diff options
Diffstat (limited to 'doc')
| -rw-r--r-- | doc/PreprocessorDirectives.dox | 4 | ||||
| -rw-r--r-- | doc/SparseLinearSystems.dox | 3 | ||||
| -rw-r--r-- | doc/UsingIntelMKL.dox | 6 |
3 files changed, 12 insertions, 1 deletions
diff --git a/doc/PreprocessorDirectives.dox b/doc/PreprocessorDirectives.dox index 0919d4190..b6d08c700 100644 --- a/doc/PreprocessorDirectives.dox +++ b/doc/PreprocessorDirectives.dox @@ -122,6 +122,10 @@ run time. However, these assertions do cost time and can thus be turned off. this threshold raises a compile time assertion. Use 0 to set no limit. Default is 128 KB. - \b \c EIGEN_NO_CUDA - disables CUDA support when defined. Might be useful in .cu files for which Eigen is used on the host only, and never called from device code. + - \b \c EIGEN_STRONG_INLINE - This macro is used to qualify critical functions and methods that we expect the compiler to inline. + By default it is defined to \c __forceinline for MSVC and ICC, and to \c inline for other compilers. A tipical usage is to + define it to \c inline for MSVC users wanting faster compilation times, at the risk of performance degradations in some rare + cases for which MSVC inliner fails to do a good job. - \c EIGEN_DONT_ALIGN - Deprecated, it is a synonym for \c EIGEN_MAX_ALIGN_BYTES=0. It disables alignment completely. %Eigen will not try to align its objects and does not expect that any objects passed to it are aligned. This will turn off vectorization if \b EIGEN_UNALIGNED_VECTORIZE=1. Not defined by default. diff --git a/doc/SparseLinearSystems.dox b/doc/SparseLinearSystems.dox index fc33b93e7..38754e4af 100644 --- a/doc/SparseLinearSystems.dox +++ b/doc/SparseLinearSystems.dox @@ -70,6 +70,9 @@ They are summarized in the following tables: <tr><td>UmfPackLU</td><td>\link UmfPackSupport_Module UmfPackSupport \endlink</td><td>Direct LU factorization</td><td>Square</td><td>Fill-in reducing, Leverage fast dense algebra</td> <td>Requires the <a href="http://www.suitesparse.com">SuiteSparse</a> package, \b GPL </td> <td></td></tr> +<tr><td>KLU</td><td>\link KLUSupport_Module KLUSupport \endlink</td><td>Direct LU factorization</td><td>Square</td><td>Fill-in reducing, suitted for circuit simulation</td> + <td>Requires the <a href="http://www.suitesparse.com">SuiteSparse</a> package, \b GPL </td> + <td></td></tr> <tr><td>SuperLU</td><td>\link SuperLUSupport_Module SuperLUSupport \endlink</td><td>Direct LU factorization</td><td>Square</td><td>Fill-in reducing, Leverage fast dense algebra</td> <td>Requires the <a href="http://crd-legacy.lbl.gov/~xiaoye/SuperLU/">SuperLU</a> library, (BSD-like)</td> <td></td></tr> diff --git a/doc/UsingIntelMKL.dox b/doc/UsingIntelMKL.dox index 6de14afad..fc35c3cf0 100644 --- a/doc/UsingIntelMKL.dox +++ b/doc/UsingIntelMKL.dox @@ -63,7 +63,11 @@ 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> -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. +The \c EIGEN_USE_BLAS and \c EIGEN_USE_LAPACKE* macros can be combined with \c EIGEN_USE_MKL to explicitly tell Eigen that the underlying BLAS/Lapack implementation is Intel MKL. +The main effect is to enable MKL direct call feature (\c MKL_DIRECT_CALL). +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. +MKL direct call can be disabled by defining \c EIGEN_MKL_NO_DIRECT_CALL. + 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. |