[FFmpeg-devel] Unable to compile with cuda

[Dennis Mungai]

On Sat, 22 May 2021 at 11:55, Dylan Fernando <dylanf123 at gmail.com> wrote:

> I'm unable to compile on Arch Linux with cuda enabled. Command:
>
> ./configure --enable-opencl --enable-vulkan --enable-libglslang
> --disable-stripping --enable-nonfree --enable-cuda-nvcc --extra-c
> flags=-I/opt/local/cuda/include
>
> Error:
>
> ERROR: failed checking for nvcc.
>
>
>
> If you think configure made a mistake, make sure you are using the latest
> version from Git.  If the latest version fails, report the problem to the
> ffmpeg-user at ffmpeg.org mailing list or IRC #ffmpeg on irc.freenode.net.
> Include the log file "ffbuild/config.log" produced by configure as this
> will help
> solve the problem.
>
> Attached full log
>
> Thanks,
> Dylan
>
>
Hello,

Could you  try:

./configure --enable-opencl --enable-vulkan --enable-libglslang
--disable-stripping --enable-nonfree --enable-cuda-nvcc --extra-c
flags=-I/opt/local/cuda/include --nvccflags="-gencode
arch=compute_52,code=sm_52 -O2"

Or something closer to your GPU's supported compute level?
The default flags are --nvccflags="-gencode arch=compute_30,code=sm_30 -O2"
, which  are deprecated/dropped in newer  CUDA releases.
See the default compile targets for CUDA 11 here
https://docs.nvidia.com/cuda/archive/11.0/cuda-toolkit-release-notes/#cuda-compiler-new-features

Warm regards,
Dennis.