4 CUDA/Ada is an Ada binding to NVIDIA's CUDA parallel computing platform and
5 programming model. This project was developed during the course of the master
6 seminar "Program Analysis and Transformation" at the University of Applied
12 --------------------------------------------------------------------------------
13 Copyright (C) 2011, 2012 Reto Buerki <reet@codelabs.ch>
14 Copyright (C) 2011, 2012 Adrian-Ken Rueegsegger <ken@codelabs.ch>
15 University of Applied Sciences Rapperswil
17 This program is free software: you can redistribute it and/or modify it under
18 the terms of the GNU General Public License as published by the Free Software
19 Foundation, either version 3 of the License, or (at your option) any later
21 --------------------------------------------------------------------------------
26 The paper about CUDA/Ada can be found here:
27 http://www.codelabs.ch/cuda-ada/cuda-ada-article.pdf.
35 The current release version of CUDA/Ada is available at
36 http://www.codelabs.ch/download.
40 To verify the integrity and authenticity of the distribution tarball, import
41 the key http://www.codelabs.ch/keys/0x3DC359DEpub.asc and type the following
44 $ gpg --verify libcudaada-{version}.tar.bz2.sig
46 The key fingerprint of the public key ('0x3DC359DE') is:
48 Key fingerprint = 752C 4EBC 115D 5EAD 75F7 0F34 A0AE 8AD7 3DC3 59DE
52 The current development version of CUDA/Ada is available through its git
55 $ git clone http://git.codelabs.ch/git/cuda-ada.git
57 A browsable version of the repository is also available here:
58 http://git.codelabs.ch/?p=cuda-ada.git.
63 To compile CUDA/Ada on your system, you need to have the following software
67 http://www.gnu.org/software/gnat/gnat.html
69 * Ahven (Test-Framework):
70 http://ahven.stronglytyped.org/
72 * CUDA Toolkit for Linux 10.10:
73 http://developer.nvidia.com/cuda-downloads
75 * NVIDIA Developer Drivers for Linux:
76 http://developer.nvidia.com/cuda-downloads
81 CUDA/Ada contains an unit test suite which can be run by entering the following
86 Of course, you need a CUDA aware GPU for this to work.
91 CUDA/Ada provides benchmarking code that measures matrix addition and
92 multiplication in Ada, CUDA/Ada and native CUDA C. The benchmarks can be run by
93 issuing the following command:
97 This will print the cumulated execution times of twenty successive matrix
98 operations for the different implementations.