July 2014 - FFmpeg-trac - lists.ffmpeg.org

Re: [FFmpeg-trac] #2516(undetermined:new): pgssub produce warnings and are blinking
by FFmpeg 12 Jul '14

12 Jul '14

#2516: pgssub produce warnings and are blinking -------------------------------------+------------------------------------- Reporter: cehoyos | Owner: Type: defect | Status: new Priority: important | Component: Version: git-master | undetermined Keywords: pgssub | Resolution: regression videolan | Blocked By: Blocking: | Reproduced by developer: 0 Analyzed by developer: 0 | -------------------------------------+------------------------------------- Comment (by Cigaes): I am afraid this change would break rendering of DVB subtiles, since an empty subtitle frame is used to mark the end of the previous one. I need to have a look at this issue. (Fortunately, I will have more time soon to get back working on ffmpeg.) -- Ticket URL: <https://trac.ffmpeg.org/ticket/2516#comment:8> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #3243(avcodec:closed): ff_put_h264_chroma_mc8_neon iOS
by FFmpeg 12 Jul '14

12 Jul '14

#3243: ff_put_h264_chroma_mc8_neon iOS -------------------------------------+------------------------------------- Reporter: tarunseera | Owner: Type: defect | Status: closed Priority: important | Component: avcodec Version: 2.1.1 | Resolution: Keywords: crash | needs_more_info Blocking: | Blocked By: Analyzed by developer: 0 | Reproduced by developer: 0 -------------------------------------+------------------------------------- Changes (by cehoyos): * keywords: => crash * priority: normal => important * status: new => closed * resolution: => needs_more_info Comment: Please reopen this ticket if you can add any missing information. -- Ticket URL: <https://trac.ffmpeg.org/ticket/3243#comment:2> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #2516(undetermined:new): pgssub produce warnings and are blinking
by FFmpeg 12 Jul '14

12 Jul '14

#2516: pgssub produce warnings and are blinking -------------------------------------+------------------------------------- Reporter: cehoyos | Owner: Type: defect | Status: new Priority: important | Component: Version: git-master | undetermined Keywords: pgssub | Resolution: regression videolan | Blocked By: Blocking: | Reproduced by developer: 0 Analyzed by developer: 0 | -------------------------------------+------------------------------------- Comment (by michael): for ffmpeg the flickering can be fixed with: this needs more investigation though {{{ diff --git a/ffmpeg.c b/ffmpeg.c index 5a47510..508a630 100644 --- a/ffmpeg.c +++ b/ffmpeg.c @@ -2025,11 +2025,11 @@ static int transcode_subtitles(InputStream *ist, AVPacket *pkt, int *got_output) if (!*got_output) return ret; - sub2video_update(ist, &subtitle); - if (!subtitle.num_rects) goto out; + sub2video_update(ist, &subtitle); + ist->frames_decoded++; for (i = 0; i < nb_output_streams; i++) { }}} -- Ticket URL: <https://trac.ffmpeg.org/ticket/2516#comment:7> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #3115(avcodec:open): hevc: crash with threads 1 (fuzzed file)
by FFmpeg 11 Jul '14

11 Jul '14

#3115: hevc: crash with threads 1 (fuzzed file) -------------------------------------+------------------------------------- Reporter: ami_stuff | Owner: Type: defect | Status: open Priority: important | Component: avcodec Version: git-master | Resolution: Keywords: hevc crash | Blocked By: regression | Reproduced by developer: 1 Blocking: | Analyzed by developer: 0 | -------------------------------------+------------------------------------- Comment (by cehoyos): Both Helgrind and clang-tsan report possible race conditions. -- Ticket URL: <https://trac.ffmpeg.org/ticket/3115#comment:8> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #3115(avcodec:open): hevc: crash with threads 1 (fuzzed file)
by FFmpeg 11 Jul '14

11 Jul '14

#3115: hevc: crash with threads 1 (fuzzed file) -------------------------------------+------------------------------------- Reporter: ami_stuff | Owner: Type: defect | Status: open Priority: important | Component: avcodec Version: git-master | Resolution: Keywords: hevc crash | Blocked By: regression | Reproduced by developer: 1 Blocking: | Analyzed by developer: 0 | -------------------------------------+------------------------------------- Changes (by cehoyos): * keywords: hevc crash => hevc crash regression Comment: Regression since cb148e56 -- Ticket URL: <https://trac.ffmpeg.org/ticket/3115#comment:7> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #3115(avcodec:open): hevc: crash with threads 1 (fuzzed file)
by FFmpeg 11 Jul '14

11 Jul '14

#3115: hevc: crash with threads 1 (fuzzed file) ------------------------------------+----------------------------------- Reporter: ami_stuff | Owner: Type: defect | Status: open Priority: important | Component: avcodec Version: git-master | Resolution: Keywords: hevc crash | Blocked By: Blocking: | Reproduced by developer: 1 Analyzed by developer: 0 | ------------------------------------+----------------------------------- Comment (by michael): Replying to [comment:5 michael]: > not reproduceable under valgrind or address sanitizer > dumping all alloc & free addresses on a run that crashed shows no mismatches so this isnt a double free or freeing of a corrupted pointer, at least not in that single affected run also this one was with -threads 1 -- Ticket URL: <https://trac.ffmpeg.org/ticket/3115#comment:6> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #3115(avcodec:open): hevc: crash with threads 1 (fuzzed file)
by FFmpeg 11 Jul '14

11 Jul '14

#3115: hevc: crash with threads 1 (fuzzed file) ------------------------------------+----------------------------------- Reporter: ami_stuff | Owner: Type: defect | Status: open Priority: important | Component: avcodec Version: git-master | Resolution: Keywords: hevc crash | Blocked By: Blocking: | Reproduced by developer: 1 Analyzed by developer: 0 | ------------------------------------+----------------------------------- Comment (by michael): not reproduceable under valgrind or address sanitizer dumping all alloc & free addresses on a run that crashed shows no mismatches so this isnt a double free or freeing of a corrupted pointer, at least not in that single affected run is this a regression ? -- Ticket URL: <https://trac.ffmpeg.org/ticket/3115#comment:5> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

#3745(avcodec:new): error with libopenjpeg encoding
by FFmpeg 11 Jul '14

11 Jul '14

#3745: error with libopenjpeg encoding -------------------------------------+------------------------------------- Reporter: dericed | Owner: Type: defect | Status: new Priority: normal | Component: avcodec Version: git- | Keywords: libopenjpeg master | Blocking: Blocked By: | Analyzed by developer: 0 Reproduced by developer: 0 | -------------------------------------+------------------------------------- Summary of the bug: When using ffmpeg with libopenjpeg to make lossless encodings I use framemd5 to verify the losslessness. I found this sample that causes a glitch. How to reproduce: This command makes both a jpeg2000 output and a framemd5 output of the source file, which is available here https://archive.org/download/v210_cant_be_jpeg2000d/v210_cant_be_jpeg2000d.…. {{{ ffmpeg -y -i v210_cant_be_jpeg2000d.mov -c:v libopenjpeg -an isthisjpeg2000lossless.mov -f framemd5 -an v210_cant_be_jpeg2000d.framemd5 ffmpeg version N-64075-g47c8d1e Copyright (c) 2000-2014 the FFmpeg developers built on Jun 19 2014 10:18:25 with Apple LLVM version 5.1 (clang-503.0.40) (based on LLVM 3.4svn) configuration: --prefix=/usr/local/Cellar/ffmpeg/HEAD --enable-shared --enable-pthreads --enable-gpl --enable-version3 --enable-nonfree --enable-hardcoded-tables --enable-avresample --enable-vda --cc=clang --host-cflags= --host-ldflags= --enable-libx264 --enable-libfaac --enable- libmp3lame --enable-libxvid --enable-libfreetype --enable-ffplay --enable- libopenjpeg --extra- cflags='-I/usr/local/Cellar/openjpeg/1.5.1_1/include/openjpeg-1.5 ' libavutil 52. 89.100 / 52. 89.100 libavcodec 55. 67.100 / 55. 67.100 libavformat 55. 44.100 / 55. 44.100 libavdevice 55. 13.101 / 55. 13.101 libavfilter 4. 8.100 / 4. 8.100 libavresample 1. 3. 0 / 1. 3. 0 libswscale 2. 6.100 / 2. 6.100 libswresample 0. 19.100 / 0. 19.100 libpostproc 52. 3.100 / 52. 3.100 Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'v210_cant_be_jpeg2000d.mov': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt encoder : Lavf55.44.100 Duration: 00:00:01.02, start: 0.021333, bitrate: 219267 kb/s Stream #0:0(eng): Video: v210 (v210 / 0x30313276), yuv422p10le, 720x486, 223725 kb/s, 29.97 fps, 29.97 tbr, 30k tbn, 30k tbc (default) Metadata: handler_name : DataHandler encoder : Lavc55.68.100 v210 Stream #0:1(eng): Audio: aac (mp4a / 0x6134706D), 48000 Hz, stereo, fltp, 123 kb/s (default) Metadata: handler_name : DataHandler [mov @ 0x7fb81b00e400] Using AVStream.codec.time_base as a timebase hint to the muxer is deprecated. Set AVStream.time_base instead. [framemd5 @ 0x7fb81b01f200] Using AVStream.codec.time_base as a timebase hint to the muxer is deprecated. Set AVStream.time_base instead. Output #0, mov, to 'isthisjpeg2000lossless.mov': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt encoder : Lavf55.44.100 Stream #0:0(eng): Video: jpeg2000 (libopenjpeg) (mjp2 / 0x32706A6D), yuv422p10le, 720x486, q=2-31, 200 kb/s, 29.97 fps, 30k tbn, 29.97 tbc (default) Metadata: handler_name : DataHandler encoder : Lavc55.67.100 libopenjpeg Output #1, framemd5, to 'v210_cant_be_jpeg2000d.framemd5': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt encoder : Lavf55.44.100 Stream #1:0(eng): Video: rawvideo (Y3[10][10] / 0xA0A3359), yuv422p10le, 720x486, q=2-31, 200 kb/s, 29.97 fps, 29.97 tbn, 29.97 tbc (default) Metadata: handler_name : DataHandler encoder : Lavc55.67.100 rawvideo Stream mapping: Stream #0:0 -> #0:0 (v210 (native) -> jpeg2000 (libopenjpeg)) Stream #0:0 -> #1:0 (v210 (native) -> rawvideo (native)) Press [q] to stop, [?] for help frame= 31 fps=8.8 q=0.0 Lq=0.0 size= 13587kB time=00:00:01.03 bitrate=107608.2kbits/s dup=1 drop=0 video:54593kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown }}} To verify the losslessness I then make a framemd5 from the j2k output: {{{ ffmpeg -y -vcodec libopenjpeg -i isthisjpeg2000lossless.mov -f framemd5 -an isthisjpeg2000lossless.framemd5 ffmpeg version N-64075-g47c8d1e Copyright (c) 2000-2014 the FFmpeg developers built on Jun 19 2014 10:18:25 with Apple LLVM version 5.1 (clang-503.0.40) (based on LLVM 3.4svn) configuration: --prefix=/usr/local/Cellar/ffmpeg/HEAD --enable-shared --enable-pthreads --enable-gpl --enable-version3 --enable-nonfree --enable-hardcoded-tables --enable-avresample --enable-vda --cc=clang --host-cflags= --host-ldflags= --enable-libx264 --enable-libfaac --enable- libmp3lame --enable-libxvid --enable-libfreetype --enable-ffplay --enable- libopenjpeg --extra- cflags='-I/usr/local/Cellar/openjpeg/1.5.1_1/include/openjpeg-1.5 ' libavutil 52. 89.100 / 52. 89.100 libavcodec 55. 67.100 / 55. 67.100 libavformat 55. 44.100 / 55. 44.100 libavdevice 55. 13.101 / 55. 13.101 libavfilter 4. 8.100 / 4. 8.100 libavresample 1. 3. 0 / 1. 3. 0 libswscale 2. 6.100 / 2. 6.100 libswresample 0. 19.100 / 0. 19.100 libpostproc 52. 3.100 / 52. 3.100 Input #0, mov,mp4,m4a,3gp,3g2,mj2, from 'isthisjpeg2000lossless.mov': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt encoder : Lavf55.44.100 Duration: 00:00:01.04, start: 0.000000, bitrate: 107542 kb/s Stream #0:0(eng): Video: jpeg2000 (mjp2 / 0x32706A6D), yuv422p10le, 720x486, 107601 kb/s, 29.97 fps, 29.97 tbr, 30k tbn, 30k tbc (default) Metadata: handler_name : DataHandler encoder : Lavc55.67.100 libopenjpeg [framemd5 @ 0x7f859d811e00] Using AVStream.codec.time_base as a timebase hint to the muxer is deprecated. Set AVStream.time_base instead. Output #0, framemd5, to 'isthisjpeg2000lossless.framemd5': Metadata: major_brand : qt minor_version : 512 compatible_brands: qt encoder : Lavf55.44.100 Stream #0:0(eng): Video: rawvideo (Y3[10][10] / 0xA0A3359), yuv422p10le, 720x486, q=2-31, 200 kb/s, 29.97 fps, 29.97 tbn, 29.97 tbc (default) Metadata: handler_name : DataHandler encoder : Lavc55.67.100 rawvideo Stream mapping: Stream #0:0 -> #0:0 (jpeg2000 (libopenjpeg) -> rawvideo (native)) Press [q] to stop, [?] for help frame= 31 fps= 15 q=0.0 Lsize= 3kB time=00:00:01.03 bitrate= 20.3kbits/s video:42373kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: unknown }}} Then check the similarity of the framemd5 outputs: {{{ diff v210_cant_be_jpeg2000d.framemd5 isthisjpeg2000lossless.framemd5 6a7 > 0, 0, 0, 1, 1399680, b49bc926bc8d4734e861297e9bf63c96 23c24 < 0, 17, 17, 1, 1399680, 698ea483de551f4efc0cc9fbdebdcf61 --- > 0, 17, 17, 1, 1399680, 160cda3daf1402700944d2eaa322687e }}} With mov I often see this extra frame at the beginning, but I'm more concerned about frame 17. When I look at frame 17 in the j2k output it has bright red and green colors over the frame. The Y and U planes seem accurate but the V plane seems very odd. I posted an image of it here: https://twitter.com/dericed/status/482195897456275457 the image on the right shows U on the left and V on the right. Very odd. I reran the process hoping it was a fluke but got a consistently glitchy output for jpeg2000, though when I transcode the same source to ffv1 the framemd5s matched. -- Ticket URL: <https://trac.ffmpeg.org/ticket/3745> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 9

Re: [FFmpeg-trac] #2441(avformat:closed): Program detection in transport stream broken, part 2
by FFmpeg 11 Jul '14

11 Jul '14

#2441: Program detection in transport stream broken, part 2 -------------------------------------+------------------------------------- Reporter: cehoyos | Owner: Type: defect | Status: closed Priority: important | Component: avformat Version: git-master | Resolution: fixed Keywords: mpegts | Blocked By: regression | Reproduced by developer: 1 Blocking: | Analyzed by developer: 0 | -------------------------------------+------------------------------------- Changes (by michael): * status: new => closed * resolution: => fixed * reproduced: 0 => 1 Comment: Fixed in c6c172d1738b187fda1ad80cec357910be01ff58 -- Ticket URL: <https://trac.ffmpeg.org/ticket/2441#comment:4> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0

Re: [FFmpeg-trac] #2807(build system:closed): libopenjpeg v1.5.1 library discovery failed
by FFmpeg 11 Jul '14

11 Jul '14

#2807: libopenjpeg v1.5.1 library discovery failed -------------------------------------+------------------------------------- Reporter: floodkoff | Owner: Type: defect | Status: closed Priority: normal | Component: build Version: 2.0 | system Keywords: libopenjpeg | Resolution: Blocking: | worksforme Analyzed by developer: 0 | Blocked By: | Reproduced by developer: 0 -------------------------------------+------------------------------------- Changes (by cehoyos): * status: new => closed * resolution: => worksforme Comment: The error is: {{{ i686-w64-mingw32-gcc -D_ISOC99_SOURCE -D_FILE_OFFSET_BITS=64 -D_LARGEFILE_SOURCE -U__STRICT_ANSI__ -I/home/mybestfriend/openjpeg-1.5.1/libopenjpeg -std=c99 -include ./compat/msvcrt/snprintf.h -fomit-frame-pointer -c -o /tmp/ffconf.19XcSWiA.o /tmp/ffconf.5cH5dlGQ.c i686-w64-mingw32-gcc -L/home/mybestfriend/openjpeg-1.5.1/libopenjpeg/.libs -Wl,--nxcompat -Wl,--dynamicbase -Wl,--as-needed -o /tmp/ffconf.q7296gAS.exe /tmp/ffconf.19XcSWiA.o -lopenjpeg -lm -lpsapi -ladvapi32 -lshell32 /tmp/ffconf.19XcSWiA.o:ffconf.5cH5dlGQ.c:(.text+0xc): undefined reference to `_opj_version' }}} I often use a similar configure line and I can confirm that openjpeg-1.5.1 does contain this symbol here: {{{ $ nm ../openjpeg-1.5.1/libopenjpeg/.libs/libopenjpeg.so.1.5.0 |grep opj_version 00000000000118c0 T opj_version $ nm ../openjpeg-1.5.1/libopenjpeg/.libs/libopenjpeg.a |grep opj_version 0000000000000000 T opj_version }}} Either your library was miscompiled or your toolchain has a problem afaict. -- Ticket URL: <https://trac.ffmpeg.org/ticket/2807#comment:1> FFmpeg <https://ffmpeg.org> FFmpeg issue tracker

1 0