ffmpeg/libswscale/rgb2rgb_template.c

/*
 * software RGB to RGB converter
 * pluralize by software PAL8 to RGB converter
 *              software YUV to YUV converter
 *              software YUV to RGB converter
 * Written by Nick Kurshev.
 * palette & YUV & runtime CPU stuff by Michael (michaelni@gmx.at)
 * lot of big-endian byte order fixes by Alex Beregszaszi
 *
 * 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 <stddef.h>

#include "libavutil/attributes.h"

static inline void rgb24tobgr32_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint8_t *dest      = dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
#if HAVE_BIGENDIAN
        /* RGB24 (= R, G, B) -> RGB32 (= A, B, G, R) */
        *dest++  = 255;
        *dest++  = s[2];
        *dest++  = s[1];
        *dest++  = s[0];
        s       += 3;
#else
        *dest++  = *s++;
        *dest++  = *s++;
        *dest++  = *s++;
        *dest++  = 255;
#endif
    }
}

static inline void rgb32tobgr24_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint8_t *dest      = dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
#if HAVE_BIGENDIAN
        /* RGB32 (= A, B, G, R) -> RGB24 (= R, G, B) */
        s++;
        dest[2]  = *s++;
        dest[1]  = *s++;
        dest[0]  = *s++;
        dest    += 3;
#else
        *dest++  = *s++;
        *dest++  = *s++;
        *dest++  = *s++;
        s++;
#endif
    }
}

/*
 * original by Strepto/Astral
 * ported to gcc & bugfixed: A'rpi
 * MMXEXT, 3DNOW optimization by Nick Kurshev
 * 32-bit C version, and and&add trick by Michael Niedermayer
 */
static inline void rgb15to16_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    register uint8_t *d         = dst;
    register const uint8_t *s   = src;
    register const uint8_t *end = s + src_size;
    const uint8_t *mm_end       = end - 3;

    while (s < mm_end) {
        register unsigned x = *((const uint32_t *)s);
        *((uint32_t *)d)    = (x & 0x7FFF7FFF) + (x & 0x7FE07FE0);
        d += 4;
        s += 4;
    }
    if (s < end) {
        register unsigned short x = *((const uint16_t *)s);
        *((uint16_t *)d)          = (x & 0x7FFF) + (x & 0x7FE0);
    }
}

static inline void rgb16to15_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    register uint8_t *d         = dst;
    register const uint8_t *s   = src;
    register const uint8_t *end = s + src_size;
    const uint8_t *mm_end       = end - 3;

    while (s < mm_end) {
        register uint32_t x  = *((const uint32_t *)s);
        *((uint32_t *)d)     = ((x >> 1) & 0x7FE07FE0) | (x & 0x001F001F);
        s                   += 4;
        d                   += 4;
    }
    if (s < end) {
        register uint16_t x = *((const uint16_t *)s);
        *((uint16_t *)d)    = ((x >> 1) & 0x7FE0) | (x & 0x001F);
    }
}

static inline void rgb32to16_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        register int rgb  = *(const uint32_t *)s;
        s                += 4;
        *d++              = ((rgb & 0xFF)     >> 3) +
                            ((rgb & 0xFC00)   >> 5) +
                            ((rgb & 0xF80000) >> 8);
    }
}

static inline void rgb32tobgr16_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        register int rgb  = *(const uint32_t *)s;
        s                += 4;
        *d++              = ((rgb & 0xF8)     << 8) +
                            ((rgb & 0xFC00)   >> 5) +
                            ((rgb & 0xF80000) >> 19);
    }
}

static inline void rgb32to15_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        register int rgb  = *(const uint32_t *)s;
        s                += 4;
        *d++              = ((rgb & 0xFF)     >> 3) +
                            ((rgb & 0xF800)   >> 6) +
                            ((rgb & 0xF80000) >> 9);
    }
}

static inline void rgb32tobgr15_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        register int rgb  = *(const uint32_t *)s;
        s                += 4;
        *d++              = ((rgb & 0xF8)     <<  7) +
                            ((rgb & 0xF800)   >>  6) +
                            ((rgb & 0xF80000) >> 19);
    }
}

static inline void rgb24tobgr16_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        const int b = *s++;
        const int g = *s++;
        const int r = *s++;
        *d++        = (b >> 3) | ((g & 0xFC) << 3) | ((r & 0xF8) << 8);
    }
}

static inline void rgb24to16_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        const int r = *s++;
        const int g = *s++;
        const int b = *s++;
        *d++        = (b >> 3) | ((g & 0xFC) << 3) | ((r & 0xF8) << 8);
    }
}

static inline void rgb24tobgr15_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        const int b = *s++;
        const int g = *s++;
        const int r = *s++;
        *d++        = (b >> 3) | ((g & 0xF8) << 2) | ((r & 0xF8) << 7);
    }
}

static inline void rgb24to15_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    uint16_t *d        = (uint16_t *)dst;
    const uint8_t *s   = src;
    const uint8_t *end = s + src_size;

    while (s < end) {
        const int r = *s++;
        const int g = *s++;
        const int b = *s++;
        *d++        = (b >> 3) | ((g & 0xF8) << 2) | ((r & 0xF8) << 7);
    }
}

static inline void rgb15tobgr24_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint8_t *d          = dst;
    const uint16_t *s   = (const uint16_t *)src;
    const uint16_t *end = s + src_size / 2;

    while (s < end) {
        register uint16_t bgr = *s++;
        *d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
        *d++ = ((bgr&0x03E0)>>2) | ((bgr&0x03E0)>> 7);
        *d++ = ((bgr&0x7C00)>>7) | ((bgr&0x7C00)>>12);
    }
}

static inline void rgb16tobgr24_c(const uint8_t *src, uint8_t *dst,
                                  int src_size)
{
    uint8_t *d          = (uint8_t *)dst;
    const uint16_t *s   = (const uint16_t *)src;
    const uint16_t *end = s + src_size / 2;

    while (s < end) {
        register uint16_t bgr = *s++;
        *d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
        *d++ = ((bgr&0x07E0)>>3) | ((bgr&0x07E0)>> 9);
        *d++ = ((bgr&0xF800)>>8) | ((bgr&0xF800)>>13);
    }
}

static inline void rgb15to32_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    uint8_t *d          = dst;
    const uint16_t *s   = (const uint16_t *)src;
    const uint16_t *end = s + src_size / 2;

    while (s < end) {
        register uint16_t bgr = *s++;
#if HAVE_BIGENDIAN
        *d++ = 255;
        *d++ = ((bgr&0x7C00)>>7) | ((bgr&0x7C00)>>12);
        *d++ = ((bgr&0x03E0)>>2) | ((bgr&0x03E0)>> 7);
        *d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
#else
        *d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
        *d++ = ((bgr&0x03E0)>>2) | ((bgr&0x03E0)>> 7);
        *d++ = ((bgr&0x7C00)>>7) | ((bgr&0x7C00)>>12);
        *d++ = 255;
#endif
    }
}

static inline void rgb16to32_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    uint8_t *d          = dst;
    const uint16_t *s   = (const uint16_t *)src;
    const uint16_t *end = s + src_size / 2;

    while (s < end) {
        register uint16_t bgr = *s++;
#if HAVE_BIGENDIAN
        *d++ = 255;
        *d++ = ((bgr&0xF800)>>8) | ((bgr&0xF800)>>13);
        *d++ = ((bgr&0x07E0)>>3) | ((bgr&0x07E0)>> 9);
        *d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
#else
        *d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
        *d++ = ((bgr&0x07E0)>>3) | ((bgr&0x07E0)>> 9);
        *d++ = ((bgr&0xF800)>>8) | ((bgr&0xF800)>>13);
        *d++ = 255;
#endif
    }
}

static inline void shuffle_bytes_2103_c(const uint8_t *src, uint8_t *dst,
                                        int src_size)
{
    int idx          = 15  - src_size;
    const uint8_t *s = src - idx;
    uint8_t *d       = dst - idx;

    for (; idx < 15; idx += 4) {
        register unsigned v   = *(const uint32_t *)&s[idx], g = v & 0xff00ff00;
        v                    &= 0xff00ff;
        *(uint32_t *)&d[idx]  = (v >> 16) + g + (v << 16);
    }
}

static inline void shuffle_bytes_0321_c(const uint8_t *src, uint8_t *dst,
                                        int src_size)
{
    int idx          = 15  - src_size;
    const uint8_t *s = src - idx;
    uint8_t *d       = dst - idx;

    for (; idx < 15; idx += 4) {
        register unsigned v   = *(const uint32_t *)&s[idx], g = v & 0x00ff00ff;
        v                    &= 0xff00ff00;
        *(uint32_t *)&d[idx]  = (v >> 16) + g + (v << 16);
    }
}

#define DEFINE_SHUFFLE_BYTES(name, a, b, c, d)                          \
static void shuffle_bytes_##name (const uint8_t *src,                   \
                                        uint8_t *dst, int src_size)     \
{                                                                       \
    int i;                                                              \
                                                                        \
    for (i = 0; i < src_size; i += 4) {                                 \
        dst[i + 0] = src[i + a];                                        \
        dst[i + 1] = src[i + b];                                        \
        dst[i + 2] = src[i + c];                                        \
        dst[i + 3] = src[i + d];                                        \
    }                                                                   \
}

DEFINE_SHUFFLE_BYTES(1230_c, 1, 2, 3, 0)
DEFINE_SHUFFLE_BYTES(3012_c, 3, 0, 1, 2)
DEFINE_SHUFFLE_BYTES(3210_c, 3, 2, 1, 0)
DEFINE_SHUFFLE_BYTES(3102_c, 3, 1, 0, 2)
DEFINE_SHUFFLE_BYTES(2013_c, 2, 0, 1, 3)
DEFINE_SHUFFLE_BYTES(2130_c, 2, 1, 3, 0)
DEFINE_SHUFFLE_BYTES(1203_c, 1, 2, 0, 3)

static inline void rgb24tobgr24_c(const uint8_t *src, uint8_t *dst, int src_size)
{
    unsigned i;

    for (i = 0; i < src_size; i += 3) {
        register uint8_t x = src[i + 2];
        dst[i + 1]         = src[i + 1];
        dst[i + 2]         = src[i + 0];
        dst[i + 0]         = x;
    }
}

static inline void yuvPlanartoyuy2_c(const uint8_t *ysrc, const uint8_t *usrc,
                                     const uint8_t *vsrc, uint8_t *dst,
                                     int width, int height,
                                     int lumStride, int chromStride,
                                     int dstStride, int vertLumPerChroma)
{
    int y, i;
    const int chromWidth = width >> 1;

    for (y = 0; y < height; y++) {
#if HAVE_FAST_64BIT
        uint64_t *ldst = (uint64_t *)dst;
        const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc;
        for (i = 0; i < chromWidth; i += 2) {
            uint64_t k = yc[0] + (uc[0] << 8) +
                         (yc[1] << 16) + ((unsigned) vc[0] << 24);
            uint64_t l = yc[2] + (uc[1] << 8) +
                         (yc[3] << 16) + ((unsigned) vc[1] << 24);
            *ldst++ = k + (l << 32);
            yc     += 4;
            uc     += 2;
            vc     += 2;
        }

#else
        int *idst = (int32_t *)dst;
        const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc;

        for (i = 0; i < chromWidth; i++) {
#if HAVE_BIGENDIAN
            *idst++ = (yc[0] << 24) + (uc[0] << 16) +
                      (yc[1] <<  8) + (vc[0] <<  0);
#else
            *idst++ = yc[0] + (uc[0] << 8) +
                      (yc[1] << 16) + (vc[0] << 24);
#endif
            yc += 2;
            uc++;
            vc++;
        }
#endif
        if ((y & (vertLumPerChroma - 1)) == vertLumPerChroma - 1) {
            usrc += chromStride;
            vsrc += chromStride;
        }
        ysrc += lumStride;
        dst  += dstStride;
    }
}

/**
 * Height should be a multiple of 2 and width should be a multiple of 16.
 * (If this is a problem for anyone then tell me, and I will fix it.)
 */
static inline void yv12toyuy2_c(const uint8_t *ysrc, const uint8_t *usrc,
                                const uint8_t *vsrc, uint8_t *dst,
                                int width, int height, int lumStride,
                                int chromStride, int dstStride)
{
    //FIXME interpolate chroma
    yuvPlanartoyuy2_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
                      chromStride, dstStride, 2);
}

static inline void yuvPlanartouyvy_c(const uint8_t *ysrc, const uint8_t *usrc,
                                     const uint8_t *vsrc, uint8_t *dst,
                                     int width, int height,
                                     int lumStride, int chromStride,
                                     int dstStride, int vertLumPerChroma)
{
    int y, i;
    const int chromWidth = width >> 1;

    for (y = 0; y < height; y++) {
#if HAVE_FAST_64BIT
        uint64_t *ldst = (uint64_t *)dst;
        const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc;
        for (i = 0; i < chromWidth; i += 2) {
            uint64_t k = uc[0] + (yc[0] << 8) +
                         (vc[0] << 16) + ((unsigned) yc[1] << 24);
            uint64_t l = uc[1] + (yc[2] << 8) +
                         (vc[1] << 16) + ((unsigned) yc[3] << 24);
            *ldst++ = k + (l << 32);
            yc     += 4;
            uc     += 2;
            vc     += 2;
        }

#else
        int *idst = (int32_t *)dst;
        const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc;

        for (i = 0; i < chromWidth; i++) {
#if HAVE_BIGENDIAN
            *idst++ = (uc[0] << 24) + (yc[0] << 16) +
                      (vc[0] <<  8) + (yc[1] <<  0);
#else
            *idst++ = uc[0] + (yc[0] << 8) +
                      (vc[0] << 16) + (yc[1] << 24);
#endif
            yc += 2;
            uc++;
            vc++;
        }
#endif
        if ((y & (vertLumPerChroma - 1)) == vertLumPerChroma - 1) {
            usrc += chromStride;
            vsrc += chromStride;
        }
        ysrc += lumStride;
        dst  += dstStride;
    }
}

/**
 * Height should be a multiple of 2 and width should be a multiple of 16
 * (If this is a problem for anyone then tell me, and I will fix it.)
 */
static inline void yv12touyvy_c(const uint8_t *ysrc, const uint8_t *usrc,
                                const uint8_t *vsrc, uint8_t *dst,
                                int width, int height, int lumStride,
                                int chromStride, int dstStride)
{
    //FIXME interpolate chroma
    yuvPlanartouyvy_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
                      chromStride, dstStride, 2);
}

/**
 * Width should be a multiple of 16.
 */
static inline void yuv422ptouyvy_c(const uint8_t *ysrc, const uint8_t *usrc,
                                   const uint8_t *vsrc, uint8_t *dst,
                                   int width, int height, int lumStride,
                                   int chromStride, int dstStride)
{
    yuvPlanartouyvy_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
                      chromStride, dstStride, 1);
}

/**
 * Width should be a multiple of 16.
 */
static inline void yuv422ptoyuy2_c(const uint8_t *ysrc, const uint8_t *usrc,
                                   const uint8_t *vsrc, uint8_t *dst,
                                   int width, int height, int lumStride,
                                   int chromStride, int dstStride)
{
    yuvPlanartoyuy2_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
                      chromStride, dstStride, 1);
}

/**
 * Height should be a multiple of 2 and width should be a multiple of 16.
 * (If this is a problem for anyone then tell me, and I will fix it.)
 */
static inline void yuy2toyv12_c(const uint8_t *src, uint8_t *ydst,
                                uint8_t *udst, uint8_t *vdst,
                                int width, int height, int lumStride,
                                int chromStride, int srcStride)
{
    int y;
    const int chromWidth = width >> 1;

    for (y = 0; y < height; y += 2) {
        int i;
        for (i = 0; i < chromWidth; i++) {
            ydst[2 * i + 0] = src[4 * i + 0];
            udst[i]         = src[4 * i + 1];
            ydst[2 * i + 1] = src[4 * i + 2];
            vdst[i]         = src[4 * i + 3];
        }
        ydst += lumStride;
        src  += srcStride;

        for (i = 0; i < chromWidth; i++) {
            ydst[2 * i + 0] = src[4 * i + 0];
            ydst[2 * i + 1] = src[4 * i + 2];
        }
        udst += chromStride;
        vdst += chromStride;
        ydst += lumStride;
        src  += srcStride;
    }
}

static inline void planar2x_c(const uint8_t *src, uint8_t *dst, int srcWidth,
                              int srcHeight, int srcStride, int dstStride)
{
    int x, y;

    dst[0] = src[0];

    // first line
    for (x = 0; x < srcWidth - 1; x++) {
        dst[2 * x + 1] = (3 * src[x] + src[x + 1]) >> 2;
        dst[2 * x + 2] = (src[x] + 3 * src[x + 1]) >> 2;
    }
    dst[2 * srcWidth - 1] = src[srcWidth - 1];

    dst += dstStride;

    for (y = 1; y < srcHeight; y++) {
        const int mmxSize = 1;

        dst[0]         = (src[0] * 3 + src[srcStride]) >> 2;
        dst[dstStride] = (src[0] + 3 * src[srcStride]) >> 2;

        for (x = mmxSize - 1; x < srcWidth - 1; x++) {
            dst[2 * x + 1]             = (src[x + 0] * 3 + src[x + srcStride + 1]) >> 2;
            dst[2 * x + dstStride + 2] = (src[x + 0] + 3 * src[x + srcStride + 1]) >> 2;
            dst[2 * x + dstStride + 1] = (src[x + 1] + 3 * src[x + srcStride])     >> 2;
            dst[2 * x + 2]             = (src[x + 1] * 3 + src[x + srcStride])     >> 2;
        }
        dst[srcWidth * 2 - 1]             = (src[srcWidth - 1] * 3 + src[srcWidth - 1 + srcStride]) >> 2;
        dst[srcWidth * 2 - 1 + dstStride] = (src[srcWidth - 1] + 3 * src[srcWidth - 1 + srcStride]) >> 2;

        dst += dstStride * 2;
        src += srcStride;
    }

    // last line
    dst[0] = src[0];

    for (x = 0; x < srcWidth - 1; x++) {
        dst[2 * x + 1] = (src[x] * 3 + src[x + 1]) >> 2;
        dst[2 * x + 2] = (src[x] + 3 * src[x + 1]) >> 2;
    }
    dst[2 * srcWidth - 1] = src[srcWidth - 1];
}

/**
 * Height should be a multiple of 2 and width should be a multiple of 16.
 * (If this is a problem for anyone then tell me, and I will fix it.)
 * Chrominance data is only taken from every second line, others are ignored.
 * FIXME: Write HQ version.
 */
static inline void uyvytoyv12_c(const uint8_t *src, uint8_t *ydst,
                                uint8_t *udst, uint8_t *vdst,
                                int width, int height, int lumStride,
                                int chromStride, int srcStride)
{
    int y;
    const int chromWidth = width >> 1;

    for (y = 0; y < height; y += 2) {
        int i;
        for (i = 0; i < chromWidth; i++) {
            udst[i]         = src[4 * i + 0];
            ydst[2 * i + 0] = src[4 * i + 1];
            vdst[i]         = src[4 * i + 2];
            ydst[2 * i + 1] = src[4 * i + 3];
        }
        ydst += lumStride;
        src  += srcStride;

        for (i = 0; i < chromWidth; i++) {
            ydst[2 * i + 0] = src[4 * i + 1];
            ydst[2 * i + 1] = src[4 * i + 3];
        }
        udst += chromStride;
        vdst += chromStride;
        ydst += lumStride;
        src  += srcStride;
    }
}

/**
 * width should be a multiple of 2.
 * (If this is a problem for anyone then tell me, and I will fix it.)
 */
void ff_rgb24toyv12_c(const uint8_t *src, uint8_t *ydst, uint8_t *udst,
                   uint8_t *vdst, int width, int height, int lumStride,
                   int chromStride, int srcStride, const int32_t *rgb2yuv)
{
    int32_t ry = rgb2yuv[RY_IDX], gy = rgb2yuv[GY_IDX], by = rgb2yuv[BY_IDX];
    int32_t ru = rgb2yuv[RU_IDX], gu = rgb2yuv[GU_IDX], bu = rgb2yuv[BU_IDX];
    int32_t rv = rgb2yuv[RV_IDX], gv = rgb2yuv[GV_IDX], bv = rgb2yuv[BV_IDX];
    int y;
    const int chromWidth = width >> 1;
    const uint8_t *src1 = src;
    const uint8_t *src2 = src1 + srcStride;
    uint8_t *ydst1 = ydst;
    uint8_t *ydst2 = ydst + lumStride;

    for (y = 0; y < height; y += 2) {
        int i;
        if (y + 1 == height) {
            ydst2 = ydst1;
            src2  = src1;
        }

        for (i = 0; i < chromWidth; i++) {
            unsigned int b11 = src1[6 * i + 0];
            unsigned int g11 = src1[6 * i + 1];
            unsigned int r11 = src1[6 * i + 2];
            unsigned int b12 = src1[6 * i + 3];
            unsigned int g12 = src1[6 * i + 4];
            unsigned int r12 = src1[6 * i + 5];
            unsigned int b21 = src2[6 * i + 0];
            unsigned int g21 = src2[6 * i + 1];
            unsigned int r21 = src2[6 * i + 2];
            unsigned int b22 = src2[6 * i + 3];
            unsigned int g22 = src2[6 * i + 4];
            unsigned int r22 = src2[6 * i + 5];

            unsigned int Y11 = ((ry * r11 + gy * g11 + by * b11) >> RGB2YUV_SHIFT) + 16;
            unsigned int Y12 = ((ry * r12 + gy * g12 + by * b12) >> RGB2YUV_SHIFT) + 16;
            unsigned int Y21 = ((ry * r21 + gy * g21 + by * b21) >> RGB2YUV_SHIFT) + 16;
            unsigned int Y22 = ((ry * r22 + gy * g22 + by * b22) >> RGB2YUV_SHIFT) + 16;

            unsigned int bx = (b11 + b12 + b21 + b22) >> 2;
            unsigned int gx = (g11 + g12 + g21 + g22) >> 2;
            unsigned int rx = (r11 + r12 + r21 + r22) >> 2;

            unsigned int U  = ((ru * rx + gu * gx + bu * bx) >> RGB2YUV_SHIFT) + 128;
            unsigned int V  = ((rv * rx + gv * gx + bv * bx) >> RGB2YUV_SHIFT) + 128;

            ydst1[2 * i + 0] = Y11;
            ydst1[2 * i + 1] = Y12;
            ydst2[2 * i + 0] = Y21;
            ydst2[2 * i + 1] = Y22;
            udst[i]          = U;
            vdst[i]          = V;
        }
        src1  += srcStride * 2;
        src2  += srcStride * 2;
        ydst1 += lumStride * 2;
        ydst2 += lumStride * 2;
        udst  += chromStride;
        vdst  += chromStride;
    }
}

static void interleaveBytes_c(const uint8_t *src1, const uint8_t *src2,
                              uint8_t *dest, int width, int height,
                              int src1Stride, int src2Stride, int dstStride)
{
    int h;

    for (h = 0; h < height; h++) {
        int w;
        for (w = 0; w < width; w++) {
            dest[2 * w + 0] = src1[w];
            dest[2 * w + 1] = src2[w];
        }
        dest += dstStride;
        src1 += src1Stride;
        src2 += src2Stride;
    }
}

static void deinterleaveBytes_c(const uint8_t *src, uint8_t *dst1, uint8_t *dst2,
                                int width, int height, int srcStride,
                                int dst1Stride, int dst2Stride)
{
    int h;

    for (h = 0; h < height; h++) {
        int w;
        for (w = 0; w < width; w++) {
            dst1[w] = src[2 * w + 0];
            dst2[w] = src[2 * w + 1];
        }
        src  += srcStride;
        dst1 += dst1Stride;
        dst2 += dst2Stride;
    }
}

static inline void vu9_to_vu12_c(const uint8_t *src1, const uint8_t *src2,
                                 uint8_t *dst1, uint8_t *dst2,
                                 int width, int height,
                                 int srcStride1, int srcStride2,
                                 int dstStride1, int dstStride2)
{
    int x, y;
    int w = width  / 2;
    int h = height / 2;

    for (y = 0; y < h; y++) {
        const uint8_t *s1 = src1 + srcStride1 * (y >> 1);
        uint8_t *d        = dst1 + dstStride1 *  y;
        for (x = 0; x < w; x++)
            d[2 * x] = d[2 * x + 1] = s1[x];
    }
    for (y = 0; y < h; y++) {
        const uint8_t *s2 = src2 + srcStride2 * (y >> 1);
        uint8_t *d        = dst2 + dstStride2 *  y;
        for (x = 0; x < w; x++)
            d[2 * x] = d[2 * x + 1] = s2[x];
    }
}

static inline void yvu9_to_yuy2_c(const uint8_t *src1, const uint8_t *src2,
                                  const uint8_t *src3, uint8_t *dst,
                                  int width, int height,
                                  int srcStride1, int srcStride2,
                                  int srcStride3, int dstStride)
{
    int x, y;
    int w = width / 2;
    int h = height;

    for (y = 0; y < h; y++) {
        const uint8_t *yp = src1 + srcStride1 *  y;
        const uint8_t *up = src2 + srcStride2 * (y >> 2);
        const uint8_t *vp = src3 + srcStride3 * (y >> 2);
        uint8_t *d        = dst  + dstStride  *  y;
        for (x = 0; x < w; x++) {
            const int x2 = x << 2;
            d[8 * x + 0] = yp[x2];
            d[8 * x + 1] = up[x];
            d[8 * x + 2] = yp[x2 + 1];
            d[8 * x + 3] = vp[x];
            d[8 * x + 4] = yp[x2 + 2];
            d[8 * x + 5] = up[x];
            d[8 * x + 6] = yp[x2 + 3];
            d[8 * x + 7] = vp[x];
        }
    }
}

static void extract_even_c(const uint8_t *src, uint8_t *dst, int count)
{
    dst   +=  count;
    src   +=  count * 2;
    count  = -count;
    while (count < 0) {
        dst[count] = src[2 * count];
        count++;
    }
}

static void extract_even2_c(const uint8_t *src, uint8_t *dst0, uint8_t *dst1,
                            int count)
{
    dst0  +=  count;
    dst1  +=  count;
    src   +=  count * 4;
    count  = -count;
    while (count < 0) {
        dst0[count] = src[4 * count + 0];
        dst1[count] = src[4 * count + 2];
        count++;
    }
}

static void extract_even2avg_c(const uint8_t *src0, const uint8_t *src1,
                               uint8_t *dst0, uint8_t *dst1, int count)
{
    dst0  +=  count;
    dst1  +=  count;
    src0  +=  count * 4;
    src1  +=  count * 4;
    count  = -count;
    while (count < 0) {
        dst0[count] = (src0[4 * count + 0] + src1[4 * count + 0]) >> 1;
        dst1[count] = (src0[4 * count + 2] + src1[4 * count + 2]) >> 1;
        count++;
    }
}

static void extract_odd2_c(const uint8_t *src, uint8_t *dst0, uint8_t *dst1,
                           int count)
{
    dst0  +=  count;
    dst1  +=  count;
    src   +=  count * 4;
    count  = -count;
    src++;
    while (count < 0) {
        dst0[count] = src[4 * count + 0];
        dst1[count] = src[4 * count + 2];
        count++;
    }
}

static void extract_odd2avg_c(const uint8_t *src0, const uint8_t *src1,
                              uint8_t *dst0, uint8_t *dst1, int count)
{
    dst0  +=  count;
    dst1  +=  count;
    src0  +=  count * 4;
    src1  +=  count * 4;
    count  = -count;
    src0++;
    src1++;
    while (count < 0) {
        dst0[count] = (src0[4 * count + 0] + src1[4 * count + 0]) >> 1;
        dst1[count] = (src0[4 * count + 2] + src1[4 * count + 2]) >> 1;
        count++;
    }
}

static void yuyvtoyuv420_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst,
                           const uint8_t *src, int width, int height,
                           int lumStride, int chromStride, int srcStride)
{
    int y;
    const int chromWidth = AV_CEIL_RSHIFT(width, 1);

    for (y = 0; y < height; y++) {
        extract_even_c(src, ydst, width);
        if (y & 1) {
            extract_odd2avg_c(src - srcStride, src, udst, vdst, chromWidth);
            udst += chromStride;
            vdst += chromStride;
        }

        src  += srcStride;
        ydst += lumStride;
    }
}

static void yuyvtoyuv422_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst,
                           const uint8_t *src, int width, int height,
                           int lumStride, int chromStride, int srcStride)
{
    int y;
    const int chromWidth = AV_CEIL_RSHIFT(width, 1);

    for (y = 0; y < height; y++) {
        extract_even_c(src, ydst, width);
        extract_odd2_c(src, udst, vdst, chromWidth);

        src  += srcStride;
        ydst += lumStride;
        udst += chromStride;
        vdst += chromStride;
    }
}

static void uyvytoyuv420_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst,
                           const uint8_t *src, int width, int height,
                           int lumStride, int chromStride, int srcStride)
{
    int y;
    const int chromWidth = AV_CEIL_RSHIFT(width, 1);

    for (y = 0; y < height; y++) {
        extract_even_c(src + 1, ydst, width);
        if (y & 1) {
            extract_even2avg_c(src - srcStride, src, udst, vdst, chromWidth);
            udst += chromStride;
            vdst += chromStride;
        }

        src  += srcStride;
        ydst += lumStride;
    }
}

static void uyvytoyuv422_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst,
                           const uint8_t *src, int width, int height,
                           int lumStride, int chromStride, int srcStride)
{
    int y;
    const int chromWidth = AV_CEIL_RSHIFT(width, 1);

    for (y = 0; y < height; y++) {
        extract_even_c(src + 1, ydst, width);
        extract_even2_c(src, udst, vdst, chromWidth);

        src  += srcStride;
        ydst += lumStride;
        udst += chromStride;
        vdst += chromStride;
    }
}

static av_cold void rgb2rgb_init_c(void)
{
    rgb15to16          = rgb15to16_c;
    rgb15tobgr24       = rgb15tobgr24_c;
    rgb15to32          = rgb15to32_c;
    rgb16tobgr24       = rgb16tobgr24_c;
    rgb16to32          = rgb16to32_c;
    rgb16to15          = rgb16to15_c;
    rgb24tobgr16       = rgb24tobgr16_c;
    rgb24tobgr15       = rgb24tobgr15_c;
    rgb24tobgr32       = rgb24tobgr32_c;
    rgb32to16          = rgb32to16_c;
    rgb32to15          = rgb32to15_c;
    rgb32tobgr24       = rgb32tobgr24_c;
    rgb24to15          = rgb24to15_c;
    rgb24to16          = rgb24to16_c;
    rgb24tobgr24       = rgb24tobgr24_c;
#if HAVE_BIGENDIAN
    shuffle_bytes_0321 = shuffle_bytes_2103_c;
    shuffle_bytes_2103 = shuffle_bytes_0321_c;
#else
    shuffle_bytes_0321 = shuffle_bytes_0321_c;
    shuffle_bytes_2103 = shuffle_bytes_2103_c;
#endif
    shuffle_bytes_1230 = shuffle_bytes_1230_c;
    shuffle_bytes_3012 = shuffle_bytes_3012_c;
    shuffle_bytes_3210 = shuffle_bytes_3210_c;
    shuffle_bytes_3102 = shuffle_bytes_3102_c;
    shuffle_bytes_2013 = shuffle_bytes_2013_c;
    shuffle_bytes_2130 = shuffle_bytes_2130_c;
    shuffle_bytes_1203 = shuffle_bytes_1203_c;
    rgb32tobgr16       = rgb32tobgr16_c;
    rgb32tobgr15       = rgb32tobgr15_c;
    yv12toyuy2         = yv12toyuy2_c;
    yv12touyvy         = yv12touyvy_c;
    yuv422ptoyuy2      = yuv422ptoyuy2_c;
    yuv422ptouyvy      = yuv422ptouyvy_c;
    yuy2toyv12         = yuy2toyv12_c;
    planar2x           = planar2x_c;
    ff_rgb24toyv12     = ff_rgb24toyv12_c;
    interleaveBytes    = interleaveBytes_c;
    deinterleaveBytes  = deinterleaveBytes_c;
    vu9_to_vu12        = vu9_to_vu12_c;
    yvu9_to_yuy2       = yvu9_to_yuy2_c;

    uyvytoyuv420       = uyvytoyuv420_c;
    uyvytoyuv422       = uyvytoyuv422_c;
    yuyvtoyuv420       = yuyvtoyuv420_c;
    yuyvtoyuv422       = yuyvtoyuv422_c;
}