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 "types_ocl.c"
19 #include "include/rp_gpu.h"
22 #define COMPARE_S "check_single_comp4.c"
23 #define COMPARE_M "check_multi_comp4.c"
25 __constant u32 k_sha256[64] =
27 SHA256C00, SHA256C01, SHA256C02, SHA256C03,
28 SHA256C04, SHA256C05, SHA256C06, SHA256C07,
29 SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
30 SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
31 SHA256C10, SHA256C11, SHA256C12, SHA256C13,
32 SHA256C14, SHA256C15, SHA256C16, SHA256C17,
33 SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
34 SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
35 SHA256C20, SHA256C21, SHA256C22, SHA256C23,
36 SHA256C24, SHA256C25, SHA256C26, SHA256C27,
37 SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
38 SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
39 SHA256C30, SHA256C31, SHA256C32, SHA256C33,
40 SHA256C34, SHA256C35, SHA256C36, SHA256C37,
41 SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
42 SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
45 static void sha256_transform (u32 digest[8], const u32 w[16])
73 #define ROUND_EXPAND() \
75 w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
76 w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
77 w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
78 w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
79 w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
80 w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
81 w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
82 w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
83 w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
84 w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
85 wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
86 wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
87 wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
88 wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
89 we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
90 wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
93 #define ROUND_STEP(i) \
95 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
96 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
97 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
98 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
99 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
100 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
101 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
102 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
103 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
104 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
105 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
106 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
107 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
108 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
109 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
110 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
116 for (int i = 16; i < 64; i += 16)
118 ROUND_EXPAND (); ROUND_STEP (i);
131 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_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)
137 const u32 lid = get_local_id (0);
143 const u32 gid = get_global_id (0);
145 if (gid >= gid_max) return;
149 pw_buf0[0] = pws[gid].i[ 0];
150 pw_buf0[1] = pws[gid].i[ 1];
151 pw_buf0[2] = pws[gid].i[ 2];
152 pw_buf0[3] = pws[gid].i[ 3];
156 pw_buf1[0] = pws[gid].i[ 4];
157 pw_buf1[1] = pws[gid].i[ 5];
158 pw_buf1[2] = pws[gid].i[ 6];
159 pw_buf1[3] = pws[gid].i[ 7];
161 const u32 pw_len = pws[gid].pw_len;
167 const u32 salt_buf0 = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
168 const u32 salt_buf1 = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
169 const u32 salt_buf2 = swap32 (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
175 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
199 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
206 make_unicode (w0, w0_t, w1_t);
208 make_unicode (w1, w2_t, w3_t);
212 w_t[ 0] = swap32 (w0_t[0]);
213 w_t[ 1] = swap32 (w0_t[1]);
214 w_t[ 2] = swap32 (w0_t[2]);
215 w_t[ 3] = swap32 (w0_t[3]);
216 w_t[ 4] = swap32 (w1_t[0]);
217 w_t[ 5] = swap32 (w1_t[1]);
218 w_t[ 6] = swap32 (w1_t[2]);
219 w_t[ 7] = swap32 (w1_t[3]);
220 w_t[ 8] = swap32 (w2_t[0]);
221 w_t[ 9] = swap32 (w2_t[1]);
222 w_t[10] = swap32 (w2_t[2]);
223 w_t[11] = swap32 (w2_t[3]);
224 w_t[12] = swap32 (w3_t[0]);
225 w_t[13] = swap32 (w3_t[1]);
226 w_t[14] = swap32 (w3_t[2]);
227 w_t[15] = swap32 (w3_t[3]);
229 w_t[ 0] = w_t[ 0] >> 8;
230 w_t[ 1] = w_t[ 1] >> 8;
231 w_t[ 2] = w_t[ 2] >> 8;
232 w_t[ 3] = w_t[ 3] >> 8;
233 w_t[ 4] = w_t[ 4] >> 8;
234 w_t[ 5] = w_t[ 5] >> 8;
235 w_t[ 6] = w_t[ 6] >> 8;
236 w_t[ 7] = w_t[ 7] >> 8;
237 w_t[ 8] = w_t[ 8] >> 8;
238 w_t[ 9] = w_t[ 9] >> 8;
239 w_t[10] = w_t[10] >> 8;
240 w_t[11] = w_t[11] >> 8;
241 w_t[12] = w_t[12] >> 8;
242 w_t[13] = w_t[13] >> 8;
243 w_t[14] = w_t[14] >> 8;
244 w_t[15] = w_t[15] >> 8;
248 digest[0] = SHA256M_A;
249 digest[1] = SHA256M_B;
250 digest[2] = SHA256M_C;
251 digest[3] = SHA256M_D;
252 digest[4] = SHA256M_E;
253 digest[5] = SHA256M_F;
254 digest[6] = SHA256M_G;
255 digest[7] = SHA256M_H;
257 sha256_transform (digest, w_t); // 0 - 64
276 sha256_transform (digest, w_t); // 64 - 128
277 sha256_transform (digest, w_t); // 128 - 192
278 sha256_transform (digest, w_t); // 192 - 256
279 sha256_transform (digest, w_t); // 256 - 320
280 sha256_transform (digest, w_t); // 320 - 384
281 sha256_transform (digest, w_t); // 384 - 448
283 w_t[15] = 0 | salt_buf0 >> 16;
285 sha256_transform (digest, w_t); // 448 - 512
287 w_t[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
288 w_t[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
289 w_t[ 2] = salt_buf2 << 16 | 0;
290 w_t[15] = (510 + 8) * 8;
292 sha256_transform (digest, w_t); // 512 - 576
294 const u32 r0 = digest[3];
295 const u32 r1 = digest[7];
296 const u32 r2 = digest[2];
297 const u32 r3 = digest[6];
303 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_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)
307 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_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)
311 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_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)
317 const u32 lid = get_local_id (0);
323 const u32 gid = get_global_id (0);
325 if (gid >= gid_max) return;
329 pw_buf0[0] = pws[gid].i[ 0];
330 pw_buf0[1] = pws[gid].i[ 1];
331 pw_buf0[2] = pws[gid].i[ 2];
332 pw_buf0[3] = pws[gid].i[ 3];
336 pw_buf1[0] = pws[gid].i[ 4];
337 pw_buf1[1] = pws[gid].i[ 5];
338 pw_buf1[2] = pws[gid].i[ 6];
339 pw_buf1[3] = pws[gid].i[ 7];
341 const u32 pw_len = pws[gid].pw_len;
347 const u32 salt_buf0 = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
348 const u32 salt_buf1 = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
349 const u32 salt_buf2 = swap32 (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
355 const u32 search[4] =
357 digests_buf[digests_offset].digest_buf[DGST_R0],
358 digests_buf[digests_offset].digest_buf[DGST_R1],
359 digests_buf[digests_offset].digest_buf[DGST_R2],
360 digests_buf[digests_offset].digest_buf[DGST_R3]
367 for (u32 il_pos = 0; il_pos < rules_cnt; il_pos++)
391 const u32 out_len = apply_rules (rules_buf[il_pos].cmds, w0, w1, pw_len);
398 make_unicode (w0, w0_t, w1_t);
400 make_unicode (w1, w2_t, w3_t);
404 w_t[ 0] = swap32 (w0_t[0]);
405 w_t[ 1] = swap32 (w0_t[1]);
406 w_t[ 2] = swap32 (w0_t[2]);
407 w_t[ 3] = swap32 (w0_t[3]);
408 w_t[ 4] = swap32 (w1_t[0]);
409 w_t[ 5] = swap32 (w1_t[1]);
410 w_t[ 6] = swap32 (w1_t[2]);
411 w_t[ 7] = swap32 (w1_t[3]);
412 w_t[ 8] = swap32 (w2_t[0]);
413 w_t[ 9] = swap32 (w2_t[1]);
414 w_t[10] = swap32 (w2_t[2]);
415 w_t[11] = swap32 (w2_t[3]);
416 w_t[12] = swap32 (w3_t[0]);
417 w_t[13] = swap32 (w3_t[1]);
418 w_t[14] = swap32 (w3_t[2]);
419 w_t[15] = swap32 (w3_t[3]);
421 w_t[ 0] = w_t[ 0] >> 8;
422 w_t[ 1] = w_t[ 1] >> 8;
423 w_t[ 2] = w_t[ 2] >> 8;
424 w_t[ 3] = w_t[ 3] >> 8;
425 w_t[ 4] = w_t[ 4] >> 8;
426 w_t[ 5] = w_t[ 5] >> 8;
427 w_t[ 6] = w_t[ 6] >> 8;
428 w_t[ 7] = w_t[ 7] >> 8;
429 w_t[ 8] = w_t[ 8] >> 8;
430 w_t[ 9] = w_t[ 9] >> 8;
431 w_t[10] = w_t[10] >> 8;
432 w_t[11] = w_t[11] >> 8;
433 w_t[12] = w_t[12] >> 8;
434 w_t[13] = w_t[13] >> 8;
435 w_t[14] = w_t[14] >> 8;
436 w_t[15] = w_t[15] >> 8;
440 digest[0] = SHA256M_A;
441 digest[1] = SHA256M_B;
442 digest[2] = SHA256M_C;
443 digest[3] = SHA256M_D;
444 digest[4] = SHA256M_E;
445 digest[5] = SHA256M_F;
446 digest[6] = SHA256M_G;
447 digest[7] = SHA256M_H;
449 sha256_transform (digest, w_t); // 0 - 64
468 sha256_transform (digest, w_t); // 64 - 128
469 sha256_transform (digest, w_t); // 128 - 192
470 sha256_transform (digest, w_t); // 192 - 256
471 sha256_transform (digest, w_t); // 256 - 320
472 sha256_transform (digest, w_t); // 320 - 384
473 sha256_transform (digest, w_t); // 384 - 448
475 w_t[15] = 0 | salt_buf0 >> 16;
477 sha256_transform (digest, w_t); // 448 - 512
479 w_t[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
480 w_t[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
481 w_t[ 2] = salt_buf2 << 16 | 0;
482 w_t[15] = (510 + 8) * 8;
484 sha256_transform (digest, w_t); // 512 - 576
486 const u32 r0 = digest[3];
487 const u32 r1 = digest[7];
488 const u32 r2 = digest[2];
489 const u32 r3 = digest[6];
495 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_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)
499 __kernel void __attribute__((reqd_work_group_size (64, 1, 1))) m08000_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)