Recherche avancée

Médias (0)

Mot : - Tags -/presse-papier

Aucun média correspondant à vos critères n’est disponible sur le site.

Autres articles (43)

Sur d’autres sites (8557)

  • online free media hosting for live streaming

    30 novembre 2013, par Abdul Ali

    wanted to ask two things :

    1- How can we put the output of FFMPEG to a stream (online address to put the stream).

    2- Is there any free service (for testing purpose) to use FFMPEG and pass the output to it for live streaming .

    apologies if am unable to explain properly what is intended.

    to summarize, wish to convert images to video using FFMPEG (have tried the image to video conversion locally and seems to be working) and put the output to an online resource for live streaming and possibly also have VOD (so users who later logon can view at least from some point behind which they have missed).

    regards,

  • What could be causing processes to be left behind ?

    11 novembre 2015, par Alex

    I’m using the Process class to spawn a process (in particular, I’m using ffmpeg.exe to convert some video files). This ffmpeg process spawns more processes (on my computer, the total is 4, I’m guessing one per CPU core ?). If it matters, I’m doing this in a Windows service, although the problem also occurs when just debugging in Visual Studio, so I don’t think it does matter.

    The main ffmpeg process runs as expected and exits, causing the WaitForExit() call to return. Except, unlike when run normally (i.e. in a command prompt), the three processes that were spawned hang around. And they keep hanging around until the process that spawned the original ffmpeg.exe process (i.e. my service) is ended.

    Any ideas what this could be about ?

    My `ProcessStartInfo looks like this :

    _processStartInfo = new ProcessStartInfo(process, string.Join(" ", parameters))
    {
       UseShellExecute = false,
       CreateNoWindow = true,
       RedirectStandardOutput = true,
       RedirectStandardError = true
    };

    I launch the Process like this :

    _process = Process.Start(_processStartInfo);

    _outputReceivedHandler = (sender, e) =>
    {
       if (OutputData != null) OutputData(e.Data);
    };
    _process.OutputDataReceived += _outputReceivedHandler;

    _errorReceivedHandler = (sender, e) =>
    {
       if (ErrorData != null) ErrorData(e.Data);
    };
    _process.ErrorDataReceived += _errorReceivedHandler;

    _exitHandler = (sender, e) =>
    {
       if (Exited != null) Exited();
    };
    _process.EnableRaisingEvents = true;
    _process.Exited += _exitHandler;

    _process.Start();

    _process.BeginOutputReadLine();
    _process.BeginErrorReadLine();

    If it’s relevant, OutputData, ErrorData and Exited are Action<string></string>, Action<string></string> and Action, respectively.

    The reason I’m keeping the various handlers around is so that I can do this :

    private bool _disposed;
    public void Dispose()
    {
       Dispose(true);
       GC.SuppressFinalize(this);
    }

    protected virtual void Dispose(bool disposing)
    {
       if (_disposed || !disposing) return;

       if (_outputReceivedHandler != null) _process.OutputDataReceived -= _outputReceivedHandler;
       if (_errorReceivedHandler != null) _process.ErrorDataReceived -= _errorReceivedHandler;
       if (_exitHandler != null) _process.Exited -= _exitHandler;
       if (_process != null) _process.Dispose();

       _disposed = true;
    }

    Though it hasn’t made a difference whether or not I use Dispose(), the problem still occurs.

  • swscale : arm : fix NEON hscale init

    7 mai 2020, par Josh de Kock
    swscale : arm : fix NEON hscale init
    

    The NEON hscale function only supports X8 filter sizes and should only
    be selected when these are being used. At the moment filterAlign is
    set to 8 but in the future when extra NEON assembly for specific sizes is
    added they will need to have checks here too.

    The immediate usecase for this change is making the hscale checkasm
    test easier and without NEON specific edge-cases (x86 already has these
    guards).

    This applies the same fix from 718c8f9aa59751bb490e2688acf2b5cb68fd5ad1
    on the 32 bit arm version of the function, fixing fate-checkasm-sw_scale
    there.

    Signed-off-by : Martin Storsjö <martin@martin.st>

    • [DH] libswscale/arm/swscale.c