mpegvideoencdsp.c 7.09 KB
Newer Older
1
/*
2
 * This file is part of FFmpeg.
3
 *
4
 * FFmpeg is free software; you can redistribute it and/or
5 6 7 8
 * 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.
 *
9
 * FFmpeg is distributed in the hope that it will be useful,
10 11 12 13 14
 * 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
15
 * License along with FFmpeg; if not, write to the Free Software
16 17 18 19 20
 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
 */

#include <assert.h>
#include <stdint.h>
21
#include <string.h>
22 23

#include "config.h"
24
#include "libavutil/avassert.h"
25
#include "libavutil/attributes.h"
26
#include "libavutil/imgutils.h"
27
#include "avcodec.h"
28
#include "me_cmp.h"
29 30 31 32 33 34 35 36 37 38 39 40 41 42
#include "mpegvideoencdsp.h"

static int try_8x8basis_c(int16_t rem[64], int16_t weight[64],
                          int16_t basis[64], int scale)
{
    int i;
    unsigned int sum = 0;

    for (i = 0; i < 8 * 8; i++) {
        int b = rem[i] + ((basis[i] * scale +
                           (1 << (BASIS_SHIFT - RECON_SHIFT - 1))) >>
                          (BASIS_SHIFT - RECON_SHIFT));
        int w = weight[i];
        b >>= RECON_SHIFT;
43
        av_assert2(-512 < b && b < 512);
44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59

        sum += (w * b) * (w * b) >> 4;
    }
    return sum >> 2;
}

static void add_8x8basis_c(int16_t rem[64], int16_t basis[64], int scale)
{
    int i;

    for (i = 0; i < 8 * 8; i++)
        rem[i] += (basis[i] * scale +
                   (1 << (BASIS_SHIFT - RECON_SHIFT - 1))) >>
                  (BASIS_SHIFT - RECON_SHIFT);
}

60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
static int pix_sum_c(uint8_t *pix, int line_size)
{
    int s = 0, i, j;

    for (i = 0; i < 16; i++) {
        for (j = 0; j < 16; j += 8) {
            s   += pix[0];
            s   += pix[1];
            s   += pix[2];
            s   += pix[3];
            s   += pix[4];
            s   += pix[5];
            s   += pix[6];
            s   += pix[7];
            pix += 8;
        }
        pix += line_size - 16;
    }
    return s;
}

static int pix_norm1_c(uint8_t *pix, int line_size)
{
    int s = 0, i, j;
84
    const uint32_t *sq = ff_square_tab + 256;
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

    for (i = 0; i < 16; i++) {
        for (j = 0; j < 16; j += 8) {
#if HAVE_FAST_64BIT
            register uint64_t x = *(uint64_t *) pix;
            s += sq[x         & 0xff];
            s += sq[(x >>  8) & 0xff];
            s += sq[(x >> 16) & 0xff];
            s += sq[(x >> 24) & 0xff];
            s += sq[(x >> 32) & 0xff];
            s += sq[(x >> 40) & 0xff];
            s += sq[(x >> 48) & 0xff];
            s += sq[(x >> 56) & 0xff];
#else
            register uint32_t x = *(uint32_t *) pix;
            s += sq[x         & 0xff];
            s += sq[(x >>  8) & 0xff];
            s += sq[(x >> 16) & 0xff];
            s += sq[(x >> 24) & 0xff];
            x  = *(uint32_t *) (pix + 4);
            s += sq[x         & 0xff];
            s += sq[(x >>  8) & 0xff];
            s += sq[(x >> 16) & 0xff];
            s += sq[(x >> 24) & 0xff];
#endif
            pix += 8;
        }
        pix += line_size - 16;
    }
    return s;
}

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
/* draw the edges of width 'w' of an image of size width, height */
// FIXME: Check that this is OK for MPEG-4 interlaced.
static void draw_edges_8_c(uint8_t *buf, int wrap, int width, int height,
                           int w, int h, int sides)
{
    uint8_t *ptr = buf, *last_line;
    int i;

    /* left and right */
    for (i = 0; i < height; i++) {
        memset(ptr - w, ptr[0], w);
        memset(ptr + width, ptr[width - 1], w);
        ptr += wrap;
    }

    /* top and bottom + corners */
    buf -= w;
    last_line = buf + (height - 1) * wrap;
    if (sides & EDGE_TOP)
        for (i = 0; i < h; i++)
            // top
            memcpy(buf - (i + 1) * wrap, buf, width + w + w);
    if (sides & EDGE_BOTTOM)
        for (i = 0; i < h; i++)
            // bottom
            memcpy(last_line + (i + 1) * wrap, last_line, width + w + w);
}

145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
/* 2x2 -> 1x1 */
static void shrink22(uint8_t *dst, int dst_wrap,
                     const uint8_t *src, int src_wrap,
                     int width, int height)
{
    int w;
    const uint8_t *s1, *s2;
    uint8_t *d;

    for (; height > 0; height--) {
        s1 = src;
        s2 = s1 + src_wrap;
        d = dst;
        for (w = width; w >= 4; w -= 4) {
            d[0] = (s1[0] + s1[1] + s2[0] + s2[1] + 2) >> 2;
            d[1] = (s1[2] + s1[3] + s2[2] + s2[3] + 2) >> 2;
            d[2] = (s1[4] + s1[5] + s2[4] + s2[5] + 2) >> 2;
            d[3] = (s1[6] + s1[7] + s2[6] + s2[7] + 2) >> 2;
            s1 += 8;
            s2 += 8;
            d += 4;
        }
        for (; w > 0; w--) {
            d[0] = (s1[0] + s1[1] + s2[0] + s2[1] + 2) >> 2;
            s1 += 2;
            s2 += 2;
            d++;
        }
        src += 2 * src_wrap;
        dst += dst_wrap;
    }
}

/* 4x4 -> 1x1 */
static void shrink44(uint8_t *dst, int dst_wrap,
                     const uint8_t *src, int src_wrap,
                     int width, int height)
{
    int w;
    const uint8_t *s1, *s2, *s3, *s4;
    uint8_t *d;

    for (; height > 0; height--) {
        s1 = src;
        s2 = s1 + src_wrap;
        s3 = s2 + src_wrap;
        s4 = s3 + src_wrap;
        d = dst;
        for (w = width; w > 0; w--) {
            d[0] = (s1[0] + s1[1] + s1[2] + s1[3] +
                    s2[0] + s2[1] + s2[2] + s2[3] +
                    s3[0] + s3[1] + s3[2] + s3[3] +
                    s4[0] + s4[1] + s4[2] + s4[3] + 8) >> 4;
            s1 += 4;
            s2 += 4;
            s3 += 4;
            s4 += 4;
            d++;
        }
        src += 4 * src_wrap;
        dst += dst_wrap;
    }
}

/* 8x8 -> 1x1 */
static void shrink88(uint8_t *dst, int dst_wrap,
                     const uint8_t *src, int src_wrap,
                     int width, int height)
{
    int w, i;

    for (; height > 0; height--) {
        for(w = width;w > 0; w--) {
            int tmp = 0;
            for (i = 0; i < 8; i++) {
                tmp += src[0] + src[1] + src[2] + src[3] +
                       src[4] + src[5] + src[6] + src[7];
                src += src_wrap;
            }
            *(dst++) = (tmp + 32) >> 6;
            src += 8 - 8 * src_wrap;
        }
        src += 8 * src_wrap - 8 * width;
        dst += dst_wrap - width;
    }
}

232 233 234 235 236 237
av_cold void ff_mpegvideoencdsp_init(MpegvideoEncDSPContext *c,
                                     AVCodecContext *avctx)
{
    c->try_8x8basis = try_8x8basis_c;
    c->add_8x8basis = add_8x8basis_c;

238
    c->shrink[0] = av_image_copy_plane;
239 240 241
    c->shrink[1] = shrink22;
    c->shrink[2] = shrink44;
    c->shrink[3] = shrink88;
242 243 244 245

    c->pix_sum   = pix_sum_c;
    c->pix_norm1 = pix_norm1_c;

246 247
    c->draw_edges = draw_edges_8_c;

248 249 250 251
    if (ARCH_ARM)
        ff_mpegvideoencdsp_init_arm(c, avctx);
    if (ARCH_PPC)
        ff_mpegvideoencdsp_init_ppc(c, avctx);
252 253
    if (ARCH_X86)
        ff_mpegvideoencdsp_init_x86(c, avctx);
254 255
    if (ARCH_MIPS)
        ff_mpegvideoencdsp_init_mips(c, avctx);
256
}