You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@mxnet.apache.org by GitBox <gi...@apache.org> on 2020/01/16 20:20:10 UTC

[GitHub] [incubator-mxnet] larroy edited a comment on issue #17293: [Build] Add a reasonable default for CMAKE_CUDA_COMPILER in *nix

larroy edited a comment on issue #17293: [Build] Add a reasonable default for CMAKE_CUDA_COMPILER in *nix
URL: https://github.com/apache/incubator-mxnet/pull/17293#issuecomment-575328978
 
 
   What do you suggest wrt https://cmake.org/cmake/help/v3.13/envvar/CUDACXX.html should we check if this is set? I understand your concern.  Please suggest a better approach, I'm not a CMake expert, but this was working before without needing to set any paths. Even though you are right about the documentation from Nvidia.
   
   I can compile pytorch just fine without needing to do any additional changes to PATHs, or environments. When there's a single cuda version installed or symlinked to /usr/local/cuda we should just pick up that one unless specified otherwise.
   
   Please propose changes or alternatives.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
users@infra.apache.org


With regards,
Apache Git Services