cavsdsp.c.svn-base

ffmpeg最新源码

/*
 * Chinese AVS video (AVS1-P2, JiZhun profile) decoder.
 *
 * DSP functions
 *
 * Copyright (c) 2006  Stefan Gehrer <[e-mail]>
 *
 * 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 <stdio.h>
#include "dsputil.h"

/*****************************************************************************
 *
 * in-loop deblocking filter
 *
 ****************************************************************************/

#define P2 p0_p[-3*stride]
#define P1 p0_p[-2*stride]
#define P0 p0_p[-1*stride]
#define Q0 p0_p[ 0*stride]
#define Q1 p0_p[ 1*stride]
#define Q2 p0_p[ 2*stride]

static inline void loop_filter_l2(uint8_t *p0_p,int stride,int alpha, int beta) {
    int p0 = P0;
    int q0 = Q0;

    if(abs(p0-q0)<alpha && abs(P1-p0)<beta && abs(Q1-q0)<beta) {
        int s = p0 + q0 + 2;
        alpha = (alpha>>2) + 2;
        if(abs(P2-p0) < beta && abs(p0-q0) < alpha) {
            P0 = (P1 + p0 + s) >> 2;
            P1 = (2*P1 + s) >> 2;
        } else
            P0 = (2*P1 + s) >> 2;
        if(abs(Q2-q0) < beta && abs(q0-p0) < alpha) {
            Q0 = (Q1 + q0 + s) >> 2;
            Q1 = (2*Q1 + s) >> 2;
        } else
            Q0 = (2*Q1 + s) >> 2;
    }
}

static inline void loop_filter_l1(uint8_t *p0_p, int stride, int alpha, int beta, int tc) {
    int p0 = P0;
    int q0 = Q0;

    if(abs(p0-q0)<alpha && abs(P1-p0)<beta && abs(Q1-q0)<beta) {
        int delta = av_clip(((q0-p0)*3+P1-Q1+4)>>3,-tc, tc);
        P0 = av_clip_uint8(p0+delta);
        Q0 = av_clip_uint8(q0-delta);
        if(abs(P2-p0)<beta) {
            delta = av_clip(((P0-P1)*3+P2-Q0+4)>>3, -tc, tc);
            P1 = av_clip_uint8(P1+delta);
        }
        if(abs(Q2-q0)<beta) {
            delta = av_clip(((Q1-Q0)*3+P0-Q2+4)>>3, -tc, tc);
            Q1 = av_clip_uint8(Q1-delta);
        }
    }
}

static inline void loop_filter_c2(uint8_t *p0_p,int stride,int alpha, int beta) {
    int p0 = P0;
    int q0 = Q0;

    if(abs(p0-q0)<alpha && abs(P1-p0)<beta && abs(Q1-q0)<beta) {
        int s = p0 + q0 + 2;
        alpha = (alpha>>2) + 2;
        if(abs(P2-p0) < beta && abs(p0-q0) < alpha) {
            P0 = (P1 + p0 + s) >> 2;
        } else
            P0 = (2*P1 + s) >> 2;
        if(abs(Q2-q0) < beta && abs(q0-p0) < alpha) {
            Q0 = (Q1 + q0 + s) >> 2;
        } else
            Q0 = (2*Q1 + s) >> 2;
    }
}

static inline void loop_filter_c1(uint8_t *p0_p,int stride,int alpha, int beta,
                                  int tc) {
    if(abs(P0-Q0)<alpha && abs(P1-P0)<beta && abs(Q1-Q0)<beta) {
        int delta = av_clip(((Q0-P0)*3+P1-Q1+4)>>3, -tc, tc);
        P0 = av_clip_uint8(P0+delta);
        Q0 = av_clip_uint8(Q0-delta);
    }
}

#undef P0
#undef P1
#undef P2
#undef Q0
#undef Q1
#undef Q2

static void cavs_filter_lv_c(uint8_t *d, int stride, int alpha, int beta, int tc,
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<16;i++)
            loop_filter_l2(d + i*stride,1,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<8;i++)
                loop_filter_l1(d + i*stride,1,alpha,beta,tc);
        if (bs2)
            for(i=8;i<16;i++)
                loop_filter_l1(d + i*stride,1,alpha,beta,tc);
    }
}

static void cavs_filter_lh_c(uint8_t *d, int stride, int alpha, int beta, int tc,
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<16;i++)
            loop_filter_l2(d + i,stride,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<8;i++)
                loop_filter_l1(d + i,stride,alpha,beta,tc);
        if (bs2)
            for(i=8;i<16;i++)
                loop_filter_l1(d + i,stride,alpha,beta,tc);
    }
}

static void cavs_filter_cv_c(uint8_t *d, int stride, int alpha, int beta, int tc,
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<8;i++)
            loop_filter_c2(d + i*stride,1,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<4;i++)
                loop_filter_c1(d + i*stride,1,alpha,beta,tc);
        if (bs2)
            for(i=4;i<8;i++)
                loop_filter_c1(d + i*stride,1,alpha,beta,tc);
    }
}

static void cavs_filter_ch_c(uint8_t *d, int stride, int alpha, int beta, int tc,
                           int bs1, int bs2) {
    int i;
    if(bs1==2)
        for(i=0;i<8;i++)
            loop_filter_c2(d + i,stride,alpha,beta);
    else {
        if(bs1)
            for(i=0;i<4;i++)
                loop_filter_c1(d + i,stride,alpha,beta,tc);
        if (bs2)
            for(i=4;i<8;i++)
                loop_filter_c1(d + i,stride,alpha,beta,tc);
    }
}

/*****************************************************************************
 *
 * inverse transform
 *
 ****************************************************************************/

static void cavs_idct8_add_c(uint8_t *dst, DCTELEM *block, int stride) {
    int i;
    DCTELEM (*src)[8] = (DCTELEM(*)[8])block;
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;

    src[0][0] += 8;

    for( i = 0; i < 8; i++ ) {
        const int a0 =  3*src[i][1] - (src[i][7]<<1);
        const int a1 =  3*src[i][3] + (src[i][5]<<1);
        const int a2 =  (src[i][3]<<1) - 3*src[i][5];
        const int a3 =  (src[i][1]<<1) + 3*src[i][7];

        const int b4 = ((a0 + a1 + a3)<<1) + a1;
        const int b5 = ((a0 - a1 + a2)<<1) + a0;
        const int b6 = ((a3 - a2 - a1)<<1) + a3;
        const int b7 = ((a0 - a2 - a3)<<1) - a2;

        const int a7 = (src[i][2]<<2) - 10*src[i][6];
        const int a6 = (src[i][6]<<2) + 10*src[i][2];
        const int a5 = ((src[i][0] - src[i][4]) << 3) + 4;
        const int a4 = ((src[i][0] + src[i][4]) << 3) + 4;

        const int b0 = a4 + a6;
        const int b1 = a5 + a7;
        const int b2 = a5 - a7;
        const int b3 = a4 - a6;

        src[i][0] = (b0 + b4) >> 3;
        src[i][1] = (b1 + b5) >> 3;
        src[i][2] = (b2 + b6) >> 3;
        src[i][3] = (b3 + b7) >> 3;
        src[i][4] = (b3 - b7) >> 3;
        src[i][5] = (b2 - b6) >> 3;
        src[i][6] = (b1 - b5) >> 3;
        src[i][7] = (b0 - b4) >> 3;
    }
    for( i = 0; i < 8; i++ ) {
        const int a0 =  3*src[1][i] - (src[7][i]<<1);
        const int a1 =  3*src[3][i] + (src[5][i]<<1);
        const int a2 =  (src[3][i]<<1) - 3*src[5][i];
        const int a3 =  (src[1][i]<<1) + 3*src[7][i];

        const int b4 = ((a0 + a1 + a3)<<1) + a1;
        const int b5 = ((a0 - a1 + a2)<<1) + a0;
        const int b6 = ((a3 - a2 - a1)<<1) + a3;
        const int b7 = ((a0 - a2 - a3)<<1) - a2;

        const int a7 = (src[2][i]<<2) - 10*src[6][i];
        const int a6 = (src[6][i]<<2) + 10*src[2][i];
        const int a5 = (src[0][i] - src[4][i]) << 3;
        const int a4 = (src[0][i] + src[4][i]) << 3;

        const int b0 = a4 + a6;
        const int b1 = a5 + a7;
        const int b2 = a5 - a7;
        const int b3 = a4 - a6;

        dst[i + 0*stride] = cm[ dst[i + 0*stride] + ((b0 + b4) >> 7)];
        dst[i + 1*stride] = cm[ dst[i + 1*stride] + ((b1 + b5) >> 7)];
        dst[i + 2*stride] = cm[ dst[i + 2*stride] + ((b2 + b6) >> 7)];
        dst[i + 3*stride] = cm[ dst[i + 3*stride] + ((b3 + b7) >> 7)];
        dst[i + 4*stride] = cm[ dst[i + 4*stride] + ((b3 - b7) >> 7)];
        dst[i + 5*stride] = cm[ dst[i + 5*stride] + ((b2 - b6) >> 7)];
        dst[i + 6*stride] = cm[ dst[i + 6*stride] + ((b1 - b5) >> 7)];
        dst[i + 7*stride] = cm[ dst[i + 7*stride] + ((b0 - b4) >> 7)];
    }
    memset(block,0,64*sizeof(DCTELEM));
}

/*****************************************************************************
 *
 * motion compensation
 *
 ****************************************************************************/

#define CAVS_SUBPIX(OPNAME, OP, NAME, A, B, C, D, E, F) \
static void OPNAME ## cavs_filt8_h_ ## NAME(uint8_t *dst, uint8_t *src, int dstStride, int srcStride){\
    const int h=8;\
    uint8_t *cm = ff_cropTbl + MAX_NEG_CROP;\
    int i;\
    for(i=0; i<h; i++)\
    {\
        OP(dst[0], A*src[-2] + B*src[-1] + C*src[0] + D*src[1] + E*src[2] + F*src[3]);\
        OP(dst[1], A*src[-1] + B*src[ 0] + C*src[1] + D*src[2] + E*src[3] + F*src[4]);\
        OP(dst[2], A*src[ 0] + B*src[ 1] + C*src[2] + D*src[3] + E*src[4] + F*src[5]);\
        OP(dst[3], A*src[ 1] + B*src[ 2] + C*src[3] + D*src[4] + E*src[5] + F*src[6]);\
        OP(dst[4], A*src[ 2] + B*src[ 3] + C*src[4] + D*src[5] + E*src[6] + F*src[7]);\
        OP(dst[5], A*src[ 3] + B*src[ 4] + C*src[5] + D*src[6] + E*src[7] + F*src[8]);\
        OP(dst[6], A*src[ 4] + B*src[ 5] + C*src[6] + D*src[7] + E*src[8] + F*src[9]);\
        OP(dst[7], A*src[ 5] + B*src[ 6] + C*src[7] + D*src[8] + E*src[9] + F*src[10]);\