[Pw_forum] Fortran runtime error during scf in 4.0cvs3

Yaser Rehem yrehem at mac.com
Thu Apr 17 20:34:00 CEST 2008


It ran fine with 2 and 3 processors as well.  Is there some upper  
limit when the square of an integer issue kicks in?
On Apr 17, 2008, at 11:25 AM, Paolo Giannozzi wrote:

>
> On Apr 17, 2008, at 20:13 , Yaser Rehem wrote:
>
>> I upgraded to g95-0.91 and recompiled pwscf.  I still get the same
>> error messages when attempted the scf with 6 processors.  I played a
>> little and discovered that it runs fine with upto  4 processors, but
>> fails when I attempt to use more than 4.
>
> the problem exists only of the number of processors used in the
> parallel subspace diagonalization (must be the square of an integer)
> is not the same as the number of processor for R- and G-space
> parallelization. So 4=2x2 is fine, 16=4x4 is also, 6 is not
>
> Paolo
> ---
> Paolo Giannozzi, Dept of Physics, University of Udine
> via delle Scienze 208, 33100 Udine, Italy
> Phone +39-0432-558216, fax +39-0432-558222
>
>
>



More information about the Pw_forum mailing list