1 /*
2 * Template for the Discrete Cosine Transform for 32 samples
3 * Copyright (c) 2001, 2002 Fabrice Bellard
4 *
5 * This file is part of FFmpeg.
6 *
7 * FFmpeg is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
11 *
12 * FFmpeg is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with FFmpeg; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20 */
21
25
26 #ifdef CHECKED
27 #define SUINT int
28 #define SUINT32 int32_t
29 #else
30 #define SUINT unsigned
31 #define SUINT32 uint32_t
32 #endif
33
34 #if DCT32_FLOAT
35 # define dct32 ff_dct32_float
36 # define FIXHR(x) ((float)(x))
37 # define MULH3(x, y, s) ((s)*(y)*(x))
38 # define INTFLOAT float
39 # define SUINTFLOAT float
40 #else
41 # define dct32 ff_dct32_fixed
42 # define FIXHR(a) ((int)((a) * (1LL<<32) + 0.5))
43 # define MULH3(x, y, s) MULH((s)*(x), y)
45 # define SUINTFLOAT SUINT
46 #endif
47
48
49 /* tab[i][j] = 1.0 / (2.0 * cos(pi*(2*k+1) / 2^(6 - j))) */
50
51 /* cos(i*pi/64) */
52
53 #define COS0_0 FIXHR(0.50060299823519630134/2)
54 #define COS0_1 FIXHR(0.50547095989754365998/2)
55 #define COS0_2 FIXHR(0.51544730992262454697/2)
56 #define COS0_3 FIXHR(0.53104259108978417447/2)
57 #define COS0_4 FIXHR(0.55310389603444452782/2)
58 #define COS0_5 FIXHR(0.58293496820613387367/2)
59 #define COS0_6 FIXHR(0.62250412303566481615/2)
60 #define COS0_7 FIXHR(0.67480834145500574602/2)
61 #define COS0_8 FIXHR(0.74453627100229844977/2)
62 #define COS0_9 FIXHR(0.83934964541552703873/2)
63 #define COS0_10 FIXHR(0.97256823786196069369/2)
64 #define COS0_11 FIXHR(1.16943993343288495515/4)
65 #define COS0_12 FIXHR(1.48416461631416627724/4)
66 #define COS0_13 FIXHR(2.05778100995341155085/8)
67 #define COS0_14 FIXHR(3.40760841846871878570/8)
68 #define COS0_15 FIXHR(10.19000812354805681150/32)
69
70 #define COS1_0 FIXHR(0.50241928618815570551/2)
71 #define COS1_1 FIXHR(0.52249861493968888062/2)
72 #define COS1_2 FIXHR(0.56694403481635770368/2)
73 #define COS1_3 FIXHR(0.64682178335999012954/2)
74 #define COS1_4 FIXHR(0.78815462345125022473/2)
75 #define COS1_5 FIXHR(1.06067768599034747134/4)
76 #define COS1_6 FIXHR(1.72244709823833392782/4)
77 #define COS1_7 FIXHR(5.10114861868916385802/16)
78
79 #define COS2_0 FIXHR(0.50979557910415916894/2)
80 #define COS2_1 FIXHR(0.60134488693504528054/2)
81 #define COS2_2 FIXHR(0.89997622313641570463/2)
82 #define COS2_3 FIXHR(2.56291544774150617881/8)
83
84 #define COS3_0 FIXHR(0.54119610014619698439/2)
85 #define COS3_1 FIXHR(1.30656296487637652785/4)
86
87 #define COS4_0 FIXHR(M_SQRT1_2/2)
88
89 /* butterfly operator */
90 #define BF(a, b, c, s)\
91 {\
92 tmp0 = val##a + val##b;\
93 tmp1 = val##a - val##b;\
94 val##a = tmp0;\
95 val##b = MULH3(tmp1, c, 1<<(s));\
96 }
97
98 #define BF0(a, b, c, s)\
99 {\
100 tmp0 = tab[a] + tab[b];\
101 tmp1 = tab[a] - tab[b];\
102 val##a = tmp0;\
103 val##b = MULH3(tmp1, c, 1<<(s));\
104 }
105
106 #define BF1(a, b, c, d)\
107 {\
108 BF(a, b, COS4_0, 1);\
109 BF(c, d,-COS4_0, 1);\
110 val##c += val##d;\
111 }
112
113 #define BF2(a, b, c, d)\
114 {\
115 BF(a, b, COS4_0, 1);\
116 BF(c, d,-COS4_0, 1);\
117 val##c += val##d;\
118 val##a += val##c;\
119 val##c += val##b;\
120 val##b += val##d;\
121 }
122
123 #define ADD(a, b) val##a += val##b
124
125 /* DCT32 without 1/sqrt(2) coef zero scaling. */
127 {
130
131 SUINTFLOAT val0 , val1 , val2 , val3 , val4 , val5 , val6 , val7 ,
132 val8 , val9 , val10, val11, val12, val13, val14, val15,
133 val16, val17, val18, val19, val20, val21, val22, val23,
134 val24, val25, val26, val27, val28, val29, val30, val31;
135
136 /* pass 1 */
139 /* pass 2 */
142 /* pass 1 */
145 /* pass 2 */
148 /* pass 3 */
153 /* pass 1 */
156 /* pass 2 */
159 /* pass 1 */
162 /* pass 2 */
165 /* pass 3 */
170 /* pass 4 */
179
180
181
182 /* pass 1 */
185 /* pass 2 */
188 /* pass 1 */
191 /* pass 2 */
194 /* pass 3 */
199
200 /* pass 1 */
203 /* pass 2 */
206 /* pass 1 */
209 /* pass 2 */
212 /* pass 3 */
217 /* pass 4 */
226
227 /* pass 5 */
236
237 /* pass 6 */
238
246
263
271
272 out[ 1] = val16 + val24;
273 out[17] = val17 + val25;
274 out[ 9] = val18 + val26;
275 out[25] = val19 + val27;
276 out[ 5] = val20 + val28;
277 out[21] = val21 + val29;
278 out[13] = val22 + val30;
279 out[29] = val23 + val31;
280 out[ 3] = val24 + val20;
281 out[19] = val25 + val21;
282 out[11] = val26 + val22;
283 out[27] = val27 + val23;
284 out[ 7] = val28 + val18;
285 out[23] = val29 + val19;
286 out[15] = val30 + val17;
288 }