mirror/eigen
2
0
Fork 0
mirror of https://gitlab.com/libeigen/eigen.git synced 2024-12-15 07:10:37 +08:00
Code Issues Packages Projects Releases Wiki Activity

Clarify MKL_DIRECT_CALL doc.

Browse Source
This commit is contained in:
Gael Guennebaud 2017-08-17 22:12:26 +02:00
parent f727844658
commit 2810ba194b
1 changed files with 1 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
doc/UsingIntelMKL.dox
View File

@ -63,7 +63,7 @@ In addition you can choose which parts will be substituted by defining one or mu
<tr><td>\c EIGEN_USE_MKL_ALL </td><td>Defines \c EIGEN_USE_BLAS, \c EIGEN_USE_LAPACKE, and \c EIGEN_USE_MKL_VML </td></tr>
</table>
In order to be able to use \b MKL_DIRECT_CALL for BLAS level 2 and 3 routines, then the macro \c EIGEN_USE_MKL must also be defined in the case none of the other \c EIGEN_USE_MKL_* macros has been defined. This is needed to tell Eigen that the BLAS backend is the MKL and that the MKL interface must be used instead of the generic F77 one.
The options can be combined with \b MKL_DIRECT_CALL to enable MKL direct call feature. This may help to increase performance of some MKL BLAS (?GEMM, ?GEMV, ?TRSM, ?AXPY and ?DOT) and LAPACK (LU, Cholesky and QR) routines for very small matrices. To make it work properly, the macro \c EIGEN_USE_MKL must also be defined in the case none of the other \c EIGEN_USE_MKL_* macros has been defined.
Note that the BLAS and LAPACKE backends can be enabled for any F77 compatible BLAS and LAPACK libraries. See this \link TopicUsingBlasLapack page \endlink for the details.