diff options
Diffstat (limited to 'libswscale/swscale.c')
-rw-r--r-- | libswscale/swscale.c | 529 |
1 files changed, 343 insertions, 186 deletions
diff --git a/libswscale/swscale.c b/libswscale/swscale.c index f5c4e88688..d53af2771d 100644 --- a/libswscale/swscale.c +++ b/libswscale/swscale.c @@ -1,20 +1,20 @@ /* * 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 */ @@ -80,7 +80,7 @@ untested special converters (x)==PIX_FMT_PAL8 \ || (x)==PIX_FMT_YUYV422 \ || (x)==PIX_FMT_UYVY422 \ - || (x)==PIX_FMT_Y400A \ + || (x)==PIX_FMT_GRAY8A \ || isAnyRGB(x) \ ) @@ -121,6 +121,63 @@ add BGR4 output support write special BGR->BGR scaler */ +#if ARCH_X86 +DECLARE_ASM_CONST(8, uint64_t, bF8)= 0xF8F8F8F8F8F8F8F8LL; +DECLARE_ASM_CONST(8, uint64_t, bFC)= 0xFCFCFCFCFCFCFCFCLL; +DECLARE_ASM_CONST(8, uint64_t, w10)= 0x0010001000100010LL; +DECLARE_ASM_CONST(8, uint64_t, w02)= 0x0002000200020002LL; +DECLARE_ASM_CONST(8, uint64_t, bm00001111)=0x00000000FFFFFFFFLL; +DECLARE_ASM_CONST(8, uint64_t, bm00000111)=0x0000000000FFFFFFLL; +DECLARE_ASM_CONST(8, uint64_t, bm11111000)=0xFFFFFFFFFF000000LL; +DECLARE_ASM_CONST(8, uint64_t, bm01010101)=0x00FF00FF00FF00FFLL; + +const DECLARE_ALIGNED(8, uint64_t, ff_dither4)[2] = { + 0x0103010301030103LL, + 0x0200020002000200LL,}; + +const DECLARE_ALIGNED(8, uint64_t, ff_dither8)[2] = { + 0x0602060206020602LL, + 0x0004000400040004LL,}; + +DECLARE_ASM_CONST(8, uint64_t, b16Mask)= 0x001F001F001F001FLL; +DECLARE_ASM_CONST(8, uint64_t, g16Mask)= 0x07E007E007E007E0LL; +DECLARE_ASM_CONST(8, uint64_t, r16Mask)= 0xF800F800F800F800LL; +DECLARE_ASM_CONST(8, uint64_t, b15Mask)= 0x001F001F001F001FLL; +DECLARE_ASM_CONST(8, uint64_t, g15Mask)= 0x03E003E003E003E0LL; +DECLARE_ASM_CONST(8, uint64_t, r15Mask)= 0x7C007C007C007C00LL; + +DECLARE_ALIGNED(8, const uint64_t, ff_M24A) = 0x00FF0000FF0000FFLL; +DECLARE_ALIGNED(8, const uint64_t, ff_M24B) = 0xFF0000FF0000FF00LL; +DECLARE_ALIGNED(8, const uint64_t, ff_M24C) = 0x0000FF0000FF0000LL; + +#ifdef FAST_BGR2YV12 +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YCoeff) = 0x000000210041000DULL; +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UCoeff) = 0x0000FFEEFFDC0038ULL; +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2VCoeff) = 0x00000038FFD2FFF8ULL; +#else +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YCoeff) = 0x000020E540830C8BULL; +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UCoeff) = 0x0000ED0FDAC23831ULL; +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2VCoeff) = 0x00003831D0E6F6EAULL; +#endif /* FAST_BGR2YV12 */ +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2YOffset) = 0x1010101010101010ULL; +DECLARE_ALIGNED(8, const uint64_t, ff_bgr2UVOffset) = 0x8080808080808080ULL; +DECLARE_ALIGNED(8, const uint64_t, ff_w1111) = 0x0001000100010001ULL; + +DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toY1Coeff) = 0x0C88000040870C88ULL; +DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toY2Coeff) = 0x20DE4087000020DEULL; +DECLARE_ASM_CONST(8, uint64_t, ff_rgb24toY1Coeff) = 0x20DE0000408720DEULL; +DECLARE_ASM_CONST(8, uint64_t, ff_rgb24toY2Coeff) = 0x0C88408700000C88ULL; +DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toYOffset) = 0x0008400000084000ULL; + +DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toUV)[2][4] = { + {0x38380000DAC83838ULL, 0xECFFDAC80000ECFFULL, 0xF6E40000D0E3F6E4ULL, 0x3838D0E300003838ULL}, + {0xECFF0000DAC8ECFFULL, 0x3838DAC800003838ULL, 0x38380000D0E33838ULL, 0xF6E4D0E30000F6E4ULL}, +}; + +DECLARE_ASM_CONST(8, uint64_t, ff_bgr24toUVOffset)= 0x0040400000404000ULL; + +#endif /* ARCH_X86 */ + DECLARE_ALIGNED(8, static const uint8_t, dither_2x2_4)[2][8]={ { 1, 3, 1, 3, 1, 3, 1, 3, }, { 2, 0, 2, 0, 2, 0, 2, 0, }, @@ -209,6 +266,99 @@ DECLARE_ALIGNED(8, const uint8_t, dither_8x8_220)[8][8]={ }; #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,}, +}}; + +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 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, @@ -271,54 +421,77 @@ static av_always_inline void yuv2yuvX16inC_template(const int16_t *lumFilter, co } } -#define yuv2NBPS(bits, BE_LE, is_be) \ -static void yuv2yuvX ## bits ## BE_LE ## _c(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, int dstW, int chrDstW) \ -{ \ - yuv2yuvX16inC_template(lumFilter, lumSrc, lumFilterSize, \ - chrFilter, chrSrc, chrFilterSize, \ - alpSrc, \ - dest, uDest, vDest, aDest, \ - dstW, chrDstW, is_be, bits); \ +static av_always_inline void yuv2yuvXNinC_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, + int dstW, int chrDstW, int big_endian, int depth) +{ + //FIXME Optimize (just quickly written not optimized..) + int i; + + for (i = 0; i < dstW; i++) { + int val = 1 << (26-depth); + int j; + + for (j = 0; j < lumFilterSize; j++) + val += lumSrc[j][i] * lumFilter[j]; + + if (big_endian) { + AV_WB16(&dest[i], av_clip(val >> (27-depth), 0, (1<<depth)-1)); + } else { + AV_WL16(&dest[i], av_clip(val >> (27-depth), 0, (1<<depth)-1)); + } + } + + if (uDest) { + for (i = 0; i < chrDstW; i++) { + int u = 1 << (26-depth); + int v = 1 << (26-depth); + int j; + + for (j = 0; j < chrFilterSize; j++) { + u += chrSrc[j][i ] * chrFilter[j]; + v += chrSrc[j][i + VOFW] * chrFilter[j]; + } + + if (big_endian) { + AV_WB16(&uDest[i], av_clip(u >> (27-depth), 0, (1<<depth)-1)); + AV_WB16(&vDest[i], av_clip(v >> (27-depth), 0, (1<<depth)-1)); + } else { + AV_WL16(&uDest[i], av_clip(u >> (27-depth), 0, (1<<depth)-1)); + AV_WL16(&vDest[i], av_clip(v >> (27-depth), 0, (1<<depth)-1)); + } + } + } } -yuv2NBPS( 9, BE, 1); -yuv2NBPS( 9, LE, 0); -yuv2NBPS(10, BE, 1); -yuv2NBPS(10, LE, 0); -yuv2NBPS(16, BE, 1); -yuv2NBPS(16, LE, 0); static inline void yuv2yuvX16inC(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, int dstW, int chrDstW, enum PixelFormat dstFormat) { -#define conv16(bits) \ - if (isBE(dstFormat)) { \ - yuv2yuvX ## bits ## BE_c(lumFilter, lumSrc, lumFilterSize, \ - chrFilter, chrSrc, chrFilterSize, \ - alpSrc, \ - dest, uDest, vDest, aDest, \ - dstW, chrDstW); \ - } else { \ - yuv2yuvX ## bits ## LE_c(lumFilter, lumSrc, lumFilterSize, \ - chrFilter, chrSrc, chrFilterSize, \ - alpSrc, \ - dest, uDest, vDest, aDest, \ - dstW, chrDstW); \ - } - if (is16BPS(dstFormat)) { - conv16(16); - } else if (av_pix_fmt_descriptors[dstFormat].comp[0].depth_minus1 == 8) { - conv16(9); + if (isNBPS(dstFormat)) { + const int depth = av_pix_fmt_descriptors[dstFormat].comp[0].depth_minus1+1; + yuv2yuvXNinC_template(lumFilter, lumSrc, lumFilterSize, + chrFilter, chrSrc, chrFilterSize, + alpSrc, + dest, uDest, vDest, aDest, + dstW, chrDstW, isBE(dstFormat), depth); } else { - conv16(10); + if (isBE(dstFormat)) { + yuv2yuvX16inC_template(lumFilter, lumSrc, lumFilterSize, + chrFilter, chrSrc, chrFilterSize, + alpSrc, + dest, uDest, vDest, aDest, + dstW, chrDstW, 1, 16); + } else { + yuv2yuvX16inC_template(lumFilter, lumSrc, lumFilterSize, + chrFilter, chrSrc, chrFilterSize, + alpSrc, + dest, uDest, vDest, aDest, + dstW, chrDstW, 0, 16); + } } -#undef conv16 } static inline void yuv2yuvXinC(const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize, @@ -1133,6 +1306,16 @@ BGR2UV(uint16_t, bgr15ToUV, 0, 0, 0, 0, 0x001F, 0x03E0, 0x7C00, RU<<10, GU<< 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; @@ -1182,9 +1365,9 @@ static inline void monoblack2Y(uint8_t *dst, const uint8_t *src, long width, uin #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 @@ -1227,14 +1410,17 @@ static inline void monoblack2Y(uint8_t *dst, const uint8_t *src, long width, uin #define COMPILE_TEMPLATE_AMD3DNOW 0 #define COMPILE_TEMPLATE_ALTIVEC 0 +#if COMPILE_C +#define RENAME(a) a ## _C #include "swscale_template.c" +#endif #if COMPILE_ALTIVEC #undef RENAME #undef COMPILE_TEMPLATE_ALTIVEC #define COMPILE_TEMPLATE_ALTIVEC 1 #define RENAME(a) a ## _altivec -#include "ppc/swscale_template.c" +#include "swscale_template.c" #endif #if ARCH_X86 @@ -1249,7 +1435,7 @@ static inline void monoblack2Y(uint8_t *dst, const uint8_t *src, long width, uin #define COMPILE_TEMPLATE_MMX2 0 #define COMPILE_TEMPLATE_AMD3DNOW 0 #define RENAME(a) a ## _MMX -#include "x86/swscale_template.c" +#include "swscale_template.c" #endif //MMX2 versions @@ -1262,7 +1448,7 @@ static inline void monoblack2Y(uint8_t *dst, const uint8_t *src, long width, uin #define COMPILE_TEMPLATE_MMX2 1 #define COMPILE_TEMPLATE_AMD3DNOW 0 #define RENAME(a) a ## _MMX2 -#include "x86/swscale_template.c" +#include "swscale_template.c" #endif //3DNOW versions @@ -1275,36 +1461,44 @@ static inline void monoblack2Y(uint8_t *dst, const uint8_t *src, long width, uin #define COMPILE_TEMPLATE_MMX2 0 #define COMPILE_TEMPLATE_AMD3DNOW 1 #define RENAME(a) a ## _3DNow -#include "x86/swscale_template.c" +#include "swscale_template.c" #endif #endif //ARCH_X86 SwsFunc ff_getSwsFunc(SwsContext *c) { - sws_init_swScale_c(c); - #if CONFIG_RUNTIME_CPUDETECT + int flags = c->flags; + #if ARCH_X86 // ordered per speed fastest first - if (c->flags & SWS_CPU_CAPS_MMX2) { + if (flags & SWS_CPU_CAPS_MMX2) { sws_init_swScale_MMX2(c); return swScale_MMX2; - } else if (c->flags & SWS_CPU_CAPS_3DNOW) { + } else if (flags & SWS_CPU_CAPS_3DNOW) { sws_init_swScale_3DNow(c); return swScale_3DNow; - } else if (c->flags & SWS_CPU_CAPS_MMX) { + } else if (flags & SWS_CPU_CAPS_MMX) { sws_init_swScale_MMX(c); return swScale_MMX; + } else { + sws_init_swScale_C(c); + return swScale_C; } #else #if COMPILE_ALTIVEC - if (c->flags & SWS_CPU_CAPS_ALTIVEC) { + if (flags & SWS_CPU_CAPS_ALTIVEC) { sws_init_swScale_altivec(c); return swScale_altivec; + } else { + sws_init_swScale_C(c); + return swScale_C; } #endif + sws_init_swScale_C(c); + return swScale_C; #endif /* ARCH_X86 */ #else //CONFIG_RUNTIME_CPUDETECT #if COMPILE_TEMPLATE_MMX2 @@ -1319,10 +1513,11 @@ SwsFunc ff_getSwsFunc(SwsContext *c) #elif COMPILE_TEMPLATE_ALTIVEC sws_init_swScale_altivec(c); return swScale_altivec; +#else + sws_init_swScale_C(c); + return swScale_C; #endif #endif //!CONFIG_RUNTIME_CPUDETECT - - return swScale_c; } static void copyPlane(const uint8_t *src, int srcStride, @@ -1491,7 +1686,7 @@ static int palToRgbWrapper(SwsContext *c, const uint8_t* src[], int srcStride[], 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; @@ -1610,7 +1805,7 @@ static int rgbToRgbWrapper(SwsContext *c, const uint8_t* src[], int srcStride[], 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; @@ -1682,6 +1877,28 @@ static int packedCopyWrapper(SwsContext *c, const uint8_t* src[], int srcStride[ return srcSliceH; } +#define DITHER_COPY(dst, dstStride, src, srcStride, bswap, dbswap)\ + uint16_t scale= dither_scale[dst_depth-1][src_depth-1];\ + int shift= src_depth-dst_depth + dither_scale[src_depth-2][dst_depth-1];\ + for (i = 0; i < height; i++) {\ + uint8_t *dither= dithers[src_depth-9][i&7];\ + for (j = 0; j < length-7; j+=8){\ + dst[j+0] = dbswap((bswap(src[j+0]) + dither[0])*scale>>shift);\ + dst[j+1] = dbswap((bswap(src[j+1]) + dither[1])*scale>>shift);\ + dst[j+2] = dbswap((bswap(src[j+2]) + dither[2])*scale>>shift);\ + dst[j+3] = dbswap((bswap(src[j+3]) + dither[3])*scale>>shift);\ + dst[j+4] = dbswap((bswap(src[j+4]) + dither[4])*scale>>shift);\ + dst[j+5] = dbswap((bswap(src[j+5]) + dither[5])*scale>>shift);\ + dst[j+6] = dbswap((bswap(src[j+6]) + dither[6])*scale>>shift);\ + dst[j+7] = dbswap((bswap(src[j+7]) + dither[7])*scale>>shift);\ + }\ + for (; j < length; j++)\ + dst[j] = dbswap((bswap(src[j]) + dither[j&7])*scale>>shift);\ + dst += dstStride;\ + src += srcStride;\ + } + + static int planarCopyWrapper(SwsContext *c, const uint8_t* src[], int srcStride[], int srcSliceY, int srcSliceH, uint8_t* dst[], int dstStride[]) { @@ -1701,141 +1918,72 @@ static int planarCopyWrapper(SwsContext *c, const uint8_t* src[], int srcStride[ length*=2; fillPlane(dst[plane], dstStride[plane], length, height, y, (plane==3) ? 255 : 128); } else { - if(is9_OR_10BPS(c->srcFormat)) { + if(isNBPS(c->srcFormat) || isNBPS(c->dstFormat) + || (is16BPS(c->srcFormat) != is16BPS(c->dstFormat)) + ) { const int src_depth = av_pix_fmt_descriptors[c->srcFormat].comp[plane].depth_minus1+1; const int dst_depth = av_pix_fmt_descriptors[c->dstFormat].comp[plane].depth_minus1+1; const uint16_t *srcPtr2 = (const uint16_t*)srcPtr; + uint16_t *dstPtr2 = (uint16_t*)dstPtr; - if (is16BPS(c->dstFormat)) { - uint16_t *dstPtr2 = (uint16_t*)dstPtr; -#define COPY9_OR_10TO16(rfunc, wfunc) \ - for (i = 0; i < height; i++) { \ - for (j = 0; j < length; j++) { \ - int srcpx = rfunc(&srcPtr2[j]); \ - wfunc(&dstPtr2[j], (srcpx<<(16-src_depth)) | (srcpx>>(2*src_depth-16))); \ - } \ - dstPtr2 += dstStride[plane]/2; \ - srcPtr2 += srcStride[plane]/2; \ - } - if (isBE(c->dstFormat)) { - if (isBE(c->srcFormat)) { - COPY9_OR_10TO16(AV_RB16, AV_WB16); - } else { - COPY9_OR_10TO16(AV_RL16, AV_WB16); - } + if (dst_depth == 8) { + if(isBE(c->srcFormat) == HAVE_BIGENDIAN){ + DITHER_COPY(dstPtr, dstStride[plane], srcPtr2, srcStride[plane]/2, , ) } else { - if (isBE(c->srcFormat)) { - COPY9_OR_10TO16(AV_RB16, AV_WL16); - } else { - COPY9_OR_10TO16(AV_RL16, AV_WL16); - } - } - } else if (is9_OR_10BPS(c->dstFormat)) { - uint16_t *dstPtr2 = (uint16_t*)dstPtr; -#define COPY9_OR_10TO9_OR_10(loop) \ - for (i = 0; i < height; i++) { \ - for (j = 0; j < length; j++) { \ - loop; \ - } \ - dstPtr2 += dstStride[plane]/2; \ - srcPtr2 += srcStride[plane]/2; \ - } -#define COPY9_OR_10TO9_OR_10_2(rfunc, wfunc) \ - if (dst_depth > src_depth) { \ - COPY9_OR_10TO9_OR_10(int srcpx = rfunc(&srcPtr2[j]); \ - wfunc(&dstPtr2[j], (srcpx << 1) | (srcpx >> 9))); \ - } else if (dst_depth < src_depth) { \ - COPY9_OR_10TO9_OR_10(wfunc(&dstPtr2[j], rfunc(&srcPtr2[j]) >> 1)); \ - } else { \ - COPY9_OR_10TO9_OR_10(wfunc(&dstPtr2[j], rfunc(&srcPtr2[j]))); \ + DITHER_COPY(dstPtr, dstStride[plane], srcPtr2, srcStride[plane]/2, av_bswap16, ) } - if (isBE(c->dstFormat)) { - if (isBE(c->srcFormat)) { - COPY9_OR_10TO9_OR_10_2(AV_RB16, AV_WB16); + } else if (src_depth == 8) { + for (i = 0; i < height; i++) { + if(isBE(c->dstFormat)){ + for (j = 0; j < length; j++) + AV_WB16(&dstPtr2[j], (srcPtr[j]<<(dst_depth-8)) | + (srcPtr[j]>>(2*8-dst_depth))); } else { - COPY9_OR_10TO9_OR_10_2(AV_RL16, AV_WB16); + for (j = 0; j < length; j++) + AV_WL16(&dstPtr2[j], (srcPtr[j]<<(dst_depth-8)) | + (srcPtr[j]>>(2*8-dst_depth))); } - } else { - if (isBE(c->srcFormat)) { - COPY9_OR_10TO9_OR_10_2(AV_RB16, AV_WL16); + dstPtr2 += dstStride[plane]/2; + srcPtr += srcStride[plane]; + } + } else if (src_depth <= dst_depth) { + for (i = 0; i < height; i++) { +#define COPY_UP(r,w) \ + for (j = 0; j < length; j++){ \ + unsigned int v= r(&srcPtr2[j]);\ + w(&dstPtr2[j], (v<<(dst_depth-src_depth)) | \ + (v>>(2*src_depth-dst_depth)));\ + } + if(isBE(c->srcFormat)){ + if(isBE(c->dstFormat)){ + COPY_UP(AV_RB16, AV_WB16) + } else { + COPY_UP(AV_RB16, AV_WL16) + } } else { - COPY9_OR_10TO9_OR_10_2(AV_RL16, AV_WL16); + if(isBE(c->dstFormat)){ + COPY_UP(AV_RL16, AV_WB16) + } else { + COPY_UP(AV_RL16, AV_WL16) + } } + dstPtr2 += dstStride[plane]/2; + srcPtr2 += srcStride[plane]/2; } } else { - // FIXME Maybe dither instead. -#define COPY9_OR_10TO8(rfunc) \ - for (i = 0; i < height; i++) { \ - for (j = 0; j < length; j++) { \ - dstPtr[j] = rfunc(&srcPtr2[j])>>(src_depth-8); \ - } \ - dstPtr += dstStride[plane]; \ - srcPtr2 += srcStride[plane]/2; \ - } - if (isBE(c->srcFormat)) { - COPY9_OR_10TO8(AV_RB16); - } else { - COPY9_OR_10TO8(AV_RL16); - } - } - } else if(is9_OR_10BPS(c->dstFormat)) { - const int dst_depth = av_pix_fmt_descriptors[c->dstFormat].comp[plane].depth_minus1+1; - uint16_t *dstPtr2 = (uint16_t*)dstPtr; - - if (is16BPS(c->srcFormat)) { - const uint16_t *srcPtr2 = (const uint16_t*)srcPtr; -#define COPY16TO9_OR_10(rfunc, wfunc) \ - for (i = 0; i < height; i++) { \ - for (j = 0; j < length; j++) { \ - wfunc(&dstPtr2[j], rfunc(&srcPtr2[j])>>(16-dst_depth)); \ - } \ - dstPtr2 += dstStride[plane]/2; \ - srcPtr2 += srcStride[plane]/2; \ - } - if (isBE(c->dstFormat)) { - if (isBE(c->srcFormat)) { - COPY16TO9_OR_10(AV_RB16, AV_WB16); + if(isBE(c->srcFormat) == HAVE_BIGENDIAN){ + if(isBE(c->dstFormat) == HAVE_BIGENDIAN){ + DITHER_COPY(dstPtr2, dstStride[plane]/2, srcPtr2, srcStride[plane]/2, , ) } else { - COPY16TO9_OR_10(AV_RL16, AV_WB16); + DITHER_COPY(dstPtr2, dstStride[plane]/2, srcPtr2, srcStride[plane]/2, , av_bswap16) } - } else { - if (isBE(c->srcFormat)) { - COPY16TO9_OR_10(AV_RB16, AV_WL16); + }else{ + if(isBE(c->dstFormat) == HAVE_BIGENDIAN){ + DITHER_COPY(dstPtr2, dstStride[plane]/2, srcPtr2, srcStride[plane]/2, av_bswap16, ) } else { - COPY16TO9_OR_10(AV_RL16, AV_WL16); + DITHER_COPY(dstPtr2, dstStride[plane]/2, srcPtr2, srcStride[plane]/2, av_bswap16, av_bswap16) } } - } else /* 8bit */ { -#define COPY8TO9_OR_10(wfunc) \ - for (i = 0; i < height; i++) { \ - for (j = 0; j < length; j++) { \ - const int srcpx = srcPtr[j]; \ - wfunc(&dstPtr2[j], (srcpx<<(dst_depth-8)) | (srcpx >> (16-dst_depth))); \ - } \ - dstPtr2 += dstStride[plane]/2; \ - srcPtr += srcStride[plane]; \ - } - if (isBE(c->dstFormat)) { - COPY8TO9_OR_10(AV_WB16); - } else { - COPY8TO9_OR_10(AV_WL16); - } - } - } else if(is16BPS(c->srcFormat) && !is16BPS(c->dstFormat)) { - if (!isBE(c->srcFormat)) srcPtr++; - for (i=0; i<height; i++) { - for (j=0; j<length; j++) dstPtr[j] = srcPtr[j<<1]; - srcPtr+= srcStride[plane]; - dstPtr+= dstStride[plane]; - } - } else if(!is16BPS(c->srcFormat) && is16BPS(c->dstFormat)) { - for (i=0; i<height; i++) { - for (j=0; j<length; j++) { - dstPtr[ j<<1 ] = srcPtr[j]; - dstPtr[(j<<1)+1] = srcPtr[j]; - } - srcPtr+= srcStride[plane]; - dstPtr+= dstStride[plane]; } } else if(is16BPS(c->srcFormat) && is16BPS(c->dstFormat) && isBE(c->srcFormat) != isBE(c->dstFormat)) { @@ -2061,9 +2209,10 @@ int sws_scale(SwsContext *c, const uint8_t* const src[], const int srcStride[], 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; @@ -2079,7 +2228,7 @@ int sws_scale(SwsContext *c, const uint8_t* const src[], const int srcStride[], 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); @@ -2090,33 +2239,33 @@ int sws_scale(SwsContext *c, const uint8_t* const src[], const int srcStride[], 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); } } } @@ -2161,6 +2310,14 @@ int sws_scale(SwsContext *c, const uint8_t* const src[], const int srcStride[], } } +#if LIBSWSCALE_VERSION_MAJOR < 1 +int sws_scale_ordered(SwsContext *c, const uint8_t* const src[], int srcStride[], int srcSliceY, + int srcSliceH, uint8_t* dst[], int dstStride[]) +{ + return sws_scale(c, src, srcStride, srcSliceY, srcSliceH, dst, dstStride); +} +#endif + /* Convert the palette to the same packed 32-bit format as the palette */ void sws_convertPalette8ToPacked32(const uint8_t *src, uint8_t *dst, long num_pixels, const uint8_t *palette) { |