ffmpeg configure配置选项

用于查看ffmpeg的能力

选项	说明
	print this message
	显示所有可用的解码器（h264/mjpeg等）
	显示所有可用的编码器（h264/mjpeg等）
	显示所有支持的硬编解码器（h264_videotoolbox/h264_mediacodec等）
	显示所有支持解复用的容器（mp4/h264等）
	显示所有支持复用的容器（mp4/h264等）
	show all available parsers
	显示所有支持的传输协议（rtmp/rtp等）
	显示所有可用的格式转换（h264_mp4toannexb/aac_adtstoasc等）
	显示所有支持的输入设备（alsa/v4l2等）
	显示所有支持的输出设备（alsa/opengl等）
	显示支持的所有过滤器（scale/volume/fps/allyuv等）

编译配置

选项	说明
	配置过程中的log输出文件，默认输出到当前位置的ffbuild/config.log文件
	配置过程中不输出日志
	把配置过程中的任何警告当做致命的错误处理
	设定安装的跟目录，如果不指定默认是/usr
	设置可执行程序的安装位置，默认是[PREFIX/bin]
	设置测试程序以及数据的安装位置，默认是[PREFIX/share/ffmpeg]
	设置文档的安装目录，默认是[PREFIX/share/doc/ffmpeg]
	设置静态库的安装位置，默认是[PREFIX/lib]
	设置动态库的安装位置，默认是[LIBDIR]
	设置头文件的安装位置，默认是[PREFIX/include]；一般来说用于依赖此头文件来开发就够了
	设置man文件的安装目录，默认是[PREFIX/share/man]
	设置pkgconfig的安装目录，默认是[LIBDIR/pkgconfig]，只有`--enable-shared`使能的时候这个选项才有效
	use rpath to allow installing libraries in paths not part of the dynamic linker search path use rpath when linking programs [USE WITH CARE]
	Darwin directory name for installed targets

选择许可证，FFMPEG默认许可证LGPL 2.1，如果需要加gpl的库需要使用gpl的许可证，例如libx264就是gpl的，如果需要加入libx264则需要--enable-gpl。

选项	说明
	allow use of GPL code, the resulting libs and binaries will be under GPL [no]
	upgrade (L)GPL to version 3 [no]
	allow use of nonfree code, the resulting libs and binaries will be unredistributable [no]

编译选项的配置

选项	说明
	不生产静态库，默认生成静态库
	生成动态库，默认不生成动态库
	optimize for size instead of speed，默认开启
	disable detecting cpu capabilities at runtime (smaller binary)，默认开启
	enable full grayscale support (slower color)
	disable alpha channel support in swscale
	禁止编译所有库和可执行程序
	增加主版本号

可执行程序开启选项，默认编译ffmpeg中的所有可执行程序，包括ffmpeg、ffplay、ffprobe、ffserver，不过Mac平台默认情况下不生成ffplay，目前暂未知道啥原因。

选项	说明
	do not build command line programs
	disable ffmpeg build
	disable ffplay build
	disable ffprobe build
	disable ffserver build

离线文档选择

选项	说明
	do not build documentation
	do not build HTML documentation pages
	do not build man documentation pages
	do not build POD documentation pages
	do not build text documentation pages

除了avresample模块，默认编译所有模块。一般来说用于轻量化ffmpeg库的大小，可以仅仅开启指定某些组件的某些功能。

选项	说明
	disable libavdevice build
	disable libavcodec build
	disable libavformat build
	disable libswresample build
	disable libswscale build
	disable libpostproc build
	disable libavfilter build
	enable libavresample build [no]
	disable pthreads [autodetect]
	disable Win32 threads [autodetect]
	disable OS/2 threads [autodetect]
	disable network support [no]
	disable DCT code
	disable DWT code
	disable error resilience code
	disable LSP code
	disable LZO decoder code
	disable MDCT code
	disable RDFT code
	disable FFT code
	disable floating point AAN (I)DCT code
	disable pixel utils in libavutil

可以用于设定开启指定功能，例如禁止所有encoders，在这里可以开启特定的encoders（x264、aac等）

选项	说明
	disable all components listed below
	disable encoder NAME
	enable encoder NAME
	disable all encoders
	disable decoder NAME
	enable decoder NAME
	disable all decoders
	disable hwaccel NAME
	enable hwaccel NAME
	disable all hwaccels
	disable muxer NAME
	enable muxer NAME
	disable all muxers
	disable demuxer NAME
	enable demuxer NAME
	disable all demuxers
	enable parser NAME
	disable parser NAME
	disable all parsers
	enable bitstream filter NAME
	disable bitstream filter NAME
	disable all bitstream filters
	enable protocol NAME
	disable protocol NAME
	disable all protocols
	enable input device NAME
	disable input device NAME
	disable input devices
	enable output device NAME
	disable output device NAME
	disable output devices
	disable all devices
	enable filter NAME
	disable filter NAME
	disable all filters

ffmpeg提供的一些功能是由其他扩展库支持的，如果需要使用需要明确声明，确定编译的第三方库的目标架构--arch相同就好了，在编译ffmpeg的时候需加入第三方库的头文和库搜索路径（通过extra-cflags和extra-ldflags指定即可），剩下的事ffmpeg都给你做好了。

ffmpeg集成libx264官方文档

 $ git clone http://source.ffmpeg.org/git/ffmpeg.git $ cd x264 $ ./configure --prefix=$FFMPEG_PREFIX --enable-static --enable-shared $ make -j8 && make install

选项	说明
	enable reading of AviSynth script files [no]
	disable bzlib [autodetect]
	enable audio fingerprinting with chromaprint [no]
	enable frei0r video filtering [no]
	enable gcrypt, needed for rtmp(t)e support if openssl, librtmp or gmp is not used [no]
	enable gmp, needed for rtmp(t)e support if openssl or librtmp is not used [no]
	enable gnutls, needed for https support if openssl is not used [no]
	disable iconv [autodetect]
	enable JNI support [no]
	enable LADSPA audio filtering [no]
	enable libass subtitles rendering, needed for subtitles and ass filter [no]
	enable BluRay reading using libbluray [no]
	enable bs2b DSP library [no]
	enable textual display using libcaca [no]
	enable CELT decoding via libcelt [no]
	enable audio CD grabbing with libcdio [no]
	enable IIDC-1394 grabbing using libdc1394 and libraw1394 [no]
	enable libebur128 for EBU R128 measurement, needed for loudnorm filter [no]
	enable AAC de/encoding via libfdk-aac [no]
	enable flite (voice synthesis) support via libflite [no]
	enable libfontconfig, useful for drawtext filter [no]
	enable libfreetype, needed for drawtext filter [no]
	enable libfribidi, improves drawtext filter [no]
	enable Game Music Emu via libgme [no]
	enable GSM de/encoding via libgsm [no]
	enable iec61883 via libiec61883 [no]
	enable iLBC de/encoding via libilbc [no]
	enable HEVC encoding via libkvazaar [no]
	enable ModPlug via libmodplug [no]
	enable MP3 encoding via libmp3lame [no]
	enable NUT (de)muxing via libnut, native (de)muxer exists [no]
	enable AMR-NB de/encoding via libopencore-amrnb [no]
	enable AMR-WB decoding via libopencore-amrwb [no]
	enable video filtering via libopencv [no]
	enable H.264 encoding via OpenH264 [no]
	enable JPEG 2000 de/encoding via OpenJPEG [no]
	enable decoding tracked files via libopenmpt [no]
	enable Opus de/encoding via libopus [no]
	enable Pulseaudio input via libpulse [no]
	enable rubberband needed for rubberband filter [no]
	enable RTMP[E] support via librtmp [no]
	enable Dirac de/encoding via libschroedinger [no]
	enable fixed-point MP3 encoding via libshine [no]
	enable Samba protocol via libsmbclient [no]
	enable Snappy compression, needed for hap encoding [no]
	enable Include libsoxr resampling [no]
	enable Speex de/encoding via libspeex [no]
	enable SFTP protocol via libssh [no]
	enable Tesseract, needed for ocr filter [no]
	enable Theora encoding via libtheora [no]
	enable MP2 encoding via libtwolame [no]
	enable libv4l2/v4l-utils [no]
	enable video stabilization using vid.stab [no]
	enable AMR-WB encoding via libvo-amrwbenc [no]
	enable Vorbis en/decoding via libvorbis, native implementation exists [no]
	enable VP8 and VP9 de/encoding via libvpx [no]
	enable wavpack encoding via libwavpack [no]
	enable WebP encoding via libwebp [no]
	enable H.264 encoding via x264 [no]
	enable HEVC encoding via x265 [no]
	enable AVS encoding via xavs [no]
	enable X11 grabbing using XCB [autodetect]
	enable X11 grabbing shm communication [autodetect]
	enable X11 grabbing mouse rendering [autodetect]
	enable X11 grabbing shape rendering [autodetect]
	enable Xvid encoding via xvidcore, native MPEG-4/Xvid encoder exists [no]
	enable z.lib, needed for zscale filter [no]
	enable message passing via libzmq [no]
	enable teletext support via libzvbi [no]
	disable lzma [autodetect]
	enable Blackmagic DeckLink I/O support [no]
	enable Android MediaCodec support [no]
	enable NetCDF, needed for sofalizer filter [no]
	enable OpenAL 1.1 capture support [no]
	enable OpenCL code
	enable OpenGL rendering [no]
	enable openssl, needed for https support if gnutls is not used [no]
	disable SChannel SSP, needed for TLS support on Windows if openssl and gnutls are not used [autodetect]
	disable sdl2 [autodetect]
	disable Secure Transport, needed for TLS support on OSX if openssl and gnutls are not used [autodetect]
	enable X11 grabbing (legacy) [no]
	disable xlib [autodetect]
	disable zlib [autodetect]

ffmpeg默认实现了移动端（Android和IOS）的硬编解码，可以选择disable的都是默认开启的，可以关闭，可以选择enable的都是需要自己解决依赖的。

选项	说明
	disable Apple AudioToolbox code [autodetect]
	enable dynamically linked Nvidia CUDA code [no]
	enable Nvidia CUVID support [autodetect]
	disable Microsoft Direct3D 11 video acceleration code [autodetect]
	disable Microsoft DirectX 9 video acceleration code [autodetect]
	enable Intel MediaSDK (AKA Quick Sync Video) code via libmfx [no]
	enable Nvidia Performance Primitives-based code [no]
	enable Broadcom Multi-Media Abstraction Layer (Raspberry Pi) via MMAL [no]
	disable Nvidia video encoding code [autodetect]
	enable OpenMAX IL code [no]
	enable OpenMAX IL code for Raspberry Pi [no]
	disable Video Acceleration API (mainly Unix/Intel) code [autodetect]
	disable Apple Video Decode Acceleration code [autodetect]
	disable Nvidia Video Decode and Presentation API for Unix code [autodetect]
	disable VideoToolbox code [autodetect]

ffmpeg代码本身是支持跨平台的，要编译不同的平台需要配置不同平台的交叉编译工具链。ffmpeg都是c代码，所以不需要配置c++的sysroot。常用的就几个arch,cpu,cross-prefix,enable-cross-compile,sysroot,target-os,extra-cflags,extra-ldflags,enable-pic。现在Android和IOS几乎没有armv5的设备了，所以如果编译这两个平台配置armv7和armv8就好了。

选项	说明
	选择目标架构[armv7a/aarch64/x86/x86_64等]
	选择目标cpu[armv7-a/armv8-a/x86/x86_64]
	设定交叉编译工具链的前缀,不算gcc/nm/as命令，例如android 32位的交叉编译链`$ndk_dir/toolchains/arm-linux-androideabi-$toolchain_version/prebuilt/linux-$host_arch/bin/arm-linux-androideabi-`
	program name suffix []
	如果目标平台和编译平台不同则需要使能它
	交叉工具链的头文件和库位，例如Android 32位位置`$ndk_dir/platforms/android-14/arch-arm`
	location of cross-build system headers
	设置目标系统
	command to run executables on target
	path to view of build directory on target
	path to samples directory on target
	force fixed dir/prefix instead of mktemp for checks
	set tool defaults according to NAME
	use nm tool NM [nm -g]
	use archive tool AR [ar]
	use assembler AS []
	use symbolic link tool LN_S [ln -s -f]
	use strip tool STRIP [strip]
	use windows resource compiler WINDRES [windres]
	use yasm-compatible assembler EXE [yasm]
	use C compiler CC [gcc]
	use C compiler CXX [g++]
	use ObjC compiler OCC [gcc]
	use dependency generator DEPCC [gcc]
	use linker LD []
	use pkg-config tool PKGCONFIG [pkg-config]
	pass additional flags to pkgconf []
	use ranlib RANLIB [ranlib]
	use DOXYGEN to generate API doc [doxygen]
	use host C compiler HOSTCC
	use HCFLAGS when compiling for host
	use HCPPFLAGS when compiling for host
	use host linker HOSTLD
	use HLDFLAGS when linking for host
	use libs HLIBS when linking for host
	compiler host OS []
	设置cflags，如果是Android平台可以根据ndk内的设定,arm-linux-androideabi-4.6/setup.mk，建议参考你当前的setup来配置
	add ECFLAGS to CXXFLAGS []
	add FLAGS to OBJCFLAGS []
	参考cflags
	add ELDFLAGS to LDEXEFLAGS []
	add ELDFLAGS to LDLIBFLAGS []
	add ELIBS []
	version string suffix []
	override optimization-related compiler flags
	library name suffix []
	build position-independent code
	compile for Thumb instruction set
	use link-time optimization
	override the environment variables

选项	说明
	prefix malloc and related names with PREFIX
	use a supported custom allocator
	disable symbol versioning
	use hardcoded tables instead of runtime generation
	disable buffer boundary checking in bitreaders (faster, but may crash)
	emulate memalign, interferes with memory debuggers
	the max filter size swscale uses [256]

默认开启各个平台的汇编优化，有些嵌入式平台可能并不能完整的支持架构的所有汇编指令，所以需要关闭。（自己理解的，没有实战）

选项	说明
	disable all assembly optimizations
	disable AltiVec optimizations
	disable VSX optimizations
	disable POWER8 optimizations
	disable 3DNow! optimizations
	disable 3DNow! extended optimizations
	disable MMX optimizations
	disable MMXEXT optimizations
	disable SSE optimizations
	disable SSE2 optimizations
	disable SSE3 optimizations
	disable SSSE3 optimizations
	disable SSE4 optimizations
	disable SSE4.2 optimizations
	disable AVX optimizations
	disable XOP optimizations
	disable FMA3 optimizations
	disable FMA4 optimizations
	disable AVX2 optimizations
	disable AESNI optimizations
	disable armv5te optimizations
	disable armv6 optimizations
	disable armv6t2 optimizations
	disable VFP optimizations
	disable NEON optimizations
	disable use of inline assembly
	disable use of nasm/yasm assembly
	disable MIPS DSP ASE R1 optimizations
	disable MIPS DSP ASE R2 optimizations
	disable MSA optimizations
	disable floating point MIPS optimizations
	disable Loongson SIMD optimizations
	consider unaligned accesses slow

调试用的一些开关

选项	说明
	disable debugging symbols
	set the debug level []
	disable compiler optimizations
	enable more compiler warnings
	disable stripping of executables and shared libraries
	0(default), 1 or 2, amount of assertion testing, 2 causes a slowdown at runtime.
	fill heap uninitialized allocated space with arbitrary data

	Trap arithmetic overflows
	location of test samples for FATE, if not set use $FATE_SAMPLES at make invocation time.
	check NEON registers for clobbering (should be used only for debugging purposes)
	check XMM registers for clobbering (Win64-only; should be used only for debugging purposes)
	randomly enable/disable components

	randomly enable/disable specific components or
	component groups. LIST is a comma-separated list of NAME[:PROB] entries where NAME is a component (group) and PROB the probability associated with

https://blog.csdn.net/momo0853/article/details/78043903

文章来源: ffmpeg configure配置选项

标签

ffmpeg