On Fri, Jun 13, 2014 at 2:22 PM, Docin <nicod.lyon+...@gmail.com> wrote:
>
> Hello everybody.
> At first, thank you for your grreat job!
>
> I encounter a probleme when trying to render a project generated with kdenlive
>
> My system:
> Intel(R) Core(TM) i5 CPU         760  @ 2.80GHz  (4 cores)
> Ubuntu 14.04 LTS
> 3.13.0-30-generic (i686)
>
> ram/swap
> free -mh
>              total       used       free     shared    buffers     cached
> Mem:          3,9G       3,1G       829M        68M        91M       1,3G
> -/+ buffers/cache:       1,8G       2,2G
> Swap:          11G       259M        11G
>
> My packages version
> ii  ffmpeg                                                      
> 4:0.8.12-0ubuntu0.12.04.1                           all          Multimedia 
> player, server, encoder and transcoder (transitional package)
> ii  kdenlive                                                    
> 0.9.8-1ubuntu1~sunab~trusty4                        i386         non-linear 
> video editor
> ii  kdenlive-data                                               
> 0.9.8-1ubuntu1~sunab~trusty4                        all          non-linear 
> video editor (data files)
> ii  kdenlive-dbg                                                
> 0.9.8-1ubuntu1~sunab~trusty4                        i386         non-linear 
> video editor (debugging symbols)
> ii  libaacs0:i386                                               0.7.0-1       
>                                       i386         free-and-libre 
> implementation of AACS
> ii  libfaac0:i386                                               1.28-6        
>                                       i386         AAC audio encoder (library)
> ii  libmlt++-dev                                                0.9.0-3       
>                                       i386         MLT multimedia framework 
> C++ wrapper (development)
> ii  libmlt++3                                                   0.9.0-3       
>                                       i386         MLT multimedia framework 
> C++ wrapper (runtime)
> ii  libmlt-data                                                 0.9.0-3       
>                                       all          multimedia framework (data)
> ii  libmlt-dbg                                                  0.9.0-3       
>                                       i386         multimedia framework 
> (debugging symbols)
> ii  libmlt-dev                                                  0.9.0-3       
>                                       i386         multimedia framework 
> (development)
> ii  libmlt6                                                     0.9.0-3       
>                                       i386         multimedia framework 
> (runtime)
> ii  libpostproc52                                               
> 6:0.git20120821-4                                   i386         FFmpeg 
> derived postprocessing library
> ii  libvo-aacenc0:i386                                          0.1.3-1       
>                                       i386         VisualOn AAC encoder 
> library
> ii  libx264-142:i386                                            
> 2:0.142.2389+git956c8d8-2                           i386         x264 video 
> coding library
> ii  libx264-dev:i386                                            
> 2:0.142.2389+git956c8d8-2                           i386         development 
> files for libx264
> ii  melt                                                        0.9.0-3       
>                                       i386         command line media player 
> and video editor
>
> My rendering command :
> /usr/bin/melt -verbose /home/foobar//NYv5.mlt -profile atsc_1080p_2997 
> -consumer avformat:/home/foobar/NY1.mp4 progress=1 acodec=aac ar=48000 
> ab=192k vcodec=libx264 vb=10000k aspect=@16/9 threads=4 real_time=-1
>
> when trying to render with libx264, on this PC, it always fail.

Thank you for providing good information. Unfortunately, it is still
very hard to say or know the source of the problem.

>
> I tried to analyse gdb backtrace:
>
> #0  __memcpy_ssse3_rep () at 
> ../sysdeps/i386/i686/multiarch/memcpy-ssse3-rep.S:181
> #1  0xb5bcbd06 in avcodec_decode_audio3 () from 
> /usr/lib/i386-linux-gnu/i686/cmov/libavcodec.so.54
> #2  0xb6f37f9b in decode_audio (fps=29,970029970029969, timecode=<optimized 
> out>, samples=1602, channels=2, pkt=..., ignore=0xae608be4, self=0x40d1c60) 
> at producer_avformat.c:2077

The crash occurs when decoding audio. More than likely, MLT gave
libavcodec some bad input, but this backtrace does not give us the
full details of all state. At this point, it does not appear to be a
problem related to encoding with x264. Are you also saying that you
can encode to something other than H.264?

There has not been many changes in producer_avformat.c since the 0.9.0
release, but there is one change related to audio. Even though it was
only observed on Windows, I suppose it could theoretically be a
problem elsewhere:
https://github.com/mltframework/mlt/commit/31cb473c0599cadfcab3e8d78d3585595751c720

> #3  producer_get_audio (frame=frame@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608d1c,
>     samples=samples@entry=0xae608fb8) at producer_avformat.c:2318
> #4  0xb77109d9 in mlt_frame_get_audio (self=self@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608d1c,
>     samples=samples@entry=0xae608fb8) at mlt_frame.c:711
> #5  0xb43e4ff6 in filter_get_audio (frame=frame@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608fb0,
>     samples=samples@entry=0xae608fb8) at filter_audiochannels.c:33
> #6  0xb77109d9 in mlt_frame_get_audio (self=self@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608fb0,
>     samples=samples@entry=0xae608fb8) at mlt_frame.c:711
> #7  0xb652ad93 in resample_get_audio (frame=frame@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608fb0,
>     samples=samples@entry=0xae608fb8) at filter_resample.c:54
> #8  0xb77109d9 in mlt_frame_get_audio (self=self@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608fb0,
>     samples=samples@entry=0xae608fb8) at mlt_frame.c:711
> #9  0xb6526d84 in filter_get_audio (frame=frame@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608fb0,
>     samples=samples@entry=0xae608fb8) at filter_volume.c:192
> #10 0xb77109d9 in mlt_frame_get_audio (self=self@entry=0x6bbea530, 
> buffer=buffer@entry=0xae608fa0, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae608fa8, channels=channels@entry=0xae608fb0,
>     samples=samples@entry=0xae608fb8) at mlt_frame.c:711
> #11 0xb43f3cb7 in combine_audio (samples=0xae6090cc, channels=0xae6090c4, 
> frequency=0xae6090bc, format=0xae609320, buffer=0xae6090b4, that=0x6bbea530, 
> this=0xadbc1370) at transition_mix.c:114
> #12 transition_get_audio (frame=frame@entry=0xadbc1370, 
> buffer=buffer@entry=0xae6090b4, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae6090bc, channels=channels@entry=0xae6090c4,
>     samples=samples@entry=0xae6090cc) at transition_mix.c:197
> #13 0xb77109d9 in mlt_frame_get_audio (self=self@entry=0xadbc1370, 
> buffer=buffer@entry=0xae6090b4, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae6090bc, channels=channels@entry=0xae6090c4,
>     samples=samples@entry=0xae6090cc) at mlt_frame.c:711
> #14 0xb43f3ce9 in combine_audio (samples=0xae60932c, channels=0xae609324, 
> frequency=0xae609328, format=0xae609320, buffer=0xae609330, that=0x392ede10, 
> this=0xadbc1370) at transition_mix.c:115
> #15 transition_get_audio (frame=frame@entry=0xadbc1370, 
> buffer=buffer@entry=0xae609330, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae609328, channels=channels@entry=0xae609324,
>     samples=samples@entry=0xae60932c) at transition_mix.c:197
> #16 0xb77109d9 in mlt_frame_get_audio (self=0xadbc1370, 
> buffer=buffer@entry=0xae609330, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae609328, channels=channels@entry=0xae609324,
>     samples=samples@entry=0xae60932c) at mlt_frame.c:711
> #17 0xb7727f0e in producer_get_audio (self=self@entry=0x25dce1f0, 
> buffer=buffer@entry=0xae609330, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae609328, channels=channels@entry=0xae609324,
>     samples=samples@entry=0xae60932c) at mlt_tractor.c:296
> #18 0xb77109d9 in mlt_frame_get_audio (self=0x25dce1f0, 
> buffer=buffer@entry=0xae609330, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae609328, channels=channels@entry=0xae609324,
>     samples=samples@entry=0xae60932c) at mlt_frame.c:711
> #19 0xb7727f0e in producer_get_audio (self=self@entry=0xadba0130, 
> buffer=buffer@entry=0xae609330, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae609328, channels=channels@entry=0xae609324,
>     samples=samples@entry=0xae60932c) at mlt_tractor.c:296
> #20 0xb77109d9 in mlt_frame_get_audio (self=self@entry=0xadba0130, 
> buffer=buffer@entry=0xae609330, format=format@entry=0xae609320, 
> frequency=frequency@entry=0xae609328, channels=channels@entry=0xae609324,
>     samples=samples@entry=0xae60932c) at mlt_frame.c:711
> #21 0xb7724dc9 in consumer_read_ahead_thread (arg=0xb929a958) at 
> mlt_consumer.c:861
> #22 0xb76f3f70 in start_thread (arg=0xae609b40) at pthread_create.c:312
> #23 0xb762970e in clone () at ../sysdeps/unix/sysv/linux/i386/clone.S:129
>
>
> Have you any idea of the problem ?

I suggest you closely monitor the progress of the render job. Note the
time or percentage of when the failure occurs and try to correlate
that to your timeline. Is there something on the timeline here that
could be causing a problem? It could be as simple as a clip that has
some data in it that confuses the code and triggers the bug. It is not
necessarily effects or fancy edits.

-- 
+-DRD-+

------------------------------------------------------------------------------
HPCC Systems Open Source Big Data Platform from LexisNexis Risk Solutions
Find What Matters Most in Your Big Data with HPCC Systems
Open Source. Fast. Scalable. Simple. Ideal for Dirty Data.
Leverages Graph Analysis for Fast Processing & Easy Data Exploration
http://p.sf.net/sfu/hpccsystems
_______________________________________________
Mlt-devel mailing list
Mlt-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mlt-devel

Reply via email to