dca: factorize scaling in inverse ADPCM
[libav.git] / libavcodec / x86 / dcadsp_init.c
CommitLineData
5b59a9fc
CG
1/*
2 * Copyright (c) 2012-2014 Christophe Gisquet <christophe.gisquet@gmail.com>
3 *
4 * This file is part of Libav.
5 *
6 * Libav 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 * Libav 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 Libav; if not, write to the Free Software
18 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19 */
20
21#include "libavutil/attributes.h"
22#include "libavutil/cpu.h"
23#include "libavutil/x86/cpu.h"
24#include "libavcodec/dcadsp.h"
25
26void ff_int8x8_fmul_int32_sse(float *dst, const int8_t *src, int scale);
27void ff_int8x8_fmul_int32_sse2(float *dst, const int8_t *src, int scale);
28void ff_int8x8_fmul_int32_sse4(float *dst, const int8_t *src, int scale);
ad507d79
CG
29void ff_dca_lfe_fir0_sse(float *out, const float *in, const float *coefs);
30void ff_dca_lfe_fir1_sse(float *out, const float *in, const float *coefs);
5b59a9fc
CG
31
32av_cold void ff_dcadsp_init_x86(DCADSPContext *s)
33{
34 int cpu_flags = av_get_cpu_flags();
35
36 if (EXTERNAL_SSE(cpu_flags)) {
37#if ARCH_X86_32
38 s->int8x8_fmul_int32 = ff_int8x8_fmul_int32_sse;
39#endif
ad507d79
CG
40 s->lfe_fir[0] = ff_dca_lfe_fir0_sse;
41 s->lfe_fir[1] = ff_dca_lfe_fir1_sse;
5b59a9fc
CG
42 }
43
44 if (EXTERNAL_SSE2(cpu_flags)) {
45 s->int8x8_fmul_int32 = ff_int8x8_fmul_int32_sse2;
46 }
47
48 if (EXTERNAL_SSE4(cpu_flags)) {
49 s->int8x8_fmul_int32 = ff_int8x8_fmul_int32_sse4;
50 }
51}
08e3ea60
CG
52
53void ff_synth_filter_inner_sse2(float *synth_buf_ptr, float synth_buf2[32],
54 const float window[512],
55 float out[32], intptr_t offset, float scale);
56
57static void synth_filter_sse2(FFTContext *imdct,
58 float *synth_buf_ptr, int *synth_buf_offset,
59 float synth_buf2[32], const float window[512],
60 float out[32], const float in[32], float scale)
61{
62 float *synth_buf= synth_buf_ptr + *synth_buf_offset;
63
64 imdct->imdct_half(imdct, synth_buf, in);
65
66 ff_synth_filter_inner_sse2(synth_buf, synth_buf2, window,
67 out, *synth_buf_offset, scale);
68
69 *synth_buf_offset = (*synth_buf_offset - 32) & 511;
70}
71
72av_cold void ff_synth_filter_init_x86(SynthFilterContext *s)
73{
74 int cpu_flags = av_get_cpu_flags();
75
76 if (EXTERNAL_SSE2(cpu_flags)) {
77 s->synth_filter_float = synth_filter_sse2;
78 }
79}