2 * Authors.....: Jens Steube <jens.steube@gmail.com>
3 * Gabriele Gristina <matrix@hashcat.net>
10 #include "inc_hash_constants.h"
11 #include "inc_vendor.cl"
18 #include "inc_hash_functions.cl"
22 #include "inc_types.cl"
23 #include "inc_common.cl"
25 #define COMPARE_S "inc_comp_single.cl"
26 #define COMPARE_M "inc_comp_multi.cl"
28 __constant char lotus64_table[] = "0123456789ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz+/";
30 #define uint_to_hex_upper8(i) l_bin2asc[(i)]
32 __constant u32 lotus_magic_table[256] =
34 0xbd, 0x56, 0xea, 0xf2, 0xa2, 0xf1, 0xac, 0x2a,
35 0xb0, 0x93, 0xd1, 0x9c, 0x1b, 0x33, 0xfd, 0xd0,
36 0x30, 0x04, 0xb6, 0xdc, 0x7d, 0xdf, 0x32, 0x4b,
37 0xf7, 0xcb, 0x45, 0x9b, 0x31, 0xbb, 0x21, 0x5a,
38 0x41, 0x9f, 0xe1, 0xd9, 0x4a, 0x4d, 0x9e, 0xda,
39 0xa0, 0x68, 0x2c, 0xc3, 0x27, 0x5f, 0x80, 0x36,
40 0x3e, 0xee, 0xfb, 0x95, 0x1a, 0xfe, 0xce, 0xa8,
41 0x34, 0xa9, 0x13, 0xf0, 0xa6, 0x3f, 0xd8, 0x0c,
42 0x78, 0x24, 0xaf, 0x23, 0x52, 0xc1, 0x67, 0x17,
43 0xf5, 0x66, 0x90, 0xe7, 0xe8, 0x07, 0xb8, 0x60,
44 0x48, 0xe6, 0x1e, 0x53, 0xf3, 0x92, 0xa4, 0x72,
45 0x8c, 0x08, 0x15, 0x6e, 0x86, 0x00, 0x84, 0xfa,
46 0xf4, 0x7f, 0x8a, 0x42, 0x19, 0xf6, 0xdb, 0xcd,
47 0x14, 0x8d, 0x50, 0x12, 0xba, 0x3c, 0x06, 0x4e,
48 0xec, 0xb3, 0x35, 0x11, 0xa1, 0x88, 0x8e, 0x2b,
49 0x94, 0x99, 0xb7, 0x71, 0x74, 0xd3, 0xe4, 0xbf,
50 0x3a, 0xde, 0x96, 0x0e, 0xbc, 0x0a, 0xed, 0x77,
51 0xfc, 0x37, 0x6b, 0x03, 0x79, 0x89, 0x62, 0xc6,
52 0xd7, 0xc0, 0xd2, 0x7c, 0x6a, 0x8b, 0x22, 0xa3,
53 0x5b, 0x05, 0x5d, 0x02, 0x75, 0xd5, 0x61, 0xe3,
54 0x18, 0x8f, 0x55, 0x51, 0xad, 0x1f, 0x0b, 0x5e,
55 0x85, 0xe5, 0xc2, 0x57, 0x63, 0xca, 0x3d, 0x6c,
56 0xb4, 0xc5, 0xcc, 0x70, 0xb2, 0x91, 0x59, 0x0d,
57 0x47, 0x20, 0xc8, 0x4f, 0x58, 0xe0, 0x01, 0xe2,
58 0x16, 0x38, 0xc4, 0x6f, 0x3b, 0x0f, 0x65, 0x46,
59 0xbe, 0x7e, 0x2d, 0x7b, 0x82, 0xf9, 0x40, 0xb5,
60 0x1d, 0x73, 0xf8, 0xeb, 0x26, 0xc7, 0x87, 0x97,
61 0x25, 0x54, 0xb1, 0x28, 0xaa, 0x98, 0x9d, 0xa5,
62 0x64, 0x6d, 0x7a, 0xd4, 0x10, 0x81, 0x44, 0xef,
63 0x49, 0xd6, 0xae, 0x2e, 0xdd, 0x76, 0x5c, 0x2f,
64 0xa7, 0x1c, 0xc9, 0x09, 0x69, 0x9a, 0x83, 0xcf,
65 0x29, 0x39, 0xb9, 0xe9, 0x4c, 0xff, 0x43, 0xab,
69 #define BOX1(S,i) (S)[(i)]
71 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1])
73 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3])
75 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7])
77 #define BOX1(S,i) (u32x) ((S)[(i).s0], (S)[(i).s1], (S)[(i).s2], (S)[(i).s3], (S)[(i).s4], (S)[(i).s5], (S)[(i).s6], (S)[(i).s7], (S)[(i).s8], (S)[(i).s9], (S)[(i).sa], (S)[(i).sb], (S)[(i).sc], (S)[(i).sd], (S)[(i).se], (S)[(i).sf])
80 void lotus_mix (u32x *in, __local u32 *s_lotus_magic_table)
84 for (int i = 0; i < 18; i++)
88 for (int j = 0; j < 12; j++)
93 p = (p + s--) & 0xff; p = ((tmp_in >> 0) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 0;
94 p = (p + s--) & 0xff; p = ((tmp_in >> 8) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 8;
95 p = (p + s--) & 0xff; p = ((tmp_in >> 16) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 16;
96 p = (p + s--) & 0xff; p = ((tmp_in >> 24) & 0xff) ^ BOX1 (s_lotus_magic_table, p); tmp_out |= p << 24;
103 void lotus_transform_password (u32x in[4], u32x out[4], __local u32 *s_lotus_magic_table)
105 u32x t = out[3] >> 24;
112 for (int i = 0; i < 4; i++)
114 t ^= (in[i] >> 0) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 0; t = ((out[i] >> 0) & 0xff);
115 t ^= (in[i] >> 8) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 8; t = ((out[i] >> 8) & 0xff);
116 t ^= (in[i] >> 16) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 16; t = ((out[i] >> 16) & 0xff);
117 t ^= (in[i] >> 24) & 0xff; c = BOX1 (s_lotus_magic_table, t); out[i] ^= c << 24; t = ((out[i] >> 24) & 0xff);
121 void pad (u32 w[4], const u32 len)
123 const u32 val = 16 - len;
125 const u32 mask1 = val << 24;
127 const u32 mask2 = val << 16
130 const u32 mask3 = val << 8
134 const u32 mask4 = val << 0
141 case 0: w[0] = mask4;
146 case 1: w[0] |= mask3;
151 case 2: w[0] |= mask2;
156 case 3: w[0] |= mask1;
161 case 4: w[1] = mask4;
165 case 5: w[1] |= mask3;
169 case 6: w[1] |= mask2;
173 case 7: w[1] |= mask1;
177 case 8: w[2] = mask4;
180 case 9: w[2] |= mask3;
183 case 10: w[2] |= mask2;
186 case 11: w[2] |= mask1;
189 case 12: w[3] = mask4;
191 case 13: w[3] |= mask3;
193 case 14: w[3] |= mask2;
195 case 15: w[3] |= mask1;
200 void mdtransform_norecalc (u32x state[4], u32x block[4], __local u32 *s_lotus_magic_table)
212 x[ 8] = state[0] ^ block[0];
213 x[ 9] = state[1] ^ block[1];
214 x[10] = state[2] ^ block[2];
215 x[11] = state[3] ^ block[3];
217 lotus_mix (x, s_lotus_magic_table);
225 void mdtransform (u32x state[4], u32x checksum[4], u32x block[4], __local u32 *s_lotus_magic_table)
227 mdtransform_norecalc (state, block, s_lotus_magic_table);
229 lotus_transform_password (block, checksum, s_lotus_magic_table);
232 void domino_big_md (const u32x saved_key[16], const u32 size, u32x state[4], __local u32 *s_lotus_magic_table)
251 for (curpos = 0, idx = 0; curpos + 16 < size; curpos += 16, idx += 4)
253 block[0] = saved_key[idx + 0];
254 block[1] = saved_key[idx + 1];
255 block[2] = saved_key[idx + 2];
256 block[3] = saved_key[idx + 3];
258 mdtransform (state, checksum, block, s_lotus_magic_table);
261 block[0] = saved_key[idx + 0];
262 block[1] = saved_key[idx + 1];
263 block[2] = saved_key[idx + 2];
264 block[3] = saved_key[idx + 3];
266 mdtransform (state, checksum, block, s_lotus_magic_table);
268 mdtransform_norecalc (state, checksum, s_lotus_magic_table);
271 void sha1_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[5])
299 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w0_t);
300 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w1_t);
301 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w2_t);
302 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w3_t);
303 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w4_t);
304 SHA1_STEP (SHA1_F0o, A, B, C, D, E, w5_t);
305 SHA1_STEP (SHA1_F0o, E, A, B, C, D, w6_t);
306 SHA1_STEP (SHA1_F0o, D, E, A, B, C, w7_t);
307 SHA1_STEP (SHA1_F0o, C, D, E, A, B, w8_t);
308 SHA1_STEP (SHA1_F0o, B, C, D, E, A, w9_t);
309 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wa_t);
310 SHA1_STEP (SHA1_F0o, E, A, B, C, D, wb_t);
311 SHA1_STEP (SHA1_F0o, D, E, A, B, C, wc_t);
312 SHA1_STEP (SHA1_F0o, C, D, E, A, B, wd_t);
313 SHA1_STEP (SHA1_F0o, B, C, D, E, A, we_t);
314 SHA1_STEP (SHA1_F0o, A, B, C, D, E, wf_t);
315 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F0o, E, A, B, C, D, w0_t);
316 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F0o, D, E, A, B, C, w1_t);
317 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F0o, C, D, E, A, B, w2_t);
318 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F0o, B, C, D, E, A, w3_t);
323 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w4_t);
324 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w5_t);
325 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w6_t);
326 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w7_t);
327 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w8_t);
328 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w9_t);
329 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wa_t);
330 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wb_t);
331 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wc_t);
332 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wd_t);
333 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, we_t);
334 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wf_t);
335 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w0_t);
336 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w1_t);
337 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w2_t);
338 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w3_t);
339 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w4_t);
340 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w5_t);
341 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w6_t);
342 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w7_t);
347 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w8_t);
348 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w9_t);
349 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wa_t);
350 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wb_t);
351 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wc_t);
352 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, wd_t);
353 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, we_t);
354 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, wf_t);
355 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w0_t);
356 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w1_t);
357 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w2_t);
358 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w3_t);
359 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w4_t);
360 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, w5_t);
361 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, w6_t);
362 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F2o, A, B, C, D, E, w7_t);
363 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F2o, E, A, B, C, D, w8_t);
364 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F2o, D, E, A, B, C, w9_t);
365 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F2o, C, D, E, A, B, wa_t);
366 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F2o, B, C, D, E, A, wb_t);
371 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wc_t);
372 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wd_t);
373 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, we_t);
374 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, wf_t);
375 w0_t = rotl32 ((wd_t ^ w8_t ^ w2_t ^ w0_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w0_t);
376 w1_t = rotl32 ((we_t ^ w9_t ^ w3_t ^ w1_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w1_t);
377 w2_t = rotl32 ((wf_t ^ wa_t ^ w4_t ^ w2_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w2_t);
378 w3_t = rotl32 ((w0_t ^ wb_t ^ w5_t ^ w3_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w3_t);
379 w4_t = rotl32 ((w1_t ^ wc_t ^ w6_t ^ w4_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w4_t);
380 w5_t = rotl32 ((w2_t ^ wd_t ^ w7_t ^ w5_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, w5_t);
381 w6_t = rotl32 ((w3_t ^ we_t ^ w8_t ^ w6_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, w6_t);
382 w7_t = rotl32 ((w4_t ^ wf_t ^ w9_t ^ w7_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, w7_t);
383 w8_t = rotl32 ((w5_t ^ w0_t ^ wa_t ^ w8_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, w8_t);
384 w9_t = rotl32 ((w6_t ^ w1_t ^ wb_t ^ w9_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, w9_t);
385 wa_t = rotl32 ((w7_t ^ w2_t ^ wc_t ^ wa_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wa_t);
386 wb_t = rotl32 ((w8_t ^ w3_t ^ wd_t ^ wb_t), 1u); SHA1_STEP (SHA1_F1, A, B, C, D, E, wb_t);
387 wc_t = rotl32 ((w9_t ^ w4_t ^ we_t ^ wc_t), 1u); SHA1_STEP (SHA1_F1, E, A, B, C, D, wc_t);
388 wd_t = rotl32 ((wa_t ^ w5_t ^ wf_t ^ wd_t), 1u); SHA1_STEP (SHA1_F1, D, E, A, B, C, wd_t);
389 we_t = rotl32 ((wb_t ^ w6_t ^ w0_t ^ we_t), 1u); SHA1_STEP (SHA1_F1, C, D, E, A, B, we_t);
390 wf_t = rotl32 ((wc_t ^ w7_t ^ w1_t ^ wf_t), 1u); SHA1_STEP (SHA1_F1, B, C, D, E, A, wf_t);
399 void hmac_sha1_pad (u32 w0[4], u32 w1[4], u32 w2[4], u32 w3[4], u32 ipad[5], u32 opad[5])
401 w0[0] = w0[0] ^ 0x36363636;
402 w0[1] = w0[1] ^ 0x36363636;
403 w0[2] = w0[2] ^ 0x36363636;
404 w0[3] = w0[3] ^ 0x36363636;
405 w1[0] = w1[0] ^ 0x36363636;
406 w1[1] = w1[1] ^ 0x36363636;
407 w1[2] = w1[2] ^ 0x36363636;
408 w1[3] = w1[3] ^ 0x36363636;
409 w2[0] = w2[0] ^ 0x36363636;
410 w2[1] = w2[1] ^ 0x36363636;
411 w2[2] = w2[2] ^ 0x36363636;
412 w2[3] = w2[3] ^ 0x36363636;
413 w3[0] = w3[0] ^ 0x36363636;
414 w3[1] = w3[1] ^ 0x36363636;
415 w3[2] = w3[2] ^ 0x36363636;
416 w3[3] = w3[3] ^ 0x36363636;
424 sha1_transform (w0, w1, w2, w3, ipad);
426 w0[0] = w0[0] ^ 0x6a6a6a6a;
427 w0[1] = w0[1] ^ 0x6a6a6a6a;
428 w0[2] = w0[2] ^ 0x6a6a6a6a;
429 w0[3] = w0[3] ^ 0x6a6a6a6a;
430 w1[0] = w1[0] ^ 0x6a6a6a6a;
431 w1[1] = w1[1] ^ 0x6a6a6a6a;
432 w1[2] = w1[2] ^ 0x6a6a6a6a;
433 w1[3] = w1[3] ^ 0x6a6a6a6a;
434 w2[0] = w2[0] ^ 0x6a6a6a6a;
435 w2[1] = w2[1] ^ 0x6a6a6a6a;
436 w2[2] = w2[2] ^ 0x6a6a6a6a;
437 w2[3] = w2[3] ^ 0x6a6a6a6a;
438 w3[0] = w3[0] ^ 0x6a6a6a6a;
439 w3[1] = w3[1] ^ 0x6a6a6a6a;
440 w3[2] = w3[2] ^ 0x6a6a6a6a;
441 w3[3] = w3[3] ^ 0x6a6a6a6a;
449 sha1_transform (w0, w1, w2, w3, opad);
452 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])
460 sha1_transform (w0, w1, w2, w3, digest);
477 w3[3] = (64 + 20) * 8;
485 sha1_transform (w0, w1, w2, w3, digest);
488 void base64_encode (u8 *base64_hash, const u32 len, const u8 *base64_plain)
490 u8 *out_ptr = (u8 *) base64_hash;
491 u8 *in_ptr = (u8 *) base64_plain;
495 for (i = 0; i < len; i += 3)
497 char out_val0 = lotus64_table [ ((in_ptr[0] >> 2) & 0x3f)];
498 char out_val1 = lotus64_table [((in_ptr[0] << 4) & 0x30) | ((in_ptr[1] >> 4) & 0x0f)];
499 char out_val2 = lotus64_table [((in_ptr[1] << 2) & 0x3c) | ((in_ptr[2] >> 6) & 0x03)];
500 char out_val3 = lotus64_table [ ((in_ptr[2] >> 0) & 0x3f)];
502 out_ptr[0] = out_val0 & 0x7f;
503 out_ptr[1] = out_val1 & 0x7f;
504 out_ptr[2] = out_val2 & 0x7f;
505 out_ptr[3] = out_val3 & 0x7f;
512 void lotus6_base64_encode (u8 base64_hash[24], const u32 salt0, const u32 salt1, u32 a, u32 b, u32 c)
514 uchar4 salt0c = as_uchar4 (salt0);
515 uchar4 salt1c = as_uchar4 (salt1);
525 u8 tmp[24]; // size 22 (=pw_len) is needed but base64 needs size divisible by 4
528 * Copy $salt.$digest to a tmp buffer
533 base64_plain[ 0] = salt0c.s0;
534 base64_plain[ 1] = salt0c.s1;
535 base64_plain[ 2] = salt0c.s2;
536 base64_plain[ 3] = salt0c.s3;
537 base64_plain[ 3] -= -4; // dont ask!
538 base64_plain[ 4] = salt1c.s0;
539 base64_plain[ 5] = ac.s0;
540 base64_plain[ 6] = ac.s1;
541 base64_plain[ 7] = ac.s2;
542 base64_plain[ 8] = ac.s3;
543 base64_plain[ 9] = bc.s0;
544 base64_plain[10] = bc.s1;
545 base64_plain[11] = bc.s2;
546 base64_plain[12] = bc.s3;
547 base64_plain[13] = cc.s0;
548 base64_plain[14] = cc.s1;
549 base64_plain[15] = cc.s2;
552 * base64 encode the $salt.$digest string
555 base64_encode (tmp + 2, 14, base64_plain);
557 base64_hash[ 0] = '(';
558 base64_hash[ 1] = 'G';
559 base64_hash[ 2] = tmp[ 2];
560 base64_hash[ 3] = tmp[ 3];
561 base64_hash[ 4] = tmp[ 4];
562 base64_hash[ 5] = tmp[ 5];
563 base64_hash[ 6] = tmp[ 6];
564 base64_hash[ 7] = tmp[ 7];
565 base64_hash[ 8] = tmp[ 8];
566 base64_hash[ 9] = tmp[ 9];
567 base64_hash[10] = tmp[10];
568 base64_hash[11] = tmp[11];
569 base64_hash[12] = tmp[12];
570 base64_hash[13] = tmp[13];
571 base64_hash[14] = tmp[14];
572 base64_hash[15] = tmp[15];
573 base64_hash[16] = tmp[16];
574 base64_hash[17] = tmp[17];
575 base64_hash[18] = tmp[18];
576 base64_hash[19] = tmp[19];
577 base64_hash[20] = tmp[20];
578 base64_hash[21] = ')';
581 __kernel void m09100_init (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global lotus8_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 wpa_t *wpa_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
587 const u32 gid = get_global_id (0);
588 const u32 lid = get_local_id (0);
589 const u32 lsz = get_local_size (0);
595 __local u32 s_lotus_magic_table[256];
597 for (u32 i = lid; i < 256; i += lsz)
599 s_lotus_magic_table[i] = lotus_magic_table[i];
602 __local u32 l_bin2asc[256];
604 for (u32 i = lid; i < 256; i += lsz)
606 const u32 i0 = (i >> 0) & 15;
607 const u32 i1 = (i >> 4) & 15;
609 l_bin2asc[i] = ((i0 < 10) ? '0' + i0 : 'A' - 10 + i0) << 8
610 | ((i1 < 10) ? '0' + i1 : 'A' - 10 + i1) << 0;
613 barrier (CLK_LOCAL_MEM_FENCE);
615 if (gid >= gid_max) return;
623 w[ 0] = pws[gid].i[ 0];
624 w[ 1] = pws[gid].i[ 1];
625 w[ 2] = pws[gid].i[ 2];
626 w[ 3] = pws[gid].i[ 3];
628 w[ 4] = pws[gid].i[ 4];
629 w[ 5] = pws[gid].i[ 5];
630 w[ 6] = pws[gid].i[ 6];
631 w[ 7] = pws[gid].i[ 7];
633 w[ 8] = pws[gid].i[ 8];
634 w[ 9] = pws[gid].i[ 9];
635 w[10] = pws[gid].i[10];
636 w[11] = pws[gid].i[11];
638 w[12] = pws[gid].i[12];
639 w[13] = pws[gid].i[13];
640 w[14] = pws[gid].i[14];
641 w[15] = pws[gid].i[15];
647 u32 pw_len = pws[gid].pw_len;
651 pad (&w[ 0], pw_len & 0xf);
653 else if (pw_len < 32)
655 pad (&w[ 4], pw_len & 0xf);
657 else if (pw_len < 48)
659 pad (&w[ 8], pw_len & 0xf);
661 else if (pw_len < 64)
663 pad (&w[12], pw_len & 0xf);
670 u32 salt_len = salt_bufs[salt_pos].salt_len;
674 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
675 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
676 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
677 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
681 salt_buf1[0] = 0x01000000;
682 salt_buf1[1] = 0x00000080;
700 const u32 salt0 = salt_buf0[0];
701 const u32 salt1 = (salt_buf0[1] & 0xff) | ('(' << 8);
704 * Lotus 6 hash - SEC_pwddigest_V2
733 domino_big_md (w_tmp, pw_len, state, s_lotus_magic_table);
735 const u32 w0_t = uint_to_hex_upper8 ((state[0] >> 0) & 255) << 0
736 | uint_to_hex_upper8 ((state[0] >> 8) & 255) << 16;
737 const u32 w1_t = uint_to_hex_upper8 ((state[0] >> 16) & 255) << 0
738 | uint_to_hex_upper8 ((state[0] >> 24) & 255) << 16;
739 const u32 w2_t = uint_to_hex_upper8 ((state[1] >> 0) & 255) << 0
740 | uint_to_hex_upper8 ((state[1] >> 8) & 255) << 16;
741 const u32 w3_t = uint_to_hex_upper8 ((state[1] >> 16) & 255) << 0
742 | uint_to_hex_upper8 ((state[1] >> 24) & 255) << 16;
743 const u32 w4_t = uint_to_hex_upper8 ((state[2] >> 0) & 255) << 0
744 | uint_to_hex_upper8 ((state[2] >> 8) & 255) << 16;
745 const u32 w5_t = uint_to_hex_upper8 ((state[2] >> 16) & 255) << 0
746 | uint_to_hex_upper8 ((state[2] >> 24) & 255) << 16;
747 const u32 w6_t = uint_to_hex_upper8 ((state[3] >> 0) & 255) << 0
748 | uint_to_hex_upper8 ((state[3] >> 8) & 255) << 16;
750 const u32 pade = 0x0e0e0e0e;
753 w_tmp[ 1] = salt1 | w0_t << 16;
754 w_tmp[ 2] = w0_t >> 16 | w1_t << 16;
755 w_tmp[ 3] = w1_t >> 16 | w2_t << 16;
756 w_tmp[ 4] = w2_t >> 16 | w3_t << 16;
757 w_tmp[ 5] = w3_t >> 16 | w4_t << 16;
758 w_tmp[ 6] = w4_t >> 16 | w5_t << 16;
759 w_tmp[ 7] = w5_t >> 16 | w6_t << 16;
760 w_tmp[ 8] = w6_t >> 16 | pade << 16;
774 domino_big_md (w_tmp, 34, state, s_lotus_magic_table);
788 lotus6_base64_encode (base64_hash, salt_buf0[0], salt_buf0[1], a, b, c);
791 * PBKDF2 - HMACSHA1 - 1st iteration
799 w0[0] = (base64_hash[ 0] << 24) | (base64_hash[ 1] << 16) | (base64_hash[ 2] << 8) | base64_hash[ 3];
800 w0[1] = (base64_hash[ 4] << 24) | (base64_hash[ 5] << 16) | (base64_hash[ 6] << 8) | base64_hash[ 7];
801 w0[2] = (base64_hash[ 8] << 24) | (base64_hash[ 9] << 16) | (base64_hash[10] << 8) | base64_hash[11];
802 w0[3] = (base64_hash[12] << 24) | (base64_hash[13] << 16) | (base64_hash[14] << 8) | base64_hash[15];
803 w1[0] = (base64_hash[16] << 24) | (base64_hash[17] << 16) | (base64_hash[18] << 8) | base64_hash[19];
804 w1[1] = (base64_hash[20] << 24) | (base64_hash[21] << 16);
823 hmac_sha1_pad (w0, w1, w2, w3, ipad, opad);
825 tmps[gid].ipad[0] = ipad[0];
826 tmps[gid].ipad[1] = ipad[1];
827 tmps[gid].ipad[2] = ipad[2];
828 tmps[gid].ipad[3] = ipad[3];
829 tmps[gid].ipad[4] = ipad[4];
831 tmps[gid].opad[0] = opad[0];
832 tmps[gid].opad[1] = opad[1];
833 tmps[gid].opad[2] = opad[2];
834 tmps[gid].opad[3] = opad[3];
835 tmps[gid].opad[4] = opad[4];
837 w0[0] = salt_buf0[0];
838 w0[1] = salt_buf0[1];
839 w0[2] = salt_buf0[2];
840 w0[3] = salt_buf0[3];
841 w1[0] = salt_buf1[0];
842 w1[1] = salt_buf1[1];
843 w1[2] = salt_buf1[2];
844 w1[3] = salt_buf1[3];
845 w2[0] = salt_buf2[0];
846 w2[1] = salt_buf2[1];
847 w2[2] = salt_buf2[2];
848 w2[3] = salt_buf2[3];
849 w3[0] = salt_buf3[0];
850 w3[1] = salt_buf3[1];
851 w3[2] = salt_buf3[2];
852 //w3[3] = salt_buf3[3];
854 w0[0] = swap32 (w0[0]);
855 w0[1] = swap32 (w0[1]);
856 w0[2] = swap32 (w0[2]);
857 w0[3] = swap32 (w0[3]);
858 w1[0] = swap32 (w1[0]);
859 w1[1] = swap32 (w1[1]);
860 w1[2] = swap32 (w1[2]);
861 w1[3] = swap32 (w1[3]);
862 w2[0] = swap32 (w2[0]);
863 w2[1] = swap32 (w2[1]);
864 w2[2] = swap32 (w2[2]);
865 w2[3] = swap32 (w2[3]);
866 w3[0] = swap32 (w3[0]);
867 w3[1] = swap32 (w3[1]);
868 w3[2] = swap32 (w3[2]);
869 w3[3] = (64 + salt_len + 4) * 8;
873 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
875 tmps[gid].dgst[0] = dgst[0];
876 tmps[gid].dgst[1] = dgst[1];
877 tmps[gid].dgst[2] = dgst[2];
878 tmps[gid].dgst[3] = dgst[3];
879 tmps[gid].dgst[4] = dgst[4];
881 tmps[gid].out[0] = dgst[0];
882 tmps[gid].out[1] = dgst[1];
883 tmps[gid].out[2] = dgst[2];
884 tmps[gid].out[3] = dgst[3];
885 tmps[gid].out[4] = dgst[4];
888 __kernel void m09100_loop (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global lotus8_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 wpa_t *wpa_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
890 const u32 gid = get_global_id (0);
892 if (gid >= gid_max) return;
897 ipad[0] = tmps[gid].ipad[0];
898 ipad[1] = tmps[gid].ipad[1];
899 ipad[2] = tmps[gid].ipad[2];
900 ipad[3] = tmps[gid].ipad[3];
901 ipad[4] = tmps[gid].ipad[4];
903 opad[0] = tmps[gid].opad[0];
904 opad[1] = tmps[gid].opad[1];
905 opad[2] = tmps[gid].opad[2];
906 opad[3] = tmps[gid].opad[3];
907 opad[4] = tmps[gid].opad[4];
912 dgst[0] = tmps[gid].dgst[0];
913 dgst[1] = tmps[gid].dgst[1];
914 dgst[2] = tmps[gid].dgst[2];
915 dgst[3] = tmps[gid].dgst[3];
916 dgst[4] = tmps[gid].dgst[4];
918 out[0] = tmps[gid].out[0];
919 out[1] = tmps[gid].out[1];
920 out[2] = tmps[gid].out[2];
921 out[3] = tmps[gid].out[3];
922 out[4] = tmps[gid].out[4];
924 for (u32 j = 0; j < loop_cnt; j++)
946 w3[3] = (64 + 20) * 8;
948 hmac_sha1_run (w0, w1, w2, w3, ipad, opad, dgst);
957 tmps[gid].dgst[0] = dgst[0];
958 tmps[gid].dgst[1] = dgst[1];
959 tmps[gid].dgst[2] = dgst[2];
960 tmps[gid].dgst[3] = dgst[3];
961 tmps[gid].dgst[4] = dgst[4];
963 tmps[gid].out[0] = out[0];
964 tmps[gid].out[1] = out[1];
965 tmps[gid].out[2] = out[2];
966 tmps[gid].out[3] = out[3];
967 tmps[gid].out[4] = out[4];
970 __kernel void m09100_comp (__global pw_t *pws, __global kernel_rule_t *rules_buf, __global comb_t *combs_buf, __global bf_t *bfs_buf, __global lotus8_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 wpa_t *wpa_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
976 const u32 gid = get_global_id (0);
978 if (gid >= gid_max) return;
980 const u32 lid = get_local_id (0);
986 const u32 r0 = tmps[gid].out[DGST_R0];
987 const u32 r1 = tmps[gid].out[DGST_R1];