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"
32 #define STR(s) AV_TOSTRING(s) //AV_STRINGIFY is too long
34 #define MAX_FILTER_SIZE 256
39 #define VOFW 2048 // faster on PPC and not tested on others
45 #define ALT32_CORR (-1)
62 typedef int (*SwsFunc
)(struct SwsContext
*context
, const uint8_t* src
[],
63 int srcStride
[], int srcSliceY
, int srcSliceH
,
64 uint8_t* dst
[], int dstStride
[]);
66 /* This struct should be aligned on at least a 32-byte boundary. */
67 typedef struct SwsContext
{
69 * info on struct for av_log
71 const AVClass
*av_class
;
74 * Note that src, dst, srcStride, dstStride will be copied in the
75 * sws_scale() wrapper so they can be freely modified here.
78 int srcW
; ///< Width of source luma/alpha planes.
79 int srcH
; ///< Height of source luma/alpha planes.
80 int dstH
; ///< Height of destination luma/alpha planes.
81 int chrSrcW
; ///< Width of source chroma planes.
82 int chrSrcH
; ///< Height of source chroma planes.
83 int chrDstW
; ///< Width of destination chroma planes.
84 int chrDstH
; ///< Height of destination chroma planes.
87 enum PixelFormat dstFormat
; ///< Destination pixel format.
88 enum PixelFormat srcFormat
; ///< Source pixel format.
89 int dstFormatBpp
; ///< Number of bits per pixel of the destination pixel format.
90 int srcFormatBpp
; ///< Number of bits per pixel of the source pixel format.
91 int chrSrcHSubSample
; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in source image.
92 int chrSrcVSubSample
; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in source image.
93 int chrDstHSubSample
; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in destination image.
94 int chrDstVSubSample
; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in destination image.
95 int vChrDrop
; ///< Binary logarithm of extra vertical subsampling factor in source image chroma planes specified by user.
96 int sliceDir
; ///< Direction that slices are fed to the scaler (1 = top-to-bottom, -1 = bottom-to-top).
97 double param
[2]; ///< Input parameters for scaling algorithms that need them.
99 uint32_t pal_yuv
[256];
100 uint32_t pal_rgb
[256];
103 * @name Scaled horizontal lines ring buffer.
104 * The horizontal scaler keeps just enough scaled lines in a ring buffer
105 * so they may be passed to the vertical scaler. The pointers to the
106 * allocated buffers for each line are duplicated in sequence in the ring
107 * buffer to simplify indexing and avoid wrapping around between lines
108 * inside the vertical scaler code. The wrapping is done before the
109 * vertical scaler is called.
112 int16_t **lumPixBuf
; ///< Ring buffer for scaled horizontal luma plane lines to be fed to the vertical scaler.
113 int16_t **chrPixBuf
; ///< Ring buffer for scaled horizontal chroma plane lines to be fed to the vertical scaler.
114 int16_t **alpPixBuf
; ///< Ring buffer for scaled horizontal alpha plane lines to be fed to the vertical scaler.
115 int vLumBufSize
; ///< Number of vertical luma/alpha lines allocated in the ring buffer.
116 int vChrBufSize
; ///< Number of vertical chroma lines allocated in the ring buffer.
117 int lastInLumBuf
; ///< Last scaled horizontal luma/alpha line from source in the ring buffer.
118 int lastInChrBuf
; ///< Last scaled horizontal chroma line from source in the ring buffer.
119 int lumBufIndex
; ///< Index in ring buffer of the last scaled horizontal luma/alpha line from source.
120 int chrBufIndex
; ///< Index in ring buffer of the last scaled horizontal chroma line from source.
123 uint8_t formatConvBuffer
[VOF
]; //FIXME dynamic allocation, but we have to change a lot of code for this to be useful
126 * @name Horizontal and vertical filters.
127 * To better understand the following fields, here is a pseudo-code of
128 * their usage in filtering a horizontal line:
130 * for (i = 0; i < width; i++) {
132 * for (j = 0; j < filterSize; j++)
133 * dst[i] += src[ filterPos[i] + j ] * filter[ filterSize * i + j ];
134 * dst[i] >>= FRAC_BITS; // The actual implementation is fixed-point.
139 int16_t *hLumFilter
; ///< Array of horizontal filter coefficients for luma/alpha planes.
140 int16_t *hChrFilter
; ///< Array of horizontal filter coefficients for chroma planes.
141 int16_t *vLumFilter
; ///< Array of vertical filter coefficients for luma/alpha planes.
142 int16_t *vChrFilter
; ///< Array of vertical filter coefficients for chroma planes.
143 int16_t *hLumFilterPos
; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes.
144 int16_t *hChrFilterPos
; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes.
145 int16_t *vLumFilterPos
; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes.
146 int16_t *vChrFilterPos
; ///< Array of vertical filter starting positions for each dst[i] for chroma planes.
147 int hLumFilterSize
; ///< Horizontal filter size for luma/alpha pixels.
148 int hChrFilterSize
; ///< Horizontal filter size for chroma pixels.
149 int vLumFilterSize
; ///< Vertical filter size for luma/alpha pixels.
150 int vChrFilterSize
; ///< Vertical filter size for chroma pixels.
153 int lumMmx2FilterCodeSize
; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for luma/alpha planes.
154 int chrMmx2FilterCodeSize
; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code size for chroma planes.
155 uint8_t *lumMmx2FilterCode
; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code for luma/alpha planes.
156 uint8_t *chrMmx2FilterCode
; ///< Runtime-generated MMX2 horizontal fast bilinear scaler code for chroma planes.
160 int dstY
; ///< Last destination vertical line output from last slice.
161 int flags
; ///< Flags passed by the user to select scaler algorithm, optimizations, subsampling, etc...
162 void * yuvTable
; // pointer to the yuv->rgb table start so it can be freed()
163 uint8_t * table_rV
[256];
164 uint8_t * table_gU
[256];
166 uint8_t * table_bU
[256];
169 int contrast
, brightness
, saturation
; // for sws_getColorspaceDetails
170 int srcColorspaceTable
[4];
171 int dstColorspaceTable
[4];
172 int srcRange
; ///< 0 = MPG YUV range, 1 = JPG YUV range (source image).
173 int dstRange
; ///< 0 = MPG YUV range, 1 = JPG YUV range (destination image).
174 int yuv2rgb_y_offset
;
176 int yuv2rgb_v2r_coeff
;
177 int yuv2rgb_v2g_coeff
;
178 int yuv2rgb_u2g_coeff
;
179 int yuv2rgb_u2b_coeff
;
181 #define RED_DITHER "0*8"
182 #define GREEN_DITHER "1*8"
183 #define BLUE_DITHER "2*8"
184 #define Y_COEFF "3*8"
185 #define VR_COEFF "4*8"
186 #define UB_COEFF "5*8"
187 #define VG_COEFF "6*8"
188 #define UG_COEFF "7*8"
189 #define Y_OFFSET "8*8"
190 #define U_OFFSET "9*8"
191 #define V_OFFSET "10*8"
192 #define LUM_MMX_FILTER_OFFSET "11*8"
193 #define CHR_MMX_FILTER_OFFSET "11*8+4*4*256"
194 #define DSTW_OFFSET "11*8+4*4*256*2" //do not change, it is hardcoded in the ASM
195 #define ESP_OFFSET "11*8+4*4*256*2+8"
196 #define VROUNDER_OFFSET "11*8+4*4*256*2+16"
197 #define U_TEMP "11*8+4*4*256*2+24"
198 #define V_TEMP "11*8+4*4*256*2+32"
199 #define Y_TEMP "11*8+4*4*256*2+40"
200 #define ALP_MMX_FILTER_OFFSET "11*8+4*4*256*2+48"
202 DECLARE_ALIGNED(8, uint64_t, redDither
);
203 DECLARE_ALIGNED(8, uint64_t, greenDither
);
204 DECLARE_ALIGNED(8, uint64_t, blueDither
);
206 DECLARE_ALIGNED(8, uint64_t, yCoeff
);
207 DECLARE_ALIGNED(8, uint64_t, vrCoeff
);
208 DECLARE_ALIGNED(8, uint64_t, ubCoeff
);
209 DECLARE_ALIGNED(8, uint64_t, vgCoeff
);
210 DECLARE_ALIGNED(8, uint64_t, ugCoeff
);
211 DECLARE_ALIGNED(8, uint64_t, yOffset
);
212 DECLARE_ALIGNED(8, uint64_t, uOffset
);
213 DECLARE_ALIGNED(8, uint64_t, vOffset
);
214 int32_t lumMmxFilter
[4*MAX_FILTER_SIZE
];
215 int32_t chrMmxFilter
[4*MAX_FILTER_SIZE
];
216 int dstW
; ///< Width of destination luma/alpha planes.
217 DECLARE_ALIGNED(8, uint64_t, esp
);
218 DECLARE_ALIGNED(8, uint64_t, vRounder
);
219 DECLARE_ALIGNED(8, uint64_t, u_temp
);
220 DECLARE_ALIGNED(8, uint64_t, v_temp
);
221 DECLARE_ALIGNED(8, uint64_t, y_temp
);
222 int32_t alpMmxFilter
[4*MAX_FILTER_SIZE
];
225 vector
signed short CY
;
226 vector
signed short CRV
;
227 vector
signed short CBU
;
228 vector
signed short CGU
;
229 vector
signed short CGV
;
230 vector
signed short OY
;
231 vector
unsigned short CSHIFT
;
232 vector
signed short *vYCoeffsBank
, *vCCoeffsBank
;
236 DECLARE_ALIGNED(4, uint32_t, oy
);
237 DECLARE_ALIGNED(4, uint32_t, oc
);
238 DECLARE_ALIGNED(4, uint32_t, zero
);
239 DECLARE_ALIGNED(4, uint32_t, cy
);
240 DECLARE_ALIGNED(4, uint32_t, crv
);
241 DECLARE_ALIGNED(4, uint32_t, rmask
);
242 DECLARE_ALIGNED(4, uint32_t, cbu
);
243 DECLARE_ALIGNED(4, uint32_t, bmask
);
244 DECLARE_ALIGNED(4, uint32_t, cgu
);
245 DECLARE_ALIGNED(4, uint32_t, cgv
);
246 DECLARE_ALIGNED(4, uint32_t, gmask
);
250 DECLARE_ALIGNED(8, uint64_t, sparc_coeffs
)[10];
253 /* function pointers for swScale() */
254 void (*yuv2nv12X
)(struct SwsContext
*c
,
255 const int16_t *lumFilter
, const int16_t **lumSrc
, int lumFilterSize
,
256 const int16_t *chrFilter
, const int16_t **chrSrc
, int chrFilterSize
,
257 uint8_t *dest
, uint8_t *uDest
,
258 int dstW
, int chrDstW
, int dstFormat
);
259 void (*yuv2yuv1
)(struct SwsContext
*c
,
260 const int16_t *lumSrc
, const int16_t *chrSrc
, const int16_t *alpSrc
,
262 uint8_t *uDest
, uint8_t *vDest
, uint8_t *aDest
,
263 long dstW
, long chrDstW
);
264 void (*yuv2yuvX
)(struct SwsContext
*c
,
265 const int16_t *lumFilter
, const int16_t **lumSrc
, int lumFilterSize
,
266 const int16_t *chrFilter
, const int16_t **chrSrc
, int chrFilterSize
,
267 const int16_t **alpSrc
,
269 uint8_t *uDest
, uint8_t *vDest
, uint8_t *aDest
,
270 long dstW
, long chrDstW
);
271 void (*yuv2packed1
)(struct SwsContext
*c
,
272 const uint16_t *buf0
,
273 const uint16_t *uvbuf0
, const uint16_t *uvbuf1
,
274 const uint16_t *abuf0
,
276 int dstW
, int uvalpha
, int dstFormat
, int flags
, int y
);
277 void (*yuv2packed2
)(struct SwsContext
*c
,
278 const uint16_t *buf0
, const uint16_t *buf1
,
279 const uint16_t *uvbuf0
, const uint16_t *uvbuf1
,
280 const uint16_t *abuf0
, const uint16_t *abuf1
,
282 int dstW
, int yalpha
, int uvalpha
, int y
);
283 void (*yuv2packedX
)(struct SwsContext
*c
,
284 const int16_t *lumFilter
, const int16_t **lumSrc
, int lumFilterSize
,
285 const int16_t *chrFilter
, const int16_t **chrSrc
, int chrFilterSize
,
286 const int16_t **alpSrc
, uint8_t *dest
,
287 long dstW
, long dstY
);
289 void (*lumToYV12
)(uint8_t *dst
, const uint8_t *src
,
290 long width
, uint32_t *pal
); ///< Unscaled conversion of luma plane to YV12 for horizontal scaler.
291 void (*alpToYV12
)(uint8_t *dst
, const uint8_t *src
,
292 long width
, uint32_t *pal
); ///< Unscaled conversion of alpha plane to YV12 for horizontal scaler.
293 void (*chrToYV12
)(uint8_t *dstU
, uint8_t *dstV
,
294 const uint8_t *src1
, const uint8_t *src2
,
295 long width
, uint32_t *pal
); ///< Unscaled conversion of chroma planes to YV12 for horizontal scaler.
296 void (*hyscale_fast
)(struct SwsContext
*c
,
297 int16_t *dst
, long dstWidth
,
298 const uint8_t *src
, int srcW
, int xInc
);
299 void (*hcscale_fast
)(struct SwsContext
*c
,
300 int16_t *dst
, long dstWidth
,
301 const uint8_t *src1
, const uint8_t *src2
,
304 void (*hScale
)(int16_t *dst
, int dstW
, const uint8_t *src
, int srcW
,
305 int xInc
, const int16_t *filter
, const int16_t *filterPos
,
308 void (*lumConvertRange
)(uint16_t *dst
, int width
); ///< Color range conversion function for luma plane if needed.
309 void (*chrConvertRange
)(uint16_t *dst
, int width
); ///< Color range conversion function for chroma planes if needed.
311 int lumSrcOffset
; ///< Offset given to luma src pointers passed to horizontal input functions.
312 int chrSrcOffset
; ///< Offset given to chroma src pointers passed to horizontal input functions.
313 int alpSrcOffset
; ///< Offset given to alpha src pointers passed to horizontal input functions.
315 int needs_hcscale
; ///< Set if there are chroma planes to be converted.
318 //FIXME check init (where 0)
320 SwsFunc
ff_yuv2rgb_get_func_ptr(SwsContext
*c
);
321 int ff_yuv2rgb_c_init_tables(SwsContext
*c
, const int inv_table
[4],
322 int fullRange
, int brightness
,
323 int contrast
, int saturation
);
325 void ff_yuv2rgb_init_tables_altivec(SwsContext
*c
, const int inv_table
[4],
326 int brightness
, int contrast
, int saturation
);
327 SwsFunc
ff_yuv2rgb_init_mmx(SwsContext
*c
);
328 SwsFunc
ff_yuv2rgb_init_vis(SwsContext
*c
);
329 SwsFunc
ff_yuv2rgb_init_mlib(SwsContext
*c
);
330 SwsFunc
ff_yuv2rgb_init_altivec(SwsContext
*c
);
331 SwsFunc
ff_yuv2rgb_get_func_ptr_bfin(SwsContext
*c
);
332 void ff_bfin_get_unscaled_swscale(SwsContext
*c
);
333 void ff_yuv2packedX_altivec(SwsContext
*c
,
334 const int16_t *lumFilter
, const int16_t **lumSrc
, int lumFilterSize
,
335 const int16_t *chrFilter
, const int16_t **chrSrc
, int chrFilterSize
,
336 uint8_t *dest
, int dstW
, int dstY
);
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_RGB48BE \
345 || (x)==PIX_FMT_RGB48LE \
346 || (x)==PIX_FMT_YUV420P16LE \
347 || (x)==PIX_FMT_YUV422P16LE \
348 || (x)==PIX_FMT_YUV444P16LE \
349 || (x)==PIX_FMT_YUV420P16BE \
350 || (x)==PIX_FMT_YUV422P16BE \
351 || (x)==PIX_FMT_YUV444P16BE \
353 #define isBE(x) ((x)&1)
354 #define isPlanar8YUV(x) ( \
355 (x)==PIX_FMT_YUV410P \
356 || (x)==PIX_FMT_YUV420P \
357 || (x)==PIX_FMT_YUVA420P \
358 || (x)==PIX_FMT_YUV411P \
359 || (x)==PIX_FMT_YUV422P \
360 || (x)==PIX_FMT_YUV444P \
361 || (x)==PIX_FMT_YUV440P \
362 || (x)==PIX_FMT_NV12 \
363 || (x)==PIX_FMT_NV21 \
365 #define isPlanarYUV(x) ( \
367 || (x)==PIX_FMT_YUV420P16LE \
368 || (x)==PIX_FMT_YUV422P16LE \
369 || (x)==PIX_FMT_YUV444P16LE \
370 || (x)==PIX_FMT_YUV420P16BE \
371 || (x)==PIX_FMT_YUV422P16BE \
372 || (x)==PIX_FMT_YUV444P16BE \
375 (x)==PIX_FMT_UYVY422 \
376 || (x)==PIX_FMT_YUYV422 \
379 #define isGray(x) ( \
381 || (x)==PIX_FMT_Y400A \
382 || (x)==PIX_FMT_GRAY16BE \
383 || (x)==PIX_FMT_GRAY16LE \
385 #define isGray16(x) ( \
386 (x)==PIX_FMT_GRAY16BE \
387 || (x)==PIX_FMT_GRAY16LE \
389 #define isRGBinInt(x) ( \
390 (x)==PIX_FMT_RGB48BE \
391 || (x)==PIX_FMT_RGB48LE \
392 || (x)==PIX_FMT_RGB32 \
393 || (x)==PIX_FMT_RGB32_1 \
394 || (x)==PIX_FMT_RGB24 \
395 || (x)==PIX_FMT_RGB565BE \
396 || (x)==PIX_FMT_RGB565LE \
397 || (x)==PIX_FMT_RGB555BE \
398 || (x)==PIX_FMT_RGB555LE \
399 || (x)==PIX_FMT_RGB444BE \
400 || (x)==PIX_FMT_RGB444LE \
401 || (x)==PIX_FMT_RGB8 \
402 || (x)==PIX_FMT_RGB4 \
403 || (x)==PIX_FMT_RGB4_BYTE \
404 || (x)==PIX_FMT_MONOBLACK \
405 || (x)==PIX_FMT_MONOWHITE \
407 #define isBGRinInt(x) ( \
409 || (x)==PIX_FMT_BGR32_1 \
410 || (x)==PIX_FMT_BGR24 \
411 || (x)==PIX_FMT_BGR565BE \
412 || (x)==PIX_FMT_BGR565LE \
413 || (x)==PIX_FMT_BGR555BE \
414 || (x)==PIX_FMT_BGR555LE \
415 || (x)==PIX_FMT_BGR444BE \
416 || (x)==PIX_FMT_BGR444LE \
417 || (x)==PIX_FMT_BGR8 \
418 || (x)==PIX_FMT_BGR4 \
419 || (x)==PIX_FMT_BGR4_BYTE \
420 || (x)==PIX_FMT_MONOBLACK \
421 || (x)==PIX_FMT_MONOWHITE \
423 #define isRGBinBytes(x) ( \
424 (x)==PIX_FMT_RGB48BE \
425 || (x)==PIX_FMT_RGB48LE \
426 || (x)==PIX_FMT_RGBA \
427 || (x)==PIX_FMT_ARGB \
428 || (x)==PIX_FMT_RGB24 \
430 #define isBGRinBytes(x) ( \
432 || (x)==PIX_FMT_ABGR \
433 || (x)==PIX_FMT_BGR24 \
435 #define isAnyRGB(x) ( \
439 #define isALPHA(x) ( \
441 || (x)==PIX_FMT_BGR32_1 \
442 || (x)==PIX_FMT_RGB32 \
443 || (x)==PIX_FMT_RGB32_1 \
444 || (x)==PIX_FMT_Y400A \
445 || (x)==PIX_FMT_YUVA420P \
447 #define usePal(x) ((av_pix_fmt_descriptors[x].flags & PIX_FMT_PAL) || (x) == PIX_FMT_Y400A)
449 extern const uint64_t ff_dither4
[2];
450 extern const uint64_t ff_dither8
[2];
452 extern const AVClass sws_context_class
;
455 * Sets c->swScale to an unscaled converter if one exists for the specific
456 * source and destination formats, bit depths, flags, etc.
458 void ff_get_unscaled_swscale(SwsContext
*c
);
461 * Returns the SWS_CPU_CAPS for the optimized code compiled into swscale.
463 int ff_hardcodedcpuflags(void);
466 * Returns function pointer to fastest main scaler path function depending
467 * on architecture and available optimizations.
469 SwsFunc
ff_getSwsFunc(SwsContext
*c
);
471 #endif /* SWSCALE_SWSCALE_INTERNAL_H */