2 * Copyright (C) 2001-2003 Michael Niedermayer <michaelni@gmx.at>
4 * This file is part of FFmpeg.
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.
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.
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
21 #ifndef SWSCALE_SWSCALE_INTERNAL_H
22 #define SWSCALE_SWSCALE_INTERNAL_H
30 #include "libavutil/avutil.h"
31 #include "libavutil/log.h"
32 #include "libavutil/pixfmt.h"
34 #define STR(s) AV_TOSTRING(s) //AV_STRINGIFY is too long
36 #define FAST_BGR2YV12 //use 7-bit instead of 15-bit coefficients
38 #define MAX_FILTER_SIZE 256
43 #define ALT32_CORR (-1)
60 typedef int (*SwsFunc)(struct SwsContext *context, const uint8_t* src[],
61 int srcStride[], int srcSliceY, int srcSliceH,
62 uint8_t* dst[], int dstStride[]);
65 * Write one line of horizontally scaled Y/U/V/A to planar output
66 * without any additional vertical scaling (or point-scaling).
68 * @param c SWS scaling context
69 * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output,
70 * 19-bit for 16bit output (in int32_t)
71 * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output,
72 * 19-bit for 16bit output (in int32_t)
73 * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output,
74 * 19-bit for 16bit output (in int32_t)
75 * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output,
76 * 19-bit for 16bit output (in int32_t)
77 * @param dest pointer to the 4 output planes (Y/U/V/A). For >8bit
78 * output, this is in uint16_t
79 * @param dstW width of dest[0], dest[3], lumSrc and alpSrc in pixels
80 * @param chrDstW width of dest[1], dest[2], chrUSrc and chrVSrc
82 typedef void (*yuv2planar1_fn) (struct SwsContext *c,
83 const int16_t *lumSrc, const int16_t *chrUSrc,
84 const int16_t *chrVSrc, const int16_t *alpSrc,
85 uint8_t *dest[4], int dstW, int chrDstW);
87 * Write one line of horizontally scaled Y/U/V/A to planar output
88 * with multi-point vertical scaling between input pixels.
90 * @param c SWS scaling context
91 * @param lumFilter vertical luma/alpha scaling coefficients, 12bit [0,4096]
92 * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output,
93 * 19-bit for 16bit output (in int32_t)
94 * @param lumFilterSize number of vertical luma/alpha input lines to scale
95 * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096]
96 * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output,
97 * 19-bit for 16bit output (in int32_t)
98 * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output,
99 * 19-bit for 16bit output (in int32_t)
100 * @param chrFilterSize number of vertical chroma input lines to scale
101 * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output,
102 * 19-bit for 16bit output (in int32_t)
103 * @param dest pointer to the 4 output planes (Y/U/V/A). For >8bit
104 * output, this is in uint16_t
105 * @param dstW width of dest[0], dest[3], lumSrc and alpSrc in pixels
106 * @param chrDstW width of dest[1], dest[2], chrUSrc and chrVSrc
108 typedef void (*yuv2planarX_fn) (struct SwsContext *c, const int16_t *lumFilter,
109 const int16_t **lumSrc, int lumFilterSize,
110 const int16_t *chrFilter, const int16_t **chrUSrc,
111 const int16_t **chrVSrc, int chrFilterSize,
112 const int16_t **alpSrc, uint8_t *dest[4],
113 int dstW, int chrDstW);
115 * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB
116 * output without any additional vertical scaling (or point-scaling). Note
117 * that this function may do chroma scaling, see the "uvalpha" argument.
119 * @param c SWS scaling context
120 * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output,
121 * 19-bit for 16bit output (in int32_t)
122 * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output,
123 * 19-bit for 16bit output (in int32_t)
124 * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output,
125 * 19-bit for 16bit output (in int32_t)
126 * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output,
127 * 19-bit for 16bit output (in int32_t)
128 * @param dest pointer to the output plane. For 16bit output, this is
130 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
131 * to write into dest[]
132 * @param uvalpha chroma scaling coefficient for the second line of chroma
133 * pixels, either 2048 or 0. If 0, one chroma input is used
134 * for 2 output pixels (or if the SWS_FLAG_FULL_CHR_INT flag
135 * is set, it generates 1 output pixel). If 2048, two chroma
136 * input pixels should be averaged for 2 output pixels (this
137 * only happens if SWS_FLAG_FULL_CHR_INT is not set)
138 * @param y vertical line number for this output. This does not need
139 * to be used to calculate the offset in the destination,
140 * but can be used to generate comfort noise using dithering
141 * for some output formats.
143 typedef void (*yuv2packed1_fn) (struct SwsContext *c, const int16_t *lumSrc,
144 const int16_t *chrUSrc[2], const int16_t *chrVSrc[2],
145 const int16_t *alpSrc, uint8_t *dest,
146 int dstW, int uvalpha, int y);
148 * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB
149 * output by doing bilinear scaling between two input lines.
151 * @param c SWS scaling context
152 * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output,
153 * 19-bit for 16bit output (in int32_t)
154 * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output,
155 * 19-bit for 16bit output (in int32_t)
156 * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output,
157 * 19-bit for 16bit output (in int32_t)
158 * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output,
159 * 19-bit for 16bit output (in int32_t)
160 * @param dest pointer to the output plane. For 16bit output, this is
162 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
163 * to write into dest[]
164 * @param yalpha luma/alpha scaling coefficients for the second input line.
165 * The first line's coefficients can be calculated by using
167 * @param uvalpha chroma scaling coefficient for the second input line. The
168 * first line's coefficients can be calculated by using
170 * @param y vertical line number for this output. This does not need
171 * to be used to calculate the offset in the destination,
172 * but can be used to generate comfort noise using dithering
173 * for some output formats.
175 typedef void (*yuv2packed2_fn) (struct SwsContext *c, const int16_t *lumSrc[2],
176 const int16_t *chrUSrc[2], const int16_t *chrVSrc[2],
177 const int16_t *alpSrc[2], uint8_t *dest,
178 int dstW, int yalpha, int uvalpha, int y);
180 * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB
181 * output by doing multi-point vertical scaling between input pixels.
183 * @param c SWS scaling context
184 * @param lumFilter vertical luma/alpha scaling coefficients, 12bit [0,4096]
185 * @param lumSrc scaled luma (Y) source data, 15bit for 8-10bit output,
186 * 19-bit for 16bit output (in int32_t)
187 * @param lumFilterSize number of vertical luma/alpha input lines to scale
188 * @param chrFilter vertical chroma scaling coefficients, 12bit [0,4096]
189 * @param chrUSrc scaled chroma (U) source data, 15bit for 8-10bit output,
190 * 19-bit for 16bit output (in int32_t)
191 * @param chrVSrc scaled chroma (V) source data, 15bit for 8-10bit output,
192 * 19-bit for 16bit output (in int32_t)
193 * @param chrFilterSize number of vertical chroma input lines to scale
194 * @param alpSrc scaled alpha (A) source data, 15bit for 8-10bit output,
195 * 19-bit for 16bit output (in int32_t)
196 * @param dest pointer to the output plane. For 16bit output, this is
198 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
199 * to write into dest[]
200 * @param y vertical line number for this output. This does not need
201 * to be used to calculate the offset in the destination,
202 * but can be used to generate comfort noise using dithering
203 * or some output formats.
205 typedef void (*yuv2packedX_fn) (struct SwsContext *c, const int16_t *lumFilter,
206 const int16_t **lumSrc, int lumFilterSize,
207 const int16_t *chrFilter, const int16_t **chrUSrc,
208 const int16_t **chrVSrc, int chrFilterSize,
209 const int16_t **alpSrc, uint8_t *dest,
212 /* This struct should be aligned on at least a 32-byte boundary. */
213 typedef struct SwsContext {
215 * info on struct for av_log
217 const AVClass *av_class;
220 * Note that src, dst, srcStride, dstStride will be copied in the
221 * sws_scale() wrapper so they can be freely modified here.
224 int srcW; ///< Width of source luma/alpha planes.
225 int srcH; ///< Height of source luma/alpha planes.
226 int dstH; ///< Height of destination luma/alpha planes.
227 int chrSrcW; ///< Width of source chroma planes.
228 int chrSrcH; ///< Height of source chroma planes.
229 int chrDstW; ///< Width of destination chroma planes.
230 int chrDstH; ///< Height of destination chroma planes.
231 int lumXInc, chrXInc;
232 int lumYInc, chrYInc;
233 enum PixelFormat dstFormat; ///< Destination pixel format.
234 enum PixelFormat srcFormat; ///< Source pixel format.
235 int dstFormatBpp; ///< Number of bits per pixel of the destination pixel format.
236 int srcFormatBpp; ///< Number of bits per pixel of the source pixel format.
238 int chrSrcHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in source image.
239 int chrSrcVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in source image.
240 int chrDstHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in destination image.
241 int chrDstVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in destination image.
242 int vChrDrop; ///< Binary logarithm of extra vertical subsampling factor in source image chroma planes specified by user.
243 int sliceDir; ///< Direction that slices are fed to the scaler (1 = top-to-bottom, -1 = bottom-to-top).
244 double param[2]; ///< Input parameters for scaling algorithms that need them.
246 uint32_t pal_yuv[256];
247 uint32_t pal_rgb[256];
250 * @name Scaled horizontal lines ring buffer.
251 * The horizontal scaler keeps just enough scaled lines in a ring buffer
252 * so they may be passed to the vertical scaler. The pointers to the
253 * allocated buffers for each line are duplicated in sequence in the ring
254 * buffer to simplify indexing and avoid wrapping around between lines
255 * inside the vertical scaler code. The wrapping is done before the
256 * vertical scaler is called.
259 int16_t **lumPixBuf; ///< Ring buffer for scaled horizontal luma plane lines to be fed to the vertical scaler.
260 int16_t **chrUPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler.
261 int16_t **chrVPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler.
262 int16_t **alpPixBuf; ///< Ring buffer for scaled horizontal alpha plane lines to be fed to the vertical scaler.
263 int vLumBufSize; ///< Number of vertical luma/alpha lines allocated in the ring buffer.
264 int vChrBufSize; ///< Number of vertical chroma lines allocated in the ring buffer.
265 int lastInLumBuf; ///< Last scaled horizontal luma/alpha line from source in the ring buffer.
266 int lastInChrBuf; ///< Last scaled horizontal chroma line from source in the ring buffer.
267 int lumBufIndex; ///< Index in ring buffer of the last scaled horizontal luma/alpha line from source.
268 int chrBufIndex; ///< Index in ring buffer of the last scaled horizontal chroma line from source.
271 uint8_t *formatConvBuffer;
274 * @name Horizontal and vertical filters.
275 * To better understand the following fields, here is a pseudo-code of
276 * their usage in filtering a horizontal line:
278 * for (i = 0; i < width; i++) {
280 * for (j = 0; j < filterSize; j++)
281 * dst[i] += src[ filterPos[i] + j ] * filter[ filterSize * i + j ];
282 * dst[i] >>= FRAC_BITS; // The actual implementation is fixed-point.
287 int16_t *hLumFilter; ///< Array of horizontal filter coefficients for luma/alpha planes.
288 int16_t *hChrFilter; ///< Array of horizontal filter coefficients for chroma planes.
289 int16_t *vLumFilter; ///< Array of vertical filter coefficients for luma/alpha planes.
290 int16_t *vChrFilter; ///< Array of vertical filter coefficients for chroma planes.
291 int16_t *hLumFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes.
292 int16_t *hChrFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes.
293 int16_t *vLumFilterPos; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes.
294 int16_t *vChrFilterPos; ///< Array of vertical filter starting positions for each dst[i] for chroma planes.
295 int hLumFilterSize; ///< Horizontal filter size for luma/alpha pixels.
296 int hChrFilterSize; ///< Horizontal filter size for chroma pixels.
297 int vLumFilterSize; ///< Vertical filter size for luma/alpha pixels.
298 int vChrFilterSize; ///< Vertical filter size for chroma pixels.
301 int lumMmx2FilterCodeSize; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for luma/alpha planes.
302 int chrMmx2FilterCodeSize; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for chroma planes.
303 uint8_t *lumMmx2FilterCode; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code for luma/alpha planes.
304 uint8_t *chrMmx2FilterCode; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code for chroma planes.
308 int dstY; ///< Last destination vertical line output from last slice.
309 int flags; ///< Flags passed by the user to select scaler algorithm, optimizations, subsampling, etc...
310 void * yuvTable; // pointer to the yuv->rgb table start so it can be freed()
311 uint8_t * table_rV[256];
312 uint8_t * table_gU[256];
314 uint8_t * table_bU[256];
317 int contrast, brightness, saturation; // for sws_getColorspaceDetails
318 int srcColorspaceTable[4];
319 int dstColorspaceTable[4];
320 int srcRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (source image).
321 int dstRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (destination image).
322 int yuv2rgb_y_offset;
324 int yuv2rgb_v2r_coeff;
325 int yuv2rgb_v2g_coeff;
326 int yuv2rgb_u2g_coeff;
327 int yuv2rgb_u2b_coeff;
329 #define RED_DITHER "0*8"
330 #define GREEN_DITHER "1*8"
331 #define BLUE_DITHER "2*8"
332 #define Y_COEFF "3*8"
333 #define VR_COEFF "4*8"
334 #define UB_COEFF "5*8"
335 #define VG_COEFF "6*8"
336 #define UG_COEFF "7*8"
337 #define Y_OFFSET "8*8"
338 #define U_OFFSET "9*8"
339 #define V_OFFSET "10*8"
340 #define LUM_MMX_FILTER_OFFSET "11*8"
341 #define CHR_MMX_FILTER_OFFSET "11*8+4*4*256"
342 #define DSTW_OFFSET "11*8+4*4*256*2" //do not change, it is hardcoded in the ASM
343 #define ESP_OFFSET "11*8+4*4*256*2+8"
344 #define VROUNDER_OFFSET "11*8+4*4*256*2+16"
345 #define U_TEMP "11*8+4*4*256*2+24"
346 #define V_TEMP "11*8+4*4*256*2+32"
347 #define Y_TEMP "11*8+4*4*256*2+40"
348 #define ALP_MMX_FILTER_OFFSET "11*8+4*4*256*2+48"
349 #define UV_OFF "11*8+4*4*256*3+48"
350 #define UV_OFFx2 "11*8+4*4*256*3+56"
351 #define DITHER16 "11*8+4*4*256*3+64"
352 #define DITHER32 "11*8+4*4*256*3+80"
354 DECLARE_ALIGNED(8, uint64_t, redDither);
355 DECLARE_ALIGNED(8, uint64_t, greenDither);
356 DECLARE_ALIGNED(8, uint64_t, blueDither);
358 DECLARE_ALIGNED(8, uint64_t, yCoeff);
359 DECLARE_ALIGNED(8, uint64_t, vrCoeff);
360 DECLARE_ALIGNED(8, uint64_t, ubCoeff);
361 DECLARE_ALIGNED(8, uint64_t, vgCoeff);
362 DECLARE_ALIGNED(8, uint64_t, ugCoeff);
363 DECLARE_ALIGNED(8, uint64_t, yOffset);
364 DECLARE_ALIGNED(8, uint64_t, uOffset);
365 DECLARE_ALIGNED(8, uint64_t, vOffset);
366 int32_t lumMmxFilter[4*MAX_FILTER_SIZE];
367 int32_t chrMmxFilter[4*MAX_FILTER_SIZE];
368 int dstW; ///< Width of destination luma/alpha planes.
369 DECLARE_ALIGNED(8, uint64_t, esp);
370 DECLARE_ALIGNED(8, uint64_t, vRounder);
371 DECLARE_ALIGNED(8, uint64_t, u_temp);
372 DECLARE_ALIGNED(8, uint64_t, v_temp);
373 DECLARE_ALIGNED(8, uint64_t, y_temp);
374 int32_t alpMmxFilter[4*MAX_FILTER_SIZE];
375 // alignment of these values is not necessary, but merely here
376 // to maintain the same offset across x8632 and x86-64. Once we
377 // use proper offset macros in the asm, they can be removed.
378 DECLARE_ALIGNED(8, ptrdiff_t, uv_off); ///< offset (in pixels) between u and v planes
379 DECLARE_ALIGNED(8, ptrdiff_t, uv_offx2); ///< offset (in bytes) between u and v planes
380 DECLARE_ALIGNED(8, uint16_t, dither16)[8];
381 DECLARE_ALIGNED(8, uint32_t, dither32)[8];
383 const uint8_t *chrDither8, *lumDither8;
386 vector signed short CY;
387 vector signed short CRV;
388 vector signed short CBU;
389 vector signed short CGU;
390 vector signed short CGV;
391 vector signed short OY;
392 vector unsigned short CSHIFT;
393 vector signed short *vYCoeffsBank, *vCCoeffsBank;
397 DECLARE_ALIGNED(4, uint32_t, oy);
398 DECLARE_ALIGNED(4, uint32_t, oc);
399 DECLARE_ALIGNED(4, uint32_t, zero);
400 DECLARE_ALIGNED(4, uint32_t, cy);
401 DECLARE_ALIGNED(4, uint32_t, crv);
402 DECLARE_ALIGNED(4, uint32_t, rmask);
403 DECLARE_ALIGNED(4, uint32_t, cbu);
404 DECLARE_ALIGNED(4, uint32_t, bmask);
405 DECLARE_ALIGNED(4, uint32_t, cgu);
406 DECLARE_ALIGNED(4, uint32_t, cgv);
407 DECLARE_ALIGNED(4, uint32_t, gmask);
411 DECLARE_ALIGNED(8, uint64_t, sparc_coeffs)[10];
414 /* function pointers for swScale() */
415 yuv2planar1_fn yuv2yuv1;
416 yuv2planarX_fn yuv2yuvX;
417 yuv2packed1_fn yuv2packed1;
418 yuv2packed2_fn yuv2packed2;
419 yuv2packedX_fn yuv2packedX;
421 void (*lumToYV12)(uint8_t *dst, const uint8_t *src, const uint8_t *src2, const uint8_t *src3,
422 int width, uint32_t *pal); ///< Unscaled conversion of luma plane to YV12 for horizontal scaler.
423 void (*alpToYV12)(uint8_t *dst, const uint8_t *src, const uint8_t *src2, const uint8_t *src3,
424 int width, uint32_t *pal); ///< Unscaled conversion of alpha plane to YV12 for horizontal scaler.
425 void (*chrToYV12)(uint8_t *dstU, uint8_t *dstV,
426 const uint8_t *src1, const uint8_t *src2, const uint8_t *src3,
427 int width, uint32_t *pal); ///< Unscaled conversion of chroma planes to YV12 for horizontal scaler.
429 * Scale one horizontal line of input data using a bilinear filter
430 * to produce one line of output data. Compared to SwsContext->hScale(),
431 * please take note of the following caveats when using these:
432 * - Scaling is done using only 7bit instead of 14bit coefficients.
433 * - You can use no more than 5 input pixels to produce 4 output
434 * pixels. Therefore, this filter should not be used for downscaling
435 * by more than ~20% in width (because that equals more than 5/4th
436 * downscaling and thus more than 5 pixels input per 4 pixels output).
437 * - In general, bilinear filters create artifacts during downscaling
438 * (even when <20%), because one output pixel will span more than one
439 * input pixel, and thus some pixels will need edges of both neighbor
440 * pixels to interpolate the output pixel. Since you can use at most
441 * two input pixels per output pixel in bilinear scaling, this is
442 * impossible and thus downscaling by any size will create artifacts.
443 * To enable this type of scaling, set SWS_FLAG_FAST_BILINEAR
444 * in SwsContext->flags.
447 void (*hyscale_fast)(struct SwsContext *c,
448 int16_t *dst, int dstWidth,
449 const uint8_t *src, int srcW, int xInc);
450 void (*hcscale_fast)(struct SwsContext *c,
451 int16_t *dst1, int16_t *dst2, int dstWidth,
452 const uint8_t *src1, const uint8_t *src2,
457 * Scale one horizontal line of input data using a filter over the input
458 * lines, to produce one (differently sized) line of output data.
460 * @param dst pointer to destination buffer for horizontally scaled
461 * data. If the number of bits per component of one
462 * destination pixel (SwsContext->dstBpc) is <= 10, data
463 * will be 15bpc in 16bits (int16_t) width. Else (i.e.
464 * SwsContext->dstBpc == 16), data will be 19bpc in
465 * 32bits (int32_t) width.
466 * @param dstW width of destination image
467 * @param src pointer to source data to be scaled. If the number of
468 * bits per component of a source pixel (SwsContext->srcBpc)
469 * is 8, this is 8bpc in 8bits (uint8_t) width. Else
470 * (i.e. SwsContext->dstBpc > 8), this is native depth
471 * in 16bits (uint16_t) width. In other words, for 9-bit
472 * YUV input, this is 9bpc, for 10-bit YUV input, this is
473 * 10bpc, and for 16-bit RGB or YUV, this is 16bpc.
474 * @param filter filter coefficients to be used per output pixel for
475 * scaling. This contains 14bpp filtering coefficients.
476 * Guaranteed to contain dstW * filterSize entries.
477 * @param filterPos position of the first input pixel to be used for
478 * each output pixel during scaling. Guaranteed to
479 * contain dstW entries.
480 * @param filterSize the number of input coefficients to be used (and
481 * thus the number of input pixels to be used) for
482 * creating a single output pixel. Is aligned to 4
483 * (and input coefficients thus padded with zeroes)
484 * to simplify creating SIMD code.
487 void (*hyScale)(struct SwsContext *c, int16_t *dst, int dstW, const uint8_t *src,
488 const int16_t *filter, const int16_t *filterPos,
490 void (*hcScale)(struct SwsContext *c, int16_t *dst, int dstW, const uint8_t *src,
491 const int16_t *filter, const int16_t *filterPos,
495 void (*lumConvertRange)(int16_t *dst, int width); ///< Color range conversion function for luma plane if needed.
496 void (*chrConvertRange)(int16_t *dst1, int16_t *dst2, int width); ///< Color range conversion function for chroma planes if needed.
498 int needs_hcscale; ///< Set if there are chroma planes to be converted.
501 //FIXME check init (where 0)
503 SwsFunc ff_yuv2rgb_get_func_ptr(SwsContext *c);
504 int ff_yuv2rgb_c_init_tables(SwsContext *c, const int inv_table[4],
505 int fullRange, int brightness,
506 int contrast, int saturation);
508 void ff_yuv2rgb_init_tables_altivec(SwsContext *c, const int inv_table[4],
509 int brightness, int contrast, int saturation);
510 void updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufIndex,
511 int lastInLumBuf, int lastInChrBuf);
513 SwsFunc ff_yuv2rgb_init_mmx(SwsContext *c);
514 SwsFunc ff_yuv2rgb_init_vis(SwsContext *c);
515 SwsFunc ff_yuv2rgb_init_mlib(SwsContext *c);
516 SwsFunc ff_yuv2rgb_init_altivec(SwsContext *c);
517 SwsFunc ff_yuv2rgb_get_func_ptr_bfin(SwsContext *c);
518 void ff_bfin_get_unscaled_swscale(SwsContext *c);
520 #if FF_API_SWS_FORMAT_NAME
522 * @deprecated Use av_get_pix_fmt_name() instead.
525 const char *sws_format_name(enum PixelFormat format);
528 //FIXME replace this with something faster
529 #define is16BPS(x) ( \
530 (x)==PIX_FMT_GRAY16BE \
531 || (x)==PIX_FMT_GRAY16LE \
532 || (x)==PIX_FMT_BGR48BE \
533 || (x)==PIX_FMT_BGR48LE \
534 || (x)==PIX_FMT_RGB48BE \
535 || (x)==PIX_FMT_RGB48LE \
536 || (x)==PIX_FMT_BGRA64BE \
537 || (x)==PIX_FMT_BGRA64LE \
538 || (x)==PIX_FMT_RGBA64BE \
539 || (x)==PIX_FMT_RGBA64LE \
540 || (x)==PIX_FMT_YUV420P16LE \
541 || (x)==PIX_FMT_YUV422P16LE \
542 || (x)==PIX_FMT_YUV444P16LE \
543 || (x)==PIX_FMT_YUV420P16BE \
544 || (x)==PIX_FMT_YUV422P16BE \
545 || (x)==PIX_FMT_YUV444P16BE \
547 #define isNBPS(x) ( \
548 (x)==PIX_FMT_YUV420P9LE \
549 || (x)==PIX_FMT_YUV420P9BE \
550 || (x)==PIX_FMT_YUV444P9BE \
551 || (x)==PIX_FMT_YUV444P9LE \
552 || (x)==PIX_FMT_YUV422P10BE \
553 || (x)==PIX_FMT_YUV422P10LE \
554 || (x)==PIX_FMT_YUV444P10BE \
555 || (x)==PIX_FMT_YUV444P10LE \
556 || (x)==PIX_FMT_YUV420P10LE \
557 || (x)==PIX_FMT_YUV420P10BE \
558 || (x)==PIX_FMT_YUV422P10LE \
559 || (x)==PIX_FMT_YUV422P10BE \
561 #define is9_OR_10BPS isNBPS //for ronald
562 #define isBE(x) ((x)&1)
563 #define isPlanar8YUV(x) ( \
564 (x)==PIX_FMT_YUV410P \
565 || (x)==PIX_FMT_YUV420P \
566 || (x)==PIX_FMT_YUVA420P \
567 || (x)==PIX_FMT_YUV411P \
568 || (x)==PIX_FMT_YUV422P \
569 || (x)==PIX_FMT_YUV444P \
570 || (x)==PIX_FMT_YUV440P \
571 || (x)==PIX_FMT_NV12 \
572 || (x)==PIX_FMT_NV21 \
574 #define isPlanarYUV(x) ( \
576 || (x)==PIX_FMT_YUV420P9LE \
577 || (x)==PIX_FMT_YUV444P9LE \
578 || (x)==PIX_FMT_YUV420P10LE \
579 || (x)==PIX_FMT_YUV422P10LE \
580 || (x)==PIX_FMT_YUV444P10LE \
581 || (x)==PIX_FMT_YUV420P16LE \
582 || (x)==PIX_FMT_YUV422P10LE \
583 || (x)==PIX_FMT_YUV422P16LE \
584 || (x)==PIX_FMT_YUV444P16LE \
585 || (x)==PIX_FMT_YUV420P9BE \
586 || (x)==PIX_FMT_YUV444P9BE \
587 || (x)==PIX_FMT_YUV420P10BE \
588 || (x)==PIX_FMT_YUV422P10BE \
589 || (x)==PIX_FMT_YUV444P10BE \
590 || (x)==PIX_FMT_YUV420P16BE \
591 || (x)==PIX_FMT_YUV422P10BE \
592 || (x)==PIX_FMT_YUV422P16BE \
593 || (x)==PIX_FMT_YUV444P16BE \
596 #define isPlanar(x) ( \
598 || (x)==PIX_FMT_GBR24P \
602 (x)==PIX_FMT_UYVY422 \
603 || (x)==PIX_FMT_YUYV422 \
606 #define isGray(x) ( \
608 || (x)==PIX_FMT_GRAY8A \
609 || (x)==PIX_FMT_GRAY16BE \
610 || (x)==PIX_FMT_GRAY16LE \
612 #define isGray16(x) ( \
613 (x)==PIX_FMT_GRAY16BE \
614 || (x)==PIX_FMT_GRAY16LE \
616 #define isRGBinInt(x) ( \
617 (x)==PIX_FMT_RGB48BE \
618 || (x)==PIX_FMT_RGB48LE \
619 || (x)==PIX_FMT_RGBA64BE \
620 || (x)==PIX_FMT_RGBA64LE \
621 || (x)==PIX_FMT_RGB32 \
622 || (x)==PIX_FMT_RGB32_1 \
623 || (x)==PIX_FMT_RGB24 \
624 || (x)==PIX_FMT_RGB565BE \
625 || (x)==PIX_FMT_RGB565LE \
626 || (x)==PIX_FMT_RGB555BE \
627 || (x)==PIX_FMT_RGB555LE \
628 || (x)==PIX_FMT_RGB444BE \
629 || (x)==PIX_FMT_RGB444LE \
630 || (x)==PIX_FMT_RGB8 \
631 || (x)==PIX_FMT_RGB4 \
632 || (x)==PIX_FMT_RGB4_BYTE \
633 || (x)==PIX_FMT_MONOBLACK \
634 || (x)==PIX_FMT_MONOWHITE \
636 #define isBGRinInt(x) ( \
637 (x)==PIX_FMT_BGR48BE \
638 || (x)==PIX_FMT_BGR48LE \
639 || (x)==PIX_FMT_BGRA64BE \
640 || (x)==PIX_FMT_BGRA64LE \
641 || (x)==PIX_FMT_BGR32 \
642 || (x)==PIX_FMT_BGR32_1 \
643 || (x)==PIX_FMT_BGR24 \
644 || (x)==PIX_FMT_BGR565BE \
645 || (x)==PIX_FMT_BGR565LE \
646 || (x)==PIX_FMT_BGR555BE \
647 || (x)==PIX_FMT_BGR555LE \
648 || (x)==PIX_FMT_BGR444BE \
649 || (x)==PIX_FMT_BGR444LE \
650 || (x)==PIX_FMT_BGR8 \
651 || (x)==PIX_FMT_BGR4 \
652 || (x)==PIX_FMT_BGR4_BYTE \
653 || (x)==PIX_FMT_MONOBLACK \
654 || (x)==PIX_FMT_MONOWHITE \
656 #define isRGBinBytes(x) ( \
657 (x)==PIX_FMT_RGB48BE \
658 || (x)==PIX_FMT_RGB48LE \
659 || (x)==PIX_FMT_RGBA64BE \
660 || (x)==PIX_FMT_RGBA64LE \
661 || (x)==PIX_FMT_RGBA \
662 || (x)==PIX_FMT_ARGB \
663 || (x)==PIX_FMT_RGB24 \
665 #define isBGRinBytes(x) ( \
666 (x)==PIX_FMT_BGR48BE \
667 || (x)==PIX_FMT_BGR48LE \
668 || (x)==PIX_FMT_BGRA64BE \
669 || (x)==PIX_FMT_BGRA64LE \
670 || (x)==PIX_FMT_BGRA \
671 || (x)==PIX_FMT_ABGR \
672 || (x)==PIX_FMT_BGR24 \
674 #define isAnyRGB(x) ( \
677 || (x)==PIX_FMT_GBR24P \
679 #define isALPHA(x) ( \
680 (x)==PIX_FMT_BGRA64BE \
681 || (x)==PIX_FMT_BGRA64LE \
682 || (x)==PIX_FMT_RGBA64BE \
683 || (x)==PIX_FMT_RGBA64LE \
684 || (x)==PIX_FMT_BGR32 \
685 || (x)==PIX_FMT_BGR32_1 \
686 || (x)==PIX_FMT_RGB32 \
687 || (x)==PIX_FMT_RGB32_1 \
688 || (x)==PIX_FMT_PAL8 \
689 || (x)==PIX_FMT_GRAY8A \
690 || (x)==PIX_FMT_YUVA420P \
692 #define isPacked(x) ( \
694 || (x)==PIX_FMT_YUYV422 \
695 || (x)==PIX_FMT_UYVY422 \
696 || (x)==PIX_FMT_Y400A \
700 #define usePal(x) ((av_pix_fmt_descriptors[x].flags & PIX_FMT_PAL) || (x) == PIX_FMT_GRAY8A)
702 extern const uint64_t ff_dither4[2];
703 extern const uint64_t ff_dither8[2];
704 extern const uint8_t dithers[8][8][8];
705 extern const uint16_t dither_scale[15][16];
708 extern const AVClass sws_context_class;
711 * Sets c->swScale to an unscaled converter if one exists for the specific
712 * source and destination formats, bit depths, flags, etc.
714 void ff_get_unscaled_swscale(SwsContext *c);
716 void ff_swscale_get_unscaled_altivec(SwsContext *c);
719 * Returns function pointer to fastest main scaler path function depending
720 * on architecture and available optimizations.
722 SwsFunc ff_getSwsFunc(SwsContext *c);
724 void ff_sws_init_swScale_altivec(SwsContext *c);
725 void ff_sws_init_swScale_mmx(SwsContext *c);
727 #endif /* SWSCALE_SWSCALE_INTERNAL_H */