Accelerated Parallel Processing Math Libraries (APPML) is now open-source!
clMath is the new name of this open-source project. The source is available on GitHub at: https://github.com/clMathLibraries. Please read our blog to learn more about this exciting development. We will continue to support binary releases of clMath on developer.amd.com.
AMD Accelerated Parallel Processing Math Libraries are software libraries containing FFT and BLAS functions written in OpenCL and designed to run on AMD GPUs. The libraries support running on CPU devices to facilitate debugging and multicore programming. APPML 1.10 is the most current generally available version of the library. Example programs are included to illustrate usage of these libraries. Additional sample programs are available separately (see list of files for download below).
APPML v1.10 libraries are now available! This new version adds the complete set of L1 routines for BLAS. Please read the release notes in the library package for known issues. Documentation on new functions is included in the pdf manual. Users should download and install the latest AMD graphics drivers for use on AMD GPUs.
- Complex & real FFTs
- 1D, 2D, 3D
- Transform lengths – powers of 2, 3, 5 or any combination thereof
- The complete set of BLAS level 1, 2 & 3 routines is implemented. Please see Netlib BLAS for the list of routines.
For more information on supported graphics cards, see the AMD APP System Requirements.
OpenCL version of MAGMA is available
clMAGMA 1.0 is available from the Innovative Computing Lab at the University of Tennessee, Knoxville. This version supports the following features:
- LU, QR, and Cholesky factorizations in both real and complex arithmetic (single and double);
- Linear and least squares solvers based on correspondingly the LU/Cholesky and QR factorizations in both real and complex arithmetic (single and double);
- Reductions to Hessenberg, bidiagonal, and tridiagonal forms using orthgonal similarity transformations in both real and complex arithmetic (single and double);
- Eigen and singular value problem solvers in both real and complex arithmetic (single and double);
- Orthogonal transformation routines.
The following components are needed to build an application with OpenCL MAGMA:
- clMAGMA from MAGMA
- APPML BLAS: from the table below
- CPU math library (ACML recommended for AMD platforms)