2 * Copyright (C) 2004-2010 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
27 typedef short IDWTELEM;
29 #define MAX_DWT_SUPPORT 8
30 #define MAX_DECOMPOSITIONS 8
33 IDWTELEM *b[MAX_DWT_SUPPORT];
42 /** Used to minimize the amount of memory used in order to optimize cache performance. **/
43 typedef struct slice_buffer_s {
44 IDWTELEM * * line; ///< For use by idwt and predict_slices.
45 IDWTELEM * * data_stack; ///< Used for internal purposes.
50 IDWTELEM * base_buffer; ///< Buffer that this structure is caching.
55 // Possible prototypes for vertical_compose functions
56 typedef void (*vertical_compose_2tap)(IDWTELEM *b0, IDWTELEM *b1, int width);
57 typedef void (*vertical_compose_3tap)(IDWTELEM *b0, IDWTELEM *b1, IDWTELEM *b2, int width);
58 typedef void (*vertical_compose_5tap)(IDWTELEM *b0, IDWTELEM *b1, IDWTELEM *b2, IDWTELEM *b3, IDWTELEM *b4, int width);
59 typedef void (*vertical_compose_9tap)(IDWTELEM *dst, IDWTELEM *b[8], int width);
61 typedef struct DWTContext {
67 int decomposition_count;
70 void (*spatial_compose)(struct DWTContext *cs, int level, int width, int height, int stride);
71 void (*vertical_compose_l0)(void);
72 void (*vertical_compose_h0)(void);
73 void (*vertical_compose_l1)(void);
74 void (*vertical_compose_h1)(void);
75 void (*vertical_compose)(void); ///< one set of lowpass and highpass combined
76 void (*horizontal_compose)(IDWTELEM *b, IDWTELEM *tmp, int width);
78 void (*vertical_compose97i)(IDWTELEM *b0, IDWTELEM *b1, IDWTELEM *b2, IDWTELEM *b3, IDWTELEM *b4, IDWTELEM *b5, int width);
79 void (*horizontal_compose97i)(IDWTELEM *b, int width);
80 void (*inner_add_yblock)(const uint8_t *obmc, const int obmc_stride, uint8_t * * block, int b_w, int b_h, int src_x, int src_y, int src_stride, slice_buffer * sb, int add, uint8_t * dst8);
82 DWTCompose cs[MAX_DECOMPOSITIONS];
98 // -1 if an error occurred, e.g. the dwt_type isn't recognized
99 int ff_spatial_idwt_init2(DWTContext *d, IDWTELEM *buffer, int width, int height,
100 int stride, enum dwt_type type, int decomposition_count,
103 int ff_spatial_idwt2(IDWTELEM *buffer, int width, int height, int stride,
104 enum dwt_type type, int decomposition_count, IDWTELEM *temp);
106 void ff_spatial_idwt_slice2(DWTContext *d, int y);
108 // shared stuff for simd optimiztions
109 #define COMPOSE_53iL0(b0, b1, b2)\
110 (b1 - ((b0 + b2 + 2) >> 2))
112 #define COMPOSE_DIRAC53iH0(b0, b1, b2)\
113 (b1 + ((b0 + b2 + 1) >> 1))
115 #define COMPOSE_DD97iH0(b0, b1, b2, b3, b4)\
116 (b2 + ((-b0 + 9*b1 + 9*b3 - b4 + 8) >> 4))
118 #define COMPOSE_DD137iL0(b0, b1, b2, b3, b4)\
119 (b2 - ((-b0 + 9*b1 + 9*b3 - b4 + 16) >> 5))
121 #define COMPOSE_HAARiL0(b0, b1)\
122 (b0 - ((b1 + 1) >> 1))
124 #define COMPOSE_HAARiH0(b0, b1)\
127 #define COMPOSE_FIDELITYiL0(b0, b1, b2, b3, b4, b5, b6, b7, b8)\
128 (b4 - ((-8*(b0+b8) + 21*(b1+b7) - 46*(b2+b6) + 161*(b3+b5) + 128) >> 8))
130 #define COMPOSE_FIDELITYiH0(b0, b1, b2, b3, b4, b5, b6, b7, b8)\
131 (b4 + ((-2*(b0+b8) + 10*(b1+b7) - 25*(b2+b6) + 81*(b3+b5) + 128) >> 8))
133 #define COMPOSE_DAUB97iL1(b0, b1, b2)\
134 (b1 - ((1817*(b0 + b2) + 2048) >> 12))
136 #define COMPOSE_DAUB97iH1(b0, b1, b2)\
137 (b1 - (( 113*(b0 + b2) + 64) >> 7))
139 #define COMPOSE_DAUB97iL0(b0, b1, b2)\
140 (b1 + (( 217*(b0 + b2) + 2048) >> 12))
142 #define COMPOSE_DAUB97iH0(b0, b1, b2)\
143 (b1 + ((6497*(b0 + b2) + 2048) >> 12))
220 #define slice_buffer_get_line(slice_buf, line_num) ((slice_buf)->line[line_num] ? (slice_buf)->line[line_num] : ff_slice_buffer_load_line((slice_buf), (line_num)))
221 //#define slice_buffer_get_line(slice_buf, line_num) (ff_slice_buffer_load_line((slice_buf), (line_num)))
223 void ff_slice_buffer_init(slice_buffer * buf, int line_count, int max_allocated_lines, int line_width, IDWTELEM * base_buffer);
224 void ff_slice_buffer_release(slice_buffer * buf, int line);
225 void ff_slice_buffer_flush(slice_buffer * buf);
226 void ff_slice_buffer_destroy(slice_buffer * buf);
227 IDWTELEM * ff_slice_buffer_load_line(slice_buffer * buf, int line);
229 void ff_snow_vertical_compose97i(IDWTELEM *b0, IDWTELEM *b1, IDWTELEM *b2, IDWTELEM *b3, IDWTELEM *b4, IDWTELEM *b5, int width);
230 void ff_snow_horizontal_compose97i(IDWTELEM *b, int width);
231 void ff_snow_inner_add_yblock(const uint8_t *obmc, const int obmc_stride, uint8_t * * block, int b_w, int b_h, int src_x, int src_y, int src_stride, slice_buffer * sb, int add, uint8_t * dst8);
233 int ff_w53_32_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h);
234 int ff_w97_32_c(void *v, uint8_t * pix1, uint8_t * pix2, int line_size, int h);
236 void ff_spatial_dwt(int *buffer, int width, int height, int stride, int type, int decomposition_count);
238 void ff_spatial_idwt_buffered_init(DWTCompose *cs, slice_buffer * sb, int width, int height, int stride_line, int type, int decomposition_count);
239 void ff_spatial_idwt_buffered_slice(DWTContext *dsp, DWTCompose *cs, slice_buffer * slice_buf, int width, int height, int stride_line, int type, int decomposition_count, int y);
240 void ff_spatial_idwt(IDWTELEM *buffer, int width, int height, int stride, int type, int decomposition_count);
242 void ff_dwt_init(DWTContext *c);
243 void ff_dwt_init_x86(DWTContext *c);
245 #endif /* AVCODEC_DWT_H */