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 #include "version.h"
26
27 #include "libavutil/avassert.h"
28 #include "libavutil/avutil.h"
29 #include "libavutil/common.h"
30 #include "libavutil/intreadwrite.h"
31 #include "libavutil/log.h"
32 #include "libavutil/pixfmt.h"
33 #include "libavutil/pixdesc.h"
34 #include "libavutil/ppc/util_altivec.h"
35
36 #define STR(s) AV_TOSTRING(s) // AV_STRINGIFY is too long
37
38 #define YUVRGB_TABLE_HEADROOM 512
39 #define YUVRGB_TABLE_LUMA_HEADROOM 512
40
41 #define MAX_FILTER_SIZE SWS_MAX_FILTER_SIZE
42
43 #define DITHER1XBPP
44
45 #if HAVE_BIGENDIAN
46 #define ALT32_CORR (-1)
47 #else
48 #define ALT32_CORR 1
49 #endif
50
51 #if ARCH_X86_64
52 # define APCK_PTR2 8
53 # define APCK_COEF 16
54 # define APCK_SIZE 24
55 #else
56 # define APCK_PTR2 4
57 # define APCK_COEF 8
58 # define APCK_SIZE 16
59 #endif
60
61 #define RETCODE_USE_CASCADE -12345
62
63 struct SwsContext;
64
65 typedef enum SwsDither {
66 SWS_DITHER_NONE = 0,
67 SWS_DITHER_AUTO,
68 SWS_DITHER_BAYER,
69 SWS_DITHER_ED,
70 SWS_DITHER_A_DITHER,
71 SWS_DITHER_X_DITHER,
72 NB_SWS_DITHER,
73 } SwsDither;
74
75 typedef enum SwsAlphaBlend {
76 SWS_ALPHA_BLEND_NONE = 0,
77 SWS_ALPHA_BLEND_UNIFORM,
78 SWS_ALPHA_BLEND_CHECKERBOARD,
79 SWS_ALPHA_BLEND_NB,
80 } SwsAlphaBlend;
81
82 typedef int (*SwsFunc)(struct SwsContext *context, const uint8_t *src[],
83 int srcStride[], int srcSliceY, int srcSliceH,
84 uint8_t *dst[], int dstStride[]);
85
86 /**
87 * Write one line of horizontally scaled data to planar output
88 * without any additional vertical scaling (or point-scaling).
89 *
90 * @param src scaled source data, 15 bits for 8-10-bit output,
91 * 19 bits for 16-bit output (in int32_t)
92 * @param dest pointer to the output plane. For >8-bit
93 * output, this is in uint16_t
94 * @param dstW width of destination in pixels
95 * @param dither ordered dither array of type int16_t and size 8
96 * @param offset Dither offset
97 */
98 typedef void (*yuv2planar1_fn)(const int16_t *src, uint8_t *dest, int dstW,
99 const uint8_t *dither, int offset);
100
101 /**
102 * Write one line of horizontally scaled data to planar output
103 * with multi-point vertical scaling between input pixels.
104 *
105 * @param filter vertical luma/alpha scaling coefficients, 12 bits [0,4096]
106 * @param src scaled luma (Y) or alpha (A) source data, 15 bits for
107 * 8-10-bit output, 19 bits for 16-bit output (in int32_t)
108 * @param filterSize number of vertical input lines to scale
109 * @param dest pointer to output plane. For >8-bit
110 * output, this is in uint16_t
111 * @param dstW width of destination pixels
112 * @param offset Dither offset
113 */
114 typedef void (*yuv2planarX_fn)(const int16_t *filter, int filterSize,
115 const int16_t **src, uint8_t *dest, int dstW,
116 const uint8_t *dither, int offset);
117
118 /**
119 * Write one line of horizontally scaled chroma to interleaved output
120 * with multi-point vertical scaling between input pixels.
121 *
122 * @param c SWS scaling context
123 * @param chrFilter vertical chroma scaling coefficients, 12 bits [0,4096]
124 * @param chrUSrc scaled chroma (U) source data, 15 bits for 8-10-bit
125 * output, 19 bits for 16-bit output (in int32_t)
126 * @param chrVSrc scaled chroma (V) source data, 15 bits for 8-10-bit
127 * output, 19 bits for 16-bit output (in int32_t)
128 * @param chrFilterSize number of vertical chroma input lines to scale
129 * @param dest pointer to the output plane. For >8-bit
130 * output, this is in uint16_t
131 * @param dstW width of chroma planes
132 */
133 typedef void (*yuv2interleavedX_fn)(struct SwsContext *c,
134 const int16_t *chrFilter,
135 int chrFilterSize,
136 const int16_t **chrUSrc,
137 const int16_t **chrVSrc,
138 uint8_t *dest, int dstW);
139
140 /**
141 * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB
142 * output without any additional vertical scaling (or point-scaling). Note
143 * that this function may do chroma scaling, see the "uvalpha" argument.
144 *
145 * @param c SWS scaling context
146 * @param lumSrc scaled luma (Y) source data, 15 bits for 8-10-bit output,
147 * 19 bits for 16-bit output (in int32_t)
148 * @param chrUSrc scaled chroma (U) source data, 15 bits for 8-10-bit output,
149 * 19 bits for 16-bit output (in int32_t)
150 * @param chrVSrc scaled chroma (V) source data, 15 bits for 8-10-bit output,
151 * 19 bits for 16-bit output (in int32_t)
152 * @param alpSrc scaled alpha (A) source data, 15 bits for 8-10-bit output,
153 * 19 bits for 16-bit output (in int32_t)
154 * @param dest pointer to the output plane. For 16-bit output, this is
155 * uint16_t
156 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
157 * to write into dest[]
158 * @param uvalpha chroma scaling coefficient for the second line of chroma
159 * pixels, either 2048 or 0. If 0, one chroma input is used
160 * for 2 output pixels (or if the SWS_FLAG_FULL_CHR_INT flag
161 * is set, it generates 1 output pixel). If 2048, two chroma
162 * input pixels should be averaged for 2 output pixels (this
163 * only happens if SWS_FLAG_FULL_CHR_INT is not set)
164 * @param y vertical line number for this output. This does not need
165 * to be used to calculate the offset in the destination,
166 * but can be used to generate comfort noise using dithering
167 * for some output formats.
168 */
169 typedef void (*yuv2packed1_fn)(struct SwsContext *c, const int16_t *lumSrc,
170 const int16_t *chrUSrc[2],
171 const int16_t *chrVSrc[2],
172 const int16_t *alpSrc, uint8_t *dest,
173 int dstW, int uvalpha, int y);
174 /**
175 * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB
176 * output by doing bilinear scaling between two input lines.
177 *
178 * @param c SWS scaling context
179 * @param lumSrc scaled luma (Y) source data, 15 bits for 8-10-bit output,
180 * 19 bits for 16-bit output (in int32_t)
181 * @param chrUSrc scaled chroma (U) source data, 15 bits for 8-10-bit output,
182 * 19 bits for 16-bit output (in int32_t)
183 * @param chrVSrc scaled chroma (V) source data, 15 bits for 8-10-bit output,
184 * 19 bits for 16-bit output (in int32_t)
185 * @param alpSrc scaled alpha (A) source data, 15 bits for 8-10-bit output,
186 * 19 bits for 16-bit output (in int32_t)
187 * @param dest pointer to the output plane. For 16-bit output, this is
188 * uint16_t
189 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
190 * to write into dest[]
191 * @param yalpha luma/alpha scaling coefficients for the second input line.
192 * The first line's coefficients can be calculated by using
193 * 4096 - yalpha
194 * @param uvalpha chroma scaling coefficient for the second input line. The
195 * first line's coefficients can be calculated by using
196 * 4096 - uvalpha
197 * @param y vertical line number for this output. This does not need
198 * to be used to calculate the offset in the destination,
199 * but can be used to generate comfort noise using dithering
200 * for some output formats.
201 */
202 typedef void (*yuv2packed2_fn)(struct SwsContext *c, const int16_t *lumSrc[2],
203 const int16_t *chrUSrc[2],
204 const int16_t *chrVSrc[2],
205 const int16_t *alpSrc[2],
206 uint8_t *dest,
207 int dstW, int yalpha, int uvalpha, int y);
208 /**
209 * Write one line of horizontally scaled Y/U/V/A to packed-pixel YUV/RGB
210 * output by doing multi-point vertical scaling between input pixels.
211 *
212 * @param c SWS scaling context
213 * @param lumFilter vertical luma/alpha scaling coefficients, 12 bits [0,4096]
214 * @param lumSrc scaled luma (Y) source data, 15 bits for 8-10-bit output,
215 * 19 bits for 16-bit output (in int32_t)
216 * @param lumFilterSize number of vertical luma/alpha input lines to scale
217 * @param chrFilter vertical chroma scaling coefficients, 12 bits [0,4096]
218 * @param chrUSrc scaled chroma (U) source data, 15 bits for 8-10-bit output,
219 * 19 bits for 16-bit output (in int32_t)
220 * @param chrVSrc scaled chroma (V) source data, 15 bits for 8-10-bit output,
221 * 19 bits for 16-bit output (in int32_t)
222 * @param chrFilterSize number of vertical chroma input lines to scale
223 * @param alpSrc scaled alpha (A) source data, 15 bits for 8-10-bit output,
224 * 19 bits for 16-bit output (in int32_t)
225 * @param dest pointer to the output plane. For 16-bit output, this is
226 * uint16_t
227 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
228 * to write into dest[]
229 * @param y vertical line number for this output. This does not need
230 * to be used to calculate the offset in the destination,
231 * but can be used to generate comfort noise using dithering
232 * or some output formats.
233 */
234 typedef void (*yuv2packedX_fn)(struct SwsContext *c, const int16_t *lumFilter,
235 const int16_t **lumSrc, int lumFilterSize,
236 const int16_t *chrFilter,
237 const int16_t **chrUSrc,
238 const int16_t **chrVSrc, int chrFilterSize,
239 const int16_t **alpSrc, uint8_t *dest,
240 int dstW, int y);
241
242 /**
243 * Write one line of horizontally scaled Y/U/V/A to YUV/RGB
244 * output by doing multi-point vertical scaling between input pixels.
245 *
246 * @param c SWS scaling context
247 * @param lumFilter vertical luma/alpha scaling coefficients, 12 bits [0,4096]
248 * @param lumSrc scaled luma (Y) source data, 15 bits for 8-10-bit output,
249 * 19 bits for 16-bit output (in int32_t)
250 * @param lumFilterSize number of vertical luma/alpha input lines to scale
251 * @param chrFilter vertical chroma scaling coefficients, 12 bits [0,4096]
252 * @param chrUSrc scaled chroma (U) source data, 15 bits for 8-10-bit output,
253 * 19 bits for 16-bit output (in int32_t)
254 * @param chrVSrc scaled chroma (V) source data, 15 bits for 8-10-bit output,
255 * 19 bits for 16-bit output (in int32_t)
256 * @param chrFilterSize number of vertical chroma input lines to scale
257 * @param alpSrc scaled alpha (A) source data, 15 bits for 8-10-bit output,
258 * 19 bits for 16-bit output (in int32_t)
259 * @param dest pointer to the output planes. For 16-bit output, this is
260 * uint16_t
261 * @param dstW width of lumSrc and alpSrc in pixels, number of pixels
262 * to write into dest[]
263 * @param y vertical line number for this output. This does not need
264 * to be used to calculate the offset in the destination,
265 * but can be used to generate comfort noise using dithering
266 * or some output formats.
267 */
268 typedef void (*yuv2anyX_fn)(struct SwsContext *c, const int16_t *lumFilter,
269 const int16_t **lumSrc, int lumFilterSize,
270 const int16_t *chrFilter,
271 const int16_t **chrUSrc,
272 const int16_t **chrVSrc, int chrFilterSize,
273 const int16_t **alpSrc, uint8_t **dest,
274 int dstW, int y);
275
276 struct SwsSlice;
277 struct SwsFilterDescriptor;
278
279 /* This struct should be aligned on at least a 32-byte boundary. */
280 typedef struct SwsContext {
281 /**
282 * info on struct for av_log
283 */
284 const AVClass *av_class;
285
286 /**
287 * Note that src, dst, srcStride, dstStride will be copied in the
288 * sws_scale() wrapper so they can be freely modified here.
289 */
290 SwsFunc swscale;
291 int srcW; ///< Width of source luma/alpha planes.
292 int srcH; ///< Height of source luma/alpha planes.
293 int dstH; ///< Height of destination luma/alpha planes.
294 int chrSrcW; ///< Width of source chroma planes.
295 int chrSrcH; ///< Height of source chroma planes.
296 int chrDstW; ///< Width of destination chroma planes.
297 int chrDstH; ///< Height of destination chroma planes.
298 int lumXInc, chrXInc;
299 int lumYInc, chrYInc;
300 enum AVPixelFormat dstFormat; ///< Destination pixel format.
301 enum AVPixelFormat srcFormat; ///< Source pixel format.
302 int dstFormatBpp; ///< Number of bits per pixel of the destination pixel format.
303 int srcFormatBpp; ///< Number of bits per pixel of the source pixel format.
304 int dstBpc, srcBpc;
305 int chrSrcHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in source image.
306 int chrSrcVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in source image.
307 int chrDstHSubSample; ///< Binary logarithm of horizontal subsampling factor between luma/alpha and chroma planes in destination image.
308 int chrDstVSubSample; ///< Binary logarithm of vertical subsampling factor between luma/alpha and chroma planes in destination image.
309 int vChrDrop; ///< Binary logarithm of extra vertical subsampling factor in source image chroma planes specified by user.
310 int sliceDir; ///< Direction that slices are fed to the scaler (1 = top-to-bottom, -1 = bottom-to-top).
311 double param[2]; ///< Input parameters for scaling algorithms that need them.
312
313 /* The cascaded_* fields allow spliting a scaler task into multiple
314 * sequential steps, this is for example used to limit the maximum
315 * downscaling factor that needs to be supported in one scaler.
316 */
317 struct SwsContext *cascaded_context[3];
318 int cascaded_tmpStride[4];
319 uint8_t *cascaded_tmp[4];
320 int cascaded1_tmpStride[4];
321 uint8_t *cascaded1_tmp[4];
322 int cascaded_mainindex;
323
324 double gamma_value;
325 int gamma_flag;
326 int is_internal_gamma;
327 uint16_t *gamma;
328 uint16_t *inv_gamma;
329
330 int numDesc;
331 int descIndex[2];
332 int numSlice;
333 struct SwsSlice *slice;
334 struct SwsFilterDescriptor *desc;
335
336 uint32_t pal_yuv[256];
337 uint32_t pal_rgb[256];
338
339 float uint2float_lut[256];
340
341 /**
342 * @name Scaled horizontal lines ring buffer.
343 * The horizontal scaler keeps just enough scaled lines in a ring buffer
344 * so they may be passed to the vertical scaler. The pointers to the
345 * allocated buffers for each line are duplicated in sequence in the ring
346 * buffer to simplify indexing and avoid wrapping around between lines
347 * inside the vertical scaler code. The wrapping is done before the
348 * vertical scaler is called.
349 */
350 //@{
351 int lastInLumBuf; ///< Last scaled horizontal luma/alpha line from source in the ring buffer.
352 int lastInChrBuf; ///< Last scaled horizontal chroma line from source in the ring buffer.
353 //@}
354
355 uint8_t *formatConvBuffer;
356 int needAlpha;
357
358 /**
359 * @name Horizontal and vertical filters.
360 * To better understand the following fields, here is a pseudo-code of
361 * their usage in filtering a horizontal line:
362 * @code
363 * for (i = 0; i < width; i++) {
364 * dst[i] = 0;
365 * for (j = 0; j < filterSize; j++)
366 * dst[i] += src[ filterPos[i] + j ] * filter[ filterSize * i + j ];
367 * dst[i] >>= FRAC_BITS; // The actual implementation is fixed-point.
368 * }
369 * @endcode
370 */
371 //@{
372 int16_t *hLumFilter; ///< Array of horizontal filter coefficients for luma/alpha planes.
373 int16_t *hChrFilter; ///< Array of horizontal filter coefficients for chroma planes.
374 int16_t *vLumFilter; ///< Array of vertical filter coefficients for luma/alpha planes.
375 int16_t *vChrFilter; ///< Array of vertical filter coefficients for chroma planes.
376 int32_t *hLumFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for luma/alpha planes.
377 int32_t *hChrFilterPos; ///< Array of horizontal filter starting positions for each dst[i] for chroma planes.
378 int32_t *vLumFilterPos; ///< Array of vertical filter starting positions for each dst[i] for luma/alpha planes.
379 int32_t *vChrFilterPos; ///< Array of vertical filter starting positions for each dst[i] for chroma planes.
380 int hLumFilterSize; ///< Horizontal filter size for luma/alpha pixels.
381 int hChrFilterSize; ///< Horizontal filter size for chroma pixels.
382 int vLumFilterSize; ///< Vertical filter size for luma/alpha pixels.
383 int vChrFilterSize; ///< Vertical filter size for chroma pixels.
384 //@}
385
386 int lumMmxextFilterCodeSize; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code size for luma/alpha planes.
387 int chrMmxextFilterCodeSize; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code size for chroma planes.
388 uint8_t *lumMmxextFilterCode; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code for luma/alpha planes.
389 uint8_t *chrMmxextFilterCode; ///< Runtime-generated MMXEXT horizontal fast bilinear scaler code for chroma planes.
390
391 int canMMXEXTBeUsed;
392 int warned_unuseable_bilinear;
393
394 int dstY; ///< Last destination vertical line output from last slice.
395 int flags; ///< Flags passed by the user to select scaler algorithm, optimizations, subsampling, etc...
396 void *yuvTable; // pointer to the yuv->rgb table start so it can be freed()
397 // alignment ensures the offset can be added in a single
398 // instruction on e.g. ARM
399 DECLARE_ALIGNED(16, int, table_gV)[256 + 2*YUVRGB_TABLE_HEADROOM];
400 uint8_t *table_rV[256 + 2*YUVRGB_TABLE_HEADROOM];
401 uint8_t *table_gU[256 + 2*YUVRGB_TABLE_HEADROOM];
402 uint8_t *table_bU[256 + 2*YUVRGB_TABLE_HEADROOM];
403 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
404 #define RY_IDX 0
405 #define GY_IDX 1
406 #define BY_IDX 2
407 #define RU_IDX 3
408 #define GU_IDX 4
409 #define BU_IDX 5
410 #define RV_IDX 6
411 #define GV_IDX 7
412 #define BV_IDX 8
413 #define RGB2YUV_SHIFT 15
414
415 int *dither_error[4];
416
417 //Colorspace stuff
418 int contrast, brightness, saturation; // for sws_getColorspaceDetails
419 int srcColorspaceTable[4];
420 int dstColorspaceTable[4];
421 int srcRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (source image).
422 int dstRange; ///< 0 = MPG YUV range, 1 = JPG YUV range (destination image).
423 int src0Alpha;
424 int dst0Alpha;
425 int srcXYZ;
426 int dstXYZ;
427 int src_h_chr_pos;
428 int dst_h_chr_pos;
429 int src_v_chr_pos;
430 int dst_v_chr_pos;
431 int yuv2rgb_y_offset;
432 int yuv2rgb_y_coeff;
433 int yuv2rgb_v2r_coeff;
434 int yuv2rgb_v2g_coeff;
435 int yuv2rgb_u2g_coeff;
436 int yuv2rgb_u2b_coeff;
437
438 #define RED_DITHER "0*8"
439 #define GREEN_DITHER "1*8"
440 #define BLUE_DITHER "2*8"
441 #define Y_COEFF "3*8"
442 #define VR_COEFF "4*8"
443 #define UB_COEFF "5*8"
444 #define VG_COEFF "6*8"
445 #define UG_COEFF "7*8"
446 #define Y_OFFSET "8*8"
447 #define U_OFFSET "9*8"
448 #define V_OFFSET "10*8"
449 #define LUM_MMX_FILTER_OFFSET "11*8"
450 #define CHR_MMX_FILTER_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)
451 #define DSTW_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2"
452 #define ESP_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+8"
453 #define VROUNDER_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+16"
454 #define U_TEMP "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+24"
455 #define V_TEMP "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+32"
456 #define Y_TEMP "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+40"
457 #define ALP_MMX_FILTER_OFFSET "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*2+48"
458 #define UV_OFF_PX "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+48"
459 #define UV_OFF_BYTE "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+56"
460 #define DITHER16 "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+64"
461 #define DITHER32 "11*8+4*4*"AV_STRINGIFY(MAX_FILTER_SIZE)"*3+80"
462 #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
463
464 DECLARE_ALIGNED(8, uint64_t, redDither);
465 DECLARE_ALIGNED(8, uint64_t, greenDither);
466 DECLARE_ALIGNED(8, uint64_t, blueDither);
467
468 DECLARE_ALIGNED(8, uint64_t, yCoeff);
469 DECLARE_ALIGNED(8, uint64_t, vrCoeff);
470 DECLARE_ALIGNED(8, uint64_t, ubCoeff);
471 DECLARE_ALIGNED(8, uint64_t, vgCoeff);
472 DECLARE_ALIGNED(8, uint64_t, ugCoeff);
473 DECLARE_ALIGNED(8, uint64_t, yOffset);
474 DECLARE_ALIGNED(8, uint64_t, uOffset);
475 DECLARE_ALIGNED(8, uint64_t, vOffset);
476 int32_t lumMmxFilter[4 * MAX_FILTER_SIZE];
477 int32_t chrMmxFilter[4 * MAX_FILTER_SIZE];
478 int dstW; ///< Width of destination luma/alpha planes.
479 DECLARE_ALIGNED(8, uint64_t, esp);
480 DECLARE_ALIGNED(8, uint64_t, vRounder);
481 DECLARE_ALIGNED(8, uint64_t, u_temp);
482 DECLARE_ALIGNED(8, uint64_t, v_temp);
483 DECLARE_ALIGNED(8, uint64_t, y_temp);
484 int32_t alpMmxFilter[4 * MAX_FILTER_SIZE];
485 // alignment of these values is not necessary, but merely here
486 // to maintain the same offset across x8632 and x86-64. Once we
487 // use proper offset macros in the asm, they can be removed.
488 DECLARE_ALIGNED(8, ptrdiff_t, uv_off); ///< offset (in pixels) between u and v planes
489 DECLARE_ALIGNED(8, ptrdiff_t, uv_offx2); ///< offset (in bytes) between u and v planes
490 DECLARE_ALIGNED(8, uint16_t, dither16)[8];
491 DECLARE_ALIGNED(8, uint32_t, dither32)[8];
492
493 const uint8_t *chrDither8, *lumDither8;
494
495 #if HAVE_ALTIVEC
496 vector signed short CY;
497 vector signed short CRV;
498 vector signed short CBU;
499 vector signed short CGU;
500 vector signed short CGV;
501 vector signed short OY;
502 vector unsigned short CSHIFT;
503 vector signed short *vYCoeffsBank, *vCCoeffsBank;
504 #endif
505
506 int use_mmx_vfilter;
507
508 /* pre defined color-spaces gamma */
509 #define XYZ_GAMMA (2.6f)
510 #define RGB_GAMMA (2.2f)
511 int16_t *xyzgamma;
512 int16_t *rgbgamma;
513 int16_t *xyzgammainv;
514 int16_t *rgbgammainv;
515 int16_t xyz2rgb_matrix[3][4];
516 int16_t rgb2xyz_matrix[3][4];
517
518 /* function pointers for swscale() */
519 yuv2planar1_fn yuv2plane1;
520 yuv2planarX_fn yuv2planeX;
521 yuv2interleavedX_fn yuv2nv12cX;
522 yuv2packed1_fn yuv2packed1;
523 yuv2packed2_fn yuv2packed2;
524 yuv2packedX_fn yuv2packedX;
525 yuv2anyX_fn yuv2anyX;
526
527 /// Unscaled conversion of luma plane to YV12 for horizontal scaler.
528 void (*lumToYV12)(uint8_t *dst, const uint8_t *src, const uint8_t *src2, const uint8_t *src3,
529 int width, uint32_t *pal);
530 /// Unscaled conversion of alpha plane to YV12 for horizontal scaler.
531 void (*alpToYV12)(uint8_t *dst, const uint8_t *src, const uint8_t *src2, const uint8_t *src3,
532 int width, uint32_t *pal);
533 /// Unscaled conversion of chroma planes to YV12 for horizontal scaler.
534 void (*chrToYV12)(uint8_t *dstU, uint8_t *dstV,
535 const uint8_t *src1, const uint8_t *src2, const uint8_t *src3,
536 int width, uint32_t *pal);
537
538 /**
539 * Functions to read planar input, such as planar RGB, and convert
540 * internally to Y/UV/A.
541 */
542 /** @{ */
543 void (*readLumPlanar)(uint8_t *dst, const uint8_t *src[4], int width, int32_t *rgb2yuv);
544 void (*readChrPlanar)(uint8_t *dstU, uint8_t *dstV, const uint8_t *src[4],
545 int width, int32_t *rgb2yuv);
546 void (*readAlpPlanar)(uint8_t *dst, const uint8_t *src[4], int width, int32_t *rgb2yuv);
547 /** @} */
548
549 /**
550 * Scale one horizontal line of input data using a bilinear filter
551 * to produce one line of output data. Compared to SwsContext->hScale(),
552 * please take note of the following caveats when using these:
553 * - Scaling is done using only 7 bits instead of 14-bit coefficients.
554 * - You can use no more than 5 input pixels to produce 4 output
555 * pixels. Therefore, this filter should not be used for downscaling
556 * by more than ~20% in width (because that equals more than 5/4th
557 * downscaling and thus more than 5 pixels input per 4 pixels output).
558 * - In general, bilinear filters create artifacts during downscaling
559 * (even when <20%), because one output pixel will span more than one
560 * input pixel, and thus some pixels will need edges of both neighbor
561 * pixels to interpolate the output pixel. Since you can use at most
562 * two input pixels per output pixel in bilinear scaling, this is
563 * impossible and thus downscaling by any size will create artifacts.
564 * To enable this type of scaling, set SWS_FLAG_FAST_BILINEAR
565 * in SwsContext->flags.
566 */
567 /** @{ */
568 void (*hyscale_fast)(struct SwsContext *c,
569 int16_t *dst, int dstWidth,
570 const uint8_t *src, int srcW, int xInc);
571 void (*hcscale_fast)(struct SwsContext *c,
572 int16_t *dst1, int16_t *dst2, int dstWidth,
573 const uint8_t *src1, const uint8_t *src2,
574 int srcW, int xInc);
575 /** @} */
576
577 /**
578 * Scale one horizontal line of input data using a filter over the input
579 * lines, to produce one (differently sized) line of output data.
580 *
581 * @param dst pointer to destination buffer for horizontally scaled
582 * data. If the number of bits per component of one
583 * destination pixel (SwsContext->dstBpc) is <= 10, data
584 * will be 15 bpc in 16 bits (int16_t) width. Else (i.e.
585 * SwsContext->dstBpc == 16), data will be 19bpc in
586 * 32 bits (int32_t) width.
587 * @param dstW width of destination image
588 * @param src pointer to source data to be scaled. If the number of
589 * bits per component of a source pixel (SwsContext->srcBpc)
590 * is 8, this is 8bpc in 8 bits (uint8_t) width. Else
591 * (i.e. SwsContext->dstBpc > 8), this is native depth
592 * in 16 bits (uint16_t) width. In other words, for 9-bit
593 * YUV input, this is 9bpc, for 10-bit YUV input, this is
594 * 10bpc, and for 16-bit RGB or YUV, this is 16bpc.
595 * @param filter filter coefficients to be used per output pixel for
596 * scaling. This contains 14bpp filtering coefficients.
597 * Guaranteed to contain dstW * filterSize entries.
598 * @param filterPos position of the first input pixel to be used for
599 * each output pixel during scaling. Guaranteed to
600 * contain dstW entries.
601 * @param filterSize the number of input coefficients to be used (and
602 * thus the number of input pixels to be used) for
603 * creating a single output pixel. Is aligned to 4
604 * (and input coefficients thus padded with zeroes)
605 * to simplify creating SIMD code.
606 */
607 /** @{ */
608 void (*hyScale)(struct SwsContext *c, int16_t *dst, int dstW,
609 const uint8_t *src, const int16_t *filter,
610 const int32_t *filterPos, int filterSize);
611 void (*hcScale)(struct SwsContext *c, int16_t *dst, int dstW,
612 const uint8_t *src, const int16_t *filter,
613 const int32_t *filterPos, int filterSize);
614 /** @} */
615
616 /// Color range conversion function for luma plane if needed.
617 void (*lumConvertRange)(int16_t *dst, int width);
618 /// Color range conversion function for chroma planes if needed.
619 void (*chrConvertRange)(int16_t *dst1, int16_t *dst2, int width);
620
621 int needs_hcscale; ///< Set if there are chroma planes to be converted.
622
623 SwsDither dither;
624
625 SwsAlphaBlend alphablend;
626 } SwsContext;
627 //FIXME check init (where 0)
628
629 SwsFunc ff_yuv2rgb_get_func_ptr(SwsContext *c);
630 int ff_yuv2rgb_c_init_tables(SwsContext *c, const int inv_table[4],
631 int fullRange, int brightness,
632 int contrast, int saturation);
633 void ff_yuv2rgb_init_tables_ppc(SwsContext *c, const int inv_table[4],
634 int brightness, int contrast, int saturation);
635
636 void ff_updateMMXDitherTables(SwsContext *c, int dstY);
637
638 av_cold void ff_sws_init_range_convert(SwsContext *c);
639
640 SwsFunc ff_yuv2rgb_init_x86(SwsContext *c);
641 SwsFunc ff_yuv2rgb_init_ppc(SwsContext *c);
642
is16BPS(enum AVPixelFormat pix_fmt)643 static av_always_inline int is16BPS(enum AVPixelFormat pix_fmt)
644 {
645 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
646 av_assert0(desc);
647 return desc->comp[0].depth == 16;
648 }
649
is32BPS(enum AVPixelFormat pix_fmt)650 static av_always_inline int is32BPS(enum AVPixelFormat pix_fmt)
651 {
652 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
653 av_assert0(desc);
654 return desc->comp[0].depth == 32;
655 }
656
isNBPS(enum AVPixelFormat pix_fmt)657 static av_always_inline int isNBPS(enum AVPixelFormat pix_fmt)
658 {
659 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
660 av_assert0(desc);
661 return desc->comp[0].depth >= 9 && desc->comp[0].depth <= 14;
662 }
663
isBE(enum AVPixelFormat pix_fmt)664 static av_always_inline int isBE(enum AVPixelFormat pix_fmt)
665 {
666 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
667 av_assert0(desc);
668 return desc->flags & AV_PIX_FMT_FLAG_BE;
669 }
670
isYUV(enum AVPixelFormat pix_fmt)671 static av_always_inline int isYUV(enum AVPixelFormat pix_fmt)
672 {
673 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
674 av_assert0(desc);
675 return !(desc->flags & AV_PIX_FMT_FLAG_RGB) && desc->nb_components >= 2;
676 }
677
isPlanarYUV(enum AVPixelFormat pix_fmt)678 static av_always_inline int isPlanarYUV(enum AVPixelFormat pix_fmt)
679 {
680 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
681 av_assert0(desc);
682 return ((desc->flags & AV_PIX_FMT_FLAG_PLANAR) && isYUV(pix_fmt));
683 }
684
685 /*
686 * Identity semi-planar YUV formats. Specifically, those are YUV formats
687 * where the second and third components (U & V) are on the same plane.
688 */
isSemiPlanarYUV(enum AVPixelFormat pix_fmt)689 static av_always_inline int isSemiPlanarYUV(enum AVPixelFormat pix_fmt)
690 {
691 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
692 av_assert0(desc);
693 return (isPlanarYUV(pix_fmt) && desc->comp[1].plane == desc->comp[2].plane);
694 }
695
isRGB(enum AVPixelFormat pix_fmt)696 static av_always_inline int isRGB(enum AVPixelFormat pix_fmt)
697 {
698 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
699 av_assert0(desc);
700 return (desc->flags & AV_PIX_FMT_FLAG_RGB);
701 }
702
isGray(enum AVPixelFormat pix_fmt)703 static av_always_inline int isGray(enum AVPixelFormat pix_fmt)
704 {
705 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
706 av_assert0(desc);
707 return !(desc->flags & AV_PIX_FMT_FLAG_PAL) &&
708 !(desc->flags & AV_PIX_FMT_FLAG_HWACCEL) &&
709 desc->nb_components <= 2 &&
710 pix_fmt != AV_PIX_FMT_MONOBLACK &&
711 pix_fmt != AV_PIX_FMT_MONOWHITE;
712 }
713
isRGBinInt(enum AVPixelFormat pix_fmt)714 static av_always_inline int isRGBinInt(enum AVPixelFormat pix_fmt)
715 {
716 return pix_fmt == AV_PIX_FMT_RGB48BE ||
717 pix_fmt == AV_PIX_FMT_RGB48LE ||
718 pix_fmt == AV_PIX_FMT_RGB32 ||
719 pix_fmt == AV_PIX_FMT_RGB32_1 ||
720 pix_fmt == AV_PIX_FMT_RGB24 ||
721 pix_fmt == AV_PIX_FMT_RGB565BE ||
722 pix_fmt == AV_PIX_FMT_RGB565LE ||
723 pix_fmt == AV_PIX_FMT_RGB555BE ||
724 pix_fmt == AV_PIX_FMT_RGB555LE ||
725 pix_fmt == AV_PIX_FMT_RGB444BE ||
726 pix_fmt == AV_PIX_FMT_RGB444LE ||
727 pix_fmt == AV_PIX_FMT_RGB8 ||
728 pix_fmt == AV_PIX_FMT_RGB4 ||
729 pix_fmt == AV_PIX_FMT_RGB4_BYTE ||
730 pix_fmt == AV_PIX_FMT_RGBA64BE ||
731 pix_fmt == AV_PIX_FMT_RGBA64LE ||
732 pix_fmt == AV_PIX_FMT_MONOBLACK ||
733 pix_fmt == AV_PIX_FMT_MONOWHITE;
734 }
735
isBGRinInt(enum AVPixelFormat pix_fmt)736 static av_always_inline int isBGRinInt(enum AVPixelFormat pix_fmt)
737 {
738 return pix_fmt == AV_PIX_FMT_BGR48BE ||
739 pix_fmt == AV_PIX_FMT_BGR48LE ||
740 pix_fmt == AV_PIX_FMT_BGR32 ||
741 pix_fmt == AV_PIX_FMT_BGR32_1 ||
742 pix_fmt == AV_PIX_FMT_BGR24 ||
743 pix_fmt == AV_PIX_FMT_BGR565BE ||
744 pix_fmt == AV_PIX_FMT_BGR565LE ||
745 pix_fmt == AV_PIX_FMT_BGR555BE ||
746 pix_fmt == AV_PIX_FMT_BGR555LE ||
747 pix_fmt == AV_PIX_FMT_BGR444BE ||
748 pix_fmt == AV_PIX_FMT_BGR444LE ||
749 pix_fmt == AV_PIX_FMT_BGR8 ||
750 pix_fmt == AV_PIX_FMT_BGR4 ||
751 pix_fmt == AV_PIX_FMT_BGR4_BYTE ||
752 pix_fmt == AV_PIX_FMT_BGRA64BE ||
753 pix_fmt == AV_PIX_FMT_BGRA64LE ||
754 pix_fmt == AV_PIX_FMT_MONOBLACK ||
755 pix_fmt == AV_PIX_FMT_MONOWHITE;
756 }
757
isBayer(enum AVPixelFormat pix_fmt)758 static av_always_inline int isBayer(enum AVPixelFormat pix_fmt)
759 {
760 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
761 av_assert0(desc);
762 return !!(desc->flags & AV_PIX_FMT_FLAG_BAYER);
763 }
764
isAnyRGB(enum AVPixelFormat pix_fmt)765 static av_always_inline int isAnyRGB(enum AVPixelFormat pix_fmt)
766 {
767 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
768 av_assert0(desc);
769 return (desc->flags & AV_PIX_FMT_FLAG_RGB) ||
770 pix_fmt == AV_PIX_FMT_MONOBLACK || pix_fmt == AV_PIX_FMT_MONOWHITE;
771 }
772
isFloat(enum AVPixelFormat pix_fmt)773 static av_always_inline int isFloat(enum AVPixelFormat pix_fmt)
774 {
775 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
776 av_assert0(desc);
777 return desc->flags & AV_PIX_FMT_FLAG_FLOAT;
778 }
779
isALPHA(enum AVPixelFormat pix_fmt)780 static av_always_inline int isALPHA(enum AVPixelFormat pix_fmt)
781 {
782 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
783 av_assert0(desc);
784 if (pix_fmt == AV_PIX_FMT_PAL8)
785 return 1;
786 return desc->flags & AV_PIX_FMT_FLAG_ALPHA;
787 }
788
isPacked(enum AVPixelFormat pix_fmt)789 static av_always_inline int isPacked(enum AVPixelFormat pix_fmt)
790 {
791 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
792 av_assert0(desc);
793 return (desc->nb_components >= 2 && !(desc->flags & AV_PIX_FMT_FLAG_PLANAR)) ||
794 pix_fmt == AV_PIX_FMT_PAL8 ||
795 pix_fmt == AV_PIX_FMT_MONOBLACK || pix_fmt == AV_PIX_FMT_MONOWHITE;
796 }
797
isPlanar(enum AVPixelFormat pix_fmt)798 static av_always_inline int isPlanar(enum AVPixelFormat pix_fmt)
799 {
800 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
801 av_assert0(desc);
802 return (desc->nb_components >= 2 && (desc->flags & AV_PIX_FMT_FLAG_PLANAR));
803 }
804
isPackedRGB(enum AVPixelFormat pix_fmt)805 static av_always_inline int isPackedRGB(enum AVPixelFormat pix_fmt)
806 {
807 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
808 av_assert0(desc);
809 return ((desc->flags & (AV_PIX_FMT_FLAG_PLANAR | AV_PIX_FMT_FLAG_RGB)) == AV_PIX_FMT_FLAG_RGB);
810 }
811
isPlanarRGB(enum AVPixelFormat pix_fmt)812 static av_always_inline int isPlanarRGB(enum AVPixelFormat pix_fmt)
813 {
814 const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(pix_fmt);
815 av_assert0(desc);
816 return ((desc->flags & (AV_PIX_FMT_FLAG_PLANAR | AV_PIX_FMT_FLAG_RGB)) ==
817 (AV_PIX_FMT_FLAG_PLANAR | AV_PIX_FMT_FLAG_RGB));
818 }
819
usePal(enum AVPixelFormat pix_fmt)820 static av_always_inline int usePal(enum AVPixelFormat pix_fmt)
821 {
822 switch (pix_fmt) {
823 case AV_PIX_FMT_PAL8:
824 case AV_PIX_FMT_BGR4_BYTE:
825 case AV_PIX_FMT_BGR8:
826 case AV_PIX_FMT_GRAY8:
827 case AV_PIX_FMT_RGB4_BYTE:
828 case AV_PIX_FMT_RGB8:
829 return 1;
830 default:
831 return 0;
832 }
833 }
834
835 extern const uint64_t ff_dither4[2];
836 extern const uint64_t ff_dither8[2];
837
838 extern const uint8_t ff_dither_2x2_4[3][8];
839 extern const uint8_t ff_dither_2x2_8[3][8];
840 extern const uint8_t ff_dither_4x4_16[5][8];
841 extern const uint8_t ff_dither_8x8_32[9][8];
842 extern const uint8_t ff_dither_8x8_73[9][8];
843 extern const uint8_t ff_dither_8x8_128[9][8];
844 extern const uint8_t ff_dither_8x8_220[9][8];
845
846 extern const int32_t ff_yuv2rgb_coeffs[11][4];
847
848 extern const AVClass ff_sws_context_class;
849
850 /**
851 * Set c->swscale to an unscaled converter if one exists for the specific
852 * source and destination formats, bit depths, flags, etc.
853 */
854 void ff_get_unscaled_swscale(SwsContext *c);
855 void ff_get_unscaled_swscale_ppc(SwsContext *c);
856 void ff_get_unscaled_swscale_arm(SwsContext *c);
857 void ff_get_unscaled_swscale_aarch64(SwsContext *c);
858
859 /**
860 * Return function pointer to fastest main scaler path function depending
861 * on architecture and available optimizations.
862 */
863 SwsFunc ff_getSwsFunc(SwsContext *c);
864
865 void ff_sws_init_input_funcs(SwsContext *c);
866 void ff_sws_init_output_funcs(SwsContext *c,
867 yuv2planar1_fn *yuv2plane1,
868 yuv2planarX_fn *yuv2planeX,
869 yuv2interleavedX_fn *yuv2nv12cX,
870 yuv2packed1_fn *yuv2packed1,
871 yuv2packed2_fn *yuv2packed2,
872 yuv2packedX_fn *yuv2packedX,
873 yuv2anyX_fn *yuv2anyX);
874 void ff_sws_init_swscale_ppc(SwsContext *c);
875 void ff_sws_init_swscale_vsx(SwsContext *c);
876 void ff_sws_init_swscale_x86(SwsContext *c);
877 void ff_sws_init_swscale_aarch64(SwsContext *c);
878 void ff_sws_init_swscale_arm(SwsContext *c);
879
880 void ff_hyscale_fast_c(SwsContext *c, int16_t *dst, int dstWidth,
881 const uint8_t *src, int srcW, int xInc);
882 void ff_hcscale_fast_c(SwsContext *c, int16_t *dst1, int16_t *dst2,
883 int dstWidth, const uint8_t *src1,
884 const uint8_t *src2, int srcW, int xInc);
885 int ff_init_hscaler_mmxext(int dstW, int xInc, uint8_t *filterCode,
886 int16_t *filter, int32_t *filterPos,
887 int numSplits);
888 void ff_hyscale_fast_mmxext(SwsContext *c, int16_t *dst,
889 int dstWidth, const uint8_t *src,
890 int srcW, int xInc);
891 void ff_hcscale_fast_mmxext(SwsContext *c, int16_t *dst1, int16_t *dst2,
892 int dstWidth, const uint8_t *src1,
893 const uint8_t *src2, int srcW, int xInc);
894
895 /**
896 * Allocate and return an SwsContext.
897 * This is like sws_getContext() but does not perform the init step, allowing
898 * the user to set additional AVOptions.
899 *
900 * @see sws_getContext()
901 */
902 struct SwsContext *sws_alloc_set_opts(int srcW, int srcH, enum AVPixelFormat srcFormat,
903 int dstW, int dstH, enum AVPixelFormat dstFormat,
904 int flags, const double *param);
905
906 int ff_sws_alphablendaway(SwsContext *c, const uint8_t *src[],
907 int srcStride[], int srcSliceY, int srcSliceH,
908 uint8_t *dst[], int dstStride[]);
909
fillPlane16(uint8_t * plane,int stride,int width,int height,int y,int alpha,int bits,const int big_endian)910 static inline void fillPlane16(uint8_t *plane, int stride, int width, int height, int y,
911 int alpha, int bits, const int big_endian)
912 {
913 int i, j;
914 uint8_t *ptr = plane + stride * y;
915 int v = alpha ? 0xFFFF>>(16-bits) : (1<<(bits-1));
916 for (i = 0; i < height; i++) {
917 #define FILL(wfunc) \
918 for (j = 0; j < width; j++) {\
919 wfunc(ptr+2*j, v);\
920 }
921 if (big_endian) {
922 FILL(AV_WB16);
923 } else {
924 FILL(AV_WL16);
925 }
926 ptr += stride;
927 }
928 #undef FILL
929 }
930
fillPlane32(uint8_t * plane,int stride,int width,int height,int y,int alpha,int bits,const int big_endian,int is_float)931 static inline void fillPlane32(uint8_t *plane, int stride, int width, int height, int y,
932 int alpha, int bits, const int big_endian, int is_float)
933 {
934 int i, j;
935 uint8_t *ptr = plane + stride * y;
936 uint32_t v;
937 uint32_t onef32 = 0x3f800000;
938 if (is_float)
939 v = alpha ? onef32 : 0;
940 else
941 v = alpha ? 0xFFFFFFFF>>(32-bits) : (1<<(bits-1));
942
943 for (i = 0; i < height; i++) {
944 #define FILL(wfunc) \
945 for (j = 0; j < width; j++) {\
946 wfunc(ptr+4*j, v);\
947 }
948 if (big_endian) {
949 FILL(AV_WB32);
950 } else {
951 FILL(AV_WL32);
952 }
953 ptr += stride;
954 }
955 #undef FILL
956 }
957
958
959 #define MAX_SLICE_PLANES 4
960
961 /// Slice plane
962 typedef struct SwsPlane
963 {
964 int available_lines; ///< max number of lines that can be hold by this plane
965 int sliceY; ///< index of first line
966 int sliceH; ///< number of lines
967 uint8_t **line; ///< line buffer
968 uint8_t **tmp; ///< Tmp line buffer used by mmx code
969 } SwsPlane;
970
971 /**
972 * Struct which defines a slice of an image to be scaled or an output for
973 * a scaled slice.
974 * A slice can also be used as intermediate ring buffer for scaling steps.
975 */
976 typedef struct SwsSlice
977 {
978 int width; ///< Slice line width
979 int h_chr_sub_sample; ///< horizontal chroma subsampling factor
980 int v_chr_sub_sample; ///< vertical chroma subsampling factor
981 int is_ring; ///< flag to identify if this slice is a ring buffer
982 int should_free_lines; ///< flag to identify if there are dynamic allocated lines
983 enum AVPixelFormat fmt; ///< planes pixel format
984 SwsPlane plane[MAX_SLICE_PLANES]; ///< color planes
985 } SwsSlice;
986
987 /**
988 * Struct which holds all necessary data for processing a slice.
989 * A processing step can be a color conversion or horizontal/vertical scaling.
990 */
991 typedef struct SwsFilterDescriptor
992 {
993 SwsSlice *src; ///< Source slice
994 SwsSlice *dst; ///< Output slice
995
996 int alpha; ///< Flag for processing alpha channel
997 void *instance; ///< Filter instance data
998
999 /// Function for processing input slice sliceH lines starting from line sliceY
1000 int (*process)(SwsContext *c, struct SwsFilterDescriptor *desc, int sliceY, int sliceH);
1001 } SwsFilterDescriptor;
1002
1003 // warp input lines in the form (src + width*i + j) to slice format (line[i][j])
1004 // relative=true means first line src[x][0] otherwise first line is src[x][lum/crh Y]
1005 int ff_init_slice_from_src(SwsSlice * s, uint8_t *src[4], int stride[4], int srcW, int lumY, int lumH, int chrY, int chrH, int relative);
1006
1007 // Initialize scaler filter descriptor chain
1008 int ff_init_filters(SwsContext *c);
1009
1010 // Free all filter data
1011 int ff_free_filters(SwsContext *c);
1012
1013 /*
1014 function for applying ring buffer logic into slice s
1015 It checks if the slice can hold more @lum lines, if yes
1016 do nothing otherwise remove @lum least used lines.
1017 It applies the same procedure for @chr lines.
1018 */
1019 int ff_rotate_slice(SwsSlice *s, int lum, int chr);
1020
1021 /// initializes gamma conversion descriptor
1022 int ff_init_gamma_convert(SwsFilterDescriptor *desc, SwsSlice * src, uint16_t *table);
1023
1024 /// initializes lum pixel format conversion descriptor
1025 int ff_init_desc_fmt_convert(SwsFilterDescriptor *desc, SwsSlice * src, SwsSlice *dst, uint32_t *pal);
1026
1027 /// initializes lum horizontal scaling descriptor
1028 int ff_init_desc_hscale(SwsFilterDescriptor *desc, SwsSlice *src, SwsSlice *dst, uint16_t *filter, int * filter_pos, int filter_size, int xInc);
1029
1030 /// initializes chr pixel format conversion descriptor
1031 int ff_init_desc_cfmt_convert(SwsFilterDescriptor *desc, SwsSlice * src, SwsSlice *dst, uint32_t *pal);
1032
1033 /// initializes chr horizontal scaling descriptor
1034 int ff_init_desc_chscale(SwsFilterDescriptor *desc, SwsSlice *src, SwsSlice *dst, uint16_t *filter, int * filter_pos, int filter_size, int xInc);
1035
1036 int ff_init_desc_no_chr(SwsFilterDescriptor *desc, SwsSlice * src, SwsSlice *dst);
1037
1038 /// initializes vertical scaling descriptors
1039 int ff_init_vscale(SwsContext *c, SwsFilterDescriptor *desc, SwsSlice *src, SwsSlice *dst);
1040
1041 /// setup vertical scaler functions
1042 void ff_init_vscale_pfn(SwsContext *c, yuv2planar1_fn yuv2plane1, yuv2planarX_fn yuv2planeX,
1043 yuv2interleavedX_fn yuv2nv12cX, yuv2packed1_fn yuv2packed1, yuv2packed2_fn yuv2packed2,
1044 yuv2packedX_fn yuv2packedX, yuv2anyX_fn yuv2anyX, int use_mmx);
1045
1046 //number of extra lines to process
1047 #define MAX_LINES_AHEAD 4
1048
1049 #endif /* SWSCALE_SWSCALE_INTERNAL_H */
1050