2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
28 #include "include/kernel_functions.c"
29 #include "types_amd.c"
30 #include "common_amd.c"
33 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
34 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
38 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
39 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
43 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
44 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
47 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
75 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
76 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
77 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
78 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
79 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
80 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
81 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
82 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
83 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
84 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
85 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
86 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
87 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
88 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
89 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
90 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
91 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
92 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
93 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
94 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
99 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
100 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
101 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
102 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
103 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
104 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
105 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
106 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
107 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
108 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
109 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
110 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
111 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
112 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
113 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
114 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
115 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
116 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
117 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
118 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
123 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
124 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
125 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
126 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
127 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
128 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
129 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
130 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
131 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
132 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
133 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
134 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
135 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
136 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
137 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
138 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
139 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
140 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
141 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
142 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
147 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
148 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
149 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
150 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
151 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
152 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
153 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
154 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
155 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
156 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
157 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
158 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
159 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
160 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
161 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
162 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
163 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
164 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
165 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
166 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
175 static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
177 w0[0] = w0[0] ^ 0x36363636;
178 w0[1] = w0[1] ^ 0x36363636;
179 w0[2] = w0[2] ^ 0x36363636;
180 w0[3] = w0[3] ^ 0x36363636;
181 w1[0] = w1[0] ^ 0x36363636;
182 w1[1] = w1[1] ^ 0x36363636;
183 w1[2] = w1[2] ^ 0x36363636;
184 w1[3] = w1[3] ^ 0x36363636;
185 w2[0] = w2[0] ^ 0x36363636;
186 w2[1] = w2[1] ^ 0x36363636;
187 w2[2] = w2[2] ^ 0x36363636;
188 w2[3] = w2[3] ^ 0x36363636;
189 w3[0] = w3[0] ^ 0x36363636;
190 w3[1] = w3[1] ^ 0x36363636;
191 w3[2] = w3[2] ^ 0x36363636;
192 w3[3] = w3[3] ^ 0x36363636;
200 sha1_transform (w0, w1, w2, w3, ipad);
202 w0[0] = w0[0] ^ 0x6a6a6a6a;
203 w0[1] = w0[1] ^ 0x6a6a6a6a;
204 w0[2] = w0[2] ^ 0x6a6a6a6a;
205 w0[3] = w0[3] ^ 0x6a6a6a6a;
206 w1[0] = w1[0] ^ 0x6a6a6a6a;
207 w1[1] = w1[1] ^ 0x6a6a6a6a;
208 w1[2] = w1[2] ^ 0x6a6a6a6a;
209 w1[3] = w1[3] ^ 0x6a6a6a6a;
210 w2[0] = w2[0] ^ 0x6a6a6a6a;
211 w2[1] = w2[1] ^ 0x6a6a6a6a;
212 w2[2] = w2[2] ^ 0x6a6a6a6a;
213 w2[3] = w2[3] ^ 0x6a6a6a6a;
214 w3[0] = w3[0] ^ 0x6a6a6a6a;
215 w3[1] = w3[1] ^ 0x6a6a6a6a;
216 w3[2] = w3[2] ^ 0x6a6a6a6a;
217 w3[3] = w3[3] ^ 0x6a6a6a6a;
225 sha1_transform (w0, w1, w2, w3, opad);
228 static void hmac_sha1_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5], u32x digest[5])
236 sha1_transform (w0, w1, w2, w3, digest);
253 w3[3] = (64 + 20) * 8;
261 sha1_transform (w0, w1, w2, w3, digest);
264 __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)
270 const u32 lid = get_local_id (0);
276 const u32 gid = get_global_id (0);
278 if (gid >= gid_max) return;
282 wordl0[0] = pws[gid].i[ 0];
283 wordl0[1] = pws[gid].i[ 1];
284 wordl0[2] = pws[gid].i[ 2];
285 wordl0[3] = pws[gid].i[ 3];
289 wordl1[0] = pws[gid].i[ 4];
290 wordl1[1] = pws[gid].i[ 5];
291 wordl1[2] = pws[gid].i[ 6];
292 wordl1[3] = pws[gid].i[ 7];
308 const u32 pw_l_len = pws[gid].pw_len;
310 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
312 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
319 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
325 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
327 const u32 pw_r_len = combs_buf[il_pos].pw_len;
329 const u32 pw_len = pw_l_len + pw_r_len;
333 wordr0[0] = combs_buf[il_pos].i[0];
334 wordr0[1] = combs_buf[il_pos].i[1];
335 wordr0[2] = combs_buf[il_pos].i[2];
336 wordr0[3] = combs_buf[il_pos].i[3];
340 wordr1[0] = combs_buf[il_pos].i[4];
341 wordr1[1] = combs_buf[il_pos].i[5];
342 wordr1[2] = combs_buf[il_pos].i[6];
343 wordr1[3] = combs_buf[il_pos].i[7];
359 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
361 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
366 w0[0] = wordl0[0] | wordr0[0];
367 w0[1] = wordl0[1] | wordr0[1];
368 w0[2] = wordl0[2] | wordr0[2];
369 w0[3] = wordl0[3] | wordr0[3];
373 w1[0] = wordl1[0] | wordr1[0];
374 w1[1] = wordl1[1] | wordr1[1];
375 w1[2] = wordl1[2] | wordr1[2];
376 w1[3] = wordl1[3] | wordr1[3];
380 w2[0] = wordl2[0] | wordr2[0];
381 w2[1] = wordl2[1] | wordr2[1];
382 w2[2] = wordl2[2] | wordr2[2];
383 w2[3] = wordl2[3] | wordr2[3];
387 w3[0] = wordl3[0] | wordr3[0];
388 w3[1] = wordl3[1] | wordr3[1];
389 w3[2] = wordl3[2] | wordr3[2];
390 w3[3] = wordl3[3] | wordr3[3];
398 w0_t[0] = swap_workaround (w0[0]);
399 w0_t[1] = swap_workaround (w0[1]);
400 w0_t[2] = swap_workaround (w0[2]);
401 w0_t[3] = swap_workaround (w0[3]);
405 w1_t[0] = swap_workaround (w1[0]);
406 w1_t[1] = swap_workaround (w1[1]);
407 w1_t[2] = swap_workaround (w1[2]);
408 w1_t[3] = swap_workaround (w1[3]);
427 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
429 int esalt_size = esalt_len;
434 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
436 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
437 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
438 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
439 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
440 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
441 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
442 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
443 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
444 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
445 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
446 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
447 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
448 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
449 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
450 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
451 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
453 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
456 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
457 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
458 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
459 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
460 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
461 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
462 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
463 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
464 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
465 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
466 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
467 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
468 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
469 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
471 w3_t[3] = (64 + esalt_size) * 8;
475 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
477 const u32x r0 = digest[3];
478 const u32x r1 = digest[4];
479 const u32x r2 = digest[2];
480 const u32x r3 = digest[1];
482 #include VECT_COMPARE_M
486 __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)
490 __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)
494 __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)
500 const u32 lid = get_local_id (0);
506 const u32 gid = get_global_id (0);
508 if (gid >= gid_max) return;
512 wordl0[0] = pws[gid].i[ 0];
513 wordl0[1] = pws[gid].i[ 1];
514 wordl0[2] = pws[gid].i[ 2];
515 wordl0[3] = pws[gid].i[ 3];
519 wordl1[0] = pws[gid].i[ 4];
520 wordl1[1] = pws[gid].i[ 5];
521 wordl1[2] = pws[gid].i[ 6];
522 wordl1[3] = pws[gid].i[ 7];
538 const u32 pw_l_len = pws[gid].pw_len;
540 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
542 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
549 const u32 esalt_len = rakp_bufs[salt_pos].salt_len;
555 const u32 search[4] =
557 digests_buf[digests_offset].digest_buf[DGST_R0],
558 digests_buf[digests_offset].digest_buf[DGST_R1],
559 digests_buf[digests_offset].digest_buf[DGST_R2],
560 digests_buf[digests_offset].digest_buf[DGST_R3]
567 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
569 const u32 pw_r_len = combs_buf[il_pos].pw_len;
571 const u32 pw_len = pw_l_len + pw_r_len;
575 wordr0[0] = combs_buf[il_pos].i[0];
576 wordr0[1] = combs_buf[il_pos].i[1];
577 wordr0[2] = combs_buf[il_pos].i[2];
578 wordr0[3] = combs_buf[il_pos].i[3];
582 wordr1[0] = combs_buf[il_pos].i[4];
583 wordr1[1] = combs_buf[il_pos].i[5];
584 wordr1[2] = combs_buf[il_pos].i[6];
585 wordr1[3] = combs_buf[il_pos].i[7];
601 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
603 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
608 w0[0] = wordl0[0] | wordr0[0];
609 w0[1] = wordl0[1] | wordr0[1];
610 w0[2] = wordl0[2] | wordr0[2];
611 w0[3] = wordl0[3] | wordr0[3];
615 w1[0] = wordl1[0] | wordr1[0];
616 w1[1] = wordl1[1] | wordr1[1];
617 w1[2] = wordl1[2] | wordr1[2];
618 w1[3] = wordl1[3] | wordr1[3];
622 w2[0] = wordl2[0] | wordr2[0];
623 w2[1] = wordl2[1] | wordr2[1];
624 w2[2] = wordl2[2] | wordr2[2];
625 w2[3] = wordl2[3] | wordr2[3];
629 w3[0] = wordl3[0] | wordr3[0];
630 w3[1] = wordl3[1] | wordr3[1];
631 w3[2] = wordl3[2] | wordr3[2];
632 w3[3] = wordl3[3] | wordr3[3];
640 w0_t[0] = swap_workaround (w0[0]);
641 w0_t[1] = swap_workaround (w0[1]);
642 w0_t[2] = swap_workaround (w0[2]);
643 w0_t[3] = swap_workaround (w0[3]);
647 w1_t[0] = swap_workaround (w1[0]);
648 w1_t[1] = swap_workaround (w1[1]);
649 w1_t[2] = swap_workaround (w1[2]);
650 w1_t[3] = swap_workaround (w1[3]);
669 hmac_sha1_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
671 int esalt_size = esalt_len;
676 for (esalt_left = esalt_size, esalt_off = 0; esalt_left >= 56; esalt_left -= 64, esalt_off += 16)
678 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
679 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
680 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
681 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
682 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
683 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
684 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
685 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
686 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
687 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
688 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
689 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
690 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
691 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
692 w3_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 14];
693 w3_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 15];
695 sha1_transform (w0_t, w1_t, w2_t, w3_t, ipad);
698 w0_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 0];
699 w0_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 1];
700 w0_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 2];
701 w0_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 3];
702 w1_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 4];
703 w1_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 5];
704 w1_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 6];
705 w1_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 7];
706 w2_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 8];
707 w2_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 9];
708 w2_t[2] = rakp_bufs[salt_pos].salt_buf[esalt_off + 10];
709 w2_t[3] = rakp_bufs[salt_pos].salt_buf[esalt_off + 11];
710 w3_t[0] = rakp_bufs[salt_pos].salt_buf[esalt_off + 12];
711 w3_t[1] = rakp_bufs[salt_pos].salt_buf[esalt_off + 13];
713 w3_t[3] = (64 + esalt_size) * 8;
717 hmac_sha1_run (w0_t, w1_t, w2_t, w3_t, ipad, opad, digest);
719 const u32x r0 = digest[3];
720 const u32x r1 = digest[4];
721 const u32x r2 = digest[2];
722 const u32x r3 = digest[1];
724 #include VECT_COMPARE_S
728 __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)
732 __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)