Matrix-Matrix Product Experiments with BLAZE
The implementation of the GEMM algorithm is based on BLIS: A Framework for Rapidly Instantiating BLAS Functionality and adopted from ulmBLAS.
|
GEMM C++ Implementation |
The tar-ball test_blaze.tgz contains the files:
$shell> tar cfz test_blaze.tgz session*/*.h session*/*.cc $shell> tar tfvz test_blaze.tgz -rw-rw-r-- lehn/num 13573 2016-02-16 15:55 session1/avx.h -rw-rw-r-- lehn/num 42644 2016-02-16 16:04 session1/blisavx.h -rw-rw-r-- lehn/num 33580 2016-02-16 19:58 session1/fma.h -rw-rw-r-- lehn/num 1573 2016-02-16 15:55 session1/gccvec.h -rw-rw-r-- lehn/num 12122 2016-02-16 17:22 session1/gemm.h -rw-rw-r-- lehn/num 4703 2016-02-16 20:26 session1/bench_gemm.cc -rw-rw-r-- lehn/num 1531 2016-02-16 13:23 session1/test_gemm.cc $shell>