summaryrefslogtreecommitdiffstats
path: root/libavutil/float_dsp.h
blob: 9c664592bd550ec18c09ce9d0406f399bf97f1af (plain)
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
/*
 * 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
 */

#ifndef AVUTIL_FLOAT_DSP_H
#define AVUTIL_FLOAT_DSP_H

#include "config.h"

typedef struct AVFloatDSPContext {
    /**
     * Calculate the entry wise product of two vectors of floats and store the result in
     * a vector of floats.
     *
     * @param dst  output vector
     *             constraints: 32-byte aligned
     * @param src0 first input vector
     *             constraints: 32-byte aligned
     * @param src1 second input vector
     *             constraints: 32-byte aligned
     * @param len  number of elements in the input
     *             constraints: multiple of 16
     */
    void (*vector_fmul)(float *dst, const float *src0, const float *src1,
                        int len);

    /**
     * Multiply a vector of floats by a scalar float and add to
     * destination vector.  Source and destination vectors must
     * overlap exactly or not at all.
     *
     * @param dst result vector
     *            constraints: 32-byte aligned
     * @param src input vector
     *            constraints: 32-byte aligned
     * @param mul scalar value
     * @param len length of vector
     *            constraints: multiple of 16
     */
    void (*vector_fmac_scalar)(float *dst, const float *src, float mul,
                               int len);

    /**
     * Multiply a vector of doubles by a scalar double and add to
     * destination vector.  Source and destination vectors must
     * overlap exactly or not at all.
     *
     * @param dst result vector
     *            constraints: 32-byte aligned
     * @param src input vector
     *            constraints: 32-byte aligned
     * @param mul scalar value
     * @param len length of vector
     *            constraints: multiple of 16
     */
    void (*vector_dmac_scalar)(double *dst, const double *src, double mul,
                               int len);

    /**
     * Multiply a vector of floats by a scalar float.  Source and
     * destination vectors must overlap exactly or not at all.
     *
     * @param dst result vector
     *            constraints: 16-byte aligned
     * @param src input vector
     *            constraints: 16-byte aligned
     * @param mul scalar value
     * @param len length of vector
     *            constraints: multiple of 4
     */
    void (*vector_fmul_scalar)(float *dst, const float *src, float mul,
                               int len);

    /**
     * Multiply a vector of double by a scalar double.  Source and
     * destination vectors must overlap exactly or not at all.
     *
     * @param dst result vector
     *            constraints: 32-byte aligned
     * @param src input vector
     *            constraints: 32-byte aligned
     * @param mul scalar value
     * @param len length of vector
     *            constraints: multiple of 8
     */
    void (*vector_dmul_scalar)(double *dst, const double *src, double mul,
                               int len);

    /**
     * Overlap/add with window function.
     * Used primarily by MDCT-based audio codecs.
     * Source and destination vectors must overlap exactly or not at all.
     *
     * @param dst  result vector
     *             constraints: 16-byte aligned
     * @param src0 first source vector
     *             constraints: 16-byte aligned
     * @param src1 second source vector
     *             constraints: 16-byte aligned
     * @param win  half-window vector
     *             constraints: 16-byte aligned
     * @param len  length of vector
     *             constraints: multiple of 4
     */
    void (*vector_fmul_window)(float *dst, const float *src0,
                               const float *src1, const float *win, int len);

    /**
     * Calculate the entry wise product of two vectors of floats, add a third vector of
     * floats and store the result in a vector of floats.
     *
     * @param dst  output vector
     *             constraints: 32-byte aligned
     * @param src0 first input vector
     *             constraints: 32-byte aligned
     * @param src1 second input vector
     *             constraints: 32-byte aligned
     * @param src2 third input vector
     *             constraints: 32-byte aligned
     * @param len  number of elements in the input
     *             constraints: multiple of 16
     */
    void (*vector_fmul_add)(float *dst, const float *src0, const float *src1,
                            const float *src2, int len);

    /**
     * Calculate the entry wise product of two vectors of floats, and store the result
     * in a vector of floats. The second vector of floats is iterated over
     * in reverse order.
     *
     * @param dst  output vector
     *             constraints: 32-byte aligned
     * @param src0 first input vector
     *             constraints: 32-byte aligned
     * @param src1 second input vector
     *             constraints: 32-byte aligned
     * @param len  number of elements in the input
     *             constraints: multiple of 16
     */
    void (*vector_fmul_reverse)(float *dst, const float *src0,
                                const float *src1, int len);

    /**
     * Calculate the sum and difference of two vectors of floats.
     *
     * @param v1  first input vector, sum output, 16-byte aligned
     * @param v2  second input vector, difference output, 16-byte aligned
     * @param len length of vectors, multiple of 4
     */
    void (*butterflies_float)(float *av_restrict v1, float *av_restrict v2, int len);

    /**
     * Calculate the scalar product of two vectors of floats.
     *
     * @param v1  first vector, 16-byte aligned
     * @param v2  second vector, 16-byte aligned
     * @param len length of vectors, multiple of 4
     *
     * @return sum of elementwise products
     */
    float (*scalarproduct_float)(const float *v1, const float *v2, int len);

    /**
     * Calculate the entry wise product of two vectors of doubles and store the result in
     * a vector of doubles.
     *
     * @param dst  output vector
     *             constraints: 32-byte aligned
     * @param src0 first input vector
     *             constraints: 32-byte aligned
     * @param src1 second input vector
     *             constraints: 32-byte aligned
     * @param len  number of elements in the input
     *             constraints: multiple of 16
     */
    void (*vector_dmul)(double *dst, const double *src0, const double *src1,
                        int len);
} AVFloatDSPContext;

/**
 * Return the scalar product of two vectors.
 *
 * @param v1  first input vector
 * @param v2  first input vector
 * @param len number of elements
 *
 * @return sum of elementwise products
 */
float avpriv_scalarproduct_float_c(const float *v1, const float *v2, int len);

void ff_float_dsp_init_aarch64(AVFloatDSPContext *fdsp);
void ff_float_dsp_init_arm(AVFloatDSPContext *fdsp);
void ff_float_dsp_init_ppc(AVFloatDSPContext *fdsp, int strict);
void ff_float_dsp_init_x86(AVFloatDSPContext *fdsp);
void ff_float_dsp_init_mips(AVFloatDSPContext *fdsp);

/**
 * Allocate a float DSP context.
 *
 * @param strict  setting to non-zero avoids using functions which may not be IEEE-754 compliant
 */
AVFloatDSPContext *avpriv_float_dsp_alloc(int strict);

#endif /* AVUTIL_FLOAT_DSP_H */
OpenPOWER on IntegriCloud