Recherche avancée

Sur d’autres sites (7356)

• NVidia driver libraries in nvidia/cuda image

  16 février 2018, par Islam Sabyrgaliyev

  I want to run ffmpeg with cuvid hw-accelerated decoding in the container based on official nvidia/cuda image. Ffmpeg is not able to find libnvcuvid.so, although there are all required cuda libs.
  The output of ldconfig -p | grep libnv from the container :

  libnvrtc.so (libc6,x86-64) => /usr/local/cuda-9.0/targets/x86_64-linux/lib/libnvrtc.so
  
  libnvrtc-builtins.so (libc6,x86-64) => /usr/local/cuda-9.0/targets/x86_64-linux/lib/libnvrtc-builtins.so
  
  libnvidia-ptxjitcompiler.so.1 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libnvidia-ptxjitcompiler.so.1
  
  libnvidia-opencl.so.1 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libnvidia-opencl.so.1
  
  libnvidia-ml.so.1 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libnvidia-ml.so.1
  
  libnvidia-fatbinaryloader.so.390.12 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libnvidia-fatbinaryloader.so.390.12
  
  libnvidia-compiler.so.390.12 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libnvidia-compiler.so.390.12
  
  libnvidia-cfg.so.1 (libc6,x86-64) => /usr/lib/x86_64-linux-gnu/libnvidia-cfg.so.1
  
  libnvgraph.so (libc6,x86-64) => /usr/local/cuda-9.0/targets/x86_64-linux/lib/libnvgraph.so
  
  libnvblas.so (libc6,x86-64) => /usr/local/cuda-9.0/targets/x86_64-linux/lib/libnvblas.so
  
  libnvToolsExt.so (libc6,x86-64) => /usr/local/cuda-9.0/targets/x86_64-linux/lib/libnvToolsExt.so

  Should I just copy libnvcuvid.so from the host ? Wouldn’t it break if underlying driver version changes ?

• How to set a timeout for FFmpegFrameGrabber grab() method in JavaCV

  18 janvier 2019, par Null Pointer

  I’m trying to create a player in Java using JavaCV. But .grab() method takes different amounts of time for each grab. How can I discard a grab() operation when it takes too much time ?

  In detail : I’m working on a project to display rtsp streams. To do this, I use JavaCV and its FFmpegFrameGrabber class. Inside a loop I call its .grab() method to grab frames and then I print them on a panel.
  (Not knowing the reason) grab() method takes much more time in some calls. For example :

  1. grab : 15ms,
  2. grab : 13ms,
  3. grab : 14ms,
  4. grab : 95ms,
  5. grab : 12ms, ...

  I wish not to wait for a grab, in case it is taking too long like in this fourth call.

  (Sorry for not providing a working code, it’s all over different huge classes)

  I use the following grabber options (selected by some other colleague) :

  grabber.setImageHeight(480);
  
  grabber.setImageWidth(640);
  
  grabber.setOption("reconnect", "1");
  
  grabber.setOption("reconnect_at_eof", "1");
  
  grabber.setOption("reconnect_streamed", "1");
  
  grabber.setOption("reconnect_delay_max", "2");
  
  grabber.setOption("preset", "veryfast");
  
  grabber.setOption("probesize", "192");
  
  grabber.setOption("tune", "zerolatency");
  
  grabber.setFrameRate(30.0);
  
  grabber.setOption("buffer_size", "" + this.bufferSize);
  
  grabber.setOption("max_delay", "500000");
  
  grabber.setOption("stimeout", String.valueOf(6000000));
  
  grabber.setOption("loglevel", "quiet");
  
  grabber.start();

  Also I tried to do the grab() in another thread and not wait for that to end in case it takes too long. But this time I get an EXCEPTION_ACCESS_VIOLATION when I call .grab() method before the previous one did not end yet. "Calling .release() ; and .restart() ; methods before trying to grab again" also did not solve the access violation.

  Is there a setting option to cancel unordinarily long grabs ? (Or wouldn’t it work without these frames (because of a reason that I don’t know))

  Thanks

• Working directory in Dockerfile is not what is expected to be

  31 mai 2022, par user1765862

  I have following Dockerfile

  


  FROM public.ecr.aws/lambda/dotnet:6 AS base

FROM mcr.microsoft.com/dotnet/sdk:6.0-bullseye-slim as build
WORKDIR /src
COPY ["AWSServerless.csproj", "AWSServerless/"]
RUN dotnet restore "AWSServerless/AWSServerless.csproj"

WORKDIR "/src/AWSServerless"
COPY . .
RUN dotnet build "AWSServerless.csproj" --configuration Release --output /app/build

FROM build AS publish

RUN apt-get update && apt-get install -y apt-utils libgdiplus libc6-dev

RUN apt-get install -y ffmpeg

RUN dotnet publish "AWSServerless.csproj" \
            --configuration Release \ 
            --runtime linux-x64 \
            --self-contained false \ 
            --output /app/publish \
            -p:PublishReadyToRun=true  

FROM base AS final
WORKDIR /var/task

CMD ["AWSServerless::AWSServerless.LambdaEntryPoint::FunctionHandlerAsync"]
COPY --from=publish /app/publish .


  


  As per ffmpeg docs In order to use ffmpeg I need to set its binary folder
for example with

  


  {
  "BinaryFolder": "/var/task",
  "TemporaryFilesFolder": "/tmp"
}


  


  Error

  


  


  An error occurred trying to start process './ffmpeg' with working
directory '/var/task'. No such file or directory

  


  


  What is the working directory here ?

  


  Update :
Actual error is being thrown when I try to execute following command (FFMpegCore package) from my api controller

  


  public class MyController : ControllerBase&#xA;{&#xA;    public async Task<string> Get()&#xA;    {    &#xA;         await FFMpegArguments&#xA;               .FromPipeInput(new StreamPipeSource(myfile))&#xA;               .OutputToPipe(new StreamPipeSink(outputStream), options => options&#xA;                  .WithVideoCodec("vp9")&#xA;                  .ForceFormat("webm"))&#xA;                  .ProcessAsynchronously();&#xA;         ...&#xA;    }    &#xA;}&#xA;</string>