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"
17 __kernel void m00020_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)
23 const u32 lid = get_local_id (0);
29 const u32 gid = get_global_id (0);
31 if (gid >= gid_max) return;
36 pw_buf0[0] = pws[gid].i[0];
37 pw_buf0[1] = pws[gid].i[1];
38 pw_buf0[2] = pws[gid].i[2];
39 pw_buf0[3] = pws[gid].i[3];
40 pw_buf1[0] = pws[gid].i[4];
41 pw_buf1[1] = pws[gid].i[5];
42 pw_buf1[2] = pws[gid].i[6];
43 pw_buf1[3] = pws[gid].i[7];
45 const u32 pw_l_len = pws[gid].pw_len;
56 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
57 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
58 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
59 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
60 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
61 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
62 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
63 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
64 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
65 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
66 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
67 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
68 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
69 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
70 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
71 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
73 const u32 salt_len = salt_bufs[salt_pos].salt_len;
79 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
81 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
83 const u32x pw_len = pw_l_len + pw_r_len;
86 * concat password candidate
89 u32x wordl0[4] = { 0 };
90 u32x wordl1[4] = { 0 };
91 u32x wordl2[4] = { 0 };
92 u32x wordl3[4] = { 0 };
94 wordl0[0] = pw_buf0[0];
95 wordl0[1] = pw_buf0[1];
96 wordl0[2] = pw_buf0[2];
97 wordl0[3] = pw_buf0[3];
98 wordl1[0] = pw_buf1[0];
99 wordl1[1] = pw_buf1[1];
100 wordl1[2] = pw_buf1[2];
101 wordl1[3] = pw_buf1[3];
103 u32x wordr0[4] = { 0 };
104 u32x wordr1[4] = { 0 };
105 u32x wordr2[4] = { 0 };
106 u32x wordr3[4] = { 0 };
108 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
109 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
110 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
111 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
112 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
113 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
114 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
115 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
117 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
119 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
123 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
131 w0[0] = wordl0[0] | wordr0[0];
132 w0[1] = wordl0[1] | wordr0[1];
133 w0[2] = wordl0[2] | wordr0[2];
134 w0[3] = wordl0[3] | wordr0[3];
135 w1[0] = wordl1[0] | wordr1[0];
136 w1[1] = wordl1[1] | wordr1[1];
137 w1[2] = wordl1[2] | wordr1[2];
138 w1[3] = wordl1[3] | wordr1[3];
139 w2[0] = wordl2[0] | wordr2[0];
140 w2[1] = wordl2[1] | wordr2[1];
141 w2[2] = wordl2[2] | wordr2[2];
142 w2[3] = wordl2[3] | wordr2[3];
143 w3[0] = wordl3[0] | wordr3[0];
144 w3[1] = wordl3[1] | wordr3[1];
145 w3[2] = wordl3[2] | wordr3[2];
146 w3[3] = wordl3[3] | wordr3[3];
152 switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
154 const u32x pw_salt_len = pw_len + salt_len;
156 w0[0] |= salt_buf0[0];
157 w0[1] |= salt_buf0[1];
158 w0[2] |= salt_buf0[2];
159 w0[3] |= salt_buf0[3];
160 w1[0] |= salt_buf1[0];
161 w1[1] |= salt_buf1[1];
162 w1[2] |= salt_buf1[2];
163 w1[3] |= salt_buf1[3];
164 w2[0] |= salt_buf2[0];
165 w2[1] |= salt_buf2[1];
166 w2[2] |= salt_buf2[2];
167 w2[3] |= salt_buf2[3];
168 w3[0] |= salt_buf3[0];
169 w3[1] |= salt_buf3[1];
170 w3[2] = pw_salt_len * 8;
182 MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
183 MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
184 MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
185 MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
186 MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
187 MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
188 MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
189 MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
190 MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
191 MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
192 MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
193 MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
194 MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
195 MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
196 MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
197 MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
199 MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
200 MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
201 MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
202 MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
203 MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
204 MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
205 MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
206 MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
207 MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
208 MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
209 MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
210 MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
211 MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
212 MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
213 MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
214 MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
216 MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
217 MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
218 MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
219 MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
220 MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
221 MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
222 MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
223 MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
224 MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
225 MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
226 MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
227 MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
228 MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
229 MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
230 MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
231 MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
233 MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
234 MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
235 MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
236 MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
237 MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
238 MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
239 MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
240 MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
241 MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
242 MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
243 MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
244 MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
245 MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
246 MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
247 MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
248 MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
250 COMPARE_M_SIMD (a, d, c, b);
254 __kernel void m00020_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)
258 __kernel void m00020_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)
262 __kernel void m00020_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)
268 const u32 lid = get_local_id (0);
274 const u32 gid = get_global_id (0);
276 if (gid >= gid_max) return;
281 pw_buf0[0] = pws[gid].i[0];
282 pw_buf0[1] = pws[gid].i[1];
283 pw_buf0[2] = pws[gid].i[2];
284 pw_buf0[3] = pws[gid].i[3];
285 pw_buf1[0] = pws[gid].i[4];
286 pw_buf1[1] = pws[gid].i[5];
287 pw_buf1[2] = pws[gid].i[6];
288 pw_buf1[3] = pws[gid].i[7];
290 const u32 pw_l_len = pws[gid].pw_len;
301 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[ 0];
302 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[ 1];
303 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[ 2];
304 salt_buf0[3] = salt_bufs[salt_pos].salt_buf[ 3];
305 salt_buf1[0] = salt_bufs[salt_pos].salt_buf[ 4];
306 salt_buf1[1] = salt_bufs[salt_pos].salt_buf[ 5];
307 salt_buf1[2] = salt_bufs[salt_pos].salt_buf[ 6];
308 salt_buf1[3] = salt_bufs[salt_pos].salt_buf[ 7];
309 salt_buf2[0] = salt_bufs[salt_pos].salt_buf[ 8];
310 salt_buf2[1] = salt_bufs[salt_pos].salt_buf[ 9];
311 salt_buf2[2] = salt_bufs[salt_pos].salt_buf[10];
312 salt_buf2[3] = salt_bufs[salt_pos].salt_buf[11];
313 salt_buf3[0] = salt_bufs[salt_pos].salt_buf[12];
314 salt_buf3[1] = salt_bufs[salt_pos].salt_buf[13];
315 salt_buf3[2] = salt_bufs[salt_pos].salt_buf[14];
316 salt_buf3[3] = salt_bufs[salt_pos].salt_buf[15];
318 const u32 salt_len = salt_bufs[salt_pos].salt_len;
324 const u32 search[4] =
326 digests_buf[digests_offset].digest_buf[DGST_R0],
327 digests_buf[digests_offset].digest_buf[DGST_R1],
328 digests_buf[digests_offset].digest_buf[DGST_R2],
329 digests_buf[digests_offset].digest_buf[DGST_R3]
336 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
338 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
340 const u32x pw_len = pw_l_len + pw_r_len;
343 * concat password candidate
346 u32x wordl0[4] = { 0 };
347 u32x wordl1[4] = { 0 };
348 u32x wordl2[4] = { 0 };
349 u32x wordl3[4] = { 0 };
351 wordl0[0] = pw_buf0[0];
352 wordl0[1] = pw_buf0[1];
353 wordl0[2] = pw_buf0[2];
354 wordl0[3] = pw_buf0[3];
355 wordl1[0] = pw_buf1[0];
356 wordl1[1] = pw_buf1[1];
357 wordl1[2] = pw_buf1[2];
358 wordl1[3] = pw_buf1[3];
360 u32x wordr0[4] = { 0 };
361 u32x wordr1[4] = { 0 };
362 u32x wordr2[4] = { 0 };
363 u32x wordr3[4] = { 0 };
365 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
366 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
367 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
368 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
369 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
370 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
371 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
372 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
374 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
376 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
380 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
388 w0[0] = wordl0[0] | wordr0[0];
389 w0[1] = wordl0[1] | wordr0[1];
390 w0[2] = wordl0[2] | wordr0[2];
391 w0[3] = wordl0[3] | wordr0[3];
392 w1[0] = wordl1[0] | wordr1[0];
393 w1[1] = wordl1[1] | wordr1[1];
394 w1[2] = wordl1[2] | wordr1[2];
395 w1[3] = wordl1[3] | wordr1[3];
396 w2[0] = wordl2[0] | wordr2[0];
397 w2[1] = wordl2[1] | wordr2[1];
398 w2[2] = wordl2[2] | wordr2[2];
399 w2[3] = wordl2[3] | wordr2[3];
400 w3[0] = wordl3[0] | wordr3[0];
401 w3[1] = wordl3[1] | wordr3[1];
402 w3[2] = wordl3[2] | wordr3[2];
403 w3[3] = wordl3[3] | wordr3[3];
409 switch_buffer_by_offset_le (w0, w1, w2, w3, salt_len);
411 const u32x pw_salt_len = pw_len + salt_len;
413 w0[0] |= salt_buf0[0];
414 w0[1] |= salt_buf0[1];
415 w0[2] |= salt_buf0[2];
416 w0[3] |= salt_buf0[3];
417 w1[0] |= salt_buf1[0];
418 w1[1] |= salt_buf1[1];
419 w1[2] |= salt_buf1[2];
420 w1[3] |= salt_buf1[3];
421 w2[0] |= salt_buf2[0];
422 w2[1] |= salt_buf2[1];
423 w2[2] |= salt_buf2[2];
424 w2[3] |= salt_buf2[3];
425 w3[0] |= salt_buf3[0];
426 w3[1] |= salt_buf3[1];
427 w3[2] = pw_salt_len * 8;
439 MD5_STEP (MD5_Fo, a, b, c, d, w0[0], MD5C00, MD5S00);
440 MD5_STEP (MD5_Fo, d, a, b, c, w0[1], MD5C01, MD5S01);
441 MD5_STEP (MD5_Fo, c, d, a, b, w0[2], MD5C02, MD5S02);
442 MD5_STEP (MD5_Fo, b, c, d, a, w0[3], MD5C03, MD5S03);
443 MD5_STEP (MD5_Fo, a, b, c, d, w1[0], MD5C04, MD5S00);
444 MD5_STEP (MD5_Fo, d, a, b, c, w1[1], MD5C05, MD5S01);
445 MD5_STEP (MD5_Fo, c, d, a, b, w1[2], MD5C06, MD5S02);
446 MD5_STEP (MD5_Fo, b, c, d, a, w1[3], MD5C07, MD5S03);
447 MD5_STEP (MD5_Fo, a, b, c, d, w2[0], MD5C08, MD5S00);
448 MD5_STEP (MD5_Fo, d, a, b, c, w2[1], MD5C09, MD5S01);
449 MD5_STEP (MD5_Fo, c, d, a, b, w2[2], MD5C0a, MD5S02);
450 MD5_STEP (MD5_Fo, b, c, d, a, w2[3], MD5C0b, MD5S03);
451 MD5_STEP (MD5_Fo, a, b, c, d, w3[0], MD5C0c, MD5S00);
452 MD5_STEP (MD5_Fo, d, a, b, c, w3[1], MD5C0d, MD5S01);
453 MD5_STEP (MD5_Fo, c, d, a, b, w3[2], MD5C0e, MD5S02);
454 MD5_STEP (MD5_Fo, b, c, d, a, w3[3], MD5C0f, MD5S03);
456 MD5_STEP (MD5_Go, a, b, c, d, w0[1], MD5C10, MD5S10);
457 MD5_STEP (MD5_Go, d, a, b, c, w1[2], MD5C11, MD5S11);
458 MD5_STEP (MD5_Go, c, d, a, b, w2[3], MD5C12, MD5S12);
459 MD5_STEP (MD5_Go, b, c, d, a, w0[0], MD5C13, MD5S13);
460 MD5_STEP (MD5_Go, a, b, c, d, w1[1], MD5C14, MD5S10);
461 MD5_STEP (MD5_Go, d, a, b, c, w2[2], MD5C15, MD5S11);
462 MD5_STEP (MD5_Go, c, d, a, b, w3[3], MD5C16, MD5S12);
463 MD5_STEP (MD5_Go, b, c, d, a, w1[0], MD5C17, MD5S13);
464 MD5_STEP (MD5_Go, a, b, c, d, w2[1], MD5C18, MD5S10);
465 MD5_STEP (MD5_Go, d, a, b, c, w3[2], MD5C19, MD5S11);
466 MD5_STEP (MD5_Go, c, d, a, b, w0[3], MD5C1a, MD5S12);
467 MD5_STEP (MD5_Go, b, c, d, a, w2[0], MD5C1b, MD5S13);
468 MD5_STEP (MD5_Go, a, b, c, d, w3[1], MD5C1c, MD5S10);
469 MD5_STEP (MD5_Go, d, a, b, c, w0[2], MD5C1d, MD5S11);
470 MD5_STEP (MD5_Go, c, d, a, b, w1[3], MD5C1e, MD5S12);
471 MD5_STEP (MD5_Go, b, c, d, a, w3[0], MD5C1f, MD5S13);
473 MD5_STEP (MD5_H , a, b, c, d, w1[1], MD5C20, MD5S20);
474 MD5_STEP (MD5_H , d, a, b, c, w2[0], MD5C21, MD5S21);
475 MD5_STEP (MD5_H , c, d, a, b, w2[3], MD5C22, MD5S22);
476 MD5_STEP (MD5_H , b, c, d, a, w3[2], MD5C23, MD5S23);
477 MD5_STEP (MD5_H , a, b, c, d, w0[1], MD5C24, MD5S20);
478 MD5_STEP (MD5_H , d, a, b, c, w1[0], MD5C25, MD5S21);
479 MD5_STEP (MD5_H , c, d, a, b, w1[3], MD5C26, MD5S22);
480 MD5_STEP (MD5_H , b, c, d, a, w2[2], MD5C27, MD5S23);
481 MD5_STEP (MD5_H , a, b, c, d, w3[1], MD5C28, MD5S20);
482 MD5_STEP (MD5_H , d, a, b, c, w0[0], MD5C29, MD5S21);
483 MD5_STEP (MD5_H , c, d, a, b, w0[3], MD5C2a, MD5S22);
484 MD5_STEP (MD5_H , b, c, d, a, w1[2], MD5C2b, MD5S23);
485 MD5_STEP (MD5_H , a, b, c, d, w2[1], MD5C2c, MD5S20);
486 MD5_STEP (MD5_H , d, a, b, c, w3[0], MD5C2d, MD5S21);
487 MD5_STEP (MD5_H , c, d, a, b, w3[3], MD5C2e, MD5S22);
488 MD5_STEP (MD5_H , b, c, d, a, w0[2], MD5C2f, MD5S23);
490 MD5_STEP (MD5_I , a, b, c, d, w0[0], MD5C30, MD5S30);
491 MD5_STEP (MD5_I , d, a, b, c, w1[3], MD5C31, MD5S31);
492 MD5_STEP (MD5_I , c, d, a, b, w3[2], MD5C32, MD5S32);
493 MD5_STEP (MD5_I , b, c, d, a, w1[1], MD5C33, MD5S33);
494 MD5_STEP (MD5_I , a, b, c, d, w3[0], MD5C34, MD5S30);
495 MD5_STEP (MD5_I , d, a, b, c, w0[3], MD5C35, MD5S31);
496 MD5_STEP (MD5_I , c, d, a, b, w2[2], MD5C36, MD5S32);
497 MD5_STEP (MD5_I , b, c, d, a, w0[1], MD5C37, MD5S33);
498 MD5_STEP (MD5_I , a, b, c, d, w2[0], MD5C38, MD5S30);
499 MD5_STEP (MD5_I , d, a, b, c, w3[3], MD5C39, MD5S31);
500 MD5_STEP (MD5_I , c, d, a, b, w1[2], MD5C3a, MD5S32);
501 MD5_STEP (MD5_I , b, c, d, a, w3[1], MD5C3b, MD5S33);
502 MD5_STEP (MD5_I , a, b, c, d, w1[0], MD5C3c, MD5S30);
504 if (MATCHES_NONE_VS (a, search[0])) continue;
506 MD5_STEP (MD5_I , d, a, b, c, w2[3], MD5C3d, MD5S31);
507 MD5_STEP (MD5_I , c, d, a, b, w0[2], MD5C3e, MD5S32);
508 MD5_STEP (MD5_I , b, c, d, a, w2[1], MD5C3f, MD5S33);
510 COMPARE_S_SIMD (a, d, c, b);
514 __kernel void m00020_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)
518 __kernel void m00020_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)