Commit fc13a896 authored by Michael Niedermayer's avatar Michael Niedermayer

Merge remote-tracking branch 'qatar/master'

* qatar/master:
  dsputil: Separate h264 qpel

Conflicts:
	libavcodec/dsputil_template.c
	libavcodec/h264.c
Merged-by: 's avatarMichael Niedermayer <michaelni@gmx.at>
parents 70b0aeba e9d81735
......@@ -50,6 +50,7 @@ OBJS-$(CONFIG_FFT) += avfft.o fft_fixed.o fft_float.o \
OBJS-$(CONFIG_GOLOMB) += golomb.o
OBJS-$(CONFIG_H264DSP) += h264dsp.o h264idct.o
OBJS-$(CONFIG_H264PRED) += h264pred.o
OBJS-$(CONFIG_H264QPEL) += h264qpel.o
OBJS-$(CONFIG_HUFFMAN) += huffman.o
OBJS-$(CONFIG_LIBXVID) += libxvid_rc.o
OBJS-$(CONFIG_LPC) += lpc.o
......
......@@ -26,6 +26,7 @@ ARMV6-OBJS-$(CONFIG_VP8_DECODER) += arm/vp8_armv6.o \
OBJS-$(CONFIG_H264DSP) += arm/h264dsp_init_arm.o
OBJS-$(CONFIG_H264PRED) += arm/h264pred_init_arm.o
OBJS-$(CONFIG_H264QPEL) += arm/h264qpel_init_arm.o
OBJS-$(CONFIG_RV30_DECODER) += arm/rv34dsp_init_arm.o
OBJS-$(CONFIG_RV40_DECODER) += arm/rv34dsp_init_arm.o \
......@@ -70,6 +71,8 @@ NEON-OBJS-$(CONFIG_H264DSP) += arm/h264dsp_neon.o \
NEON-OBJS-$(CONFIG_H264PRED) += arm/h264pred_neon.o \
NEON-OBJS-$(CONFIG_H264QPEL) += arm/h264qpel_neon.o \
NEON-OBJS-$(CONFIG_AC3DSP) += arm/ac3dsp_neon.o
NEON-OBJS-$(CONFIG_AAC_DECODER) += arm/sbrdsp_neon.o \
......
......@@ -63,74 +63,6 @@ void ff_add_pixels_clamped_neon(const int16_t *, uint8_t *, int);
void ff_put_pixels_clamped_neon(const int16_t *, uint8_t *, int);
void ff_put_signed_pixels_clamped_neon(const int16_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_chroma_mc8_neon(uint8_t *, uint8_t *, int, int, int, int);
void ff_put_h264_chroma_mc4_neon(uint8_t *, uint8_t *, int, int, int, int);
void ff_put_h264_chroma_mc2_neon(uint8_t *, uint8_t *, int, int, int, int);
......@@ -214,74 +146,6 @@ void ff_dsputil_init_neon(DSPContext *c, AVCodecContext *avctx)
c->avg_h264_chroma_pixels_tab[0] = ff_avg_h264_chroma_mc8_neon;
c->avg_h264_chroma_pixels_tab[1] = ff_avg_h264_chroma_mc4_neon;
c->avg_h264_chroma_pixels_tab[2] = ff_avg_h264_chroma_mc2_neon;
c->put_h264_qpel_pixels_tab[0][ 0] = ff_put_h264_qpel16_mc00_neon;
c->put_h264_qpel_pixels_tab[0][ 1] = ff_put_h264_qpel16_mc10_neon;
c->put_h264_qpel_pixels_tab[0][ 2] = ff_put_h264_qpel16_mc20_neon;
c->put_h264_qpel_pixels_tab[0][ 3] = ff_put_h264_qpel16_mc30_neon;
c->put_h264_qpel_pixels_tab[0][ 4] = ff_put_h264_qpel16_mc01_neon;
c->put_h264_qpel_pixels_tab[0][ 5] = ff_put_h264_qpel16_mc11_neon;
c->put_h264_qpel_pixels_tab[0][ 6] = ff_put_h264_qpel16_mc21_neon;
c->put_h264_qpel_pixels_tab[0][ 7] = ff_put_h264_qpel16_mc31_neon;
c->put_h264_qpel_pixels_tab[0][ 8] = ff_put_h264_qpel16_mc02_neon;
c->put_h264_qpel_pixels_tab[0][ 9] = ff_put_h264_qpel16_mc12_neon;
c->put_h264_qpel_pixels_tab[0][10] = ff_put_h264_qpel16_mc22_neon;
c->put_h264_qpel_pixels_tab[0][11] = ff_put_h264_qpel16_mc32_neon;
c->put_h264_qpel_pixels_tab[0][12] = ff_put_h264_qpel16_mc03_neon;
c->put_h264_qpel_pixels_tab[0][13] = ff_put_h264_qpel16_mc13_neon;
c->put_h264_qpel_pixels_tab[0][14] = ff_put_h264_qpel16_mc23_neon;
c->put_h264_qpel_pixels_tab[0][15] = ff_put_h264_qpel16_mc33_neon;
c->put_h264_qpel_pixels_tab[1][ 0] = ff_put_h264_qpel8_mc00_neon;
c->put_h264_qpel_pixels_tab[1][ 1] = ff_put_h264_qpel8_mc10_neon;
c->put_h264_qpel_pixels_tab[1][ 2] = ff_put_h264_qpel8_mc20_neon;
c->put_h264_qpel_pixels_tab[1][ 3] = ff_put_h264_qpel8_mc30_neon;
c->put_h264_qpel_pixels_tab[1][ 4] = ff_put_h264_qpel8_mc01_neon;
c->put_h264_qpel_pixels_tab[1][ 5] = ff_put_h264_qpel8_mc11_neon;
c->put_h264_qpel_pixels_tab[1][ 6] = ff_put_h264_qpel8_mc21_neon;
c->put_h264_qpel_pixels_tab[1][ 7] = ff_put_h264_qpel8_mc31_neon;
c->put_h264_qpel_pixels_tab[1][ 8] = ff_put_h264_qpel8_mc02_neon;
c->put_h264_qpel_pixels_tab[1][ 9] = ff_put_h264_qpel8_mc12_neon;
c->put_h264_qpel_pixels_tab[1][10] = ff_put_h264_qpel8_mc22_neon;
c->put_h264_qpel_pixels_tab[1][11] = ff_put_h264_qpel8_mc32_neon;
c->put_h264_qpel_pixels_tab[1][12] = ff_put_h264_qpel8_mc03_neon;
c->put_h264_qpel_pixels_tab[1][13] = ff_put_h264_qpel8_mc13_neon;
c->put_h264_qpel_pixels_tab[1][14] = ff_put_h264_qpel8_mc23_neon;
c->put_h264_qpel_pixels_tab[1][15] = ff_put_h264_qpel8_mc33_neon;
c->avg_h264_qpel_pixels_tab[0][ 0] = ff_avg_h264_qpel16_mc00_neon;
c->avg_h264_qpel_pixels_tab[0][ 1] = ff_avg_h264_qpel16_mc10_neon;
c->avg_h264_qpel_pixels_tab[0][ 2] = ff_avg_h264_qpel16_mc20_neon;
c->avg_h264_qpel_pixels_tab[0][ 3] = ff_avg_h264_qpel16_mc30_neon;
c->avg_h264_qpel_pixels_tab[0][ 4] = ff_avg_h264_qpel16_mc01_neon;
c->avg_h264_qpel_pixels_tab[0][ 5] = ff_avg_h264_qpel16_mc11_neon;
c->avg_h264_qpel_pixels_tab[0][ 6] = ff_avg_h264_qpel16_mc21_neon;
c->avg_h264_qpel_pixels_tab[0][ 7] = ff_avg_h264_qpel16_mc31_neon;
c->avg_h264_qpel_pixels_tab[0][ 8] = ff_avg_h264_qpel16_mc02_neon;
c->avg_h264_qpel_pixels_tab[0][ 9] = ff_avg_h264_qpel16_mc12_neon;
c->avg_h264_qpel_pixels_tab[0][10] = ff_avg_h264_qpel16_mc22_neon;
c->avg_h264_qpel_pixels_tab[0][11] = ff_avg_h264_qpel16_mc32_neon;
c->avg_h264_qpel_pixels_tab[0][12] = ff_avg_h264_qpel16_mc03_neon;
c->avg_h264_qpel_pixels_tab[0][13] = ff_avg_h264_qpel16_mc13_neon;
c->avg_h264_qpel_pixels_tab[0][14] = ff_avg_h264_qpel16_mc23_neon;
c->avg_h264_qpel_pixels_tab[0][15] = ff_avg_h264_qpel16_mc33_neon;
c->avg_h264_qpel_pixels_tab[1][ 0] = ff_avg_h264_qpel8_mc00_neon;
c->avg_h264_qpel_pixels_tab[1][ 1] = ff_avg_h264_qpel8_mc10_neon;
c->avg_h264_qpel_pixels_tab[1][ 2] = ff_avg_h264_qpel8_mc20_neon;
c->avg_h264_qpel_pixels_tab[1][ 3] = ff_avg_h264_qpel8_mc30_neon;
c->avg_h264_qpel_pixels_tab[1][ 4] = ff_avg_h264_qpel8_mc01_neon;
c->avg_h264_qpel_pixels_tab[1][ 5] = ff_avg_h264_qpel8_mc11_neon;
c->avg_h264_qpel_pixels_tab[1][ 6] = ff_avg_h264_qpel8_mc21_neon;
c->avg_h264_qpel_pixels_tab[1][ 7] = ff_avg_h264_qpel8_mc31_neon;
c->avg_h264_qpel_pixels_tab[1][ 8] = ff_avg_h264_qpel8_mc02_neon;
c->avg_h264_qpel_pixels_tab[1][ 9] = ff_avg_h264_qpel8_mc12_neon;
c->avg_h264_qpel_pixels_tab[1][10] = ff_avg_h264_qpel8_mc22_neon;
c->avg_h264_qpel_pixels_tab[1][11] = ff_avg_h264_qpel8_mc32_neon;
c->avg_h264_qpel_pixels_tab[1][12] = ff_avg_h264_qpel8_mc03_neon;
c->avg_h264_qpel_pixels_tab[1][13] = ff_avg_h264_qpel8_mc13_neon;
c->avg_h264_qpel_pixels_tab[1][14] = ff_avg_h264_qpel8_mc23_neon;
c->avg_h264_qpel_pixels_tab[1][15] = ff_avg_h264_qpel8_mc33_neon;
}
c->vector_clipf = ff_vector_clipf_neon;
......
This diff is collapsed.
/*
* ARM NEON optimised DSP functions
* Copyright (c) 2008 Mans Rullgard <mans@mansr.com>
*
* 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 "config.h"
#include "libavcodec/h264qpel.h"
void ff_put_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
void ff_put_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc00_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc10_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc20_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc30_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc01_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc11_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc21_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc31_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc02_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc12_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc22_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc32_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc03_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc13_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc23_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel16_mc33_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc00_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc10_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc20_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc30_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc01_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc11_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc21_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc31_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc02_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc12_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc22_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc32_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc03_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc13_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc23_neon(uint8_t *, uint8_t *, int);
void ff_avg_h264_qpel8_mc33_neon(uint8_t *, uint8_t *, int);
void ff_h264qpel_init_arm(H264QpelContext *c, int bit_depth)
{
#if HAVE_NEON
const int high_bit_depth = bit_depth > 8;
if (!high_bit_depth) {
c->put_h264_qpel_pixels_tab[0][ 0] = ff_put_h264_qpel16_mc00_neon;
c->put_h264_qpel_pixels_tab[0][ 1] = ff_put_h264_qpel16_mc10_neon;
c->put_h264_qpel_pixels_tab[0][ 2] = ff_put_h264_qpel16_mc20_neon;
c->put_h264_qpel_pixels_tab[0][ 3] = ff_put_h264_qpel16_mc30_neon;
c->put_h264_qpel_pixels_tab[0][ 4] = ff_put_h264_qpel16_mc01_neon;
c->put_h264_qpel_pixels_tab[0][ 5] = ff_put_h264_qpel16_mc11_neon;
c->put_h264_qpel_pixels_tab[0][ 6] = ff_put_h264_qpel16_mc21_neon;
c->put_h264_qpel_pixels_tab[0][ 7] = ff_put_h264_qpel16_mc31_neon;
c->put_h264_qpel_pixels_tab[0][ 8] = ff_put_h264_qpel16_mc02_neon;
c->put_h264_qpel_pixels_tab[0][ 9] = ff_put_h264_qpel16_mc12_neon;
c->put_h264_qpel_pixels_tab[0][10] = ff_put_h264_qpel16_mc22_neon;
c->put_h264_qpel_pixels_tab[0][11] = ff_put_h264_qpel16_mc32_neon;
c->put_h264_qpel_pixels_tab[0][12] = ff_put_h264_qpel16_mc03_neon;
c->put_h264_qpel_pixels_tab[0][13] = ff_put_h264_qpel16_mc13_neon;
c->put_h264_qpel_pixels_tab[0][14] = ff_put_h264_qpel16_mc23_neon;
c->put_h264_qpel_pixels_tab[0][15] = ff_put_h264_qpel16_mc33_neon;
c->put_h264_qpel_pixels_tab[1][ 0] = ff_put_h264_qpel8_mc00_neon;
c->put_h264_qpel_pixels_tab[1][ 1] = ff_put_h264_qpel8_mc10_neon;
c->put_h264_qpel_pixels_tab[1][ 2] = ff_put_h264_qpel8_mc20_neon;
c->put_h264_qpel_pixels_tab[1][ 3] = ff_put_h264_qpel8_mc30_neon;
c->put_h264_qpel_pixels_tab[1][ 4] = ff_put_h264_qpel8_mc01_neon;
c->put_h264_qpel_pixels_tab[1][ 5] = ff_put_h264_qpel8_mc11_neon;
c->put_h264_qpel_pixels_tab[1][ 6] = ff_put_h264_qpel8_mc21_neon;
c->put_h264_qpel_pixels_tab[1][ 7] = ff_put_h264_qpel8_mc31_neon;
c->put_h264_qpel_pixels_tab[1][ 8] = ff_put_h264_qpel8_mc02_neon;
c->put_h264_qpel_pixels_tab[1][ 9] = ff_put_h264_qpel8_mc12_neon;
c->put_h264_qpel_pixels_tab[1][10] = ff_put_h264_qpel8_mc22_neon;
c->put_h264_qpel_pixels_tab[1][11] = ff_put_h264_qpel8_mc32_neon;
c->put_h264_qpel_pixels_tab[1][12] = ff_put_h264_qpel8_mc03_neon;
c->put_h264_qpel_pixels_tab[1][13] = ff_put_h264_qpel8_mc13_neon;
c->put_h264_qpel_pixels_tab[1][14] = ff_put_h264_qpel8_mc23_neon;
c->put_h264_qpel_pixels_tab[1][15] = ff_put_h264_qpel8_mc33_neon;
c->avg_h264_qpel_pixels_tab[0][ 0] = ff_avg_h264_qpel16_mc00_neon;
c->avg_h264_qpel_pixels_tab[0][ 1] = ff_avg_h264_qpel16_mc10_neon;
c->avg_h264_qpel_pixels_tab[0][ 2] = ff_avg_h264_qpel16_mc20_neon;
c->avg_h264_qpel_pixels_tab[0][ 3] = ff_avg_h264_qpel16_mc30_neon;
c->avg_h264_qpel_pixels_tab[0][ 4] = ff_avg_h264_qpel16_mc01_neon;
c->avg_h264_qpel_pixels_tab[0][ 5] = ff_avg_h264_qpel16_mc11_neon;
c->avg_h264_qpel_pixels_tab[0][ 6] = ff_avg_h264_qpel16_mc21_neon;
c->avg_h264_qpel_pixels_tab[0][ 7] = ff_avg_h264_qpel16_mc31_neon;
c->avg_h264_qpel_pixels_tab[0][ 8] = ff_avg_h264_qpel16_mc02_neon;
c->avg_h264_qpel_pixels_tab[0][ 9] = ff_avg_h264_qpel16_mc12_neon;
c->avg_h264_qpel_pixels_tab[0][10] = ff_avg_h264_qpel16_mc22_neon;
c->avg_h264_qpel_pixels_tab[0][11] = ff_avg_h264_qpel16_mc32_neon;
c->avg_h264_qpel_pixels_tab[0][12] = ff_avg_h264_qpel16_mc03_neon;
c->avg_h264_qpel_pixels_tab[0][13] = ff_avg_h264_qpel16_mc13_neon;
c->avg_h264_qpel_pixels_tab[0][14] = ff_avg_h264_qpel16_mc23_neon;
c->avg_h264_qpel_pixels_tab[0][15] = ff_avg_h264_qpel16_mc33_neon;
c->avg_h264_qpel_pixels_tab[1][ 0] = ff_avg_h264_qpel8_mc00_neon;
c->avg_h264_qpel_pixels_tab[1][ 1] = ff_avg_h264_qpel8_mc10_neon;
c->avg_h264_qpel_pixels_tab[1][ 2] = ff_avg_h264_qpel8_mc20_neon;
c->avg_h264_qpel_pixels_tab[1][ 3] = ff_avg_h264_qpel8_mc30_neon;
c->avg_h264_qpel_pixels_tab[1][ 4] = ff_avg_h264_qpel8_mc01_neon;
c->avg_h264_qpel_pixels_tab[1][ 5] = ff_avg_h264_qpel8_mc11_neon;
c->avg_h264_qpel_pixels_tab[1][ 6] = ff_avg_h264_qpel8_mc21_neon;
c->avg_h264_qpel_pixels_tab[1][ 7] = ff_avg_h264_qpel8_mc31_neon;
c->avg_h264_qpel_pixels_tab[1][ 8] = ff_avg_h264_qpel8_mc02_neon;
c->avg_h264_qpel_pixels_tab[1][ 9] = ff_avg_h264_qpel8_mc12_neon;
c->avg_h264_qpel_pixels_tab[1][10] = ff_avg_h264_qpel8_mc22_neon;
c->avg_h264_qpel_pixels_tab[1][11] = ff_avg_h264_qpel8_mc32_neon;
c->avg_h264_qpel_pixels_tab[1][12] = ff_avg_h264_qpel8_mc03_neon;
c->avg_h264_qpel_pixels_tab[1][13] = ff_avg_h264_qpel8_mc13_neon;
c->avg_h264_qpel_pixels_tab[1][14] = ff_avg_h264_qpel8_mc23_neon;
c->avg_h264_qpel_pixels_tab[1][15] = ff_avg_h264_qpel8_mc33_neon;
}
#endif /* HAVE_NEON */
}
This diff is collapsed.
......@@ -41,7 +41,6 @@
uint8_t ff_cropTbl[256 + 2 * MAX_NEG_CROP] = {0, };
uint32_t ff_squareTbl[512] = {0, };
#define pixeltmp int16_t
#define BIT_DEPTH 9
#include "dsputil_template.c"
#undef BIT_DEPTH
......@@ -50,8 +49,6 @@ uint32_t ff_squareTbl[512] = {0, };
#include "dsputil_template.c"
#undef BIT_DEPTH
#undef pixeltmp
#define pixeltmp int32_t
#define BIT_DEPTH 12
#include "dsputil_template.c"
#undef BIT_DEPTH
......@@ -60,11 +57,8 @@ uint32_t ff_squareTbl[512] = {0, };
#include "dsputil_template.c"
#undef BIT_DEPTH
#undef pixeltmp
#define pixeltmp int16_t
#define BIT_DEPTH 8
#include "dsputil_template.c"
#undef pixeltmp
// 0x7f7f7f7f or 0x7f7f7f7f7f7f7f7f or whatever, depending on the cpu's native arithmetic size
#define pb_7f (~0UL/255 * 0x7f)
......@@ -2892,24 +2886,6 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx)
#define FUNC(f, depth) f ## _ ## depth
#define FUNCC(f, depth) f ## _ ## depth ## _c
#define dspfunc2(PFX, IDX, NUM, depth)\
c->PFX ## _pixels_tab[IDX][ 0] = FUNCC(PFX ## NUM ## _mc00, depth);\
c->PFX ## _pixels_tab[IDX][ 1] = FUNCC(PFX ## NUM ## _mc10, depth);\
c->PFX ## _pixels_tab[IDX][ 2] = FUNCC(PFX ## NUM ## _mc20, depth);\
c->PFX ## _pixels_tab[IDX][ 3] = FUNCC(PFX ## NUM ## _mc30, depth);\
c->PFX ## _pixels_tab[IDX][ 4] = FUNCC(PFX ## NUM ## _mc01, depth);\
c->PFX ## _pixels_tab[IDX][ 5] = FUNCC(PFX ## NUM ## _mc11, depth);\
c->PFX ## _pixels_tab[IDX][ 6] = FUNCC(PFX ## NUM ## _mc21, depth);\
c->PFX ## _pixels_tab[IDX][ 7] = FUNCC(PFX ## NUM ## _mc31, depth);\
c->PFX ## _pixels_tab[IDX][ 8] = FUNCC(PFX ## NUM ## _mc02, depth);\
c->PFX ## _pixels_tab[IDX][ 9] = FUNCC(PFX ## NUM ## _mc12, depth);\
c->PFX ## _pixels_tab[IDX][10] = FUNCC(PFX ## NUM ## _mc22, depth);\
c->PFX ## _pixels_tab[IDX][11] = FUNCC(PFX ## NUM ## _mc32, depth);\
c->PFX ## _pixels_tab[IDX][12] = FUNCC(PFX ## NUM ## _mc03, depth);\
c->PFX ## _pixels_tab[IDX][13] = FUNCC(PFX ## NUM ## _mc13, depth);\
c->PFX ## _pixels_tab[IDX][14] = FUNCC(PFX ## NUM ## _mc23, depth);\
c->PFX ## _pixels_tab[IDX][15] = FUNCC(PFX ## NUM ## _mc33, depth)
#define BIT_DEPTH_FUNCS(depth, dct)\
c->get_pixels = FUNCC(get_pixels ## dct , depth);\
c->draw_edges = FUNCC(draw_edges , depth);\
......@@ -2923,15 +2899,7 @@ av_cold void ff_dsputil_init(DSPContext* c, AVCodecContext *avctx)
c->put_h264_chroma_pixels_tab[2] = FUNCC(put_h264_chroma_mc2 , depth);\
c->avg_h264_chroma_pixels_tab[0] = FUNCC(avg_h264_chroma_mc8 , depth);\
c->avg_h264_chroma_pixels_tab[1] = FUNCC(avg_h264_chroma_mc4 , depth);\
c->avg_h264_chroma_pixels_tab[2] = FUNCC(avg_h264_chroma_mc2 , depth);\
\
dspfunc2(put_h264_qpel, 0, 16, depth);\
dspfunc2(put_h264_qpel, 1, 8, depth);\
dspfunc2(put_h264_qpel, 2, 4, depth);\
dspfunc2(put_h264_qpel, 3, 2, depth);\
dspfunc2(avg_h264_qpel, 0, 16, depth);\
dspfunc2(avg_h264_qpel, 1, 8, depth);\
dspfunc2(avg_h264_qpel, 2, 4, depth);
c->avg_h264_chroma_pixels_tab[2] = FUNCC(avg_h264_chroma_mc2 , depth)
switch (avctx->bits_per_raw_sample) {
case 9:
......
......@@ -331,9 +331,6 @@ typedef struct DSPContext {
h264_chroma_mc_func put_h264_chroma_pixels_tab[3];
h264_chroma_mc_func avg_h264_chroma_pixels_tab[3];
qpel_mc_func put_h264_qpel_pixels_tab[4][16];
qpel_mc_func avg_h264_qpel_pixels_tab[3][16];
me_cmp_func pix_abs[2][4];
/* huffyuv specific */
......
This diff is collapsed.
......@@ -986,10 +986,9 @@ static av_cold void common_init(H264Context *h)
s->avctx->bits_per_raw_sample = 8;
h->cur_chroma_format_idc = 1;
ff_h264dsp_init(&h->h264dsp,
s->avctx->bits_per_raw_sample, h->cur_chroma_format_idc);
ff_h264_pred_init(&h->hpc, s->codec_id,
s->avctx->bits_per_raw_sample, h->cur_chroma_format_idc);
ff_h264dsp_init(&h->h264dsp, 8, 1);
ff_h264qpel_init(&h->h264qpel, 8);
ff_h264_pred_init(&h->hpc, s->codec_id, 8, 1);
h->dequant_coeff_pps = -1;
s->unrestricted_mv = 1;
......@@ -2469,6 +2468,7 @@ static int h264_set_parameter_from_sps(H264Context *h)
ff_h264dsp_init(&h->h264dsp, h->sps.bit_depth_luma,
h->sps.chroma_format_idc);
ff_h264qpel_init(&h->h264qpel, h->sps.bit_depth_luma);
ff_h264_pred_init(&h->hpc, s->codec_id, h->sps.bit_depth_luma,
h->sps.chroma_format_idc);
s->dsp.dct_bits = h->sps.bit_depth_luma > 8 ? 32 : 16;
......@@ -2625,6 +2625,7 @@ static int h264_slice_header_init(H264Context *h, int reinit)
memcpy(c, h->s.thread_context[i], sizeof(MpegEncContext));
memset(&c->s + 1, 0, sizeof(H264Context) - sizeof(MpegEncContext));
c->h264dsp = h->h264dsp;
c->h264qpel = h->h264qpel;
c->sps = h->sps;
c->pps = h->pps;
c->pixel_shift = h->pixel_shift;
......@@ -2666,8 +2667,8 @@ static int decode_slice_header(H264Context *h, H264Context *h0)
int must_reinit;
int needs_reinit = 0;
s->me.qpel_put = s->dsp.put_h264_qpel_pixels_tab;
s->me.qpel_avg = s->dsp.avg_h264_qpel_pixels_tab;
s->me.qpel_put = h->h264qpel.put_h264_qpel_pixels_tab;
s->me.qpel_avg = h->h264qpel.avg_h264_qpel_pixels_tab;
first_mb_in_slice = get_ue_golomb_long(&s->gb);
......
......@@ -33,6 +33,7 @@
#include "mpegvideo.h"
#include "h264dsp.h"
#include "h264pred.h"
#include "h264qpel.h"
#include "rectangle.h"
#define interlaced_dct interlaced_dct_is_a_bad_name
......@@ -257,6 +258,7 @@ typedef struct MMCO {
typedef struct H264Context {
MpegEncContext s;
H264DSPContext h264dsp;
H264QpelContext h264qpel;
int pixel_shift; ///< 0 for 8-bit H264, 1 for high-bit-depth H264
int chroma_qp[2]; // QPc
......
/*
* H.26L/H.264/AVC/JVT/14496-10/... encoder/decoder
* Copyright (c) 2003-2010 Michael Niedermayer <michaelni@gmx.at>
*
* 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 "h264qpel.h"
#define pixeltmp int16_t
#define BIT_DEPTH 8
#include "h264qpel_template.c"
#undef BIT_DEPTH
#define BIT_DEPTH 9
#include "h264qpel_template.c"
#undef BIT_DEPTH
#define BIT_DEPTH 10
#include "h264qpel_template.c"
#undef BIT_DEPTH
#undef pixeltmp
#define pixeltmp int32_t
#define BIT_DEPTH 12
#include "h264qpel_template.c"
#undef BIT_DEPTH
#define BIT_DEPTH 14
#include "h264qpel_template.c"
#undef BIT_DEPTH
void ff_h264qpel_init(H264QpelContext *c, int bit_depth)
{
#undef FUNCC
#define FUNCC(f, depth) f ## _ ## depth ## _c
#define dspfunc2(PFX, IDX, NUM, depth) \
c->PFX ## _pixels_tab[IDX][ 0] = FUNCC(PFX ## NUM ## _mc00, depth); \
c->PFX ## _pixels_tab[IDX][ 1] = FUNCC(PFX ## NUM ## _mc10, depth); \
c->PFX ## _pixels_tab[IDX][ 2] = FUNCC(PFX ## NUM ## _mc20, depth); \
c->PFX ## _pixels_tab[IDX][ 3] = FUNCC(PFX ## NUM ## _mc30, depth); \
c->PFX ## _pixels_tab[IDX][ 4] = FUNCC(PFX ## NUM ## _mc01, depth); \
c->PFX ## _pixels_tab[IDX][ 5] = FUNCC(PFX ## NUM ## _mc11, depth); \
c->PFX ## _pixels_tab[IDX][ 6] = FUNCC(PFX ## NUM ## _mc21, depth); \
c->PFX ## _pixels_tab[IDX][ 7] = FUNCC(PFX ## NUM ## _mc31, depth); \
c->PFX ## _pixels_tab[IDX][ 8] = FUNCC(PFX ## NUM ## _mc02, depth); \
c->PFX ## _pixels_tab[IDX][ 9] = FUNCC(PFX ## NUM ## _mc12, depth); \
c->PFX ## _pixels_tab[IDX][10] = FUNCC(PFX ## NUM ## _mc22, depth); \
c->PFX ## _pixels_tab[IDX][11] = FUNCC(PFX ## NUM ## _mc32, depth); \
c->PFX ## _pixels_tab[IDX][12] = FUNCC(PFX ## NUM ## _mc03, depth); \
c->PFX ## _pixels_tab[IDX][13] = FUNCC(PFX ## NUM ## _mc13, depth); \
c->PFX ## _pixels_tab[IDX][14] = FUNCC(PFX ## NUM ## _mc23, depth); \
c->PFX ## _pixels_tab[IDX][15] = FUNCC(PFX ## NUM ## _mc33, depth)
#define SET_QPEL(depth) \
dspfunc2(put_h264_qpel, 0, 16, depth); \
dspfunc2(put_h264_qpel, 1, 8, depth); \
dspfunc2(put_h264_qpel, 2, 4, depth); \
dspfunc2(put_h264_qpel, 3, 2, depth); \
dspfunc2(avg_h264_qpel, 0, 16, depth); \
dspfunc2(avg_h264_qpel, 1, 8, depth); \
dspfunc2(avg_h264_qpel, 2, 4, depth)
switch (bit_depth) {
default:
SET_QPEL(8);
break;
case 9:
SET_QPEL(9);
break;
case 10:
SET_QPEL(10);
break;
case 12:
SET_QPEL(12);
break;
case 14:
SET_QPEL(14);
break;
}
if (ARCH_ARM)
ff_h264qpel_init_arm(c, bit_depth);
if (ARCH_PPC)
ff_h264qpel_init_ppc(c, bit_depth);
if (ARCH_X86)
ff_h264qpel_init_x86(c, bit_depth);
}
/*
* H.26L/H.264/AVC/JVT/14496-10/... encoder/decoder
* Copyright (c) 2003-2010 Michael Niedermayer <michaelni@gmx.at>
*
* 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
*/
#ifndef AVCODEC_H264QPEL_H
#define AVCODEC_H264QPEL_H
#include "dsputil.h"
typedef struct H264QpelContext {
qpel_mc_func put_h264_qpel_pixels_tab[4][16];
qpel_mc_func avg_h264_qpel_pixels_tab[4][16];
} H264QpelContext;
void ff_h264qpel_init(H264QpelContext *c, int bit_depth);
void ff_h264qpel_init_arm(H264QpelContext *c, int bit_depth);
void ff_h264qpel_init_ppc(H264QpelContext *c, int bit_depth);
void ff_h264qpel_init_x86(H264QpelContext *c, int bit_depth);
#endif /* AVCODEC_H264QPEL_H */
This diff is collapsed.
/*
* Copyright (c) 2000, 2001 Fabrice Bellard
* Copyright (c) 2002-2004 Michael Niedermayer <michaelni@gmx.at>
*
* 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
*/
#define DEF_HPEL(OPNAME, OP) \
static inline void FUNCC(OPNAME ## _pixels2)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
int i;\
for(i=0; i<h; i++){\
OP(*((pixel2*)(block )), AV_RN2P(pixels ));\
pixels+=line_size;\
block +=line_size;\
}\
}\
static inline void FUNCC(OPNAME ## _pixels4)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
int i;\
for(i=0; i<h; i++){\
OP(*((pixel4*)(block )), AV_RN4P(pixels ));\
pixels+=line_size;\
block +=line_size;\
}\
}\
static inline void FUNCC(OPNAME ## _pixels8)(uint8_t *block, const uint8_t *pixels, int line_size, int h){\
int i;\
for(i=0; i<h; i++){\
OP(*((pixel4*)(block )), AV_RN4P(pixels ));\
OP(*((pixel4*)(block+4*sizeof(pixel))), AV_RN4P(pixels+4*sizeof(pixel)));\
pixels+=line_size;\
block +=line_size;\
}\
}\
\
static inline void FUNC(OPNAME ## _pixels8_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
int i;\
for(i=0; i<h; i++){\
pixel4 a,b;\
a= AV_RN4P(&src1[i*src_stride1 ]);\
b= AV_RN4P(&src2[i*src_stride2 ]);\
OP(*((pixel4*)&dst[i*dst_stride ]), rnd_avg_pixel4(a, b));\
a= AV_RN4P(&src1[i*src_stride1+4*sizeof(pixel)]);\
b= AV_RN4P(&src2[i*src_stride2+4*sizeof(pixel)]);\
OP(*((pixel4*)&dst[i*dst_stride+4*sizeof(pixel)]), rnd_avg_pixel4(a, b));\
}\
}\
\
static inline void FUNC(OPNAME ## _pixels4_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
int i;\
for(i=0; i<h; i++){\
pixel4 a,b;\
a= AV_RN4P(&src1[i*src_stride1 ]);\
b= AV_RN4P(&src2[i*src_stride2 ]);\
OP(*((pixel4*)&dst[i*dst_stride ]), rnd_avg_pixel4(a, b));\
}\
}\
\
static inline void FUNC(OPNAME ## _pixels2_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
int i;\
for(i=0; i<h; i++){\
pixel4 a,b;\
a= AV_RN2P(&src1[i*src_stride1 ]);\
b= AV_RN2P(&src2[i*src_stride2 ]);\
OP(*((pixel2*)&dst[i*dst_stride ]), rnd_avg_pixel4(a, b));\
}\
}\
\
static inline void FUNC(OPNAME ## _pixels16_l2)(uint8_t *dst, const uint8_t *src1, const uint8_t *src2, int dst_stride, \
int src_stride1, int src_stride2, int h){\
FUNC(OPNAME ## _pixels8_l2)(dst , src1 , src2 , dst_stride, src_stride1, src_stride2, h);\
FUNC(OPNAME ## _pixels8_l2)(dst+8*sizeof(pixel), src1+8*sizeof(pixel), src2+8*sizeof(pixel), dst_stride, src_stride1, src_stride2, h);\
}\
\
CALL_2X_PIXELS(FUNCC(OPNAME ## _pixels16) , FUNCC(OPNAME ## _pixels8) , 8*sizeof(pixel))
#define op_avg(a, b) a = rnd_avg_pixel4(a, b)
#define op_put(a, b) a = b
DEF_HPEL(avg, op_avg)
DEF_HPEL(put, op_put)
#undef op_avg
#undef op_put
OBJS += ppc/dsputil_ppc.o \
ppc/videodsp_ppc.o \
OBJS-$(CONFIG_H264QPEL) += ppc/h264_qpel.o
OBJS-$(CONFIG_VORBIS_DECODER) += ppc/vorbisdsp_altivec.o
OBJS-$(CONFIG_VP3DSP) += ppc/vp3dsp_altivec.o
......
This diff is collapsed.
This diff is collapsed.
......@@ -325,13 +325,13 @@ static void vc1_inv_trans_8x4_altivec(uint8_t *dest, int stride, int16_t *block)
#define OP_U8_ALTIVEC PUT_OP_U8_ALTIVEC
#define PREFIX_no_rnd_vc1_chroma_mc8_altivec put_no_rnd_vc1_chroma_mc8_altivec
#include "h264_altivec_template.c"
#include "h264_qpel_template.c"
#undef OP_U8_ALTIVEC
#undef PREFIX_no_rnd_vc1_chroma_mc8_altivec
#define OP_U8_ALTIVEC AVG_OP_U8_ALTIVEC
#define PREFIX_no_rnd_vc1_chroma_mc8_altivec avg_no_rnd_vc1_chroma_mc8_altivec
#include "h264_altivec_template.c"
#include "h264_qpel_template.c"
#undef OP_U8_ALTIVEC
#undef PREFIX_no_rnd_vc1_chroma_mc8_altivec
......
......@@ -26,6 +26,7 @@
#include "avcodec.h"
#include "dsputil.h"
#include "h264qpel.h"
#include "rv34dsp.h"
#define RV30_LOWPASS(OPNAME, OP) \
......@@ -253,10 +254,12 @@ RV30_MC(avg_, 8)
RV30_MC(avg_, 16)
av_cold void ff_rv30dsp_init(RV34DSPContext *c, DSPContext* dsp) {
H264QpelContext qpel;
ff_rv34dsp_init(c, dsp);
ff_h264qpel_init(&qpel, 8);
c->put_pixels_tab[0][ 0] = dsp->put_h264_qpel_pixels_tab[0][0];
c->put_pixels_tab[0][ 0] = qpel.put_h264_qpel_pixels_tab[0][0];
c->put_pixels_tab[0][ 1] = put_rv30_tpel16_mc10_c;
c->put_pixels_tab[0][ 2] = put_rv30_tpel16_mc20_c;
c->put_pixels_tab[0][ 4] = put_rv30_tpel16_mc01_c;
......@@ -265,7 +268,7 @@ av_cold void ff_rv30dsp_init(RV34DSPContext *c, DSPContext* dsp) {
c->put_pixels_tab[0][ 8] = put_rv30_tpel16_mc02_c;
c->put_pixels_tab[0][ 9] = put_rv30_tpel16_mc12_c;
c->put_pixels_tab[0][10] = put_rv30_tpel16_mc22_c;
c->avg_pixels_tab[0][ 0] = dsp->avg_h264_qpel_pixels_tab[0][0];
c->avg_pixels_tab[0][ 0] = qpel.avg_h264_qpel_pixels_tab[0][0];
c->avg_pixels_tab[0][ 1] = avg_rv30_tpel16_mc10_c;
c->avg_pixels_tab[0][ 2] = avg_rv30_tpel16_mc20_c;
c->avg_pixels_tab[0][ 4] = avg_rv30_tpel16_mc01_c;
......@@ -274,7 +277,7 @@ av_cold void ff_rv30dsp_init(RV34DSPContext *c, DSPContext* dsp) {
c->avg_pixels_tab[0][ 8] = avg_rv30_tpel16_mc02_c;
c->avg_pixels_tab[0][ 9] = avg_rv30_tpel16_mc12_c;
c->avg_pixels_tab[0][10] = avg_rv30_tpel16_mc22_c;
c->put_pixels_tab[1][ 0] = dsp->put_h264_qpel_pixels_tab[1][0];
c->put_pixels_tab[1][ 0] = qpel.put_h264_qpel_pixels_tab[1][0];
c->put_pixels_tab[1][ 1] = put_rv30_tpel8_mc10_c;
c->put_pixels_tab[1][ 2] = put_rv30_tpel8_mc20_c;
c->put_pixels_tab[1][ 4] = put_rv30_tpel8_mc01_c;
......@@ -283,7 +286,7 @@ av_cold void ff_rv30dsp_init(RV34DSPContext *c, DSPContext* dsp) {
c->put_pixels_tab[1][ 8] = put_rv30_tpel8_mc02_c;
c->put_pixels_tab[1][ 9] = put_rv30_tpel8_mc12_c;
c->put_pixels_tab[1][10] = put_rv30_tpel8_mc22_c;
c->avg_pixels_tab[1][ 0] = dsp->avg_h264_qpel_pixels_tab[1][0];
c->avg_pixels_tab[1][ 0] = qpel.avg_h264_qpel_pixels_tab[1][0];
c->avg_pixels_tab[1][ 1] = avg_rv30_tpel8_mc10_c;
c->avg_pixels_tab[1][ 2] = avg_rv30_tpel8_mc20_c;
c->avg_pixels_tab[1][ 4] = avg_rv30_tpel8_mc01_c;
......
......@@ -26,6 +26,7 @@
#include "avcodec.h"
#include "dsputil.h"
#include "h264qpel.h"
#include "rv34dsp.h"
#include "libavutil/avassert.h"
#include "libavutil/common.h"
......@@ -519,18 +520,20 @@ static int rv40_v_loop_filter_strength(uint8_t *src, ptrdiff_t stride,
}
av_cold void ff_rv40dsp_init(RV34DSPContext *c, DSPContext* dsp) {
H264QpelContext qpel;
ff_rv34dsp_init(c, dsp);
ff_h264qpel_init(&qpel, 8);
c->put_pixels_tab[0][ 0] = dsp->put_h264_qpel_pixels_tab[0][0];
c->put_pixels_tab[0][ 0] = qpel.put_h264_qpel_pixels_tab[0][0];
c->put_pixels_tab[0][ 1] = put_rv40_qpel16_mc10_c;
c->put_pixels_tab[0][ 2] = dsp->put_h264_qpel_pixels_tab[0][2];
c->put_pixels_tab[0][ 2] = qpel.put_h264_qpel_pixels_tab[0][2];
c->put_pixels_tab[0][ 3] = put_rv40_qpel16_mc30_c;
c->put_pixels_tab[0][ 4] = put_rv40_qpel16_mc01_c;
c->put_pixels_tab[0][ 5] = put_rv40_qpel16_mc11_c;
c->put_pixels_tab[0][ 6] = put_rv40_qpel16_mc21_c;
c->put_pixels_tab[0][ 7] = put_rv40_qpel16_mc31_c;
c->put_pixels_tab[0][ 8] = dsp->put_h264_qpel_pixels_tab[0][8];
c->put_pixels_tab[0][ 8] = qpel.put_h264_qpel_pixels_tab[0][8];
c->put_pixels_tab[0][ 9] = put_rv40_qpel16_mc12_c;
c->put_pixels_tab[0][10] = put_rv40_qpel16_mc22_c;
c->put_pixels_tab[0][11] = put_rv40_qpel16_mc32_c;
......@@ -538,15 +541,15 @@ av_cold void ff_rv40dsp_init(RV34DSPContext *c, DSPContext* dsp) {
c->put_pixels_tab[0][13] = put_rv40_qpel16_mc13_c;
c->put_pixels_tab[0][14] = put_rv40_qpel16_mc23_c;
c->put_pixels_tab[0][15] = ff_put_rv40_qpel16_mc33_c;
c->avg_pixels_tab[0][ 0] = dsp->avg_h264_qpel_pixels_tab[0][0];
c->avg_pixels_tab[0][ 0] = qpel.avg_h264_qpel_pixels_tab[0][0];
c->avg_pixels_tab[0][ 1] = avg_rv40_qpel16_mc10_c;
c->avg_pixels_tab[0][ 2] = dsp->avg_h264_qpel_pixels_tab[0][2];
c->avg_pixels_tab[0][ 2] = qpel.avg_h264_qpel_pixels_tab[0][2];
c->avg_pixels_tab[0][ 3] = avg_rv40_qpel16_mc30_c;
c->avg_pixels_tab[0][ 4] = avg_rv40_qpel16_mc01_c;
c->avg_pixels_tab[0][ 5] = avg_rv40_qpel16_mc11_c;
c->avg_pixels_tab[0][ 6] = avg_rv40_qpel16_mc21_c;
c->avg_pixels_tab[0][ 7] = avg_rv40_qpel16_mc31_c;
c->avg_pixels_tab[0][ 8] = dsp->avg_h264_qpel_pixels_tab[0][8];
c->avg_pixels_tab[0][ 8] = qpel.avg_h264_qpel_pixels_tab[0][8];
c->avg_pixels_tab[0][ 9] = avg_rv40_qpel16_mc12_c;
c->avg_pixels_tab[0][10] = avg_rv40_qpel16_mc22_c;
c->avg_pixels_tab[0][11] = avg_rv40_qpel16_mc32_c;
......@@ -554,15 +557,15 @@ av_cold void ff_rv40dsp_init(RV34DSPContext *c, DSPContext* dsp) {
c->avg_pixels_tab[0][13] = avg_rv40_qpel16_mc13_c;
c->avg_pixels_tab[0][14] = avg_rv40_qpel16_mc23_c;
c->avg_pixels_tab[0][15] = ff_avg_rv40_qpel16_mc33_c;
c->put_pixels_tab[1][ 0] = dsp->put_h264_qpel_pixels_tab[1][0];
c->put_pixels_tab[1][ 0] = qpel.put_h264_qpel_pixels_tab[1][0];
c->put_pixels_tab[1][ 1] = put_rv40_qpel8_mc10_c;
c->put_pixels_tab[1][ 2] = dsp->put_h264_qpel_pixels_tab[1][2];
c->put_pixels_tab[1][ 2] = qpel.put_h264_qpel_pixels_tab[1][2];
c->put_pixels_tab[1][ 3] = put_rv40_qpel8_mc30_c;
c->put_pixels_tab[1][ 4] = put_rv40_qpel8_mc01_c;
c->put_pixels_tab[1][ 5] = put_rv40_qpel8_mc11_c;
c->put_pixels_tab[1][ 6] = put_rv40_qpel8_mc21_c;
c->put_pixels_tab[1][ 7] = put_rv40_qpel8_mc31_c;
c->put_pixels_tab[1][ 8] = dsp->put_h264_qpel_pixels_tab[1][8];
c->put_pixels_tab[1][ 8] = qpel.put_h264_qpel_pixels_tab[1][8];
c->put_pixels_tab[1][ 9] = put_rv40_qpel8_mc12_c;
c->put_pixels_tab[1][10] = put_rv40_qpel8_mc22_c;
c->put_pixels_tab[1][11] = put_rv40_qpel8_mc32_c;
......@@ -570,15 +573,15 @@ av_cold void ff_rv40dsp_init(RV34DSPContext *c, DSPContext* dsp) {
c->put_pixels_tab[1][13] = put_rv40_qpel8_mc13_c;
c->put_pixels_tab[1][14] = put_rv40_qpel8_mc23_c;
c->put_pixels_tab[1][15] = ff_put_rv40_qpel8_mc33_c;
c->avg_pixels_tab[1][ 0] = dsp->avg_h264_qpel_pixels_tab[1][0];
c->avg_pixels_tab[1][ 0] = qpel.avg_h264_qpel_pixels_tab[1][0];
c->avg_pixels_tab[1][ 1] = avg_rv40_qpel8_mc10_c;
c->avg_pixels_tab[1][ 2] = dsp->avg_h264_qpel_pixels_tab[1][2];
c->avg_pixels_tab[1][ 2] = qpel.avg_h264_qpel_pixels_tab[1][2];
c->avg_pixels_tab[1][ 3] = avg_rv40_qpel8_mc30_c;
c->avg_pixels_tab[1][ 4] = avg_rv40_qpel8_mc01_c;
c->avg_pixels_tab[1][ 5] = avg_rv40_qpel8_mc11_c;
c->avg_pixels_tab[1][ 6] = avg_rv40_qpel8_mc21_c;
c->avg_pixels_tab[1][ 7] = avg_rv40_qpel8_mc31_c;
c->avg_pixels_tab[1][ 8] = dsp->avg_h264_qpel_pixels_tab[1][8];
c->avg_pixels_tab[1][ 8] = qpel.avg_h264_qpel_pixels_tab[1][8];
c->avg_pixels_tab[1][ 9] = avg_rv40_qpel8_mc12_c;
c->avg_pixels_tab[1][10] = avg_rv40_qpel8_mc22_c;
c->avg_pixels_tab[1][11] = avg_rv40_qpel8_mc32_c;
......
......@@ -396,15 +396,6 @@ void ff_dsputil_init_align(DSPContext* c, AVCodecContext *avctx)
dspfunc(avg_qpel, 1, 8);
/* dspfunc(avg_no_rnd_qpel, 1, 8); */
if (!high_bit_depth) {
dspfunc(put_h264_qpel, 0, 16);
dspfunc(put_h264_qpel, 1, 8);
dspfunc(put_h264_qpel, 2, 4);
dspfunc(avg_h264_qpel, 0, 16);
dspfunc(avg_h264_qpel, 1, 8);
dspfunc(avg_h264_qpel, 2, 4);
}
#undef dspfunc
if (!high_bit_depth) {
c->put_h264_chroma_pixels_tab[0]= put_h264_chroma_mc8_sh4;
......
This diff is collapsed.
......@@ -361,18 +361,18 @@ void ff_snow_pred_block(SnowContext *s, uint8_t *dst, uint8_t *tmp, int stride,
else if(b_w==32){
int y;
for(y=0; y<b_h; y+=16){
s->dsp.put_h264_qpel_pixels_tab[0][dy+(dx>>2)](dst + y*stride, src + 3 + (y+3)*stride,stride);
s->dsp.put_h264_qpel_pixels_tab[0][dy+(dx>>2)](dst + 16 + y*stride, src + 19 + (y+3)*stride,stride);
s->h264qpel.put_h264_qpel_pixels_tab[0][dy+(dx>>2)](dst + y*stride, src + 3 + (y+3)*stride,stride);
s->h264qpel.put_h264_qpel_pixels_tab[0][dy+(dx>>2)](dst + 16 + y*stride, src + 19 + (y+3)*stride,stride);
}
}else if(b_w==b_h)
s->dsp.put_h264_qpel_pixels_tab[tab_index ][dy+(dx>>2)](dst,src + 3 + 3*stride,stride);
s->h264qpel.put_h264_qpel_pixels_tab[tab_index ][dy+(dx>>2)](dst,src + 3 + 3*stride,stride);
else if(b_w==2*b_h){
s->dsp.put_h264_qpel_pixels_tab[tab_index+1][dy+(dx>>2)](dst ,src + 3 + 3*stride,stride);
s->dsp.put_h264_qpel_pixels_tab[tab_index+1][dy+(dx>>2)](dst+b_h,src + 3 + b_h + 3*stride,stride);
s->h264qpel.put_h264_qpel_pixels_tab[tab_index+1][dy+(dx>>2)](dst ,src + 3 + 3*stride,stride);
s->h264qpel.put_h264_qpel_pixels_tab[tab_index+1][dy+(dx>>2)](dst+b_h,src + 3 + b_h + 3*stride,stride);
}else{
av_assert2(2*b_w==b_h);
s->dsp.put_h264_qpel_pixels_tab[tab_index ][dy+(dx>>2)](dst ,src + 3 + 3*stride ,stride);
s->dsp.put_h264_qpel_pixels_tab[tab_index ][dy+(dx>>2)](dst+b_w*stride,src + 3 + 3*stride+b_w*stride,stride);
s->h264qpel.put_h264_qpel_pixels_tab[tab_index ][dy+(dx>>2)](dst ,src + 3 + 3*stride ,stride);
s->h264qpel.put_h264_qpel_pixels_tab[tab_index ][dy+(dx>>2)](dst+b_w*stride,src + 3 + 3*stride+b_w*stride,stride);
}
}
}
......@@ -403,14 +403,15 @@ av_cold int ff_snow_common_init(AVCodecContext *avctx){
ff_dsputil_init(&s->dsp, avctx);
ff_videodsp_init(&s->vdsp, 8);
ff_dwt_init(&s->dwt);
ff_h264qpel_init(&s->h264qpel, 8);
#define mcf(dx,dy)\
s->dsp.put_qpel_pixels_tab [0][dy+dx/4]=\
s->dsp.put_no_rnd_qpel_pixels_tab[0][dy+dx/4]=\
s->dsp.put_h264_qpel_pixels_tab[0][dy+dx/4];\
s->h264qpel.put_h264_qpel_pixels_tab[0][dy+dx/4];\
s->dsp.put_qpel_pixels_tab [1][dy+dx/4]=\
s->dsp.put_no_rnd_qpel_pixels_tab[1][dy+dx/4]=\
s->dsp.put_h264_qpel_pixels_tab[1][dy+dx/4];
s->h264qpel.put_h264_qpel_pixels_tab[1][dy+dx/4];
mcf( 0, 0)
mcf( 4, 0)
......
......@@ -28,6 +28,7 @@
#include "rangecoder.h"
#include "mathops.h"
#include "mpegvideo.h"
#include "h264qpel.h"
#define MID_STATE 128
......@@ -109,6 +110,7 @@ typedef struct SnowContext{
RangeCoder c;
DSPContext dsp;
VideoDSPContext vdsp;
H264QpelContext h264qpel;
SnowDWTContext dwt;
AVFrame new_picture;
AVFrame input_picture; ///< new_picture with the internal linesizes
......
This diff is collapsed.
This diff is collapsed.
This diff is collapsed.
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment