diff options
| author |  Gael Guennebaud <g.gael@free.fr> | 2016-07-25 18:41:26 +0200 |
|---|---|---|
| committer | Gael Guennebaud <g.gael@free.fr> | 2016-07-25 18:41:26 +0200 |
| commit | c9425492c8e050cc19002faaa5d0017d19c2ac76 (patch) | |
| tree | 5d6b1ab5094be42e7408189c2b8ce33372f2e498 /doc/UsingIntelMKL.dox | |
| parent | 0592b4cfbf0995261cad06add94cf5a799747e77 (diff) | |
Update doc.
Diffstat (limited to 'doc/UsingIntelMKL.dox')
| -rw-r--r-- | doc/UsingIntelMKL.dox | 12 |
1 files changed, 7 insertions, 5 deletions
diff --git a/doc/UsingIntelMKL.dox b/doc/UsingIntelMKL.dox index dbe559e53..a5c1a20b9 100644 --- a/doc/UsingIntelMKL.dox +++ b/doc/UsingIntelMKL.dox @@ -32,11 +32,13 @@ namespace Eigen { -/** \page TopicUsingIntelMKL Using Intel® Math Kernel Library from Eigen +/** \page TopicUsingIntelMKL Using BLAS/LAPACK and Intel® Math Kernel Library from Eigen -\section TopicUsingIntelMKL_Intro Eigen and Intel® Math Kernel Library (Intel® MKL) +<!-- \section TopicUsingIntelMKL_Intro Eigen and Intel® Math Kernel Library (Intel® MKL) --> + +Since %Eigen version 3.1 and later, users can benefit from built-in Intel MKL optimizations with an installed copy of Intel MKL 10.3 (or later). +Since %Eigen version 3.3 and later, <strong>any BLAS or LAPACK libraries</strong> can be used as backends. -Since Eigen version 3.1 and later, users can benefit from built-in Intel MKL optimizations with an installed copy of Intel MKL 10.3 (or later). <a href="http://eigen.tuxfamily.org/Counter/redirect_to_mkl.php"> Intel MKL </a> provides highly optimized multi-threaded mathematical routines for x86-compatible architectures. Intel MKL is available on Linux, Mac and Windows for both Intel64 and IA32 architectures. @@ -56,8 +58,8 @@ In addition you can choose which parts will be substituted by defining one or mu <table class="manual"> <tr><td>\c EIGEN_USE_BLAS </td><td>Enables the use of external BLAS level 2 and 3 routines (compatible with any F77 BLAS interface, not only Intel MKL)</td></tr> -<tr class="alt"><td>\c EIGEN_USE_LAPACKE </td><td>Enables the use of external Lapack routines via the <a href="http://www.netlib.org/lapack/lapacke.html">Intel Lapacke</a> C interface to Lapack (currently works with Intel MKL only)</td></tr> -<tr><td>\c EIGEN_USE_LAPACKE_STRICT </td><td>Same as \c EIGEN_USE_LAPACKE but algorithm of lower robustness are disabled. This currently concerns only JacobiSVD which otherwise would be replaced by \c gesvd that is less robust than Jacobi rotations.</td></tr> +<tr class="alt"><td>\c EIGEN_USE_LAPACKE </td><td>Enables the use of external Lapack routines via the <a href="http://www.netlib.org/lapack/lapacke.html">Lapacke</a> C interface to Lapack (compatible with any F77 LAPACK interface)</td></tr> +<tr><td>\c EIGEN_USE_LAPACKE_STRICT </td><td>Same as \c EIGEN_USE_LAPACKE but algorithm of lower robustness are disabled. \n This currently concerns only JacobiSVD which otherwise would be replaced by \c gesvd that is less robust than Jacobi rotations.</td></tr> <tr class="alt"><td>\c EIGEN_USE_MKL_VML </td><td>Enables the use of Intel VML (vector operations)</td></tr> <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> |