| fcfbc150 |
/* |
| 8a322796 |
* 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 |
| 4fadc2b4 |
* |
| d026b45e |
* This file is part of FFmpeg.
* |
| 819ee683 |
* 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. |
| 4fadc2b4 |
* |
| d026b45e |
* FFmpeg is distributed in the hope that it will be useful, |
| 4fadc2b4 |
* but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 819ee683 |
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details. |
| 4fadc2b4 |
* |
| 819ee683 |
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software |
| b19bcbaa |
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
| a3aece93 |
*/
|
| 0d9f3d85 |
#include <stddef.h>
|
| 2b677ffc |
#include "libavutil/attributes.h"
|
| 89d94b21 |
static inline void rgb24tobgr32_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| b234ae81 |
{ |
| 89d94b21 |
uint8_t *dest = dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size; |
| d1adad3c |
|
| dd68318c |
while (s < end) { |
| 7d73d1c3 |
#if HAVE_BIGENDIAN |
| 89d94b21 |
/* RGB24 (= R, G, B) -> RGB32 (= A, B, G, R) */
*dest++ = 255;
*dest++ = s[2];
*dest++ = s[1];
*dest++ = s[0];
s += 3; |
| 7d73d1c3 |
#else |
| 89d94b21 |
*dest++ = *s++;
*dest++ = *s++;
*dest++ = *s++;
*dest++ = 255; |
| 7d73d1c3 |
#endif |
| 6e42e6c4 |
} |
| b234ae81 |
} |
| 59ac5a93 |
|
| 89d94b21 |
static inline void rgb32tobgr24_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| 59ac5a93 |
{ |
| 89d94b21 |
uint8_t *dest = dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size; |
| d1adad3c |
|
| dd68318c |
while (s < end) { |
| a898cdc9 |
#if HAVE_BIGENDIAN |
| 89d94b21 |
/* RGB32 (= A, B, G, R) -> RGB24 (= R, G, B) */ |
| 6e42e6c4 |
s++; |
| 89d94b21 |
dest[2] = *s++;
dest[1] = *s++;
dest[0] = *s++;
dest += 3; |
| 6cb38650 |
#else |
| 89d94b21 |
*dest++ = *s++;
*dest++ = *s++;
*dest++ = *s++; |
| 6e42e6c4 |
s++; |
| 6cb38650 |
#endif |
| 6e42e6c4 |
} |
| 59ac5a93 |
} |
| b238eb2e |
|
| a3aece93 |
/* |
| 89d94b21 |
* original by Strepto/Astral
* ported to gcc & bugfixed: A'rpi |
| 652f5185 |
* MMXEXT, 3DNOW optimization by Nick Kurshev |
| 89d94b21 |
* 32-bit C version, and and&add trick by Michael Niedermayer
*/ |
| b8e89339 |
static inline void rgb15to16_c(const uint8_t *src, uint8_t *dst, int src_size) |
| b238eb2e |
{ |
| 89d94b21 |
register uint8_t *d = dst;
register const uint8_t *s = src; |
| 72b95764 |
register const uint8_t *end = s + src_size;
const uint8_t *mm_end = end - 3;
|
| dd68318c |
while (s < mm_end) { |
| 89d94b21 |
register unsigned x = *((const uint32_t *)s);
*((uint32_t *)d) = (x & 0x7FFF7FFF) + (x & 0x7FE07FE0);
d += 4;
s += 4; |
| 0d9f3d85 |
} |
| dd68318c |
if (s < end) { |
| 89d94b21 |
register unsigned short x = *((const uint16_t *)s);
*((uint16_t *)d) = (x & 0x7FFF) + (x & 0x7FE0); |
| 0d9f3d85 |
} |
| b238eb2e |
} |
| fcfbc150 |
|
| b8e89339 |
static inline void rgb16to15_c(const uint8_t *src, uint8_t *dst, int src_size) |
| ac4d0aea |
{ |
| 89d94b21 |
register uint8_t *d = dst;
register const uint8_t *s = src; |
| 72b95764 |
register const uint8_t *end = s + src_size;
const uint8_t *mm_end = end - 3; |
| d1adad3c |
|
| dd68318c |
while (s < mm_end) { |
| 89d94b21 |
register uint32_t x = *((const uint32_t *)s);
*((uint32_t *)d) = ((x >> 1) & 0x7FE07FE0) | (x & 0x001F001F);
s += 4;
d += 4; |
| ac4d0aea |
} |
| dd68318c |
if (s < end) { |
| 89d94b21 |
register uint16_t x = *((const uint16_t *)s);
*((uint16_t *)d) = ((x >> 1) & 0x7FE0) | (x & 0x001F); |
| ac4d0aea |
}
}
|
| b8e89339 |
static inline void rgb32to16_c(const uint8_t *src, uint8_t *dst, int src_size) |
| fcfbc150 |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size; |
| d1adad3c |
|
| dd68318c |
while (s < end) { |
| 89d94b21 |
register int rgb = *(const uint32_t *)s;
s += 4;
*d++ = ((rgb & 0xFF) >> 3) +
((rgb & 0xFC00) >> 5) +
((rgb & 0xF80000) >> 8); |
| 6e42e6c4 |
} |
| fcfbc150 |
}
|
| 89d94b21 |
static inline void rgb32tobgr16_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| ac4d0aea |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 89d94b21 |
register int rgb = *(const uint32_t *)s;
s += 4;
*d++ = ((rgb & 0xF8) << 8) +
((rgb & 0xFC00) >> 5) +
((rgb & 0xF80000) >> 19); |
| 6e42e6c4 |
} |
| ac4d0aea |
}
|
| b8e89339 |
static inline void rgb32to15_c(const uint8_t *src, uint8_t *dst, int src_size) |
| fcfbc150 |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 89d94b21 |
register int rgb = *(const uint32_t *)s;
s += 4;
*d++ = ((rgb & 0xFF) >> 3) +
((rgb & 0xF800) >> 6) +
((rgb & 0xF80000) >> 9); |
| 6e42e6c4 |
} |
| fcfbc150 |
}
|
| 89d94b21 |
static inline void rgb32tobgr15_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| ac4d0aea |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 89d94b21 |
register int rgb = *(const uint32_t *)s;
s += 4;
*d++ = ((rgb & 0xF8) << 7) +
((rgb & 0xF800) >> 6) +
((rgb & 0xF80000) >> 19); |
| 6e42e6c4 |
} |
| ac4d0aea |
}
|
| 89d94b21 |
static inline void rgb24tobgr16_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| 996e1a7c |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 6e42e6c4 |
const int b = *s++;
const int g = *s++;
const int r = *s++; |
| 89d94b21 |
*d++ = (b >> 3) | ((g & 0xFC) << 3) | ((r & 0xF8) << 8); |
| 6e42e6c4 |
} |
| 996e1a7c |
}
|
| b8e89339 |
static inline void rgb24to16_c(const uint8_t *src, uint8_t *dst, int src_size) |
| ac4d0aea |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 6e42e6c4 |
const int r = *s++;
const int g = *s++;
const int b = *s++; |
| 89d94b21 |
*d++ = (b >> 3) | ((g & 0xFC) << 3) | ((r & 0xF8) << 8); |
| 6e42e6c4 |
} |
| ac4d0aea |
}
|
| 89d94b21 |
static inline void rgb24tobgr15_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| 996e1a7c |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 6e42e6c4 |
const int b = *s++;
const int g = *s++;
const int r = *s++; |
| 89d94b21 |
*d++ = (b >> 3) | ((g & 0xF8) << 2) | ((r & 0xF8) << 7); |
| 6e42e6c4 |
} |
| 0d9f3d85 |
}
|
| b8e89339 |
static inline void rgb24to15_c(const uint8_t *src, uint8_t *dst, int src_size) |
| ac4d0aea |
{ |
| 89d94b21 |
uint16_t *d = (uint16_t *)dst;
const uint8_t *s = src; |
| 72b95764 |
const uint8_t *end = s + src_size;
|
| dd68318c |
while (s < end) { |
| 6e42e6c4 |
const int r = *s++;
const int g = *s++;
const int b = *s++; |
| 89d94b21 |
*d++ = (b >> 3) | ((g & 0xF8) << 2) | ((r & 0xF8) << 7); |
| 6e42e6c4 |
} |
| ac4d0aea |
}
|
| 89d94b21 |
static inline void rgb15tobgr24_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| 0d9f3d85 |
{ |
| 89d94b21 |
uint8_t *d = dst;
const uint16_t *s = (const uint16_t *)src; |
| 72b95764 |
const uint16_t *end = s + src_size / 2;
|
| dd68318c |
while (s < end) { |
| 72b95764 |
register uint16_t bgr = *s++; |
| e99f1a8c |
*d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
*d++ = ((bgr&0x03E0)>>2) | ((bgr&0x03E0)>> 7); |
| 0827222b |
*d++ = ((bgr&0x7C00)>>7) | ((bgr&0x7C00)>>12); |
| 6e42e6c4 |
} |
| 0d9f3d85 |
}
|
| 89d94b21 |
static inline void rgb16tobgr24_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| 0d9f3d85 |
{ |
| 89d94b21 |
uint8_t *d = (uint8_t *)dst;
const uint16_t *s = (const uint16_t *)src; |
| 72b95764 |
const uint16_t *end = s + src_size / 2;
|
| dd68318c |
while (s < end) { |
| 72b95764 |
register uint16_t bgr = *s++; |
| e99f1a8c |
*d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
*d++ = ((bgr&0x07E0)>>3) | ((bgr&0x07E0)>> 9); |
| 0827222b |
*d++ = ((bgr&0xF800)>>8) | ((bgr&0xF800)>>13); |
| 6e42e6c4 |
} |
| 0d9f3d85 |
}
|
| b8e89339 |
static inline void rgb15to32_c(const uint8_t *src, uint8_t *dst, int src_size) |
| 0d9f3d85 |
{ |
| 89d94b21 |
uint8_t *d = dst;
const uint16_t *s = (const uint16_t *)src; |
| 72b95764 |
const uint16_t *end = s + src_size / 2;
|
| dd68318c |
while (s < end) { |
| 72b95764 |
register uint16_t bgr = *s++; |
| a898cdc9 |
#if HAVE_BIGENDIAN |
| f8a138be |
*d++ = 255; |
| 0827222b |
*d++ = ((bgr&0x7C00)>>7) | ((bgr&0x7C00)>>12); |
| e99f1a8c |
*d++ = ((bgr&0x03E0)>>2) | ((bgr&0x03E0)>> 7);
*d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2); |
| 6cb38650 |
#else |
| e99f1a8c |
*d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
*d++ = ((bgr&0x03E0)>>2) | ((bgr&0x03E0)>> 7); |
| 0827222b |
*d++ = ((bgr&0x7C00)>>7) | ((bgr&0x7C00)>>12); |
| f8a138be |
*d++ = 255; |
| deb2277c |
#endif |
| 6e42e6c4 |
} |
| 0d9f3d85 |
} |
| 996e1a7c |
|
| b8e89339 |
static inline void rgb16to32_c(const uint8_t *src, uint8_t *dst, int src_size) |
| 0d9f3d85 |
{ |
| 89d94b21 |
uint8_t *d = dst;
const uint16_t *s = (const uint16_t *)src; |
| 72b95764 |
const uint16_t *end = s + src_size / 2;
|
| dd68318c |
while (s < end) { |
| 72b95764 |
register uint16_t bgr = *s++; |
| a898cdc9 |
#if HAVE_BIGENDIAN |
| f8a138be |
*d++ = 255; |
| 0827222b |
*d++ = ((bgr&0xF800)>>8) | ((bgr&0xF800)>>13); |
| e99f1a8c |
*d++ = ((bgr&0x07E0)>>3) | ((bgr&0x07E0)>> 9);
*d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2); |
| 6cb38650 |
#else |
| e99f1a8c |
*d++ = ((bgr&0x001F)<<3) | ((bgr&0x001F)>> 2);
*d++ = ((bgr&0x07E0)>>3) | ((bgr&0x07E0)>> 9); |
| 0827222b |
*d++ = ((bgr&0xF800)>>8) | ((bgr&0xF800)>>13); |
| f8a138be |
*d++ = 255; |
| 6cb38650 |
#endif |
| 6e42e6c4 |
} |
| 996e1a7c |
} |
| fcfbc150 |
|
| 89d94b21 |
static inline void shuffle_bytes_2103_c(const uint8_t *src, uint8_t *dst,
int src_size) |
| 99969243 |
{ |
| 89d94b21 |
int idx = 15 - src_size;
const uint8_t *s = src - idx;
uint8_t *d = dst - idx;
for (; idx < 15; idx += 4) {
register int v = *(const uint32_t *)&s[idx], g = v & 0xff00ff00;
v &= 0xff00ff;
*(uint32_t *)&d[idx] = (v >> 16) + g + (v << 16); |
| 6e42e6c4 |
} |
| 99969243 |
}
|
| b8e89339 |
static inline void rgb24tobgr24_c(const uint8_t *src, uint8_t *dst, int src_size) |
| 74d35835 |
{ |
| 6e42e6c4 |
unsigned i; |
| 89d94b21 |
for (i = 0; i < src_size; i += 3) { |
| 72b95764 |
register uint8_t x = src[i + 2]; |
| 89d94b21 |
dst[i + 1] = src[i + 1];
dst[i + 2] = src[i + 0];
dst[i + 0] = x; |
| 6e42e6c4 |
} |
| 74d35835 |
}
|
| d1adad3c |
static inline void yuvPlanartoyuy2_c(const uint8_t *ysrc, const uint8_t *usrc,
const uint8_t *vsrc, uint8_t *dst, |
| b8e89339 |
int width, int height,
int lumStride, int chromStride,
int dstStride, int vertLumPerChroma) |
| d9d58d17 |
{ |
| 72b95764 |
int y, i; |
| d1adad3c |
const int chromWidth = width >> 1; |
| 89d94b21 |
for (y = 0; y < height; y++) { |
| d1adad3c |
#if HAVE_FAST_64BIT |
| 89d94b21 |
uint64_t *ldst = (uint64_t *)dst; |
| 6e42e6c4 |
const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc; |
| dd68318c |
for (i = 0; i < chromWidth; i += 2) { |
| 72b95764 |
uint64_t k = yc[0] + (uc[0] << 8) + |
| a3097260 |
(yc[1] << 16) + (unsigned)(vc[0] << 24); |
| 72b95764 |
uint64_t l = yc[2] + (uc[1] << 8) + |
| a3097260 |
(yc[3] << 16) + (unsigned)(vc[1] << 24); |
| 6e42e6c4 |
*ldst++ = k + (l << 32); |
| 89d94b21 |
yc += 4;
uc += 2;
vc += 2; |
| 6e42e6c4 |
} |
| 0d9f3d85 |
#else |
| 89d94b21 |
int *idst = (int32_t *)dst; |
| 6e42e6c4 |
const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc; |
| 89d94b21 |
|
| dd68318c |
for (i = 0; i < chromWidth; i++) { |
| a898cdc9 |
#if HAVE_BIGENDIAN |
| 89d94b21 |
*idst++ = (yc[0] << 24) + (uc[0] << 16) +
(yc[1] << 8) + (vc[0] << 0); |
| da7f8893 |
#else |
| 6e42e6c4 |
*idst++ = yc[0] + (uc[0] << 8) + |
| 89d94b21 |
(yc[1] << 16) + (vc[0] << 24); |
| 6e42e6c4 |
#endif
yc += 2;
uc++;
vc++;
}
#endif |
| 89d94b21 |
if ((y & (vertLumPerChroma - 1)) == vertLumPerChroma - 1) { |
| 6e42e6c4 |
usrc += chromStride;
vsrc += chromStride;
}
ysrc += lumStride;
dst += dstStride;
} |
| d9d58d17 |
}
|
| dabcdbc4 |
/** |
| 8a322796 |
* 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.) |
| dabcdbc4 |
*/ |
| d1adad3c |
static inline void yv12toyuy2_c(const uint8_t *ysrc, const uint8_t *usrc,
const uint8_t *vsrc, uint8_t *dst, |
| 89d94b21 |
int width, int height, int lumStride,
int chromStride, int dstStride) |
| b1ec5875 |
{ |
| 6e42e6c4 |
//FIXME interpolate chroma |
| d1adad3c |
yuvPlanartoyuy2_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
chromStride, dstStride, 2); |
| caeaabe7 |
}
|
| d1adad3c |
static inline void yuvPlanartouyvy_c(const uint8_t *ysrc, const uint8_t *usrc,
const uint8_t *vsrc, uint8_t *dst, |
| b8e89339 |
int width, int height,
int lumStride, int chromStride,
int dstStride, int vertLumPerChroma) |
| caeaabe7 |
{ |
| 72b95764 |
int y, i; |
| d1adad3c |
const int chromWidth = width >> 1; |
| 89d94b21 |
for (y = 0; y < height; y++) { |
| 02a6a6ee |
#if HAVE_FAST_64BIT |
| 89d94b21 |
uint64_t *ldst = (uint64_t *)dst; |
| 6e42e6c4 |
const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc; |
| dd68318c |
for (i = 0; i < chromWidth; i += 2) { |
| 72b95764 |
uint64_t k = uc[0] + (yc[0] << 8) + |
| a07e9d72 |
(vc[0] << 16) + (unsigned)(yc[1] << 24); |
| 72b95764 |
uint64_t l = uc[1] + (yc[2] << 8) + |
| a07e9d72 |
(vc[1] << 16) + (unsigned)(yc[3] << 24); |
| 6e42e6c4 |
*ldst++ = k + (l << 32); |
| 89d94b21 |
yc += 4;
uc += 2;
vc += 2; |
| 6e42e6c4 |
} |
| caeaabe7 |
#else |
| 89d94b21 |
int *idst = (int32_t *)dst; |
| 6e42e6c4 |
const uint8_t *yc = ysrc, *uc = usrc, *vc = vsrc; |
| 89d94b21 |
|
| dd68318c |
for (i = 0; i < chromWidth; i++) { |
| a898cdc9 |
#if HAVE_BIGENDIAN |
| 89d94b21 |
*idst++ = (uc[0] << 24) + (yc[0] << 16) +
(vc[0] << 8) + (yc[1] << 0); |
| da7f8893 |
#else |
| 6e42e6c4 |
*idst++ = uc[0] + (yc[0] << 8) + |
| 89d94b21 |
(vc[0] << 16) + (yc[1] << 24); |
| 6e42e6c4 |
#endif
yc += 2;
uc++;
vc++;
}
#endif |
| 89d94b21 |
if ((y & (vertLumPerChroma - 1)) == vertLumPerChroma - 1) { |
| 6e42e6c4 |
usrc += chromStride;
vsrc += chromStride;
}
ysrc += lumStride; |
| 89d94b21 |
dst += dstStride; |
| 6e42e6c4 |
} |
| caeaabe7 |
}
/** |
| 8a322796 |
* 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.) |
| caeaabe7 |
*/ |
| d1adad3c |
static inline void yv12touyvy_c(const uint8_t *ysrc, const uint8_t *usrc,
const uint8_t *vsrc, uint8_t *dst, |
| 89d94b21 |
int width, int height, int lumStride,
int chromStride, int dstStride) |
| caeaabe7 |
{ |
| 6e42e6c4 |
//FIXME interpolate chroma |
| d1adad3c |
yuvPlanartouyvy_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
chromStride, dstStride, 2); |
| b1ec5875 |
}
/** |
| 594ff7cc |
* Width should be a multiple of 16. |
| b1ec5875 |
*/ |
| d1adad3c |
static inline void yuv422ptouyvy_c(const uint8_t *ysrc, const uint8_t *usrc,
const uint8_t *vsrc, uint8_t *dst, |
| 89d94b21 |
int width, int height, int lumStride,
int chromStride, int dstStride) |
| a6100f39 |
{ |
| d1adad3c |
yuvPlanartouyvy_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
chromStride, dstStride, 1); |
| a6100f39 |
}
/**
* Width should be a multiple of 16.
*/ |
| d1adad3c |
static inline void yuv422ptoyuy2_c(const uint8_t *ysrc, const uint8_t *usrc,
const uint8_t *vsrc, uint8_t *dst, |
| 89d94b21 |
int width, int height, int lumStride,
int chromStride, int dstStride) |
| b1ec5875 |
{ |
| d1adad3c |
yuvPlanartoyuy2_c(ysrc, usrc, vsrc, dst, width, height, lumStride,
chromStride, dstStride, 1); |
| b1ec5875 |
}
/** |
| 8a322796 |
* 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.) |
| b1ec5875 |
*/ |
| d1adad3c |
static inline void yuy2toyv12_c(const uint8_t *src, uint8_t *ydst,
uint8_t *udst, uint8_t *vdst, |
| 89d94b21 |
int width, int height, int lumStride,
int chromStride, int srcStride) |
| d9d58d17 |
{ |
| b8e89339 |
int y; |
| d1adad3c |
const int chromWidth = width >> 1; |
| 89d94b21 |
for (y = 0; y < height; y += 2) { |
| b8e89339 |
int i; |
| 89d94b21 |
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]; |
| 6e42e6c4 |
}
ydst += lumStride;
src += srcStride;
|
| 89d94b21 |
for (i = 0; i < chromWidth; i++) {
ydst[2 * i + 0] = src[4 * i + 0];
ydst[2 * i + 1] = src[4 * i + 2]; |
| 6e42e6c4 |
}
udst += chromStride;
vdst += chromStride;
ydst += lumStride;
src += srcStride;
} |
| 42b5fcb8 |
} |
| 81c0590e |
|
| b8e89339 |
static inline void planar2x_c(const uint8_t *src, uint8_t *dst, int srcWidth,
int srcHeight, int srcStride, int dstStride) |
| b241cbf2 |
{ |
| 89d94b21 |
int x, y; |
| 6a4970ab |
|
| 89d94b21 |
dst[0] = src[0]; |
| 6a4970ab |
|
| 6e42e6c4 |
// first line |
| 89d94b21 |
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; |
| 6e42e6c4 |
} |
| 89d94b21 |
dst[2 * srcWidth - 1] = src[srcWidth - 1]; |
| 6a4970ab |
|
| 89d94b21 |
dst += dstStride; |
| b241cbf2 |
|
| 89d94b21 |
for (y = 1; y < srcHeight; y++) { |
| d1adad3c |
const int mmxSize = 1; |
| 40b433b6 |
|
| 89d94b21 |
dst[0] = (src[0] * 3 + src[srcStride]) >> 2;
dst[dstStride] = (src[0] + 3 * src[srcStride]) >> 2; |
| 6e42e6c4 |
|
| 89d94b21 |
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; |
| 6e42e6c4 |
} |
| 89d94b21 |
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; |
| 6e42e6c4 |
|
| 89d94b21 |
dst += dstStride * 2;
src += srcStride; |
| 6e42e6c4 |
} |
| 6a4970ab |
|
| 6e42e6c4 |
// last line |
| 89d94b21 |
dst[0] = src[0]; |
| 6a4970ab |
|
| 89d94b21 |
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; |
| 6e42e6c4 |
} |
| 89d94b21 |
dst[2 * srcWidth - 1] = src[srcWidth - 1]; |
| b241cbf2 |
}
|
| 81c0590e |
/** |
| 8a322796 |
* 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. |
| 594ff7cc |
* FIXME: Write HQ version. |
| 81c0590e |
*/ |
| d1adad3c |
static inline void uyvytoyv12_c(const uint8_t *src, uint8_t *ydst,
uint8_t *udst, uint8_t *vdst, |
| 89d94b21 |
int width, int height, int lumStride,
int chromStride, int srcStride) |
| 81c0590e |
{ |
| b8e89339 |
int y; |
| d1adad3c |
const int chromWidth = width >> 1; |
| 89d94b21 |
for (y = 0; y < height; y += 2) { |
| b8e89339 |
int i; |
| 89d94b21 |
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]; |
| 6e42e6c4 |
}
ydst += lumStride;
src += srcStride;
|
| 89d94b21 |
for (i = 0; i < chromWidth; i++) {
ydst[2 * i + 0] = src[4 * i + 1];
ydst[2 * i + 1] = src[4 * i + 3]; |
| 6e42e6c4 |
}
udst += chromStride;
vdst += chromStride;
ydst += lumStride;
src += srcStride;
} |
| 81c0590e |
}
|
| 1de97d84 |
/** |
| 8a322796 |
* Height should be a multiple of 2 and width should be a multiple of 2.
* (If this is a problem for anyone then tell me, and I will fix it.)
* Chrominance data is only taken from every second line, |
| 594ff7cc |
* others are ignored in the C version.
* FIXME: Write HQ version. |
| 1de97d84 |
*/ |
| a37fd7f9 |
void ff_rgb24toyv12_c(const uint8_t *src, uint8_t *ydst, uint8_t *udst, |
| b8e89339 |
uint8_t *vdst, int width, int height, int lumStride, |
| a37fd7f9 |
int chromStride, int srcStride, int32_t *rgb2yuv) |
| 1de97d84 |
{ |
| a37fd7f9 |
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]; |
| b8e89339 |
int y; |
| d1adad3c |
const int chromWidth = width >> 1; |
| 72b95764 |
for (y = 0; y < height; y += 2) { |
| b8e89339 |
int i; |
| 89d94b21 |
for (i = 0; i < chromWidth; i++) {
unsigned int b = src[6 * i + 0];
unsigned int g = src[6 * i + 1];
unsigned int r = src[6 * i + 2]; |
| 6e42e6c4 |
|
| a37fd7f9 |
unsigned int Y = ((ry * r + gy * g + by * b) >> RGB2YUV_SHIFT) + 16;
unsigned int V = ((rv * r + gv * g + bv * b) >> RGB2YUV_SHIFT) + 128;
unsigned int U = ((ru * r + gu * g + bu * b) >> RGB2YUV_SHIFT) + 128; |
| 6e42e6c4 |
udst[i] = U;
vdst[i] = V; |
| 89d94b21 |
ydst[2 * i] = Y; |
| 6e42e6c4 |
|
| 89d94b21 |
b = src[6 * i + 3];
g = src[6 * i + 4];
r = src[6 * i + 5]; |
| 6e42e6c4 |
|
| a37fd7f9 |
Y = ((ry * r + gy * g + by * b) >> RGB2YUV_SHIFT) + 16; |
| 89d94b21 |
ydst[2 * i + 1] = Y; |
| 6e42e6c4 |
}
ydst += lumStride;
src += srcStride;
|
| e99f1a8c |
if (y+1 == height) |
| 9a5624a0 |
break;
|
| 89d94b21 |
for (i = 0; i < chromWidth; i++) {
unsigned int b = src[6 * i + 0];
unsigned int g = src[6 * i + 1];
unsigned int r = src[6 * i + 2]; |
| 6e42e6c4 |
|
| a37fd7f9 |
unsigned int Y = ((ry * r + gy * g + by * b) >> RGB2YUV_SHIFT) + 16; |
| 6e42e6c4 |
|
| 89d94b21 |
ydst[2 * i] = Y; |
| 6e42e6c4 |
|
| 89d94b21 |
b = src[6 * i + 3];
g = src[6 * i + 4];
r = src[6 * i + 5]; |
| 6e42e6c4 |
|
| a37fd7f9 |
Y = ((ry * r + gy * g + by * b) >> RGB2YUV_SHIFT) + 16; |
| 89d94b21 |
ydst[2 * i + 1] = Y; |
| 6e42e6c4 |
}
udst += chromStride;
vdst += chromStride;
ydst += lumStride;
src += srcStride;
} |
| 1de97d84 |
} |
| 5d55fdb4 |
|
| d1adad3c |
static void interleaveBytes_c(const uint8_t *src1, const uint8_t *src2, |
| 89d94b21 |
uint8_t *dest, int width, int height,
int src1Stride, int src2Stride, int dstStride) |
| dd68318c |
{ |
| b8e89339 |
int h; |
| 5d55fdb4 |
|
| 89d94b21 |
for (h = 0; h < height; h++) { |
| b8e89339 |
int w; |
| 89d94b21 |
for (w = 0; w < width; w++) {
dest[2 * w + 0] = src1[w];
dest[2 * w + 1] = src2[w]; |
| 6e42e6c4 |
}
dest += dstStride; |
| 40b433b6 |
src1 += src1Stride;
src2 += src2Stride; |
| 6e42e6c4 |
} |
| 5d55fdb4 |
} |
| ac4d0aea |
|
| ef627bf9 |
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];
} |
| 9047491f |
src += srcStride; |
| ef627bf9 |
dst1 += dst1Stride;
dst2 += dst2Stride;
}
}
|
| d1adad3c |
static inline void vu9_to_vu12_c(const uint8_t *src1, const uint8_t *src2,
uint8_t *dst1, uint8_t *dst2, |
| b8e89339 |
int width, int height,
int srcStride1, int srcStride2,
int dstStride1, int dstStride2) |
| ac4d0aea |
{ |
| 72b95764 |
int x, y;
int w = width / 2;
int h = height / 2;
|
| 89d94b21 |
for (y = 0; y < h; y++) {
const uint8_t *s1 = src1 + srcStride1 * (y >> 1);
uint8_t *d = dst1 + dstStride1 * y; |
| 72b95764 |
for (x = 0; x < w; x++)
d[2 * x] = d[2 * x + 1] = s1[x]; |
| ac4d0aea |
} |
| 89d94b21 |
for (y = 0; y < h; y++) {
const uint8_t *s2 = src2 + srcStride2 * (y >> 1);
uint8_t *d = dst2 + dstStride2 * y; |
| 72b95764 |
for (x = 0; x < w; x++)
d[2 * x] = d[2 * x + 1] = s2[x]; |
| ac4d0aea |
}
}
|
| d1adad3c |
static inline void yvu9_to_yuy2_c(const uint8_t *src1, const uint8_t *src2,
const uint8_t *src3, uint8_t *dst, |
| b8e89339 |
int width, int height,
int srcStride1, int srcStride2,
int srcStride3, int dstStride) |
| ac4d0aea |
{ |
| 72b95764 |
int x, y;
int w = width / 2;
int h = height;
|
| 89d94b21 |
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; |
| 72b95764 |
for (x = 0; x < w; x++) { |
| 89d94b21 |
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]; |
| 9b734d44 |
} |
| ac4d0aea |
}
} |
| df170b81 |
|
| d1adad3c |
static void extract_even_c(const uint8_t *src, uint8_t *dst, int count) |
| 0411072e |
{ |
| 89d94b21 |
dst += count;
src += count * 2;
count = -count;
while (count < 0) {
dst[count] = src[2 * count]; |
| 0411072e |
count++;
}
}
|
| d1adad3c |
static void extract_even2_c(const uint8_t *src, uint8_t *dst0, uint8_t *dst1,
int count) |
| 0411072e |
{ |
| 89d94b21 |
dst0 += count;
dst1 += count;
src += count * 4;
count = -count;
while (count < 0) {
dst0[count] = src[4 * count + 0];
dst1[count] = src[4 * count + 2]; |
| 0411072e |
count++;
}
}
|
| d1adad3c |
static void extract_even2avg_c(const uint8_t *src0, const uint8_t *src1,
uint8_t *dst0, uint8_t *dst1, int count) |
| 870a4adb |
{ |
| 89d94b21 |
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; |
| 870a4adb |
count++;
}
}
|
| d1adad3c |
static void extract_odd2_c(const uint8_t *src, uint8_t *dst0, uint8_t *dst1,
int count) |
| 0411072e |
{ |
| 89d94b21 |
dst0 += count;
dst1 += count;
src += count * 4;
count = -count; |
| 3d99e069 |
src++; |
| 89d94b21 |
while (count < 0) {
dst0[count] = src[4 * count + 0];
dst1[count] = src[4 * count + 2]; |
| 0411072e |
count++;
}
}
|
| d1adad3c |
static void extract_odd2avg_c(const uint8_t *src0, const uint8_t *src1,
uint8_t *dst0, uint8_t *dst1, int count) |
| 870a4adb |
{ |
| 89d94b21 |
dst0 += count;
dst1 += count;
src0 += count * 4;
src1 += count * 4;
count = -count; |
| 3d99e069 |
src0++;
src1++; |
| 89d94b21 |
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; |
| 870a4adb |
count++;
}
}
|
| d1adad3c |
static void yuyvtoyuv420_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst, |
| b8e89339 |
const uint8_t *src, int width, int height,
int lumStride, int chromStride, int srcStride) |
| 0411072e |
{ |
| b8e89339 |
int y; |
| 570d63ee |
const int chromWidth = FF_CEIL_RSHIFT(width, 1); |
| 0411072e |
|
| 89d94b21 |
for (y = 0; y < height; y++) { |
| d1adad3c |
extract_even_c(src, ydst, width); |
| 89d94b21 |
if (y & 1) { |
| d1adad3c |
extract_odd2avg_c(src - srcStride, src, udst, vdst, chromWidth); |
| 89d94b21 |
udst += chromStride;
vdst += chromStride; |
| 0411072e |
}
|
| 89d94b21 |
src += srcStride;
ydst += lumStride; |
| 0411072e |
}
}
|
| d1adad3c |
static void yuyvtoyuv422_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst, |
| b8e89339 |
const uint8_t *src, int width, int height,
int lumStride, int chromStride, int srcStride) |
| 0411072e |
{ |
| b8e89339 |
int y; |
| 570d63ee |
const int chromWidth = FF_CEIL_RSHIFT(width, 1); |
| 0411072e |
|
| 89d94b21 |
for (y = 0; y < height; y++) { |
| d1adad3c |
extract_even_c(src, ydst, width);
extract_odd2_c(src, udst, vdst, chromWidth); |
| 0411072e |
|
| 89d94b21 |
src += srcStride;
ydst += lumStride;
udst += chromStride;
vdst += chromStride; |
| 0411072e |
}
}
|
| d1adad3c |
static void uyvytoyuv420_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst, |
| b8e89339 |
const uint8_t *src, int width, int height,
int lumStride, int chromStride, int srcStride) |
| 0411072e |
{ |
| b8e89339 |
int y; |
| 570d63ee |
const int chromWidth = FF_CEIL_RSHIFT(width, 1); |
| 0411072e |
|
| 89d94b21 |
for (y = 0; y < height; y++) { |
| d1adad3c |
extract_even_c(src + 1, ydst, width); |
| 89d94b21 |
if (y & 1) { |
| d1adad3c |
extract_even2avg_c(src - srcStride, src, udst, vdst, chromWidth); |
| 89d94b21 |
udst += chromStride;
vdst += chromStride; |
| 0411072e |
}
|
| 89d94b21 |
src += srcStride;
ydst += lumStride; |
| 0411072e |
}
}
|
| d1adad3c |
static void uyvytoyuv422_c(uint8_t *ydst, uint8_t *udst, uint8_t *vdst, |
| b8e89339 |
const uint8_t *src, int width, int height,
int lumStride, int chromStride, int srcStride) |
| 0411072e |
{ |
| b8e89339 |
int y; |
| 570d63ee |
const int chromWidth = FF_CEIL_RSHIFT(width, 1); |
| 0411072e |
|
| 89d94b21 |
for (y = 0; y < height; y++) { |
| d1adad3c |
extract_even_c(src + 1, ydst, width);
extract_even2_c(src, udst, vdst, chromWidth); |
| 0411072e |
|
| 89d94b21 |
src += srcStride;
ydst += lumStride;
udst += chromStride;
vdst += chromStride; |
| 0411072e |
}
}
|
| 2b677ffc |
static av_cold void rgb2rgb_init_c(void) |
| d1adad3c |
{
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;
shuffle_bytes_2103 = shuffle_bytes_2103_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; |
| a37fd7f9 |
ff_rgb24toyv12 = ff_rgb24toyv12_c; |
| d1adad3c |
interleaveBytes = interleaveBytes_c; |
| ef627bf9 |
deinterleaveBytes = deinterleaveBytes_c; |
| d1adad3c |
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; |
| df170b81 |
} |