[Wien] Intel mkl dft+fftw3

[Laurence Marks]

No, I made the file. However, as I mentioned in the original email
there are some unresolved references

ipo: warning #11021: unresolved fftw_mpi_execute_dft
ipo: warning #11021: unresolved fftw_mpi_plan_dft_3d_f03
ipo: warning #11021: unresolved fftw_alloc_complex
ipo: warning #11021: unresolved fftw_mpi_local_size_3d_f03
ipo: warning #11021: unresolved fftw_mpi_init

At least via ar -t I cannot find them in any of the mkl libraries,
although I did not look that hard (ar -t is probably not the best
command).

On Tue, Feb 19, 2013 at 8:18 AM, Gavin Abo <gsabo at crimson.ua.edu> wrote:
> Are you saying that you cannot create libfftw3x_cdft.a on your system?
>
> The file is created on my Debian squeeze system, but I used:
>
> username at computername:/opt/intel/composer_xe_2013.1.117/mkl/interfaces/fftw3x_cdft$
> sudo make libintel64 mpi=mpich2 compiler=gnu
>
>
> On 2/19/2013 6:05 AM, Laurence Marks wrote:
>
> Dear Luis,
>
> Thanks for the options, but they do not quite do what I want - which is to
> replace -lfftw3_mpi by the mkl cdft eqivalents. The documentation claims
> that the mpi version is integrated. It might be better (or worse). I think I
> managed to do it once before with fftw2 then ran into bugs in cdft so gave
> up.
>
> ---------------------------
> Professor Laurence Marks
> Department of Materials Science and Engineering
> Northwestern University
> www.numis.northwestern.edu 1-847-491-3996
> "Research is to see what everybody else has seen, and to think what nobody
> else has thought"
> Albert Szent-Gyorgi
>
>



-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu 1-847-491-3996
"Research is to see what everybody else has seen, and to think what
nobody else has thought"
Albert Szent-Gyorgi