1/*
2 * Copyright (c) 2007 Luca Barbato <lu_zero@gentoo.org>
3 *
4 * This file is part of FFmpeg.
5 *
6 * FFmpeg is free software; you can redistribute it and/or
7 * modify it under the terms of the GNU Lesser General Public
8 * License as published by the Free Software Foundation; either
9 * version 2.1 of the License, or (at your option) any later version.
10 *
11 * FFmpeg is distributed in the hope that it will be useful,
12 * but WITHOUT ANY WARRANTY; without even the implied warranty of
13 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
14 * Lesser General Public License for more details.
15 *
16 * You should have received a copy of the GNU Lesser General Public
17 * License along with FFmpeg; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19 */
20
21/**
22 * @file
23 * miscellaneous audio operations
24 */
25
26#include "config.h"
27
28#include "libavutil/attributes.h"
29#include "libavutil/cpu.h"
30#include "libavutil/ppc/cpu.h"
31#include "libavutil/ppc/util_altivec.h"
32
33#include "libavcodec/audiodsp.h"
34
35#if HAVE_ALTIVEC
36
37static int32_t scalarproduct_int16_altivec(const int16_t *v1, const int16_t *v2,
38                                           int order)
39{
40    int i;
41    LOAD_ZERO;
42    register vec_s16 vec1;
43    register vec_s32 res = vec_splat_s32(0), t;
44    int32_t ires;
45
46    for (i = 0; i < order; i += 8) {
47        vec1 = vec_unaligned_load(v1);
48        t    = vec_msum(vec1, vec_ld(0, v2), zero_s32v);
49        res  = vec_sums(t, res);
50        v1  += 8;
51        v2  += 8;
52    }
53    res = vec_splat(res, 3);
54    vec_ste(res, 0, &ires);
55
56    return ires;
57}
58
59#endif /* HAVE_ALTIVEC */
60
61#if HAVE_VSX
62
63static int32_t scalarproduct_int16_vsx(const int16_t *v1, const int16_t *v2, int order)
64{
65    int i;
66    LOAD_ZERO;
67    register vec_s16 vec1;
68    register vec_s32 res = vec_splat_s32(0), t;
69    int32_t ires;
70
71    for (i = 0; i < order; i += 8) {
72        vec1 = vec_vsx_ld(0, v1);
73        t    = vec_msum(vec1, vec_ld(0, v2), zero_s32v);
74        res  = vec_sums(t, res);
75        v1  += 8;
76        v2  += 8;
77    }
78    res = vec_splat(res, 3);
79    vec_ste(res, 0, &ires);
80
81    return ires;
82}
83
84#endif /* HAVE_VSX */
85
86av_cold void ff_audiodsp_init_ppc(AudioDSPContext *c)
87{
88#if HAVE_ALTIVEC
89    if (!PPC_ALTIVEC(av_get_cpu_flags()))
90        return;
91
92    c->scalarproduct_int16 = scalarproduct_int16_altivec;
93#endif /* HAVE_ALTIVEC */
94
95#if HAVE_VSX
96    if (!PPC_VSX(av_get_cpu_flags()))
97        return;
98
99    c->scalarproduct_int16 = scalarproduct_int16_vsx;
100#endif /* HAVE_VSX */
101}
102