CUDA Runtime API or CUDA Driver API

Dear all,

this is my first post in this forum and I would like to congratulate people who make it possible.

My question is simple… It is possible to indicate the pgcc compiler using compilation flags that only CUDA Runtime API functions have to be used?

I mean, when I write “pgcc -acc -ta=nvidia nbody_openacc nbodyacc.o check.o -L/usr/local/cuda-6.0/lib64 -lcudart” I want that only Runtime API functions are used.

Thanks in advance

Best regards

Hi Adriancsg,

Welcome to the PGI Forums!

My question is simple… It is possible to indicate the pgcc compiler using compilation flags that only CUDA Runtime API functions have to be used?

No sorry, there isn’t a way to switch to using the CUDA runtime API.

  • Mat

Dear mkcolg,

Thank you so much!

Adrian