[petsc-dev] MKL discussion

[Satish Balay]

Well it might stop mails from future changes to MKL that might break
current configure code.

But most current e-mail is regarding missing/incorrect
--with-blas-lapack-dir option. Those issues will exist with configure
finding 'mkl_link_tool' anyway. Or perhaps the current code that
detects MKL can be simplified with mkl_link_tool.

Satish

On Fri, 1 Feb 2013, Jed Brown wrote:

> I've been delinquent here. If I had added support for mkl_link_tool when I
> first said I would, it would just work and users would stop mailing us
> about it.
> 
> http://software.intel.com/en-us/articles/intel-mkl-command-line-link-tool
> 
> 
> On Fri, Feb 1, 2013 at 4:14 PM, Barry Smith <bsmith at mcs.anl.gov> wrote:
> 
> >
> >   Satish,
> >
> >     The "BLAS/LAPACK" part of installation.html should discuss using the
> > MKL libraries and how to use Intel's tool for determining the libraries to
> > use. Currently this is not discussed there at all so users won't have a
> > clue.
> >
> >    Thanks
> >
> >      Barry
> >
> > From installation.html
> >
> > BLAS/LAPACK
> >
> > These packages provide some basic numeric kernels used by PETSc.
> >
> >         • Configure will automatically look for blas/lapack in certain
> > standard locations, on most systems you should not need to provide any
> > information about BLAS/LAPACK in the ./configure command.
> >         • One can use the following options to let configure
> > download/install blas/lapack automatically.
> >                 • --download-f-blas-lapack [when fortran compiler is
> > present]
> >                 • --download-f2cblaslapack [when configuring without a
> > fortran compiler - i.e --with-fc=0]
> >         • Alternatively one can use other options like one of the
> > following.
> >                 • --with-blas-lapack-lib=libsunperf.a
> >                 • --with-blas-lib=libblas.a --with-lapack-lib=liblapack.a
> >                 • --with-blas-lapack-dir=opt/intel/mkl72
> > Notes:
>