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
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
|
/*
* Copyright (c) 2010 Alex Converse <alex.converse@gmail.com>
*
* 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
*
* Note: Rounding-to-nearest used unless otherwise stated
*
*/
#include <stdint.h>
#include "config.h"
#include "libavutil/attributes.h"
#include "aacpsdsp.h"
static void ps_add_squares_c(INTFLOAT *restrict dst,
const INTFLOAT (*src)[2], int n)
{
for (int i = 0; i < n; i++)
dst[i] += (UINTFLOAT)AAC_MADD28(src[i][0], src[i][0], src[i][1], src[i][1]);
}
static void ps_mul_pair_single_c(INTFLOAT (*restrict dst)[2],
INTFLOAT (*src0)[2], INTFLOAT *src1,
int n)
{
for (int i = 0; i < n; i++) {
dst[i][0] = AAC_MUL16(src0[i][0], src1[i]);
dst[i][1] = AAC_MUL16(src0[i][1], src1[i]);
}
}
static void ps_hybrid_analysis_c(INTFLOAT (*restrict out)[2],
INTFLOAT (*in)[2],
const INTFLOAT (*filter)[8][2],
ptrdiff_t stride, int n)
{
INT64FLOAT inre0[6], inre1[6], inim0[6], inim1[6];
for (int j = 0; j < 6; j++) {
inre0[j] = in[j][0] + in[12 - j][0];
inre1[j] = in[j][1] - in[12 - j][1];
inim0[j] = in[j][1] + in[12 - j][1];
inim1[j] = in[j][0] - in[12 - j][0];
}
for (int i = 0; i < n; i++) {
INT64FLOAT sum_re = (INT64FLOAT)filter[i][6][0] * in[6][0];
INT64FLOAT sum_im = (INT64FLOAT)filter[i][6][0] * in[6][1];
for (int j = 0; j < 6; j++) {
sum_re += (INT64FLOAT)filter[i][j][0] * inre0[j] -
(INT64FLOAT)filter[i][j][1] * inre1[j];
sum_im += (INT64FLOAT)filter[i][j][0] * inim0[j] +
(INT64FLOAT)filter[i][j][1] * inim1[j];
}
#if USE_FIXED
out[i * stride][0] = (int)((sum_re + 0x40000000) >> 31);
out[i * stride][1] = (int)((sum_im + 0x40000000) >> 31);
#else
out[i * stride][0] = sum_re;
out[i * stride][1] = sum_im;
#endif /* USE_FIXED */
}
}
static void ps_hybrid_analysis_ileave_c(INTFLOAT (*restrict out)[32][2],
INTFLOAT L[2][38][64],
int i, int len)
{
for (; i < 64; i++) {
for (int j = 0; j < len; j++) {
out[i][j][0] = L[0][j][i];
out[i][j][1] = L[1][j][i];
}
}
}
static void ps_hybrid_synthesis_deint_c(INTFLOAT out[2][38][64],
INTFLOAT (*restrict in)[32][2],
int i, int len)
{
for (; i < 64; i++) {
for (int n = 0; n < len; n++) {
out[0][n][i] = in[i][n][0];
out[1][n][i] = in[i][n][1];
}
}
}
static void ps_decorrelate_c(INTFLOAT (*out)[2], INTFLOAT (*delay)[2],
INTFLOAT (*ap_delay)[PS_QMF_TIME_SLOTS + PS_MAX_AP_DELAY][2],
const INTFLOAT phi_fract[2], const INTFLOAT (*Q_fract)[2],
const INTFLOAT *transient_gain,
INTFLOAT g_decay_slope,
int len)
{
static const INTFLOAT a[] = { Q31(0.65143905753106f),
Q31(0.56471812200776f),
Q31(0.48954165955695f) };
INTFLOAT ag[PS_AP_LINKS];
int m, n;
for (m = 0; m < PS_AP_LINKS; m++)
ag[m] = AAC_MUL30(a[m], g_decay_slope);
for (n = 0; n < len; n++) {
INTFLOAT in_re = AAC_MSUB30(delay[n][0], phi_fract[0], delay[n][1], phi_fract[1]);
INTFLOAT in_im = AAC_MADD30(delay[n][0], phi_fract[1], delay[n][1], phi_fract[0]);
for (m = 0; m < PS_AP_LINKS; m++) {
INTFLOAT a_re = AAC_MUL31(ag[m], in_re);
INTFLOAT a_im = AAC_MUL31(ag[m], in_im);
INTFLOAT link_delay_re = ap_delay[m][n+2-m][0];
INTFLOAT link_delay_im = ap_delay[m][n+2-m][1];
INTFLOAT fractional_delay_re = Q_fract[m][0];
INTFLOAT fractional_delay_im = Q_fract[m][1];
INTFLOAT apd_re = in_re;
INTFLOAT apd_im = in_im;
in_re = AAC_MSUB30(link_delay_re, fractional_delay_re,
link_delay_im, fractional_delay_im);
in_re -= (UINTFLOAT)a_re;
in_im = AAC_MADD30(link_delay_re, fractional_delay_im,
link_delay_im, fractional_delay_re);
in_im -= (UINTFLOAT)a_im;
ap_delay[m][n+5][0] = apd_re + (UINTFLOAT)AAC_MUL31(ag[m], in_re);
ap_delay[m][n+5][1] = apd_im + (UINTFLOAT)AAC_MUL31(ag[m], in_im);
}
out[n][0] = AAC_MUL16(transient_gain[n], in_re);
out[n][1] = AAC_MUL16(transient_gain[n], in_im);
}
}
static void ps_stereo_interpolate_c(INTFLOAT (*l)[2], INTFLOAT (*r)[2],
INTFLOAT h[2][4], INTFLOAT h_step[2][4],
int len)
{
INTFLOAT h0 = h[0][0];
INTFLOAT h1 = h[0][1];
INTFLOAT h2 = h[0][2];
INTFLOAT h3 = h[0][3];
UINTFLOAT hs0 = h_step[0][0];
UINTFLOAT hs1 = h_step[0][1];
UINTFLOAT hs2 = h_step[0][2];
UINTFLOAT hs3 = h_step[0][3];
int n;
for (n = 0; n < len; n++) {
//l is s, r is d
INTFLOAT l_re = l[n][0];
INTFLOAT l_im = l[n][1];
INTFLOAT r_re = r[n][0];
INTFLOAT r_im = r[n][1];
h0 += hs0;
h1 += hs1;
h2 += hs2;
h3 += hs3;
l[n][0] = AAC_MADD30(h0, l_re, h2, r_re);
l[n][1] = AAC_MADD30(h0, l_im, h2, r_im);
r[n][0] = AAC_MADD30(h1, l_re, h3, r_re);
r[n][1] = AAC_MADD30(h1, l_im, h3, r_im);
}
}
static void ps_stereo_interpolate_ipdopd_c(INTFLOAT (*l)[2], INTFLOAT (*r)[2],
INTFLOAT h[2][4], INTFLOAT h_step[2][4],
int len)
{
INTFLOAT h00 = h[0][0], h10 = h[1][0];
INTFLOAT h01 = h[0][1], h11 = h[1][1];
INTFLOAT h02 = h[0][2], h12 = h[1][2];
INTFLOAT h03 = h[0][3], h13 = h[1][3];
UINTFLOAT hs00 = h_step[0][0], hs10 = h_step[1][0];
UINTFLOAT hs01 = h_step[0][1], hs11 = h_step[1][1];
UINTFLOAT hs02 = h_step[0][2], hs12 = h_step[1][2];
UINTFLOAT hs03 = h_step[0][3], hs13 = h_step[1][3];
int n;
for (n = 0; n < len; n++) {
//l is s, r is d
INTFLOAT l_re = l[n][0];
INTFLOAT l_im = l[n][1];
INTFLOAT r_re = r[n][0];
INTFLOAT r_im = r[n][1];
h00 += hs00;
h01 += hs01;
h02 += hs02;
h03 += hs03;
h10 += hs10;
h11 += hs11;
h12 += hs12;
h13 += hs13;
l[n][0] = AAC_MSUB30_V8(h00, l_re, h02, r_re, h10, l_im, h12, r_im);
l[n][1] = AAC_MADD30_V8(h00, l_im, h02, r_im, h10, l_re, h12, r_re);
r[n][0] = AAC_MSUB30_V8(h01, l_re, h03, r_re, h11, l_im, h13, r_im);
r[n][1] = AAC_MADD30_V8(h01, l_im, h03, r_im, h11, l_re, h13, r_re);
}
}
av_cold void AAC_RENAME(ff_psdsp_init)(PSDSPContext *s)
{
s->add_squares = ps_add_squares_c;
s->mul_pair_single = ps_mul_pair_single_c;
s->hybrid_analysis = ps_hybrid_analysis_c;
s->hybrid_analysis_ileave = ps_hybrid_analysis_ileave_c;
s->hybrid_synthesis_deint = ps_hybrid_synthesis_deint_c;
s->decorrelate = ps_decorrelate_c;
s->stereo_interpolate[0] = ps_stereo_interpolate_c;
s->stereo_interpolate[1] = ps_stereo_interpolate_ipdopd_c;
#if !USE_FIXED
#if ARCH_ARM
ff_psdsp_init_arm(s);
#elif ARCH_AARCH64
ff_psdsp_init_aarch64(s);
#elif ARCH_RISCV
ff_psdsp_init_riscv(s);
#elif ARCH_X86
ff_psdsp_init_x86(s);
#endif
#endif /* !USE_FIXED */
}
|
