FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
dcadsp_init.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2012-2014 Christophe Gisquet <christophe.gisquet@gmail.com>
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 #include "libavutil/attributes.h"
22 #include "libavutil/cpu.h"
23 #include "libavutil/x86/cpu.h"
24 #include "libavcodec/dcadsp.h"
25 
26 void ff_decode_hf_sse(float dst[DCA_SUBBANDS][8], const int vq_num[DCA_SUBBANDS],
27  const int8_t hf_vq[1024][32], intptr_t vq_offset,
28  int scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end);
29 void ff_decode_hf_sse2(float dst[DCA_SUBBANDS][8], const int vq_num[DCA_SUBBANDS],
30  const int8_t hf_vq[1024][32], intptr_t vq_offset,
31  int scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end);
32 void ff_decode_hf_sse4(float dst[DCA_SUBBANDS][8], const int vq_num[DCA_SUBBANDS],
33  const int8_t hf_vq[1024][32], intptr_t vq_offset,
34  int scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end);
35 void ff_dca_lfe_fir0_sse(float *out, const float *in, const float *coefs);
36 void ff_dca_lfe_fir1_sse(float *out, const float *in, const float *coefs);
37 void ff_dca_lfe_fir0_fma3(float *out, const float *in, const float *coefs);
38 
40 {
41  int cpu_flags = av_get_cpu_flags();
42 
43  if (EXTERNAL_SSE(cpu_flags)) {
44 #if ARCH_X86_32
46 #endif
49  }
50 
51  if (EXTERNAL_SSE2(cpu_flags)) {
53  }
54 
55  if (EXTERNAL_SSE4(cpu_flags)) {
57  }
58 
59  if (EXTERNAL_FMA3(cpu_flags)) {
61  }
62 }
63 
64 
65 #define SYNTH_FILTER_FUNC(opt) \
66 void ff_synth_filter_inner_##opt(float *synth_buf_ptr, float synth_buf2[32], \
67  const float window[512], \
68  float out[32], intptr_t offset, float scale); \
69 static void synth_filter_##opt(FFTContext *imdct, \
70  float *synth_buf_ptr, int *synth_buf_offset, \
71  float synth_buf2[32], const float window[512], \
72  float out[32], const float in[32], float scale) \
73 { \
74  float *synth_buf= synth_buf_ptr + *synth_buf_offset; \
75  \
76  imdct->imdct_half(imdct, synth_buf, in); \
77  \
78  ff_synth_filter_inner_##opt(synth_buf, synth_buf2, window, \
79  out, *synth_buf_offset, scale); \
80  \
81  *synth_buf_offset = (*synth_buf_offset - 32) & 511; \
82 } \
83 
84 #if HAVE_YASM
85 #if ARCH_X86_32
87 #endif
91 #endif /* HAVE_YASM */
92 
94 {
95 #if HAVE_YASM
96  int cpu_flags = av_get_cpu_flags();
97 
98 #if ARCH_X86_32
99  if (EXTERNAL_SSE(cpu_flags)) {
100  s->synth_filter_float = synth_filter_sse;
101  }
102 #endif
103  if (EXTERNAL_SSE2(cpu_flags)) {
104  s->synth_filter_float = synth_filter_sse2;
105  }
106  if (EXTERNAL_AVX_FAST(cpu_flags)) {
107  s->synth_filter_float = synth_filter_avx;
108  }
109  if (EXTERNAL_FMA3(cpu_flags) && !(cpu_flags & AV_CPU_FLAG_AVXSLOW)) {
110  s->synth_filter_float = synth_filter_fma3;
111  }
112 #endif /* HAVE_YASM */
113 }
const char * s
Definition: avisynth_c.h:631
void ff_decode_hf_sse(float dst[DCA_SUBBANDS][8], const int vq_num[DCA_SUBBANDS], const int8_t hf_vq[1024][32], intptr_t vq_offset, int scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end)
void(* lfe_fir[2])(float *out, const float *in, const float *coefs)
Definition: dcadsp.h:28
static int sse(MpegEncContext *s, uint8_t *src1, uint8_t *src2, int w, int h, int stride)
#define EXTERNAL_SSE(flags)
Definition: cpu.h:55
av_cold void ff_synth_filter_init_x86(SynthFilterContext *s)
Definition: dcadsp_init.c:93
Macro definitions for various function/variable attributes.
#define EXTERNAL_SSE4(flags)
Definition: cpu.h:63
#define av_cold
Definition: attributes.h:74
static av_cold int end(AVCodecContext *avctx)
Definition: avrndec.c:67
void ff_dca_lfe_fir0_fma3(float *out, const float *in, const float *coefs)
void ff_decode_hf_sse2(float dst[DCA_SUBBANDS][8], const int vq_num[DCA_SUBBANDS], const int8_t hf_vq[1024][32], intptr_t vq_offset, int scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end)
#define EXTERNAL_AVX_FAST(flags)
Definition: cpu.h:66
#define EXTERNAL_SSE2(flags)
Definition: cpu.h:56
void(* decode_hf)(float dst[DCA_SUBBANDS][8], const int32_t vq_num[DCA_SUBBANDS], const int8_t hf_vq[1024][32], intptr_t vq_offset, int32_t scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end)
Definition: dcadsp.h:35
void ff_dca_lfe_fir1_sse(float *out, const float *in, const float *coefs)
#define AV_CPU_FLAG_AVXSLOW
AVX supported, but slow when using YMM registers (e.g. Bulldozer)
Definition: cpu.h:46
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(constint16_t *) pi >>8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(constint32_t *) pi >>24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(constfloat *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(constfloat *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(constfloat *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(constdouble *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(constdouble *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(constdouble *) pi *(1U<< 31))))#defineSET_CONV_FUNC_GROUP(ofmt, ifmt) staticvoidset_generic_function(AudioConvert *ac){}voidff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enumAVSampleFormatout_fmt, enumAVSampleFormatin_fmt, intchannels, intsample_rate, intapply_map){AudioConvert *ac;intin_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) returnNULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt)>2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);returnNULL;}returnac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}elseif(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;elseac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);returnac;}intff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){intuse_generic=1;intlen=in->nb_samples;intp;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%dsamples-audio_convert:%sto%s(dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));returnff_convert_dither(ac-> in
int av_get_cpu_flags(void)
Return the flags which specify extensions supported by the CPU.
Definition: cpu.c:76
#define SYNTH_FILTER_FUNC(opt)
Definition: dcadsp_init.c:65
void ff_dca_lfe_fir0_sse(float *out, const float *in, const float *coefs)
void(* synth_filter_float)(FFTContext *imdct, float *synth_buf_ptr, int *synth_buf_offset, float synth_buf2[32], const float window[512], float out[32], const float in[32], float scale)
Definition: synth_filter.h:27
#define DCA_SUBBANDS
Definition: dcadsp.h:25
#define EXTERNAL_FMA3(flags)
Definition: cpu.h:69
av_cold void ff_dcadsp_init_x86(DCADSPContext *s)
Definition: dcadsp_init.c:39
void ff_decode_hf_sse4(float dst[DCA_SUBBANDS][8], const int vq_num[DCA_SUBBANDS], const int8_t hf_vq[1024][32], intptr_t vq_offset, int scale[DCA_SUBBANDS][2], intptr_t start, intptr_t end)
uint8_t pi<< 24) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0f/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_U8, uint8_t,(*(constuint8_t *) pi-0x80)*(1.0/(1<< 7))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S16, int16_t,(*(constint16_t *) pi >>8)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0f/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S16, int16_t,*(constint16_t *) pi *(1.0/(1<< 15))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_S32, int32_t,(*(constint32_t *) pi >>24)+0x80) CONV_FUNC_GROUP(AV_SAMPLE_FMT_FLT, float, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0f/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_DBL, double, AV_SAMPLE_FMT_S32, int32_t,*(constint32_t *) pi *(1.0/(1U<< 31))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_FLT, float, av_clip_uint8(lrintf(*(constfloat *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_FLT, float, av_clip_int16(lrintf(*(constfloat *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_FLT, float, av_clipl_int32(llrintf(*(constfloat *) pi *(1U<< 31)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_U8, uint8_t, AV_SAMPLE_FMT_DBL, double, av_clip_uint8(lrint(*(constdouble *) pi *(1<< 7))+0x80)) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S16, int16_t, AV_SAMPLE_FMT_DBL, double, av_clip_int16(lrint(*(constdouble *) pi *(1<< 15)))) CONV_FUNC_GROUP(AV_SAMPLE_FMT_S32, int32_t, AV_SAMPLE_FMT_DBL, double, av_clipl_int32(llrint(*(constdouble *) pi *(1U<< 31))))#defineSET_CONV_FUNC_GROUP(ofmt, ifmt) staticvoidset_generic_function(AudioConvert *ac){}voidff_audio_convert_free(AudioConvert **ac){if(!*ac) return;ff_dither_free(&(*ac) ->dc);av_freep(ac);}AudioConvert *ff_audio_convert_alloc(AVAudioResampleContext *avr, enumAVSampleFormatout_fmt, enumAVSampleFormatin_fmt, intchannels, intsample_rate, intapply_map){AudioConvert *ac;intin_planar, out_planar;ac=av_mallocz(sizeof(*ac));if(!ac) returnNULL;ac->avr=avr;ac->out_fmt=out_fmt;ac->in_fmt=in_fmt;ac->channels=channels;ac->apply_map=apply_map;if(avr->dither_method!=AV_RESAMPLE_DITHER_NONE &&av_get_packed_sample_fmt(out_fmt)==AV_SAMPLE_FMT_S16 &&av_get_bytes_per_sample(in_fmt)>2){ac->dc=ff_dither_alloc(avr, out_fmt, in_fmt, channels, sample_rate, apply_map);if(!ac->dc){av_free(ac);returnNULL;}returnac;}in_planar=ff_sample_fmt_is_planar(in_fmt, channels);out_planar=ff_sample_fmt_is_planar(out_fmt, channels);if(in_planar==out_planar){ac->func_type=CONV_FUNC_TYPE_FLAT;ac->planes=in_planar?ac->channels:1;}elseif(in_planar) ac->func_type=CONV_FUNC_TYPE_INTERLEAVE;elseac->func_type=CONV_FUNC_TYPE_DEINTERLEAVE;set_generic_function(ac);if(ARCH_AARCH64) ff_audio_convert_init_aarch64(ac);if(ARCH_ARM) ff_audio_convert_init_arm(ac);if(ARCH_X86) ff_audio_convert_init_x86(ac);returnac;}intff_audio_convert(AudioConvert *ac, AudioData *out, AudioData *in){intuse_generic=1;intlen=in->nb_samples;intp;if(ac->dc){av_log(ac->avr, AV_LOG_TRACE,"%dsamples-audio_convert:%sto%s(dithered)\n", len, av_get_sample_fmt_name(ac->in_fmt), av_get_sample_fmt_name(ac->out_fmt));returnff_convert_dither(ac-> out
void INT64 start
Definition: avisynth_c.h:553