160 w1[3] = w1[3] ^ 0x36363636;
161 w2[0] = w2[0] ^ 0x36363636;
162 w2[1] = w2[1] ^ 0x36363636;
163 w2[2] = w2[2] ^ 0x36363636;
164 w2[3] = w2[3] ^ 0x36363636;
165 w3[0] = w3[0] ^ 0x36363636;
166 w3[1] = w3[1] ^ 0x36363636;
167 w3[2] = w3[2] ^ 0x36363636;
168 w3[3] = w3[3] ^ 0x36363636;
176 sha1_transform (w0, w1, w2, w3, ipad);
178 w0[0] = w0[0] ^ 0x6a6a6a6a;
179 w0[1] = w0[1] ^ 0x6a6a6a6a;
180 w0[2] = w0[2] ^ 0x6a6a6a6a;
181 w0[3] = w0[3] ^ 0x6a6a6a6a;
182 w1[0] = w1[0] ^ 0x6a6a6a6a;
183 w1[1] = w1[1] ^ 0x6a6a6a6a;
184 w1[2] = w1[2] ^ 0x6a6a6a6a;
185 w1[3] = w1[3] ^ 0x6a6a6a6a;
186 w2[0] = w2[0] ^ 0x6a6a6a6a;
187 w2[1] = w2[1] ^ 0x6a6a6a6a;
188 w2[2] = w2[2] ^ 0x6a6a6a6a;
189 w2[3] = w2[3] ^ 0x6a6a6a6a;
190 w3[0] = w3[0] ^ 0x6a6a6a6a;
191 w3[1] = w3[1] ^ 0x6a6a6a6a;
192 w3[2] = w3[2] ^ 0x6a6a6a6a;
193 w3[3] = w3[3] ^ 0x6a6a6a6a;
201 sha1_transform (w0, w1, w2, w3, opad);
204 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])
212 sha1_transform (w0, w1, w2, w3, digest);
229 w3[3] = (64 + 20) * 8;
237 sha1_transform (w0, w1, w2, w3, digest);
240 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07300_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 rakp_t *rakp_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)
246 const u32 lid = get_local_id (0);
252 const u32 gid = get_global_id (0);
254 if (gid >= gid_max) return;
258 wordl0[0] = pws[gid].i[ 0];
259 wordl0[1] = pws[gid].i[ 1];
260 wordl0[2] = pws[gid].i[ 2];
261 wordl0[3] = pws[gid].i[ 3];
265 wordl1[0] = pws[gid].i[ 4];
266 wordl1[1] = pws[gid].i[ 5];
267 wordl1[2] = pws[gid].i[ 6];
268 wordl1[3] = pws[gid].i[ 7];
284 const u32 pw_l_len = pws[gid].pw_len;
286 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
288 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
295 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
301 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
303 const u32 pw_r_len = combs_buf[il_pos].pw_len;
305 const u32 pw_len = pw_l_len + pw_r_len;
309 wordr0[0] = combs_buf[il_pos].i[0];
310 wordr0[1] = combs_buf[il_pos].i[1];
311 wordr0[2] = combs_buf[il_pos].i[2];
312 wordr0[3] = combs_buf[il_pos].i[3];
316 wordr1[0] = combs_buf[il_pos].i[4];
317 wordr1[1] = combs_buf[il_pos].i[5];
318 wordr1[2] = combs_buf[il_pos].i[6];
319 wordr1[3] = combs_buf[il_pos].i[7];
335 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
337 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_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];
365 w3[2] = wordl3[2] | wordr3[2];
366 w3[3] = wordl3[3] | wordr3[3];
374 w0_t[0] = swap32 (w0[0]);
375 w0_t[1] = swap32 (w0[1]);
376 w0_t[2] = swap32 (w0[2]);
377 w0_t[3] = swap32 (w0[3]);
381 w1_t[0] = swap32 (w1[0]);
382 w1_t[1] = swap32 (w1[1]);
383 w1_t[2] = swap32 (w1[2]);
384 w1_t[3] = swap32 (w1[3]);
403 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
405 int esalt_size = esalt_len;
410 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
412 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
413 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
414 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
415 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
416 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
417 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
418 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
419 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
420 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
421 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
422 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
423 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
424 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
425 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
426 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
427 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
429 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
432 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
433 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
434 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
435 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
436 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
437 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
438 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
439 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
440 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
441 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
442 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
443 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
444 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
445 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
447 w3_t[3] = (64 + esalt_size) * 8;
451 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
453 const u32 r0 = digest[3];
454 const u32 r1 = digest[4];
455 const u32 r2 = digest[2];
456 const u32 r3 = digest[1];
462 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07300_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 rakp_t *rakp_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)
466 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07300_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 rakp_t *rakp_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)
470 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07300_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 rakp_t *rakp_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)
476 const u32 lid = get_local_id (0);
482 const u32 gid = get_global_id (0);
484 if (gid >= gid_max) return;
488 wordl0[0] = pws[gid].i[ 0];
489 wordl0[1] = pws[gid].i[ 1];
490 wordl0[2] = pws[gid].i[ 2];
491 wordl0[3] = pws[gid].i[ 3];
495 wordl1[0] = pws[gid].i[ 4];
496 wordl1[1] = pws[gid].i[ 5];
497 wordl1[2] = pws[gid].i[ 6];
498 wordl1[3] = pws[gid].i[ 7];
514 const u32 pw_l_len = pws[gid].pw_len;
516 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
518 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
525 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
531 const u32 search[4] =
533 digests_buf[digests_offset].digest_buf[DGST_R0],
534 digests_buf[digests_offset].digest_buf[DGST_R1],
535 digests_buf[digests_offset].digest_buf[DGST_R2],
536 digests_buf[digests_offset].digest_buf[DGST_R3]
543 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
545 const u32 pw_r_len = combs_buf[il_pos].pw_len;
547 const u32 pw_len = pw_l_len + pw_r_len;
551 wordr0[0] = combs_buf[il_pos].i[0];
552 wordr0[1] = combs_buf[il_pos].i[1];
553 wordr0[2] = combs_buf[il_pos].i[2];
554 wordr0[3] = combs_buf[il_pos].i[3];
558 wordr1[0] = combs_buf[il_pos].i[4];
559 wordr1[1] = combs_buf[il_pos].i[5];
560 wordr1[2] = combs_buf[il_pos].i[6];
561 wordr1[3] = combs_buf[il_pos].i[7];
577 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
579 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
584 w0[0] = wordl0[0] | wordr0[0];
585 w0[1] = wordl0[1] | wordr0[1];
586 w0[2] = wordl0[2] | wordr0[2];
587 w0[3] = wordl0[3] | wordr0[3];
591 w1[0] = wordl1[0] | wordr1[0];
592 w1[1] = wordl1[1] | wordr1[1];
593 w1[2] = wordl1[2] | wordr1[2];
594 w1[3] = wordl1[3] | wordr1[3];
598 w2[0] = wordl2[0] | wordr2[0];
599 w2[1] = wordl2[1] | wordr2[1];
600 w2[2] = wordl2[2] | wordr2[2];
601 w2[3] = wordl2[3] | wordr2[3];
605 w3[0] = wordl3[0] | wordr3[0];
606 w3[1] = wordl3[1] | wordr3[1];
607 w3[2] = wordl3[2] | wordr3[2];
608 w3[3] = wordl3[3] | wordr3[3];
616 w0_t[0] = swap32 (w0[0]);
617 w0_t[1] = swap32 (w0[1]);
618 w0_t[2] = swap32 (w0[2]);
619 w0_t[3] = swap32 (w0[3]);
623 w1_t[0] = swap32 (w1[0]);
624 w1_t[1] = swap32 (w1[1]);
625 w1_t[2] = swap32 (w1[2]);
626 w1_t[3] = swap32 (w1[3]);
645 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
647 int esalt_size = esalt_len;
652 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
654 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
655 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
656 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
657 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
658 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
659 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
660 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
661 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
662 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
663 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
664 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
665 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
666 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
667 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
668 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
669 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
671 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
674 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
675 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
676 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
677 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
678 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
679 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
680 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
681 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
682 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
683 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
684 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
685 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
686 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
687 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
689 w3_t[3] = (64 + esalt_size) * 8;
693 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
695 const u32 r0 = digest[3];
696 const u32 r1 = digest[4];
697 const u32 r2 = digest[2];
698 const u32 r3 = digest[1];
704 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07300_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 rakp_t *rakp_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)
708 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07300_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 rakp_t *rakp_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)