| 1 | /* |
| 2 | * Copyright (C) 2001-2011 Michael Niedermayer <michaelni@gmx.at> |
| 3 | * |
| 4 | * This file is part of FFmpeg. |
| 5 | * |
| 6 | * FFmpeg is free software; you can redistribute it and/or |
| 7 | * modify it under the terms of the GNU Lesser General Public |
| 8 | * License as published by the Free Software Foundation; either |
| 9 | * version 2.1 of the License, or (at your option) any later version. |
| 10 | * |
| 11 | * FFmpeg is distributed in the hope that it will be useful, |
| 12 | * but WITHOUT ANY WARRANTY; without even the implied warranty of |
| 13 | * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU |
| 14 | * Lesser General Public License for more details. |
| 15 | * |
| 16 | * You should have received a copy of the GNU Lesser General Public |
| 17 | * License along with FFmpeg; if not, write to the Free Software |
| 18 | * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA |
| 19 | */ |
| 20 | |
| 21 | #ifndef SWSCALE_SWSCALE_INTERNAL_H |
| 22 | #define SWSCALE_SWSCALE_INTERNAL_H |
| 23 | |
| 24 | #include "config.h" |
| 25 | |
| 26 | #if HAVE_ALTIVEC_H |
| 27 | #include <altivec.h> |
| 28 | #endif |
| 29 | |
| 30 | #include "version.h" |
| 31 | |
| 32 | #include "libavutil/avassert.h" |
| 33 | #include "libavutil/avutil.h" |
| 34 | #include "libavutil/common.h" |
| 35 | #include "libavutil/intreadwrite.h" |
| 36 | #include "libavutil/log.h" |
| 37 | #include "libavutil/pixfmt.h" |
| 38 | #include "libavutil/pixdesc.h" |
| 39 | |
| 40 | #define STR(s) AV_TOSTRING(s) // AV_STRINGIFY is too long |
| 41 | |
| 42 | #define YUVRGB_TABLE_HEADROOM 128 |
| 43 | |
| 44 | #define MAX_FILTER_SIZE SWS_MAX_FILTER_SIZE |
| 45 | |
| 46 | #define DITHER1XBPP |
| 47 | |
| 48 | #if HAVE_BIGENDIAN |
| 49 | #define ALT32_CORR (-1) |
| 50 | #else |
| 51 | #define ALT32_CORR 1 |
| 52 | #endif |
| 53 | |
| 54 | #if ARCH_X86_64 |
| 55 | # define APCK_PTR2 8 |
| 56 | # define APCK_COEF 16 |
| 57 | # define APCK_SIZE 24 |
| 58 | #else |
| 59 | # define APCK_PTR2 4 |
| 60 | # define APCK_COEF 8 |
| 61 | # define APCK_SIZE 16 |
| 62 | #endif |
| 63 | |
| 64 | struct SwsContext; |
| 65 | |
| 66 | typedef enum SwsDither { |
| 67 | SWS_DITHER_NONE = 0, |
| 68 | SWS_DITHER_AUTO, |
| 69 | SWS_DITHER_BAYER, |
| 70 | SWS_DITHER_ED, |
| 71 | SWS_DITHER_A_DITHER, |
| 72 | SWS_DITHER_X_DITHER, |
| 73 | NB_SWS_DITHER, |
| 74 | } SwsDither; |
| 75 | |
| 76 | typedef int (*SwsFunc)(struct SwsContext *context, const uint8_t *src[], |
| 77 | int srcStride[], int srcSliceY, int srcSliceH, |
| 78 | uint8_t *dst[], int dstStride[]); |
| 79 | |
| 80 | /** |
| 81 | * Write one line of horizontally scaled data to planar output |
| 82 | * without any additional vertical scaling (or point-scaling). |
| 83 | * |
| 84 | * @param src scaled source data, 15bit for 8-10bit output, |
| 85 | * 19-bit for 16bit output (in int32_t) |
| 86 | * @param dest pointer to the output plane. For >8bit |
| 87 | * output, this is in uint16_t |
| 88 | * @param dstW width of destination in pixels |
| 89 | * @param dither ordered dither array of type int16_t and size 8 |
| 90 | * @param offset Dither offset |
| 91 | */ |
| 92 | typedef void (*yuv2planar1_fn)(const int16_t *src, uint8_t *dest, int dstW, |
| 93 | const uint8_t *dither, int offset); |
| 94 | |
| 95 | /** |
| 96 | * Write one line of horizontally scaled data to planar output |
| 97 | * with multi-point vertical scaling between input pixels. |
| 98 | * |
| 99 | * @param filter vertical luma/alpha scaling coefficients, 12bit [0,4096] |
| 100 | * @param src scaled luma (Y) or alpha (A) source data, 15bit for 8-10bit output, |
| 101 | * 19-bit for 16bit output (in int32_t) |
| 102 | * @param filterSize number of vertical input lines to scale |
| 103 | * @param dest pointer to output plane. For >8bit |
| 104 | * output, this is in uint16_t |
| 105 | * @param dstW width of destination pixels |
| 106 | * @param offset Dither offset |
| 107 | */ |
| 108 | typedef void (*yuv2planarX_fn)(const int16_t *filter, int filterSize, |
| 109 | const int16_t **src, uint8_t *dest, int dstW, |
| 110 | const uint8_t *dither, int offset); |
| 111 | |
| 112 | /** |
| 113 | * Write one line of horizontally scaled chroma to interleaved output |
| 114 | * with multi-point vertical scaling between input pixels. |
| 115 | * |
| 116 | * @param c SWS scaling context |
| 117 | * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096] |
| 118 | * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, |
| 119 | * 19-bit for 16bit output (in int32_t) |
| 120 | * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, |
| 121 | * 19-bit for 16bit output (in int32_t) |
| 122 | * @param chrFilterSize number of vertical chroma input lines to scale |
| 123 | * @param dest pointer to the output plane. For >8bit |
| 124 | * output, this is in uint16_t |
| 125 | * @param dstW width of chroma planes |
| 126 | */ |
| 127 | typedef void (*yuv2interleavedX_fn)(struct SwsContext *c, |
| 128 | const int16_t *chrFilter, |
| 129 | int chrFilterSize, |
| 130 | const int16_t **chrUSrc, |
| 131 | const int16_t **chrVSrc, |
| 132 | uint8_t *dest, int dstW); |
| 133 | |
| 134 | /** |
| 135 | * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB |
| 136 | * output without any additional vertical scaling (or point-scaling). Note |
| 137 | * that this function may do chroma scaling, see the "uvalpha" argument. |
| 138 | * |
| 139 | * @param c SWS scaling context |
| 140 | * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, |
| 141 | * 19-bit for 16bit output (in int32_t) |
| 142 | * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, |
| 143 | * 19-bit for 16bit output (in int32_t) |
| 144 | * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, |
| 145 | * 19-bit for 16bit output (in int32_t) |
| 146 | * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, |
| 147 | * 19-bit for 16bit output (in int32_t) |
| 148 | * @param dest pointer to the output plane. For 16bit output, this is |
| 149 | * uint16_t |
| 150 | * @param dstW width of lumSrc and alpSrc in pixels, number of pixels |
| 151 | * to write into dest[] |
| 152 | * @param uvalpha chroma scaling coefficient for the second line of chroma |
| 153 | * pixels, either 2048 or 0. If 0, one chroma input is used |
| 154 | * for 2 output pixels (or if the SWS_FLAG_FULL_CHR_INT flag |
| 155 | * is set, it generates 1 output pixel). If 2048, two chroma |
| 156 | * input pixels should be averaged for 2 output pixels (this |
| 157 | * only happens if SWS_FLAG_FULL_CHR_INT is not set) |
| 158 | * @param y vertical line number for this output. This does not need |
| 159 | * to be used to calculate the offset in the destination, |
| 160 | * but can be used to generate comfort noise using dithering |
| 161 | * for some output formats. |
| 162 | */ |
| 163 | typedef void (*yuv2packed1_fn)(struct SwsContext *c, const int16_t *lumSrc, |
| 164 | const int16_t *chrUSrc[2], |
| 165 | const int16_t *chrVSrc[2], |
| 166 | const int16_t *alpSrc, uint8_t *dest, |
| 167 | int dstW, int uvalpha, int y); |
| 168 | /** |
| 169 | * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB |
| 170 | * output by doing bilinear scaling between two input lines. |
| 171 | * |
| 172 | * @param c SWS scaling context |
| 173 | * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, |
| 174 | * 19-bit for 16bit output (in int32_t) |
| 175 | * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, |
| 176 | * 19-bit for 16bit output (in int32_t) |
| 177 | * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, |
| 178 | * 19-bit for 16bit output (in int32_t) |
| 179 | * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, |
| 180 | * 19-bit for 16bit output (in int32_t) |
| 181 | * @param dest pointer to the output plane. For 16bit output, this is |
| 182 | * uint16_t |
| 183 | * @param dstW width of lumSrc and alpSrc in pixels, number of pixels |
| 184 | * to write into dest[] |
| 185 | * @param yalpha luma/alpha scaling coefficients for the second input line. |
| 186 | * The first line's coefficients can be calculated by using |
| 187 | * 4096 - yalpha |
| 188 | * @param uvalpha chroma scaling coefficient for the second input line. The |
| 189 | * first line's coefficients can be calculated by using |
| 190 | * 4096 - uvalpha |
| 191 | * @param y vertical line number for this output. This does not need |
| 192 | * to be used to calculate the offset in the destination, |
| 193 | * but can be used to generate comfort noise using dithering |
| 194 | * for some output formats. |
| 195 | */ |
| 196 | typedef void (*yuv2packed2_fn)(struct SwsContext *c, const int16_t *lumSrc[2], |
| 197 | const int16_t *chrUSrc[2], |
| 198 | const int16_t *chrVSrc[2], |
| 199 | const int16_t *alpSrc[2], |
| 200 | uint8_t *dest, |
| 201 | int dstW, int yalpha, int uvalpha, int y); |
| 202 | /** |
| 203 | * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB |
| 204 | * output by doing multi-point vertical scaling between input pixels. |
| 205 | * |
| 206 | * @param c SWS scaling context |
| 207 | * @param lumFilter vertical luma/alpha scaling coefficients, 12bit [0,4096] |
| 208 | * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, |
| 209 | * 19-bit for 16bit output (in int32_t) |
| 210 | * @param lumFilterSize number of vertical luma/alpha input lines to scale |
| 211 | * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096] |
| 212 | * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, |
| 213 | * 19-bit for 16bit output (in int32_t) |
| 214 | * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, |
| 215 | * 19-bit for 16bit output (in int32_t) |
| 216 | * @param chrFilterSize number of vertical chroma input lines to scale |
| 217 | * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, |
| 218 | * 19-bit for 16bit output (in int32_t) |
| 219 | * @param dest pointer to the output plane. For 16bit output, this is |
| 220 | * uint16_t |
| 221 | * @param dstW width of lumSrc and alpSrc in pixels, number of pixels |
| 222 | * to write into dest[] |
| 223 | * @param y vertical line number for this output. This does not need |
| 224 | * to be used to calculate the offset in the destination, |
| 225 | * but can be used to generate comfort noise using dithering |
| 226 | * or some output formats. |
| 227 | */ |
| 228 | typedef void (*yuv2packedX_fn)(struct SwsContext *c, const int16_t *lumFilter, |
| 229 | const int16_t **lumSrc, int lumFilterSize, |
| 230 | const int16_t *chrFilter, |
| 231 | const int16_t **chrUSrc, |
| 232 | const int16_t **chrVSrc, int chrFilterSize, |
| 233 | const int16_t **alpSrc, uint8_t *dest, |
| 234 | int dstW, int y); |
| 235 | |
| 236 | /** |
| 237 | * Write one line of horizontally scaled Y/U/V/A to YUV/RGB |
| 238 | * output by doing multi-point vertical scaling between input pixels. |
| 239 | * |
| 240 | * @param c SWS scaling context |
| 241 | * @param lumFilter vertical luma/alpha scaling coefficients, 12bit [0,4096] |
| 242 | * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output, |
| 243 | * 19-bit for 16bit output (in int32_t) |
| 244 | * @param lumFilterSize number of vertical luma/alpha input lines to scale |
| 245 | * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096] |
| 246 | * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output, |
| 247 | * 19-bit for 16bit output (in int32_t) |
| 248 | * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output, |
| 249 | * 19-bit for 16bit output (in int32_t) |
| 250 | * @param chrFilterSize number of vertical chroma input lines to scale |
| 251 | * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output, |
| 252 | * 19-bit for 16bit output (in int32_t) |
| 253 | * @param dest pointer to the output planes. For 16bit output, this is |
| 254 | * uint16_t |
| 255 | * @param dstW width of lumSrc and alpSrc in pixels, number of pixels |
| 256 | * to write into dest[] |
| 257 | * @param y vertical line number for this output. This does not need |
| 258 | * to be used to calculate the offset in the destination, |
| 259 | * but can be used to generate comfort noise using dithering |
| 260 | * or some output formats. |
| 261 | */ |
| 262 | typedef void (*yuv2anyX_fn)(struct SwsContext *c, const int16_t *lumFilter, |
| 263 | const int16_t **lumSrc, int lumFilterSize, |
| 264 | const int16_t *chrFilter, |
| 265 | const int16_t **chrUSrc, |
| 266 | const int16_t **chrVSrc, int chrFilterSize, |
| 267 | const int16_t **alpSrc, uint8_t **dest, |
| 268 | int dstW, int y); |
| 269 | |
| 270 | /* This struct should be aligned on at least a 32-byte boundary. */ |
| 271 | typedef struct SwsContext { |
| 272 | /** |
| 273 | * info on struct for av_log |
| 274 | */ |
| 275 | const AVClass *av_class; |
| 276 | |
| 277 | /** |
| 278 | * Note that src, dst, srcStride, dstStride will be copied in the |
| 279 | * sws_scale() wrapper so they can be freely modified here. |
| 280 | */ |
| 281 | SwsFunc swscale; |
| 282 | int srcW; ///< Width of source luma/alpha planes. |
| 283 | int srcH; ///< Height of source luma/alpha planes. |
| 284 | int dstH; ///< Height of destination luma/alpha planes. |
| 285 | int chrSrcW; ///< Width of source chroma planes. |
| 286 | int chrSrcH; ///< Height of source chroma planes. |
| 287 | int chrDstW; ///< Width of destination chroma planes. |
| 288 | int chrDstH; ///< Height of destination chroma planes. |
| 289 | int lumXInc, chrXInc; |
| 290 | int lumYInc, chrYInc; |
| 291 | enum AVPixelFormat dstFormat; ///< Destination pixel format. |
| 292 | enum AVPixelFormat srcFormat; ///< Source pixel format. |
| 293 | int dstFormatBpp; ///< Number of bits per pixel of the destination pixel format. |
| 294 | int srcFormatBpp; ///< Number of bits per pixel of the source pixel format. |
| 295 | int dstBpc, srcBpc; |
| 296 | int chrSrcHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in source image. |
| 297 | int chrSrcVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in source image. |
| 298 | int chrDstHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in destination image. |
| 299 | int chrDstVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in destination image. |
| 300 | int vChrDrop; ///< Binary logarithm of extra vertical subsampling factor in source image chroma planes specified by user. |
| 301 | int sliceDir; ///< Direction that slices are fed to the scaler (1 = top-to-bottom, -1 = bottom-to-top). |
| 302 | double param[2]; ///< Input parameters for scaling algorithms that need them. |
| 303 | |
| 304 | uint32_t pal_yuv[256]; |
| 305 | uint32_t pal_rgb[256]; |
| 306 | |
| 307 | /** |
| 308 | * @name Scaled horizontal lines ring buffer. |
| 309 | * The horizontal scaler keeps just enough scaled lines in a ring buffer |
| 310 | * so they may be passed to the vertical scaler. The pointers to the |
| 311 | * allocated buffers for each line are duplicated in sequence in the ring |
| 312 | * buffer to simplify indexing and avoid wrapping around between lines |
| 313 | * inside the vertical scaler code. The wrapping is done before the |
| 314 | * vertical scaler is called. |
| 315 | */ |
| 316 | //@{ |
| 317 | int16_t **lumPixBuf; ///< Ring buffer for scaled horizontal luma plane lines to be fed to the vertical scaler. |
| 318 | int16_t **chrUPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler. |
| 319 | int16_t **chrVPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler. |
| 320 | int16_t **alpPixBuf; ///< Ring buffer for scaled horizontal alpha plane lines to be fed to the vertical scaler. |
| 321 | int vLumBufSize; ///< Number of vertical luma/alpha lines allocated in the ring buffer. |
| 322 | int vChrBufSize; ///< Number of vertical chroma lines allocated in the ring buffer. |
| 323 | int lastInLumBuf; ///< Last scaled horizontal luma/alpha line from source in the ring buffer. |
| 324 | int lastInChrBuf; ///< Last scaled horizontal chroma line from source in the ring buffer. |
| 325 | int lumBufIndex; ///< Index in ring buffer of the last scaled horizontal luma/alpha line from source. |
| 326 | int chrBufIndex; ///< Index in ring buffer of the last scaled horizontal chroma line from source. |
| 327 | //@} |
| 328 | |
| 329 | uint8_t *formatConvBuffer; |
| 330 | |
| 331 | /** |
| 332 | * @name Horizontal and vertical filters. |
| 333 | * To better understand the following fields, here is a pseudo-code of |
| 334 | * their usage in filtering a horizontal line: |
| 335 | * @code |
| 336 | * for (i = 0; i < width; i++) { |
| 337 | * dst[i] = 0; |
| 338 | * for (j = 0; j < filterSize; j++) |
| 339 | * dst[i] += src[ filterPos[i] + j ] * filter[ filterSize * i + j ]; |
| 340 | * dst[i] >>= FRAC_BITS; // The actual implementation is fixed-point. |
| 341 | * } |
| 342 | * @endcode |
| 343 | */ |
| 344 | //@{ |
| 345 | int16_t *hLumFilter; ///< Array of horizontal filter coefficients for luma/alpha planes. |
| 346 | int16_t *hChrFilter; ///< Array of horizontal filter coefficients for chroma planes. |
| 347 | int16_t *vLumFilter; ///< Array of vertical filter coefficients for luma/alpha planes. |
| 348 | int16_t *vChrFilter; ///< Array of vertical filter coefficients for chroma planes. |
| 349 | int32_t *hLumFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes. |
| 350 | int32_t *hChrFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes. |
| 351 | int32_t *vLumFilterPos; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes. |
| 352 | int32_t *vChrFilterPos; ///< Array of vertical filter starting positions for each dst[i] for chroma planes. |
| 353 | int hLumFilterSize; ///< Horizontal filter size for luma/alpha pixels. |
| 354 | int hChrFilterSize; ///< Horizontal filter size for chroma pixels. |
| 355 | int vLumFilterSize; ///< Vertical filter size for luma/alpha pixels. |
| 356 | int vChrFilterSize; ///< Vertical filter size for chroma pixels. |
| 357 | //@} |
| 358 | |
| 359 | int lumMmxextFilterCodeSize; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code size for luma/alpha planes. |
| 360 | int chrMmxextFilterCodeSize; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code size for chroma planes. |
| 361 | uint8_t *lumMmxextFilterCode; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code for luma/alpha planes. |
| 362 | uint8_t *chrMmxextFilterCode; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code for chroma planes. |
| 363 | |
| 364 | int canMMXEXTBeUsed; |
| 365 | |
| 366 | int dstY; ///< Last destination vertical line output from last slice. |
| 367 | int flags; ///< Flags passed by the user to select scaler algorithm, optimizations, subsampling, etc... |
| 368 | void *yuvTable; // pointer to the yuv->rgb table start so it can be freed() |
| 369 | // alignment ensures the offset can be added in a single |
| 370 | // instruction on e.g. ARM |
| 371 | DECLARE_ALIGNED(16, int, table_gV)[256 + 2*YUVRGB_TABLE_HEADROOM]; |
| 372 | uint8_t *table_rV[256 + 2*YUVRGB_TABLE_HEADROOM]; |
| 373 | uint8_t *table_gU[256 + 2*YUVRGB_TABLE_HEADROOM]; |
| 374 | uint8_t *table_bU[256 + 2*YUVRGB_TABLE_HEADROOM]; |
| 375 | DECLARE_ALIGNED(16, int32_t, input_rgb2yuv_table)[16+40*4]; // This table can contain both C and SIMD formatted values, the C vales are always at the XY_IDX points |
| 376 | #define RY_IDX 0 |
| 377 | #define GY_IDX 1 |
| 378 | #define BY_IDX 2 |
| 379 | #define RU_IDX 3 |
| 380 | #define GU_IDX 4 |
| 381 | #define BU_IDX 5 |
| 382 | #define RV_IDX 6 |
| 383 | #define GV_IDX 7 |
| 384 | #define BV_IDX 8 |
| 385 | #define RGB2YUV_SHIFT 15 |
| 386 | |
| 387 | int *dither_error[4]; |
| 388 | |
| 389 | //Colorspace stuff |
| 390 | int contrast, brightness, saturation; // for sws_getColorspaceDetails |
| 391 | int srcColorspaceTable[4]; |
| 392 | int dstColorspaceTable[4]; |
| 393 | int srcRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (source image). |
| 394 | int dstRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (destination image). |
| 395 | int src0Alpha; |
| 396 | int dst0Alpha; |
| 397 | int srcXYZ; |
| 398 | int dstXYZ; |
| 399 | int src_h_chr_pos; |
| 400 | int dst_h_chr_pos; |
| 401 | int src_v_chr_pos; |
| 402 | int dst_v_chr_pos; |
| 403 | int yuv2rgb_y_offset; |
| 404 | int yuv2rgb_y_coeff; |
| 405 | int yuv2rgb_v2r_coeff; |
| 406 | int yuv2rgb_v2g_coeff; |
| 407 | int yuv2rgb_u2g_coeff; |
| 408 | int yuv2rgb_u2b_coeff; |
| 409 | |
| 410 | #define RED_DITHER "0*8" |
| 411 | #define GREEN_DITHER "1*8" |
| 412 | #define BLUE_DITHER "2*8" |
| 413 | #define Y_COEFF "3*8" |
| 414 | #define VR_COEFF "4*8" |
| 415 | #define UB_COEFF "5*8" |
| 416 | #define VG_COEFF "6*8" |
| 417 | #define UG_COEFF "7*8" |
| 418 | #define Y_OFFSET "8*8" |
| 419 | #define U_OFFSET "9*8" |
| 420 | #define V_OFFSET "10*8" |
| 421 | #define LUM_MMX_FILTER_OFFSET "11*8" |
| 422 | #define CHR_MMX_FILTER_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE) |
| 423 | #define DSTW_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2" |
| 424 | #define ESP_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+8" |
| 425 | #define VROUNDER_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+16" |
| 426 | #define U_TEMP "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+24" |
| 427 | #define V_TEMP "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+32" |
| 428 | #define Y_TEMP "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+40" |
| 429 | #define ALP_MMX_FILTER_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+48" |
| 430 | #define UV_OFF_PX "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+48" |
| 431 | #define UV_OFF_BYTE "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+56" |
| 432 | #define DITHER16 "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+64" |
| 433 | #define DITHER32 "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+80" |
| 434 | #define DITHER32_INT (11*8+4*4*MAX_FILTER_SIZE*3+80) // value equal to above, used for checking that the struct hasn't been changed by mistake |
| 435 | |
| 436 | DECLARE_ALIGNED(8, uint64_t, redDither); |
| 437 | DECLARE_ALIGNED(8, uint64_t, greenDither); |
| 438 | DECLARE_ALIGNED(8, uint64_t, blueDither); |
| 439 | |
| 440 | DECLARE_ALIGNED(8, uint64_t, yCoeff); |
| 441 | DECLARE_ALIGNED(8, uint64_t, vrCoeff); |
| 442 | DECLARE_ALIGNED(8, uint64_t, ubCoeff); |
| 443 | DECLARE_ALIGNED(8, uint64_t, vgCoeff); |
| 444 | DECLARE_ALIGNED(8, uint64_t, ugCoeff); |
| 445 | DECLARE_ALIGNED(8, uint64_t, yOffset); |
| 446 | DECLARE_ALIGNED(8, uint64_t, uOffset); |
| 447 | DECLARE_ALIGNED(8, uint64_t, vOffset); |
| 448 | int32_t lumMmxFilter[4 * MAX_FILTER_SIZE]; |
| 449 | int32_t chrMmxFilter[4 * MAX_FILTER_SIZE]; |
| 450 | int dstW; ///< Width of destination luma/alpha planes. |
| 451 | DECLARE_ALIGNED(8, uint64_t, esp); |
| 452 | DECLARE_ALIGNED(8, uint64_t, vRounder); |
| 453 | DECLARE_ALIGNED(8, uint64_t, u_temp); |
| 454 | DECLARE_ALIGNED(8, uint64_t, v_temp); |
| 455 | DECLARE_ALIGNED(8, uint64_t, y_temp); |
| 456 | int32_t alpMmxFilter[4 * MAX_FILTER_SIZE]; |
| 457 | // alignment of these values is not necessary, but merely here |
| 458 | // to maintain the same offset across x8632 and x86-64. Once we |
| 459 | // use proper offset macros in the asm, they can be removed. |
| 460 | DECLARE_ALIGNED(8, ptrdiff_t, uv_off); ///< offset (in pixels) between u and v planes |
| 461 | DECLARE_ALIGNED(8, ptrdiff_t, uv_offx2); ///< offset (in bytes) between u and v planes |
| 462 | DECLARE_ALIGNED(8, uint16_t, dither16)[8]; |
| 463 | DECLARE_ALIGNED(8, uint32_t, dither32)[8]; |
| 464 | |
| 465 | const uint8_t *chrDither8, *lumDither8; |
| 466 | |
| 467 | #if HAVE_ALTIVEC |
| 468 | vector signed short CY; |
| 469 | vector signed short CRV; |
| 470 | vector signed short CBU; |
| 471 | vector signed short CGU; |
| 472 | vector signed short CGV; |
| 473 | vector signed short OY; |
| 474 | vector unsigned short CSHIFT; |
| 475 | vector signed short *vYCoeffsBank, *vCCoeffsBank; |
| 476 | #endif |
| 477 | |
| 478 | int use_mmx_vfilter; |
| 479 | |
| 480 | /* pre defined color-spaces gamma */ |
| 481 | #define XYZ_GAMMA (2.6f) |
| 482 | #define RGB_GAMMA (2.2f) |
| 483 | int16_t *xyzgamma; |
| 484 | int16_t *rgbgamma; |
| 485 | int16_t *xyzgammainv; |
| 486 | int16_t *rgbgammainv; |
| 487 | int16_t xyz2rgb_matrix[3][4]; |
| 488 | int16_t rgb2xyz_matrix[3][4]; |
| 489 | |
| 490 | /* function pointers for swscale() */ |
| 491 | yuv2planar1_fn yuv2plane1; |
| 492 | yuv2planarX_fn yuv2planeX; |
| 493 | yuv2interleavedX_fn yuv2nv12cX; |
| 494 | yuv2packed1_fn yuv2packed1; |
| 495 | yuv2packed2_fn yuv2packed2; |
| 496 | yuv2packedX_fn yuv2packedX; |
| 497 | yuv2anyX_fn yuv2anyX; |
| 498 | |
| 499 | /// Unscaled conversion of luma plane to YV12 for horizontal scaler. |
| 500 | void (*lumToYV12)(uint8_t *dst, const uint8_t *src, const uint8_t *src2, const uint8_t *src3, |
| 501 | int width, uint32_t *pal); |
| 502 | /// Unscaled conversion of alpha plane to YV12 for horizontal scaler. |
| 503 | void (*alpToYV12)(uint8_t *dst, const uint8_t *src, const uint8_t *src2, const uint8_t *src3, |
| 504 | int width, uint32_t *pal); |
| 505 | /// Unscaled conversion of chroma planes to YV12 for horizontal scaler. |
| 506 | void (*chrToYV12)(uint8_t *dstU, uint8_t *dstV, |
| 507 | const uint8_t *src1, const uint8_t *src2, const uint8_t *src3, |
| 508 | int width, uint32_t *pal); |
| 509 | |
| 510 | /** |
| 511 | * Functions to read planar input, such as planar RGB, and convert |
| 512 | * internally to Y/UV/A. |
| 513 | */ |
| 514 | /** @{ */ |
| 515 | void (*readLumPlanar)(uint8_t *dst, const uint8_t *src[4], int width, int32_t *rgb2yuv); |
| 516 | void (*readChrPlanar)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src[4], |
| 517 | int width, int32_t *rgb2yuv); |
| 518 | void (*readAlpPlanar)(uint8_t *dst, const uint8_t *src[4], int width, int32_t *rgb2yuv); |
| 519 | /** @} */ |
| 520 | |
| 521 | /** |
| 522 | * Scale one horizontal line of input data using a bilinear filter |
| 523 | * to produce one line of output data. Compared to SwsContext->hScale(), |
| 524 | * please take note of the following caveats when using these: |
| 525 | * - Scaling is done using only 7bit instead of 14bit coefficients. |
| 526 | * - You can use no more than 5 input pixels to produce 4 output |
| 527 | * pixels. Therefore, this filter should not be used for downscaling |
| 528 | * by more than ~20% in width (because that equals more than 5/4th |
| 529 | * downscaling and thus more than 5 pixels input per 4 pixels output). |
| 530 | * - In general, bilinear filters create artifacts during downscaling |
| 531 | * (even when <20%), because one output pixel will span more than one |
| 532 | * input pixel, and thus some pixels will need edges of both neighbor |
| 533 | * pixels to interpolate the output pixel. Since you can use at most |
| 534 | * two input pixels per output pixel in bilinear scaling, this is |
| 535 | * impossible and thus downscaling by any size will create artifacts. |
| 536 | * To enable this type of scaling, set SWS_FLAG_FAST_BILINEAR |
| 537 | * in SwsContext->flags. |
| 538 | */ |
| 539 | /** @{ */ |
| 540 | void (*hyscale_fast)(struct SwsContext *c, |
| 541 | int16_t *dst, int dstWidth, |
| 542 | const uint8_t *src, int srcW, int xInc); |
| 543 | void (*hcscale_fast)(struct SwsContext *c, |
| 544 | int16_t *dst1, int16_t *dst2, int dstWidth, |
| 545 | const uint8_t *src1, const uint8_t *src2, |
| 546 | int srcW, int xInc); |
| 547 | /** @} */ |
| 548 | |
| 549 | /** |
| 550 | * Scale one horizontal line of input data using a filter over the input |
| 551 | * lines, to produce one (differently sized) line of output data. |
| 552 | * |
| 553 | * @param dst pointer to destination buffer for horizontally scaled |
| 554 | * data. If the number of bits per component of one |
| 555 | * destination pixel (SwsContext->dstBpc) is <= 10, data |
| 556 | * will be 15bpc in 16bits (int16_t) width. Else (i.e. |
| 557 | * SwsContext->dstBpc == 16), data will be 19bpc in |
| 558 | * 32bits (int32_t) width. |
| 559 | * @param dstW width of destination image |
| 560 | * @param src pointer to source data to be scaled. If the number of |
| 561 | * bits per component of a source pixel (SwsContext->srcBpc) |
| 562 | * is 8, this is 8bpc in 8bits (uint8_t) width. Else |
| 563 | * (i.e. SwsContext->dstBpc > 8), this is native depth |
| 564 | * in 16bits (uint16_t) width. In other words, for 9-bit |
| 565 | * YUV input, this is 9bpc, for 10-bit YUV input, this is |
| 566 | * 10bpc, and for 16-bit RGB or YUV, this is 16bpc. |
| 567 | * @param filter filter coefficients to be used per output pixel for |
| 568 | * scaling. This contains 14bpp filtering coefficients. |
| 569 | * Guaranteed to contain dstW * filterSize entries. |
| 570 | * @param filterPos position of the first input pixel to be used for |
| 571 | * each output pixel during scaling. Guaranteed to |
| 572 | * contain dstW entries. |
| 573 | * @param filterSize the number of input coefficients to be used (and |
| 574 | * thus the number of input pixels to be used) for |
| 575 | * creating a single output pixel. Is aligned to 4 |
| 576 | * (and input coefficients thus padded with zeroes) |
| 577 | * to simplify creating SIMD code. |
| 578 | */ |
| 579 | /** @{ */ |
| 580 | void (*hyScale)(struct SwsContext *c, int16_t *dst, int dstW, |
| 581 | const uint8_t *src, const int16_t *filter, |
| 582 | const int32_t *filterPos, int filterSize); |
| 583 | void (*hcScale)(struct SwsContext *c, int16_t *dst, int dstW, |
| 584 | const uint8_t *src, const int16_t *filter, |
| 585 | const int32_t *filterPos, int filterSize); |
| 586 | /** @} */ |
| 587 | |
| 588 | /// Color range conversion function for luma plane if needed. |
| 589 | void (*lumConvertRange)(int16_t *dst, int width); |
| 590 | /// Color range conversion function for chroma planes if needed. |
| 591 | void (*chrConvertRange)(int16_t *dst1, int16_t *dst2, int width); |
| 592 | |
| 593 | int needs_hcscale; ///< Set if there are chroma planes to be converted. |
| 594 | |
| 595 | SwsDither dither; |
| 596 | } SwsContext; |
| 597 | //FIXME check init (where 0) |
| 598 | |
| 599 | SwsFunc ff_yuv2rgb_get_func_ptr(SwsContext *c); |
| 600 | int ff_yuv2rgb_c_init_tables(SwsContext *c, const int inv_table[4], |
| 601 | int fullRange, int brightness, |
| 602 | int contrast, int saturation); |
| 603 | void ff_yuv2rgb_init_tables_ppc(SwsContext *c, const int inv_table[4], |
| 604 | int brightness, int contrast, int saturation); |
| 605 | |
| 606 | void updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufIndex, |
| 607 | int lastInLumBuf, int lastInChrBuf); |
| 608 | |
| 609 | av_cold void ff_sws_init_range_convert(SwsContext *c); |
| 610 | |
| 611 | SwsFunc ff_yuv2rgb_init_x86(SwsContext *c); |
| 612 | SwsFunc ff_yuv2rgb_init_ppc(SwsContext *c); |
| 613 | |
| 614 | #if FF_API_SWS_FORMAT_NAME |
| 615 | /** |
| 616 | * @deprecated Use av_get_pix_fmt_name() instead. |
| 617 | */ |
| 618 | attribute_deprecated |
| 619 | const char *sws_format_name(enum AVPixelFormat format); |
| 620 | #endif |
| 621 | |
| 622 | static av_always_inline int is16BPS(enum AVPixelFormat pix_fmt) |
| 623 | { |
| 624 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 625 | av_assert0(desc); |
| 626 | return desc->comp[0].depth_minus1 == 15; |
| 627 | } |
| 628 | |
| 629 | static av_always_inline int is9_OR_10BPS(enum AVPixelFormat pix_fmt) |
| 630 | { |
| 631 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 632 | av_assert0(desc); |
| 633 | return desc->comp[0].depth_minus1 >= 8 && desc->comp[0].depth_minus1 <= 13; |
| 634 | } |
| 635 | |
| 636 | #define isNBPS(x) is9_OR_10BPS(x) |
| 637 | |
| 638 | static av_always_inline int isBE(enum AVPixelFormat pix_fmt) |
| 639 | { |
| 640 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 641 | av_assert0(desc); |
| 642 | return desc->flags & AV_PIX_FMT_FLAG_BE; |
| 643 | } |
| 644 | |
| 645 | static av_always_inline int isYUV(enum AVPixelFormat pix_fmt) |
| 646 | { |
| 647 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 648 | av_assert0(desc); |
| 649 | return !(desc->flags & AV_PIX_FMT_FLAG_RGB) && desc->nb_components >= 2; |
| 650 | } |
| 651 | |
| 652 | static av_always_inline int isPlanarYUV(enum AVPixelFormat pix_fmt) |
| 653 | { |
| 654 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 655 | av_assert0(desc); |
| 656 | return ((desc->flags & AV_PIX_FMT_FLAG_PLANAR) && isYUV(pix_fmt)); |
| 657 | } |
| 658 | |
| 659 | static av_always_inline int isRGB(enum AVPixelFormat pix_fmt) |
| 660 | { |
| 661 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 662 | av_assert0(desc); |
| 663 | return (desc->flags & AV_PIX_FMT_FLAG_RGB); |
| 664 | } |
| 665 | |
| 666 | #if 0 // FIXME |
| 667 | #define isGray(x) \ |
| 668 | (!(av_pix_fmt_desc_get(x)->flags & AV_PIX_FMT_FLAG_PAL) && \ |
| 669 | av_pix_fmt_desc_get(x)->nb_components <= 2) |
| 670 | #else |
| 671 | #define isGray(x) \ |
| 672 | ((x) == AV_PIX_FMT_GRAY8 || \ |
| 673 | (x) == AV_PIX_FMT_YA8 || \ |
| 674 | (x) == AV_PIX_FMT_GRAY16BE || \ |
| 675 | (x) == AV_PIX_FMT_GRAY16LE || \ |
| 676 | (x) == AV_PIX_FMT_YA16BE || \ |
| 677 | (x) == AV_PIX_FMT_YA16LE) |
| 678 | #endif |
| 679 | |
| 680 | #define isRGBinInt(x) \ |
| 681 | ( \ |
| 682 | (x) == AV_PIX_FMT_RGB48BE || \ |
| 683 | (x) == AV_PIX_FMT_RGB48LE || \ |
| 684 | (x) == AV_PIX_FMT_RGB32 || \ |
| 685 | (x) == AV_PIX_FMT_RGB32_1 || \ |
| 686 | (x) == AV_PIX_FMT_RGB24 || \ |
| 687 | (x) == AV_PIX_FMT_RGB565BE || \ |
| 688 | (x) == AV_PIX_FMT_RGB565LE || \ |
| 689 | (x) == AV_PIX_FMT_RGB555BE || \ |
| 690 | (x) == AV_PIX_FMT_RGB555LE || \ |
| 691 | (x) == AV_PIX_FMT_RGB444BE || \ |
| 692 | (x) == AV_PIX_FMT_RGB444LE || \ |
| 693 | (x) == AV_PIX_FMT_RGB8 || \ |
| 694 | (x) == AV_PIX_FMT_RGB4 || \ |
| 695 | (x) == AV_PIX_FMT_RGB4_BYTE || \ |
| 696 | (x) == AV_PIX_FMT_RGBA64BE || \ |
| 697 | (x) == AV_PIX_FMT_RGBA64LE || \ |
| 698 | (x) == AV_PIX_FMT_MONOBLACK || \ |
| 699 | (x) == AV_PIX_FMT_MONOWHITE \ |
| 700 | ) |
| 701 | #define isBGRinInt(x) \ |
| 702 | ( \ |
| 703 | (x) == AV_PIX_FMT_BGR48BE || \ |
| 704 | (x) == AV_PIX_FMT_BGR48LE || \ |
| 705 | (x) == AV_PIX_FMT_BGR32 || \ |
| 706 | (x) == AV_PIX_FMT_BGR32_1 || \ |
| 707 | (x) == AV_PIX_FMT_BGR24 || \ |
| 708 | (x) == AV_PIX_FMT_BGR565BE || \ |
| 709 | (x) == AV_PIX_FMT_BGR565LE || \ |
| 710 | (x) == AV_PIX_FMT_BGR555BE || \ |
| 711 | (x) == AV_PIX_FMT_BGR555LE || \ |
| 712 | (x) == AV_PIX_FMT_BGR444BE || \ |
| 713 | (x) == AV_PIX_FMT_BGR444LE || \ |
| 714 | (x) == AV_PIX_FMT_BGR8 || \ |
| 715 | (x) == AV_PIX_FMT_BGR4 || \ |
| 716 | (x) == AV_PIX_FMT_BGR4_BYTE || \ |
| 717 | (x) == AV_PIX_FMT_BGRA64BE || \ |
| 718 | (x) == AV_PIX_FMT_BGRA64LE || \ |
| 719 | (x) == AV_PIX_FMT_MONOBLACK || \ |
| 720 | (x) == AV_PIX_FMT_MONOWHITE \ |
| 721 | ) |
| 722 | |
| 723 | #define isRGBinBytes(x) ( \ |
| 724 | (x) == AV_PIX_FMT_RGB48BE \ |
| 725 | || (x) == AV_PIX_FMT_RGB48LE \ |
| 726 | || (x) == AV_PIX_FMT_RGBA64BE \ |
| 727 | || (x) == AV_PIX_FMT_RGBA64LE \ |
| 728 | || (x) == AV_PIX_FMT_RGBA \ |
| 729 | || (x) == AV_PIX_FMT_ARGB \ |
| 730 | || (x) == AV_PIX_FMT_RGB24 \ |
| 731 | ) |
| 732 | #define isBGRinBytes(x) ( \ |
| 733 | (x) == AV_PIX_FMT_BGR48BE \ |
| 734 | || (x) == AV_PIX_FMT_BGR48LE \ |
| 735 | || (x) == AV_PIX_FMT_BGRA64BE \ |
| 736 | || (x) == AV_PIX_FMT_BGRA64LE \ |
| 737 | || (x) == AV_PIX_FMT_BGRA \ |
| 738 | || (x) == AV_PIX_FMT_ABGR \ |
| 739 | || (x) == AV_PIX_FMT_BGR24 \ |
| 740 | ) |
| 741 | |
| 742 | #define isBayer(x) ( \ |
| 743 | (x)==AV_PIX_FMT_BAYER_BGGR8 \ |
| 744 | || (x)==AV_PIX_FMT_BAYER_BGGR16LE \ |
| 745 | || (x)==AV_PIX_FMT_BAYER_BGGR16BE \ |
| 746 | || (x)==AV_PIX_FMT_BAYER_RGGB8 \ |
| 747 | || (x)==AV_PIX_FMT_BAYER_RGGB16LE \ |
| 748 | || (x)==AV_PIX_FMT_BAYER_RGGB16BE \ |
| 749 | || (x)==AV_PIX_FMT_BAYER_GBRG8 \ |
| 750 | || (x)==AV_PIX_FMT_BAYER_GBRG16LE \ |
| 751 | || (x)==AV_PIX_FMT_BAYER_GBRG16BE \ |
| 752 | || (x)==AV_PIX_FMT_BAYER_GRBG8 \ |
| 753 | || (x)==AV_PIX_FMT_BAYER_GRBG16LE \ |
| 754 | || (x)==AV_PIX_FMT_BAYER_GRBG16BE \ |
| 755 | ) |
| 756 | |
| 757 | #define isAnyRGB(x) \ |
| 758 | ( \ |
| 759 | isBayer(x) || \ |
| 760 | isRGBinInt(x) || \ |
| 761 | isBGRinInt(x) || \ |
| 762 | isRGB(x) \ |
| 763 | ) |
| 764 | |
| 765 | static av_always_inline int isALPHA(enum AVPixelFormat pix_fmt) |
| 766 | { |
| 767 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 768 | av_assert0(desc); |
| 769 | if (pix_fmt == AV_PIX_FMT_PAL8) |
| 770 | return 1; |
| 771 | return desc->flags & AV_PIX_FMT_FLAG_ALPHA; |
| 772 | } |
| 773 | |
| 774 | #if 1 |
| 775 | #define isPacked(x) ( \ |
| 776 | (x)==AV_PIX_FMT_PAL8 \ |
| 777 | || (x)==AV_PIX_FMT_YUYV422 \ |
| 778 | || (x)==AV_PIX_FMT_YVYU422 \ |
| 779 | || (x)==AV_PIX_FMT_UYVY422 \ |
| 780 | || (x)==AV_PIX_FMT_YA8 \ |
| 781 | || (x)==AV_PIX_FMT_YA16LE \ |
| 782 | || (x)==AV_PIX_FMT_YA16BE \ |
| 783 | || isRGBinInt(x) \ |
| 784 | || isBGRinInt(x) \ |
| 785 | ) |
| 786 | #else |
| 787 | static av_always_inline int isPacked(enum AVPixelFormat pix_fmt) |
| 788 | { |
| 789 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 790 | av_assert0(desc); |
| 791 | return ((desc->nb_components >= 2 && !(desc->flags & AV_PIX_FMT_FLAG_PLANAR)) || |
| 792 | pix_fmt == AV_PIX_FMT_PAL8); |
| 793 | } |
| 794 | |
| 795 | #endif |
| 796 | static av_always_inline int isPlanar(enum AVPixelFormat pix_fmt) |
| 797 | { |
| 798 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 799 | av_assert0(desc); |
| 800 | return (desc->nb_components >= 2 && (desc->flags & AV_PIX_FMT_FLAG_PLANAR)); |
| 801 | } |
| 802 | |
| 803 | static av_always_inline int isPackedRGB(enum AVPixelFormat pix_fmt) |
| 804 | { |
| 805 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 806 | av_assert0(desc); |
| 807 | return ((desc->flags & (AV_PIX_FMT_FLAG_PLANAR | AV_PIX_FMT_FLAG_RGB)) == AV_PIX_FMT_FLAG_RGB); |
| 808 | } |
| 809 | |
| 810 | static av_always_inline int isPlanarRGB(enum AVPixelFormat pix_fmt) |
| 811 | { |
| 812 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 813 | av_assert0(desc); |
| 814 | return ((desc->flags & (AV_PIX_FMT_FLAG_PLANAR | AV_PIX_FMT_FLAG_RGB)) == |
| 815 | (AV_PIX_FMT_FLAG_PLANAR | AV_PIX_FMT_FLAG_RGB)); |
| 816 | } |
| 817 | |
| 818 | static av_always_inline int usePal(enum AVPixelFormat pix_fmt) |
| 819 | { |
| 820 | const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt); |
| 821 | av_assert0(desc); |
| 822 | return (desc->flags & AV_PIX_FMT_FLAG_PAL) || (desc->flags & AV_PIX_FMT_FLAG_PSEUDOPAL); |
| 823 | } |
| 824 | |
| 825 | extern const uint64_t ff_dither4[2]; |
| 826 | extern const uint64_t ff_dither8[2]; |
| 827 | |
| 828 | extern const uint8_t ff_dither_2x2_4[3][8]; |
| 829 | extern const uint8_t ff_dither_2x2_8[3][8]; |
| 830 | extern const uint8_t ff_dither_4x4_16[5][8]; |
| 831 | extern const uint8_t ff_dither_8x8_32[9][8]; |
| 832 | extern const uint8_t ff_dither_8x8_73[9][8]; |
| 833 | extern const uint8_t ff_dither_8x8_128[9][8]; |
| 834 | extern const uint8_t ff_dither_8x8_220[9][8]; |
| 835 | |
| 836 | extern const int32_t ff_yuv2rgb_coeffs[8][4]; |
| 837 | |
| 838 | extern const AVClass sws_context_class; |
| 839 | |
| 840 | /** |
| 841 | * Set c->swscale to an unscaled converter if one exists for the specific |
| 842 | * source and destination formats, bit depths, flags, etc. |
| 843 | */ |
| 844 | void ff_get_unscaled_swscale(SwsContext *c); |
| 845 | void ff_get_unscaled_swscale_ppc(SwsContext *c); |
| 846 | void ff_get_unscaled_swscale_arm(SwsContext *c); |
| 847 | |
| 848 | /** |
| 849 | * Return function pointer to fastest main scaler path function depending |
| 850 | * on architecture and available optimizations. |
| 851 | */ |
| 852 | SwsFunc ff_getSwsFunc(SwsContext *c); |
| 853 | |
| 854 | void ff_sws_init_input_funcs(SwsContext *c); |
| 855 | void ff_sws_init_output_funcs(SwsContext *c, |
| 856 | yuv2planar1_fn *yuv2plane1, |
| 857 | yuv2planarX_fn *yuv2planeX, |
| 858 | yuv2interleavedX_fn *yuv2nv12cX, |
| 859 | yuv2packed1_fn *yuv2packed1, |
| 860 | yuv2packed2_fn *yuv2packed2, |
| 861 | yuv2packedX_fn *yuv2packedX, |
| 862 | yuv2anyX_fn *yuv2anyX); |
| 863 | void ff_sws_init_swscale_ppc(SwsContext *c); |
| 864 | void ff_sws_init_swscale_x86(SwsContext *c); |
| 865 | |
| 866 | void ff_hyscale_fast_c(SwsContext *c, int16_t *dst, int dstWidth, |
| 867 | const uint8_t *src, int srcW, int xInc); |
| 868 | void ff_hcscale_fast_c(SwsContext *c, int16_t *dst1, int16_t *dst2, |
| 869 | int dstWidth, const uint8_t *src1, |
| 870 | const uint8_t *src2, int srcW, int xInc); |
| 871 | int ff_init_hscaler_mmxext(int dstW, int xInc, uint8_t *filterCode, |
| 872 | int16_t *filter, int32_t *filterPos, |
| 873 | int numSplits); |
| 874 | void ff_hyscale_fast_mmxext(SwsContext *c, int16_t *dst, |
| 875 | int dstWidth, const uint8_t *src, |
| 876 | int srcW, int xInc); |
| 877 | void ff_hcscale_fast_mmxext(SwsContext *c, int16_t *dst1, int16_t *dst2, |
| 878 | int dstWidth, const uint8_t *src1, |
| 879 | const uint8_t *src2, int srcW, int xInc); |
| 880 | |
| 881 | static inline void fillPlane16(uint8_t *plane, int stride, int width, int height, int y, |
| 882 | int alpha, int bits, const int big_endian) |
| 883 | { |
| 884 | int i, j; |
| 885 | uint8_t *ptr = plane + stride * y; |
| 886 | int v = alpha ? 0xFFFF>>(15-bits) : (1<<bits); |
| 887 | for (i = 0; i < height; i++) { |
| 888 | #define FILL(wfunc) \ |
| 889 | for (j = 0; j < width; j++) {\ |
| 890 | wfunc(ptr+2*j, v);\ |
| 891 | } |
| 892 | if (big_endian) { |
| 893 | FILL(AV_WB16); |
| 894 | } else { |
| 895 | FILL(AV_WL16); |
| 896 | } |
| 897 | ptr += stride; |
| 898 | } |
| 899 | } |
| 900 | |
| 901 | #endif /* SWSCALE_SWSCALE_INTERNAL_H */ |