| blob | 7f24a1540821053d66c1460db83453a1febfb80e |
1 /*
2 * Serpent Cipher 8-way parallel algorithm (x86_64/SSE2)
3 *
4 * Copyright (C) 2011 Jussi Kivilinna <jussi.kivilinna@mbnet.fi>
5 *
6 * Based on crypto/serpent.c by
7 * Copyright (C) 2002 Dag Arne Osvik <osvik@ii.uib.no>
8 * 2003 Herbert Valerio Riedel <hvr@gnu.org>
9 *
10 * This program is free software; you can redistribute it and/or modify
11 * it under the terms of the GNU General Public License as published by
12 * the Free Software Foundation; either version 2 of the License, or
13 * (at your option) any later version.
14 *
15 * This program is distributed in the hope that it will be useful,
16 * but WITHOUT ANY WARRANTY; without even the implied warranty of
17 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
18 * GNU General Public License for more details.
19 *
20 * You should have received a copy of the GNU General Public License
21 * along with this program; if not, write to the Free Software
22 * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307
23 * USA
24 *
25 */
27 .file "serpent-sse2-x86_64-asm_64.S"
28 .text
30 #define CTX %rdi
32 /**********************************************************************
33 8-way SSE2 serpent
34 **********************************************************************/
35 #define RA1 %xmm0
36 #define RB1 %xmm1
37 #define RC1 %xmm2
38 #define RD1 %xmm3
39 #define RE1 %xmm4
41 #define RA2 %xmm5
42 #define RB2 %xmm6
43 #define RC2 %xmm7
44 #define RD2 %xmm8
45 #define RE2 %xmm9
47 #define RNOT %xmm10
49 #define RK0 %xmm11
50 #define RK1 %xmm12
51 #define RK2 %xmm13
52 #define RK3 %xmm14
54 #define S0_1(x0, x1, x2, x3, x4) \
55 movdqa x3, x4; \
56 por x0, x3; \
57 pxor x4, x0; \
58 pxor x2, x4; \
59 pxor RNOT, x4; \
60 pxor x1, x3; \
61 pand x0, x1; \
62 pxor x4, x1; \
63 pxor x0, x2;
64 #define S0_2(x0, x1, x2, x3, x4) \
65 pxor x3, x0; \
66 por x0, x4; \
67 pxor x2, x0; \
68 pand x1, x2; \
69 pxor x2, x3; \
70 pxor RNOT, x1; \
71 pxor x4, x2; \
72 pxor x2, x1;
74 #define S1_1(x0, x1, x2, x3, x4) \
75 movdqa x1, x4; \
76 pxor x0, x1; \
77 pxor x3, x0; \
78 pxor RNOT, x3; \
79 pand x1, x4; \
80 por x1, x0; \
81 pxor x2, x3; \
82 pxor x3, x0; \
83 pxor x3, x1;
84 #define S1_2(x0, x1, x2, x3, x4) \
85 pxor x4, x3; \
86 por x4, x1; \
87 pxor x2, x4; \
88 pand x0, x2; \
89 pxor x1, x2; \
90 por x0, x1; \
91 pxor RNOT, x0; \
92 pxor x2, x0; \
93 pxor x1, x4;
95 #define S2_1(x0, x1, x2, x3, x4) \
96 pxor RNOT, x3; \
97 pxor x0, x1; \
98 movdqa x0, x4; \
99 pand x2, x0; \
100 pxor x3, x0; \
101 por x4, x3; \
102 pxor x1, x2; \
103 pxor x1, x3; \
104 pand x0, x1;
105 #define S2_2(x0, x1, x2, x3, x4) \
106 pxor x2, x0; \
107 pand x3, x2; \
108 por x1, x3; \
109 pxor RNOT, x0; \
110 pxor x0, x3; \
111 pxor x0, x4; \
112 pxor x2, x0; \
113 por x2, x1;
115 #define S3_1(x0, x1, x2, x3, x4) \
116 movdqa x1, x4; \
117 pxor x3, x1; \
118 por x0, x3; \
119 pand x0, x4; \
120 pxor x2, x0; \
121 pxor x1, x2; \
122 pand x3, x1; \
123 pxor x3, x2; \
124 por x4, x0; \
125 pxor x3, x4;
126 #define S3_2(x0, x1, x2, x3, x4) \
127 pxor x0, x1; \
128 pand x3, x0; \
129 pand x4, x3; \
130 pxor x2, x3; \
131 por x1, x4; \
132 pand x1, x2; \
133 pxor x3, x4; \
134 pxor x3, x0; \
135 pxor x2, x3;
137 #define S4_1(x0, x1, x2, x3, x4) \
138 movdqa x3, x4; \
139 pand x0, x3; \
140 pxor x4, x0; \
141 pxor x2, x3; \
142 por x4, x2; \
143 pxor x1, x0; \
144 pxor x3, x4; \
145 por x0, x2; \
146 pxor x1, x2;
147 #define S4_2(x0, x1, x2, x3, x4) \
148 pand x0, x1; \
149 pxor x4, x1; \
150 pand x2, x4; \
151 pxor x3, x2; \
152 pxor x0, x4; \
153 por x1, x3; \
154 pxor RNOT, x1; \
155 pxor x0, x3;
157 #define S5_1(x0, x1, x2, x3, x4) \
158 movdqa x1, x4; \
159 por x0, x1; \
160 pxor x1, x2; \
161 pxor RNOT, x3; \
162 pxor x0, x4; \
163 pxor x2, x0; \
164 pand x4, x1; \
165 por x3, x4; \
166 pxor x0, x4;
167 #define S5_2(x0, x1, x2, x3, x4) \
168 pand x3, x0; \
169 pxor x3, x1; \
170 pxor x2, x3; \
171 pxor x1, x0; \
172 pand x4, x2; \
173 pxor x2, x1; \
174 pand x0, x2; \
175 pxor x2, x3;
177 #define S6_1(x0, x1, x2, x3, x4) \
178 movdqa x1, x4; \
179 pxor x0, x3; \
180 pxor x2, x1; \
181 pxor x0, x2; \
182 pand x3, x0; \
183 por x3, x1; \
184 pxor RNOT, x4; \
185 pxor x1, x0; \
186 pxor x2, x1;
187 #define S6_2(x0, x1, x2, x3, x4) \
188 pxor x4, x3; \
189 pxor x0, x4; \
190 pand x0, x2; \
191 pxor x1, x4; \
192 pxor x3, x2; \
193 pand x1, x3; \
194 pxor x0, x3; \
195 pxor x2, x1;
197 #define S7_1(x0, x1, x2, x3, x4) \
198 pxor RNOT, x1; \
199 movdqa x1, x4; \
200 pxor RNOT, x0; \
201 pand x2, x1; \
202 pxor x3, x1; \
203 por x4, x3; \
204 pxor x2, x4; \
205 pxor x3, x2; \
206 pxor x0, x3; \
207 por x1, x0;
208 #define S7_2(x0, x1, x2, x3, x4) \
209 pand x0, x2; \
210 pxor x4, x0; \
211 pxor x3, x4; \
212 pand x0, x3; \
213 pxor x1, x4; \
214 pxor x4, x2; \
215 pxor x1, x3; \
216 por x0, x4; \
217 pxor x1, x4;
219 #define SI0_1(x0, x1, x2, x3, x4) \
220 movdqa x3, x4; \
221 pxor x0, x1; \
222 por x1, x3; \
223 pxor x1, x4; \
224 pxor RNOT, x0; \
225 pxor x3, x2; \
226 pxor x0, x3; \
227 pand x1, x0; \
228 pxor x2, x0;
229 #define SI0_2(x0, x1, x2, x3, x4) \
230 pand x3, x2; \
231 pxor x4, x3; \
232 pxor x3, x2; \
233 pxor x3, x1; \
234 pand x0, x3; \
235 pxor x0, x1; \
236 pxor x2, x0; \
237 pxor x3, x4;
239 #define SI1_1(x0, x1, x2, x3, x4) \
240 pxor x3, x1; \
241 movdqa x0, x4; \
242 pxor x2, x0; \
243 pxor RNOT, x2; \
244 por x1, x4; \
245 pxor x3, x4; \
246 pand x1, x3; \
247 pxor x2, x1; \
248 pand x4, x2;
249 #define SI1_2(x0, x1, x2, x3, x4) \
250 pxor x1, x4; \
251 por x3, x1; \
252 pxor x0, x3; \
253 pxor x0, x2; \
254 por x4, x0; \
255 pxor x4, x2; \
256 pxor x0, x1; \
257 pxor x1, x4;
259 #define SI2_1(x0, x1, x2, x3, x4) \
260 pxor x1, x2; \
261 movdqa x3, x4; \
262 pxor RNOT, x3; \
263 por x2, x3; \
264 pxor x4, x2; \
265 pxor x0, x4; \
266 pxor x1, x3; \
267 por x2, x1; \
268 pxor x0, x2;
269 #define SI2_2(x0, x1, x2, x3, x4) \
270 pxor x4, x1; \
271 por x3, x4; \
272 pxor x3, x2; \
273 pxor x2, x4; \
274 pand x1, x2; \
275 pxor x3, x2; \
276 pxor x4, x3; \
277 pxor x0, x4;
279 #define SI3_1(x0, x1, x2, x3, x4) \
280 pxor x1, x2; \
281 movdqa x1, x4; \
282 pand x2, x1; \
283 pxor x0, x1; \
284 por x4, x0; \
285 pxor x3, x4; \
286 pxor x3, x0; \
287 por x1, x3; \
288 pxor x2, x1;
289 #define SI3_2(x0, x1, x2, x3, x4) \
290 pxor x3, x1; \
291 pxor x2, x0; \
292 pxor x3, x2; \
293 pand x1, x3; \
294 pxor x0, x1; \
295 pand x2, x0; \
296 pxor x3, x4; \
297 pxor x0, x3; \
298 pxor x1, x0;
300 #define SI4_1(x0, x1, x2, x3, x4) \
301 pxor x3, x2; \
302 movdqa x0, x4; \
303 pand x1, x0; \
304 pxor x2, x0; \
305 por x3, x2; \
306 pxor RNOT, x4; \
307 pxor x0, x1; \
308 pxor x2, x0; \
309 pand x4, x2;
310 #define SI4_2(x0, x1, x2, x3, x4) \
311 pxor x0, x2; \
312 por x4, x0; \
313 pxor x3, x0; \
314 pand x2, x3; \
315 pxor x3, x4; \
316 pxor x1, x3; \
317 pand x0, x1; \
318 pxor x1, x4; \
319 pxor x3, x0;
321 #define SI5_1(x0, x1, x2, x3, x4) \
322 movdqa x1, x4; \
323 por x2, x1; \
324 pxor x4, x2; \
325 pxor x3, x1; \
326 pand x4, x3; \
327 pxor x3, x2; \
328 por x0, x3; \
329 pxor RNOT, x0; \
330 pxor x2, x3; \
331 por x0, x2;
332 #define SI5_2(x0, x1, x2, x3, x4) \
333 pxor x1, x4; \
334 pxor x4, x2; \
335 pand x0, x4; \
336 pxor x1, x0; \
337 pxor x3, x1; \
338 pand x2, x0; \
339 pxor x3, x2; \
340 pxor x2, x0; \
341 pxor x4, x2; \
342 pxor x3, x4;
344 #define SI6_1(x0, x1, x2, x3, x4) \
345 pxor x2, x0; \
346 movdqa x0, x4; \
347 pand x3, x0; \
348 pxor x3, x2; \
349 pxor x2, x0; \
350 pxor x1, x3; \
351 por x4, x2; \
352 pxor x3, x2; \
353 pand x0, x3;
354 #define SI6_2(x0, x1, x2, x3, x4) \
355 pxor RNOT, x0; \
356 pxor x1, x3; \
357 pand x2, x1; \
358 pxor x0, x4; \
359 pxor x4, x3; \
360 pxor x2, x4; \
361 pxor x1, x0; \
362 pxor x0, x2;
364 #define SI7_1(x0, x1, x2, x3, x4) \
365 movdqa x3, x4; \
366 pand x0, x3; \
367 pxor x2, x0; \
368 por x4, x2; \
369 pxor x1, x4; \
370 pxor RNOT, x0; \
371 por x3, x1; \
372 pxor x0, x4; \
373 pand x2, x0; \
374 pxor x1, x0;
375 #define SI7_2(x0, x1, x2, x3, x4) \
376 pand x2, x1; \
377 pxor x2, x3; \
378 pxor x3, x4; \
379 pand x3, x2; \
380 por x0, x3; \
381 pxor x4, x1; \
382 pxor x4, x3; \
383 pand x0, x4; \
384 pxor x2, x4;
386 #define get_key(i, j, t) \
387 movd (4*(i)+(j))*4(CTX), t; \
388 pshufd $0, t, t;
390 #define K2(x0, x1, x2, x3, x4, i) \
391 get_key(i, 0, RK0); \
392 get_key(i, 1, RK1); \
393 get_key(i, 2, RK2); \
394 get_key(i, 3, RK3); \
395 pxor RK0, x0 ## 1; \
396 pxor RK1, x1 ## 1; \
397 pxor RK2, x2 ## 1; \
398 pxor RK3, x3 ## 1; \
399 pxor RK0, x0 ## 2; \
400 pxor RK1, x1 ## 2; \
401 pxor RK2, x2 ## 2; \
402 pxor RK3, x3 ## 2;
404 #define LK2(x0, x1, x2, x3, x4, i) \
405 movdqa x0 ## 1, x4 ## 1; \
406 pslld $13, x0 ## 1; \
407 psrld $(32 - 13), x4 ## 1; \
408 por x4 ## 1, x0 ## 1; \
409 pxor x0 ## 1, x1 ## 1; \
410 movdqa x2 ## 1, x4 ## 1; \
411 pslld $3, x2 ## 1; \
412 psrld $(32 - 3), x4 ## 1; \
413 por x4 ## 1, x2 ## 1; \
414 pxor x2 ## 1, x1 ## 1; \
415 movdqa x0 ## 2, x4 ## 2; \
416 pslld $13, x0 ## 2; \
417 psrld $(32 - 13), x4 ## 2; \
418 por x4 ## 2, x0 ## 2; \
419 pxor x0 ## 2, x1 ## 2; \
420 movdqa x2 ## 2, x4 ## 2; \
421 pslld $3, x2 ## 2; \
422 psrld $(32 - 3), x4 ## 2; \
423 por x4 ## 2, x2 ## 2; \
424 pxor x2 ## 2, x1 ## 2; \
425 movdqa x1 ## 1, x4 ## 1; \
426 pslld $1, x1 ## 1; \
427 psrld $(32 - 1), x4 ## 1; \
428 por x4 ## 1, x1 ## 1; \
429 movdqa x0 ## 1, x4 ## 1; \
430 pslld $3, x4 ## 1; \
431 pxor x2 ## 1, x3 ## 1; \
432 pxor x4 ## 1, x3 ## 1; \
433 movdqa x3 ## 1, x4 ## 1; \
434 get_key(i, 1, RK1); \
435 movdqa x1 ## 2, x4 ## 2; \
436 pslld $1, x1 ## 2; \
437 psrld $(32 - 1), x4 ## 2; \
438 por x4 ## 2, x1 ## 2; \
439 movdqa x0 ## 2, x4 ## 2; \
440 pslld $3, x4 ## 2; \
441 pxor x2 ## 2, x3 ## 2; \
442 pxor x4 ## 2, x3 ## 2; \
443 movdqa x3 ## 2, x4 ## 2; \
444 get_key(i, 3, RK3); \
445 pslld $7, x3 ## 1; \
446 psrld $(32 - 7), x4 ## 1; \
447 por x4 ## 1, x3 ## 1; \
448 movdqa x1 ## 1, x4 ## 1; \
449 pslld $7, x4 ## 1; \
450 pxor x1 ## 1, x0 ## 1; \
451 pxor x3 ## 1, x0 ## 1; \
452 pxor x3 ## 1, x2 ## 1; \
453 pxor x4 ## 1, x2 ## 1; \
454 get_key(i, 0, RK0); \
455 pslld $7, x3 ## 2; \
456 psrld $(32 - 7), x4 ## 2; \
457 por x4 ## 2, x3 ## 2; \
458 movdqa x1 ## 2, x4 ## 2; \
459 pslld $7, x4 ## 2; \
460 pxor x1 ## 2, x0 ## 2; \
461 pxor x3 ## 2, x0 ## 2; \
462 pxor x3 ## 2, x2 ## 2; \
463 pxor x4 ## 2, x2 ## 2; \
464 get_key(i, 2, RK2); \
465 pxor RK1, x1 ## 1; \
466 pxor RK3, x3 ## 1; \
467 movdqa x0 ## 1, x4 ## 1; \
468 pslld $5, x0 ## 1; \
469 psrld $(32 - 5), x4 ## 1; \
470 por x4 ## 1, x0 ## 1; \
471 movdqa x2 ## 1, x4 ## 1; \
472 pslld $22, x2 ## 1; \
473 psrld $(32 - 22), x4 ## 1; \
474 por x4 ## 1, x2 ## 1; \
475 pxor RK0, x0 ## 1; \
476 pxor RK2, x2 ## 1; \
477 pxor RK1, x1 ## 2; \
478 pxor RK3, x3 ## 2; \
479 movdqa x0 ## 2, x4 ## 2; \
480 pslld $5, x0 ## 2; \
481 psrld $(32 - 5), x4 ## 2; \
482 por x4 ## 2, x0 ## 2; \
483 movdqa x2 ## 2, x4 ## 2; \
484 pslld $22, x2 ## 2; \
485 psrld $(32 - 22), x4 ## 2; \
486 por x4 ## 2, x2 ## 2; \
487 pxor RK0, x0 ## 2; \
488 pxor RK2, x2 ## 2;
490 #define KL2(x0, x1, x2, x3, x4, i) \
491 pxor RK0, x0 ## 1; \
492 pxor RK2, x2 ## 1; \
493 movdqa x0 ## 1, x4 ## 1; \
494 psrld $5, x0 ## 1; \
495 pslld $(32 - 5), x4 ## 1; \
496 por x4 ## 1, x0 ## 1; \
497 pxor RK3, x3 ## 1; \
498 pxor RK1, x1 ## 1; \
499 movdqa x2 ## 1, x4 ## 1; \
500 psrld $22, x2 ## 1; \
501 pslld $(32 - 22), x4 ## 1; \
502 por x4 ## 1, x2 ## 1; \
503 pxor x3 ## 1, x2 ## 1; \
504 pxor RK0, x0 ## 2; \
505 pxor RK2, x2 ## 2; \
506 movdqa x0 ## 2, x4 ## 2; \
507 psrld $5, x0 ## 2; \
508 pslld $(32 - 5), x4 ## 2; \
509 por x4 ## 2, x0 ## 2; \
510 pxor RK3, x3 ## 2; \
511 pxor RK1, x1 ## 2; \
512 movdqa x2 ## 2, x4 ## 2; \
513 psrld $22, x2 ## 2; \
514 pslld $(32 - 22), x4 ## 2; \
515 por x4 ## 2, x2 ## 2; \
516 pxor x3 ## 2, x2 ## 2; \
517 pxor x3 ## 1, x0 ## 1; \
518 movdqa x1 ## 1, x4 ## 1; \
519 pslld $7, x4 ## 1; \
520 pxor x1 ## 1, x0 ## 1; \
521 pxor x4 ## 1, x2 ## 1; \
522 movdqa x1 ## 1, x4 ## 1; \
523 psrld $1, x1 ## 1; \
524 pslld $(32 - 1), x4 ## 1; \
525 por x4 ## 1, x1 ## 1; \
526 pxor x3 ## 2, x0 ## 2; \
527 movdqa x1 ## 2, x4 ## 2; \
528 pslld $7, x4 ## 2; \
529 pxor x1 ## 2, x0 ## 2; \
530 pxor x4 ## 2, x2 ## 2; \
531 movdqa x1 ## 2, x4 ## 2; \
532 psrld $1, x1 ## 2; \
533 pslld $(32 - 1), x4 ## 2; \
534 por x4 ## 2, x1 ## 2; \
535 movdqa x3 ## 1, x4 ## 1; \
536 psrld $7, x3 ## 1; \
537 pslld $(32 - 7), x4 ## 1; \
538 por x4 ## 1, x3 ## 1; \
539 pxor x0 ## 1, x1 ## 1; \
540 movdqa x0 ## 1, x4 ## 1; \
541 pslld $3, x4 ## 1; \
542 pxor x4 ## 1, x3 ## 1; \
543 movdqa x0 ## 1, x4 ## 1; \
544 movdqa x3 ## 2, x4 ## 2; \
545 psrld $7, x3 ## 2; \
546 pslld $(32 - 7), x4 ## 2; \
547 por x4 ## 2, x3 ## 2; \
548 pxor x0 ## 2, x1 ## 2; \
549 movdqa x0 ## 2, x4 ## 2; \
550 pslld $3, x4 ## 2; \
551 pxor x4 ## 2, x3 ## 2; \
552 movdqa x0 ## 2, x4 ## 2; \
553 psrld $13, x0 ## 1; \
554 pslld $(32 - 13), x4 ## 1; \
555 por x4 ## 1, x0 ## 1; \
556 pxor x2 ## 1, x1 ## 1; \
557 pxor x2 ## 1, x3 ## 1; \
558 movdqa x2 ## 1, x4 ## 1; \
559 psrld $3, x2 ## 1; \
560 pslld $(32 - 3), x4 ## 1; \
561 por x4 ## 1, x2 ## 1; \
562 psrld $13, x0 ## 2; \
563 pslld $(32 - 13), x4 ## 2; \
564 por x4 ## 2, x0 ## 2; \
565 pxor x2 ## 2, x1 ## 2; \
566 pxor x2 ## 2, x3 ## 2; \
567 movdqa x2 ## 2, x4 ## 2; \
568 psrld $3, x2 ## 2; \
569 pslld $(32 - 3), x4 ## 2; \
570 por x4 ## 2, x2 ## 2;
572 #define S(SBOX, x0, x1, x2, x3, x4) \
573 SBOX ## _1(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
574 SBOX ## _2(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
575 SBOX ## _1(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \
576 SBOX ## _2(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2);
578 #define SP(SBOX, x0, x1, x2, x3, x4, i) \
579 get_key(i, 0, RK0); \
580 SBOX ## _1(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
581 get_key(i, 2, RK2); \
582 SBOX ## _1(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \
583 get_key(i, 3, RK3); \
584 SBOX ## _2(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
585 get_key(i, 1, RK1); \
586 SBOX ## _2(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \
588 #define transpose_4x4(x0, x1, x2, x3, t1, t2, t3) \
589 movdqa x2, t3; \
590 movdqa x0, t1; \
591 unpcklps x3, t3; \
592 movdqa x0, t2; \
593 unpcklps x1, t1; \
594 unpckhps x1, t2; \
595 movdqa t3, x1; \
596 unpckhps x3, x2; \
597 movdqa t1, x0; \
598 movhlps t1, x1; \
599 movdqa t2, t1; \
600 movlhps t3, x0; \
601 movlhps x2, t1; \
602 movhlps t2, x2; \
603 movdqa x2, x3; \
604 movdqa t1, x2;
606 #define read_blocks(in, x0, x1, x2, x3, t0, t1, t2) \
607 movdqu (0*4*4)(in), x0; \
608 movdqu (1*4*4)(in), x1; \
609 movdqu (2*4*4)(in), x2; \
610 movdqu (3*4*4)(in), x3; \
611 \
612 transpose_4x4(x0, x1, x2, x3, t0, t1, t2)
614 #define write_blocks(out, x0, x1, x2, x3, t0, t1, t2) \
615 transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \
616 \
617 movdqu x0, (0*4*4)(out); \
618 movdqu x1, (1*4*4)(out); \
619 movdqu x2, (2*4*4)(out); \
620 movdqu x3, (3*4*4)(out);
622 #define xor_blocks(out, x0, x1, x2, x3, t0, t1, t2) \
623 transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \
624 \
625 movdqu (0*4*4)(out), t0; \
626 pxor t0, x0; \
627 movdqu x0, (0*4*4)(out); \
628 movdqu (1*4*4)(out), t0; \
629 pxor t0, x1; \
630 movdqu x1, (1*4*4)(out); \
631 movdqu (2*4*4)(out), t0; \
632 pxor t0, x2; \
633 movdqu x2, (2*4*4)(out); \
634 movdqu (3*4*4)(out), t0; \
635 pxor t0, x3; \
636 movdqu x3, (3*4*4)(out);
638 .align 8
639 .global __serpent_enc_blk_8way
640 .type __serpent_enc_blk_8way,@function;
642 __serpent_enc_blk_8way:
643 /* input:
644 * %rdi: ctx, CTX
645 * %rsi: dst
646 * %rdx: src
647 * %rcx: bool, if true: xor output
648 */
650 pcmpeqd RNOT, RNOT;
652 leaq (4*4*4)(%rdx), %rax;
653 read_blocks(%rdx, RA1, RB1, RC1, RD1, RK0, RK1, RK2);
654 read_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
656 K2(RA, RB, RC, RD, RE, 0);
657 S(S0, RA, RB, RC, RD, RE); LK2(RC, RB, RD, RA, RE, 1);
658 S(S1, RC, RB, RD, RA, RE); LK2(RE, RD, RA, RC, RB, 2);
659 S(S2, RE, RD, RA, RC, RB); LK2(RB, RD, RE, RC, RA, 3);
660 S(S3, RB, RD, RE, RC, RA); LK2(RC, RA, RD, RB, RE, 4);
661 S(S4, RC, RA, RD, RB, RE); LK2(RA, RD, RB, RE, RC, 5);
662 S(S5, RA, RD, RB, RE, RC); LK2(RC, RA, RD, RE, RB, 6);
663 S(S6, RC, RA, RD, RE, RB); LK2(RD, RB, RA, RE, RC, 7);
664 S(S7, RD, RB, RA, RE, RC); LK2(RC, RA, RE, RD, RB, 8);
665 S(S0, RC, RA, RE, RD, RB); LK2(RE, RA, RD, RC, RB, 9);
666 S(S1, RE, RA, RD, RC, RB); LK2(RB, RD, RC, RE, RA, 10);
667 S(S2, RB, RD, RC, RE, RA); LK2(RA, RD, RB, RE, RC, 11);
668 S(S3, RA, RD, RB, RE, RC); LK2(RE, RC, RD, RA, RB, 12);
669 S(S4, RE, RC, RD, RA, RB); LK2(RC, RD, RA, RB, RE, 13);
670 S(S5, RC, RD, RA, RB, RE); LK2(RE, RC, RD, RB, RA, 14);
671 S(S6, RE, RC, RD, RB, RA); LK2(RD, RA, RC, RB, RE, 15);
672 S(S7, RD, RA, RC, RB, RE); LK2(RE, RC, RB, RD, RA, 16);
673 S(S0, RE, RC, RB, RD, RA); LK2(RB, RC, RD, RE, RA, 17);
674 S(S1, RB, RC, RD, RE, RA); LK2(RA, RD, RE, RB, RC, 18);
675 S(S2, RA, RD, RE, RB, RC); LK2(RC, RD, RA, RB, RE, 19);
676 S(S3, RC, RD, RA, RB, RE); LK2(RB, RE, RD, RC, RA, 20);
677 S(S4, RB, RE, RD, RC, RA); LK2(RE, RD, RC, RA, RB, 21);
678 S(S5, RE, RD, RC, RA, RB); LK2(RB, RE, RD, RA, RC, 22);
679 S(S6, RB, RE, RD, RA, RC); LK2(RD, RC, RE, RA, RB, 23);
680 S(S7, RD, RC, RE, RA, RB); LK2(RB, RE, RA, RD, RC, 24);
681 S(S0, RB, RE, RA, RD, RC); LK2(RA, RE, RD, RB, RC, 25);
682 S(S1, RA, RE, RD, RB, RC); LK2(RC, RD, RB, RA, RE, 26);
683 S(S2, RC, RD, RB, RA, RE); LK2(RE, RD, RC, RA, RB, 27);
684 S(S3, RE, RD, RC, RA, RB); LK2(RA, RB, RD, RE, RC, 28);
685 S(S4, RA, RB, RD, RE, RC); LK2(RB, RD, RE, RC, RA, 29);
686 S(S5, RB, RD, RE, RC, RA); LK2(RA, RB, RD, RC, RE, 30);
687 S(S6, RA, RB, RD, RC, RE); LK2(RD, RE, RB, RC, RA, 31);
688 S(S7, RD, RE, RB, RC, RA); K2(RA, RB, RC, RD, RE, 32);
690 leaq (4*4*4)(%rsi), %rax;
692 testb %cl, %cl;
693 jnz __enc_xor8;
695 write_blocks(%rsi, RA1, RB1, RC1, RD1, RK0, RK1, RK2);
696 write_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
698 ret;
700 __enc_xor8:
701 xor_blocks(%rsi, RA1, RB1, RC1, RD1, RK0, RK1, RK2);
702 xor_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
704 ret;
706 .align 8
707 .global serpent_dec_blk_8way
708 .type serpent_dec_blk_8way,@function;
710 serpent_dec_blk_8way:
711 /* input:
712 * %rdi: ctx, CTX
713 * %rsi: dst
714 * %rdx: src
715 */
717 pcmpeqd RNOT, RNOT;
719 leaq (4*4*4)(%rdx), %rax;
720 read_blocks(%rdx, RA1, RB1, RC1, RD1, RK0, RK1, RK2);
721 read_blocks(%rax, RA2, RB2, RC2, RD2, RK0, RK1, RK2);
723 K2(RA, RB, RC, RD, RE, 32);
724 SP(SI7, RA, RB, RC, RD, RE, 31); KL2(RB, RD, RA, RE, RC, 31);
725 SP(SI6, RB, RD, RA, RE, RC, 30); KL2(RA, RC, RE, RB, RD, 30);
726 SP(SI5, RA, RC, RE, RB, RD, 29); KL2(RC, RD, RA, RE, RB, 29);
727 SP(SI4, RC, RD, RA, RE, RB, 28); KL2(RC, RA, RB, RE, RD, 28);
728 SP(SI3, RC, RA, RB, RE, RD, 27); KL2(RB, RC, RD, RE, RA, 27);
729 SP(SI2, RB, RC, RD, RE, RA, 26); KL2(RC, RA, RE, RD, RB, 26);
730 SP(SI1, RC, RA, RE, RD, RB, 25); KL2(RB, RA, RE, RD, RC, 25);
731 SP(SI0, RB, RA, RE, RD, RC, 24); KL2(RE, RC, RA, RB, RD, 24);
732 SP(SI7, RE, RC, RA, RB, RD, 23); KL2(RC, RB, RE, RD, RA, 23);
733 SP(SI6, RC, RB, RE, RD, RA, 22); KL2(RE, RA, RD, RC, RB, 22);
734 SP(SI5, RE, RA, RD, RC, RB, 21); KL2(RA, RB, RE, RD, RC, 21);
735 SP(SI4, RA, RB, RE, RD, RC, 20); KL2(RA, RE, RC, RD, RB, 20);
736 SP(SI3, RA, RE, RC, RD, RB, 19); KL2(RC, RA, RB, RD, RE, 19);
737 SP(SI2, RC, RA, RB, RD, RE, 18); KL2(RA, RE, RD, RB, RC, 18);
738 SP(SI1, RA, RE, RD, RB, RC, 17); KL2(RC, RE, RD, RB, RA, 17);
739 SP(SI0, RC, RE, RD, RB, RA, 16); KL2(RD, RA, RE, RC, RB, 16);
740 SP(SI7, RD, RA, RE, RC, RB, 15); KL2(RA, RC, RD, RB, RE, 15);
741 SP(SI6, RA, RC, RD, RB, RE, 14); KL2(RD, RE, RB, RA, RC, 14);
742 SP(SI5, RD, RE, RB, RA, RC, 13); KL2(RE, RC, RD, RB, RA, 13);
743 SP(SI4, RE, RC, RD, RB, RA, 12); KL2(RE, RD, RA, RB, RC, 12);
744 SP(SI3, RE, RD, RA, RB, RC, 11); KL2(RA, RE, RC, RB, RD, 11);
745 SP(SI2, RA, RE, RC, RB, RD, 10); KL2(RE, RD, RB, RC, RA, 10);
746 SP(SI1, RE, RD, RB, RC, RA, 9); KL2(RA, RD, RB, RC, RE, 9);
747 SP(SI0, RA, RD, RB, RC, RE, 8); KL2(RB, RE, RD, RA, RC, 8);
748 SP(SI7, RB, RE, RD, RA, RC, 7); KL2(RE, RA, RB, RC, RD, 7);
749 SP(SI6, RE, RA, RB, RC, RD, 6); KL2(RB, RD, RC, RE, RA, 6);
750 SP(SI5, RB, RD, RC, RE, RA, 5); KL2(RD, RA, RB, RC, RE, 5);
751 SP(SI4, RD, RA, RB, RC, RE, 4); KL2(RD, RB, RE, RC, RA, 4);
752 SP(SI3, RD, RB, RE, RC, RA, 3); KL2(RE, RD, RA, RC, RB, 3);
753 SP(SI2, RE, RD, RA, RC, RB, 2); KL2(RD, RB, RC, RA, RE, 2);
754 SP(SI1, RD, RB, RC, RA, RE, 1); KL2(RE, RB, RC, RA, RD, 1);
755 S(SI0, RE, RB, RC, RA, RD); K2(RC, RD, RB, RE, RA, 0);
757 leaq (4*4*4)(%rsi), %rax;
758 write_blocks(%rsi, RC1, RD1, RB1, RE1, RK0, RK1, RK2);
759 write_blocks(%rax, RC2, RD2, RB2, RE2, RK0, RK1, RK2);
761 ret;