aboutsummaryrefslogtreecommitdiffstats
path: root/libavcodec/celp_filters.c
blob: 4f627e009280e1919e7bcffe8239a4702efcc39a (plain) (blame)
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
/*
 * various filters for ACELP-based codecs
 *
 * Copyright (c) 2008 Vladimir Voroshilov
 *
 * 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 <stdint.h>
#include <string.h>

#include "config.h"
#include "celp_filters.h"
#include "libavutil/avassert.h"
#include "libavutil/common.h"

void ff_celp_convolve_circ(int16_t* fc_out, const int16_t* fc_in,
                           const int16_t* filter, int len)
{
    int i, k;

    memset(fc_out, 0, len * sizeof(int16_t));

    /* Since there are few pulses over an entire subframe (i.e. almost
       all fc_in[i] are zero) it is faster to loop over fc_in first. */
    for (i = 0; i < len; i++) {
        if (fc_in[i]) {
            for (k = 0; k < i; k++)
                fc_out[k] += (fc_in[i] * filter[len + k - i]) >> 15;

            for (k = i; k < len; k++)
                fc_out[k] += (fc_in[i] * filter[      k - i]) >> 15;
        }
    }
}

void ff_celp_circ_addf(float *out, const float *in,
                       const float *lagged, int lag, float fac, int n)
{
    int k;
    for (k = 0; k < lag; k++)
        out[k] = in[k] + fac * lagged[n + k - lag];
    for (; k < n; k++)
        out[k] = in[k] + fac * lagged[    k - lag];
}

int ff_celp_lp_synthesis_filter(int16_t *out, const int16_t *filter_coeffs,
                                const int16_t *in, int buffer_length,
                                int filter_length, int stop_on_overflow,
                                int shift, int rounder)
{
    int i,n;

    for (n = 0; n < buffer_length; n++) {
        int sum = rounder, sum1;
        for (i = 1; i <= filter_length; i++)
            sum -= (unsigned)(filter_coeffs[i-1] * out[n-i]);

        sum1 = ((sum >> 12) + in[n]) >> shift;
        sum  = av_clip_int16(sum1);

        if (stop_on_overflow && sum != sum1)
            return 1;

        out[n] = sum;
    }

    return 0;
}

void ff_celp_lp_synthesis_filterf(float *out, const float *filter_coeffs,
                                  const float* in, int buffer_length,
                                  int filter_length)
{
    int i,n;

#if 0 // Unoptimized code path for improved readability
    for (n = 0; n < buffer_length; n++) {
        out[n] = in[n];
        for (i = 1; i <= filter_length; i++)
            out[n] -= filter_coeffs[i-1] * out[n-i];
    }
#else
    float out0, out1, out2, out3;
    float old_out0, old_out1, old_out2, old_out3;
    float a,b,c;

    a = filter_coeffs[0];
    b = filter_coeffs[1];
    c = filter_coeffs[2];
    b -= filter_coeffs[0] * filter_coeffs[0];
    c -= filter_coeffs[1] * filter_coeffs[0];
    c -= filter_coeffs[0] * b;

    av_assert2((filter_length&1)==0 && filter_length>=4);

    old_out0 = out[-4];
    old_out1 = out[-3];
    old_out2 = out[-2];
    old_out3 = out[-1];
    for (n = 0; n <= buffer_length - 4; n+=4) {
        float tmp0,tmp1,tmp2;
        float val;

        out0 = in[0];
        out1 = in[1];
        out2 = in[2];
        out3 = in[3];

        out0 -= filter_coeffs[2] * old_out1;
        out1 -= filter_coeffs[2] * old_out2;
        out2 -= filter_coeffs[2] * old_out3;

        out0 -= filter_coeffs[1] * old_out2;
        out1 -= filter_coeffs[1] * old_out3;

        out0 -= filter_coeffs[0] * old_out3;

        val = filter_coeffs[3];

        out0 -= val * old_out0;
        out1 -= val * old_out1;
        out2 -= val * old_out2;
        out3 -= val * old_out3;

        for (i = 5; i < filter_length; i += 2) {
            old_out3 = out[-i];
            val = filter_coeffs[i-1];

            out0 -= val * old_out3;
            out1 -= val * old_out0;
            out2 -= val * old_out1;
            out3 -= val * old_out2;

            old_out2 = out[-i-1];

            val = filter_coeffs[i];

            out0 -= val * old_out2;
            out1 -= val * old_out3;
            out2 -= val * old_out0;
            out3 -= val * old_out1;

            FFSWAP(float, old_out0, old_out2);
            old_out1 = old_out3;
        }

        tmp0 = out0;
        tmp1 = out1;
        tmp2 = out2;

        out3 -= a * tmp2;
        out2 -= a * tmp1;
        out1 -= a * tmp0;

        out3 -= b * tmp1;
        out2 -= b * tmp0;

        out3 -= c * tmp0;


        out[0] = out0;
        out[1] = out1;
        out[2] = out2;
        out[3] = out3;

        old_out0 = out0;
        old_out1 = out1;
        old_out2 = out2;
        old_out3 = out3;

        out += 4;
        in  += 4;
    }

    out -= n;
    in -= n;
    for (; n < buffer_length; n++) {
        out[n] = in[n];
        for (i = 1; i <= filter_length; i++)
            out[n] -= filter_coeffs[i-1] * out[n-i];
    }
#endif
}

void ff_celp_lp_zero_synthesis_filterf(float *out, const float *filter_coeffs,
                                       const float *in, int buffer_length,
                                       int filter_length)
{
    int i,n;

    for (n = 0; n < buffer_length; n++) {
        out[n] = in[n];
        for (i = 1; i <= filter_length; i++)
            out[n] += filter_coeffs[i-1] * in[n-i];
    }
}

void ff_celp_filter_init(CELPFContext *c)
{
    c->celp_lp_synthesis_filterf        = ff_celp_lp_synthesis_filterf;
    c->celp_lp_zero_synthesis_filterf   = ff_celp_lp_zero_synthesis_filterf;

#if HAVE_MIPSFPU
    ff_celp_filter_init_mips(c);
#endif
}