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_M "check_multi_vect1_comp4.c"
37 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
40 static void md4_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
47 MD4_STEP (MD4_Fo, a, b, c, d, w0[0], MD4C00, MD4S00);
48 MD4_STEP (MD4_Fo, d, a, b, c, w0[1], MD4C00, MD4S01);
49 MD4_STEP (MD4_Fo, c, d, a, b, w0[2], MD4C00, MD4S02);
50 MD4_STEP (MD4_Fo, b, c, d, a, w0[3], MD4C00, MD4S03);
51 MD4_STEP (MD4_Fo, a, b, c, d, w1[0], MD4C00, MD4S00);
52 MD4_STEP (MD4_Fo, d, a, b, c, w1[1], MD4C00, MD4S01);
53 MD4_STEP (MD4_Fo, c, d, a, b, w1[2], MD4C00, MD4S02);
54 MD4_STEP (MD4_Fo, b, c, d, a, w1[3], MD4C00, MD4S03);
55 MD4_STEP (MD4_Fo, a, b, c, d, w2[0], MD4C00, MD4S00);
56 MD4_STEP (MD4_Fo, d, a, b, c, w2[1], MD4C00, MD4S01);
57 MD4_STEP (MD4_Fo, c, d, a, b, w2[2], MD4C00, MD4S02);
58 MD4_STEP (MD4_Fo, b, c, d, a, w2[3], MD4C00, MD4S03);
59 MD4_STEP (MD4_Fo, a, b, c, d, w3[0], MD4C00, MD4S00);
60 MD4_STEP (MD4_Fo, d, a, b, c, w3[1], MD4C00, MD4S01);
61 MD4_STEP (MD4_Fo, c, d, a, b, w3[2], MD4C00, MD4S02);
62 MD4_STEP (MD4_Fo, b, c, d, a, w3[3], MD4C00, MD4S03);
64 MD4_STEP (MD4_Go, a, b, c, d, w0[0], MD4C01, MD4S10);
65 MD4_STEP (MD4_Go, d, a, b, c, w1[0], MD4C01, MD4S11);
66 MD4_STEP (MD4_Go, c, d, a, b, w2[0], MD4C01, MD4S12);
67 MD4_STEP (MD4_Go, b, c, d, a, w3[0], MD4C01, MD4S13);
68 MD4_STEP (MD4_Go, a, b, c, d, w0[1], MD4C01, MD4S10);
69 MD4_STEP (MD4_Go, d, a, b, c, w1[1], MD4C01, MD4S11);
70 MD4_STEP (MD4_Go, c, d, a, b, w2[1], MD4C01, MD4S12);
71 MD4_STEP (MD4_Go, b, c, d, a, w3[1], MD4C01, MD4S13);
72 MD4_STEP (MD4_Go, a, b, c, d, w0[2], MD4C01, MD4S10);
73 MD4_STEP (MD4_Go, d, a, b, c, w1[2], MD4C01, MD4S11);
74 MD4_STEP (MD4_Go, c, d, a, b, w2[2], MD4C01, MD4S12);
75 MD4_STEP (MD4_Go, b, c, d, a, w3[2], MD4C01, MD4S13);
76 MD4_STEP (MD4_Go, a, b, c, d, w0[3], MD4C01, MD4S10);
77 MD4_STEP (MD4_Go, d, a, b, c, w1[3], MD4C01, MD4S11);
78 MD4_STEP (MD4_Go, c, d, a, b, w2[3], MD4C01, MD4S12);
79 MD4_STEP (MD4_Go, b, c, d, a, w3[3], MD4C01, MD4S13);
81 MD4_STEP (MD4_H , a, b, c, d, w0[0], MD4C02, MD4S20);
82 MD4_STEP (MD4_H , d, a, b, c, w2[0], MD4C02, MD4S21);
83 MD4_STEP (MD4_H , c, d, a, b, w1[0], MD4C02, MD4S22);
84 MD4_STEP (MD4_H , b, c, d, a, w3[0], MD4C02, MD4S23);
85 MD4_STEP (MD4_H , a, b, c, d, w0[2], MD4C02, MD4S20);
86 MD4_STEP (MD4_H , d, a, b, c, w2[2], MD4C02, MD4S21);
87 MD4_STEP (MD4_H , c, d, a, b, w1[2], MD4C02, MD4S22);
88 MD4_STEP (MD4_H , b, c, d, a, w3[2], MD4C02, MD4S23);
89 MD4_STEP (MD4_H , a, b, c, d, w0[1], MD4C02, MD4S20);
90 MD4_STEP (MD4_H , d, a, b, c, w2[1], MD4C02, MD4S21);
91 MD4_STEP (MD4_H , c, d, a, b, w1[1], MD4C02, MD4S22);
92 MD4_STEP (MD4_H , b, c, d, a, w3[1], MD4C02, MD4S23);
93 MD4_STEP (MD4_H , a, b, c, d, w0[3], MD4C02, MD4S20);
94 MD4_STEP (MD4_H , d, a, b, c, w2[3], MD4C02, MD4S21);
95 MD4_STEP (MD4_H , c, d, a, b, w1[3], MD4C02, MD4S22);
96 MD4_STEP (MD4_H , b, c, d, a, w3[3], MD4C02, MD4S23);
104 static void sha1_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[5])
132 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
133 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
134 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
135 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
136 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
137 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
138 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
139 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
140 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
141 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
142 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
143 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
144 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
145 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
146 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
147 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
148 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
149 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
150 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
151 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
156 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
157 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
158 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
159 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
160 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
161 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
162 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
163 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
164 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
165 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
166 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
167 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
168 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
169 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
170 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
171 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
172 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
173 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
174 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
175 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
180 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
181 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
182 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
183 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
184 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
185 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
186 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
187 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
188 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
189 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
190 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
191 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
192 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
193 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
194 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
195 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
196 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
197 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
198 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
199 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
204 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
205 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
206 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
207 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
208 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
209 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
210 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
211 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
212 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
213 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
214 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
215 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
216 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
217 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
218 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
219 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
220 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
221 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
222 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
223 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
232 static void hmac_sha1_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[5], u32x opad[5])
234 w0[0] = w0[0] ^ 0x36363636;
235 w0[1] = w0[1] ^ 0x36363636;
236 w0[2] = w0[2] ^ 0x36363636;
237 w0[3] = w0[3] ^ 0x36363636;
238 w1[0] = w1[0] ^ 0x36363636;
239 w1[1] = w1[1] ^ 0x36363636;
240 w1[2] = w1[2] ^ 0x36363636;
241 w1[3] = w1[3] ^ 0x36363636;
242 w2[0] = w2[0] ^ 0x36363636;
243 w2[1] = w2[1] ^ 0x36363636;
244 w2[2] = w2[2] ^ 0x36363636;
245 w2[3] = w2[3] ^ 0x36363636;
246 w3[0] = w3[0] ^ 0x36363636;
247 w3[1] = w3[1] ^ 0x36363636;
248 w3[2] = w3[2] ^ 0x36363636;
249 w3[3] = w3[3] ^ 0x36363636;
257 sha1_transform (w0, w1, w2, w3, ipad);
259 w0[0] = w0[0] ^ 0x6a6a6a6a;
260 w0[1] = w0[1] ^ 0x6a6a6a6a;
261 w0[2] = w0[2] ^ 0x6a6a6a6a;
262 w0[3] = w0[3] ^ 0x6a6a6a6a;
263 w1[0] = w1[0] ^ 0x6a6a6a6a;
264 w1[1] = w1[1] ^ 0x6a6a6a6a;
265 w1[2] = w1[2] ^ 0x6a6a6a6a;
266 w1[3] = w1[3] ^ 0x6a6a6a6a;
267 w2[0] = w2[0] ^ 0x6a6a6a6a;
268 w2[1] = w2[1] ^ 0x6a6a6a6a;
269 w2[2] = w2[2] ^ 0x6a6a6a6a;
270 w2[3] = w2[3] ^ 0x6a6a6a6a;
271 w3[0] = w3[0] ^ 0x6a6a6a6a;
272 w3[1] = w3[1] ^ 0x6a6a6a6a;
273 w3[2] = w3[2] ^ 0x6a6a6a6a;
274 w3[3] = w3[3] ^ 0x6a6a6a6a;
282 sha1_transform (w0, w1, w2, w3, opad);
285 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])
293 sha1_transform (w0, w1, w2, w3, digest);
310 w3[3] = (64 + 20) * 8;
318 sha1_transform (w0, w1, w2, w3, digest);
321 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m02100_init (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
327 const u32 gid = get_global_id (0);
329 if (gid >= gid_max) return;
333 w0[0] = pws[gid].i[ 0];
334 w0[1] = pws[gid].i[ 1];
335 w0[2] = pws[gid].i[ 2];
336 w0[3] = pws[gid].i[ 3];
359 const u32 pw_len = pws[gid].pw_len;
365 u32 salt_len = salt_bufs[salt_pos].salt_len;
371 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
372 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
373 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
374 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[3];
375 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[4];
376 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[5];
377 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[6];
378 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[7];
379 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[8];
380 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[9];
388 append_0x80_1 (w0, pw_len);
390 make_unicode (w0, w0, w1);
392 w3[2] = pw_len * 2 * 8;
396 digest_md4[0] = MD4M_A;
397 digest_md4[1] = MD4M_B;
398 digest_md4[2] = MD4M_C;
399 digest_md4[3] = MD4M_D;
401 md4_transform (w0, w1, w2, w3, digest_md4);
403 w0[0] = digest_md4[0];
404 w0[1] = digest_md4[1];
405 w0[2] = digest_md4[2];
406 w0[3] = digest_md4[3];
407 w1[0] = salt_buf0[0];
408 w1[1] = salt_buf0[1];
409 w1[2] = salt_buf0[2];
410 w1[3] = salt_buf0[3];
411 w2[0] = salt_buf1[0];
412 w2[1] = salt_buf1[1];
413 w2[2] = salt_buf1[2];
414 w2[3] = salt_buf1[3];
415 w3[0] = salt_buf2[0];
416 w3[1] = salt_buf2[1];
417 w3[2] = (16 + salt_len) * 8;
420 append_0x80_4 (w0, w1, w2, w3, 16 + salt_len);
422 digest_md4[0] = MD4M_A;
423 digest_md4[1] = MD4M_B;
424 digest_md4[2] = MD4M_C;
425 digest_md4[3] = MD4M_D;
427 md4_transform (w0, w1, w2, w3, digest_md4);
433 w0[0] = swap_workaround (digest_md4[0]);
434 w0[1] = swap_workaround (digest_md4[1]);
435 w0[2] = swap_workaround (digest_md4[2]);
436 w0[3] = swap_workaround (digest_md4[3]);
453 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
455 tmps[gid].ipad[0] = ipad[0];
456 tmps[gid].ipad[1] = ipad[1];
457 tmps[gid].ipad[2] = ipad[2];
458 tmps[gid].ipad[3] = ipad[3];
459 tmps[gid].ipad[4] = ipad[4];
461 tmps[gid].opad[0] = opad[0];
462 tmps[gid].opad[1] = opad[1];
463 tmps[gid].opad[2] = opad[2];
464 tmps[gid].opad[3] = opad[3];
465 tmps[gid].opad[4] = opad[4];
471 w0[0] = salt_buf0[0];
472 w0[1] = salt_buf0[1];
473 w0[2] = salt_buf0[2];
474 w0[3] = salt_buf0[3];
475 w1[0] = salt_buf1[0];
476 w1[1] = salt_buf1[1];
477 w1[2] = salt_buf1[2];
478 w1[3] = salt_buf1[3];
479 w2[0] = salt_buf2[0];
480 w2[1] = salt_buf2[1];
486 w3[3] = (64 + salt_len + 4) * 8;
488 append_0x01_4 (w0, w1, w2, w3, salt_len + 3);
489 append_0x80_4 (w0, w1, w2, w3, salt_len + 4);
491 w0[0] = swap_workaround (w0[0]);
492 w0[1] = swap_workaround (w0[1]);
493 w0[2] = swap_workaround (w0[2]);
494 w0[3] = swap_workaround (w0[3]);
495 w1[0] = swap_workaround (w1[0]);
496 w1[1] = swap_workaround (w1[1]);
497 w1[2] = swap_workaround (w1[2]);
498 w1[3] = swap_workaround (w1[3]);
499 w2[0] = swap_workaround (w2[0]);
500 w2[1] = swap_workaround (w2[1]);
501 w2[2] = swap_workaround (w2[2]);
502 w2[3] = swap_workaround (w2[3]);
503 w3[0] = swap_workaround (w3[0]);
504 w3[1] = swap_workaround (w3[1]);
508 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, digest);
510 tmps[gid].dgst[0] = digest[0];
511 tmps[gid].dgst[1] = digest[1];
512 tmps[gid].dgst[2] = digest[2];
513 tmps[gid].dgst[3] = digest[3];
514 tmps[gid].dgst[4] = digest[4];
516 tmps[gid].out[0] = digest[0];
517 tmps[gid].out[1] = digest[1];
518 tmps[gid].out[2] = digest[2];
519 tmps[gid].out[3] = digest[3];
522 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m02100_loop (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
528 const u32 gid = get_global_id (0);
530 if (gid >= gid_max) return;
535 ipad[0] = tmps[gid].ipad[0];
536 ipad[1] = tmps[gid].ipad[1];
537 ipad[2] = tmps[gid].ipad[2];
538 ipad[3] = tmps[gid].ipad[3];
539 ipad[4] = tmps[gid].ipad[4];
541 opad[0] = tmps[gid].opad[0];
542 opad[1] = tmps[gid].opad[1];
543 opad[2] = tmps[gid].opad[2];
544 opad[3] = tmps[gid].opad[3];
545 opad[4] = tmps[gid].opad[4];
554 dgst[0] = tmps[gid].dgst[0];
555 dgst[1] = tmps[gid].dgst[1];
556 dgst[2] = tmps[gid].dgst[2];
557 dgst[3] = tmps[gid].dgst[3];
558 dgst[4] = tmps[gid].dgst[4];
560 out[0] = tmps[gid].out[0];
561 out[1] = tmps[gid].out[1];
562 out[2] = tmps[gid].out[2];
563 out[3] = tmps[gid].out[3];
565 for (u32 i = 0; i < loop_cnt; i++)
587 w3[3] = (64 + 20) * 8;
589 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
597 tmps[gid].dgst[0] = dgst[0];
598 tmps[gid].dgst[1] = dgst[1];
599 tmps[gid].dgst[2] = dgst[2];
600 tmps[gid].dgst[3] = dgst[3];
601 tmps[gid].dgst[4] = dgst[4];
603 tmps[gid].out[0] = out[0];
604 tmps[gid].out[1] = out[1];
605 tmps[gid].out[2] = out[2];
606 tmps[gid].out[3] = out[3];
609 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m02100_comp (__global pw_t *pws, __global gpu_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global dcc2_tmp_t *tmps, __global void *hooks, __global u32 *bitmaps_buf_s1_a, __global u32 *bitmaps_buf_s1_b, __global u32 *bitmaps_buf_s1_c, __global u32 *bitmaps_buf_s1_d, __global u32 *bitmaps_buf_s2_a, __global u32 *bitmaps_buf_s2_b, __global u32 *bitmaps_buf_s2_c, __global u32 *bitmaps_buf_s2_d, __global plain_t *plains_buf, __global digest_t *digests_buf, __global u32 *hashes_shown, __global salt_t *salt_bufs, __global 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)
615 const u32 gid = get_global_id (0);
617 if (gid >= gid_max) return;
619 const u32 lid = get_local_id (0);
621 const u32x r0 = tmps[gid].out[DGST_R0];
622 const u32x r1 = tmps[gid].out[DGST_R1];
623 const u32x r2 = tmps[gid].out[DGST_R2];
624 const u32x r3 = tmps[gid].out[DGST_R3];
628 #include VECT_COMPARE_M