/usr/share/psychtoolbox-3/PsychGPGPU/Contents.m is in psychtoolbox-3-common 3.0.14.20170103+git6-g605ff5c.dfsg1-1build1.
This file is owned by root:root, with mode 0o644.
The actual contents of the file can be viewed below.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 | % Psychtoolbox:PsychGPGPU
%
% General Purpose Graphics Processing Unit (GPGPU) computing suppport.
%
% This folder contains functions which employ massively parallel computing
% hardware, typically modern GPU's, to accelerate computing tasks in the
% context of stimulus presentation. The compute hardware is controlled via
% computing api's like NVidia CUDA or the cross-platform, cross-vendor api
% OpenCL.
%
% The current initial implementation as of April 2013 only supports CUDA
% compute capable NVidia GPUs in combination with the free and open-source
% GPUmat toolkit for 64-Bit Matlab. Support for GNU/Octave and other GPU
% and compute toolkits, e.g., OpenCL based systems, will follow in later
% releases. Currently only 64-Bit Matlab is supported. While it would be
% possible to also provide 32-Bit support, we don't expect much demand for
% it, so will not deliver 32-Bit support for the time being.
%
%
% Requirements:
%
% 1. A NVidia GPU with CUDA compute capabilities. This is any OpenGL3/4
% capable GPU, starting with the GeForce-8000 series or later.
%
% 2. The freely downloadable CUDA SDK Version 5.0 or later from NVidia:
% https://developer.nvidia.com/cuda-downloads
%
% 3. The free and open-source GPUmat toolbox for Matlab from SourceForge:
% http://sourceforge.net/projects/gpumat
%
% 4. A 64-Bit version of Matlab for Linux, Windows or MacOSX.
%
% Demos can be found in the Psychtoolbox/PsychDemos/GPGPUDemos folder.
%
% This folder contains support functions for GPU computing:
%
% GPUTypeFromToGL - Helper function to transfer data efficiently
% between CUDA/GPUmat data types and
% OpenGL/Psychtoolbox.
%
% memcpyCudaOpenGL.cpp - Source code for the memcpyCudaOpenGL mex files.
% makememcpyCudaOpenGL.m - Build script to build mex files against Matlab
% and CUDA 5.0 SDK.
% memcpyCudaOpenGL.mex* - Mex files for 64-Bit Matlab.
%
|