BLAS Library for OpenCL

I use the conjugate gradient solver without preconditioners to solve a linear system Ax=b, where A is a sparse matrix. This method is iterative and uses some BLAS functions like Dot Product, Scalar Product, xAXPY and xGEMV (SpMV for sparse matrix).I've started to develop these functions for the OpenCL language and I've decided to share them.

Right now, the following BLAS level 1 functions are available:
sDOT :: single precision dot product or scalar product (dot<-xy)
sNRM2 :: single precision vector 2-norm
sSCAL :: single precision product of vector by scalar (x<-ax)
sAXPY :: single precision AXPY (y<-ax + y) You can download the OpenCL code which was tested on NVIDIA Tesla C870 and GPU Computing SDK 2.3

SourceForge Project

Please join up with your contribution!

Update: OpenCL BLAS now is a discontinued project.

ATI Stream Software Development Kit (SDK) v2.0 Beta Program With OpenCL™ 1.0 Support

With ATI Stream SDK, AMD/ATI provides a way to program OpenCL on its cards. I didn't download it yet, but you can get more information on:

http://developer.amd.com/GPU/ATISTREAMSDKBETAPROGRAM/Pages/default.aspx

and on the OpenCL/ATI forum:
http://forums.amd.com/devforum/categories.cfm?catid=390&entercat=y

I'm going to test it and I will post here an overview.

GPU and Matlab

If you like programming Matlab-like environment, I suggest the freeware GPUMat from GP-you Group. You can explore the power of GPUs, BLAS and FFT libraries on NVIDIA cards. You can get more information on:
GP-you Group