DPLASMA is a highly optimized, accelerator-aware, implementation of a dense linear algebra package for distributed heterogeneous systems. It is designed to deliver sustained performance for distributed systems where each node featuring multiple sockets of multicore processors, and if available, accelerators, using the PaRSEC runtime as a backend.
The choice made here is to allow the developer to specify if v2 is needed (by including it before the dplasmaaux.h), otherwise include by default cublas v1.
The choice made here is to allow the developer to specify if v2 is needed (by including it before the dplasmaaux.h), otherwise include by default cublas v1.