//#undef ARCH_X86
#define DITHER1XBPP
-#define FAST_BGR2YV12 // use 7 bit coefficients instead of 15 bit
-
#define isPacked(x) ( \
(x)==PIX_FMT_PAL8 \
|| (x)==PIX_FMT_YUYV422 \
|| (x)==PIX_FMT_UYVY422 \
- || (x)==PIX_FMT_Y400A \
+ || (x)==PIX_FMT_GRAY8A \
|| isAnyRGB(x) \
)
};
#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 av_always_inline void yuv2yuvX16inC_template(const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize,
const int16_t *chrFilter, const int16_t **chrSrc, int chrFilterSize,
const int16_t **alpSrc, uint16_t *dest, uint16_t *uDest, uint16_t *vDest, uint16_t *aDest,
dest+=12;\
}\
break;\
+ case PIX_FMT_BGR48BE:\
+ case PIX_FMT_BGR48LE:\
+ func(uint8_t,0)\
+ ((uint8_t*)dest)[ 0] = ((uint8_t*)dest)[ 1] = b[Y1];\
+ ((uint8_t*)dest)[ 2] = ((uint8_t*)dest)[ 3] = g[Y1];\
+ ((uint8_t*)dest)[ 4] = ((uint8_t*)dest)[ 5] = r[Y1];\
+ ((uint8_t*)dest)[ 6] = ((uint8_t*)dest)[ 7] = b[Y2];\
+ ((uint8_t*)dest)[ 8] = ((uint8_t*)dest)[ 9] = g[Y2];\
+ ((uint8_t*)dest)[10] = ((uint8_t*)dest)[11] = r[Y2];\
+ dest+=12;\
+ }\
+ break;\
case PIX_FMT_RGBA:\
case PIX_FMT_BGRA:\
if (CONFIG_SMALL) {\
}
}
+static inline void bgr48ToY(uint8_t *dst, const uint8_t *src, long width,
+ uint32_t *unused)
+{
+ int i;
+ for (i = 0; i < width; i++) {
+ int b = src[i*6+0];
+ int g = src[i*6+2];
+ int r = src[i*6+4];
+
+ dst[i] = (RY*r + GY*g + BY*b + (33<<(RGB2YUV_SHIFT-1))) >> RGB2YUV_SHIFT;
+ }
+}
+
+static inline void bgr48ToUV(uint8_t *dstU, uint8_t *dstV,
+ const uint8_t *src1, const uint8_t *src2,
+ long width, uint32_t *unused)
+{
+ int i;
+ for (i = 0; i < width; i++) {
+ int b = src1[6*i + 0];
+ int g = src1[6*i + 2];
+ int r = src1[6*i + 4];
+
+ 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;
+ }
+}
+
+static inline void bgr48ToUV_half(uint8_t *dstU, uint8_t *dstV,
+ const uint8_t *src1, const uint8_t *src2,
+ long width, uint32_t *unused)
+{
+ int i;
+ for (i = 0; i < width; i++) {
+ int b= src1[12*i + 0] + src1[12*i + 6];
+ int g= src1[12*i + 2] + src1[12*i + 8];
+ int r= src1[12*i + 4] + src1[12*i + 10];
+
+ 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);
+ }
+}
+
#define BGR2Y(type, name, shr, shg, shb, maskr, maskg, maskb, RY, GY, BY, S)\
static inline void name(uint8_t *dst, const uint8_t *src, long width, uint32_t *unused)\
{\
BGR2UV(uint16_t, rgb16ToUV, 0, 0, 0, 0, 0xF800, 0x07E0, 0x001F, RU , GU<<5, BU<<11, RV , GV<<5, BV<<11, RGB2YUV_SHIFT+8)
BGR2UV(uint16_t, rgb15ToUV, 0, 0, 0, 0, 0x7C00, 0x03E0, 0x001F, RU , GU<<5, BU<<10, RV , GV<<5, BV<<10, RGB2YUV_SHIFT+7)
+static inline void palToA(uint8_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] >> 24;
+ }
+}
+
static inline void palToY(uint8_t *dst, const uint8_t *src, long width, uint32_t *pal)
{
int i;
#if CONFIG_RUNTIME_CPUDETECT
# define COMPILE_C 1
# if ARCH_X86
-# define COMPILE_MMX HAVE_MMX
-# define COMPILE_MMX2 HAVE_MMX2
-# define COMPILE_3DNOW HAVE_AMD3DNOW
+# define COMPILE_MMX 1
+# define COMPILE_MMX2 1
+# define COMPILE_3DNOW 1
# elif ARCH_PPC
# define COMPILE_ALTIVEC HAVE_ALTIVEC
# endif
uint8_t *dstPtr= dst[0] + dstStride[0]*srcSliceY;
const uint8_t *srcPtr= src[0];
- if (srcFormat == PIX_FMT_Y400A) {
+ if (srcFormat == PIX_FMT_GRAY8A) {
switch (dstFormat) {
case PIX_FMT_RGB32 : conv = gray8aToPacked32; break;
case PIX_FMT_BGR32 : conv = gray8aToPacked32; break;
if ((dstFormat == PIX_FMT_RGB32_1 || dstFormat == PIX_FMT_BGR32_1) && !isRGBA32(srcFormat))
dstPtr += ALT32_CORR;
- if (dstStride[0]*srcBpp == srcStride[0]*dstBpp && srcStride[0] > 0)
+ if (dstStride[0]*srcBpp == srcStride[0]*dstBpp && srcStride[0] > 0 && !(srcStride[0]%srcBpp))
conv(srcPtr, dstPtr + dstStride[0]*srcSliceY, srcSliceH*srcStride[0]);
else {
int i;
length*=2;
fillPlane(dst[plane], dstStride[plane], length, height, y, (plane==3) ? 255 : 128);
} else {
- if(is16BPS(c->srcFormat) && !is16BPS(c->dstFormat)) {
+ if(isNBPS(c->srcFormat)) {
+ const int depth = av_pix_fmt_descriptors[c->srcFormat].comp[plane].depth_minus1+1;
+ uint16_t *srcPtr2 = (uint16_t*)srcPtr;
+
+ if (is16BPS(c->dstFormat)) {
+ uint16_t *dstPtr2 = (uint16_t*)dstPtr;
+ for (i = 0; i < height; i++) {
+ for (j = 0; j < length; j++)
+ dstPtr2[j] = (srcPtr2[j]<<(16-depth)) | (srcPtr2[j]>>(2*depth-16));
+ dstPtr2 += dstStride[plane]/2;
+ srcPtr2 += srcStride[plane]/2;
+ }
+ } else {
+ for (i = 0; i < height; i++) {
+ uint8_t *dither= dithers[depth-9][i&7];
+ for (j = 0; j < length-7; j+=8){
+ dstPtr[j+0] = (srcPtr2[j+0] + dither[0])>>(depth-8);
+ dstPtr[j+1] = (srcPtr2[j+1] + dither[1])>>(depth-8);
+ dstPtr[j+2] = (srcPtr2[j+2] + dither[2])>>(depth-8);
+ dstPtr[j+3] = (srcPtr2[j+3] + dither[3])>>(depth-8);
+ dstPtr[j+4] = (srcPtr2[j+4] + dither[4])>>(depth-8);
+ dstPtr[j+5] = (srcPtr2[j+5] + dither[5])>>(depth-8);
+ dstPtr[j+6] = (srcPtr2[j+6] + dither[6])>>(depth-8);
+ dstPtr[j+7] = (srcPtr2[j+7] + dither[7])>>(depth-8);
+ }
+ for (; j < length; j++)
+ dstPtr[j] = (srcPtr2[j] + dither[j&7])>>(depth-8);
+ dstPtr += dstStride[plane];
+ srcPtr2 += srcStride[plane]/2;
+ }
+ }
+ } else if(is16BPS(c->srcFormat) && !is16BPS(c->dstFormat)) {
+ //FIXME add dither
if (!isBE(c->srcFormat)) srcPtr++;
for (i=0; i<height; i++) {
for (j=0; j<length; j++) dstPtr[j] = srcPtr[j<<1];
&& srcFormat != PIX_FMT_MONOWHITE && dstFormat != PIX_FMT_MONOWHITE
&& srcFormat != PIX_FMT_RGB48LE && dstFormat != PIX_FMT_RGB48LE
&& srcFormat != PIX_FMT_RGB48BE && dstFormat != PIX_FMT_RGB48BE
+ && srcFormat != PIX_FMT_BGR48LE && dstFormat != PIX_FMT_BGR48LE
+ && srcFormat != PIX_FMT_BGR48BE && dstFormat != PIX_FMT_BGR48BE
&& (!needsDither || (c->flags&(SWS_FAST_BILINEAR|SWS_POINT))))
c->swScale= rgbToRgbWrapper;
if (usePal(c->srcFormat)) {
for (i=0; i<256; i++) {
- int p, r, g, b,y,u,v;
+ int p, r, g, b, y, u, v, a = 0xff;
if(c->srcFormat == PIX_FMT_PAL8) {
p=((const uint32_t*)(src[1]))[i];
+ a= (p>>24)&0xFF;
r= (p>>16)&0xFF;
g= (p>> 8)&0xFF;
b= p &0xFF;
r= (i>>3 )*255;
g= ((i>>1)&3)*85;
b= (i&1 )*255;
- } else if(c->srcFormat == PIX_FMT_GRAY8 || c->srcFormat == PIX_FMT_Y400A) {
+ } else if(c->srcFormat == PIX_FMT_GRAY8 || c->srcFormat == PIX_FMT_GRAY8A) {
r = g = b = i;
} else {
assert(c->srcFormat == PIX_FMT_BGR4_BYTE);
y= av_clip_uint8((RY*r + GY*g + BY*b + ( 33<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
u= av_clip_uint8((RU*r + GU*g + BU*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
v= av_clip_uint8((RV*r + GV*g + BV*b + (257<<(RGB2YUV_SHIFT-1)))>>RGB2YUV_SHIFT);
- c->pal_yuv[i]= y + (u<<8) + (v<<16);
+ c->pal_yuv[i]= y + (u<<8) + (v<<16) + (a<<24);
switch(c->dstFormat) {
case PIX_FMT_BGR32:
#if !HAVE_BIGENDIAN
case PIX_FMT_RGB24:
#endif
- c->pal_rgb[i]= r + (g<<8) + (b<<16);
+ c->pal_rgb[i]= r + (g<<8) + (b<<16) + (a<<24);
break;
case PIX_FMT_BGR32_1:
#if HAVE_BIGENDIAN
case PIX_FMT_BGR24:
#endif
- c->pal_rgb[i]= (r + (g<<8) + (b<<16)) << 8;
+ c->pal_rgb[i]= a + (r<<8) + (g<<16) + (b<<24);
break;
case PIX_FMT_RGB32_1:
#if HAVE_BIGENDIAN
case PIX_FMT_RGB24:
#endif
- c->pal_rgb[i]= (b + (g<<8) + (r<<16)) << 8;
+ c->pal_rgb[i]= a + (b<<8) + (g<<16) + (r<<24);
break;
case PIX_FMT_RGB32:
#if !HAVE_BIGENDIAN
case PIX_FMT_BGR24:
#endif
default:
- c->pal_rgb[i]= b + (g<<8) + (r<<16);
+ c->pal_rgb[i]= b + (g<<8) + (r<<16) + (a<<24);
}
}
}
projects / ffmpeg / blobdiff