FFmbc 0.6 rc4 - output to two video files causes Segmentation fault

[GoogleCodeExporter]

Hi Baptiste,

I'm trying to output to two files at the same time. If I run these two commands 
separately then ffmbc encodes as I expect:

~/src/ffmbc/FFmbc-0.6-rc4b/ffmbc -y -i 576i25/Clock10tone.mov -vf scale=640:480 
-r 25 out1.mp4

~/src/ffmbc/FFmbc-0.6-rc4b/ffmbc -y -i 576i25/Clock10tone.mov -vf scale=320:240 
-r 25 out2.mp4

But if I combine that into one command I (often) get a seg-fault.

~/src/ffmbc/FFmbc-0.6-rc4b/ffmbc -y -i 576i25/Clock10tone.mov -vf scale=640:480 
-r 25 out1.mp4 -vf scale=320:240 -r 25 out2.mp4
FFmpeg version FFmbc-0.6-rc4, Copyright (c) 2000-2011 the FFmpeg developers
  built on Apr 20 2011 12:18:33 with gcc 4.4.3
  configuration: --cpu=i686 --enable-static --extra-cflags='--static -I$PREFIX/include' --extra-libs='-static -L$PREFIX/lib' --prefix= --enable-ffplay --enable-ffprobe --enable-ffserver --enable-gpl --enable-nonfree --enable-pthreads --enable-x11grab --enable-libxvid --enable-libvorbis --enable-libopenjpeg --enable-libspeex --enable-libtheora --enable-libfaac --enable-libgsm --enable-libdirac --enable-libmp3lame --enable-libx264 --enable-libschroedinger --enable-libvpx
  libavutil    50. 38. 0 / 50. 38. 0
  libavcodec   52.112. 1 / 52.112. 1
  libavformat  52. 99. 0 / 52. 99. 0
  libavdevice  52.  2. 3 / 52.  2. 3
  libavfilter   1. 76. 0 /  1. 76. 0
  libswscale    0. 12. 0 /  0. 12. 0
Input #0, mov,mp4,m4a,3gp,3g2,mj2, from '576i25/Clock10tone.mov':
  Metadata:
    major_brand: qt  
    minor_version: 537199360
    compatible_brands: qt  
    creation_time: 2009-07-16 17:23:31
  Duration: 00:00:10.00, start: 0.000000, bitrate: 30334 kb/s
    Stream #0.0(eng): Video: dvvideo, yuv420p, 720x576i bff [PAR 64:45 DAR 16:9], 28800 kb/s, 25.00 fps
    Stream #0.1(eng): Audio: pcm_s16le, 48000 Hz, 2 channels, s16, 1536 kb/s
[scale @ 0x9e0a0a0] w:720 h:576 fmt:yuv420p -> w:640 h:480 fmt:yuv420p 
flags:0x4 interlaced:0
[scale @ 0x9e20560] w:720 h:576 fmt:yuv420p -> w:320 h:240 fmt:yuv420p 
flags:0x4 interlaced:0
Output #0, mp4, to 'out1.mp4':
  Metadata:
    encoder: FFmbc 0.6
    Stream #0.0(und): Video: mpeg4, yuv420p, 640x480p [PAR 4:3 DAR 16:9], 200 kb/s, 25.00 fps
    Stream #0.1(und): Audio: libfaac, 48000 Hz, 2 channels, s16, 64 kb/s
Output #1, mp4, to 'out2.mp4':
  Metadata:
    encoder: FFmbc 0.6
    Stream #1.0(und): Video: mpeg4, yuv420p, 320x240p [PAR 4:3 DAR 16:9], 200 kb/s, 25.00 fps
    Stream #1.1(und): Audio: libfaac, 48000 Hz, 2 channels, s16, 64 kb/s
Stream mapping:
  Stream #0.0 -> #0.0
  Stream #0.1 -> #0.1
  Stream #0.0 -> #1.0
  Stream #0.1 -> #1.1
Press [q] to stop encoding
Segmentation fault

Running that through valgrind it is more likely to work, but sometimes will 
fail like this:

valgrind ~/src/ffmbc/FFmbc-0.6-rc4b/ffmbc_g -y -i 576i25/Clock10tone.mov -vf 
scale=640:480 -r 25 out1.mp4 -vf scale=320:240 -r 25 out2.mp4
[...]
==21667== Invalid read of size 4
==21667==    at 0x8A1FE1F: memcpy (in 
/usr/local/ssd/himslm01/src/ffmbc/FFmbc-0.6-rc4b/ffmbc_g)
==21667==    by 0x85DF1DA: av_image_copy (string3.h:52)
==21667==    by 0x80A0F95: request_frame (vsrc_buffer.c:131)
==21667==    by 0x808F60E: avfilter_request_frame (avfilter.c:362)
==21667==    by 0x808787D: get_filtered_video_frame (cmdutils.c:813)
==21667==    by 0x80816DF: output_packet (ffmbc.c:1850)
==21667==    by 0x8086208: T.733 (ffmbc.c:3070)
==21667==  Address 0x58a2000 is not stack'd, malloc'd or (recently) free'd
==21667== 
==21667== 
==21667== Process terminating with default action of signal 11 (SIGSEGV)
==21667==  Access not within mapped region at address 0x58A2000
==21667==    at 0x8A1FE1F: memcpy (in 
/usr/local/ssd/himslm01/src/ffmbc/FFmbc-0.6-rc4b/ffmbc_g)
==21667==    by 0x85DF1DA: av_image_copy (string3.h:52)
==21667==    by 0x80A0F95: request_frame (vsrc_buffer.c:131)
==21667==    by 0x808F60E: avfilter_request_frame (avfilter.c:362)
==21667==    by 0x808787D: get_filtered_video_frame (cmdutils.c:813)
==21667==    by 0x80816DF: output_packet (ffmbc.c:1850)
==21667==    by 0x8086208: T.733 (ffmbc.c:3070)
==21667==  If you believe this happened as a result of a stack
==21667==  overflow in your program's main thread (unlikely but
==21667==  possible), you can try to increase the size of the
==21667==  main thread stack using the --main-stacksize= flag.
==21667==  The main thread stack size used in this run was 8388608.
==21667== 
==21667== HEAP SUMMARY:
==21667==     in use at exit: 0 bytes in 0 blocks
==21667==   total heap usage: 0 allocs, 0 frees, 0 bytes allocated
==21667== 
==21667== All heap blocks were freed -- no leaks are possible
==21667== 
==21667== For counts of detected and suppressed errors, rerun with: -v
==21667== Use --track-origins=yes to see where uninitialised values come from
==21667== ERROR SUMMARY: 4041 errors from 373 contexts (suppressed: 0 from 0)
Segmentation fault

Original issue reported on code.google.com by mark.him...@gmail.com on 3 May 2011 at 3:51

[GoogleCodeExporter]

I should point out that it is the addition of one or more -vf scale filters 
that causes the seg-fault, adding other filters (pad and yadif have been 
tested) don't appear to cause the seg-fault.

Original comment by mark.him...@gmail.com on 3 May 2011 at 4:15

[GoogleCodeExporter]

Fixed, I updated the tarball

Original comment by baptiste...@gmail.com on 4 May 2011 at 10:32

• Changed state: Fixed

[GoogleCodeExporter]

Confirmed fixed in current 0.6-rc4

Original comment by mark.him...@gmail.com on 6 May 2011 at 12:21