* cudaPackages.cudnn: migrate to redist cudaPackages
* cudaPackages.cudnn: fix missing zlib in rpath
* cudaPackages.cudnn: remove the propagated cudatoolkit
* cudaPackages.cudnn: use autoPatchelfHook
instead of custom find ... -exec ldd | grep routine
mark libcudnn_cnn_infer.so as needed for libcudnn.so on cudnn>=8.0.5
- a hint for autoPatchelf, as an alternative to manually adding $ORIGIN
* cudaPackages.cudnn: use automatic hooks for rpath
as a more common way to use addOpenGLRunpath and autoPatchelf with cudaPackages
* cudaPackages.cudnn: consume individual cuda packages
...since cudnn is part of the cuda package set
- introduces the scary useCudatoolkitRunfile function argument
to discourage usage of the runfile-based cudatoolkit
- instead of the rather hidden useRedist term in let ... in
- repeats cudatoolkit_root pattern after cuda_joined in pytorch &c
(the "toolchain view")
- redist packages are marked optional to support cuda<11.4 where the
attributes for redist packages do not exist
* cudaPackages.cudnn: update to pname+version
Co-authored-by: Sandro <sandro.jaeckel@gmail.com>
Co-authored-by: Sandro <sandro.jaeckel@gmail.com>
* python3Packages.fenics: fix build, pin to older boost
Looking at upstream, there are various issues with newer boost.
(At least some of them have been since fixed)
For now, fix the build by using a version of boost that works
with the current version.
Error here was complaining about `std::min_element`,
which is no longer available, apparently, due to newer boost
no longer (transitively) including <algorithm>.
This was added in C++17, so I'm not sure the cmake flag
specifying dolfin built with C++11 makes sense or is used.
Leaving for now :).
* nixos/tests/fenics: fix name of machine/node in script
Still fails for now.
* python3Packages.fenics: fix accidentally changed strings in subst
Looks like in migration to pkg-config this was erroneously
changed from `pkgconfig` (python package, and source string)
to `pkg-config` (nix package name, tool name).
(see 9bb3fccb5b)
Fixes the NixOS test.
There are many different versions of the `cudatoolkit` and related
cuda packages, and it can be tricky to ensure they remain compatible.
- `cudaPackages` is now a package set with `cudatoolkit`, `cudnn`, `cutensor`, `nccl`, as well as `cudatoolkit` split into smaller packages ("redist");
- expressions should now use `cudaPackages` as parameter instead of the individual cuda packages;
- `makeScope` is now used, so it is possible to use `.overrideScope'` to set e.g. a different `cudnn` version;
- `release-cuda.nix` is introduced to easily evaluate cuda packages using hydra.
