2 * Author......: Jens Steube <jens.steube@gmail.com>
10 #include "inc_vendor.cl"
11 #include "inc_hash_constants.h"
12 #include "inc_hash_functions.cl"
13 #include "inc_types.cl"
14 #include "inc_common.cl"
15 #include "inc_simd.cl"
18 void md5_transform (const u32x w0[4], const u32x w1[4], const u32x w2[4], const u32x w3[4], u32x digest[4])
42 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
43 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
44 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
45 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
46 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
47 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
48 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
49 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
50 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
51 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
52 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
53 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
54 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
55 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
56 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
57 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
59 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
60 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
61 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
62 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
63 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
64 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
65 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
66 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
67 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
68 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
69 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
70 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
71 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
72 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
73 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
74 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
76 MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
77 MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
78 MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
79 MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
80 MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
81 MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
82 MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
83 MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
84 MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
85 MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
86 MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
87 MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
88 MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
89 MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
90 MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
91 MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
93 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
94 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
95 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
96 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
97 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
98 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
99 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
100 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
101 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
102 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
103 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
104 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
105 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
106 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
107 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
108 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
116 void hmac_md5_pad (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4])
118 w0[0] = w0[0] ^ 0x36363636;
119 w0[1] = w0[1] ^ 0x36363636;
120 w0[2] = w0[2] ^ 0x36363636;
121 w0[3] = w0[3] ^ 0x36363636;
122 w1[0] = w1[0] ^ 0x36363636;
123 w1[1] = w1[1] ^ 0x36363636;
124 w1[2] = w1[2] ^ 0x36363636;
125 w1[3] = w1[3] ^ 0x36363636;
126 w2[0] = w2[0] ^ 0x36363636;
127 w2[1] = w2[1] ^ 0x36363636;
128 w2[2] = w2[2] ^ 0x36363636;
129 w2[3] = w2[3] ^ 0x36363636;
130 w3[0] = w3[0] ^ 0x36363636;
131 w3[1] = w3[1] ^ 0x36363636;
132 w3[2] = w3[2] ^ 0x36363636;
133 w3[3] = w3[3] ^ 0x36363636;
140 md5_transform (w0, w1, w2, w3, ipad);
142 w0[0] = w0[0] ^ 0x6a6a6a6a;
143 w0[1] = w0[1] ^ 0x6a6a6a6a;
144 w0[2] = w0[2] ^ 0x6a6a6a6a;
145 w0[3] = w0[3] ^ 0x6a6a6a6a;
146 w1[0] = w1[0] ^ 0x6a6a6a6a;
147 w1[1] = w1[1] ^ 0x6a6a6a6a;
148 w1[2] = w1[2] ^ 0x6a6a6a6a;
149 w1[3] = w1[3] ^ 0x6a6a6a6a;
150 w2[0] = w2[0] ^ 0x6a6a6a6a;
151 w2[1] = w2[1] ^ 0x6a6a6a6a;
152 w2[2] = w2[2] ^ 0x6a6a6a6a;
153 w2[3] = w2[3] ^ 0x6a6a6a6a;
154 w3[0] = w3[0] ^ 0x6a6a6a6a;
155 w3[1] = w3[1] ^ 0x6a6a6a6a;
156 w3[2] = w3[2] ^ 0x6a6a6a6a;
157 w3[3] = w3[3] ^ 0x6a6a6a6a;
164 md5_transform (w0, w1, w2, w3, opad);
167 void hmac_md5_run (u32x w0[4], u32x w1[4], u32x w2[4], u32x w3[4], u32x ipad[4], u32x opad[4], u32x digest[4])
174 md5_transform (w0, w1, w2, w3, digest);
190 w3[2] = (64 + 16) * 8;
198 md5_transform (w0, w1, w2, w3, digest);
201 __kernel void m00060_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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)
207 const u32 lid = get_local_id (0);
213 const u32 gid = get_global_id (0);
215 if (gid >= gid_max) return;
220 pw_buf0[0] = pws[gid].i[0];
221 pw_buf0[1] = pws[gid].i[1];
222 pw_buf0[2] = pws[gid].i[2];
223 pw_buf0[3] = pws[gid].i[3];
224 pw_buf1[0] = pws[gid].i[4];
225 pw_buf1[1] = pws[gid].i[5];
226 pw_buf1[2] = pws[gid].i[6];
227 pw_buf1[3] = pws[gid].i[7];
229 const u32 pw_l_len = pws[gid].pw_len;
240 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
241 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
242 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
243 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
244 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
245 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
246 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
247 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
248 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
249 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
250 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
251 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
252 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
253 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
254 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
255 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
257 const u32 salt_len = salt_bufs[salt_pos].salt_len;
268 w0_t[0] = salt_buf0[0];
269 w0_t[1] = salt_buf0[1];
270 w0_t[2] = salt_buf0[2];
271 w0_t[3] = salt_buf0[3];
272 w1_t[0] = salt_buf1[0];
273 w1_t[1] = salt_buf1[1];
274 w1_t[2] = salt_buf1[2];
275 w1_t[3] = salt_buf1[3];
276 w2_t[0] = salt_buf2[0];
277 w2_t[1] = salt_buf2[1];
278 w2_t[2] = salt_buf2[2];
279 w2_t[3] = salt_buf2[3];
280 w3_t[0] = salt_buf3[0];
281 w3_t[1] = salt_buf3[1];
282 w3_t[2] = salt_buf3[2];
283 w3_t[3] = salt_buf3[3];
288 hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
294 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
296 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
298 const u32x pw_len = pw_l_len + pw_r_len;
301 * concat password candidate
304 u32x wordl0[4] = { 0 };
305 u32x wordl1[4] = { 0 };
306 u32x wordl2[4] = { 0 };
307 u32x wordl3[4] = { 0 };
309 wordl0[0] = pw_buf0[0];
310 wordl0[1] = pw_buf0[1];
311 wordl0[2] = pw_buf0[2];
312 wordl0[3] = pw_buf0[3];
313 wordl1[0] = pw_buf1[0];
314 wordl1[1] = pw_buf1[1];
315 wordl1[2] = pw_buf1[2];
316 wordl1[3] = pw_buf1[3];
318 u32x wordr0[4] = { 0 };
319 u32x wordr1[4] = { 0 };
320 u32x wordr2[4] = { 0 };
321 u32x wordr3[4] = { 0 };
323 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
324 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
325 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
326 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
327 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
328 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
329 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
330 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
332 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
334 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
338 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
346 w0[0] = wordl0[0] | wordr0[0];
347 w0[1] = wordl0[1] | wordr0[1];
348 w0[2] = wordl0[2] | wordr0[2];
349 w0[3] = wordl0[3] | wordr0[3];
350 w1[0] = wordl1[0] | wordr1[0];
351 w1[1] = wordl1[1] | wordr1[1];
352 w1[2] = wordl1[2] | wordr1[2];
353 w1[3] = wordl1[3] | wordr1[3];
355 pw_len = mangle(w0, w1, pw_len);
356 append_0x80_2x4_VV (w0, w1, pw_len);
365 //w2[0] = wordl2[0] | wordr2[0];
366 //w2[1] = wordl2[1] | wordr2[1];
367 //w2[2] = wordl2[2] | wordr2[2];
368 //w2[3] = wordl2[3] | wordr2[3];
369 //w3[0] = wordl3[0] | wordr3[0];
370 //w3[1] = wordl3[1] | wordr3[1];
372 w3[2] = (64 + pw_len) * 8;
377 hmac_md5_run (w0, w1, w2, w3, ipad, opad, digest);
379 COMPARE_M_SIMD (digest[0], digest[3], digest[2], digest[1]);
383 __kernel void m00060_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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)
387 __kernel void m00060_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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)
391 __kernel void m00060_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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)
397 const u32 lid = get_local_id (0);
403 const u32 gid = get_global_id (0);
405 if (gid >= gid_max) return;
410 pw_buf0[0] = pws[gid].i[0];
411 pw_buf0[1] = pws[gid].i[1];
412 pw_buf0[2] = pws[gid].i[2];
413 pw_buf0[3] = pws[gid].i[3];
414 pw_buf1[0] = pws[gid].i[4];
415 pw_buf1[1] = pws[gid].i[5];
416 pw_buf1[2] = pws[gid].i[6];
417 pw_buf1[3] = pws[gid].i[7];
419 const u32 pw_l_len = pws[gid].pw_len;
430 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
431 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
432 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
433 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
434 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
435 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
436 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
437 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
438 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
439 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
440 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
441 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
442 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
443 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
444 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
445 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
447 const u32 salt_len = salt_bufs[salt_pos].salt_len;
458 w0_t[0] = salt_buf0[0];
459 w0_t[1] = salt_buf0[1];
460 w0_t[2] = salt_buf0[2];
461 w0_t[3] = salt_buf0[3];
462 w1_t[0] = salt_buf1[0];
463 w1_t[1] = salt_buf1[1];
464 w1_t[2] = salt_buf1[2];
465 w1_t[3] = salt_buf1[3];
466 w2_t[0] = salt_buf2[0];
467 w2_t[1] = salt_buf2[1];
468 w2_t[2] = salt_buf2[2];
469 w2_t[3] = salt_buf2[3];
470 w3_t[0] = salt_buf3[0];
471 w3_t[1] = salt_buf3[1];
472 w3_t[2] = salt_buf3[2];
473 w3_t[3] = salt_buf3[3];
478 hmac_md5_pad (w0_t, w1_t, w2_t, w3_t, ipad, opad);
484 const u32 search[4] =
486 digests_buf[digests_offset].digest_buf[DGST_R0],
487 digests_buf[digests_offset].digest_buf[DGST_R1],
488 digests_buf[digests_offset].digest_buf[DGST_R2],
489 digests_buf[digests_offset].digest_buf[DGST_R3]
496 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
498 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
500 const u32x pw_len = pw_l_len + pw_r_len;
503 * concat password candidate
506 u32x wordl0[4] = { 0 };
507 u32x wordl1[4] = { 0 };
508 u32x wordl2[4] = { 0 };
509 u32x wordl3[4] = { 0 };
511 wordl0[0] = pw_buf0[0];
512 wordl0[1] = pw_buf0[1];
513 wordl0[2] = pw_buf0[2];
514 wordl0[3] = pw_buf0[3];
515 wordl1[0] = pw_buf1[0];
516 wordl1[1] = pw_buf1[1];
517 wordl1[2] = pw_buf1[2];
518 wordl1[3] = pw_buf1[3];
520 u32x wordr0[4] = { 0 };
521 u32x wordr1[4] = { 0 };
522 u32x wordr2[4] = { 0 };
523 u32x wordr3[4] = { 0 };
525 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
526 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
527 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
528 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
529 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
530 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
531 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
532 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
534 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
536 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
540 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
548 w0[0] = wordl0[0] | wordr0[0];
549 w0[1] = wordl0[1] | wordr0[1];
550 w0[2] = wordl0[2] | wordr0[2];
551 w0[3] = wordl0[3] | wordr0[3];
552 w1[0] = wordl1[0] | wordr1[0];
553 w1[1] = wordl1[1] | wordr1[1];
554 w1[2] = wordl1[2] | wordr1[2];
555 w1[3] = wordl1[3] | wordr1[3];
557 pw_len = mangle(w0, w1, pw_len);
558 append_0x80_2x4_VV (w0, w1, pw_len);
567 //w2[0] = wordl2[0] | wordr2[0];
568 //w2[1] = wordl2[1] | wordr2[1];
569 //w2[2] = wordl2[2] | wordr2[2];
570 //w2[3] = wordl2[3] | wordr2[3];
571 //w3[0] = wordl3[0] | wordr3[0];
572 //w3[1] = wordl3[1] | wordr3[1];
574 w3[2] = (64 + pw_len) * 8;
579 hmac_md5_run (w0, w1, w2, w3, ipad, opad, digest);
581 COMPARE_S_SIMD (digest[0], digest[3], digest[2], digest[1]);
585 __kernel void m00060_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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)
589 __kernel void m00060_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_scryptV0_buf, __global u32 *d_scryptV1_buf, __global u32 *d_scryptV2_buf, __global u32 *d_scryptV3_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)