1*2874c5fdSThomas Gleixner/* SPDX-License-Identifier: GPL-2.0-or-later */ 2cf1521a1SJussi Kivilinna/* 3cf1521a1SJussi Kivilinna * Shared glue code for 128bit block ciphers, AVX2 assembler macros 4cf1521a1SJussi Kivilinna * 5cf1521a1SJussi Kivilinna * Copyright © 2012-2013 Jussi Kivilinna <jussi.kivilinna@mbnet.fi> 6cf1521a1SJussi Kivilinna */ 7cf1521a1SJussi Kivilinna 8cf1521a1SJussi Kivilinna#define load_16way(src, x0, x1, x2, x3, x4, x5, x6, x7) \ 9cf1521a1SJussi Kivilinna vmovdqu (0*32)(src), x0; \ 10cf1521a1SJussi Kivilinna vmovdqu (1*32)(src), x1; \ 11cf1521a1SJussi Kivilinna vmovdqu (2*32)(src), x2; \ 12cf1521a1SJussi Kivilinna vmovdqu (3*32)(src), x3; \ 13cf1521a1SJussi Kivilinna vmovdqu (4*32)(src), x4; \ 14cf1521a1SJussi Kivilinna vmovdqu (5*32)(src), x5; \ 15cf1521a1SJussi Kivilinna vmovdqu (6*32)(src), x6; \ 16cf1521a1SJussi Kivilinna vmovdqu (7*32)(src), x7; 17cf1521a1SJussi Kivilinna 18cf1521a1SJussi Kivilinna#define store_16way(dst, x0, x1, x2, x3, x4, x5, x6, x7) \ 19cf1521a1SJussi Kivilinna vmovdqu x0, (0*32)(dst); \ 20cf1521a1SJussi Kivilinna vmovdqu x1, (1*32)(dst); \ 21cf1521a1SJussi Kivilinna vmovdqu x2, (2*32)(dst); \ 22cf1521a1SJussi Kivilinna vmovdqu x3, (3*32)(dst); \ 23cf1521a1SJussi Kivilinna vmovdqu x4, (4*32)(dst); \ 24cf1521a1SJussi Kivilinna vmovdqu x5, (5*32)(dst); \ 25cf1521a1SJussi Kivilinna vmovdqu x6, (6*32)(dst); \ 26cf1521a1SJussi Kivilinna vmovdqu x7, (7*32)(dst); 27cf1521a1SJussi Kivilinna 28cf1521a1SJussi Kivilinna#define store_cbc_16way(src, dst, x0, x1, x2, x3, x4, x5, x6, x7, t0) \ 29cf1521a1SJussi Kivilinna vpxor t0, t0, t0; \ 30cf1521a1SJussi Kivilinna vinserti128 $1, (src), t0, t0; \ 31cf1521a1SJussi Kivilinna vpxor t0, x0, x0; \ 32cf1521a1SJussi Kivilinna vpxor (0*32+16)(src), x1, x1; \ 33cf1521a1SJussi Kivilinna vpxor (1*32+16)(src), x2, x2; \ 34cf1521a1SJussi Kivilinna vpxor (2*32+16)(src), x3, x3; \ 35cf1521a1SJussi Kivilinna vpxor (3*32+16)(src), x4, x4; \ 36cf1521a1SJussi Kivilinna vpxor (4*32+16)(src), x5, x5; \ 37cf1521a1SJussi Kivilinna vpxor (5*32+16)(src), x6, x6; \ 38cf1521a1SJussi Kivilinna vpxor (6*32+16)(src), x7, x7; \ 39cf1521a1SJussi Kivilinna store_16way(dst, x0, x1, x2, x3, x4, x5, x6, x7); 40