[clblas] 15/67: Merge pull request #133 from TimmyLiu/develop

Ghislain Vaillant ghisvail-guest at moszumanska.debian.org
Tue Oct 27 08:02:10 UTC 2015


This is an automated email from the git hooks/post-receive script.

ghisvail-guest pushed a commit to branch master
in repository clblas.

commit ecd89f95d43eb522cce2a7f1a2efa3b638a1f607
Merge: ce984ae 5a74faf
Author: Timmy <timmy.liu at amd.com>
Date:   Mon Aug 17 13:24:48 2015 -0500

    Merge pull request #133 from TimmyLiu/develop
    
    fix the performance drop of SGEMM column major NT or row major TN when lda and ldb are big multiples of 1024 such as 4096, 5120, 6144, 7168, 8192

 src/library/CMakeLists.txt                         |   6 +
 src/library/bingen.cmake                           |   1 +
 src/library/blas/functor/hawaii.cc                 |  19 +
 .../blas/functor/hawaii_sgemmBig1024Kernel.cc      | 506 +++++++++++++++++++++
 .../blas/functor/hawaii_sgemmSplitKernel.cc        | 147 ++++++
 .../functor/include/hawaii_sgemmBig1024Kernel.h    |  48 ++
 .../blas/gens/clTemplates/sgemm_gcn_bigMatrices.cl | 264 +++++++++++
 src/library/blas/xgemm.cc                          |   1 +
 8 files changed, 992 insertions(+)

-- 
Alioth's /usr/local/bin/git-commit-notice on /srv/git.debian.org/git/debian-science/packages/clblas.git



More information about the debian-science-commits mailing list