Summary: | =media-video/ffmpeg-4.1.4[hardcoded-tables] fails to build: undefined reference to `avpriv_request_sample' | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Sergey S. Starikoff <Ikonta> |
Component: | Current packages | Assignee: | Gentoo Media-video project <media-video> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | arthurzam, dharding, jnerin |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | All | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Bug Depends on: | |||
Bug Blocks: | 653678 | ||
Attachments: |
build.log.xz
environment.xz |
Description
Sergey S. Starikoff
2019-07-18 07:13:30 UTC
Created attachment 583374 [details]
environment.xz
I'm using this USE combination: USE="X alsa bzip2 encode gpl hardcoded-tables iconv libass librtmp mp3 network openssl postproc threads vaapi vorbis x264 x265 zlib (-altivec) -amr -amrenc (-appkit) -bluray -bs2b -cdio -chromaprint -chromium -codec2 -cpudetection -debug -doc -fdk -flite -fontconfig -frei0r -fribidi -gcrypt -gme -gmp -gnutls -gsm -iec61883 -ieee1394 -jack -jpeg2k -kvazaar -ladspa -libaom -libcaca -libdrm -libilbc -libressl -libsoxr -libv4l -libxml2 -lv2 -lzma (-mipsdspr1) (-mipsdspr2) (-mipsfpu) (-mmal) -modplug -openal -opencl -opengl -openh264 -opus -oss -pic -pulseaudio -rubberband -samba -sdl -snappy -speex -srt -ssh -static-libs -svg -test -theora -truetype -twolame -v4l -vdpau -vpx -wavpack -webp -xcb -xvid -zeromq -zimg -zvbi" ABI_X86="(64) -32 (-x32)" CPU_FLAGS_X86="aes avx avx2 fma3 mmx mmxext sse sse2 sse3 sse4_1 sse4_2 ssse3 -3dnow -3dnowext -fma4 -xop" FFTOOLS="aviocat cws2fws ffescape ffeval ffhash fourcc2pixfmt graph2dot ismindex pktdumper qt-faststart sidxindex trasher" VIDEO_CARDS="-nvidia" Also fails with the mentioned above. But, using the patch c8232e50074f6f9f9b0674d0a5433f49d73a4e50, which can be found here https://github.com/FFmpeg/FFmpeg/commit/c8232e50074f6f9f9b0674d0a5433f49d73a4e50 . This fixed the building and passed the tests run. But, this patch sets the call to empty using a define, so I'm not sure if it is the correct way. Now running my system with this patch, hope it works. Does USE="hardcoded-tables" trigger this bug, then? Flag disabled: With USE="-hardcoded-tables" and without the patch: compile and test passes. With USE="-hardcoded-tables" and with the patch: compile and test passes. Flag enabled: With USE="hardcoded-tables" and without the patch: compile fails with this bug. With USE="hardcoded-tables" and with the patch: compile and test passes. So in conclusion, it happens only with USE="hardcoded-tables", and the patch should fix it (I hope the tests cover this functionality and there are no regressions). What is also interesting is that this patch is older that the 4.1.4 release, so it should have been included in upstream - so what happened? I can confirm, USE="hardcoded-tables" triggers this bug. You should add this to the ffmpeg-4.0 tracker, so that people actually notice it. (In reply to tt_1 from comment #5) > I can confirm, USE="hardcoded-tables" triggers this bug. > > You should add this to the ffmpeg-4.0 tracker, so that people actually > notice it. Probably, upstream reporting requires some more knowledge of package's build system, than I have. Could you do it? this is fixed in 4.2 Maybe it is a good idea to mark this bug in version 4.1.4 (maybe 4.1.*) in the tree? Maybe remove the USE flag itself for those versions? Maybe mask it for all affected versions? I'm afraid of some package hark dependency on <ffmpeg-4.2 and failing to build and repeating this bug. |