Statistics
| Branch: | Revision:

ffmpeg / libavcodec / x86 / ac3dsp_mmx.c @ 2912e87a

History | View | Annotate | Download (3.43 KB)

1
/*
2
 * x86-optimized AC-3 DSP utils
3
 * Copyright (c) 2011 Justin Ruggles
4
 *
5
 * This file is part of Libav.
6
 *
7
 * Libav 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
 * Libav 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 Libav; 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/x86_cpu.h"
23
#include "dsputil_mmx.h"
24
#include "libavcodec/ac3dsp.h"
25

    
26
extern void ff_ac3_exponent_min_mmx   (uint8_t *exp, int num_reuse_blocks, int nb_coefs);
27
extern void ff_ac3_exponent_min_mmxext(uint8_t *exp, int num_reuse_blocks, int nb_coefs);
28
extern void ff_ac3_exponent_min_sse2  (uint8_t *exp, int num_reuse_blocks, int nb_coefs);
29

    
30
extern int ff_ac3_max_msb_abs_int16_mmx   (const int16_t *src, int len);
31
extern int ff_ac3_max_msb_abs_int16_mmxext(const int16_t *src, int len);
32
extern int ff_ac3_max_msb_abs_int16_sse2  (const int16_t *src, int len);
33
extern int ff_ac3_max_msb_abs_int16_ssse3 (const int16_t *src, int len);
34

    
35
extern void ff_ac3_lshift_int16_mmx (int16_t *src, unsigned int len, unsigned int shift);
36
extern void ff_ac3_lshift_int16_sse2(int16_t *src, unsigned int len, unsigned int shift);
37

    
38
extern void ff_ac3_rshift_int32_mmx (int32_t *src, unsigned int len, unsigned int shift);
39
extern void ff_ac3_rshift_int32_sse2(int32_t *src, unsigned int len, unsigned int shift);
40

    
41
extern void ff_float_to_fixed24_3dnow(int32_t *dst, const float *src, unsigned int len);
42
extern void ff_float_to_fixed24_sse  (int32_t *dst, const float *src, unsigned int len);
43
extern void ff_float_to_fixed24_sse2 (int32_t *dst, const float *src, unsigned int len);
44

    
45
av_cold void ff_ac3dsp_init_x86(AC3DSPContext *c, int bit_exact)
46
{
47
    int mm_flags = av_get_cpu_flags();
48

    
49
#if HAVE_YASM
50
    if (mm_flags & AV_CPU_FLAG_MMX) {
51
        c->ac3_exponent_min = ff_ac3_exponent_min_mmx;
52
        c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmx;
53
        c->ac3_lshift_int16 = ff_ac3_lshift_int16_mmx;
54
        c->ac3_rshift_int32 = ff_ac3_rshift_int32_mmx;
55
    }
56
    if (mm_flags & AV_CPU_FLAG_3DNOW && HAVE_AMD3DNOW) {
57
        if (!bit_exact) {
58
            c->float_to_fixed24 = ff_float_to_fixed24_3dnow;
59
        }
60
    }
61
    if (mm_flags & AV_CPU_FLAG_MMX2 && HAVE_MMX2) {
62
        c->ac3_exponent_min = ff_ac3_exponent_min_mmxext;
63
        c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_mmxext;
64
    }
65
    if (mm_flags & AV_CPU_FLAG_SSE && HAVE_SSE) {
66
        c->float_to_fixed24 = ff_float_to_fixed24_sse;
67
    }
68
    if (mm_flags & AV_CPU_FLAG_SSE2 && HAVE_SSE) {
69
        c->ac3_exponent_min = ff_ac3_exponent_min_sse2;
70
        c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_sse2;
71
        c->float_to_fixed24 = ff_float_to_fixed24_sse2;
72
        if (!(mm_flags & AV_CPU_FLAG_SSE2SLOW)) {
73
            c->ac3_lshift_int16 = ff_ac3_lshift_int16_sse2;
74
            c->ac3_rshift_int32 = ff_ac3_rshift_int32_sse2;
75
        }
76
    }
77
    if (mm_flags & AV_CPU_FLAG_SSSE3 && HAVE_SSSE3) {
78
        c->ac3_max_msb_abs_int16 = ff_ac3_max_msb_abs_int16_ssse3;
79
    }
80
#endif
81
}