How to compile namd with Nvidia-hpc-sdk toolkit and nvcc compiler


We have one workstation with Quadro RTX 6000 graphic card and want to compile namd-2.15 version
(NAnoscale Molecular Dynamics) application with nvidia-hpc-sdk toolkit which supports nvcc compiler.
When we tried to compile with this option we saw that the error message like this.
root@test:/opt/NAMD_2.15alpha1_Source-AVX512/charm-6.10.2# ./build charm++ multicore-linux-x86_64 nvcc -j16 --with-production
Error> option: nvcc is not supported in this version!
Supported compilers: clang craycc gcc icc iccstatic msvc pgcc xlc xlc64
Supported options: bigemulator bigsim common cuda flang gfortran ifort nolb omp ooc papi pedantic perftools persistent pgf90 tsan

There is no suitable file in arch directory of the NAMD 's source code. So we see in this site High Performance Computing (HPC) SDK | NVIDIA andHPC-SDK was developed for namd and other application to increase the performance benchmark test.
How we can compile namd with nvcc compile or is there any way to do this ?
Processor: Intel® Xeon® Gold 5118 CPU @ 2.30GHz
GPU : Quadro RTX 6000
OS: Ubuntu 18.04 LTS

Thank you.

I’ve not built NAMD myself so would need to refer you to the NAMD developers for help on how to build their software.

Though in the information you posted, I see that they include “pgcc”, the PGI C compiler, as a supported compiler. PGI was rebranded to be the NVIDIA HPC Compiler (pgcc was renamed to nvc). Though we still ship pgcc as part of the SDK so you might try using it.

nvcc is the CUDA C compiler so most likely used for some files when the “cuda” option is selected.