Projet

Général

Profil

Install YoGA » Historique » Version 3

Damien Gratadour, 28/10/2013 08:03

1 1 Damien Gratadour
h1. Install YoGA
2
3
This page will guide you through the process of getting YoGA up and running on your machine
4
5 3 Damien Gratadour
h2. System requirements
6 1 Damien Gratadour
7 3 Damien Gratadour
Your system must be running Linux or Mac OS and should host a NVIDIA video card.
8 1 Damien Gratadour
9
h2. Install NVIDIA's devdriver, toolkit and SDK
10
11 2 Damien Gratadour
Go to the "nvidia cuda website":http://developer.nvidia.com/cuda-toolkit and get the CUDA framework and driver installed, following the installation instruction for your operating system.
12 1 Damien Gratadour
13
h2. Install Yorick
14
15
You get can Yorick from here : http://dhmunro.github.com/yorick-doc/ . This webpage will help you as well : http://www.maumae.net/yorick/doc/index.php. The associated forum provides a lot of help.
16
17
18
h2. Download & Install YoGA
19
20 2 Damien Gratadour
Get the latest version on the SVN repository or if you do not have access, contact us. 
21 1 Damien Gratadour
22 2 Damien Gratadour
The YoGA distribution is splitted into two parts corresponding to 2 levels of use the high-level and the low-level. The latter is the YoGA library and corresponding C++ API. The corresponding files are located in the trunk/libyoga directory. Once compiled, a libyoga.so file will be created that can be dynamically linked to your software. The high-level corresponds to the YoGA plugin for Yorick. The corresponding files are located in the trunk/yorick directory. Once compiled, a file yoga.so will be created and various files will be installed in your Yorick distribution so as to be able to load the yoga plugin in any of your Yorick session. 
23
24
then go to the newly created Yoga directory and update the Makefile with a:
25 1 Damien Gratadour
26
<pre>
27
yorick -batch make.i
28
</pre>
29
30
check the path to access the CUDA libs (by default : /usr/local/cuda).
31
32
To get the full flavor of Yoga, you need an additional CUDA library : "cudpp":http://code.google.com/p/cudpp/. The following command will do everything from downloading the package to moving the proper files to the proper place for you : 
33
34
<pre>
35
make cudpp
36
</pre>
37
38
then a standard : 
39
40
<pre>
41
make && make install
42
</pre>
43
44
will get you up and running !
45
46
There are some check scripts available in check_yoga.i that allow you to check your installation.