1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-23 12:43:46 +02:00
FFmpeg/libswresample/x86/rematrix_init.c
James Almer 9937362c54 x86/swr: use lavu helper macros to check CPU extensions
Signed-off-by: James Almer <jamrial@gmail.com>
Signed-off-by: Michael Niedermayer <michaelni@gmx.at>
2014-07-04 02:12:16 +02:00

84 lines
3.2 KiB
C

/*
* Copyright (C) 2012 Michael Niedermayer (michaelni@gmx.at)
*
* This file is part of libswresample
*
* libswresample is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* libswresample is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with libswresample; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "libavutil/x86/cpu.h"
#include "libswresample/swresample_internal.h"
#define D(type, simd) \
mix_1_1_func_type ff_mix_1_1_a_## type ## _ ## simd;\
mix_2_1_func_type ff_mix_2_1_a_## type ## _ ## simd;
D(float, sse)
D(float, avx)
D(int16, mmx)
D(int16, sse2)
av_cold void swri_rematrix_init_x86(struct SwrContext *s){
#if HAVE_YASM
int mm_flags = av_get_cpu_flags();
int nb_in = av_get_channel_layout_nb_channels(s->in_ch_layout);
int nb_out = av_get_channel_layout_nb_channels(s->out_ch_layout);
int num = nb_in * nb_out;
int i,j;
s->mix_1_1_simd = NULL;
s->mix_2_1_simd = NULL;
if (s->midbuf.fmt == AV_SAMPLE_FMT_S16P){
if(EXTERNAL_MMX(mm_flags)) {
s->mix_1_1_simd = ff_mix_1_1_a_int16_mmx;
s->mix_2_1_simd = ff_mix_2_1_a_int16_mmx;
}
if(EXTERNAL_SSE2(mm_flags)) {
s->mix_1_1_simd = ff_mix_1_1_a_int16_sse2;
s->mix_2_1_simd = ff_mix_2_1_a_int16_sse2;
}
s->native_simd_matrix = av_mallocz(2 * num * sizeof(int16_t));
s->native_simd_one = av_mallocz(2 * sizeof(int16_t));
for(i=0; i<nb_out; i++){
int sh = 0;
for(j=0; j<nb_in; j++)
sh = FFMAX(sh, FFABS(((int*)s->native_matrix)[i * nb_in + j]));
sh = FFMAX(av_log2(sh) - 14, 0);
for(j=0; j<nb_in; j++) {
((int16_t*)s->native_simd_matrix)[2*(i * nb_in + j)+1] = 15 - sh;
((int16_t*)s->native_simd_matrix)[2*(i * nb_in + j)] =
((((int*)s->native_matrix)[i * nb_in + j]) + (1<<sh>>1)) >> sh;
}
}
((int16_t*)s->native_simd_one)[1] = 14;
((int16_t*)s->native_simd_one)[0] = 16384;
} else if(s->midbuf.fmt == AV_SAMPLE_FMT_FLTP){
if(EXTERNAL_SSE(mm_flags)) {
s->mix_1_1_simd = ff_mix_1_1_a_float_sse;
s->mix_2_1_simd = ff_mix_2_1_a_float_sse;
}
if(EXTERNAL_AVX(mm_flags)) {
s->mix_1_1_simd = ff_mix_1_1_a_float_avx;
s->mix_2_1_simd = ff_mix_2_1_a_float_avx;
}
s->native_simd_matrix = av_mallocz(num * sizeof(float));
memcpy(s->native_simd_matrix, s->native_matrix, num * sizeof(float));
s->native_simd_one = av_mallocz(sizeof(float));
memcpy(s->native_simd_one, s->native_one, sizeof(float));
}
#endif
}