libflame is a portable library for dense matrix computations, providing complete functionality present in LAPACK, developed by current and former members of the Science of High-Performance Computing (SHPC) group in the Institute for Computational Engineering and Sciences at The University of Texas at Austin. In addtion, AMD is actively optimizing some of the key routines in libFLAME for AMD Zen core based architectures in the "amd" fork of libFLAME hosted on github: https://github.com/amd/libflame
For detailed instructions on how to configure, build, install, and link against libflame on AMD CPUs, please refer to the AOCL User Guide on https://developer.amd.com/amd-aocl
Upstream repository contains libflame reference manual that has all the information you need to get started with libflame, including installation directions, usage examples, and a complete API reference. If you have LaTeX installed on your system, you may simply change into the 'docs/libflame' subdirectory of the top-level directory of the libflame source tree and build the document from its source. You may also find a copy of the document here on github.
You can keep in touch with developers and other users of the project by joining the following mailing list:
- For any issues/suggestion in the "amd" fork of libFLAME, please email to toolchainsupport@amd.com
- libflame-discuss Please join and post to this mailing list if you have general questions or feedback regarding libflame.
Also, please read the LICENSE file for information on copying and distributing this software.