NVRTC Example#
08_NVRTC/nvrtc_potrs.cpp
This example shows how to use cuSolverDx with NVRTC runtime compilation and nvJitLink runtime linking.
The problem description built from cuSolverDx operators is defined only in device code. The header cusolverdx.hpp is included only in the device code that is passed to NVRTC to produce LTO IR. nvJitLink is then used to link that IR with cuSolverDx’s LTO library, optimize the linked modules, and generate the cubin for the target GPU architecture.