torchvision: Does not use the same compiler as torch when building with CUDA

[ConnorBaker]

Describe the bug

Unlike torch, torchvision does not explicitly set the C/C++ compiler when building with CUDA. This is problematic because we can end up linking libraries built by different versions of the compiler built with different language standards. Additionally, we can potentially end up using a compiler which isn't supported by the version of NVCC being used.

torch does this in the preConfigure phase:

nixpkgs/pkgs/development/python-modules/torch/default.nix

Lines 148 to 151 in e5bde65

	preConfigure = lib.optionalString cudaSupport ''
	export TORCH_CUDA_ARCH_LIST="${cudaFlags.cudaCapabilitiesSemiColonString}"
	export CC=${cudatoolkit.cc}/bin/gcc CXX=${cudatoolkit.cc}/bin/g++
	'' + lib.optionalString (cudaSupport && cudnn != null) ''

torchvision can do it in the same place or in preBuild:

nixpkgs/pkgs/development/python-modules/torchvision/default.nix

Lines 47 to 50 in e5bde65

	preBuild = lib.optionalString cudaSupport ''
	export TORCH_CUDA_ARCH_LIST="${cudaFlags.cudaCapabilitiesSemiColonString}"
	export FORCE_CUDA=1
	'';

Notify maintainers

@ericsagnes

cc @NixOS/cuda-maintainers

Metadata

Please run nix-shell -p nix-info --run "nix-info -m" and paste the result.

[user@system:~]$ nix-shell -p nix-info --run "nix-info -m"
 - system: `"x86_64-linux"`
 - host os: `Linux 6.1.12-200.fc37.x86_64, Fedora Linux, 37 (Workstation Edition), nobuild`
 - multi-user?: `no`
 - sandbox: `no`
 - version: `nix-env (Nix) 2.14.0pre20230208_ec78896`
 - channels(connorbaker): `"nixpkgs"`
 - nixpkgs: `/home/connorbaker/.nix-defexpr/channels/nixpkgs`

[ConnorBaker]

Would be closed by #218035

[SomeoneSerge]

Duplicate of: #217878 (unless I'm mistaken)

[nixos-discourse]

This issue has been mentioned on NixOS Discourse. There might be relevant details there:

https://discourse.nixos.org/t/tweag-nix-dev-update-45/26397/1