2 * Author......: Jens Steube <jens.steube@gmail.com>
10 #include "include/constants.h"
11 #include "include/kernel_vendor.h"
18 #include "include/kernel_functions.c"
19 #include "OpenCL/types_ocl.c"
20 #include "OpenCL/common.c"
21 #include "OpenCL/simd.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 __kernel void m08300_m04 (__global pw_t *pws, __global kernel_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 void *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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
157 const u32 lid = get_local_id (0);
163 const u32 gid = get_global_id (0);
165 if (gid >= gid_max) return;
170 pws0[0] = pws[gid].i[0];
171 pws0[1] = pws[gid].i[1];
172 pws0[2] = pws[gid].i[2];
173 pws0[3] = pws[gid].i[3];
174 pws1[0] = pws[gid].i[4];
175 pws1[1] = pws[gid].i[5];
176 pws1[2] = pws[gid].i[6];
177 pws1[3] = pws[gid].i[7];
179 const u32 pw_l_len = pws[gid].pw_len;
185 const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
189 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
190 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
191 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
192 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
196 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
197 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
198 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
199 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
201 const u32 salt_len = salt_bufs[salt_pos].salt_len;
205 domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
206 domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
207 domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
208 domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
212 domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
213 domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
214 domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
217 const u32 domain_len = salt_bufs[salt_pos].salt_buf_pc[ 7];
223 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
225 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
227 const u32x pw_len = pw_l_len + pw_r_len;
229 u32x wordr0[4] = { 0 };
230 u32x wordr1[4] = { 0 };
231 u32x wordr2[4] = { 0 };
232 u32x wordr3[4] = { 0 };
234 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
235 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
236 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
237 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
238 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
239 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
240 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
241 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
243 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
245 switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
253 w0[0] = wordl0[0] | wordr0[0];
254 w0[1] = wordl0[1] | wordr0[1];
255 w0[2] = wordl0[2] | wordr0[2];
256 w0[3] = wordl0[3] | wordr0[3];
257 w1[0] = wordl1[0] | wordr1[0];
258 w1[1] = wordl1[1] | wordr1[1];
259 w1[2] = wordl1[2] | wordr1[2];
260 w1[3] = wordl1[3] | wordr1[3];
261 w2[0] = wordl2[0] | wordr2[0];
262 w2[1] = wordl2[1] | wordr2[1];
263 w2[2] = wordl2[2] | wordr2[2];
264 w2[3] = wordl2[3] | wordr2[3];
265 w3[0] = wordl3[0] | wordr3[0];
266 w3[1] = wordl3[1] | wordr3[1];
267 w3[2] = wordl3[2] | wordr3[2];
268 w3[3] = wordl3[3] | wordr3[3];
298 switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, 1);
300 w0_t[0] |= pw_len & 0xff;
308 s0[0] = salt_buf0[0];
309 s0[1] = salt_buf0[1];
310 s0[2] = salt_buf0[2];
311 s0[3] = salt_buf0[3];
315 s1[0] = salt_buf1[0];
316 s1[1] = salt_buf1[1];
317 s1[2] = salt_buf1[2];
318 s1[3] = salt_buf1[3];
334 switch_buffer_by_offset_le (s0, s1, s2, s3, 1 + pw_len + domain_len + 1);
338 d0[0] = domain_buf0[0];
339 d0[1] = domain_buf0[1];
340 d0[2] = domain_buf0[2];
341 d0[3] = domain_buf0[3];
345 d1[0] = domain_buf1[0];
346 d1[1] = domain_buf1[1];
347 d1[2] = domain_buf1[2];
364 switch_buffer_by_offset_le (d0, d1, d2, d3, 1 + pw_len);
372 w0_t2[0] = swap32 (w0_t[0] | d0[0] | s0[0]);
373 w0_t2[1] = swap32 (w0_t[1] | d0[1] | s0[1]);
374 w0_t2[2] = swap32 (w0_t[2] | d0[2] | s0[2]);
375 w0_t2[3] = swap32 (w0_t[3] | d0[3] | s0[3]);
379 w1_t2[0] = swap32 (w1_t[0] | d1[0] | s1[0]);
380 w1_t2[1] = swap32 (w1_t[1] | d1[1] | s1[1]);
381 w1_t2[2] = swap32 (w1_t[2] | d1[2] | s1[2]);
382 w1_t2[3] = swap32 (w1_t[3] | d1[3] | s1[3]);
386 w2_t2[0] = swap32 (w2_t[0] | d2[0] | s2[0]);
387 w2_t2[1] = swap32 (w2_t[1] | d2[1] | s2[1]);
388 w2_t2[2] = swap32 (w2_t[2] | d2[2] | s2[2]);
389 w2_t2[3] = swap32 (w2_t[3] | d2[3] | s2[3]);
393 w3_t2[0] = swap32 (w3_t[0] | d3[0] | s3[0]);
394 w3_t2[1] = swap32 (w3_t[1] | d3[1] | s3[1]);
396 w3_t2[3] = (1 + pw_len + domain_len + 1 + salt_len) * 8;
406 sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
410 for (u32 i = 0; i < salt_iter; i++)
414 w0_t3[0] = digest[0];
415 w0_t3[1] = digest[1];
416 w0_t3[2] = digest[2];
417 w0_t3[3] = digest[3];
421 w1_t3[0] = digest[4];
422 w1_t3[1] = swap32 (salt_buf0[0]);
423 w1_t3[2] = swap32 (salt_buf0[1]);
424 w1_t3[3] = swap32 (salt_buf0[2]);
428 w2_t3[0] = swap32 (salt_buf0[3]);
429 w2_t3[1] = swap32 (salt_buf1[0]);
430 w2_t3[2] = swap32 (salt_buf1[1]);
431 w2_t3[3] = swap32 (salt_buf1[2]);
435 w3_t3[0] = swap32 (salt_buf1[3]);
438 w3_t3[3] = (20 + salt_len) * 8;
446 sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
449 COMPARE_M_SIMD (digest[3], digest[4], digest[2], digest[1]);
453 __kernel void m08300_m08 (__global pw_t *pws, __global kernel_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 void *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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
457 __kernel void m08300_m16 (__global pw_t *pws, __global kernel_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 void *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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
461 __kernel void m08300_s04 (__global pw_t *pws, __global kernel_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 void *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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
467 const u32 lid = get_local_id (0);
473 const u32 gid = get_global_id (0);
475 if (gid >= gid_max) return;
480 pws0[0] = pws[gid].i[0];
481 pws0[1] = pws[gid].i[1];
482 pws0[2] = pws[gid].i[2];
483 pws0[3] = pws[gid].i[3];
484 pws1[0] = pws[gid].i[4];
485 pws1[1] = pws[gid].i[5];
486 pws1[2] = pws[gid].i[6];
487 pws1[3] = pws[gid].i[7];
489 const u32 pw_l_len = pws[gid].pw_len;
495 const u32 salt_iter = salt_bufs[salt_pos].salt_iter;
499 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
500 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
501 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
502 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
506 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
507 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
508 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
509 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
511 const u32 salt_len = salt_bufs[salt_pos].salt_len;
515 domain_buf0[0] = salt_bufs[salt_pos].salt_buf_pc[ 0];
516 domain_buf0[1] = salt_bufs[salt_pos].salt_buf_pc[ 1];
517 domain_buf0[2] = salt_bufs[salt_pos].salt_buf_pc[ 2];
518 domain_buf0[3] = salt_bufs[salt_pos].salt_buf_pc[ 3];
522 domain_buf1[0] = salt_bufs[salt_pos].salt_buf_pc[ 4];
523 domain_buf1[1] = salt_bufs[salt_pos].salt_buf_pc[ 5];
524 domain_buf1[2] = salt_bufs[salt_pos].salt_buf_pc[ 6];
527 const u32 domain_len = salt_bufs[salt_pos].salt_buf_pc[ 7];
533 const u32 search[4] =
535 digests_buf[digests_offset].digest_buf[DGST_R0],
536 digests_buf[digests_offset].digest_buf[DGST_R1],
537 digests_buf[digests_offset].digest_buf[DGST_R2],
538 digests_buf[digests_offset].digest_buf[DGST_R3]
545 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos += VECT_SIZE)
547 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
549 const u32x pw_len = pw_l_len + pw_r_len;
551 u32x wordr0[4] = { 0 };
552 u32x wordr1[4] = { 0 };
553 u32x wordr2[4] = { 0 };
554 u32x wordr3[4] = { 0 };
556 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
557 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
558 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
559 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
560 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
561 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
562 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
563 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
565 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
567 switch_buffer_by_offset_le (wordr0, wordr1, wordr2, wordr3, pw_l_len);
575 w0[0] = wordl0[0] | wordr0[0];
576 w0[1] = wordl0[1] | wordr0[1];
577 w0[2] = wordl0[2] | wordr0[2];
578 w0[3] = wordl0[3] | wordr0[3];
579 w1[0] = wordl1[0] | wordr1[0];
580 w1[1] = wordl1[1] | wordr1[1];
581 w1[2] = wordl1[2] | wordr1[2];
582 w1[3] = wordl1[3] | wordr1[3];
583 w2[0] = wordl2[0] | wordr2[0];
584 w2[1] = wordl2[1] | wordr2[1];
585 w2[2] = wordl2[2] | wordr2[2];
586 w2[3] = wordl2[3] | wordr2[3];
587 w3[0] = wordl3[0] | wordr3[0];
588 w3[1] = wordl3[1] | wordr3[1];
589 w3[2] = wordl3[2] | wordr3[2];
590 w3[3] = wordl3[3] | wordr3[3];
620 switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, 1);
622 w0_t[0] |= pw_len & 0xff;
630 s0[0] = salt_buf0[0];
631 s0[1] = salt_buf0[1];
632 s0[2] = salt_buf0[2];
633 s0[3] = salt_buf0[3];
637 s1[0] = salt_buf1[0];
638 s1[1] = salt_buf1[1];
639 s1[2] = salt_buf1[2];
640 s1[3] = salt_buf1[3];
656 switch_buffer_by_offset_le (s0, s1, s2, s3, 1 + pw_len + domain_len + 1);
660 d0[0] = domain_buf0[0];
661 d0[1] = domain_buf0[1];
662 d0[2] = domain_buf0[2];
663 d0[3] = domain_buf0[3];
667 d1[0] = domain_buf1[0];
668 d1[1] = domain_buf1[1];
669 d1[2] = domain_buf1[2];
686 switch_buffer_by_offset_le (d0, d1, d2, d3, 1 + pw_len);
694 w0_t2[0] = swap32 (w0_t[0] | d0[0] | s0[0]);
695 w0_t2[1] = swap32 (w0_t[1] | d0[1] | s0[1]);
696 w0_t2[2] = swap32 (w0_t[2] | d0[2] | s0[2]);
697 w0_t2[3] = swap32 (w0_t[3] | d0[3] | s0[3]);
701 w1_t2[0] = swap32 (w1_t[0] | d1[0] | s1[0]);
702 w1_t2[1] = swap32 (w1_t[1] | d1[1] | s1[1]);
703 w1_t2[2] = swap32 (w1_t[2] | d1[2] | s1[2]);
704 w1_t2[3] = swap32 (w1_t[3] | d1[3] | s1[3]);
708 w2_t2[0] = swap32 (w2_t[0] | d2[0] | s2[0]);
709 w2_t2[1] = swap32 (w2_t[1] | d2[1] | s2[1]);
710 w2_t2[2] = swap32 (w2_t[2] | d2[2] | s2[2]);
711 w2_t2[3] = swap32 (w2_t[3] | d2[3] | s2[3]);
715 w3_t2[0] = swap32 (w3_t[0] | d3[0] | s3[0]);
716 w3_t2[1] = swap32 (w3_t[1] | d3[1] | s3[1]);
718 w3_t2[3] = (1 + pw_len + domain_len + 1 + salt_len) * 8;
728 sha1_transform (w0_t2, w1_t2, w2_t2, w3_t2, digest);
732 for (u32 i = 0; i < salt_iter; i++)
736 w0_t3[0] = digest[0];
737 w0_t3[1] = digest[1];
738 w0_t3[2] = digest[2];
739 w0_t3[3] = digest[3];
743 w1_t3[0] = digest[4];
744 w1_t3[1] = swap32 (salt_buf0[0]);
745 w1_t3[2] = swap32 (salt_buf0[1]);
746 w1_t3[3] = swap32 (salt_buf0[2]);
750 w2_t3[0] = swap32 (salt_buf0[3]);
751 w2_t3[1] = swap32 (salt_buf1[0]);
752 w2_t3[2] = swap32 (salt_buf1[1]);
753 w2_t3[3] = swap32 (salt_buf1[2]);
757 w3_t3[0] = swap32 (salt_buf1[3]);
760 w3_t3[3] = (20 + salt_len) * 8;
768 sha1_transform (w0_t3, w1_t3, w2_t3, w3_t3, digest);
771 COMPARE_S_SIMD (digest[3], digest[4], digest[2], digest[1]);
775 __kernel void m08300_s08 (__global pw_t *pws, __global kernel_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 void *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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
779 __kernel void m08300_s16 (__global pw_t *pws, __global kernel_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 void *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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)