R interface to CUDA libraries
Turing C R TeX Cuda C++ Other
Latest commit 2479a3b Jul 27, 2016 @duncantl drive and run-time apis
Permalink
Failed to load latest commit information.
Paper drive and run-time apis Jul 27, 2016
R comment about a potential new feature Jul 26, 2016
TU don't exclude CUfunction Jul 26, 2016
Web updated the web page Jul 18, 2013
explorations updates May 18, 2014
inst ... Jul 26, 2016
man added nvccCmd as an alias Jul 18, 2016
src newly generated code Jul 17, 2016
tests got these to work. Jul 26, 2016
.Rbuildignore .. Jul 26, 2016
.gitignore ... Jul 16, 2016
Books updates for libNVVM and other matters Jun 30, 2013
Changes exploring overhead in .cuda for outputs and copyToDevice Apr 15, 2014
DESCRIPTION testing the package Jul 16, 2016
GenerationNotes generated the documentation May 24, 2013
LICENSE fixed URLs in documentation Jul 16, 2016
Links fixes for R CMD check Jul 16, 2016
Linux updates for libNVVM and other matters Jun 30, 2013
NAMESPACE fixes for R CMD check Jul 16, 2016
NEWS memory bindings generated May 18, 2013
Notes generated bindings for lots more routines. Still others we could do b… May 19, 2013
README simple start Apr 19, 2013
README.md update Jul 16, 2016
Todo.xml ... Jul 26, 2016
checkSDKVersion.c added test for SDK version May 14, 2013
cleanup cleanup Jul 16, 2016
configure version information from CUDA Jul 18, 2016
configure.in version information from CUDA Jul 18, 2016
cudaversion.c C code to create executable to compute version information at install… Jul 18, 2016
deprecated generated bindings for lots more routines. Still others we could do b… May 19, 2013
nvccOpts.R fixes for R CMD check Jul 16, 2016
showSizeofs.cu newly generated code Jul 17, 2016
sizeofs.cu computing the sizeof structs so we can allcate them on the device May 21, 2013

README.md

This RCUDA package provides a reasonably comprehensive interface to the CUDA SDK. This allows one to work with a GPU via R code. We can invoke CUDA code on the GPU, put data on the GPU and bring it back into R. We have access to devices, contexts, streams, events, and all of the CUDA concepts.

The package is now updated to work with CUDA SDK 7.5.

The package requires the RAutoGenRunTime package to be installed. This is available on github and omegahat.net and omegahat.ucdavis.edu.

You may need to set LD_LIBRARY_PATH - or DYLD_LIBRARY_PATH on OS X - to allow the system find the CUDA libraries. They are often installed in /usr/local/cuda/lib.

The code in the package is a mix of manually written and programmatically generated code. The R code in TU/clang.R is used to generate the code and can be re-run for different versions of CUDA. TU/clang.R