1 /*
2 * x86-optimized AC-3 DSP functions
3 * Copyright (c) 2011 Justin Ruggles
4 *
5 * This file is part of FFmpeg.
6 *
7 * FFmpeg is free software; you can redistribute it and/or
8 * modify it under the terms of the GNU Lesser General Public
9 * License as published by the Free Software Foundation; either
10 * version 2.1 of the License, or (at your option) any later version.
11 *
12 * FFmpeg is distributed in the hope that it will be useful,
13 * but WITHOUT ANY WARRANTY; without even the implied warranty of
14 * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
15 * Lesser General Public License for more details.
16 *
17 * You should have received a copy of the GNU Lesser General Public
18 * License along with FFmpeg; if not, write to the Free Software
19 * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
20 */
21
22 #include "libavutil/attributes.h"
23 #include "libavutil/mem.h"
24 #include "libavutil/x86/asm.h"
25 #include "libavutil/x86/cpu.h"
26 #include "libavcodec/ac3.h"
27 #include "libavcodec/ac3dsp.h"
28
29 void ff_ac3_exponent_min_mmx (uint8_t *exp, int num_reuse_blocks, int nb_coefs);
30 void ff_ac3_exponent_min_mmxext(uint8_t *exp, int num_reuse_blocks, int nb_coefs);
31 void ff_ac3_exponent_min_sse2 (uint8_t *exp, int num_reuse_blocks, int nb_coefs);
32
33 void ff_float_to_fixed24_3dnow(int32_t *dst, const float *src, unsigned int len);
34 void ff_float_to_fixed24_sse (int32_t *dst, const float *src, unsigned int len);
35 void ff_float_to_fixed24_sse2 (int32_t *dst, const float *src, unsigned int len);
36
37 int ff_ac3_compute_mantissa_size_sse2(uint16_t mant_cnt[6][16]);
38
39 void ff_ac3_extract_exponents_sse2 (uint8_t *exp, int32_t *coef, int nb_coefs);
40 void ff_ac3_extract_exponents_ssse3(uint8_t *exp, int32_t *coef, int nb_coefs);
41
ff_ac3dsp_init_x86(AC3DSPContext * c,int bit_exact)42 av_cold void ff_ac3dsp_init_x86(AC3DSPContext *c, int bit_exact)
43 {
44 int cpu_flags = av_get_cpu_flags();
45
46 if (EXTERNAL_MMX(cpu_flags)) {
47 c->ac3_exponent_min = ff_ac3_exponent_min_mmx;
48 }
49 if (EXTERNAL_AMD3DNOW(cpu_flags)) {
50 if (!bit_exact) {
51 c->float_to_fixed24 = ff_float_to_fixed24_3dnow;
52 }
53 }
54 if (EXTERNAL_MMXEXT(cpu_flags)) {
55 c->ac3_exponent_min = ff_ac3_exponent_min_mmxext;
56 }
57 if (EXTERNAL_SSE(cpu_flags)) {
58 c->float_to_fixed24 = ff_float_to_fixed24_sse;
59 }
60 if (EXTERNAL_SSE2(cpu_flags)) {
61 c->ac3_exponent_min = ff_ac3_exponent_min_sse2;
62 c->float_to_fixed24 = ff_float_to_fixed24_sse2;
63 c->compute_mantissa_size = ff_ac3_compute_mantissa_size_sse2;
64 c->extract_exponents = ff_ac3_extract_exponents_sse2;
65 }
66
67 if (EXTERNAL_SSSE3(cpu_flags)) {
68 if (!(cpu_flags & AV_CPU_FLAG_ATOM))
69 c->extract_exponents = ff_ac3_extract_exponents_ssse3;
70 }
71 }
72
73 #define DOWNMIX_FUNC_OPT(ch, opt) \
74 void ff_ac3_downmix_ ## ch ## _to_1_ ## opt(float **samples, \
75 float **matrix, int len); \
76 void ff_ac3_downmix_ ## ch ## _to_2_ ## opt(float **samples, \
77 float **matrix, int len);
78
79 #define DOWNMIX_FUNCS(opt) \
80 DOWNMIX_FUNC_OPT(3, opt) \
81 DOWNMIX_FUNC_OPT(4, opt) \
82 DOWNMIX_FUNC_OPT(5, opt) \
83 DOWNMIX_FUNC_OPT(6, opt)
84
85 DOWNMIX_FUNCS(sse)
DOWNMIX_FUNCS(avx)86 DOWNMIX_FUNCS(avx)
87 DOWNMIX_FUNCS(fma3)
88
89 void ff_ac3dsp_set_downmix_x86(AC3DSPContext *c)
90 {
91 int cpu_flags = av_get_cpu_flags();
92
93 #define SET_DOWNMIX(ch, suf, SUF) \
94 if (ch == c->in_channels) { \
95 if (EXTERNAL_ ## SUF (cpu_flags)) { \
96 if (c->out_channels == 1) \
97 c->downmix = ff_ac3_downmix_ ## ch ## _to_1_ ## suf; \
98 else \
99 c->downmix = ff_ac3_downmix_ ## ch ## _to_2_ ## suf; \
100 } \
101 }
102
103 #define SET_DOWNMIX_ALL(suf, SUF) \
104 SET_DOWNMIX(3, suf, SUF) \
105 SET_DOWNMIX(4, suf, SUF) \
106 SET_DOWNMIX(5, suf, SUF) \
107 SET_DOWNMIX(6, suf, SUF)
108
109 SET_DOWNMIX_ALL(sse, SSE)
110 if (!(cpu_flags & AV_CPU_FLAG_AVXSLOW)) {
111 SET_DOWNMIX_ALL(avx, AVX)
112 SET_DOWNMIX_ALL(fma3, FMA3)
113 }
114 }
115