Edward Smyth 43d36b9f66 AOCL_ENABLE_INSTRUCTIONS improvements 2
Use of AOCL_ENABLE_INSTRUCTIONS in dgemm tiny code path is
unnecessary and incorrectly caused AVX512 code to be run
on zen4 and later processors when AOCL_ENABLE_INSTRUCTIONS=avx2
or equivalent options was selected.

Replace with code to select kernel in a similar way to other
dgemm code paths and other APIs. Note that at present AVX2 code
is used the smallest matrix sizes on all zen platforms.

AMD-Internal: [CPUPL-5078]
Change-Id: Ie6b4895461cbbb915d2b48b92fc063f5cd6adb85
2024-06-25 04:57:38 -04:00
2024-06-24 08:50:36 -04:00
2024-06-24 08:50:36 -04:00
2019-05-23 12:51:17 -05:00
2023-11-23 08:54:31 -05:00
2019-10-02 10:16:22 +01:00
2021-03-22 17:42:33 -05:00
2023-11-10 13:05:12 -05:00
2018-08-07 14:21:07 -05:00
2023-11-23 08:54:31 -05:00
2023-05-25 14:46:33 +00:00

AOCL-BLAS library

AOCL-BLAS is AMD's optimized version of BLAS targeted for AMD EPYC and Ryzen CPUs. It is developed as a forked version of BLIS (https://github.com/flame/blis), which is developed by 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 and other collaborators (including AMD). All known features and functionalities of BLIS are retained and supported in AOCL-BLAS library. AOCL-BLAS is regularly updated with the improvements from the upstream repository.

AOCL BLAS is optimized with SSE2, AVX2, AVX512 instruction sets which would be enabled based on the target Zen architecture using the dynamic dispatch feature. All prominent Level 3, Level 2 and Level 1 APIs are designed and optimized for specific paths targeting different size spectrums e.g., Small, Medium and Large sizes. These algorithms are designed and customized to exploit the architectural improvements of the target platform.

For detailed instructions on how to configure, build, install, and link against AOCL-BLAS on AMD CPUs, please refer to the AOCL User Guide located on AMD developer portal.

The upstream repository (https://github.com/flame/blis) contains further information on BLIS, including background information on BLIS design, usage examples, and a complete BLIS API reference.

AOCL-BLAS is developed and maintained by AMD. You can contact us on the email-id toolchainsupport@amd.com. You can also raise any issue/suggestion on the git-hub repository at https://github.com/amd/blis/issues.

Description
BLAS-like Library Instantiation Software Framework
Readme BSD-3-Clause 72 MiB
Languages
C 86.3%
C++ 9.5%
Fortran 1.9%
Makefile 0.8%
MATLAB 0.5%
Other 0.9%