x86/crypto/serpent-sse2-x86_64-asm_64.S

41 #define S0_1(x0, x1, x2, x3, x4) \
43 	por x0,			x3; \
44 	pxor x4,		x0; \
48 	pand x0,		x1; \
50 	pxor x0,		x2;
51 #define S0_2(x0, x1, x2, x3, x4) \
52 	pxor x3,		x0; \
53 	por x0,			x4; \
54 	pxor x2,		x0; \
61 #define S1_1(x0, x1, x2, x3, x4) \
63 	pxor x0,		x1; \
64 	pxor x3,		x0; \
67 	por x1,			x0; \
69 	pxor x3,		x0; \
71 #define S1_2(x0, x1, x2, x3, x4) \
75 	pand x0,		x2; \
77 	por x0,			x1; \
78 	pxor RNOT,		x0; \
79 	pxor x2,		x0; \
82 #define S2_1(x0, x1, x2, x3, x4) \
84 	pxor x0,		x1; \
85 	movdqa x0,		x4; \
86 	pand x2,		x0; \
87 	pxor x3,		x0; \
91 	pand x0,		x1;
92 #define S2_2(x0, x1, x2, x3, x4) \
93 	pxor x2,		x0; \
96 	pxor RNOT,		x0; \
97 	pxor x0,		x3; \
98 	pxor x0,		x4; \
99 	pxor x2,		x0; \
102 #define S3_1(x0, x1, x2, x3, x4) \
105 	por x0,			x3; \
106 	pand x0,		x4; \
107 	pxor x2,		x0; \
111 	por x4,			x0; \
113 #define S3_2(x0, x1, x2, x3, x4) \
114 	pxor x0,		x1; \
115 	pand x3,		x0; \
121 	pxor x3,		x0; \
124 #define S4_1(x0, x1, x2, x3, x4) \
126 	pand x0,		x3; \
127 	pxor x4,		x0; \
130 	pxor x1,		x0; \
132 	por x0,			x2; \
134 #define S4_2(x0, x1, x2, x3, x4) \
135 	pand x0,		x1; \
139 	pxor x0,		x4; \
142 	pxor x0,		x3;
144 #define S5_1(x0, x1, x2, x3, x4) \
146 	por x0,			x1; \
149 	pxor x0,		x4; \
150 	pxor x2,		x0; \
153 	pxor x0,		x4;
154 #define S5_2(x0, x1, x2, x3, x4) \
155 	pand x3,		x0; \
158 	pxor x1,		x0; \
161 	pand x0,		x2; \
164 #define S6_1(x0, x1, x2, x3, x4) \
166 	pxor x0,		x3; \
168 	pxor x0,		x2; \
169 	pand x3,		x0; \
172 	pxor x1,		x0; \
174 #define S6_2(x0, x1, x2, x3, x4) \
176 	pxor x0,		x4; \
177 	pand x0,		x2; \
181 	pxor x0,		x3; \
184 #define S7_1(x0, x1, x2, x3, x4) \
187 	pxor RNOT,		x0; \
193 	pxor x0,		x3; \
194 	por x1,			x0;
195 #define S7_2(x0, x1, x2, x3, x4) \
196 	pand x0,		x2; \
197 	pxor x4,		x0; \
199 	pand x0,		x3; \
203 	por x0,			x4; \
206 #define SI0_1(x0, x1, x2, x3, x4) \
208 	pxor x0,		x1; \
211 	pxor RNOT,		x0; \
213 	pxor x0,		x3; \
214 	pand x1,		x0; \
215 	pxor x2,		x0;
216 #define SI0_2(x0, x1, x2, x3, x4) \
221 	pand x0,		x3; \
222 	pxor x0,		x1; \
223 	pxor x2,		x0; \
226 #define SI1_1(x0, x1, x2, x3, x4) \
228 	movdqa x0,		x4; \
229 	pxor x2,		x0; \
236 #define SI1_2(x0, x1, x2, x3, x4) \
239 	pxor x0,		x3; \
240 	pxor x0,		x2; \
241 	por x4,			x0; \
243 	pxor x0,		x1; \
246 #define SI2_1(x0, x1, x2, x3, x4) \
252 	pxor x0,		x4; \
255 	pxor x0,		x2;
256 #define SI2_2(x0, x1, x2, x3, x4) \
264 	pxor x0,		x4;
266 #define SI3_1(x0, x1, x2, x3, x4) \
270 	pxor x0,		x1; \
271 	por x4,			x0; \
273 	pxor x3,		x0; \
276 #define SI3_2(x0, x1, x2, x3, x4) \
278 	pxor x2,		x0; \
281 	pxor x0,		x1; \
282 	pand x2,		x0; \
284 	pxor x0,		x3; \
285 	pxor x1,		x0;
287 #define SI4_1(x0, x1, x2, x3, x4) \
289 	movdqa x0,		x4; \
290 	pand x1,		x0; \
291 	pxor x2,		x0; \
294 	pxor x0,		x1; \
295 	pxor x2,		x0; \
297 #define SI4_2(x0, x1, x2, x3, x4) \
298 	pxor x0,		x2; \
299 	por x4,			x0; \
300 	pxor x3,		x0; \
304 	pand x0,		x1; \
306 	pxor x3,		x0;
308 #define SI5_1(x0, x1, x2, x3, x4) \
315 	por x0,			x3; \
316 	pxor RNOT,		x0; \
318 	por x0,			x2;
319 #define SI5_2(x0, x1, x2, x3, x4) \
322 	pand x0,		x4; \
323 	pxor x1,		x0; \
325 	pand x2,		x0; \
327 	pxor x2,		x0; \
331 #define SI6_1(x0, x1, x2, x3, x4) \
332 	pxor x2,		x0; \
333 	movdqa x0,		x4; \
334 	pand x3,		x0; \
336 	pxor x2,		x0; \
340 	pand x0,		x3;
341 #define SI6_2(x0, x1, x2, x3, x4) \
342 	pxor RNOT,		x0; \
345 	pxor x0,		x4; \
348 	pxor x1,		x0; \
349 	pxor x0,		x2;
351 #define SI7_1(x0, x1, x2, x3, x4) \
353 	pand x0,		x3; \
354 	pxor x2,		x0; \
357 	pxor RNOT,		x0; \
359 	pxor x0,		x4; \
360 	pand x2,		x0; \
361 	pxor x1,		x0;
362 #define SI7_2(x0, x1, x2, x3, x4) \
367 	por x0,			x3; \
370 	pand x0,		x4; \
377 #define K2(x0, x1, x2, x3, x4, i) \
382 	pxor RK0,		x0 ## 1; \
386 		pxor RK0,		x0 ## 2; \
391 #define LK2(x0, x1, x2, x3, x4, i) \
392 	movdqa x0 ## 1,		x4 ## 1; \
393 	pslld $13,		x0 ## 1; \
395 	por x4 ## 1,		x0 ## 1; \
396 	pxor x0 ## 1,		x1 ## 1; \
402 		movdqa x0 ## 2,		x4 ## 2; \
403 		pslld $13,		x0 ## 2; \
405 		por x4 ## 2,		x0 ## 2; \
406 		pxor x0 ## 2,		x1 ## 2; \
416 	movdqa x0 ## 1,		x4 ## 1; \
426 		movdqa x0 ## 2,		x4 ## 2; \
437 	pxor x1 ## 1,		x0 ## 1; \
438 	pxor x3 ## 1,		x0 ## 1; \
447 		pxor x1 ## 2,		x0 ## 2; \
448 		pxor x3 ## 2,		x0 ## 2; \
454 	movdqa x0 ## 1,		x4 ## 1; \
455 	pslld $5,		x0 ## 1; \
457 	por x4 ## 1,		x0 ## 1; \
462 	pxor RK0,		x0 ## 1; \
466 		movdqa x0 ## 2,		x4 ## 2; \
467 		pslld $5,		x0 ## 2; \
469 		por x4 ## 2,		x0 ## 2; \
474 		pxor RK0,		x0 ## 2; \
477 #define KL2(x0, x1, x2, x3, x4, i) \
478 	pxor RK0,		x0 ## 1; \
480 	movdqa x0 ## 1,		x4 ## 1; \
481 	psrld $5,		x0 ## 1; \
483 	por x4 ## 1,		x0 ## 1; \
491 		pxor RK0,		x0 ## 2; \
493 		movdqa x0 ## 2,		x4 ## 2; \
494 		psrld $5,		x0 ## 2; \
496 		por x4 ## 2,		x0 ## 2; \
504 	pxor x3 ## 1,		x0 ## 1; \
507 	pxor x1 ## 1,		x0 ## 1; \
513 		pxor x3 ## 2,		x0 ## 2; \
516 		pxor x1 ## 2,		x0 ## 2; \
526 	pxor x0 ## 1,		x1 ## 1; \
527 	movdqa x0 ## 1,		x4 ## 1; \
530 	movdqa x0 ## 1,		x4 ## 1; \
535 		pxor x0 ## 2,		x1 ## 2; \
536 		movdqa x0 ## 2,		x4 ## 2; \
539 		movdqa x0 ## 2,		x4 ## 2; \
540 	psrld $13,		x0 ## 1; \
542 	por x4 ## 1,		x0 ## 1; \
549 		psrld $13,		x0 ## 2; \
551 		por x4 ## 2,		x0 ## 2; \
559 #define S(SBOX, x0, x1, x2, x3, x4) \
560 	SBOX ## _1(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
561 	SBOX ## _2(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
562 	SBOX ## _1(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \
563 	SBOX ## _2(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2);
565 #define SP(SBOX, x0, x1, x2, x3, x4, i) \
567 	SBOX ## _1(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
569 	SBOX ## _1(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \
571 	SBOX ## _2(x0 ## 1, x1 ## 1, x2 ## 1, x3 ## 1, x4 ## 1); \
573 	SBOX ## _2(x0 ## 2, x1 ## 2, x2 ## 2, x3 ## 2, x4 ## 2); \
575 #define transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \
576 	movdqa x0,		t2; \
577 	punpckldq x1,		x0; \
582 	movdqa x0,		x1; \
583 	punpcklqdq t1,		x0; \
590 #define read_blocks(in, x0, x1, x2, x3, t0, t1, t2) \
591 	movdqu (0*4*4)(in),	x0; \
596 	transpose_4x4(x0, x1, x2, x3, t0, t1, t2)
598 #define write_blocks(out, x0, x1, x2, x3, t0, t1, t2) \
599 	transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \
601 	movdqu x0,		(0*4*4)(out); \
606 #define xor_blocks(out, x0, x1, x2, x3, t0, t1, t2) \
607 	transpose_4x4(x0, x1, x2, x3, t0, t1, t2) \
610 	pxor t0,		x0; \
611 	movdqu x0,		(0*4*4)(out); \