2 * Author......: Jens Steube <jens.steube@gmail.com>
8 #include "include/constants.h"
9 #include "include/kernel_vendor.h"
24 #include "include/kernel_functions.c"
26 #include "common_nv.c"
28 #include "gpu_aes256_nv.c"
29 #include "gpu_twofish256_nv.c"
30 #include "gpu_serpent256_nv.c"
32 __device__ __constant__ u64 k_sha512[80] =
34 SHA512C00, SHA512C01, SHA512C02, SHA512C03,
35 SHA512C04, SHA512C05, SHA512C06, SHA512C07,
36 SHA512C08, SHA512C09, SHA512C0a, SHA512C0b,
37 SHA512C0c, SHA512C0d, SHA512C0e, SHA512C0f,
38 SHA512C10, SHA512C11, SHA512C12, SHA512C13,
39 SHA512C14, SHA512C15, SHA512C16, SHA512C17,
40 SHA512C18, SHA512C19, SHA512C1a, SHA512C1b,
41 SHA512C1c, SHA512C1d, SHA512C1e, SHA512C1f,
42 SHA512C20, SHA512C21, SHA512C22, SHA512C23,
43 SHA512C24, SHA512C25, SHA512C26, SHA512C27,
44 SHA512C28, SHA512C29, SHA512C2a, SHA512C2b,
45 SHA512C2c, SHA512C2d, SHA512C2e, SHA512C2f,
46 SHA512C30, SHA512C31, SHA512C32, SHA512C33,
47 SHA512C34, SHA512C35, SHA512C36, SHA512C37,
48 SHA512C38, SHA512C39, SHA512C3a, SHA512C3b,
49 SHA512C3c, SHA512C3d, SHA512C3e, SHA512C3f,
50 SHA512C40, SHA512C41, SHA512C42, SHA512C43,
51 SHA512C44, SHA512C45, SHA512C46, SHA512C47,
52 SHA512C48, SHA512C49, SHA512C4a, SHA512C4b,
53 SHA512C4c, SHA512C4d, SHA512C4e, SHA512C4f,
56 __device__ static void sha512_transform (const u64 w[16], u64 dgst[8])
84 #define ROUND_EXPAND() \
86 w0_t = SHA512_EXPAND (we_t, w9_t, w1_t, w0_t); \
87 w1_t = SHA512_EXPAND (wf_t, wa_t, w2_t, w1_t); \
88 w2_t = SHA512_EXPAND (w0_t, wb_t, w3_t, w2_t); \
89 w3_t = SHA512_EXPAND (w1_t, wc_t, w4_t, w3_t); \
90 w4_t = SHA512_EXPAND (w2_t, wd_t, w5_t, w4_t); \
91 w5_t = SHA512_EXPAND (w3_t, we_t, w6_t, w5_t); \
92 w6_t = SHA512_EXPAND (w4_t, wf_t, w7_t, w6_t); \
93 w7_t = SHA512_EXPAND (w5_t, w0_t, w8_t, w7_t); \
94 w8_t = SHA512_EXPAND (w6_t, w1_t, w9_t, w8_t); \
95 w9_t = SHA512_EXPAND (w7_t, w2_t, wa_t, w9_t); \
96 wa_t = SHA512_EXPAND (w8_t, w3_t, wb_t, wa_t); \
97 wb_t = SHA512_EXPAND (w9_t, w4_t, wc_t, wb_t); \
98 wc_t = SHA512_EXPAND (wa_t, w5_t, wd_t, wc_t); \
99 wd_t = SHA512_EXPAND (wb_t, w6_t, we_t, wd_t); \
100 we_t = SHA512_EXPAND (wc_t, w7_t, wf_t, we_t); \
101 wf_t = SHA512_EXPAND (wd_t, w8_t, w0_t, wf_t); \
104 #define ROUND_STEP(i) \
106 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w0_t, k_sha512[i + 0]); \
107 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w1_t, k_sha512[i + 1]); \
108 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, w2_t, k_sha512[i + 2]); \
109 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, w3_t, k_sha512[i + 3]); \
110 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, w4_t, k_sha512[i + 4]); \
111 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, w5_t, k_sha512[i + 5]); \
112 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, w6_t, k_sha512[i + 6]); \
113 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, w7_t, k_sha512[i + 7]); \
114 SHA512_STEP (SHA512_F0o, SHA512_F1o, a, b, c, d, e, f, g, h, w8_t, k_sha512[i + 8]); \
115 SHA512_STEP (SHA512_F0o, SHA512_F1o, h, a, b, c, d, e, f, g, w9_t, k_sha512[i + 9]); \
116 SHA512_STEP (SHA512_F0o, SHA512_F1o, g, h, a, b, c, d, e, f, wa_t, k_sha512[i + 10]); \
117 SHA512_STEP (SHA512_F0o, SHA512_F1o, f, g, h, a, b, c, d, e, wb_t, k_sha512[i + 11]); \
118 SHA512_STEP (SHA512_F0o, SHA512_F1o, e, f, g, h, a, b, c, d, wc_t, k_sha512[i + 12]); \
119 SHA512_STEP (SHA512_F0o, SHA512_F1o, d, e, f, g, h, a, b, c, wd_t, k_sha512[i + 13]); \
120 SHA512_STEP (SHA512_F0o, SHA512_F1o, c, d, e, f, g, h, a, b, we_t, k_sha512[i + 14]); \
121 SHA512_STEP (SHA512_F0o, SHA512_F1o, b, c, d, e, f, g, h, a, wf_t, k_sha512[i + 15]); \
126 for (int i = 16; i < 80; i += 16)
128 ROUND_EXPAND (); ROUND_STEP (i);
141 __device__ static void hmac_run (const u64 w1[16], const u64 ipad[8], const u64 opad[8], u64 dgst[8])
152 sha512_transform (w1, dgst);
164 w[ 8] = 0x8000000000000000;
171 w[15] = (128 + 64) * 8;
182 sha512_transform (w, dgst);
185 __device__ static void hmac_init (u64 w[16], u64 ipad[8], u64 opad[8])
187 w[ 0] ^= 0x3636363636363636;
188 w[ 1] ^= 0x3636363636363636;
189 w[ 2] ^= 0x3636363636363636;
190 w[ 3] ^= 0x3636363636363636;
191 w[ 4] ^= 0x3636363636363636;
192 w[ 5] ^= 0x3636363636363636;
193 w[ 6] ^= 0x3636363636363636;
194 w[ 7] ^= 0x3636363636363636;
195 w[ 8] ^= 0x3636363636363636;
196 w[ 9] ^= 0x3636363636363636;
197 w[10] ^= 0x3636363636363636;
198 w[11] ^= 0x3636363636363636;
199 w[12] ^= 0x3636363636363636;
200 w[13] ^= 0x3636363636363636;
201 w[14] ^= 0x3636363636363636;
202 w[15] ^= 0x3636363636363636;
213 sha512_transform (w, ipad);
215 w[ 0] ^= 0x6a6a6a6a6a6a6a6a;
216 w[ 1] ^= 0x6a6a6a6a6a6a6a6a;
217 w[ 2] ^= 0x6a6a6a6a6a6a6a6a;
218 w[ 3] ^= 0x6a6a6a6a6a6a6a6a;
219 w[ 4] ^= 0x6a6a6a6a6a6a6a6a;
220 w[ 5] ^= 0x6a6a6a6a6a6a6a6a;
221 w[ 6] ^= 0x6a6a6a6a6a6a6a6a;
222 w[ 7] ^= 0x6a6a6a6a6a6a6a6a;
223 w[ 8] ^= 0x6a6a6a6a6a6a6a6a;
224 w[ 9] ^= 0x6a6a6a6a6a6a6a6a;
225 w[10] ^= 0x6a6a6a6a6a6a6a6a;
226 w[11] ^= 0x6a6a6a6a6a6a6a6a;
227 w[12] ^= 0x6a6a6a6a6a6a6a6a;
228 w[13] ^= 0x6a6a6a6a6a6a6a6a;
229 w[14] ^= 0x6a6a6a6a6a6a6a6a;
230 w[15] ^= 0x6a6a6a6a6a6a6a6a;
241 sha512_transform (w, opad);
244 __device__ static u32 u8add (const u32 a, const u32 b)
246 const u32 a1 = (a >> 0) & 0xff;
247 const u32 a2 = (a >> 8) & 0xff;
248 const u32 a3 = (a >> 16) & 0xff;
249 const u32 a4 = (a >> 24) & 0xff;
251 const u32 b1 = (b >> 0) & 0xff;
252 const u32 b2 = (b >> 8) & 0xff;
253 const u32 b3 = (b >> 16) & 0xff;
254 const u32 b4 = (b >> 24) & 0xff;
256 const u32 r1 = (a1 + b1) & 0xff;
257 const u32 r2 = (a2 + b2) & 0xff;
258 const u32 r3 = (a3 + b3) & 0xff;
259 const u32 r4 = (a4 + b4) & 0xff;
261 const u32 r = r1 << 0
269 extern "C" __global__ void __launch_bounds__ (256, 1) m06221_init (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, tc64_tmp_t *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const tc_t *esalt_bufs, u32 *d_return_buf, 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)
275 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
277 if (gid >= gid_max) return;
281 w0[0] = pws[gid].i[ 0];
282 w0[1] = pws[gid].i[ 1];
283 w0[2] = pws[gid].i[ 2];
284 w0[3] = pws[gid].i[ 3];
288 w1[0] = pws[gid].i[ 4];
289 w1[1] = pws[gid].i[ 5];
290 w1[2] = pws[gid].i[ 6];
291 w1[3] = pws[gid].i[ 7];
295 w2[0] = pws[gid].i[ 8];
296 w2[1] = pws[gid].i[ 9];
297 w2[2] = pws[gid].i[10];
298 w2[3] = pws[gid].i[11];
302 w3[0] = pws[gid].i[12];
303 w3[1] = pws[gid].i[13];
304 w3[2] = pws[gid].i[14];
305 w3[3] = pws[gid].i[15];
311 w0[0] = u8add (w0[0], esalt_bufs[salt_pos].keyfile_buf[ 0]);
312 w0[1] = u8add (w0[1], esalt_bufs[salt_pos].keyfile_buf[ 1]);
313 w0[2] = u8add (w0[2], esalt_bufs[salt_pos].keyfile_buf[ 2]);
314 w0[3] = u8add (w0[3], esalt_bufs[salt_pos].keyfile_buf[ 3]);
315 w1[0] = u8add (w1[0], esalt_bufs[salt_pos].keyfile_buf[ 4]);
316 w1[1] = u8add (w1[1], esalt_bufs[salt_pos].keyfile_buf[ 5]);
317 w1[2] = u8add (w1[2], esalt_bufs[salt_pos].keyfile_buf[ 6]);
318 w1[3] = u8add (w1[3], esalt_bufs[salt_pos].keyfile_buf[ 7]);
319 w2[0] = u8add (w2[0], esalt_bufs[salt_pos].keyfile_buf[ 8]);
320 w2[1] = u8add (w2[1], esalt_bufs[salt_pos].keyfile_buf[ 9]);
321 w2[2] = u8add (w2[2], esalt_bufs[salt_pos].keyfile_buf[10]);
322 w2[3] = u8add (w2[3], esalt_bufs[salt_pos].keyfile_buf[11]);
323 w3[0] = u8add (w3[0], esalt_bufs[salt_pos].keyfile_buf[12]);
324 w3[1] = u8add (w3[1], esalt_bufs[salt_pos].keyfile_buf[13]);
325 w3[2] = u8add (w3[2], esalt_bufs[salt_pos].keyfile_buf[14]);
326 w3[3] = u8add (w3[3], esalt_bufs[salt_pos].keyfile_buf[15]);
336 salt_buf[ 0] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 0])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 1]);
337 salt_buf[ 1] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 2])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 3]);
338 salt_buf[ 2] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 4])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 5]);
339 salt_buf[ 3] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 6])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 7]);
340 salt_buf[ 4] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 8])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[ 9]);
341 salt_buf[ 5] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[10])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[11]);
342 salt_buf[ 6] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[12])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[13]);
343 salt_buf[ 7] = ((u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[14])) << 32 | (u64) swap_workaround (esalt_bufs[salt_pos].salt_buf[15]);
351 salt_buf[15] = (128 + 64 + 4) * 8;
353 const u32 truecrypt_mdlen = salt_bufs[0].truecrypt_mdlen;
357 w[ 0] = ((u64) swap_workaround (w0[0])) << 32 | (u64) swap_workaround (w0[1]);
358 w[ 1] = ((u64) swap_workaround (w0[2])) << 32 | (u64) swap_workaround (w0[3]);
359 w[ 2] = ((u64) swap_workaround (w1[0])) << 32 | (u64) swap_workaround (w1[1]);
360 w[ 3] = ((u64) swap_workaround (w1[2])) << 32 | (u64) swap_workaround (w1[3]);
361 w[ 4] = ((u64) swap_workaround (w2[0])) << 32 | (u64) swap_workaround (w2[1]);
362 w[ 5] = ((u64) swap_workaround (w2[2])) << 32 | (u64) swap_workaround (w2[3]);
363 w[ 6] = ((u64) swap_workaround (w3[0])) << 32 | (u64) swap_workaround (w3[1]);
364 w[ 7] = ((u64) swap_workaround (w3[2])) << 32 | (u64) swap_workaround (w3[3]);
377 hmac_init (w, ipad, opad);
379 tmps[gid].ipad[0] = ipad[0];
380 tmps[gid].ipad[1] = ipad[1];
381 tmps[gid].ipad[2] = ipad[2];
382 tmps[gid].ipad[3] = ipad[3];
383 tmps[gid].ipad[4] = ipad[4];
384 tmps[gid].ipad[5] = ipad[5];
385 tmps[gid].ipad[6] = ipad[6];
386 tmps[gid].ipad[7] = ipad[7];
388 tmps[gid].opad[0] = opad[0];
389 tmps[gid].opad[1] = opad[1];
390 tmps[gid].opad[2] = opad[2];
391 tmps[gid].opad[3] = opad[3];
392 tmps[gid].opad[4] = opad[4];
393 tmps[gid].opad[5] = opad[5];
394 tmps[gid].opad[6] = opad[6];
395 tmps[gid].opad[7] = opad[7];
397 for (u32 i = 0, j = 1; i < (truecrypt_mdlen / 8 / 8); i += 8, j += 1)
399 salt_buf[8] = (u64) j << 32 | (u64) 0x80000000;
403 hmac_run (salt_buf, ipad, opad, dgst);
405 tmps[gid].dgst[i + 0] = dgst[0];
406 tmps[gid].dgst[i + 1] = dgst[1];
407 tmps[gid].dgst[i + 2] = dgst[2];
408 tmps[gid].dgst[i + 3] = dgst[3];
409 tmps[gid].dgst[i + 4] = dgst[4];
410 tmps[gid].dgst[i + 5] = dgst[5];
411 tmps[gid].dgst[i + 6] = dgst[6];
412 tmps[gid].dgst[i + 7] = dgst[7];
414 tmps[gid].out[i + 0] = dgst[0];
415 tmps[gid].out[i + 1] = dgst[1];
416 tmps[gid].out[i + 2] = dgst[2];
417 tmps[gid].out[i + 3] = dgst[3];
418 tmps[gid].out[i + 4] = dgst[4];
419 tmps[gid].out[i + 5] = dgst[5];
420 tmps[gid].out[i + 6] = dgst[6];
421 tmps[gid].out[i + 7] = dgst[7];
425 extern "C" __global__ void __launch_bounds__ (256, 1) m06221_loop (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, tc64_tmp_t *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const tc_t *esalt_bufs, u32 *d_return_buf, 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)
427 const u32 truecrypt_mdlen = salt_bufs[0].truecrypt_mdlen;
429 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
431 if (gid >= gid_max) return;
435 ipad[0] = tmps[gid].ipad[0];
436 ipad[1] = tmps[gid].ipad[1];
437 ipad[2] = tmps[gid].ipad[2];
438 ipad[3] = tmps[gid].ipad[3];
439 ipad[4] = tmps[gid].ipad[4];
440 ipad[5] = tmps[gid].ipad[5];
441 ipad[6] = tmps[gid].ipad[6];
442 ipad[7] = tmps[gid].ipad[7];
446 opad[0] = tmps[gid].opad[0];
447 opad[1] = tmps[gid].opad[1];
448 opad[2] = tmps[gid].opad[2];
449 opad[3] = tmps[gid].opad[3];
450 opad[4] = tmps[gid].opad[4];
451 opad[5] = tmps[gid].opad[5];
452 opad[6] = tmps[gid].opad[6];
453 opad[7] = tmps[gid].opad[7];
455 for (u32 i = 0; i < (truecrypt_mdlen / 8 / 8); i += 8)
459 dgst[0] = tmps[gid].dgst[i + 0];
460 dgst[1] = tmps[gid].dgst[i + 1];
461 dgst[2] = tmps[gid].dgst[i + 2];
462 dgst[3] = tmps[gid].dgst[i + 3];
463 dgst[4] = tmps[gid].dgst[i + 4];
464 dgst[5] = tmps[gid].dgst[i + 5];
465 dgst[6] = tmps[gid].dgst[i + 6];
466 dgst[7] = tmps[gid].dgst[i + 7];
470 out[0] = tmps[gid].out[i + 0];
471 out[1] = tmps[gid].out[i + 1];
472 out[2] = tmps[gid].out[i + 2];
473 out[3] = tmps[gid].out[i + 3];
474 out[4] = tmps[gid].out[i + 4];
475 out[5] = tmps[gid].out[i + 5];
476 out[6] = tmps[gid].out[i + 6];
477 out[7] = tmps[gid].out[i + 7];
479 for (u32 j = 0; j < loop_cnt; j++)
491 w[ 8] = 0x8000000000000000;
498 w[15] = (128 + 64) * 8;
500 hmac_run (w, ipad, opad, dgst);
512 tmps[gid].dgst[i + 0] = dgst[0];
513 tmps[gid].dgst[i + 1] = dgst[1];
514 tmps[gid].dgst[i + 2] = dgst[2];
515 tmps[gid].dgst[i + 3] = dgst[3];
516 tmps[gid].dgst[i + 4] = dgst[4];
517 tmps[gid].dgst[i + 5] = dgst[5];
518 tmps[gid].dgst[i + 6] = dgst[6];
519 tmps[gid].dgst[i + 7] = dgst[7];
521 tmps[gid].out[i + 0] = out[0];
522 tmps[gid].out[i + 1] = out[1];
523 tmps[gid].out[i + 2] = out[2];
524 tmps[gid].out[i + 3] = out[3];
525 tmps[gid].out[i + 4] = out[4];
526 tmps[gid].out[i + 5] = out[5];
527 tmps[gid].out[i + 6] = out[6];
528 tmps[gid].out[i + 7] = out[7];
532 extern "C" __global__ void __launch_bounds__ (256, 1) m06221_comp (const pw_t *pws, const gpu_rule_t *rules_buf, const comb_t *combs_buf, const bf_t *bfs_buf, tc64_tmp_t *tmps, void *hooks, const u32 *bitmaps_buf_s1_a, const u32 *bitmaps_buf_s1_b, const u32 *bitmaps_buf_s1_c, const u32 *bitmaps_buf_s1_d, const u32 *bitmaps_buf_s2_a, const u32 *bitmaps_buf_s2_b, const u32 *bitmaps_buf_s2_c, const u32 *bitmaps_buf_s2_d, plain_t *plains_buf, const digest_t *digests_buf, u32 *hashes_shown, const salt_t *salt_bufs, const tc_t *esalt_bufs, u32 *d_return_buf, 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)
538 const u32 gid = (blockIdx.x * blockDim.x) + threadIdx.x;
539 const u32 lid = threadIdx.x;
541 if (gid >= gid_max) return;
545 ukey1[0] = swap_workaround (h32_from_64 (tmps[gid].out[ 0]));
546 ukey1[1] = swap_workaround (l32_from_64 (tmps[gid].out[ 0]));
547 ukey1[2] = swap_workaround (h32_from_64 (tmps[gid].out[ 1]));
548 ukey1[3] = swap_workaround (l32_from_64 (tmps[gid].out[ 1]));
549 ukey1[4] = swap_workaround (h32_from_64 (tmps[gid].out[ 2]));
550 ukey1[5] = swap_workaround (l32_from_64 (tmps[gid].out[ 2]));
551 ukey1[6] = swap_workaround (h32_from_64 (tmps[gid].out[ 3]));
552 ukey1[7] = swap_workaround (l32_from_64 (tmps[gid].out[ 3]));
556 ukey2[0] = swap_workaround (h32_from_64 (tmps[gid].out[ 4]));
557 ukey2[1] = swap_workaround (l32_from_64 (tmps[gid].out[ 4]));
558 ukey2[2] = swap_workaround (h32_from_64 (tmps[gid].out[ 5]));
559 ukey2[3] = swap_workaround (l32_from_64 (tmps[gid].out[ 5]));
560 ukey2[4] = swap_workaround (h32_from_64 (tmps[gid].out[ 6]));
561 ukey2[5] = swap_workaround (l32_from_64 (tmps[gid].out[ 6]));
562 ukey2[6] = swap_workaround (h32_from_64 (tmps[gid].out[ 7]));
563 ukey2[7] = swap_workaround (l32_from_64 (tmps[gid].out[ 7]));
567 data[0] = esalt_bufs[0].data_buf[0];
568 data[1] = esalt_bufs[0].data_buf[1];
569 data[2] = esalt_bufs[0].data_buf[2];
570 data[3] = esalt_bufs[0].data_buf[3];
580 aes256_decrypt_xts (ukey1, ukey2, tmp, tmp);
582 if (((tmp[0] == 0x45555254) && (tmp[3] == 0)) || ((tmp[0] == 0x45555254) && ((tmp[1] >> 16) <= 5)))
584 mark_hash_s0 (plains_buf, hashes_shown, 0, gid, 0);
586 d_return_buf[lid] = 1;
596 serpent256_decrypt_xts (ukey1, ukey2, tmp, tmp);
598 if (((tmp[0] == 0x45555254) && (tmp[3] == 0)) || ((tmp[0] == 0x45555254) && ((tmp[1] >> 16) <= 5)))
600 mark_hash_s0 (plains_buf, hashes_shown, 0, gid, 0);
602 d_return_buf[lid] = 1;
612 twofish256_decrypt_xts (ukey1, ukey2, tmp, tmp);
614 if (((tmp[0] == 0x45555254) && (tmp[3] == 0)) || ((tmp[0] == 0x45555254) && ((tmp[1] >> 16) <= 5)))
616 mark_hash_s0 (plains_buf, hashes_shown, 0, gid, 0);
618 d_return_buf[lid] = 1;