1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-28 20:53:54 +02:00
FFmpeg/libavcodec/vp8dsp.h

109 lines
4.7 KiB
C
Raw Normal View History

/*
* Copyright (C) 2010 David Conrad
* Copyright (C) 2010 Ronald S. Bultje
*
* This file is part of FFmpeg.
*
* FFmpeg 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.
*
* FFmpeg 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 FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
/**
* @file
* VP8 compatible video decoder
*/
#ifndef AVCODEC_VP8DSP_H
#define AVCODEC_VP8DSP_H
#include <stddef.h>
#include <stdint.h>
typedef void (*vp8_mc_func)(uint8_t *dst /* align 8 */, ptrdiff_t dstStride,
uint8_t *src /* align 1 */, ptrdiff_t srcStride,
int h, int x, int y);
typedef struct VP8DSPContext {
void (*vp8_luma_dc_wht)(int16_t block[4][4][16], int16_t dc[16]);
void (*vp8_luma_dc_wht_dc)(int16_t block[4][4][16], int16_t dc[16]);
void (*vp8_idct_add)(uint8_t *dst, int16_t block[16], ptrdiff_t stride);
void (*vp8_idct_dc_add)(uint8_t *dst, int16_t block[16], ptrdiff_t stride);
void (*vp8_idct_dc_add4y)(uint8_t *dst, int16_t block[4][16],
ptrdiff_t stride);
void (*vp8_idct_dc_add4uv)(uint8_t *dst, int16_t block[4][16],
ptrdiff_t stride);
// loop filter applied to edges between macroblocks
void (*vp8_v_loop_filter16y)(uint8_t *dst, ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_h_loop_filter16y)(uint8_t *dst, ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_v_loop_filter8uv)(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_h_loop_filter8uv)(uint8_t *dstU, uint8_t *dstV, ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
// loop filter applied to inner macroblock edges
void (*vp8_v_loop_filter16y_inner)(uint8_t *dst, ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_h_loop_filter16y_inner)(uint8_t *dst, ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_v_loop_filter8uv_inner)(uint8_t *dstU, uint8_t *dstV,
ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_h_loop_filter8uv_inner)(uint8_t *dstU, uint8_t *dstV,
ptrdiff_t stride,
int flim_E, int flim_I, int hev_thresh);
void (*vp8_v_loop_filter_simple)(uint8_t *dst, ptrdiff_t stride, int flim);
void (*vp8_h_loop_filter_simple)(uint8_t *dst, ptrdiff_t stride, int flim);
/**
* first dimension: 4-log2(width)
* second dimension: 0 if no vertical interpolation is needed;
* 1 4-tap vertical interpolation filter (my & 1)
* 2 6-tap vertical interpolation filter (!(my & 1))
* third dimension: same as second dimension, for horizontal interpolation
* so something like put_vp8_epel_pixels_tab[4-log2(width)][2*!!my-(my&1)][2*!!mx-(mx&1)](..., mx, my)
*/
vp8_mc_func put_vp8_epel_pixels_tab[3][3][3];
vp8_mc_func put_vp8_bilinear_pixels_tab[3][3][3];
} VP8DSPContext;
void ff_put_vp8_pixels16_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
int h, int x, int y);
void ff_put_vp8_pixels8_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
int h, int x, int y);
void ff_put_vp8_pixels4_c(uint8_t *dst, uint8_t *src, ptrdiff_t stride,
int h, int x, int y);
void ff_vp7dsp_init(VP8DSPContext *c);
void ff_vp78dsp_init(VP8DSPContext *c);
libavcodec: vp8 neon optimizations for aarch64 Partial port of the ARM Neon for aarch64. Benchmarks from fate: benchmarking with Linux Perf Monitoring API nop: 58.6 checkasm: using random seed 1760970128 NEON: - vp8dsp.idct [OK] - vp8dsp.mc [OK] - vp8dsp.loopfilter [OK] checkasm: all 21 tests passed vp8_idct_add_c: 201.6 vp8_idct_add_neon: 83.1 vp8_idct_dc_add_c: 107.6 vp8_idct_dc_add_neon: 33.8 vp8_idct_dc_add4y_c: 426.4 vp8_idct_dc_add4y_neon: 59.4 vp8_loop_filter8uv_h_c: 688.1 vp8_loop_filter8uv_h_neon: 216.3 vp8_loop_filter8uv_inner_h_c: 649.3 vp8_loop_filter8uv_inner_h_neon: 195.3 vp8_loop_filter8uv_inner_v_c: 544.8 vp8_loop_filter8uv_inner_v_neon: 131.3 vp8_loop_filter8uv_v_c: 706.1 vp8_loop_filter8uv_v_neon: 141.1 vp8_loop_filter16y_h_c: 668.8 vp8_loop_filter16y_h_neon: 242.8 vp8_loop_filter16y_inner_h_c: 647.3 vp8_loop_filter16y_inner_h_neon: 224.6 vp8_loop_filter16y_inner_v_c: 647.8 vp8_loop_filter16y_inner_v_neon: 128.8 vp8_loop_filter16y_v_c: 721.8 vp8_loop_filter16y_v_neon: 154.3 vp8_loop_filter_simple_h_c: 387.8 vp8_loop_filter_simple_h_neon: 187.6 vp8_loop_filter_simple_v_c: 384.1 vp8_loop_filter_simple_v_neon: 78.6 vp8_put_epel8_h4v4_c: 3971.1 vp8_put_epel8_h4v4_neon: 855.1 vp8_put_epel8_h4v6_c: 5060.1 vp8_put_epel8_h4v6_neon: 989.6 vp8_put_epel8_h6v4_c: 4320.8 vp8_put_epel8_h6v4_neon: 1007.3 vp8_put_epel8_h6v6_c: 5449.3 vp8_put_epel8_h6v6_neon: 1158.1 vp8_put_epel16_h6_c: 6683.8 vp8_put_epel16_h6_neon: 831.8 vp8_put_epel16_h6v6_c: 11110.8 vp8_put_epel16_h6v6_neon: 2214.8 vp8_put_epel16_v6_c: 7024.8 vp8_put_epel16_v6_neon: 799.6 vp8_put_pixels8_c: 112.8 vp8_put_pixels8_neon: 78.1 vp8_put_pixels16_c: 131.3 vp8_put_pixels16_neon: 129.8 Signed-off-by: Magnus Röös <mla2.roos@gmail.com>
2019-01-31 18:04:27 +02:00
void ff_vp78dsp_init_aarch64(VP8DSPContext *c);
void ff_vp78dsp_init_arm(VP8DSPContext *c);
void ff_vp78dsp_init_ppc(VP8DSPContext *c);
void ff_vp78dsp_init_x86(VP8DSPContext *c);
void ff_vp8dsp_init(VP8DSPContext *c);
libavcodec: vp8 neon optimizations for aarch64 Partial port of the ARM Neon for aarch64. Benchmarks from fate: benchmarking with Linux Perf Monitoring API nop: 58.6 checkasm: using random seed 1760970128 NEON: - vp8dsp.idct [OK] - vp8dsp.mc [OK] - vp8dsp.loopfilter [OK] checkasm: all 21 tests passed vp8_idct_add_c: 201.6 vp8_idct_add_neon: 83.1 vp8_idct_dc_add_c: 107.6 vp8_idct_dc_add_neon: 33.8 vp8_idct_dc_add4y_c: 426.4 vp8_idct_dc_add4y_neon: 59.4 vp8_loop_filter8uv_h_c: 688.1 vp8_loop_filter8uv_h_neon: 216.3 vp8_loop_filter8uv_inner_h_c: 649.3 vp8_loop_filter8uv_inner_h_neon: 195.3 vp8_loop_filter8uv_inner_v_c: 544.8 vp8_loop_filter8uv_inner_v_neon: 131.3 vp8_loop_filter8uv_v_c: 706.1 vp8_loop_filter8uv_v_neon: 141.1 vp8_loop_filter16y_h_c: 668.8 vp8_loop_filter16y_h_neon: 242.8 vp8_loop_filter16y_inner_h_c: 647.3 vp8_loop_filter16y_inner_h_neon: 224.6 vp8_loop_filter16y_inner_v_c: 647.8 vp8_loop_filter16y_inner_v_neon: 128.8 vp8_loop_filter16y_v_c: 721.8 vp8_loop_filter16y_v_neon: 154.3 vp8_loop_filter_simple_h_c: 387.8 vp8_loop_filter_simple_h_neon: 187.6 vp8_loop_filter_simple_v_c: 384.1 vp8_loop_filter_simple_v_neon: 78.6 vp8_put_epel8_h4v4_c: 3971.1 vp8_put_epel8_h4v4_neon: 855.1 vp8_put_epel8_h4v6_c: 5060.1 vp8_put_epel8_h4v6_neon: 989.6 vp8_put_epel8_h6v4_c: 4320.8 vp8_put_epel8_h6v4_neon: 1007.3 vp8_put_epel8_h6v6_c: 5449.3 vp8_put_epel8_h6v6_neon: 1158.1 vp8_put_epel16_h6_c: 6683.8 vp8_put_epel16_h6_neon: 831.8 vp8_put_epel16_h6v6_c: 11110.8 vp8_put_epel16_h6v6_neon: 2214.8 vp8_put_epel16_v6_c: 7024.8 vp8_put_epel16_v6_neon: 799.6 vp8_put_pixels8_c: 112.8 vp8_put_pixels8_neon: 78.1 vp8_put_pixels16_c: 131.3 vp8_put_pixels16_neon: 129.8 Signed-off-by: Magnus Röös <mla2.roos@gmail.com>
2019-01-31 18:04:27 +02:00
void ff_vp8dsp_init_aarch64(VP8DSPContext *c);
void ff_vp8dsp_init_arm(VP8DSPContext *c);
void ff_vp8dsp_init_x86(VP8DSPContext *c);
void ff_vp8dsp_init_mips(VP8DSPContext *c);
#define IS_VP7 1
#define IS_VP8 0
#endif /* AVCODEC_VP8DSP_H */