Chapter 7: Interfacing with CUDA C code and CUDA libraries
Abstract
This chapter deals with interface with CUDA C source code and libraries distributed with the Nvidia HPC SDK from CUDA Fortran.
Keywords
iso_c_binding module; ignore_tkr; cuBLAS; GEMM; Tensor cores; cuSPARSE; Compressed sparse row (CSR); Cholesky; cuTENSOR; Thrust; Sort; Unlimited polymorphism
This chapter covers how to interface CUDA Fortran with CUDA C, including both CUDA C source code and many of the CUDA libraries distributed with the NVIDIA HPC SDK. As far as the CUDA libraries are concerned, in this chapter we cover: cuBLAS, a CUDA implementation of the Basic Linear Algebra Subprograms (BLAS) library; cuSPARSE, a library of linear algebra routines that accommodates sparse matrices; ...
Get CUDA Fortran for Scientists and Engineers, 2nd Edition now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
