aboutsummaryrefslogtreecommitdiffstats
path: root/libswresample/resample_template.c
blob: 5151f585fc45973b6f0a731d617b002655ac11a6 (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
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
/*
 * audio resampling
 * Copyright (c) 2004-2012 Michael Niedermayer <michaelni@gmx.at>
 *
 * 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
 */

/**
 * @file
 * audio resampling
 * @author Michael Niedermayer <michaelni@gmx.at>
 */

#if    defined(TEMPLATE_RESAMPLE_DBL)     \
    || defined(TEMPLATE_RESAMPLE_DBL_SSE2)

#    define FILTER_SHIFT 0
#    define DELEM  double
#    define FELEM  double
#    define FELEM2 double
#    define FELEML double
#    define OUT(d, v) d = v

#    if defined(TEMPLATE_RESAMPLE_DBL)
#        define RENAME(N) N ## _double
#    elif defined(TEMPLATE_RESAMPLE_DBL_SSE2)
#        define COMMON_CORE COMMON_CORE_DBL_SSE2
#        define LINEAR_CORE LINEAR_CORE_DBL_SSE2
#        define RENAME(N) N ## _double_sse2
#    endif

#elif    defined(TEMPLATE_RESAMPLE_FLT)     \
      || defined(TEMPLATE_RESAMPLE_FLT_SSE) \
      || defined(TEMPLATE_RESAMPLE_FLT_AVX)

#    define FILTER_SHIFT 0
#    define DELEM  float
#    define FELEM  float
#    define FELEM2 float
#    define FELEML float
#    define OUT(d, v) d = v

#    if defined(TEMPLATE_RESAMPLE_FLT)
#        define RENAME(N) N ## _float
#    elif defined(TEMPLATE_RESAMPLE_FLT_SSE)
#        define COMMON_CORE COMMON_CORE_FLT_SSE
#        define LINEAR_CORE LINEAR_CORE_FLT_SSE
#        define RENAME(N) N ## _float_sse
#    elif defined(TEMPLATE_RESAMPLE_FLT_AVX)
#        define COMMON_CORE COMMON_CORE_FLT_AVX
#        define LINEAR_CORE LINEAR_CORE_FLT_AVX
#        define RENAME(N) N ## _float_avx
#    endif

#elif defined(TEMPLATE_RESAMPLE_S32)
#    define RENAME(N) N ## _int32
#    define FILTER_SHIFT 30
#    define DELEM  int32_t
#    define FELEM  int32_t
#    define FELEM2 int64_t
#    define FELEML int64_t
#    define FELEM_MAX INT32_MAX
#    define FELEM_MIN INT32_MIN
#    define OUT(d, v) v = (v + (1<<(FILTER_SHIFT-1)))>>FILTER_SHIFT;\
                      d = (uint64_t)(v + 0x80000000) > 0xFFFFFFFF ? (v>>63) ^ 0x7FFFFFFF : v

#elif    defined(TEMPLATE_RESAMPLE_S16)      \
      || defined(TEMPLATE_RESAMPLE_S16_MMX2) \
      || defined(TEMPLATE_RESAMPLE_S16_SSE2)

#    define FILTER_SHIFT 15
#    define DELEM  int16_t
#    define FELEM  int16_t
#    define FELEM2 int32_t
#    define FELEML int64_t
#    define FELEM_MAX INT16_MAX
#    define FELEM_MIN INT16_MIN
#    define OUT(d, v) v = (v + (1<<(FILTER_SHIFT-1)))>>FILTER_SHIFT;\
                      d = (unsigned)(v + 32768) > 65535 ? (v>>31) ^ 32767 : v

#    if defined(TEMPLATE_RESAMPLE_S16)
#        define RENAME(N) N ## _int16
#    elif defined(TEMPLATE_RESAMPLE_S16_MMX2)
#        define COMMON_CORE COMMON_CORE_INT16_MMX2
#        define LINEAR_CORE LINEAR_CORE_INT16_MMX2
#        define RENAME(N) N ## _int16_mmx2
#    elif defined(TEMPLATE_RESAMPLE_S16_SSE2)
#        define COMMON_CORE COMMON_CORE_INT16_SSE2
#        define LINEAR_CORE LINEAR_CORE_INT16_SSE2
#        define RENAME(N) N ## _int16_sse2
#    endif

#endif

int RENAME(swri_resample)(ResampleContext *c, DELEM *dst, const DELEM *src, int *consumed, int src_size, int dst_size, int update_ctx){
    int dst_index, i;
    int index= c->index;
    int frac= c->frac;
    int dst_incr_frac= c->dst_incr % c->src_incr;
    int dst_incr=      c->dst_incr / c->src_incr;

    av_assert1(c->filter_shift == FILTER_SHIFT);
    av_assert1(c->felem_size == sizeof(FELEM));

    if (c->filter_length == 1 && c->phase_shift == 0) {
        int64_t index2= (1LL<<32)*c->frac/c->src_incr + (1LL<<32)*index;
        int64_t incr= (1LL<<32) * c->dst_incr / c->src_incr;
        int new_size = (src_size * (int64_t)c->src_incr - frac + c->dst_incr - 1) / c->dst_incr;

        dst_size= FFMIN(dst_size, new_size);

        for(dst_index=0; dst_index < dst_size; dst_index++){
            dst[dst_index] = src[index2>>32];
            index2 += incr;
        }
        index += dst_index * dst_incr;
        index += (frac + dst_index * (int64_t)dst_incr_frac) / c->src_incr;
        frac   = (frac + dst_index * (int64_t)dst_incr_frac) % c->src_incr;
        av_assert2(index >= 0);
        *consumed= index;
        index = 0;
    } else if (index >= 0 &&
               src_size*(int64_t)c->src_incr < (INT64_MAX >> (c->phase_shift+1))) {
        int64_t end_index = (1LL + src_size - c->filter_length) << c->phase_shift;
        int64_t delta_frac = (end_index - index) * c->src_incr - c->frac;
        int delta_n = (delta_frac + c->dst_incr - 1) / c->dst_incr;
        int n = FFMIN(dst_size, delta_n);
        int sample_index;

        if (!c->linear) {
            sample_index = index >> c->phase_shift;
            index &= c->phase_mask;
            for (dst_index = 0; dst_index < n; dst_index++) {
                FELEM *filter = ((FELEM *) c->filter_bank) + c->filter_alloc * index;

#ifdef COMMON_CORE
                COMMON_CORE
#else
                FELEM2 val=0;
                for (i = 0; i < c->filter_length; i++) {
                    val += src[sample_index + i] * (FELEM2)filter[i];
                }
                OUT(dst[dst_index], val);
#endif

                frac += dst_incr_frac;
                index += dst_incr;
                if (frac >= c->src_incr) {
                    frac -= c->src_incr;
                    index++;
                }
                sample_index += index >> c->phase_shift;
                index &= c->phase_mask;
            }
        } else {
            sample_index = index >> c->phase_shift;
            index &= c->phase_mask;
            for (dst_index = 0; dst_index < n; dst_index++) {
                FELEM *filter = ((FELEM *) c->filter_bank) + c->filter_alloc * index;
                FELEM2 val=0, v2 = 0;

#ifdef LINEAR_CORE
                LINEAR_CORE
#else
                for (i = 0; i < c->filter_length; i++) {
                    val += src[sample_index + i] * (FELEM2)filter[i];
                    v2  += src[sample_index + i] * (FELEM2)filter[i + c->filter_alloc];
                }
#endif
                val += (v2 - val) * (FELEML) frac / c->src_incr;
                OUT(dst[dst_index], val);

                frac += dst_incr_frac;
                index += dst_incr;
                if (frac >= c->src_incr) {
                    frac -= c->src_incr;
                    index++;
                }
                sample_index += index >> c->phase_shift;
                index &= c->phase_mask;
            }
        }

        *consumed = sample_index;
    } else {
        int sample_index = 0;
        for(dst_index=0; dst_index < dst_size; dst_index++){
            FELEM *filter;
            FELEM2 val=0;

            sample_index += index >> c->phase_shift;
            index &= c->phase_mask;
            filter = ((FELEM*)c->filter_bank) + c->filter_alloc*index;

            if(sample_index + c->filter_length > src_size || -sample_index >= src_size){
                break;
            }else if(sample_index < 0){
                for(i=0; i<c->filter_length; i++)
                    val += src[FFABS(sample_index + i)] * (FELEM2)filter[i];
                OUT(dst[dst_index], val);
            }else if(c->linear){
                FELEM2 v2=0;
#ifdef LINEAR_CORE
                LINEAR_CORE
#else
                for(i=0; i<c->filter_length; i++){
                    val += src[sample_index + i] * (FELEM2)filter[i];
                    v2  += src[sample_index + i] * (FELEM2)filter[i + c->filter_alloc];
                }
#endif
                val+=(v2-val)*(FELEML)frac / c->src_incr;
                OUT(dst[dst_index], val);
            }else{
#ifdef COMMON_CORE
                COMMON_CORE
#else
                for(i=0; i<c->filter_length; i++){
                    val += src[sample_index + i] * (FELEM2)filter[i];
                }
                OUT(dst[dst_index], val);
#endif
            }

            frac += dst_incr_frac;
            index += dst_incr;
            if(frac >= c->src_incr){
                frac -= c->src_incr;
                index++;
            }
        }
        *consumed= FFMAX(sample_index, 0);
        index += FFMIN(sample_index, 0) << c->phase_shift;
    }

    if(update_ctx){
        c->frac= frac;
        c->index= index;
        c->dst_incr= dst_incr_frac + c->src_incr*dst_incr;
    }

    return dst_index;
}

#undef COMMON_CORE
#undef LINEAR_CORE
#undef RENAME
#undef FILTER_SHIFT
#undef DELEM
#undef FELEM
#undef FELEM2
#undef FELEML
#undef FELEM_MAX
#undef FELEM_MIN
#undef OUT