-
Notifications
You must be signed in to change notification settings - Fork 406
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
clang/8+cuda/10.0 build error with c++17 #2809
Comments
You didn't enable CUDA_LAMBDA something is broken with the logic it should enable it automatically |
I shouldn't have to enable lambdas to compile and run kokkos tests. This was with the test_all_sandia script, I have no issues with c++11 or c++14. Edit: We are testing this configuration nightly with c++11 and c++14 |
For additional reference, this is the resultant cmake line call invoked by the generate_makefile.bash call from the reproducer instructions:
|
adding -DKokkos_ENABLE_CUDA_LAMBDA=ON resolves the problem. The test is insufficiently guarded. It is "if defined(KOKKOS_CLASS_LAMBDA)" which doesn't in and of itself means that that includes host device ... |
that said: for Clang we should just enable that by default, and so we will probably for any NVCC which supports C++17 |
fix is coming including defaults change |
Confirmed this works on kokkos-dev-2, added nightly test https://jenkins-son.sandia.gov/view/Kokkos/job/Kokkos_develop_kokkosdev2_clang8_cuda_cpp17/ |
Error message:
Sample reproducer instructions from kokkos-dev-2 (Volta70 Cuda Arch):
The text was updated successfully, but these errors were encountered: