2 * Copyright (C) 2001-2003 Michael Niedermayer <michaelni@gmx.at>
4 * This file is part of Libav.
6 * Libav 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 * Libav 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 Libav; 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"
32 #define STR(s) AV_TOSTRING(s) //AV_STRINGIFY is too long
34 #define FAST_BGR2YV12 //use 7-bit instead of 15-bit coefficients
36 #define MAX_FILTER_SIZE 256
39 #define ALT32_CORR (-1)
56 typedef int (*SwsFunc)(struct SwsContext *context, const uint8_t* src[],
57 int srcStride[], int srcSliceY, int srcSliceH,
58 uint8_t* dst[], int dstStride[]);
60 typedef void (*yuv2planar1_fn) (struct SwsContext *c,
61 const int16_t *lumSrc, const int16_t *chrUSrc,
62 const int16_t *chrVSrc, const int16_t *alpSrc,
64 uint8_t *uDest, uint8_t *vDest, uint8_t *aDest,
65 int dstW, int chrDstW);
66 typedef void (*yuv2planarX_fn) (struct SwsContext *c,
67 const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize,
68 const int16_t *chrFilter, const int16_t **chrUSrc,
69 const int16_t **chrVSrc, int chrFilterSize,
70 const int16_t **alpSrc,
72 uint8_t *uDest, uint8_t *vDest, uint8_t *aDest,
73 int dstW, int chrDstW);
74 typedef void (*yuv2packed1_fn) (struct SwsContext *c,
76 const uint16_t *ubuf0, const uint16_t *ubuf1,
77 const uint16_t *vbuf0, const uint16_t *vbuf1,
78 const uint16_t *abuf0,
80 int dstW, int uvalpha, int dstFormat, int flags, int y);
81 typedef void (*yuv2packed2_fn) (struct SwsContext *c,
82 const uint16_t *buf0, const uint16_t *buf1,
83 const uint16_t *ubuf0, const uint16_t *ubuf1,
84 const uint16_t *vbuf0, const uint16_t *vbuf1,
85 const uint16_t *abuf0, const uint16_t *abuf1,
87 int dstW, int yalpha, int uvalpha, int y);
88 typedef void (*yuv2packedX_fn) (struct SwsContext *c,
89 const int16_t *lumFilter, const int16_t **lumSrc, int lumFilterSize,
90 const int16_t *chrFilter, const int16_t **chrUSrc,
91 const int16_t **chrVSrc, int chrFilterSize,
92 const int16_t **alpSrc, uint8_t *dest,
95 /* This struct should be aligned on at least a 32-byte boundary. */
96 typedef struct SwsContext {
98 * info on struct for av_log
100 const AVClass *av_class;
103 * Note that src, dst, srcStride, dstStride will be copied in the
104 * sws_scale() wrapper so they can be freely modified here.
107 int srcW; ///< Width of source luma/alpha planes.
108 int srcH; ///< Height of source luma/alpha planes.
109 int dstH; ///< Height of destination luma/alpha planes.
110 int chrSrcW; ///< Width of source chroma planes.
111 int chrSrcH; ///< Height of source chroma planes.
112 int chrDstW; ///< Width of destination chroma planes.
113 int chrDstH; ///< Height of destination chroma planes.
114 int lumXInc, chrXInc;
115 int lumYInc, chrYInc;
116 enum PixelFormat dstFormat; ///< Destination pixel format.
117 enum PixelFormat srcFormat; ///< Source pixel format.
118 int dstFormatBpp; ///< Number of bits per pixel of the destination pixel format.
119 int srcFormatBpp; ///< Number of bits per pixel of the source pixel format.
120 int chrSrcHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in source image.
121 int chrSrcVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in source image.
122 int chrDstHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in destination image.
123 int chrDstVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in destination image.
124 int vChrDrop; ///< Binary logarithm of extra vertical subsampling factor in source image chroma planes specified by user.
125 int sliceDir; ///< Direction that slices are fed to the scaler (1 = top-to-bottom, -1 = bottom-to-top).
126 double param[2]; ///< Input parameters for scaling algorithms that need them.
128 uint32_t pal_yuv[256];
129 uint32_t pal_rgb[256];
132 * @name Scaled horizontal lines ring buffer.
133 * The horizontal scaler keeps just enough scaled lines in a ring buffer
134 * so they may be passed to the vertical scaler. The pointers to the
135 * allocated buffers for each line are duplicated in sequence in the ring
136 * buffer to simplify indexing and avoid wrapping around between lines
137 * inside the vertical scaler code. The wrapping is done before the
138 * vertical scaler is called.
141 int16_t **lumPixBuf; ///< Ring buffer for scaled horizontal luma plane lines to be fed to the vertical scaler.
142 int16_t **chrUPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler.
143 int16_t **chrVPixBuf; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler.
144 int16_t **alpPixBuf; ///< Ring buffer for scaled horizontal alpha plane lines to be fed to the vertical scaler.
145 int vLumBufSize; ///< Number of vertical luma/alpha lines allocated in the ring buffer.
146 int vChrBufSize; ///< Number of vertical chroma lines allocated in the ring buffer.
147 int lastInLumBuf; ///< Last scaled horizontal luma/alpha line from source in the ring buffer.
148 int lastInChrBuf; ///< Last scaled horizontal chroma line from source in the ring buffer.
149 int lumBufIndex; ///< Index in ring buffer of the last scaled horizontal luma/alpha line from source.
150 int chrBufIndex; ///< Index in ring buffer of the last scaled horizontal chroma line from source.
153 uint8_t *formatConvBuffer;
156 * @name Horizontal and vertical filters.
157 * To better understand the following fields, here is a pseudo-code of
158 * their usage in filtering a horizontal line:
160 * for (i = 0; i < width; i++) {
162 * for (j = 0; j < filterSize; j++)
163 * dst[i] += src[ filterPos[i] + j ] * filter[ filterSize * i + j ];
164 * dst[i] >>= FRAC_BITS; // The actual implementation is fixed-point.
169 int16_t *hLumFilter; ///< Array of horizontal filter coefficients for luma/alpha planes.
170 int16_t *hChrFilter; ///< Array of horizontal filter coefficients for chroma planes.
171 int16_t *vLumFilter; ///< Array of vertical filter coefficients for luma/alpha planes.
172 int16_t *vChrFilter; ///< Array of vertical filter coefficients for chroma planes.
173 int16_t *hLumFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes.
174 int16_t *hChrFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes.
175 int16_t *vLumFilterPos; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes.
176 int16_t *vChrFilterPos; ///< Array of vertical filter starting positions for each dst[i] for chroma planes.
177 int hLumFilterSize; ///< Horizontal filter size for luma/alpha pixels.
178 int hChrFilterSize; ///< Horizontal filter size for chroma pixels.
179 int vLumFilterSize; ///< Vertical filter size for luma/alpha pixels.
180 int vChrFilterSize; ///< Vertical filter size for chroma pixels.
183 int lumMmx2FilterCodeSize; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for luma/alpha planes.
184 int chrMmx2FilterCodeSize; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for chroma planes.
185 uint8_t *lumMmx2FilterCode; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code for luma/alpha planes.
186 uint8_t *chrMmx2FilterCode; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code for chroma planes.
190 int dstY; ///< Last destination vertical line output from last slice.
191 int flags; ///< Flags passed by the user to select scaler algorithm, optimizations, subsampling, etc...
192 void * yuvTable; // pointer to the yuv->rgb table start so it can be freed()
193 uint8_t * table_rV[256];
194 uint8_t * table_gU[256];
196 uint8_t * table_bU[256];
199 int contrast, brightness, saturation; // for sws_getColorspaceDetails
200 int srcColorspaceTable[4];
201 int dstColorspaceTable[4];
202 int srcRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (source image).
203 int dstRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (destination image).
204 int yuv2rgb_y_offset;
206 int yuv2rgb_v2r_coeff;
207 int yuv2rgb_v2g_coeff;
208 int yuv2rgb_u2g_coeff;
209 int yuv2rgb_u2b_coeff;
211 #define RED_DITHER "0*8"
212 #define GREEN_DITHER "1*8"
213 #define BLUE_DITHER "2*8"
214 #define Y_COEFF "3*8"
215 #define VR_COEFF "4*8"
216 #define UB_COEFF "5*8"
217 #define VG_COEFF "6*8"
218 #define UG_COEFF "7*8"
219 #define Y_OFFSET "8*8"
220 #define U_OFFSET "9*8"
221 #define V_OFFSET "10*8"
222 #define LUM_MMX_FILTER_OFFSET "11*8"
223 #define CHR_MMX_FILTER_OFFSET "11*8+4*4*256"
224 #define DSTW_OFFSET "11*8+4*4*256*2" //do not change, it is hardcoded in the ASM
225 #define ESP_OFFSET "11*8+4*4*256*2+8"
226 #define VROUNDER_OFFSET "11*8+4*4*256*2+16"
227 #define U_TEMP "11*8+4*4*256*2+24"
228 #define V_TEMP "11*8+4*4*256*2+32"
229 #define Y_TEMP "11*8+4*4*256*2+40"
230 #define ALP_MMX_FILTER_OFFSET "11*8+4*4*256*2+48"
231 #define UV_OFF "11*8+4*4*256*3+48"
232 #define UV_OFFx2 "11*8+4*4*256*3+56"
234 DECLARE_ALIGNED(8, uint64_t, redDither);
235 DECLARE_ALIGNED(8, uint64_t, greenDither);
236 DECLARE_ALIGNED(8, uint64_t, blueDither);
238 DECLARE_ALIGNED(8, uint64_t, yCoeff);
239 DECLARE_ALIGNED(8, uint64_t, vrCoeff);
240 DECLARE_ALIGNED(8, uint64_t, ubCoeff);
241 DECLARE_ALIGNED(8, uint64_t, vgCoeff);
242 DECLARE_ALIGNED(8, uint64_t, ugCoeff);
243 DECLARE_ALIGNED(8, uint64_t, yOffset);
244 DECLARE_ALIGNED(8, uint64_t, uOffset);
245 DECLARE_ALIGNED(8, uint64_t, vOffset);
246 int32_t lumMmxFilter[4*MAX_FILTER_SIZE];
247 int32_t chrMmxFilter[4*MAX_FILTER_SIZE];
248 int dstW; ///< Width of destination luma/alpha planes.
249 DECLARE_ALIGNED(8, uint64_t, esp);
250 DECLARE_ALIGNED(8, uint64_t, vRounder);
251 DECLARE_ALIGNED(8, uint64_t, u_temp);
252 DECLARE_ALIGNED(8, uint64_t, v_temp);
253 DECLARE_ALIGNED(8, uint64_t, y_temp);
254 int32_t alpMmxFilter[4*MAX_FILTER_SIZE];
255 DECLARE_ALIGNED(8, ptrdiff_t, uv_off); ///< offset (in pixels) between u and v planes
256 DECLARE_ALIGNED(8, ptrdiff_t, uv_offx2); ///< offset (in bytes) between u and v planes
259 vector signed short CY;
260 vector signed short CRV;
261 vector signed short CBU;
262 vector signed short CGU;
263 vector signed short CGV;
264 vector signed short OY;
265 vector unsigned short CSHIFT;
266 vector signed short *vYCoeffsBank, *vCCoeffsBank;
270 DECLARE_ALIGNED(4, uint32_t, oy);
271 DECLARE_ALIGNED(4, uint32_t, oc);
272 DECLARE_ALIGNED(4, uint32_t, zero);
273 DECLARE_ALIGNED(4, uint32_t, cy);
274 DECLARE_ALIGNED(4, uint32_t, crv);
275 DECLARE_ALIGNED(4, uint32_t, rmask);
276 DECLARE_ALIGNED(4, uint32_t, cbu);
277 DECLARE_ALIGNED(4, uint32_t, bmask);
278 DECLARE_ALIGNED(4, uint32_t, cgu);
279 DECLARE_ALIGNED(4, uint32_t, cgv);
280 DECLARE_ALIGNED(4, uint32_t, gmask);
284 DECLARE_ALIGNED(8, uint64_t, sparc_coeffs)[10];
287 /* function pointers for swScale() */
288 yuv2planar1_fn yuv2yuv1;
289 yuv2planarX_fn yuv2yuvX;
290 yuv2packed1_fn yuv2packed1;
291 yuv2packed2_fn yuv2packed2;
292 yuv2packedX_fn yuv2packedX;
294 void (*lumToYV12)(uint8_t *dst, const uint8_t *src,
295 int width, uint32_t *pal); ///< Unscaled conversion of luma plane to YV12 for horizontal scaler.
296 void (*alpToYV12)(uint8_t *dst, const uint8_t *src,
297 int width, uint32_t *pal); ///< Unscaled conversion of alpha plane to YV12 for horizontal scaler.
298 void (*chrToYV12)(uint8_t *dstU, uint8_t *dstV,
299 const uint8_t *src1, const uint8_t *src2,
300 int width, uint32_t *pal); ///< Unscaled conversion of chroma planes to YV12 for horizontal scaler.
301 void (*hyscale_fast)(struct SwsContext *c,
302 int16_t *dst, int dstWidth,
303 const uint8_t *src, int srcW, int xInc);
304 void (*hcscale_fast)(struct SwsContext *c,
305 int16_t *dst1, int16_t *dst2, int dstWidth,
306 const uint8_t *src1, const uint8_t *src2,
309 void (*hScale)(int16_t *dst, int dstW, const uint8_t *src, int srcW,
310 int xInc, const int16_t *filter, const int16_t *filterPos,
313 void (*lumConvertRange)(int16_t *dst, int width); ///< Color range conversion function for luma plane if needed.
314 void (*chrConvertRange)(int16_t *dst1, int16_t *dst2, int width); ///< Color range conversion function for chroma planes if needed.
316 int needs_hcscale; ///< Set if there are chroma planes to be converted.
319 //FIXME check init (where 0)
321 SwsFunc ff_yuv2rgb_get_func_ptr(SwsContext *c);
322 int ff_yuv2rgb_c_init_tables(SwsContext *c, const int inv_table[4],
323 int fullRange, int brightness,
324 int contrast, int saturation);
326 void ff_yuv2rgb_init_tables_altivec(SwsContext *c, const int inv_table[4],
327 int brightness, int contrast, int saturation);
328 void updateMMXDitherTables(SwsContext *c, int dstY, int lumBufIndex, int chrBufIndex,
329 int lastInLumBuf, int lastInChrBuf);
331 SwsFunc ff_yuv2rgb_init_mmx(SwsContext *c);
332 SwsFunc ff_yuv2rgb_init_vis(SwsContext *c);
333 SwsFunc ff_yuv2rgb_init_mlib(SwsContext *c);
334 SwsFunc ff_yuv2rgb_init_altivec(SwsContext *c);
335 SwsFunc ff_yuv2rgb_get_func_ptr_bfin(SwsContext *c);
336 void ff_bfin_get_unscaled_swscale(SwsContext *c);
338 const char *sws_format_name(enum PixelFormat format);
340 //FIXME replace this with something faster
341 #define is16BPS(x) ( \
342 (x)==PIX_FMT_GRAY16BE \
343 || (x)==PIX_FMT_GRAY16LE \
344 || (x)==PIX_FMT_BGR48BE \
345 || (x)==PIX_FMT_BGR48LE \
346 || (x)==PIX_FMT_RGB48BE \
347 || (x)==PIX_FMT_RGB48LE \
348 || (x)==PIX_FMT_YUV420P16LE \
349 || (x)==PIX_FMT_YUV422P16LE \
350 || (x)==PIX_FMT_YUV444P16LE \
351 || (x)==PIX_FMT_YUV420P16BE \
352 || (x)==PIX_FMT_YUV422P16BE \
353 || (x)==PIX_FMT_YUV444P16BE \
355 #define is9_OR_10BPS(x) ( \
356 (x)==PIX_FMT_YUV420P9LE \
357 || (x)==PIX_FMT_YUV420P9BE \
358 || (x)==PIX_FMT_YUV444P9BE \
359 || (x)==PIX_FMT_YUV444P9LE \
360 || (x)==PIX_FMT_YUV422P10BE \
361 || (x)==PIX_FMT_YUV422P10LE \
362 || (x)==PIX_FMT_YUV444P10BE \
363 || (x)==PIX_FMT_YUV444P10LE \
364 || (x)==PIX_FMT_YUV420P10LE \
365 || (x)==PIX_FMT_YUV420P10BE \
367 #define isBE(x) ((x)&1)
368 #define isPlanar8YUV(x) ( \
369 (x)==PIX_FMT_YUV410P \
370 || (x)==PIX_FMT_YUV420P \
371 || (x)==PIX_FMT_YUVA420P \
372 || (x)==PIX_FMT_YUV411P \
373 || (x)==PIX_FMT_YUV422P \
374 || (x)==PIX_FMT_YUV444P \
375 || (x)==PIX_FMT_YUV440P \
376 || (x)==PIX_FMT_NV12 \
377 || (x)==PIX_FMT_NV21 \
379 #define isPlanarYUV(x) ( \
381 || (x)==PIX_FMT_YUV420P9LE \
382 || (x)==PIX_FMT_YUV444P9LE \
383 || (x)==PIX_FMT_YUV420P10LE \
384 || (x)==PIX_FMT_YUV422P10LE \
385 || (x)==PIX_FMT_YUV444P10LE \
386 || (x)==PIX_FMT_YUV420P16LE \
387 || (x)==PIX_FMT_YUV422P16LE \
388 || (x)==PIX_FMT_YUV444P16LE \
389 || (x)==PIX_FMT_YUV420P9BE \
390 || (x)==PIX_FMT_YUV444P9BE \
391 || (x)==PIX_FMT_YUV420P10BE \
392 || (x)==PIX_FMT_YUV422P10BE \
393 || (x)==PIX_FMT_YUV444P10BE \
394 || (x)==PIX_FMT_YUV420P16BE \
395 || (x)==PIX_FMT_YUV422P16BE \
396 || (x)==PIX_FMT_YUV444P16BE \
399 (x)==PIX_FMT_UYVY422 \
400 || (x)==PIX_FMT_YUYV422 \
403 #define isGray(x) ( \
405 || (x)==PIX_FMT_Y400A \
406 || (x)==PIX_FMT_GRAY16BE \
407 || (x)==PIX_FMT_GRAY16LE \
409 #define isGray16(x) ( \
410 (x)==PIX_FMT_GRAY16BE \
411 || (x)==PIX_FMT_GRAY16LE \
413 #define isRGBinInt(x) ( \
414 (x)==PIX_FMT_RGB48BE \
415 || (x)==PIX_FMT_RGB48LE \
416 || (x)==PIX_FMT_RGB32 \
417 || (x)==PIX_FMT_RGB32_1 \
418 || (x)==PIX_FMT_RGB24 \
419 || (x)==PIX_FMT_RGB565BE \
420 || (x)==PIX_FMT_RGB565LE \
421 || (x)==PIX_FMT_RGB555BE \
422 || (x)==PIX_FMT_RGB555LE \
423 || (x)==PIX_FMT_RGB444BE \
424 || (x)==PIX_FMT_RGB444LE \
425 || (x)==PIX_FMT_RGB8 \
426 || (x)==PIX_FMT_RGB4 \
427 || (x)==PIX_FMT_RGB4_BYTE \
428 || (x)==PIX_FMT_MONOBLACK \
429 || (x)==PIX_FMT_MONOWHITE \
431 #define isBGRinInt(x) ( \
432 (x)==PIX_FMT_BGR48BE \
433 || (x)==PIX_FMT_BGR48LE \
434 || (x)==PIX_FMT_BGR32 \
435 || (x)==PIX_FMT_BGR32_1 \
436 || (x)==PIX_FMT_BGR24 \
437 || (x)==PIX_FMT_BGR565BE \
438 || (x)==PIX_FMT_BGR565LE \
439 || (x)==PIX_FMT_BGR555BE \
440 || (x)==PIX_FMT_BGR555LE \
441 || (x)==PIX_FMT_BGR444BE \
442 || (x)==PIX_FMT_BGR444LE \
443 || (x)==PIX_FMT_BGR8 \
444 || (x)==PIX_FMT_BGR4 \
445 || (x)==PIX_FMT_BGR4_BYTE \
446 || (x)==PIX_FMT_MONOBLACK \
447 || (x)==PIX_FMT_MONOWHITE \
449 #define isRGBinBytes(x) ( \
450 (x)==PIX_FMT_RGB48BE \
451 || (x)==PIX_FMT_RGB48LE \
452 || (x)==PIX_FMT_RGBA \
453 || (x)==PIX_FMT_ARGB \
454 || (x)==PIX_FMT_RGB24 \
456 #define isBGRinBytes(x) ( \
457 (x)==PIX_FMT_BGR48BE \
458 || (x)==PIX_FMT_BGR48LE \
459 || (x)==PIX_FMT_BGRA \
460 || (x)==PIX_FMT_ABGR \
461 || (x)==PIX_FMT_BGR24 \
463 #define isAnyRGB(x) ( \
467 #define isALPHA(x) ( \
469 || (x)==PIX_FMT_BGR32_1 \
470 || (x)==PIX_FMT_RGB32 \
471 || (x)==PIX_FMT_RGB32_1 \
472 || (x)==PIX_FMT_Y400A \
473 || (x)==PIX_FMT_YUVA420P \
475 #define isPacked(x) ( \
477 || (x)==PIX_FMT_YUYV422 \
478 || (x)==PIX_FMT_UYVY422 \
479 || (x)==PIX_FMT_Y400A \
482 #define usePal(x) ((av_pix_fmt_descriptors[x].flags & PIX_FMT_PAL) || (x) == PIX_FMT_Y400A)
484 extern const uint64_t ff_dither4[2];
485 extern const uint64_t ff_dither8[2];
487 extern const AVClass sws_context_class;
490 * Sets c->swScale to an unscaled converter if one exists for the specific
491 * source and destination formats, bit depths, flags, etc.
493 void ff_get_unscaled_swscale(SwsContext *c);
495 void ff_swscale_get_unscaled_altivec(SwsContext *c);
498 * Returns function pointer to fastest main scaler path function depending
499 * on architecture and available optimizations.
501 SwsFunc ff_getSwsFunc(SwsContext *c);
503 void ff_sws_init_swScale_altivec(SwsContext *c);
504 void ff_sws_init_swScale_mmx(SwsContext *c);
506 #endif /* SWSCALE_SWSCALE_INTERNAL_H */