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"
31 #include "include/rp_gpu.h"
35 #define VECT_COMPARE_S "check_single_vect1_comp4.c"
36 #define VECT_COMPARE_M "check_multi_vect1_comp4.c"
39 #define GETCHAR(a,p) (((a)[(p) / 4] >> (((p) & 3) * 8)) & 0xff)
40 #define PUTCHAR(a,p,c) ((a)[(p) / 4] = (((a)[(p) / 4] & ~(0xff << (((p) & 3) * 8))) | ((c) << (((p) & 3) * 8))))
42 #define SETSHIFTEDINT(a,n,v) \
44 const u32 s = ((n) & 3) * 8; \
45 const u64 x = (u64) (v) << s; \
46 (a)[((n)/4)+0] &= ~(0xff << ((n & 3) * 8)); \
47 (a)[((n)/4)+0] |= x; \
48 (a)[((n)/4)+1] = x >> 32; \
51 __constant u32 sapb_trans_tbl[256] =
53 // first value hack for 0 byte as part of an optimization
54 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
55 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
56 0x3f, 0x40, 0x41, 0x50, 0x43, 0x44, 0x45, 0x4b, 0x47, 0x48, 0x4d, 0x4e, 0x54, 0x51, 0x53, 0x46,
57 0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x56, 0x55, 0x5c, 0x49, 0x5d, 0x4a,
58 0x42, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
59 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x58, 0x5b, 0x59, 0xff, 0x52,
60 0x4c, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
61 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x57, 0x5e, 0x5a, 0x4f, 0xff,
62 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
63 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
64 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
65 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
66 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
67 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
68 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
69 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff
72 __constant u32 bcodeArray[48] =
74 0x14, 0x77, 0xf3, 0xd4, 0xbb, 0x71, 0x23, 0xd0, 0x03, 0xff, 0x47, 0x93, 0x55, 0xaa, 0x66, 0x91,
75 0xf2, 0x88, 0x6b, 0x99, 0xbf, 0xcb, 0x32, 0x1a, 0x19, 0xd9, 0xa7, 0x82, 0x22, 0x49, 0xa2, 0x51,
76 0xe2, 0xb7, 0x33, 0x71, 0x8b, 0x9f, 0x5d, 0x01, 0x44, 0x70, 0xae, 0x11, 0xef, 0x28, 0xf0, 0x0d
79 static u32x sapb_trans (const u32x in)
84 out |= (sapb_trans_tbl[(in >> 0) & 0xff]) << 0;
85 out |= (sapb_trans_tbl[(in >> 8) & 0xff]) << 8;
86 out |= (sapb_trans_tbl[(in >> 16) & 0xff]) << 16;
87 out |= (sapb_trans_tbl[(in >> 24) & 0xff]) << 24;
93 static u32x walld0rf_magic (const u32x w0[4], const u32 pw_len, const u32x salt_buf0[4], const u32 salt_len, const u32x a, const u32x b, const u32x c, const u32x d, u32x t[16])
112 u32 sum20 = ((a >> 24) & 3)
120 const u32 w[2] = { w0[0], w0[1] };
122 const u32 s[3] = { salt_buf0[0], salt_buf0[1], salt_buf0[2] };
124 u32 saved_key[4] = { a, b, c, d };
130 // we can assume this because the password must be at least 3
131 // and the username must be at least 1 so we can save the if ()
137 t0 |= bcodeArray[47] << 0;
138 t0 |= (w[0] & 0xff) << 8;
139 t0 |= (s[0] & 0xff) << 16;
140 t0 |= bcodeArray[ 1] << 24;
148 t0 |= (w[0] & 0xff) << 0;
149 t0 |= (s[0] & 0xff) << 8;
150 t0 |= bcodeArray[ 0] << 16;
159 // because the following code can increase i2 by a maximum of 5,
160 // there is an overflow potential of 4 before it comes to the next test for i2 >= sum20
161 // we need to truncate in that case
163 while ((i1 < pw_len) && (i3 < salt_len))
169 if (GETCHAR (saved_key, 15 - i1) & 1)
171 x0 |= bcodeArray[48 - 1 - i1] << 0; i2++;
172 x0 |= GETCHAR (w, i1) << 8; i2++; i1++;
173 x0 |= GETCHAR (s, i3) << 16; i2++; i3++;
174 x0 |= bcodeArray[i2 - i1 - i3] << 24; i2++; i2++;
178 x0 |= GETCHAR (w, i1) << 0; i2++; i1++;
179 x0 |= GETCHAR (s, i3) << 8; i2++; i3++;
180 x0 |= bcodeArray[i2 - i1 - i3] << 16; i2++; i2++;
183 SETSHIFTEDINT (t, i2_sav, x0);
191 while ((i1 < pw_len) || (i3 < salt_len))
193 if (i1 < pw_len) // max 8
195 if (GETCHAR (saved_key, 15 - i1) & 1)
197 PUTCHAR (t, i2, bcodeArray[48 - 1 - i1]);
202 PUTCHAR (t, i2, GETCHAR (w, i1));
209 PUTCHAR (t, i2, GETCHAR (s, i3));
215 PUTCHAR (t, i2, bcodeArray[i2 - i1 - i3]);
228 PUTCHAR (t, i2, bcodeArray[i2 - i1 - i3]);
237 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07700_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 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];
267 const u32 pw_len = pws[gid].pw_len;
275 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
276 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
277 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
280 const u32 salt_len = salt_bufs[salt_pos].salt_len;
282 salt_buf0[0] = sapb_trans (salt_buf0[0]);
283 salt_buf0[1] = sapb_trans (salt_buf0[1]);
284 salt_buf0[2] = sapb_trans (salt_buf0[2]);
290 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
320 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
322 if (out_len > 8) continue; // otherwise it overflows in waldorf function
324 w0[0] = sapb_trans (w0[0]);
325 w0[1] = sapb_trans (w0[1]);
333 s0[0] = salt_buf0[0];
334 s0[1] = salt_buf0[1];
335 s0[2] = salt_buf0[2];
359 switch_buffer_by_offset (s0, s1, s2, s3, out_len);
361 const u32 pw_salt_len = out_len + salt_len;
365 t[ 0] = s0[0] | w0[0];
366 t[ 1] = s0[1] | w0[1];
379 t[14] = pw_salt_len * 8;
382 PUTCHAR (t, pw_salt_len, 0x80);
393 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
394 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
395 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
396 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
397 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
398 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
399 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
400 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
401 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
402 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
403 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
404 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
405 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
406 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
407 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
408 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
410 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
411 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
412 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
413 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
414 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
415 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
416 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
417 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
418 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
419 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
420 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
421 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
422 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
423 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
424 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
425 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
427 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
428 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
429 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
430 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
431 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
432 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
433 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
434 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
435 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
436 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
437 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
438 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
439 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
440 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
441 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
442 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
444 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
445 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
446 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
447 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
448 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
449 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
450 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
451 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
452 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
453 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
454 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
455 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
456 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
457 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
458 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
459 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
466 const u32x sum20 = walld0rf_magic (w0, out_len, salt_buf0, salt_len, a, b, c, d, t);
468 SETSHIFTEDINT (t, sum20, 0x80);
477 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
478 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
479 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
480 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
481 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
482 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
483 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
484 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
485 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
486 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
487 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
488 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
489 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
490 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
491 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
492 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
494 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
495 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
496 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
497 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
498 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
499 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
500 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
501 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
502 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
503 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
504 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
505 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
506 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
507 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
508 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
509 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
511 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
512 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
513 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
514 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
515 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
516 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
517 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
518 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
519 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
520 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
521 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
522 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
523 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
524 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
525 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
526 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
528 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
529 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
530 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
531 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
532 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
533 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
534 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
535 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
536 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
537 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
538 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
539 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
540 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
541 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
542 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
543 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
558 #include VECT_COMPARE_M
562 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07700_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
566 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07700_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
570 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07700_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
576 const u32 lid = get_local_id (0);
582 const u32 gid = get_global_id (0);
584 if (gid >= gid_max) return;
588 pw_buf0[0] = pws[gid].i[ 0];
589 pw_buf0[1] = pws[gid].i[ 1];
600 const u32 pw_len = pws[gid].pw_len;
608 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
609 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
610 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
613 const u32 salt_len = salt_bufs[salt_pos].salt_len;
615 salt_buf0[0] = sapb_trans (salt_buf0[0]);
616 salt_buf0[1] = sapb_trans (salt_buf0[1]);
617 salt_buf0[2] = sapb_trans (salt_buf0[2]);
623 const u32 search[4] =
625 digests_buf[digests_offset].digest_buf[DGST_R0],
626 digests_buf[digests_offset].digest_buf[DGST_R1],
627 digests_buf[digests_offset].digest_buf[DGST_R2],
628 digests_buf[digests_offset].digest_buf[DGST_R3]
635 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
665 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
667 if (out_len > 8) continue; // otherwise it overflows in waldorf function
669 w0[0] = sapb_trans (w0[0]);
670 w0[1] = sapb_trans (w0[1]);
678 s0[0] = salt_buf0[0];
679 s0[1] = salt_buf0[1];
680 s0[2] = salt_buf0[2];
704 switch_buffer_by_offset (s0, s1, s2, s3, out_len);
706 const u32 pw_salt_len = out_len + salt_len;
710 t[ 0] = s0[0] | w0[0];
711 t[ 1] = s0[1] | w0[1];
724 t[14] = pw_salt_len * 8;
727 PUTCHAR (t, pw_salt_len, 0x80);
738 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
739 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
740 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
741 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
742 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
743 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
744 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
745 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
746 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
747 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
748 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
749 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
750 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
751 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
752 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
753 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
755 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
756 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
757 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
758 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
759 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
760 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
761 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
762 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
763 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
764 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
765 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
766 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
767 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
768 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
769 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
770 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
772 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
773 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
774 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
775 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
776 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
777 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
778 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
779 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
780 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
781 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
782 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
783 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
784 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
785 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
786 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
787 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
789 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
790 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
791 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
792 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
793 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
794 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
795 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
796 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
797 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
798 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
799 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
800 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
801 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
802 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
803 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
804 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
811 const u32x sum20 = walld0rf_magic (w0, out_len, salt_buf0, salt_len, a, b, c, d, t);
813 SETSHIFTEDINT (t, sum20, 0x80);
822 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
823 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
824 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
825 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
826 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
827 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
828 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
829 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
830 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
831 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
832 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
833 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
834 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
835 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
836 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
837 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
839 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
840 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
841 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
842 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
843 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
844 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
845 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
846 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
847 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
848 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
849 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
850 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
851 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
852 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
853 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
854 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
856 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
857 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
858 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
859 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
860 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
861 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
862 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
863 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
864 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
865 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
866 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
867 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
868 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
869 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
870 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
871 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
873 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
874 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
875 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
876 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
877 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
878 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
879 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
880 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
881 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
882 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
883 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
884 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
885 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
886 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
887 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
888 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
903 #include VECT_COMPARE_S
907 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07700_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
911 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m07700_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 rules_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)