Frage

I am writing my first ffmpeg application using the tutorial linked to from the ffmpeg.org documentation page [tutorial] [sample]. Here is my initial code:

const char * fnmVideoIn = argv [1];
const char * fnmImageOut = argv [2];

av_register_all ();

// [] Open the file
AVFormatContext * pcxFormat;
if (avformat_open_input (&pcxFormat, fnmVideoIn, NULL, NULL)) {
    fprintf (stderr, "Could not open file %s for reading\n",
            fnmVideoIn);
    return -1;
}

// [] Get stream information
if (avformat_find_stream_info (pcxFormat, NULL) < 0) {
    fprintf (stderr, "Could not find stream info\n");
    return -1;
}

// [log] print stream info
av_dump_format (pcxFormat, 0, fnmVideoIn, 0);

and all is well. The program runs without errors and dumps the video info properly. but then I got to the next step...

...
// [log] print stream info
av_dump_format (pcxFormat, 0, fnmVideoIn, 0);

int ixVideoStream = -1, ixStrm;
for (ixStrm = 0; ixStrm < pcxFormat->nb_streams; ++ixStrm) {
    if (pcxFormat->streams [ixStrm]->codec->codec_type == AVMEDIA_TYPE_VIDEO) {
        ixVideoStream = ixStrm;
        break;
    }
}

and it exits with a segfault. gdb says:

(gdb) r test.mp4 out
...
Program received signal SIGSEGV, Segmentation fault.
0xb7f572c5 in avformat_open_input () from /usr/lib/libavformat.so.53

how is it possible that avformat_open_input segfaults because of code added after it is called? I even tested it with a regular i=0; i<100 for-loop and it still segfaults! is this a bug? For reference, my system is:

$ gcc --version
gcc (GCC) 4.7.0 20120414 (prerelease)

$ uname -srvmpio
Linux 3.3.4-1-ARCH #1 SMP PREEMPT Sat Apr 28 06:04:27 UTC 2012 i686 Intel(R) Core(TM)2 Duo CPU P7450 @ 2.13GHz GenuineIntel GNU/Linux

$ ffmpeg -version
ffmpeg version 0.10.2
built on Mar 17 2012 08:53:01 with gcc 4.6.3
configuration: --prefix=/usr --enable-libmp3lame --enable-libvorbis --enable-libxvid --enable-libx264 --enable-libvpx --enable-libtheora --enable-libgsm --enable-libspeex --enable-postproc --enable-shared --enable-x11grab --enable-libopencore_amrnb --enable-libopencore_amrwb --enable-libschroedinger --enable-libopenjpeg --enable-librtmp --enable-libpulse --enable-gpl --enable-version3 --enable-runtime-cpudetect --disable-debug --disable-static
libavutil      51. 35.100 / 51. 35.100
libavcodec     53. 61.100 / 53. 61.100
libavformat    53. 32.100 / 53. 32.100
libavdevice    53.  4.100 / 53.  4.100
libavfilter     2. 61.100 /  2. 61.100
libswscale      2.  1.100 /  2.  1.100
libswresample   0.  6.100 /  0.  6.100
libpostproc    52.  0.100 / 52.  0.100
War es hilfreich?

Lösung

I am not sure of the etiquette on questions whose solutions were found elsewhere, but I'll self-answer here anyway, for posterity.

Anyway, the first argument to avformat_open_input is a pointer to a pointer to an AVFormatContext structure allocated using avformat_alloc_context, or if you want the function to allocate it for you, a null pointer.

Here, I've made the mistake of providing the function with an uninitialized pointer, which triggers the occasional segmentation fault errors. The connection with the for-loop was only incidental, and probably has something to do with how the compiler structures the resulting machine code.

Lizenziert unter: CC-BY-SA mit Zuschreibung
Nicht verbunden mit StackOverflow
scroll top