2 * Author......: Jens Steube <jens.steube@gmail.com>
8 //too much register pressure
9 //#define NEW_SIMD_CODE
11 #include "inc_hash_constants.h"
12 #include "inc_vendor.cl"
19 #include "inc_hash_functions.cl"
20 #include "inc_types.cl"
21 #include "inc_common.cl"
24 #include "inc_simd.cl"
26 #define GETCHAR(a,p) (((a)[(p) / 4] >> (((p) & 3) * 8)) & 0xff)
27 #define PUTCHAR(a,p,c) ((a)[(p) / 4] = (((a)[(p) / 4] & ~(0xff << (((p) & 3) * 8))) | ((c) << (((p) & 3) * 8))))
29 #define SETSHIFTEDINT(a,n,v) \
31 const u32 s = ((n) & 3) * 8; \
32 const u64 x = (u64) (v) << s; \
33 (a)[((n)/4)+0] &= ~(0xff << ((n & 3) * 8)); \
34 (a)[((n)/4)+0] |= x; \
35 (a)[((n)/4)+1] = x >> 32; \
38 __constant u32 sapb_trans_tbl[256] =
40 // first value hack for 0 byte as part of an optimization
41 0x00, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
42 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
43 0x3f, 0x40, 0x41, 0x50, 0x43, 0x44, 0x45, 0x4b, 0x47, 0x48, 0x4d, 0x4e, 0x54, 0x51, 0x53, 0x46,
44 0x35, 0x36, 0x37, 0x38, 0x39, 0x3a, 0x3b, 0x3c, 0x3d, 0x3e, 0x56, 0x55, 0x5c, 0x49, 0x5d, 0x4a,
45 0x42, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
46 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x58, 0x5b, 0x59, 0xff, 0x52,
47 0x4c, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07, 0x08, 0x09, 0x0a, 0x0b, 0x0c, 0x0d, 0x0e, 0x0f,
48 0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x19, 0x1a, 0x57, 0x5e, 0x5a, 0x4f, 0xff,
49 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
50 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
51 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
52 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
53 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff,
54 0xff, 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 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff, 0xff
59 __constant u32 bcodeArray[48] =
61 0x14, 0x77, 0xf3, 0xd4, 0xbb, 0x71, 0x23, 0xd0, 0x03, 0xff, 0x47, 0x93, 0x55, 0xaa, 0x66, 0x91,
62 0xf2, 0x88, 0x6b, 0x99, 0xbf, 0xcb, 0x32, 0x1a, 0x19, 0xd9, 0xa7, 0x82, 0x22, 0x49, 0xa2, 0x51,
63 0xe2, 0xb7, 0x33, 0x71, 0x8b, 0x9f, 0x5d, 0x01, 0x44, 0x70, 0xae, 0x11, 0xef, 0x28, 0xf0, 0x0d
66 u32 sapb_trans (const u32 in)
70 out |= (sapb_trans_tbl[(in >> 0) & 0xff]) << 0;
71 out |= (sapb_trans_tbl[(in >> 8) & 0xff]) << 8;
72 out |= (sapb_trans_tbl[(in >> 16) & 0xff]) << 16;
73 out |= (sapb_trans_tbl[(in >> 24) & 0xff]) << 24;
78 u32 walld0rf_magic (const u32 w0[4], const u32 pw_len, const u32 salt_buf0[4], const u32 salt_len, const u32 a, const u32 b, const u32 c, const u32 d, u32 t[16])
97 u32 sum20 = ((a >> 24) & 3)
105 const u32 w[2] = { w0[0], w0[1] };
107 const u32 s[3] = { salt_buf0[0], salt_buf0[1], salt_buf0[2] };
109 u32 saved_key[4] = { a, b, c, d };
115 // we can assume this because the password must be at least 3
116 // and the username must be at least 1 so we can save the if ()
122 t0 |= bcodeArray[47] << 0;
123 t0 |= (w[0] & 0xff) << 8;
124 t0 |= (s[0] & 0xff) << 16;
125 t0 |= bcodeArray[ 1] << 24;
133 t0 |= (w[0] & 0xff) << 0;
134 t0 |= (s[0] & 0xff) << 8;
135 t0 |= bcodeArray[ 0] << 16;
144 // because the following code can increase i2 by a maximum of 5,
145 // there is an overflow potential of 4 before it comes to the next test for i2 >= sum20
146 // we need to truncate in that case
148 while ((i1 < pw_len) && (i3 < salt_len))
154 if (GETCHAR (saved_key, 15 - i1) & 1)
156 x0 |= bcodeArray[48 - 1 - i1] << 0; i2++;
157 x0 |= GETCHAR (w, i1) << 8; i2++; i1++;
158 x0 |= GETCHAR (s, i3) << 16; i2++; i3++;
159 x0 |= bcodeArray[i2 - i1 - i3] << 24; i2++; i2++;
163 x0 |= GETCHAR (w, i1) << 0; i2++; i1++;
164 x0 |= GETCHAR (s, i3) << 8; i2++; i3++;
165 x0 |= bcodeArray[i2 - i1 - i3] << 16; i2++; i2++;
168 SETSHIFTEDINT (t, i2_sav, x0);
176 while ((i1 < pw_len) || (i3 < salt_len))
178 if (i1 < pw_len) // max 8
180 if (GETCHAR (saved_key, 15 - i1) & 1)
182 PUTCHAR (t, i2, bcodeArray[48 - 1 - i1]);
187 PUTCHAR (t, i2, GETCHAR (w, i1));
194 PUTCHAR (t, i2, GETCHAR (s, i3));
200 PUTCHAR (t, i2, bcodeArray[i2 - i1 - i3]);
213 PUTCHAR (t, i2, bcodeArray[i2 - i1 - i3]);
222 __kernel void m07700_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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
228 const u32 lid = get_local_id (0);
234 const u32 gid = get_global_id (0);
236 if (gid >= gid_max) return;
241 pw_buf0[0] = pws[gid].i[0];
242 pw_buf0[1] = pws[gid].i[1];
243 pw_buf0[2] = pws[gid].i[2];
244 pw_buf0[3] = pws[gid].i[3];
245 pw_buf1[0] = pws[gid].i[4];
246 pw_buf1[1] = pws[gid].i[5];
247 pw_buf1[2] = pws[gid].i[6];
248 pw_buf1[3] = pws[gid].i[7];
250 const u32 pw_len = pws[gid].pw_len;
258 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
259 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
260 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
263 const u32 salt_len = salt_bufs[salt_pos].salt_len;
265 salt_buf0[0] = sapb_trans (salt_buf0[0]);
266 salt_buf0[1] = sapb_trans (salt_buf0[1]);
267 salt_buf0[2] = sapb_trans (salt_buf0[2]);
273 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
280 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
282 if (out_len > 8) continue; // otherwise it overflows in waldorf function
288 w0[0] = sapb_trans (w0[0]);
289 w0[1] = sapb_trans (w0[1]);
300 s0[0] = salt_buf0[0];
301 s0[1] = salt_buf0[1];
302 s0[2] = salt_buf0[2];
317 switch_buffer_by_offset_le (s0, s1, s2, s3, out_len);
319 const u32 pw_salt_len = out_len + salt_len;
323 t[ 0] = s0[0] | w0[0];
324 t[ 1] = s0[1] | w0[1];
337 t[14] = pw_salt_len * 8;
340 PUTCHAR (t, pw_salt_len, 0x80);
351 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
352 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
353 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
354 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
355 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
356 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
357 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
358 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
359 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
360 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
361 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
362 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
363 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
364 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
365 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
366 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
368 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
369 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
370 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
371 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
372 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
373 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
374 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
375 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
376 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
377 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
378 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
379 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
380 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
381 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
382 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
383 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
385 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
386 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
387 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
388 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
389 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
390 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
391 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
392 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
393 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
394 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
395 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
396 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
397 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
398 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
399 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
400 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
402 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
403 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
404 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
405 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
406 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
407 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
408 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
409 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
410 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
411 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
412 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
413 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
414 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
415 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
416 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
417 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
424 const u32 sum20 = walld0rf_magic (w0, out_len, salt_buf0, salt_len, a, b, c, d, t);
426 SETSHIFTEDINT (t, sum20, 0x80);
435 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
436 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
437 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
438 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
439 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
440 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
441 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
442 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
443 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
444 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
445 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
446 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
447 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
448 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
449 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
450 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
452 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
453 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
454 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
455 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
456 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
457 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
458 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
459 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
460 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
461 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
462 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
463 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
464 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
465 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
466 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
467 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
469 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
470 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
471 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
472 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
473 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
474 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
475 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
476 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
477 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
478 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
479 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
480 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
481 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
482 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
483 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
484 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
486 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
487 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
488 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
489 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
490 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
491 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
492 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
493 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
494 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
495 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
496 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
497 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
498 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
499 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
500 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
501 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
513 COMPARE_M_SIMD (a, b, c, d);
517 __kernel void m07700_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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
521 __kernel void m07700_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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
525 __kernel void m07700_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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
531 const u32 lid = get_local_id (0);
537 const u32 gid = get_global_id (0);
539 if (gid >= gid_max) return;
544 pw_buf0[0] = pws[gid].i[0];
545 pw_buf0[1] = pws[gid].i[1];
546 pw_buf0[2] = pws[gid].i[2];
547 pw_buf0[3] = pws[gid].i[3];
548 pw_buf1[0] = pws[gid].i[4];
549 pw_buf1[1] = pws[gid].i[5];
550 pw_buf1[2] = pws[gid].i[6];
551 pw_buf1[3] = pws[gid].i[7];
553 const u32 pw_len = pws[gid].pw_len;
561 salt_buf0[0] = salt_bufs[salt_pos].salt_buf[0];
562 salt_buf0[1] = salt_bufs[salt_pos].salt_buf[1];
563 salt_buf0[2] = salt_bufs[salt_pos].salt_buf[2];
566 const u32 salt_len = salt_bufs[salt_pos].salt_len;
568 salt_buf0[0] = sapb_trans (salt_buf0[0]);
569 salt_buf0[1] = sapb_trans (salt_buf0[1]);
570 salt_buf0[2] = sapb_trans (salt_buf0[2]);
576 const u32 search[4] =
578 digests_buf[digests_offset].digest_buf[DGST_R0],
579 digests_buf[digests_offset].digest_buf[DGST_R1],
588 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
595 const u32x out_len = apply_rules_vect (pw_buf0, pw_buf1, pw_len, rules_buf, il_pos, w0, w1);
597 if (out_len > 8) continue; // otherwise it overflows in waldorf function
603 w0[0] = sapb_trans (w0[0]);
604 w0[1] = sapb_trans (w0[1]);
615 s0[0] = salt_buf0[0];
616 s0[1] = salt_buf0[1];
617 s0[2] = salt_buf0[2];
632 switch_buffer_by_offset_le (s0, s1, s2, s3, out_len);
634 const u32 pw_salt_len = out_len + salt_len;
638 t[ 0] = s0[0] | w0[0];
639 t[ 1] = s0[1] | w0[1];
652 t[14] = pw_salt_len * 8;
655 PUTCHAR (t, pw_salt_len, 0x80);
666 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
667 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
668 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
669 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
670 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
671 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
672 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
673 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
674 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
675 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
676 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
677 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
678 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
679 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
680 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
681 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
683 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
684 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
685 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
686 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
687 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
688 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
689 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
690 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
691 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
692 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
693 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
694 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
695 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
696 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
697 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
698 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
700 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
701 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
702 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
703 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
704 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
705 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
706 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
707 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
708 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
709 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
710 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
711 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
712 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
713 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
714 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
715 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
717 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
718 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
719 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
720 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
721 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
722 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
723 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
724 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
725 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
726 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
727 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
728 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
729 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
730 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
731 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
732 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
739 const u32 sum20 = walld0rf_magic (w0, out_len, salt_buf0, salt_len, a, b, c, d, t);
741 SETSHIFTEDINT (t, sum20, 0x80);
750 MD5_STEP (MD5_Fo, a, b, c, d, t[ 0], MD5C00, MD5S00);
751 MD5_STEP (MD5_Fo, d, a, b, c, t[ 1], MD5C01, MD5S01);
752 MD5_STEP (MD5_Fo, c, d, a, b, t[ 2], MD5C02, MD5S02);
753 MD5_STEP (MD5_Fo, b, c, d, a, t[ 3], MD5C03, MD5S03);
754 MD5_STEP (MD5_Fo, a, b, c, d, t[ 4], MD5C04, MD5S00);
755 MD5_STEP (MD5_Fo, d, a, b, c, t[ 5], MD5C05, MD5S01);
756 MD5_STEP (MD5_Fo, c, d, a, b, t[ 6], MD5C06, MD5S02);
757 MD5_STEP (MD5_Fo, b, c, d, a, t[ 7], MD5C07, MD5S03);
758 MD5_STEP (MD5_Fo, a, b, c, d, t[ 8], MD5C08, MD5S00);
759 MD5_STEP (MD5_Fo, d, a, b, c, t[ 9], MD5C09, MD5S01);
760 MD5_STEP (MD5_Fo, c, d, a, b, t[10], MD5C0a, MD5S02);
761 MD5_STEP (MD5_Fo, b, c, d, a, t[11], MD5C0b, MD5S03);
762 MD5_STEP (MD5_Fo, a, b, c, d, t[12], MD5C0c, MD5S00);
763 MD5_STEP (MD5_Fo, d, a, b, c, t[13], MD5C0d, MD5S01);
764 MD5_STEP (MD5_Fo, c, d, a, b, t[14], MD5C0e, MD5S02);
765 MD5_STEP (MD5_Fo, b, c, d, a, t[15], MD5C0f, MD5S03);
767 MD5_STEP (MD5_Go, a, b, c, d, t[ 1], MD5C10, MD5S10);
768 MD5_STEP (MD5_Go, d, a, b, c, t[ 6], MD5C11, MD5S11);
769 MD5_STEP (MD5_Go, c, d, a, b, t[11], MD5C12, MD5S12);
770 MD5_STEP (MD5_Go, b, c, d, a, t[ 0], MD5C13, MD5S13);
771 MD5_STEP (MD5_Go, a, b, c, d, t[ 5], MD5C14, MD5S10);
772 MD5_STEP (MD5_Go, d, a, b, c, t[10], MD5C15, MD5S11);
773 MD5_STEP (MD5_Go, c, d, a, b, t[15], MD5C16, MD5S12);
774 MD5_STEP (MD5_Go, b, c, d, a, t[ 4], MD5C17, MD5S13);
775 MD5_STEP (MD5_Go, a, b, c, d, t[ 9], MD5C18, MD5S10);
776 MD5_STEP (MD5_Go, d, a, b, c, t[14], MD5C19, MD5S11);
777 MD5_STEP (MD5_Go, c, d, a, b, t[ 3], MD5C1a, MD5S12);
778 MD5_STEP (MD5_Go, b, c, d, a, t[ 8], MD5C1b, MD5S13);
779 MD5_STEP (MD5_Go, a, b, c, d, t[13], MD5C1c, MD5S10);
780 MD5_STEP (MD5_Go, d, a, b, c, t[ 2], MD5C1d, MD5S11);
781 MD5_STEP (MD5_Go, c, d, a, b, t[ 7], MD5C1e, MD5S12);
782 MD5_STEP (MD5_Go, b, c, d, a, t[12], MD5C1f, MD5S13);
784 MD5_STEP (MD5_H , a, b, c, d, t[ 5], MD5C20, MD5S20);
785 MD5_STEP (MD5_H , d, a, b, c, t[ 8], MD5C21, MD5S21);
786 MD5_STEP (MD5_H , c, d, a, b, t[11], MD5C22, MD5S22);
787 MD5_STEP (MD5_H , b, c, d, a, t[14], MD5C23, MD5S23);
788 MD5_STEP (MD5_H , a, b, c, d, t[ 1], MD5C24, MD5S20);
789 MD5_STEP (MD5_H , d, a, b, c, t[ 4], MD5C25, MD5S21);
790 MD5_STEP (MD5_H , c, d, a, b, t[ 7], MD5C26, MD5S22);
791 MD5_STEP (MD5_H , b, c, d, a, t[10], MD5C27, MD5S23);
792 MD5_STEP (MD5_H , a, b, c, d, t[13], MD5C28, MD5S20);
793 MD5_STEP (MD5_H , d, a, b, c, t[ 0], MD5C29, MD5S21);
794 MD5_STEP (MD5_H , c, d, a, b, t[ 3], MD5C2a, MD5S22);
795 MD5_STEP (MD5_H , b, c, d, a, t[ 6], MD5C2b, MD5S23);
796 MD5_STEP (MD5_H , a, b, c, d, t[ 9], MD5C2c, MD5S20);
797 MD5_STEP (MD5_H , d, a, b, c, t[12], MD5C2d, MD5S21);
798 MD5_STEP (MD5_H , c, d, a, b, t[15], MD5C2e, MD5S22);
799 MD5_STEP (MD5_H , b, c, d, a, t[ 2], MD5C2f, MD5S23);
801 MD5_STEP (MD5_I , a, b, c, d, t[ 0], MD5C30, MD5S30);
802 MD5_STEP (MD5_I , d, a, b, c, t[ 7], MD5C31, MD5S31);
803 MD5_STEP (MD5_I , c, d, a, b, t[14], MD5C32, MD5S32);
804 MD5_STEP (MD5_I , b, c, d, a, t[ 5], MD5C33, MD5S33);
805 MD5_STEP (MD5_I , a, b, c, d, t[12], MD5C34, MD5S30);
806 MD5_STEP (MD5_I , d, a, b, c, t[ 3], MD5C35, MD5S31);
807 MD5_STEP (MD5_I , c, d, a, b, t[10], MD5C36, MD5S32);
808 MD5_STEP (MD5_I , b, c, d, a, t[ 1], MD5C37, MD5S33);
809 MD5_STEP (MD5_I , a, b, c, d, t[ 8], MD5C38, MD5S30);
810 MD5_STEP (MD5_I , d, a, b, c, t[15], MD5C39, MD5S31);
811 MD5_STEP (MD5_I , c, d, a, b, t[ 6], MD5C3a, MD5S32);
812 MD5_STEP (MD5_I , b, c, d, a, t[13], MD5C3b, MD5S33);
813 MD5_STEP (MD5_I , a, b, c, d, t[ 4], MD5C3c, MD5S30);
814 MD5_STEP (MD5_I , d, a, b, c, t[11], MD5C3d, MD5S31);
815 MD5_STEP (MD5_I , c, d, a, b, t[ 2], MD5C3e, MD5S32);
816 MD5_STEP (MD5_I , b, c, d, a, t[ 9], MD5C3f, MD5S33);
828 COMPARE_S_SIMD (a, b, c, d);
832 __kernel void m07700_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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
836 __kernel void m07700_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_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 il_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)