diff --git a/multimedia/ffmpeg4/Makefile b/multimedia/ffmpeg4/Makefile index e5cb726577fb..d288339f368c 100644 --- a/multimedia/ffmpeg4/Makefile +++ b/multimedia/ffmpeg4/Makefile @@ -1,648 +1,644 @@ PORTNAME= ffmpeg -PORTVERSION= 4.4.4 -PORTREVISION= 30 +PORTVERSION= 4.4.6 CATEGORIES= multimedia audio net MASTER_SITES= https://ffmpeg.org/releases/ PKGNAMESUFFIX= 4 -MAINTAINER= multimedia@FreeBSD.org +MAINTAINER= monwarez@mailoo.org COMMENT= Realtime audio/video encoder/converter and streaming server (legacy 4.* series) WWW= https://ffmpeg.org/ LICENSE= GPLv2+ LGPL21+ LICENSE_COMB= multi -DEPRECATED= Outdated and unsupported version, consider migrating to multimedia/ffmpeg -EXPIRATION_DATE=2026-06-30 - BUILD_DEPENDS= ${BUILD_DEPENDS_${ARCH}} BUILD_DEPENDS_aarch64= as:devel/binutils BUILD_DEPENDS_amd64= nasm:devel/nasm BUILD_DEPENDS_armv6= as:devel/binutils BUILD_DEPENDS_armv7= as:devel/binutils BUILD_DEPENDS_i386= nasm:devel/nasm HAS_CONFIGURE= yes CONFIGURE_LOG= ffbuild/config.log USES= compiler:c11 cpe gmake localbase:ldflags perl5 \ pkgconfig shebangfix tar:xz USE_LDCONFIG= yes USE_PERL5= build SHEBANG_FILES= doc/texi2pod.pl NOPRECIOUSMAKEVARS= yes # ARCH .ifdef PKGNAMESUFFIX PORTSCOUT= limit:^4\. PREFIX= ${LOCALBASE}/${PKGBASE} # avoid conflict with the default .endif # Option CHROMAPRINT disabled, it cannot work and people are baffled. OPTIONS_DEFINE= ALSA AMR_NB AMR_WB AOM ARIBB24 ASM ASS BS2B CACA CDIO \ CODEC2 DAV1D DAVS2 DC1394 DEBUG DOCS DRM FDK_AAC FLITE \ FONTCONFIG FREETYPE FREI0R FRIBIDI GLSLANG GME GSM ICONV ILBC \ JACK KLVANC KVAZAAR LADSPA LAME LENSFUN LIBBLURAY LIBRSVG2 \ LIBXML2 LTO LV2 MODPLUG MYSOFA NETWORK NVENC OPENAL OPENCL OPENGL \ OPENH264 OPENJPEG OPENMPT OPENVINO OPTIMIZED_CFLAGS OPUS POCKETSPHINX \ PULSEAUDIO RAV1E RABBITMQ RIST RTCPU RUBBERBAND SDL SMB SNAPPY \ SNDIO SOXR SPEEX SRT SSH SVTAV1 SVTHEVC SVTVP9 TENSORFLOW \ TESSERACT THEORA TWOLAME UAVS3D V4L VAAPI VAPOURSYNTH VDPAU VIDSTAB \ VMAF VO_AMRWBENC VORBIS VPX VULKAN WEBP X264 X265 \ XAVS2 XCB XVID XVIDEO ZIMG ZMQ ZVBI # intel-media-sdk only for i386/amd64 OPTIONS_DEFINE_amd64= MFX OPTIONS_DEFINE_i386= MFX OPTIONS_DEFAULT= AOM ASM ASS DAV1D DRM FONTCONFIG FREETYPE FREI0R GMP \ GNUTLS ICONV LAME LIBXML2 NETWORK OPTIMIZED_CFLAGS OPUS \ RTCPU SVTAV1 V4L VAAPI VDPAU VMAF VORBIS VPX WEBP X264 \ X265 XCB # i386 is too register-starved for LTO (PR257124) OPTIONS_EXCLUDE_i386= LTO .if !exists(${.CURDIR:H:H}/multimedia/davs2) # https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/5985a1bf7233 OPTIONS_EXCLUDE+= DAVS2 .endif .if !exists(${.CURDIR:H:H}/multimedia/uavs3d) # https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/c952db9d68d3 OPTIONS_EXCLUDE+= UAVS3D .endif .if !exists(${.CURDIR:H:H}/multimedia/xavs2) # https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/85a921f4618 OPTIONS_EXCLUDE+= XAVS2 .endif .if !exists(${.CURDIR:H:H}/multimedia/klvanc) # https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/8732dfa9954 OPTIONS_EXCLUDE+= KLVANC .endif .if !exists(${.CURDIR:H:H}/science/openvino) # https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/ff37ebaf30e6 OPTIONS_EXCLUDE+= OPENVINO .endif OPTIONS_RADIO= RTMP SSL OPTIONS_RADIO_RTMP= GCRYPT GMP OPTIONS_RADIO_SSL= GNUTLS MBEDTLS OPENSSL OPTIONS_GROUP= LICENSE OPTIONS_GROUP_LICENSE= GPL3 NONFREE ASS_DESC= Subtitles rendering via libass AOM_DESC= AV1 video encoding/decoding via libaom ARIBB24_DESC= ARIB text and caption decoding via libaribb24 BS2B_DESC= Bauer Stereophonic-to-Binaural filter CHROMAPRINT_DESC= Audio fingerprinting with chromaprint CODEC2_DESC= Codec 2 audio encoding/decoding via libcodec2 DAV1D_DESC= AV1 video decoding via libdav1d DAVS2_DESC= AVS2 decoding via libdavs2 DC1394_DESC= IIDC-1394 grabbing using libdc1394 DRM_DESC= KMS grabbing using libdrm FDK_AAC_DESC= AAC audio encoding via Fraunhofer FDK FLITE_DESC= Voice synthesis support via libflite GLSLANG_DESC= GLSL->SPIRV compilation via libglslang GME_DESC= Game Music Emu demuxer GPL3_DESC= Allow (L)GPL version 3 code(cs) ILBC_DESC= Internet Low Bit Rate codec KLVANC_DESC= Kernel Labs VANC processing KVAZAAR_DESC= H.265 video codec support via Kvazaar LICENSE_DESC= Licensing options LV2_DESC= LV2 audio filtering MFX_DESC= Intel MediaSDK (aka Quick Sync Video) MYSOFA_DESC= SOFAlizer binaural filter NETWORK_DESC= Networking support NONFREE_DESC= Allow use of nonfree code NVENC_DESC= NVIDIA decoder/encoder with CUDA support OPENMPT_DESC= Decoding tracked files via libopenmpt OPENH264_DESC= H.264 video codec support via OpenH264 OPENVINO_DESC= OpenVINO as a DNN module backend for DNN based filters like dnn_processing POCKETSPHINX_DESC= Automatic Speech Recognition via PocketSphinx RAV1E_DESC= AV1 encoding via librav1e RABBITMQ_DESC= AMQP 0-9-1 protocol via RabbitMQ RIST_DESC= Reliable Internet Stream Transport protocol via librist RUBBERBAND_DESC=Time-stretching and pitch-shifting with librubberband RTCPU_DESC= Detect CPU capabilities at runtime RTMP_DESC= RTMP(T)E protocol support SRT_DESC= Haivision SRT protocol via libsrt SVTAV1_DESC= AV1 encoding via SVT-AV1 SVTHEVC_DESC= HEVC encoding via SVT-HEVC SVTVP9_DESC= VP9 encoding via SVT-VP9 TENSORFLOW_DESC=TensorFlow as a DNN module backend for DNN based filters like sr TESSERACT_DESC= Optical Character Recognition via Tesseract THEORA_DESC= Encoding support for theora via libtheora UAVS3D_DESC= AVS3 decoding via libuavs3d VAPOURSYNTH_DESC= VapourSynth demuxer VIDSTAB_DESC= Video stabilization filter VMAF_DESC= VMAF filter via libvmaf VULKAN_DESC= Vulkan filters XAVS2_DESC= AVS2 encoding via libxavs2 XCB_DESC= X11 grabbing using XCB XVID_DESC= Encoding support for MPEG-4 ASP via libxvid ZIMG_DESC= "z" library video scaling filter ZMQ_DESC= Message passing via libzmq${ZMQ_VERSION} ZVBI_DESC= Teletext support via libzvbi OPTIONS_SUB= yes # alsa ALSA_LIB_DEPENDS= libasound.so:audio/alsa-lib ALSA_CONFIGURE_ENABLE= alsa # Opencore AMR NB AMR_NB_LIB_DEPENDS= libopencore-amrnb.so:audio/opencore-amr AMR_NB_CONFIGURE_ENABLE= libopencore-amrnb AMR_NB_IMPLIES= GPL3 # Opencore AMR WB AMR_WB_LIB_DEPENDS= libopencore-amrwb.so:audio/opencore-amr AMR_WB_CONFIGURE_ENABLE= libopencore-amrwb AMR_WB_IMPLIES= GPL3 # aom AOM_LIB_DEPENDS= libaom.so:multimedia/aom AOM_CONFIGURE_ENABLE= libaom # arib ARIBB24_LIB_DEPENDS= libaribb24.so:multimedia/aribb24 ARIBB24_CONFIGURE_ENABLE= libaribb24 # asm support ASM_CONFIGURE_ENABLE= asm # ass ASS_LIB_DEPENDS= libass.so:multimedia/libass ASS_CONFIGURE_ENABLE= libass # bs2b BS2B_LIB_DEPENDS= libbs2b.so:audio/libbs2b BS2B_CONFIGURE_ENABLE= libbs2b # caca CACA_LIB_DEPENDS= libcaca.so:graphics/libcaca CACA_CONFIGURE_ENABLE= libcaca # cdio CDIO_LIB_DEPENDS= libcdio_paranoia.so:sysutils/libcdio-paranoia CDIO_CONFIGURE_ENABLE= libcdio # chromaprint CHROMAPRINT_BROKEN= Dependency loop CHROMAPRINT_LIB_DEPENDS= libchromaprint.so:audio/chromaprint CHROMAPRINT_CONFIGURE_ENABLE= chromaprint # codec2 CODEC2_LIB_DEPENDS= libcodec2.so:audio/codec2 CODEC2_CONFIGURE_ENABLE= libcodec2 # dav1d DAV1D_LIB_DEPENDS= libdav1d.so:multimedia/dav1d DAV1D_CONFIGURE_ENABLE= libdav1d # davs2 DAVS2_LIB_DEPENDS= libdavs2.so:multimedia/davs2 DAVS2_CONFIGURE_ENABLE= libdavs2 # dc1394 DC1394_LIB_DEPENDS= libdc1394.so:multimedia/libdc1394 DC1394_CONFIGURE_ENABLE= libdc1394 # debugging DEBUG_CONFIGURE_ON= --disable-stripping DEBUG_CONFIGURE_OFF= --disable-debug # docs DOCS_BUILD_DEPENDS= texi2html:textproc/texi2html DOCS_CONFIGURE_ENABLE= htmlpages DOCS_BINARY_ALIAS= makeinfo=${FALSE} # force texi2html # drm DRM_LIB_DEPENDS= libdrm.so:graphics/libdrm DRM_CONFIGURE_ENABLE= libdrm # fdk_aac FDK_AAC_LIB_DEPENDS= libfdk-aac.so:audio/fdk-aac FDK_AAC_CONFIGURE_ENABLE= libfdk-aac FDK_AAC_IMPLIES= NONFREE # flite FLITE_LIB_DEPENDS= libflite.so:audio/flite FLITE_CONFIGURE_ENABLE= libflite # fontconfig FONTCONFIG_LIB_DEPENDS= libfontconfig.so:x11-fonts/fontconfig FONTCONFIG_CONFIGURE_ENABLE= fontconfig # freetype FREETYPE_LIB_DEPENDS= libfreetype.so:print/freetype2 FREETYPE_CONFIGURE_ENABLE= libfreetype # frei0r FREI0R_BUILD_DEPENDS= ${LOCALBASE}/include/frei0r.h:graphics/frei0r FREI0R_CONFIGURE_ENABLE= frei0r # fribidi FRIBIDI_LIB_DEPENDS= libfribidi.so:converters/fribidi FRIBIDI_CONFIGURE_ENABLE= libfribidi # gcrypt GCRYPT_LIB_DEPENDS= libgcrypt.so:security/libgcrypt GCRYPT_CONFIGURE_ENABLE= gcrypt # glslang GLSLANG_BUILD_DEPENDS= glslang>0:graphics/glslang \ spirv-tools>0:graphics/spirv-tools GLSLANG_CONFIGURE_ENABLE= libglslang GLSLANG_LDFLAGS= -lm GLSLANG_IMPLIES= VULKAN # gnutls GNUTLS_LIB_DEPENDS= libgnutls.so:security/gnutls GNUTLS_CONFIGURE_ENABLE= gnutls # gme GME_LIB_DEPENDS= libgme.so:audio/libgme GME_CONFIGURE_ENABLE= libgme # gmp GMP_LIB_DEPENDS= libgmp.so:math/gmp GMP_CONFIGURE_ENABLE= gmp GMP_IMPLIES= GPL3 # gsm GSM_LIB_DEPENDS= libgsm.so:audio/gsm GSM_CONFIGURE_ENABLE= libgsm # iconv ICONV_USES= iconv ICONV_CONFIGURE_ENABLE= iconv # ilbc ILBC_LIB_DEPENDS= libilbc.so:net/libilbc ILBC_CONFIGURE_ENABLE= libilbc # jack JACK_LIB_DEPENDS= libjack.so:audio/jack JACK_CONFIGURE_ENABLE= libjack # klvanc KLVANC_LIB_DEPENDS= libklvanc.so:multimedia/libklvanc KLVANC_CONFIGURE_ENABLE= libklvanc # kvazaar KVAZAAR_LIB_DEPENDS= libkvazaar.so:multimedia/kvazaar KVAZAAR_CONFIGURE_ENABLE= libkvazaar # ladspa LADSPA_BUILD_DEPENDS= ${LOCALBASE}/include/ladspa.h:audio/ladspa LADSPA_RUN_DEPENDS= ${LOCALBASE}/lib/ladspa/amp.so:audio/ladspa LADSPA_CONFIGURE_ENABLE= ladspa # lame LAME_LIB_DEPENDS= libmp3lame.so:audio/lame LAME_CONFIGURE_ENABLE= libmp3lame # lensfun LENSFUN_LIB_DEPENDS= liblensfun.so:graphics/lensfun LENSFUN_CONFIGURE_ENABLE= liblensfun # libbluray LIBBLURAY_LIB_DEPENDS= libbluray.so:multimedia/libbluray LIBBLURAY_CONFIGURE_ENABLE= libbluray # librsvg LIBRSVG2_USES= gnome LIBRSVG2_USE= GNOME=cairo,librsvg2 LIBRSVG2_CONFIGURE_ENABLE= librsvg # libxml LIBXML2_LIB_DEPENDS= libxml2.so:textproc/libxml2 LIBXML2_CONFIGURE_ENABLE= libxml2 # lto LTO_CONFIGURE_ENABLE= lto # lv2 LV2_BUILD_DEPENDS= lv2>0:audio/lv2 LV2_LIB_DEPENDS= liblilv-0.so:audio/lilv LV2_CONFIGURE_ENABLE= lv2 # libv4l V4L_BUILD_DEPENDS= v4l_compat>0:multimedia/v4l_compat V4L_LIB_DEPENDS= libv4l2.so:multimedia/libv4l V4L_CONFIGURE_ENABLE= libv4l2 V4L_CONFIGURE_OFF= --disable-indev=v4l2 \ --disable-outdev=v4l2 # mbedtls MBEDTLS_LIB_DEPENDS= libmbedtls.so:security/mbedtls2 MBEDTLS_CONFIGURE_ENABLE= mbedtls # mfx MFX_LIB_DEPENDS= libmfx.so:multimedia/intel-media-sdk MFX_CONFIGURE_ENABLE= libmfx # modplug MODPLUG_LIB_DEPENDS= libmodplug.so:audio/libmodplug MODPLUG_CONFIGURE_ENABLE= libmodplug # mysofa MYSOFA_LIB_DEPENDS= libmysofa.so:audio/libmysofa MYSOFA_CONFIGURE_ENABLE= libmysofa # network NETWORK_CONFIGURE_ENABLE= network # nvenc NVENC_BROKEN= Fails to configure or build with ffnvcodec-headers >= 13.0 NVENC_BUILD_DEPENDS= ${LOCALBASE}/include/ffnvcodec/nvEncodeAPI.h:multimedia/ffnvcodec-headers NVENC_CONFIGURE_ENABLE= nvenc # OpenAL OPENAL_LIB_DEPENDS= libopenal.so:audio/openal-soft OPENAL_CONFIGURE_ENABLE= openal # opencl OPENCL_BUILD_DEPENDS= ${LOCALBASE}/include/CL/opencl.h:devel/opencl OPENCL_LIB_DEPENDS= libOpenCL.so:devel/ocl-icd OPENCL_CONFIGURE_ENABLE= opencl # opengl OPENGL_USES= gl OPENGL_USE= GL=gl OPENGL_CONFIGURE_ENABLE= opengl # openh264 OPENH264_LIB_DEPENDS= libopenh264.so:multimedia/openh264 OPENH264_CONFIGURE_ENABLE= libopenh264 # openjpeg OPENJPEG_LIB_DEPENDS= libopenjp2.so:graphics/openjpeg OPENJPEG_CONFIGURE_ENABLE= libopenjpeg # openmpt OPENMPT_LIB_DEPENDS= libopenmpt.so:audio/libopenmpt OPENMPT_CONFIGURE_ENABLE= libopenmpt # openssl/libtls OPENSSL_USES= ssl OPENSSL_CONFIGURE_ENABLE= ${"${SSL_DEFAULT:Mlibressl*}"!="":?libtls:openssl} OPENSSL_IMPLIES= NONFREE # openvino OPENVINO_LIB_DEPENDS= libinference_engine_c_api.so:science/openvino OPENVINO_CONFIGURE_ENABLE= libopenvino # optimizations OPTIMIZED_CFLAGS_CONFIGURE_ENABLE= optimizations # opus OPUS_LIB_DEPENDS= libopus.so:audio/opus OPUS_CONFIGURE_ENABLE= libopus # pocketsphinx POCKETSPHINX_LIB_DEPENDS= libpocketsphinx.so:audio/pocketsphinx POCKETSPHINX_CONFIGURE_ENABLE= pocketsphinx POCKETSPHINX_BROKEN= pocketsphinx < 5prealpha is not supported # pulseaudio PULSEAUDIO_LIB_DEPENDS= libpulse.so:audio/pulseaudio PULSEAUDIO_CONFIGURE_ENABLE= libpulse # rav1e RAV1E_LIB_DEPENDS= librav1e.so:multimedia/librav1e RAV1E_CONFIGURE_ENABLE= librav1e # rabbitmq RABBITMQ_LIB_DEPENDS= librabbitmq.so:net/rabbitmq-c RABBITMQ_CONFIGURE_ENABLE= librabbitmq # rist RIST_LIB_DEPENDS= librist.so:multimedia/librist RIST_CONFIGURE_ENABLE= librist # rubberband RUBBERBAND_LIB_DEPENDS= librubberband.so:audio/rubberband RUBBERBAND_CONFIGURE_ENABLE= librubberband # rtcpu RTCPU_CONFIGURE_ENABLE= runtime-cpudetect # sdl SDL_USES= sdl SDL_USE= SDL=sdl2 SDL_CONFIGURE_ENABLE= sdl2 # smbclient SMB_USES= samba:lib SMB_CONFIGURE_ENABLE= libsmbclient SMB_IMPLIES= GPL3 # snappy SNAPPY_LIB_DEPENDS= libsnappy.so:archivers/snappy SNAPPY_CONFIGURE_ENABLE= libsnappy # sndio SNDIO_LIB_DEPENDS= libsndio.so:audio/sndio SNDIO_CONFIGURE_ENABLE= sndio # soxr SOXR_LIB_DEPENDS= libsoxr.so:audio/libsoxr SOXR_CONFIGURE_ENABLE= libsoxr # speex SPEEX_LIB_DEPENDS= libspeex.so:audio/speex SPEEX_CONFIGURE_ENABLE= libspeex # ssh SSH_LIB_DEPENDS= libssh.so:security/libssh SSH_CONFIGURE_ENABLE= libssh # srt SRT_LIB_DEPENDS= libsrt.so:www/srt SRT_CONFIGURE_ENABLE= libsrt # svt-av1 SVTAV1_LIB_DEPENDS= libSvtAv1Enc.so:multimedia/svt-av1 SVTAV1_CONFIGURE_ON= --enable-libsvtav1 # svt-hevc SVTHEVC_LIB_DEPENDS= libSvtHevcEnc.so:multimedia/svt-hevc SVTHEVC_CONFIGURE_ON= --enable-libsvthevc SVTHEVC_PATCH_SITES= https://github.com/OpenVisualCloud/SVT-HEVC/raw/v1.5.0-3-g86b58f77/ffmpeg_plugin/:svthevc SVTHEVC_PATCHFILES= 0001-lavc-svt_hevc-add-libsvt-hevc-encoder-wrapper.patch:-p1:svthevc \ 0002-doc-Add-libsvt_hevc-encoder-docs.patch:-p1:svthevc .if make(makesum) .MAKEFLAGS: WITH+=SVTHEVC .endif # svt-vp9 SVTVP9_LIB_DEPENDS= libSvtVp9Enc.so:multimedia/svt-vp9 SVTVP9_CONFIGURE_ON= --enable-libsvtvp9 SVTVP9_PATCH_SITES= https://github.com/OpenVisualCloud/SVT-VP9/raw/v0.3.0-4-gabd5c59/ffmpeg_plugin/:svtvp9 SVTVP9_PATCHFILES= master-0001-Add-ability-for-ffmpeg-to-run-svt-vp9.patch:-p1:svtvp9 .if make(makesum) .MAKEFLAGS: WITH+=SVTVP9 .endif # tensorflow TENSORFLOW_LIB_DEPENDS= libtensorflow.so:science/py-tensorflow TENSORFLOW_CONFIGURE_ENABLE= libtensorflow # tesseract TESSERACT_LIB_DEPENDS= libtesseract.so:graphics/tesseract TESSERACT_CONFIGURE_ENABLE= libtesseract # theora THEORA_LIB_DEPENDS= libtheora.so:multimedia/libtheora THEORA_CONFIGURE_ENABLE= libtheora # twolame TWOLAME_LIB_DEPENDS= libtwolame.so:audio/twolame TWOLAME_CONFIGURE_ENABLE= libtwolame # uavs3d UAVS3D_LIB_DEPENDS= libuavs3d.so:multimedia/uavs3d UAVS3D_CONFIGURE_ENABLE= libuavs3d # vaapi VAAPI_LIB_DEPENDS= libva.so:multimedia/libva VAAPI_CONFIGURE_ENABLE= vaapi # vapoursynth VAPOURSYNTH_LIB_DEPENDS= libvapoursynth-script.so:multimedia/vapoursynth VAPOURSYNTH_CONFIGURE_ENABLE= vapoursynth # vdpau VDPAU_USES= xorg VDPAU_USE= XORG=x11 VDPAU_LIB_DEPENDS= libvdpau.so:multimedia/libvdpau VDPAU_CONFIGURE_ENABLE= vdpau # vmaf VMAF_LIB_DEPENDS= libvmaf.so:multimedia/vmaf VMAF_CONFIGURE_ENABLE= libvmaf # vo-amrwbenc VO_AMRWBENC_LIB_DEPENDS= libvo-amrwbenc.so:audio/vo-amrwbenc VO_AMRWBENC_CONFIGURE_ENABLE= libvo-amrwbenc VO_AMRWBENC_IMPLIES= GPL3 # vid.stab VIDSTAB_LIB_DEPENDS= libvidstab.so:multimedia/vid.stab VIDSTAB_CONFIGURE_ENABLE= libvidstab # vorbis VORBIS_LIB_DEPENDS= libvorbisenc.so:audio/libvorbis VORBIS_CONFIGURE_ENABLE= libvorbis # vp8 VPX_LIB_DEPENDS= libvpx.so:multimedia/libvpx VPX_CONFIGURE_ENABLE= libvpx # vulkan VULKAN_BUILD_DEPENDS= vulkan-headers>0:graphics/vulkan-headers VULKAN_LIB_DEPENDS= libvulkan.so:graphics/vulkan-loader VULKAN_CONFIGURE_ENABLE= vulkan # webp WEBP_LIB_DEPENDS= libwebp.so:graphics/webp WEBP_CONFIGURE_ENABLE= libwebp # x264 X264_LIB_DEPENDS= libx264.so:multimedia/libx264 X264_CONFIGURE_ENABLE= libx264 # x265 X265_LIB_DEPENDS= libx265.so:multimedia/x265 X265_CONFIGURE_ENABLE= libx265 # xavs2 XAVS2_LIB_DEPENDS= libxavs2.so:multimedia/xavs2 XAVS2_CONFIGURE_ENABLE= libxavs2 # xcb XCB_USES= xorg XCB_USE= XORG=xcb XCB_CONFIGURE_ENABLE= libxcb # xvid XVID_LIB_DEPENDS= libxvidcore.so:multimedia/xvid XVID_CONFIGURE_ENABLE= libxvid # xv XVIDEO_USES= xorg XVIDEO_USE= XORG=x11,xext,xv XVIDEO_CONFIGURE_OFF= --disable-outdev=xv # zimg ZIMG_LIB_DEPENDS= libzimg.so:graphics/sekrit-twc-zimg ZIMG_CONFIGURE_ENABLE= libzimg # zmq ZMQ_LIB_DEPENDS= libzmq.so:net/libzmq${ZMQ_VERSION} ZMQ_CONFIGURE_ENABLE= libzmq ZMQ_VERSION?= 4 # zvbi ZVBI_LIB_DEPENDS= libzvbi.so:devel/libzvbi ZVBI_CONFIGURE_ENABLE= libzvbi # License knobs GPL3_CONFIGURE_ENABLE= version3 GPL3_VARS= LICENSE="GPLv3+ LGPL3+" LICENSE_FILE_GPLv3+ = ${WRKSRC}/COPYING.GPLv3 LICENSE_FILE_LGPL3+ = ${WRKSRC}/COPYING.LGPLv3 NONFREE_CONFIGURE_ENABLE=nonfree INSTALL_TARGET= install-progs install-doc install-data \ install-libs install-headers install-examples DATADIR= ${PREFIX}/share/${PORTNAME}${PKGNAMESUFFIX} DOCSDIR= ${PREFIX}/share/doc/${PORTNAME}${PKGNAMESUFFIX} MAKE_ENV+= V=1 LDFLAGS_aarch64=-Wl,-z,notext LDFLAGS_armv6= -Wl,-z,notext LDFLAGS_armv7= -Wl,-z,notext LDFLAGS_i386= -Wl,-z,notext CONFIGURE_ARGS+=--prefix="${PREFIX}" \ --mandir="${PREFIX}/share/man" \ --datadir="${DATADIR}" \ --docdir="${DOCSDIR}" \ --pkgconfigdir="${PREFIX}/libdata/pkgconfig" \ --disable-static \ --disable-libcelt \ --enable-shared \ --enable-pic \ --enable-gpl \ --enable-avresample \ --cc="${CC}" \ --cxx="${CXX}" DOC_FILES= Changelog CREDITS INSTALL.md LICENSE.md MAINTAINERS \ README.md RELEASE_NOTES # under doc subdirectory DOC_DOCFILES= APIchanges *.txt PORTDOCS= * .include .if ${PORT_OPTIONS:MNONFREE} LICENSE+= NONFREE LICENSE_COMB= multi LICENSE_NAME_NONFREE= Non free code LICENSE_TEXT_NONFREE= enabling OPENSSL or FDK_AAC restricts redistribution LICENSE_PERMS_NONFREE= auto-accept .endif post-install: (cd ${WRKSRC} && ${COPYTREE_SHARE} \ "${DOC_FILES}" ${STAGEDIR}${DOCSDIR}) (cd ${WRKSRC}/doc && ${COPYTREE_SHARE} \ "${DOC_DOCFILES}" ${STAGEDIR}${DOCSDIR}) .include diff --git a/multimedia/ffmpeg4/distinfo b/multimedia/ffmpeg4/distinfo index 287f9dba3d86..ab830e49aada 100644 --- a/multimedia/ffmpeg4/distinfo +++ b/multimedia/ffmpeg4/distinfo @@ -1,9 +1,9 @@ -TIMESTAMP = 1681325571 -SHA256 (ffmpeg-4.4.4.tar.xz) = e80b380d595c809060f66f96a5d849511ef4a76a26b76eacf5778b94c3570309 -SIZE (ffmpeg-4.4.4.tar.xz) = 9565584 +TIMESTAMP = 1773592763 +SHA256 (ffmpeg-4.4.6.tar.xz) = 2290461f467c08ab801731ed412d8e724a5511d6c33173654bd9c1d2e25d0617 +SIZE (ffmpeg-4.4.6.tar.xz) = 9575932 SHA256 (0001-lavc-svt_hevc-add-libsvt-hevc-encoder-wrapper.patch) = 376ea23e9d876390fbca936c80cb437bb9daa6232ff65df0ea91673e18b739ff SIZE (0001-lavc-svt_hevc-add-libsvt-hevc-encoder-wrapper.patch) = 24259 SHA256 (0002-doc-Add-libsvt_hevc-encoder-docs.patch) = d16589efc4be2602a9f436e9938cfee0cc7b5ccd7a4a6b41c0c9a59ae0859fda SIZE (0002-doc-Add-libsvt_hevc-encoder-docs.patch) = 5336 SHA256 (master-0001-Add-ability-for-ffmpeg-to-run-svt-vp9.patch) = b7d722dfce20b73e9d5c73d55ffe041bbdc92a3c4a5c5d766b6b3040671b4052 SIZE (master-0001-Add-ability-for-ffmpeg-to-run-svt-vp9.patch) = 33497 diff --git a/multimedia/ffmpeg4/files/patch-configure b/multimedia/ffmpeg4/files/patch-configure index 6135beab7ee8..87000d4090d5 100644 --- a/multimedia/ffmpeg4/files/patch-configure +++ b/multimedia/ffmpeg4/files/patch-configure @@ -1,43 +1,22 @@ ---- configure.orig 2023-04-12 18:52:51 UTC +https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/3d29724c008d + +--- configure.orig 2026-03-15 17:13:37 UTC +++ configure -@@ -3784,13 +3784,7 @@ target_os_default=$(tolower $(uname -s)) - host_os=$target_os_default - - # machine --if test "$target_os_default" = aix; then -- arch_default=$(uname -p) -- strip_default="strip -X32_64" -- nm_default="nm -g -X32_64" --else -- arch_default=$(uname -m) --fi -+arch_default=$(uname -p) - cpu="generic" - intrinsics="none" - -@@ -5387,6 +5381,7 @@ case $target_os in - disable symver - ;; - freebsd) -+ enable section_data_rel_ro - ;; - bsd/os) - add_extralibs -lpoll -lgnugetopt -@@ -6369,7 +6369,7 @@ enabled libfribidi && require_pkg_config libfri - enabled libfontconfig && require_pkg_config libfontconfig fontconfig "fontconfig/fontconfig.h" FcInit - enabled libfreetype && require_pkg_config libfreetype freetype2 "ft2build.h FT_FREETYPE_H" FT_Init_FreeType - enabled libfribidi && require_pkg_config libfribidi fribidi fribidi.h fribidi_version_info --enabled libglslang && require_cpp libglslang glslang/SPIRV/GlslangToSpv.h "glslang::TIntermediate*" -lglslang -lMachineIndependent -lOSDependent -lHLSL -lOGLCompiler -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -lstdc++ -+enabled libglslang && require_cpp libglslang glslang/SPIRV/GlslangToSpv.h "glslang::TIntermediate*" -lglslang -lMachineIndependent -lOSDependent -lGenericCodeGen -lSPVRemapper -lSPIRV -lSPIRV-Tools-opt -lSPIRV-Tools -lpthread -lstdc++ - enabled libgme && { check_pkg_config libgme libgme gme/gme.h gme_new_emu || - require libgme gme/gme.h gme_new_emu -lgme -lstdc++; } - enabled libgsm && { for gsm_hdr in "gsm.h" "gsm/gsm.h"; do -@@ -6758,7 +6753,7 @@ enabled vulkan && - - if enabled x86; then - case $target_os in -- mingw32*|mingw64*|win32|win64|linux|cygwin*) -+ freebsd|mingw32*|mingw64*|win32|win64|linux|cygwin*) - ;; - *) - disable ffnvcodec cuvid nvdec nvenc +@@ -3668,7 +3668,7 @@ vidstabtransform_filter_deps="libvidstab" + vaguedenoiser_filter_deps="gpl" + vidstabdetect_filter_deps="libvidstab" + vidstabtransform_filter_deps="libvidstab" +-libvmaf_filter_deps="libvmaf pthreads" ++libvmaf_filter_deps="libvmaf" + zmq_filter_deps="libzmq" + zoompan_filter_deps="swscale" + zscale_filter_deps="libzimg const_nan" +@@ -6460,7 +6460,7 @@ enabled libvidstab && require_pkg_config libvid + enabled libuavs3d && require_pkg_config libuavs3d "uavs3d >= 1.1.41" uavs3d.h uavs3d_decode + enabled libv4l2 && require_pkg_config libv4l2 libv4l2 libv4l2.h v4l2_ioctl + enabled libvidstab && require_pkg_config libvidstab "vidstab >= 0.98" vid.stab/libvidstab.h vsMotionDetectInit +-enabled libvmaf && require_pkg_config libvmaf "libvmaf >= 1.5.2" libvmaf.h compute_vmaf ++enabled libvmaf && require_pkg_config libvmaf "libvmaf >= 2.0.0" libvmaf.h vmaf_init + enabled libvo_amrwbenc && require libvo_amrwbenc vo-amrwbenc/enc_if.h E_IF_init -lvo-amrwbenc + enabled libvorbis && require_pkg_config libvorbis vorbis vorbis/codec.h vorbis_info_init && + require_pkg_config libvorbisenc vorbisenc vorbis/vorbisenc.h vorbis_encode_init diff --git a/multimedia/ffmpeg4/files/patch-doc-Makefile b/multimedia/ffmpeg4/files/patch-doc-Makefile index 52ab4033ad98..118989b96e85 100644 --- a/multimedia/ffmpeg4/files/patch-doc-Makefile +++ b/multimedia/ffmpeg4/files/patch-doc-Makefile @@ -1,11 +1,11 @@ ---- doc/Makefile.orig 2015-06-19 20:44:35 UTC +--- doc/Makefile.orig 2025-03-12 11:48:35 UTC +++ doc/Makefile -@@ -145,7 +145,7 @@ install-html: $(HTMLPAGES) +@@ -119,7 +119,7 @@ ifdef CONFIG_MANPAGES endif ifdef CONFIG_MANPAGES -install-progs-$(CONFIG_DOC): install-man +#install-progs-$(CONFIG_DOC): install-man install-man: $(MANPAGES) $(Q)mkdir -p "$(MANDIR)/man1" diff --git a/multimedia/ffmpeg4/files/patch-libavcodec_libsvtav1.c b/multimedia/ffmpeg4/files/patch-libavcodec_libsvtav1.c index 36eb954b63e1..77d50cf0ccea 100644 --- a/multimedia/ffmpeg4/files/patch-libavcodec_libsvtav1.c +++ b/multimedia/ffmpeg4/files/patch-libavcodec_libsvtav1.c @@ -1,16 +1,361 @@ -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/d1ed5c06e3ed - ---- libavcodec/libsvtav1.c.orig 2024-08-01 23:22:48 UTC +--- libavcodec/libsvtav1.c.orig 2025-05-17 13:52:31 UTC +++ libavcodec/libsvtav1.c -@@ -389,7 +389,11 @@ static av_cold int eb_enc_init(AVCodecContext *avctx) +@@ -60,17 +60,20 @@ typedef struct SvtContext { + EOS_STATUS eos_flag; - svt_enc->eos_flag = EOS_NOT_REACHED; + // User options. ++ AVDictionary *svtav1_opts; ++#if FF_API_SVTAV1_OPTS + int hierarchical_level; + int la_depth; +- int enc_mode; +- int rc_mode; + int scd; +- int qp; -+#if SVT_AV1_CHECK_VERSION(3, 0, 0) -+ svt_ret = svt_av1_enc_init_handle(&svt_enc->svt_handle, &svt_enc->enc_params); + int tier; + + int tile_columns; + int tile_rows; ++#endif ++ int enc_mode; ++ int crf; ++ int qp; + } SvtContext; + + static const struct { +@@ -147,11 +150,136 @@ static int config_enc_params(EbSvtAv1EncConfiguration + { + SvtContext *svt_enc = avctx->priv_data; + const AVPixFmtDescriptor *desc; ++ AVDictionaryEntry *en = NULL; + ++ // Update param from options ++#if FF_API_SVTAV1_OPTS ++ if (svt_enc->hierarchical_level >= 0) ++ param->hierarchical_levels = svt_enc->hierarchical_level; ++ if (svt_enc->tier >= 0) ++ param->tier = svt_enc->tier; ++ if (svt_enc->scd >= 0) ++ param->scene_change_detection = svt_enc->scd; ++ if (svt_enc->tile_columns >= 0) ++ param->tile_columns = svt_enc->tile_columns; ++ if (svt_enc->tile_rows >= 0) ++ param->tile_rows = svt_enc->tile_rows; ++ ++ if (svt_enc->la_depth >= 0) ++ param->look_ahead_distance = svt_enc->la_depth; ++#endif ++ ++ if (svt_enc->enc_mode >= 0) ++ param->enc_mode = svt_enc->enc_mode; ++ ++ if (avctx->bit_rate) { ++ param->target_bit_rate = avctx->bit_rate; ++ if (avctx->rc_max_rate != avctx->bit_rate) ++ param->rate_control_mode = 1; ++ else ++ param->rate_control_mode = 2; ++ ++ param->max_qp_allowed = avctx->qmax; ++ param->min_qp_allowed = avctx->qmin; ++ } ++ param->max_bit_rate = avctx->rc_max_rate; ++ if (avctx->bit_rate && avctx->rc_buffer_size) ++ param->maximum_buffer_size_ms = avctx->rc_buffer_size * 1000LL / avctx->bit_rate; ++ ++ if (svt_enc->crf > 0) { ++ param->qp = svt_enc->crf; ++ param->rate_control_mode = 0; ++ } else if (svt_enc->qp > 0) { ++ param->qp = svt_enc->qp; ++ param->rate_control_mode = 0; ++#if SVT_AV1_CHECK_VERSION(4, 0, 0) ++ param->aq_mode = 0; +#else - svt_ret = svt_av1_enc_init_handle(&svt_enc->svt_handle, svt_enc, &svt_enc->enc_params); ++ param->enable_adaptive_quantization = 0; ++#endif ++ } ++ ++ desc = av_pix_fmt_desc_get(avctx->pix_fmt); ++ param->color_primaries = avctx->color_primaries; ++ param->matrix_coefficients = (desc->flags & AV_PIX_FMT_FLAG_RGB) ? ++ AVCOL_SPC_RGB : avctx->colorspace; ++ param->transfer_characteristics = avctx->color_trc; ++ ++ if (avctx->color_range != AVCOL_RANGE_UNSPECIFIED) ++ param->color_range = avctx->color_range == AVCOL_RANGE_JPEG; ++ else ++ param->color_range = !!(desc->flags & AV_PIX_FMT_FLAG_RGB); ++ ++#if SVT_AV1_CHECK_VERSION(1, 0, 0) ++ if (avctx->chroma_sample_location != AVCHROMA_LOC_UNSPECIFIED) { ++ const char *name = ++ av_chroma_location_name(avctx->chroma_sample_location); ++ ++ switch (avctx->chroma_sample_location) { ++ case AVCHROMA_LOC_LEFT: ++ param->chroma_sample_position = EB_CSP_VERTICAL; ++ break; ++ case AVCHROMA_LOC_TOPLEFT: ++ param->chroma_sample_position = EB_CSP_COLOCATED; ++ break; ++ default: ++ if (!name) ++ break; ++ ++ av_log(avctx, AV_LOG_WARNING, ++ "Specified chroma sample location %s is unsupported " ++ "on the AV1 bit stream level. Usage of a container that " ++ "allows passing this information - such as Matroska - " ++ "is recommended.\n", ++ name); ++ break; ++ } ++ } +#endif ++ ++ if (avctx->profile != FF_PROFILE_UNKNOWN) ++ param->profile = avctx->profile; ++ ++ if (avctx->level != FF_LEVEL_UNKNOWN) ++ param->level = avctx->level; ++ ++ if (avctx->gop_size > 0) ++ param->intra_period_length = avctx->gop_size - 1; ++ ++ if (avctx->framerate.num > 0 && avctx->framerate.den > 0) { ++ param->frame_rate_numerator = avctx->framerate.num; ++ param->frame_rate_denominator = avctx->framerate.den; ++ } else { ++ param->frame_rate_numerator = avctx->time_base.den; ++ param->frame_rate_denominator = avctx->time_base.num * avctx->ticks_per_frame; ++ } ++ ++ /* 2 = IDR, closed GOP, 1 = CRA, open GOP */ ++ param->intra_refresh_type = avctx->flags & AV_CODEC_FLAG_CLOSED_GOP ? 2 : 1; ++ ++#if SVT_AV1_CHECK_VERSION(0, 9, 1) ++ while ((en = av_dict_get(svt_enc->svtav1_opts, "", en, AV_DICT_IGNORE_SUFFIX))) { ++ EbErrorType ret = svt_av1_enc_parse_parameter(param, en->key, en->value); ++ if (ret != EB_ErrorNone) { ++ int level = (avctx->err_recognition & AV_EF_EXPLODE) ? AV_LOG_ERROR : AV_LOG_WARNING; ++ av_log(avctx, level, "Error parsing option %s: %s.\n", en->key, en->value); ++ if (avctx->err_recognition & AV_EF_EXPLODE) ++ return AVERROR(EINVAL); ++ } ++ } ++#else ++ if ((en = av_dict_get(svt_enc->svtav1_opts, "", NULL, AV_DICT_IGNORE_SUFFIX))) { ++ int level = (avctx->err_recognition & AV_EF_EXPLODE) ? AV_LOG_ERROR : AV_LOG_WARNING; ++ av_log(avctx, level, "svt-params needs libavcodec to be compiled with SVT-AV1 " ++ "headers >= 0.9.1.\n"); ++ if (avctx->err_recognition & AV_EF_EXPLODE) ++ return AVERROR(ENOSYS); ++ } ++#endif ++ + param->source_width = avctx->width; + param->source_height = avctx->height; + +- desc = av_pix_fmt_desc_get(avctx->pix_fmt); + param->encoder_bit_depth = desc->comp[0].depth; + + if (desc->log2_chroma_w == 1 && desc->log2_chroma_h == 1) +@@ -165,12 +293,6 @@ static int config_enc_params(EbSvtAv1EncConfiguration + return AVERROR(EINVAL); + } + +- if (avctx->profile != FF_PROFILE_UNKNOWN) +- param->profile = avctx->profile; +- +- if (avctx->level != FF_LEVEL_UNKNOWN) +- param->level = avctx->level; +- + if ((param->encoder_color_format == EB_YUV422 || param->encoder_bit_depth > 10) + && param->profile != FF_PROFILE_AV1_PROFESSIONAL ) { + av_log(avctx, AV_LOG_WARNING, "Forcing Professional profile\n"); +@@ -180,40 +302,21 @@ static int config_enc_params(EbSvtAv1EncConfiguration + param->profile = FF_PROFILE_AV1_HIGH; + } + +- // Update param from options +- param->hierarchical_levels = svt_enc->hierarchical_level; +- param->enc_mode = svt_enc->enc_mode; +- param->tier = svt_enc->tier; +- param->rate_control_mode = svt_enc->rc_mode; +- param->scene_change_detection = svt_enc->scd; +- param->qp = svt_enc->qp; ++ avctx->bit_rate = param->rate_control_mode > 0 ? ++ param->target_bit_rate : 0; ++ avctx->rc_max_rate = param->max_bit_rate; ++ avctx->rc_buffer_size = param->maximum_buffer_size_ms * avctx->bit_rate / 1000LL; + +- param->target_bit_rate = avctx->bit_rate; ++ if (avctx->bit_rate || avctx->rc_max_rate || avctx->rc_buffer_size) { ++ AVCPBProperties *cpb_props = ff_add_cpb_side_data(avctx); ++ if (!cpb_props) ++ return AVERROR(ENOMEM); + +- if (avctx->gop_size > 0) +- param->intra_period_length = avctx->gop_size - 1; +- +- if (avctx->framerate.num > 0 && avctx->framerate.den > 0) { +- param->frame_rate_numerator = avctx->framerate.num; +- param->frame_rate_denominator = avctx->framerate.den; +- } else { +- param->frame_rate_numerator = avctx->time_base.den; +- param->frame_rate_denominator = avctx->time_base.num * avctx->ticks_per_frame; ++ cpb_props->buffer_size = avctx->rc_buffer_size; ++ cpb_props->max_bitrate = avctx->rc_max_rate; ++ cpb_props->avg_bitrate = avctx->bit_rate; + } + +- if (param->rate_control_mode) { +- param->max_qp_allowed = avctx->qmax; +- param->min_qp_allowed = avctx->qmin; +- } +- +- param->intra_refresh_type = 2; /* Real keyframes only */ +- +- if (svt_enc->la_depth >= 0) +- param->look_ahead_distance = svt_enc->la_depth; +- +- param->tile_columns = svt_enc->tile_columns; +- param->tile_rows = svt_enc->tile_rows; +- + return 0; + } + +@@ -262,7 +365,7 @@ static av_cold int eb_enc_init(AVCodecContext *avctx) + + svt_enc->eos_flag = EOS_NOT_REACHED; + +- svt_ret = svt_av1_enc_init_handle(&svt_enc->svt_handle, svt_enc, &svt_enc->enc_params); ++ svt_ret = svt_av1_enc_init_handle(&svt_enc->svt_handle, &svt_enc->enc_params); if (svt_ret != EB_ErrorNone) { return svt_print_error(avctx, svt_ret, "Error initializing encoder handle"); } +@@ -326,11 +429,8 @@ static int eb_send_frame(AVCodecContext *avctx, const + if (svt_enc->eos_flag == EOS_SENT) + return 0; + +- headerPtrLast.n_alloc_len = 0; +- headerPtrLast.n_filled_len = 0; +- headerPtrLast.n_tick_count = 0; +- headerPtrLast.p_app_private = NULL; +- headerPtrLast.p_buffer = NULL; ++ memset(&headerPtrLast, 0, sizeof(headerPtrLast)); ++ headerPtrLast.pic_type = EB_AV1_INVALID_PICTURE; + headerPtrLast.flags = EB_BUFFERFLAG_EOS; + + svt_av1_enc_send_picture(svt_enc->svt_handle, &headerPtrLast); +@@ -346,6 +446,16 @@ static int eb_send_frame(AVCodecContext *avctx, const + headerPtr->p_app_private = NULL; + headerPtr->pts = frame->pts; + ++ switch (frame->pict_type) { ++ case AV_PICTURE_TYPE_I: ++ headerPtr->pic_type = EB_AV1_KEY_PICTURE; ++ break; ++ default: ++ // Actually means auto, or default. ++ headerPtr->pic_type = EB_AV1_INVALID_PICTURE; ++ break; ++ } ++ + svt_av1_enc_send_picture(svt_enc->svt_handle, headerPtr); + + return 0; +@@ -468,21 +578,22 @@ static const AVOption options[] = { + #define OFFSET(x) offsetof(SvtContext, x) + #define VE AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM + static const AVOption options[] = { +- { "hielevel", "Hierarchical prediction levels setting", OFFSET(hierarchical_level), +- AV_OPT_TYPE_INT, { .i64 = 4 }, 3, 4, VE , "hielevel"}, ++#if FF_API_SVTAV1_OPTS ++ { "hielevel", "Hierarchical prediction levels setting (Deprecated, use svtav1-params)", OFFSET(hierarchical_level), ++ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 4, VE | AV_OPT_FLAG_DEPRECATED , "hielevel"}, + { "3level", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 3 }, INT_MIN, INT_MAX, VE, "hielevel" }, + { "4level", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 4 }, INT_MIN, INT_MAX, VE, "hielevel" }, + +- { "la_depth", "Look ahead distance [0, 120]", OFFSET(la_depth), +- AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 120, VE }, ++ { "la_depth", "Look ahead distance [0, 120] (Deprecated, use svtav1-params)", OFFSET(la_depth), ++ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 120, VE | AV_OPT_FLAG_DEPRECATED }, + +- { "preset", "Encoding preset [0, 8]", +- OFFSET(enc_mode), AV_OPT_TYPE_INT, { .i64 = MAX_ENC_PRESET }, 0, MAX_ENC_PRESET, VE }, +- +- { "tier", "Set operating point tier", OFFSET(tier), +- AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, VE, "tier" }, ++ { "tier", "Set operating point tier (Deprecated, use svtav1-params)", OFFSET(tier), ++ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE | AV_OPT_FLAG_DEPRECATED, "tier" }, + { "main", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, 0, 0, VE, "tier" }, + { "high", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 1 }, 0, 0, VE, "tier" }, ++#endif ++ { "preset", "Encoding preset", ++ OFFSET(enc_mode), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, MAX_ENC_PRESET, VE }, + + FF_AV1_PROFILE_OPTS + +@@ -514,21 +625,20 @@ static const AVOption options[] = { + { LEVEL("7.3", 73) }, + #undef LEVEL + +- { "rc", "Bit rate control mode", OFFSET(rc_mode), +- AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 3, VE , "rc"}, +- { "cqp", "Constant quantizer", 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, INT_MIN, INT_MAX, VE, "rc" }, +- { "vbr", "Variable Bit Rate, use a target bitrate for the entire stream", 0, AV_OPT_TYPE_CONST, { .i64 = 1 }, INT_MIN, INT_MAX, VE, "rc" }, +- { "cvbr", "Constrained Variable Bit Rate, use a target bitrate for each GOP", 0, AV_OPT_TYPE_CONST,{ .i64 = 2 }, INT_MIN, INT_MAX, VE, "rc" }, ++ { "crf", "Constant Rate Factor value", OFFSET(crf), ++ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 63, VE }, ++ { "qp", "Initial Quantizer level value", OFFSET(qp), ++ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 63, VE }, ++#if FF_API_SVTAV1_OPTS ++ { "sc_detection", "Scene change detection (Deprecated, use svtav1-params)", OFFSET(scd), ++ AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VE | AV_OPT_FLAG_DEPRECATED }, + +- { "qp", "Quantizer to use with cqp rate control mode", OFFSET(qp), +- AV_OPT_TYPE_INT, { .i64 = 50 }, 0, 63, VE }, ++ { "tile_columns", "Log2 of number of tile columns to use (Deprecated, use svtav1-params)", OFFSET(tile_columns), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 4, VE | AV_OPT_FLAG_DEPRECATED }, ++ { "tile_rows", "Log2 of number of tile rows to use (Deprecated, use svtav1-params)", OFFSET(tile_rows), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 6, VE | AV_OPT_FLAG_DEPRECATED }, ++#endif + +- { "sc_detection", "Scene change detection", OFFSET(scd), +- AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE }, ++ { "svtav1-params", "Set the SVT-AV1 configuration using a :-separated list of key=value parameters", OFFSET(svtav1_opts), AV_OPT_TYPE_DICT, { 0 }, 0, 0, VE }, + +- { "tile_columns", "Log2 of number of tile columns to use", OFFSET(tile_columns), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 4, VE}, +- { "tile_rows", "Log2 of number of tile rows to use", OFFSET(tile_rows), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 6, VE}, +- + {NULL}, + }; + +@@ -540,9 +650,10 @@ static const AVCodecDefault eb_enc_defaults[] = { + }; + + static const AVCodecDefault eb_enc_defaults[] = { +- { "b", "7M" }, ++ { "b", "0" }, ++ { "flags", "+cgop" }, + { "g", "-1" }, +- { "qmin", "0" }, ++ { "qmin", "1" }, + { "qmax", "63" }, + { NULL }, + }; +@@ -557,12 +668,11 @@ AVCodec ff_libsvtav1_encoder = { + .receive_packet = eb_receive_packet, + .close = eb_enc_close, + .capabilities = AV_CODEC_CAP_DELAY | AV_CODEC_CAP_OTHER_THREADS, +- .caps_internal = FF_CODEC_CAP_AUTO_THREADS, ++ .caps_internal = FF_CODEC_CAP_AUTO_THREADS | FF_CODEC_CAP_INIT_CLEANUP, + .pix_fmts = (const enum AVPixelFormat[]){ AV_PIX_FMT_YUV420P, + AV_PIX_FMT_YUV420P10, + AV_PIX_FMT_NONE }, + .priv_class = &class, + .defaults = eb_enc_defaults, +- .caps_internal = FF_CODEC_CAP_INIT_CLEANUP, + .wrapper_name = "libsvtav1", + }; diff --git a/multimedia/ffmpeg4/files/patch-libavdevice_v4l2.c b/multimedia/ffmpeg4/files/patch-libavdevice_v4l2.c deleted file mode 100644 index 8518fadce3af..000000000000 --- a/multimedia/ffmpeg4/files/patch-libavdevice_v4l2.c +++ /dev/null @@ -1,18 +0,0 @@ -- ioctl() prototype on BSDs follows Version 7 AT&T UNIX - ---- libavdevice/v4l2.c.orig 2019-04-21 04:37:17 UTC -+++ libavdevice/v4l2.c -@@ -95,10 +95,10 @@ struct video_data { - int (*open_f)(const char *file, int oflag, ...); - int (*close_f)(int fd); - int (*dup_f)(int fd); --#ifdef __GLIBC__ -- int (*ioctl_f)(int fd, unsigned long int request, ...); --#else -+#if defined(__sun) || defined(__BIONIC__) || defined(__musl__) /* POSIX-like */ - int (*ioctl_f)(int fd, int request, ...); -+#else -+ int (*ioctl_f)(int fd, unsigned long int request, ...); - #endif - ssize_t (*read_f)(int fd, void *buffer, size_t n); - void *(*mmap_f)(void *start, size_t length, int prot, int flags, int fd, int64_t offset); diff --git a/multimedia/ffmpeg4/files/patch-libavformat_rtsp.c b/multimedia/ffmpeg4/files/patch-libavformat_rtsp.c index 7cd751dbd79c..e56281ecea34 100644 --- a/multimedia/ffmpeg4/files/patch-libavformat_rtsp.c +++ b/multimedia/ffmpeg4/files/patch-libavformat_rtsp.c @@ -1,44 +1,44 @@ ---- libavformat/rtsp.c.orig 2016-10-27 16:17:40 UTC +--- libavformat/rtsp.c.orig 2025-05-17 13:52:31 UTC +++ libavformat/rtsp.c -@@ -1614,7 +1614,12 @@ int ff_rtsp_make_setup_request(AVFormatC +@@ -1672,7 +1672,12 @@ int ff_rtsp_make_setup_request(AVFormatContext *s, con } if (ttl > 0) snprintf(optbuf, sizeof(optbuf), "?ttl=%d", ttl); - getnameinfo((struct sockaddr*) &addr, sizeof(addr), + getnameinfo((struct sockaddr*) &addr, +#if HAVE_STRUCT_SOCKADDR_SA_LEN + ((struct sockaddr*) &addr)->sa_len, +#else + sizeof(addr), +#endif namebuf, sizeof(namebuf), NULL, 0, NI_NUMERICHOST); ff_url_join(url, sizeof(url), "rtp", NULL, namebuf, port, "%s", optbuf); -@@ -1830,8 +1835,13 @@ redirect: +@@ -1906,8 +1911,13 @@ redirect: goto fail; } if (!getpeername(tcp_fd, (struct sockaddr*) &peer, &peer_len)) { - getnameinfo((struct sockaddr*) &peer, peer_len, host, sizeof(host), - NULL, 0, NI_NUMERICHOST); + getnameinfo((struct sockaddr*) &peer, +#if HAVE_STRUCT_SOCKADDR_SA_LEN + ((struct sockaddr*) &peer)->sa_len, +#else + peer_len, +#endif + host, sizeof(host), NULL, 0, NI_NUMERICHOST); } /* request options supported by the server; this also detects server -@@ -2310,7 +2320,11 @@ static int sdp_read_header(AVFormatConte +@@ -2425,7 +2435,11 @@ static int sdp_read_header(AVFormatContext *s) AVDictionary *opts = map_to_opts(rt); err = getnameinfo((struct sockaddr*) &rtsp_st->sdp_ip, +#if HAVE_STRUCT_SOCKADDR_SA_LEN + ((struct sockaddr*) &rtsp_st->sdp_ip)->sa_len, +#else sizeof(rtsp_st->sdp_ip), +#endif namebuf, sizeof(namebuf), NULL, 0, NI_NUMERICHOST); if (err) { av_log(s, AV_LOG_ERROR, "getnameinfo: %s\n", gai_strerror(err)); diff --git a/multimedia/ffmpeg4/files/patch-libavutil-common.h b/multimedia/ffmpeg4/files/patch-libavutil-common.h index 2c8c84e6d372..4baccfda1089 100644 --- a/multimedia/ffmpeg4/files/patch-libavutil-common.h +++ b/multimedia/ffmpeg4/files/patch-libavutil-common.h @@ -1,13 +1,13 @@ ---- libavutil/common.h.orig 2015-06-19 20:44:53 UTC +--- libavutil/common.h.orig 2025-05-17 13:52:31 UTC +++ libavutil/common.h -@@ -49,6 +49,10 @@ +@@ -50,6 +50,10 @@ # define AV_NE(be, le) (le) #endif +#ifndef UINT64_C +#define UINT64_C(c) (c ## UL) +#endif + //rounded division & shift #define RSHIFT(a,b) ((a) > 0 ? ((a) + ((1<<(b))>>1))>>(b) : ((a) + ((1<<(b))>>1)-1)>>(b)) /* assume b>0 */ diff --git a/multimedia/ffmpeg4/files/patch-libavutil-x86-asm.h b/multimedia/ffmpeg4/files/patch-libavutil-x86-asm.h index 71e31a029955..7f9a7d5e59f1 100644 --- a/multimedia/ffmpeg4/files/patch-libavutil-x86-asm.h +++ b/multimedia/ffmpeg4/files/patch-libavutil-x86-asm.h @@ -1,11 +1,11 @@ ---- libavutil/x86/asm.h.orig 2015-06-19 20:44:54 UTC +--- libavutil/x86/asm.h.orig 2023-11-09 23:38:51 UTC +++ libavutil/x86/asm.h -@@ -71,7 +71,7 @@ typedef int32_t x86_reg; +@@ -72,7 +72,7 @@ typedef int x86_reg; typedef int x86_reg; #endif -#define HAVE_7REGS (ARCH_X86_64 || (HAVE_EBX_AVAILABLE && HAVE_EBP_AVAILABLE)) +#define HAVE_7REGS (ARCH_X86_64 || (HAVE_EBX_AVAILABLE && HAVE_EBP_AVAILABLE && !defined(__clang__))) #define HAVE_6REGS (ARCH_X86_64 || (HAVE_EBX_AVAILABLE || HAVE_EBP_AVAILABLE)) #if ARCH_X86_64 && defined(PIC) diff --git a/multimedia/ffmpeg4/files/patch-libswscale_ppc_yuv2rgb__altivec.c b/multimedia/ffmpeg4/files/patch-libswscale_ppc_yuv2rgb__altivec.c deleted file mode 100644 index 38ff9087b842..000000000000 --- a/multimedia/ffmpeg4/files/patch-libswscale_ppc_yuv2rgb__altivec.c +++ /dev/null @@ -1,15 +0,0 @@ -Error: -libswscale/ppc/yuv2rgb_altivec.c:288:36: error: redeclaration of 'vec_xl' must have the 'overloadable' attribute -static inline vector unsigned char vec_xl(signed long long offset, const ubyte *addr) - ---- libswscale/ppc/yuv2rgb_altivec.c.orig 2021-04-08 21:28:40 UTC -+++ libswscale/ppc/yuv2rgb_altivec.c -@@ -284,7 +284,7 @@ static inline void cvtyuvtoRGB(SwsContext *c, vector s - * ------------------------------------------------------------------------------ - */ - --#if !HAVE_VSX -+#if !HAVE_VSX && !defined(__clang__) - static inline vector unsigned char vec_xl(signed long long offset, const ubyte *addr) - { - const vector unsigned char *v_addr = (const vector unsigned char *) (addr + offset); diff --git a/multimedia/ffmpeg4/files/patch-svtav1 b/multimedia/ffmpeg4/files/patch-svtav1 index dc32623e0d6a..65596fd6cbd2 100644 --- a/multimedia/ffmpeg4/files/patch-svtav1 +++ b/multimedia/ffmpeg4/files/patch-svtav1 @@ -1,472 +1,66 @@ -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/04b89e8ae33b -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/64e2fb3f9d89 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/0463f5d6d56d -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/c5f314309067 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/c33b4048859a -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/a2b090da7932 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/1dddb930aaf0 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/4e47ebf38b97 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/50bc87263576 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/d794b36a7788 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/51c0b9e829be -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/e3c4442b249a -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/6fd1533057ff -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/ded0334d214f -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/70887d44ffa3 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/fe100bc556d7 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/fb70e0611bd7 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/031f1561cd28 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/1c6fd7d756af -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/96748ac54f99 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/fe196fd29a67 -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/a5d4c398b411 - ---- configure.orig 2021-10-24 20:47:11 UTC -+++ configure -@@ -6430,7 +6430,7 @@ enabled libsrt && require_pkg_config libsrt - enabled libssh && require_pkg_config libssh libssh libssh/sftp.h sftp_init - enabled libspeex && require_pkg_config libspeex speex speex/speex.h speex_decoder_init - enabled libsrt && require_pkg_config libsrt "srt >= 1.3.0" srt/srt.h srt_socket --enabled libsvtav1 && require_pkg_config libsvtav1 "SvtAv1Enc >= 0.8.4" EbSvtAv1Enc.h svt_av1_enc_init_handle -+enabled libsvtav1 && require_pkg_config libsvtav1 "SvtAv1Enc >= 0.9.0" EbSvtAv1Enc.h svt_av1_enc_init_handle - enabled libtensorflow && require libtensorflow tensorflow/c/c_api.h TF_Version -ltensorflow - enabled libtesseract && require_pkg_config libtesseract tesseract tesseract/capi.h TessBaseAPICreate - enabled libtheora && require libtheora theora/theoraenc.h th_info_init -ltheoraenc -ltheoradec -logg ---- doc/encoders.texi.orig 2021-10-24 20:47:07 UTC +--- doc/encoders.texi.orig 2025-05-17 13:52:31 UTC +++ doc/encoders.texi @@ -1754,28 +1754,15 @@ Set the operating point tier. @item tier Set the operating point tier. -@item rc -Set the rate control mode to use. - -Possible modes: -@table @option -@item cqp -Constant quantizer: use fixed values of qindex (dependent on the frame type) -throughout the stream. This mode is the default. - -@item vbr -Variable bitrate: use a target bitrate for the whole stream. - -@item cvbr -Constrained variable bitrate: use a target bitrate for each GOP. -@end table - @item qmax Set the maximum quantizer to use when using a bitrate mode. @item qmin Set the minimum quantizer to use when using a bitrate mode. +@item crf +Constant rate factor value used in crf rate control mode (0-63). + @item qp Set the quantizer used in cqp rate control mode (0-63). @@ -1786,14 +1773,18 @@ Set number of frames to look ahead (0-120). Set number of frames to look ahead (0-120). @item preset -Set the quality-speed tradeoff, in the range 0 to 8. Higher values are -faster but lower quality. Defaults to 8 (highest speed). +Set the quality-speed tradeoff, in the range 0 to 13. Higher values are +faster but lower quality. @item tile_rows Set log2 of the number of rows of tiles to use (0-6). @item tile_columns Set log2 of the number of columns of tiles to use (0-4). + +@item svtav1-params +Set SVT-AV1 options using a list of @var{key}=@var{value} pairs separated +by ":". See the SVT-AV1 encoder user guide for a list of accepted parameters. @end table ---- libavcodec/libsvtav1.c.orig 2021-10-24 20:47:07 UTC -+++ libavcodec/libsvtav1.c -@@ -60,17 +60,20 @@ typedef struct SvtContext { - EOS_STATUS eos_flag; - - // User options. -+ AVDictionary *svtav1_opts; -+#if FF_API_SVTAV1_OPTS - int hierarchical_level; - int la_depth; -- int enc_mode; -- int rc_mode; - int scd; -- int qp; - - int tier; - - int tile_columns; - int tile_rows; -+#endif -+ int enc_mode; -+ int crf; -+ int qp; - } SvtContext; - - static const struct { -@@ -120,16 +123,12 @@ static int alloc_buffer(EbSvtAv1EncConfiguration *conf - - static int alloc_buffer(EbSvtAv1EncConfiguration *config, SvtContext *svt_enc) - { -- const int pack_mode_10bit = -- (config->encoder_bit_depth > 8) && (config->compressed_ten_bit_format == 0) ? 1 : 0; -- const size_t luma_size_8bit = -- config->source_width * config->source_height * (1 << pack_mode_10bit); -- const size_t luma_size_10bit = -- (config->encoder_bit_depth > 8 && pack_mode_10bit == 0) ? luma_size_8bit : 0; -+ const size_t luma_size = config->source_width * config->source_height * -+ (config->encoder_bit_depth > 8 ? 2 : 1); - - EbSvtIOFormat *in_data; - -- svt_enc->raw_size = (luma_size_8bit + luma_size_10bit) * 3 / 2; -+ svt_enc->raw_size = luma_size * 3 / 2; - - // allocate buffer for in and out - svt_enc->in_buf = av_mallocz(sizeof(*svt_enc->in_buf)); -@@ -151,11 +150,136 @@ static int config_enc_params(EbSvtAv1EncConfiguration - { - SvtContext *svt_enc = avctx->priv_data; - const AVPixFmtDescriptor *desc; -+ AVDictionaryEntry *en = NULL; - -+ // Update param from options -+#if FF_API_SVTAV1_OPTS -+ if (svt_enc->hierarchical_level >= 0) -+ param->hierarchical_levels = svt_enc->hierarchical_level; -+ if (svt_enc->tier >= 0) -+ param->tier = svt_enc->tier; -+ if (svt_enc->scd >= 0) -+ param->scene_change_detection = svt_enc->scd; -+ if (svt_enc->tile_columns >= 0) -+ param->tile_columns = svt_enc->tile_columns; -+ if (svt_enc->tile_rows >= 0) -+ param->tile_rows = svt_enc->tile_rows; -+ -+ if (svt_enc->la_depth >= 0) -+ param->look_ahead_distance = svt_enc->la_depth; -+#endif -+ -+ if (svt_enc->enc_mode >= 0) -+ param->enc_mode = svt_enc->enc_mode; -+ -+ if (avctx->bit_rate) { -+ param->target_bit_rate = avctx->bit_rate; -+ if (avctx->rc_max_rate != avctx->bit_rate) -+ param->rate_control_mode = 1; -+ else -+ param->rate_control_mode = 2; -+ -+ param->max_qp_allowed = avctx->qmax; -+ param->min_qp_allowed = avctx->qmin; -+ } -+ param->max_bit_rate = avctx->rc_max_rate; -+ if (avctx->bit_rate && avctx->rc_buffer_size) -+ param->maximum_buffer_size_ms = avctx->rc_buffer_size * 1000LL / avctx->bit_rate; -+ -+ if (svt_enc->crf > 0) { -+ param->qp = svt_enc->crf; -+ param->rate_control_mode = 0; -+ } else if (svt_enc->qp > 0) { -+ param->qp = svt_enc->qp; -+ param->rate_control_mode = 0; -+#if SVT_AV1_CHECK_VERSION(4, 0, 0) -+ param->aq_mode = 0; -+#else -+ param->enable_adaptive_quantization = 0; -+#endif -+ } -+ -+ desc = av_pix_fmt_desc_get(avctx->pix_fmt); -+ param->color_primaries = avctx->color_primaries; -+ param->matrix_coefficients = (desc->flags & AV_PIX_FMT_FLAG_RGB) ? -+ AVCOL_SPC_RGB : avctx->colorspace; -+ param->transfer_characteristics = avctx->color_trc; -+ -+ if (avctx->color_range != AVCOL_RANGE_UNSPECIFIED) -+ param->color_range = avctx->color_range == AVCOL_RANGE_JPEG; -+ else -+ param->color_range = !!(desc->flags & AV_PIX_FMT_FLAG_RGB); -+ -+#if SVT_AV1_CHECK_VERSION(1, 0, 0) -+ if (avctx->chroma_sample_location != AVCHROMA_LOC_UNSPECIFIED) { -+ const char *name = -+ av_chroma_location_name(avctx->chroma_sample_location); -+ -+ switch (avctx->chroma_sample_location) { -+ case AVCHROMA_LOC_LEFT: -+ param->chroma_sample_position = EB_CSP_VERTICAL; -+ break; -+ case AVCHROMA_LOC_TOPLEFT: -+ param->chroma_sample_position = EB_CSP_COLOCATED; -+ break; -+ default: -+ if (!name) -+ break; -+ -+ av_log(avctx, AV_LOG_WARNING, -+ "Specified chroma sample location %s is unsupported " -+ "on the AV1 bit stream level. Usage of a container that " -+ "allows passing this information - such as Matroska - " -+ "is recommended.\n", -+ name); -+ break; -+ } -+ } -+#endif -+ -+ if (avctx->profile != FF_PROFILE_UNKNOWN) -+ param->profile = avctx->profile; -+ -+ if (avctx->level != FF_LEVEL_UNKNOWN) -+ param->level = avctx->level; -+ -+ if (avctx->gop_size > 0) -+ param->intra_period_length = avctx->gop_size - 1; -+ -+ if (avctx->framerate.num > 0 && avctx->framerate.den > 0) { -+ param->frame_rate_numerator = avctx->framerate.num; -+ param->frame_rate_denominator = avctx->framerate.den; -+ } else { -+ param->frame_rate_numerator = avctx->time_base.den; -+ param->frame_rate_denominator = avctx->time_base.num * avctx->ticks_per_frame; -+ } -+ -+ /* 2 = IDR, closed GOP, 1 = CRA, open GOP */ -+ param->intra_refresh_type = avctx->flags & AV_CODEC_FLAG_CLOSED_GOP ? 2 : 1; -+ -+#if SVT_AV1_CHECK_VERSION(0, 9, 1) -+ while ((en = av_dict_get(svt_enc->svtav1_opts, "", en, AV_DICT_IGNORE_SUFFIX))) { -+ EbErrorType ret = svt_av1_enc_parse_parameter(param, en->key, en->value); -+ if (ret != EB_ErrorNone) { -+ int level = (avctx->err_recognition & AV_EF_EXPLODE) ? AV_LOG_ERROR : AV_LOG_WARNING; -+ av_log(avctx, level, "Error parsing option %s: %s.\n", en->key, en->value); -+ if (avctx->err_recognition & AV_EF_EXPLODE) -+ return AVERROR(EINVAL); -+ } -+ } -+#else -+ if ((en = av_dict_get(svt_enc->svtav1_opts, "", NULL, AV_DICT_IGNORE_SUFFIX))) { -+ int level = (avctx->err_recognition & AV_EF_EXPLODE) ? AV_LOG_ERROR : AV_LOG_WARNING; -+ av_log(avctx, level, "svt-params needs libavcodec to be compiled with SVT-AV1 " -+ "headers >= 0.9.1.\n"); -+ if (avctx->err_recognition & AV_EF_EXPLODE) -+ return AVERROR(ENOSYS); -+ } -+#endif -+ - param->source_width = avctx->width; - param->source_height = avctx->height; - -- desc = av_pix_fmt_desc_get(avctx->pix_fmt); - param->encoder_bit_depth = desc->comp[0].depth; - - if (desc->log2_chroma_w == 1 && desc->log2_chroma_h == 1) -@@ -169,12 +289,6 @@ static int config_enc_params(EbSvtAv1EncConfiguration - return AVERROR(EINVAL); - } - -- if (avctx->profile != FF_PROFILE_UNKNOWN) -- param->profile = avctx->profile; -- -- if (avctx->level != FF_LEVEL_UNKNOWN) -- param->level = avctx->level; -- - if ((param->encoder_color_format == EB_YUV422 || param->encoder_bit_depth > 10) - && param->profile != FF_PROFILE_AV1_PROFESSIONAL ) { - av_log(avctx, AV_LOG_WARNING, "Forcing Professional profile\n"); -@@ -184,40 +298,21 @@ static int config_enc_params(EbSvtAv1EncConfiguration - param->profile = FF_PROFILE_AV1_HIGH; - } - -- // Update param from options -- param->hierarchical_levels = svt_enc->hierarchical_level; -- param->enc_mode = svt_enc->enc_mode; -- param->tier = svt_enc->tier; -- param->rate_control_mode = svt_enc->rc_mode; -- param->scene_change_detection = svt_enc->scd; -- param->qp = svt_enc->qp; -+ avctx->bit_rate = param->rate_control_mode > 0 ? -+ param->target_bit_rate : 0; -+ avctx->rc_max_rate = param->max_bit_rate; -+ avctx->rc_buffer_size = param->maximum_buffer_size_ms * avctx->bit_rate / 1000LL; - -- param->target_bit_rate = avctx->bit_rate; -+ if (avctx->bit_rate || avctx->rc_max_rate || avctx->rc_buffer_size) { -+ AVCPBProperties *cpb_props = ff_add_cpb_side_data(avctx); -+ if (!cpb_props) -+ return AVERROR(ENOMEM); - -- if (avctx->gop_size > 0) -- param->intra_period_length = avctx->gop_size - 1; -- -- if (avctx->framerate.num > 0 && avctx->framerate.den > 0) { -- param->frame_rate_numerator = avctx->framerate.num; -- param->frame_rate_denominator = avctx->framerate.den; -- } else { -- param->frame_rate_numerator = avctx->time_base.den; -- param->frame_rate_denominator = avctx->time_base.num * avctx->ticks_per_frame; -+ cpb_props->buffer_size = avctx->rc_buffer_size; -+ cpb_props->max_bitrate = avctx->rc_max_rate; -+ cpb_props->avg_bitrate = avctx->bit_rate; - } - -- if (param->rate_control_mode) { -- param->max_qp_allowed = avctx->qmax; -- param->min_qp_allowed = avctx->qmin; -- } -- -- param->intra_refresh_type = 2; /* Real keyframes only */ -- -- if (svt_enc->la_depth >= 0) -- param->look_ahead_distance = svt_enc->la_depth; -- -- param->tile_columns = svt_enc->tile_columns; -- param->tile_rows = svt_enc->tile_rows; -- - return 0; - } - -@@ -330,11 +425,8 @@ static int eb_send_frame(AVCodecContext *avctx, const - if (svt_enc->eos_flag == EOS_SENT) - return 0; - -- headerPtrLast.n_alloc_len = 0; -- headerPtrLast.n_filled_len = 0; -- headerPtrLast.n_tick_count = 0; -- headerPtrLast.p_app_private = NULL; -- headerPtrLast.p_buffer = NULL; -+ memset(&headerPtrLast, 0, sizeof(headerPtrLast)); -+ headerPtrLast.pic_type = EB_AV1_INVALID_PICTURE; - headerPtrLast.flags = EB_BUFFERFLAG_EOS; - - svt_av1_enc_send_picture(svt_enc->svt_handle, &headerPtrLast); -@@ -350,6 +442,16 @@ static int eb_send_frame(AVCodecContext *avctx, const - headerPtr->p_app_private = NULL; - headerPtr->pts = frame->pts; - -+ switch (frame->pict_type) { -+ case AV_PICTURE_TYPE_I: -+ headerPtr->pic_type = EB_AV1_KEY_PICTURE; -+ break; -+ default: -+ // Actually means auto, or default. -+ headerPtr->pic_type = EB_AV1_INVALID_PICTURE; -+ break; -+ } -+ - svt_av1_enc_send_picture(svt_enc->svt_handle, headerPtr); - - return 0; -@@ -472,21 +574,22 @@ static const AVOption options[] = { - #define OFFSET(x) offsetof(SvtContext, x) - #define VE AV_OPT_FLAG_VIDEO_PARAM | AV_OPT_FLAG_ENCODING_PARAM - static const AVOption options[] = { -- { "hielevel", "Hierarchical prediction levels setting", OFFSET(hierarchical_level), -- AV_OPT_TYPE_INT, { .i64 = 4 }, 3, 4, VE , "hielevel"}, -+#if FF_API_SVTAV1_OPTS -+ { "hielevel", "Hierarchical prediction levels setting (Deprecated, use svtav1-params)", OFFSET(hierarchical_level), -+ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 4, VE | AV_OPT_FLAG_DEPRECATED , "hielevel"}, - { "3level", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 3 }, INT_MIN, INT_MAX, VE, "hielevel" }, - { "4level", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 4 }, INT_MIN, INT_MAX, VE, "hielevel" }, - -- { "la_depth", "Look ahead distance [0, 120]", OFFSET(la_depth), -- AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 120, VE }, -+ { "la_depth", "Look ahead distance [0, 120] (Deprecated, use svtav1-params)", OFFSET(la_depth), -+ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 120, VE | AV_OPT_FLAG_DEPRECATED }, - -- { "preset", "Encoding preset [0, 8]", -- OFFSET(enc_mode), AV_OPT_TYPE_INT, { .i64 = MAX_ENC_PRESET }, 0, MAX_ENC_PRESET, VE }, -- -- { "tier", "Set operating point tier", OFFSET(tier), -- AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 1, VE, "tier" }, -+ { "tier", "Set operating point tier (Deprecated, use svtav1-params)", OFFSET(tier), -+ AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 1, VE | AV_OPT_FLAG_DEPRECATED, "tier" }, - { "main", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, 0, 0, VE, "tier" }, - { "high", NULL, 0, AV_OPT_TYPE_CONST, { .i64 = 1 }, 0, 0, VE, "tier" }, -+#endif -+ { "preset", "Encoding preset", -+ OFFSET(enc_mode), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, MAX_ENC_PRESET, VE }, - - FF_AV1_PROFILE_OPTS - -@@ -518,21 +621,20 @@ static const AVOption options[] = { - { LEVEL("7.3", 73) }, - #undef LEVEL - -- { "rc", "Bit rate control mode", OFFSET(rc_mode), -- AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 3, VE , "rc"}, -- { "cqp", "Constant quantizer", 0, AV_OPT_TYPE_CONST, { .i64 = 0 }, INT_MIN, INT_MAX, VE, "rc" }, -- { "vbr", "Variable Bit Rate, use a target bitrate for the entire stream", 0, AV_OPT_TYPE_CONST, { .i64 = 1 }, INT_MIN, INT_MAX, VE, "rc" }, -- { "cvbr", "Constrained Variable Bit Rate, use a target bitrate for each GOP", 0, AV_OPT_TYPE_CONST,{ .i64 = 2 }, INT_MIN, INT_MAX, VE, "rc" }, -+ { "crf", "Constant Rate Factor value", OFFSET(crf), -+ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 63, VE }, -+ { "qp", "Initial Quantizer level value", OFFSET(qp), -+ AV_OPT_TYPE_INT, { .i64 = 0 }, 0, 63, VE }, -+#if FF_API_SVTAV1_OPTS -+ { "sc_detection", "Scene change detection (Deprecated, use svtav1-params)", OFFSET(scd), -+ AV_OPT_TYPE_BOOL, { .i64 = -1 }, -1, 1, VE | AV_OPT_FLAG_DEPRECATED }, - -- { "qp", "Quantizer to use with cqp rate control mode", OFFSET(qp), -- AV_OPT_TYPE_INT, { .i64 = 50 }, 0, 63, VE }, -+ { "tile_columns", "Log2 of number of tile columns to use (Deprecated, use svtav1-params)", OFFSET(tile_columns), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 4, VE | AV_OPT_FLAG_DEPRECATED }, -+ { "tile_rows", "Log2 of number of tile rows to use (Deprecated, use svtav1-params)", OFFSET(tile_rows), AV_OPT_TYPE_INT, { .i64 = -1 }, -1, 6, VE | AV_OPT_FLAG_DEPRECATED }, -+#endif - -- { "sc_detection", "Scene change detection", OFFSET(scd), -- AV_OPT_TYPE_BOOL, { .i64 = 0 }, 0, 1, VE }, -+ { "svtav1-params", "Set the SVT-AV1 configuration using a :-separated list of key=value parameters", OFFSET(svtav1_opts), AV_OPT_TYPE_DICT, { 0 }, 0, 0, VE }, - -- { "tile_columns", "Log2 of number of tile columns to use", OFFSET(tile_columns), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 4, VE}, -- { "tile_rows", "Log2 of number of tile rows to use", OFFSET(tile_rows), AV_OPT_TYPE_INT, {.i64 = 0}, 0, 6, VE}, -- - {NULL}, - }; - -@@ -544,9 +646,10 @@ static const AVCodecDefault eb_enc_defaults[] = { - }; - - static const AVCodecDefault eb_enc_defaults[] = { -- { "b", "7M" }, -+ { "b", "0" }, -+ { "flags", "+cgop" }, - { "g", "-1" }, -- { "qmin", "0" }, -+ { "qmin", "1" }, - { "qmax", "63" }, - { NULL }, - }; -@@ -561,12 +664,11 @@ AVCodec ff_libsvtav1_encoder = { - .receive_packet = eb_receive_packet, - .close = eb_enc_close, - .capabilities = AV_CODEC_CAP_DELAY | AV_CODEC_CAP_OTHER_THREADS, -- .caps_internal = FF_CODEC_CAP_AUTO_THREADS, -+ .caps_internal = FF_CODEC_CAP_AUTO_THREADS | FF_CODEC_CAP_INIT_CLEANUP, - .pix_fmts = (const enum AVPixelFormat[]){ AV_PIX_FMT_YUV420P, - AV_PIX_FMT_YUV420P10, - AV_PIX_FMT_NONE }, - .priv_class = &class, - .defaults = eb_enc_defaults, -- .caps_internal = FF_CODEC_CAP_INIT_CLEANUP, - .wrapper_name = "libsvtav1", - }; ---- libavcodec/version.h.orig 2021-10-24 20:47:07 UTC +--- libavcodec/version.h.orig 2025-05-17 13:52:31 UTC +++ libavcodec/version.h @@ -168,5 +168,8 @@ #ifndef FF_API_INIT_PACKET #define FF_API_INIT_PACKET (LIBAVCODEC_VERSION_MAJOR < 60) #endif +#ifndef FF_API_SVTAV1_OPTS +#define FF_API_SVTAV1_OPTS (LIBAVCODEC_VERSION_MAJOR < 60) +#endif #endif /* AVCODEC_VERSION_H */ diff --git a/multimedia/ffmpeg4/files/patch-vmaf b/multimedia/ffmpeg4/files/patch-vmaf index abe772274d3c..0ecfe4377b8b 100644 --- a/multimedia/ffmpeg4/files/patch-vmaf +++ b/multimedia/ffmpeg4/files/patch-vmaf @@ -1,951 +1,929 @@ -https://git.ffmpeg.org/gitweb/ffmpeg.git/commitdiff/3d29724c008d - ---- configure.orig 2021-10-24 20:47:11 UTC -+++ configure -@@ -3663,7 +3663,7 @@ vidstabtransform_filter_deps="libvidstab" - vaguedenoiser_filter_deps="gpl" - vidstabdetect_filter_deps="libvidstab" - vidstabtransform_filter_deps="libvidstab" --libvmaf_filter_deps="libvmaf pthreads" -+libvmaf_filter_deps="libvmaf" - zmq_filter_deps="libzmq" - zoompan_filter_deps="swscale" - zscale_filter_deps="libzimg const_nan" -@@ -6441,7 +6441,7 @@ enabled libvidstab && require_pkg_config libvid - enabled libuavs3d && require_pkg_config libuavs3d "uavs3d >= 1.1.41" uavs3d.h uavs3d_decode - enabled libv4l2 && require_pkg_config libv4l2 libv4l2 libv4l2.h v4l2_ioctl - enabled libvidstab && require_pkg_config libvidstab "vidstab >= 0.98" vid.stab/libvidstab.h vsMotionDetectInit --enabled libvmaf && require_pkg_config libvmaf "libvmaf >= 1.5.2" libvmaf.h compute_vmaf -+enabled libvmaf && require_pkg_config libvmaf "libvmaf >= 2.0.0" libvmaf.h vmaf_init - enabled libvo_amrwbenc && require libvo_amrwbenc vo-amrwbenc/enc_if.h E_IF_init -lvo-amrwbenc - enabled libvorbis && require_pkg_config libvorbis vorbis vorbis/codec.h vorbis_info_init && - require_pkg_config libvorbisenc vorbisenc vorbis/vorbisenc.h vorbis_encode_init ---- doc/filters.texi.orig 2021-10-24 20:47:07 UTC +--- doc/filters.texi.orig 2025-05-17 13:52:31 UTC +++ doc/filters.texi @@ -13867,66 +13867,58 @@ ffmpeg -i input.mov -vf lensfun=make=Canon:model="Cano @section libvmaf -Obtain the VMAF (Video Multi-Method Assessment Fusion) -score between two input videos. +Calulate the VMAF (Video Multi-Method Assessment Fusion) score for a +reference/distorted pair of input videos. The obtained VMAF score is printed through the logging system. It requires Netflix's vmaf library (libvmaf) as a pre-requisite. After installing the library it can be enabled using: @code{./configure --enable-libvmaf}. -If no model path is specified it uses the default model: @code{vmaf_v0.6.1.pkl}. The filter has following options: @table @option +@item model +A `|` delimited list of vmaf models. Each model can be configured with a number of parameters. +Default value: @code{"version=vmaf_v0.6.1"} + @item model_path -Set the model path which is to be used for SVM. -Default value: @code{"/usr/local/share/model/vmaf_v0.6.1.pkl"} +Deprecated, use model='path=...'. -@item log_path -Set the file path to be used to store logs. - -@item log_fmt -Set the format of the log file (csv, json or xml). - @item enable_transform -This option can enable/disable the @code{score_transform} applied to the final predicted VMAF score, -if you have specified score_transform option in the input parameter file passed to @code{run_vmaf_training.py} -Default value: @code{false} +Deprecated, use model='enable_transform=true'. @item phone_model -Invokes the phone model which will generate VMAF scores higher than in the -regular model, which is more suitable for laptop, TV, etc. viewing conditions. -Default value: @code{false} +Deprecated, use model='enable_transform=true'. +@item enable_conf_interval +Deprecated, use model='enable_conf_interval=true'. + +@item feature +A `|` delimited list of features. Each feature can be configured with a number of parameters. + @item psnr -Enables computing psnr along with vmaf. -Default value: @code{false} +Deprecated, use feature='name=psnr'. @item ssim -Enables computing ssim along with vmaf. -Default value: @code{false} +Deprecated, use feature='name=ssim'. @item ms_ssim -Enables computing ms_ssim along with vmaf. -Default value: @code{false} +Deprecated, use feature='name=ms_ssim'. -@item pool -Set the pool method to be used for computing vmaf. -Options are @code{min}, @code{harmonic_mean} or @code{mean} (default). +@item log_path +Set the file path to be used to store log files. +@item log_fmt +Set the format of the log file (xml, json, csv, or sub). + @item n_threads -Set number of threads to be used when computing vmaf. -Default value: @code{0}, which makes use of all available logical processors. +Set number of threads to be used when initializing libvmaf. +Default value: @code{0}, no threads. @item n_subsample -Set interval for frame subsampling used when computing vmaf. -Default value: @code{1} - -@item enable_conf_interval -Enables confidence interval. -Default value: @code{false} +Set frame subsampling interval to be used. @end table This filter also supports the @ref{framesync} options. @@ -13934,23 +13926,31 @@ This filter also supports the @ref{framesync} options. @subsection Examples @itemize @item -On the below examples the input file @file{main.mpg} being processed is -compared with the reference file @file{ref.mpg}. +In the examples below, a distorted video @file{distorted.mpg} is +compared with a reference file @file{reference.mpg}. +@item +Basic usage: @example -ffmpeg -i main.mpg -i ref.mpg -lavfi libvmaf -f null - +ffmpeg -i distorted.mpg -i reference.mpg -lavfi libvmaf=log_path=output.xml -f null - @end example @item -Example with options: +Example with multiple models: @example -ffmpeg -i main.mpg -i ref.mpg -lavfi libvmaf="psnr=1:log_fmt=json" -f null - +ffmpeg -i distorted.mpg -i reference.mpg -lavfi libvmaf='model=version=vmaf_v0.6.1\\:name=vmaf|version=vmaf_v0.6.1neg\\:name=vmaf_neg' -f null - @end example @item +Example with multiple addtional features: +@example +ffmpeg -i distorted.mpg -i reference.mpg -lavfi libvmaf='feature=name=psnr|name=ciede' -f null - +@end example + +@item Example with options and different containers: @example -ffmpeg -i main.mpg -i ref.mkv -lavfi "[0:v]settb=AVTB,setpts=PTS-STARTPTS[main];[1:v]settb=AVTB,setpts=PTS-STARTPTS[ref];[main][ref]libvmaf=psnr=1:log_fmt=json" -f null - +ffmpeg -i distorted.mpg -i reference.mkv -lavfi "[0:v]settb=AVTB,setpts=PTS-STARTPTS[main];[1:v]settb=AVTB,setpts=PTS-STARTPTS[ref];[main][ref]libvmaf=log_fmt=json:log_path=output.json" -f null - @end example @end itemize ---- libavfilter/vf_libvmaf.c.orig 2021-10-24 20:47:07 UTC +--- libavfilter/vf_libvmaf.c.orig 2025-05-17 13:52:31 UTC +++ libavfilter/vf_libvmaf.c @@ -24,8 +24,8 @@ * Calculate the VMAF between two input videos. */ -#include #include + #include "libavutil/avstring.h" #include "libavutil/opt.h" #include "libavutil/pixdesc.h" @@ -39,23 +39,9 @@ typedef struct LIBVMAFContext { typedef struct LIBVMAFContext { const AVClass *class; FFFrameSync fs; - const AVPixFmtDescriptor *desc; - int width; - int height; - double vmaf_score; - int vmaf_thread_created; - pthread_t vmaf_thread; - pthread_mutex_t lock; - pthread_cond_t cond; - int eof; - AVFrame *gmain; - AVFrame *gref; - int frame_set; char *model_path; char *log_path; char *log_fmt; - int disable_clip; - int disable_avx; int enable_transform; int phone_model; int psnr; @@ -65,185 +51,488 @@ typedef struct LIBVMAFContext { int n_threads; int n_subsample; int enable_conf_interval; - int error; + char *model_cfg; + char *feature_cfg; + VmafContext *vmaf; + VmafModel **model; + unsigned model_cnt; + unsigned frame_cnt; + unsigned bpc; } LIBVMAFContext; #define OFFSET(x) offsetof(LIBVMAFContext, x) #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM static const AVOption libvmaf_options[] = { - {"model_path", "Set the model to be used for computing vmaf.", OFFSET(model_path), AV_OPT_TYPE_STRING, {.str="/usr/local/share/model/vmaf_v0.6.1.pkl"}, 0, 1, FLAGS}, - {"log_path", "Set the file path to be used to store logs.", OFFSET(log_path), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 1, FLAGS}, - {"log_fmt", "Set the format of the log (csv, json or xml).", OFFSET(log_fmt), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 1, FLAGS}, - {"enable_transform", "Enables transform for computing vmaf.", OFFSET(enable_transform), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS}, - {"phone_model", "Invokes the phone model that will generate higher VMAF scores.", OFFSET(phone_model), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS}, - {"psnr", "Enables computing psnr along with vmaf.", OFFSET(psnr), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS}, - {"ssim", "Enables computing ssim along with vmaf.", OFFSET(ssim), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS}, - {"ms_ssim", "Enables computing ms-ssim along with vmaf.", OFFSET(ms_ssim), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS}, + {"model_path", "use model='path=...'.", OFFSET(model_path), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, + {"log_path", "Set the file path to be used to write log.", OFFSET(log_path), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 1, FLAGS}, + {"log_fmt", "Set the format of the log (csv, json, xml, or sub).", OFFSET(log_fmt), AV_OPT_TYPE_STRING, {.str="xml"}, 0, 1, FLAGS}, + {"enable_transform", "use model='enable_transform=true'.", OFFSET(enable_transform), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, + {"phone_model", "use model='enable_transform=true'.", OFFSET(phone_model), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, + {"psnr", "use feature='name=psnr'.", OFFSET(psnr), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, + {"ssim", "use feature='name=ssim'.", OFFSET(ssim), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, + {"ms_ssim", "use feature='name=ms_ssim'.", OFFSET(ms_ssim), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, {"pool", "Set the pool method to be used for computing vmaf.", OFFSET(pool), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 1, FLAGS}, {"n_threads", "Set number of threads to be used when computing vmaf.", OFFSET(n_threads), AV_OPT_TYPE_INT, {.i64=0}, 0, UINT_MAX, FLAGS}, {"n_subsample", "Set interval for frame subsampling used when computing vmaf.", OFFSET(n_subsample), AV_OPT_TYPE_INT, {.i64=1}, 1, UINT_MAX, FLAGS}, - {"enable_conf_interval", "Enables confidence interval.", OFFSET(enable_conf_interval), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS}, + {"enable_conf_interval", "model='enable_conf_interval=true'.", OFFSET(enable_conf_interval), AV_OPT_TYPE_BOOL, {.i64=0}, 0, 1, FLAGS|AV_OPT_FLAG_DEPRECATED}, + {"model", "Set the model to be used for computing vmaf.", OFFSET(model_cfg), AV_OPT_TYPE_STRING, {.str="version=vmaf_v0.6.1"}, 0, 1, FLAGS}, + {"feature", "Set the feature to be used for computing vmaf.", OFFSET(feature_cfg), AV_OPT_TYPE_STRING, {.str=NULL}, 0, 1, FLAGS}, { NULL } }; FRAMESYNC_DEFINE_CLASS(libvmaf, LIBVMAFContext, fs); -#define read_frame_fn(type, bits) \ - static int read_frame_##bits##bit(float *ref_data, float *main_data, \ - float *temp_data, int stride, void *ctx) \ -{ \ - LIBVMAFContext *s = (LIBVMAFContext *) ctx; \ - int ret; \ - \ - pthread_mutex_lock(&s->lock); \ - \ - while (!s->frame_set && !s->eof) { \ - pthread_cond_wait(&s->cond, &s->lock); \ - } \ - \ - if (s->frame_set) { \ - int ref_stride = s->gref->linesize[0]; \ - int main_stride = s->gmain->linesize[0]; \ - \ - const type *ref_ptr = (const type *) s->gref->data[0]; \ - const type *main_ptr = (const type *) s->gmain->data[0]; \ - \ - float *ptr = ref_data; \ - float factor = 1.f / (1 << (bits - 8)); \ - \ - int h = s->height; \ - int w = s->width; \ - \ - int i,j; \ - \ - for (i = 0; i < h; i++) { \ - for ( j = 0; j < w; j++) { \ - ptr[j] = ref_ptr[j] * factor; \ - } \ - ref_ptr += ref_stride / sizeof(*ref_ptr); \ - ptr += stride / sizeof(*ptr); \ - } \ - \ - ptr = main_data; \ - \ - for (i = 0; i < h; i++) { \ - for (j = 0; j < w; j++) { \ - ptr[j] = main_ptr[j] * factor; \ - } \ - main_ptr += main_stride / sizeof(*main_ptr); \ - ptr += stride / sizeof(*ptr); \ - } \ - } \ - \ - ret = !s->frame_set; \ - \ - av_frame_unref(s->gref); \ - av_frame_unref(s->gmain); \ - s->frame_set = 0; \ - \ - pthread_cond_signal(&s->cond); \ - pthread_mutex_unlock(&s->lock); \ - \ - if (ret) { \ - return 2; \ - } \ - \ - return 0; \ +static enum VmafPixelFormat pix_fmt_map(enum AVPixelFormat av_pix_fmt) +{ + switch (av_pix_fmt) { + case AV_PIX_FMT_YUV420P: + case AV_PIX_FMT_YUV420P10LE: + case AV_PIX_FMT_YUV420P12LE: + case AV_PIX_FMT_YUV420P16LE: + return VMAF_PIX_FMT_YUV420P; + case AV_PIX_FMT_YUV422P: + case AV_PIX_FMT_YUV422P10LE: + case AV_PIX_FMT_YUV422P12LE: + case AV_PIX_FMT_YUV422P16LE: + return VMAF_PIX_FMT_YUV422P; + case AV_PIX_FMT_YUV444P: + case AV_PIX_FMT_YUV444P10LE: + case AV_PIX_FMT_YUV444P12LE: + case AV_PIX_FMT_YUV444P16LE: + return VMAF_PIX_FMT_YUV444P; + default: + return VMAF_PIX_FMT_UNKNOWN; + } } -read_frame_fn(uint8_t, 8); -read_frame_fn(uint16_t, 10); +static int copy_picture_data(AVFrame *src, VmafPicture *dst, unsigned bpc) +{ + int err = vmaf_picture_alloc(dst, pix_fmt_map(src->format), bpc, + src->width, src->height); + if (err) + return AVERROR(ENOMEM); -static void compute_vmaf_score(LIBVMAFContext *s) + for (unsigned i = 0; i < 3; i++) { + uint8_t *src_data = src->data[i]; + uint8_t *dst_data = dst->data[i]; + for (unsigned j = 0; j < dst->h[i]; j++) { + memcpy(dst_data, src_data, sizeof(*dst_data) * dst->w[i]); + src_data += src->linesize[i]; + dst_data += dst->stride[i]; + } + } + + return 0; +} + +static int do_vmaf(FFFrameSync *fs) { - int (*read_frame)(float *ref_data, float *main_data, float *temp_data, - int stride, void *ctx); - char *format; + AVFilterContext *ctx = fs->parent; + LIBVMAFContext *s = ctx->priv; + VmafPicture pic_ref, pic_dist; + AVFrame *ref, *dist; + int err = 0; - if (s->desc->comp[0].depth <= 8) { - read_frame = read_frame_8bit; - } else { - read_frame = read_frame_10bit; + int ret = ff_framesync_dualinput_get(fs, &dist, &ref); + if (ret < 0) + return ret; + if (ctx->is_disabled || !ref) + return ff_filter_frame(ctx->outputs[0], dist); + + err = copy_picture_data(ref, &pic_ref, s->bpc); + if (err) { + av_log(s, AV_LOG_ERROR, "problem during vmaf_picture_alloc.\n"); + return AVERROR(ENOMEM); } - format = (char *) s->desc->name; + err = copy_picture_data(dist, &pic_dist, s->bpc); + if (err) { + av_log(s, AV_LOG_ERROR, "problem during vmaf_picture_alloc.\n"); + vmaf_picture_unref(&pic_ref); + return AVERROR(ENOMEM); + } - s->error = compute_vmaf(&s->vmaf_score, format, s->width, s->height, - read_frame, s, s->model_path, s->log_path, - s->log_fmt, 0, 0, s->enable_transform, - s->phone_model, s->psnr, s->ssim, - s->ms_ssim, s->pool, - s->n_threads, s->n_subsample, s->enable_conf_interval); + err = vmaf_read_pictures(s->vmaf, &pic_ref, &pic_dist, s->frame_cnt++); + if (err) { + av_log(s, AV_LOG_ERROR, "problem during vmaf_read_pictures.\n"); + return AVERROR(EINVAL); + } + + return ff_filter_frame(ctx->outputs[0], dist); } -static void *call_vmaf(void *ctx) + +static AVDictionary **delimited_dict_parse(char *str, unsigned *cnt) { - LIBVMAFContext *s = (LIBVMAFContext *) ctx; - compute_vmaf_score(s); - if (!s->error) { - av_log(ctx, AV_LOG_INFO, "VMAF score: %f\n",s->vmaf_score); - } else { - pthread_mutex_lock(&s->lock); - pthread_cond_signal(&s->cond); - pthread_mutex_unlock(&s->lock); + AVDictionary **dict = NULL; + char *str_copy = NULL; + char *saveptr = NULL; + unsigned cnt2; + int err = 0; + + if (!str) + return NULL; + + cnt2 = 1; + for (char *p = str; *p; p++) { + if (*p == '|') + cnt2++; } - pthread_exit(NULL); + + dict = av_calloc(cnt2, sizeof(*dict)); + if (!dict) + goto fail; + + str_copy = av_strdup(str); + if (!str_copy) + goto fail; + + *cnt = 0; + for (unsigned i = 0; i < cnt2; i++) { + char *s = av_strtok(i == 0 ? str_copy : NULL, "|", &saveptr); + if (!s) + continue; + err = av_dict_parse_string(&dict[(*cnt)++], s, "=", ":", 0); + if (err) + goto fail; + } + + av_free(str_copy); + return dict; + +fail: + if (dict) { + for (unsigned i = 0; i < *cnt; i++) { + if (dict[i]) + av_dict_free(&dict[i]); + } + av_free(dict); + } + + av_free(str_copy); + *cnt = 0; return NULL; } -static int do_vmaf(FFFrameSync *fs) +static int parse_features(AVFilterContext *ctx) { - AVFilterContext *ctx = fs->parent; LIBVMAFContext *s = ctx->priv; - AVFrame *master, *ref; - int ret; + AVDictionary **dict = NULL; + unsigned dict_cnt; + int err = 0; - ret = ff_framesync_dualinput_get(fs, &master, &ref); - if (ret < 0) - return ret; - if (!ref) - return ff_filter_frame(ctx->outputs[0], master); + if (!s->feature_cfg) + return 0; - pthread_mutex_lock(&s->lock); + dict = delimited_dict_parse(s->feature_cfg, &dict_cnt); + if (!dict) { + av_log(ctx, AV_LOG_ERROR, + "could not parse feature config: %s\n", s->feature_cfg); + return AVERROR(EINVAL); + } - while (s->frame_set && !s->error) { - pthread_cond_wait(&s->cond, &s->lock); + for (unsigned i = 0; i < dict_cnt; i++) { + char *feature_name = NULL; + VmafFeatureDictionary *feature_opts_dict = NULL; + AVDictionaryEntry *e = NULL; + + while (e = av_dict_get(dict[i], "", e, AV_DICT_IGNORE_SUFFIX)) { + if (av_stristr(e->key, "name")) { + feature_name = e->value; + continue; + } + + err = vmaf_feature_dictionary_set(&feature_opts_dict, e->key, + e->value); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "could not set feature option: %s.%s=%s\n", + feature_name, e->key, e->value); + goto exit; + } + } + + err = vmaf_use_feature(s->vmaf, feature_name, feature_opts_dict); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem during vmaf_use_feature: %s\n", feature_name); + goto exit; + } } - if (s->error) { +exit: + for (unsigned i = 0; i < dict_cnt; i++) { + if (dict[i]) + av_dict_free(&dict[i]); + } + av_free(dict); + return err; +} + +static int parse_models(AVFilterContext *ctx) +{ + LIBVMAFContext *s = ctx->priv; + AVDictionary **dict; + unsigned dict_cnt; + int err = 0; + + if (!s->model_cfg) return 0; + + dict_cnt = 0; + dict = delimited_dict_parse(s->model_cfg, &dict_cnt); + if (!dict) { av_log(ctx, AV_LOG_ERROR, - "libvmaf encountered an error, check log for details\n"); - pthread_mutex_unlock(&s->lock); + "could not parse model config: %s\n", s->model_cfg); return AVERROR(EINVAL); } - av_frame_ref(s->gref, ref); - av_frame_ref(s->gmain, master); + s->model_cnt = dict_cnt; + s->model = av_calloc(s->model_cnt, sizeof(*s->model)); + if (!s->model) + return AVERROR(ENOMEM); - s->frame_set = 1; + for (unsigned i = 0; i < dict_cnt; i++) { + VmafModelConfig model_cfg = { 0 }; + AVDictionaryEntry *e = NULL; + char *version = NULL; + char *path = NULL; - pthread_cond_signal(&s->cond); - pthread_mutex_unlock(&s->lock); + while (e = av_dict_get(dict[i], "", e, AV_DICT_IGNORE_SUFFIX)) { + if (av_stristr(e->key, "disable_clip")) { + model_cfg.flags |= av_stristr(e->value, "true") ? + VMAF_MODEL_FLAG_DISABLE_CLIP : 0; + continue; + } - return ff_filter_frame(ctx->outputs[0], master); + if (av_stristr(e->key, "enable_transform")) { + model_cfg.flags |= av_stristr(e->value, "true") ? + VMAF_MODEL_FLAG_ENABLE_TRANSFORM : 0; + continue; + } + + if (av_stristr(e->key, "name")) { + model_cfg.name = e->value; + continue; + } + + if (av_stristr(e->key, "version")) { + version = e->value; + continue; + } + + if (av_stristr(e->key, "path")) { + path = e->value; + continue; + } + } + + if (version) { + err = vmaf_model_load(&s->model[i], &model_cfg, version); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "could not load libvmaf model with version: %s\n", + version); + goto exit; + } + } + + if (path && !s->model[i]) { + err = vmaf_model_load_from_path(&s->model[i], &model_cfg, path); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "could not load libvmaf model with path: %s\n", + path); + goto exit; + } + } + + if (!s->model[i]) { + av_log(ctx, AV_LOG_ERROR, + "could not load libvmaf model with config: %s\n", + s->model_cfg); + goto exit; + } + + while (e = av_dict_get(dict[i], "", e, AV_DICT_IGNORE_SUFFIX)) { + VmafFeatureDictionary *feature_opts_dict = NULL; + char *feature_opt = NULL; + + char *feature_name = av_strtok(e->key, ".", &feature_opt); + if (!feature_opt) + continue; + + err = vmaf_feature_dictionary_set(&feature_opts_dict, + feature_opt, e->value); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "could not set feature option: %s.%s=%s\n", + feature_name, feature_opt, e->value); + err = AVERROR(EINVAL); + goto exit; + } + + err = vmaf_model_feature_overload(s->model[i], feature_name, + feature_opts_dict); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "could not overload feature: %s\n", feature_name); + err = AVERROR(EINVAL); + goto exit; + } + } + } + + for (unsigned i = 0; i < s->model_cnt; i++) { + err = vmaf_use_features_from_model(s->vmaf, s->model[i]); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem during vmaf_use_features_from_model\n"); + err = AVERROR(EINVAL); + goto exit; + } + } + +exit: + for (unsigned i = 0; i < dict_cnt; i++) { + if (dict[i]) + av_dict_free(&dict[i]); + } + av_free(dict); + return err; } +static enum VmafLogLevel log_level_map(int log_level) +{ + switch (log_level) { + case AV_LOG_QUIET: + return VMAF_LOG_LEVEL_NONE; + case AV_LOG_ERROR: + return VMAF_LOG_LEVEL_ERROR; + case AV_LOG_WARNING: + return VMAF_LOG_LEVEL_WARNING; + case AV_LOG_INFO: + return VMAF_LOG_LEVEL_INFO; + case AV_LOG_DEBUG: + return VMAF_LOG_LEVEL_DEBUG; + default: + return VMAF_LOG_LEVEL_INFO; + } +} + +static int parse_deprecated_options(AVFilterContext *ctx) +{ + LIBVMAFContext *s = ctx->priv; + VmafModel *model = NULL; + VmafModelCollection *model_collection = NULL; + enum VmafModelFlags flags = VMAF_MODEL_FLAGS_DEFAULT; + int err = 0; + + VmafModelConfig model_cfg = { + .name = "vmaf", + .flags = flags, + }; + + if (s->enable_transform || s->phone_model) + flags |= VMAF_MODEL_FLAG_ENABLE_TRANSFORM; + + if (!s->model_path) + goto extra_metrics_only; + + if (s->enable_conf_interval) { + err = vmaf_model_collection_load_from_path(&model, &model_collection, + &model_cfg, s->model_path); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading model file: %s\n", s->model_path); + goto exit; + } + + err = vmaf_use_features_from_model_collection(s->vmaf, model_collection); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading feature extractors from model file: %s\n", + s->model_path); + goto exit; + } + } else { + err = vmaf_model_load_from_path(&model, &model_cfg, s->model_path); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading model file: %s\n", s->model_path); + goto exit; + } + err = vmaf_use_features_from_model(s->vmaf, model); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading feature extractors from model file: %s\n", + s->model_path); + goto exit; + } + } + +extra_metrics_only: + if (s->psnr) { + VmafFeatureDictionary *d = NULL; + vmaf_feature_dictionary_set(&d, "enable_chroma", "false"); + + err = vmaf_use_feature(s->vmaf, "psnr", d); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading feature extractor: psnr\n"); + goto exit; + } + } + + if (s->ssim) { + err = vmaf_use_feature(s->vmaf, "float_ssim", NULL); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading feature extractor: ssim\n"); + goto exit; + } + } + + if (s->ms_ssim) { + err = vmaf_use_feature(s->vmaf, "float_ms_ssim", NULL); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem loading feature extractor: ms_ssim\n"); + goto exit; + } + } + +exit: + return err; +} + static av_cold int init(AVFilterContext *ctx) { LIBVMAFContext *s = ctx->priv; + int err = 0; - s->gref = av_frame_alloc(); - s->gmain = av_frame_alloc(); - if (!s->gref || !s->gmain) - return AVERROR(ENOMEM); + VmafConfiguration cfg = { + .log_level = log_level_map(av_log_get_level()), + .n_subsample = s->n_subsample, + .n_threads = s->n_threads, + }; - s->error = 0; + err = vmaf_init(&s->vmaf, cfg); + if (err) + return AVERROR(EINVAL); - s->vmaf_thread_created = 0; - pthread_mutex_init(&s->lock, NULL); - pthread_cond_init (&s->cond, NULL); + err = parse_deprecated_options(ctx); + if (err) + return err; + err = parse_models(ctx); + if (err) + return err; + + err = parse_features(ctx); + if (err) + return err; + s->fs.on_event = do_vmaf; return 0; } @@ -265,31 +554,36 @@ static int config_input_ref(AVFilterLink *inlink) static int config_input_ref(AVFilterLink *inlink) { - AVFilterContext *ctx = inlink->dst; + AVFilterContext *ctx = inlink->dst; LIBVMAFContext *s = ctx->priv; - int th; + const AVPixFmtDescriptor *desc; + int err = 0; - if (ctx->inputs[0]->w != ctx->inputs[1]->w || - ctx->inputs[0]->h != ctx->inputs[1]->h) { - av_log(ctx, AV_LOG_ERROR, "Width and height of input videos must be same.\n"); - return AVERROR(EINVAL); + if (ctx->inputs[0]->w != ctx->inputs[1]->w) { + av_log(ctx, AV_LOG_ERROR, "input width must match.\n"); + err |= AVERROR(EINVAL); } if (ctx->inputs[0]->format != ctx->inputs[1]->format) { av_log(ctx, AV_LOG_ERROR, "Inputs must be of same pixel format.\n"); return AVERROR(EINVAL); } - s->desc = av_pix_fmt_desc_get(inlink->format); - s->width = ctx->inputs[0]->w; - s->height = ctx->inputs[0]->h; + if (ctx->inputs[0]->h != ctx->inputs[1]->h) { + av_log(ctx, AV_LOG_ERROR, "input height must match.\n"); + err |= AVERROR(EINVAL); + } - th = pthread_create(&s->vmaf_thread, NULL, call_vmaf, (void *) s); - if (th) { - av_log(ctx, AV_LOG_ERROR, "Thread creation failed.\n"); - return AVERROR(EINVAL); + if (ctx->inputs[0]->format != ctx->inputs[1]->format) { + av_log(ctx, AV_LOG_ERROR, "input pix_fmt must match.\n"); + err |= AVERROR(EINVAL); } - s->vmaf_thread_created = 1; + if (err) + return err; + + desc = av_pix_fmt_desc_get(inlink->format); + s->bpc = desc->comp[0].depth; + return 0; } @@ -320,28 +614,80 @@ static int activate(AVFilterContext *ctx) return ff_framesync_activate(&s->fs); } +static enum VmafOutputFormat log_fmt_map(const char *log_fmt) +{ + if (log_fmt) { + if (av_stristr(log_fmt, "xml")) + return VMAF_OUTPUT_FORMAT_XML; + if (av_stristr(log_fmt, "json")) + return VMAF_OUTPUT_FORMAT_JSON; + if (av_stristr(log_fmt, "csv")) + return VMAF_OUTPUT_FORMAT_CSV; + if (av_stristr(log_fmt, "sub")) + return VMAF_OUTPUT_FORMAT_SUB; + } + + return VMAF_OUTPUT_FORMAT_XML; +} + +static enum VmafPoolingMethod pool_method_map(const char *pool_method) +{ + if (pool_method) { + if (av_stristr(pool_method, "min")) + return VMAF_POOL_METHOD_MIN; + if (av_stristr(pool_method, "mean")) + return VMAF_POOL_METHOD_MEAN; + if (av_stristr(pool_method, "harmonic_mean")) + return VMAF_POOL_METHOD_HARMONIC_MEAN; + } + + return VMAF_POOL_METHOD_MEAN; +} + static av_cold void uninit(AVFilterContext *ctx) { LIBVMAFContext *s = ctx->priv; + int err = 0; ff_framesync_uninit(&s->fs); - pthread_mutex_lock(&s->lock); - s->eof = 1; - pthread_cond_signal(&s->cond); - pthread_mutex_unlock(&s->lock); + if (!s->frame_cnt) + goto clean_up; - if (s->vmaf_thread_created) - { - pthread_join(s->vmaf_thread, NULL); - s->vmaf_thread_created = 0; + err = vmaf_read_pictures(s->vmaf, NULL, NULL, 0); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem flushing libvmaf context.\n"); } - av_frame_free(&s->gref); - av_frame_free(&s->gmain); + for (unsigned i = 0; i < s->model_cnt; i++) { + double vmaf_score; + err = vmaf_score_pooled(s->vmaf, s->model[i], pool_method_map(s->pool), + &vmaf_score, 0, s->frame_cnt - 1); + if (err) { + av_log(ctx, AV_LOG_ERROR, + "problem getting pooled vmaf score.\n"); + } - pthread_mutex_destroy(&s->lock); - pthread_cond_destroy(&s->cond); + av_log(ctx, AV_LOG_INFO, "VMAF score: %f\n", vmaf_score); + } + + if (s->vmaf) { + if (s->log_path && !err) + vmaf_write_output(s->vmaf, s->log_path, log_fmt_map(s->log_fmt)); + } + +clean_up: + if (s->model) { + for (unsigned i = 0; i < s->model_cnt; i++) { + if (s->model[i]) + vmaf_model_destroy(s->model[i]); + } + av_free(s->model); + } + + if (s->vmaf) + vmaf_close(s->vmaf); } static const AVFilterPad libvmaf_inputs[] = {