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"
19 #include "include/rp_gpu.h"
20 #include "OpenCL/rp.c"
22 #define COMPARE_S "OpenCL/check_single_comp4.c"
23 #define COMPARE_M "OpenCL/check_multi_comp4.c"
25 #define uint_to_hex_lower8_le(i) l_bin2asc[(i)]
27 static void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
55 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
56 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
57 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
58 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
59 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
60 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
61 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
62 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
63 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
64 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
65 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
66 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
67 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
68 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
69 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
70 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
71 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
72 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
73 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
74 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
79 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
80 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
81 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
82 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
83 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
84 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
85 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
86 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
87 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
88 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
89 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
90 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
91 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
92 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
93 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
94 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
95 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
96 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
97 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
98 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
103 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
104 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
105 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
106 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
107 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
108 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
109 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
110 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
111 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
112 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
113 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
114 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
115 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
116 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
117 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
118 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
119 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
120 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
121 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
122 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
127 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
128 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
129 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
130 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
131 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
132 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
133 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
134 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
135 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
136 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
137 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
138 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
139 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
140 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
141 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
142 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
143 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
144 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
145 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
146 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
155 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
161 const u32 lid = get_local_id (0);
167 const u32 gid = get_global_id (0);
171 pw_buf0[0] = pws[gid].i[ 0];
172 pw_buf0[1] = pws[gid].i[ 1];
173 pw_buf0[2] = pws[gid].i[ 2];
174 pw_buf0[3] = pws[gid].i[ 3];
178 pw_buf1[0] = pws[gid].i[ 4];
179 pw_buf1[1] = pws[gid].i[ 5];
180 pw_buf1[2] = pws[gid].i[ 6];
181 pw_buf1[3] = pws[gid].i[ 7];
183 const u32 pw_len = pws[gid].pw_len;
189 __local u32 l_bin2asc[256];
191 const u32 lid4 = lid * 4;
193 const u32 lid40 = lid4 + 0;
194 const u32 lid41 = lid4 + 1;
195 const u32 lid42 = lid4 + 2;
196 const u32 lid43 = lid4 + 3;
198 const u32 v400 = (lid40 >> 0) & 15;
199 const u32 v401 = (lid40 >> 4) & 15;
200 const u32 v410 = (lid41 >> 0) & 15;
201 const u32 v411 = (lid41 >> 4) & 15;
202 const u32 v420 = (lid42 >> 0) & 15;
203 const u32 v421 = (lid42 >> 4) & 15;
204 const u32 v430 = (lid43 >> 0) & 15;
205 const u32 v431 = (lid43 >> 4) & 15;
207 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
208 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
209 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
210 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
211 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
212 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
213 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
214 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
216 barrier (CLK_LOCAL_MEM_FENCE);
218 if (gid >= gid_max) return;
226 salt_buf0[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
227 salt_buf0[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
228 salt_buf0[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 2]);
229 salt_buf0[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 3]);
233 salt_buf1[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 4]);
234 salt_buf1[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 5]);
235 salt_buf1[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 6]);
236 salt_buf1[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 7]);
240 salt_buf2[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 8]);
241 salt_buf2[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 9]);
245 const u32 salt_len = salt_bufs[salt_pos].salt_len;
251 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
281 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
283 append_0x80_2x4 (w0, w1, out_len);
287 w0_t[0] = swap32 (w0[0]);
288 w0_t[1] = swap32 (w0[1]);
289 w0_t[2] = swap32 (w0[2]);
290 w0_t[3] = swap32 (w0[3]);
294 w1_t[0] = swap32 (w1[0]);
295 w1_t[1] = swap32 (w1[1]);
296 w1_t[2] = swap32 (w1[2]);
297 w1_t[3] = swap32 (w1[3]);
301 w2_t[0] = swap32 (w2[0]);
302 w2_t[1] = swap32 (w2[1]);
303 w2_t[2] = swap32 (w2[2]);
304 w2_t[3] = swap32 (w2[3]);
308 w3_t[0] = swap32 (w3[0]);
309 w3_t[1] = swap32 (w3[1]);
311 w3_t[3] = pw_len * 8;
321 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
335 w0_t[0] = salt_buf0[0];
336 w0_t[1] = salt_buf0[1];
337 w0_t[2] = salt_buf0[2];
338 w0_t[3] = salt_buf0[3];
339 w1_t[0] = salt_buf1[0];
340 w1_t[1] = salt_buf1[1];
341 w1_t[2] = salt_buf1[2];
342 w1_t[3] = salt_buf1[3];
343 w2_t[0] = salt_buf2[0];
344 w2_t[1] = salt_buf2[1];
345 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
346 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
347 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
348 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
349 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
350 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
351 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
352 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
353 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
354 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
355 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
356 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
364 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
366 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
367 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
368 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
369 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
370 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
371 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
372 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
373 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
374 w1_t[0] = 0x80000000;
387 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
395 w0_t[0] = salt_buf0[0];
396 w0_t[1] = salt_buf0[1];
397 w0_t[2] = salt_buf0[2];
398 w0_t[3] = salt_buf0[3];
399 w1_t[0] = salt_buf1[0];
400 w1_t[1] = salt_buf1[1];
401 w1_t[2] = salt_buf1[2];
402 w1_t[3] = salt_buf1[3];
403 w2_t[0] = salt_buf2[0];
404 w2_t[1] = salt_buf2[1];
405 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
406 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
407 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
408 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
409 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
410 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
411 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
412 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
413 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
414 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
415 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
416 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
424 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
426 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
427 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
428 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
429 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
430 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
431 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
432 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
433 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
434 w1_t[0] = 0x80000000;
447 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
449 const u32 r0 = digest[3];
450 const u32 r1 = digest[4];
451 const u32 r2 = digest[2];
452 const u32 r3 = digest[1];
458 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
462 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 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 rules_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))) m08400_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 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
472 const u32 lid = get_local_id (0);
478 const u32 gid = get_global_id (0);
482 pw_buf0[0] = pws[gid].i[ 0];
483 pw_buf0[1] = pws[gid].i[ 1];
484 pw_buf0[2] = pws[gid].i[ 2];
485 pw_buf0[3] = pws[gid].i[ 3];
489 pw_buf1[0] = pws[gid].i[ 4];
490 pw_buf1[1] = pws[gid].i[ 5];
491 pw_buf1[2] = pws[gid].i[ 6];
492 pw_buf1[3] = pws[gid].i[ 7];
494 const u32 pw_len = pws[gid].pw_len;
500 __local u32 l_bin2asc[256];
502 const u32 lid4 = lid * 4;
504 const u32 lid40 = lid4 + 0;
505 const u32 lid41 = lid4 + 1;
506 const u32 lid42 = lid4 + 2;
507 const u32 lid43 = lid4 + 3;
509 const u32 v400 = (lid40 >> 0) & 15;
510 const u32 v401 = (lid40 >> 4) & 15;
511 const u32 v410 = (lid41 >> 0) & 15;
512 const u32 v411 = (lid41 >> 4) & 15;
513 const u32 v420 = (lid42 >> 0) & 15;
514 const u32 v421 = (lid42 >> 4) & 15;
515 const u32 v430 = (lid43 >> 0) & 15;
516 const u32 v431 = (lid43 >> 4) & 15;
518 l_bin2asc[lid40] = ((v400 < 10) ? '0' + v400 : 'a' - 10 + v400) << 0
519 | ((v401 < 10) ? '0' + v401 : 'a' - 10 + v401) << 8;
520 l_bin2asc[lid41] = ((v410 < 10) ? '0' + v410 : 'a' - 10 + v410) << 0
521 | ((v411 < 10) ? '0' + v411 : 'a' - 10 + v411) << 8;
522 l_bin2asc[lid42] = ((v420 < 10) ? '0' + v420 : 'a' - 10 + v420) << 0
523 | ((v421 < 10) ? '0' + v421 : 'a' - 10 + v421) << 8;
524 l_bin2asc[lid43] = ((v430 < 10) ? '0' + v430 : 'a' - 10 + v430) << 0
525 | ((v431 < 10) ? '0' + v431 : 'a' - 10 + v431) << 8;
527 barrier (CLK_LOCAL_MEM_FENCE);
529 if (gid >= gid_max) return;
537 salt_buf0[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
538 salt_buf0[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
539 salt_buf0[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 2]);
540 salt_buf0[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 3]);
544 salt_buf1[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 4]);
545 salt_buf1[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 5]);
546 salt_buf1[2] = swap32 (salt_bufs[salt_pos].salt_buf[ 6]);
547 salt_buf1[3] = swap32 (salt_bufs[salt_pos].salt_buf[ 7]);
551 salt_buf2[0] = swap32 (salt_bufs[salt_pos].salt_buf[ 8]);
552 salt_buf2[1] = swap32 (salt_bufs[salt_pos].salt_buf[ 9]);
556 const u32 salt_len = salt_bufs[salt_pos].salt_len;
562 const u32 search[4] =
564 digests_buf[digests_offset].digest_buf[DGST_R0],
565 digests_buf[digests_offset].digest_buf[DGST_R1],
566 digests_buf[digests_offset].digest_buf[DGST_R2],
567 digests_buf[digests_offset].digest_buf[DGST_R3]
574 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
604 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
606 append_0x80_2x4 (w0, w1, out_len);
610 w0_t[0] = swap32 (w0[0]);
611 w0_t[1] = swap32 (w0[1]);
612 w0_t[2] = swap32 (w0[2]);
613 w0_t[3] = swap32 (w0[3]);
617 w1_t[0] = swap32 (w1[0]);
618 w1_t[1] = swap32 (w1[1]);
619 w1_t[2] = swap32 (w1[2]);
620 w1_t[3] = swap32 (w1[3]);
624 w2_t[0] = swap32 (w2[0]);
625 w2_t[1] = swap32 (w2[1]);
626 w2_t[2] = swap32 (w2[2]);
627 w2_t[3] = swap32 (w2[3]);
631 w3_t[0] = swap32 (w3[0]);
632 w3_t[1] = swap32 (w3[1]);
634 w3_t[3] = pw_len * 8;
644 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
658 w0_t[0] = salt_buf0[0];
659 w0_t[1] = salt_buf0[1];
660 w0_t[2] = salt_buf0[2];
661 w0_t[3] = salt_buf0[3];
662 w1_t[0] = salt_buf1[0];
663 w1_t[1] = salt_buf1[1];
664 w1_t[2] = salt_buf1[2];
665 w1_t[3] = salt_buf1[3];
666 w2_t[0] = salt_buf2[0];
667 w2_t[1] = salt_buf2[1];
668 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
669 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
670 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
671 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
672 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
673 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
674 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
675 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
676 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
677 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
678 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
679 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
687 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
689 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
690 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
691 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
692 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
693 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
694 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
695 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
696 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
697 w1_t[0] = 0x80000000;
708 w3_t[3] = (salt_len + 40) * 8;
710 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
718 w0_t[0] = salt_buf0[0];
719 w0_t[1] = salt_buf0[1];
720 w0_t[2] = salt_buf0[2];
721 w0_t[3] = salt_buf0[3];
722 w1_t[0] = salt_buf1[0];
723 w1_t[1] = salt_buf1[1];
724 w1_t[2] = salt_buf1[2];
725 w1_t[3] = salt_buf1[3];
726 w2_t[0] = salt_buf2[0];
727 w2_t[1] = salt_buf2[1];
728 w2_t[2] = uint_to_hex_lower8_le ((a >> 16) & 255) << 0
729 | uint_to_hex_lower8_le ((a >> 24) & 255) << 16;
730 w2_t[3] = uint_to_hex_lower8_le ((a >> 0) & 255) << 0
731 | uint_to_hex_lower8_le ((a >> 8) & 255) << 16;
732 w3_t[0] = uint_to_hex_lower8_le ((b >> 16) & 255) << 0
733 | uint_to_hex_lower8_le ((b >> 24) & 255) << 16;
734 w3_t[1] = uint_to_hex_lower8_le ((b >> 0) & 255) << 0
735 | uint_to_hex_lower8_le ((b >> 8) & 255) << 16;
736 w3_t[2] = uint_to_hex_lower8_le ((c >> 16) & 255) << 0
737 | uint_to_hex_lower8_le ((c >> 24) & 255) << 16;
738 w3_t[3] = uint_to_hex_lower8_le ((c >> 0) & 255) << 0
739 | uint_to_hex_lower8_le ((c >> 8) & 255) << 16;
747 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
749 w0_t[0] = uint_to_hex_lower8_le ((d >> 16) & 255) << 0
750 | uint_to_hex_lower8_le ((d >> 24) & 255) << 16;
751 w0_t[1] = uint_to_hex_lower8_le ((d >> 0) & 255) << 0
752 | uint_to_hex_lower8_le ((d >> 8) & 255) << 16;
753 w0_t[2] = uint_to_hex_lower8_le ((e >> 16) & 255) << 0
754 | uint_to_hex_lower8_le ((e >> 24) & 255) << 16;
755 w0_t[3] = uint_to_hex_lower8_le ((e >> 0) & 255) << 0
756 | uint_to_hex_lower8_le ((e >> 8) & 255) << 16;
757 w1_t[0] = 0x80000000;
768 w3_t[3] = (salt_len + 40) * 8;
770 sha1_transform (w0_t, w1_t, w2_t, w3_t, digest);
772 const u32 r0 = digest[3];
773 const u32 r1 = digest[4];
774 const u32 r2 = digest[2];
775 const u32 r3 = digest[1];
781 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
785 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08400_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 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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)