Skip to content

Latest commit

 

History

History
18 lines (12 loc) · 712 Bytes

README.md

File metadata and controls

18 lines (12 loc) · 712 Bytes

FDD3360

Course on Applied GPU programming at KTH

This repository contains the solution/codes for the assignments in the course.

The codes are written in Fortran. cudaFortran is supported by PGI. The codes are compiled as follows:

module load pgi cuda to load the PGI compiler.

To compile the cudaFortran code: pgfortran -Mcuda=cc3x cuda_file.cuf -o cuda_file.out

Update: Only Assignment 3 - exercise 3 and bonus exercise features cudaC code. This is because of the asynchronous copy of derived data type where, in cudaFortran the datatype has to be generic. As a temporary fix, switched to cudaC in only those exercises. A better solution should be available and shall be updated, when possible.