libnvrtc

NOTE:
This package is not open source and does not contain any source code. Instead, in order to "build" the exported target(s) it downloads (potentially large) pre-built binaries provided by NVIDIA for the target platform.

CI for this package is disabled due to the above.
Supported platforms/compilers are Windows/MSVC. Linux binaries exist but are currently not available through this package.

NVRTC is a runtime compilation library for CUDA C++. It accepts CUDA C++ source code in character string form and creates handles that can be used to obtain the PTX. The PTX string generated by NVRTC can be loaded by cuModuleLoadData and cuModuleLoadDataEx, and linked with other modules by using the nvJitLink library or using cuLinkAddData of the CUDA Driver API. This facility can often provide optimizations and performance not possible in a purely offline static compilation.

Usage

To start using libnvrtc in your project, add the following depends value to your manifest, adjusting the version constraint as appropriate:

depends: libnvrtc ^12.8.0

Then import the library in your buildfile:

import libs = libnvrtc%libs{nvrtc}

Importable targets

This package provides the following importable targets:

libs{nvrtc}