173 		     y0, y1, y2, y3,			\  argument
187 	vmovdqu64 (11 * 64)(rio), y3;			\
196 		      y0, y1, y2, y3,			\  argument
201 			 y0, y1, y2, y3,		\
216 	vmovdqu64 y3, 3 * 64(mem_cd);			\
224 		     y0, y1, y2, y3,			\  argument
238 	vmovdqu64 y3, 11 * 64(mem);			\
270 		       y0, y1, y2, y3,			\  argument
297 	vpxorq t0, y3, y3;				\
331 			     y0, y1, y2, y3,		\  argument
360 	vgf2p8affineqb $(tf_x2_const), t4, y3, y3;	\
362 	vgf2p8affineinvqb $0, t2, y3, y3;		\
383 		       y0, y1, y2, y3,			\  argument
389 	vpxorq y3, x7, x7;				\
395 	vpxorq y7, y3, y3;				\
413 	vpxorq x3, y3, y3;				\
419 	vpxorq y3, x7, x7;
423 		     y0, y1, y2, y3,			\  argument
429 		       y0, y1, y2, y3, y4, y5, y6, y7,	\
434 			     y2, y3, y0, y1,		\
441 	aria_diff_m(y0, y1, y2, y3, z0, z1, z2, z3);	\
445 		       y0, y1, y2, y3,			\
457 		       y0, y1, y2, y3,			\
463 		     y0, y1, y2, y3,			\  argument
469 		       y0, y1, y2, y3, y4, y5, y6, y7,	\
474 			     y0, y1, y2, y3,		\
481 	aria_diff_m(y0, y1, y2, y3, z0, z1, z2, z3);	\
485 		       y0, y1, y2, y3,			\
491 	 * T2 = y0, y1, y2, y3, -> y2, y3, y0, y1	\
497 		       y2, y3, y0, y1,			\
502 		     y0, y1, y2, y3,			\  argument
509 		       y0, y1, y2, y3,			\
514 			     y2, y3, y0, y1,		\
520 		       y0, y1, y2, y3,			\