[MPlayer-users] lavcac3enc=1 Crashing mplayer

Thu Jan 10 19:10:06 CET 2008

2008/1/10, Matthew Nichols <mjnichol at uwaterloo.ca>:
> I'm using the latest from SVN (from about 8 hours ago). From my tests so
> far, it seems that lavcac3enc (or a closely related component) is
> causing mplayer to crash. Most often, the crash happens within 1 second,
> and is the result of a realloc call failing:
>
> ....
> Starting playback...
> [libaf] Reallocating memory in module lavcac3enc, old len = 0, new len =
> 21857
> Increasing filtered audio buffer size from 0 to 18432
> Increasing filtered audio buffer size from 18432 to 43008
> *** glibc detected *** /home/mjnichol/mplayer/mplayer: realloc():
> invalid pointer: 0x0000000000f4ac00 ***
> *** glibc detected *** ./mplayer: realloc(): invalid pointer:
> 0x0000000000f4ac10 ***
> ======= Backtrace: =========
> /lib64/libc.so.6(realloc+0x363)[0x37f82763e3]
> ./mplayer(decode_audio+0x200)[0x4a3e90]
> ./mplayer(main+0x163a)[0x43f34a]
> /lib64/libc.so.6(__libc_start_main+0xf4)[0x37f821e074]
> ./mplayer(cos+0x279)[0x43b3d9]
>
>
> The crash does NOT happen if:
>
> The input file is not more than 2 channels OR
> You pass 0 as the first parameter to lavcac3enc (telling it output a raw
> AC3 stream)
> (Also I have not had any crashes when I don't use lavcac3enc)
>
> So, my guess is that the code that adds the extra information to the
> stream to allow it to do SPDIF passthrough is at fault. But for some
> reason, is only affected if the input file is more than 2 channels.
>
> If you'd like to try this out, you can easily create a multi-channel wav
> file using:
>
> sox -M a.wav a.wav a.wav a.wav a.wav a.wav test.wav
>
> where a.wav is any (preferably short) wav file you can file on your hard
> drive.
>
> Then call
>
> mplayer -af lavcac3enc=1 test.wav
>
> (but make sure to turn down/off your speakers if your sound card doesn't
> support SPDIF passthrough properly)
>
> Sometimes mplayer doesn't crash until the file is finished playing, and
> in that case, a call to "free" causes the crash. If I specify -ao
> alsa:device=hw=1.1 (my SPDIF output device), it seems to crash on
> realloc, whereas if I don't, it seems to crash on "free"
>
> Backtrace when "realloc" causes crash:
>
> (gdb) bt
> #0  0x00000037f8230ec5 in raise () from /lib64/libc.so.6
> #1  0x00000037f8232970 in abort () from /lib64/libc.so.6
> #2  0x00000037f826b0db in __libc_message () from /lib64/libc.so.6
> #3  0x00000037f82763e3 in realloc () from /lib64/libc.so.6
> #4  0x00000000004a3e90 in decode_audio (sh_audio=0xe5aab0, minlen=65536)
>      at dec_audio.c:399
> #5  0x000000000043f34a in main (argc=<value optimized out>,
>      argv=<value optimized out>) at mplayer.c:1807
>
> Backtrace when "free" causes crash:
>
> (gdb) bt
> #0  0x00000037f8230ec5 in raise () from /lib64/libc.so.6
> #1  0x00000037f8232970 in abort () from /lib64/libc.so.6
> #2  0x00000037f826b0db in __libc_message () from /lib64/libc.so.6
> #3  0x00000037f8272832 in _int_free () from /lib64/libc.so.6
> #4  0x00000037f8275f2c in free () from /lib64/libc.so.6
> #5  0x00000000004fdc1a in uninit (af=0xe5ac70) at af_lavcac3enc.c:142
> #6  0x00000000004ed89f in af_remove (s=0xe5abe0, af=0xe5ac70) at af.c:230
> #7  0x00000000004ed8e8 in af_uninit (s=0x7dba) at af.c:341
> #8  0x00000000004a406a in uninit_audio (sh_audio=0xe5aab0) at
> dec_audio.c:290
> #9  0x000000000043cdfc in uninit_player (mask=<value optimized out>)
>      at mplayer.c:584
> #10 0x00000000004406ea in main (argc=<value optimized out>, argv=0x0)
>      at mplayer.c:3650
>
>
> System info:
>
> Distribution: Fedora 8
>
> [mjnichol at lakota mplayer]$ uname -a
> Linux lakota.----.---- 2.6.23.9-85.fc8 #1 SMP Fri Dec 7 15:49:36 EST
> 2007 x86_64 x86_64 x86_64 GNU/Linux
>
> [mjnichol at lakota mplayer]$ ls -l /lib/libc[.-]*
> -rwxr-xr-x 1 root root 1692524 2007-10-18 04:49 /lib/libc-2.7.so
> lrwxrwxrwx 1 root root      11 2007-11-07 18:15 /lib/libc.so.6 ->
> libc-2.7.so
>
> [mjnichol at lakota mplayer]$ gcc -v
> Using built-in specs.
> Target: x86_64-redhat-linux
> Configured with: ../configure --prefix=/usr --mandir=/usr/share/man
> --infodir=/usr/share/info --enable-shared --enable-threads=posix
> --enable-checking=release --with-system-zlib --enable-__cxa_atexit
> --disable-libunwind-exceptions
> --enable-languages=c,c++,objc,obj-c++,java,fortran,ada
> --enable-java-awt=gtk --disable-dssi --enable-plugin
> --with-java-home=/usr/lib/jvm/java-1.5.0-gcj-1.5.0.0/jre
> --enable-libgcj-multifile --enable-java-maintainer-mode
> --with-ecj-jar=/usr/share/java/eclipse-ecj.jar --with-cpu=generic
> --host=x86_64-redhat-linux
> Thread model: posix
> gcc version 4.1.2 20070925 (Red Hat 4.1.2-33)
>
> [mjnichol at lakota mplayer]$ ld -v
> GNU ld version 2.17.50.0.18-1 20070731
>
> [mjnichol at lakota mplayer]$ as -version
> GNU assembler version 2.17.50.0.18-1 20070731
>
> [mjnichol at lakota mplayer]$ cat /proc/cpuinfo
> processor       : 0
> vendor_id       : GenuineIntel
> cpu family      : 6
> model           : 15
> model name      : Intel(R) Core(TM)2 Quad CPU    Q6600  @ 2.40GHz
> stepping        : 11
> cpu MHz         : 2997.003
> cache size      : 4096 KB
> physical id     : 0
> siblings        : 4
> core id         : 0
> cpu cores       : 4
> fpu             : yes
> fpu_exception   : yes
> cpuid level     : 10
> wp              : yes
> flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
> mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe
> syscall nx lm constant_tsc arch_perfmon pebs bts rep_good pni monitor
> ds_cpl vmx est tm2 ssse3 cx16 xtpr lahf_lm
> bogomips        : 5997.07
> clflush size    : 64
> cache_alignment : 64
> address sizes   : 36 bits physical, 48 bits virtual
> power management:
>
> (repeated 3 more times)

I can not reproduct it, I do not have sox so I test with my 5.1 wav.
Need mplayer's full output messages, please add '-v' in the mplayer's
cmdline and run until crash and post the log or attach it here.
and also what's in your mplayer's config file, maybe related.

-- 
Ulion