1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
|
/*
* RV60 dsp routines
*
* This file is part of FFmpeg.
*
* FFmpeg is free software; you can redistribute it and/or
* modify it under the terms of the GNU Lesser General Public
* License as published by the Free Software Foundation; either
* version 2.1 of the License, or (at your option) any later version.
*
* FFmpeg is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Lesser General Public License for more details.
*
* You should have received a copy of the GNU Lesser General Public
* License along with FFmpeg; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
*/
#include "rv60dsp.h"
#include "libavutil/common.h"
void ff_rv60_idct4x4_add(const int16_t * block, uint8_t * dst, int dst_stride)
{
int tmp[16];
#define IDCT4X4(src, src_stride, src_step, dst, dst_stride, dst_step) \
for (int y = 0; y < 4; y++) { \
int a = src[y*src_stride + 0*src_step]; \
int b = src[y*src_stride + 1*src_step]; \
int c = src[y*src_stride + 2*src_step]; \
int d = src[y*src_stride + 3*src_step]; \
int t0 = 13 * (a + c); \
int t1 = 13 * (a - c); \
int t2 = 7 * b - 17 * d; \
int t3 = 7 * d + 17 * b; \
STORE(dst[y*dst_stride + 0*dst_step], (t0 + t3 + 16) >> 5); \
STORE(dst[y*dst_stride + 1*dst_step], (t1 + t2 + 16) >> 5); \
STORE(dst[y*dst_stride + 2*dst_step], (t1 - t2 + 16) >> 5); \
STORE(dst[y*dst_stride + 3*dst_step], (t0 - t3 + 16) >> 5); \
}
#define STORE(a, b) a = b
IDCT4X4(block, 1, 4, tmp, 1, 4)
#undef STORE
#define STORE(a, b) a = av_clip_uint8(a + (b))
IDCT4X4(tmp, 4, 1, dst, dst_stride, 1)
#undef STORE
}
void ff_rv60_idct8x8_add(const int16_t * block, uint8_t * dst, int dst_stride)
{
int tmp[64];
#define IDCT8X8(src, src_stride, src_step, dst, dst_stride, dst_step) \
for (int y = 0; y < 8; y++) { \
int a = src[y*src_stride + 0*src_step]; \
int b = src[y*src_stride + 1*src_step]; \
int c = src[y*src_stride + 2*src_step]; \
int d = src[y*src_stride + 3*src_step]; \
int e = src[y*src_stride + 4*src_step]; \
int f = src[y*src_stride + 5*src_step]; \
int g = src[y*src_stride + 6*src_step]; \
int h = src[y*src_stride + 7*src_step]; \
int t0 = 37 * (a + e); \
int t1 = 37 * (a - e); \
int t2 = 48 * c + 20 * g; \
int t3 = 20 * c - 48 * g; \
int t4 = t0 + t2; \
int t5 = t0 - t2; \
int t6 = t1 + t3; \
int t7 = t1 - t3; \
int t8 = 51 * b + 43 * d + 29 * f + 10 * h; \
int t9 = 43 * b - 10 * d - 51 * f - 29 * h; \
int ta = 29 * b - 51 * d + 10 * f + 43 * h; \
int tb = 10 * b - 29 * d + 43 * f - 51 * h; \
STORE(dst[y*dst_stride + 0*dst_step], (t4 + t8 + 64) >> 7); \
STORE(dst[y*dst_stride + 1*dst_step], (t6 + t9 + 64) >> 7); \
STORE(dst[y*dst_stride + 2*dst_step], (t7 + ta + 64) >> 7); \
STORE(dst[y*dst_stride + 3*dst_step], (t5 + tb + 64) >> 7); \
STORE(dst[y*dst_stride + 4*dst_step], (t5 - tb + 64) >> 7); \
STORE(dst[y*dst_stride + 5*dst_step], (t7 - ta + 64) >> 7); \
STORE(dst[y*dst_stride + 6*dst_step], (t6 - t9 + 64) >> 7); \
STORE(dst[y*dst_stride + 7*dst_step], (t4 - t8 + 64) >> 7); \
}
#define STORE(a, b) a = b
IDCT8X8(block, 1, 8, tmp, 1, 8)
#undef STORE
#define STORE(a, b) a = av_clip_uint8(a + (b))
IDCT8X8(tmp, 8, 1, dst, dst_stride, 1)
#undef STORE
}
void ff_rv60_idct16x16_add(const int16_t * block, uint8_t * dst, int dst_stride)
{
int16_t tmp[256];
#define IDCT16X16(src, src_stride, src_step, dst, dst_stride, dst_step) \
for (int y = 0; y < 16; y++) { \
int a = src[y*src_stride + 0*src_step]; \
int b = src[y*src_stride + 1*src_step]; \
int c = src[y*src_stride + 2*src_step]; \
int d = src[y*src_stride + 3*src_step]; \
int e = src[y*src_stride + 4*src_step]; \
int f = src[y*src_stride + 5*src_step]; \
int g = src[y*src_stride + 6*src_step]; \
int h = src[y*src_stride + 7*src_step]; \
int i = src[y*src_stride + 8*src_step]; \
int j = src[y*src_stride + 9*src_step]; \
int k = src[y*src_stride + 10*src_step]; \
int l = src[y*src_stride + 11*src_step]; \
int m = src[y*src_stride + 12*src_step]; \
int n = src[y*src_stride + 13*src_step]; \
int o = src[y*src_stride + 14*src_step]; \
int p = src[y*src_stride + 15*src_step]; \
int t0 = 26 * (a + i); \
int t1 = 26 * (a - i); \
int t2 = 14 * e - 34 * m; \
int t3 = 34 * e + 14 * m; \
int t4 = t0 + t3; \
int t5 = t0 - t3; \
int t6 = t1 + t2; \
int t7 = t1 - t2; \
int tmp00 = 31 * c - 7 * g - 36 * k - 20 * o; \
int tmp01 = 36 * c + 31 * g + 20 * k + 7 * o; \
int tmp02 = 20 * c - 36 * g + 7 * k + 31 * o; \
int tmp03 = 7 * c - 20 * g + 31 * k - 36 * o; \
int tm0 = t4 + tmp01; \
int tm1 = t4 - tmp01; \
int tm2 = t5 + tmp03; \
int tm3 = t5 - tmp03; \
int tm4 = t6 + tmp00; \
int tm5 = t6 - tmp00; \
int tm6 = t7 + tmp02; \
int tm7 = t7 - tmp02; \
int tt0 = 37 * b + 35 * d + 32 * f + 28 * h + 23 * j + 17 * l + 11 * n + 4 * p; \
int tt1 = 35 * b + 23 * d + 4 * f - 17 * h - 32 * j - 37 * l - 28 * n - 11 * p; \
int tt2 = 32 * b + 4 * d - 28 * f - 35 * h - 11 * j + 23 * l + 37 * n + 17 * p; \
int tt3 = 28 * b - 17 * d - 35 * f + 4 * h + 37 * j + 11 * l - 32 * n - 23 * p; \
int tt4 = 23 * b - 32 * d - 11 * f + 37 * h - 4 * j - 35 * l + 17 * n + 28 * p; \
int tt5 = 17 * b - 37 * d + 23 * f + 11 * h - 35 * j + 28 * l + 4 * n - 32 * p; \
int tt6 = 11 * b - 28 * d + 37 * f - 32 * h + 17 * j + 4 * l - 23 * n + 35 * p; \
int tt7 = 4 * b - 11 * d + 17 * f - 23 * h + 28 * j - 32 * l + 35 * n - 37 * p; \
STORE(dst[y*dst_stride+ 0*dst_step], (tm0 + tt0 + 64) >> 7); \
STORE(dst[y*dst_stride+ 1*dst_step], (tm4 + tt1 + 64) >> 7); \
STORE(dst[y*dst_stride+ 2*dst_step], (tm6 + tt2 + 64) >> 7); \
STORE(dst[y*dst_stride+ 3*dst_step], (tm2 + tt3 + 64) >> 7); \
STORE(dst[y*dst_stride+ 4*dst_step], (tm3 + tt4 + 64) >> 7); \
STORE(dst[y*dst_stride+ 5*dst_step], (tm7 + tt5 + 64) >> 7); \
STORE(dst[y*dst_stride+ 6*dst_step], (tm5 + tt6 + 64) >> 7); \
STORE(dst[y*dst_stride+ 7*dst_step], (tm1 + tt7 + 64) >> 7); \
STORE(dst[y*dst_stride+ 8*dst_step], (tm1 - tt7 + 64) >> 7); \
STORE(dst[y*dst_stride+ 9*dst_step], (tm5 - tt6 + 64) >> 7); \
STORE(dst[y*dst_stride+ 10*dst_step], (tm7 - tt5 + 64) >> 7); \
STORE(dst[y*dst_stride+ 11*dst_step], (tm3 - tt4 + 64) >> 7); \
STORE(dst[y*dst_stride+ 12*dst_step], (tm2 - tt3 + 64) >> 7); \
STORE(dst[y*dst_stride+ 13*dst_step], (tm6 - tt2 + 64) >> 7); \
STORE(dst[y*dst_stride+ 14*dst_step], (tm4 - tt1 + 64) >> 7); \
STORE(dst[y*dst_stride+ 15*dst_step], (tm0 - tt0 + 64) >> 7); \
}
#define STORE(a, x) a = av_clip_intp2(x, 15)
IDCT16X16(block, 1, 16, tmp, 1, 16)
#undef STORE
#define STORE(a, x) a = av_clip_uint8(a + (x))
IDCT16X16(tmp, 16, 1, dst, dst_stride, 1)
#undef STORE
}
|
