1 /*
2  * Copyright (C) 2001-2003 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 FFMPEG_SWSCALE_INTERNAL_H
22 #define FFMPEG_SWSCALE_INTERNAL_H
23 
24 #include "config.h"
25 
26 #ifdef HAVE_ALTIVEC_H
27 #include <altivec.h>
28 #endif
29 
30 #include "avutil.h"
31 
32 #define MAX_FILTER_SIZE 256
33 
34 #define VOFW 2048
35 #define VOF  (VOFW*2)
36 
37 typedef int (*SwsFunc)(struct SwsContext *context, uint8_t* src[], int srcStride[], int srcSliceY,
38              int srcSliceH, uint8_t* dst[], int dstStride[]);
39 
40 /* this struct should be aligned on at least 32-byte boundary */
41 typedef struct SwsContext{
42     /**
43      * info on struct for av_log
44      */
45     const AVClass *av_class;
46 
47     /**
48      * Note that src, dst, srcStride, dstStride will be copied in the
49      * sws_scale() wrapper so they can be freely modified here.
50      */
51     SwsFunc swScale;
52     int srcW, srcH, dstH;
53     int chrSrcW, chrSrcH, chrDstW, chrDstH;
54     int lumXInc, chrXInc;
55     int lumYInc, chrYInc;
56     int dstFormat, srcFormat;               ///< format 4:2:0 type is always YV12
57     int origDstFormat, origSrcFormat;       ///< format
58     int chrSrcHSubSample, chrSrcVSubSample;
59     int chrIntHSubSample, chrIntVSubSample;
60     int chrDstHSubSample, chrDstVSubSample;
61     int vChrDrop;
62     int sliceDir;
63     double param[2];
64 
65     int16_t **lumPixBuf;
66     int16_t **chrPixBuf;
67     int16_t *hLumFilter;
68     int16_t *hLumFilterPos;
69     int16_t *hChrFilter;
70     int16_t *hChrFilterPos;
71     int16_t *vLumFilter;
72     int16_t *vLumFilterPos;
73     int16_t *vChrFilter;
74     int16_t *vChrFilterPos;
75 
76     uint8_t formatConvBuffer[VOF]; //FIXME dynamic alloc, but we have to change a lot of code for this to be useful
77 
78     int hLumFilterSize;
79     int hChrFilterSize;
80     int vLumFilterSize;
81     int vChrFilterSize;
82     int vLumBufSize;
83     int vChrBufSize;
84 
85     uint8_t *funnyYCode;
86     uint8_t *funnyUVCode;
87     int32_t *lumMmx2FilterPos;
88     int32_t *chrMmx2FilterPos;
89     int16_t *lumMmx2Filter;
90     int16_t *chrMmx2Filter;
91 
92     int canMMX2BeUsed;
93 
94     int lastInLumBuf;
95     int lastInChrBuf;
96     int lumBufIndex;
97     int chrBufIndex;
98     int dstY;
99     int flags;
100     void * yuvTable;            // pointer to the yuv->rgb table start so it can be freed()
101     uint8_t * table_rV[256];
102     uint8_t * table_gU[256];
103     int    table_gV[256];
104     uint8_t * table_bU[256];
105 
106     //Colorspace stuff
107     int contrast, brightness, saturation;    // for sws_getColorspaceDetails
108     int srcColorspaceTable[4];
109     int dstColorspaceTable[4];
110     int srcRange, dstRange;
111 
112 #define RED_DITHER            "0*8"
113 #define GREEN_DITHER          "1*8"
114 #define BLUE_DITHER           "2*8"
115 #define Y_COEFF               "3*8"
116 #define VR_COEFF              "4*8"
117 #define UB_COEFF              "5*8"
118 #define VG_COEFF              "6*8"
119 #define UG_COEFF              "7*8"
120 #define Y_OFFSET              "8*8"
121 #define U_OFFSET              "9*8"
122 #define V_OFFSET              "10*8"
123 #define LUM_MMX_FILTER_OFFSET "11*8"
124 #define CHR_MMX_FILTER_OFFSET "11*8+4*4*256"
125 #define DSTW_OFFSET           "11*8+4*4*256*2" //do not change, it is hardcoded in the asm
126 #define ESP_OFFSET            "11*8+4*4*256*2+8"
127 #define VROUNDER_OFFSET       "11*8+4*4*256*2+16"
128 #define U_TEMP                "11*8+4*4*256*2+24"
129 #define V_TEMP                "11*8+4*4*256*2+32"
130 
131     uint64_t redDither   __attribute__((aligned(8)));
132     uint64_t greenDither __attribute__((aligned(8)));
133     uint64_t blueDither  __attribute__((aligned(8)));
134 
135     uint64_t yCoeff      __attribute__((aligned(8)));
136     uint64_t vrCoeff     __attribute__((aligned(8)));
137     uint64_t ubCoeff     __attribute__((aligned(8)));
138     uint64_t vgCoeff     __attribute__((aligned(8)));
139     uint64_t ugCoeff     __attribute__((aligned(8)));
140     uint64_t yOffset     __attribute__((aligned(8)));
141     uint64_t uOffset     __attribute__((aligned(8)));
142     uint64_t vOffset     __attribute__((aligned(8)));
143     int32_t  lumMmxFilter[4*MAX_FILTER_SIZE];
144     int32_t  chrMmxFilter[4*MAX_FILTER_SIZE];
145     int dstW;
146     uint64_t esp          __attribute__((aligned(8)));
147     uint64_t vRounder     __attribute__((aligned(8)));
148     uint64_t u_temp       __attribute__((aligned(8)));
149     uint64_t v_temp       __attribute__((aligned(8)));
150 
151 #ifdef HAVE_ALTIVEC
152 
153   vector signed short   CY;
154   vector signed short   CRV;
155   vector signed short   CBU;
156   vector signed short   CGU;
157   vector signed short   CGV;
158   vector signed short   OY;
159   vector unsigned short CSHIFT;
160   vector signed short   *vYCoeffsBank, *vCCoeffsBank;
161 
162 #endif
163 
164 
165 #ifdef ARCH_BFIN
166     uint32_t oy           __attribute__((aligned(4)));
167     uint32_t oc           __attribute__((aligned(4)));
168     uint32_t zero         __attribute__((aligned(4)));
169     uint32_t cy           __attribute__((aligned(4)));
170     uint32_t crv          __attribute__((aligned(4)));
171     uint32_t rmask        __attribute__((aligned(4)));
172     uint32_t cbu          __attribute__((aligned(4)));
173     uint32_t bmask        __attribute__((aligned(4)));
174     uint32_t cgu          __attribute__((aligned(4)));
175     uint32_t cgv          __attribute__((aligned(4)));
176     uint32_t gmask        __attribute__((aligned(4)));
177 #endif
178 
179 #ifdef HAVE_VIS
180     uint64_t sparc_coeffs[10] __attribute__((aligned(8)));
181 #endif
182 
183 } SwsContext;
184 //FIXME check init (where 0)
185 
186 SwsFunc yuv2rgb_get_func_ptr (SwsContext *c);
187 int yuv2rgb_c_init_tables (SwsContext *c, const int inv_table[4], int fullRange, int brightness, int contrast, int saturation);
188 
189 void yuv2rgb_altivec_init_tables (SwsContext *c, const int inv_table[4],int brightness,int contrast, int saturation);
190 SwsFunc yuv2rgb_init_altivec (SwsContext *c);
191 void altivec_yuv2packedX (SwsContext *c,
192                           int16_t *lumFilter, int16_t **lumSrc, int lumFilterSize,
193                           int16_t *chrFilter, int16_t **chrSrc, int chrFilterSize,
194                           uint8_t *dest, int dstW, int dstY);
195 
196 const char *sws_format_name(int format);
197 
198 //FIXME replace this with something faster
199 #define isPlanarYUV(x)  (           \
200            (x)==PIX_FMT_YUV410P     \
201         || (x)==PIX_FMT_YUV420P     \
202         || (x)==PIX_FMT_YUV411P     \
203         || (x)==PIX_FMT_YUV422P     \
204         || (x)==PIX_FMT_YUV444P     \
205         || (x)==PIX_FMT_YUV440P     \
206         || (x)==PIX_FMT_NV12        \
207         || (x)==PIX_FMT_NV21        \
208     )
209 #define isYUV(x)        (           \
210            (x)==PIX_FMT_UYVY422     \
211         || (x)==PIX_FMT_YUYV422     \
212         || isPlanarYUV(x)           \
213     )
214 #define isGray(x)       (           \
215            (x)==PIX_FMT_GRAY8       \
216         || (x)==PIX_FMT_GRAY16BE    \
217         || (x)==PIX_FMT_GRAY16LE    \
218     )
219 #define isGray16(x)     (           \
220            (x)==PIX_FMT_GRAY16BE    \
221         || (x)==PIX_FMT_GRAY16LE    \
222     )
223 #define isRGB(x)        (           \
224            (x)==PIX_FMT_BGR32       \
225         || (x)==PIX_FMT_RGB24       \
226         || (x)==PIX_FMT_RGB565      \
227         || (x)==PIX_FMT_RGB555      \
228         || (x)==PIX_FMT_RGB8        \
229         || (x)==PIX_FMT_RGB4        \
230         || (x)==PIX_FMT_RGB4_BYTE   \
231         || (x)==PIX_FMT_MONOBLACK   \
232     )
233 #define isBGR(x)        (           \
234            (x)==PIX_FMT_RGB32       \
235         || (x)==PIX_FMT_BGR24       \
236         || (x)==PIX_FMT_BGR565      \
237         || (x)==PIX_FMT_BGR555      \
238         || (x)==PIX_FMT_BGR8        \
239         || (x)==PIX_FMT_BGR4        \
240         || (x)==PIX_FMT_BGR4_BYTE   \
241         || (x)==PIX_FMT_MONOBLACK   \
242     )
243 
fmt_depth(int fmt)244 static inline int fmt_depth(int fmt)
245 {
246     switch(fmt) {
247         case PIX_FMT_BGRA:
248         case PIX_FMT_ABGR:
249         case PIX_FMT_RGBA:
250         case PIX_FMT_ARGB:
251             return 32;
252         case PIX_FMT_BGR24:
253         case PIX_FMT_RGB24:
254             return 24;
255         case PIX_FMT_BGR565:
256         case PIX_FMT_RGB565:
257         case PIX_FMT_GRAY16BE:
258         case PIX_FMT_GRAY16LE:
259             return 16;
260         case PIX_FMT_BGR555:
261         case PIX_FMT_RGB555:
262             return 15;
263         case PIX_FMT_BGR8:
264         case PIX_FMT_RGB8:
265             return 8;
266         case PIX_FMT_BGR4:
267         case PIX_FMT_RGB4:
268         case PIX_FMT_BGR4_BYTE:
269         case PIX_FMT_RGB4_BYTE:
270             return 4;
271         case PIX_FMT_MONOBLACK:
272             return 1;
273         default:
274             return 0;
275     }
276 }
277 
278 extern const DECLARE_ALIGNED(8, uint64_t, ff_dither4[2]);
279 extern const DECLARE_ALIGNED(8, uint64_t, ff_dither8[2]);
280 
281 #endif /* FFMPEG_SWSCALE_INTERNAL_H */
282