1 /*****************************************************************************
2 * dct.h: h264 encoder library
3 *****************************************************************************
4 * Copyright (C) 2004-2008 Loren Merritt <lorenm@u.washington.edu>
6 * This program is free software; you can redistribute it and/or modify
7 * it under the terms of the GNU General Public License as published by
8 * the Free Software Foundation; either version 2 of the License, or
9 * (at your option) any later version.
11 * This program 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
14 * GNU General Public License for more details.
16 * You should have received a copy of the GNU General Public License
17 * along with this program; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02111, USA.
19 *****************************************************************************/
24 /* the inverse of the scaling factors introduced by 8x8 fdct */
25 #define W(i) (i==0 ? FIX8(1.0000) :\
26 i==1 ? FIX8(0.8859) :\
27 i==2 ? FIX8(1.6000) :\
28 i==3 ? FIX8(0.9415) :\
29 i==4 ? FIX8(1.2651) :\
30 i==5 ? FIX8(1.1910) :0)
31 static const uint16_t x264_dct8_weight_tab[64] = {
32 W(0), W(3), W(4), W(3), W(0), W(3), W(4), W(3),
33 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1),
34 W(4), W(5), W(2), W(5), W(4), W(5), W(2), W(5),
35 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1),
37 W(0), W(3), W(4), W(3), W(0), W(3), W(4), W(3),
38 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1),
39 W(4), W(5), W(2), W(5), W(4), W(5), W(2), W(5),
40 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1)
45 #define W(i) (i==0 ? FIX8(3.125) :\
48 static const uint16_t x264_dct4_weight2_tab[16] = {
49 W(0), W(1), W(0), W(1),
50 W(1), W(2), W(1), W(2),
51 W(0), W(1), W(0), W(1),
52 W(1), W(2), W(1), W(2)
56 #define W(i) (i==0 ? FIX8(1.00000) :\
57 i==1 ? FIX8(0.78487) :\
58 i==2 ? FIX8(2.56132) :\
59 i==3 ? FIX8(0.88637) :\
60 i==4 ? FIX8(1.60040) :\
61 i==5 ? FIX8(1.41850) :0)
62 static const uint16_t x264_dct8_weight2_tab[64] = {
63 W(0), W(3), W(4), W(3), W(0), W(3), W(4), W(3),
64 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1),
65 W(4), W(5), W(2), W(5), W(4), W(5), W(2), W(5),
66 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1),
68 W(0), W(3), W(4), W(3), W(0), W(3), W(4), W(3),
69 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1),
70 W(4), W(5), W(2), W(5), W(4), W(5), W(2), W(5),
71 W(3), W(1), W(5), W(1), W(3), W(1), W(5), W(1)
75 extern int x264_dct4_weight2_zigzag[2][16]; // [2] = {frame, field}
76 extern int x264_dct8_weight2_zigzag[2][64];
80 // pix1 stride = FENC_STRIDE
81 // pix2 stride = FDEC_STRIDE
82 // p_dst stride = FDEC_STRIDE
83 void (*sub4x4_dct) ( int16_t dct[4][4], uint8_t *pix1, uint8_t *pix2 );
84 void (*add4x4_idct) ( uint8_t *p_dst, int16_t dct[4][4] );
86 void (*sub8x8_dct) ( int16_t dct[4][4][4], uint8_t *pix1, uint8_t *pix2 );
87 void (*add8x8_idct) ( uint8_t *p_dst, int16_t dct[4][4][4] );
89 void (*sub16x16_dct) ( int16_t dct[16][4][4], uint8_t *pix1, uint8_t *pix2 );
90 void (*add16x16_idct)( uint8_t *p_dst, int16_t dct[16][4][4] );
92 void (*sub8x8_dct8) ( int16_t dct[8][8], uint8_t *pix1, uint8_t *pix2 );
93 void (*add8x8_idct8) ( uint8_t *p_dst, int16_t dct[8][8] );
95 void (*sub16x16_dct8) ( int16_t dct[4][8][8], uint8_t *pix1, uint8_t *pix2 );
96 void (*add16x16_idct8)( uint8_t *p_dst, int16_t dct[4][8][8] );
98 void (*dct4x4dc) ( int16_t d[4][4] );
99 void (*idct4x4dc)( int16_t d[4][4] );
101 void (*dct2x2dc) ( int16_t d[2][2] );
102 void (*idct2x2dc)( int16_t d[2][2] );
104 } x264_dct_function_t;
108 void (*scan_8x8)( int16_t level[64], int16_t dct[8][8] );
109 void (*scan_4x4)( int16_t level[16], int16_t dct[4][4] );
110 void (*sub_4x4)( int16_t level[16], const uint8_t *p_src, uint8_t *p_dst );
112 } x264_zigzag_function_t;
114 void x264_dct_init( int cpu, x264_dct_function_t *dctf );
115 void x264_dct_init_weights( void );
116 void x264_zigzag_init( int cpu, x264_zigzag_function_t *pf, int b_interlaced );