2 * Authors.....: Jens Steube <jens.steube@gmail.com>
3 * Gabriele Gristina <matrix@hashcat.net>
12 #include "include/constants.h"
13 #include "include/kernel_vendor.h"
20 #include "include/kernel_functions.c"
21 #include "OpenCL/types_ocl.c"
22 #include "OpenCL/common.c"
23 #include "OpenCL/simd.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 #define SHA256_S0_S(x) (rotl32_S ((x), 25u) ^ rotl32_S ((x), 14u) ^ SHIFT_RIGHT_32 ((x), 3u))
46 #define SHA256_S1_S(x) (rotl32_S ((x), 15u) ^ rotl32_S ((x), 13u) ^ SHIFT_RIGHT_32 ((x), 10u))
48 #define SHA256_EXPAND_S(x,y,z,w) (SHA256_S1_S (x) + y + SHA256_S0_S (z) + w)
50 void sha256_transform (u32x digest[8], const u32x w[16])
78 #define ROUND_EXPAND() \
80 w0_t = SHA256_EXPAND (we_t, w9_t, w1_t, w0_t); \
81 w1_t = SHA256_EXPAND (wf_t, wa_t, w2_t, w1_t); \
82 w2_t = SHA256_EXPAND (w0_t, wb_t, w3_t, w2_t); \
83 w3_t = SHA256_EXPAND (w1_t, wc_t, w4_t, w3_t); \
84 w4_t = SHA256_EXPAND (w2_t, wd_t, w5_t, w4_t); \
85 w5_t = SHA256_EXPAND (w3_t, we_t, w6_t, w5_t); \
86 w6_t = SHA256_EXPAND (w4_t, wf_t, w7_t, w6_t); \
87 w7_t = SHA256_EXPAND (w5_t, w0_t, w8_t, w7_t); \
88 w8_t = SHA256_EXPAND (w6_t, w1_t, w9_t, w8_t); \
89 w9_t = SHA256_EXPAND (w7_t, w2_t, wa_t, w9_t); \
90 wa_t = SHA256_EXPAND (w8_t, w3_t, wb_t, wa_t); \
91 wb_t = SHA256_EXPAND (w9_t, w4_t, wc_t, wb_t); \
92 wc_t = SHA256_EXPAND (wa_t, w5_t, wd_t, wc_t); \
93 wd_t = SHA256_EXPAND (wb_t, w6_t, we_t, wd_t); \
94 we_t = SHA256_EXPAND (wc_t, w7_t, wf_t, we_t); \
95 wf_t = SHA256_EXPAND (wd_t, w8_t, w0_t, wf_t); \
98 #define ROUND_STEP(i) \
100 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha256[i + 0]); \
101 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha256[i + 1]); \
102 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha256[i + 2]); \
103 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha256[i + 3]); \
104 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha256[i + 4]); \
105 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha256[i + 5]); \
106 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha256[i + 6]); \
107 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha256[i + 7]); \
108 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha256[i + 8]); \
109 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha256[i + 9]); \
110 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha256[i + 10]); \
111 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha256[i + 11]); \
112 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha256[i + 12]); \
113 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha256[i + 13]); \
114 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, we_t, k_sha256[i + 14]); \
115 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha256[i + 15]); \
123 for (int i = 16; i < 64; i += 16)
125 ROUND_EXPAND (); ROUND_STEP (i);
138 void sha256_transform_z (u32x digest[8])
149 #define ROUND_STEP_Z(i) \
151 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, 0, k_sha256[i + 0]); \
152 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, 0, k_sha256[i + 1]); \
153 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, 0, k_sha256[i + 2]); \
154 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, 0, k_sha256[i + 3]); \
155 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, 0, k_sha256[i + 4]); \
156 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, 0, k_sha256[i + 5]); \
157 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, 0, k_sha256[i + 6]); \
158 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, 0, k_sha256[i + 7]); \
159 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, 0, k_sha256[i + 8]); \
160 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, 0, k_sha256[i + 9]); \
161 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, 0, k_sha256[i + 10]); \
162 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, 0, k_sha256[i + 11]); \
163 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, 0, k_sha256[i + 12]); \
164 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, 0, k_sha256[i + 13]); \
165 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, 0, k_sha256[i + 14]); \
166 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, 0, k_sha256[i + 15]); \
174 for (int i = 16; i < 64; i += 16)
189 void sha256_transform_s (u32x digest[8], __local u32 *w)
200 #define ROUND_STEP_S(i) \
202 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w[i + 0], k_sha256[i + 0]); \
203 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w[i + 1], k_sha256[i + 1]); \
204 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w[i + 2], k_sha256[i + 2]); \
205 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w[i + 3], k_sha256[i + 3]); \
206 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w[i + 4], k_sha256[i + 4]); \
207 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w[i + 5], k_sha256[i + 5]); \
208 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w[i + 6], k_sha256[i + 6]); \
209 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w[i + 7], k_sha256[i + 7]); \
210 SHA256_STEP (SHA256_F0o, SHA256_F1o, a, b, c, d, e, f, g, h, w[i + 8], k_sha256[i + 8]); \
211 SHA256_STEP (SHA256_F0o, SHA256_F1o, h, a, b, c, d, e, f, g, w[i + 9], k_sha256[i + 9]); \
212 SHA256_STEP (SHA256_F0o, SHA256_F1o, g, h, a, b, c, d, e, f, w[i + 10], k_sha256[i + 10]); \
213 SHA256_STEP (SHA256_F0o, SHA256_F1o, f, g, h, a, b, c, d, e, w[i + 11], k_sha256[i + 11]); \
214 SHA256_STEP (SHA256_F0o, SHA256_F1o, e, f, g, h, a, b, c, d, w[i + 12], k_sha256[i + 12]); \
215 SHA256_STEP (SHA256_F0o, SHA256_F1o, d, e, f, g, h, a, b, c, w[i + 13], k_sha256[i + 13]); \
216 SHA256_STEP (SHA256_F0o, SHA256_F1o, c, d, e, f, g, h, a, b, w[i + 14], k_sha256[i + 14]); \
217 SHA256_STEP (SHA256_F0o, SHA256_F1o, b, c, d, e, f, g, h, a, w[i + 15], k_sha256[i + 15]); \
225 for (int i = 16; i < 64; i += 16)
240 __kernel void m08000_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)
246 const u32 gid = get_global_id (0);
247 const u32 lid = get_local_id (0);
248 const u32 lsz = get_local_size (0);
254 const u32 salt_buf0 = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
255 const u32 salt_buf1 = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
256 const u32 salt_buf2 = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
259 * precompute final msg blocks
262 __local u32 w_s1[64];
263 __local u32 w_s2[64];
265 for (u32 i = lid; i < 64; i += lsz)
271 barrier (CLK_LOCAL_MEM_FENCE);
275 w_s1[15] = 0 | salt_buf0 >> 16;
280 for (int i = 16; i < 64; i++)
282 w_s1[i] = SHA256_EXPAND_S (w_s1[i - 2], w_s1[i - 7], w_s1[i - 15], w_s1[i - 16]);
285 w_s2[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
286 w_s2[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
287 w_s2[ 2] = salt_buf2 << 16 | 0;
288 w_s2[15] = (510 + 8) * 8;
293 for (int i = 16; i < 64; i++)
295 w_s2[i] = SHA256_EXPAND_S (w_s2[i - 2], w_s2[i - 7], w_s2[i - 15], w_s2[i - 16]);
299 barrier (CLK_LOCAL_MEM_FENCE);
301 if (gid >= gid_max) return;
310 pw_buf0[0] = pws[gid].i[0];
311 pw_buf0[1] = pws[gid].i[1];
312 pw_buf0[2] = pws[gid].i[2];
313 pw_buf0[3] = pws[gid].i[3];
314 pw_buf1[0] = pws[gid].i[4];
315 pw_buf1[1] = pws[gid].i[5];
316 pw_buf1[2] = pws[gid].i[6];
317 pw_buf1[3] = pws[gid].i[7];
319 const u32 pw_l_len = pws[gid].pw_len;
325 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
327 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
329 const u32x pw_len = pw_l_len + pw_r_len;
332 * concat password candidate
335 u32x wordl0[4] = { 0 };
336 u32x wordl1[4] = { 0 };
337 u32x wordl2[4] = { 0 };
338 u32x wordl3[4] = { 0 };
340 wordl0[0] = pw_buf0[0];
341 wordl0[1] = pw_buf0[1];
342 wordl0[2] = pw_buf0[2];
343 wordl0[3] = pw_buf0[3];
344 wordl1[0] = pw_buf1[0];
345 wordl1[1] = pw_buf1[1];
346 wordl1[2] = pw_buf1[2];
347 wordl1[3] = pw_buf1[3];
349 u32x wordr0[4] = { 0 };
350 u32x wordr1[4] = { 0 };
351 u32x wordr2[4] = { 0 };
352 u32x wordr3[4] = { 0 };
354 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
355 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
356 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
357 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
358 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
359 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
360 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
361 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
363 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
365 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
369 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
375 w0[0] = wordl0[0] | wordr0[0];
376 w0[1] = wordl0[1] | wordr0[1];
377 w0[2] = wordl0[2] | wordr0[2];
378 w0[3] = wordl0[3] | wordr0[3];
379 w1[0] = wordl1[0] | wordr1[0];
380 w1[1] = wordl1[1] | wordr1[1];
381 w1[2] = wordl1[2] | wordr1[2];
382 w1[3] = wordl1[3] | wordr1[3];
393 make_unicode (w0, w0_t, w1_t);
394 make_unicode (w1, w2_t, w3_t);
398 w_t[ 0] = swap32 (w0_t[0]);
399 w_t[ 1] = swap32 (w0_t[1]);
400 w_t[ 2] = swap32 (w0_t[2]);
401 w_t[ 3] = swap32 (w0_t[3]);
402 w_t[ 4] = swap32 (w1_t[0]);
403 w_t[ 5] = swap32 (w1_t[1]);
404 w_t[ 6] = swap32 (w1_t[2]);
405 w_t[ 7] = swap32 (w1_t[3]);
406 w_t[ 8] = swap32 (w2_t[0]);
407 w_t[ 9] = swap32 (w2_t[1]);
408 w_t[10] = swap32 (w2_t[2]);
409 w_t[11] = swap32 (w2_t[3]);
410 w_t[12] = swap32 (w3_t[0]);
411 w_t[13] = swap32 (w3_t[1]);
412 w_t[14] = swap32 (w3_t[2]);
413 w_t[15] = swap32 (w3_t[3]);
415 w_t[ 0] = w_t[ 0] >> 8;
416 w_t[ 1] = w_t[ 1] >> 8;
417 w_t[ 2] = w_t[ 2] >> 8;
418 w_t[ 3] = w_t[ 3] >> 8;
419 w_t[ 4] = w_t[ 4] >> 8;
420 w_t[ 5] = w_t[ 5] >> 8;
421 w_t[ 6] = w_t[ 6] >> 8;
422 w_t[ 7] = w_t[ 7] >> 8;
423 w_t[ 8] = w_t[ 8] >> 8;
424 w_t[ 9] = w_t[ 9] >> 8;
425 w_t[10] = w_t[10] >> 8;
426 w_t[11] = w_t[11] >> 8;
427 w_t[12] = w_t[12] >> 8;
428 w_t[13] = w_t[13] >> 8;
429 w_t[14] = w_t[14] >> 8;
430 w_t[15] = w_t[15] >> 8;
434 digest[0] = SHA256M_A;
435 digest[1] = SHA256M_B;
436 digest[2] = SHA256M_C;
437 digest[3] = SHA256M_D;
438 digest[4] = SHA256M_E;
439 digest[5] = SHA256M_F;
440 digest[6] = SHA256M_G;
441 digest[7] = SHA256M_H;
443 sha256_transform (digest, w_t); // 0 - 64
444 sha256_transform_z (digest); // 64 - 128
445 sha256_transform_z (digest); // 128 - 192
446 sha256_transform_z (digest); // 192 - 256
447 sha256_transform_z (digest); // 256 - 320
448 sha256_transform_z (digest); // 320 - 384
449 sha256_transform_z (digest); // 384 - 448
450 sha256_transform_s (digest, w_s1); // 448 - 512
451 sha256_transform_s (digest, w_s2); // 512 - 576
453 COMPARE_M_SIMD (digest[3], digest[7], digest[2], digest[6]);
457 __kernel void m08000_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)
461 __kernel void m08000_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)
465 __kernel void m08000_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)
471 const u32 gid = get_global_id (0);
472 const u32 lid = get_local_id (0);
473 const u32 lsz = get_local_size (0);
479 const u32 salt_buf0 = swap32_S (salt_bufs[salt_pos].salt_buf[ 0]);
480 const u32 salt_buf1 = swap32_S (salt_bufs[salt_pos].salt_buf[ 1]);
481 const u32 salt_buf2 = swap32_S (salt_bufs[salt_pos].salt_buf[ 2]); // 0x80
484 * precompute final msg blocks
487 __local u32 w_s1[64];
488 __local u32 w_s2[64];
490 for (u32 i = lid; i < 64; i += lsz)
496 barrier (CLK_LOCAL_MEM_FENCE);
500 w_s1[15] = 0 | salt_buf0 >> 16;
505 for (int i = 16; i < 64; i++)
507 w_s1[i] = SHA256_EXPAND_S (w_s1[i - 2], w_s1[i - 7], w_s1[i - 15], w_s1[i - 16]);
510 w_s2[ 0] = salt_buf0 << 16 | salt_buf1 >> 16;
511 w_s2[ 1] = salt_buf1 << 16 | salt_buf2 >> 16;
512 w_s2[ 2] = salt_buf2 << 16 | 0;
513 w_s2[15] = (510 + 8) * 8;
518 for (int i = 16; i < 64; i++)
520 w_s2[i] = SHA256_EXPAND_S (w_s2[i - 2], w_s2[i - 7], w_s2[i - 15], w_s2[i - 16]);
524 barrier (CLK_LOCAL_MEM_FENCE);
526 if (gid >= gid_max) return;
535 pw_buf0[0] = pws[gid].i[0];
536 pw_buf0[1] = pws[gid].i[1];
537 pw_buf0[2] = pws[gid].i[2];
538 pw_buf0[3] = pws[gid].i[3];
539 pw_buf1[0] = pws[gid].i[4];
540 pw_buf1[1] = pws[gid].i[5];
541 pw_buf1[2] = pws[gid].i[6];
542 pw_buf1[3] = pws[gid].i[7];
544 const u32 pw_l_len = pws[gid].pw_len;
550 const u32 search[4] =
552 digests_buf[digests_offset].digest_buf[DGST_R0],
553 digests_buf[digests_offset].digest_buf[DGST_R1],
554 digests_buf[digests_offset].digest_buf[DGST_R2],
555 digests_buf[digests_offset].digest_buf[DGST_R3]
562 for (u32 il_pos = 0; il_pos < il_cnt; il_pos += VECT_SIZE)
564 const u32x pw_r_len = pwlenx_create_combt (combs_buf, il_pos);
566 const u32x pw_len = pw_l_len + pw_r_len;
569 * concat password candidate
572 u32x wordl0[4] = { 0 };
573 u32x wordl1[4] = { 0 };
574 u32x wordl2[4] = { 0 };
575 u32x wordl3[4] = { 0 };
577 wordl0[0] = pw_buf0[0];
578 wordl0[1] = pw_buf0[1];
579 wordl0[2] = pw_buf0[2];
580 wordl0[3] = pw_buf0[3];
581 wordl1[0] = pw_buf1[0];
582 wordl1[1] = pw_buf1[1];
583 wordl1[2] = pw_buf1[2];
584 wordl1[3] = pw_buf1[3];
586 u32x wordr0[4] = { 0 };
587 u32x wordr1[4] = { 0 };
588 u32x wordr2[4] = { 0 };
589 u32x wordr3[4] = { 0 };
591 wordr0[0] = ix_create_combt (combs_buf, il_pos, 0);
592 wordr0[1] = ix_create_combt (combs_buf, il_pos, 1);
593 wordr0[2] = ix_create_combt (combs_buf, il_pos, 2);
594 wordr0[3] = ix_create_combt (combs_buf, il_pos, 3);
595 wordr1[0] = ix_create_combt (combs_buf, il_pos, 4);
596 wordr1[1] = ix_create_combt (combs_buf, il_pos, 5);
597 wordr1[2] = ix_create_combt (combs_buf, il_pos, 6);
598 wordr1[3] = ix_create_combt (combs_buf, il_pos, 7);
600 if (combs_mode == COMBINATOR_MODE_BASE_LEFT)
602 switch_buffer_by_offset_le_VV (wordr0, wordr1, wordr2, wordr3, pw_l_len);
606 switch_buffer_by_offset_le_VV (wordl0, wordl1, wordl2, wordl3, pw_r_len);
612 w0[0] = wordl0[0] | wordr0[0];
613 w0[1] = wordl0[1] | wordr0[1];
614 w0[2] = wordl0[2] | wordr0[2];
615 w0[3] = wordl0[3] | wordr0[3];
616 w1[0] = wordl1[0] | wordr1[0];
617 w1[1] = wordl1[1] | wordr1[1];
618 w1[2] = wordl1[2] | wordr1[2];
619 w1[3] = wordl1[3] | wordr1[3];
630 make_unicode (w0, w0_t, w1_t);
631 make_unicode (w1, w2_t, w3_t);
635 w_t[ 0] = swap32 (w0_t[0]);
636 w_t[ 1] = swap32 (w0_t[1]);
637 w_t[ 2] = swap32 (w0_t[2]);
638 w_t[ 3] = swap32 (w0_t[3]);
639 w_t[ 4] = swap32 (w1_t[0]);
640 w_t[ 5] = swap32 (w1_t[1]);
641 w_t[ 6] = swap32 (w1_t[2]);
642 w_t[ 7] = swap32 (w1_t[3]);
643 w_t[ 8] = swap32 (w2_t[0]);
644 w_t[ 9] = swap32 (w2_t[1]);
645 w_t[10] = swap32 (w2_t[2]);
646 w_t[11] = swap32 (w2_t[3]);
647 w_t[12] = swap32 (w3_t[0]);
648 w_t[13] = swap32 (w3_t[1]);
649 w_t[14] = swap32 (w3_t[2]);
650 w_t[15] = swap32 (w3_t[3]);
652 w_t[ 0] = w_t[ 0] >> 8;
653 w_t[ 1] = w_t[ 1] >> 8;
654 w_t[ 2] = w_t[ 2] >> 8;
655 w_t[ 3] = w_t[ 3] >> 8;
656 w_t[ 4] = w_t[ 4] >> 8;
657 w_t[ 5] = w_t[ 5] >> 8;
658 w_t[ 6] = w_t[ 6] >> 8;
659 w_t[ 7] = w_t[ 7] >> 8;
660 w_t[ 8] = w_t[ 8] >> 8;
661 w_t[ 9] = w_t[ 9] >> 8;
662 w_t[10] = w_t[10] >> 8;
663 w_t[11] = w_t[11] >> 8;
664 w_t[12] = w_t[12] >> 8;
665 w_t[13] = w_t[13] >> 8;
666 w_t[14] = w_t[14] >> 8;
667 w_t[15] = w_t[15] >> 8;
671 digest[0] = SHA256M_A;
672 digest[1] = SHA256M_B;
673 digest[2] = SHA256M_C;
674 digest[3] = SHA256M_D;
675 digest[4] = SHA256M_E;
676 digest[5] = SHA256M_F;
677 digest[6] = SHA256M_G;
678 digest[7] = SHA256M_H;
680 sha256_transform (digest, w_t); // 0 - 64
681 sha256_transform_z (digest); // 64 - 128
682 sha256_transform_z (digest); // 128 - 192
683 sha256_transform_z (digest); // 192 - 256
684 sha256_transform_z (digest); // 256 - 320
685 sha256_transform_z (digest); // 320 - 384
686 sha256_transform_z (digest); // 384 - 448
687 sha256_transform_s (digest, w_s1); // 448 - 512
688 sha256_transform_s (digest, w_s2); // 512 - 576
690 COMPARE_S_SIMD (digest[3], digest[7], digest[2], digest[6]);
694 __kernel void m08000_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)
698 __kernel void m08000_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)