The LLVM libc++ as a device C++ run time library

libcu++ people have been working on upstreaming their implementations but do you C++ runtime community have a road map as to when it will be completed? I’m interested in offloading C++ code to the device currently NVIDIA’s. Lots of libraries need to be taken care by the experts.

I’m wondering if the libcu++ people really frequent this forum. Maybe it’d be easier to ask them directly at https://github.com/NVIDIA/libcudacxx/issues

1 Like

They’re saying they upstream their implementations, thus I thought I should ask the maintainers here first how they see the progress.

@rengolin but thanks, I’ll post a question there.

Where do they say that? I’d be curious about plans (and community interaction) as well.

Sorry, I thought it is claimed on their project README file on GitHub but not. Can Bryce jump in here and comment? (I don’t know his account cant mention him)

@jdoerfert as a lead of the OpenMP offloading project, to support C++, how would you like libc++ to evolve in the near future?