yading@10: /*
yading@10:  * ARM NEON optimised DSP functions
yading@10:  * Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
yading@10:  *
yading@10:  * This file is part of FFmpeg.
yading@10:  *
yading@10:  * FFmpeg is free software; you can redistribute it and/or
yading@10:  * modify it under the terms of the GNU Lesser General Public
yading@10:  * License as published by the Free Software Foundation; either
yading@10:  * version 2.1 of the License, or (at your option) any later version.
yading@10:  *
yading@10:  * FFmpeg is distributed in the hope that it will be useful,
yading@10:  * but WITHOUT ANY WARRANTY; without even the implied warranty of
yading@10:  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
yading@10:  * Lesser General Public License for more details.
yading@10:  *
yading@10:  * You should have received a copy of the GNU Lesser General Public
yading@10:  * License along with FFmpeg; if not, write to the Free Software
yading@10:  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
yading@10:  */
yading@10: 
yading@10: #include <stddef.h>
yading@10: #include <stdint.h>
yading@10: 
yading@10: #include "libavutil/attributes.h"
yading@10: #include "hpeldsp_arm.h"
yading@10: 
yading@10: void ff_put_pixels16_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels16_x2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels16_y2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels16_xy2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_x2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_y2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_xy2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels16_x2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels16_y2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels16_xy2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_x2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_y2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_put_pixels8_xy2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: 
yading@10: void ff_avg_pixels16_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels16_x2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels16_y2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels16_xy2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels8_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels8_x2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels8_y2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels8_xy2_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels16_x2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels16_y2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: void ff_avg_pixels16_xy2_no_rnd_neon(uint8_t *, const uint8_t *, ptrdiff_t, int);
yading@10: 
yading@10: av_cold void ff_hpeldsp_init_neon(HpelDSPContext *c, int flags)
yading@10: {
yading@10:     c->put_pixels_tab[0][0] = ff_put_pixels16_neon;
yading@10:     c->put_pixels_tab[0][1] = ff_put_pixels16_x2_neon;
yading@10:     c->put_pixels_tab[0][2] = ff_put_pixels16_y2_neon;
yading@10:     c->put_pixels_tab[0][3] = ff_put_pixels16_xy2_neon;
yading@10:     c->put_pixels_tab[1][0] = ff_put_pixels8_neon;
yading@10:     c->put_pixels_tab[1][1] = ff_put_pixels8_x2_neon;
yading@10:     c->put_pixels_tab[1][2] = ff_put_pixels8_y2_neon;
yading@10:     c->put_pixels_tab[1][3] = ff_put_pixels8_xy2_neon;
yading@10: 
yading@10:     c->put_no_rnd_pixels_tab[0][0] = ff_put_pixels16_neon;
yading@10:     c->put_no_rnd_pixels_tab[0][1] = ff_put_pixels16_x2_no_rnd_neon;
yading@10:     c->put_no_rnd_pixels_tab[0][2] = ff_put_pixels16_y2_no_rnd_neon;
yading@10:     c->put_no_rnd_pixels_tab[0][3] = ff_put_pixels16_xy2_no_rnd_neon;
yading@10:     c->put_no_rnd_pixels_tab[1][0] = ff_put_pixels8_neon;
yading@10:     c->put_no_rnd_pixels_tab[1][1] = ff_put_pixels8_x2_no_rnd_neon;
yading@10:     c->put_no_rnd_pixels_tab[1][2] = ff_put_pixels8_y2_no_rnd_neon;
yading@10:     c->put_no_rnd_pixels_tab[1][3] = ff_put_pixels8_xy2_no_rnd_neon;
yading@10: 
yading@10:     c->avg_pixels_tab[0][0] = ff_avg_pixels16_neon;
yading@10:     c->avg_pixels_tab[0][1] = ff_avg_pixels16_x2_neon;
yading@10:     c->avg_pixels_tab[0][2] = ff_avg_pixels16_y2_neon;
yading@10:     c->avg_pixels_tab[0][3] = ff_avg_pixels16_xy2_neon;
yading@10:     c->avg_pixels_tab[1][0] = ff_avg_pixels8_neon;
yading@10:     c->avg_pixels_tab[1][1] = ff_avg_pixels8_x2_neon;
yading@10:     c->avg_pixels_tab[1][2] = ff_avg_pixels8_y2_neon;
yading@10:     c->avg_pixels_tab[1][3] = ff_avg_pixels8_xy2_neon;
yading@10: 
yading@10:     c->avg_no_rnd_pixels_tab[0] = ff_avg_pixels16_neon;
yading@10:     c->avg_no_rnd_pixels_tab[1] = ff_avg_pixels16_x2_no_rnd_neon;
yading@10:     c->avg_no_rnd_pixels_tab[2] = ff_avg_pixels16_y2_no_rnd_neon;
yading@10:     c->avg_no_rnd_pixels_tab[3] = ff_avg_pixels16_xy2_no_rnd_neon;
yading@10: }