FFmpeg
 All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Macros Groups Pages
vf_gradfun_init.c
Go to the documentation of this file.
1 /*
2  * Copyright (C) 2009 Loren Merritt <lorenm@u.washington.edu>
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 "config.h"
22 #include "libavutil/attributes.h"
23 #include "libavutil/cpu.h"
24 #include "libavutil/mem.h"
25 #include "libavutil/x86/cpu.h"
26 #include "libavfilter/gradfun.h"
27 
28 void ff_gradfun_filter_line_mmxext(intptr_t x, uint8_t *dst, const uint8_t *src,
29  const uint16_t *dc, int thresh,
30  const uint16_t *dithers);
31 void ff_gradfun_filter_line_ssse3(intptr_t x, uint8_t *dst, const uint8_t *src,
32  const uint16_t *dc, int thresh,
33  const uint16_t *dithers);
34 
35 void ff_gradfun_blur_line_movdqa_sse2(intptr_t x, uint16_t *buf,
36  const uint16_t *buf1, uint16_t *dc,
37  const uint8_t *src1, const uint8_t *src2);
38 void ff_gradfun_blur_line_movdqu_sse2(intptr_t x, uint16_t *buf,
39  const uint16_t *buf1, uint16_t *dc,
40  const uint8_t *src1, const uint8_t *src2);
41 
42 #if HAVE_X86ASM
43 static void gradfun_filter_line_mmxext(uint8_t *dst, const uint8_t *src,
44  const uint16_t *dc,
45  int width, int thresh,
46  const uint16_t *dithers)
47 {
48  intptr_t x;
49  if (width & 3) {
50  x = width & ~3;
51  ff_gradfun_filter_line_c(dst + x, src + x, dc + x / 2,
52  width - x, thresh, dithers);
53  width = x;
54  }
55  x = -width;
56  ff_gradfun_filter_line_mmxext(x, dst + width, src + width, dc + width / 2,
57  thresh, dithers);
58 }
59 
60 static void gradfun_filter_line_ssse3(uint8_t *dst, const uint8_t *src, const uint16_t *dc,
61  int width, int thresh,
62  const uint16_t *dithers)
63 {
64  intptr_t x;
65  if (width & 7) {
66  // could be 10% faster if I somehow eliminated this
67  x = width & ~7;
68  ff_gradfun_filter_line_c(dst + x, src + x, dc + x / 2,
69  width - x, thresh, dithers);
70  width = x;
71  }
72  x = -width;
73  ff_gradfun_filter_line_ssse3(x, dst + width, src + width, dc + width / 2,
74  thresh, dithers);
75 }
76 
77 static void gradfun_blur_line_sse2(uint16_t *dc, uint16_t *buf, const uint16_t *buf1,
78  const uint8_t *src, int src_linesize, int width)
79 {
80  intptr_t x = -2 * width;
81  if (((intptr_t) src | src_linesize) & 15)
82  ff_gradfun_blur_line_movdqu_sse2(x, buf + width, buf1 + width,
83  dc + width, src + width * 2,
84  src + width * 2 + src_linesize);
85  else
86  ff_gradfun_blur_line_movdqa_sse2(x, buf + width, buf1 + width,
87  dc + width, src + width * 2,
88  src + width * 2 + src_linesize);
89 }
90 #endif /* HAVE_X86ASM */
91 
93 {
94 #if HAVE_X86ASM
96 
97  if (EXTERNAL_MMXEXT(cpu_flags))
98  gf->filter_line = gradfun_filter_line_mmxext;
99  if (EXTERNAL_SSSE3(cpu_flags))
100  gf->filter_line = gradfun_filter_line_ssse3;
101 
102  if (EXTERNAL_SSE2(cpu_flags))
103  gf->blur_line = gradfun_blur_line_sse2;
104 #endif /* HAVE_X86ASM */
105 }
Memory handling functions.
Holds instance-specific information for gradfun.
Definition: gradfun.h:28
void ff_gradfun_blur_line_movdqu_sse2(intptr_t x, uint16_t *buf, const uint16_t *buf1, uint16_t *dc, const uint8_t *src1, const uint8_t *src2)
static atomic_int cpu_flags
Definition: cpu.c:50
void ff_gradfun_filter_line_ssse3(intptr_t x, uint8_t *dst, const uint8_t *src, const uint16_t *dc, int thresh, const uint16_t *dithers)
#define src
Definition: vp8dsp.c:254
Macro definitions for various function/variable attributes.
uint8_t
#define av_cold
Definition: attributes.h:82
#define EXTERNAL_SSE2(flags)
Definition: cpu.h:59
uint16_t width
Definition: gdv.c:47
void ff_gradfun_blur_line_movdqa_sse2(intptr_t x, uint16_t *buf, const uint16_t *buf1, uint16_t *dc, const uint8_t *src1, const uint8_t *src2)
static const uint8_t dithers[8][8][8]
void(* blur_line)(uint16_t *dc, uint16_t *buf, const uint16_t *buf1, const uint8_t *src, int src_linesize, int width)
Definition: gradfun.h:39
void ff_gradfun_filter_line_mmxext(intptr_t x, uint8_t *dst, const uint8_t *src, const uint16_t *dc, int thresh, const uint16_t *dithers)
void ff_gradfun_filter_line_c(uint8_t *dst, const uint8_t *src, const uint16_t *dc, int width, int thresh, const uint16_t *dithers)
Definition: vf_gradfun.c:57
void(* filter_line)(uint8_t *dst, const uint8_t *src, const uint16_t *dc, int width, int thresh, const uint16_t *dithers)
DSP functions.
Definition: gradfun.h:38
#define src1
Definition: h264pred.c:139
void * buf
Definition: avisynth_c.h:690
#define EXTERNAL_SSSE3(flags)
Definition: cpu.h:65
int av_get_cpu_flags(void)
Return the flags which specify extensions supported by the CPU.
Definition: cpu.c:92
#define EXTERNAL_MMXEXT(flags)
Definition: cpu.h:57
av_cold void ff_gradfun_init_x86(GradFunContext *gf)
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-> dc