Intel® oneAPI Math Kernel Library
Ask questions and share information with other developers who use Intel® Math Kernel Library.
Announcements
FPGA community forums and blogs on community.intel.com are migrating to the new Altera Community and are read-only. For urgent support needs during this transition, please visit the FPGA Design Resources page or contact an Altera Authorized Distributor.

how to use mkl in linux?

beblo33yahoo_com
Beginner
961 Views
I installed the mkl linux version package but now I want to know how can I use the libraries from the makefile in fortran?
0 Kudos
2 Replies
ArturGuzik
Valued Contributor I
961 Views
I installed the mkl linux version package but now I want to know how can I use the libraries from the makefile in fortran?
Hi,

what (exactly) do you mean? Linking your application with MKL using makefile on Linux? The following commands are needed to link and include (linker can find the libraries)

-L -I

MKL path usually something like /opt/intel/mkl/10.0.xxx/lib/

then you list required libraries, say,

-lmkl_solver_lp64, .....

I believe a clear explanation you'll find in user's guide (chapter 5 - Linking Your Application with Intel Math Kernel Library)

an example from there:

static linking of user code myprog.f, parallel version of sparse solver, and parallel Intel MKL supporting LP64 interface.

ifort myprog.f -L$MKLPATH -I$MKLINCLUDE -lmkl_solver_lp64_sequential.a
-Wl,--start-group $MKLPATH/libmkl_intel_lp64.a $MKLPATH/libmkl_sequential.a
$MKLPATH/libmkl_core.a -Wl,--end-group -lpthread

A.
0 Kudos
TimP
Honored Contributor III
961 Views
I installed the mkl linux version package but now I want to know how can I use the libraries from the makefile in fortran?
As there is a forum dedicated to MKL, and this forum is dedicated to Windows, this is not the place to get advice. You must first consult the MKL notes in the docs folder, particularly if you have no idea which details you must specify when asking a question. You do need to let us know whether you are making a 32- or 64-bit build, and your intentions about 32- or 64-bit integer and floating point data types.
0 Kudos
Reply