X-Git-Url: https://git.sesse.net/?a=blobdiff_plain;f=libswscale%2Fswscale_internal.h;h=2a7d2dbd1dff89f8b9666958c12a9f1f6e9d9cdc;hb=36a8c43073114cffc6ea147178f227a95e40a59f;hp=9492303301e999b7966fb7ec150328b31b51f107;hpb=62ee0e6a977e1990c9853630c7dea1415b38bb28;p=ffmpeg diff --git a/libswscale/swscale_internal.h b/libswscale/swscale_internal.h index 9492303301e..2a7d2dbd1df 100644 --- a/libswscale/swscale_internal.h +++ b/libswscale/swscale_internal.h @@ -30,10 +30,11 @@ #include "libavutil/avutil.h" #include "libavutil/log.h" #include "libavutil/pixfmt.h" +#include "libavutil/pixdesc.h" -#define STR(s) AV_TOSTRING(s) //AV_STRINGIFY is too long +#define STR(s) AV_TOSTRING(s) // AV_STRINGIFY is too long -#define FAST_BGR2YV12 //use 7-bit instead of 15-bit coefficients +#define FAST_BGR2YV12 // use 7-bit instead of 15-bit coefficients #define MAX_FILTER_SIZE 256 @@ -44,72 +45,91 @@ #endif #if ARCH_X86_64 -# define APCK_PTR2 8 +# define APCK_PTR2 8 # define APCK_COEF 16 # define APCK_SIZE 24 #else -# define APCK_PTR2 4 -# define APCK_COEF 8 +# define APCK_PTR2 4 +# define APCK_COEF 8 # define APCK_SIZE 16 #endif struct SwsContext; -typedef int (*SwsFunc)(struct SwsContext *context, const uint8_t* src[], +typedef int (*SwsFunc)(struct SwsContext *context, const uint8_t *src[], int srcStride[], int srcSliceY, int srcSliceH, - uint8_t* dst[], int dstStride[]); + uint8_t *dst[], int dstStride[]); /** - * Write one line of horizontally scaled Y/U/V/A to planar output + * Write one line of horizontally scaled data to planar output * without any additional vertical scaling (or point-scaling). * - * @param c SWS scaling context - * @param lumSrc scaled luma (Y) source data, 15bit for 8bit output - * @param chrUSrc scaled chroma (U) source data, 15bit for 8bit output - * @param chrVSrc scaled chroma (V) source data, 15bit for 8bit output - * @param alpSrc scaled alpha (A) source data, 15bit for 8bit output - * @param dest pointer to the 4 output planes (Y/U/V/A) - * @param dstW width of dest[0], dest[3], lumSrc and alpSrc in pixels - * @param chrDstW width of dest[1], dest[2], chrUSrc and chrVSrc + * @param src scaled source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param dest pointer to the output plane. For >8bit + * output, this is in uint16_t + * @param dstW width of destination in pixels + * @param dither ordered dither array of type int16_t and size 8 + * @param offset Dither offset */ -typedef void (*yuv2planar1_fn) (struct 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); +typedef void (*yuv2planar1_fn)(const int16_t *src, uint8_t *dest, int dstW, + const uint8_t *dither, int offset); + /** - * Write one line of horizontally scaled Y/U/V/A to planar output + * Write one line of horizontally scaled data to planar output + * with multi-point vertical scaling between input pixels. + * + * @param filter vertical luma/alpha scaling coefficients, 12bit [0,4096] + * @param src scaled luma (Y) or alpha (A) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param filterSize number of vertical input lines to scale + * @param dest pointer to output plane. For >8bit + * output, this is in uint16_t + * @param dstW width of destination pixels + * @param offset Dither offset + */ +typedef void (*yuv2planarX_fn)(const int16_t *filter, int filterSize, + const int16_t **src, uint8_t *dest, int dstW, + const uint8_t *dither, int offset); + +/** + * Write one line of horizontally scaled chroma to interleaved output * with multi-point vertical scaling between input pixels. * * @param c SWS scaling context - * @param lumFilter vertical luma/alpha scaling coefficients, 12bit [0,4096] - * @param lumSrc scaled luma (Y) source data, 15bit for 8bit output - * @param lumFilterSize number of vertical luma/alpha input lines to scale * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096] - * @param chrUSrc scaled chroma (U) source data, 15bit for 8bit output - * @param chrVSrc scaled chroma (V) source data, 15bit for 8bit output + * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) * @param chrFilterSize number of vertical chroma input lines to scale - * @param alpSrc scaled alpha (A) source data, 15bit for 8bit output - * @param dest pointer to the 4 output planes (Y/U/V/A) - * @param dstW width of dest[0], dest[3], lumSrc and alpSrc in pixels - * @param chrDstW width of dest[1], dest[2], chrUSrc and chrVSrc + * @param dest pointer to the output plane. For >8bit + * output, this is in uint16_t + * @param dstW width of chroma planes */ -typedef void (*yuv2planarX_fn) (struct 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[4], - int dstW, int chrDstW); +typedef void (*yuv2interleavedX_fn)(struct SwsContext *c, + const int16_t *chrFilter, + int chrFilterSize, + const int16_t **chrUSrc, + const int16_t **chrVSrc, + uint8_t *dest, int dstW); + /** * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB * output without any additional vertical scaling (or point-scaling). Note * that this function may do chroma scaling, see the "uvalpha" argument. * * @param c SWS scaling context - * @param lumSrc scaled luma (Y) source data, 15bit for 8bit output - * @param chrUSrc scaled chroma (U) source data, 15bit for 8bit output - * @param chrVSrc scaled chroma (V) source data, 15bit for 8bit output - * @param alpSrc scaled alpha (A) source data, 15bit for 8bit output - * @param dest pointer to the output plane + * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param dest pointer to the output plane. For 16bit output, this is + * uint16_t * @param dstW width of lumSrc and alpSrc in pixels, number of pixels * to write into dest[] * @param uvalpha chroma scaling coefficient for the second line of chroma @@ -123,20 +143,26 @@ typedef void (*yuv2planarX_fn) (struct SwsContext *c, const int16_t *lumFilter, * but can be used to generate comfort noise using dithering * for some output formats. */ -typedef void (*yuv2packed1_fn) (struct SwsContext *c, const int16_t *lumSrc, - const int16_t *chrUSrc[2], const int16_t *chrVSrc[2], - const int16_t *alpSrc, uint8_t *dest, - int dstW, int uvalpha, int y); +typedef void (*yuv2packed1_fn)(struct SwsContext *c, const int16_t *lumSrc, + const int16_t *chrUSrc[2], + const int16_t *chrVSrc[2], + const int16_t *alpSrc, uint8_t *dest, + int dstW, int uvalpha, int y); /** * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB * output by doing bilinear scaling between two input lines. * * @param c SWS scaling context - * @param lumSrc scaled luma (Y) source data, 15bit for 8bit output - * @param chrUSrc scaled chroma (U) source data, 15bit for 8bit output - * @param chrVSrc scaled chroma (V) source data, 15bit for 8bit output - * @param alpSrc scaled alpha (A) source data, 15bit for 8bit output - * @param dest pointer to the output plane + * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param dest pointer to the output plane. For 16bit output, this is + * uint16_t * @param dstW width of lumSrc and alpSrc in pixels, number of pixels * to write into dest[] * @param yalpha luma/alpha scaling coefficients for the second input line. @@ -150,24 +176,31 @@ typedef void (*yuv2packed1_fn) (struct SwsContext *c, const int16_t *lumSrc, * but can be used to generate comfort noise using dithering * for some output formats. */ -typedef void (*yuv2packed2_fn) (struct SwsContext *c, const int16_t *lumSrc[2], - const int16_t *chrUSrc[2], const int16_t *chrVSrc[2], - const int16_t *alpSrc[2], uint8_t *dest, - int dstW, int yalpha, int uvalpha, int y); +typedef void (*yuv2packed2_fn)(struct SwsContext *c, const int16_t *lumSrc[2], + const int16_t *chrUSrc[2], + const int16_t *chrVSrc[2], + const int16_t *alpSrc[2], + uint8_t *dest, + int dstW, int yalpha, int uvalpha, int y); /** * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB * output by doing multi-point vertical scaling between input pixels. * * @param c SWS scaling context * @param lumFilter vertical luma/alpha scaling coefficients, 12bit [0,4096] - * @param lumSrc scaled luma (Y) source data, 15bit for 8bit output + * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) * @param lumFilterSize number of vertical luma/alpha input lines to scale * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096] - * @param chrUSrc scaled chroma (U) source data, 15bit for 8bit output - * @param chrVSrc scaled chroma (V) source data, 15bit for 8bit output + * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) * @param chrFilterSize number of vertical chroma input lines to scale - * @param alpSrc scaled alpha (A) source data, 15bit for 8bit output - * @param dest pointer to the output plane + * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, + * 19-bit for 16bit output (in int32_t) + * @param dest pointer to the output plane. For 16bit output, this is + * uint16_t * @param dstW width of lumSrc and alpSrc in pixels, number of pixels * to write into dest[] * @param y vertical line number for this output. This does not need @@ -175,12 +208,13 @@ typedef void (*yuv2packed2_fn) (struct SwsContext *c, const int16_t *lumSrc[2], * but can be used to generate comfort noise using dithering * or some output formats. */ -typedef void (*yuv2packedX_fn) (struct 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); +typedef void (*yuv2packedX_fn)(struct 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); /* This struct should be aligned on at least a 32-byte boundary. */ typedef struct SwsContext { @@ -207,7 +241,7 @@ typedef struct SwsContext { enum PixelFormat srcFormat; ///< Source pixel format. int dstFormatBpp; ///< Number of bits per pixel of the destination pixel format. int srcFormatBpp; ///< Number of bits per pixel of the source pixel format. - int scalingBpp; + int dstBpc, srcBpc; int chrSrcHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in source image. int chrSrcVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in source image. int chrDstHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in destination image. @@ -233,12 +267,12 @@ typedef struct SwsContext { int16_t **chrUPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler. int16_t **chrVPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler. int16_t **alpPixBuf; ///< Ring buffer for scaled horizontal alpha plane lines to be fed to the vertical scaler. - int vLumBufSize; ///< Number of vertical luma/alpha lines allocated in the ring buffer. - int vChrBufSize; ///< Number of vertical chroma lines allocated in the ring buffer. - int lastInLumBuf; ///< Last scaled horizontal luma/alpha line from source in the ring buffer. - int lastInChrBuf; ///< Last scaled horizontal chroma line from source in the ring buffer. - int lumBufIndex; ///< Index in ring buffer of the last scaled horizontal luma/alpha line from source. - int chrBufIndex; ///< Index in ring buffer of the last scaled horizontal chroma line from source. + int vLumBufSize; ///< Number of vertical luma/alpha lines allocated in the ring buffer. + int vChrBufSize; ///< Number of vertical chroma lines allocated in the ring buffer. + int lastInLumBuf; ///< Last scaled horizontal luma/alpha line from source in the ring buffer. + int lastInChrBuf; ///< Last scaled horizontal chroma line from source in the ring buffer. + int lumBufIndex; ///< Index in ring buffer of the last scaled horizontal luma/alpha line from source. + int chrBufIndex; ///< Index in ring buffer of the last scaled horizontal chroma line from source. //@} uint8_t *formatConvBuffer; @@ -261,14 +295,14 @@ typedef struct SwsContext { int16_t *hChrFilter; ///< Array of horizontal filter coefficients for chroma planes. int16_t *vLumFilter; ///< Array of vertical filter coefficients for luma/alpha planes. int16_t *vChrFilter; ///< Array of vertical filter coefficients for chroma planes. - int16_t *hLumFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes. - int16_t *hChrFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes. - int16_t *vLumFilterPos; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes. - int16_t *vChrFilterPos; ///< Array of vertical filter starting positions for each dst[i] for chroma planes. - int hLumFilterSize; ///< Horizontal filter size for luma/alpha pixels. - int hChrFilterSize; ///< Horizontal filter size for chroma pixels. - int vLumFilterSize; ///< Vertical filter size for luma/alpha pixels. - int vChrFilterSize; ///< Vertical filter size for chroma pixels. + int32_t *hLumFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes. + int32_t *hChrFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes. + int32_t *vLumFilterPos; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes. + int32_t *vChrFilterPos; ///< Array of vertical filter starting positions for each dst[i] for chroma planes. + int hLumFilterSize; ///< Horizontal filter size for luma/alpha pixels. + int hChrFilterSize; ///< Horizontal filter size for chroma pixels. + int vLumFilterSize; ///< Vertical filter size for luma/alpha pixels. + int vChrFilterSize; ///< Vertical filter size for chroma pixels. //@} int lumMmx2FilterCodeSize; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for luma/alpha planes. @@ -280,11 +314,11 @@ typedef struct SwsContext { int dstY; ///< Last destination vertical line output from last slice. int flags; ///< Flags passed by the user to select scaler algorithm, optimizations, subsampling, etc... - void * yuvTable; // pointer to the yuv->rgb table start so it can be freed() - uint8_t * table_rV[256]; - uint8_t * table_gU[256]; - int table_gV[256]; - uint8_t * table_bU[256]; + void *yuvTable; // pointer to the yuv->rgb table start so it can be freed() + uint8_t *table_rV[256]; + uint8_t *table_gU[256]; + int table_gV[256]; + uint8_t *table_bU[256]; //Colorspace stuff int contrast, brightness, saturation; // for sws_getColorspaceDetails @@ -336,15 +370,15 @@ typedef struct SwsContext { DECLARE_ALIGNED(8, uint64_t, yOffset); DECLARE_ALIGNED(8, uint64_t, uOffset); DECLARE_ALIGNED(8, uint64_t, vOffset); - int32_t lumMmxFilter[4*MAX_FILTER_SIZE]; - int32_t chrMmxFilter[4*MAX_FILTER_SIZE]; + int32_t lumMmxFilter[4 * MAX_FILTER_SIZE]; + int32_t chrMmxFilter[4 * MAX_FILTER_SIZE]; int dstW; ///< Width of destination luma/alpha planes. DECLARE_ALIGNED(8, uint64_t, esp); DECLARE_ALIGNED(8, uint64_t, vRounder); DECLARE_ALIGNED(8, uint64_t, u_temp); DECLARE_ALIGNED(8, uint64_t, v_temp); DECLARE_ALIGNED(8, uint64_t, y_temp); - int32_t alpMmxFilter[4*MAX_FILTER_SIZE]; + int32_t alpMmxFilter[4 * MAX_FILTER_SIZE]; // alignment of these values is not necessary, but merely here // to maintain the same offset across x8632 and x86-64. Once we // use proper offset macros in the asm, they can be removed. @@ -363,7 +397,7 @@ typedef struct SwsContext { vector signed short CGV; vector signed short OY; vector unsigned short CSHIFT; - vector signed short *vYCoeffsBank, *vCCoeffsBank; + vector signed short *vYCoeffsBank, *vCCoeffsBank; #endif #if ARCH_BFIN @@ -385,19 +419,34 @@ typedef struct SwsContext { #endif /* function pointers for swScale() */ - yuv2planar1_fn yuv2yuv1; - yuv2planarX_fn yuv2yuvX; + yuv2planar1_fn yuv2plane1; + yuv2planarX_fn yuv2planeX; + yuv2interleavedX_fn yuv2nv12cX; yuv2packed1_fn yuv2packed1; yuv2packed2_fn yuv2packed2; yuv2packedX_fn yuv2packedX; + /// Unscaled conversion of luma plane to YV12 for horizontal scaler. void (*lumToYV12)(uint8_t *dst, const uint8_t *src, - int width, uint32_t *pal); ///< Unscaled conversion of luma plane to YV12 for horizontal scaler. + int width, uint32_t *pal); + /// Unscaled conversion of alpha plane to YV12 for horizontal scaler. void (*alpToYV12)(uint8_t *dst, const uint8_t *src, - int width, uint32_t *pal); ///< Unscaled conversion of alpha plane to YV12 for horizontal scaler. + int width, uint32_t *pal); + /// Unscaled conversion of chroma planes to YV12 for horizontal scaler. void (*chrToYV12)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src1, const uint8_t *src2, - int width, uint32_t *pal); ///< Unscaled conversion of chroma planes to YV12 for horizontal scaler. + int width, uint32_t *pal); + + /** + * Functions to read planar input, such as planar RGB, and convert + * internally to Y/UV. + */ + /** @{ */ + void (*readLumPlanar)(uint8_t *dst, const uint8_t *src[4], int width); + void (*readChrPlanar)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src[4], + int width); + /** @} */ + /** * Scale one horizontal line of input data using a bilinear filter * to produce one line of output data. Compared to SwsContext->hScale(), @@ -431,17 +480,19 @@ typedef struct SwsContext { * lines, to produce one (differently sized) line of output data. * * @param dst pointer to destination buffer for horizontally scaled - * data. If the scaling depth (SwsContext->scalingBpp) is - * 8, data will be 15bpp in 16bits (int16_t) width. If - * scaling depth is 16, data will be 19bpp in 32bpp - * (int32_t) width. + * data. If the number of bits per component of one + * destination pixel (SwsContext->dstBpc) is <= 10, data + * will be 15bpc in 16bits (int16_t) width. Else (i.e. + * SwsContext->dstBpc == 16), data will be 19bpc in + * 32bits (int32_t) width. * @param dstW width of destination image - * @param src pointer to source data to be scaled. If scaling depth - * is 8, this is 8bpp in 8bpp (uint8_t) width. If scaling - * depth is 16, this is native depth in 16bbp (uint16_t) - * width. In other words, for 9-bit YUV input, this is - * 9bpp, for 10-bit YUV input, this is 10bpp, and for - * 16-bit RGB or YUV, this is 16bpp. + * @param src pointer to source data to be scaled. If the number of + * bits per component of a source pixel (SwsContext->srcBpc) + * is 8, this is 8bpc in 8bits (uint8_t) width. Else + * (i.e. SwsContext->dstBpc > 8), this is native depth + * in 16bits (uint16_t) width. In other words, for 9-bit + * YUV input, this is 9bpc, for 10-bit YUV input, this is + * 10bpc, and for 16-bit RGB or YUV, this is 16bpc. * @param filter filter coefficients to be used per output pixel for * scaling. This contains 14bpp filtering coefficients. * Guaranteed to contain dstW * filterSize entries. @@ -454,24 +505,21 @@ typedef struct SwsContext { * (and input coefficients thus padded with zeroes) * to simplify creating SIMD code. */ - void (*hScale)(struct SwsContext *c, int16_t *dst, int dstW, const uint8_t *src, - const int16_t *filter, const int16_t *filterPos, - int filterSize); - - void (*lumConvertRange)(int16_t *dst, int width); ///< Color range conversion function for luma plane if needed. - void (*chrConvertRange)(int16_t *dst1, int16_t *dst2, int width); ///< Color range conversion function for chroma planes if needed. + /** @{ */ + void (*hyScale)(struct SwsContext *c, int16_t *dst, int dstW, + const uint8_t *src, const int16_t *filter, + const int32_t *filterPos, int filterSize); + void (*hcScale)(struct SwsContext *c, int16_t *dst, int dstW, + const uint8_t *src, const int16_t *filter, + const int32_t *filterPos, int filterSize); + /** @} */ - /** - * dst[..] = (src[..] << 8) | src[..]; - */ - void (*scale8To16Rv)(uint16_t *dst, const uint8_t *src, int len); - /** - * dst[..] = src[..] >> 4; - */ - void (*scale19To15Fw)(int16_t *dst, const int32_t *src, int len); + /// Color range conversion function for luma plane if needed. + void (*lumConvertRange)(int16_t *dst, int width); + /// Color range conversion function for chroma planes if needed. + void (*chrConvertRange)(int16_t *dst1, int16_t *dst2, int width); int needs_hcscale; ///< Set if there are chroma planes to be converted. - } SwsContext; //FIXME check init (where 0) @@ -487,156 +535,109 @@ void updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufI SwsFunc ff_yuv2rgb_init_mmx(SwsContext *c); SwsFunc ff_yuv2rgb_init_vis(SwsContext *c); -SwsFunc ff_yuv2rgb_init_mlib(SwsContext *c); SwsFunc ff_yuv2rgb_init_altivec(SwsContext *c); SwsFunc ff_yuv2rgb_get_func_ptr_bfin(SwsContext *c); void ff_bfin_get_unscaled_swscale(SwsContext *c); const char *sws_format_name(enum PixelFormat format); -//FIXME replace this with something faster -#define is16BPS(x) ( \ - (x)==PIX_FMT_GRAY16BE \ - || (x)==PIX_FMT_GRAY16LE \ - || (x)==PIX_FMT_BGR48BE \ - || (x)==PIX_FMT_BGR48LE \ - || (x)==PIX_FMT_RGB48BE \ - || (x)==PIX_FMT_RGB48LE \ - || (x)==PIX_FMT_YUV420P16LE \ - || (x)==PIX_FMT_YUV422P16LE \ - || (x)==PIX_FMT_YUV444P16LE \ - || (x)==PIX_FMT_YUV420P16BE \ - || (x)==PIX_FMT_YUV422P16BE \ - || (x)==PIX_FMT_YUV444P16BE \ - ) -#define is9_OR_10BPS(x) ( \ - (x)==PIX_FMT_YUV420P9LE \ - || (x)==PIX_FMT_YUV420P9BE \ - || (x)==PIX_FMT_YUV444P9BE \ - || (x)==PIX_FMT_YUV444P9LE \ - || (x)==PIX_FMT_YUV422P10BE \ - || (x)==PIX_FMT_YUV422P10LE \ - || (x)==PIX_FMT_YUV444P10BE \ - || (x)==PIX_FMT_YUV444P10LE \ - || (x)==PIX_FMT_YUV420P10LE \ - || (x)==PIX_FMT_YUV420P10BE \ - ) -#define isBE(x) ((x)&1) -#define isPlanar8YUV(x) ( \ - (x)==PIX_FMT_YUV410P \ - || (x)==PIX_FMT_YUV420P \ - || (x)==PIX_FMT_YUVA420P \ - || (x)==PIX_FMT_YUV411P \ - || (x)==PIX_FMT_YUV422P \ - || (x)==PIX_FMT_YUV444P \ - || (x)==PIX_FMT_YUV440P \ - || (x)==PIX_FMT_NV12 \ - || (x)==PIX_FMT_NV21 \ - ) -#define isPlanarYUV(x) ( \ - isPlanar8YUV(x) \ - || (x)==PIX_FMT_YUV420P9LE \ - || (x)==PIX_FMT_YUV444P9LE \ - || (x)==PIX_FMT_YUV420P10LE \ - || (x)==PIX_FMT_YUV422P10LE \ - || (x)==PIX_FMT_YUV444P10LE \ - || (x)==PIX_FMT_YUV420P16LE \ - || (x)==PIX_FMT_YUV422P16LE \ - || (x)==PIX_FMT_YUV444P16LE \ - || (x)==PIX_FMT_YUV420P9BE \ - || (x)==PIX_FMT_YUV444P9BE \ - || (x)==PIX_FMT_YUV420P10BE \ - || (x)==PIX_FMT_YUV422P10BE \ - || (x)==PIX_FMT_YUV444P10BE \ - || (x)==PIX_FMT_YUV420P16BE \ - || (x)==PIX_FMT_YUV422P16BE \ - || (x)==PIX_FMT_YUV444P16BE \ - ) -#define isYUV(x) ( \ - (x)==PIX_FMT_UYVY422 \ - || (x)==PIX_FMT_YUYV422 \ - || isPlanarYUV(x) \ - ) -#define isGray(x) ( \ - (x)==PIX_FMT_GRAY8 \ - || (x)==PIX_FMT_Y400A \ - || (x)==PIX_FMT_GRAY16BE \ - || (x)==PIX_FMT_GRAY16LE \ - ) -#define isGray16(x) ( \ - (x)==PIX_FMT_GRAY16BE \ - || (x)==PIX_FMT_GRAY16LE \ - ) -#define isRGBinInt(x) ( \ - (x)==PIX_FMT_RGB48BE \ - || (x)==PIX_FMT_RGB48LE \ - || (x)==PIX_FMT_RGB32 \ - || (x)==PIX_FMT_RGB32_1 \ - || (x)==PIX_FMT_RGB24 \ - || (x)==PIX_FMT_RGB565BE \ - || (x)==PIX_FMT_RGB565LE \ - || (x)==PIX_FMT_RGB555BE \ - || (x)==PIX_FMT_RGB555LE \ - || (x)==PIX_FMT_RGB444BE \ - || (x)==PIX_FMT_RGB444LE \ - || (x)==PIX_FMT_RGB8 \ - || (x)==PIX_FMT_RGB4 \ - || (x)==PIX_FMT_RGB4_BYTE \ - || (x)==PIX_FMT_MONOBLACK \ - || (x)==PIX_FMT_MONOWHITE \ - ) -#define isBGRinInt(x) ( \ - (x)==PIX_FMT_BGR48BE \ - || (x)==PIX_FMT_BGR48LE \ - || (x)==PIX_FMT_BGR32 \ - || (x)==PIX_FMT_BGR32_1 \ - || (x)==PIX_FMT_BGR24 \ - || (x)==PIX_FMT_BGR565BE \ - || (x)==PIX_FMT_BGR565LE \ - || (x)==PIX_FMT_BGR555BE \ - || (x)==PIX_FMT_BGR555LE \ - || (x)==PIX_FMT_BGR444BE \ - || (x)==PIX_FMT_BGR444LE \ - || (x)==PIX_FMT_BGR8 \ - || (x)==PIX_FMT_BGR4 \ - || (x)==PIX_FMT_BGR4_BYTE \ - || (x)==PIX_FMT_MONOBLACK \ - || (x)==PIX_FMT_MONOWHITE \ - ) -#define isRGBinBytes(x) ( \ - (x)==PIX_FMT_RGB48BE \ - || (x)==PIX_FMT_RGB48LE \ - || (x)==PIX_FMT_RGBA \ - || (x)==PIX_FMT_ARGB \ - || (x)==PIX_FMT_RGB24 \ - ) -#define isBGRinBytes(x) ( \ - (x)==PIX_FMT_BGR48BE \ - || (x)==PIX_FMT_BGR48LE \ - || (x)==PIX_FMT_BGRA \ - || (x)==PIX_FMT_ABGR \ - || (x)==PIX_FMT_BGR24 \ - ) -#define isAnyRGB(x) ( \ - isRGBinInt(x) \ - || isBGRinInt(x) \ - ) -#define isALPHA(x) ( \ - (x)==PIX_FMT_BGR32 \ - || (x)==PIX_FMT_BGR32_1 \ - || (x)==PIX_FMT_RGB32 \ - || (x)==PIX_FMT_RGB32_1 \ - || (x)==PIX_FMT_Y400A \ - || (x)==PIX_FMT_YUVA420P \ - ) -#define isPacked(x) ( \ - (x)==PIX_FMT_PAL8 \ - || (x)==PIX_FMT_YUYV422 \ - || (x)==PIX_FMT_UYVY422 \ - || (x)==PIX_FMT_Y400A \ - || isAnyRGB(x) \ - ) -#define usePal(x) ((av_pix_fmt_descriptors[x].flags & PIX_FMT_PAL) || (x) == PIX_FMT_Y400A) +#define is16BPS(x) \ + (av_pix_fmt_descriptors[x].comp[0].depth_minus1 == 15) + +#define is9_OR_10BPS(x) \ + (av_pix_fmt_descriptors[x].comp[0].depth_minus1 == 8 || \ + av_pix_fmt_descriptors[x].comp[0].depth_minus1 == 9) + +#define isBE(x) \ + (av_pix_fmt_descriptors[x].flags & PIX_FMT_BE) + +#define isYUV(x) \ + (!(av_pix_fmt_descriptors[x].flags & PIX_FMT_RGB) && \ + av_pix_fmt_descriptors[x].nb_components >= 2) + +#define isPlanarYUV(x) \ + ((av_pix_fmt_descriptors[x].flags & PIX_FMT_PLANAR) && \ + isYUV(x)) + +#define isRGB(x) \ + (av_pix_fmt_descriptors[x].flags & PIX_FMT_RGB) + +#if 0 // FIXME +#define isGray(x) \ + (!(av_pix_fmt_descriptors[x].flags & PIX_FMT_PAL) && \ + av_pix_fmt_descriptors[x].nb_components <= 2) +#else +#define isGray(x) \ + ((x) == PIX_FMT_GRAY8 || \ + (x) == PIX_FMT_Y400A || \ + (x) == PIX_FMT_GRAY16BE || \ + (x) == PIX_FMT_GRAY16LE) +#endif + +#define isRGBinInt(x) \ + ((x) == PIX_FMT_RGB48BE || \ + (x) == PIX_FMT_RGB48LE || \ + (x) == PIX_FMT_RGB32 || \ + (x) == PIX_FMT_RGB32_1 || \ + (x) == PIX_FMT_RGB24 || \ + (x) == PIX_FMT_RGB565BE || \ + (x) == PIX_FMT_RGB565LE || \ + (x) == PIX_FMT_RGB555BE || \ + (x) == PIX_FMT_RGB555LE || \ + (x) == PIX_FMT_RGB444BE || \ + (x) == PIX_FMT_RGB444LE || \ + (x) == PIX_FMT_RGB8 || \ + (x) == PIX_FMT_RGB4 || \ + (x) == PIX_FMT_RGB4_BYTE || \ + (x) == PIX_FMT_MONOBLACK || \ + (x) == PIX_FMT_MONOWHITE) + +#define isBGRinInt(x) \ + ((x) == PIX_FMT_BGR48BE || \ + (x) == PIX_FMT_BGR48LE || \ + (x) == PIX_FMT_BGR32 || \ + (x) == PIX_FMT_BGR32_1 || \ + (x) == PIX_FMT_BGR24 || \ + (x) == PIX_FMT_BGR565BE || \ + (x) == PIX_FMT_BGR565LE || \ + (x) == PIX_FMT_BGR555BE || \ + (x) == PIX_FMT_BGR555LE || \ + (x) == PIX_FMT_BGR444BE || \ + (x) == PIX_FMT_BGR444LE || \ + (x) == PIX_FMT_BGR8 || \ + (x) == PIX_FMT_BGR4 || \ + (x) == PIX_FMT_BGR4_BYTE || \ + (x) == PIX_FMT_MONOBLACK || \ + (x) == PIX_FMT_MONOWHITE) + +#define isAnyRGB(x) \ + (isRGBinInt(x) || \ + isBGRinInt(x)) + +#define isALPHA(x) \ + (av_pix_fmt_descriptors[x].nb_components == 2 || \ + av_pix_fmt_descriptors[x].nb_components == 4) + +#define isPacked(x) \ + ((av_pix_fmt_descriptors[x].nb_components >= 2 && \ + !(av_pix_fmt_descriptors[x].flags & PIX_FMT_PLANAR)) || \ + (x) == PIX_FMT_PAL8) + +#define isPlanar(x) \ + (av_pix_fmt_descriptors[x].nb_components >= 2 && \ + (av_pix_fmt_descriptors[x].flags & PIX_FMT_PLANAR)) + +#define isPackedRGB(x) \ + ((av_pix_fmt_descriptors[x].flags & \ + (PIX_FMT_PLANAR | PIX_FMT_RGB)) == PIX_FMT_RGB) + +#define isPlanarRGB(x) \ + ((av_pix_fmt_descriptors[x].flags & \ + (PIX_FMT_PLANAR | PIX_FMT_RGB)) == (PIX_FMT_PLANAR | PIX_FMT_RGB)) + +#define usePal(x) ((av_pix_fmt_descriptors[x].flags & PIX_FMT_PAL) || \ + (av_pix_fmt_descriptors[x].flags & PIX_FMT_PSEUDOPAL) || \ + (x) == PIX_FMT_Y400A) extern const uint64_t ff_dither4[2]; extern const uint64_t ff_dither8[2]; @@ -644,7 +645,7 @@ extern const uint64_t ff_dither8[2]; extern const AVClass sws_context_class; /** - * Sets c->swScale to an unscaled converter if one exists for the specific + * Set c->swScale to an unscaled converter if one exists for the specific * source and destination formats, bit depths, flags, etc. */ void ff_get_unscaled_swscale(SwsContext *c); @@ -652,11 +653,19 @@ void ff_get_unscaled_swscale(SwsContext *c); void ff_swscale_get_unscaled_altivec(SwsContext *c); /** - * Returns function pointer to fastest main scaler path function depending + * Return function pointer to fastest main scaler path function depending * on architecture and available optimizations. */ SwsFunc ff_getSwsFunc(SwsContext *c); +void ff_sws_init_input_funcs(SwsContext *c); +void ff_sws_init_output_funcs(SwsContext *c, + yuv2planar1_fn *yuv2plane1, + yuv2planarX_fn *yuv2planeX, + yuv2interleavedX_fn *yuv2nv12cX, + yuv2packed1_fn *yuv2packed1, + yuv2packed2_fn *yuv2packed2, + yuv2packedX_fn *yuv2packedX); void ff_sws_init_swScale_altivec(SwsContext *c); void ff_sws_init_swScale_mmx(SwsContext *c);