/*
* Copyright (C) 2001-2003 Michael Niedermayer <michaelni@gmx.at>
*
- * This file is part of Libav.
+ * This file is part of FFmpeg.
*
- * Libav is free software; you can redistribute it and/or
+ * 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.
*
- * Libav is distributed in the hope that it will be useful,
+ * 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 Libav; if not, write to the Free Software
+ * License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "swscale.h"
#include "swscale_internal.h"
#include "rgb2rgb.h"
+#include "libavutil/avassert.h"
#include "libavutil/intreadwrite.h"
#include "libavutil/cpu.h"
#include "libavutil/avutil.h"
#include "libavutil/bswap.h"
#include "libavutil/pixdesc.h"
-#define DITHER1XBPP
#define RGB2YUV_SHIFT 15
#define BY ( (int)(0.114*219/255*(1<<RGB2YUV_SHIFT)+0.5))
};
#endif
+DECLARE_ALIGNED(8, const uint8_t, dithers)[8][8][8]={
+{
+ { 0, 1, 0, 1, 0, 1, 0, 1,},
+ { 1, 0, 1, 0, 1, 0, 1, 0,},
+ { 0, 1, 0, 1, 0, 1, 0, 1,},
+ { 1, 0, 1, 0, 1, 0, 1, 0,},
+ { 0, 1, 0, 1, 0, 1, 0, 1,},
+ { 1, 0, 1, 0, 1, 0, 1, 0,},
+ { 0, 1, 0, 1, 0, 1, 0, 1,},
+ { 1, 0, 1, 0, 1, 0, 1, 0,},
+},{
+ { 1, 2, 1, 2, 1, 2, 1, 2,},
+ { 3, 0, 3, 0, 3, 0, 3, 0,},
+ { 1, 2, 1, 2, 1, 2, 1, 2,},
+ { 3, 0, 3, 0, 3, 0, 3, 0,},
+ { 1, 2, 1, 2, 1, 2, 1, 2,},
+ { 3, 0, 3, 0, 3, 0, 3, 0,},
+ { 1, 2, 1, 2, 1, 2, 1, 2,},
+ { 3, 0, 3, 0, 3, 0, 3, 0,},
+},{
+ { 2, 4, 3, 5, 2, 4, 3, 5,},
+ { 6, 0, 7, 1, 6, 0, 7, 1,},
+ { 3, 5, 2, 4, 3, 5, 2, 4,},
+ { 7, 1, 6, 0, 7, 1, 6, 0,},
+ { 2, 4, 3, 5, 2, 4, 3, 5,},
+ { 6, 0, 7, 1, 6, 0, 7, 1,},
+ { 3, 5, 2, 4, 3, 5, 2, 4,},
+ { 7, 1, 6, 0, 7, 1, 6, 0,},
+},{
+ { 4, 8, 7, 11, 4, 8, 7, 11,},
+ { 12, 0, 15, 3, 12, 0, 15, 3,},
+ { 6, 10, 5, 9, 6, 10, 5, 9,},
+ { 14, 2, 13, 1, 14, 2, 13, 1,},
+ { 4, 8, 7, 11, 4, 8, 7, 11,},
+ { 12, 0, 15, 3, 12, 0, 15, 3,},
+ { 6, 10, 5, 9, 6, 10, 5, 9,},
+ { 14, 2, 13, 1, 14, 2, 13, 1,},
+},{
+ { 9, 17, 15, 23, 8, 16, 14, 22,},
+ { 25, 1, 31, 7, 24, 0, 30, 6,},
+ { 13, 21, 11, 19, 12, 20, 10, 18,},
+ { 29, 5, 27, 3, 28, 4, 26, 2,},
+ { 8, 16, 14, 22, 9, 17, 15, 23,},
+ { 24, 0, 30, 6, 25, 1, 31, 7,},
+ { 12, 20, 10, 18, 13, 21, 11, 19,},
+ { 28, 4, 26, 2, 29, 5, 27, 3,},
+},{
+ { 18, 34, 30, 46, 17, 33, 29, 45,},
+ { 50, 2, 62, 14, 49, 1, 61, 13,},
+ { 26, 42, 22, 38, 25, 41, 21, 37,},
+ { 58, 10, 54, 6, 57, 9, 53, 5,},
+ { 16, 32, 28, 44, 19, 35, 31, 47,},
+ { 48, 0, 60, 12, 51, 3, 63, 15,},
+ { 24, 40, 20, 36, 27, 43, 23, 39,},
+ { 56, 8, 52, 4, 59, 11, 55, 7,},
+},{
+ { 18, 34, 30, 46, 17, 33, 29, 45,},
+ { 50, 2, 62, 14, 49, 1, 61, 13,},
+ { 26, 42, 22, 38, 25, 41, 21, 37,},
+ { 58, 10, 54, 6, 57, 9, 53, 5,},
+ { 16, 32, 28, 44, 19, 35, 31, 47,},
+ { 48, 0, 60, 12, 51, 3, 63, 15,},
+ { 24, 40, 20, 36, 27, 43, 23, 39,},
+ { 56, 8, 52, 4, 59, 11, 55, 7,},
+},{
+ { 36, 68, 60, 92, 34, 66, 58, 90,},
+ { 100, 4,124, 28, 98, 2,122, 26,},
+ { 52, 84, 44, 76, 50, 82, 42, 74,},
+ { 116, 20,108, 12,114, 18,106, 10,},
+ { 32, 64, 56, 88, 38, 70, 62, 94,},
+ { 96, 0,120, 24,102, 6,126, 30,},
+ { 48, 80, 40, 72, 54, 86, 46, 78,},
+ { 112, 16,104, 8,118, 22,110, 14,},
+}};
+
+static const uint8_t flat64[8]={64,64,64,64,64,64,64,64};
+
+const uint16_t dither_scale[15][16]={
+{ 2, 3, 3, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5, 5,},
+{ 2, 3, 7, 7, 13, 13, 25, 25, 25, 25, 25, 25, 25, 25, 25, 25,},
+{ 3, 3, 4, 15, 15, 29, 57, 57, 57, 113, 113, 113, 113, 113, 113, 113,},
+{ 3, 4, 4, 5, 31, 31, 61, 121, 241, 241, 241, 241, 481, 481, 481, 481,},
+{ 3, 4, 5, 5, 6, 63, 63, 125, 249, 497, 993, 993, 993, 993, 993, 1985,},
+{ 3, 5, 6, 6, 6, 7, 127, 127, 253, 505, 1009, 2017, 4033, 4033, 4033, 4033,},
+{ 3, 5, 6, 7, 7, 7, 8, 255, 255, 509, 1017, 2033, 4065, 8129,16257,16257,},
+{ 3, 5, 6, 8, 8, 8, 8, 9, 511, 511, 1021, 2041, 4081, 8161,16321,32641,},
+{ 3, 5, 7, 8, 9, 9, 9, 9, 10, 1023, 1023, 2045, 4089, 8177,16353,32705,},
+{ 3, 5, 7, 8, 10, 10, 10, 10, 10, 11, 2047, 2047, 4093, 8185,16369,32737,},
+{ 3, 5, 7, 8, 10, 11, 11, 11, 11, 11, 12, 4095, 4095, 8189,16377,32753,},
+{ 3, 5, 7, 9, 10, 12, 12, 12, 12, 12, 12, 13, 8191, 8191,16381,32761,},
+{ 3, 5, 7, 9, 10, 12, 13, 13, 13, 13, 13, 13, 14,16383,16383,32765,},
+{ 3, 5, 7, 9, 10, 12, 14, 14, 14, 14, 14, 14, 14, 15,32767,32767,},
+{ 3, 5, 7, 9, 11, 12, 14, 15, 15, 15, 15, 15, 15, 15, 16,65535,},
+};
+
static av_always_inline void
-yuv2yuvX16_c_template(const int16_t *lumFilter, const int32_t **lumSrc,
+yuv2yuvX16_c_template(const int16_t *lumFilter, const int16_t **lumSrc,
int lumFilterSize, const int16_t *chrFilter,
- const int32_t **chrUSrc, const int32_t **chrVSrc,
- int chrFilterSize, const int32_t **alpSrc,
+ const int16_t **chrUSrc, const int16_t **chrVSrc,
+ int chrFilterSize, const int16_t **alpSrc,
uint16_t *dest[4], int dstW, int chrDstW,
int big_endian, int output_bits)
{
//FIXME Optimize (just quickly written not optimized..)
int i;
+ int shift = 11 + 16 - output_bits;
uint16_t *yDest = dest[0], *uDest = dest[1], *vDest = dest[2],
*aDest = CONFIG_SWSCALE_ALPHA ? dest[3] : NULL;
- int shift = 15 + 16 - output_bits;
#define output_pixel(pos, val) \
if (big_endian) { \
} \
}
for (i = 0; i < dstW; i++) {
- int val = 1 << (30-output_bits);
+ int val = 1 << (26-output_bits);
int j;
for (j = 0; j < lumFilterSize; j++)
if (uDest) {
for (i = 0; i < chrDstW; i++) {
- int u = 1 << (30-output_bits);
- int v = 1 << (30-output_bits);
+ int u = 1 << (26-output_bits);
+ int v = 1 << (26-output_bits);
int j;
for (j = 0; j < chrFilterSize; j++) {
if (CONFIG_SWSCALE_ALPHA && aDest) {
for (i = 0; i < dstW; i++) {
- int val = 1 << (30-output_bits);
+ int val = 1 << (26-output_bits);
int j;
for (j = 0; j < lumFilterSize; j++)
#define yuv2NBPS(bits, BE_LE, is_be) \
static void yuv2yuvX ## bits ## BE_LE ## _c(SwsContext *c, const int16_t *lumFilter, \
- const int16_t **_lumSrc, int lumFilterSize, \
- const int16_t *chrFilter, const int16_t **_chrUSrc, \
- const int16_t **_chrVSrc, \
- int chrFilterSize, const int16_t **_alpSrc, \
+ const int16_t **lumSrc, int lumFilterSize, \
+ const int16_t *chrFilter, const int16_t **chrUSrc, \
+ const int16_t **chrVSrc, \
+ int chrFilterSize, const int16_t **alpSrc, \
uint8_t *_dest[4], int dstW, int chrDstW) \
{ \
- const int32_t **lumSrc = (const int32_t **) _lumSrc, \
- **chrUSrc = (const int32_t **) _chrUSrc, \
- **chrVSrc = (const int32_t **) _chrVSrc, \
- **alpSrc = (const int32_t **) _alpSrc; \
yuv2yuvX16_c_template(lumFilter, lumSrc, lumFilterSize, \
chrFilter, chrUSrc, chrVSrc, chrFilterSize, \
alpSrc, (uint16_t **) _dest, \
const int16_t *chrFilter, const int16_t **chrUSrc,
const int16_t **chrVSrc,
int chrFilterSize, const int16_t **alpSrc,
- uint8_t *dest[4], int dstW, int chrDstW)
+ uint8_t *dest[4], int dstW, int chrDstW,
+ const uint8_t *lumDither, const uint8_t *chrDither)
{
uint8_t *yDest = dest[0], *uDest = dest[1], *vDest = dest[2],
*aDest = CONFIG_SWSCALE_ALPHA ? dest[3] : NULL;
//FIXME Optimize (just quickly written not optimized..)
for (i=0; i<dstW; i++) {
- int val=1<<18;
+ int val = lumDither[i&7] << 12;
int j;
for (j=0; j<lumFilterSize; j++)
val += lumSrc[j][i] * lumFilter[j];
if (uDest)
for (i=0; i<chrDstW; i++) {
- int u=1<<18;
- int v=1<<18;
+ int u = chrDither[i&7] << 12;
+ int v = chrDither[(i+3)&7] << 12;
int j;
for (j=0; j<chrFilterSize; j++) {
u += chrUSrc[j][i] * chrFilter[j];
if (CONFIG_SWSCALE_ALPHA && aDest)
for (i=0; i<dstW; i++) {
- int val=1<<18;
+ int val = lumDither[i&7] << 12;
int j;
for (j=0; j<lumFilterSize; j++)
val += alpSrc[j][i] * lumFilter[j];
static void yuv2yuv1_c(SwsContext *c, const int16_t *lumSrc,
const int16_t *chrUSrc, const int16_t *chrVSrc,
const int16_t *alpSrc,
- uint8_t *dest[4], int dstW, int chrDstW)
+ uint8_t *dest[4], int dstW, int chrDstW,
+ const uint8_t *lumDither, const uint8_t *chrDither)
{
uint8_t *yDest = dest[0], *uDest = dest[1], *vDest = dest[2],
*aDest = CONFIG_SWSCALE_ALPHA ? dest[3] : NULL;
int i;
for (i=0; i<dstW; i++) {
- int val= (lumSrc[i]+64)>>7;
+ int val= (lumSrc[i]+lumDither[i&7])>>7;
yDest[i]= av_clip_uint8(val);
}
if (uDest)
for (i=0; i<chrDstW; i++) {
- int u=(chrUSrc[i]+64)>>7;
- int v=(chrVSrc[i]+64)>>7;
+ int u=(chrUSrc[i]+chrDither[i&7])>>7;
+ int v=(chrVSrc[i]+chrDither[(i+3)&7])>>7;
uDest[i]= av_clip_uint8(u);
vDest[i]= av_clip_uint8(v);
}
if (CONFIG_SWSCALE_ALPHA && aDest)
for (i=0; i<dstW; i++) {
- int val= (alpSrc[i]+64)>>7;
+ int val= (alpSrc[i]+lumDither[i&7])>>7;
aDest[i]= av_clip_uint8(val);
}
}
const int16_t *chrFilter, const int16_t **chrUSrc,
const int16_t **chrVSrc, int chrFilterSize,
const int16_t **alpSrc, uint8_t *dest[4],
- int dstW, int chrDstW)
+ int dstW, int chrDstW,
+ const uint8_t *lumDither, const uint8_t *chrDither)
{
uint8_t *yDest = dest[0], *uDest = dest[1];
enum PixelFormat dstFormat = c->dstFormat;
//FIXME Optimize (just quickly written not optimized..)
int i;
for (i=0; i<dstW; i++) {
- int val=1<<18;
+ int val = lumDither[i&7]<<12;
int j;
for (j=0; j<lumFilterSize; j++)
val += lumSrc[j][i] * lumFilter[j];
if (dstFormat == PIX_FMT_NV12)
for (i=0; i<chrDstW; i++) {
- int u=1<<18;
- int v=1<<18;
+ int u = chrDither[i&7]<<12;
+ int v = chrDither[(i+3)&7]<<12;
int j;
for (j=0; j<chrFilterSize; j++) {
u += chrUSrc[j][i] * chrFilter[j];
}
else
for (i=0; i<chrDstW; i++) {
- int u=1<<18;
- int v=1<<18;
+ int u = chrDither[i&7]<<12;
+ int v = chrDither[(i+3)&7]<<12;
int j;
for (j=0; j<chrFilterSize; j++) {
u += chrUSrc[j][i] * chrFilter[j];
static av_always_inline void
yuv2gray16_X_c_template(SwsContext *c, const int16_t *lumFilter,
- const int32_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int32_t **chrUSrc,
- const int32_t **chrVSrc, int chrFilterSize,
- const int32_t **alpSrc, uint16_t *dest, int dstW,
+ const int16_t **lumSrc, int lumFilterSize,
+ const int16_t *chrFilter, const int16_t **chrUSrc,
+ const int16_t **chrVSrc, int chrFilterSize,
+ const int16_t **alpSrc, uint8_t *dest, int dstW,
int y, enum PixelFormat target)
{
int i;
for (i = 0; i < (dstW >> 1); i++) {
int j;
- int Y1 = 1 << 14;
- int Y2 = 1 << 14;
+ int Y1 = 1 << 18;
+ int Y2 = 1 << 18;
+ const int i2 = 2 * i;
for (j = 0; j < lumFilterSize; j++) {
- Y1 += lumSrc[j][i * 2] * lumFilter[j];
- Y2 += lumSrc[j][i * 2 + 1] * lumFilter[j];
+ Y1 += lumSrc[j][i2] * lumFilter[j];
+ Y2 += lumSrc[j][i2+1] * lumFilter[j];
}
- Y1 >>= 15;
- Y2 >>= 15;
+ Y1 >>= 11;
+ Y2 >>= 11;
if ((Y1 | Y2) & 0x10000) {
Y1 = av_clip_uint16(Y1);
Y2 = av_clip_uint16(Y2);
}
- output_pixel(&dest[i * 2 + 0], Y1);
- output_pixel(&dest[i * 2 + 1], Y2);
+ output_pixel(&dest[2 * i2 + 0], Y1);
+ output_pixel(&dest[2 * i2 + 2], Y2);
}
}
static av_always_inline void
-yuv2gray16_2_c_template(SwsContext *c, const int32_t *buf[2],
- const int32_t *ubuf[2], const int32_t *vbuf[2],
- const int32_t *abuf[2], uint16_t *dest, int dstW,
+yuv2gray16_2_c_template(SwsContext *c, const int16_t *buf[2],
+ const int16_t *ubuf[2], const int16_t *vbuf[2],
+ const int16_t *abuf[2], uint8_t *dest, int dstW,
int yalpha, int uvalpha, int y,
enum PixelFormat target)
{
int yalpha1 = 4095 - yalpha;
int i;
- const int32_t *buf0 = buf[0], *buf1 = buf[1];
+ const int16_t *buf0 = buf[0], *buf1 = buf[1];
for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = (buf0[i * 2 ] * yalpha1 + buf1[i * 2 ] * yalpha) >> 15;
- int Y2 = (buf0[i * 2 + 1] * yalpha1 + buf1[i * 2 + 1] * yalpha) >> 15;
+ const int i2 = 2 * i;
+ int Y1 = (buf0[i2 ] * yalpha1 + buf1[i2 ] * yalpha) >> 11;
+ int Y2 = (buf0[i2+1] * yalpha1 + buf1[i2+1] * yalpha) >> 11;
- output_pixel(&dest[i * 2 + 0], Y1);
- output_pixel(&dest[i * 2 + 1], Y2);
+ output_pixel(&dest[2 * i2 + 0], Y1);
+ output_pixel(&dest[2 * i2 + 2], Y2);
}
}
static av_always_inline void
-yuv2gray16_1_c_template(SwsContext *c, const int32_t *buf0,
- const int32_t *ubuf[2], const int32_t *vbuf[2],
- const int32_t *abuf0, uint16_t *dest, int dstW,
+yuv2gray16_1_c_template(SwsContext *c, const int16_t *buf0,
+ const int16_t *ubuf[2], const int16_t *vbuf[2],
+ const int16_t *abuf0, uint8_t *dest, int dstW,
int uvalpha, int y, enum PixelFormat target)
{
int i;
for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = buf0[i * 2 ] << 1;
- int Y2 = buf0[i * 2 + 1] << 1;
+ const int i2 = 2 * i;
+ int Y1 = buf0[i2 ] << 1;
+ int Y2 = buf0[i2+1] << 1;
- output_pixel(&dest[i * 2 + 0], Y1);
- output_pixel(&dest[i * 2 + 1], Y2);
+ output_pixel(&dest[2 * i2 + 0], Y1);
+ output_pixel(&dest[2 * i2 + 2], Y2);
}
}
#undef output_pixel
-#define YUV2PACKED16WRAPPER(name, base, ext, fmt) \
+#define YUV2PACKEDWRAPPER(name, base, ext, fmt) \
static void name ## ext ## _X_c(SwsContext *c, const int16_t *lumFilter, \
- const int16_t **_lumSrc, int lumFilterSize, \
- const int16_t *chrFilter, const int16_t **_chrUSrc, \
- const int16_t **_chrVSrc, int chrFilterSize, \
- const int16_t **_alpSrc, uint8_t *_dest, int dstW, \
+ const int16_t **lumSrc, int lumFilterSize, \
+ const int16_t *chrFilter, const int16_t **chrUSrc, \
+ const int16_t **chrVSrc, int chrFilterSize, \
+ const int16_t **alpSrc, uint8_t *dest, int dstW, \
int y) \
{ \
- const int32_t **lumSrc = (const int32_t **) _lumSrc, \
- **chrUSrc = (const int32_t **) _chrUSrc, \
- **chrVSrc = (const int32_t **) _chrVSrc, \
- **alpSrc = (const int32_t **) _alpSrc; \
- uint16_t *dest = (uint16_t *) _dest; \
name ## base ## _X_c_template(c, lumFilter, lumSrc, lumFilterSize, \
chrFilter, chrUSrc, chrVSrc, chrFilterSize, \
alpSrc, dest, dstW, y, fmt); \
} \
\
-static void name ## ext ## _2_c(SwsContext *c, const int16_t *_buf[2], \
- const int16_t *_ubuf[2], const int16_t *_vbuf[2], \
- const int16_t *_abuf[2], uint8_t *_dest, int dstW, \
+static void name ## ext ## _2_c(SwsContext *c, const int16_t *buf[2], \
+ const int16_t *ubuf[2], const int16_t *vbuf[2], \
+ const int16_t *abuf[2], uint8_t *dest, int dstW, \
int yalpha, int uvalpha, int y) \
{ \
- const int32_t **buf = (const int32_t **) _buf, \
- **ubuf = (const int32_t **) _ubuf, \
- **vbuf = (const int32_t **) _vbuf, \
- **abuf = (const int32_t **) _abuf; \
- uint16_t *dest = (uint16_t *) _dest; \
name ## base ## _2_c_template(c, buf, ubuf, vbuf, abuf, \
dest, dstW, yalpha, uvalpha, y, fmt); \
} \
\
-static void name ## ext ## _1_c(SwsContext *c, const int16_t *_buf0, \
- const int16_t *_ubuf[2], const int16_t *_vbuf[2], \
- const int16_t *_abuf0, uint8_t *_dest, int dstW, \
+static void name ## ext ## _1_c(SwsContext *c, const int16_t *buf0, \
+ const int16_t *ubuf[2], const int16_t *vbuf[2], \
+ const int16_t *abuf0, uint8_t *dest, int dstW, \
int uvalpha, int y) \
{ \
- const int32_t *buf0 = (const int32_t *) _buf0, \
- **ubuf = (const int32_t **) _ubuf, \
- **vbuf = (const int32_t **) _vbuf, \
- *abuf0 = (const int32_t *) _abuf0; \
- uint16_t *dest = (uint16_t *) _dest; \
name ## base ## _1_c_template(c, buf0, ubuf, vbuf, abuf0, dest, \
dstW, uvalpha, y, fmt); \
}
-YUV2PACKED16WRAPPER(yuv2gray16,, LE, PIX_FMT_GRAY16LE);
-YUV2PACKED16WRAPPER(yuv2gray16,, BE, PIX_FMT_GRAY16BE);
+YUV2PACKEDWRAPPER(yuv2gray16,, LE, PIX_FMT_GRAY16LE);
+YUV2PACKEDWRAPPER(yuv2gray16,, BE, PIX_FMT_GRAY16BE);
#define output_pixel(pos, acc) \
if (target == PIX_FMT_MONOBLACK) { \
#undef output_pixel
-#define YUV2PACKEDWRAPPER(name, base, ext, fmt) \
-static void name ## ext ## _X_c(SwsContext *c, const int16_t *lumFilter, \
- const int16_t **lumSrc, int lumFilterSize, \
- const int16_t *chrFilter, const int16_t **chrUSrc, \
- const int16_t **chrVSrc, int chrFilterSize, \
- const int16_t **alpSrc, uint8_t *dest, int dstW, \
- int y) \
-{ \
- name ## base ## _X_c_template(c, lumFilter, lumSrc, lumFilterSize, \
- chrFilter, chrUSrc, chrVSrc, chrFilterSize, \
- alpSrc, dest, dstW, y, fmt); \
-} \
- \
-static void name ## ext ## _2_c(SwsContext *c, const int16_t *buf[2], \
- const int16_t *ubuf[2], const int16_t *vbuf[2], \
- const int16_t *abuf[2], uint8_t *dest, int dstW, \
- int yalpha, int uvalpha, int y) \
-{ \
- name ## base ## _2_c_template(c, buf, ubuf, vbuf, abuf, \
- dest, dstW, yalpha, uvalpha, y, fmt); \
-} \
- \
-static void name ## ext ## _1_c(SwsContext *c, const int16_t *buf0, \
- const int16_t *ubuf[2], const int16_t *vbuf[2], \
- const int16_t *abuf0, uint8_t *dest, int dstW, \
- int uvalpha, int y) \
-{ \
- name ## base ## _1_c_template(c, buf0, ubuf, vbuf, \
- abuf0, dest, dstW, uvalpha, \
- y, fmt); \
-}
-
YUV2PACKEDWRAPPER(yuv2mono,, white, PIX_FMT_MONOWHITE);
YUV2PACKEDWRAPPER(yuv2mono,, black, PIX_FMT_MONOBLACK);
YUV2PACKEDWRAPPER(yuv2, 422, yuyv422, PIX_FMT_YUYV422);
YUV2PACKEDWRAPPER(yuv2, 422, uyvy422, PIX_FMT_UYVY422);
-#define R_B ((target == PIX_FMT_RGB48LE || target == PIX_FMT_RGB48BE) ? R : B)
-#define B_R ((target == PIX_FMT_RGB48LE || target == PIX_FMT_RGB48BE) ? B : R)
-#define output_pixel(pos, val) \
- if (isBE(target)) { \
- AV_WB16(pos, val); \
- } else { \
- AV_WL16(pos, val); \
- }
+#define r_b ((target == PIX_FMT_RGB48LE || target == PIX_FMT_RGB48BE) ? r : b)
+#define b_r ((target == PIX_FMT_RGB48LE || target == PIX_FMT_RGB48BE) ? b : r)
static av_always_inline void
yuv2rgb48_X_c_template(SwsContext *c, const int16_t *lumFilter,
- const int32_t **lumSrc, int lumFilterSize,
- const int16_t *chrFilter, const int32_t **chrUSrc,
- const int32_t **chrVSrc, int chrFilterSize,
- const int32_t **alpSrc, uint16_t *dest, int dstW,
+ const int16_t **lumSrc, int lumFilterSize,
+ const int16_t *chrFilter, const int16_t **chrUSrc,
+ const int16_t **chrVSrc, int chrFilterSize,
+ const int16_t **alpSrc, uint8_t *dest, int dstW,
int y, enum PixelFormat target)
{
int i;
for (i = 0; i < (dstW >> 1); i++) {
int j;
- int Y1 = 0;
- int Y2 = 0;
- int U = -128 << 23; // 19
- int V = -128 << 23;
- int R, G, B;
+ int Y1 = 1 << 18;
+ int Y2 = 1 << 18;
+ int U = 1 << 18;
+ int V = 1 << 18;
+ const uint8_t *r, *g, *b;
for (j = 0; j < lumFilterSize; j++) {
Y1 += lumSrc[j][i * 2] * lumFilter[j];
U += chrUSrc[j][i] * chrFilter[j];
V += chrVSrc[j][i] * chrFilter[j];
}
+ Y1 >>= 19;
+ Y2 >>= 19;
+ U >>= 19;
+ V >>= 19;
+ if ((Y1 | Y2 | U | V) & 0x100) {
+ Y1 = av_clip_uint8(Y1);
+ Y2 = av_clip_uint8(Y2);
+ U = av_clip_uint8(U);
+ V = av_clip_uint8(V);
+ }
- // 8bit: 12+15=27; 16-bit: 12+19=31
- Y1 >>= 14; // 10
- Y2 >>= 14;
- U >>= 14;
- V >>= 14;
-
- // 8bit: 27 -> 17bit, 16bit: 31 - 14 = 17bit
- Y1 -= c->yuv2rgb_y_offset;
- Y2 -= c->yuv2rgb_y_offset;
- Y1 *= c->yuv2rgb_y_coeff;
- Y2 *= c->yuv2rgb_y_coeff;
- Y1 += 1 << 13; // 21
- Y2 += 1 << 13;
- // 8bit: 17 + 13bit = 30bit, 16bit: 17 + 13bit = 30bit
-
- R = V * c->yuv2rgb_v2r_coeff;
- G = V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
- B = U * c->yuv2rgb_u2b_coeff;
+ /* FIXME fix tables so that clipping is not needed and then use _NOCLIP*/
+ r = (const uint8_t *) c->table_rV[V];
+ g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]);
+ b = (const uint8_t *) c->table_bU[U];
- // 8bit: 30 - 22 = 8bit, 16bit: 30bit - 14 = 16bit
- output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
- output_pixel(&dest[1], av_clip_uintp2( G + Y1, 30) >> 14);
- output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
- output_pixel(&dest[3], av_clip_uintp2(R_B + Y2, 30) >> 14);
- output_pixel(&dest[4], av_clip_uintp2( G + Y2, 30) >> 14);
- output_pixel(&dest[5], av_clip_uintp2(B_R + Y2, 30) >> 14);
- dest += 6;
+ dest[ 0] = dest[ 1] = r_b[Y1];
+ dest[ 2] = dest[ 3] = g[Y1];
+ dest[ 4] = dest[ 5] = b_r[Y1];
+ dest[ 6] = dest[ 7] = r_b[Y2];
+ dest[ 8] = dest[ 9] = g[Y2];
+ dest[10] = dest[11] = b_r[Y2];
+ dest += 12;
}
}
static av_always_inline void
-yuv2rgb48_2_c_template(SwsContext *c, const int32_t *buf[2],
- const int32_t *ubuf[2], const int32_t *vbuf[2],
- const int32_t *abuf[2], uint16_t *dest, int dstW,
+yuv2rgb48_2_c_template(SwsContext *c, const int16_t *buf[2],
+ const int16_t *ubuf[2], const int16_t *vbuf[2],
+ const int16_t *abuf[2], uint8_t *dest, int dstW,
int yalpha, int uvalpha, int y,
enum PixelFormat target)
{
- const int32_t *buf0 = buf[0], *buf1 = buf[1],
+ const int16_t *buf0 = buf[0], *buf1 = buf[1],
*ubuf0 = ubuf[0], *ubuf1 = ubuf[1],
*vbuf0 = vbuf[0], *vbuf1 = vbuf[1];
int yalpha1 = 4095 - yalpha;
int i;
for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = (buf0[i * 2] * yalpha1 + buf1[i * 2] * yalpha) >> 14;
- int Y2 = (buf0[i * 2 + 1] * yalpha1 + buf1[i * 2 + 1] * yalpha) >> 14;
- int U = (ubuf0[i] * uvalpha1 + ubuf1[i] * uvalpha + (-128 << 23)) >> 14;
- int V = (vbuf0[i] * uvalpha1 + vbuf1[i] * uvalpha + (-128 << 23)) >> 14;
- int R, G, B;
-
- Y1 -= c->yuv2rgb_y_offset;
- Y2 -= c->yuv2rgb_y_offset;
- Y1 *= c->yuv2rgb_y_coeff;
- Y2 *= c->yuv2rgb_y_coeff;
- Y1 += 1 << 13;
- Y2 += 1 << 13;
-
- R = V * c->yuv2rgb_v2r_coeff;
- G = V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
- B = U * c->yuv2rgb_u2b_coeff;
+ int Y1 = (buf0[i * 2] * yalpha1 + buf1[i * 2] * yalpha) >> 19;
+ int Y2 = (buf0[i * 2 + 1] * yalpha1 + buf1[i * 2 + 1] * yalpha) >> 19;
+ int U = (ubuf0[i] * uvalpha1 + ubuf1[i] * uvalpha) >> 19;
+ int V = (vbuf0[i] * uvalpha1 + vbuf1[i] * uvalpha) >> 19;
+ const uint8_t *r = (const uint8_t *) c->table_rV[V],
+ *g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]),
+ *b = (const uint8_t *) c->table_bU[U];
- output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
- output_pixel(&dest[1], av_clip_uintp2( G + Y1, 30) >> 14);
- output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
- output_pixel(&dest[3], av_clip_uintp2(R_B + Y2, 30) >> 14);
- output_pixel(&dest[4], av_clip_uintp2( G + Y2, 30) >> 14);
- output_pixel(&dest[5], av_clip_uintp2(B_R + Y2, 30) >> 14);
- dest += 6;
+ dest[ 0] = dest[ 1] = r_b[Y1];
+ dest[ 2] = dest[ 3] = g[Y1];
+ dest[ 4] = dest[ 5] = b_r[Y1];
+ dest[ 6] = dest[ 7] = r_b[Y2];
+ dest[ 8] = dest[ 9] = g[Y2];
+ dest[10] = dest[11] = b_r[Y2];
+ dest += 12;
}
}
static av_always_inline void
-yuv2rgb48_1_c_template(SwsContext *c, const int32_t *buf0,
- const int32_t *ubuf[2], const int32_t *vbuf[2],
- const int32_t *abuf0, uint16_t *dest, int dstW,
+yuv2rgb48_1_c_template(SwsContext *c, const int16_t *buf0,
+ const int16_t *ubuf[2], const int16_t *vbuf[2],
+ const int16_t *abuf0, uint8_t *dest, int dstW,
int uvalpha, int y, enum PixelFormat target)
{
- const int32_t *ubuf0 = ubuf[0], *ubuf1 = ubuf[1],
+ const int16_t *ubuf0 = ubuf[0], *ubuf1 = ubuf[1],
*vbuf0 = vbuf[0], *vbuf1 = vbuf[1];
int i;
if (uvalpha < 2048) {
for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = (buf0[i * 2] ) >> 2;
- int Y2 = (buf0[i * 2 + 1]) >> 2;
- int U = (ubuf0[i] + (-128 << 11)) >> 2;
- int V = (vbuf0[i] + (-128 << 11)) >> 2;
- int R, G, B;
-
- Y1 -= c->yuv2rgb_y_offset;
- Y2 -= c->yuv2rgb_y_offset;
- Y1 *= c->yuv2rgb_y_coeff;
- Y2 *= c->yuv2rgb_y_coeff;
- Y1 += 1 << 13;
- Y2 += 1 << 13;
-
- R = V * c->yuv2rgb_v2r_coeff;
- G = V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
- B = U * c->yuv2rgb_u2b_coeff;
-
- output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
- output_pixel(&dest[1], av_clip_uintp2( G + Y1, 30) >> 14);
- output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
- output_pixel(&dest[3], av_clip_uintp2(R_B + Y2, 30) >> 14);
- output_pixel(&dest[4], av_clip_uintp2( G + Y2, 30) >> 14);
- output_pixel(&dest[5], av_clip_uintp2(B_R + Y2, 30) >> 14);
- dest += 6;
+ int Y1 = buf0[i * 2] >> 7;
+ int Y2 = buf0[i * 2 + 1] >> 7;
+ int U = ubuf1[i] >> 7;
+ int V = vbuf1[i] >> 7;
+ const uint8_t *r = (const uint8_t *) c->table_rV[V],
+ *g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]),
+ *b = (const uint8_t *) c->table_bU[U];
+
+ dest[ 0] = dest[ 1] = r_b[Y1];
+ dest[ 2] = dest[ 3] = g[Y1];
+ dest[ 4] = dest[ 5] = b_r[Y1];
+ dest[ 6] = dest[ 7] = r_b[Y2];
+ dest[ 8] = dest[ 9] = g[Y2];
+ dest[10] = dest[11] = b_r[Y2];
+ dest += 12;
}
} else {
for (i = 0; i < (dstW >> 1); i++) {
- int Y1 = (buf0[i * 2] ) >> 2;
- int Y2 = (buf0[i * 2 + 1]) >> 2;
- int U = (ubuf0[i] + ubuf1[i] + (-128 << 11)) >> 3;
- int V = (vbuf0[i] + vbuf1[i] + (-128 << 11)) >> 3;
- int R, G, B;
-
- Y1 -= c->yuv2rgb_y_offset;
- Y2 -= c->yuv2rgb_y_offset;
- Y1 *= c->yuv2rgb_y_coeff;
- Y2 *= c->yuv2rgb_y_coeff;
- Y1 += 1 << 13;
- Y2 += 1 << 13;
-
- R = V * c->yuv2rgb_v2r_coeff;
- G = V * c->yuv2rgb_v2g_coeff + U * c->yuv2rgb_u2g_coeff;
- B = U * c->yuv2rgb_u2b_coeff;
-
- output_pixel(&dest[0], av_clip_uintp2(R_B + Y1, 30) >> 14);
- output_pixel(&dest[1], av_clip_uintp2( G + Y1, 30) >> 14);
- output_pixel(&dest[2], av_clip_uintp2(B_R + Y1, 30) >> 14);
- output_pixel(&dest[3], av_clip_uintp2(R_B + Y2, 30) >> 14);
- output_pixel(&dest[4], av_clip_uintp2( G + Y2, 30) >> 14);
- output_pixel(&dest[5], av_clip_uintp2(B_R + Y2, 30) >> 14);
- dest += 6;
+ int Y1 = buf0[i * 2] >> 7;
+ int Y2 = buf0[i * 2 + 1] >> 7;
+ int U = (ubuf0[i] + ubuf1[i]) >> 8;
+ int V = (vbuf0[i] + vbuf1[i]) >> 8;
+ const uint8_t *r = (const uint8_t *) c->table_rV[V],
+ *g = (const uint8_t *)(c->table_gU[U] + c->table_gV[V]),
+ *b = (const uint8_t *) c->table_bU[U];
+
+ dest[ 0] = dest[ 1] = r_b[Y1];
+ dest[ 2] = dest[ 3] = g[Y1];
+ dest[ 4] = dest[ 5] = b_r[Y1];
+ dest[ 6] = dest[ 7] = r_b[Y2];
+ dest[ 8] = dest[ 9] = g[Y2];
+ dest[10] = dest[11] = b_r[Y2];
+ dest += 12;
}
}
}
-#undef output_pixel
#undef r_b
#undef b_r
-YUV2PACKED16WRAPPER(yuv2, rgb48, rgb48be, PIX_FMT_RGB48BE);
-YUV2PACKED16WRAPPER(yuv2, rgb48, rgb48le, PIX_FMT_RGB48LE);
-YUV2PACKED16WRAPPER(yuv2, rgb48, bgr48be, PIX_FMT_BGR48BE);
-YUV2PACKED16WRAPPER(yuv2, rgb48, bgr48le, PIX_FMT_BGR48LE);
+YUV2PACKEDWRAPPER(yuv2, rgb48, rgb48be, PIX_FMT_RGB48BE);
+//YUV2PACKEDWRAPPER(yuv2, rgb48, rgb48le, PIX_FMT_RGB48LE);
+YUV2PACKEDWRAPPER(yuv2, rgb48, bgr48be, PIX_FMT_BGR48BE);
+//YUV2PACKEDWRAPPER(yuv2, rgb48, bgr48le, PIX_FMT_BGR48LE);
static av_always_inline void
yuv2rgb_write(uint8_t *_dest, int i, int Y1, int Y2,
U = av_clip_uint8(U);
V = av_clip_uint8(V);
}
- if (hasAlpha) {
+ if (hasAlpha) {\
A1 = 1 << 18;
A2 = 1 << 18;
for (j = 0; j < lumFilterSize; j++) {
for (i = 0; i < dstW; i++) {
int j;
- int Y = 0;
- int U = -128 << 19;
- int V = -128 << 19;
+ int Y = 1<<9;
+ int U = (1<<9)-(128 << 19);
+ int V = (1<<9)-(128 << 19);
int av_unused A;
int R, G, B;
U >>= 10;
V >>= 10;
if (hasAlpha) {
- A = 1 << 21;
+ A = 1 << 18;
for (j = 0; j < lumFilterSize; j++) {
A += alpSrc[j][i] * lumFilter[j];
}
dest[1] = B >> 22;
dest[2] = G >> 22;
dest[3] = R >> 22;
- dest += 4;
break;
case PIX_FMT_BGR24:
dest[0] = B >> 22;
#define b ((origin == PIX_FMT_BGR48BE || origin == PIX_FMT_BGR48LE) ? r_b : b_r)
static av_always_inline void
-rgb48ToY_c_template(uint16_t *dst, const uint16_t *src, int width,
+rgb48ToY_c_template(int16_t *dst, const uint16_t *src, int width,
enum PixelFormat origin)
{
int i;
for (i = 0; i < width; i++) {
- unsigned int r_b = input_pixel(&src[i*3+0]);
- unsigned int g = input_pixel(&src[i*3+1]);
- unsigned int b_r = input_pixel(&src[i*3+2]);
+ int r_b = input_pixel(&src[i*3+0]);
+ int g = input_pixel(&src[i*3+1]);
+ int b_r = input_pixel(&src[i*3+2]);
- dst[i] = (RY*r + GY*g + BY*b + (0x2001<<(RGB2YUV_SHIFT-1))) >> RGB2YUV_SHIFT;
+ dst[i] = (RY*r + GY*g + BY*b + (32<<(RGB2YUV_SHIFT-1+8)) + (1<<(RGB2YUV_SHIFT-7+8))) >> (RGB2YUV_SHIFT-6+8);
}
}
static av_always_inline void
-rgb48ToUV_c_template(uint16_t *dstU, uint16_t *dstV,
+rgb48ToUV_c_template(int16_t *dstU, int16_t *dstV,
const uint16_t *src1, const uint16_t *src2,
int width, enum PixelFormat origin)
{
int g = input_pixel(&src1[i*3+1]);
int b_r = input_pixel(&src1[i*3+2]);
- dstU[i] = (RU*r + GU*g + BU*b + (0x10001<<(RGB2YUV_SHIFT-1))) >> RGB2YUV_SHIFT;
- dstV[i] = (RV*r + GV*g + BV*b + (0x10001<<(RGB2YUV_SHIFT-1))) >> RGB2YUV_SHIFT;
+ dstU[i] = (RU*r + GU*g + BU*b + (256<<(RGB2YUV_SHIFT-1+8)) + (1<<(RGB2YUV_SHIFT-7+8))) >> (RGB2YUV_SHIFT-6+8);
+ dstV[i] = (RV*r + GV*g + BV*b + (256<<(RGB2YUV_SHIFT-1+8)) + (1<<(RGB2YUV_SHIFT-7+8))) >> (RGB2YUV_SHIFT-6+8);
}
}
static av_always_inline void
-rgb48ToUV_half_c_template(uint16_t *dstU, uint16_t *dstV,
+rgb48ToUV_half_c_template(int16_t *dstU, int16_t *dstV,
const uint16_t *src1, const uint16_t *src2,
int width, enum PixelFormat origin)
{
int i;
assert(src1==src2);
for (i = 0; i < width; i++) {
- int r_b = (input_pixel(&src1[6 * i + 0]) + input_pixel(&src1[6 * i + 3]) + 1) >> 1;
- int g = (input_pixel(&src1[6 * i + 1]) + input_pixel(&src1[6 * i + 4]) + 1) >> 1;
- int b_r = (input_pixel(&src1[6 * i + 2]) + input_pixel(&src1[6 * i + 5]) + 1) >> 1;
+ int r_b = (input_pixel(&src1[6*i + 0])) + (input_pixel(&src1[6*i + 3]));
+ int g = (input_pixel(&src1[6*i + 1])) + (input_pixel(&src1[6*i + 4]));
+ int b_r = (input_pixel(&src1[6*i + 2])) + (input_pixel(&src1[6*i + 5]));
- dstU[i]= (RU*r + GU*g + BU*b + (0x10001<<(RGB2YUV_SHIFT-1))) >> RGB2YUV_SHIFT;
- dstV[i]= (RV*r + GV*g + BV*b + (0x10001<<(RGB2YUV_SHIFT-1))) >> RGB2YUV_SHIFT;
+ dstU[i]= (RU*r + GU*g + BU*b + (256U<<(RGB2YUV_SHIFT+8)) + (1<<(RGB2YUV_SHIFT-6+8))) >> (RGB2YUV_SHIFT-5+8);
+ dstV[i]= (RV*r + GV*g + BV*b + (256U<<(RGB2YUV_SHIFT+8)) + (1<<(RGB2YUV_SHIFT-6+8))) >> (RGB2YUV_SHIFT-5+8);
}
}
#undef input_pixel
#define rgb48funcs(pattern, BE_LE, origin) \
-static void pattern ## 48 ## BE_LE ## ToY_c(uint8_t *_dst, const uint8_t *_src, \
+static void pattern ## 48 ## BE_LE ## ToY_c(uint8_t *dst, const uint8_t *src, \
int width, uint32_t *unused) \
{ \
- const uint16_t *src = (const uint16_t *) _src; \
- uint16_t *dst = (uint16_t *) _dst; \
rgb48ToY_c_template(dst, src, width, origin); \
} \
\
-static void pattern ## 48 ## BE_LE ## ToUV_c(uint8_t *_dstU, uint8_t *_dstV, \
- const uint8_t *_src1, const uint8_t *_src2, \
+static void pattern ## 48 ## BE_LE ## ToUV_c(uint8_t *dstU, uint8_t *dstV, \
+ const uint8_t *src1, const uint8_t *src2, \
int width, uint32_t *unused) \
{ \
- const uint16_t *src1 = (const uint16_t *) _src1, \
- *src2 = (const uint16_t *) _src2; \
- uint16_t *dstU = (uint16_t *) _dstU, *dstV = (uint16_t *) _dstV; \
rgb48ToUV_c_template(dstU, dstV, src1, src2, width, origin); \
} \
\
-static void pattern ## 48 ## BE_LE ## ToUV_half_c(uint8_t *_dstU, uint8_t *_dstV, \
- const uint8_t *_src1, const uint8_t *_src2, \
+static void pattern ## 48 ## BE_LE ## ToUV_half_c(uint8_t *dstU, uint8_t *dstV, \
+ const uint8_t *src1, const uint8_t *src2, \
int width, uint32_t *unused) \
{ \
- const uint16_t *src1 = (const uint16_t *) _src1, \
- *src2 = (const uint16_t *) _src2; \
- uint16_t *dstU = (uint16_t *) _dstU, *dstV = (uint16_t *) _dstV; \
rgb48ToUV_half_c_template(dstU, dstV, src1, src2, width, origin); \
}
(isBE(origin) ? AV_RB16(&src[(i)*2]) : AV_RL16(&src[(i)*2])))
static av_always_inline void
-rgb16_32ToY_c_template(uint8_t *dst, const uint8_t *src,
+rgb16_32ToY_c_template(int16_t *dst, const uint8_t *src,
int width, enum PixelFormat origin,
int shr, int shg, int shb, int shp,
int maskr, int maskg, int maskb,
int rsh, int gsh, int bsh, int S)
{
const int ry = RY << rsh, gy = GY << gsh, by = BY << bsh,
- rnd = 33 << (S - 1);
+ rnd = (32<<((S)-1)) + (1<<(S-7));
int i;
for (i = 0; i < width; i++) {
int g = (px & maskg) >> shg;
int r = (px & maskr) >> shr;
- dst[i] = (ry * r + gy * g + by * b + rnd) >> S;
+ dst[i] = (ry * r + gy * g + by * b + rnd) >> ((S)-6);
}
}
static av_always_inline void
-rgb16_32ToUV_c_template(uint8_t *dstU, uint8_t *dstV,
+rgb16_32ToUV_c_template(int16_t *dstU, int16_t *dstV,
const uint8_t *src, int width,
enum PixelFormat origin,
int shr, int shg, int shb, int shp,
{
const int ru = RU << rsh, gu = GU << gsh, bu = BU << bsh,
rv = RV << rsh, gv = GV << gsh, bv = BV << bsh,
- rnd = 257 << (S - 1);
+ rnd = (256<<((S)-1)) + (1<<(S-7));
int i;
for (i = 0; i < width; i++) {
int g = (px & maskg) >> shg;
int r = (px & maskr) >> shr;
- dstU[i] = (ru * r + gu * g + bu * b + rnd) >> S;
- dstV[i] = (rv * r + gv * g + bv * b + rnd) >> S;
+ dstU[i] = (ru * r + gu * g + bu * b + rnd) >> ((S)-6);
+ dstV[i] = (rv * r + gv * g + bv * b + rnd) >> ((S)-6);
}
}
static av_always_inline void
-rgb16_32ToUV_half_c_template(uint8_t *dstU, uint8_t *dstV,
+rgb16_32ToUV_half_c_template(int16_t *dstU, int16_t *dstV,
const uint8_t *src, int width,
enum PixelFormat origin,
int shr, int shg, int shb, int shp,
{
const int ru = RU << rsh, gu = GU << gsh, bu = BU << bsh,
rv = RV << rsh, gv = GV << gsh, bv = BV << bsh,
- rnd = 257 << S, maskgx = ~(maskr | maskb);
+ rnd = (256U<<(S)) + (1<<(S-6)), maskgx = ~(maskr | maskb);
int i;
maskr |= maskr << 1; maskb |= maskb << 1; maskg |= maskg << 1;
}
r = (rb & maskr) >> shr;
- dstU[i] = (ru * r + gu * g + bu * b + rnd) >> (S + 1);
- dstV[i] = (rv * r + gv * g + bv * b + rnd) >> (S + 1);
+ dstU[i] = (ru * r + gu * g + bu * b + (unsigned)rnd) >> ((S)-6+1);
+ dstV[i] = (rv * r + gv * g + bv * b + (unsigned)rnd) >> ((S)-6+1);
}
}
rgb16_32_wrapper(PIX_FMT_RGB565BE, rgb16be, 0, 0, 0, 0, 0xF800, 0x07E0, 0x001F, 0, 5, 11, RGB2YUV_SHIFT+8);
rgb16_32_wrapper(PIX_FMT_RGB555BE, rgb15be, 0, 0, 0, 0, 0x7C00, 0x03E0, 0x001F, 0, 5, 10, RGB2YUV_SHIFT+7);
-static void abgrToA_c(uint8_t *dst, const uint8_t *src, int width, uint32_t *unused)
+static void abgrToA_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
{
int i;
for (i=0; i<width; i++) {
- dst[i]= src[4*i];
+ dst[i]= src[4*i]<<6;
}
}
-static void rgbaToA_c(uint8_t *dst, const uint8_t *src, int width, uint32_t *unused)
+static void rgbaToA_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
{
int i;
for (i=0; i<width; i++) {
- dst[i]= src[4*i+3];
+ dst[i]= src[4*i+3]<<6;
}
}
-static void palToY_c(uint8_t *dst, const uint8_t *src, int width, uint32_t *pal)
+static void palToA_c(int16_t *dst, const uint8_t *src, int width, uint32_t *pal)
{
int i;
for (i=0; i<width; i++) {
int d= src[i];
- dst[i]= pal[d] & 0xFF;
+ dst[i]= (pal[d] >> 24)<<6;
}
}
-static void palToUV_c(uint8_t *dstU, uint8_t *dstV,
- const uint8_t *src1, const uint8_t *src2,
- int width, uint32_t *pal)
+static void palToY_c(int16_t *dst, const uint8_t *src, long width, uint32_t *pal)
+{
+ int i;
+ for (i=0; i<width; i++) {
+ int d= src[i];
+
+ dst[i]= (pal[d] & 0xFF)<<6;
+ }
+}
+
+static void palToUV_c(uint16_t *dstU, int16_t *dstV,
+ const uint8_t *src1, const uint8_t *src2,
+ int width, uint32_t *pal)
{
int i;
assert(src1 == src2);
for (i=0; i<width; i++) {
int p= pal[src1[i]];
- dstU[i]= p>>8;
- dstV[i]= p>>16;
+ dstU[i]= (uint8_t)(p>> 8)<<6;
+ dstV[i]= (uint8_t)(p>>16)<<6;
}
}
-static void monowhite2Y_c(uint8_t *dst, const uint8_t *src,
- int width, uint32_t *unused)
+static void monowhite2Y_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
{
int i, j;
for (i=0; i<width/8; i++) {
int d= ~src[i];
for(j=0; j<8; j++)
- dst[8*i+j]= ((d>>(7-j))&1)*255;
+ dst[8*i+j]= ((d>>(7-j))&1)*16383;
+ }
+ if(width&7){
+ int d= ~src[i];
+ for(j=0; j<(width&7); j++)
+ dst[8*i+j]= ((d>>(7-j))&1)*16383;
}
}
-static void monoblack2Y_c(uint8_t *dst, const uint8_t *src,
- int width, uint32_t *unused)
+static void monoblack2Y_c(int16_t *dst, const uint8_t *src, int width, uint32_t *unused)
{
int i, j;
for (i=0; i<width/8; i++) {
int d= src[i];
for(j=0; j<8; j++)
- dst[8*i+j]= ((d>>(7-j))&1)*255;
+ dst[8*i+j]= ((d>>(7-j))&1)*16383;
+ }
+ if(width&7){
+ int d= src[i];
+ for(j=0; j<(width&7); j++)
+ dst[8*i+j]= ((d>>(7-j))&1)*16383;
}
}
assert(src1 == src2);
}
-static void bswap16Y_c(uint8_t *_dst, const uint8_t *_src, int width, uint32_t *unused)
-{
- int i;
- const uint16_t *src = (const uint16_t *) _src;
- uint16_t *dst = (uint16_t *) _dst;
- for (i=0; i<width; i++) {
- dst[i] = av_bswap16(src[i]);
- }
-}
-
-static void bswap16UV_c(uint8_t *_dstU, uint8_t *_dstV, const uint8_t *_src1,
- const uint8_t *_src2, int width, uint32_t *unused)
+static void LEToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
+ const uint8_t *src2, int width, uint32_t *unused)
{
int i;
- const uint16_t *src1 = (const uint16_t *) _src1,
- *src2 = (const uint16_t *) _src2;
- uint16_t *dstU = (uint16_t *) _dstU, *dstV = (uint16_t *) _dstV;
for (i=0; i<width; i++) {
- dstU[i] = av_bswap16(src1[i]);
- dstV[i] = av_bswap16(src2[i]);
+ dstU[i]= src1[2*i + 1];
+ dstV[i]= src2[2*i + 1];
}
}
assert(src1 == src2);
}
+static void BEToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
+ const uint8_t *src2, int width, uint32_t *unused)
+{
+ int i;
+ for (i=0; i<width; i++) {
+ dstU[i]= src1[2*i];
+ dstV[i]= src2[2*i];
+ }
+}
+
static av_always_inline void nvXXtoUV_c(uint8_t *dst1, uint8_t *dst2,
const uint8_t *src, int width)
{
#define input_pixel(pos) (isBE(origin) ? AV_RB16(pos) : AV_RL16(pos))
-static void bgr24ToY_c(uint8_t *dst, const uint8_t *src,
+static void bgr24ToY_c(int16_t *dst, const uint8_t *src,
int width, uint32_t *unused)
{
int i;
int g= src[i*3+1];
int r= src[i*3+2];
- dst[i]= ((RY*r + GY*g + BY*b + (33<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
+ dst[i]= ((RY*r + GY*g + BY*b + (32<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6));
}
}
-static void bgr24ToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
+static void bgr24ToUV_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
const uint8_t *src2, int width, uint32_t *unused)
{
int i;
int g= src1[3*i + 1];
int r= src1[3*i + 2];
- dstU[i]= (RU*r + GU*g + BU*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
- dstV[i]= (RV*r + GV*g + BV*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
+ dstU[i]= (RU*r + GU*g + BU*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
+ dstV[i]= (RV*r + GV*g + BV*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
}
assert(src1 == src2);
}
-static void bgr24ToUV_half_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
+static void bgr24ToUV_half_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
const uint8_t *src2, int width, uint32_t *unused)
{
int i;
int g= src1[6*i + 1] + src1[6*i + 4];
int r= src1[6*i + 2] + src1[6*i + 5];
- dstU[i]= (RU*r + GU*g + BU*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
- dstV[i]= (RV*r + GV*g + BV*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
+ dstU[i]= (RU*r + GU*g + BU*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
+ dstV[i]= (RV*r + GV*g + BV*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
}
assert(src1 == src2);
}
-static void rgb24ToY_c(uint8_t *dst, const uint8_t *src, int width,
+static void rgb24ToY_c(int16_t *dst, const uint8_t *src, int width,
uint32_t *unused)
{
int i;
int g= src[i*3+1];
int b= src[i*3+2];
- dst[i]= ((RY*r + GY*g + BY*b + (33<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
+ dst[i]= ((RY*r + GY*g + BY*b + (32<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6));
}
}
-static void rgb24ToUV_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
+static void rgb24ToUV_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
const uint8_t *src2, int width, uint32_t *unused)
{
int i;
int g= src1[3*i + 1];
int b= src1[3*i + 2];
- dstU[i]= (RU*r + GU*g + BU*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
- dstV[i]= (RV*r + GV*g + BV*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT;
+ dstU[i]= (RU*r + GU*g + BU*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
+ dstV[i]= (RV*r + GV*g + BV*b + (256<<(RGB2YUV_SHIFT-1)) + (1<<(RGB2YUV_SHIFT-7)))>>(RGB2YUV_SHIFT-6);
}
}
-static void rgb24ToUV_half_c(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1,
- const uint8_t *src2, int width, uint32_t *unused)
+static void rgb24ToUV_half_c(int16_t *dstU, int16_t *dstV, const uint8_t *src1,
+ const uint8_t *src2, int width, uint32_t *unused)
{
int i;
assert(src1==src2);
int g= src1[6*i + 1] + src1[6*i + 4];
int b= src1[6*i + 2] + src1[6*i + 5];
- dstU[i]= (RU*r + GU*g + BU*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
- dstV[i]= (RV*r + GV*g + BV*b + (257<<RGB2YUV_SHIFT))>>(RGB2YUV_SHIFT+1);
- }
-}
-
-static void hScale16_c(SwsContext *c, int16_t *_dst, int dstW, const uint8_t *_src,
- const int16_t *filter,
- const int16_t *filterPos, int filterSize)
-{
- int i;
- int32_t *dst = (int32_t *) _dst;
- const uint16_t *src = (const uint16_t *) _src;
- int bits = av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1;
- int sh = (bits <= 7) ? 11 : (bits - 4);
-
- for (i = 0; i < dstW; i++) {
- int j;
- int srcPos = filterPos[i];
- unsigned int val = 0;
-
- for (j = 0; j < filterSize; j++) {
- val += src[srcPos + j] * filter[filterSize * i + j];
- }
- // filter=14 bit, input=16 bit, output=30 bit, >> 11 makes 19 bit
- dst[i] = FFMIN(val >> sh, (1 << 19) - 1);
+ dstU[i]= (RU*r + GU*g + BU*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
+ dstV[i]= (RV*r + GV*g + BV*b + (256<<RGB2YUV_SHIFT) + (1<<(RGB2YUV_SHIFT-6)))>>(RGB2YUV_SHIFT-5);
}
}
// bilinear / bicubic scaling
-static void hScale_c(SwsContext *c, int16_t *dst, int dstW, const uint8_t *src,
+static void hScale_c(int16_t *dst, int dstW, const uint8_t *src,
const int16_t *filter, const int16_t *filterPos,
int filterSize)
{
}
}
+static inline void hScale16_c(int16_t *dst, int dstW, const uint16_t *src, int srcW, int xInc,
+ const int16_t *filter, const int16_t *filterPos, long filterSize, int shift)
+{
+ int i, j;
+
+ for (i=0; i<dstW; i++) {
+ int srcPos= filterPos[i];
+ int val=0;
+ for (j=0; j<filterSize; j++) {
+ val += ((int)src[srcPos + j])*filter[filterSize*i + j];
+ }
+ dst[i] = FFMIN(val>>shift, (1<<15)-1); // the cubic equation does overflow ...
+ }
+}
+
+static inline void hScale16X_c(int16_t *dst, int dstW, const uint16_t *src, int srcW, int xInc,
+ const int16_t *filter, const int16_t *filterPos, long filterSize, int shift)
+{
+ int i, j;
+ for (i=0; i<dstW; i++) {
+ int srcPos= filterPos[i];
+ int val=0;
+ for (j=0; j<filterSize; j++) {
+ val += ((int)av_bswap16(src[srcPos + j]))*filter[filterSize*i + j];
+ }
+ dst[i] = FFMIN(val>>shift, (1<<15)-1); // the cubic equation does overflow ...
+ }
+}
+
//FIXME all pal and rgb srcFormats could do this convertion as well
//FIXME all scalers more complex than bilinear could do half of this transform
static void chrRangeToJpeg_c(int16_t *dstU, int16_t *dstV, int width)
dst[i] = (dst[i]*14071 + 33561947)>>14;
}
-static void chrRangeToJpeg16_c(int16_t *_dstU, int16_t *_dstV, int width)
-{
- int i;
- int32_t *dstU = (int32_t *) _dstU;
- int32_t *dstV = (int32_t *) _dstV;
- for (i = 0; i < width; i++) {
- dstU[i] = (FFMIN(dstU[i],30775<<4)*4663 - (9289992<<4))>>12; //-264
- dstV[i] = (FFMIN(dstV[i],30775<<4)*4663 - (9289992<<4))>>12; //-264
- }
-}
-static void chrRangeFromJpeg16_c(int16_t *_dstU, int16_t *_dstV, int width)
-{
- int i;
- int32_t *dstU = (int32_t *) _dstU;
- int32_t *dstV = (int32_t *) _dstV;
- for (i = 0; i < width; i++) {
- dstU[i] = (dstU[i]*1799 + (4081085<<4))>>11; //1469
- dstV[i] = (dstV[i]*1799 + (4081085<<4))>>11; //1469
- }
-}
-static void lumRangeToJpeg16_c(int16_t *_dst, int width)
-{
- int i;
- int32_t *dst = (int32_t *) _dst;
- for (i = 0; i < width; i++)
- dst[i] = (FFMIN(dst[i],30189<<4)*4769 - (39057361<<2))>>12;
-}
-static void lumRangeFromJpeg16_c(int16_t *_dst, int width)
-{
- int i;
- int32_t *dst = (int32_t *) _dst;
- for (i = 0; i < width; i++)
- dst[i] = (dst[i]*14071 + (33561947<<4))>>14;
-}
-
static void hyscale_fast_c(SwsContext *c, int16_t *dst, int dstWidth,
const uint8_t *src, int srcW, int xInc)
{
dst[i]= (src[xx]<<7) + (src[xx+1] - src[xx])*xalpha;
xpos+=xInc;
}
-}
-
-static void scale8To16Rv_c(uint16_t *_dst, const uint8_t *src, int len)
-{
- int i;
- uint8_t *dst = (uint8_t *) _dst;
- for (i = len - 1; i >= 0; i--) {
- dst[i * 2] = dst[i * 2 + 1] = src[i];
- }
-}
-
-static void scale19To15Fw_c(int16_t *dst, const int32_t *src, int len)
-{
- int i;
- for (i = 0; i < len; i++) {
- dst[i] = src[i] >> 4;
- }
+ for (i=dstWidth-1; (i*xInc)>>16 >=srcW-1; i--)
+ dst[i] = src[srcW-1]*128;
}
// *** horizontal scale Y line to temp buffer
-static av_always_inline void hyscale(SwsContext *c, int16_t *dst, int dstWidth,
+static av_always_inline void hyscale(SwsContext *c, uint16_t *dst, int dstWidth,
const uint8_t *src, int srcW, int xInc,
const int16_t *hLumFilter,
const int16_t *hLumFilterPos, int hLumFilterSize,
src= formatConvBuffer;
}
- if (av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1 < 8 && c->scalingBpp == 16) {
- c->scale8To16Rv((uint16_t *) formatConvBuffer, src, srcW);
- src = formatConvBuffer;
- }
-
- if (!c->hyscale_fast) {
- c->hScale(c, dst, dstWidth, src, hLumFilter, hLumFilterPos, hLumFilterSize);
+ if (c->hScale16) {
+ int shift= isAnyRGB(c->srcFormat) || c->srcFormat==PIX_FMT_PAL8 ? 13 : av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1;
+ c->hScale16(dst, dstWidth, (const uint16_t*)src, srcW, xInc, hLumFilter, hLumFilterPos, hLumFilterSize, shift);
+ } else if (!c->hyscale_fast) {
+ c->hScale(dst, dstWidth, src, hLumFilter, hLumFilterPos, hLumFilterSize);
} else { // fast bilinear upscale / crap downscale
c->hyscale_fast(c, dst, dstWidth, src, srcW, xInc);
}
if (convertRange)
convertRange(dst, dstWidth);
-
- if (av_pix_fmt_descriptors[c->dstFormat].comp[0].depth_minus1 < 8 && c->scalingBpp == 16) {
- c->scale19To15Fw(dst, (int32_t *) dst, dstWidth);
- }
}
static void hcscale_fast_c(SwsContext *c, int16_t *dst1, int16_t *dst2,
dst2[i]=(src2[xx]*(xalpha^127)+src2[xx+1]*xalpha);
xpos+=xInc;
}
+ for (i=dstWidth-1; (i*xInc)>>16 >=srcW-1; i--) {
+ dst1[i] = src1[srcW-1]*128;
+ dst2[i] = src2[srcW-1]*128;
+ }
}
-static av_always_inline void hcscale(SwsContext *c, int16_t *dst1, int16_t *dst2, int dstWidth,
+static av_always_inline void hcscale(SwsContext *c, uint16_t *dst1, uint16_t *dst2, int dstWidth,
const uint8_t *src1, const uint8_t *src2,
int srcW, int xInc, const int16_t *hChrFilter,
const int16_t *hChrFilterPos, int hChrFilterSize,
uint8_t *formatConvBuffer, uint32_t *pal)
{
if (c->chrToYV12) {
- uint8_t *buf2 = formatConvBuffer + FFALIGN(srcW * c->scalingBpp >> 3, 16);
+ uint8_t *buf2 = formatConvBuffer + FFALIGN(srcW*2+78, 16);
c->chrToYV12(formatConvBuffer, buf2, src1, src2, srcW, pal);
src1= formatConvBuffer;
src2= buf2;
}
- if (av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1 < 8 && c->scalingBpp == 16) {
- uint8_t *buf2 = (formatConvBuffer + FFALIGN(srcW * 2, 16));
- c->scale8To16Rv((uint16_t *) formatConvBuffer, src1, srcW);
- c->scale8To16Rv((uint16_t *) buf2, src2, srcW);
- src1 = formatConvBuffer;
- src2 = buf2;
- }
-
- if (!c->hcscale_fast) {
- c->hScale(c, dst1, dstWidth, src1, hChrFilter, hChrFilterPos, hChrFilterSize);
- c->hScale(c, dst2, dstWidth, src2, hChrFilter, hChrFilterPos, hChrFilterSize);
+ if (c->hScale16) {
+ int shift= isAnyRGB(c->srcFormat) || c->srcFormat==PIX_FMT_PAL8 ? 13 : av_pix_fmt_descriptors[c->srcFormat].comp[0].depth_minus1;
+ c->hScale16(dst1, dstWidth, (const uint16_t*)src1, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize, shift);
+ c->hScale16(dst2, dstWidth, (const uint16_t*)src2, srcW, xInc, hChrFilter, hChrFilterPos, hChrFilterSize, shift);
+ } else if (!c->hcscale_fast) {
+ c->hScale(dst1, dstWidth, src1, hChrFilter, hChrFilterPos, hChrFilterSize);
+ c->hScale(dst2, dstWidth, src2, hChrFilter, hChrFilterPos, hChrFilterSize);
} else { // fast bilinear upscale / crap downscale
c->hcscale_fast(c, dst1, dst2, dstWidth, src1, src2, srcW, xInc);
}
if (c->chrConvertRange)
c->chrConvertRange(dst1, dst2, dstWidth);
-
- if (av_pix_fmt_descriptors[c->dstFormat].comp[0].depth_minus1 < 8 && c->scalingBpp == 16) {
- c->scale19To15Fw(dst1, (int32_t *) dst1, dstWidth);
- c->scale19To15Fw(dst2, (int32_t *) dst2, dstWidth);
- }
}
static av_always_inline void
*yuv2packedX = yuv2bgr24_full_X_c;
break;
}
+ if(!*yuv2packedX)
+ goto YUV_PACKED;
} else {
+ YUV_PACKED:
switch (dstFormat) {
case PIX_FMT_GRAY16BE:
*yuv2packed1 = yuv2gray16BE_1_c;
*yuv2packedX = yuv2uyvy422_X_c;
break;
case PIX_FMT_RGB48LE:
- *yuv2packed1 = yuv2rgb48le_1_c;
- *yuv2packed2 = yuv2rgb48le_2_c;
- *yuv2packedX = yuv2rgb48le_X_c;
- break;
+ //*yuv2packed1 = yuv2rgb48le_1_c;
+ //*yuv2packed2 = yuv2rgb48le_2_c;
+ //*yuv2packedX = yuv2rgb48le_X_c;
+ //break;
case PIX_FMT_RGB48BE:
*yuv2packed1 = yuv2rgb48be_1_c;
*yuv2packed2 = yuv2rgb48be_2_c;
*yuv2packedX = yuv2rgb48be_X_c;
break;
case PIX_FMT_BGR48LE:
- *yuv2packed1 = yuv2bgr48le_1_c;
- *yuv2packed2 = yuv2bgr48le_2_c;
- *yuv2packedX = yuv2bgr48le_X_c;
- break;
+ //*yuv2packed1 = yuv2bgr48le_1_c;
+ //*yuv2packed2 = yuv2bgr48le_2_c;
+ //*yuv2packedX = yuv2bgr48le_X_c;
+ //break;
case PIX_FMT_BGR48BE:
*yuv2packed1 = yuv2bgr48be_1_c;
*yuv2packed2 = yuv2bgr48be_2_c;
const int chrSrcSliceH= -((-srcSliceH) >> c->chrSrcVSubSample);
int lastDstY;
uint32_t *pal=c->pal_yuv;
+ int should_dither= isNBPS(c->srcFormat) || is16BPS(c->srcFormat);
yuv2planar1_fn yuv2yuv1 = c->yuv2yuv1;
yuv2planarX_fn yuv2yuvX = c->yuv2yuvX;
yuv2packed1_fn yuv2packed1 = c->yuv2packed1;
dst[2] + dstStride[2] * chrDstY,
(CONFIG_SWSCALE_ALPHA && alpPixBuf) ? dst[3] + dstStride[3] * dstY : NULL,
};
+ const uint8_t *lumDither= should_dither ? dithers[7][dstY &7] : flat64;
+ const uint8_t *chrDither= should_dither ? dithers[7][chrDstY&7] : flat64;
const int firstLumSrcY= vLumFilterPos[dstY]; //First line needed as input
const int firstLumSrcY2= vLumFilterPos[FFMIN(dstY | ((1<<c->chrDstVSubSample) - 1), dstH-1)];
const int16_t **chrUSrcPtr= (const int16_t **) chrUPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
const int16_t **chrVSrcPtr= (const int16_t **) chrVPixBuf + chrBufIndex + firstChrSrcY - lastInChrBuf + vChrBufSize;
const int16_t **alpSrcPtr= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? (const int16_t **) alpPixBuf + lumBufIndex + firstLumSrcY - lastInLumBuf + vLumBufSize : NULL;
+
if (isPlanarYUV(dstFormat) || dstFormat==PIX_FMT_GRAY8) { //YV12 like
const int chrSkipMask= (1<<c->chrDstVSubSample)-1;
if ((dstY&chrSkipMask) || isGray(dstFormat))
if (c->yuv2yuv1 && vLumFilterSize == 1 && vChrFilterSize == 1) { // unscaled YV12
const int16_t *alpBuf= (CONFIG_SWSCALE_ALPHA && alpPixBuf) ? alpSrcPtr[0] : NULL;
yuv2yuv1(c, lumSrcPtr[0], chrUSrcPtr[0], chrVSrcPtr[0], alpBuf,
- dest, dstW, chrDstW);
+ dest, dstW, chrDstW, lumDither, chrDither);
} else { //General YV12
yuv2yuvX(c, vLumFilter + dstY * vLumFilterSize,
lumSrcPtr, vLumFilterSize,
vChrFilter + chrDstY * vChrFilterSize,
chrUSrcPtr, chrVSrcPtr, vChrFilterSize,
- alpSrcPtr, dest, dstW, chrDstW);
+ alpSrcPtr, dest, dstW, chrDstW, lumDither, chrDither);
}
} else {
assert(lumSrcPtr + vLumFilterSize - 1 < lumPixBuf + vLumBufSize*2);
&c->yuv2packed1, &c->yuv2packed2,
&c->yuv2packedX);
+ c->hScale = hScale_c;
+
+ if (c->flags & SWS_FAST_BILINEAR) {
+ c->hyscale_fast = hyscale_fast_c;
+ c->hcscale_fast = hcscale_fast_c;
+ }
+
c->chrToYV12 = NULL;
switch(srcFormat) {
case PIX_FMT_YUYV422 : c->chrToYV12 = yuy2ToUV_c; break;
case PIX_FMT_PAL8 :
case PIX_FMT_BGR4_BYTE:
case PIX_FMT_RGB4_BYTE: c->chrToYV12 = palToUV_c; break;
-#if HAVE_BIGENDIAN
- case PIX_FMT_YUV444P9LE:
- case PIX_FMT_YUV420P9LE:
- case PIX_FMT_YUV422P10LE:
- case PIX_FMT_YUV444P10LE:
- case PIX_FMT_YUV420P10LE:
- case PIX_FMT_YUV420P16LE:
- case PIX_FMT_YUV422P16LE:
- case PIX_FMT_YUV444P16LE: c->chrToYV12 = bswap16UV_c; break;
-#else
+ case PIX_FMT_GRAY16BE :
case PIX_FMT_YUV444P9BE:
case PIX_FMT_YUV420P9BE:
case PIX_FMT_YUV444P10BE:
case PIX_FMT_YUV420P10BE:
case PIX_FMT_YUV420P16BE:
case PIX_FMT_YUV422P16BE:
- case PIX_FMT_YUV444P16BE: c->chrToYV12 = bswap16UV_c; break;
-#endif
+ case PIX_FMT_YUV444P16BE: c->hScale16= HAVE_BIGENDIAN ? hScale16_c : hScale16X_c; break;
+ case PIX_FMT_GRAY16LE :
+ case PIX_FMT_YUV444P9LE:
+ case PIX_FMT_YUV420P9LE:
+ case PIX_FMT_YUV422P10LE:
+ case PIX_FMT_YUV420P10LE:
+ case PIX_FMT_YUV444P10LE:
+ case PIX_FMT_YUV420P16LE:
+ case PIX_FMT_YUV422P16LE:
+ case PIX_FMT_YUV444P16LE: c->hScale16= HAVE_BIGENDIAN ? hScale16X_c : hScale16_c; break;
}
if (c->chrSrcHSubSample) {
switch(srcFormat) {
c->lumToYV12 = NULL;
c->alpToYV12 = NULL;
switch (srcFormat) {
-#if HAVE_BIGENDIAN
- case PIX_FMT_YUV444P9LE:
- case PIX_FMT_YUV420P9LE:
- case PIX_FMT_YUV444P10LE:
- case PIX_FMT_YUV422P10LE:
- case PIX_FMT_YUV420P10LE:
- case PIX_FMT_YUV420P16LE:
- case PIX_FMT_YUV422P16LE:
- case PIX_FMT_YUV444P16LE:
- case PIX_FMT_GRAY16LE: c->lumToYV12 = bswap16Y_c; break;
-#else
- case PIX_FMT_YUV444P9BE:
- case PIX_FMT_YUV420P9BE:
- case PIX_FMT_YUV444P10BE:
- case PIX_FMT_YUV422P10BE:
- case PIX_FMT_YUV420P10BE:
- case PIX_FMT_YUV420P16BE:
- case PIX_FMT_YUV422P16BE:
- case PIX_FMT_YUV444P16BE:
- case PIX_FMT_GRAY16BE: c->lumToYV12 = bswap16Y_c; break;
-#endif
case PIX_FMT_YUYV422 :
- case PIX_FMT_Y400A : c->lumToYV12 = yuy2ToY_c; break;
- case PIX_FMT_UYVY422 : c->lumToYV12 = uyvyToY_c; break;
+ case PIX_FMT_GRAY8A :
+ c->lumToYV12 = yuy2ToY_c; break;
+ case PIX_FMT_UYVY422 :
+ c->lumToYV12 = uyvyToY_c; break;
case PIX_FMT_BGR24 : c->lumToYV12 = bgr24ToY_c; break;
case PIX_FMT_BGR565LE : c->lumToYV12 = bgr16leToY_c; break;
case PIX_FMT_BGR565BE : c->lumToYV12 = bgr16beToY_c; break;
case PIX_FMT_ABGR:
case PIX_FMT_ARGB: c->alpToYV12 = abgrToA_c; break;
case PIX_FMT_Y400A: c->alpToYV12 = uyvyToY_c; break;
+ case PIX_FMT_PAL8 : c->alpToYV12 = palToA_c; break;
}
}
- if (c->scalingBpp == 8) {
- c->hScale = hScale_c;
- if (c->flags & SWS_FAST_BILINEAR) {
- c->hyscale_fast = hyscale_fast_c;
- c->hcscale_fast = hcscale_fast_c;
- }
+ if(isAnyRGB(c->srcFormat) || c->srcFormat == PIX_FMT_PAL8)
+ c->hScale16= hScale16_c;
if (c->srcRange != c->dstRange && !isAnyRGB(c->dstFormat)) {
if (c->srcRange) {
c->chrConvertRange = chrRangeToJpeg_c;
}
}
- } else {
- c->hScale = hScale16_c;
- c->scale19To15Fw = scale19To15Fw_c;
- c->scale8To16Rv = scale8To16Rv_c;
-
- if (c->srcRange != c->dstRange && !isAnyRGB(c->dstFormat)) {
- if (c->srcRange) {
- c->lumConvertRange = lumRangeFromJpeg16_c;
- c->chrConvertRange = chrRangeFromJpeg16_c;
- } else {
- c->lumConvertRange = lumRangeToJpeg16_c;
- c->chrConvertRange = chrRangeToJpeg16_c;
- }
- }
- }
if (!(isGray(srcFormat) || isGray(c->dstFormat) ||
srcFormat == PIX_FMT_MONOBLACK || srcFormat == PIX_FMT_MONOWHITE))