Recherche avancée

Sur d’autres sites (10785)

• Memory leak when using av_frame_get_buffer()

  19 juillet 2018, par Alec Matthews

  I am making a simple video player with ffmpeg. I have noticed that there is a memory leak originating in libavutil. Because ffmpeg is a mature library I assume that I am allocating a new frame incorrectly. The documentation is also vague about freeing the buffer that is created when you call av_frame_get_buffer(). Below is the code I am using to decode the video and queue it up for display on the UI thread.

  DWORD WINAPI DecoderThread(LPVOID lpParam)
  
  {
  
      AVFrame *frame = NULL;
  
      AVPacket pkt;
  
  
  
      SwsContext *swsCtx = NULL;
  
      UINT8 *buffer = NULL;
  
      INT iNumBytes = 0;
  
  
  
      INT result = 0;
  
  
  
      frame = av_frame_alloc();
  
  
  
      av_init_packet(&pkt);
  
      pkt.data = NULL;
  
      pkt.size = 0;
  
  
  
      // Create scaling context
  
      swsCtx = sws_getContext(codecCtx->width, codecCtx->height, codecCtx->pix_fmt, codecCtx->width, codecCtx->height, AV_PIX_FMT_BGR24, SWS_BICUBIC, NULL, NULL, NULL);
  
  
  
     while (av_read_frame(fmtCtx, &pkt) >= 0) {
  
          if (pkt.stream_index == videoStream) {
  
              result = avcodec_send_packet(codecCtx, &pkt);
  
  
  
              while (result >= 0) {
  
                  result = avcodec_receive_frame(codecCtx, frame);
  
                  if (result == AVERROR(EAGAIN) || result == AVERROR_EOF) {
  
                      break;
  
                  } else if (result < 0) {
  
                      // another error.
  
                  }
  
                  // Create a new frame to store the RGB24 data.
  
                  AVFrame *pFrameRGB = av_frame_alloc();
  
                  // Allocate space for the new RGB image.
  
                  //av_image_alloc(pFrameRGB->data, pFrameRGB->linesize, codecCtx->width, codecCtx->height, AV_PIX_FMT_BGR24, 1);
  
                  // Copy all of the properties from the YUV420P frame.
  
                  av_frame_copy_props(pFrameRGB, frame);
  
                  pFrameRGB->width = frame->width;
  
                  pFrameRGB->height = frame->height;
  
                  pFrameRGB->format = AV_PIX_FMT_BGR24;
  
                  av_frame_get_buffer(pFrameRGB, 0);
  
  
  
                  // Convert fram from YUV420P to BGR24 for display.
  
                  sws_scale(swsCtx, (const UINT8* const *) frame->data, frame->linesize, 0, codecCtx->height, pFrameRGB->data, pFrameRGB->linesize);
  
  
  
                  // Queue thr BGR frame for drawing by the main thread.
  
                  AddItemToFrameQueue(pFrameRGB);
  
  
  
                  av_frame_unref(frame);
  
              }
  
          }
  
         while (GetQueueSize() > 100) {
  
              Sleep(10);
  
         }
  
     }
  
  
  
      CloseFrameQueue();
  
  
  
      av_frame_free(&frame);
  
     avcodec_close(codecCtx);
  
     avformat_close_input(&fmtCtx);
  
  
  
      return 0;
  
  }

  Is there a better way to allocate a new frame for holding the post sws_scale() transformation ?

  There is a similar stackoverflow question that uses mostly depreciated function calls. I can’t seem to find any answers that conform to the new version of ffmpeg in the documentation. Any help would be appreciated.

• Using gcovr with FFmpeg

  6 septembre 2010, par Multimedia Mike — FATE Server

  When I started investigating code coverage tools to analyze FFmpeg, I knew there had to be an easier way to do what I was trying to do (obtain code coverage statistics on a macro level for the entire project). I was hoping there was a way to ask the GNU gcov tool to do this directly. John K informed me in the comments of a tool called gcovr. Like my tool from the previous post, gcovr is a Python script that aggregates data collected by gcov. gcovr proves to be a little more competent than my tool.

  Results
  Here is the spreadsheet of results, reflecting FATE code coverage as of this writing. All FFmpeg source files are on the same sheet this time, including header files, sorted by percent covered (ascending), then total lines (descending).

  Methodology
  I wasn’t easily able to work with the default output from the gcovr tool. So I modified it into a tool called gcovr-csv which creates data that spreadsheets can digest more easily.

  • Build FFmpeg using the '-fprofile-arcs -ftest-coverage' in both the extra cflags and extra ldflags configuration options
  • 'make'
  • 'make fate'
  • From build directory : 'gcovr-csv > output.csv'
  • Massage the data a bit, deleting information about system header files (assuming you don’t care how much of /usr/include/stdlib.h is covered — 66%, BTW)

  Leftovers
  I became aware of some spreadsheet limitations thanks to this tool :

  1. OpenOffice can’t process percent values correctly– it imports the percent data from the CSV file but sorts it alphabetically rather than numerically.
  2. Google Spreadsheet expects CSV to really be comma-delimited– forget about any other delimiters. Also, line length is an issue which is why I needed my tool to omit the uncovered ine number ranges, which it does in its default state.

• Revision 30146 : ajouts d’options non encore fonctionnelles dans le CFG

  24 juillet 2009, par kent1@… — Log

  ajouts d’options non encore fonctionnelles dans le CFG