2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
16 #include "include/kernel_functions.c"
17 #include "OpenCL/types_ocl.c"
18 #include "OpenCL/common.c"
19 #include "include/rp_kernel.h"
20 #include "OpenCL/rp.c"
22 #define COMPARE_S "OpenCL/check_single_comp4.c"
23 #define COMPARE_M "OpenCL/check_multi_comp4.c"
25 __constant u32 padding[8] =
45 static void swap (__local RC4_KEY *rc4_key, const u8 i, const u8 j)
50 rc4_key->S[i] = rc4_key->S[j];
54 static void rc4_init_16 (__local RC4_KEY *rc4_key, const u32 data[4])
59 __local u32 *ptr = (__local u32 *) rc4_key->S;
62 for (u32 i = 0; i < 64; i++)
67 const u32 d0 = data[0] >> 0;
68 const u32 d1 = data[0] >> 8;
69 const u32 d2 = data[0] >> 16;
70 const u32 d3 = data[0] >> 24;
71 const u32 d4 = data[1] >> 0;
76 for (u32 i = 0; i < 255; i += 5)
78 j += rc4_key->S[i + 0] + d0; swap (rc4_key, i + 0, j);
79 j += rc4_key->S[i + 1] + d1; swap (rc4_key, i + 1, j);
80 j += rc4_key->S[i + 2] + d2; swap (rc4_key, i + 2, j);
81 j += rc4_key->S[i + 3] + d3; swap (rc4_key, i + 3, j);
82 j += rc4_key->S[i + 4] + d4; swap (rc4_key, i + 4, j);
85 j += rc4_key->S[255] + d0; swap (rc4_key, 255, j);
88 static u8 rc4_next_16 (__local RC4_KEY *rc4_key, u8 i, u8 j, __constant u32 in[4], u32 out[4])
91 for (u32 k = 0; k < 4; k++)
100 swap (rc4_key, i, j);
102 idx = rc4_key->S[i] + rc4_key->S[j];
104 xor4 |= rc4_key->S[idx] << 0;
109 swap (rc4_key, i, j);
111 idx = rc4_key->S[i] + rc4_key->S[j];
113 xor4 |= rc4_key->S[idx] << 8;
118 swap (rc4_key, i, j);
120 idx = rc4_key->S[i] + rc4_key->S[j];
122 xor4 |= rc4_key->S[idx] << 16;
127 swap (rc4_key, i, j);
129 idx = rc4_key->S[i] + rc4_key->S[j];
131 xor4 |= rc4_key->S[idx] << 24;
133 out[k] = in[k] ^ xor4;
139 static void md5_transform (const u32 w0[4], const u32 w1[4], const u32 w2[4], const u32 w3[4], u32 digest[4])
163 MD5_STEP (MD5_Fo, a, b, c, d, w0_t, MD5C00, MD5S00);
164 MD5_STEP (MD5_Fo, d, a, b, c, w1_t, MD5C01, MD5S01);
165 MD5_STEP (MD5_Fo, c, d, a, b, w2_t, MD5C02, MD5S02);
166 MD5_STEP (MD5_Fo, b, c, d, a, w3_t, MD5C03, MD5S03);
167 MD5_STEP (MD5_Fo, a, b, c, d, w4_t, MD5C04, MD5S00);
168 MD5_STEP (MD5_Fo, d, a, b, c, w5_t, MD5C05, MD5S01);
169 MD5_STEP (MD5_Fo, c, d, a, b, w6_t, MD5C06, MD5S02);
170 MD5_STEP (MD5_Fo, b, c, d, a, w7_t, MD5C07, MD5S03);
171 MD5_STEP (MD5_Fo, a, b, c, d, w8_t, MD5C08, MD5S00);
172 MD5_STEP (MD5_Fo, d, a, b, c, w9_t, MD5C09, MD5S01);
173 MD5_STEP (MD5_Fo, c, d, a, b, wa_t, MD5C0a, MD5S02);
174 MD5_STEP (MD5_Fo, b, c, d, a, wb_t, MD5C0b, MD5S03);
175 MD5_STEP (MD5_Fo, a, b, c, d, wc_t, MD5C0c, MD5S00);
176 MD5_STEP (MD5_Fo, d, a, b, c, wd_t, MD5C0d, MD5S01);
177 MD5_STEP (MD5_Fo, c, d, a, b, we_t, MD5C0e, MD5S02);
178 MD5_STEP (MD5_Fo, b, c, d, a, wf_t, MD5C0f, MD5S03);
180 MD5_STEP (MD5_Go, a, b, c, d, w1_t, MD5C10, MD5S10);
181 MD5_STEP (MD5_Go, d, a, b, c, w6_t, MD5C11, MD5S11);
182 MD5_STEP (MD5_Go, c, d, a, b, wb_t, MD5C12, MD5S12);
183 MD5_STEP (MD5_Go, b, c, d, a, w0_t, MD5C13, MD5S13);
184 MD5_STEP (MD5_Go, a, b, c, d, w5_t, MD5C14, MD5S10);
185 MD5_STEP (MD5_Go, d, a, b, c, wa_t, MD5C15, MD5S11);
186 MD5_STEP (MD5_Go, c, d, a, b, wf_t, MD5C16, MD5S12);
187 MD5_STEP (MD5_Go, b, c, d, a, w4_t, MD5C17, MD5S13);
188 MD5_STEP (MD5_Go, a, b, c, d, w9_t, MD5C18, MD5S10);
189 MD5_STEP (MD5_Go, d, a, b, c, we_t, MD5C19, MD5S11);
190 MD5_STEP (MD5_Go, c, d, a, b, w3_t, MD5C1a, MD5S12);
191 MD5_STEP (MD5_Go, b, c, d, a, w8_t, MD5C1b, MD5S13);
192 MD5_STEP (MD5_Go, a, b, c, d, wd_t, MD5C1c, MD5S10);
193 MD5_STEP (MD5_Go, d, a, b, c, w2_t, MD5C1d, MD5S11);
194 MD5_STEP (MD5_Go, c, d, a, b, w7_t, MD5C1e, MD5S12);
195 MD5_STEP (MD5_Go, b, c, d, a, wc_t, MD5C1f, MD5S13);
197 MD5_STEP (MD5_H , a, b, c, d, w5_t, MD5C20, MD5S20);
198 MD5_STEP (MD5_H , d, a, b, c, w8_t, MD5C21, MD5S21);
199 MD5_STEP (MD5_H , c, d, a, b, wb_t, MD5C22, MD5S22);
200 MD5_STEP (MD5_H , b, c, d, a, we_t, MD5C23, MD5S23);
201 MD5_STEP (MD5_H , a, b, c, d, w1_t, MD5C24, MD5S20);
202 MD5_STEP (MD5_H , d, a, b, c, w4_t, MD5C25, MD5S21);
203 MD5_STEP (MD5_H , c, d, a, b, w7_t, MD5C26, MD5S22);
204 MD5_STEP (MD5_H , b, c, d, a, wa_t, MD5C27, MD5S23);
205 MD5_STEP (MD5_H , a, b, c, d, wd_t, MD5C28, MD5S20);
206 MD5_STEP (MD5_H , d, a, b, c, w0_t, MD5C29, MD5S21);
207 MD5_STEP (MD5_H , c, d, a, b, w3_t, MD5C2a, MD5S22);
208 MD5_STEP (MD5_H , b, c, d, a, w6_t, MD5C2b, MD5S23);
209 MD5_STEP (MD5_H , a, b, c, d, w9_t, MD5C2c, MD5S20);
210 MD5_STEP (MD5_H , d, a, b, c, wc_t, MD5C2d, MD5S21);
211 MD5_STEP (MD5_H , c, d, a, b, wf_t, MD5C2e, MD5S22);
212 MD5_STEP (MD5_H , b, c, d, a, w2_t, MD5C2f, MD5S23);
214 MD5_STEP (MD5_I , a, b, c, d, w0_t, MD5C30, MD5S30);
215 MD5_STEP (MD5_I , d, a, b, c, w7_t, MD5C31, MD5S31);
216 MD5_STEP (MD5_I , c, d, a, b, we_t, MD5C32, MD5S32);
217 MD5_STEP (MD5_I , b, c, d, a, w5_t, MD5C33, MD5S33);
218 MD5_STEP (MD5_I , a, b, c, d, wc_t, MD5C34, MD5S30);
219 MD5_STEP (MD5_I , d, a, b, c, w3_t, MD5C35, MD5S31);
220 MD5_STEP (MD5_I , c, d, a, b, wa_t, MD5C36, MD5S32);
221 MD5_STEP (MD5_I , b, c, d, a, w1_t, MD5C37, MD5S33);
222 MD5_STEP (MD5_I , a, b, c, d, w8_t, MD5C38, MD5S30);
223 MD5_STEP (MD5_I , d, a, b, c, wf_t, MD5C39, MD5S31);
224 MD5_STEP (MD5_I , c, d, a, b, w6_t, MD5C3a, MD5S32);
225 MD5_STEP (MD5_I , b, c, d, a, wd_t, MD5C3b, MD5S33);
226 MD5_STEP (MD5_I , a, b, c, d, w4_t, MD5C3c, MD5S30);
227 MD5_STEP (MD5_I , d, a, b, c, wb_t, MD5C3d, MD5S31);
228 MD5_STEP (MD5_I , c, d, a, b, w2_t, MD5C3e, MD5S32);
229 MD5_STEP (MD5_I , b, c, d, a, w9_t, MD5C3f, MD5S33);
237 __kernel void m10400_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 pdf_t *pdf_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)
243 const u32 lid = get_local_id (0);
249 const u32 gid = get_global_id (0);
251 if (gid >= gid_max) return;
255 pw_buf0[0] = pws[gid].i[ 0];
256 pw_buf0[1] = pws[gid].i[ 1];
257 pw_buf0[2] = pws[gid].i[ 2];
258 pw_buf0[3] = pws[gid].i[ 3];
262 pw_buf1[0] = pws[gid].i[ 4];
263 pw_buf1[1] = pws[gid].i[ 5];
264 pw_buf1[2] = pws[gid].i[ 6];
265 pw_buf1[3] = pws[gid].i[ 7];
267 const u32 pw_len = pws[gid].pw_len;
273 __local RC4_KEY rc4_keys[64];
275 __local RC4_KEY *rc4_key = &rc4_keys[lid];
283 o_buf[0] = pdf_bufs[salt_pos].o_buf[0];
284 o_buf[1] = pdf_bufs[salt_pos].o_buf[1];
285 o_buf[2] = pdf_bufs[salt_pos].o_buf[2];
286 o_buf[3] = pdf_bufs[salt_pos].o_buf[3];
287 o_buf[4] = pdf_bufs[salt_pos].o_buf[4];
288 o_buf[5] = pdf_bufs[salt_pos].o_buf[5];
289 o_buf[6] = pdf_bufs[salt_pos].o_buf[6];
290 o_buf[7] = pdf_bufs[salt_pos].o_buf[7];
292 u32 P = pdf_bufs[salt_pos].P;
296 id_buf[0] = pdf_bufs[salt_pos].id_buf[0];
297 id_buf[1] = pdf_bufs[salt_pos].id_buf[1];
298 id_buf[2] = pdf_bufs[salt_pos].id_buf[2];
299 id_buf[3] = pdf_bufs[salt_pos].id_buf[3];
305 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
335 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
342 // max length supported by pdf11 is 32
344 w0_t[0] = padding[0];
345 w0_t[1] = padding[1];
346 w0_t[2] = padding[2];
347 w0_t[3] = padding[3];
348 w1_t[0] = padding[4];
349 w1_t[1] = padding[5];
350 w1_t[2] = padding[6];
351 w1_t[3] = padding[7];
361 switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, pw_len);
364 // truncate at 32 is wanted, not a bug!
391 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
410 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
417 key[1] = digest[1] & 0xff;
421 rc4_init_16 (rc4_key, key);
425 rc4_next_16 (rc4_key, 0, 0, padding, out);
427 const u32 r0 = out[0];
428 const u32 r1 = out[1];
429 const u32 r2 = out[2];
430 const u32 r3 = out[3];
436 __kernel void m10400_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 pdf_t *pdf_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)
440 __kernel void m10400_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 pdf_t *pdf_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)
444 __kernel void m10400_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 pdf_t *pdf_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)
450 const u32 lid = get_local_id (0);
456 const u32 gid = get_global_id (0);
458 if (gid >= gid_max) return;
462 pw_buf0[0] = pws[gid].i[ 0];
463 pw_buf0[1] = pws[gid].i[ 1];
464 pw_buf0[2] = pws[gid].i[ 2];
465 pw_buf0[3] = pws[gid].i[ 3];
469 pw_buf1[0] = pws[gid].i[ 4];
470 pw_buf1[1] = pws[gid].i[ 5];
471 pw_buf1[2] = pws[gid].i[ 6];
472 pw_buf1[3] = pws[gid].i[ 7];
474 const u32 pw_len = pws[gid].pw_len;
480 __local RC4_KEY rc4_keys[64];
482 __local RC4_KEY *rc4_key = &rc4_keys[lid];
488 const u32 search[4] =
490 digests_buf[digests_offset].digest_buf[DGST_R0],
491 digests_buf[digests_offset].digest_buf[DGST_R1],
492 digests_buf[digests_offset].digest_buf[DGST_R2],
493 digests_buf[digests_offset].digest_buf[DGST_R3]
502 o_buf[0] = pdf_bufs[salt_pos].o_buf[0];
503 o_buf[1] = pdf_bufs[salt_pos].o_buf[1];
504 o_buf[2] = pdf_bufs[salt_pos].o_buf[2];
505 o_buf[3] = pdf_bufs[salt_pos].o_buf[3];
506 o_buf[4] = pdf_bufs[salt_pos].o_buf[4];
507 o_buf[5] = pdf_bufs[salt_pos].o_buf[5];
508 o_buf[6] = pdf_bufs[salt_pos].o_buf[6];
509 o_buf[7] = pdf_bufs[salt_pos].o_buf[7];
511 u32 P = pdf_bufs[salt_pos].P;
515 id_buf[0] = pdf_bufs[salt_pos].id_buf[0];
516 id_buf[1] = pdf_bufs[salt_pos].id_buf[1];
517 id_buf[2] = pdf_bufs[salt_pos].id_buf[2];
518 id_buf[3] = pdf_bufs[salt_pos].id_buf[3];
524 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
554 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
561 // max length supported by pdf11 is 32
563 w0_t[0] = padding[0];
564 w0_t[1] = padding[1];
565 w0_t[2] = padding[2];
566 w0_t[3] = padding[3];
567 w1_t[0] = padding[4];
568 w1_t[1] = padding[5];
569 w1_t[2] = padding[6];
570 w1_t[3] = padding[7];
580 switch_buffer_by_offset_le (w0_t, w1_t, w2_t, w3_t, pw_len);
583 // truncate at 32 is wanted, not a bug!
610 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
629 md5_transform (w0_t, w1_t, w2_t, w3_t, digest);
636 key[1] = digest[1] & 0xff;
640 rc4_init_16 (rc4_key, key);
644 rc4_next_16 (rc4_key, 0, 0, padding, out);
646 const u32 r0 = out[0];
647 const u32 r1 = out[1];
648 const u32 r2 = out[2];
649 const u32 r3 = out[3];
655 __kernel void m10400_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 pdf_t *pdf_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)
659 __kernel void m10400_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 pdf_t *pdf_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)