Skip to Content.
Sympa Menu

forum - Re: [abinit-forum] abinip compilation problems on itanium 2

forum@abinit.org

Subject: The ABINIT Users Mailing List ( CLOSED )

List archive

Re: [abinit-forum] abinip compilation problems on itanium 2


Chronological Thread 
  • From: "Jose C. Conesa" <jcconesa@icp.csic.es>
  • To: forum@abinit.org
  • Subject: Re: [abinit-forum] abinip compilation problems on itanium 2
  • Date: Tue, 31 May 2005 10:31:07 +0200
  • Priority: normal

Hello,
On 31 May 2005 at 6:45, kunj.tandon@geind.ge.com wrote:

> Dear all
>
> I use the abinit-4.4.4 source file to get abinip and abinis working on a Intel Itanium 2 server (Red Hat Linux Advanced Server release 2.1AS (Derry),  2.4.18-e.31smp kernel ).
>
> Copied the makefile_macros.IFC_mikami from the  Machine_dept_files directory and started compiling using "make abinip"
>
> This ends with a error saying "chmod: getting attributes of abinip, No such file or directory".
>
> Please let me know if I have to use different makefile_macros or different compiling options to get it working.
>
> I use Intel Fortran compiler 8.0.
>
> TIA
>
I compiled abinip in a similar machine (Itanium2 biprocessor with RedHat Advanced Server 3.0), using as well makefile_macros.IFC_mikami. The fortran compiler is Intel's efc. For implementation of parallelism, I had installed MPICH version 1.2.6. For the compilation of abinip, I just needed to make the following:

a) make a link to the system MPI mpi.f file, as said in the instructions. In my case this was made with:
  ln -s /usr/local/mpich-1.2.6/include/mpif.h mpif.h

b) In the makefile_macros:
  -adjust the fortran compiler definition (FC) to the mpif90 generated in the installation of MPI.
  -uncomment (remove #) the definitions of flag variables (FFLAGS, etc.) which contain -parallel, and comment out their alternatives
  -change, if necessary, the directory name mpich-intel for the appropriate one (in my case it was mpich-1.2.6) where it appears (3 times).

Then "make abinip" ran more or less OK; some warnings appeared but the result seems correct (it works).

Good luck,

--
Jose C. Conesa
Instituto de Catalisis y Petroleoquimica, CSIC
c/ Marie Curie 2, Campus de Cantoblanco, 28049 Madrid, Spain
Phone: (34)915854766    Fax:(34)915854766    e-mail:jcconesa@icp.csic.es



Archive powered by MHonArc 2.6.16.

Top of Page