File size: 7,273 Bytes
8ead80b |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 |
/*
* Copyright (c) 2011 Janne Grunau <[email protected]>
*
* 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
*/
#include <stdint.h>
#include "libavutil/attributes.h"
#include "libavcodec/avcodec.h"
#include "libavcodec/rv34dsp.h"
#include "libavutil/arm/cpu.h"
#define DECL_QPEL3(type, w, pos) \
void ff_ ## type ## _rv40_qpel ## w ## _mc ## pos ## _neon(uint8_t *dst, \
const uint8_t *src, \
ptrdiff_t stride)
#define DECL_QPEL2(w, pos) \
DECL_QPEL3(put, w, pos); \
DECL_QPEL3(avg, w, pos)
#define DECL_QPEL_XY(x, y) \
DECL_QPEL2(16, x ## y); \
DECL_QPEL2(8, x ## y)
#define DECL_QPEL_Y(y) \
DECL_QPEL_XY(0, y); \
DECL_QPEL_XY(1, y); \
DECL_QPEL_XY(2, y); \
DECL_QPEL_XY(3, y); \
DECL_QPEL_Y(0);
DECL_QPEL_Y(1);
DECL_QPEL_Y(2);
DECL_QPEL_Y(3);
void ff_put_rv40_chroma_mc8_neon(uint8_t *, const uint8_t *, ptrdiff_t, int, int, int);
void ff_put_rv40_chroma_mc4_neon(uint8_t *, const uint8_t *, ptrdiff_t, int, int, int);
void ff_avg_rv40_chroma_mc8_neon(uint8_t *, const uint8_t *, ptrdiff_t, int, int, int);
void ff_avg_rv40_chroma_mc4_neon(uint8_t *, const uint8_t *, ptrdiff_t, int, int, int);
void ff_rv40_weight_func_16_neon(uint8_t *, uint8_t *, uint8_t *, int, int, ptrdiff_t);
void ff_rv40_weight_func_8_neon(uint8_t *, uint8_t *, uint8_t *, int, int, ptrdiff_t);
int ff_rv40_h_loop_filter_strength_neon(uint8_t *src, ptrdiff_t stride,
int beta, int beta2, int edge,
int *p1, int *q1);
int ff_rv40_v_loop_filter_strength_neon(uint8_t *src, ptrdiff_t stride,
int beta, int beta2, int edge,
int *p1, int *q1);
void ff_rv40_h_weak_loop_filter_neon(uint8_t *src, ptrdiff_t stride, int filter_p1,
int filter_q1, int alpha, int beta,
int lim_p0q0, int lim_q1, int lim_p1);
void ff_rv40_v_weak_loop_filter_neon(uint8_t *src, ptrdiff_t stride, int filter_p1,
int filter_q1, int alpha, int beta,
int lim_p0q0, int lim_q1, int lim_p1);
static av_cold void rv40dsp_init_neon(RV34DSPContext *c)
{
c->put_pixels_tab[0][ 1] = ff_put_rv40_qpel16_mc10_neon;
c->put_pixels_tab[0][ 3] = ff_put_rv40_qpel16_mc30_neon;
c->put_pixels_tab[0][ 4] = ff_put_rv40_qpel16_mc01_neon;
c->put_pixels_tab[0][ 5] = ff_put_rv40_qpel16_mc11_neon;
c->put_pixels_tab[0][ 6] = ff_put_rv40_qpel16_mc21_neon;
c->put_pixels_tab[0][ 7] = ff_put_rv40_qpel16_mc31_neon;
c->put_pixels_tab[0][ 9] = ff_put_rv40_qpel16_mc12_neon;
c->put_pixels_tab[0][10] = ff_put_rv40_qpel16_mc22_neon;
c->put_pixels_tab[0][11] = ff_put_rv40_qpel16_mc32_neon;
c->put_pixels_tab[0][12] = ff_put_rv40_qpel16_mc03_neon;
c->put_pixels_tab[0][13] = ff_put_rv40_qpel16_mc13_neon;
c->put_pixels_tab[0][14] = ff_put_rv40_qpel16_mc23_neon;
c->put_pixels_tab[0][15] = ff_put_rv40_qpel16_mc33_neon;
c->avg_pixels_tab[0][ 1] = ff_avg_rv40_qpel16_mc10_neon;
c->avg_pixels_tab[0][ 3] = ff_avg_rv40_qpel16_mc30_neon;
c->avg_pixels_tab[0][ 4] = ff_avg_rv40_qpel16_mc01_neon;
c->avg_pixels_tab[0][ 5] = ff_avg_rv40_qpel16_mc11_neon;
c->avg_pixels_tab[0][ 6] = ff_avg_rv40_qpel16_mc21_neon;
c->avg_pixels_tab[0][ 7] = ff_avg_rv40_qpel16_mc31_neon;
c->avg_pixels_tab[0][ 9] = ff_avg_rv40_qpel16_mc12_neon;
c->avg_pixels_tab[0][10] = ff_avg_rv40_qpel16_mc22_neon;
c->avg_pixels_tab[0][11] = ff_avg_rv40_qpel16_mc32_neon;
c->avg_pixels_tab[0][12] = ff_avg_rv40_qpel16_mc03_neon;
c->avg_pixels_tab[0][13] = ff_avg_rv40_qpel16_mc13_neon;
c->avg_pixels_tab[0][14] = ff_avg_rv40_qpel16_mc23_neon;
c->avg_pixels_tab[0][15] = ff_avg_rv40_qpel16_mc33_neon;
c->put_pixels_tab[1][ 1] = ff_put_rv40_qpel8_mc10_neon;
c->put_pixels_tab[1][ 3] = ff_put_rv40_qpel8_mc30_neon;
c->put_pixels_tab[1][ 4] = ff_put_rv40_qpel8_mc01_neon;
c->put_pixels_tab[1][ 5] = ff_put_rv40_qpel8_mc11_neon;
c->put_pixels_tab[1][ 6] = ff_put_rv40_qpel8_mc21_neon;
c->put_pixels_tab[1][ 7] = ff_put_rv40_qpel8_mc31_neon;
c->put_pixels_tab[1][ 9] = ff_put_rv40_qpel8_mc12_neon;
c->put_pixels_tab[1][10] = ff_put_rv40_qpel8_mc22_neon;
c->put_pixels_tab[1][11] = ff_put_rv40_qpel8_mc32_neon;
c->put_pixels_tab[1][12] = ff_put_rv40_qpel8_mc03_neon;
c->put_pixels_tab[1][13] = ff_put_rv40_qpel8_mc13_neon;
c->put_pixels_tab[1][14] = ff_put_rv40_qpel8_mc23_neon;
c->put_pixels_tab[1][15] = ff_put_rv40_qpel8_mc33_neon;
c->avg_pixels_tab[1][ 1] = ff_avg_rv40_qpel8_mc10_neon;
c->avg_pixels_tab[1][ 3] = ff_avg_rv40_qpel8_mc30_neon;
c->avg_pixels_tab[1][ 4] = ff_avg_rv40_qpel8_mc01_neon;
c->avg_pixels_tab[1][ 5] = ff_avg_rv40_qpel8_mc11_neon;
c->avg_pixels_tab[1][ 6] = ff_avg_rv40_qpel8_mc21_neon;
c->avg_pixels_tab[1][ 7] = ff_avg_rv40_qpel8_mc31_neon;
c->avg_pixels_tab[1][ 9] = ff_avg_rv40_qpel8_mc12_neon;
c->avg_pixels_tab[1][10] = ff_avg_rv40_qpel8_mc22_neon;
c->avg_pixels_tab[1][11] = ff_avg_rv40_qpel8_mc32_neon;
c->avg_pixels_tab[1][12] = ff_avg_rv40_qpel8_mc03_neon;
c->avg_pixels_tab[1][13] = ff_avg_rv40_qpel8_mc13_neon;
c->avg_pixels_tab[1][14] = ff_avg_rv40_qpel8_mc23_neon;
c->avg_pixels_tab[1][15] = ff_avg_rv40_qpel8_mc33_neon;
c->put_chroma_pixels_tab[0] = ff_put_rv40_chroma_mc8_neon;
c->put_chroma_pixels_tab[1] = ff_put_rv40_chroma_mc4_neon;
c->avg_chroma_pixels_tab[0] = ff_avg_rv40_chroma_mc8_neon;
c->avg_chroma_pixels_tab[1] = ff_avg_rv40_chroma_mc4_neon;
c->rv40_weight_pixels_tab[0][0] = ff_rv40_weight_func_16_neon;
c->rv40_weight_pixels_tab[0][1] = ff_rv40_weight_func_8_neon;
c->rv40_loop_filter_strength[0] = ff_rv40_h_loop_filter_strength_neon;
c->rv40_loop_filter_strength[1] = ff_rv40_v_loop_filter_strength_neon;
c->rv40_weak_loop_filter[0] = ff_rv40_h_weak_loop_filter_neon;
c->rv40_weak_loop_filter[1] = ff_rv40_v_weak_loop_filter_neon;
}
av_cold void ff_rv40dsp_init_arm(RV34DSPContext *c)
{
int cpu_flags = av_get_cpu_flags();
if (have_neon(cpu_flags))
rv40dsp_init_neon(c);
}
|