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_S "check_single_vect1_comp4.c"
34 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
38 #define VECT_COMPARE_S "check_single_vect2_comp4.c"
39 #define VECT_COMPARE_M "check_multi_vect2_comp4.c"
43 #define VECT_COMPARE_S "check_single_vect4_comp4.c"
44 #define VECT_COMPARE_M "check_multi_vect4_comp4.c"
47 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01000_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
53 const u32 lid = get_local_id (0);
59 const u32 gid = get_global_id (0);
61 if (gid >= gid_max) return;
65 wordl0[0] = pws[gid].i[ 0];
66 wordl0[1] = pws[gid].i[ 1];
67 wordl0[2] = pws[gid].i[ 2];
68 wordl0[3] = pws[gid].i[ 3];
72 wordl1[0] = pws[gid].i[ 4];
73 wordl1[1] = pws[gid].i[ 5];
74 wordl1[2] = pws[gid].i[ 6];
75 wordl1[3] = pws[gid].i[ 7];
91 const u32 pw_l_len = pws[gid].pw_len;
93 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
95 append_0x80_2 (wordl0, wordl1, pw_l_len);
97 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
104 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
106 const u32 pw_r_len = combs_buf[il_pos].pw_len;
108 const u32 pw_len = pw_l_len + pw_r_len;
112 wordr0[0] = combs_buf[il_pos].i[0];
113 wordr0[1] = combs_buf[il_pos].i[1];
114 wordr0[2] = combs_buf[il_pos].i[2];
115 wordr0[3] = combs_buf[il_pos].i[3];
119 wordr1[0] = combs_buf[il_pos].i[4];
120 wordr1[1] = combs_buf[il_pos].i[5];
121 wordr1[2] = combs_buf[il_pos].i[6];
122 wordr1[3] = combs_buf[il_pos].i[7];
138 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
140 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
145 w0[0] = wordl0[0] | wordr0[0];
146 w0[1] = wordl0[1] | wordr0[1];
147 w0[2] = wordl0[2] | wordr0[2];
148 w0[3] = wordl0[3] | wordr0[3];
152 w1[0] = wordl1[0] | wordr1[0];
153 w1[1] = wordl1[1] | wordr1[1];
154 w1[2] = wordl1[2] | wordr1[2];
155 w1[3] = wordl1[3] | wordr1[3];
176 make_unicode (w0, w0_t, w1_t);
177 make_unicode (w1, w2_t, w3_t);
179 w3_t[2] = pw_len * 8 * 2;
188 MD4_STEP (MD4_Fo, a, b, c, d, w0_t[0], MD4C00, MD4S00);
189 MD4_STEP (MD4_Fo, d, a, b, c, w0_t[1], MD4C00, MD4S01);
190 MD4_STEP (MD4_Fo, c, d, a, b, w0_t[2], MD4C00, MD4S02);
191 MD4_STEP (MD4_Fo, b, c, d, a, w0_t[3], MD4C00, MD4S03);
192 MD4_STEP (MD4_Fo, a, b, c, d, w1_t[0], MD4C00, MD4S00);
193 MD4_STEP (MD4_Fo, d, a, b, c, w1_t[1], MD4C00, MD4S01);
194 MD4_STEP (MD4_Fo, c, d, a, b, w1_t[2], MD4C00, MD4S02);
195 MD4_STEP (MD4_Fo, b, c, d, a, w1_t[3], MD4C00, MD4S03);
196 MD4_STEP (MD4_Fo, a, b, c, d, w2_t[0], MD4C00, MD4S00);
197 MD4_STEP (MD4_Fo, d, a, b, c, w2_t[1], MD4C00, MD4S01);
198 MD4_STEP (MD4_Fo, c, d, a, b, w2_t[2], MD4C00, MD4S02);
199 MD4_STEP (MD4_Fo, b, c, d, a, w2_t[3], MD4C00, MD4S03);
200 MD4_STEP (MD4_Fo, a, b, c, d, w3_t[0], MD4C00, MD4S00);
201 MD4_STEP (MD4_Fo, d, a, b, c, w3_t[1], MD4C00, MD4S01);
202 MD4_STEP (MD4_Fo, c, d, a, b, w3_t[2], MD4C00, MD4S02);
203 MD4_STEP (MD4_Fo, b, c, d, a, w3_t[3], MD4C00, MD4S03);
205 MD4_STEP (MD4_Go, a, b, c, d, w0_t[0], MD4C01, MD4S10);
206 MD4_STEP (MD4_Go, d, a, b, c, w1_t[0], MD4C01, MD4S11);
207 MD4_STEP (MD4_Go, c, d, a, b, w2_t[0], MD4C01, MD4S12);
208 MD4_STEP (MD4_Go, b, c, d, a, w3_t[0], MD4C01, MD4S13);
209 MD4_STEP (MD4_Go, a, b, c, d, w0_t[1], MD4C01, MD4S10);
210 MD4_STEP (MD4_Go, d, a, b, c, w1_t[1], MD4C01, MD4S11);
211 MD4_STEP (MD4_Go, c, d, a, b, w2_t[1], MD4C01, MD4S12);
212 MD4_STEP (MD4_Go, b, c, d, a, w3_t[1], MD4C01, MD4S13);
213 MD4_STEP (MD4_Go, a, b, c, d, w0_t[2], MD4C01, MD4S10);
214 MD4_STEP (MD4_Go, d, a, b, c, w1_t[2], MD4C01, MD4S11);
215 MD4_STEP (MD4_Go, c, d, a, b, w2_t[2], MD4C01, MD4S12);
216 MD4_STEP (MD4_Go, b, c, d, a, w3_t[2], MD4C01, MD4S13);
217 MD4_STEP (MD4_Go, a, b, c, d, w0_t[3], MD4C01, MD4S10);
218 MD4_STEP (MD4_Go, d, a, b, c, w1_t[3], MD4C01, MD4S11);
219 MD4_STEP (MD4_Go, c, d, a, b, w2_t[3], MD4C01, MD4S12);
220 MD4_STEP (MD4_Go, b, c, d, a, w3_t[3], MD4C01, MD4S13);
222 MD4_STEP (MD4_H1, a, b, c, d, w0_t[0], MD4C02, MD4S20);
223 MD4_STEP (MD4_H2, d, a, b, c, w2_t[0], MD4C02, MD4S21);
224 MD4_STEP (MD4_H1, c, d, a, b, w1_t[0], MD4C02, MD4S22);
225 MD4_STEP (MD4_H2, b, c, d, a, w3_t[0], MD4C02, MD4S23);
226 MD4_STEP (MD4_H1, a, b, c, d, w0_t[2], MD4C02, MD4S20);
227 MD4_STEP (MD4_H2, d, a, b, c, w2_t[2], MD4C02, MD4S21);
228 MD4_STEP (MD4_H1, c, d, a, b, w1_t[2], MD4C02, MD4S22);
229 MD4_STEP (MD4_H2, b, c, d, a, w3_t[2], MD4C02, MD4S23);
230 MD4_STEP (MD4_H1, a, b, c, d, w0_t[1], MD4C02, MD4S20);
231 MD4_STEP (MD4_H2, d, a, b, c, w2_t[1], MD4C02, MD4S21);
232 MD4_STEP (MD4_H1, c, d, a, b, w1_t[1], MD4C02, MD4S22);
233 MD4_STEP (MD4_H2, b, c, d, a, w3_t[1], MD4C02, MD4S23);
234 MD4_STEP (MD4_H1, a, b, c, d, w0_t[3], MD4C02, MD4S20);
235 MD4_STEP (MD4_H2, d, a, b, c, w2_t[3], MD4C02, MD4S21);
236 MD4_STEP (MD4_H1, c, d, a, b, w1_t[3], MD4C02, MD4S22);
237 MD4_STEP (MD4_H2, b, c, d, a, w3_t[3], MD4C02, MD4S23);
244 #include VECT_COMPARE_M
248 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01000_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
252 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01000_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
256 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01000_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
262 const u32 lid = get_local_id (0);
268 const u32 gid = get_global_id (0);
270 if (gid >= gid_max) return;
274 wordl0[0] = pws[gid].i[ 0];
275 wordl0[1] = pws[gid].i[ 1];
276 wordl0[2] = pws[gid].i[ 2];
277 wordl0[3] = pws[gid].i[ 3];
281 wordl1[0] = pws[gid].i[ 4];
282 wordl1[1] = pws[gid].i[ 5];
283 wordl1[2] = pws[gid].i[ 6];
284 wordl1[3] = pws[gid].i[ 7];
300 const u32 pw_l_len = pws[gid].pw_len;
302 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
304 append_0x80_2 (wordl0, wordl1, pw_l_len);
306 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
313 const u32 search[4] =
315 digests_buf[digests_offset].digest_buf[DGST_R0],
316 digests_buf[digests_offset].digest_buf[DGST_R1],
317 digests_buf[digests_offset].digest_buf[DGST_R2],
318 digests_buf[digests_offset].digest_buf[DGST_R3]
325 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
327 const u32 pw_r_len = combs_buf[il_pos].pw_len;
329 const u32 pw_len = pw_l_len + pw_r_len;
333 wordr0[0] = combs_buf[il_pos].i[0];
334 wordr0[1] = combs_buf[il_pos].i[1];
335 wordr0[2] = combs_buf[il_pos].i[2];
336 wordr0[3] = combs_buf[il_pos].i[3];
340 wordr1[0] = combs_buf[il_pos].i[4];
341 wordr1[1] = combs_buf[il_pos].i[5];
342 wordr1[2] = combs_buf[il_pos].i[6];
343 wordr1[3] = combs_buf[il_pos].i[7];
359 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
361 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
366 w0[0] = wordl0[0] | wordr0[0];
367 w0[1] = wordl0[1] | wordr0[1];
368 w0[2] = wordl0[2] | wordr0[2];
369 w0[3] = wordl0[3] | wordr0[3];
373 w1[0] = wordl1[0] | wordr1[0];
374 w1[1] = wordl1[1] | wordr1[1];
375 w1[2] = wordl1[2] | wordr1[2];
376 w1[3] = wordl1[3] | wordr1[3];
397 make_unicode (w0, w0_t, w1_t);
398 make_unicode (w1, w2_t, w3_t);
400 w3_t[2] = pw_len * 8 * 2;
409 MD4_STEP (MD4_Fo, a, b, c, d, w0_t[0], MD4C00, MD4S00);
410 MD4_STEP (MD4_Fo, d, a, b, c, w0_t[1], MD4C00, MD4S01);
411 MD4_STEP (MD4_Fo, c, d, a, b, w0_t[2], MD4C00, MD4S02);
412 MD4_STEP (MD4_Fo, b, c, d, a, w0_t[3], MD4C00, MD4S03);
413 MD4_STEP (MD4_Fo, a, b, c, d, w1_t[0], MD4C00, MD4S00);
414 MD4_STEP (MD4_Fo, d, a, b, c, w1_t[1], MD4C00, MD4S01);
415 MD4_STEP (MD4_Fo, c, d, a, b, w1_t[2], MD4C00, MD4S02);
416 MD4_STEP (MD4_Fo, b, c, d, a, w1_t[3], MD4C00, MD4S03);
417 MD4_STEP (MD4_Fo, a, b, c, d, w2_t[0], MD4C00, MD4S00);
418 MD4_STEP (MD4_Fo, d, a, b, c, w2_t[1], MD4C00, MD4S01);
419 MD4_STEP (MD4_Fo, c, d, a, b, w2_t[2], MD4C00, MD4S02);
420 MD4_STEP (MD4_Fo, b, c, d, a, w2_t[3], MD4C00, MD4S03);
421 MD4_STEP (MD4_Fo, a, b, c, d, w3_t[0], MD4C00, MD4S00);
422 MD4_STEP (MD4_Fo, d, a, b, c, w3_t[1], MD4C00, MD4S01);
423 MD4_STEP (MD4_Fo, c, d, a, b, w3_t[2], MD4C00, MD4S02);
424 MD4_STEP (MD4_Fo, b, c, d, a, w3_t[3], MD4C00, MD4S03);
426 MD4_STEP (MD4_Go, a, b, c, d, w0_t[0], MD4C01, MD4S10);
427 MD4_STEP (MD4_Go, d, a, b, c, w1_t[0], MD4C01, MD4S11);
428 MD4_STEP (MD4_Go, c, d, a, b, w2_t[0], MD4C01, MD4S12);
429 MD4_STEP (MD4_Go, b, c, d, a, w3_t[0], MD4C01, MD4S13);
430 MD4_STEP (MD4_Go, a, b, c, d, w0_t[1], MD4C01, MD4S10);
431 MD4_STEP (MD4_Go, d, a, b, c, w1_t[1], MD4C01, MD4S11);
432 MD4_STEP (MD4_Go, c, d, a, b, w2_t[1], MD4C01, MD4S12);
433 MD4_STEP (MD4_Go, b, c, d, a, w3_t[1], MD4C01, MD4S13);
434 MD4_STEP (MD4_Go, a, b, c, d, w0_t[2], MD4C01, MD4S10);
435 MD4_STEP (MD4_Go, d, a, b, c, w1_t[2], MD4C01, MD4S11);
436 MD4_STEP (MD4_Go, c, d, a, b, w2_t[2], MD4C01, MD4S12);
437 MD4_STEP (MD4_Go, b, c, d, a, w3_t[2], MD4C01, MD4S13);
438 MD4_STEP (MD4_Go, a, b, c, d, w0_t[3], MD4C01, MD4S10);
439 MD4_STEP (MD4_Go, d, a, b, c, w1_t[3], MD4C01, MD4S11);
440 MD4_STEP (MD4_Go, c, d, a, b, w2_t[3], MD4C01, MD4S12);
441 MD4_STEP (MD4_Go, b, c, d, a, w3_t[3], MD4C01, MD4S13);
443 MD4_STEP (MD4_H1, a, b, c, d, w0_t[0], MD4C02, MD4S20);
444 MD4_STEP (MD4_H2, d, a, b, c, w2_t[0], MD4C02, MD4S21);
445 MD4_STEP (MD4_H1, c, d, a, b, w1_t[0], MD4C02, MD4S22);
446 MD4_STEP (MD4_H2, b, c, d, a, w3_t[0], MD4C02, MD4S23);
447 MD4_STEP (MD4_H1, a, b, c, d, w0_t[2], MD4C02, MD4S20);
448 MD4_STEP (MD4_H2, d, a, b, c, w2_t[2], MD4C02, MD4S21);
449 MD4_STEP (MD4_H1, c, d, a, b, w1_t[2], MD4C02, MD4S22);
450 MD4_STEP (MD4_H2, b, c, d, a, w3_t[2], MD4C02, MD4S23);
451 MD4_STEP (MD4_H1, a, b, c, d, w0_t[1], MD4C02, MD4S20);
452 MD4_STEP (MD4_H2, d, a, b, c, w2_t[1], MD4C02, MD4S21);
453 MD4_STEP (MD4_H1, c, d, a, b, w1_t[1], MD4C02, MD4S22);
454 MD4_STEP (MD4_H2, b, c, d, a, w3_t[1], MD4C02, MD4S23);
455 MD4_STEP (MD4_H1, a, b, c, d, w0_t[3], MD4C02, MD4S20);
456 MD4_STEP (MD4_H2, d, a, b, c, w2_t[3], MD4C02, MD4S21);
457 MD4_STEP (MD4_H1, c, d, a, b, w1_t[3], MD4C02, MD4S22);
458 MD4_STEP (MD4_H2, b, c, d, a, w3_t[3], MD4C02, MD4S23);
465 #include VECT_COMPARE_S
469 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01000_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
473 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m01000_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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)