Chapter 17

CUDA FORTRAN

With special contributions fromGreg Ruetsch and Massimiliano Fatica

Chapter Outline

17.1 CUDA FORTRAN and CUDA C Differences

17.2 A First CUDA FORTRAN Program

17.3 Multidimensional Array in CUDA FORTRAN

17.4 Overloading Host/Device Routines With Generic Interfaces

17.5 Calling CUDA C Via Iso_C_Binding

17.6 Kernel Loop Directives and Reduction Operations

17.7 Dynamic Shared Memory

17.8 Asynchronous Data Transfers

17.9 Compilation and Profiling

17.10 Calling Thrust from CUDA FORTRAN

17.11 Exercises

This chapter gives an introduction to CUDA FORTRAN, the FORTRAN interface to the CUDA architecture. CUDA FORTRAN was developed in 2009 as a joint effort between the Portland Group (PGI) and NVIDIA. CUDA FORTRAN shares much in common ...

Get Programming Massively Parallel Processors, 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.