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"
20 #define COMPARE_S "OpenCL/check_single_comp4.c"
21 #define COMPARE_M "OpenCL/check_multi_comp4.c"
23 __constant u32 k_sha256[64] =
25 SHA256C00, SHA256C01, SHA256C02, SHA256C03,
26 SHA256C04, SHA256C05, SHA256C06, SHA256C07,
27 SHA256C08, SHA256C09, SHA256C0a, SHA256C0b,
28 SHA256C0c, SHA256C0d, SHA256C0e, SHA256C0f,
29 SHA256C10, SHA256C11, SHA256C12, SHA256C13,
30 SHA256C14, SHA256C15, SHA256C16, SHA256C17,
31 SHA256C18, SHA256C19, SHA256C1a, SHA256C1b,
32 SHA256C1c, SHA256C1d, SHA256C1e, SHA256C1f,
33 SHA256C20, SHA256C21, SHA256C22, SHA256C23,
34 SHA256C24, SHA256C25, SHA256C26, SHA256C27,
35 SHA256C28, SHA256C29, SHA256C2a, SHA256C2b,
36 SHA256C2c, SHA256C2d, SHA256C2e, SHA256C2f,
37 SHA256C30, SHA256C31, SHA256C32, SHA256C33,
38 SHA256C34, SHA256C35, SHA256C36, SHA256C37,
39 SHA256C38, SHA256C39, SHA256C3a, SHA256C3b,
40 SHA256C3c, SHA256C3d, SHA256C3e, SHA256C3f,
43 static void sha256_transform (u32 digest[8], const u32 w[16])
71 #define ROUND_EXPAND() \
73 w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
74 w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
75 w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
76 w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
77 w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
78 w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
79 w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
80 w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
81 w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
82 w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
83 wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
84 wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
85 wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
86 wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
87 we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
88 wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
91 #define ROUND_STEP(i) \
93 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
94 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
95 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
96 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
97 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
98 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
99 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
100 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
101 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
102 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
103 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
104 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
105 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
106 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
107 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
108 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
114 for (int i = 16; i < 64; i += 16)
116 ROUND_EXPAND (); ROUND_STEP (i);
129 static void sha256_transform_z (u32 digest[8])
140 #define ROUND_STEP_Z(i) \
142 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, 0, k_sha256[i + 0]); \
143 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, 0, k_sha256[i + 1]); \
144 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, 0, k_sha256[i + 2]); \
145 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, 0, k_sha256[i + 3]); \
146 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, 0, k_sha256[i + 4]); \
147 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, 0, k_sha256[i + 5]); \
148 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, 0, k_sha256[i + 6]); \
149 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, 0, k_sha256[i + 7]); \
150 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, 0, k_sha256[i + 8]); \
151 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, 0, k_sha256[i + 9]); \
152 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, 0, k_sha256[i + 10]); \
153 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, 0, k_sha256[i + 11]); \
154 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, 0, k_sha256[i + 12]); \
155 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, 0, k_sha256[i + 13]); \
156 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, 0, k_sha256[i + 14]); \
157 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, 0, k_sha256[i + 15]); \
163 for (int i = 16; i < 64; i += 16)
178 static void sha256_transform_s (u32 digest[8], __local u32 w[64])
189 #define ROUND_STEP_S(i) \
191 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w[i + 0], k_sha256[i + 0]); \
192 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w[i + 1], k_sha256[i + 1]); \
193 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w[i + 2], k_sha256[i + 2]); \
194 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w[i + 3], k_sha256[i + 3]); \
195 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w[i + 4], k_sha256[i + 4]); \
196 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w[i + 5], k_sha256[i + 5]); \
197 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w[i + 6], k_sha256[i + 6]); \
198 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w[i + 7], k_sha256[i + 7]); \
199 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w[i + 8], k_sha256[i + 8]); \
200 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w[i + 9], k_sha256[i + 9]); \
201 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w[i + 10], k_sha256[i + 10]); \
202 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w[i + 11], k_sha256[i + 11]); \
203 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w[i + 12], k_sha256[i + 12]); \
204 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w[i + 13], k_sha256[i + 13]); \
205 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w[i + 14], k_sha256[i + 14]); \
206 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w[i + 15], k_sha256[i + 15]); \
212 for (int i = 16; i < 64; i += 16)
227 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
233 const u32 lid = get_local_id (0);
239 const u32 gid = get_global_id (0);
243 wordl0[0] = pws[gid].i[ 0];
244 wordl0[1] = pws[gid].i[ 1];
245 wordl0[2] = pws[gid].i[ 2];
246 wordl0[3] = pws[gid].i[ 3];
250 wordl1[0] = pws[gid].i[ 4];
251 wordl1[1] = pws[gid].i[ 5];
252 wordl1[2] = pws[gid].i[ 6];
253 wordl1[3] = pws[gid].i[ 7];
269 const u32 pw_l_len = pws[gid].pw_len;
271 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
273 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
280 const u32 salt_buf0 = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
281 const u32 salt_buf1 = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
282 const u32 salt_buf2 = swap32 (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
285 * precompute final msg blocks
288 __local u32 w_s1[64];
289 __local u32 w_s2[64];
294 barrier (CLK_LOCAL_MEM_FENCE);
298 w_s1[15] = 0 | salt_buf0 >> 16;
301 for (int i = 16; i < 64; i++)
303 w_s1[i] = SHA256_EXPAND (w_s1[i - 2], w_s1[i - 7], w_s1[i - 15], w_s1[i - 16]);
308 w_s2[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
309 w_s2[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
310 w_s2[ 2] = salt_buf2 << 16 | 0;
311 w_s2[15] = (510 + 8) * 8;
314 for (int i = 16; i < 64; i++)
316 w_s2[i] = SHA256_EXPAND (w_s2[i - 2], w_s2[i - 7], w_s2[i - 15], w_s2[i - 16]);
320 barrier (CLK_LOCAL_MEM_FENCE);
322 if (gid >= gid_max) return;
328 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
335 wordr0[0] = combs_buf[il_pos].i[0];
336 wordr0[1] = combs_buf[il_pos].i[1];
337 wordr0[2] = combs_buf[il_pos].i[2];
338 wordr0[3] = combs_buf[il_pos].i[3];
339 wordr1[0] = combs_buf[il_pos].i[4];
340 wordr1[1] = combs_buf[il_pos].i[5];
341 wordr1[2] = combs_buf[il_pos].i[6];
342 wordr1[3] = combs_buf[il_pos].i[7];
352 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
354 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
362 w0[0] = wordl0[0] | wordr0[0];
363 w0[1] = wordl0[1] | wordr0[1];
364 w0[2] = wordl0[2] | wordr0[2];
365 w0[3] = wordl0[3] | wordr0[3];
366 w1[0] = wordl1[0] | wordr1[0];
367 w1[1] = wordl1[1] | wordr1[1];
368 w1[2] = wordl1[2] | wordr1[2];
369 w1[3] = wordl1[3] | wordr1[3];
370 w2[0] = wordl2[0] | wordr2[0];
371 w2[1] = wordl2[1] | wordr2[1];
372 w2[2] = wordl2[2] | wordr2[2];
373 w2[3] = wordl2[3] | wordr2[3];
374 w3[0] = wordl3[0] | wordr3[0];
375 w3[1] = wordl3[1] | wordr3[1];
376 w3[2] = wordl3[2] | wordr3[2];
377 w3[3] = wordl3[3] | wordr3[3];
384 make_unicode (w0, w0_t, w1_t);
386 make_unicode (w1, w2_t, w3_t);
390 w_t[ 0] = swap32 (w0_t[0]);
391 w_t[ 1] = swap32 (w0_t[1]);
392 w_t[ 2] = swap32 (w0_t[2]);
393 w_t[ 3] = swap32 (w0_t[3]);
394 w_t[ 4] = swap32 (w1_t[0]);
395 w_t[ 5] = swap32 (w1_t[1]);
396 w_t[ 6] = swap32 (w1_t[2]);
397 w_t[ 7] = swap32 (w1_t[3]);
398 w_t[ 8] = swap32 (w2_t[0]);
399 w_t[ 9] = swap32 (w2_t[1]);
400 w_t[10] = swap32 (w2_t[2]);
401 w_t[11] = swap32 (w2_t[3]);
402 w_t[12] = swap32 (w3_t[0]);
403 w_t[13] = swap32 (w3_t[1]);
404 w_t[14] = swap32 (w3_t[2]);
405 w_t[15] = swap32 (w3_t[3]);
407 w_t[ 0] = w_t[ 0] >> 8;
408 w_t[ 1] = w_t[ 1] >> 8;
409 w_t[ 2] = w_t[ 2] >> 8;
410 w_t[ 3] = w_t[ 3] >> 8;
411 w_t[ 4] = w_t[ 4] >> 8;
412 w_t[ 5] = w_t[ 5] >> 8;
413 w_t[ 6] = w_t[ 6] >> 8;
414 w_t[ 7] = w_t[ 7] >> 8;
415 w_t[ 8] = w_t[ 8] >> 8;
416 w_t[ 9] = w_t[ 9] >> 8;
417 w_t[10] = w_t[10] >> 8;
418 w_t[11] = w_t[11] >> 8;
419 w_t[12] = w_t[12] >> 8;
420 w_t[13] = w_t[13] >> 8;
421 w_t[14] = w_t[14] >> 8;
422 w_t[15] = w_t[15] >> 8;
426 digest[0] = SHA256M_A;
427 digest[1] = SHA256M_B;
428 digest[2] = SHA256M_C;
429 digest[3] = SHA256M_D;
430 digest[4] = SHA256M_E;
431 digest[5] = SHA256M_F;
432 digest[6] = SHA256M_G;
433 digest[7] = SHA256M_H;
435 sha256_transform (digest, w_t); // 0 - 64
436 sha256_transform_z (digest); // 64 - 128
437 sha256_transform_z (digest); // 128 - 192
438 sha256_transform_z (digest); // 192 - 256
439 sha256_transform_z (digest); // 256 - 320
440 sha256_transform_z (digest); // 320 - 384
441 sha256_transform_z (digest); // 384 - 448
442 sha256_transform_s (digest, w_s1); // 448 - 512
443 sha256_transform_s (digest, w_s2); // 512 - 576
445 const u32 r0 = digest[3];
446 const u32 r1 = digest[7];
447 const u32 r2 = digest[2];
448 const u32 r3 = digest[6];
454 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
458 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
462 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
468 const u32 lid = get_local_id (0);
474 const u32 gid = get_global_id (0);
478 wordl0[0] = pws[gid].i[ 0];
479 wordl0[1] = pws[gid].i[ 1];
480 wordl0[2] = pws[gid].i[ 2];
481 wordl0[3] = pws[gid].i[ 3];
485 wordl1[0] = pws[gid].i[ 4];
486 wordl1[1] = pws[gid].i[ 5];
487 wordl1[2] = pws[gid].i[ 6];
488 wordl1[3] = pws[gid].i[ 7];
504 const u32 pw_l_len = pws[gid].pw_len;
506 if (combs_mode == COMBINATOR_MODE_BASE_RIGHT)
508 switch_buffer_by_offset (wordl0, wordl1, wordl2, wordl3, combs_buf[0].pw_len);
515 const u32 salt_buf0 = swap32 (salt_bufs[salt_pos].salt_buf[ 0]);
516 const u32 salt_buf1 = swap32 (salt_bufs[salt_pos].salt_buf[ 1]);
517 const u32 salt_buf2 = swap32 (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
523 const u32 search[4] =
525 digests_buf[digests_offset].digest_buf[DGST_R0],
526 digests_buf[digests_offset].digest_buf[DGST_R1],
527 digests_buf[digests_offset].digest_buf[DGST_R2],
528 digests_buf[digests_offset].digest_buf[DGST_R3]
532 * precompute final msg blocks
535 __local u32 w_s1[64];
536 __local u32 w_s2[64];
541 barrier (CLK_LOCAL_MEM_FENCE);
545 w_s1[15] = 0 | salt_buf0 >> 16;
548 for (int i = 16; i < 64; i++)
550 w_s1[i] = SHA256_EXPAND (w_s1[i - 2], w_s1[i - 7], w_s1[i - 15], w_s1[i - 16]);
555 w_s2[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
556 w_s2[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
557 w_s2[ 2] = salt_buf2 << 16 | 0;
558 w_s2[15] = (510 + 8) * 8;
561 for (int i = 16; i < 64; i++)
563 w_s2[i] = SHA256_EXPAND (w_s2[i - 2], w_s2[i - 7], w_s2[i - 15], w_s2[i - 16]);
567 barrier (CLK_LOCAL_MEM_FENCE);
569 if (gid >= gid_max) return;
575 for (u32 il_pos = 0; il_pos < combs_cnt; il_pos++)
582 wordr0[0] = combs_buf[il_pos].i[0];
583 wordr0[1] = combs_buf[il_pos].i[1];
584 wordr0[2] = combs_buf[il_pos].i[2];
585 wordr0[3] = combs_buf[il_pos].i[3];
586 wordr1[0] = combs_buf[il_pos].i[4];
587 wordr1[1] = combs_buf[il_pos].i[5];
588 wordr1[2] = combs_buf[il_pos].i[6];
589 wordr1[3] = combs_buf[il_pos].i[7];
599 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
601 switch_buffer_by_offset (wordr0, wordr1, wordr2, wordr3, pw_l_len);
609 w0[0] = wordl0[0] | wordr0[0];
610 w0[1] = wordl0[1] | wordr0[1];
611 w0[2] = wordl0[2] | wordr0[2];
612 w0[3] = wordl0[3] | wordr0[3];
613 w1[0] = wordl1[0] | wordr1[0];
614 w1[1] = wordl1[1] | wordr1[1];
615 w1[2] = wordl1[2] | wordr1[2];
616 w1[3] = wordl1[3] | wordr1[3];
617 w2[0] = wordl2[0] | wordr2[0];
618 w2[1] = wordl2[1] | wordr2[1];
619 w2[2] = wordl2[2] | wordr2[2];
620 w2[3] = wordl2[3] | wordr2[3];
621 w3[0] = wordl3[0] | wordr3[0];
622 w3[1] = wordl3[1] | wordr3[1];
623 w3[2] = wordl3[2] | wordr3[2];
624 w3[3] = wordl3[3] | wordr3[3];
631 make_unicode (w0, w0_t, w1_t);
633 make_unicode (w1, w2_t, w3_t);
637 w_t[ 0] = swap32 (w0_t[0]);
638 w_t[ 1] = swap32 (w0_t[1]);
639 w_t[ 2] = swap32 (w0_t[2]);
640 w_t[ 3] = swap32 (w0_t[3]);
641 w_t[ 4] = swap32 (w1_t[0]);
642 w_t[ 5] = swap32 (w1_t[1]);
643 w_t[ 6] = swap32 (w1_t[2]);
644 w_t[ 7] = swap32 (w1_t[3]);
645 w_t[ 8] = swap32 (w2_t[0]);
646 w_t[ 9] = swap32 (w2_t[1]);
647 w_t[10] = swap32 (w2_t[2]);
648 w_t[11] = swap32 (w2_t[3]);
649 w_t[12] = swap32 (w3_t[0]);
650 w_t[13] = swap32 (w3_t[1]);
651 w_t[14] = swap32 (w3_t[2]);
652 w_t[15] = swap32 (w3_t[3]);
654 w_t[ 0] = w_t[ 0] >> 8;
655 w_t[ 1] = w_t[ 1] >> 8;
656 w_t[ 2] = w_t[ 2] >> 8;
657 w_t[ 3] = w_t[ 3] >> 8;
658 w_t[ 4] = w_t[ 4] >> 8;
659 w_t[ 5] = w_t[ 5] >> 8;
660 w_t[ 6] = w_t[ 6] >> 8;
661 w_t[ 7] = w_t[ 7] >> 8;
662 w_t[ 8] = w_t[ 8] >> 8;
663 w_t[ 9] = w_t[ 9] >> 8;
664 w_t[10] = w_t[10] >> 8;
665 w_t[11] = w_t[11] >> 8;
666 w_t[12] = w_t[12] >> 8;
667 w_t[13] = w_t[13] >> 8;
668 w_t[14] = w_t[14] >> 8;
669 w_t[15] = w_t[15] >> 8;
673 digest[0] = SHA256M_A;
674 digest[1] = SHA256M_B;
675 digest[2] = SHA256M_C;
676 digest[3] = SHA256M_D;
677 digest[4] = SHA256M_E;
678 digest[5] = SHA256M_F;
679 digest[6] = SHA256M_G;
680 digest[7] = SHA256M_H;
682 sha256_transform (digest, w_t); // 0 - 64
683 sha256_transform_z (digest); // 64 - 128
684 sha256_transform_z (digest); // 128 - 192
685 sha256_transform_z (digest); // 192 - 256
686 sha256_transform_z (digest); // 256 - 320
687 sha256_transform_z (digest); // 320 - 384
688 sha256_transform_z (digest); // 384 - 448
689 sha256_transform_s (digest, w_s1); // 448 - 512
690 sha256_transform_s (digest, w_s2); // 512 - 576
692 const u32 r0 = digest[3];
693 const u32 r1 = digest[7];
694 const u32 r2 = digest[2];
695 const u32 r3 = digest[6];
701 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)
705 __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 combs_cnt, const u32 digests_cnt, const u32 digests_offset, const u32 combs_mode, const u32 gid_max)