1 ;******************************************************************************
2 ;* optimized bswap buffer functions
3 ;* Copyright (c) 2008 Loren Merritt
4 ;* Copyright (c) 2003-2013 Michael Niedermayer
5 ;* Copyright (c) 2013 Daniel Kang
7 ;* This file is part of FFmpeg.
9 ;* FFmpeg is free software; you can redistribute it and/or
10 ;* modify it under the terms of the GNU Lesser General Public
11 ;* License as published by the Free Software Foundation; either
12 ;* version 2.1 of the License, or (at your option) any later version.
14 ;* FFmpeg is distributed in the hope that it will be useful,
15 ;* but WITHOUT ANY WARRANTY; without even the implied warranty of
16 ;* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
17 ;* Lesser General Public License for more details.
19 ;* You should have received a copy of the GNU Lesser General Public
20 ;* License along with FFmpeg; if not, write to the Free Software
21 ;* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
22 ;******************************************************************************
24 %include "libavutil/x86/x86util.asm"
27 pb_bswap32: db 3, 2, 1, 0, 7, 6, 5, 4, 11, 10, 9, 8, 15, 14, 13, 12
33 ; %1 = aligned/unaligned
44 mov%1 m1, [r1 + mmsize]
45 %if cpuflag(ssse3)||cpuflag(avx2)
49 mov%1 [r0 + mmsize], m1
51 pshuflw m0, m0, 10110001b
52 pshuflw m1, m1, 10110001b
53 pshufhw m0, m0, 10110001b
54 pshufhw m1, m1, 10110001b
90 pshuflw m0, m0, 10110001b
91 pshufhw m0, m0, 10110001b
102 ; void ff_bswap_buf(uint32_t *dst, const uint32_t *src, int w);
104 %if cpuflag(ssse3)||cpuflag(avx2)
105 cglobal bswap32_buf, 3,4,3
108 vbroadcasti128 m2, [pb_bswap32]
110 mova m2, [pb_bswap32]
113 cglobal bswap32_buf, 3,4,5
160 %if HAVE_AVX2_EXTERNAL